You are on page 1of 428

RHEOLOGICAL METHODS IN

FOOD PROCESS ENGINEERING


Secon d Edit ion
Jame s F. St e ffe , Ph. D. , P. E.
Profes s or of Food Proces s Engineering
Dept. of Food Science and Human Nutrition
Dept. of Agricultural Engineering
Michigan State Univers ity
Fre e man Pre s s
2807 St ill Va lley Dr .
Ea s t La n s in g, MI 48823
USA
Pr of. J a mes F. St effe
209 Fa r r a ll Ha ll
Mich iga n St a t e Un iver s it y
Ea s t La n s in g, MI 48824-1323
USA
Ph on e: 517-353-4544
FAX: 517-432-2892
E-ma il: s t effe@ms u .edu
URL: www.egr .ms u .edu / ~s t effe/
Copyr igh t 1992, 1996 by J a mes F. St effe.
All r igh t s r es er ved. No pa r t of t h is wor k ma y be r epr odu ced,
s t or ed in a r et r ieva l s ys t em, or t r a n s mit t ed, in a n y for m or by a n y
mea n s , elect r on ic, mech a n ica l, ph ot ocopyin g, r ecor din g, or ot h er -
wis e, wit h ou t t h e pr ior wr it t en per mis s ion of t h e a u t h or .
Pr in t ed on a cid-fr ee pa per in t h e Un it ed St a t es of Amer ica
Secon d Pr in t in g
Libr a r y of Con gr es s Ca t a log Ca r d Nu mber : 96-83538
In t er n a t ion a l St a n da r d Book Nu mber : 0-9632036-1-4
Fre e man Pre s s
2807 St ill Va lley Dr .
Ea s t La n s in g, MI 48823
USA
Table of Cont e nt s
Preface ix
Chapter 1. Introduction to Rheology 1
1.1. Overview ...................................................................................... 1
1.2. Rheological Instruments for Fluids .......................................... 2
1.3. Stress and Strain .......................................................................... 4
1.4. Solid Behavior ............................................................................. 8
1.5. Fluid Behavior in Steady Shear Flow ....................................... 13
1.5.1. Time-Independent Material Functions ............................ 13
1.5.2. Time-Dependent Material Functions ............................... 27
1.5.3. Modeling Rheological Behavior of Fluids ....................... 32
1.6. Yield Stress Phenomena ............................................................. 35
1.7. Extensional Flow ......................................................................... 39
1.8. Viscoelastic Material Functions ................................................ 47
1.9. Attacking Problems in Rheological Testing ............................ 49
1.10. Interfacial Rheology ................................................................. 53
1.11. Electrorheology ......................................................................... 55
1.12. Viscometers for Process Control and Monitoring ................ 57
1.13. Empirical Measurement Methods for Foods ........................ 63
1.14. Example Problems .................................................................... 77
1.14.1. Carrageenan Gum Solution ............................................. 77
1.14.2. Concentrated Corn Starch Solution ................................ 79
1.14.3. Milk Chocolate .................................................................. 81
1.14.4. Falling Ball Viscometer for Honey ................................. 82
1.14.5. Orange Juice Concentrate ................................................ 86
1.14.6. Influence of the Yield Stress in Coating Food ............... 91
Chapter 2. Tube Viscometry 94
2.1. Introduction ................................................................................. 94
2.2. Rabinowitsch-Mooney Equation .............................................. 97
2.3. Laminar Flow Velocity Profiles ................................................ 103
2.4. Laminar Flow Criteria ................................................................ 107
2.5. Data Corrections ......................................................................... 110
2.6. Yield Stress Evaluation .............................................................. 121
2.7. Jet Expansion ............................................................................... 121
2.8. Slit Viscometry ............................................................................ 122
2.9. Glass Capillary (U-Tube) Viscometers .................................... 125
2.10. Pipeline Design Calculations .................................................. 128
2.11. Velocity Profiles In Turbulent Flow ....................................... 138
2.12. Example Problems .................................................................... 141
2.12.1. Conservation of Momentum Equations ........................ 141
2.12.2. Capillary Viscometry - Soy Dough ................................ 143
2.12.3. Tube Viscometry - 1.5% CMC ......................................... 146
2.12.4. Casson Model: Flow Rate Equation ............................... 149
2.12.5. Slit Viscometry - Corn Syrup .......................................... 150
2.12.6. Friction Losses in Pumping ............................................. 152
2.12.7. Turbulent Flow - Newtonian Fluid ................................ 155
2.12.8. Turbulent Flow - Power Law Fluid ................................ 156
v
Chapter 3. Rotational Viscometry 158
3.1. Introduction ................................................................................. 158
3.2. Concentric Cylinder Viscometry .............................................. 158
3.2.1. Derivation of the Basic Equation ...................................... 158
3.2.2. Shear Rate Calculations ...................................................... 163
3.2.3. Finite Bob in an Infinite Cup ............................................. 168
3.3. Cone and Plate Viscometry ....................................................... 169
3.4. Parallel Plate Viscometry (Torsional Flow) ............................ 172
3.5. Corrections: Concentric Cylinder ............................................. 174
3.6. Corrections: Cone and Plate, and Parallel Plate ..................... 182
3.7. Mixer Viscometry ....................................................................... 185
3.7.1. Mixer Viscometry: Power Law Fluids ............................. 190
3.7.2. Mixer Viscometry: Bingham Plastic Fluids ..................... 199
3.7.3. Yield Stress Calculation: Vane Method ........................... 200
3.7.4. Investigating Rheomalaxis ................................................ 208
3.7.5. Defining An Effective Viscosity ........................................ 210
3.8. Example Problems ...................................................................... 210
3.8.1. Bob Speed for a Bingham Plastic Fluid ............................ 210
3.8.2. Simple Shear in Power Law Fluids .................................. 212
3.8.3. Newtonian Fluid in a Concentric Cylinder ..................... 213
3.8.4. Representative (Average) Shear Rate .............................. 214
3.8.5. Concentric Cylinder Viscometer: Power Law Fluid ...... 216
3.8.6. Concentric Cylinder Data - Tomato Ketchup ................. 218
3.8.7. Infinite Cup - Single Point Test ......................................... 221
3.8.8. Infinite Cup Approximation - Power Law Fluid ........... 221
3.8.9. Infinite Cup - Salad Dressing ............................................ 223
3.8.10. Infinite Cup - Yield Stress Materials .............................. 225
3.8.11. Cone and Plate - Speed and Torque Range ................... 226
3.8.12. Cone and Plate - Salad Dressing ..................................... 227
3.8.13. Parallel Plate - Methylcellulose Solution ....................... 229
3.8.14. End Effect Calculation for a Cylindrical Bob ................ 231
3.8.15. Bob Angle for a Mooney-Couette Viscometer .............. 233
3.8.16. Viscous Heating ................................................................ 235
3.8.17. Cavitation in Concentric Cylinder Systems .................. 236
3.8.18. Mixer Viscometry .............................................................. 237
3.8.19. Vane Method - Sizing the Viscometer ........................... 243
3.8.20. Vane Method to Find Yield Stresses .............................. 244
3.8.21. Vane Rotation in Yield Stress Calculation .................... 247
3.8.22. Rheomalaxis from Mixer Viscometer Data ................... 250
Chapter 4. Extensional Flow 255
4.1. Introduction ................................................................................. 255
4.2. Uniaxial Extension ...................................................................... 255
4.3. Biaxial Extension ......................................................................... 258
4.4. Flow Through a Converging Die .............................................. 263
4.4.1. Cogswells Equations ......................................................... 264
4.4.2. Gibsons Equations ............................................................. 268
4.4.3. Empirical Method ............................................................... 271
4.5. Opposing Jets .............................................................................. 272
4.6. Spinning ....................................................................................... 274
4.7. Tubeless Siphon (Fano Flow) .................................................... 276
vi
4.8. Steady Shear Properties from Squeezing Flow Data ............. 276
4.8.1. Lubricated Squeezing Flow ............................................... 277
4.8.2. Nonlubricated Squeezing Flow ........................................ 279
4.9. Example Problems ...................................................................... 283
4.9.1. Biaxial Extension of Processed Cheese Spread ............... 283
4.9.2. Biaxial Extension of Butter ................................................ 286
4.9.3. 45 Converging Die, Cogswells Method ........................ 287
4.9.4. 45 Converging Die, Gibsons Method ............................ 289
4.9.5. Lubricated Squeezing Flow of Peanut Butter ................. 291
Chapter 5. Viscoelasticity 294
5.1. Introduction ................................................................................. 294
5.2. Transient Tests for Viscoelasticity ............................................ 297
5.2.1. Mechanical Analogues ....................................................... 298
5.2.2. Step Strain (Stress Relaxation) .......................................... 299
5.2.3. Creep and Recovery ........................................................... 304
5.2.4. Start-Up Flow (Stress Overshoot) ..................................... 310
5.3. Oscillatory Testing ...................................................................... 312
5.4. Typical Oscillatory Data ............................................................ 324
5.5. Deborah Number ........................................................................ 332
5.6. Experimental Difficulties in Oscillatory Testing of Food ..... 336
5.7. Viscometric and Linear Viscoelastic Functions ...................... 338
5.8. Example Problems ...................................................................... 341
5.8.1. Generalized Maxwell Model of Stress Relaxation ........ 341
5.8.2. Linearized Stress Relaxation ............................................. 342
5.8.3. Analysis of Creep Compliance Data ................................ 343
5.8.4. Plotting Oscillatory Data ................................................... 348
6. Appendices 350
6.1. Conversion Factors and SI Prefixes .......................................... 350
6.2. Greek Alphabet ........................................................................... 351
6.3. Mathematics: Roots, Powers, and Logarithms ....................... 352
6.4. Linear Regression Analysis of Two Variables ........................ 353
6.5. Hookean Properties .................................................................... 357
6.6. Steady Shear and Normal Stress Difference ........................... 358
6.7. Yield Stress of Fluid Foods ........................................................ 359
6.8. Newtonian Fluids ....................................................................... 361
6.9. Dairy, Fish and Meat Products ................................................. 366
6.10. Oils and Miscellaneous Products ........................................... 367
6.11. Fruit and Vegetable Products ................................................. 368
6.12. Polymer Melts ........................................................................... 371
6.13. Cosmetic and Toiletry Products ............................................. 372
6.14. Energy of Activation for Flow for Fluid Foods .................... 374
6.15. Extensional Viscosities of Newtonian Fluids ........................ 375
6.16. Extensional Viscosities of Non-Newtonian Fluids .............. 376
6.17. Fanning Friction Factors: Bingham Plastics .......................... 377
6.18. Fanning Friction Factors: Power Law Fluids ........................ 378
6.19. Creep (Burgers Model) of Salad Dressing ............................. 379
6.20. Oscillatory Data for Butter ...................................................... 380
6.21. Oscillatory Data Iota-Carrageenan Gel ................................. 381
6.22. Storage and Loss Moduli of Fluid Foods .............................. 382

vii
Nomenclature ......................................................................................... 385
Bibliography ........................................................................................... 393
Index ......................................................................................................... 412
viii
Pre fac e
Gr owt h and development of t his wor k spr ang fr om t he need t o
pr ovide educat ional mat er ial for food engineer s and food scient ist s. The
fir st edit ion was conceived as a t ext book and t he wor k cont inues t o be
used in gr aduat e level cour ses at var ious univer sit ies. It s gr eat est
appeal, however , was t o individuals solving pr act ical day-t o-day pr ob-
lems. Hence, t he second edit ion, a significant ly expanded and r evised
ver sion of t he or iginal wor k, is aimed pr imar ily at t he r heological
pr act it ioner (par t icular ly t he indust r ial pr act it ioner ) seeking a br oad
under st anding of t he subject mat t er . The over all goal of t he t ext is t o
pr esent t he infor mat ion needed t o answer t hr ee quest ions when facing
pr oblems in food r heology: 1. What pr oper t ies should be measur ed? 2.
What t ype and degr ee of defor mat ion should be induced in t he mea-
sur ement ? 3. How should exper iment al dat a be analyzed t o gener at e
pr act ical infor mat ion? Alt hough t he main focus of t he book is food,
scient ist s and engineer s in ot her fields will find t he wor k a convenient
r efer ence for st andar d r heological met hods and t ypical dat a.
Over all, t he wor k pr esent s t he t heor y of r heological t est ing and
pr ovides t he analyt ical t ools needed t o det er mine r heological pr oper t ies
fr om exper iment al dat a. Met hods appr opr iat e for common food indust r y
applicat ions ar e pr esent ed. All st andar d measur ement t echniques for
fluid and semi-solid foods ar e included. Select ed met hods for solids ar e
also pr esent ed. Result s fr om numer ous fields, par t icular ly polymer
r heology, ar e used t o addr ess t he flow behavior of food. Mat hemat ical
r elat ionships, der ived fr om simple for ce balances, pr ovide a funda-
ment al view of r heological t est ing. Only a backgr ound in basic calculus
and element ar y st at ist ics (mainly r egr ession analysis) is needed t o
under st and t he mat er ial. The t ext cont ains numer ous pr act ical example
pr oblems, involving act ual exper iment al dat a, t o enhance compr ehen-
sion and t he execut ion of concept s pr esent ed. This feat ur e makes t he
wor k convenient for self-st udy.
Specific explanat ions of key t opics in r heology ar e pr esent ed in
Chapt er 1: basic concept s of st r ess and st r ain; elast ic solids showing
Hookean and non-Hookean behavior ; viscomet r ic funct ions including
nor mal st r ess differ ences; modeling fluid behavior as a funct ion of shear
r at e, t emper at ur e, and composit ion; yield st r ess phenomena, ext en-
sional flow; and viscoelast ic behavior . Efficient met hods of at t acking
pr oblems and t ypical inst r ument s used t o measur e fluid pr oper t ies ar e
discussed along wit h an examinat ion of pr oblems involving int er facial
ix
r heology, elect r or heology, and on-line viscomet r y for cont r ol and mon-
it or ing of food pr ocessing oper at ions. Common met hods and empir ical
inst r ument s ut ilized in t he food indust r y ar e also int r oduced: Text ur e
Pr ofile Analysis, Compr ession-Ext r usion Cell, War ner -Br at zler Shear
Cell, Bost wick Consist omet er , Adams Consist omet er , Amylogr aph,
Far inogr aph, Mixogr aph, Ext ensigr aph, Alveogr aph, Kr amer Shear
Cell, Br ookfield disks and T-bar s, Cone Penet r omet er , Hoeppler Vis-
comet er , Zhan Viscomet er , Br abender -FMC Consist omet er .
The basic equat ions of t ube (or capillar y) viscomet r y, such as t he
Rabinowit sch-Mooney equat ion, ar e der ived and applied in Chapt er 2.
Laminar flow cr it er ia and velocit y pr ofiles ar e evaluat ed along wit h dat a
cor r ect ion met hods for many sour ces of er r or : kinet ic ener gy losses, end
effect s, slip (wall effect s), viscous heat ing, and hole pr essur e. Tech-
niques for glass capillar y and slit viscomet er s ar e consider ed in det ail.
A sect ion on pipeline design calculat ions has been included t o facilit at e
t he const r uct ion of lar ge scale t ube viscomet er s and t he design of fluid
pumping syst ems.
A gener al for mat , analogous t o t hat used in Chapt er 2, is cont inued
in Chapt er 3 t o pr ovide cont inuit y in subject mat t er development . The
main foci of t he chapt er cent er ar ound t he t heor et ical pr inciples and
exper iment al pr ocedur es r elat ed t o t hr ee t r adit ional t ypes of r ot at ional
viscomet er s: concent r ic cylinder , cone and plat e, and par allel plat e.
Mat hemat ical analyses of dat a ar e discussed in det ail. Er r or s due t o
end effect s, viscous heat ing, slip, Taylor vor t ices, cavit at ion, and cone
t r uncat ions ar e invest igat ed. Numer ous met hods in mixer viscomet r y,
t echniques having significant pot ent ial t o solve many food r heology
pr oblems, ar e also pr esent ed: slope and mat ching viscosit y met hods t o
evaluat e aver age shear r at e, det er minat ion of power law and Bingham
plast ic fluid pr oper t ies. The vane met hod of yield st r ess evaluat ion,
using bot h t he slope and single point met hods, along wit h a consider -
at ion of vane r ot at ion dur ing t est ing, is explor ed in det ail.
The exper iment al met hods t o det er mine ext ensional viscosit y ar e
explained in Chapt er 4. Techniques pr esent ed involve uniaxial ext en-
sion bet ween r ot at ing clamps, biaxial ext ensional flow achieved by
squeezing mat er ial bet ween lubr icat ed par allel plat es, opposing jet s,
spinning, and t ubeless siphon (Fano) flow. Relat ed pr ocedur es,
involving lubr icat ed and nonlubr icat ed squeezing, t o det er mine shear
flow behavior ar e also pr esent ed. Calculat ing ext ensional viscosit y fr om
flows t hr ough t aper ed conver gences and flat ent r y dies is given a
t hor ough examinat ion.
x
Essent ial concept s in viscoelast icit y and st andar d met hods of
invest igat ing t he phenomenon ar e invest igat ed in Chapt er 5. The full
scope of viscoelast ic mat er ial funct ions det er mined in t r ansient and
oscillat or y t est ing ar e discussed. Mechanical analogues of r heological
behavior ar e given as a means of analyzing cr eep and st r ess r elaxat ion
dat a. Theor et ical aspect s of oscillat or y t est ing, t ypical dat a, and a
discussion of t he var ious modes of oper at ing commer cial inst r ument s
-st r ain, fr equency, t ime, and t emper at ur e sweep modes- ar e pr esent ed.
The Debor ah number concept , and how it can be used t o dist inguish
liquid fr om solid-like behavior , is int r oduced. St ar t -up flow (st r ess
over shoot ) and t he r elat ionship bet ween st eady shear and oscillat or y
pr oper t ies ar e also discussed. Conver sion fact or s, mat hemat ical r ela-
t ionships, linear r egr ession analysis, and t ypical r heological dat a for
food as well as cosmet ics and polymer s ar e pr ovided in t he Appendices.
Nomenclat ur e is convenient ly summar ized at t he end of t he t ext and a
lar ge bibliogr aphy is fur nished t o dir ect r eader s t o addit ional infor -
mat ion.
J .F. St effe
J une, 1996
xi
De dic at ion
To Susan, J ust inn, and Dana.
xiii
Chapt e r 1 . Int roduc t ion t o Rhe ology
1.1. Overvi ew
The fir st use of t he wor d "r heology" is cr edit ed t o Eugene C. Bingham
(cir ca 1928) who also descr ibed t he mot t o of t he subject as
("pant a r hei," fr om t he wor ks of Her aclit us, a pr e-Socr at ic Gr eek phi-
losopher act ive about 500 B.C.) meaning "ever yt hing flows" (Reiner ,
1964). Rheology is now well est ablished as t he science of t he defor mat ion
and flow of mat t er : It is t he st udy of t he manner in which mat er ials
r espond t o applied st r ess or st r ain. All mat er ials have r heological
pr oper t ies and t he ar ea is r elevant in many fields of st udy: geology and
mining (Cr ist escu, 1989), concr et e t echnology (Tat t er sall and Banfill,
1983), soil mechanics (Haghighi et al., 1987; Vyalov, 1986), plast ics
pr ocessing (Dealy and Wissbur n, 1990), polymer s and composit es
(Neilsen and Landel, 1994; Yanovsky, 1993), t r ibology (st udy of lubr i-
cat ion, fr ict ion and wear ), paint flow and pigment disper sion (Pat t on,
1964), blood (Dint enfass, 1985), bioengineer ing (Skalak and Chien,
1987), int er facial r heology (Edwar ds et al., 1991), st r uct ur al mat er ials
(Callist er , 1991), elect r or heology (Block and Kelly, 1988), psychor -
heology (Dr ake, 1987), cosmet ics and t oilet r ies (Laba, 1993b), and
pr essur e sensit ive adhesion (Saunder s et al., 1992). The focus of t his
wor k is food wher e under st anding r heology is cr it ical in opt imizing
pr oduct development effor t s, pr ocessing met hodology and final pr oduct
qualit y. To t he ext ent possible, st andar d nomenclat ur e (Dealy, 1994)
has been used in t he t ext .
One can t hink of food r heology as t he mat er ial science of food. Ther e
ar e numer ous ar eas wher e r heological dat a ar e needed in t he food
indust r y:
a. Pr ocess engineer ing calculat ions involving a wide r ange of equip-
ment such as pipelines, pumps, ext r uder s, mixer s, coat er s, heat
exchanger s, homogenizer s, calender s, and on-line viscomet er s;
b. Det er mining ingr edient funct ionalit y in pr oduct development ;
c. Int er mediat e or final pr oduct qualit y cont r ol;
d. Shelf life t est ing;
e. Evaluat ion of food t ext ur e by cor r elat ion t o sensor y dat a;
f. Analysis of r heological equat ions of st at e or const it ut ive equat ions.
Many of t he unique r heological pr oper t ies of var ious foods have been
summar ized in books by Rao and St effe (1992), and Weiper t et al. (1993).

2 Chapter 1. Introduction to Rheology
Fundament al r heological pr oper t ies ar e independent of t he inst r u-
ment on which t hey ar e measur ed so differ ent inst r ument s will yield
t he same r esult s. This is an ideal concept and differ ent inst r ument s
r ar ely yield ident ical r esult s; however , t he idea is one which dist in-
guishes t r ue r heological mat er ial pr oper t ies fr om subject ive (empir ical
and gener ally inst r ument dependent , t hough fr equent ly useful)
mat er ial char act er izat ions. Examples of inst r ument s giving subject ive
r esult s include t he following (Bour ne, 1982): Far inogr aph, Mixogr aph,
Ext ensogr aph, Viscoamlyogr aph, and t he Bost wick Consist omet er .
Empir ical t est ing devices and met hods, including Text ur e Pr ofile
Analysis, ar e consider ed in mor e det ail in Sec. 1.13.
1.2. Rheologi cal Instruments for Flui ds
Common inst r ument s, capable of measur ing fundament al r heolog-
ical pr oper t ies of fluid and semi-solid foods, may be placed int o t wo
gener al cat egor ies (Fig. 1.1): r ot at ional t ype and t ube t ype. Most ar e
commer cially available, ot her s (mixer and pipe viscomet er s) ar e easily
fabr icat ed. Cost s var y t r emendously fr om t he inexpensive glass capil-
lar y viscomet er t o a ver y expensive r ot at ional inst r ument capable of
measur ing dynamic pr oper t ies and nor mal st r ess differ ences. Solid
foods ar e oft en t est ed in compr ession (bet ween par allel plat es), t ension,
or t or sion. Inst r ument s which measur e r heological pr oper t ies ar e called
r heomet er s. Viscomet er is a mor e limit ing t er m r efer r ing t o devices
t hat only measur e viscosit y.
Rot at ional inst r ument s may be oper at ed in t he st eady shear (con-
st ant angular velocit y) or oscillat or y (dynamic) mode. Some r ot at ional
inst r ument s funct ion in t he cont r olled st r ess mode facilit at ing t he
collect ion of cr eep dat a, t he analysis of mat er ials at ver y low shear r at es,
and t he invest igat ion of yield st r esses. This infor mat ion is needed t o
under st and t he int er nal st r uct ur e of mat er ials. The cont r olled r at e
mode is most useful in obt aining dat a r equir ed in pr ocess engineer ing
calculat ions. Mechanical differ ences bet ween cont r olled r at e and con-
t r olled st r ess inst r ument s ar e discussed in Sec. 3.7.3. Rot at ional sys-
t ems ar e gener ally used t o invest igat e t ime-dependent behavior because
t ube syst ems only allow one pass of t he mat er ial t hr ough t he appar at us.
A det ailed discussion of oscillat or y t est ing, t he pr imar y met hod of
det er mining t he viscoelast ic behavior of food, is pr ovided in Chapt er 5.
1.2 Rheological Instruments for Fluids 3
Figure 1.1. Common rheological instruments divided into two major categories:
rotational and tube type.
Ther e ar e advant ages and disadvant ages associat ed wit h each
inst r ument . Gr avit y oper at ed glass capillar ies, such as t he Cannon-
Fenske t ype shown in Fig. 1.1, ar e only suit able for Newt onian fluids
because t he shear r at e var ies dur ing dischar ge. Cone and plat e syst ems
ar e limit ed t o moder at e shear r at es but calculat ions (for small cone
angles) ar e simple. Pipe and mixer viscomet er s can handle much lar ger
par t icles t han cone and plat e, or par allel plat e, devices. Pr oblems
associat ed wit h slip and degr adat ion in st r uct ur ally sensit ive mat er ials
ar e minimized wit h mixer viscomet er s. High pr essur e capillar ies
oper at e at high shear r at es but gener ally involve a significant end
pr essur e cor r ect ion. Pipe viscomet er s can be const r uct ed t o wit hst and
t he r igor s of t he pr oduct ion or pilot plant envir onment .
Rotational Type
Parallel Plate Cone and Plate
Concentric Cylinder
Mixer
Tube Type
Pipe Glass Capillary High Pressure Capillary
4 Chapter 1. Introduction to Rheology
All t he inst r ument s pr esent ed in Fig. 1.1 ar e "volume loaded" devices
wit h cont ainer dimensions t hat ar e cr it ical in t he det er minat ion of
r heological pr oper t ies. Anot her common t ype of inst r ument , known as
a vibr at ional viscomet er , uses t he pr inciple of "sur face loading" wher e
t he sur face of an immer sed pr obe (usually a spher e or a r od) gener at es
a shear wave which dissipat es in t he sur r ounding medium. A lar ge
enough cont ainer is used so shear for ces do not r each t he wall and r eflect
back t o t he pr obe. Measur ement s depend only on abilit y of t he sur -
r ounding fluid t o damp pr obe vibr at ion. The damping char act er ist ic of
a fluid is a funct ion of t he pr oduct of t he fluid viscosit y (of Newt onian
fluids) and t he densit y. Vibr at ional viscomet er s ar e popular as in-line
inst r ument s for pr ocess cont r ol syst ems (see Sec. 1.12). It is difficult t o
use t hese unit s t o evaluat e fundament al r heological pr oper t ies of non-
Newt onian fluids (Fer r y, 1977), but t he subject ive r esult s obt ained oft en
cor r elat e well wit h impor t ant food qualit y at t r ibut es. The coagulat ion
t ime and cur d fir mness of r ennet ed milk, for example, have been suc-
cessfully invest igat ed using a vibr at ional viscomet er (Shar ma et al.,
1989).
Inst r ument s used t o evaluat e t he ext ensional viscosit y of mat er ials
ar e discussed in Chapt er 4. Pulling or st r et ching a sample bet ween
t oot hed gear s, sucking mat er ial int o opposing jet s, spinning, or
exploit ing t he open siphon phenomenon can gener at e dat a for calcu-
lat ing t ensile ext ensional viscosit y. Infor mat ion t o det er mine biaxial
ext ensional viscosit y is pr ovided by compr essing samples bet ween
lubr icat ed par allel plat es. Shear viscosit y can also be evaluat ed fr om
unlubr icat ed squeezing flow bet ween par allel plat es. A number of
met hods ar e available t o calculat e an aver age ext ensional viscosit y fr om
dat a descr ibing flow t hr ough a conver gence int o a capillar y die or slit .
1.3. Stress and Strai n
Since r heology is t he st udy of t he defor mat ion of mat t er , it is essent ial
t o have a good under st anding of st r ess and st r ain. Consider a r ect an-
gular bar t hat , due t o a t ensile for ce, is slight ly elongat ed (Fig. 1.2). The
init ial lengt h of t he bar is and t he elongat ed lengt h is wher e
wit h r epr esent ing t he incr ease in lengt h. This defor mat ion
may be t hought of in t er ms of Cauchy st r ain (also called engineer ing
st r ain):
L
o
L
L L
o
+ L L
1.3 Stress and Strain 5
[1.1]
or Hencky st r ain (also called t r ue st r ain) which is det er mined by
evaluat ing an int egr al fr om t o :
[1.2]
Figure 1.2. Linear extension of a rectangular bar.
Cauchy and Hencky st r ains ar e bot h zer o when t he mat er ial is
unst r ained and appr oximat ely equal at small st r ains. The choice of
which st r ain measur e t o use is lar gely a mat t er of convenience and one
can be calculat ed fr om t he ot her :
[1.3]
is pr efer r ed for calculat ing st r ain r esult ing fr om a lar ge defor mat ion.
Anot her t ype of defor mat ion commonly found in r heology is simple
shear . The idea can be illust r at ed wit h a r ect angular bar (Fig. 1.3) of
height . The lower sur face is st at ionar y and t he upper plat e is linear ly
displaced by an amount equal t o . Each element is subject t o t he same
level of defor mat ion so t he size of t he element is not r elevant . The angle
of shear , , may be calculat ed as

c

L
L
o

L L
o
L
o

L
L
o
1
L
o
L

L
o
L
dL
L
ln(L/L
o
)
L L
0

h
ln(1 +
c
)

h
h
L

6 Chapter 1. Introduction to Rheology


[1.4]
Wit h small defor mat ions, t he angle of shear (in r adians) is equal t o t he
shear st r ain (also symbolized by ), .
Figure 1.3. Shear deformation of a rectangular bar.
Figure 1.4. Typical stresses on a material element.
St r ess, defined as a for ce per unit ar ea and usually expr essed in
Pascal (N/m
2
), may be t ensile, compr essive, or shear . Nine separ at e
quant it ies ar e r equir ed t o complet ely descr ibe t he st at e of st r ess in a
mat er ial. A small element (Fig. 1.4) may be consider ed in t er ms of
tan()
L
h
tan
h
L
11
21
23
22
33
1
2
3
x
x
x
1.3 Stress and Strain 7
Car t esian coor dinat es ( ). St r ess is indicat ed as wher e t he fir st
subscr ipt r efer s t o t he or ient at ion of t he face upon which t he for ce act s
and t he second subscr ipt r efer s t o t he dir ect ion of t he for ce. Ther efor e,
is a nor mal st r ess act ing in t he plane per pendicular t o in t he
dir ect ion of and is a shear st r ess act ing in t he plane per pendicular
t o in t he dir ect ion of . Nor mal st r esses ar e consider ed posit ive when
t hey act out war d (act ing t o cr eat e a t ensile st r ess) and negat ive when
t hey act inwar d (act ing t o cr eat e a compr essive st r ess).
St r ess component s may be summar ized as a st r ess t ensor wr it t en
in t he for m of a mat r ix:
[1.5]
A r elat ed t ensor for st r ain can also be expr essed in mat r ix for m. Basic
laws of mechanics, consider ing t he moment about t he axis under
equilibr ium condit ions, can be used t o pr ove t hat t he st r ess mat r ix is
symmet r ical:
[1.6]
so
[1.7]
[1.8]
[1.9]
meaning t her e ar e only six independent component s in t he st r ess t ensor
r epr esent ed by Eq. [1.5].
Equat ions t hat show t he r elat ionship bet ween st r ess and st r ain ar e
eit her called r heological equat ions of st at e or const it ut ive equat ions. In
complex mat er ials t hese equat ions may include ot her var iables such as
t ime, t emper at ur e, and pr essur e. A modulus is defined as t he r at io of
st r ess t o st r ain while a compliance is defined as t he r at io of st r ain t o
st r ess. The wor d r heogr am r efer s t oa gr aph of a r heological r elat ionship.
x
1
, x
2
, x
3

ij

11
x
1
x
1

23
x
2
x
3

ij

11

12

13

21

22

23

31

32

33
1
1
1
1
]

ij

ji

12

21

31

13

32

23
8 Chapter 1. Introduction to Rheology
1.4. Soli d Behavi or
When for ce is applied t o a solid mat er ial and t he r esult ing st r ess
ver sus st r ain cur ve is a st r aight line t hr ough t he or igin, t he mat er ial is
obeying Hookes law. The r elat ionship may be st at ed for shear st r ess
and shear st r ain as
[1.10]
wher e is t he shear modulus. Hookean mat er ials do not flow and ar e
linear ly elast ic. St r ess r emains const ant unt il t he st r ain is r emoved
and t he mat er ial r et ur ns t o it s or iginal shape. Somet imes shape
r ecover y, t hough complet e, is delayed by some at omist ic pr ocess. This
t ime-dependent , or delayed, elast ic behavior is known as anelast icit y.
Hookes law can be used t o descr ibe t he behavior of many solids (st eel,
egg shell, dr y past a, har d candy, et c.) when subject ed t o small st r ains,
t ypically less t han 0.01. Lar ge st r ains oft en pr oduce br it t le fr act ur e or
non-linear behavior .
The behavior of a Hookean solid may be invest igat ed by st udying
t he uniaxial compr ession of a cylindr ical sample (Fig. 1.5). If a mat er ial
is compr essed so t hat it exper iences a change in lengt h and r adius, t hen
t he nor mal st r ess and st r ain may be calculat ed:
[1.11]
[1.12]
Figure 1.5 Uniaxial compression of a cylindrical sample.

12
G
G

22

F
A

F
(R
o
)
2

c

h
h
o
R
h
h
o
R R
h
Initial Shape Compressed Shape
o
o
1.4 Solid Behavior 9
This infor mat ion can be used t o det er mine Youngs modulus ( ), also
called t he modulus of elast icit y, of t he sample:
[1.13]
If t he defor mat ions ar e lar ge, Hencky st r ain ( ) should be used t o
calculat e st r ain and t he ar ea t er m needed in t he st r ess calculat ion
should be adjust ed for t he change in r adius caused by compr ession:
[1.14]
A cr it ical assumpt ion in t hese calculat ions is t hat t he sample r emains
cylindr ical in shape. For t his r eason lubr icat ed cont act sur faces ar e
oft en r ecommended when t est ing mat er ials such as food gels.
Youngs modulus may also be det er mined by flexur al t est ing of
beams. In one such t est , a cant ilever beam of known lengt h (a) is
deflect ed a dist ance (d) when a for ce (F) is applied t o t he fr ee end of t he
beam. This infor mat ion can be used t o calculat e Youngs modulus for
mat er ials having a r ect angular or cir cular cr ossect ional ar ea (Fig. 1.6).
Similar calculat ions can be per for med in a t hr ee-point bending t est (Fig.
1.7) wher e deflect ion (d) is measur ed when a mat er ial is subject ed t o a
for ce (F) placed midway bet ween t wo suppor t s. Calculat ions ar e sight ly
differ ent depending on wet her -or -not t he t est mat er ial has a r ect angular
or cir cular shape. Ot her simple beam t est s, such as t he double cant ilever
or four -point bending t est , yield compar able r esult s. Flexur al t est ing
may have applicat ion t o solid foods having a well defined geomet r y such
as dr y past a or har d candy.
In addit ion t o Youngs modulus, Poissons r at io ( ) can be defined
fr om compr ession dat a (Fig. 1.5):
[1.15]
Poissons r at io may r ange fr om 0 t o 0.5. Typically, var ies fr om 0.0 for
r igid like mat er ials cont aining lar ge amount s of air t o near 0.5 for liquid
like mat er ials. Values fr om 0.2 t o 0.5 ar e common for biological
mat er ials wit h 0.5 r epr esent ing an incompr essible subst ance like pot at o
E
E

22

22

F
(R
o
+ R)
2


lateral strain
axial strain

R/R
o
h/h
o

10 Chapter 1. Introduction to Rheology


flesh. Tissues wit h a high level of cellular gas, such as apple flesh, would
have values closer t o 0.2. Met als usually have Poisson r at ios bet ween
0.25 and 0.35.
Figure 1.6 Deflection of a cantilever beam to determine Youngs modulus.
Figure 1.7 Three-point beam bending test to determine Youngs modulus (b, h,
and D are defined on Fig. 1.6).
E = 4Fa /(dbh )
E = 64Fa /(3d D )
Rectangular Crossection Circular Crossection
3 3 4 3
h
b D
d
F
a
E = Fa /(4dbh )
E = 4Fa /(3d D )
Rectangular Crossection Circular Crossection
4 3
3 3
F
d
a
1.4 Solid Behavior 11
If a mat er ial is subject ed t o a unifor m change in ext er nal pr essur e,
it may exper ience a volumet r ic change. These quant it ies ar e used t o
define t he bulk modulus ( ):
[1.16]
The bulk modulus of dough is appr oximat ely Pa while t he value for
st eel is Pa. Anot her common pr oper t y, bulk compr essibilit y, is
defined as t he r ecipr ocal of bulk modulus.
When t wo mat er ial const ant s descr ibing t he behavior of a Hookean
solid ar e known, t he ot her t wo can be calculat ed using any of t he fol-
lowing t heor et ical r elat ionships:
[1.17]
[1.18]
[1.19]
Numer ous exper iment al t echniques, applicable t o food mat er ials, may
be used t o det er mine Hookean mat er ial const ant s. Met hods include
t est ing in t ension, compr ession and t or sion (Mohsenin, 1986; Pola-
kowski and Ripling, 1966; Dally and Ripley, 1965). Hookean pr oper t ies
of t ypical mat er ials ar e pr esent ed in t he Appendix [6.5].
Linear -elast ic and non-linear elast ic mat er ials (like r ubber ) bot h
r et ur n t o t heir or iginal shape when t he st r ain is r emoved. Food may
be solid in nat ur e but not Hookean. A compar ison of cur ves for linear
elast ic (Hookean), elast oplast ic and non-linear elast ic mat er ials (Fig.
1.8) shows a number of similar it ies and differ ences. The elast oplast ic
mat er ial has Hookean t ype behavior below t he yield st r ess ( ) but flows
t o pr oduce per manent defor mat ion above t hat value. Mar gar ine and
but t er , at r oom t emper at ur e, may behave as elast oplast ic subst ances.
Invest igat ion of t his t ype of mat er ial, as a solid or a fluid, depends on
t he shear st r ess being above or below (see Sec. 1.6 for a mor e det ailed
discussion of t he yield st r ess concept and Appendix [6.7] for t ypical yield
st r ess values). Fur t her mor e, t o fully dist inguish fluid fr om solid like
behavior , t he char act er ist ic t ime of t he mat er ial and t he char act er ist ic
t ime of t he defor mat ion pr ocess involved must be consider ed simult a-
K
K
change in pressure
volumetric strain

change in pressure
(change in volume/original volume)
10
6
10
11
1
E

1
3G
+
1
9K
E 3K(1 2) 2G(1 + )

3K E
6K

E 2G
2G

o
12 Chapter 1. Introduction to Rheology
neously. The Debor ah number has been defined t o addr ess t his issue.
A det ailed discussion of t he concept , including an example involving
silly put t y (t he "r eal solid-liquid") is pr esent ed in Sec. 5.5.
Figure 1.8. Deformation curves for linear elastic (Hookean), elastoplastic and
non-linear elastic materials.
Food r heologist s also find t he failur e behavior of solid food (par t ic-
ular ly, br it t le mat er ials and fir m gels) t o be ver y useful because t hese
dat a somet imes cor r elat e well wit h t he conclusions of human sensor y
panels (Hamann, 1983; Mont ejano et al., 1985; Kawanar i et al., 1981).
The following t er minology (t aken fr om Amer ican Societ y for Test ing and
Mat er ials, St andar d E-6) is useful in descr ibing t he lar ge defor mat ion
behavior involved in t he mechanical failur e of food:
elast ic limit - t he gr eat est st r ess which a mat er ial is capable of sus-
t aining wit hout any per manent st r ain r emaining upon complet e
r elease of st r ess;
pr opor t ional limit - t he gr eat est st r ess which a mat er ial is capable of
sust aining wit hout any deviat ion fr om Hookes Law;
compr essive st r engt h - t he maximum compr essive st r ess a mat er ial
is capable of sust aining;
shear st r engt h - t he maximum shear st r ess a mat er ial is capable of
sust aining;
t ensile st r engt h - t he maximum t ensile st r ess a mat er ial is capable
of sust aining;
yield point - t he fir st st r ess in a t est wher e t he incr ease in st r ain
occur s wit hout an incr ease in st r ess;
o
Linear Elastic Elastoplastic Non-Linear Elastic
Permanent
Deformation
12 12 12
1.5.1 Time-Independent Material Functions 13
yield st r engt h - t he engineer ing st r ess at which a mat er ial exhibit s
a specified limit ing deviat ion fr om t he pr opor t ionalit y of st r ess t o
st r ain.
A t ypical char act er ist ic of br it t le solids is t hat t hey br eak when given
a small defor mat ion. Failur e t est ing and fr act ur e mechanics in st r uc-
t ur al solids ar e well developed ar eas of mat er ial science (Callist er , 1991)
which offer much t o t he food r heologist . Evaluat ing t he st r uct ur al
failur e of solid foods in compr ession, t or sion, and sandwich shear modes
wer e summar ized by Hamann (1983). J agged for ce-defor mat ion r ela-
t ionships of cr unchy mat er ials may offer alt er nat ive t ext ur e classifi-
cat ion cr it er ia for br it t le or cr unchy foods (Ulbr icht et al., 1995; Peleg
and Nor mand, 1995).
1.5. Flui d Behavi or i n Steady Shear Flow
1.5.1. Ti me-Independent Materi al Functi ons
Vi scometri c Functi ons. Fluids may be st udied by subject ing t hem
t o cont inuous shear ing at a const ant r at e. Ideally, t his can be accom-
plished using t wo par allel plat es wit h a fluid in t he gap bet ween t hem
(Fig. 1.9). The lower plat e is fixed and t he upper plat e moves at a
const ant velocit y ( ) which can be t hought of as an incr ement al change
in posit ion divided by a small t ime per iod, . A for ce per unit ar ea
on t he plat e is r equir ed for mot ion r esult ing in a shear st r ess ( ) on
t he upper plat e which, concept ually, could also be consider ed t o be a
layer of fluid.
The flow descr ibed above is st eady simple shear and t he shear r at e
(also called t he st r ain r at e) is defined as t he r at e of change of st r ain:
[1.20]
This definit ion only applies t o st r eamline (laminar ) flow bet ween
par allel plat es. It is dir ect ly applicable t o sliding plat e viscomet er
descr ibed by Dealy and Giacomin (1988). The definit ion must be
adjust ed t o account for cur ved lines such as t hose found in t ube and
r ot at ional viscomet er s; however , t he idea of "maximum speed divided
by gap size" can be useful in est imat ing shear r at es found in par t icular
applicat ions like br ush coat ing. This idea is explor ed in mor e det ail in
Sec. 1.9.
u
L/t

21


d
dt

d
dt

L
h
_

,

u
h
14 Chapter 1. Introduction to Rheology
Figure 1.9. Velocity profile between parallel plates.
Rheological t est ing t o det er mine st eady shear behavior is conduct ed
under laminar flow condit ions. In t ur bulent flow, lit t le infor mat ion is
gener at ed t hat can be used t o det er mine mat er ial pr oper t ies. Also, t o
be meaningful, dat a must be collect ed over t he shear r at e r ange
appr opr iat e for t he pr oblem in quest ion which may var y widely in
indust r ial pr ocesses (Table 1.1): Sediment at ion of par t icles may involve
ver y low shear r at es, spr ay dr ying will involve high shear r at es, and
pipe flow of food will usually occur over a moder at e shear r at e r ange.
Ext r apolat ing exper iment al dat a over a br oad r ange of shear r at es is
not r ecommended because it may int r oduce significant er r or s when
evaluat ing r heological behavior .
Mat er ial flow must be consider ed in t hr ee dimensions t o complet ely
descr ibe t he st at e of st r ess or st r ain. In st eady, simple shear flow t he
coor dinat e syst em may be or ient ed wit h t he dir ect ion of flow sot he st r ess
t ensor given by Eq. [1.5] r educes t o
[1.21]
h
VELOCITY PROFILE
FORCE
AREA
u
u = 0
1
2
x
x

ij

11

12
0

21

22
0
0 0
33
1
1
1
1
]
1.5.1 Time-Independent Material Functions 15
Table 1.1. Shear Rates Typical of Familiar Materials and Processes
Sit u a t ion (1/ s ) Applica t ion
Sedimen t a t ion of pa r t icles in 10
-6
- 10
-3
Medicin es , pa in t s , s pices in
a s u s pen din g liqu id s a la d dr es s in g
Levelin g du e t o s u r fa ce t en - 10
-2
- 10
-1
Fr os t in g, pa in t s , pr in t in g in ks
s ion
Dr a in in g u n der gr a vit y 10
-1
- 10
1
Va t s , s ma ll food con t a in er s ,
pa in t in g a n d coa t in g
Ext r u s ion 10
0
- 10
3
Sn a ck a n d pet foods , t oot h -
pa s t e, cer ea ls , pa s t a , poly-
mer s
Ca len der in g 10
1
- 10
2
Dou gh Sh eet in g
Pou r in g fr om a bot t le 10
1
- 10
2
Foods , cos met ics , t oilet r ies
Ch ewin g a n d s wa llowin g 10
1
- 10
2
Foods
Dip coa t in g 10
1
- 10
2
Pa in t s , con fect ion er y
Mixin g a n d s t ir r in g 10
1
- 10
3
Food pr oces s in g
Pipe flow 10
0
- 10
3
Food pr oces s in g, blood flow
Ru bbin g 10
2
- 10
4
Topica l a pplica t ion of cr ea ms
a n d lot ion s
Br u s h in g 10
3
- 10
4
Br u s h pa in t in g, lips t ick, n a il
polis h
Spr a yin g 10
3
- 10
5
Spr a y dr yin g, s pr a y pa in t in g,
fu el a t omiza t ion
High s peed coa t in g 10
4
- 10
6
Pa per
Lu br ica t ion 10
3
- 10
7
Bea r in gs , ga s olin e en gin es
Simple shear flow is also called viscomet r ic flow. It includes axial
flow in a t ube, r ot at ional flow bet ween concent r ic cylinder s, r ot at ional
flow bet ween a cone and a plat e, and t or sional flow (also r ot at ional)
bet ween par allel plat es. In viscomet r ic flow, t hr ee shear -r at e-dependent
mat er ial funct ions, collect ively called viscomet r ic funct ions, ar e needed
t o complet ely est ablish t he st at e of st r ess in a fluid. These may be
descr ibed as t he viscosit y funct ion, , and t he fir st and second nor mal
st r ess coefficient s, and , defined mat hemat ically as
[1.22]
[1.23]

1
(

)
2
(

)
f (

)

21

1
f (

)

11

22
(

)
2

N
1
(

)
2
16 Chapter 1. Introduction to Rheology
[1.24]
The fir st ( ) and second ( ) nor mal st r ess differ ences ar e
oft en symbolically r epr esent ed as and , r espect ively. is always
posit ive and consider ed t o be appr oximat ely 10 t imes gr eat er t han .
Measur ement of is difficult ; for t unat ely, t he assumpt ion t hat
is usually sat isfact or y. The r at ioof , known as t he r ecover able shear
(or t he r ecover able elast ic st r ain), has pr oven t o be a useful par amet er
in modeling die swell phenomena in polymer s (Tanner , 1988). Some
dat a on t he values of fluid foods have been published (see Appendix
[6.6]).
If a fluid is Newt onian, is a const ant (equal t o t he Newt onian
viscosit y) and t he fir st and second nor mal st r ess differ ences ar e zer o.
As appr oaches zer o, elast ic fluids t end t o display Newt onian behavior .
Viscoelast ic fluids simult aneously exhibit obvious fluid-like (viscous)
and solid-like (elast ic) behavior . Manifest at ions of t his behavior , due
t o a high elast ic component , can be ver y st r ong and cr eat e difficult
pr oblems in pr ocess engineer ing design. These pr oblems ar e par t icu-
lar ly pr evalent in t he plast ic pr ocessing indust r ies but also pr esent in
pr ocessing foods such as dough, par t icular ly t hose cont aining lar ge
quant it ies of wheat pr ot ein.
Fig. 1.10 illust r at es sever al phenomena. Dur ing mixing or agit at ion,
a viscoelast ic fluid may climb t he impeller shaft in a phenomenon known
as t he Weissenber g effect (Fig. 1.10). This can be obser ved in home
mixing of cake or chocolat e br ownie bat t er . When a Newt onian fluid
emer ges fr om a long, r ound t ube int o t he air , t he emer ging jet will
nor mally cont r act ; at low Reynolds number s it may expand t oa diamet er
which is 10 t o 15% lar ger t han t he t ube diamet er . Nor mal st r ess dif-
fer ences pr esent in a viscoelast ic fluid, however , may cause jet expan-
sions (called die swell) which ar e t wo or mor e t imes t he diamet er of t he
t ube (Fig. 1.10). This behavior cont r ibut es t o t he challenge of designing
ext r uder dies t opr oduce t he desir ed shape of many pet , snack, and cer eal
foods. Melt fr act ur e, a flow inst abilit y causing dist or t ed ext r udat es, is
also a pr oblem r elat ed t o fluid viscoelast icit y. In addit ion, highly elast ic
fluids may exhibit a t ubeless siphon effect (Fig. 1.10).

2
f (

)

22

33
(

)
2

N
2
(

)
2

11

22

22

33
N
1
N
2
N
1
N
2
N
2
N
2
0
N
1
/
12
N
1
(

1.5.1 Time-Independent Material Functions 17


Figure 1.10. Weissenberg effect (fluid climbing a rotating rod), tubeless siphon
and jet swell of viscous (Newtonian) and viscoelastic fluids.
The r ecoil phenomena (Fig. 1.11), wher e t ensile for ces in t he fluid
cause par t icles t o move backwar d (snap back) when flow is st opped, may
also be obser ved in viscoelast ic fluids. Ot her impor t ant effect s include
dr ag r educt ion, ext r udat e inst abilit y, and vor t ex inhibit ion. An excel-
lent pict or ial summar y of t he behavior of viscoelast ic polymer solut ions
in var ious flow sit uat ions has been pr epar ed by Boger and Walt er s
(1993).
Nor mal st r ess dat a can be collect ed in st eady shear flow using a
number of differ ent t echniques (Dealy, 1982): exit pr essur e differ ences
in capillar y and slit flow, axial flow in an annulus, wall pr essur e in
concent r ic cylinder flow, axial t hr ust in cone and plat e as well as par allel
plat e flow. In gener al, t hese met hods have been developed for plast ic
melt s (and r elat ed polymer ic mat er ials) wit h t he pr oblems of t he plast ic
indust r ies pr oviding t he main dr iving for ce for change.
VISCOUS FLUID VISCOELASTIC FLUID
WEISSENBERG
EFFECT
TUBELESS
SIPHON
JET SWELL
18 Chapter 1. Introduction to Rheology
Cone and plat e syst ems ar e most commonly used for obt aining
pr imar y nor mal st r ess dat a and a number of commer cial inst r ument s
ar e available t o make t hese measur ement s. Obt aining accur at e dat a
for food mat er ials is complicat ed by var ious fact or s such as t he pr esence
of a yield st r ess, t ime-dependent behavior and chemical r eact ions
occur r ing dur ing pr ocessing (e.g., hydr at ion, pr ot ein denat ur at ion, and
st ar ch gelat inizat ion). Rheogoniomet er is a t er m somet imes used t o
descr ibe an inst r ument capable of measur ing bot h nor mal and shear
st r esses. Det ailed infor mat ion on t est ing viscoelast ic polymer s can be
found in numer ous books: Bir d et al. (1987), Bar nes et al. (1989), Bogue
and Whit e (1970), Dar by (1976), Macosko (1994), and Walt er s (1975).
Figure 1.11. Recoil phenomenon in viscous (Newtonian) and viscoelastic fluids.
Viscomet r ic funct ions have been ver y useful in under st anding t he
behavior of synt het ic polymer solut ions and melt s (polyet hylene, poly-
pr opylene, polyst yr ene, et c.). Fr om an indust r ial st andpoint , t he vis-
cosit y funct ion is most impor t ant in st udying fluid foods and much of
t he cur r ent wor k is applied t o t hat ar ea. To dat e, nor mal st r ess dat a
for foods have not been widely used in food pr ocess engineer ing. This
is par t ly due t ot he fact t hat ot her fact or s oft en complicat e t he evaluat ion
of t he fluid dynamics pr esent in var ious pr oblems. In food ext r usion,
for example, flashing (vapor izat ion) of wat er when t he pr oduct exit s t he
START
STOP
RECOIL
VISCOUS FLUID VISCOELASTIC FLUID
START
STOP
1.5.1 Time-Independent Material Functions 19
die makes it difficult t o pr edict t he influence of nor mal st r ess differ ences
on ext r udat e expansion. Fut ur e r esear ch may cr eat e significant
advances in t he use of nor mal st r ess dat a by t he food indust r y.
Mathemati cal Models for Inelasti c Flui ds. The elast ic behavior of
many fluid foods is small or can be neglect ed (mat er ials such as dough
ar e t he except ion) leaving t he viscosit y funct ion as t he main ar ea of
int er est . This funct ion involves shear st r ess and shear r at e: t he r ela-
t ionship bet ween t he t wo is est ablished fr om exper iment al dat a.
Behavior is visualized as a plot of shear st r ess ver sus shear r at e, and
t he r esult ing cur ve is mat hemat ically modeled using var ious funct ional
r elat ionships. The simplest t ype of subst ance t o consider is t he New-
t onian fluid wher e shear st r ess is dir ect ly pr opor t ional t o shear r at e [for
convenience t he subscr ipt on will be dr opped in t he r emainder of t he
t ext when dealing exclusively wit h one dimensional flow]:
[1.25]
wit h being t he const ant of pr opor t ionalit y appr opr iat e for a Newt onian
fluid. Using unit s of N, m
2
, m, m/s for for ce, ar ea, lengt h and velocit y
gives viscosit y as Pa s which is 1 poiseuille or 1000 cent ipoise (not e: 1
Pa s = 1000 cP = 1000 mPa s; 1 P = 100 cP). Dynamic viscosit y and
coefficient of viscosit y ar e synonyms for t he t er m "viscosit y" in r efer r ing
t o Newt onian fluids. The r ecipr ocal of viscosit y is called fluidit y.
Coefficient of viscosit y and fluidit y ar e infr equent ly used t er ms.
Newt onian fluids may also be descr ibed in t er ms of t heir kinemat ic
viscosit y ( ) which is equal t o t he dynamic viscosit y divided by densit y
( ). This is a common pr act ice for non-food mat er ials, par t icular ly
lubr icat ing oils. Viscosit y conver sion fact or s ar e available in Appendix
[6.1].
Newt onian fluids, by definit ion, have a st r aight line r elat ionship
bet ween t he shear st r ess and t he shear r at e wit h a zer o int er cept . All
fluids which do not exhibit t his behavior may be called non-Newt onian.
Looking at t ypical Newt onian fluids on a r heogr am (Fig. 1.12) r eveals
t hat t he slope of t he line incr eases wit h incr easing viscosit y.
Van Wazer et al. (1963) discussed t he sensit ivit y of t he eye in judging
viscosit y of Newt onian liquids. It is difficult for t he eye t o dist inguish
differ ences in t he r ange of 0.1 t o 10 cP. Small differ ences in viscosit y
ar e clear ly seen fr om appr oximat ely 100 t o 10,000 cP: somet hing at 800
cP may look t wice as t hick as somet hing at 400 cP. Above 100,000 cP
it is difficult t o make visual dist inct ions because t he mat er ials do not

21

/
20 Chapter 1. Introduction to Rheology
Figure 1.12. Rheograms for typical Newtonian fluids.
pour and appear , t o t he casual obser ver , as solids. As point s of r efer ence
t he following r epr esent t ypical Newt onian viscosit ies at r oom t emper -
at ur e: air , 0.01 cP; gasoline (pet r ol), 0.7 cP; wat er , 1 cP; mer cur y, 1.5
cP; coffee cr eam or bicycle oil, 10 cP; veget able oil, 100 cP; glycer ol, 1000
cP; glycer ine, 1500 cP; honey, 10,000 cP; t ar , 1,000,000 cP. Dat a for
many Newt onian fluids at differ ent t emper at ur es ar e pr esent ed in
Appendices [6.8], [6.9], and [6.10].
A gener al r elat ionship t o descr ibe t he behavior of non-Newt onian
fluids is t he Her schel-Bulkley model:
[1.26]
wher e is t he consist ency coefficient , is t he flow behavior index, and
is t he yield st r ess. This model is appr opr iat e for many fluid foods.
Eq. [1.26] is ver y convenient because Newt onian, power law (shear -
t hinning when or shear -t hickening when ) and Bing-
ham plast ic behavior may be consider ed as special cases (Table 1.2, Fig.
1.13). Wit h t he Newt onian and Bingham plast ic models, is commonly
called t he viscosit y ( ) and plast ic viscosit y ( ), r espect ively. Shear -
t hinning and shear -t hickening ar e also r efer r ed t o as pseudoplast ic and
dilat ent behavior , r espect ively; however , shear -t hinning and
0 5 10 15 20
0
0.5
1
1.5
2
2.5
3
Shear Rate, 1/s
40% fat cream, 6.9 cP
olive oil, 36.3 cP
castor oil, 231 cP
0 5 10 15 20
0
0.5
1
1.5
2
2.5
3
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
40% fat cream, 6.9 cP
olive oil, 36.3 cP
castor oil, 231 cP
K(

)
n
+
o
K n

o
0 < n < 1 1 < n <
K

pl
1.5.1 Time-Independent Material Functions 21
shear -t hickening ar e t he pr efer r ed t er ms. A t ypical example of a
shear -t hinning mat er ial is found in t he flow behavior of a 1% aqueous
solut ion of car r ageenan gum as demonst r at ed in Example Pr oblem
1.14.1. Shear -t hickening is consider ed wit h a concent r at ed cor n st ar ch
solut ion in Example Pr oblem 1.14.2.
Table 1.2. Newtonian, Power Law and Bingham Plastic Fluids as Special Cases of
the Herschel-Bulkley Model (Eq. [1.26])
Flu id K n Typica l Exa mples
Her s ch el-Bu lkley > 0 0 < n < > 0 min ced fis h pa s t e,
r a is in pa s t e
Newt on ia n > 0 1 0 wa t er , fr u it ju ice,
milk, h on ey, veget a -
ble oil
s h ea r -t h in n in g > 0 0 < n < 1 0 a pples a u ce, ba n a n a
(ps eu dopla s t ic) pu r ee, or a n ge ju ice
con cen t r a t e
s h ea r -t h icken in g > 0 1 < n < 0 s ome t ypes of
(dila t en t ) h on ey, 40% r a w
cor n s t a r ch s olu t ion
Bin gh a m pla s t ic > 0 1 > 0 t oot h pa s t e, t oma t o
pa s t e
An impor t ant char act er ist ic of t he Her schel-Bulkley and Bingham
plast ic mat er ials is t he pr esence of a yield st r ess ( ) which r epr esent s
a finit e st r ess r equir ed t o achieve flow. Below t he yield st r ess a mat er ial
exhibit s solid like char act er ist ics: It st or es ener gy at small st r ains and
does not level out under t he influence of gr avit y t o for m a flat sur face.
This char act er ist ic is ver y impor t ant in pr ocess design and qualit y
assessment for mat er ials such as but t er , yogur t and cheese spr ead. The
yield st r ess is a pr act ical, but idealized, concept t hat will r eceive addi-
t ional discussion in a lat er sect ion (Sec. 1.6). Typical yield st r ess values
may be found in Appendix [6.7].

o
22 Chapter 1. Introduction to Rheology
Figure 1.13. Curves for typical time-independent fluids.
Figure 1.14. Rheogram of idealized shear-thinning (pseudoplastic) behavior.
Newtonian
Shear-Thickening
Shear-Thinning
Herschel-Bulkley
Bingham
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Shear Rate, 1/s
Lower Region
Upper Region
Middle Region
Slope =
Slope =
1.5.1 Time-Independent Material Functions 23
Shear -t hinning behavior is ver y common in fr uit and veget able
pr oduct s, polymer melt s, as well as cosmet ic and t oilet r y pr oduct s
(Appendices [6.11], [6.12], [6.13]). Dur ing flow, t hese mat er ials may
exhibit t hr ee dist inct r egions (Fig. 1.14): a lower Newt onian r egion
wher e t he appar ent viscosit y ( ), called t he limit ing viscosit y at zer o
shear r at e, is const ant wit h changing shear r at es; a middle r egion wher e
t he appar ent viscosit y ( ) is changing (decr easing for shear -t hinning
fluids) wit h shear r at e and t he power law equat ion is a suit able model
for t he phenomenon; and an upper Newt onian r egion wher e t he slope
of t he cur ve ( ), called t he limit ing viscosit y at infinit e shear r at e, is
const ant wit h changing shear r at es. The middle r egion is most oft en
examined when consider ing t he per for mance of food pr ocessing equip-
ment . The lower Newt onian r egion may be r elevant in pr oblems
involving low shear r at es such as t hose r elat ed t o t he sediment at ion of
fine par t icles in fluids. Values of for some viscoelast ic fluids ar e given
in Table 5.4.
Numer ous fact or s influence t he select ion of t he r heological model
used t o descr ibe flow behavior of a par t icular fluid. Many models, in
addit ion t o t he power law, Bingham plast ic and Her schel-Bulkley
models, have been used t o r epr esent t he flow behavior of non-Newt onian
fluids. Some of t hem ar e summar ized in Table 1.3. The Cr oss,
Reiner -Philippoff, Van Wazer and Powell-Eyr ing models ar e useful in
modeling pseudoplast ic behavior over low, middle and high shear r at e
r anges. Some of t he equat ions, such as t he Modified Casson and t he
Gener alized Her schel-Bulkley, have pr oven useful in developing
mat hemat ical models t o solve food pr ocess engineer ing pr oblems (Ofoli
et al., 1987) involving wide shear r at e r anges. Addit ional r heological
models have been summar ized by Holdswor t h (1993).
The Casson equat ion has been adopt ed by t he Int er nat ional Office
of Cocoa and Chocolat e for int er pr et ing chocolat e flow behavior . The
Casson and Bingham plast ic models ar e similar because t hey bot h have
a yield st r ess. Each, however , will give differ ent values of t he fluid
par amet er s depending on t he dat a r ange used in t he mat hemat ical
analysis. The most r eliable value of a yield st r ess, when det er mined
fr om a mat hemat ical int er cept , is found using dat a t aken at t he lowest
shear r at es. This concept is demonst r at ed in Example Pr oblem 1.14.3
for milk chocolat e.

o
24 Chapter 1. Introduction to Rheology
Apparent Vi scosi ty. Appar ent viscosit y has a pr ecise definit ion. It
is, as not ed in Eq. [1.22], shear st r ess divided by shear r at e:
[1.27]
Wit h Newt onian fluids, t he appar ent viscosit y and t he Newt onian
viscosit y ( ) ar e ident ical but for a power law fluid is
[1.28]
Table 1.3. Rheological Models to Describe the Behavior of Time-
independent Fluids
Model (Sou r ce) Equ a t ion
*
Ca s s on (Ca s s on , 1959)
Modified Ca s s on (Mizr a h i a n d Ber k,
1972)
Ellis (Ellis , 1927)
Gen er a lized Her s ch el-Bu lkley (Ofoli et
a l., 1987)
Voca dlo (Pa r zon ka a n d Voca dlo, 1968)
Power Ser ies (Wh or low, 1992)
Ca r r ea u (Ca r r ea u , 1968)
Cr os s (Cr os s , 1965)
Va n Wa zer (Va n Wa zer , 1963)
Powell-Eyr in g (Powell a n d Eyr in g, 1944)
Rein er -Ph ilippoff (Ph ilippoff, 1935)
*
a n d a r e a r bit r a r y con s t a n t s a n d power in dices , r es pect ively, det er min ed
fr om exper imen t a l da t a .
f (


f (

)
K(

)
n

K(

)
n 1

0.5
(
o
)
0.5
+ K
1
(

)
0.5

0.5
(
o
)
0.5
+ K
1
(

)
n
1

K
1
+ K
2
()
n
1

n
1
(
o
)
n
1
+ K
1
(

)
n
2

(
o
)
1/n
1
+ K
1

_
,
n
1

K
1
+ K
2
()
3
+ K
3
()
5

K
1

+ K
2

()
3
+ K
3

()
5

+ (
o

1 + (K
1

)
21
]
(n 1)/2

+

o

1 + K
1
(

)
n


o

1 + K
1

+ K
2
(

)
n
1
+

K
1

1
K
2
_

,
sinh
1
(K
3

+

o

1 + (()
2
/K
1
)
_

K
1
, K
2
, K
3
n
1
, n
2
1.5.1 Time-Independent Material Functions 25
Appar ent viscosit ies for Bingham plast ic and Her schel-Bulkley fluids
ar e det er mined in a like manner :
[1.29]
[1.30]
decr eases wit h incr easing shear r at e in shear -t hinning and Bingham
plast ic subst ances. In Her schel-Bulkley fluids, appar ent viscosit y will
decr ease wit h higher shear r at es when , but behave in t he
opposit e manner when Appar ent viscosit y is const ant wit h
Newt onian mat er ials and incr eases wit h incr easing shear r at e in
shear -t hickening fluids (Fig. 1.15).
Figure 1.15. Apparent viscosity of time-independent fluids.
A single point appar ent viscosit y value is somet imes used as a
measur e of mout hfeel of fluid foods: The human per cept ion of t hickness
is cor r elat ed t o t he appar ent viscosit y at appr oximat ely 60 s
-1
. Appar ent
viscosit y can also be used t o illust r at e t he axiom t hat t aking single point
t est s for det er mining t he gener al behavior of non-Newt onian mat er ials
may cause ser ious pr oblems. Some qualit y cont r ol inst r ument s designed
for single point t est s may pr oduce confusing r esult s. Consider , for
f (

)
K(

) +
o

K +

o

f (

)
K(

)
n
+
o

K(

)
n 1
+

o

0<n <1.0
n >1.0.
Shear Rate, 1/s
A
p
p
a
r
e
n
t

V
i
s
c
o
s
i
t
y
,

P
a

s
Shear-Thickening
Shear-Thinning
Herschel-Bulkley
Bingham
Newtonian
Time-Independent Fluids
( 0 < n < 1.0 )
26 Chapter 1. Introduction to Rheology
example, t he t wo Bingham plast ic mat er ials shown in Fig. 1.16. The
t wo cur ves int er sect at 19.89 1/s and an inst r ument measur ing t he
appar ent viscosit y at t hat shear r at e, for each fluid, would give ident ical
r esult s: = 1.65 Pa s. However , a simple examinat ion of t he mat er ial
wit h t he hands and eyes would show t hem t o be quit e differ ent because
t he yield st r ess of one mat er ial is mor e t han 4 t imes t hat of t he ot her
mat er ial. Clear ly, numer ous dat a point s (minimum of t wo for t he power
law or Bingham plast ic models) ar e r equir ed t o evaluat e t he flow
behavior of non-Newt onian fluids.
Figure 1.16. Rheograms for two Bingham plastic fluids.
Soluti on Vi scosi ti es. It is somet imes useful t o det er mine t he visco-
sit ies of dilut e synt het ic or biopolymer solut ions. When a polymer is
dissolved in a solvent , t her e is a not iceable incr ease in t he viscosit y of
t he r esult ing solut ion. The viscosit ies of pur e solvent s and solut ions
can be measur ed and var ious values calculat ed fr om t he r esult ing dat a:
[1.31]
[1.32]

0 10 20 30 40 50 60
0
20
40
60
80
100
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Yield Stress = 25.7 Pa
Plastic Viscosity = .36 Pa s
Yield Stress = 6.0 Pa
Plastic Viscosity = 1.35 Pa s
Bingham Plastic Fluids
relative viscosity
rel


solution

solvent
specific viscosity
sp

rel
1
1.5.2 Time-Dependent Material Functions 27
[1.33]
[1.34]
[1.35]
wher e is t he mass concent r at ion of t he solut ion in unit s of g/dl or
g/100ml. Not e t hat unit s of r educed, inher ent , and int r insic viscosit y
ar e r ecipr ocal concent r at ion (usually decilit er s of solut ion per gr ams of
polymer ). The int r insic viscosit y has gr eat pr act ical value in molecular
weight det er minat ions of high polymer s (Sever s, 1962; Mor t on-J ones,
1989; Gr ulke, 1994). This concept is based on t he Mar k-Houwink
r elat ion suggest ing t hat t he int r insic viscosit y of a dilut e polymer
solut ion is pr opor t ional t o t he aver age molecular weight of t he solut e
r aised t o a power in t he r ange of 0.5 t o 0.9. Values of t he pr opor t ionalit y
const ant and t he exponent ar e well known for many polymer -solvent
combinat ions (Pr ogelf and Thr one, 1993; Rodr iquez, 1982). Solut ion
viscosit ies ar e useful in under st anding t he behavior of some biopolymer s
including aqueous solut ions of locust bean gum, guar gum, and car -
boxymet hylcellulose (Rao, 1986). The int r insic viscosit ies of numer ous
pr ot ein solut ions have been summar ized by Rha and Pr adipasena
(1986).
1.5.2. Ti me-Dependent Materi al Functi ons
Ideally, t ime-dependent mat er ials ar e consider ed t obe inelast ic wit h
a viscosit y funct ion which depends on t ime. The r esponse of t he sub-
st ance t o st r ess is inst ant aneous and t he t ime-dependent behavior is
due t o changes in t he st r uct ur e of t he mat er ial it self. In cont r ast , t ime
effect s found in viscoelast ic mat er ials ar ise because t he r esponse of
st r ess t o applied st r ain is not inst ant aneous and not associat ed wit h a
st r uct ur al change in t he mat er ial. Also, t he t ime scale of t hixot r opy may
be quit e differ ent t han t he t ime scale associat ed wit h viscoelast icit y:
The most dr amat ic effect s ar e usually obser ved in sit uat ions involving
shor t pr ocess t imes. Not e t oo, t hat r eal mat er ials may be bot h t ime-
dependent and viscoelast ic!
reduced viscosity
red


sp
C
inherent viscosity
inh

ln
rel
C
intrinsic viscosity
int

sp
C
1
1
]C 0
C
28 Chapter 1. Introduction to Rheology
Figure 1.17. Time-dependent behavior of fluids.
Separ at e t er minology has been developed t o descr ibe fluids wit h
t ime-dependent char act er ist ics. Thixot r opic and r heopect ic mat er ials
exhibit , r espect ively, decr easing and incr easing shear st r ess (and
appar ent viscosit y) over t ime at a fixed r at e of shear (Fig. 1.17). In ot her
wor ds, t hixot r opy is t ime-dependent t hinning and r heopexy is t ime-
dependent t hickening. Bot h phenomena may be ir r ever sible, r ever sible
or par t ially r ever sible. Ther e is gener al agr eement t hat t he t er m
"t hixot r opy" r efer s t o t he t ime-dependent decr ease in viscosit y, due t o
shear ing, and t he subsequent r ecover y of viscosit y when shear ing is
r emoved (Mewis, 1979). Ir r ever sible t hixot r opy, called r heomalaxis or
r heodest r uct ion, is common in food pr oduct s and may be a fact or in
evaluat ing yield st r ess as well as t he gener al flow behavior of a mat er ial.
Ant i-t hixot r opy and negat ive t hixot r opy ar e synonyms for r heopexy.
Thixot r opy in many fluid foods may be descr ibed in t er ms of t he
sol-gel t r ansit ion phenomenon. This t er minology could apply, for
example, t o st ar ch-t hickened baby food or yogur t . Aft er being man-
ufact ur ed, and placed in a cont ainer , t hese foods slowly develop a t hr ee
dimensional net wor k and may be descr ibed as gels. When subject ed t o
shear (by st andar d r heological t est ing or mixing wit h a spoon), st r uct ur e
is br oken down and t he mat er ials r each a minimum t hickness wher e
Thixotropic
Time-Independent
Rheopectic
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Time at Constant Shear Rate, s
Time-Dependent Behavior
1.5.2 Time-Dependent Material Functions 29
Figure 1.18. Thixotropic behavior observed in torque decay curves.
t hey exist in t he sol st at e. In foods t hat show r ever sibilit y, t he net wor k
is r ebuilt and t he gel st at e r eobt ained. Ir r ever sible mat er ials r emain
in t he sol st at e.
The r ange of t hixot r opic behavior is illust r at ed in Fig. 1.18. Sub-
ject ed t o a const ant shear r at e, t he shear st r ess will decay over t ime.
Dur ing t he r est per iod t he mat er ial may complet ely r ecover , par t ially
r ecover or not r ecover any of it s or iginal st r uct ur e leading t o a high,
medium, or low t or que r esponse in t he sample. Rot at ional viscomet er s
have pr oven t o be ver y useful in evaluat ing t ime-dependent fluid
behavior because (unlike t ube viscomet er s) t hey easily allow mat er ials
t o be subject ed t o alt er nat e per iods of shear and r est .
St ep (or linear ) changes in shear r at e may also be car r ied out
sequent ially wit h t he r esult ing shear st r ess obser ved bet ween st eps.
Typical r esult s ar e depict ed in Fig. 1.19. Act ual cur ve segment s (such
as 1-2 and 3-4) depend on t he r elat ive cont r ibut ion of st r uct ur al
br eakdown and buildup in t he subst ance. Plot t ing shear st r ess ver sus
shear r at e for t he incr easing and decr easing shear r at e values can be
used t ogener at e hyst er esis loops (a differ ence in t he up and down cur ves)
for t he mat er ial. The ar ea bet ween t he cur ves depends on t he t ime-
dependent nat ur e of t he subst ance: it is zer o for a t ime-independent
0
S
t
r
e
s
s
0
time
Rest
Period
S
h
e
a
r

R
a
t
e
Complete Recovery
Partial Recovery
No Recovery
Evidence of Thixotropy in Torque Decay Curves
30 Chapter 1. Introduction to Rheology
fluid. This infor mat ion may be valuable in compar ing differ ent
mat er ials, but it is somewhat subject ive because differ ent st ep change
per iods may lead t o differ ent hyst er esis loops. Similar infor mat ion can
be gener at ed by subject ing mat er ials t o st ep (or linear ) changes in shear
st r ess and obser ving t he r esult ing changes in shear r at es.
Figure 1.19. Thixotropic behavior observed from step changes in shear rate.
Tor que decay dat a (like t hat given for a pr oblem in mixer viscomet r y
descr ibed in Example Pr oblem 3.8.22) may be used t o model ir r ever sible
t hixot r opy by adding a st r uct ur al decay par amet er t o t he Her schel-
Bulkley model t o account for br eakdown (Tiu and Boger , 1974):
[1.36]
wher e , t he st r uct ur al par amet er , is a funct ion of t ime. befor e t he
onset of shear ing and equals an equilibr ium value ( ) obt ained aft er
complet e br eakdown fr om shear ing. The decay of t he st r uct ur al
par amet er wit h t ime may be assumed t o obey a second or der equat ion:
[1.37]
time
S
h
e
a
r

S
t
r
e
s
s
S
h
e
a
r

R
a
t
e
1
2
3
4
Step Changes in Shear Rate
f (,

) (
o
+ K(

)
n
)
1

e
d
dt
k
1
(
e
)
2
for >
e
1.5.2 Time-Dependent Material Functions 31
wher e is a r at e const ant t hat is a funct ion of shear r at e. Then, t he
ent ir e model is complet ely det er mined by five par amet er s: ,
and . and ar e det er mined under init ial shear ing condit ions
when and . In ot her wor ds, t hey ar e det er mined fr om t he init ial
shear st r ess in t he mat er ial, obser ved at t he beginning of a t est , for each
shear r at e consider ed.
and ar e expr essed in t er ms of t he appar ent viscosit y ( ) t o
find . Equat ing t he r heological model (Eq. [1.36]) t o t he definit ion of
appar ent viscosit y (which in t his case is a funct ion of bot h shear r at e
and t he t ime-dependent appar ent viscosit y) yields an expr ession for :
[1.38]
Eq. [1.38] is valid for all values of including at , t he equilibr ium
value of t he appar ent viscosit y. Differ ent iat ing wit h r espect t o t ime,
at a const ant shear r at e, gives
[1.39]
Using t he definit ion of , Eq. [1.37] and [1.39] may be combined
yielding
[1.40]
Consider ing t he definit ion of given by Eq. [1.38], t his may be r ewr it t en
as
[1.41]
Simplificat ion yields
[1.42]
or
k
1

o
, K, n, k
1
(

e
K, n
o
1 t 0

e
/

k
1

o
+ K(

)
n

e

e

d
dt

d
dt

o
+ K(

)
n
_

,
d/dt
k
1
(
e
)
2

d
dt

o
+ K(

)
n
_

k
1

o
+ K(

)
n
_

o
+ K(

)
n
_

,
1
1
]
2

d
dt

o
+ K(

)
n
_

,
d
dt
k
1

o
+ K(

)
n
_

,
(
e
)
2
32 Chapter 1. Introduction to Rheology
[1.43]
wher e
[1.44]
Int egr at ing Eq. [1.43] gives
[1.45]
so
[1.46]
wher e is t he init ial value of t he appar ent viscosit y calculat ed fr om
t he init ial ( and ) shear st r ess and shear r at e.
Using Eq. [1.46], a plot ver sus , at a par t icular shear r at e,
is made t o obt ain . This is done at numer ous shear r at es and t he
r esult ing infor mat ion is used t o det er mine t he r elat ion bet ween and
and, fr om t hat , t he r elat ion bet ween and . This is t he final infor -
mat ion r equir ed t o complet ely specify t he mat hemat ical model given by
Eq. [1.36] and [1.37].
The above appr oach has been used t o descr ibe t he behavior of
mayonnaise (Tiu and Boger , 1974), baby food (For d and St effe, 1986),
and but t er milk (But ler and McNult y, 1995). Mor e complex models
which include t er ms for t he r ecover y of st r uct ur e ar e also available
(Cheng, 1973; Fer guson and Kemblowski, 1991). Numer ous r heological
models t o descr ibe t ime-dependent behavior have been summar ized by
Holdswor t h (1993).
1.5.3. Modeli ng Rheologi cal Behavi or of Flui ds
Modeling pr ovides a means of r epr esent ing a lar ge quant it y of
r heological dat a in t er ms of a simple mat hemat ical expr ession. Rheo-
gr ams, summar ized in t er ms of t he Her schel-Bulkley equat ion (Eq.
[1.26]), r epr esent one example of modeling. In t his sect ion we will
expand t he idea t o include t emper at ur e and concent r at ion (or moist ur e
cont ent ) effect s int o single empir ical expr essions. Many for ms of t he
equat ions ar e possible and one mast er model, suit able for all sit uat ions,
d
dt
a
1
(
e
)
2
a
1

k
1

o
+ K(

)
n

(
e
)
2
d

0
t
a
1
dt
1

e

o

e
+ a
1
t

o
t 0 1
1/(
e
) t
a
1
a
1

k
1

1.5.3 Modeling Rheological Behavior of Fluids 33


does not exist . The equat ions cover ed her e ar e accept able for a lar ge
number of pr act ical pr oblems involving homogeneous mat er ials which
do not exper ience a phase change over t he r ange of condit ions under
consider at ion.
The influence of t emper at ur e on t he viscosit y for Newt onian fluids
can be expr essed in t er ms of an Ar r henius t ype equat ion involving t he
absolut e t emper at ur e ( ), t he univer sal gas const ant ( ), and t he ener gy
of act ivat ion for viscosit y ( ):
[1.47]
and ar e det er mined fr om exper iment al dat a. Higher values
indicat e a mor e r apid change in viscosit y wit h t emper at ur e. The ener gy
of act ivat ion for honey is evaluat ed in Example Pr oblem 1.14.4.
Consider ing an unknown viscosit y ( ) at any t emper at ur e ( ) and a
r efer ence viscosit y ( ) at a r efer ence t emper at ur e ( ), t he const ant ( )
may be eliminat ed fr om Eq. [1.47] and t he r esult ing equat ion wr it t en
in logar it hmic for m:
[1.48]
In addit ion t o modeling t he viscosit y of Newt onian fluids, an Ar r henius
r elat ionship can be used t o model t he influence of t emper at ur e on
appar ent viscosit y in power law fluids. Consider ing a const ant shear
r at e, wit h t he assumpt ion t hat t emper at ur e has a negligible influence
on t he flow behavior index, yields
[1.49]
or
[1.50]
Eq. [1.50] can be used t o find at any t emper at ur e ( ) fr om appr opr iat e
r efer ence values ( ). Act ivat ion ener gies and r efer ence viscosit ies
for a number of fluid foods ar e summar ized in Appendix [6.14].
T R
E
a
f (T) A exp

E
a
RT
_

,
E
a
A E
a
T

r
T
r
A
ln

r
_

E
a
R
_

1
T

1
T
r
_

,
ln

r
_

,

E
a
R

1
T

1
T
r
_

r
exp

E
a
R

1
T

1
T
r
_

,
_

,
T

r
, T
r
34 Chapter 1. Introduction to Rheology
The effect of t emper at ur e on viscosit y can also be modeled using a
r elat ionship known as t he Williams-Landel-Fer r y (WLF) equat ion
pr oposed by Williams et al. (1955). A r efer ence viscosit y at a r efer ence
t emper at ur e, and t he numer ical value of t wo const ant s ar e needed t o
specify t he r elat ionship. The WLF equat ion is ver y useful in modeling
t he viscosit y of amor phous foods above t he glass t r ansit ion t emper at ur e
(Roos, 1992; Roos, 1995).
The effect of shear r at e and t emper at ur e can be combined int o a
single expr ession (Har per and El Sahr igi, 1965):
[1.51]
wher e is an aver age value of t he flow behavior index based on all
t emper at ur es. Eq. [1.51] can also be expr essed in t er ms of shear st r ess:
[1.52]
The pr act ical value of Eq. [1.52] is demonst r at ed for concent r at ed
or ange juice in Example Pr oblem 1.14.5. Par amet er s in t he model
( ) wer e det er mined using a limit ed number of dat a set s t aken at
a few specific t emper at ur es. The final model (Eq. [1.52]), however , can
gener at e a r heogr am at any t emper at ur e in t he r ange. This is useful
in solving many food engineer ing pr oblems such as t hose r equir ing a
pr edict ion of t he fluid velocit y pr ofile or pr essur e dr op dur ing t ube flow.
Effect s of t emper at ur e and concent r at ion ( ) on appar ent viscosit y,
at a const ant shear r at e, can also be combined int o a single r elat ionship
(Vit ali and Rao, 1984; Cast aldo et al. 1990):
[1.53]
The t hr ee const ant s ( ) must be det er mined fr om exper iment al
dat a. Shear r at e, t emper at ur e, and concent r at ion (or moist ur e cont ent )
can also be combined int o a single expr ession (Mackey et al., 1989):
[1.54]
wher e t he influence of shear r at e is given in t er ms of a power law
funct ion. The par amet er s ( , , and ) cannot be given an exact
physical int er pr et at ion because t he sequence of st eps used in det er -
mining t hem influences t he magnit ude of t he const ant s. Equat ion
f (T,

) K
T
exp

E
a
RT
_

,
(

)
n 1
n
f (T,

) K
T
exp

E
a
RT
_

,
(

)
n
K
T
, n, E
a
C
f (T, C) K
T, C
exp

E
a
RT
_

,
C
B
K
T, C
, E
a
, B
f (T,

, C) K

, T, C
(

)
n 1
exp

E
a
RT
_

,
+ B(C)
1
1
]
K
, T, C
n E
a
B
1.6 Yield Stress Phenomena 35
par amet er s, for example, may be det er mined using st epwise r egr ession
analysis wit h t he assumpt ion t hat int er act ion effect s (such as t emper -
at ur e dependence of and ) can be neglect ed. is a const ant which
combines t he effect s of shear r at e, t emper at ur e and concent r at ion.
Rheological behavior of fluid foods is complex and influenced by
numer ous fact or s. Eq. [1.54] allows pr edict ion of appar ent viscosit y on
t he basis of shear r at e, t emper at ur e, and moist ur e cont ent . Time-
t emper at ur e hist or y and st r ain hist or y may be added t o for m a mor e
compr ehensive equat ion (Dolan et al., 1989; Dolan and St effe, 1990;
Mackey et al., 1989; Mor gan et al., 1989) applicable t o pr ot ein and st ar ch
based dough or slur r y syst ems.
The influence of t emper at ur e on t he behavior of polymer ic mat er ials
may be modeled by det er mining a shift fact or using t he pr inciple of
t ime-t emper at ur e super posit ion (Bir d et al., 1987). This t echnique is
one example of t he met hod of r educed var iables (out lined in det ail by
Fer r y, 1980) which can be expanded t o include t he effect of concent r at ion
and pr essur e on r heological behavior . Time-t emper at ur e super posit ion
equat es t he effect of t ime and t emper at ur e on r heological pr oper t ies.
The useful consequence of t he met hod is t hat mat er ial behavior can be
invest igat ed in t ime domains (usually ver y long or ver y shor t ) t hat ar e
ot her wise unavailable due t o t he exper iment al limit at ions. Time-
t emper at ur e super posit ion has pr oven t o be a valuable met hod in st u-
dying cr eep and st r ess r elaxat ion of synt het ic polymer s (Neilsen and
Landel, 1994); but t he t echnique, which is somet imes applicable t o
biological mat er ials, has not been widely applied t o foods. Da Silva et
al. (1994) found t he t ime-t emper at ur e super posit ion pr inciple was
inappr opr iat e for modeling t he t emper at ur e dependence of st or age and
loss moduli of pect in disper sions. The idea, however , can pr ovide a
useful empir ical met hod for developing mast er -cur ves of r heological
dat a for many fluid foods. The t echnique is illust r at ed in Example
Pr oblem 1.14.5 wher e a shift fact or is calculat ed for concent r at ed or ange
juice.
1.6. Yi eld Stress Phenomena
A yield st r ess ( ) may be defined as t he minimum shear st r ess
r equir ed t o init iat e flow. The exist ence of a yield st r ess has been chal-
lenged (Bar nes and Walt er s, 1985) using t he ar gument t hat ever yt hing
flows given sufficient t ime or ver y sensit ive measur ing equipment . This
concept is explor ed in mor e det ail in Sec. 5.6. Fr om a pr act ical st and-
n b K
, T, C

o
36 Chapter 1. Introduction to Rheology
point , t her e is lit t le doubt t hat is an engineer ing r ealit y (Har t net t
and Hu, 1989) which may st r ongly influence pr ocess engineer ing
calculat ions.
Table 1.4. Methods of Determining Yield Stress
Met h od Des cr ipt ion or Pa r a met er Refer en ce
Mea s u r ed
Ext r a pola t ion Sh ea r s t r es s ver s u s s h ea r r a t e Ofoli et a l. (1987)
cu r ve ext r a pola t ed t o zer o Keen t ok (1982)
s h ea r r a t e. Yos h imu r a et a l. (1987)
Ext r a pola t ion Appa r en t vis cos it y ver s u s Va n Wa zer et a l. (1963)
s h ea r s t r es s cu r ve ext r a pola t ed Ka let u n c-Gen cer a n d Peleg
t o in fin it e a ppa r en t vis cos it y. (1984)
St r es s Deca y Res idu a l s t r es s on a bob. La n g a n d Rh a (1981)
St r es s Deca y Res idu a l s t r es s in a ba ck St effe a n d Os or io (1987)
ext r u der (a n n u la r pu mp).
St r es s t o In it ia t e Flow Con t r olled s t r es s r h eomet er s J a mes et a l. (1987)
mea s u r e t h e min imu m s t r es s
r equ ir ed for flow in t r a dit ion a l
geomet r ies : con e a n d pla t e, et c.
St r es s t o In it ia t e Flow St r es s on a s moot h , r ou gh or La n g a n d Rh a (1981);
gr ooved bob. Voca dlo a n d Ch a r les (1971)
St r es s t o In it ia t e Flow St r es s t o move a n immer s ed DeKee et a l. (1980)
ver t ica l pla t e.
St r es s t o In it ia t e Flow St r es s t o ca u s e mot ion in a t u be Ch en g (1986)
vis comet er .
St r es s t o In it ia t e Flow St r es s t o cr ea t e va n e mot ion . Ngu yen a n d Boger (1985)
Qiu a n d Ra o (1988)
Yoo et a l. (1995)
St r es s t o In it ia t e Flow Size of t h e plu g flow r a diu s in La n g a n d Rh a (1981)
a n a n n u lu s .
St r es s t o In it ia t e Flow For ce t o move ma t er ia l t h r ou gh Goodr ich et a l. (1989)
a fin n ed, cylin dr ica l cell.
Dyn a mic Tes t in g Fla t r es pon s e of a n os cilla t or y Ch en g (1986)
in pu t .
Ver t ica l Pla t e Coa t in g Amou n t of flu id r ema in in g on a La n g a n d Rh a (1981)
pla t e a ft er wit h dr a wa l fr om Ch a r m (1962)
s a mple.
Squ eezin g Flow Defor ma t ion bet ween pa r a llel Ca mpa n ella a n d Peleg
cir cu la r dis ks . (1987a )
Con e Pen et r omet er Dept h of pen et r a t ion . Ta n a ka et a l. (1971)

o
1.6 Yield Stress Phenomena 37
Ther e ar e many ways t o evaluat e t he yield st r ess for fluid like
subst ances (Table 1.4) and no single, "best " t echnique can be ident ified.
Differ ent applicat ions r equir e differ ent met hods. One common met hod
of obt aining a yield st r ess value is t o ext r apolat e t he shear st r ess ver sus
shear r at e cur ve back t o t he shear st r ess int er cept at zer o shear r at e.
Values obt ained using t his met hod will be st r ongly influenced by t he
r heological model (Bingham, Her schel-Bulkley, et c.) and shear r at e
r ange select ed t o r epr esent t he dat a (Ofoli et al., 1987). This difficult y
is demonst r at ed for milk chocolat e in Example Pr oblem 1.14.3. An
alt er nat ive numer ical pr ocedur e is t o plot appar ent viscosit y ver sus
shear st r ess and det er mine fr om t he point (r elat ed t o zer o shear r at e)
wher e becomes infinit e.
Repor t ed yield st r ess values ar e act ually defined by t he r heological
t echniques and assumpt ions used in t he measur ement . An absolut e
yield st r ess is an elusive pr oper t y: It is not unusual for a yield st r ess
obt ained by one t echnique t o be ver y differ ent fr om one found using a
differ ent met hod. Cheng (1986) has wr it t en an excellent r eview of t he
yield st r ess pr oblem and shown t hat t he magnit udes of measur ed values
ar e closely associat ed wit h cr eep, st r ess gr owt h, t hixot r opy, and t he
char act er ist ic t imes of t hese t r ansient r esponses. He descr ibed a concept
of st at ic and dynamic yield st r esses t hat has gr eat pr act ical value in
r heological t est ing of fluid foods.
Many foods, such as st ar ch-t hickened baby food (St effe and For d,
1985), t hicken dur ing st or age and exhibit ir r ever sible t hixot r opic
behavior when st ir r ed befor e consumpt ion. Chemical changes (e.g.,
st ar ch r et r ogr adat ion) cause a weak gel st r uct ur e t o for m in t he mat er ial
dur ing st or age. This st r uct ur e is sensit ive and easily disr upt ed by fluid
movement . The yield st r ess, measur ed in an undist ur bed sample, is t he
st at ic yield st r ess. The yield st r ess of a complet ely br oken down sample,
oft en det er mined fr om ext r apolat ion of t he equilibr ium flow cur ve, is
t he dynamic yield st r ess (Fig. 1.20). A st at ic yield st r ess may be sig-
nificant ly higher t han t he dynamic yield st r ess. If t he mat er ial r ecover s
it s st r uct ur e dur ing a shor t per iod of t ime (uncommon in food pr oduct s),
t hen a r at e par amet er may be ut ilized t o fully descr ibe r heological
behavior .
The idea of a st at ic and a dynamic yield st r ess can be explained by
assuming t her e ar e t wo t ypes of st r uct ur e in a t hixot r opic fluid (Cheng,
1986). One st r uct ur e is insensit ive t o shear r at e and ser ves t o define
t he dynamic yield st r ess associat ed wit h t he equilibr ium flow cur ve. A

38 Chapter 1. Introduction to Rheology


Figure 1.20. Static and dynamic yield stresses.
second st r uct ur e, t he weak st r uct ur e, for ms over a cer t ain per iod of t ime
when t he sample is at r est . Combined, t he t wo st r uct ur es cause a
r esist ance t o flow which det er mines t he st at ic yield st r ess. Behavior
discussed above has been obser ved, t hough not ment ioned in t er ms of
t he st at ic and dynamic yield st r ess, for many food pr oduct s: apple sauce,
banana and peach baby food, must ar d, t omat o ket chup (Bar bosa
Canovas and Peleg, 1983); and meat and yeast ext r act s (Halomos and
Tiu, 1981). Also, t he same t er minology (and essent ially t he same
meaning) was used by Pokor ny et al. (1985) t o int er pr et r heological dat a
for mar gar ine. Yoo et al. (1995) defined a new dimensionless number ,
t he yield number defined as t he st at icyield st r ess divided by t he dynamic
yield st r ess, t o differ ent iat e yield st r esses.
An impor t ant issue in yield st r ess measur ement , par t icular ly fr om
a qualit y cont r ol st andpoint , is r epr oducibilit y of t he exper iment al dat a.
This is cr it ical when compar ing t he over all char act er ist ics of pr oduct s
made on differ ent pr oduct ion lines or in differ ent plant s. In t his sit u-
at ion t he measur ement is closely t ied t o t he applicat ion and t he absolut e
value of t he yield st r ess may be unimpor t ant . A "t r ue value (most likely
t he dynamic value)," of t he yield st r ess may be essent ial t o pr oper ly
design food pr ocessing syst ems like t hose r equir ed in t ubular t her mal
pr ocessing equipment wher e fluid velocit y pr ofiles ar e cr it ical. Typical
Equilibrium
Flow Curve
Dynamic Yield Stress
Static Yield Stress
S
h
e
a
r

S
t
r
e
s
s
Shear Rate
1.7 Extensional Flow 39
yield st r esses, including dat a showing differ ences bet ween t he st at ic
and dynamic yield values, ar e summar ized in Appendix [6.7]. The vane
met hod, a simple and pr act ical means of measur ing t he yield st r ess, is
pr esent ed in det ail in Sec. 3.7.3 and Example Pr oblems 3.8.19, 3.8.20,
and 3.8.21. Evaluat ing yield st r esses using cont r olled st r ess r heomet er s
is discussed in Sec. 3.7.3 and t he r ole of t he yield st r ess in det er mining
t he t hickness of a food coat ing is examined in Example Pr oblem 1.14.6.
1.7. Extensi onal Flow
Viscomet r ic flow may be defined as t hat which is indist inguishable
fr om st eady simple shear flow. Addit ional infor mat ion may be obt ained
fr om a differ ent t ype of flow: ext ensional flow t hat yields an ext ensional
viscosit y. Pur e ext ensional flow does not involve shear ing and is
somet imes r efer r ed t o as "shear fr ee" flow. In published lit er at ur e,
elongat ional viscosit y and Tr out on viscosit y ar e fr equent ly used syn-
onyms for ext ensional viscosit y. Similar ly, elongat ional flow is a syn-
onym for ext ensional flow.
Many food pr ocessing oper at ions involve ext ensional defor mat ion
and t he molecular or ient at ion caused by ext ension, ver sus shear , can
pr oduce unique food pr oduct s and behavior . The r eason shear and
ext ensional flow have a differ ent influence on mat er ial behavior may
be explained by t he way in which flow fields or ient long molecules of
high molecular weight . In shear flow, t he pr efer r ed or ient at ion cor r e-
sponds t o t he dir ect ion of flow; however , t he pr esence of a differ ent ial
velocit y acr oss t he flow field encour ages molecules t o r ot at e t her eby
r educing t he degr ee of st r et ching induced in molecular chains. The
t endency of molecules t o r ot at e, ver sus elongat e, depends on t he mag-
nit ude of t he shear field: Ther e is r elat ively mor e elongat ion, less
r ot at ion, at high shear r at es. In ext ensional flow, t he sit uat ion is ver y
differ ent . The pr efer r ed molecular or ient at ion is in t he dir ect ion of t he
flowfield because t her e ar e nocompet ing for ces t ocause r ot at ion. Hence,
ext ensional flow will induce t he maximum st r et ching of t he molecules
pr oducing a chain t ension t hat may r esult in a lar ge (compar ed t o shear
flow) r esist ance t o defor mat ion.
The nat ur e of t he molecule, br anched ver sus linear , may signifi-
cant ly influence flowbehavior in ext ension. In compar able fluid syst ems
(i.e., high-densit y polyet hylene, a linear molecule, ver sus low-densit y
polyet hylene, a br anched molecule) br anched molecules will cause a
fluid t o be less t ension-t hinning t hen linear molecules. A similar
40 Chapter 1. Introduction to Rheology
ar gument can be made in compar ing t he r elat ive st iffness of biopolymer
molecules: St iffer molecules ar e mor e quickly or ient ed in an ext ensional
flow field. This phenomenon may be a fact or in t he choice of a t hickening
agent for pancake syr up: St r inginess can be r educed, while maint aining
t hickness, when st iffer molecules ar e select ed as addit ives. Reduced
st r inginess leads t o what can be called a clean "cut -off" aft er pour ing
syr up fr om a bot t le. An example of a st iff molecule would be t he r od-like
biopolymer xant han compar ed t o sodium alginat e or car boxymet hyl-
cellulose which exhibit a r andon-coil-t ype confor mat ion in solut ion
(Padmanabhan, 1995).
Ext ensional flow is an impor t ant aspect of food pr ocess engineer ing
and pr evalent in many oper at ions such as dough pr ocessing. Sheet
st r et ching, as well as ext r udat e dr awing, pr ovides a good example of
ext ensional flow (Fig. 1.21). Conver ging flow int o dies, such as t hose
found in single and t win scr ew ext r uder s, involves a combinat ion of
shear and ext ensional flow; t he ext ensional component of defor mat ion
is illust r at ed in Fig. 1.21. The analysis of flow in a conver ging die (see
Sec. 4.4) allows one t o separ at e t he pr essur e dr op over t he die int o t he
shear and ext ensional component s. Conver ging flow may also be
obser ved when fluid is sucked int o a pipe or a st r aw, or when applying
a food spr ead wit h a knife.
One of t he most common examples of ext ensional flow is seen when
st r et ching war m mozzar ella cheese while pulling a slice of pizza away
fr om t he ser ving pan. Somet imes t his behavior is subject ively r efer r ed
t o as st r inginess. Asimilar obser vat ion can be made when pulling apar t
a car amel filled candy bar or a past r y wit h fr uit filling. Ext ensional
defor mat ion is also pr esent in calender ing (Fig. 1.22), a st andar d
oper at ion found in dough sheet ing. Gr avit y induced sagging (Fig. 1.22)
also embodies ext ensional defor mat ion. This may be obser ved in a
cut -off appar at us associat ed wit h fr uit filling syst ems for past r y pr od-
uct s. Ext ensional flow in t his sit uat ion is undesir able because it may
cont r ibut e t o inconsist ent levels of fill or an unsight ly pr oduct
appear ance due t o smear ed filling. Bubble gr owt h fr om t he pr oduct ion
of car bon dioxide gas occur r ing dur ing dough fer ment at ion, ext r udat e
expansion fr om t he vapor izat ion of wat er , and squeezing t o achieve
pr oduct spr eading involve ext ensional defor mat ion (Fig. 1.23). Ext en-
sional flow is also a fact or in die swell and mixing, par t icular ly dough
mixing wit h r ibbon blender s.
1.7 Extensional Flow 41
Figure 1.21. Extensional flow found in sheet stretching (or extrudate drawing)
and convergence into an extruder die.
Figure 1.22. Extensional flow in calendering and gravity induced sagging.
Sheet Stretching Extruder Die
Calendering Sagging
42 Chapter 1. Introduction to Rheology
Figure 1.23. Extensional flow found in bubble growth and squeezing flow
between lubricated plates.
Alt hough ext ensional viscosit y is clear ly a fact or in food pr ocessing,
our use of t his r heological pr oper t y in engineer ing design of pr ocesses
and equipment is st ill at an ear ly st age of development . Ext ensional
flow is also an impor t ant fact or in t he human per cept ion of t ext ur e wit h
r egar d t o t he mout hfeel and swallowing of fluid foods and fluid dr ugs.
In addit ion, many plast ic manufact ur ing oper at ions involve ext ensional
flow: compr ession moulding, t her mofor ming, blow moulding, fiber
spinning, film blowing, inject ion moulding, and ext r usion.
Ext ensional viscosit y has been measur ed for var ious food pr oduct s.
Leight on et al. (1934) used t he sagging beam met hod developed by
Tr out on (1906) t o measur e t he ext ensional viscosit y of ice cr eam.
Result s wer e pr esent ed in t er ms of appar ent viscosit y by using t he well
known Tr out on r at io showing t hat ext ensional viscosit y is equal t o t hr ee
t imes t he shear viscosit y (see Eq. [1.78]). This appear s t o be t he fir st
r epor t ed measur ement of t he ext ensional flow of a food pr oduct . Moz-
zar ella cheese has been t est ed in uniaxial t ension by Ak and Gunase-
kar an (1995). Biaxial ext ensional flow, pr oduced by squeezing mat er ial
bet ween par allel plat es, has been used in evaluat ing cheese
(Campanella et al., 1987; Casir aghi et al., 1985), wheat flour doughs
(Huang and Kokini, 1993; Wikst r m et al., 1994), gels (Bagley et al.
1985; Chr ist ianson et al. 1985), and but t er (Rohn, 1993; Shuka et al.,
1995). Dat a fr om t he Chopin Alveogr aph, a common dough t est ing
device wher e a spher ical bubble of mat er ial is for med by inflat ing a
sheet , can be int er pr et ed in t er ms of biaxial ext ensional viscosit y (Far idi
Bubble Growth Squeezing
1.7 Extensional Flow 43
and Rasper , 1987; Launay and Bur , 1977). This t echnique r equir es an
accur at e det er minat ion of t he sample geomet r y befor e and dur ing
inflat ion. Doughs have also been evaluat ed by subject ing t hem t o
uniaxial ext ension (de Br uijne et al., 1990).
The spinning t est (also called ext r udat e dr awing) was applied t o
measur e t he st r et chabilit y of melt ed Mozzar ella cheese (Cavella et al.,
1992). Ent r ance pr essur e dr op fr om conver ging flow int o a die has been
used t o evaluat e an ext ensional viscosit y for cor n meal dough (Bhat -
t achar ya et al., 1994; Padmanabhan and Bhat t achar ya, 1993; See-
t hamr aju and Bhat t achar ya, 1994) and br ead dough (Bhat t achar ya,
1993). Addit ional met hods have been pr oposed for evaluat ing t he
ext ensional behavior of polymer ic mat er ials (Fer guson and Kemb-
lowski, 1991; J ames and Walt er s, 1993; J ones et al., 1987; Macosko,
1994; Pet r ie, 1979; Tir t aamadja and Sr idhar , 1993; Walt er s, 1975):
bubble collapse, st agnat ion flow in lubr icat ed and unlubr icat ed dies,
open siphon (Fanoflow), filament st r et ching, spinning dr op t ensiomet er ,
and conver ging jet s. Ext ensional viscosit ies for some Newt onian and
non-Newt onian fluids ar e pr esent ed in Appendices [6.15] and [6.16],
r espect ively. Measur ement met hods, and example pr oblems, ar e dis-
cussed in Chapt er 4.
Types of Extensi onal Flow. Ther e ar e t hr ee basic t ypes of ext ensional
flow (Fig. 1.24): uniaxial, planar , and biaxial. Dur ing uniaxial ext ension
mat er ial is st r et ched in one dir ect ion wit h a cor r esponding size r educt ion
in t he ot her t wo dir ect ions. In planar ext ension, a flat sheet of mat er ial
is st r et ched in t he dir ect ion wit h a cor r esponding decr ease in t hickness
( decr eases) while t he widt h ( dir ect ion) r emains unchanged. Biaxial
ext ension looks like uniaxial compr ession, but it is usually t hought of
as flow which pr oduces a r adial t ensile st r ess.
Uni axi al Extensi on. Wit h a const ant densit y mat er ial in uniaxial
ext ension (Fig. 1.24), t he velocit y dist r ibut ion in Car t esian coor dinat es,
descr ibed wit h t he Hencky st r ain r at e, is
x
1
x
2
x
3
44 Chapter 1. Introduction to Rheology
Figure 1.24. Uniaxial, planar, and biaxial extension.
[1.55]
[1.56]
[1.57]
wher e . Since t his flow is axisymmet r ic, it may also be descr ibed
in cylindr ical coor dinat es (it may be helpful t o visualize t his sit uat ion
wit h t he posit ive axis aligned wit h t he axis, Fig. 1.24):
[1.58]
[1.59]
[1.60]
Pur e ext ensional flow does not involve shear defor mat ion; t her efor e, all
t he shear st r ess t er ms ar e zer o:
[1.61]
AFTER DEFORMATION
BEFORE DEFORMATION
uniaxial extension
planar extension
biaxial extension
z
r
1
2
3
x
x
x
u
1

h
x
1
u
2


h
x
2
2
u
3


h
x
3
2

h
> 0
z x
1
u
z

h
z
u
r


h
r
2
u

12

13

23

r

rz

z
0
1.7 Extensional Flow 45
St r ess is also axisymmet r ic:
[1.62]
r esult ing in one nor mal st r ess differ ence t hat can be used t o define t he
t ensile ext ensional viscosit y:
[1.63]
Mat er ials ar e consider ed t ension-t hinning (or ext ensional-t hinning) if
decr eases wit h incr easing values of . They ar e t ension-t hickening
(ext ensional-t hickening) if incr eases wit h incr easing values of .
These t er ms ar e analogous t o shear -t hinning and shear -t hickening used
pr eviously (Sec. 1.5.1) t o descr ibe changes in appar ent viscosit y wit h
shear r at e.
Bi axi al Extensi on. The velocit y dist r ibut ion pr oduced by uniaxial
compr ession causing a biaxial ext ensional flow (Fig. 1.24) can be
expr essed in Car t esian coor dinat es as
[1.64]
[1.65]
[1.66]
wher e . Since , biaxial ext ension can act ually be viewed as
a for m of t ensile defor mat ion. Uniaxial compr ession, however , should
not be viewed as being simply t he opposit e of uniaxial t ension because
t he t endency of molecules t o or ient t hemselves is st r onger in t ension
t han compr ession. Axial symmet r y allows t he above equat ions t o be
r ewr it t en in cylindr ical coor dinat es (Fig. 1.24) as
[1.67]
[1.68]
[1.69]
Biaxial ext ensional viscosit y is defined in t er ms of t he nor mal st r ess
differ ence and t he st r ain r at e:
[1.70]
Planar Extensi on. In planar ext ension (Fig. 1.24), t he velocit y
dist r ibut ion is

rr

22

33

E
f (

h
)

11

22


11

33


zz

rr

h
u
1

B
x
1
u
2
2

B
x
2
u
3

B
x
3

B
> 0

h
2

B
u
z
2

B
z
u
r

B
r
u

B
f (

B
)

11

22


33

22


rr

zz

2(
rr

zz
)

h
46 Chapter 1. Introduction to Rheology
[1.71]
[1.72]
[1.73]
This t ype of flow pr oduces t wo dist inct st r ess differ ences: and
. Planar ext ensional viscosit y is defined in t er ms of t he most
easily measur ed st r ess differ ence, :
[1.74]
It is difficult t o gener at e planar ext ensional flow and exper iment al t est s
of t his t ype ar e less common t han t hose involving t ensile or biaxial flow.
Relati on Between Extensi onal and Shear Vi scosi ti es. The fol-
lowing limit ing r elat ionships bet ween ext ensional and shear viscosit ies
can be expect ed for non-Newt onian fluids at small st r ains (Dealy, 1994;
Walt er s, 1975; Pet r ie, 1979):
[1.75]
[1.76]
[1.77]
Reliable r elat ionships for non-Newt onian fluids at lar ge st r ains have
not been developed. The above equat ions may be pr ecisely defined for
t he special case of Newt onian fluids:
[1.78]
[1.79]
[1.80]
Eq. [1.78], [1.79], and [1.80] can be used t o ver ify t he oper at ion of
ext ensional viscomet er s. Clear ly, however , a Newt onian fluid must be
ext r emely viscous t o maint ain it s shape and give t he solid-like
appear ance r equir ed in many ext ensional flow t est s. Ext ensional
behavior of low viscosit y fluids can be evaluat ed wit h t he met hod of
opposing jet s (Sec. 4.5), by spinning (Sec. 4.6), or by invest igat ing
t ubeless siphon behavior (Sec. 4.7).
u
1

h
x
1
u
2

h
x
2
u
3
0

11

22

11

33

11

22

P
f (

h
)

11

22

h
lim

h
0

E
(

h
) 3 lim

0
(

)
lim

B
0

B
(

B
) 6 lim

0
(

)
lim

h
0

P
(

h
) 4 lim

0
(

E
3

B
6

P
4
1.8 Viscoelastic Material Functions 47
Tr out on est ablished a mat hemat ical r elat ionship bet ween t ensile
ext ensional viscosit y (he called it t he coefficient of viscous t r act ion) and
shear viscosit y (Tr out on, 1906). Pr esent ly, dat a for ext ensional and
shear viscosit ies ar e oft en compar ed using a dimensionless r at io known
as t he Tr out on number ( ):
[1.81]
Since ext ensional and shear viscosit ies ar e funct ions of differ ent st r ain
r at es, a convent ional met hod of compar ison is needed t o r emove
ambiguit y. Based on a consider at ion of viscoelast ic and inelast ic fluid
behavior , J ones et al. (1987) advocat ed t he following convent ions in
comput ing t he Tr out on number s for uniaxial and planar ext ensional
flow:
[1.82]
[1.83]
meaning t hat shear viscosit ies ar e calculat ed at shear r at es equal t o
or for uniaxial or planar ext ension, r espect ively. Using t he
similar consider at ions, Huang and Kokini (1993) showed t hat t he
Tr out on number for case of biaxial ext ension should be calculat ed as
[1.84]
The Tr out on r at io for a Newt onian fluid may be det er mined fr om Eq.
[1.78], [1.79], and [1.80]: in t ensile ext ension it is equal t o 3; it is 6 and
4, r espect ively, in biaxial and planar flow. Depar t ur e fr om t hese
number s ar e due t o viscoelast ic mat er ial behavior . Exper iment al
r esult s may pr oduce consider ably higher values.
1.8. Vi scoelasti c Materi al Functi ons
Fluids t hat have a significant elast ic component may exhibit unusual
behavior (Fig. 1.10; Fig. 1.11): Weissenber g effect (r od climbing),
t ubeless siphon, jet expansion, and r ecoil. Elast ic behavior may be
evaluat ed using viscomet r ic met hods t o det er mine t he nor mal st r ess
differ ences found in st eady shear flow. Alt er nat ively, viscoelast ic
N
Tr
N
Tr

extensional viscosity
shear viscosity
(N
Tr
)
uniaxial


E
(

h
)
(

h
)
(N
Tr
)
planar


P
(

h
)
(2

h
)

h
2

h
(N
Tr
)
biaxial


B
(

B
)
(

12

B
)
48 Chapter 1. Introduction to Rheology
mat er ial funct ions may be det er mined fr om exper iment s involving t he
applicat ion of unst eady st at e defor mat ions. Gener ally, t hese dynamic
t est ing t echniques may be divided int o t wo major cat egor ies: t r ansient
and oscillat or y. Tr ansient met hods include t est s of st ar t -up flow, ces-
sat ion of st eady shear flow, st ep st r ain, cr eep, and r ecoil. In oscillat or y
t est ing, samples ar e defor med by t he applicat ion of har monically
var ying st r ain which is usually applied over a simple shear field. A
compr ehensive r eview of linear and nonlinear viscoelast ic mat er ial
funct ions is summar ized in Chapt er 5. St eady shear and linear vis-
coelast ic mat er ial funct ions can be r elat ed (see Sec. 5.7) using var ious
empir ical r elat ionships such as t he Cox-Mer z r ule, t he Gleisseles mir r or
r elat ion, and Launs r ule (Bir d et al., 1987).
Creep and Step-Strai n (Stress Relaxati on). In a cr eep t est ,
mat er ial is subject ed t o a const ant st r ess and t he cor r esponding st r ain
is measur ed as a funct ion of t ime, . The dat a ar e oft en plot t ed in
t er ms of t he shear cr eep compliance,
[1.85]
ver sus t ime. In a st ep-st r ain t est , commonly called a st r ess r elaxat ion
t est , a const ant st r ain is applied t o t he t est sample and t he changing
st r ess over t ime is measur ed, . The dat a ar e commonly pr esent ed
in t er ms of a shear st r ess r elaxat ion modulus,
[1.86]
ver sus t ime. Dat a fr om cr eep and st r ess r elaxat ion t est s can also be
descr ibed in t er ms of mechanical (spr ing and dashpot ) analogs (Moh-
senin, 1986; Sher man, 1970; Bar nes et al., 1989; Polakowski and Ripling
(1966)) which will be consider ed in mor e det ail in Chapt er 5. Cr eep and
st r ess r elaxat ion exper iment s can be conduct ed in shear , compr ession,
or t ension. Shear cr eep par amet er s for var ious cr eamy st yle salad
dr essings ar e given in Appendix [6.19].
Osci llatory Testi ng. The viscoelast ic behavior of fluids can be
det er mined fr om dynamic t est ing wher e samples ar e subject ed t o
oscillat or y mot ion when held in var ious cont ainment syst ems, usually
a cone and plat e or a par allel plat e appar at us. Typically, a sinusoidal
st r ain is applied t o t he sample causing some level of st r ess t o be
t r ansmit t ed t hr ough t he mat er ial. The magnit ude and t he t ime lag of
t he t r ansmission depend on t he viscoelast ic nat ur e of t he t est subst ance.
(t )
J f (t )

constant
(t )
G f (t )

constant
1.9 Attacking Problems in Rheological Testing 49
In viscous (mor e liquid like) mat er ials, much of t he st r ess is dissipat ed
in fr ict ional losses; it is most ly t r ansmit t ed in highly elast ic mat er ials.
Likewise, t he t ime lag (also called t he phase lag) is lar ge for highly
viscous subst ances but small for mat er ials exhibit ing a high degr ee of
elast icit y. Invest igat ing t his t ype of phenomena leads t o t he definit ion
of var ious mat er ial funct ions: complex viscosit y ( ), dynamic viscosit y
( ), complex modulus ( ), loss modulus ( ), and st or age modulus ( ).
These funct ions ar e discussed in det ail in Chapt er 5. Somet imes,
oscillat or y t est ing is r efer r ed t o as "small amplit ude oscillat or y t est ing"
because small defor mat ions must be employed t o maint ain linear vis-
coelast ic behavior . Typical oscillat or y dat a for var ious food pr oduct s
ar e summar ized in Appendices [6.20], [6.21], and [6.22].
1.9. Attacki ng Problems i n Rheologi cal Testi ng
At t acking r heological pr oblems involves a cr it ical judgement
r egar ding t he t ype of flowbehavior involved and a car eful det er minat ion
of t he appr opr iat e inst r ument s and t echniques t o use in finding a
solut ion. Asimple classificat ion of mat er ial behavior (Fig. 1.25) pr ovides
a useful fr amewor k t o appr oach r heological t est ing of an unknown fluid.
Behavior al ext r emes would be t hat of pur e Hookean behavior (ideally
elast ic mat er ial) and pur e Newt onian behavior (ideally viscous mat e-
r ial); hence, t hese cat egor ies have been placed on t he upper r ight and
left ext r emit y of t he figur e. This symbolizes t he fact t hat all r eal
mat er ials exhibit bot h viscous and elast ic behavior alt hough one t ype
of behavior is fr equent ly dominant . Wat er , for inst ance, is consider ed
Newt onian but will show some degr ee of elast icit y under condit ions
involving a ver y shor t pr ocess t ime, e.g., when a high velocit y object
impact s a body of wat er . In evaluat ing solids, one is t ypically looking
at a st r ess-st r ain r elat ionship as opposed t o a fluid wher e a shear
st r ess-shear r at e r elat ionship is st udied.
When invest igat ing t he behavior of a new fluid, one must fir st
det er mine if t he mat er ial can be consider ed inelast ic (pur ely viscous)
meaning t hat behavior associat ed wit h elast icit y (die swell, r od clim-
bing, et c.) is not impor t ant in t he applicat ion. If pur ely viscous, t he next
quest ion deals wit h t ime-dependency and involves issues of st r uct ur al
st abilit y or br eakdown when subject ed t o a shear for ce. Wit h mat er ials
t hat ar e t ime-independent , a r heogr am may be developed and differ ent
mat hemat ical equat ions (power law, Bingham, Her schel-Bulkley, or
any of t hose pr esent ed in Table 1.3) consider ed t o find a model t hat
accur at ely descr ibes flow behavior . Viscoelast ic fluids, t hose showing

G
*
G

50 Chapter 1. Introduction to Rheology


Figure 1.25. Simple classification of rheological behavior.
significant levels of bot h viscous and elast ic behavior , may be t est ed
using dynamic met hods or st at ic t echniques (cr eep or st r ess r elaxat ion)
wit h var ious mechanical models (discussed in Sec. 5.2) being used t o
int er pr et r esult s: Kelvin, Maxwell or Bur ger s models. Nor mal st r ess
differ ences can be det er mined fr om viscomet r ic flow. Ext ensional
defor mat ion may be used t o det er mine addit ional mat er ial funct ions
associat ed wit h st r et ching.
Classifying fluids is a valuable way t o concept ualize fluid behavior ;
however , it is not meant t o imply t hat t he t ypes of behavior not ed in
Fig. 1.25 ar e mut ually exclusive. A mat er ial showing elast ic behavior
(such as dough) may simult aneously be shear -t hinning and t ime-
dependent ! Ot her fact or s, like aging, may also influence r heological
behavior . Tomat o ket chup, for example, may be pr oper ly descr ibed as
a t ime-independent , shear -t hinning, fluid immediat ely aft er manufac-
t ur e but aging oft en gives t he mat er ial a weak gel like st r uct ur e causing
t he pr oduct t o exhibit t hixot r opic behavior when used by t he consumer .
This explains why agit at ing t he ket chup, by st ir r ing or shaking in t he
bot t le, makes t he condiment mor e pour able. Clear ly, t he abilit y t o
Fluid (Viscous Behavior) Solid (Elastic Behavior)
Newtonian
Non-Newtonian
Time-Dependent
Power Law Bingham Herschel-Bulkley
Rheopectic Thixotropic
Viscoelastic
Hookean Non-Hookean
Fluid-Solid
Other Models
Time-Independent
Non-Linear Elastic
Kelvin Maxwell Burgers
Structural Models
(time-independent)
1.9 Attacking Problems in Rheological Testing 51
concept ualize differ ent t ypes of r heological behavior is ver y impor t ant
in t he development , or impr ovement , of many new food pr oduct s and
pr ocesses.
Esti mati ng Shear Rates i n Practi cal Appli cati ons. It is ver y
impor t ant t hat st eady shear dat a be collect ed over t he appr opr iat e shear
r at e r ange. The minimum shear r at e is oft en zer o due t o t he pr esence
of st at ionar y equipment sur faces. Est imat es of t he maximum shear
r at e found in many pr ocessing syst ems can be obt ained fr om a cr it ical
evaluat ion of t he equipment .
In syst ems wher e fluid is closely cont ained bet ween moving
machiner y par t s, t he maximum shear r at e can be est imat ed fr om t he
velocit y differ ence divided by t he separ at ion dist ance. An example of
t his can be found wit h an anchor impeller t ur ning in a mixing vessel
(Fig. 1.26) wher e t he maximum shear r at e is calculat ed as t he t ip speed
of t he impeller divided by t he gap bet ween t he impeller and t he mixing
t ank:
[1.87]
Spr eading (but t er or mar gar ine) or br ushing (fr ost ing or paint ) oper a-
t ions ar e fr equent ly found in t he food indust r y. In t his case (Fig. 1.26),
t he maximum shear r at e can be est imat ed fr om t he velocit y of t he br ush
(or knife) divided by t he t hickness of t he coat ing:
[1.88]
Est imat ing maximum shear r at es in syst ems wit h widely spaced
moving par t s pr esent s a differ ent pr oblem. Consider , for example, a
paddle mixer in a vessel wher e so t he influence of t he wall is
negligible (Fig. 1.27). Exper iment al dat a, t aken a small dist ance fr om
t he impeller , can be used t o obt ain t he velocit y pr ofile per pendicular t o
t he axis of r ot at ion. If t hese dat a ar e available, t he maximum shear r at e
can be est imat ed fr om t he differ ent ial velocit y and t he height of t he
blade (Fig. 1.27):

max

d
D d

max
u/z
D d
52 Chapter 1. Introduction to Rheology
Figure 1.26. Maximum shear rates found in mixing with an anchor impeller
and brushing.
[1.89]
In t he absence of exper iment al dat a, t he t ip speed of t he impeller ( )
can be used in place of but t his pr ocedur e will give an over
est imat e of t he maximum shear r at e because due t o t he
fact t hat fluid moment um is r apidly dissipat ed in t he mixing vessel.
Dat a of fluid velocit y in t he vicinit y of mixing blades have been collect ed
for a r adial flat -blade t ur bine t ype mixer (Oldshue, 1983; Kout sakos
and Nienow, 1990).
The maximum shear r at e for fluid flow in a t ubular geomet r y (Fig.
1.27) can be det er mined fr om t he volumet r ic flow r at e ( ) and t he inside
r adius of t he t ube:
[1.90]
This calculat ion is exact for Newt onian fluids but must be modified for
power law mat er ials t o include t he flow behavior index:
[1.91]
d
D
Anchor Impeller in Mixing Vessel
u
z
Brushing
max
= u/z
= d
max
/(D-d)
mixing
blade

max
(u
b/2
u
b
)/(b/2)
d/2
(u
b/2
u
b
)
(u
b/2
u
b
) < (d/2)
Q

max

4Q
R
3

max

3n + 1
4n
_

,
4Q
R
3
1.10 Interfacial Rheology 53
Figure 1.27. Maximum shear rate in a mixer when (where is the vessel
diameter), and flow of a Newtonian fluid in a tube.
Eq. [1.91] is ver y helpful in est imat ing t he shear r at e r ange needed t o
det er mine r heological dat a for pipeline design calculat ions because
many fluid foods exhibit shear -t hinning behavior , . These
r elat ionships can also be used in consider ing flow t hr ough t ube t ype
syst ems such as oint ment and t oot h past e cont ainer s, fr ost ing t ubes,
and spr ay nozzels. The or igins of Eq. [1.90] and [1.91] will be clar ified
in a lat er discussion of t ube viscomet r y given in see Sec. 2.2.
1.10. Interfaci al Rheology
Bot h t he shear and ext ensional viscosit ies discussed in pr eceding
sect ions of t his chapt er ar e bulk mat er ial pr oper t ies. Flow behavior at
mat er ial int er faces, however , can be ver y differ ent t han bulk flow
behavior . Int er facial r heology is a field of st udy t hat invest igat es
defor mat ions occur r ing at fluid int er faces. Many pr act ical pr oblems
may involve int er facial r heological phenomena: for mat ion, st abilit y and
pr ocessing of foams and emulsions; spr aying and at omizat ion; select ion
of sur fact ant s; and film for mat ion. Bulk and int er facial viscosit ies can
be r elat ed using t he Boussinesq number :
u=0
u
b/2
b
u
b
Velocity Profile: Mixer Impeller
(D>>d)
max
= ( - )/(b/2)
u
b/2 b
u
d
Pump
Flow in a Tube
R
= (4 Q)/( R )
max
3
D d D
0 < n < 1
54 Chapter 1. Introduction to Rheology
[1.92]
Int er facial viscosit y (usually r epor t ed in unit s of sur face poise, sp,
equivalent t o1 g s
-1
) st r ongly affect s bulk flow behavior at fluid int er faces
when (Edwar ds et al., 1991).
Figure 1.28. Disk surface viscometer to measure interfacial shear viscosity.
Var ious t echniques ar e available t o measur e int er facial r heological
behavior . These can be divided int o indir ect met hods involving t he
examinat ion of velocit y pr ofiles or dir ect met hods involving t he mea-
sur ement of int er facial t or sion. The disk sur face viscomet er is a classical
(dir ect ) met hod of measur ing int er facial shear viscosit y (Fig. 1.28). In
t his syst em t he cup is t ur ned and t he r esult ing t or que is measur ed on
t he fixed plat e. To det er mine t he int er facial viscosit y, t he t or que con-
t r ibut ion is consider ed in t wo par t s: 1) one par t , r elat ed t o t he bulk
viscosit y, due t o fluid cont act under t he plat e; 2) one par t , r elat ed t o t he
int er facial shear viscosit y, due t o t he film induced t r act ion along t he
r im of t he plat e. Int er facial shear viscosit y may exhibit Newt onian or
non-Newt onian behavior . Addit ional inst r ument s t o st udy int er facial
r heology include t he deep-channel sur face viscomet er , biconical int er -
facial viscomet er , and var ious t ypes of knife edge viscomet er s. Edwar ds
et al. (1991) gives a det ailed summar y of numer ous measur ement
N
Bo

interfacial viscosity
bulk viscosity (length scale)
N
Bo
>1
fluid
interface
1.11 Electrorheology 55
t echniques for t he det er minat ion of int er facial shear viscosit y. Fut ur e
r esear ch in int er facial r heology will lead t o many impr oved pr oduct s
and pr ocesses in t he food indust r y.
1.11. Electrorheology
The phenomenon of elect r or heology r efer s t o changes in t he r heo-
logical behavior due t o t he imposit ion an elect r ic field on a mat er ial. It
is somet imes called t he Winslow effect aft er W.M. Winslow who
discover ed it in 1947 (Winslow, W.M. 1947. U.S. Pat ent Specificat ion
2417850). Ther e has been a gr eat deal of int er est in elect r or heological
(ER) fluids for use in var ious mechanical devices (Block and Kelly, 1988):
clut ches, br akes, hydr aulic valves, act ive or t unable damper syst ems,
wide-band-high-power vibr at or s, chucks, exer cise equipment , and
r obot ic cont r ol syst ems. In t he fut ur e, ER fluids will have a st r ong
impact on t he aut omot ive indust r y and may event ually lead t o impr oved
designs for food manufact ur ing machiner y. It may also be possible t o
develop unique pr ocessing schemes and new pr oduct s for foods t hat
exhibit an ER effect . The sear ch for indust r ially viable elect r or heo-
logical fluids has been hinder ed by t he abr asiveness and chemical
inst abilit y of candidat e mat er ials. Similar possibilit ies, and pr oblems,
ar e found wit h magnet or heological fluids wher e flow behavior may be
changed wit h t he imposit ion of a magnet ic field.
ER fluids ar e disper sions of solid par t iculat es, t ypically 0.1 t o 100 m
in diamet er , in an insulat ing (non-conduct ing) oil. At low shear r at es,
in t he absence of an elect r ic field, par t icles ar e r andomly dist r ibut ed
(Fig. 1.29) and many ER fluids will show near ly Newt onian behavior .
Wit h t he applicat ion of an elect r ic field, par t icles become polar ized,
causing par t icle alignment acr oss t he elect r ode gap cr eat ing an
enhanced fiber -like st r uct ur e. This alignment , associat ed wit h inher ent
elect r ical char ges on t he par t icles, causes ER subst ances t o t hicken
dr amat ically. Applicat ion of a volt age causes some mat er ials t o develop
high yield st r esses char act er ist ic of Bingham plast ic behavior . In fact ,
t he yield st r esses can be so high t hat flow ceases, effect ively t r ans-
for ming t he mat er ial fr om a liquid t o a solid. Result s ar e influenced by
many fact or s: nat ur e (alt er nat ing or dir ect cur r ent ) and st r engt h of t he
elect r ic field, t emper at ur e, composit ion and volume fr act ion of par t icles,
shear r at e, r heological and dielect r ic pr oper t ies of t he disper sing oil.
Ext ensive exper iment al wor k is needed t o evaluat e t he flow behavior of
a par t icular ER fluid. To fully under st and t his pr oblem involves a

56 Chapter 1. Introduction to Rheology


Figure 1.29. Particle alignment with and without an electric field (voltage per
unit length) applied across a narrow gap.
car eful st udy of t he int er play bet ween viscous, t her mal, and polar izat ion
for ces (Zukoski, 1993). Comment s similar t o t he above may be made
for magnet or heological fluids wher e t he magnet ic domains of t he par -
t icles r ot at e unt il t hey line up wit h t he applied magnet ic field.
Ther e ar e a number of food r elat ed ER fluids such as cellulose or
sodium car boxymet hyl cellulose in liquid par affin, and st ar ch or gelat ine
in olive oil. Milk chocolat e is also known t o be an ER fluid. Dauber t
and St effe (1996) obser ved an ER r esponse in t his mat er ial: milk
chocolat e r heogr ams wer e shift ed upwar d wit h an incr ease in elect r ic
field st r engt h. A t ypical example of t his is illust r at ed in Fig. 1.30. The
yield st r ess, defined by t he Casson equat ion, also incr eased wit h volt age.
Temper at ur e changes had a ver y int er est ing effect on ER behavior . In
t he pr esence of a volt age, incr easing t he t emper at ur e caused milk
chocolat e t o t hicken pr oducing an upwar d shift in t he r heogr am. It
appear s t hat par t icle polar it y was enhanced at higher t emper at ur es.
In t he absence of a volt age t he usual t r end, a decr ease in appar ent
viscosit y wit h t emper at ur e incr eases, was obser ved.
No Electric Field Applied Electric Field
+
+
+
+
-
-
-
-
1.12 Viscometers for Process Control and Monitoring 57
Figure 1.30 Typical influence of electric field strength (DC Volts/mm) on the
flow behavior of molten milk chocolate.
1.12. Vi scometers for Process Control and Moni tori ng
The goal of t his sect ion is t o pr ovide an over view of t he pr imar y
measur ement concept s and issues involved in t he use of viscomet er s for
cont inuous pr ocess cont r ol or monit or ing. All pr ocess viscomet er s used
for food pr oduct s must confor m t o appr opr iat e sanit ar y st andar ds and
accept ed pr act ices such as t he 3-A and Egg 3-A st andar ds published by
t he Int er nat ional Associat ion of Milk, Food and Envir onment al Sani-
t ar ians (Des Moines, IA). Many pr ocess cont r ol viscomet er s wer e not
designed for food applicat ions and cannot be modified for accept able
sanit ar y oper at ion. Viscomet er s discussed her e ar e t ypical of indus-
t r ially available unit s which ar e gener ally accept able for use in t he food
indust r y. The cur r ent focus is on unit s t hat evaluat e a st eady shear
viscosit y. An indust r ial syst em t o det er mine ext ensional viscosit y,
based on flow t hr ough an or ifice (see Sec. 4.4.3), is also available.
0 0.5 1 1.5 2
0
2
4
6
8
10
12
14
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
0 DC Volts 150 DC Volts 300 DC Volts
Milk Chocolate
58 Chapter 1. Introduction to Rheology
In-li ne or On-li ne Installati on. Viscomet er s for pr ocess cont r ol or
monit or ing may be inst alled using var ious measur ement schemes.
"In-line" syst ems ar e inst alled dir ect ly in t he pr ocess line, usually in a
pipe. These syst ems ar e subject t o pr ocess var iat ions, such as changes
in sample t emper at ur e, which may significant ly influence sensor out -
put . "On-line" unit s make measur ement s on a pr oduct side st r eam, also
called a by-pass loop, t aken fr om t he main pr ocess flow line. One
advant age of t his t ype of syst em is t hat sample var iables (including flow
r at e, t emper at ur e, and pr essur e) may be cont r olled dur ing t est ing. A
t hir d t ype of pr ocess cont r ol viscomet er is t he immer sion syst em
designed for use in pr ocess vessels, par t icular ly mixing t anks. The above
unit s pr ovide alt er nat ives t o "off-line" measur ement s wher e a small
sample is r emoved fr om t he pr ocess line and evaluat ed in a st andar d
labor at or y inst r ument .
Figure 1.31. Capillary, on-line viscometer using side stream flow.
Measurement Concepts. A side st r eam capillar y viscomet er is
illust r at ed in Fig. 1.31. Sample flow r at e is det er mined by t he speed of
t he gear pump. The syst em may be oper at ed in t wo modes: const ant
flow r at e or const ant pr essur e. In t he const ant pr essur e mode, t he gear
pump is dr iven at t he speed r equir ed t o maint ain a set pr essur e at t he
ent r ance t o t he capillar y. In t he const ant flow r at e mode, t he pump
main flow
side stream
gear pump
capillary
pressure
transducer
1.12 Viscometers for Process Control and Monitoring 59
speed is fixed and t he pr essur e dr op r equir ed t o for ce mat er ial t hr ough
t he capillar y is measur ed. Regar dless of t he mode of oper at ion, viscosit y
is dir ect ly pr opor t ional t o t he pr essur e dr op acr oss t he capillar y divided
by t he flow r at e t hr ough t he syst em. A commer cial ver sion of t he side
st r eam capillar y viscomet er , designed pr imar ily for polymer melt s, is
pr oduced by Goet t fer t , Inc. (Rock Hill, SC).
Figure 1.32. Concentric cylinder type in-line viscometer.
An in-line concent r ic cylinder viscomet er measur es t he t or que, on
t he inner cylinder , gener at ed by t he out er cylinder moving at a fixed
speed (Fig. 1.32). Viscosit y is pr opor t ional t o t he t or que divided by t he
speed of t he out er cylinder . The viscomet er r elies on per for at ions in t he
bob and cup t o achieve cont inuous flow t hr oughout t he annular gap
wher e fluid pr oper t ies ar e measur ed. An alt er nat ive r ot at ional vis-
comet er is illust r at ed in Fig. 1.33. In t his case, an off-set cylindr ical
element is at t ached t o a r ot at ing shaft causing a gyr at or y mot ion of t he
sensor . A cont inuous flow of t est fluid moves t hr ough a per for at ed
sheat h dur ing t est ing. Tor que r equir ed t o maint ain a fixed speed of
r ot at ion is measur ed. Viscosit y is pr opor t ional t o t he t or que r equir ed
t o maint ain a const ant speed of r ot at ion. Br ookfield Engineer ing
Labor at or ies (St ought on, MA), and C.W. Br abender Inst r ument s (S.
Hackensack, NJ ), r espect ively, pr oduce inst r ument s like t hose illus-
t r at ed in Fig. 1.32 and 1.33.
outlet inlet
transducer
stator
rotor
drive motor
torque
perforation
60 Chapter 1. Introduction to Rheology
Figure 1.33. In-line viscometer using an off-set rotating element.
Figure 1.34. In-line viscometer using a vibrating rod.
torque
transducer
flow
rotating
sensor
perforated
sheat
flow
vibrating rod
transducer
1.12 Viscometers for Process Control and Monitoring 61
Figure 1.35. In-line viscometer using a vibrating sphere.
Vibr at ional viscomet er s ar e consider ed sur face loaded syst ems
because t hey r espond t o a t hin layer of fluid at t he sur face of t he sensor .
In-line unit s may involve r od (Fig. 1.34) or spher ical (Fig. 1.35) sensor s.
In each case, t he sensor is dr iven at a fixed fr equency and t he power
r equir ed t o maint ain a pr ecise amplit ude is measur ed. Since t he
vibr at ing pr obe acceler at es t he fluid, power input is pr opor t ional t o
pr oduct of viscosit y and densit y. Densit y compensat ion can be incor -
por at ed allowing a dir ect comput at ion of Newt onian viscosit y. Spher ical
t ype vibr at ional viscomet er s ar e manufact ur ed by Namet er (Met uchen,
NJ ).
An immer sion t ype, falling body viscomet er is illust r at ed in Fig. 1.36.
A pist on is per iodically r aised allowing sample t o fill t he cylindr ical
cont ainer . Dur ing t est ing, sample is expelled fr om t he cylinder by t he
pist on which is allowed t o fall under t he influence of gr avit y. Using
r efer ence dat a for st andar d Newt onian fluids, t he falling t ime is cor -
r elat ed t osample viscosit y. This concept can also be applied t o an on-line
pr ocessing syst em. Var ious falling pist on viscomet er s ar e manufact ur ed
by Nor cr oss Cor por at ion (Newt on, MA).
Practi cal Consi derati ons i n Selecti ng a Process Control Vi s-
cometer. It is impor t ant t o have a good under st anding of t he fluid
under consider at ion befor e select ing a pr ocess cont r ol viscomet er . The
influence of t emper at ur e, ingr edient for mulat ion, and pr ocessing con-
dit ions on flow behavior must be ascer t ained befor e qualit y cont r ol or
set -point s can be accur at ely est ablished. Temper at ur e has such a st r ong
flow
transducer
vibrating sphere
62 Chapter 1. Introduction to Rheology
Figure 1.36. Falling piston, immersion type: (A) loading phase, (B) end of mea-
surement.
influence on r heological behavior t hat it is usually necessar y t o eit her
car efully cont r ol it when conduct ing measur ement s or compensat e for
it when making final calculat ions.
Pr ocess cont r ol viscomet er s ar e gener ally designed for Newt onian
fluids wher e t he viscosit y is not a funct ion of shear r at e. To evaluat e
non-Newt onian fluids, mult iple dat a point s, t aken at differ ent shear
r at es ar e r equir ed. Hence, oper at ional shear r at es must be est ablished
and mat ched t o t he capabilit ies of t he pr ocess cont r ol viscomet er . The
shear r at es can be easily est imat ed in t he capillar y viscomet er s (Fig.
1.31) or concent r ic cylinder syst ems (Fig. 1.32), but t hey ar e ver y difficult
t o calculat e in ot her s such as t he unit wit h t he offset r ot at ional element
illust r at ed in Fig. 1.33. Det er mining fundament al r heological pr oper -
t ies when shear r at es cannot be evaluat ed is ver y complex; hence, some
measuring tube
piston
fluid level
tube opening
flow
A
B
1.13 Empirical Measurement Methods for Foods 63
pr ocess cont r ol viscomet er s can only gener at e compar at ive flow
behavior dat a (not absolut e r heological pr oper t ies) for non-Newt onian
fluids.
Pr ocess cont r ol viscomet er s may pr ovide a st r ict ly empir ical
par amet er , an appar ent viscosit y, or a flow cur ve if samples ar e
defor med at mult iple shear r at es. This infor mat ion must be cor r elat ed
t o specific pr ocessing fact or s, such as t he amount of cocoa but t er in
chocolat e or t he amount of wat er in t omat o past e, befor e a cont r ol scheme
can be init iat ed. If an on-line inst r ument is used for qualit y cont r ol,
t hen qualit y must be car efully defined and dir ect ly r elat ed t o t he r he-
ological pr oper t y being measur ed. Also, t he accept able var iat ion in
qualit y must be known t o est ablish t he pr oper limit s (or set -point s)
r equir ed in developing a cont r ol st r at egy. All pr ocess cont r ol viscome-
t er s must be calibr at ed r egular ly and car efully obser ved t o ensur e
sat isfact or y long t er m per for mance.
1.13. Empi ri cal Measurement Methods for Foods
The food indust r y uses many empir ical inst r ument s (Table 1.5) t o
measur e t he flow behavior of food pr oduct s. These devices ar e not used
t o det er mine fundament al r heological pr oper t ies, but r esult s may find
diver se applicat ions: qualit y cont r ol, cor r elat ion t o sensor y dat a, or even
ser ve as official st andar ds of ident it y. Food engineer s may find it
necessar y t o r eplace empir ical devices, like t he Bost wick Consist omet er
used for pur eed foods, wit h mor e fundament al inst r ument s t o achieve
engineer ing object ives r elat ed t o pr ocess cont r ol. Wit h t he except ion of
t he melt flow indexer for molt en polymer s, all t he inst r ument s discussed
in t his sect ion ar e used for food pr oduct s.
It is impor t ant t o r ecognize t he fact t hat numer ous foods ar e so
complex it is not pr act ical, and in many cases not possible, t o measur e
t heir fundament al r heological pr oper t ies. Car r ot s, peanut s, peas, or
beans (for example) ar e non-homogeneous, nonisot r opic mat er ials wit h
complex geomet r ies. Only empir ical t est ing devices, capable of mea-
sur ing composit e mat er ial behavior , pr ovide a suit able means of char -
act er izing t hese foods. Empir ical inst r ument s ar e a valuable and well
est ablished par t of t he food indust r y. Since t hey do not measur e
fundament al pr oper t ies, t hey may appr opr iat ely be called indexer s.
Some of t he most common unit s ar e descr ibed in t his sect ion. Consult
Bour ne (1982) and Br ennan (1980) for addit ional infor mat ion on eval-
uat ing food t ext ur e.
64 Chapter 1. Introduction to Rheology
Table 1.5. Typical Empirical Testing Instruments and Applications for
Food Products (S ummarized from Bourne, 1982)
Device Common Applica t ion
Ada ms Con s is t omet er con s is t en cy of s emiflu id food pu r ees
Ar mou r Ten der omet er beef t en der n es s
Ba ker Compr es s imet er s t a len es s of br ea d
Ba lla u f Pr es s u r e Tes t er pu n ct u r e t es t in g of fr u it a n d veget a bles
BBIRA Bis cu it Text u r e Met er h a r dn es s of cookies a n d cr a cker s
Bloom Gelomet er pu n ct u r e t es t of gela t in s a n d gela t in jellies
Bos t wick Con s is t omet er flow of ba by foods a n d s imila r pu r ees
Ch a t illon Pr es s u r e Tes t er pu n ct u r e t es t in g of fr u it a n d veget a bles
Effi-Gi Pr es s u r e Tes t er pu n ct u r e t es t in g of fr u it a n d veget a bles
Ext en s igr a ph beh a vior of wh ea t dou gh
Fa r in ogr a ph ba kin g qu a lit y of wh ea t flou r
FMC Pea Ten der omet er qu a lit y a n d ma t u r it y of fr es h gr een pea s
FTC Text u r e Tes t Sys t em a t t a ch men t s for ma n y foods
GF Text u r omet er a t t a ch men t s for ma n y foods
Ha u gh Met er egg qu a lit y
Hilker -Gu t h r ie Plu mmet fir mn es s of cu lt u r ed cr ea m
In s t r on Un iver s a l Tes t in g Ma ch in e a t t a ch men t s for ma n y foods
Kr a mer Sh ea r Pr es s t en der n es s of pea s a n d ot h er pa r t icu la t e foods
Ma gn es s -Ta ylor Pr es s u r e Tes t er pu n ct u r e t es t in g of fr u it a n d veget a bles
Ma r in e Colloids Gel Tes t er pu n ct u r e t es t ma r in e ext r a ct gels
Mixogr a ph ba kin g qu a lit y of wh ea t flou r
Ot t a wa Pea Ten der omet er qu a lit y a n d ma t u r it y of fr es h gr een pea s
Ot t a wa Text u r e Mea s u r in g Sys t em a t t a ch men t s for ma n y foods
Pa bs t Text u r e Tes t er fir mn es s of pa r t icu la t e foods
Pen et r omet er fir mn es s of bu t t er a n d ma r ga r in e
Plin t Ch ees e Cu r d Tor s iomet er s et t in g of ch ees e cu r d
Res is t ogr a ph ba kin g qu a lit y of wh ea t flou r
Ridgelimit er s t iffn es s of pect in a n d fr u it jellies
St even s Compr es s ion Res pon s e a t t a ch men t s for ma n y foods
An a lyzer
Su ccu lomet er ma t u r it y a n d qu a lit y of fr es h s weet cor n
SURDD Ha r dn es s Tes t er h a r dn es s of fa t s a n d wa xes
Tor r y Br own Homogen izer t ou gh n es s of fis h
USDA Con s is t omet er con s is t en cy of s emiflu id food pu r ees
Va n Dor r a n Pr es s u r e Tes t er pu n ct u r e t es t in g of bu t t er
Wa r n er -Br a t zler Sh ea r t ou gh n es s of mea t
1.13 Empirical Measurement Methods for Foods 65
Dough Testi ng Equi pment (Fari nograph, Mi xograph, Extensi -
graph, Alveograph). Dough - a combinat ion of cer eal (usually wheat )
flour , wat er , yeast , salt , and ot her ingr edient s - is pr obably t he most
complex mat er ial facing t he food r heologist ; hence, it is not sur pr ising
t hat many empir ical inst r ument s have been developed t o evaluat e t he
flow behavior of dough. The inst r ument s may be divided int o t wo major
gr oups: t hose which measur e t he power input dur ing dough develop-
ment caused by a mixing act ion, and t hose which subject pr epar ed
(developed) dough t o an ext ensional defor mat ion. The following
discussion will descr ibe t he major inst r ument s found in each gr oup.
St andar d met hods for oper at ing t hese inst r ument s and int er pr et ing
dat a ar e published by t he Amer ican Associat ion of Cer eal Chemist s (St .
Paul, MN), and t he Int er nat ional Associat ion for Cer eal Chemist r y and
Technology.
One of t he most widely used dough mixer s is t he Fari nograph
(DAppolonia and Kuner t h, 1984). This inst r ument combines dough
ingr edient s using t wo Z-shaped mixing blades t hat r ot at e, at differ ent
speeds, in opposit e dir ect ions. Mixing is init iat ed wit h dr y flour and
wat er is added fr om a t it r at ing bur et dur ing t est ing. A dynamomet er
is used t o r ecor d t or que on t he dr ive shaft of t he mixing blades. Out put
is given as a far inogr am: a plot of an inst r ument -dependent par amet er
pr opor t ional t o t or que, expr essed as a Br abender unit (BU, also called
consist ency), ver sus t ime. The shape of t he far inogr am is int er pr et ed
in t er ms of fact or s r elat ed t o flour qualit y and t he behavior of t he dough
in t he baker y: dough development t ime, st abilit y, mixing t oler ance, and
degr ee of soft ening. The amount of wat er r equir ed t o give a consist ency
of 500 BU t o a 14%moist ur e cont ent (wet basis) flour is alsoan impor t ant
flour par amet er , known as t he far inogr aph wat er adsor pt ion, det er -
mined using t he Far inogr aph.
An alt er nat ive t o t he Far inogr aph is t he Mi xograph which involves
a planet ar y r ot at ion of ver t ical pins (lower ed int o t he dough) about
st at ionar y ver t ical pins at t ached t o t he mixing bowl. Tor que is r ecor ded
while mixing a fixed amount of flour and wat er . Result s ar e given in
t er ms of a mixogr am which is int er pr et ed in a manner analogous t o t hat
discussed for t he far inogr am.
The Extensi graph (Rasper and Pr est on, 1991) gener ally conduct s
t est s on doughs pr epar ed in t he Far inogr aph. A special molding device
shapes t he dough int o a cylindr ical specimen which is placed hor izon-
t ally int o a suppor t syst em. The ends ar e clamped fir mly in place leaving
66 Chapter 1. Introduction to Rheology
t he middle sect ion fr ee for t est ing. A hook cont act s t he middle of t he
sample and st r et ches it while moving downwar d at a const ant speed.
The for ce on t he sample, caused by t he downwar d mot ion of t he hook,
is r ecor ded. Result s ar e given as an ext ensogr am, a plot of for ce (in
Br abender unit s) ver sus t ime and ext ension, which pr ovides valuable
qualit y cont r ol infor mat ion for t he dough. It is impor t ant t o not e t hat
mat er ials wit h similar far inogr ams may have ver y differ ent ext enso-
gr ams. The effect s of oxidizing agent s and enzymes on dough behavior ,
for example, can oft en be evaluat ed wit h ext ensogr ams.
The Alveograph (Shuey and Tipples, 1980), also called t he Chopin
Ext ensigr aph, measur es dough behavior when subject ing it t o an
ext ensional defor mat ion. In t his inst r ument , a cir cular disk is cut fr om
a sheet of dough and clamped, ar ound it s cir cumfer ence, t o t he base
plat e of t he t est appar at us. Air flowing t hr ough t he base plat e causes
t he dough t o expand int o a spher ically shaped bubble which event ually
r upt ur es complet ing t he t est . The air pr essur e in t he bubble over t ime
is r ecor ded and plot t ed as an alveogr am. In r out ine t est ing, t he maxi-
mum height , over all lengt h, and t he ar ea under t he cur ve ar e t he pr i-
mar y par amet er s t aken off t he alveogr am. Alveogr aph dat a have been
used t o calculat e biaxial ext ensional viscosit y (Far idi and Rasper , 1987;
Launay and Bur , 1977).
Cone Penetrometer. St iff mat er ials --like but t er , peanut but t er , or
mar gar ine-- ar e oft en assessed for "spr eadabilit y" using cone penet r a-
t ion dat a. This inst r ument consist s of a weight ed cone t hat is posit ioned
ver t ically over t he flat sur face of t he t est sample. Cone angles of 20 or
45 degr ees ar e t ypical. In st andar d t est ing, t he cone is r eleased int o t he
sample and t he dept h of penet r at ion, aft er a fixed per iod of t ime, is
measur ed. Since t est mat er ials have a high yield st r ess, t he cone comes
t o r est quickly. Result s may be pr esent ed in t er ms of a yield value which
is dir ect ly pr opor t ional t o t he weight of t he cone assembly and inver sely
pr opor t ional t o t he dept h of penet r at ion (Haight on, 1959; Sone, 1972).
Oper at ing cone penet r omet er s wit h a const ant downwar d speed, inst ead
of a const ant weight , is also an effect ive met hod of obt aining exper -
iment al dat a (Tanaka et al., 1971). St andar d met hods for t est ing
lubr icat ing gr eases involve double angle cones: one cone, wit h a small
angle, mount ed on a second cone wit h a lar ger angle.
1.13 Empirical Measurement Methods for Foods 67
Warner-Bratzler Shear. The basic par t of t his inst r ument is made
fr om a 1 mm t hick st eel blade wit h a t r iangular hole cut fr om it . Met al
bar s, one locat ed on each side of t he blade, ser ve t wo funct ions: t hey
or ient t he blade and push t he t est sample int o t he V shaped not ch of
t he t r iangular opening. Maximum for ce t o cut t hr ough t he sample is
r ecor ded. Higher levels of cut t ing for ce ar e associat ed wit h incr easing
sample t oughness. The War ner -Br at zler Shear is ext ensively used t o
evaluat e t he t ext ur e of r aw and cooked meat s. In t hese exper iment s,
car eful sample pr epar at ion and or ient at ion ar e essent ial for obt aining
r epr oducible r esult s. Ot her foods such as car r ot s, celer y, r hubar b, and
aspar agus have also been t est ed in t his device.
Bostwi ck Consi stometer. The inst r ument is a simple device used t o
evaluat e t he flow char act er ist ics of pur eed foods including applesauce,
cat sup, and numer ous baby-food pr oduct s. These unit s ar e usually made
of st ainless st eel and consist of t wo abut t ing compar t ment s connect ed
wit h a common floor but separ at ed by a spr ing-loaded gat e. The fir st
compar t ment is 5 X 5 X 3.8 cm when t he gat e is lower ed. This sect ion
is loaded wit h fluid at t he beginning of t he t est which is init iat ed by
pr essing a t r igger t hat r eleases t he gat e. Fluid flows, under t he influence
of gr avit y, int o t he second compar t ment consist ing of an inclined t r ough
which is 5 cm wide, 24 cm long and appr oximat ely 2.5 cm high. The
floor of t he t r ough is gr aduat ed in 0.5 cm incr ement s and movement
down t he t r ough r eflect s fluid pr oper t ies. Measur ement s ar e t aken aft er
a specified t ime (t ypically 5 t o 30 s) and r epor t ed as cent imet er s of t r avel
fr om t he st ar t ing gat e. If fluid mot ion pr oduces a cur ved sur face, t he
t r avel dist ance of t he leading edge is r epor t ed. The Bost wick Consis-
t omet er is st ill widely used as a qualit y cont r ol t ool by t he food indust r y.
It is difficult t o pr ecisely r elat e Bost wick r eadings t o fundament al
r heological behavior (Ver cr usse and St effe, 1989); however , significant
pr ogr ess has been made for Newt onian and power law fluids using a
gr avit y cur r ent analysis (McCar t hy and Seymour (1994) which showed
Bost wick measur ement s t o be linear ly r elat ed t o appar ent viscosit y
divided by densit y r aised t o t he -0.2 power : .
Adams Consi stometer. This device measur es flow, due t o gr avit y,
over a hor izont al plat e made of glass, met al, or st eel. The plat e has a
ser ies of concent r ic cir cles, locat ed one-quar t er inch apar t , r adiat ing out
fr om t he geomet r ic cent er of t he plat e. A t r uncat ed cone is cent er ed on
t he plat e and loaded wit h sample, t hen ver t ically r aised by hand
allowing mat er ial t o flow r adially out war d over t he flat sur face. Aft er
(/)
0.2
68 Chapter 1. Introduction to Rheology
a fixed per iod of t ime (10 t o 30 s is t ypical), t he dist ance t r aveled in each
quadr ant is measur ed and t he aver age value is r ecor ded (in inches) as
t he "Adams Consist ency." The Adams Consist omet er is commonly used
t o evaluat e cr eam st yle cor n and similar pr oduct s.
Zhan Vi scometer. This device is a well known example of an or ifice
viscomet er . It consist s of a small cup-shaped vessel wit h a hole (or ifice)
in t he bot t om. The cup is filled wit h sample which is allowed t o dis-
char ged t hr ough t he or ifice. Time is measur ed fr om t he beginning of
dischar ge unt il t he st eady st r eam coming out of t he cup begins t o dr ip.
Dischar ge t ime is cor r elat ed t o viscosit y. Zhan t ype viscomet er s ar e
used in many qualit y cont r ol applicat ions.
Vi sco-Amylograph. The Visco-Amylogr aph (Shuey and Tipples, 1980)
was designed t o evaluat e t he behavior of st ar ch solut ions dur ing gela-
t inizat ion. It consist s of a r ot at ing bowl wit h eight ver t ical pins and a
mat ching, suspended element , wit h seven ver t ical pins. Tor que is
r ecor ded on t he upper element dur ing r ot at ion of t he bowl. The syst em
includes a t her mor egulat or which allows t he sample t o be heat ed (t he
st andar d r at e is 1.5 C per minut e) dur ing t est ing. When an aqueous
suspension of st ar ch is heat ed above t he gelat inizat ion t emper at ur e,
t he fluid t hickens dr amat ically. A complet e amylogr aph t est usually
involves four dist inct t her mal per iods while t he bowl is r ot at ed at a
const ant speed: heat ing, holding, cooling, and holding. Result s ar e
pr esent ed as an amylogr am which is a plot of t or que (given as viscosit y
in Br abender unit s) ver sus t ime. Amylogr ams have pr oven useful in
evaluat ing t he qualit y of st ar ch and it s behavior as a t hickening agent
in many food syst ems.
Rapi d Vi sco Analyser. This inst r ument gener at es dat a similar t o
t hat pr ovided by t he Visco-Amylogr aph. Small samples, t ypically 3 t o
4 gr ams of st ar ch in wat er , ar e heat ed in a small mixing vessel wit h a
pit ched paddle impeller . The aluminum mixing vessel and plast ic
impeller ar e bot h disposable. Samples ar e subject ed t o user pr ogr am
changes in t emper at ur e (heat ing, holding, and cooling) int ended t o
mat ch pr ocessing condit ions found in a par t icular applicat ion. Tor que,
or inst r ument viscosit y, ar e measur ed over t ime while t he sample is
agit at ed and pr ogr ammed t emper at ur e changes ar e execut ed. This
inst r ument , or iginally int ended t o evaluat e t he qualit y of Aust r alian
wheat , may be used t o examine t he qualit y of a wide var iet y of food
st ar ches.

1.13 Empirical Measurement Methods for Foods 69


Brookfi eld Rotati ng Di sks and T-Bars. Br ookfield Engineer ing
Labor at or ies (St ought on, MA) manufact ur es a r ot at ional inst r ument
(called t he Br ookfield Viscomet er ) t hat is ext ensively used in t he food
indust r y. The most common sensor s ar e flat disks (spindles) at t ached
t o t he inst r ument wit h a ver t ical shaft . Disks ar e available in var ious
sizes and may be r ot at ed at differ ent speeds. Tor que r equir ed t o
maint ain const ant r ot at ion is measur ed. This device can r ead t he vis-
cosit y of Newt onian fluids dir ect ly because it is calibr at ed wit h New-
t onian mat er ials (silicone oils).
Since a t hor ough analysis of shear r at e on t he disk is complex
(Williams, 1979), it is difficult t o use t he Br ookfield Viscomet er for t he
det er minat ion of non-Newt onian fluid pr oper t ies. Simplified
appr oaches suggest ed by Mit schka (1982) and Dur gueil (1987) ar e
applicable t o some foods. These t echniques use numer ous const ant s t o
conver t t or que and angular velocit y dat a int o shear st r ess and shear
r at e values, r espect ively. Br iggs (1995) successfully det er mined t he
shear -t hinning behavior of banana pur ee, salad dr essing, chocolat e
syr up, enchilada sauce, and pancake syr up using Br ookfield spindles
and t he Mit schka (1982) met hod of analysis.
Disk sensor s can be ver y useful in obt aining a r elat ive index of food
t hickness for t he pur pose of compar ing pr oduct s or making qualit y
cont r ol judgment s. In addit ion t o disks, T-shaped bar s ar e made for t he
same pur pose. Inst r ument s, equipped wit h t he T-bar s, can also be
at t ached t o t he Br ookfield Helipat h St and which allows t he ent ir e
inst r ument t o be lower ed dur ing t est ing. This causes t he sensor t o t ake
a spir al pat h t hr ough t he sample while t or que dat a ar e obt ained. The
Helipat h St and is t ypically used for t hick past es and gels wher e a
r ot at ing disk would be difficult t o inser t or cr eat e a channeling effect
dur ing measur ement .
Falli ng Ball Vi scometer. Equat ions for t he falling ball viscomet er
ar e der ived in Example Pr oblem 1.14.4. This t ype of viscomet er involves
a ver t ical t ube wher e a ball is allowed t o fall, under t he influence of
gr avit y, t hr ough a Newt onian fluid. Viscosit y is calculat ed on t he basis
of t he t ime t aken t o fall a fixed dist ance. If t he vessel diamet er is 10
t imes t he ball diamet er , wall effect s can be neglect ed.
The r ising bubble viscomet er r epr esent s anot her applicat ion of t he
falling ball concept . In t his case, a bubble of air is allowed t o ascend
t hr ough a column of sample. Rising t ime over a set dist ance is cor r elat ed
t o Newt onian viscosit y.
70 Chapter 1. Introduction to Rheology
Hoeppler Vi scometer. The Hoeppler Viscomet er is a var iat ion of t he
falling ball concept . In t his case, a ball having a diamet er slight ly
smaller t han t he cylindr ical vessel cont aining t he t est fluid, falls
t hr ough a t ube posit ioned wit h a 10 degr ee inclinat ion fr om t he ver t ical
posit ion. Time of fall over a set dist ance is cor r elat ed t o t he viscosit y of
Newt onian fluids. Unit s of t his t ype ar e somet imes called r olling ball
viscomet er s because descending spher es may r oll along t he wall of t he
viscomet er .
Brabender-FMC Consi stometer. This unit was or iginally designed
t oevaluat e cr eam st yle cor n but it has alsobeen used for ket chup, t omat o
past e, baby food, and similar pr oduct s. It is designed t o lower a t hin,
r ect angular shaped, paddle int o a sample held in a st ainless st eel cup.
The cup is r ot at ed at a single speed of 78 r pm. This mot ion cr eat es a
t or que on t he paddle t hat is r ead fr om a dial locat ed on t he t op of t he
inst r ument . Paddles ar e available in var ious sizes: 5.08 cm (2 inch) by
3.56 cm (1.4 inch) is t ypical.
Figure 1.37. Compression-extrusion and Kramer shear cells commonly used to
evaluate the behavior of particulate foods.
blade holder
shear blades
sample holder
Kramer Shear Cell
Compression-Extrusion Cell
annulus
1.13 Empirical Measurement Methods for Foods 71
Compressi on-Extrusi on Cell. This t est appar at us (Fig. 1.37) is a
common device used t o measur e t he behavior of par t iculat e foods
(Bour ne, 1982). In t he compr ession-ext r usion cell (also called a back
ext r usion cell), sample compr ession causes mat er ial t o flow t hr ough t he
annulus for med bet ween t he plunger and t he cylindr ical cont ainer .
Sample dat a consist of a cur ve r elat ing t he for ce on t he plunger ver sus
t he dist ance of plunger t r avel (or t r avel t ime). Cur ves fr om differ ent
samples, involving differ ent t r eat ment s or var iet ies, ar e compar ed t o
est ablish differ ences in pr oduct t ext ur e.
Kramer Shear Cell. The Kr amer shear cell (Fig. 1.37) is a well
est ablished t ool for evaluat ing t he composit e flowbehavior of par t iculat e
foods. At ypical syst em cont ains 10 shear blades which ar e 3.2 mm t hick
and separ at ed by a dist ance equal t o t he t hickness (3.2 mm). Bar s for m
mat ching slit s in t he t op and bot t om of t he sample holder . The sample
box is appr oximat ely 65 mm wide, long, and deep. Dur ing t est ing, t he
sample holder is filled wit h food and t he shear blades (pr oper ly aligned
wit h t he bar s in t he t op) ar e for ced t hr ough t he mat er ial unt il t hey pass
t hr ough t he bar s in t he bot t om of t he sample holder . For ce on t he r am
holding t he blades is measur ed over t ime and cor r elat ed t o pr oduct
fir mness.
Si mple Compressi on. Biological mat er ials may be evaluat ed in t er ms
of a bioyield point and a r upt ur e point (Mohsenin, 1984). A cur ve such
as t he one illust r at ed in Fig. 1.38 is t ypical for solid foods, like fr uit s
and veget ables, when a cylindr ical sample is t est ed in simple com-
pr ession. The init ial por t ion of t he cur ve (a-b) is a st r aight line up t o
t he linear limit (b). Youngs modulus may be calculat ed fr om t he st r ess
and st r ain at t hat point : . Asecant or a t angent modulus, defined
at a par t icular st r ain, may be calculat ed if t he line is cur ved (Mohsenin,
1986). The slope of t he init ial por t ion of t he cur ve is oft en t aken as an
index of fir mness. When st r ess and st r ain cannot be calculat ed, dat a
may be simply plot t ed in t er ms of for ce and defor mat ion.
The bioyield point (c) is r elat ed t o a failur e in t he micr ost r uct ur e of
t he mat er ial associat ed wit h an init ial disr upt ion of cellular st r uct ur e
(Fig. 1.38). It is obser ved at a st r ess and st r ain of and , r espect ively.
The r upt ur e point (d) of t he mat er ial, defined by and , cor r elat es t o
t he macr oscopic failur e in t he sample. Wit h mor e br it t le mat er ials t he
r upt ur e point may be ver y close t o t he bioyield point : These point s may
be widely separ at ed in t ough mat er ials. The Amer ican Societ y of

b
/
b
E

c

c

d

d
72 Chapter 1. Introduction to Rheology
Figure 1.38. Generalized compression curve for a biological solid.
Agr icult ur al Engineer s (ASAE, St . J oseph, MI) has est ablished a
st andar d met hod for t he compr ession t est ing of food mat er ials of convex
shape (ASAE St andar d S368.2). It is similar t o t he simple compr ession
t est descr ibed her e.
Texture Profi le Analysi s. Text ur e r efer s t o t he human sensat ion of
food der ived fr om it s r heological behavior dur ing mast icat ion and
swallowing. Obt aining a quant it at ive descr ipt ion of t ext ur e using
inst r ument al dat a is ver y complicat ed because no inst r ument can
duplicat e human capabilit ies. Fr om an engineer ing per spect ive, t he
mout h can be consider ed an int r icat e mechanical syst em and chemical
r eact or t hat can cr ush, wet , enzymat ically degr ade, pr essur ize, heat or
cool, pump, chemically sample for t ast e, and sense for ce and t emper a-
t ur e. In addit ion, t his "eat ing machine" has a sophist icat ed feedback
cont r ol syst em. Init ially t her e is open loop, feed for war d cont r ol t o set
pr imar y par amet er s: size of mout h opening, sur face select ion for fir st
bit e (incisor s or molar s), et c. Once t he food is in t he mout h, t her e is an
adapt ive feed back cont r ol syst em wit h a var iable gain -high wit h
unfamiliar foods, low wit h ever yday foods- t hat depends on bolus
a
b
c
d
d
c
b
b
d
c ,
Strain ( )
S
t
r
e
s
s

(



)
bioyield point
1.13 Empirical Measurement Methods for Foods 73
development dur ing mast icat ion. This pr ocess is influenced by many
fact or s: volume of t he or al cavit y, r at e of addit ion of saliva, chemical
composit ion of saliva, chemical and physical int er act ion of saliva and
t he food, r at e of chewing, t ot al number of chews, sur face ar ea in cont act
wit h t he food, movement of t he lips and cheeks, dynamic volume of t he
or al cavit y dur ing mast icat ion, r esidence t ime of t he bolus, init ial vol-
ume of t he bolus, and par t ial fluid r emoval (by swallowing) dur ing
chewing. Given t he above, it is not sur pr ising t hat lit t le pr ogr ess has
been made in cor r elat ing fundament al r heological pr oper t ies t o t he
human per cept ion of t ext ur e. Ther e has been limit ed success wit h some
fluid (Chr ist ensen, 1987; Kokini and Cussler , 1987; Sher man, 1988) and
solid foods (Hamann and Lanier , 1987; Mont ejano et al., 1985).
Over all, t her e ar e t wo met hods t o evaluat e food t ext ur e: sensor y and
inst r ument al. The sensor y met hod of developing a t ext ur e pr ofile
(Muoz et al. 1992) ut ilizes a human t ast e panel and pr ovides t he
ult imat e t est which, as discussed above, cannot be complet ely duplicat ed
by any inst r ument al pr ocedur e. Inst r ument al met hods, however , ar e
much less cost ly and t ime consuming t han sensor y t est s. Mor eover ,
t hey oft en cor r elat e t o cr it ical sensor y at t r ibut es which allow some
measur e of consumer accept abilit y. It is, however , r ar e for t hem t ost and
alone as a complet e t est . In any event , t hey can cer t ainly be ver y
valuable when used in conjunct ion wit h sensor y panels. Gener at ing
and int er pr et ing t ext ur e pr ofile infor mat ion, wit h inst r ument al or
sensor y means, is called Text ur e Pr ofile Analysis. Adouble compr ession
t est , t he most r ecognized inst r ument al means of char act er izing t he
t ext ur e of solid and semi-solid foods, is discussed below.
The idea of t ext ur e pr ofiling food was pr oposed in 1963 (Fr iedman
et al., 1963; Szczesnaik et al. 1963) and conduct ed using an inst r ument
known as t he Gener al Foods Text ur omet er . Bour ne (1968 and 1974)
adopt ed, and ext ended, t he t echnique t o t he Inst r on Univer sal Test ing
Machine wher e a food sample (bit e size pieces of food, usually a 1 cm
cube) is compr essed, t wo t imes, usually t o 80 per cent of it s or iginal
height . Compr ession is achieved using par allel plat es wher e one plat e
is fixed and t he ot her plat e moves wit h a r ecipr ocat ing linear cyclical
mot ion. Since t his t est is int ended t o r eflect t he human per cept ion of
t ext ur e, t he fir st and second compr ession cycles ar e r efer r ed t o as t he
fir st bit e and second bit e, r espect ively. A gener alized t ext ur e pr ofile
cur ve is illust r at ed in Fig. 1.39. Var ious t ext ur al par amet er s may be
det er mined fr om t he cur ve (Bour ne, 1978; Szczesnaik et al. 1963):
74 Chapter 1. Introduction to Rheology
Figure 1.39. Generalized texture profile curve obtained from the Instron Univer-
sal Testing Machine (from Bourne et al., 1978).
Figure 1.40. Texture profile curves (1 = first bite; 2 = second bite) for four food
products (from Bourne, 1978).
0
F
o
r
c
e
Time
Area 1
Area 3
Fracturability
Hardness 1
Stringiness
First Bite Second Bite
downstroke upstroke
Area 2
Hardness 2
Springiness
downstroke upstroke
0
F
o
r
c
e
Time
APPLE
0
F
o
r
c
e
Time
FRANKFURTER
0
F
o
r
c
e
Time
PRETZEL STICK
0
F
o
r
c
e
Time
CREAM CHEESE
1
1
1
1
2
2
2
2
1.13 Empirical Measurement Methods for Foods 75
Fr act ur abilit y: for ce at t he fir st major dr op in for ce cur ve. Popular
t er ms descr ibing fr act ur abilit y ar e cr umbly, cr unchy, and br it t le.
Har dness 1: for ce at maximum compr ession dur ing fir st bit e.
Popular t er ms descr ibing har dness ar e soft , fir m, and har d.
Ar ea 1 (ar ea under t he solid line up t o t he dashed line in t he fir st
compr ession cycle): wor k done on t he sample dur ing t he fir st bit e.
Adhesiveness: Ar ea 3 (ar ea under t he zer o for ce line) r epr esent ing
t he wor k, caused fr om a t ensile for ce, needed t o pull food apar t and
separ at e it fr om t he compr ession plat es. A similar adhesiveness
char act er ist ic is pr esent in t he second compr ession cycle. Popular
t er ms descr ibing adhesiveness ar e st icky, t acky, and gooey.
Adhesive For ce: maximum negat ive for ce.
St r inginess: dist ance food ext ends befor e it br eaks away fr om t he
compr ession plat es.
Har dness 2: for ce at maximum compr ession dur ing second bit e.
Ar ea 2 (ar ea under t he solid line up t o t he dashed line in t he second
compr ession cycle): wor k done on t he sample dur ing t he second bit e.
Spr inginess: dist ance or lengt h of compr ession cycle dur ing t he
second bit e. Popular t er ms descr ibing spr inginess ar e plast ic and
elast ic.
Cohesiveness: t he r at io of Ar ea 2 divided by Ar ea 1.
Gumminess: t he pr oduct of Har dness (fir st peak called Har dness
1 in Fig. 1.39) t imes Cohesiveness. Popular t er ms descr ibing
gumminess ar e shor t , mealy, past y, and gummy.
Chewiness: t he pr oduct of Gumminess t imes Cohesiveness t imes
Spr inginess which is equivalent t o Gumminess t imes Spr inginess.
Popular t er ms descr ibing chewiness ar e t ender , chewey, and t ough.
Though Chewiness and Gumminess ar e similar , t hey ar e mut ually
exclusive. The same pr oduct cannot exhibit bot h Chewiness and
Gumminess: Chewiness r efer s t o solid foods and Gumminess r efer s
t o semi-solid foods (Szczesnaik, 1995).
Typical t ext ur e pr ofile cur ves for apple t issue, fr ankfur t er , cr eam
cheese, a pr et zel st ick ar e shown in Fig. 1.40. Obser ving t hese cur ves
makes it clear t hat many foods do not exhibit all t he t ext ur al par amet er s
defined above. Text ur e Pr ofile Analysis has pr oved t o be a ver y useful
t echnique for examining food pr oduct s which fr act ur e; however , t ext ur e
76 Chapter 1. Introduction to Rheology
pr ofile cur ves ar e subject t o many var iat ions making t he pr oper int er -
pr et at ion of exper iment al dat a challenging. Mor e fundament al analyses
can be at t empt ed using t he science of fr act ur e mechanics.
Figure 1.41. Simple melt flow indexer for molten polymers.
Melt Flow Indexer for Polymers. The melt flow indexer is a weight
dr iven capillar y inst r ument used as a qualit y cont r ol device t o char ac-
t er ize t he behavior of molt en polymer s. Alt hough t he device is not
employed for food, use of t he inst r ument in t he t her moplast ics indust r y
is so widespr ead t hat anyone int er est ed in fluid r heology should have
a gener al awar eness of how it oper at es.
A simple melt flow indexer (Fig. 1.41) is cont ained wit hin a t her mal
jacket which cont r ols sample t emper at ur e. A known mass is placed on
t he plunger which dr ives a pist on downwar d causing t he sample t o be
ext r uded t hr ough t he die. Exact dimensions and oper at ing pr ocedur es
mass
piston
sample
die
jacket
extrudate
1.14.1 Carrageenan Gum Solution 77
ar e descr ibed by Amer ican Societ y for Test ing and Mat er ials (ASTM)
st andar ds. Typical dimensions involve a die diamet er of 2.1 mm, and
a r at io of die lengt h t o die diamet er equal t o 4. In a common t est ing
pr ocedur e known as "condit ion E," a sample (such as polyet hylene) is
allowed t o come t o an equilibr ium t emper at ur e of 190 C. Then, a 2.16
kg mass is placed on t he pist on and t he molt en polymer is ext r uded fr om
t he die. The weight , in gr ams, of t he ext r udat e pr oduced in 10 minut es
is t he melt flow index, or simply t he melt index, of t he polymer . Values
of t he melt flow index r eflect t he viscosit y of t he mat er ial: Low viscosit y
mat er ials, cor r esponding t o a low molecular weight , have a high melt
flow index. This infor mat ion is ver y valuable in t he t her moplast ics
indust r y because it cor r elat es well t o t he qualit y and pr ocessing char -
act er ist ics of numer ous polymer s.
1.14. Example Problems
1.14.1. Carrageenan Gum Soluti on
Rheological dat a for a 1% aqueous solut ion of car r ageenan gum at 25 C
ar e available (Table 1.6). Det er mine t he power law par amet er s and plot
t he appar ent viscosit y cur ve.
Table 1.6. Steady Shear, Rheological Data for a 1% Aqueous Solution of Carra-
geenan Gum at 25 C (Data from Prentice and Huber, 1983)
(s
-1
) (Pa ) (s
-1
) (Pa )
9.88 2.61 58.8 8.20
11.4 2.97 75.4 9.08
12.9 2.81 104.1 11.63
14.1 3.44 110.4 10.65
17.6 3.80 120.5 12.75
26.3 4.85 136.5 13.10
42.0 6.61 145.8 14.90
48.6 6.19 187.1 15.85
49.3 5.89 210.2 12.70
55.5 7.22 270.0 20.50
The dat a wer e plot t ed (Fig. 1.42) and an excellent fit was obt ained
wit h linear r egr ession analysis using t he power law model: = 0.66 Pa
s
n
and = 0.60. Ot her r heological models could be used t o fit t he
exper iment al dat a. Some (like t he Newt onian model) would r esult in




K
n
78 Chapter 1. Introduction to Rheology
gr eat er st at ist ical var iat ion while ot her s (like t he Her schel-Bulkley
model) would impr ove t he st at ist ical accur acy wit h which t he equat ion
r epr esent s t he dat a.
Figure 1.42. Rheogram for a carrageenan gum solution at 25 C.
Figure 1.43. Apparent viscosity of a carrageenan gum solution at 25 C.
Carrageenan Gum
1% Aqueous Solution
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Shear Rate, 1/s
K = 0.66 Pa s
n
n = 0.60
20
15
10
5
0
0 50 100 150 200

0 50 100 150 200


0
0.05
0.1
0.15
0.2
0.25
0.3
Shear Rate, 1/s
A
p
p
a
r
e
n
t

V
i
s
c
o
s
i
t
y
,

P
a

s
Carrageenan Gum
1% Aqueous Solution
K = 0.66 Pa s
n
n = 0.60

1.14.2 Concentrated Corn Starch Solution 79


Appar ent viscosit y is calculat ed as and plot t ed as a funct ion of
shear r at e (Fig. 1.43). Mat hemat ically, t he cur ve can be descr ibed in
t er ms of t he power law fluid model using and values det er mined
fr om t he shear st r ess ver sus shear r at e dat a:
The dat a pr esent ed in Table 1.6 have also been evaluat ed in t er ms of
t he Casson, Bingham, and Her schel-Bulkley models in Appendix 6.4
wher e t he met hod of linear r egr ession analysis is explained.
1.14.2. Concentrated Corn Starch Soluti on
Examine t he r heological infor mat ion pr esent ed (Fig. 1.44) for a con-
cent r at ed (53% wt /wt ) solut ion of r aw cor n st ar ch and wat er . St eady
shear dat a wer e collect ed using a convent ional cone (4 degr ee, 60 mm
diamet er ) and plat e appar at us.
Figure 1.44. Rheogram of a 53% (wt/wt) solution of raw corn starch and water
at 25 C.
/

K n
K(

)
n 1
.66(

)
.4
0 2 4 6 8 10
0
2
4
6
8
10
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s

(
P
a
)

o
r

A
p
p
a
r
e
n
t

V
i
s
c
o
s
i
t
y

(
P
)
shear stress apparent viscosity
53% Corn Starch and Water

80 Chapter 1. Introduction to Rheology


The cor n st ar ch solut ion is a concent r at ed suspension of st ar ch
par t icles in wat er . Examinat ion of t he appar ent viscosit y cur ve (Fig.
1.44) shows init ial shear -t hinning followed by st r ong shear -t hickening
behavior . At low shear r at es ( s
-1
), t he wat er has a lubr icat ing
effect bet ween t he par t icles and flow is r elat ively unhinder ed. Wit h
higher shear r at es ( s
-1
), incr eased r esist ance fr om par t icle t o
par t icle int er act ion causes a significant incr ease in appar ent viscosit y.
Cur ve fit t ing t he dat a, above 4.5 s
-1
, t o t he power law model yields:
wher e = 0.131 Pa s and = 1.72. A flow behavior index equal t o 1.72,
a value significant ly gr eat er t han 1.0, is a numer ical indicat ion of a lar ge
shear -t hickening effect . The above equat ion is illust r at ed as t he line
plot t ed in Fig. 1.44.
Table1.7. Rheological Datafor SwedishCommercial Milk Chocolateat 40 C(Data
from Prentice and Huber, 1983)
(s
-1
) (Pa ) (s
-1
) (Pa )
0.099 28.6 6.4 123.8
0.140 35.7 7.9 133.3
0.199 42.8 11.5 164.2
0.390 52.4 13.1 178.5
0.790 61.9 15.9 201.1
1.60 71.4 17.9 221.3
2.40 80.9 19.9 235.6
3.90 100.0
A concent r at ed cor n st ar ch solut ion can pr ovide an excellent visual
and t act ile example of shear -t hickening behavior . Fir st , a 50 t o 55%
(wt /wt ) solut ion consist ing of r aw cor n st ar ch (available in most gr ocer y
st or es) and wat er is r equir ed. Next , one must be willing t o examine t he
mat er ial wit h bar e hands. If you slowly move a finger t hr ough t he
mat er ial, t he solut ion feels and appear s like a liquid. If t he finger is
moved quickly, however , t his subst ance pr ovides much gr eat er
r esist ance and shows solid-like behavior by fr act ur ing and separ at ing
at t he higher shear r at e! The mat er ial quickly r ever t s t o a liquid-like
appear ance at t he cessat ion of movement .
0 <

< 4.5

4.5
K(

)
n
.131(

)
1.72
K n




1.14.3 Milk Chocolate 81
1.14.3. Mi lk Chocolate
Rheological dat a for milk chocolat e at 40 C ar e available (Table 1.7).
Det er mine t he Casson and Bingham plast ic model par amet er s for t his
mat er ial.
Figure 1.45. Regression analyses of 40 C milk chocolate applying the Bingham
plastic model over different shear rate ranges.
Dat a ar e pr esent for low shear st r esses and a plot (Fig. 1.45) of t his
infor mat ion suggest s t he pr esence of a yield st r ess. Casson and Bing-
ham plast ic model par amet er s wer e calculat ed over t hr ee differ ent shear
r at e r anges (Table 1.8). The r esult s clear ly indicat e t hat t he model and
shear r at e r ange cover ed in t he analysis have a st r ong influence on t he
yield st r ess (t he dynamic yield st r ess) calculat ed fr om t he int er cept of
t he r egr ession cur ve. When dat a at t he lower shear r at es ar e empha-
sized, t he calculat ed yield st r ess decr eases. This r esult s in Bingham
yield values r anging fr om 35.1 t o 52.3 Pa.

Shear Rate, 1/s


S
h
e
a
r

S
t
e
s
s
,

P
a
0 - 20 1/s

0 - 1.6 1/s
1.6 - 20 1/s
Data Range
Milk Chocolate
250
200
150
100
50
0
0 5 10 15 20 25

82 Chapter 1. Introduction to Rheology


Table 1.8. Constants for Bingham Plastic ( ) and Casson
( ) Models Used to Describe the Behavior of 40 C Milk
Chocolate Over Three Shear Rate Ranges
Ca s s on Model Bin gh a m Pla s t ic
r
2
r
2
(s
-1
) (Pa
.5
s
.5
) (Pa ) (Pa s ) (Pa )
0 - 20 2.21 29.7 .99 8.82 62.3 .99
1.6 - 20 2.14 32.3 .99 9.80 48.7 .98
0 - 1.6 3.04 17.8 .89 25.5 35.1 .84
1.14.4. Falli ng Ball Vi scometer for Honey
The effect of t emper at ur e on t he viscosit y of a Newt onian fluid can be
illust r at ed by consider ing dat a fr om a falling ball viscomet er . Der ive
(Part a) t he falling ball viscomet er equat ions and use t hem t o evaluat e
(Part b) t he viscosit y of honey fr om t he dat a given in Table 1.9. Also,
evaluat e t he influence of t emper at ur e on honey viscosit y using t he
Ar r henius equat ion.
Part a. Consider a spher e (r adius= ; densit y= ) dr opping t hr ough a
Newt onian fluid (densit y= ) ot her wise at r est . The spher e, t r aveling
downwar d at t er minal velocit y ( ), is subject t o t hr ee for ces: buoyancy
( ), dr ag ( ) and gr avit y ( ). A for ce balance
yields
[1.93]
wher e t he dr ag t er m comes fr om St okes law (Tr ans. Cambr idge Phil.
Soc. Vol. 8, 1845 and Vol. 9, 1851). Simplificat ion of Eq. [1.93] gives
[1.94]
The t ime ( ) r equir ed for t he ball t o fall a fixed dist ance ( ) is
[1.95]
so t he t er minal velocit y is

o
+
pl
(

)
n

0.5
(
o
)
0.5
+ K
1
(

)
0.5

K
1

o

pl

o
R
s

l
u
t
(4R
3

l
g)/3 (4R
3

s
g)/3 6Ru
t

4
3
R
3
_

,
g
l
+ 6Ru
t

4
3
R
3
_

,
g
s

s

l

2R
2
g
9u
t
t L
t
L
u
t
1.14.4 Falling Ball Viscometer for Honey 83
Table 1.9. Falling Ball Viscometer Data for Honey at Six Temperatures
Ba ll #1 Ba ll #2 Ba ll #3
ba ll dia met er ( ), m .00475 .00678 .00792
ba ll den s it y, kg/ m
3
7900 7900 7900
len gt h of fa ll, m 0.20 0.20 0.20
h on ey den s it y, kg/ m
3
1400 1400 1400
con t a in er dia met er ( ), m 0.036 0.036 0.036
= d/ D 0.132 0.188 0.220
Fa xen cor r ect ion fa ct or 0.727 0.617 0.559
a t 6.5 C, s 270.0 152.5 122.0
a t 12.5 C, s 108.3 58.5 46.8
a t 20.0 C, s 31.7 17.8 14.4
a t 21.5 C, s 25.7 14.4 11.4
a t 38.0 C, s 3.0 1.8 1.4
a t 48.0 C, s 1.5 1.0 0.9
a t 6.5 C, Pa s 107.79 124.04 135.41
a t 12.5 C, Pa s 43.22 47.58 51.94
a t 20.0 C, Pa s 12.67 14.51 15.98
a t 21.5 C, Pa s 10.26 11.67 12.65
a t 38.0 C, Pa s 1.20 1.46 1.55
a t 48.0 C, Pa s 0.60 0.81 0.99
a t 6.5 C, Pa s 78.38 76.59 75.68
a t 12.5 C, Pa s 31.43 29.38 29.03
a t 20.0 C, Pa s 9.21 8.96 8.93
a t 21.5 C, Pa s 7.46 7.20 7.07
a t 38.0 C, Pa s 0.87 0.90 0.87
a t 48.0 C, Pa s 0.44 0.50 0.56
a t 6.5 C, s
-1
0.94 1.16 1.24
a t 12.5 C, s
-1
2.3 3.0 3.2
a t 20.0 C, s
-1
7.9 9.9 10.5
a t 21.5 C, s
-1
9.8 12.3 13.3
a t 38.0 C, s
-1
84.2 98.3 108.2
a t 48.0 C, s
-1
168.9 176.9 168.4
[1.96]
Subst it ut ing Eq. [1.96] int o Eq. [1.94] yields
d
D

max

u
t

L
t
84 Chapter 1. Introduction to Rheology
[1.97]
so
[1.98]
wher e is t he falling ball viscomet er const ant defined as
[1.99]
Our focus in t his sect ion is on t he falling ball viscomet er . Eq. [1.94],
however , can also be used as a basis for calculat ing set t ling velocit ies
of par t icles including pigment s in paint , and chocolat e in milk.
Using Eq. [1.98] and [1.99] t he viscosit y of a Newt onian fluid can be
calculat ed dir ect ly. It is also possible t o calculat e t he viscosit y using a
r efer ence liquid, a Newt onian liquid of known viscosit y, t o eliminat e
er r or s caused fr om inaccur at e det er minat ion of t he ball r adius or dr op
lengt h. is equal for t he unknown and r efer ence liquids, t her efor e,
[1.100]
wher e t he subscr ipt s 1 and 2 r efer t o t he t est and r efer ence liquid,
r espect ively. The maximum shear r at e, locat ed at t he equat or of t he
spher e, is (Sher man, 1970):
[1.101]
Eq. [1.100] is accept able for and cases wher e t he
spher e diamet er is less t han 1/10 t he vessel diamet er . If necessar y, one
may mat hemat ically account for t he pr esence of t he wall when .
Faxen (Ar kiv. Mat . Ast r on. Fyzik. 17(27)1: 1922-1923) showed t hat t he
viscosit y calculat ed using St okes law could be cor r ect ed for wall effect s
using t he following equat ion:
[1.102]

s

l

2R
2
gt
9L

s

l
kt
k
k
2R
2
g
9L
k

1
(
s

1
)t
1


2
(
s

2
)t
2

max

3u
t
R
N
Re
u
t
d/ < 1.0
d/D >0.1

c
[1 2.104() + 2.09()
3
.95()
5
]
1.14.4 Falling Ball Viscometer for Honey 85
wher e is equal t o t he spher e diamet er ( ) divided by t he cont ainer
diamet er ( ) and is t he cor r ect ed viscosit y. The t er m in br acket s, t he
Faxen cor r ect ion fact or , is applicable up t o = 0.32. An alt er nat ive
expr ession (Chhabr a, 1992) can be used when :
[1.103]
Part b. Raw dat a for a t est involving t hr ee spher es, one cylindr ical
cont ainer and honey at six differ ent t emper at ur es ar e pr esent ed in Table
1.9. The Faxen cor r ect ion fact or , as well as uncor r ect ed (fr om Eq. [1.97])
and cor r ect ed (fr om Eq. [1.102]) viscosit ies wer e calculat ed fr om t he
exper iment al dat a (Table 1.9).
Cor r ect ed honey viscosit ies, t aken fr om Table 1.9, wer e aver aged at
each t emper at ur e (Table 1.10) and t he dat a fit (r
2
=0.99) t o t he Ar r henius
equat ion (Eq. [1.47]):
yielding
wher e: is t he t emper at ur e in degr ees Kelvin; = 1.987 cal/(g-mole K);
= 21,801 cal/ (g-mole); = 5.58 (10
-16
) Pa s. Result s may also be
expr essed in t er ms of a r efer ence viscosit y. Using a value of 3.77 Pa s,
calculat ed at 300 K wit h t he pr evious r elat ionship, yields
Table 1.10. Average of Corrected Viscosities for Honey
T T
( C) (K) (Pa s )
6.5 278.7 76.88
12.5 285.7 29.95
20.0 293.2 9.03
21.5 294.7 7.25
38.0 311.2 0.88
48.0 321.2 0.50
d
D
c

<0.97

1
1 .475
_

,
4
f (T) A exp

E
a
RT
_

,
5.58(10
16
) exp

21801
RT
_

,
T R
E
a
A
3.77 exp

21801
1.987

1
T

1
300
_

,
_

,
3.77 exp

11, 126

1
T

1
300
_

,
_

86 Chapter 1. Introduction to Rheology


The above equat ions pr ovide convenient expr essions for calculat ing t he
viscosit y of honey at any t emper at ur e fr om 6.5 C t o 48 C.
1.14.5. Orange Jui ce Concentrate
Rheological dat a for concent r at ed or ange juice, at four differ ent t em-
per at ur es, ar e given in Table 1.11. Descr ibe t his dat a using t he r heo-
logical model expr essed by Eq. [1.52]:
Also, make a mast er -cur ve of t he dat a using 9.5 C as t he r efer ence
t emper at ur e.
A r heogr am (Fig. 1.46), using t he power law model, is det er mined
at each t emper at ur e gener at ing a flow behavior index and a consist ency
coefficient for each cur ve (Table 1.12). Using t his dat a, t he aver age flow
behavior index may be calculat ed:
Table1.11. Rheological Datafor Concentrated OrangeJuice(65 Brix, 5.7%Pulp)
Made from Perna Oranges (Data from Vitali and Rao, 1984)
T = -18.8 C T = -5.4 C T = 9.5 C T = 29.2 C
(Pa ) (s
-1
) (Pa ) (s
-1
) (Pa ) (s
-1
) (Pa ) (s
-1
)
14.4 0.5 4.3 0.6 2.6 1.1 3.6 8
24.3 1 6.5 1 10.3 8 7.6 20
141.9 10 38.4 10 17.1 15 13.1 40
240.4 20 65.4 20 29.5 30 17.5 60
327.2 30 88.7 30 50.3 60 31.2 120
408.0 40 111.1 40 69.4 90 54.5 240
483.9 50 131.9 50 103.3 150 94.4 480
555.9 60 151.7 60 153.8 250 141.7 800
635.2 70 171.3 70 199.8 350 170.0 1000
692.5 80 189.4 80 242.8 450 183.2 1100

f (T,

) K
T
exp

E
a
RT
_

,
(

)
n

n
.764 + .772 + .762 + .797
4
.774

1.14.5 Orange Juice Concentrate 87


Figure 1.46. Rheograms of orange juice concentrate (Perna oranges: 65 Brix,
5.7% pulp) at four different temperatures.
Temper at ur e dependency of t he consist ency coefficient (dat a in
Table 1.12, plot t ed in Fig. 1.47) may be det er mined fr om r egr ession of
Table 1.12. Power Law Fluid Properties for Concentrated Orange Juice at Four
Temperatures
T T
( C) (K) (1/ K) (Pa s
n
) (-)
-18.8 254.4 0.003931 24.37 0.764
-5.4 267.8 0.003734 6.45 0.772
9.5 282.7 0.003538 2.25 0.762
29.2 302.4 0.003307 0.69 0.797
0.3 1 3 10 30 100 300 1,000
1
2
5
10
20
50
100
200
500
1,000
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Orange Juice Concentrate
-18.8 C -5.4 C 9.5 C 29.3 C

1/T K n

88 Chapter 1. Introduction to Rheology


Figure 1.47. Variation of the consistency coefficient of concentrated orange juice
with temperature.
[1.104]
showing t hat Pa and K. Hence, t he
complet e model, giving shear st r ess as a funct ion of t emper at ur e and
shear r at e, may be expr essed as
or , in t er ms of appar ent viscosit y, as
Since t he shear st r ess and appar ent viscosit y equat ions may be used
over t he ent ir e t emper at ur e and shear r at e r ange, t hey could be ver y
useful in solving many food pr ocess engineer ing design pr oblems r elat ed
t o t he pr oduct ion of or ange juice concent r at e. A compar ison (Fig. 1.48)
of t he full pr edict ion equat ion wit h t he or iginal dat a indicat e r easonably
0.0032 0.0033 0.0034 0.0035 0.0036 0.0037 0.0038 0.0039 0.004
0.5
1
2
5
10
20
50
K
,

P
a

s
n
1/T, K
-1
Effect of Temperature on K
K K
T
exp

E
a
RT
_

,
K
T
4.65(10
9
) s
n
E
a
/R 5668.25
4.646(10
9
) exp

5668.3
T
_

,
(

)
.774
4.646(10
9
) exp

5668.3
T
_

,
(

)
.774 1.0
1.14.5 Orange Juice Concentrate 89
good r esult s when compar ed t o t he case wher e separ at e r heogr ams ar e
gener at ed for each dat a set (Fig. 1.46). This level of accur acy is
accept able in solving most food engineer ing pr oblems.
Figure 1.48. Comparison of the raw data and prediction equation for the full
shear rate-temperature model of concentrated orange juice.
A r efer ence t emper at ur e of 9.5 C will be used in developing a
mast er -cur ve of t he exper iment al dat a. Not e t hat any value over t he
r ange of t emper at ur es consider ed could be select ed as t he r efer ence
t emper at ur e. Developing a mast er cur ve r equir es a hor izont al shift ing
of t he dat a at -18.8 C, -5.4 C, and 29.3 C t o t he 9.5 C cur ve (Fig. 1.46).
A dimensionless shift fact or ( ) is numer ically found t o account for t he
movement of each cur ve. Shear st r ess ver sus shear r at e divided by t he
shift fact or ( ) ar e plot t ed t o pr oduce t he mast er -cur ve.
In t his example, a shear st r ess of 100 Pa will be used as t he basis
for det er mining . Shear r at es at Pa ar e calculat ed using t he
const ant s pr ovided in Table 1.12. At -18.8 C, for inst ance,
0.1 0.3 1 3 10 30 100 300 1,000
1
2
5
10
20
50
100
200
500
1,000
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
-18.8 C -5.4 C 9.5 C 29.3 C


a
T

/a
T
a
T
100

90 Chapter 1. Introduction to Rheology


Table 1.13. Shear Rates (at =100 Pa) and Shift Factors
T a t 100 Pa
( C) (1/ s ) ( - )
-18.8 6.35 6.35 / 145.4 = 0.0437
-5.4 34.83 34.83 / 145.4 = 0.2395
9.5 145.4 145.4 / 145.4 = 1.000
29.3 514.0 514 / 145.4 = 3.536
and at 9.5 C, t he shear r at e is calculat ed as
Wit h t his infor mat ion, t he shift fact or at -18.8 C is found using t he shear
r at e at t he r efer ence t emper at ur e of 9.5 C:
Calculat ed shear r at es and shift fact or s for each cur ve ar e summar ized
in Table 1.13.
Table 1.14. Shifted Rheological Data for Concentrated Orange Juice (65 Brix,
5.7% Pulp) Used to Produce the Master-Curve
T=-18.8 C T=-5.4 C T=9.5 C T=29.2 C
(Pa ) (s
-1
) (Pa ) (s
-1
) (Pa ) (s
-1
) (Pa ) (s
-1
)
14.4 11.4 4.3 2.5 2.6 1.1 3.6 2.3
24.3 22.8 6.5 4.2 10.3 8.0 7.6 5.7
141.9 228.8 38.4 41.8 17.1 15.0 13.1 11.3
240.4 457.7 65.4 83.5 29.5 30.0 17.5 17.0
327.2 686.5 88.7 125.3 50.3 60.0 31.2 33.9
408.0 915.3 111.1 167.0 69.4 90.0 54.5 67.9
483.9 1144.2 131.9 208.8 103.3 150.0 94.4 135.7
555.9 1373.0 151.7 250.5 153.8 250.0 141.7 226.2
635.2 1601.8 171.3 292.3 199.8 350.0 170.0 282.8
692.5 1830.7 189.4 334.0 242.8 450.0 183.2 311.1

a
T

K
_

,
1/.764

100
24.37
_

,
1/.764
6.35 s
1

K
_

,
1/.762

100
2.25
_

,
1/.762
145.4 s
1

a
T
6.35/145.4 0.0437


/a
T


/a
T


/a
T


/a
T
1.14.6 Influence of the Yield Stress in Coating Food 91
Figure 1.49. Plot of shear stress versus providing a master-curve of con-
centrated orange juice having a reference temperature of 9.5 C.
Raw dat a, shift ed using appr opr iat e values, ar e given in Table
1.14 and plot t ed t o pr oduce a mast er -cur ve (Fig. 1.49). Hor izont al
shift ing causes t he dat a t o over lap on a single line. Addit ional analysis
could be per for med t o det er mine as a funct ion of t emper at ur e by
plot t ing t he infor mat ion pr ovided in Table 1.13. Mast er -cur ves can be
ver y useful in compar ing dat a fr om differ ent pr oduct s such as concen-
t r at ed or ange juice made fr om differ ent var iet ies of or anges.
1.14.6. Influence of the Yi eld Stress i n Coati ng Food
Consider t he r ole of t he fluid yield st r ess ( ) in coat ing food by examining
pot ent ial flow and coat ing t hickness on an inclined plane.
Assuming t he shear st r ess of t he air on t he fr ee sur face is negligible,
a for ce balance on a fluid on an inclined plane (Fig. 1.50) yields an
equat ion giving shear st r ess as a funct ion of (Chur chill, 1988):
1 3 10 30 100 300 1,000 3,000
1
2
5
10
20
50
100
200
500
1,000
Shear Rate / a , 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Master-Curve
T
-18.8 C -5.4 C 9.5 C 29.2 C

/ a
T

a
T
a
T

o
y
92 Chapter 1. Introduction to Rheology
Figure 1.50. Coating material on an inclined plane.
[1.105]
wher e is t he dist ance per pendicular t o t he plane, is t he height or
t hickness of t he coat ing fluid layer , is t he densit y of t he fluid, is t he
angle of inclinat ion, and is t he acceler at ion due t o gr avit y. This
equat ion shows t hat t he shear st r ess is a maximum on t he sur face of
t he wall (t he inclined plane) and decr eases in moving t owar d t he fr ee
sur face of t he coat ing; hence, t he maximum shear st r ess may be
expr essed as
[1.106]
When t he maximum shear st r ess exceeds t he yield st r ess ( ),
gr avit y alone will cause flow down t he wall pr oducing a gr oss sagging
phenomenon known as slumping in t he paint indust r y (Pat t on, 1964).
Similar pr oblems can be obser ved in food coat ings such as milk
chocolat e, vanilla fr ost ing, and bar becue sauce.
If , t he fluid will r emain on t he sur face of t he object being
coat ed. The maximum coat ing t hickness possible, wit hout mat er ial
flowing off t he sur face, may be calculat ed in t er ms of t he yield st r ess
and t he angle of inclinat ion:
[1.107]
h
y
f ( y) g(h y) sin
y h

g

max
f (0) gh sin

max
>
o

o

max
h
max


o
gsin
1.14.6 Influence of the Yield Stress in Coating Food 93
The yield st r ess concept is a ver y useful t ool for examining t he t hickness
of food coat ings. Equat ions given above also apply t o a ver t ical wall
wher e r esult ing in
[1.108]
sinsin90 1
h
max


o
g
Chapt e r 2 . Tube Vis c ome t ry
2.1. Introducti on
Tube viscomet er s ar e ver y useful in collect ing r heological dat a.
These inst r ument s may be placed int o t hr ee basic cat egor ies: glass
capillar ies (Fig. 2.1), oft en called U-t ube viscomet er s because of t heir
r esemblance t o t he let t er U; high pr essur e capillar ies (Fig. 2.2); and pipe
viscomet er s (Fig. 2.3). All est ablish a pr essur e differ ence t o cr eat e flow.
The major differ ence bet ween a capillar y and a pipe viscomet er is t he
diamet er of t he t ube. Alt hough t her e is no clear ly defined size at which
a t ube should be called a capillar y or a pipe, some guidelines can be
offer ed.
Figure 2.1. Ostwald and Canon-Fenske glass capillary (U-tube) viscometers.
Diamet er s in commer cial capillar y inst r ument s t ypically r ange fr om
0.1 t o 4 mm wit h a var iat ion in ent r ance angles (Fig. 2.2) of 15 t o 90
degr ees. Pipe viscomet er s ar e usually built "on-sit e" so size var ies
widely. Some may be as small as 7mm in diamet er but values gr eat er
V
L
1
2
Etched Lines
Capillary
Efflux Vessel
Receiving Vessel
Cannon-Fenske Ostwald
2.1 Introduction 95
t han 12 mm (t ypically 12 t o 32mm) ar e not uncommon in food appli-
cat ions. values in t ube viscomet er s r ange fr om 2 t o 400: The smaller
values ar e found in t he capillar y unit s but ar e seldom seen in pipe
syst ems.
Figure 2.2. High pressure capillary viscometer.
In t ypical oper at ion, t he U-t ube viscomet er is filled by inver t ing it
int o a sample and sucking (int o t he side wit h t he capillar y) fluid int o
t he fixed sample bulb. The viscomet er is t ur ned upr ight , t hen placed
in a t emper at ur e cont r olled bat h and allowed t o r each t her mal equi-
libr ium. Aft er a cer t ain per iod of t ime, usually 5 t o 10 minut es, t he fluid
is allowed t o flow down t hr ough t he capillar y. A st opwat ch is st ar t ed
when fluid passes t he upper et ched line and st opped when t he fluid
sur face passes t he lower et ched line. The r esult ing t ime is consider ed
t he efflux t ime for fluid dischar ge fr om t he bulb and fluid viscosit y is
calculat ed fr om t his value.
U-t ube viscomet er s ar e designed as gr avit y oper at ed inst r ument s.
High pr essur e capillar ies, which may also be const r uct ed fr om glass
t ubes but ar e not "U-shaped," ar e t ypically pist on dr iven (Fig. 2.2) or
gas oper at ed. A pump or gas (Fig. 2.3) syst em can be used t o cr eat e a
L/D
L
R
Piston
Entrance Angle
96 Chapter 2. Tube Viscometry
dr iving for ce in pipe viscomet er s. These unit s can be oper at ed at ele-
vat ed pr essur es such as t hose found in asept ic food pr ocessing equip-
ment (Dail and St effe, 1990a and 1990b).
Figure 2.3. Gas driven pipe viscometer.
Raw dat a for t ube viscomet er s ar e pr essur e dr op and volumet r ic flow
r at e. The pr essur e dr op is det er mined fr om pr essur e t r ansducer s or , in
t he case of U-t ube viscomet er s, t he height of fluid above a r efer ence
point . In high pr essur e capillar ies, flow r at es ar e calculat ed fr om t he
assumpt ion t hat volumet r ic flow in t he pist on (or bar r el) and t he cap-
illar y ar e equivalent . Volumet r ic flow r at e may be det er mined fr om t he
mass flow r at e measur ed in pipe syst ems using a mass flow met er or a
weight t ank. Densit y is r equir ed for t his calculat ion because t he volu-
met r ic flow r at e equals t he mass flow r at e divided by t he densit y. The
quant it y of flow in a glass capillar y is fixed by t he volume ( ) of t he
efflux vessel (Fig. 2.1).
The main focus of t he cur r ent chapt er is t ime-independent fluids.
Rot at ional inst r ument s ar e super ior for t he invest igat ion of t ime-
dependent mat er ials because t he sample can be maint ained in t he t est
chamber dur ing per iodic evaluat ion. Lit t le at t ent ion is given t o
evaluat ing t he elast ic behavior of fluids fr om capillar y dat a. Theor et -
ically, it is possible t o det er mine nor mal st r ess differ ences wit h t ube (or
Air Supply
Pressure Regulator
L X
E
R
Pressure Transducers
Mixing Paddle
Temperature
Sensors
V
2.2 Rabinowitsch-Mooney Equation 97
slit ) viscomet er s fr om dat a collect ed using var ious met hods such as axial
t hr ust , r esidual exit pr essur e, or jet expansion (Walt er s, 1975; Whor low,
1992). Gener ally, t hese applicat ions have focused on polymer s wit h
minor consider at ion given t o food pr oduct s. In addit ion, commer cially
available t ube viscomet er s r equir e special modificat ions t o make t hese
measur ement s. Nor mal st r ess infor mat ion can be mor e easily calcu-
lat ed fr om axial for ce dat a gener at ed on cone and plat e viscomet er s.
Tubular syst ems involving back ext r usion (alsocalled annular pumping)
have alsobeen developed t oevaluat e power law fluids (Osor ioand St effe,
1986; St effe and Osor io, 1987).
2.2. Rabi nowi tsch-Mooney Equati on
Deri vati on of the Basi c Equati on. Numer ous assumpt ions ar e
r equir ed in developing t he Rabinowit ch-Mooney equat ion: flow is lam-
inar and st eady, end effect s ar e negligible, fluid is incompr essible,
pr oper t ies ar e not a funct ion of pr essur e or t ime, t emper at ur e is
const ant , t her e is no slip at t he wall of t he t ube meaning t hat t he velocit y
of t he fluid is zer o at t he wall-fluid int er face, and r adial and t angent ial
velocit y component s ar e zer o.
The st ar t ing point in t he der ivat ion of a t ube viscomet er equat ion is
a for ce balance. Consider a fluid flowing t hr ough a hor izont al t ube of
lengt h ( ) and inside r adius ( ). Apr essur e dr op ( ) over a fixed lengt h
( ) is causing flow. The for ce balance, equat ing t he shear st r ess causing
flow t o t he shear st r ess r esist ing flow (i.e., t he fluid), over a cor e of fluid
(Fig. 2.4) wit h r adius and lengt h yields
[2.1]
which can be solved for t he shear st r ess:
[2.2]
Eq. [2.2] can also be obt ained by st ar t ing fr om t he gener al conser vat ion
of moment um equat ions (Bir d et al., 1960; Br odkey and Her shey, 1988;
Dar by, 1976; Denn, 1980) as discussed in Example Pr oblem 2.12.1. Eq.
[2.2] depict s t he shear st r ess var ying over t he pipe fr om zer o at t he
cent er ( ) t o a maximum at t he wall ( ) wher e t he equat ion may
be wr it t en as
L R P
L
r L
(P)r
2
2rL
f (r)
(P)r
2
2rL

(P)r
2L
r 0 r R
98 Chapter 2. Tube Viscometry
Figure 2.4. Core of fluid in tube flow geometry.
[2.3]
To develop t he shear r at e equat ions, a differ ent ial flow element ( )
must be evaluat ed. This can be expr essed by consider ing t he st eady,
laminar flow of fluid moving t hr ough an annulus locat ed bet ween t he
cor e, wit h r adius , and t he posit ion :
[2.4]
wher e is t he linear velocit y at . The t ot al volumet r ic flow r at e is found
by int egr at ing Eq. [2.4] over t he r adius:
[2.5]
Recognizing t hat , so t hat becomes t he var iable of int egr a-
t ion, allows Eq. [2.5] t o be wr it t en as
[2.6]
The r ight hand side of t his equat ion can be int egr at ed by par t s:
[2.7]
and simplified by applying t he no slip boundar y condit ion which
st ipulat es t hat t he fluid velocit y is zer o at t he wall of t he pipe or ,
mat hemat ically, :
R
r
L
Flow
Flow fluid core

w

(P)R
2L
dQ
r r + dr
dQ u2rdr
u r
Q

0
Q
dQ

r 0
r R
u2r dr
2rdr dr
2
r
2
Q

r
2
0
r
2
R
2
(u) dr
2
Q ur
2

r
2
0
r
2
R
2

r
2
0
r
2
R
2
r
2
du
u 0 at r R
2.2 Rabinowitsch-Mooney Equation 99
[2.8]
To fur t her evaluat e Eq. [2.8], a number of it ems must be not ed. Fir st ,
by assuming st eady laminar flow, we know t he shear r at e is some
funct ion of t he shear st r ess:
[2.9]
or
[2.10]
The negat ive sign is r equir ed in Eq. [2.9] because we assume, as
indicat ed in Eq. [2.1], t he posit ive dir ect ion of t o be opposit e t he
dir ect ion of flow. Second, Eq. [2.2] and [2.3] can be combined t o give
[2.11]
which, when differ ent iat ed, yields
[2.12]
Taking t he expr ession for given by Eq. [2.10], and inser t ing Eq. [2.12]
for , gives
[2.13]
Using Eq. [2.13] and not ing, fr om Eq. [2.11], t hat allows
Eq. [2.8] t o be r ewr it t en as
[2.14]
Obser ve t he change in t he limit s of int egr at ion: goes fr om 0 t o as
goes fr om 0 t o . Simplifying t his expr ession gives t he final gener al
equat ion r elat ing shear st r ess and shear r at e:
[2.15]
Eq. [2.15] may be evaluat ed by differ ent iat ion using Leibnit zr ule, which
allows an int egr al of t he for m
Q [(0)R
2
u(0)
2
]

r
2
0
r
2
R
2
r
2
du

r
2
0
r
2
R
2
r
2
du

du
dr
f ()
du f () dr

w
R
dr

w
_

,
d
du
dr
du f () dr f ()

w
_

,
d
r
2
()
2
R
2
/(
w
)
2
Q

()
2
(
w
)
2
_

,
R
2

f ()
R

w
_

,
d

w
R
2
r
Q
R
3

1
(
w
)
3

w
()
2
f () d
100 Chapter 2. Tube Viscometry
[2.16]
t o be wr it t en as
[2.17]
t o differ ent iat e t he int egr al. Wr it ing, Eq. [2.15] as
[2.18]
t hen, applying Leibnit z r ule on t he r ight hand side, and differ ent iat ing
bot h sides wit h r espect t o (which is ) gives
[2.19]
Solving Eq. [2.19] for t he shear r at e at t he wall ( ) yields t he well known
Rabinowit sch-Mooney equat ion:
[2.20]
wher e t he der ivat ive is evaluat ed at a par t icular value of . Applicat ion
of t his equat ion is demonst r at ed for soy dough in Example Pr oblem
2.12.2.
Eq. [2.20] can also be expr essed in t er ms of t he appar ent wall shear
r at e, :
[2.21]
wher e . Fur t her manipulat ion gives
[2.22]
or
[2.23]
t hat can be r ewr it t en in t he following simplified for mat :
[2.24]
wher e
d
dz

'

0
z
z
2
f (z) dz

(z)
2
f (z)

Q
R
3
_

,
(
w
)
3

w
()
2
f () d

w
z
(
w
)
3

d(Q/(R
3
))
d
w
1
1
]
+

Q
R
3
_

,
3(
w
)
2
(
w
)
2
f (
w
)

w
f (
w
)

3Q
R
3
_

,
+ (
w
)

d(Q/(R
3
))
d
w
1
1
]

w
f (
w
)

3
4
_

,
+

w
4
_

d
d
w
_

,
4Q/(R
3
)

w
f (
w
)

3
4
_

,
+

w
4
_

d
d
w
_

,
1
1
]

w
f (
w
)

3
4
_

,
+

1
4
_

d(ln )
d(ln
w
)
_

,
1
1
]

3n + 1
4n
1
1
]

2.2 Rabinowitsch-Mooney Equation 101


[2.25]
showing t hat is t he point slope of t he ver sus . If t he fluid
behaves as a power law mat er ial, t he slope of t he der ivat ive is a st r aight
line and . Eq. [2.24] is a convenient for m of t he Rabinowit sch-
Mooney equat ion because power law behavior is fr equent ly obser ved
wit h fluid foods. Also, slight cur vat ur e in t he logar it hmic plot can oft en
be ignor ed. Applicat ion of Eq. [2.24] and [2.25] is illust r at ed for a 1.5%
solut ion of sodium car boxymet hylcellulose in Example Pr oblem 2.12.3.
Newtoni an Flui ds. In developing t he Rabinowit sch-Mooney equat ion
a gener al expr ession r elat ing shear st r ess t o shear r at e, Eq. [2.15], was
developed:
This can be solved for a Newt onian fluid by inser t ing t he Newt onian
definit ion for shear r at e, :
[2.26]
Int egr at ion of Eq. [2.26] gives
[2.27]
Subst it ut ing t he shear st r ess at t he wall ( ) int o Eq. [2.27]
r esult s in t he Poiseuille-Hagen equat ion:
[2.28]
Eq. [2.28] indicat es t hat t he r adius has a ver y st r ong influence on
t he behavior of t he syst em since it is r aised t o t he power four . Also, if
t his equat ion is wr it t en in t er ms of t he definit ion of a Newt onian fluid
( ), t hen t he for mula for t he shear r at e at t he wall may be
det er mined:
[2.29]
n
d(ln
w
)
d(ln )
n ln(
w
) ln()
n n
Q
R
3

1
(
w
)
3

w
()
2
f () d

f () /
Q
R
3

1
(
w
)
3

w
()
2
(/) d
1
(
w
)
3

()
3

,
d
Q
R
3

1
(
w
)
3
_

()
4
4
_

,
1
1
10


w
4

w
(P)R/(2L)
Q
(P)R
4
8L

w

4Q
R
3
102 Chapter 2. Tube Viscometry
This expr ession is ident ical t o t he one given in Eq. [1.90] (and Fig. 1.27)
for est imat ing t he maximum shear r at e of a Newt onian fluid in t ube
flow.
Power Law Flui ds. Eq. [2.15] can be solved for a power law fluid by
inser t ing int o t he equat ion:
[2.30]
Int egr at ion and subst it ut ion of gives
[2.31]
If t his equat ion is wr it t en in t er ms of t he definit ion of a power law fluid
( ), t hen t he for mula for t he shear r at e at t he wall may be
det er mined:
[2.32]
Eq. [2.32] is an exact solut ion for a power law fluid and also useful as
an est imat e of t he maximum shear r at e in t ube flow (as indicat ed in Eq.
[1.91]) for a wide r ange of fluid foods.
Bi ngham Plasti c Flui ds. In a Bingham plast ic fluid, t he shear r at e
is defined in t er ms of t he plast ic viscosit y and t he yield st r ess:
. This funct ion is discont inuous because t her e is no
shear ing flow at point s in t he t ube near t he cent er wher e t he shear
st r ess is below t he yield st r ess. Mat hemat ically, for in
t he cent r al plug r egion and for in t he out er
por t ion of t he t ube. Given t he above, it is clear t hat Eq. [2.15] must be
int egr at ed for each r egion of t he t ube t o det er mine t he t ot al volumet r ic
flow r at e:
[2.33]
Since shear r at e is zer o when t he shear st r ess is below t he yield st r ess,
Eq. [2.33] can be simplified t o
[2.34]

f () (/K)
1/n
Q
R
3

1
(
w
)
3
1
1
]

w
()
2

K
_

,
1/n
d

1
(
w
)
3
K
1/n
1
1
]

w
()
2 + 1/n
d

w
(P)R/(2L)
Q

(P)
2LK
_

,
1/n

n
3n + 1
_

,
R
(3n + 1)/n

w
K(

w
)
n

3n + 1
4n
_

4Q
R
3
_

f () (
o
)/
pl
f () 0 0 < <
o
f () (
o
)/
pl

o
<
w
Q
R
3

1
(
w
)
3
1
1
]

o
()
2
f () d+

1
(
w
)
3
1
1
]

w
()
2

pl
_

,
d
Q
R
3

1
(
w
)
3
1
1
]

w
()
2

pl
_

,
d
2.3 Laminar Flow Velocity Profiles 103
which, when int egr at ed, yields t he Buckingham-Reiner equat ion for
flow of a Bingham plast ic mat er ial in a pipe:
[2.35]
The same t echnique discussed for Bingham plast ics is used t o der ive
t he flow r at e equat ion for Casson fluids in Example Pr oblem 2.12.4.
Herschel-Bulkley Flui ds. The volumet r ic flow r at e for a Her schel-
Bulkley fluid ( ), is found using t he same met hod discussed
for t he Bingham plast ic mat er ial:
[2.36]
2.3. Lami nar Flow Veloci ty Profi les
Rheological pr oper t ies have a st r ong influence on fluid velocit y
pr ofiles in t ube flow. Under st anding t hese pr ofiles is impor t ant in
developing a clear pict ur e of inst r ument per for mance and in making
var ious food pr ocess engineer ing design calculat ions such as det er -
mining t he appr opr iat e hold t ube lengt h for a t her mal pr ocessing sys-
t em. Since t ube viscomet er s oper at e in t he laminar flow r egime, only
laminar flow velocit y pr ofiles ar e pr esent ed her e. Equat ions descr ibing
t he t ur bulent velocit y of Newt onian and power law fluids ar e discussed
lat er in Sect ion 2.11.
Combining t he shear st r ess r elat ionship (Eq. [2.2]) and t he definit ion
of a Newt onian fluid (Eq. [1.25]) wit h t he shear r at e wr it t en as
yields
[2.37]
Int egr at ing t his equat ion, using t he no slip boundar y condit ion
( ) t o det er mine t he unknown const ant , gives an expr ession
for t he velocit y pr ofile of a Newt onian fluid:
[2.38]
If t he definit ion of a power law fluid is used, Eq. [2.37] becomes
[2.39]
Q

R
4
(P)
8
pl
L
_

4
3
_

w
_

,
+

1
3
_

w
_

,
4
1
1
]
K(

)
n
+
o
Q

R
3
256
_

4n
3n + 1
_

w
K
_

,
1/n

1

o

w
_

,
1/n

1
(
o
/
w
)
2n + 1

1 +
2n
n + 1

w
_

1 +
n
o

w
_

,
1
1
]
1
1
]
du/dr

du
dr
_

,

(P)r
2L
u 0 at r R
u f (r)
(P)
4L
(R
2
r
2
)
K

du
dr
_

,
n

(P)r
2L
104 Chapter 2. Tube Viscometry
The velocit y pr ofile is found aft er int egr at ion and applicat ion of t he no
slip boundar y condit ion:
[2.40]
By combining Eq. [2.40] wit h t he volumet r ic flow r at e equat ion (Eq.
[2.31]), t he r elat ionship bet ween t he velocit y at and t he volumet r ic
aver age velocit y ( ) may be calculat ed:
[2.41]
Lower values of t he flow behavior index r esult in a flat t er velocit y pr ofile
and higher values maximize t he differ ence bet ween t he slowest and
fast est fluid element s (Fig. 2.5). The ext r eme case occur s wit h at
wher e . The cur ve (Fig. 2.5) for a Newt onian fluid is
found at .
Figure 2.5. Laminar velocity profiles for power law fluids with different values
of the flow behavior index.
The maximum velocit y, locat ed at t he cent er line wher e , may
be det er mined fr om Eq. [2.41] for any value of t he flow behavior index:
u f (r)

(P)
2LK
_

,
1/n

n
n + 1
_

,
[R
(n + 1)/n
r
(n + 1)/n
]
r
u Q/(R
2
)
u/( u )

3n + 1
n + 1
_

r
R
_

,
(n + 1)/n
1
1
]
n
r/R 0 u/ u 3.0
n 1.0
0
0.5
1
1.5
2
2.5
1 0.5 0 0.5 1
n=2.5
n=2.0
n=1.0
n=0.7
n=0.4
n=0.1
r/R
Power Law Fluid
Velocity Profiles
u

/

u
r 0
2.3 Laminar Flow Velocity Profiles 105
[2.42]
Clear ly, t he flow behavior index has a ver y st r ong influence on t he
velocit y pr ofile. Velocit y pr ofiles for power law fluids in t ur bulent flow
ar e ver y flat by compar ison (see Sect ion 2.11, Table 2.7).
In t he case of a Bingham plast ic fluid, t he for ce balance equat ion is
[2.43]
which is valid in t he out er r egions of t he pipe wher e t he fluid is subject
t o shear . Wit h t he no slip boundar y condit ion, Eq. [2.43] may be
int egr at ed t o give an expr ession for t he velocit y pr ofile bet ween t he plug
and t he wall of t he t ube:
[2.44]
for wher e . The cr it ical r adius ( ), which defines t he
out er boundar y of t he plug, may be calculat ed fr om t he yield st r ess:
[2.45]
The maximum velocit y in t he pipe, which is t he velocit y of t he plug in
t he cent er sect ion of t he pipe, is found by subst it ut ing int o Eq.
[2.44]:
[2.46]
which is valid for wher e .
Combining Eq. [2.44] or Eq. [2.46] wit h t he volumet r ic flow r at e
equat ion (Eq. [2.35]), t he r elat ionship bet ween t he velocit y at and t he
volumet r ic aver age velocit y ( ) may be calculat ed as
[2.47]
in t he shear ed por t ion of t he fluid ( ), and
[2.48]
u
max
/( u )
3n + 1
n + 1

pl

du
dr
_

,
+
o

(P)r
2L
u f (r)
(P)R
2
4
pl
L

r
R
_

,
2

2R
o
R

1
r
R
_

,
1
1
]
R
o
r R
o
R
o
R
o


o
2L
(P)
r R
o
u
max

(P)R
2
4
pl
L

1
R
o
R
_

,
2
1
1
]
0 < r R
o

o
r
u Q/(R
2
)
u/( u )
2(1 2c + 2cr/R (r/R)
2
)
1 4c/3 + c
4
/3
1 r/R c
u/( u )
2(1 c)
2
1 4c/3 + c
4
/3
106 Chapter 2. Tube Viscometry
in t he plug ( ). The value of is defined in t er ms of t he yield st r ess
or t he cr it ical r adius: . Plot t ing t hese equat ions show
t hat incr easing t he yield st r ess enlar ges t he r adius of t he plug flowing
down t he cent er of t he pipe (Fig. 2.6). The cur ve for a Newt onian fluid
is illust r at ed by t he line wit h
Figure 2.6. Laminar velocity profiles for Bingham plastic fluids with different
values of .
The same appr oach descr ibed for Bingham plast ic fluids can be used
t o det er mine t he velocit y pr ofile for a Her schel-Bulkley fluid:
[2.49]
Velocit y of t he plug may be det er mined by using Eq. [2.49] wit h ,
calculat ed fr om Eq. [2.45].
r/R c c
c
0
/
w
R
o
/R
c 0.
0
0.5
1
1.5
2
1 0.5 0 0.5 1
r/R
u

/

u
c=0.0
c=0.2
c=0.4
c=0.6
c=0.8
Bingham Plastic Fluid
Velocity Profiles
c
o
/
w
u f (r)
2L
(P) (1 + 1/n)K
1/n

(
w

o
)
1 + 1/n

(P)r
2L

o
_

,
1 + 1/n
1
1
]
r R
o
2.4 Laminar Flow Criteria 107
2.4. Lami nar Flow Cri teri a
To obt ain meaningful r heological dat a, t ube viscomet er s must
oper at e in t he laminar flow r egime. Tur bulence is r ar ely involved in
high pr essur e capillar ies but may be a pr oblem in pipe viscomet er s wit h
lar ge diamet er t ubes or when invest igat ing low viscosit y fluids.
The accept ed cr it er ion for laminar flow of Newt onian fluids in t ube
flow is t o maint ain a Reynolds number less t han 2100: wher e
. Tr ansit ion away fr om laminar flow, however , has been
obser ved for values bet ween 1225 and 3000. In addit ion, laminar
flow, under ideal condit ions, has been obser ved for number s as high
as 50,000 (Gr ovier and Aziz, 1972). Similar pr oblems exist for non-
Newt onian fluids. Hence, t he following discussion can only offer gener al
guidelines t o det er mine if t he flow r egime is laminar or t ur bulent .
As not ed above, flow is consider ed laminar for Newt onian fluids
when t he Reynolds number is below t he cr it ical value of 2100:
[2.50]
Wit h power law fluids, laminar flow occur s when (Hanks, 1963):
[2.51]
wher e t he t er m on t he r ight side of t he inequalit y is t he cr it ical power
law Reynolds number . The power law Reynolds number is defined as
[2.52]
Plot t ing Eq. [2.51] shows t he cr it ical Reynolds number incr easing
shar ply t oa maximum near , t hen dr opping off wit h gr eat er values
of t he flow behavior index (Fig. 2.7). Also, for t he special case of a
Newt onian fluid ( ) it is equal t o t he expect ed value of 2100.
Eq. [2.51] may give a conser vat ive est imat e of t he cr it ical Reynolds
number . Wor k by Campos et al. (1994) shows t hat cr it ical Reynolds
number s in t he r ange of 4000 (at ) t o 3000 (at ) may be
det er mined using a st at ist ical evaluat ion pr ocedur e. Higher number s
ar e also pr edict ed, par t icular ly for , using t he equat ion der ived
by Mishr a and Tr ipat hi (1973):
N
Re
< 2100
N
Re
D u /
N
Re
N
Re
N
Re
< 2100 (N
Re
)
critical
N
Re, PL
<
6464n(2 + n)
(2 + n)/(1 + n)
(1 + 3n)
2
(N
Re, PL
)
critical
N
Re, PL

D
n
( u )
2 n

8
n 1
K
_

4n
3n + 1
_

,
n
n 0.4
n 1
n 0.46 n 0.78
0 < n < 0.6
108 Chapter 2. Tube Viscometry
[2.53]
Eq. [2.53] pr edict s gr adually decr easing values of cr it ical Reynolds
number s fr om 3480 at t o 2357 at . Over , t he
Mishr a and Tr ipat hi (1973) equat ion pr edict s values (Fig. 2.7) ver y
similar t o t hose given by Eq. [2.51].
Figure 2.7. Variation of the critical power law Reynolds number with .
Wit h t ube flow of a Bingham plast ic fluid, st eady-st at e laminar flow
can be expect ed (Hanks, 1963) when
[2.54]
The t er m on t he r ight side of Eq. [2.54] is t he cr it ical Bingham Reynolds
number and , t he cr it ical value of , is defined as
[2.55]
(N
Re, PL
)
critical

2100(4n + 2) (5n + 3)
3(1 + 3n)
2
n 0.1 n 0.60 0.65 < n < 1.0
0 0.2 0.4 0.6 0.8 1
1,000
1,500
2,000
2,500
3,000
3,500
4,000
Flow Behavior Index
C
r
i
t
i
c
a
l

P
o
w
e
r

L
a
w

R
e
y
n
o
l
d
s

N
u
m
b
e
r
Eq.[2.51] from Hanks (1963)
Eq.[2.53] from Mishra and Tripathi (1973)
n
N
Re, B

N
He
8c
c
(1 4c
c
/3 + c
c
4
/3) (N
Re, B
)
critical
c
c
c
c
c
(1 c
c
)
3

N
He
16, 800
2.4 Laminar Flow Criteria 109
The Hedst r om number and t he Bingham Reynolds number ar e defined
by t he following equat ions:
[2.56]
[2.57]
Cr it ical values of ver sus (Eq. [2.55]), and t he cr it ical Bingham
Reynolds number ver sus t he Hedst r om number (Eq. [2.54]) ar e plot t ed
in Fig. 2.8 and 2.9, r espect ively. These illust r at ions indicat e t he
difficult y in achieving t ur bulence wit h mat er ials having a significant
yield st r ess. Not e t hat t he Hedst r om number incr eases wit h lar ger
values of t he yield st r ess. Laminar flow cr it er ion is also available (but
not exper iment ally ver ified) for Her schel-Bulkley mat er ials (Gar cia and
St effe, 1987).
Figure 2.8. Variation of with the Hedstrom number.
N
He

D
2

(
pl
)
2
N
Re, B

Du

pl
c N
He
10 100 1,000 10,000 100,000 1,000,000 10,000,000
0
0.2
0.4
0.6
0.8
1
C
c
N
He
c
c
110 Chapter 2. Tube Viscometry
2.5. Data Correcti ons
Ther e ar e numer ous measur ement er r or s which may occur in using
t ube viscomet er s (Table 2.1). Some ar e gener ally applicable and ot her s
only apply t o specific syst ems. Dat a cor r ect ions r equir ed for high
pr essur e and pipe viscomet er s ar e similar , and const it ut e t he focus of
t his sect ion. Glass capillar ies (gr avit y oper at ed U-t ube viscomet er s)
have special r equir ement s which ar e discussed in Sec. 2.9.
Figure 2.9. Variation of the critical Bingham Reynolds number with the Hed-
strom number.
Ki neti c Energy Losses. Ther e is a pr essur e loss, due t o a differ ence
in kinet ic ener gy, in high pr essur e capillar ies caused by t he acceler at ion
of t he fluid fr om t he bar r el velocit y t o t he capillar y velocit y. This
pr essur e loss ( ) may be expr essed in t er ms of t he kinet ic ener gy
cor r ect ion fact or ( ):
[2.58]
100 300 1,000 3,000 10,000 30,000 100,000 300,000 1,000,000
0
5,000
10,000
15,000
20,000

N
(
N





)
He
R
e
,
B
c
r
i
t
i
c
a
l
P

(( u
2
)
2
( u
1
)
2
)
2.5 Data Corrections 111
wher e and ar e t he aver age capillar y and bar r el velocit ies,
r espect ively. Equat ions for calculat ing ar e available (see Table 2.4 in
Sec. 2.10). Losses due t o kinet ic ener gy ar e gener ally small and ver y
difficult t o separ at e fr om ent r ance pr essur e losses. Hence, it is accept ed
pr act ice t o assume t hat kinet ic ener gy losses ar e account ed for in t he
ent r ance effect cor r ect ion.
End Effects: Entrance Correcti on. Ener gy losses due t o fluid
diver gence at t he end of a capillar y ar e small and usually neglect ed but
ent r ance losses can be ver y significant and must be evaluat ed.
Table 2.1. Sources of Error in Operating Tube Viscometers (from Van Wazer et
al., 1963)
Fa ct or Ca u s e Applica bilit y
Kin et ic en er gy los s es Los s of effect ive pr es s u r e beca u s e Gen er a l
of t h e kin et ic en er gy in t h e is s u -
in g s t r ea m.
En d effect s En er gy los s es du e t o vis cou s or Gen er a l
ela s t ic beh a vior wh en a flu id con -
ver ges or diver ges a t t h e en ds of a
ca pilla r y
Ela s t ic en er gy En er gy los s by ela s t ic defor ma - Vis coela s t ic ma t er ia ls
t ion of t h e flu id n ot r ecover ed
du r in g flow in t h e ca pilla r y
Tu r bu len ce Depa r t u r e fr om la min a r flow Gen er a l
Pr es s u r e los s es pr ior St ickin g of t h e pis t on or en er gy Cylin der -pis t on
t o t h e ca pilla r y dis s ipa t ed in t h e flow of t h e ma t e- vis comet er s
r ia l wit h in t h e cylin der befor e
en t er in g t h e ca pilla r y.
Dr a in a ge Liqu id a dh er in g t o t h e wa ll of t h e Gla s s ca pilla r y
vis comet er r es er voir . vis comet er s
Su r fa ce-t en s ion Va r ia t ion s in s u r fa ce t en s ion fr om Gla s s ca pilla r y
effect s on e t es t s u bs t a n ce t o a n ot h er . vis comet er s
Hea t effect s Con ver s ion of pr es s u r e en er gy High -s h ea r vis comet er s
in t o h ea t en er gy t h r ou gh flow.
Wa ll effect s Su r fa ce ph en omen a a t t h e Polyph a s e flu ids
flu id-wa ll in t er fa ce.
Effect of t ime- Va r ia t ion s in t h e r es iden ce t ime Th ixot r opic a n d r h eo-
depen den t pr oper t ies in t h e ca pilla r y. pect ic ma t er ia ls
The ent r ance effect cor r ect ion account s for excess pr essur e loss
occur r ing at t he opening of t he t ube viscomet er fr om sudden changes in
geomet r y caused by a conver gence, and losses in kinet ic ener gy. This
pr oblem may be exper iment ally evaluat ed using a number of t ubes
u
2
u
1

112 Chapter 2. Tube Viscometry


Figure 2.10. Bagley plot illustrating entrance effect pressure corrections deter-
mined at three flow rates.
having differ ent r at ios (Bagley, 1957). Dat a of t ot al pr essur e dr op
ver sus flow r at e ar e collect ed for each t ube. Pr essur e dr op ver sus
dat a ar e plot t ed at each flow r at e (or appar ent wall shear r at e, )
and t he lines ar e ext r apolat ed t o equal t o zer o. These figur es (Fig.
2.10) ar e somet imes called "Bagley plot s." The r esult ing pr essur e dr op
is t he ent r ance effect pr essur e loss at a par t icular flow r at e and pipe
diamet er . To achieve t he gr eat est level of accur acy, t he same pr ocedur e
must be followed for each diamet er under consider at ion by using a
number of t ubes (at least t hr ee) having t he same diamet er but differ ent
lengt hs.
Pr essur e loss at t he ent r ance has also been evaluat ed by subt r act ing
or ifice die dat a fr om capillar y dat a. These calculat ions ar e made wit h
an or ifice and capillar y of t he same diamet er using dat a t aken at an
equivalent volumet r ic flow r at e. Result s wit h soy pr ot ein isolat e show
excellent agr eement bet ween t his met hod and t he Bagley pr ocedur e
discussed pr eviously (Hyashi et al., 1991).
Ent r ance pr essur e loss is used t o cor r ect t he measur ed pr essur e dr op
values:
[2.59]
L/D
P
Q
Q
Q
Q
Q
1
2
3
P
1
Entrance Pressure Loss at Q1
P
3
L/D
L/D
4Q/(R
3
)
L/D
P P
m
P
en
2.5 Data Corrections 113
wher e is t he measur ed pr essur e dr op, and is t he ent r ance
pr essur e loss. Cor r ect ed values of ar e used in calculat ing t he shear
st r ess at t he wall, Eq. [2.3]. The above t echnique is demonst r at ed for
soy dough in Example Pr oblem 2.12.2.
Toms (1958) descr ibed an alt er nat ive met hod of eliminat ing end
effect s. The concept involves a long and shor t t ube of equivalent
diamet er and sufficient lengt h so bot h have an int er nal sect ion wher e
flow is fully developed. If t he volumet r ic flow r at e is t he same in bot h
t ubes, t hen t he exit pr essur e loss ( ) and t he ent r ance pr essur e loss
( ) ar e equal for t he long (subscr ipt L) and shor t (subscr ipt S) t ubes.
Wit h t his idea, t he pr essur e dr op acr oss each t ube may be wr it t en as
[2.60]
and
[2.61]
wher e and ar e t he pr essur e losses in t ube sect ions wher e flow
is fully developed. At const ant flow r at e, t he pr essur e gr adient s in t he
fully developed sect ions ar e equal:
[2.62]
Subt r act ing Eq. [2.61] fr om Eq. [2.60], and using Eq. [2.62] yields a
pr essur e gr adient t er m t hat is fr ee of end effect s:
[2.63]
Eq. [2.63] can be used t o calculat e t he gr adient t er m ( ) in calculat ing
t he wall shear st r ess fr om Eq. [2.3]. In pr act ice, numer ous t ubes would
be used and car e t aken t o be sur e is lar ge enough t o avoid
significant exper iment al er r or s.
Entrance Length. Wit h long t ubes, t he ent r ance cor r ect ion can oft en
be neglect ed. Pr essur e t r ansducer s can be st r at egically placed in pipe
viscomet er s so t hat t he ent r ance r egion does not influence exper iment al
dat a. Also, ent r ance lengt h infor mat ion may be needed t o design, or
evaluat e t he per for mance of, t ube viscomet er s.
P
m
P
en
P
P
ex
P
en
P
L
P
en
+ P
ex
+ P
fd, L
P
S
P
en
+ P
ex
+ P
fd, S
P
fd, L
P
fd, S
P
fd, L
L
L

P
fd, S
L
S
P
fd, L
L
L

P
L
P
S
L
L
L
S
P/L
P
L
P
S
114 Chapter 2. Tube Viscometry
Ther e is some published infor mat ion for est imat ing t he ent r ance
lengt h ( ) r equir ed t o obt ain 98% of fully developed flow which should
be consider ed a conser vat ive guideline in t he design of t ube viscomet er s.
An appr oximat e equat ion t o calculat e for Newt onian fluids in laminar
flow, based on t heor et ical and exper iment al st udies (Boger , 1982), is
[2.64]
The const ant , 0.55, account s for t he ent r ance effect at ver y low values
of t he Reynolds number . Ent r ance lengt h equat ions given for power
law and Bingham plast ic fluids discussed subsequent ly do not have an
analogous t er m.
Collins and Schowalt er (1963) pr esent ed ent r ance lengt h dat a for
pseudoplast ic fluids in laminar flow. Result s wer e given in t er ms of a
modified power law Reynolds number ( ). Ther e was a linear
r elat ionship fr om t hat may be expr essed as
[2.65]
Using Eq. [2.52], t he modified power law Reynolds number may be
wr it t en in t er ms of as
[2.66]
which can be subst it ut ed int o Eq. [2.65] yielding
[2.67]
Plot t ing Eq. [2.67], assuming const ant values, shows
incr easing wit h incr easing values of t he flow behavior index (Fig. 2.11)
when .
Ent r ance lengt h dat a have also been published for Bingham plast ic
fluids in laminar flow (Michiyosi et al., 1966). Their dat a may be
summar ized wit h t he following equat ion:
X
E
X
E
X
E
D
0.55 + 0.055 (N
Re
)
D
n
( u )
(2 n)
/K
0.1 n 1.0
2X
E
D
(.250n + .350)

D
n
( u )
(2 n)

K
_

,
N
Re, PL

D
n
( u )
(2 n)

K
_

,
N
Re, PL
(8
n 1
)

3n + 1
4n
_

,
n
X
E
DN
Re, PL
(.125n + .175) (8
n 1
)

3n + 1
4n
_

,
n
N
Re, PL
X
E
/D
0.1 < n < 1.0
2.5 Data Corrections 115
Figure 2.11. Entrance length of power law fluids.
[2.68]
wher e, r ecall, . Plot t ing Eq. [2.68] illust r at es (Fig. 2.12)
decr easing ent r ance lengt h r equir ement s wit h gr eat er values of t he
yield st r ess.
Wit h shor t or medium lengt h t ubes, ent r ance effect s must be
exper iment ally evaluat ed; however , when t ubes ar e sufficient ly long
(easily designed t his way wit h pipe viscomet er s), t he ent r ance cor r ect ion
can be neglect ed. If is in t he or der of 0.01 pr oblems should be
minimal but t ubes may be long. Der visoglu and Kokini (1986) found,
t hr ough t r ial and er r or exper iment at ion, t hat an ent r ance lengt h of 90
diamet er s was sufficient t o minimize ent r ance effect er r or s in st udying
var ious food pr oduct s including must ar d, ket chup, applesauce, and
t omat o past e. Car e should be exer cised wit h elast ic fluids because t he
ent r ance lengt h may be significant ly higher t han pr edict ed by equat ions
0.2 0.4 0.6 0.8
0.02
0.03
0.04
0.05
0.06
E
R
e
,
P
L
X
D
N
n
X
E
DN
Re, B
.0476exp(5.125c)
c
o
/
w
X
E
/L
116 Chapter 2. Tube Viscometry
Figure 2.12. Entrance length of Bingham plastic fluids.
such as t hose pr esent ed in t his sect ion (Whor low, 1992). A t ube vis-
comet er , wher e t he end effect may be neglect ed due t o high values,
is discussed in Example Pr oblem 2.12.3.
Wall Effects - Sli p Correcti on. Slip occur s when a t hin layer of fluid,
having a viscosit y lower t han t hat of t he fluid, for ms at t he wall of t he
t ube (or t he wall of any viscomet er ). This may be a pr oblem in food
suspensions like fr uit and veget able pur ees. Theor et ically, t he pr oblem
may be at t acked by adding an addit ional t er m, r epr esent ing added flow,
t o t he over all flow r at e t er m. The expr ession descr ibing t he volumet r ic
flow r at e (Eq. [2.15]) may be wr it t en as
[2.69]
At const ant values of , t he above int egr al t er m is const ant so a slip
velocit y may be int r oduced t o account for var iat ions in t he measur ed
values of flow r at e:
0 0.2 0.4 0.6 0.8
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
c
X
D
N
E
R
e
,
B
L/D
Q
without slip
Q
ws

R
3
(
w
)
3

w
()
2
f () d

w
2.5 Data Corrections 117
[2.70]
wher e is t he effect ive slip velocit y which is assumed t o be a funct ion
of t he shear st r ess at t he wall. In t he absence of slip, .
Eq. [2.70] can be expanded by defining an effect ive slip coefficient
( ) such t hat
[2.71]
or
[2.72]
Simplificat ion yields
[2.73]
which is a for m of t he equat ion oft en seen in t he lit er at ur e. J ast r zebski
(1967), however , found t hat t he slip coefficient was a funct ion of t he wall
shear st r ess and also var ied inver sely wit h t ube r adius. To account for
t his finding, a cor r ect ed slip coefficient ( ) was defined t o give a bet t er
mat hemat ical r epr esent at ion of exper iment al dat a:
[2.74]
Using , t he volumet r ic flow r at e expr ession is wr it t en as
[2.75]
may be evaluat ed fr om capillar y t ube measur ement s using var ious
t ubes (at least t hr ee) having differ ent r adii. Exper iment al dat a of
ver sus ar e plot t ed at differ ent values of . Fr om t his plot ,
values of at differ ent values of and const ant ar e obt ained.
Then, by plot t ing ver sus , at const ant , can be
det er mined as t he slope of t he line (Eq. [2.73]) at each value of . The
Q
measured
Q
m
Q
ws
+ R
2
u
s
u
s
u
s
0
u
s
/
w
Q
m

Q
ws

w
+
R
2
u
s

w
Q
m

Q
ws

w
+ R
2

Q
m
R
3

Q
ws
R
3

w
+

R

c


c
R

c
Q
m
R
3

Q
ws
R
3

w
+

c
R
2

Q
m
/(R
3

w
)
w
R
Q
m
/(R
3

w
) R
w
Q
m
/(R
3

w
) 1/R
w

w
118 Chapter 2. Tube Viscometry
r esult ing infor mat ion can be used t o evaluat e as a funct ion of . If
desir ed, t he same pr ocedur e can be followed t o find by consider ing
in place of .
The effect ive slip coefficient , or t he cor r ect ed slip coefficient , is used
t o cor r ect t he volumet r ic flow r at e dat a:
[2.76]
or
[2.77]
is used in t he Rabinowit sch-Mooney equat ion (Eq. [2.20]) t o det er -
mine t he shear r at e at t he wall.
Int r oducing an effect ive slip coefficient may be valuable in dealing
wit h suspensions such as fr uit and veget able pur ees but can give
physically meaningless r esult s (t hose pr oducing negat ive values) for
dense suspensions wher e sliding fr ict ion may be a significant fact or . In
t hese cases it may be mor e appr opr iat e t o model mat er ial movement s
as plug flow inst ead of viscous flow. This appr oach was t aken by J asber g
et al. (1979) in examining t he flow of defat t ed soy flakes in t he scr ew
channel of an ext r uder .
Kokini and Plut chok (1987) have r epor t ed values of t he cor r ect ed
slip coefficient for applesauce (Table 2.2) and slip velocit ies for pot at o
past e ar e pr esent ed in Halliday and Smit h (1995). Cor r ect ed slip
coefficient s wer e also successfully used by Shukla and Rizvi (1995) t o
evaluat e t he flow behavior of but t er .
Vi scous Heati ng. Exper iment al t est s should be designed t o avoid
significant t emper at ur e incr eases due t o viscous heat dissipat ion. Vis-
cous heat ing occur s when any fluid is shear ed. For t unat ely, ser ious
exper iment al er r or s ar e gener ally found only wit h fluids having a ver y
high viscosit y or in inst r ument s oper at ed at ver y high shear r at es.
To det er mine t he t emper at ur e r ise due t o viscous heat ing, t he
equat ions of ener gy and moment um, coupled by t he t emper at ur e
dependence of viscosit y, must be solved. It is not possible t o obt ain an
exact analyt ical solut ion t o t his pr oblem (War r en, 1988). Reasonable
est imat es ar e possible using simplifying assumpt ions and a nomogr aph
out lining a solut ion of t his t ype is available in Dealy (1982) and
Middleman (1968).

w

c
1/R
2
1/R
Q
ws
Q
m

w
R
2
Q
ws
Q
m

c

w
R
Q
ws

2.5 Data Corrections 119


Table2.2. CorrectedSlipCoefficients for ApplesauceCalculatedwith Tubes having
=65 and Radii of 0.4425 cm, 0.3235 cm and 0.2150 cm (Data from Kokini and
Plutchok, 1987)
Pa m
2
/ (Pa s )
40 .0030
54 .0051
68 .0076
81 .010
95 .013
109 .016
123 .019
136 .022
150 .025
A simple met hod t o evaluat e t he ext ent of t he pr oblem is discussed
by Whor low (1992). He shows t hat a pr essur e dr op ( ) causing a volume
( ) of mat er ial t o flow r esult s in wor k done per second equal t o .
This ener gy causes t he mean t emper at ur e of t he volume t o incr ease by
:
[2.78]
wher e is t he densit y and is t he specific heat of t he fluid. Many fluid
foods ar e aqueous syst ems having a densit y and specific heat of
appr oximat ely 1000 kg m
-3
and 4000 J kg
-1
K
-1
, r espect ively. If you
assume t hat a t emper at ur e r ise of less t han 1 C has a negligible effect
on r heological t est ing, t hen a pr essur e dr op less t han 4,000 kPa
(appr oximat ely 40 at mospher es) will not cause a pr oblem due t o viscous
heat ing! It is impor t ant t o r ealize t hat t his calculat ion will over est imat e
t he magnit ude of t he pr oblem. The effect s of viscous heat ing pr oblems
may also be evaluat ed in t er ms of t he Nahme number defined as t he
t emper at ur e r aise due t o viscous heat ing divided by t he t emper at ur e
change necessar y t o alt er viscosit y (Macosko, 1994).
Hole Pressure Error. Somet imes t he pr essur e in a t ube is measur ed
by a t r ansducer communicat ing wit h t he fluid in t he pipe t hr ough a
fluid well connect ed t o t he pipe. This pr act ice causes cur vat ur e in t he
flow st r eamlines which may cr eat e er r or s in t he pr essur e measur ement .
Wit h polymer solut ions, in par t icular , t he pr esence of a nor mal st r ess
L/D

w

c
P
V PV
T
T
PV
Vc
p

P
c
p
c
p

120 Chapter 2. Tube Viscometry


t ends t o lift t he fluid out of t he hole making t he measur ed pr essur e at
t he wall less t han t he t r ue wall pr essur e (Bir d et al. 1987; Bar nes et al.,
1989). This phenomenon is well est ablished and an inst r ument ,
designed t o measur e nor mal st r ess differ ences, has been developed
using t he concept (Lodge, 1988). It is difficult t o apply t he concept t o
fluids wit h a significant yield st r ess because t hese mat er ials would t end
t o fill r ecessed holes hinder ing t he pr oper t r ansmission of t he st r ess
signal.
In designing capillar y viscomet er s, t he hole pr essur e er r or can be
eliminat ed by using ident ical pr essur e t r ansducer s at each sensing
locat ion. Then, when pr essur e differ ences ar e calculat ed, t he hole
pr essur e er r or s cancel out . The pr oblem can also be eliminat ed by using
flush mount ed (diaphr agm t ype) pr essur e t r ansducer s. Hole pr essur e
er r or s ar e minimal for most food mat er ials.
Data Correcti on Summary. Per for ming a complet e r heological
analysis using t ube viscomet r y r equir es a gr eat deal of dat a. A t ypical
sequence of t he st eps is illust r at ed in Fig. 2.13. To evaluat e t he ent r ance
effect , t ubes having t he same r adius but differ ent lengt hs ar e r equir ed.
The ent r ance effect should be evaluat ed for each t ube r adius consider ed
befor e t he measur ed pr essur e dr op can be cor r ect ed and t he shear st r ess
at t he wall calculat ed. Dat a fr om t ubes of t he same lengt h, having
differ ent r adii, ar e needed t o evaluat e t he slip coefficient so t he flow
r at e dat a can be cor r ect ed. Next , t he Rabinowit sch-Mooney equat ion is
applied and t he r heogr am developed. As ment ioned pr eviously, viscous
heat ing and hole pr essur e er r or s ar e usually not a pr oblem wit h food
mat er ials.
Account ing for slip adds significant ly t o t he comput at ional
r equir ement s. One should not e t hat slip pr oblems will decr ease wit h
incr easing t ube sizes. Also, befor e cor r ect ing for slip, one should see if
it is pr esent . This can be done by checking r esult s fr om dat a collect ed
using t ubes of differ ent r adii - at least t wo t ubes wit h significant ly
differ ent r adii. Aft er cor r ect ing for t he end effect , r heogr ams ar e
compar ed. Allowing for differ ences in t he shear r at e r ange, t he r heo-
gr ams for t ime-independent fluids should be ident ical in t he absence of
slip.
2.6 Yield Stress Evaluation 121
Figure 2.13. Typical sequence of steps required for the analysis of tube viscome-
ter data for time-independent fluids.
2.6. Yi eld Stress Evaluati on
Yield st r ess may be det er mined in a t ube viscomet er fr om t he st r ess
t o init iat e fluid movement (Cheng, 1986). Using t his met hod, t he
minimum pr essur e ( ) r equir ed t o cause flow in a hor izont al t ube
viscomet er is measur ed. The yield st r ess is calculat ed fr om a for ce
balance on t he fluid (Eq. [2.2]) as
[2.79]
In pr act ice, t he pr essur e dr op is slowly incr eased unt il flow is obser ved.
If t he st r uct ur e of t he mat er ial causes t he yield st r ess t o exhibit
t ime-dependent char act er ist ics, t hen t he r at e of change of t he applied
pr essur e may influence r esult s.
2.7. Jet Expansi on
J et expansion (also called die swell or ext r udat e swell) can be est i-
mat ed fr om t he pr imar y nor mal st r ess differ ence. Conver sely, t he
nor mal st r ess differ ence can be est imat ed fr om t he jet expansion.
Assuming t he phenomenon is caused by elast ic r ecoil, due t o t he sudden
r emoval of t he t ube, t hen swelling can be est imat ed as (Tanner , 1988)
determine entrance correction for each radius
determine slip coefficient
correct pressure drop data
correct flow rate data
calculate shear stress at the wall
calculate shear rate at the wall
construct rheogram
P
min

o

P
min
R
2L
122 Chapter 2. Tube Viscometry
[2.80]
wher e is t he diamet er of t he capillar y and is t he final diamet er of
t he ext r udat e. The fir st nor mal st r ess differ ence divided by t he shear
st r ess ( ), a t er m called t he r ecover able shear (Tanner , 1988),
is evaluat ed at t he t ube wall. Alt hough t his equat ion is adequat e for
est imat ion pur poses, it excludes some impor t ant fact or s such as
ext ensional viscosit y and vapor izat ion (flashing) of moist ur e, which may
st r ongly influence die swell of ext r uded foods.
Newt onian fluids may exhibit jet expansion having values
r anging fr om 1.12 at low values t o 0.87 at high values (Middle-
man, 1977). Polymer melt s may behave similar ly at low shear r at es but
have number s fr om 2 t o 4 at high shear r at es (Tadmor and Gogos,
1979).
2.8. Sli t Vi scometry
The development of equat ions for slit viscomet er s is analogous t o
t he met hods used in developing t ube viscomet er equat ions. Slit s can
be const r uct ed at food pr ocessing facilit ies wit h r elat ive ease. Food
engineer s have successfully used t hem t o det er mine t he r heological
pr oper t ies of var ious ext r uded food mat er ials: maize gr it s and pot at o
flour (Senouci and Smit h, 1988a and 1988b); cor n meal (Bhat t achar ya
and Padmanabhan, 1992), r ice flour dough (Alt omar e et al., 1992), and
wheat flour dough (Gogos and Bhakuni, 1992). This infor mat ion can be
useful in monit or ing pr oduct qualit y and is r equir ed for var ious engi-
neer ing design calculat ions.
Slit size is descr ibed (Fig. 2.14) in t er ms of t he lengt h ( ) and aspect
r at io ( ) wher e and ar e t he widt h and height of t he slit , r espec-
t ively. To neglect edge effect s, t he aspect r at io should be gr eat er t han
10: . The velocit y pr ofile for a Newt onian fluid in t he slit is
[2.81]
D
e
D

1 +
1
8

11

22

12
_

,
2
1
1
]
1/6
D D
e
(
11

22
)/
12
D
e
/D
N
Re
N
Re
D
e
/D
L
w/h w h
w/h > 10
u f (x
2
)

3Q
2hw
_

1 4

x
2
h
_

,
2
1
1
]
2.8 Slit Viscometry 123
wher e at t he cent er of t he slit and at t he out er edge. The
shear r at e and shear st r ess at t he wall ( ) for a Newt onian fluid
ar e
[2.82]
and
[2.83]
wher e is t he pr essur e dr op acr oss t he slit of lengt h . The shear
st r ess equat ion, Eq. [2.83], is valid for any t ime-independent fluid.
Newt onian viscosit y may be calculat ed as
[2.84]
To calculat e t he shear r at e for non-Newt onian fluids, a gener al
equat ion r elat ing t he volumet r ic flow r at e and t he shear r at e in t he slit
is r equir ed:
[2.85]
Figure 2.14. Slit viscometer.
x
2
0 x
2
h/2
x
2
h/2

w

6Q
h
2
w

w

Ph
2L
P L

w
/

P
L
_

h
3
w
12Q
_

,
Q
w
2

w
_

,
2

w
f ()d
L
w
h
Flow
1
2
3
x
x
x
124 Chapter 2. Tube Viscometry
Eq. [2.85] is analogous t o Eq. [2.15] t hat was developed in der iving t he
Rabinowit sch-Mooney equat ion. It may be solved dir ect ly for power law
and Bingham plast ic fluids:
[2.86]
and
[2.87]
r espect ively, wit h t he st ipulat ion t hat for Eq. [2.87].
If an unknown fluid is being t est ed, a gener al solut ion t o Eq. [2.85]
(compar able t o t he Rabinowit sch-Mooney solut ion) is r equir ed:
[2.88]
wher e t he cor r ect ion fact or is defined as
[2.89]
If t he fluid is Newt onian, and, if power law, . The
t er m is called t he appar ent wall shear r at e for slit flow. Not e, t hat aft er
account ing for geomet r ical differ ences bet ween slit s and t ubes, Eq.
[2.88] and [2.89] ar e ver y similar t o Eq. [2.24] and [2.25].
The above equat ions assume fully developed flow in t he slit . If
cor r ect ions t o exper iment al dat a ar e needed, t he same met hods out lined
for high pr essur e capillar y viscomet er s ar e appr opr iat e. It is ver y
impor t ant t o evaluat e t he need for an ent r ance effect cor r ect ion when
using shor t slit s and t aking as t he pr essur e dr op over t he ent ir e slit .
A super ior met hod of obt aining exper iment al dat a is t o use flush
mount ed pr essur e t r ansducer s inst alled dir ect ly on t he slit . The
t r ansducer s (at least t hr ee, pr efer ably four ) ar e placed a sufficient
dist ance fr om t he ent r ance and exit of t he slit so t he obser ved pr essur e
dr op per unit lengt h ( ), r equir ed in Eq. [2.83], is linear (Han, 1988).
Exit pr essur e dat a fr om a slit have been used t o char act er ize pr imar y
nor mal st r ess differ ences in molt en polymer s (Han, 1988). Bhat t a-
char ya (1993) used t his met hod t o examine t he influence of glut en levels
on t he r heological behavior of br ead dough. This pr ocedur e involves
Q
wh
2
2(2 + 1/n)

w
K
_

,
1/n
Q
h
2
w
w
6
pl

1
3
2

w
_

,
+
1
2

w
_

,
3
1
1
]

w

o

w
f (
w
)

2n + 1
3n
_

6Q
h
2
w
_

,
n
d ln(Ph/(2L))
d ln(6Q/(wh
2
))

d ln(
w
)
d ln(6Q/(wh
2
))
6Q/(wh
2
) n 1.0 n n
P
P/L
2.9 Glass Capillary (U-Tube) Viscometers 125
linear ly ext r apolat ing pr essur e pr ofiles t hr ough t he exit of t he slit t o
det er mine t he exit pr essur e ( ). The fir st nor mal st r ess differ ence is
r elat ed t o t he r at e of change of t he exit pr essur e wit h r espect t o t he
shear r at e at t he wall:
[2.90]
Obt aining good exper iment al dat a using t he exit pr essur e met hod
is difficult because small pr essur e differ ences ar e measur ed. Han (1988)
ar gues t hat t he exit pr essur e met hod should pr oduce accept able r esult s
when is above a cr it ical value of appr oximat ely 25 kPa. Unfor t u-
nat ely, t he exit pr essur e met hod is pr oblemat ic for many foods because
flow does not r emain fully developed t hr ough t he exit of t he r heomet er
violat ing t he assumpt ion t hat t he pr essur e gr adient is const ant over t he
ent ir e inst r ument (Ofoli and St effe, 1993). The analysis of slit vis-
comet er dat a for cor n syr up is given in Example Pr oblem 2.12.5.
2.9. Glass Capi llary (U-Tube) Vi scometers
Glass capillar y viscomet er s (U-t ube viscomet er s) ar e designed t o be
gr avit y oper at ed, and gener ally limit ed t o use wit h Newt onian fluids
having viscosit ies in t he r ange of 0.4 t o 20,000 mPa s. Simple configu-
r at ions ar e seen in Ost wald and Cannon-Fenske t ype unit s (Fig. 2.1).
Somet imes, t hey ar e dr iven wit h applied ext er nal pr essur e t o incr ease
t he viscosit y r ange of t he inst r ument . If t he magnit ude of t he ext er nal
pr essur e makes t he magnit ude of t he st at ic pr essur e head insignificant
in compar ison, t he inst r ument s may be used for non-Newt onian fluids
by applying t he Rabinowit sch-Mooney equat ion t o t he dat a. This,
however , is a labor ious pr ocedur e and should not be consider ed "st an-
dar d pr act ice" for glass capillar y viscomet er s.
The dr iving for ce in gr avit y oper at ed glass capillar y viscomet er s is
t he hydr ost at ic head t hat var ies dur ing dischar ge. This var iat ion in
pr essur e causes a var iat ion in t he shear r at e dur ing t est ing which is
t he main r eason gr avit y oper at ed unit s ar e unsuit able for non-
Newt onian fluids. Wit h Newt onian fluids t he st ar t ing point for analysis
is t he Poiseuille-Hagen equat ion (Eq. [2.28]) wr it t en in t er ms of t he
volumet r ic aver age velocit y, :
[2.91]
P
ex

11

22
P
ex
+
w

dP
ex
d
w
_

,
P
ex

1 +
d(lnP
ex
)
d(ln
w
)
_

w
u Q/R
2
u
(P)R
2
8L
126 Chapter 2. Tube Viscometry
wher e and ar e t he r adius and lengt h of t he capillar y. Pr essur e dr op
over t he capillar y is gener at ed by a height ( ) of liquid:
[2.92]
may be t he aver age or maximum fluid height , but should be t he same
for all measur ement s. Subst it ut ion of Eq. [2.92] int o Eq. [2.91] yields
[2.93]
which can also be wr it t en as
[2.94]
The efflux t ime, defined as t he t ime t o dischar ge a fixed volume of
liquid ( ) fr om t he capillar y bulb (Fig. 2.1), is
[2.95]
so t he aver age velocit y may be expr essed in t er ms of t he dischar ge t ime:
[2.96]
Subst it ut ion of Eq. [2.96] int o Eq. [2.94] yields
[2.97]
which can be expr essed in t er ms of a const ant as
[2.98]
wher e , t he glass capillar y viscomet er const ant , is defined as
[2.99]
The value of may be comput ed dir ect ly fr om Eq. [2.99] if an accur at e
physical descr ipt ion of t he viscomet er is pr ovided. One pr oblem wit h
t his met hod is t hat small var iat ions in geomet r y, par t icular ly var iat ions
in t he r adius due t o wide manufact ur ing t oler ances or film build up aft er
r epeat ed use, may significant ly influence t he numer ical value of .
Hence, it is not sur pr ising t hat t he most common way t o use Eq. [2.98]
is t o det er mine t he pr oper t ies of an unknown fluid fr om t he known
pr oper t ies of a r efer ence fluid. This exper iment al met hod may be
implement ed by not ing t hat is t he same for bot h fluids, so
R L
h
P gh
h
u
ghR
2
8L


ghR
2
8Lu
V
t
V
u( R
2
)
u
V
R
2
t


gh R
4
t
8LV

kt
k
k
gh R
4
8LV
k
k
k
2.9 Glass Capillary (U-Tube) Viscometers 127
[2.100]
or
[2.101]
wher e subscr ipt s 1 and 2 r efer t o t he unknown and known fluids,
r espect ively. A common r efer ence mat er ial is wat er but many com-
mer cially pr oduced st andar ds, mainly silicone oils, ar e also available.
The above met hod of det er mining viscosit y is usually wit hin accept able
limit s of accur acy but does not account for t he var ious exper iment al
er r or s discussed below.
A gener al wor king equat ion which incor por at es er r or s associat ed
wit h small var iat ions in capillar y r adius, end effect s, and kinet ic ener gy
may be wr it t en in t er ms of t he kinemat ic viscosit y ( ) as (Kawat a
et al., 1991)
[2.102]
wher e (wit h unit s of m
2
/s
2
) and (wit h unit s of m
2
) ar e const ant s for
a par t icular capillar y. When t he kinemat ic viscosit ies ( ) and flow
t imes ( ) of t wo st andar d liquids ar e known, t he inst r ument const ant s
may be calculat ed:
[2.103]
and
[2.104]
If t he viscosit y r ange of a glass capillar y viscomet er is incr eased by
oper at ing t he inst r ument wit h applied ext er nal pr essur e, addit ional
cor r ect ion fact or s ar e needed (Kawat a et al., 1991; Van Wazer et al.,
1963).

1
t
1

2
t
2

2
_

t
1
t
2
_

,
/
c
1
t
c
2
t
c
1
c
2

1
,
2
t
1
, t
2
c
1


1
t
1

2
t
2
(t
1
)
2
(t
2
)
2
c
2

(
1
t
1

2
t
2
)t
1
t
2
(t
1
)
2
(t
2
)
2
128 Chapter 2. Tube Viscometry
2.10. Pi peli ne Desi gn Calculati ons
The pur pose of t his sect ion is t o pr ovide t he pr act ical infor mat ion
necessar y t o pr edict pr essur e dr op and power r equir ement s for homo-
geneous, non-t ime-dependent mat er ials in fluid handling syst ems.
Rheological pr oper t ies have a st r ong influence on t he calculat ions and
t his infor mat ion is needed t oselect pr oper pumps and r elat ed equipment
when designing lar ge scale t ube (pipe) viscomet er s or commer cial fluid
handling syst ems (St effe and Mor gan, 1986; St effe and Gar cia, 1987).
Alt hough r heological pr oper t ies can only be evaluat ed fr om dat a t aken
in t he laminar flow r egime, t he case of t ur bulent flow is also pr esent ed
t o pr ovide a t hor ough analysis of pipeline design pr oblems commonly
encount er ed by food pr ocess engineer s.
Mechani cal Energy Balance. The mechanical ener gy balance for an
incompr essible fluid in a pipe may be wr it t en as
[2.105]
wher e , t he summat ion of all fr ict ion losses, is
[2.106]
and subscr ipt s 1 and 2 r efer t o t wo specific locat ions in t he syst em. The
fr ict ion losses include t hose fr om pipes of differ ent diamet er s and a
cont r ibut ion fr om each individual valve, fit t ing, and similar par t s.
is t he wor k out put per unit mass and t he power r equir ement of t he
syst em is found by calculat ing t he pr oduct of and t he mass flow r at e.
A negat ive value of indicat es t hat wor k is being put int o t he syst em
which is t he nor mal funct ion of a pump.
Rheological pr oper t ies ar e r equir ed t o evaluat e t he mechanical
ener gy balance equat ion. Alt hough t her e ar e many mat hemat ical
models available t o descr ibe flow behavior (Table 1.3), few can be con-
sider ed pr act ical for making pr essur e dr op calculat ions involving pipe
flow. Most pumping pr oblems involving fluid foods can be solved using
t he Newt onian, power law or Bingham plast ic models. Over simplifi-
cat ion, however , can cause significant calculat ion er r or s (St effe, 1984).
Fanni ng Fri cti on Factor. The Fanning fr ict ion fact or ( ) is defined,
fr om consider at ions in dimensional analysis, as t he r at io of t he wall
shear st r ess in a pipe t o t he kinet ic ener gy per unit volume:

( u
2
)
2
( u
1
)
2

,
+ g(z
2
z
1
) +
P
2
P
1

+ F + W 0
F
F
2f ( u )
2
L
D
+
k
f
( u )
2
2
W
W
W
f
2.10 Pipeline Design Calculations 129
[2.107]
Subst it ut ing t he definit ion of t he shear st r ess at t he wall of a pipe (Eq.
[2.2]) int o Eq. [2.107] gives
[2.108]
wher e . Hence, t he ener gy loss per unit mass (needed in t he
mechanical ener gy balance) may be expr essed in t er ms of :
[2.109]
Some engineer s calculat e t he fr ict ion losses wit h t he Dar cy fr ict ion
fact or which is equal t o four t imes t he Fanning fr ict ion fact or . Pr essur e
dr op calculat ions may be adjust ed for t his differ ence. Final r esult s ar e
t he same using eit her fr ict ion fact or . Calculat ions in t his t ext deal
exclusively wit h t he Fanning fr ict ion fact or .
In laminar flow, values can be det er mined fr om t he equat ions
descr ibing t he r elat ionship bet ween pr essur e dr op and flow r at e for a
par t icular fluid. Consider , for example, a Newt onian fluid ( ). Using
Eq. [2.28], t he volumet r ic aver age velocit y for t his mat er ial, in laminar
t ube flow, may be expr essed as:
[2.110]
Simplificat ion gives an expr ession for t he pr essur e dr op per unit lengt h:
[2.111]
Subst it ut ing Eq. [2.111] int o Eq. [2.108] yields t he fr ict ion fact or :
[2.112]
which is a common equat ion appr opr iat e for pr edict ing fr ict ion fact or s
for Newt onian fluids when . Using t he same appr oach, laminar
flow fr ict ion fact or s for power law and Bingham plast ic fluids may be
calculat ed, r espect ively, fr om t he following equat ions:
[2.113]
f
2
w
( u )
2
f
(P)R
L( u )
2

(P)D
2L( u )
2
P P
2
P
1
f
P


f 2L( u )
2
D
f

u
Q
R
2

1
R
2

(P)R
4
8L
_

,

(P)D
2
32L
P
L

32u
D
2
f

P
L
1
1
]

D
2( u )
2
_

32( u )
D
2
1
1
]

D
2( u )
2
_

,

16
N
Re
N
Re
< 2100
f
16
N
Re, PL
130 Chapter 2. Tube Viscometry
and
[2.114]
The above equat ions ar e appr opr iat e when t he laminar flow cr it er ia,
pr esent ed in Sect ion 2.4, ar e sat isfied. Eq. [2.114], plot t ed in Fig. 2.15,
is an appr oximat ion for t he Fanning fr ict ion fact or based on t he
assumpt ion t hat (Heywood, 1991a). This figur e illust r at es
how t he fr ict ion fact or decr eases wit h lar ger values of t he Bingham
Reynolds number , and incr eases wit h lar ger values of t he Hedst r om
number . A lar ger ver sion of Fig. 2.15, mor e convenient for pr oblem
solving, is given in Appendix [6.17].
Figure 2.15. Fanning friction factors (from Eq. [2.114]) for Bingham plastic flu-
ids in laminar flow at different values of the Hedstrom Number.
In t ur bulent flow, fr ict ion fact or s may be det er mined fr om empir ical
equat ions (Table 2.3) for mulat ed fr om exper iment al dat a (Gr ovier and
Aziz, 1972). The equat ions ar e only applicable t o smoot h pipes which
include sanit ar y piping syst ems for food. It may be ver y difficult t o
accur at ely pr edict t r ansit ion fr om laminar t o t ur bulent flow in act ual
f
16(6N
Re, B
+ N
He
)
6N
Re, B
2
(
o
/
w
)
4
1
100 200 300 500 1,000 2,000 3,000 5,000 10,000
0.001
0.002
0.005
0.01
0.02
0.05
0.1
0.2
0.5
1
F
a
n
n
i
n
g

F
r
i
c
t
i
o
n

F
a
c
t
o
r
100 1,000 5,000 10,000 25,000 50,000 100,000 200,000
N
He
N
Re,B
2.10 Pipeline Design Calculations 131
Table2.3. Fanning Friction Factor Equations for Turbulent Flow in Smooth Tubes
Flu id Fa n n in g Fr ict ion Fa ct or
Newt on ia n
wh er e:
Power La w
wh er e:
Bin gh a m Pla s t ic
wh er e:
a n d
pr ocessing syst ems and t he equat ions given her e ar e only int ended for
use in est imat ing t he power r equir ement s for pumping. Cur ves for
power law fluids in t ur bulent flow ar e plot t ed in Fig. 2.16 (a lar ger
ver sion of t he same plot is given in Appendix [6.18]). Newt onian fluids
ar e r epr esent ed by t he cur ve wit h = 1.0.
Ki neti c Energy Evaluati on. The kinet ic ener gy t er m in t he
mechanical ener gy balance can be evaluat ed if t he kinet ic ener gy cor -
r ect ion fact or ( ) is known. In t ur bulent flow of any fluid, .
Expr essions t o comput e values for var ious fluids in laminar flow ar e
summar ized in Table 2.4. These equat ions may be given in t er ms of ,
t he flow behavior index, or which is defined as t he r at io of t he yield
st r ess ( ) t o t he shear st r ess at t he wall ( ). Equat ions pr ovided for
t he Bingham plast ic and Her schel-Bulkley cases ar e appr oximat ions.
1
f
4.0log
10
(N
Re
f ) 0.4

N
Re

Du

1
f

4
n
0.75
_

,
log
10

(N
Re, PL
)f
(1 ( n/2))1
]

0.4
n
1.2
_

,
K(

)
n
N
Re, PL

D
n
( u )
2 n

8
n 1
K
_

4n
3n + 1
_

,
n
1
f
4.53 log
10
(1 c) + 4.53 log
10
((N
Re, B
)f ) 2.3

pl

+
o
N
Re, B

Du

pl
c

o

2
o
f ( u )
2
n
2
n
c

o

w
132 Chapter 2. Tube Viscometry
Figure 2.16. Fanning friction factors (equation given in Table 2.3) for power law
fluids in turbulent flow at different values of the flow behavior index.
An exact , but cumber some, mat hemat ical equat ion for t he kinet ic
ener gy cor r ect ion fact or of a Her schel-Bulkley fluid has been published
by Osor io and St effe (1984). Values of , det er mined fr om t his equat ion,
ar e plot t ed in Fig. 2.17. This figur e r eveals some int er est ing feat ur es
of t he kinet ic ener gy cor r ect ion fact or : values go t o 2 as t he yield st r ess
appr oaches t he wall shear st r ess for all values of ; values incr ease
wit h decr easing values of . Over all, t he numer ical value of r anges
fr om 0.74 t o 2 for Her schel-Bulkley fluids (Osor io and St effe, 1984). The
minimum value of occur s at as appr oaches infinit y. KE
differ ences ar e usually small and oft en ignor ed in evaluat ing power
r equir ement s when select ing pumps.
2,000 4,000 6,000 8,000 10,000
0.002
0.004
0.006
0.008
0.01
0.012
0.014
F
a
n
n
i
n
g

F
r
i
c
t
i
o
n

F
a
c
t
o
r
N
Re,PL
n = 1.2
n = 1.0
n = 0.8
n = 0.6
n = 0.4
n = 0.3
n = 0.2
Power Law Fluids in Turbulent Flow

n
n
0.74 c 0 n
2.10 Pipeline Design Calculations 133
Table 2.4. Kinetic Energy Correction Factors for Laminar Flow in Tubes
Flu id , dimen s ion les s
Newt on ia n 1.0
Power La w
+
Bin gh a m Pla s t ic
+
Her s ch el-Bu lkley
for
a n d
for
+
Solu t ion for t h e Bin gh a m pla s t ic ma t er ia l is wit h in 2.5% of t h e t r u e s olu t ion
(Met zn er , 1956). Er r or s in u s in g t h e Her s ch el-Bu lkley s olu t ion a r e les s t h a n 3% for
0.1 1.0 bu t a s h igh a s 14.2% for 0.0 0.1 (Br iggs a n d St effe, 1995).
Fri cti on Losses: Valves, Fi tti ngs, and Si mi lar Parts. Fr ict ion loss
coefficient s ( ) must be det er mined fr om exper iment al dat a. In gener al,
published values ar e for t he t ur bulent flow of wat er t aken fr om Cr ane
(1982). An adequat e summar y of t hese number s may be found in
Sakiadis (1984). Laminar flow dat a ar e much mor e limit ed. Some ar e
available for var ious fluids: Newt onian (Kit t r edge and Rowley, 1957),
shear -t hinning (Baner jee et al., 1994; Lewicki and Skier kowski, 1988;
St effe et al., 1984) and shear -t hickening (Gr iskey and Gr een, 1971).
Over all, t he quant it y of engineer ing dat a r equir ed t o pr edict pr essur e
losses in valves and fit t ings for fluids, par t icular ly non-Newt onian
fluids, in laminar flow is insufficient .
Given t his sit uat ion, a "r ule of t humb" est imat ion pr ocedur e is
needed. Fir st some gener al obser vat ions should be made: a) The
behavior of values for Newt onian and non-Newt onian fluids is similar
(Met zner , 1961; Skelland, 1967), b) values decr ease wit h incr easing
pipe diamet er (Cr ane, 1982) -- t hey may dr op as much as 30% in going
fr om 3/4 t o 4 inch (1.9 t o 10.2 cm) pipe, c) values shar ply incr ease wit h


2(2n + 1)(5n + 3)
3(3n + 1)
2
K(

)
n

2
2 c

pl

+
o
exp(0.168 c 1.062 n c 0.954 n
.5
0.115 c
.5
+ 0.831)
0.06 n 0.38
K(

)
n
+
o
exp(0.849 c 0.296n c 0.600 n
.5
0.602 c
.5
+ 0.733)
0.38 < n 1.60
c c
k
f
k
f
k
f
k
f
134 Chapter 2. Tube Viscometry
Figure 2.17. Kinetic energy correction factors for the laminar flow of Herschel-
Bulkley fluids (from Osorio and Steffe, 1984).
decr easing Reynolds number s (Cheng, 1970; Kit t r edge and Rowley,
1957; Lewicki and Skier kowski, 1988; St effe et al., 1984) in t he laminar
flow r egime but ar e const ant in t he t ur bulent flow r egime (Sakiadis,
1984) and show lit t le change above (Kit t r edge and Rowley,
1957), d) Ent r ance pr essur e losses for power law fluids in laminar flow
decr ease wit h smaller values of t he flow behavior index (Collins and
Schowalt er , 1963), e) Ent r ance losses for Bingham plast ic fluids
decr ease wit h incr easing values of t he yield st r ess when t he wall shear
st r ess ( ) is const ant (Michiyosi et al., 1966), f) Resist ance t o flow
of non-Newt onian fluids in laminar flow, t hr ough similar valves, can be
expect ed t obe up t o133 per cent higher t han t hat obser ved for Newt onian
fluids (Ur y, 1966).
Fr ict ion loss coefficient s for many valves and fit t ings ar e summa-
r ized in Tables 2.5 and 2.6. values for t he sudden cont r act ion, or
expansion, of a Newt onian fluid in t ur bulent flow, may be found in Cr ane
(1982). The loss coefficient for a sudden cont r act ion is calculat ed in
t er ms of t he small and lar ge pipe diamet er s:
0 0.2 0.4 0.6 0.8 1
0.6
0.8
1
1.2
1.4
1.6
1.8
2
c = Yield Stress / Shear Stress at the Wall
n=0.1 n=0.3 n=0.5 n=0.7 n=0.9 n=2.0
Flow Behavior Index
N
Re
500
PR/(2L)
k
f
2.10 Pipeline Design Calculations 135
Table2.5. Friction Loss Coefficients ( Values) for theLaminar Flow of Newtonian
Fluids throughValves and Fittings (fromSakiadis (1984) withOriginal Datafrom
Kittredge and Rowley, 1957)
Type of Fit t in g or Va lve =1000 500 100
90-deg. elbow, s h or t r a diu s 0.9 1.0 7.5
Tee, s t a n da r d, a lon g r u n 0.4 0.5 2.5
Br a n ch t o lin e 1.5 1.8 4.9
Ga t e va lve 1.2 1.7 9.9
Globe va lve, compos it ion dis k 11 12 20
Plu g 12 14 19
An gle va lve 8 8.5 11
Ch eck va lve, s win g 4 4.5 17
[2.115]
Losses for a sudden enlar gement , or an exit , ar e det er mined as
[2.116]
The lar gest velocit y which is t he mean velocit y in t he smallest diamet er
pipe, should be used for bot h cont r act ions and expansions in calculat ing
t he fr ict ion loss t er m ( ) found in Eq. [2.106].
Aft er evaluat ing t he available dat a for fr ict ion loss coefficient s in
laminar and t ur bulent flow, t he following "r ule-of-t humb" guidelines,
conser vat ive for shear -t hinning fluids, ar e pr oposed for est imat ing
values:
1. For Newt onian fluids in laminar or t ur bulent flow use t he dat a of
Kit t r edge and Rowley (1957) or Sakiadis (1984), r espect ively
(Tables 2.5 and 2.6).
2. For non-Newt onian fluids above a Reynolds number ( )
of 500, use dat a for Newt onian fluids in t ur bulent flow (Table 2.6).
3. For non-Newt onian fluids in t he Reynolds number r ange of
use t he following equat ion:
k
f
N
Re
k
f
.5

D
small
D
large
_

,
2
_

,
k
f

D
small
D
large
_

,
2
_

,
2
k
f
u
2
/2
k
f
N
Re, PL
or N
Re, B
20 N 500
136 Chapter 2. Tube Viscometry
Table 2.6. Friction Loss Coefficients for the Turbulent Flow of Newtonian Fluids
through Valves and Fittings (from Sakiadis, 1984)
Type of Fit t in g of Va lve
45-deg. elbow, s t a n da r d 0.35
45-deg. elbow, lon g r a diu s 0.2
90-deg. elbow, s t a n da r d 0.75
Lon g r a diu s 0.45
Squ a r e or mit er 1.3
180-deg. ben d, clos e r et u r n 1.5
Tee, s t a n da r d, a lon g r u n , br a n ch bla n ked off 0.4
Us ed a s elbow, en t er in g r u n 1.0
Us ed a s elbow, en t er in g br a n ch 1.0
Br a n ch in g flow 1.0
Cou plin g 0.04
Un ion 0.04
Ga t e va lve, open 0.17
3/ 4 open 0.9
1/ 2 open 4.5
1/ 4 open 24.0
Dia ph r a gm va lve, open 2.3
3/ 4 open 2.6
1/ 2 open 4.3
1/ 4 open 21.0
Globe va lve, bevel s ea t , open 6.0
1/ 2 open 9.5
Compos it ion s ea t , open 6.0
1/ 2 open 8.5
Plu g dis k, open 9.0
3/ 4 open 13.0
1/ 2 open 36.0
1/ 4 open 112.0
An gle va lve, open 2.0
Y or blowoff va lve, open 3.0
Plu g cock =0 (fu lly open ) 0.0
=5 0.05
=10 0.29
=20 1.56
=40 17.3
=60 206.0
Bu t t er fly va lve =0 (fu lly open ) 0.0
=5 0.24
=10 0.52
=20 1.54
=40 10.8
=60 118.0
Ch eck va lve, s win g 2.0
Dis k 10.0
Ba ll 70.0
Foot va lve 15.0
Wa t er met er , dis k 7.0
Pis t on 15.0
Rot a r y (s t a r -s h a ped dis k) 10.0
Tu r bin e-wh eel 6.0
k
f












2.10 Pipeline Design Calculations 137
[2.117]
wher e is depending on t he t ype of fluid in quest ion.
The const ant , , is found for a par t icular valve or fit t ing (or r elat ed
it ems like cont r act ions and expansions) by mult iplying t he t ur -
bulent flow fr ict ion loss coefficient by 500:
[2.118]
Values of for many st andar d it ems may be calculat ed fr om t he
values pr ovided in Table 2.6. The value of 20 was ar bit r ar ily
set as a lower limit of in Eq. [2.117] because ver y low values of
t he Reynolds number in t hat equat ion will gener at e unr easonably
high values of t he fr ict ion loss coefficient . Values of will
cover most pr act ical applicat ions for fluid foods. Eq. [2.117] and
[2.118] may also be used for Newt onian fluids when is used
for .
The above guidelines ar e offer ed wit h caut ion and should only be used
in t he absence of act ual exper iment al dat a. Many fact or s, such as a
high ext ensional viscosit y, may significant ly influence values.
Generali zed Pressure Drop Calculati on. Met zner (1956) pr esent s
a gener alized appr oach t o r elat e flow r at e and pr essur e dr op for t ime-
independent fluids in laminar flow. The gover ning equat ion is
[2.119]
wher e
[2.120]
and ar e easily det er mined fr om a plot of t he exper iment al dat a.
Ther e is a st r ong similar it y wit h t he above equat ion and t hose descr ibing
t he flow of power law fluids in pipes. In fact , for a power law fluid,
[2.121]
k
f

A
N
N N
Re, PL
or N
Re, B
A
A (k
f
)
turbulent
(500)
A
k
f
N
N 20
N
Re
N
k
f
(P)R
2L
K

4Q
R
3
_

,
n
n
d ln((P)R/(2L))
d ln(4Q/(R
3
))

d(ln
w
)
d(ln)
K n
n n and K K

3n + 1
4n
_

,
n
138 Chapter 2. Tube Viscometry
Wit h t he gener al solut ion, may var y wit h t he shear st r ess at t he wall
and must be evaluat ed at t he par t icular value of in quest ion.
Eq. [2.119] and [2.120] have pr act ical value when consider ing dir ect
scale-up fr om dat a t aken wit h a small diamet er t ube or for cases wher e
a well defined equat ion (power law, Bingham plast ic or Her schel-
Bulkley) is not applicable. A similar met hod is available for scale-up
pr oblems involving t he t ur bulent flow of t ime-independent fluids (Lor d
et al., 1967).
Slip and t ime-dependent behavior may be a pr oblem in pr edict ing
pr essur e loss in pipes. One solut ion is t o incor por at e t hese effect s int o
t he consist ency coefficient . Houska et al. (1988) give an example of t his
t echnique for t he t r anspor t of minced meat in pipes wher e incor po-
r at es t he pr oper t y changes due t o t he aging of t he meat and wall slip
as a funct ion of pipe diamet er . An exer cise in pipeline design is
pr esent ed in Example Pr oblem 2.12.6.
2.11. Veloci ty Profi les In Turbulent Flow
Velocit y pr ofiles, cr it ical in t her mal pr ocessing syst ems (par t icular ly
in hold t ubes), ar e st r ongly influenced by r heological pr oper t ies.
Accur at ely pr edict ing velocit y pr ofiles for fluids in t ur bulent flow is
difficult . Relat ionships for Newt onian fluids ar e r eliable. Those for
power law fluids ar e available but t hey have not r eceived adequat e
exper iment al ver ificat ion for fluid foods. Appr oximat e mean, divided
by maximum, velocit ies ar e summar ized for some condit ions in Table
2.7. A det ailed discussion of laminar flow pr ofiles is pr esent ed in Sec.
2.3.
Newtoni an Flui ds. Semi-t heor et ical pr edict ion equat ions for t he
velocit y pr ofile of Newt onian fluids in t ur bulent flowar e well est ablished
and pr esent ed in t er ms of t hr ee dist inct r egions of t he pipe (Br odkey
and Her shey, 1988):
for t he viscous sublayer (called laminar sublayer in some lit er at ur e)
[2.122]
for t he t r ansit ion zone wher e t ur bulent fluct uat ions ar e gener at ed
[2.123]
and for t he t ur bulent cor e
n

w
K
u
+
y
+
y
+
5
u
+
3.05 + (11.513) log
10
( y
+
) 5 < y
+
< 30
2.11 Velocity Profiles In Turbulent Flow 139
Table 2.7. Approximate Average and Maximum Velocities for Newtonian and
Power Law Fluids in Tube Flow
Newt on ia n Flu ids
< 2100 4000 10
4
10
5
10
6
10
7
10
8
0.5 0.790 0.811 0.849 0.875 0.893 0.907
Power La w Flu ids : La min a r Flow
n 0.0 0.3 0.4 0.5 1.0 2.0
1.00 0.68 0.64 0.60 0.50 0.43 0.33
Power La w Flu ids : Tu r bu len t Flow
0.2 0.2 0.2 0.5 0.5 0.5
10
4
10
5
10
6
10
4
10
5
10
6
0.92 0.94 0.95 0.86 0.89 0.92
[2.124]
wher e
[2.125]
and
[2.126]
and
[2.127]
N
Re
u/u
max
u/u
max
(n + 1)/(3n + 1)

u/u
max
n
N
Re, PL
u/u
max
u
+
5.5 + (5.756) log
10
( y
+
) 30 y
+
u
+

u
u
*
y
+

yu
*

u
*

f
2
140 Chapter 2. Tube Viscometry
The dist ance fr om t he pipe wall, , is defined as
[2.128]
Not e t hat t he or igin of t he coor dinat e syst em is (by convent ion) locat ed
at t he wall, . Ther efor e, t he velocit y is zer o at wher e and
a maximum at t he cent er of t he pipe wher e and . The combined
velocit y equat ions const it ut e t he univer sal velocit y pr ofile.
Once t he maximum velocit y has been det er mined, t he power law
equat ion may be used t o appr oximat e at ot her locat ions:
[2.129]
Eq. [2.129] does a r easonable job pr edict ing velocit y pr ofiles in spit e of
t he fact t hat it is not dependent on t he Reynolds number . Gr ovier and
Aziz (1972) not e t hat t his equat ion is most appr opr iat e for
and . Also t he exponent may var y fr om at
t o at . Maximum velocit y, of a t ur bulent
Newt onian fluid in t ube flow, is calculat ed in Example Pr oblem 2.12.7.
Power Law Flui ds. Dodge and Met zner (1959) der ived equat ions t o
descr ibe t he velocit y pr ofile of power law fluids in t ube flow. Small er r or s
wer e cor r ect ed by Skelland (1967) and t he final equat ions pr esent ed as
[2.130]
for t he laminar sublayer and
[2.131]
for t he t ur bulent cor e, wher e incor por at es t he flow behavior index
and t he consist ency coefficient needed for t he consider at ion of power
law fluids:
[2.132]
Const ant s wer e obt ained fr om fr ict ion fact or measur ement s so t he
t hickness of t he laminar sublayer was not obt ained. These equat ions
ar e applied in Example Pr oblem 2.12.8.
An alt er nat ive equat ion for pr edict ing velocit y in t he t ur bulent cor e
for power law fluids was pr esent ed by Clapp (1961):
y
y R r
r R r R y 0
r 0 y R
1/7
u/u
max
u
u
max

y
R
_

,
1/ 7

R r
R
_

,
1/ 7
0.1 < y/R < 1.0
3000 < N
Re
< 100, 000 1/6
N
Re
4, 000 1/10 N
Re
3, 200, 000
u
+
( y
+
)
1/n
u
+

5.66
n
.75
log
10
( y
+
)
0.566
n
1.2
+
3.475
n
.75

1.960 + 0.815n 1.628n log


10

3 +
1
n
_

,
1
1
]
y
+
y
+
y
n
(u
*
)
2 n
/K
2.12.1 Conservation of Momentum Equations 141
[2.133]
This equat ion cor r elat ed well wit h exper iment al dat a in which
and .
2.12. Example Problems
2.12.1. Conservati on of Momentum Equati ons
Show t hat an equat ion descr ibing t he shear st r ess for a fluid in t ube
flow may be det er mined fr om t he conser vat ion of moment um pr inciple.
Compar e t he r esult t o Eq. [2.2] which was der ived dir ect ly fr om a for ce
balance. This example pr oblem is included t o illust r at e an alt er nat ive
met hod of obt aining r heological r elat ionships.
Fir st a number of assumpt ions must be st at ed: flow is for an
incompr essible fluid under st eady, laminar , isot her mal condit ions;
t her e ar e no ent r ance and exit effect s; and flow is par allel t o t he pipe
wall, i.e., t her e is no r adial or cir cumfer ent ial flow. Then, using cylin-
dr ical coor dinat es (Fig. 2.18),
[2.134]
The cont inuit y (conser vat ion of mass) equat ion may be wr it t en as
[2.135]
Under st eady condit ions wit h an incompr essible fluid ( ) and,
wit h t he assumpt ion t hat r adial and cir cumfer ent ial velocit ies ar e zer o,
t his equat ion r educes t o
[2.136]
indicat ing t hat conser vat ion of mass is sat isfied and velocit y is const ant
along t he (axial) dir ect ion. Consider ing t he above assumpt ions t he
moment um equat ions (Fig. 2.18) collapse t o t he following:
[2.137]
[2.138]
[2.139]
u
+

2.78
n
2.303log
10
( y
+
) +
3.80
n
0.698 < n < 0.813 5, 480 < N
Re, PL
< 42, 800
u
r
u

0 u
z
u
z
(r)

t
+
1
r

r
(ru
r
) +
1
r

(u

) +

z
(u
z
) 0
/t 0
du
z
dz
0
z
r component: 0
P
r
component: 0
1
r
P

z component: 0
P
z
+
1
r

r
(r
rz
)
142 Chapter 2. Tube Viscometry
Figure 2.18. Momentum equations in cylindrical coordinates. [Note: P is an
inclusive pressure term which ignores small variation in pressure due to varia-
tions in height ( )].
The and component s would int egr at e t o a const ant indicat ing t hat
does not var y in t hose dir ect ions. It follows t hat is a funct ion of
only:
[2.140]
Then, we can r eplace wit h , and wit h
t o wr it e t he component of t he moment um equat ion, Eq. [2.139],
as
[2.141]
or
[2.142]
wher e is r eplaced by t he const ant t er m, . This is allowed
because t he left hand side of Eq. [2.141] is a funct ion of only, and t he
z
r
r

u
r
t
+ u
r
u
r
r
+
u

r
u
r

2
r
+ u
z
u
r
z
_

,

P
r
+
1
r

r
(r
rr
) +
1
r

r
+

rz
z
+ g
r

t
+ u
r
u

r
+
u

r
u

+
u
r
u

r
+ u
z
u

z
_

,

1
r
P

+
1
r
2

r
(r
2

r
) +
1
r

z
z
+ g

u
z
t
+ u
r
u
z
r
+
u

r
u
z

+ u
z
u
z
z
_

,

P
z
+
1
r

r
(r
rz
) +
1
r

zz
z
+ g
z
gh
r
P P z
P P(z)
(1/r)(r
rz
)/r (1/r)d(r
rz
)/dr P/z
dP/dz z
1
r
d
dr
(r
rz
)
dP
dz
1
r
d
dr
(r
rz
)
( P)
L
dP/dz (P)/L
r
2.12.2 Capillary Viscometry - Soy Dough 143
r ight hand side of t he equat ion is a funct ion of only; hence, bot h sides
ar e equal t o a const ant (t he or igin of t he negat ive sign will be explained
shor t ly). Int egr at ing Eq. [2.142] yields
[2.143]
giving
[2.144]
must be zer o for t he shear st r ess t o be finit e at t he cent er line of t he
pipe, t her efor e,
[2.145]
which, wr it t en in one dimensional symbolism, is like Eq. [2.2] found
fr om a simple for ce balance:
[2.146]
Eq. [2.146] and [2.2] ar e equivalent , but appear differ ent , due t o t he
pr esence of a negat ive sign in Eq. [2.146]. This r eflect s t he sign
convent ion adopt ed in developing t he equat ions illust r at ed in Fig. 2.18
and ut ilized in var ious t ext books (Bir d et al., 1960; Denn, 1980; Br odkey
and Her shey, 1988). In t hose equat ions is negat ive so t he negat ive
sign is r equir ed t o make t he shear st r ess posit ive. In t his book, however ,
was assumed t o be a posit ive number in der iving our init ial for ce
balance (Eq. [2.1]) making a negat ive sign unnecessar y.
2.12.2. Capi llary Vi scometry - Soy Dough
Dat a for a defat t ed soy flour dough fr om a capillar y viscomet er ar e
summar ized in Table 2.8. Det er mine t he r heological pr oper t ies of t he
mat er ial.
The fir st st ep in t he analysis is t o cor r ect t he measur ed pr essur e loss
for ent r ance effect s. Pr essur e losses ar e plot t ed (Fig. 2.19) at differ ent
values of for const ant appar ent wall shear r at es, .
Regr ession analysis of t hese cur ves gives t he r equir ed ent r ance loss
cor r ect ion values at = 0. In t his par t icular pr oblem, ent r ance losses
const it ut e a ver y high per cent age, over 80% in many t r ials, of t he t ot al
pr essur e dr op acr oss t he capillar y (Table 2.8).
z

d(r
rz
)

( P)
L
r
_

,
dr

rz

( P)
2L
r +
C
1
r
C
1

rz

( P)r
2L

( P)r
2L
P
P
4Q/(R
3
) L/D
L/D
144 Chapter 2. Tube Viscometry
Table 2.8. Capillary Viscometer (D=3.18mm) Data for Dough (34.7%) at Room
Temperature Made from Defatted Soy Flour Treated to Cause Protein Denatu-
ration (Data from Morgan, 1979)
(1/ s ) (1/ s ) (-) (MPa ) (MPa ) (MPa ) (kPa ) (1/ s )
47.4 11.9 2 4.21 3.58 0.63 78.75 72.13
47.4 11.9 5 5.21 3.58 1.63 81.50 76.73
47.4 11.9 8 6.14 3.58 2.56 80.00 74.17
94.8 23.7 2 5.46 4.63 0.83 103.75 165.80
94.8 23.7 5 6.81 4.63 2.18 109.00 183.38
94.8 23.7 8 8.02 4.63 3.39 105.94 172.87
190.0 47.5 2 5.25 4.38 0.87 108.75 253.90
190.0 47.5 5 6.80 4.38 2.42 121.00 303.49
190.0 47.5 8 8.30 4.38 3.92 122.50 310.48
948.0 237.0 2 7.68 6.17 1.51 188.75 1457.45
948.0 237.0 5 10.12 6.17 3.95 197.50 1583.76
948.0 237.0 8 12.31 6.17 6.14 191.88 1500.96
*
Mea s u r ed pr es s u r e dr op;
**
En t r a n ce los s pr es s u r e cor r ect ion
***
Cor r ect ed pr es s u r e dr op.
Figure 2.19. Measured pressure drop (including entrance loss) for capillary vis-
cometer data from defatted soy flour.
4Q/(R
3
) Q/(R
3
) (P)
m
*
(P)
en
**
(P)
***
L/D
w

w
1 2 3 4 5 6 7 8 9
2
4
6
8
10
12
14
L/D
P
r
e
s
s
u
r
e

D
r
o
p
,

M
P
a
(4 Q)/( R )
3
47.4 1/s 94.8 1/s 190.0 1/s 948.0 1/s
2.12.2 Capillary Viscometry - Soy Dough 145
Figure 2.20. Capillary data for soy dough corrected for pressure loss.
To det er mine t he shear r at e, t he Rabinowit sch-Mooney equat ion
(Eq. [2.20]), must be evaluat ed:
The r elat ionship bet ween and was plot t ed (Fig. 2.20) and
det er mined (by cur ve fit t ing) as a power funct ion:
The der ivat ive of t his equat ion is
which can be inser t ed int o t he Rabinowit sch-Mooney equat ion t o
det er mine t he shear r at e at t he wall:
60 80 100 120 140 160 180 200 220
0
50
100
150
200
250
Corrected Data for Soy Dough
Q
/
(


R

)
3
,

1
/
s
w
, kPa

w
f (
w
)

3Q
R
3
_

,
+ (
w
)

d(Q/(R
3
))
d
w
1
1
]
Q/ (R
3
)
w
Q
R
3
3.1(10
6
) (
w
)
3.44
d(Q/(R
3
))
d
w
1.06(10
5
) (
w
)
2.44

3Q
R
3
_

,
+
w
[1.06(10
5
) (
w
)
2.44
]
146 Chapter 2. Tube Viscometry
Result s of t his comput at ion ar e summar ized in Table 2.8. Aft er making
t hese calculat ions, t he basic element s of t he r heogr am illust r at ing shear
st r ess ver sus shear r at e ar e available for plot t ing (Fig. 2.21). If t he
mat er ial is consider ed t o be shear -t hinning over t he shear r at e r ange
consider ed, t he following fluid par amet er s ar e calculat ed: = 23.3 kPa
s
n
, = 0.29.
Figure 2.21. Rheogram for soy dough.
2.12.3. Tube Vi scometry - 1.5%CMC
Dat a for a 1.5% aqueous solut ion of sodium car boxymet hylcellulose
(CMC) wer e collect ed at r oom t emper at ur e using a capillar y t ube vis-
comet er (Table 2.9). Det er mine t he r heological pr oper t ies of t his
mat er ial. Since t he minimum is 248, ent r ance effect s ar e assumed
t o be insignificant . Assume CMC densit y is equal t o 1003 kg/m
3
.
Regr ession analysis of ver sus or ln( ) in t he for m
of
K
n
0 500 1,000 1,500 2,000
60
80
100
120
140
160
180
200
220
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

k
P
a
Soy Dough
L/D
ln(4Q/(R
3
)) ln(
w
)
2.12.3 Tube Viscometry - 1.5% CMC 147
Table 2.9. Rheological Data for a 1.5% Aqueous Solution of Sodium Carboxy-
methylcellulose(CMC; =1003 kg/m
3
) Collected at Room Temperature(Datafrom
Middleman, 1968)
(m) (m) (Pa ) (m
3
/ s ) (Pa ) (1/ s ) (1/ s )
2.71E-03 0.944 1.379E+05 5.910E-07 98.97 302.47 409.50
2.71E-03 0.944 2.710E+05 2.950E-06 194.48 1509.78 2044.04
2.71E-03 0.944 4.103E+05 8.320E-06 294.43 4258.10 5764.89
2.71E-03 0.944 5.916E+05 2.100E-05 424.58 10747.61 14550.81
2.71E-03 0.674 1.034E+05 7.330E-07 103.96 375.14 507.89
2.71E-03 0.674 1.379E+05 1.260E-06 138.62 644.86 873.05
2.71E-03 0.674 2.724E+05 7.060E-06 273.77 3613.24 4891.84
2.71E-03 0.674 3.999E+05 1.760E-05 401.99 9007.52 12194.96
2.71E-03 0.674 5.482E+05 3.510E-05 551.00 17963.86 24320.64
1.82E-03 0.634 1.413E+05 1.920E-07 101.44 324.40 439.20
1.82E-03 0.634 1.931E+05 4.770E-07 138.55 805.94 1091.14
1.82E-03 0.634 2.586E+05 8.010E-07 185.56 1353.38 1832.29
1.82E-03 0.634 4.068E+05 2.580E-06 291.95 4359.19 5901.76
1.82E-03 0.634 5.482E+05 5.560E-06 393.39 9394.23 12718.51
1.82E-03 0.634 6.916E+05 9.340E-06 496.31 15780.95 21365.27
1.82E-03 0.452 6.206E+04 6.700E-08 62.47 113.20 156.26
1.82E-03 0.452 1.413E+05 3.500E-07 142.29 591.36 800.63
1.82E-03 0.452 2.137E+05 9.720E-07 215.16 1642.30 2223.45
1.82E-03 0.452 3.103E+05 2.720E-06 312.33 4595.74 6222.01
1.82E-03 0.452 4.103E+05 5.750E-06 412.98 9715.25 13153.14
1.82E-03 0.452 5.171E+05 9.840E-06 520.56 16625.76 22509.02
fit s t he dat a well (Fig. 2.22) and yields meaning t hat
Not e also (Fig. 2.22) t hat dat a for all t ube sizes over lap ver ifying t he
assumpt ion of a negligible ent r ance effect . Now, may be calculat ed
using t he Rabinowit sch-Mooney equat ion given in t er ms of Eq. [2.24]
and [2.25]:

4Q/(R
3
) D L P Q
w

w
ln(
w
) ln(Constant ) + n ln()
n 0.414
n
d(ln(
w
))
d(ln())
0.414

3n + 1
4n
_

3(.414) + 1
4(.414)
_

,
1.35
148 Chapter 2. Tube Viscometry
Figure 2.22. Shear stress versus apparent wall shear rate of
1.5% CMC at room temperature.
Figure 2.23. Rheogram for 1.5% CMC at room temperature.
100 300 1,000 3,000 10,000 30,000
50
100
200
300
500
1,000
Apparent Wall Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
1.5% CMC
( 4Q/(R
3
))
0 5,000 10,000 15,000 20,000 25,000 30,000
0
100
200
300
400
500
600
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
1.5% CMC
2.12.4 Casson Model: Flow Rate Equation 149
Result s ar e summar ized in Table 2.9 and plot t ed in Fig. 2.23. The power
law model was used t o r epr esent fluid behavior :
wher e = 8.14 Pa s
.414
and = 0.414. Not e, wit h t he power law model,
, as expect ed.
should be checked, t obe cer t ain flow is laminar , using Eq. [2.51]
or Eq. [2.53]. The "wor st case," at m and m
3
/s,
yields m/s and . Wit h , ,
when calculat ed using Eq. [2.53]. Since 558 < 2581, one may conclude
t hat flow is laminar for all flow r at es under consider at ion.
In t his pr oblem it was assumed t he ent r ance effect was negligible.
The act ual ent r ance lengt h can be est imat ed fr om Eq. [2.67]:
Taking t he exper iment al values used in checking for laminar flow yields
Meaning, for t his par t icular case, t he laminar flow velocit y (as descr ibed
by Eq. [2.40]) was 98% fully developed in t he 0.271cm diamet er t ube at
a dist ance of appr oximat ely 6.25cm (appr oximat ely 23 diamet er s) fr om
t he ent r ance. Since t he minimum of t his syst em is 248, it is unlikely
t he ent r ance losses will significant ly influence r esult s.
2.12.4. Casson Model: Flow Rate Equati on
Der ive t he r elat ionship bet ween volumet r ic flow r at e and pr essur e dr op
for a Casson fluid in laminar t ube flow.
The st ar t ing point in t he solut ion is Eq. [2.15] giving t he gener al
flow r at e r elat ionship in t ube flow:
The Casson equat ion, given in Table 1.3, is
[2.147]
which can easily be solved for t he shear r at e:

w
8.14(

w
)
.414
K n
n n
N
Re, PL
Q 3.510(10
5
) D 0.00271
u 6.09 N
Re, PL
558 n 0.414 (N
Re, PL
)
critical
2581
X
E
DN
Re, PL
(.125n + .175) (8
n 1
)

3n + 1
4n
_

,
n
X
E
.00271(558) (.125(.414) + .175) (8
.414 1
)

3(.414) + 1
4(.414)
1
1
]
.414
.0625 m
L/D
Q
R
3

1
(
w
)
3

w
()
2
f () d

.5

o
.5
+ K
1
(

)
.5
150 Chapter 2. Tube Viscometry
[2.148]
Eq. [2.148] may be subst it ut ed int o t he flow r at e equat ion: Recognizing
t hat t he pr esence of t he yield st r ess makes t he flow discont inuous gives
[2.149]
The fir st int egr al t er m in Eq. [2.149] is zer o because in t he cent r al
plug r egion of t he t ube wher e ; t her efor e,
[2.150]
Expansion, int egr at ion, and algebr aic manipulat ion yields t he final
solut ion:
[2.151]
Since , Eq. [2.151] pr ovides t he r elat ionship bet ween
volumet r ic flow r at e and pr essur e dr op for a Casson fluid.
2.12.5. Sli t Vi scometry - Corn Syrup
Dat a (Table 2.10) for high fr uct ose cor n syr up wer e collect ed at 28 C
using a slit viscomet er . Det er mine t he r heological pr oper t ies of t his
mat er ial. Pr essur e dr ops wer e measur ed wit h flush mount ed t r ans-
ducer s in t he r egion of t he slit wher e flow was fully developed.
The appar ent wall shear r at e ( ) and shear st r ess ( )
wer e calculat ed (Table 2.10) and plot t ed (Fig. 2.24) t o find fr om
Eq. [2.89]:

f ()

.5

o
.5
K
1
1
1
]
2
Q
w
3
K
1
2
R
3

2
f () d+

2
(
.5

o
.5
)
2
d
f () 0
0<<
o
Q
w
3
K
1
2
R
3

2
(
.5

o
.5
)
2
d
Q
R
3
K
1
2

w
4

4
o
.5

w
.5
7
+

o
3

o
4

w
4
_

1
84
_

,
_

w
PR/(2L)

6Q/(wh
2
) Ph/(2L)
n 0.97
n
d ln(
w
)
d ln(

w
)
0.97
2.12.5 Slit Viscometry - Corn Syrup 151
Table 2.10. Slit (Fig. 2.14: =0.3375 cm, = 8.89 cm, = 33.02 cm) Flow Data
of High Fructose Corn Syrup at 28 C
(m
3
/ s ) (1/ s ) (Pa / m) (kPa ) (1/ s )
6.16 E-6 44.3 2.28 3.63 44.8
8.16 E-6 54.6 3.01 4.78 55.2
9.65 E-6 64.6 3.55 5.63 65.3
10.0 E-6 73.4 4.02 6.39 74.2
12.2 E-6 81.6 4.39 6.98 82.4
13.8 E-6 92.6 4.86 7.71 93.6
15.1 E-6 101.4 5.30 8.41 102.4
16.2 E-6 108.6 5.58 8.86 109.8
Figure 2.24. Shear stress versus apparent wall shear rate for slit flow data of
high fructose corn syrup at 28 C.
A value of t his close t o 1.0 indicat es t he expect ed Newt onian
behavior . Using Eq. [2.88], t he shear r at e was evaluat ed (Table 2.10).
The r heogr am was plot t ed (Fig. 2.25) and fit wit h a linear equat ion:
h w L

6Q/(wh
2
) Q P/L
w

w
10 20 30 50 70 100
1,000
2,000
3,000
5,000
7,000
10,000
6Q/(wh ), 1/s
2
w
,

P
a
= .97
Corn Syrup
n

n
152 Chapter 2. Tube Viscometry
Showing t he viscosit y of t he sample is 78.7 Pa s. The int er cept , 374.6
Pa, is a mat hemat ical consequence of t he st at ist ical pr ocedur e and
should not be int er pr et ed as an absolut e value of t he yield st r ess.
Figure 2.25. Rheogram for high fructose corn syrup at 28 C.
2.12.6. Fri cti on Losses i n Pumpi ng
Consider t he t ypical flow pr oblem illust r at ed in Fig. 2.26. Assume t he
plug disk valve is open. The syst em has a 0.0348 m diamet er pipe wit h
a volumet r ic flow r at e of 0.00157 m
3
/s (1.97 kg/s) r esult ing in a volu-
met r ic aver age velocit y of 1.66 m/s. The fluid densit y is equal t o 1250
kg/m
3
and t he pr essur e dr op acr oss t he st r ainer is 100 kPa. Det er mine
t he fr ict ion losses in t he syst em, and calculat e t he wor k input and
pr essur e dr op acr oss t he pump for t he following t wo cases involving
power law fluids: Case 1) Assume = 5.2 Pa s
n
and = 0.45; Case 2)
Assume = 0.25 Pa s
n
and = 0.45.
Solving t he mechanical ener gy balance equat ion, Eq. [2.105], for
wor k out put yields

w
374.6 + 78.7

w
40 50 60 70 80 90 100 110
3,000
4,000
5,000
6,000
7,000
8,000
9,000
10,000
Corn Syrup
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a

K n
K n
2.12.6 Friction Losses in Pumping 153
Figure 2.26. Pumping system for a power law fluid.
Subscr ipt s 1 and 2 r efer t o t he fluid level in t he input t ank and t he exit
point of t he syst em, r espect ively. The pr essur e at point s 1 and 2 is equal
t o one at mospher e, t her efor e, . As a wor st case for pumping,
assume a near empt y t ank ( ). Also, assume a lar ge diamet er
input t ank making . These consider at ions simplify t he above
equat ion t o
wher e ( ) r epr esent s t he wor k input per unit mass. The summat ion
t er m, given by Eq. [2.106], includes fr ict ion losses in st r aight pipe, fit -
t ings, and t he st r ainer :
or
Pr essur e dr op acr oss t he pump is
1.0 m 2.0 m 1.0 m
3.0 m
3.0 m
0.5 m
Pump Strainer
Plug Disk Valve
W
( u
2
)
2

( u
1
)
2

1
+ g(z
2
z
1
) +
P
2
P
1

+ F
P
2
P
1
z
2
z
1
2.5
u
1
0
W g(z
2
z
1
) +
( u
2
)
2

2
+ F 9.81(2.5) +
(1.66)
2

2
+ F
W
F
2f ( u )
2
L
D
+
(k
f, entrance
) ( u )
2
2
+
(k
f, valve
) ( u )
2
2
+
3(k
f, elbow
) ( u )
2
2
+
100000
1250
F
2f ( u )
2
L
D
+ (k
f, entrance
+ k
f, valve
+ 3(k
f, elbow
))
( u )
2
2
+ 80.0
154 Chapter 2. Tube Viscometry
The power r equir ement would be found by calculat ing t he pr oduct of
wor k input and mass flow r at e.
Soluti on to Case 1: = 5.2 Pa s
n
and = 0.45. Using Eq. [2.52],
is calculat ed t o be 323.9, a laminar flow value of . Fr ict ion loss
coefficient s ar e det er mined fr om Eq. [2.115], Eq. [2.117], Eq. [2.118],
and Table 2.6 yielding:
and t he fr ict ion fact or is calculat ed fr om Eq. [2.113]:
Using t he above values, t he t ot al fr ict ion loss is
The kinet ic ener gy cor r ect ion fact or is found fr om t he laminar flow
equat ion for power law fluids (Table 2.4):
Wit h t his infor mat ion, t he calculat ions for wor k input and pr essur e dr op
can be complet ed:
and
(P)
p
(W)
K n N
Re, PL
N
Re, PL
k
f, entrance

(.5)500
323.9
0.77
k
f, valve

(9) (500)
323.9
13.89
k
f, elbow

(.45) (500)
323.9
.69
f
16
323.9
.0494
F
2(.0494) (1.66)
2
(10.5)
.0348
+ (0.77 + 13.89 + 3(.69))
(1.66)
2
2
+ 80.0 185.2 J/kg

2(2n + 1) (5n + 3)
3(3n + 1)
2

2(2(.45) + 1) (5(.45) + 3)
3(3(.45) + 1)
2
1.2
W 9.81(2.5) +
(1.66)
2
1.2
+ 185.2 212.0 J/kg
(P)
p
(212.0) (1250) 265 kPa
2.12.7 Turbulent Flow - Newtonian Fluid 155
Soluti on to Case 2: = 0.25 Pa s
n
and = 0.45. Using Eq. [2.52],
is calculat ed as 6,737. The cr it ical value of , det er mined fr om
Eq. [2.51], is
meaning t he flow is t ur bulent since 6,737 > 2,394. Fr ict ion loss coeffi-
cient s may be det er mined fr om Eq. [2.115], and Table 2.6: ;
; . The fr ict ion fact or is found by it er at ion of
(equat ion fr om Table 2.4)
yielding = 0.0052. Then,
In t ur bulent flow, t he kinet ic ener gy cor r ect ion fact or is equal t o 2. Then,
t he wor k input and pr essur e dr op ar e calculat ed as
and
2.12.7. Turbulent Flow - Newtoni an Flui d
A common pr oblem facing food pr ocess engineer s is t o pr edict t he
maximum velocit y found dur ing flow in t ubes. Det er mine, given t he
following dat a, t he maximum velocit y in t he pipe: Pa s;
m; m/s; kg/m
3
. Also, calculat e t he velocit y at
a point halfway bet ween t he cent er -line and t he wall of t he pipe.
The Reynolds number for a Newt onian fluid is calculat ed as
which is sufficient t o conclude t hat flow is in t he t ur bulent r egime. The
fr ict ion fact or equat ion for Newt onian fluids in t ur bulent flow (see Table
2.3) is:
K n
N
Re, PL
N
Re, PL
(N
Re, PL
)
critical

6464(.45) (2 + .45)
(2 + .45)/(1 + .45)
(1 + 3(.45))
2
2, 394.
k
f, entrance
0.55
k
f, valve
9 k
f, elbow
0.45
1

4
(.45)
0.75
_

,
log
10
[(6736.6)f
(1 (.45/2))
]

0.4
(.45)
1.2
_

,
f
F
2(.0052) (1.66)
2
(10.5)
.0348
+ (.5 + 9 + 3(.45))
(1.66)
2
2
+ 80.0 103.6 J/kg
W 9.81(2.5) +
(1.66)
2
2
+ 103.6 129.5 J/kg
(P)
p
(129.5) (1250) 162 kPa
0.012
D 0.0348 u 1.66 1250
N
Re

Du


1250(.0348)1.66
0.012
6017.5
156 Chapter 2. Tube Viscometry
Solving for gives
The velocit y is maximum at t he cent er -line wher e and . It
must be calculat ed fr om t he fr ict ion velocit y, Eq. [2.127]:
Calculat ions pr oceed using Eq. [2.126] and [2.124]:
and
The maximum velocit y may be found fr om t he definit ion (Eq. [2.125])
of t he t ur bulent velocit y, , as
Wit h t his infor mat ion, velocit ies at ot her locat ions can alsobe est imat ed.
The 1/7 power law equat ion (Eq. [2.129]) may be used t o appr oximat e
at t he velocit y halfway bet ween t he cent er -line and t he wall
( ):
giving
2.12.8. Turbulent Flow - Power Law Flui d
Det er mine t he maximum velocit y of a power law fluid in a pipe given
t he following infor mat ion: Pa s
n
; m; m/s;
kg/m
3
; .
The Reynolds number (Eq. [2.52]) is
1

f
4.0log
10
(N
Re

f ) 0.4 4.0log
10
((6017.5)

f ) 0.4
f
f 0.0089
r 0 y R
u
*
u

f
2
1.66

.0089
2
.1107 m/s
y
+

yu
*


(.0348/2) (.1107)1250
.012
200.64
u
+
5.5 + 5.756log
10
( y
+
) 5.5 + 5.756log
10
(200.64) 18.753
u
+
u/u
*
u
max
u
+
u
*
(18.753).1107 2.075 m/s
u/u
max
r (.5)R
u
u
max

y
R
_

,
1/ 7

R r
R
_

,
1/ 7
(.5)
1/ 7
.906
u (2.075) (.906) 1.879 m/s
K 0.25 D 0.0348 u 1.66
1250 n 0.45
N
Re, PL

D
n
( u )
2 n

8
n 1
K
_

4n
3n + 1
_

,
n

.0348
.45
(1.66)
2 .45
1250
8
.45 1
(.25)
_

4(.45)
3(.45) + 1
_

,
.45
6736.6
2.12.8 Turbulent Flow - Power Law Fluid 157
allowing t o be calculat ed (equat ion pr esent ed in Table 2.3) fr om:
yielding
Velocit y is maximum at t he cent er -line wher e . The fr ict ion
velocit y (Eq. [2.127]) is
is calculat ed using Eq. [2.132] as:
Then, is det er mined fr om Eq. [2.131]:
The maximum velocit y is found fr om t he definit ion (Eq. [2.125]) of t he
t ur bulent velocit y ( ) as
f
1

4
n
0.75
_

,
log
10
[(N
Re, PL
)f
(1 (n/2))
]

0.4
n
1.2
_

4
(.45)
0.75
_

,
log
10
[(6736.6)f
(1 (.45/2))
]

0.4
(.45)
1.2
_

,
f 0.0052
y R
u
*
u

f
2
1.66

.0052
2
.0846 m/s
y
+
y
+
y
n
(u
*
)
2 n
/K (.0348/2)
.45
(.0846)
2 .45
(1250)/.25 17.6
u
+
u
+

5.66
(.45)
.75
log
10
(17.6)
0.566
(.45)
1.2
+
3.475
(.45)
.75

1.960 + 0.815(.45) 1.628(.45) log


10

3 +
1
(.45)
_

,
1
1
]
22.75
u
+
u/u
*
u
max
u
+
u
*
(22.64).0846 1.92 m/s
Chapt e r 3 . Rot at ional Vis c ome t ry
3.1. Introducti on
Tr adit ional r ot at ional viscomet er s include cone and plat e, par allel
plat e, and concent r ic cylinder unit s oper at ed under st eady shear con-
dit ions Fig. 1.1). They may also be capable of oper at ing in an oscillat or y
mode which will be consider ed in t he discussion of viscoelast icit y,
Chapt er 5. Cone and plat e syst ems ar e somet imes capable of det er -
mining nor mal st r ess differ ences. Concent r ic cylinder syst ems have
been used in r esear ch t o evaluat e t hese differ ences (Padden and DeWit t ,
1954); however , commer cial inst r ument s of t his t ype ar e not available.
Mixer viscomet r y, a "less t r adit ional" met hod in r ot at ional viscomet r y,
is also pr esent ed because it has excellent ut ilit y in solving many r heo-
logical pr oblems found in t he food indust r y.
3.2. Concentri c Cyli nder Vi scometry
3.2.1. Deri vati on of the Basi c Equati on
The concent r ic cylinder viscomet er is a ver y common inst r ument
t hat will oper at e in a moder at e shear r at e r ange making it a good choice
for collect ing dat a used in many engineer ing calculat ions. A number of
assumpt ions ar e made in developing t he mat hemat ical r elat ionships
descr ibing inst r ument per for mance: flow is laminar and st eady, end
effect s ar e negligible, t est fluid is incompr essible, pr oper t ies ar e not a
funct ion of pr essur e, t emper at ur e is const ant , t her e is no slip at t he
walls of t he inst r ument , and r adial and axial velocit y component s ar e
zer o. The der ivat ion pr esent ed her e is based on a physical set up known
as t he Sear le syst em wher e t he bob r ot at es and t he cup is st at ionar y:
It is also applicable t o a Couet t e-t ype syst em in which t he cup r ot at es
and t he bob is st at ionar y. Most concent r ic cylinder viscomet er s ar e
Sear le-t ype syst ems. Unfor t unat ely, it is not uncommon for t he wor d
"Couet t e" t o be used in r efer r ing t o any concent r ic cylinder syst em.
When t he bob r ot at es at a const ant speed and t he cup is st at ionar y
(Fig. 3.1), t he inst r ument measur es t he t or que ( ) r equir ed t o maint ain
a const ant angular velocit y of t he bob ( ). The opposing t or que comes
fr om t he shear st r ess exer t ed on t he bob by t he fluid. A for ce balance
yields
[3.1]
M

M 2rhr 2hr
2

3.2.1 Derivation of the Basic Equation 159


wher e is any locat ion in t he fluid, . Solving Eq. [3.1] for t he
shear st r ess shows t hat decr eases in moving fr om t he bob t o t he cup:
[3.2]
Ut ilizing Eq. [3.2], t he shear st r ess at t he bob ( ) can be defined as
[3.3]
Figure 3.1. Typical concentric cylinder testing apparatus (based in DIN 53018)
showing a bob with recessed top and bottom to minimize end effects.
To det er mine shear r at e, consider t he linear velocit y at in t er ms
of t he angular velocit y ( ) at :
[3.4]
The der ivat ive of t he velocit y wit h r espect t o t he r adius is
[3.5]
r R
b
r R
c

f (r)
M
2hr
2
r R
b

b

M
2hR
b
2
h
R
b
R
c
r
r
u r
du
dr

r d
dr
+
160 Chapter 3. Rotational Viscometry
Since is r elat ed t o t he r ot at ion of t he ent ir e body, it does not r elat e t o
int er nal shear ing; t her efor e, Eq. [3.5] can be wr it t en as
[3.6]
Using t he definit ion of shear r at e developed in Eq. [2.9], may be defined
in t er ms of :
[3.7]
To r elat e angular velocit y t o shear st r ess, not e t hat t or que is const ant
wit h st eady flow so an expr ession for may be det er mined fr om Eq.
[3.2]:
[3.8]
Differ ent iat ing Eq. [3.8] wit h r espect t o t he shear st r ess yields
[3.9]
Subst it ut ing t he value of t or que defined by Eq. [3.1] int o Eq. [3.9] gives
[3.10]
or , wit h simplificat ion,
[3.11]
The shear r at e is some funct ion of t he shear st r ess, hence,
[3.12]
Solving Eq. [3.12] for t he differ ent ial of t he angular velocit y yields
[3.13]
which can be expr essed in t er ms of by subst it ut ing Eq. [3.11] int o
Eq. [3.13]:
[3.14]

du
dr

r d
dr


du
dr

r d
dr
r
r

M
2h
_

,
1/ 2

M
2h
_

,
1/ 2
()
1/ 2
dr
d

M
2h
_

,
1/ 2

1
2
_

,
()
3/ 2
dr
d

2hr
2

2h
_

,
1/ 2

1
2
_

,
()
3/ 2

r
2
dr
r

d
2

r
d
dr
f ()
d
dr
r
f ()
d
d
1
2
f ()
d

3.2.1 Derivation of the Basic Equation 161


Int egr at ing Eq. [3.14] over t he fluid pr esent in t he annulus r esult s in a
gener al expr ession for t he angular velocit y of t he bob ( ) as a funct ion
of t he shear st r ess in t he gap:
[3.15]
Not e t hat t he limit s of int egr at ion ar e an expr ession of t he no slip
boundar y condit ion assumed in t he der ivat ion: Angular velocit y is zer o
at t he cup (t he st at ionar y sur face), and equal t o at t he bob (t he moving
sur face). The left hand side of Eq. [3.15] is easily int egr at ed r esult ing
in t he following equat ion r elat ing angular velocit y t o shear st r ess:
[3.16]
The solut ion of Eq. [3.16] depends on which is dict at ed by t he
behavior of t he fluid in quest ion. It can be solved dir ect ly if t he funct ional
r elat ionship bet ween shear st r ess and shear r at e is known. Eq. [3.15]
is used as t he st ar t ing point in Example Pr oblem 3.8.5 t ofind t he velocit y
pr ofile of a power law fluid in a concent r ic cylinder syst em.
Eq. [3.15] r eflect s a gener al solut ion for concent r ic cylinder vis-
comet er s because t he limit s of t he int egr al could be easily changed t o
t he case wher e t he bob is st at ionar y and t he cup r ot at es (t or que is equal
in magnit ude, but opposit e in sign if measur ed on t he cup) or even a
sit uat ion wher e t he bob and cup ar e bot h r ot at ing. It is impor t ant t o
r ecognize t he fact t hat Eq. [3.16] is analogous t o t he gener al solut ion
(Eq. [2.15]) developed for t ube viscomet er s. Bot h pr ovide an over all
st ar t ing point in developing mat hemat ical r elat ionships for specific
t ypes of fluids.
Appli cati on to Newtoni an Flui ds. The r elat ionship bet ween shear
st r ess and shear r at e for a Newt onian fluid is, by definit ion,
[3.17]
Subst it ut ing t his int o t he gener al expr ession for given by Eq. [3.16]
yields
[3.18]


0
d
1
2

c
f ()
d


1
2

c
f ()
d

f ()

f ()


1
2

c
f ()
d


1
2

,
d


1
2

c
d
162 Chapter 3. Rotational Viscometry
t hen,
[3.19]
Using Eq. [3.2] for shear st r ess allows Eq. [3.19] t o be wr it t en in t er ms
of t he syst em geomet r y and t he t or que r esponse of t he inst r ument :
[3.20]
Rear r angement gives a simplified expr ession, called t he Mar gules
equat ion, descr ibing t he behavior of a Newt onian fluid in a concent r ic
cylinder syst em:
[3.21]
This equat ion clear ly indicat es t hat exper iment al dat a for Newt onian
fluids will show t or que t o be dir ect ly pr opor t ional t o bob speed.
Appli cati on to Power Law Flui ds. Wit h a power law fluid, t he
r elat ionship bet ween shear st r ess and shear r at e is
[3.22]
which can be subst it ut ed int o Eq. [3.16] yielding
[3.23]
or , aft er int egr at ion,
[3.24]
Using Eq. [3.2], an alt er nat ive expr ession for t he power law fluid is
obt ained:
[3.25]
Eq. [3.25] r eveals t hat t or que is not dir ect ly pr opor t ional t o bob speed
because it is st r ongly influenced by t he flow behavior index.
Appli cati on to Bi ngham Plasti c Flui ds. A Bingham plast ic fluid
has t he following r elat ionship bet ween shear st r ess and shear r at e:

1
2
(
b

c
)

1
2

M
2hR
b
2

M
2hR
c
2
1
1
]

M
4h

1
R
b
2

1
R
c
2
1
1
]

f ()

K
_

,
1/ n

1
2

c
f ()
d


1
2

K
_

,
1/ n
d


n
2K
1/ n
[(
b
)
1/ n
(
c
)
1/ n
]

n
2K
1/ n

M
2hR
b
2
_

,
1/ n

M
2hR
c
2
_

,
1/ n
1
1
]

n
2K
1/ n

M
2hR
b
2
_

,
1/ n

R
b
R
c
_

,
2/ n
1
1
]
3.2.2 Shear Rate Calculations 163
[3.26]
Subst it ut ing Eq. [3.26] int o Eq. [3.16] yields
[3.27]
Int egr at ion and subst it ut ion of Eq. [3.2] pr ovides t he gener al r elat ion-
ship (known as t he Reiner -Riwlin equat ion) bet ween t he t or que, angular
velocit y, and syst em geomet r y:
[3.28]
This equat ion is valid only when t he yield st r ess is exceeded at all point s
in t he fluid meaning t hat t he minimum shear st r ess must gr eat er t han
t he yield st r ess:
[3.29]
is t he minimum t or que r equir ed t o over come t he yield st r ess. If
evaluat ing fluid behavior near t he limit s descr ibed by Eq. [3.29], t he
yield st r ess should be det er mined befor e conduct ing st andar d t est s in
Sear le-t ype concent r ic cylinder viscomet er s. This can be accomplished
wit h var ious t echniques, such as t he vane met hod discussed in Sec.
3.7.3, Wit h t hat dat a, one can calculat e t he minimum r ot at ional speed
of t he bob r equir ed t o insur e shear ing t hr oughout t he cylindr ical gap
(see Example Pr oblem 3.8.1). In Couet t e syst ems, applying sufficient
t or que t o r ot at e t he cup assur es shear flow in t he ent ir e annulus because
t he minimum shear st r ess occur s at .
3.2.2. Shear Rate Calculati ons
Numer ous met hods of est imat ing shear r at es in concent r ic cylinder
viscomet er s have been pr oposed (Table 3.1) and many of t hose t ech-
niques ar e discussed in books by var ious aut hor s: Whor low (1992), Van
Wazer et al. (1963), and Dar by (1976). A few of t he mor e pr act ical
appr oximat ion t echniques ar e summar ized in t he cur r ent wor k.

f ()
(
o
)

pl

1
2

c
f ()
d


1
2

c
(
o
)

pl
d

M
4h
pl
1
1
]

1
R
b
2

1
R
c
2
1
1
]

pl
ln

R
c
R
b
_

min

M
min
2R
c
2
h
>
o
M
min
R
c
164 Chapter 3. Rotational Viscometry
Table 3.1. Some Mathematical Techniques for Analyzing Data from Concentric
Cylinder Viscometers
Solu t ion Sou r ce
Sh ea r r a t e ca lcu la t ion , mu lt iple bobs . Kr ieger , I.M. a n d S.H. Ma r on (1954)
Mea n s h ea r s t r es s , s ma ll ga ps . Moon ey, M. (1931)
Sh ea r r a t e a t t h e bob. Kr ieger , I.M. a n d H. Elr od (1953)
Flu ids followin g Ellis equ a t ion . Va n Wa zer et a l. (1963)
Flu ids followin g Her s ch el-Bu lkley equ a t ion . Va n Wa zer et a l. (1963)
Flu ids followin g Eyr in g equ a t ion . Va n Wa zer et a l. (1963)
Flu ids followin g Ca s s on equ a t ion . Mu r a t a , T. a n d S. Oka (1968)
Flu ids followin g Voca dlo equ a t ion . Pa r zon ka , W. a n d J . Voca dlo (1968)
The er r or in using appr oximat ions can be calculat ed by evaluat ing
t he exact ver sus t he appr oximat e solut ion. Accur acy is impr oved as t he
appr oximat ion becomes mor e sophist icat ed. When st udying fluid foods,
t he simple shear , Newt onian or power law appr oximat ions ar e oft en
adequat e.
Si mple Shear Approxi mati on. Wit h a ver y nar r ow annulus
( ), t he cur vat ur e of t he walls is negligible and t he syst em
appr oaches simple shear . Assuming a unifor m shear r at e acr oss t he
gap gives
[3.30]
wher e . When calculat ing shear r at es wit h Eq. [3.30] a
cor r esponding aver age shear st r ess should be used:
[3.31]
The er r or involved in using Eq. [3.30] for power law fluids is invest igat ed
in Example Pr oblem 3.8.2.
Newtoni an Approxi mati on. The shear r at e at t he bob for a New-
t onian fluid is det er mined fr om t he definit ion of a Newt onian fluid as
[3.32]
R
c
R
b
R
b

b

R
b
R
c
R
b


1
R
c
/R
b

a

1
2
(
c
+
b
)
1
2

M
2hR
c
2
+
M
2hR
b
2
1
1
]

M(1 +
2
)
4hR
c
2

b
(2)

2
1
_

,
3.2.2 Shear Rate Calculations 165
Commer cial viscomet er s fr equent ly use t his equat ion t o appr oximat e
shear r at e, oft en in t he for m of a r epr esent at ive (aver age) shear r at e.
The der ivat ion of Eq. [3.32] is pr esent ed in Example Pr oblem 3.8.3 and
t he r epr esent at ive shear r at e concept , developed using t his equat ion, is
examined in Example Pr oblem 3.8.4.
Power Law Approxi mati on. Consider ing t he definit ion of a power
law fluid, an equat ion for t he shear r at e at t he bob can be der ived as
[3.33]
The complet e development of Eq. [3.33] is given in Example Pr oblem
3.8.5.
Using Eq. [3.33] r equir es a numer ical value of t he flow behavior
index. It may be det er mined dir ect ly by consider ing t he power law
equat ion wit h Eq. [3.33] used as t he expr ession for shear r at e:
[3.34]
Taking t he logar it hm of each side, Eq. [3.34] may be wr it t en as
[3.35]
which, by evaluat ing t he der ivat ive wit h r espect t o , pr ovides a
simple expr ession for t he flow behavior index:
[3.36]
Since , Eq. [3.36] may also be wr it t en as
[3.37]
Hence, for power law fluids, is t he slope (a st r aight line) of , or
, ver sus . Once is known, Eq. [3.33] can be easily evaluat ed.
Kri eger Method. The gener al for ce balance on t he bob gave t he
r elat ionship bet ween shear st r ess and shear r at e as (Eq. [3.16])

2
n
_

2/ n

2/ n
1
1
1
]

b
K(

b
)
n

n

K
1/ n

2
n
_

2/ n

2/ n
1
1
1
]
_

,
n
ln(
b
) n ln() + n ln

K
1/ n

2
n
_

2/ n

2/ n
1
1
1
]
_

,
_

,
ln
n
d(ln
b
)
d(ln)

b
M/(2hR
b
2
)
n
d(lnM)
d(ln)
n ln(M)
ln(
b
) ln() n

1
2

b
f ()
d

166 Chapter 3. Rotational Viscometry


Taking t he der ivat ive of t his expr ession wit h r espect t o t he shear st r ess
at t he bob yields (Kr ieger and Mar on, 1952)
[3.38]
Fr om Eq. [3.1]
[3.39]
so
[3.40]
and
[3.41]
Subst it ut ion and simplificat ion, using Eq. [3.40] and [3.41], t r ansfor ms
Eq. [3.38] int o
[3.42]
Asolut ion (a suit able expr ession for shear r at e at t he bob) t oequat ion
Eq. [3.42] r equir es t he evaluat ion of an infinit e ser ies. The best solut ions
ar e t hose which involve a ser ies wher e a good appr oximat ion can be
obt ained by evaluat ing a small number of t er ms. An excellent solut ion
t o Eq. [3.42], one which t r uncat es an infinit e ser ies aft er t he fir st t er m,
was developed by Kr ieger (1968) and r ecommended by Yang and Kr ieger
(1978):
[3.43]
wher e:
[3.44]
[3.45]
and
d
d
b

1
2

f (
b
)

f (
c
)

c
_

d
c
d
b
_

,
1
1
]
M
2h
r
2
constant

c
R
c
2

b
R
b
2
d
c
d
b

R
b
R
c
_

,
2

2
d
d
b

1
2
b

f (
b
)
f (
b
)

2
_

b
f (
b
)

2
s
_

2/ s

2/ s
1
1
1
]
(1 + s
2
sg)
1/s

,
d
d
b

d(ln)
d(ln
b
)

d(ln)
d(lnM)
s
b
d(1/s)
d
b

d(1/s)
d(ln
b
)

d(1/s)
d(lnM)
3.2.2 Shear Rate Calculations 167
[3.46]
wit h , t he ar gument of t he funct ion given by Eq. [3.46], defined as
. In concent r ic cylinder viscomet er s, is t ypically in t he r ange
of 1.01 t o 1.40 wit h smaller values being mor e common. Eq. [3.43] is
accept able over t his r ange of , most accur at e at small values of , and
good in many cases for values up t o 2.0 (Yang and Kr ieger , 1978).
Figure 3.2. Plot of Eq. [3.46], , in Krieger (1968) solution for calculating
shear rate at the bob of a concentric cylinder viscometer.
The Kr ieger solut ion (Eq. [3.43]) is ver y close t o t he power law
appr oximat ion. In fact , wit h a power law fluid, and t he Kr ieger
equat ion becomes t he power law solut ion given by Eq. [3.33]. It is also
impor t ant t o not e t hat t he maximum value of is appr oximat ely 0.1
(Fig. 3.2) and, when calculat ing t he cor r ect ion fact or in Eq. [3.43], t his
number is mult iplied by ot her small number s. Hence, t he power law
expr ession for shear r at e, wit h found at each value of , is an excellent
appr oximat ion for t he shear r at e at t he bob:
g f (x)
x(e
x
(x 2) + x + 2)
2(e
x
1)
2
x
2(ln)/s

0 1 2 3 4 5 6 7
0
0.02
0.04
0.06
0.08
0.1
0.12
x
g
(
x
)
g(x)
s n
g(x)
s
b
168 Chapter 3. Rotational Viscometry
[3.47]
wher e is defined by Eq. [3.44].
Ot her solut ions t o Eq. [3.42] have been summar ized by Yang and
Kr ieger (1978). Dar by (1985) has discussed t he pr oblem of concent r ic
cylinder viscomet er dat a r educt ion for mat er ials wit h a yield st r ess and
found t he power law appr oximat ion t o give accept able r esult s in t he
major it y of pr act ical sit uat ions. Best r esult s, however , ar e obt ained
wit h pr ior knowledge of t he yield st r ess so complet e shear ing in t he
annulus can be assur ed (Nguyen and Boger , 1987). The ut ilit y of var ious
shear r at e appr oximat ion equat ions is illust r at ed in Example Pr oblem
3.8.6. This example also pr esent s t he Kr ieger and Mar on (1952) solut ion
t o Eq. [3.42] for calculat ing t he shear r at e at t he bob in a concent r ic
cylinder syst em.
3.2.3. Fi ni te Bob i n an Infi ni te Cup
When , t he case of a finit e bob in an infinit e cup, is ver y
small so Eq. [3.42] becomes
[3.48]
Solving t his expr ession for t he shear r at e at t he bob and mult iplying
numer at or and denominat or by , t hen simplifying, gives
[3.49]
Eq. [3.49] pr ovides an easy means of det er mining t he shear r at e at t he
bob in an infinit e cup. The solut ion may be useful in t est s involving
ver y lar ge beaker s or indust r ial scale food vat s. It is clear ly valid for
any t ime-independent mat er ial wit hout a yield st r ess. Remar kably, it
is also valid for fluids wit h a yield st r ess (t he yield st r ess pr oblem is
discussed lat t er in t his sect ion).
Shear r at e equat ions developed for concent r ic cylinder syst ems can
also be used for t he infinit e cup case by making t he appr opr iat e
allowance for geomet r y. This idea is illust r at ed for a power law fluid,
t omat o ket chup, in Example Pr oblem 3.8.7. Er r or s involved in using
t he infinit e cup appr oximat ion depend on geomet r y as well as fluid

2
s
_

2/ s

2/ s
1
1
1
]
s
R
c
R
b
1/
d
d
b

f (
b
)
2
b

f (
b
)

b
(2
b
)
d
d
b

2
b

,
d
d
b
(2)
d(ln)
d(ln
b
)
3.3 Cone and Plate Viscometry 169
pr oper t ies (see Example Pr oblem 3.8.8). A gener al pr oblem illust r at ing
t he infinit e cup calculat ion t echnique is given for salad dr essing in
Example Pr oblem 3.8.9.
Eq. [3.49] may also be used t o calculat e an exact solut ion for t he
shear r at e at t he bob in concent r ic cylinder syst ems (Nguyen and Boger ,
1987) when t he fluid in t he annulus is only par t ially shear ed, i.e., when
Eq. [3.29] is violat ed and t he minimum shear st r ess is less t han t he
yield st r ess. In t his sit uat ion, t he yield st r ess ( ) may be subst it ut ed
for in Eq. [3.16], t hen differ ent iat ed t o yield an expr ession similar t o
Eq. [3.38]:
[3.50]
Since t he second t er m wit hin t he br acket s is zer o, Eq. [3.50] becomes
Eq. [3.48] which can be simplified t o Eq. [3.49]. Ther efor e, t he solut ion
for a single cylinder r ot at ing in an infinit e medium and t he solut ion for
t he case of a bob r ot at ing in an annulus wit h par t ially shear ed fluid ar e
ident ical! This sur pr ising r esult means t hat Eq. [3.49] can be used in
a vat (or beaker ) cont aining mat er ial wit h a yield st r ess if t he bob is
placed sufficient ly far int o t he cont ainer so wall effect s ar e not a sour ce
of er r or . The appr opr iat e dist ance fr om t he wall can be est imat ed fr om
Eq. [3.2]. This idea is illust r at ed in Example Pr oblem 3.8.10.
3.3. Cone and Plate Vi scometry
Using a cone and plat e appar at us (Fig. 3.3), t he shear st r ess ver sus
shear r at e cur ve may usually be obt ained dir ect ly so t he calculat ions
ar e quit e simple. The inst r ument is a moder at e shear r at e device which
is inappr opr iat e for fluids wit h lar ge par t icles because t he cone angle
( ) is small, pr efer ably less t han 0.09 r ad (5 degr ees). In oper at ing a
cone and plat e viscomet er , t he apex of t he cone almost t ouches t he plat e
and fluid fills t he gap. The cone is r ot at ed at a known angular velocit y
( ) and t he r esult ing t or que ( ) is measur ed on t he fixed plat e or t hr ough
t he cone. Some inst r ument s ar e designed wit h r ot at ing plat es and fixed
cones.
Flow in a cone and plat e viscomet er can be ver y complex r equir ing
a labor ious solut ion of t he fundament al equat ions of mot ion (Walt er s,
1975). However , when using a small cone angle (less t han 5 degr ees),
sufficient ly low r ot at ional speeds, and wit h no er r or s due t o sur face

c
d
d
b

1
2

f (
b
)

f (
o
)

o
_

d
o
d
b
_

,
1
1
]

M
170 Chapter 3. Rotational Viscometry
t ension effect s at t he fr ee fluid sur face (sur face should be spher ical in
shape wit h a r adius of cur vat ur e equal t o t he cone r adius), t he shear
r at e at may be calculat ed as
[3.51]
indicat ing t hat t he shear r at e is const ant t hr oughout t he gap. This is
one of t he main advant ages of a cone and plat e viscomet er . Wit h t he
small angles found in t ypical fixt ur es, .
To develop an expr ession for shear st r ess, consider t he differ ent ial
t or que on an annular r ing of t hickness dr :
[3.52]
Figure 3.3. Cone and plate (left), and parallel plate (right) sensors.
Eq. [3.52] is int egr at ed over t he r adius t o find t he t ot al t or que r esponse:
[3.53]
Since t he shear r at e is const ant in t he gap, t he shear st r ess is also
const ant in t hat ar ea so . Then, Eq. [3.53] can be simplified t o
[3.54]
hence,
[3.55]
r


r
r tan


tan
tan
dM (2rdr)r
R
CONE ANGLE
R

0
M
dM

0
R
(2r
2
) dr
f (r)
M 2

0
R
r
2
dr

3M
2R
3
3.3 Cone and Plate Viscometry 171
This r esult shows t hat , like , is const ant t hr oughout t he gap. Using
Eq. [3.51] and [3.55], shear r at e and shear st r ess can be easily calculat ed.
By var ying t he angular velocit y, cone angle and cone r adius, a wide
var iet y of condit ions can be t est ed. If a specific model is select ed,
r heological pr oper t ies can be calculat ed dir ect ly. The following equa-
t ion, for example, would apply t o power law fluids:
[3.56]
Figure 3.4. Cone and plate system showing pressure distribution on a plate for
a viscoelastic fluid.
Fluids which have a significant elast ic component will pr oduce a
measur able pr essur e dist r ibut ion in t he dir ect ion per pendicular t o t he
shear field (Fig. 3.4). Some cone and plat e viscomet er s allow mea-
sur ement of t he r esult ing nor mal (axial dir ect ion) for ce on t he cone
making it possible t o calculat e t he fir st nor mal st r ess differ ence, not ed
in Eq. [1.23], as (Walt er s, 1995):
[3.57]

3M
2R
3
K

tan
_

,
n
plate
test fluid
cone
N
1
(
11

22
)
2(F
normal
)
R
2
172 Chapter 3. Rotational Viscometry
The nor mal for ce differ ence incr eases wit h t he shear r at e for viscoelast ic
fluids: It is equal t o zer o for Newt onian fluids. Expect ed t or que r esponse
and dat a analyses for cone and plat e syst ems ar e invest igat ed in
Example Pr oblems 3.8.11 and 3.8.12, r espect ively.
3.4. Parallel Plate Vi scometry (Torsi onal Flow)
Shear r at e in a par allel plat e appar at us (Fig. 3.3) is a funct ion of
[3.58]
so t he shear r at e at t he r im of t he plat e ( ) is
[3.59]
Shear st r ess must be det er mined fr om t he t or que r esponse of t he
inst r ument which is evaluat ed by const r uct ing a for ce balance equat ion
on t he disk and int egr at ing over t he r adius. The same pr ocedur e was
followed in t he pr evious sect ion for t he cone and plat e viscomet er .
Wit h t he cone and plat e syst em, t he shear st r ess was const ant
allowing Eq. [3.53] t o be easily evaluat ed. In a par allel plat e syst em,
however , t he shear st r ess is a funct ion of t he r adius making t he int e-
gr at ion mor e complicat ed. Eq. [3.53] may be wr it t en in a simplified for m
as
[3.60]
Using Eq. [3.58], t he var iable of int egr at ion may be changed fr om t o
. Making t he appr opr iat e subst it ut ions ( and ),
and evaluat ing t he int egr al fr om 0 t o , yields
[3.61]
Dividing each side of Eq. [3.61] by and simplifying t he r esult gives
[3.62]
or , since ,
r

f (r)
r
h

R

R
h

0
M
dM 2

0
R
(r
2
) dr
r
r
2
(

h/)
2

dr (h/)d

0
M
dM 2

,
3

R
(

)
2
d

R
3
M
2R
3

1
(

R
)
3

R
(

)
2
d

f (

)
3.4 Parallel Plate Viscometry (Torsional Flow) 173
[3.63]
Differ ent iat ing Eq. [3.63] wit h r espect t o , using Leibnit z r ule (Eq.
[2.16] and [2.17]) on t he r ight hand side, gives an independent t er m for
:
[3.64]
Simplificat ion of Eq. [3.64] pr ovides an expr ession for t he shear st r ess
at t he r im of t he plat e ( ):
[3.65]
or
[3.66]
wher e, r ecall, . This expr ession is similar in for m t o t he
Rabinowit sch-Mooney equat ion, Eq. [2.20]. Applicat ion of Eq. [3.66] is
demonst r at ed for a 3% hydr oxypr opyl met hylcellulose solut ion in
Example Pr oblem 3.8.13.
The r elat ionship bet ween and can be evaluat ed dir ect ly for
par t icular t ypes of behavior . Wit h a Newt onian fluid
[3.67]
which can be subst it ut ed int o Eq. [3.60]:
[3.68]
yielding, aft er int egr at ion and simplificat ion,
[3.69]
When wr it t en in t his for m, it is clear t hat for Newt onian
fluids. The same pr ocedur e can be followed for power law fluids giving
M
2R
3
(

R
)
3

R
(

)
2
f (

) d

R
f (

R
)
(

R
)
3
d(M/(2R
3
))
d

R
+

M
2R
3
_

,
3(

R
)
2
(

R
)
2
f (

R
)

R
f (

R
)
3M
2R
3
+

R
d(M/(2R
3
))
d

R
f (

R
)
M
2R
3

3 +
d ln(M)
d ln(

R
)
1
1
]

R
R/h
M


r
h

0
M
dM 2

0
R
r
3

h
dr
2M
R
3

R
h

R
2M/(R
3
)
174 Chapter 3. Rotational Viscometry
[3.70]
showing t hat t he shear st r ess at t he r im for a power law fluid depends
on t he numer ical value of t he flow behavior index:
[3.71]
Consider ing t he above r elat ionships, one can see t hat t he der ivat ive
t er m in Eq. [3.66] is equal t o one for Newt onian fluids and for power
law fluids.
Par allel plat e syst ems, using axial t hr ust dat a, can be used t o cal-
culat e t he second nor mal st r ess differ ence (not ed in Eq. [1.24]) pr ovided
t he fir st nor mal st r ess differ ence has been det er mined fr om cone and
plat e measur ement s (Walt er s, 1975):
[3.72]
If is known, t his equat ion can be used t o det er mine . However ,
since is ver y small compar ed t o , it is oft en r easonable t o assume
t hat t he nor mal st r ess det er mined fr om par allel plat e dat a is a good
appr oximat ion of .
3.5. Correcti ons: Concentri c Cyli nder
End Correcti on. It is impor t ant t o account for t he influence of t he
bot t om of t he cylinder on t he t or que r esponse of t he syst em. This sur face
is in cont act wit h t he fluid but not t aken int o account in t he for ce balance
given by Eq. [3.1].
To det er mine t he end cor r ect ion, t or que (or inst r ument scale divi-
sion) is measur ed at a fixed r at e of r ot at ion when t he annulus is filled
t o var ious height s (Fig. 3.5). Result ing dat a ar e plot t ed (Fig. 3.6) as
t or que ver sus t he height of fluid in cont act wit h t he immer sed lengt h
of t he bob. The cur ve should be linear wit h t he slope equal t o t he t or que
r equir ed t o maint ain t he fixed r at e of r ot at ion per unit lengt h of cylinder .
Effect ive height ( ) is det er mined fr om t he int er cept by ext r apolat ing
t o a value of zer o t or que (Fig. 3.6). This t echnique is illust r at ed in
Example Pr oblem 3.8.14 for t he t aper ed bob of a Her cules high-shear
viscomet er .
M(3 + n)
2R
3
K

R
h
_

,
n

R

M(3 + n)
2R
3
n
N
1
N
2

2F
normal
R
2
_

1 +

1
2
_

,
d(lnF
normal
)
d(ln

R
)
1
1
]
N
1
N
2
N
2
N
1
N
1
h
o
3.5 Corrections: Concentric Cylinder 175
Figure 3.5. Illustration of values used in determining end correction.
Figure 3.6. End correction for a concentric cylinder system using a graphical
technique to determine .
Effect ive height values ar e used in t he pr evious equat ions developed
for concent r ic cylinder syst ems. The Mar gules equat ion (Eq. [3.21]), for
example, would be expr essed as
h
h
h
1
2
3
h
-10 0 10 20 30 40 50
T
O
R
Q
U
E
,

S
C
A
L
E

U
N
I
T
S
x
x
x
-h
0
Height of Bob in Contact with Fluid, mm
h
o
176 Chapter 3. Rotational Viscometry
[3.73]
The value of given in Eq. [3.1] is r eplaced by which, t oget her ,
may be t hought of as t he effect ive height of t he bob. An end cor r ect ion
calculat ed for a par t icular bob wit h a st andar d Newt onian fluid pr ovides
a gener al appr oximat ion for To obt ain maximum accur acy, t he end
cor r ect ion should be evaluat ed for each fluid and r ot at ional speed under
consider at ion. This pr ocedur e, however , is ver y labor ious and not
consider ed st andar d pr act ice.
The end cor r ect ion can also be evaluat ed in t er ms of an equivalent
t or que ( ) gener at ed by a fluid in cont act wit h t he bot t om of t he sensor .
This idea is illust r at ed for t hr ee differ ent speeds in Fig. 3.7. can be
plot t ed as a funct ion of t o det er mine t he r elat ionship bet ween t he t wo
par amet er s. The t or que cor r ect ion is subt r act ed fr om t he measur ed
t or que in calculat ing t he shear st r ess at t he bob:
[3.74]
Cor r ect ing for end effect s wit h or should yield ident ical r esult s.
Var ious bob designs have been developed t o minimize end effect s.
Bobs can be made wit h a r eser voir at t he t op and a r ecessed bot t om.
This bob design, shown in Fig. 3.1, is based on a Ger man st andar d (DIN
53018) developed by t he Ger man Inst it ut e for St andar dizat ion
(Deut sches Inst it ut fr Nor mung). End effect pr oblems can also be
r educed by designing t he bot t om wit h a slight angle (called a Mooney-
Couet t e bob, Fig. 3.8) in an effor t t o make t he shear r at e at t he bot t om
equivalent t o t he shear r at e in t he annulus. The pr oper angle ( ) can
be calculat ed by equat ing t he annular shear r at e t o t he shear r at e in
t he gap (see Example Pr oblem 3.8.15). Pr oblems wit h lar ge par t iculat es
and sensor alignment limit t he usefulness of Mooney-Couet t e syst ems.
Ot her met hods, such as using a mer cur y int er face at t he bot t om of t he
bob (Pr incen, 1986), have also been pr oposed.

M
4(h + h
o
)

1
R
b
2

1
R
c
2
1
1
]
h h + h
o
h
o
M
e
M
e

b

M M
e
2hR
b
2
M
e
h
o

3.5 Corrections: Concentric Cylinder 177


Figure 3.7. End correction for a concentric cylinder system using a graphical
technique to determine a torque correction at different speeds.
Figure 3.8. Mooney-Couette bob design.
Vi scous Heati ng. Temper at ur e incr ease in a fluid dur ing r heological
t est ing can be caused by t he viscous gener at ion of heat . It may be a
ser ious pr oblem in some exper iment s because r heological pr oper t ies ar e
st r ongly influenced by t emper at ur e. The pur pose of t his sect ion is t o
pr ovide a means of det er mining if a significant t emper at ur e incr ease
1
2
e
e
Torque Due to Bottom of Bob at
3
M
M
1
h h h
1 2 3
Height of Fluid in Contact with Bob
T
o
r
q
u
e
1
178 Chapter 3. Rotational Viscometry
may occur dur ing t est ing. If t he pr oblem is ser ious (it will always exist )
appr opr iat e act ion must be t aken. Most viscomet er s ar e designed wit h
effect ive t emper at ur e cont r ol syst ems t hat minimize viscous heat ing
pr oblems by r apidly r emoving t he excess heat gener at ed dur ing t est ing.
To addr ess viscous heat ing, t he case of unifor m shear ing bet ween
par allel plat es (Fig. 1.9) may be consider ed (Dealy, 1982). A concent r ic
cylinder syst em can be appr oximat ed using t his idea when t he gap is
nar r ow ( ). This is a one dimensional pr oblem wher e it is
assumed t hat plat es ar e separ at ed by a dist ance , wit h at t he
bot t om plat e and at t he t op plat e. Also, assume t he fluid is
Newt onian wit h a viscosit y t hat does not var y wit h t emper at ur e. In
t his case t he differ ent ial equat ion r elat ing t emper at ur e and locat ion,
under st eady st at e condit ions, is
[3.75]
wher e is t he fluid t her mal conduct ivit y. is t he viscous ener gy
gener at ed per unit t ime per unit volume expr essed in unit s of J s
-1
m
-3
.
Shear r at e is consider ed t o be unifor m t hr oughout t he gap. The solut ion
t o Eq. [3.75] is
[3.76]
wher e and ar e const ant s which depend on t he boundar y condit ions
of t he pr oblem being consider ed. Solut ions for t wo differ ent cases follow.
The var ious scenar ios should be visualized in t er ms of t he pr opensit y
of heat t o move t hr ough t he cup and (or ) t he bob sur faces dur ing
shear ing.
Bot h sur faces ar e maint ained at t he same t emper at ur e ( ). In t his
sit uat ion t he boundar y condit ions ar e at , and at ,
which allow t he const ant s in Eq. [3.76] t o be det er mined: ,
. Subst it ut ing t hese values back int o Eq. [3.76] allows t he
t emper at ur e t o be expr essed as a funct ion of posit ion bet ween t he plat es:
[3.77]
R
c
R
b
R
b
s x
2
0
x
2
s
k
d
2
T
dx
2
2
(

)
2
(

)
2
k
T f (x
2
)
(

)
2
2k
x
2
2
+ C
1
x
2
+ C
2
C
1
C
2
T
o
T T
o
x
2
0 T T
o
x
2
s
C
2
T
o
C
1


()
2
s/(2k)
T f (x
2
) T
o
+


()
2
x
2
2k
(s x
2
)
3.5 Corrections: Concentric Cylinder 179
meaning t emper at ur e dist r ibut ion is par abolic and t he maximum
t emper at ur e occur s at t he midplane:
[3.78]
Hence, t he t emper at ur e r ise in t he gap is equal t o indicat ing
t hat minimizing t he size of t he gap ( ), analogous t o having a
smaller value of , is beneficial in r educing viscous heat ing pr oblems.
One sur face is adiabat ic and t he ot her sur face is maint ained at . In
t his case t he boundar y condit ions ar e at and at
allowing det er minat ion of t he const ant s: and .
Subst it ut ion of t hese values int o Eq. [3.76] yields t he t emper at ur e
dist r ibut ion funct ion:
[3.79]
The dist r ibut ion is par abolic wit h t he maximum t emper at ur e occur r ing
at t he adiabat ic sur face wher e :
[3.80]
Compar ing t his r esult t o t he case wher e bot h sur faces ar e maint ained
at indicat es t he t emper at ur e var iat ion in a sample may be four t imes
gr eat er when one sur face is consider ed adiabat ic. Eq. [3.80] has been
applied t o t omat o ket chup in Example Pr oblem 3.8.16.
Effect of t emper at ur e var iat ion on viscosit y. Pr oblems associat ed wit h
viscous heat ing will depend on t he ext ent t o which r heological pr oper t ies
ar e sensit ive t o t emper at ur e. Dealy (1982) gave an example in which
viscosit y was expr essed as an exponent ial funct ion of t emper at ur e:
[3.81]
or
[3.82]
T
max
f (s/2) T
o
+


()
2
s
2
8k


()
2
s
2
/(8k)
R
c
R
b
s
T
o
T T
o
x
2
0 dT/dx
2
0
C
1


()
2
s/k x
2
s C
2
T
o
T f (x
2
) T
o
+


()
2
x
2
2k
(2s x
2
)
x
2
s
T f (s) T
max
T
o
+


()
2
s
2
2k
T
o
f (T)
o
e
(b(T T
o
)/T
o
)
ln

,
b

T T
o
T
o
_

,
180 Chapter 3. Rotational Viscometry
wher e is t he viscosit y at and is a const ant , numer ically dependent
on t he fluid in quest ion. Assuming is t he wall t emper at ur e, t he r at io
of t he maximum t o t he minimum viscosit y is
[3.83]
Consider ing t he case when bot h walls ar e maint ained at , Eq. [3.78]
can be subst it ut ed int o Eq. [3.83] yielding:
[3.84]
Meaning, for example, if t he var iat ion in viscosit y dur ing t est ing is t o
be less t han 10%, t hen
[3.85]
making it necessar y t o maint ain t he following inequalit y:
[3.86]
In t he sit uat ion wher e one wall is maint ained at and t he ot her wall
is adiabat ic, t he viscosit y r at io may be evaluat ed by combining Eq. [3.83]
and [3.80]:
[3.87]
To maint ain a viscosit y var iat ion of less t han 10%
[3.88]
Alt hough t he above calculat ions ar e not quant it at ively exact for
non-Newt onian fluids, t hey do illust r at e t he r elat ive impor t ance of
differ ent exper iment al var iables. Analyt ical solut ions for power law
fluids in couet t e flow -wher e t he consist ency coefficient is expr essed as
a power ser ies of t emper at ur e and t he flow behavior index is assumed
t o be independent of t emper at ur e- ar e cumber some, but available
(Middleman, 1968).

o
T
o
b
T
o
ln

max

min
_

,
b

T
max
T
o
T
o
_

,
T
o
ln

max

min
_

,
b

T
o
+

()
2
s
2
/(8k) T
o
T
o
_

,

b

()
2
s
2
8kT
o
8(ln(1.10)) .76
b

()
2
s
2
kT
o
< .76
T
o
ln

max

min
_

,

b

()
2
s
2
2kT
o
b

()
2
s
2
kT
o
< 2(ln(1.10)) .19
3.5 Corrections: Concentric Cylinder 181
Wall Effects (Sli p). Wall effect s due t o separ at ion in mult iphase
mat er ials may cause er r or s in concent r ic cylinder syst ems similar t o
t hose discussed for t ube viscomet er s in Sec. 2.5. Oldr oyd (1956) sug-
gest ed t hat slip may be consider ed in t er ms of t he gener al expr ession
for angular velocit y (Eq. [3.16]) by adding a slip velocit y ( ) t hat is a
funct ion of wall shear st r ess at t he bob and t he cup:
[3.89]
In t he absence of slip, t he slip velocit y is zer o and t his equat ion r educes
t o Eq. [3.16]. Using t he met hod of Mooney (1931), it is possible t o cor r ect
for slip in concent r ic cylinder viscomet er s. The met hod r equir es
numer ous bobs because measur ement s ar e r equir ed at differ ent values
of .
A simple slip evaluat ion met hod, r equir ing t wo ser ies of measur e-
ment s in t wo differ ent measur ing set s, t hat have differ ent gap widt hs,
has also been suggest ed (Kiljanski, 1989). Cheng and Par ker (1976)
pr esent ed a met hod of det er mining wall-slip based on t he use of a smoot h
and a r ough bob. They also ur ged caut ion in invest igat ing slip because
fluids exhibit ing t hat phenomenon may have accompanying par t iculat e
behavior which may mask slip and complicat e dat a t r eat ment . Ar elat ed
pr ocedur e was pr oposed by Yoshimur a and Pr udhomme (1988). If slip
is a ser ious pr oblem, mixer viscomet r y should be evaluat ed as an
alt er nat e exper iment al met hod.
The Mooney t echnique was used by Qiu and Rao (1989) t o evaluat e
slip in apple sauce, a t ypical food disper sion of solid par t icles in a liquid.
This wor k r epr esent s one of t he few t hor ough st udies dealing wit h slip
in a food pr oduct . The invest igat or s found t hat t he wall slip cor r ect ion
did not significant ly influence t he flow behavior index (t he aver age value
for applesauce was 0.253), but incr eased t he consist ency coefficient : An
aver age consist ency coefficient equal t o 37.53 Pa s
n
was found for t ypical
applesauce and t he slip cor r ect ion caused t his value t o incr ease by an
aver age of 5% t o a value of 39.40 Pa s
n
. Qui and Rao (1989) also made
a ver y int er est ing obser vat ion when t hey said "Due t o t he for t uit ous
opposit e effect s of cor r ect ion for non-Newt onian behavior (it incr eases
t he magnit udes of shear r at es) and cor r ect ion for slip effect s (it decr eases
t he magnit udes of shear r at es), it appear s t hat for food suspensions
Newt onian shear r at es uncor r ect ed for slip may be closer t o t he shear
u
s

1
2

c
f ()
d

+
(u
s
)
bob
R
b
+
(u
s
)
cup
R
c
R
c
/R
b
182 Chapter 3. Rotational Viscometry
r at es cor r ect ed for bot h non-Newt onian behavior and for wall slip." They
caut ioned t hat t his conject ur e should be ver ified befor e applying it t o
any par t icular pr oduct .
Secondary Flow. Equat ions developed for t he analysis of r heological
dat a assume t hat t he st r eamlines ar e cir cular , i.e., flow is laminar .
When an inner cylinder r ot at es in a concent r ic cylinder syst em, t he fluid
near t he inner sur face t r ies t o move out war d due t o cent r ifugal for ces.
This movement may cr eat e non-st r eamline flow due t o t he pr esence of
"Taylor vor t ices" (G.I. Taylor (1923. Phil. Tr ans. Roy. Soc. (London), Ser .
A 223: 289). Such vor t ices may occur for Newt onian fluids when
(Whor low, 1992)
[3.90]
In a Couet t e t ype syst em wher e t he out er sur face (t he cup) is r ot at ed,
t he iner t ial for ces have a st abilizing effect and flow is laminar at much
higher shear r at es. Consult Lar son (1992) for a det ailed analysis of flow
inst abilit ies in concent r ic cylinder syst ems. Applicat ion of Eq. [3.90] is
pr esent ed in Example Pr oblem 3.8.6 and 3.8.17.
Cavi tati on. The for mat ion and collapse of vapor cavit ies, known as
cavit at ion, may occur in a high shear envir onment when t he r adial
pr essur e dr op is sufficient t o cause par t ial vapor izat ion of t he sample.
By consider ing t he Ber noulli equat ion in t er ms of t he mechanical ener gy
balance (Eq. [2.105]), one finds (Sakiadis, 1984) t hat cavit at ion will occur
when wher e is t he linear velocit y of t he bob or cup,
whichever is gr eat er . The left hand side of t he equat ion ( ) is r elat ed
t o t he pr essur e dr op ( ) acr oss t he gap: . If pr esent , cavi-
t at ion may cause er r oneous t or que r esponses in a concent r ic cylinder
viscomet er . Cavit at ion is not a significant pr oblem in food r heology
because it is usually not pr esent when laminar flow condit ions ar e
maint ained. The cavit at ion pr oblem is examined in Example Pr oblem
3.8.17.
3.6. Correcti ons: Cone and Plate, and Parallel Plate
Sour ces of er r or in cone and plat e, and par allel plat e syst ems ar e
similar and include t he following (Dealy, 1982): viscous heat ing, sec-
ondar y flow, shear r at e nonunifor mit y due t o lar ge angles, edge effect s,
and non-ideal geomet r y (also a pr oblem in concent r ic cylinder syst ems).
R
b
(R
c
R
b
)

> 41.3

R
c
R
c
R
b
u >

2(P
atm
P
vap
)/ u
u
u (2P/)
1/2
P
3.6 Corrections: Cone and Plate, and Parallel Plate 183
It is difficult , however , t o numer ically quant ify t hese pr oblems. Devi-
at ion fr om ideal geomet r y, involving eccent r icit y or incor r ect angles,
may be monit or ed by visual inspect ion. The t emper at ur e incr ease in
cone and plat e syst ems can be est imat ed in t er ms of t he Br inkman
number (Powell, 1988) defined as t he r at e of heat gener at ed by viscous
dissipat ion divided by t he r at e of heat conduct ion t o t he sur face of t he
fluid cont ainment syst em. Viscous heat ing is r ar ely a pr oblem when
t est ing biological mat er ials in cone and plat e, or par allel plat e, syst ems.
Edge Effects. One t ype of edge effect , sample skin for mat ion fr om
dehydr at ion, can be minimized by applying a t hin coat ing of oil on t he
out er sur face of t he sample. Some r heomet er s come equipped wit h a
solvent t r ap t o r educe loss of volat iles and subsequent edge effect s. A
second edge effect , known as edge failur e, may be obser ved wit h t hick
foods. Even at r at her low shear r at es, t he sample may appear t o be
r ecessed at t he cent er but flowing out on t he t op and bot t om sur faces of
t he mat er ial. Typically, a shar p dr op in t or que is obser ved at t he onset
of edge failur e. This pr oblem may be t he gover ning fact or in est ablishing
maximum shear r at es in cone and plat e, and par allel plat e t est ing.
Sli p. Slip cor r ect ion met hodology for par allel plat e syst ems, based on
a compar ison of shear st r ess ver sus shear r at e at differ ent gap set t ings,
has been pr esent ed by Yoshimur a and Pr udhomme (1988). This
analysis is similar t o t hat pr esent ed in Sec. 2.5 for evaluat ing a slip
cor r ect ion fact or for capillar y viscomet er s. In t he absence of slip, plot s
of t or que ( ) ver sus t he appar ent shear r at e at t he r im ( ), det er -
mined using a single plat e (const ant r adius) but differ ent gap height s,
will yield ident ical cur ves. This fact is ver y useful in checking for slip
because dat a set s can be collect ed, on t he same sample, by r unning r at e
sweeps at successively smaller values of .
Truncated Cones. Er r or s associat ed wit h t r uncat ed cones (Fig. 3.9)
may be invest igat ed as a deviat ion fr om ideal geomet r y. For t unat ely,
t he maximum er r or int r oduced by t he t r uncat ion can be easily est i-
mat ed. Tor que in t he conical sect ion is calculat ed fr om Eq. [3.54] using
, t he r adius of t he t r uncat ed por t ion, as t he lower limit of t he int egr al:
M R/h
h
R
T
184 Chapter 3. Rotational Viscometry
Figure 3.9. Ideal and truncated cone design.
[3.91]
The per cent maximum er r or in t or que in t he conical sect ion is
[3.92]
Aft er subst it ut ing t he appr opr iat e expr essions, Eq. [3.55] for t he ideal
t or que and Eq. [3.91] for t he appr oximat e t or que, Eq. [3.92] becomes
[3.93]
or , wit h simplificat ion,
[3.94]
R
R
Ideal Cone
Truncated Cone
T
M 2

R
T
R
r
2
dr (2/3)(R
3
R
T
3
)
% Max. Error

1
approximate torque
ideal torque
1
1
]
100
% Max. Error

1
(2/3)(R
3
R
T
3
)
(2/3)(R
3
)
1
1
]
100
% Max. Error

1
R
3
R
T
3
R
3
1
1
]
100
3.7 Mixer Viscometry 185
Using Eq. [3.94], t he maximum er r or can be calculat ed. If, for example,
is or , t hen t he per cent maximum er r or is 0.1 and 0.8,
r espect ively. In act ual oper at ion, t he t r ue er r or is less t han t he pr edict ed
value because t he t r uncat ed sect ion of t he cone also cont r ibut es some
t or que dur ing measur ement .
3.7. Mi xer Vi scometry
Ext ensive wor k has been conduct ed on mixer viscomet r y (Cast ell-
Per ez and St effe, 1992). The t echnique has been used most ly for non-
r eact ing biological mat er ials but also applied in evaluat ing t he
wor kabilit y of fr esh concr et e (Tat t er sall and Banfill, 1983), and t o
chemor heological st udies involving st ar ch gelat inizat ion (Dolan and
St effe, 1990; St effe et al., 1989). Mixer viscomet r y may be useful t o t he
food engineer in evaluat ing difficult fluids like t hose exhibit ing slip or
t ime-dependent behavior , and t hose having lar ge par t icles or par t icle
set t ling pr oblems. Some concept s, such as t he mat ching viscosit y
met hod of det er mining t he mixer viscomet er const ant , ar e also useful
in developing models t o simulat e t he shear hist or y found in complex
food pr ocessing equipment such as scr ape-sur face heat exchanger s.
An unusual t ype of mixer viscomet er , known as t he helical scr ew
r heomet er , was pr oposed by Kr aynik et al. (1984) and successfully used
for t omat o pr oduct s (Tamur a et al., 1989). The inst r ument consist s of
a helical scr ew in a t ight fit t ing bar r el and r esembles a single scr ew
ext r uder wit h a closed dischar ge. Scr ew r ot at ion maint ains par t icle
suspension, and r heological pr oper t ies ar e cor r elat ed t o pr essur e dr op
over t he lengt h of t he scr ew. The helical scr ew r heomet er is not com-
mer cially available but has pot ent ial for fut ur e applicat ions in on-line
viscomet r y.
In past year s, significant advances have been made in our under -
st anding of commer cial mixing and r elat ed pr ocesses (Har nby et al.,
1985; Holland and Chapman, 1966; Nagat a, 1975; Oldshue, 1983;
Silvest er , 1985; Skelland, 1983; Sweeney, 1978; Ulbr echt and Pat t er son,
1985; Uhl and Gar y, 1986). Or iginal ideas in mixer viscomet r y came
fr om concept s developed t oaddr ess indust r ial pr oblems soa shor t r eview
of commer cial mixing is appr opr iat e befor e examining mixer viscomet r y.
Commerci al Mi xi ng. The t er ms mixing, blending and disper sing ar e
somet imes used int er changeably. Mixing can be defined as a unit
oper at ion which involves t he int er mingling of t wo or mor e dissimilar
R
T
0.1R 0.2R
186 Chapter 3. Rotational Viscometry
mat er ials t o obt ain a desir ed degr ee of unifor mit y. It is usually
accomplished by mechanical agit at ion which cr eat es mot ion in t he
mat er ial being pr ocessed.
In mixing, an agit at or induces mat er ial flow by impar t ing iner t ial
for ces t o t he fluid which, if not significant ly dampened by viscous for ces,
cause fluid mot ion at some dist ance away fr om t he impeller . When
mat er ials ar e so t hick t hat fluid cannot be convect ed away fr om t he
st ir r er , mixing is accomplished by t he bulk movement of mat er ial due
t o physical displacement by t he agit at or .
Mixer impeller s for low viscosit y fluids may be divided int o axial and
r adial flow t ypes. Wit h axial flow impeller s, such as t he mar ine t ype
and fixed blade t ur bines, t op-t o-bot t om mot ion is pr omot ed by placing
blades at an angle of less t han 90 degr ees wit h t he angle of r ot at ion.
Radial flow impeller s, including flat and cur ved blade t ur bines, have
blades which ar e mount ed par allel t o t he ver t ical axis of t he dr ive shaft .
Anchor t ype agit at or s (Fig. 1.26) ar e commonly used for high viscosit y
mat er ials. They may have a close-clear ance bet ween t he impeller and
t he t ank t o enhance heat t r ansfer . Ext r emely high viscosit y mat er ials
including past es, dough, and meat emulsions r ely on helical r ibbons,
scr ews or kneader s for mixing.
Di mensi onal Analysi s. Mixing is a complex pr ocess t hat does not
lend it self t o r igor ous analyt ical t r eat ment . St udies on t he subject , and
most pr act ical r esult s, have come fr om dimensional analysis and sim-
ilar it y t heor y (Gupt a, 1984; Langhar r , 1980; Mur phy, 1950). The most
common t echnique involves t he Buckingham pi t heor em discussed
below.
If physical var iables influence a pr ocess, t hen t he basic equat ion
r elat ing t he var iables may be wr it t en as
[3.95]
The Buckingham pi t heor em st at es t hat a r elat ionship may be found
bet ween independent dimensionless gr oups of var iables, called pi
gr oups, which have fewer t er ms t han t he basic equat ions:
[3.96]
wher e wit h defined as
[3.97]
The pi gr oups ar e
m
f
1
(x
1
, x
2
, x
3
, x
4
, , x
m
) 0
f
2
(
1
,
2
,
3
, ,
i
) 0
i < m i
i m j
3.7 Mixer Viscometry 187
[3.98]
[3.99]
[3.100]
wher e , , and ar e const ant s det er mined fr om t he
dimensions of t he physical var iables. In most cases, is t he number of
fundament al dimensions (such as lengt h, mass and t ime) involved in
t he pr oblem. Eq. [3.96] is det er mined fr om exper iment al dat a. Reducing
t he number of var iables by dimensional analysis has an obvious
advant age of r educing t he amount of dat a r equir ed t o model mixing
syst ems.
Power Consumpti on i n Flui d Mi xi ng. In fluid mixing, dimensional
analysis shows t hat t he power number is a funct ion of many dimen-
sionless var iables:
[3.101]
wher e:
[3.102]
[3.103]
[3.104]
[3.105]
[3.106]
The power is equal t ot he pr oduct of t or que and angular velocit y: .
and ar e t he viscosit y funct ion (Eq. [1.22]) and t he fir st nor mal st r ess
coefficient (Eq. [1.23]), r espect ively. Numer ous geomet r ical dimen-
sionless number s may be consider ed for a t ypical mixer (Fig. 3.10): ,
, , , , , widt h of baffle divided by t he impeller
diamet er , number of impeller blades, impeller pit ch, and number of

1
x
1
a
1
x
2
a
2
x
3
a
3
x
m
a
m

2
x
1
b
1
x
2
b
2
x
3
b
3
x
m
b
m

3
x
1
c
1
x
2
c
2
x
3
c
3
x
m
c
m
a
1
. a
m
b
1
. b
m
c
1
. c
m
j
N
Po
f (N
Re, I
, N
Fr
, N
We
, N
Wi
, Geometric Dimensionless Numbers)
N
Po

P

3
d
5
N
Re, I

d
2

N
Fr


2
d
g
N
We


2
d
3

st
N
Wi

P M

1
d/D
(Z
1
+ Z
2
+ h)/d Z
1
/d Z
2
/d h/d W/d
188 Chapter 3. Rotational Viscometry
baffles. To be consist ent wit h t he pr evious wor k on r ot at ional visco-
met r y, t he angular speed is given in unit s of r ad/s inst ead of r ev/s or
r evs/min which ar e fr equent ly used in invest igat ing commer cial mixing.
Figure 3.10. Typical mixer viscometer apparatus with paddle type impeller and
water jacket for temperature control.
In geomet r ically similar syst ems, Eq. [3.101] may be simplified t o
[3.107]
wher e t he power number ( ) r epr esent s t he r at io of t he applied for ce
t o t he opposing iner t ial for ce. The impeller Reynolds number ( )
r epr esent s t he r at io of t he iner t ial t o t he opposing viscous for ce in t he
mixing syst em and defines laminar , t r ansit ional, and t ur bulent flow in
t he mixing vessel. It is gener ally accept ed t hat laminar flow exist s for
and t ur bulent flow is assur ed at . [Not e: if is
for mulat ed wit h speed expr essed as r ev/s inst ead of r ad/s, laminar and
t ur bulent flows ar e found at and , r espect ively.
Differ ent unit s, r ev/s ver sus r ad/s, cause t he number s t o be differ ent by
h
D
Z
d
W
1
Z
2
water jacket
N
Po
f (N
Re, I
, N
Fr
, N
We
, N
Wi
)
N
Po
N
Re, I
N
Re, I
< 63 N
Re, I
> 63, 000 N
Re, I
N
Re, I
< 10 N
Re, I
> 10, 000
3.7 Mixer Viscometry 189
a fact or or .] The t r ansit ion r egion is lar ge and depends on t he
par t icular syst em in quest ion. In some cases, t ur bulent flow may be
pr esent wit h an impeller Reynolds number as low as 1900.
The Fr oude number ( ) r eflect s t he r at io of iner t ial t o gr avit at ional
for ces and is used t o account for t he effect of vor t exing on t he power
number . It may be impor t ant in unbaffled syst ems oper at ing at high
impeller Reynolds number s but can be ignor ed at low speed, and in most
baffled syst ems. The Weber number ( ) r epr esent s t he r at io of t he
iner t ial for ce t o sur face t ension for ce. It should be consider ed in syst ems
wher e int er facial effect s ar e impor t ant such as t hose found in t wo phase
disper sions. Viscoelast ic behavior is char act er ized by t he Weissenber g
number ( ) defined as t he r at io of t he pr imar y nor mal st r ess coefficient
(Eq. [1.23]) t imes t he angular velocit y, divided by t he appar ent viscosit y
funct ion (Eq. [1.22]). The influence of liquid elast icit y on power con-
sumpt ion is not clear but t hought t o be small (Ulbr echt and Car r eau,
1985).
Wit h a single phase fluid oper at ing at low speed, or in a baffled
syst em, t he power number can be expr essed wit h t he impeller Reynolds
number alone:
[3.108]
or
[3.109]
The gener al funct ional r elat ionship may be st at ed as
[3.110]
wher e and depend on t he geomet r y of t he syst em and t he flow r egime
pr esent dur ing mixing. when , and when
. The values of and in t he int er mediat e r egion will
depend on t he par t icular mixing syst em under consider at ion.
Holland and Chapman (1966) pr ovide power ver sus Reynolds
number solut ions for a wide r ange of mixing syst ems. Relat ionships for
some st andar d mixer s ar e pr ovided in Sakiadis (1984). Power cur ves
ar e independent of scale and depend only on t he geomet r y of t he syst em.
When t he power cur ve is available for a par t icular configur at ion, it may
2
N
Fr
N
We
N
Wi
N
Po
f (N
Re, I
)
P

3
d
5
f

d
2

,
P

3
d
5
A

d
2

,
B
A B
B 1 N
Re, I
< 63 B 0
N
Re, I
> 63, 000 A B
190 Chapter 3. Rotational Viscometry
be used t o calculat e t he power r equir ement given var ious agit at or
speeds, liquid viscosit ies, and densit ies. Ot her r efer ences (Ulbr echt and
Car r eau, 1985; Nagat a, 1975) may also be consult ed for power con-
sumpt ion infor mat ion r elat ed t o mixing wit h helical scr ews, r ibbons,
and dr aught t ubes.
3.7.1. Mi xer Vi scometry: Power Law Flui ds
Mixer viscomet r y, t ypically conduct ed using flag or pit ched paddle
impeller s (Fig. 3.11), has been used ext ensively for power law fluids
(Cast ell-Per ez and St effe, 1990; Cast ell-Per ez and St effe, 1992).
Techniques have been developed for est imat ing power law fluid pr op-
er t ies t hat ar e useful for qualit y cont r ol and engineer ing design appli-
cat ions. Shear r at e est imat ion met hods have been developed t o allow
fluid pr oper t ies t o be calculat ed over an appr opr iat e shear r at e r ange.
They ar e also useful in det er mining t he degr ee of mixing act ion pr esent
when evaluat ing pr ocess per for mance char act er ist ics involving par t icle
mot ion and gas disper sion.
Power Consumpti on and Average Shear Rate
Mixer viscomet r y is an impor t ant t ool for invest igat ing power law
fluids. Wor king equat ions ar e developed by fir st consider ing a New-
t onian fluid in laminar flow ( ). When , and may be
neglect ed (sur face t ension, elast ic, and vor t exing effect s ar e
insignificant ), t he power consumpt ion equat ion (Eq. [3.110]) may be
wr it t en as
[3.111]
or , incor por at ing Eq. [3.102] and [3.103], as
[3.112]
Eq. [3.112] may be used for power law fluids ( ) if t he Newt onian
fluid viscosit y ( ) is r eplaced by an appar ent viscosit y ( )
evaluat ed at an aver age shear r at e ( ) defined as
[3.113]
wher e is t he mixer viscomet er const ant having unit s of 1/r ad (unit s
of 1/r ev ar e found in some published wor ks). is unique for any
par t icular physical syst em and must be det er mined fr om exper iment al
dat a.
N
Re, I
< 63 N
We
N
Wi
N
Fr
N
Po

A
N
Re, I
P
d
5


A
d
2

n
K(

a
)
n 1

a
k
k
k
3.7.1 Mixer Viscometry: Power Law Fluids 191
In mixer viscomet r y, t her e ar e t wo pr imar y t echniques for det er -
mining : t he "slope met hod" and t he "viscosit y mat ching met hod."
These met hods ar e pr esent ed in t he following discussion.
Figure 3.11. Typical impellers used in mixer viscometry.
Evaluati on of
Slope Method. Subst it ut ion of Eq. [3.113], and t he power law appar ent
viscosit y for t he Newt onian viscosit y, int o a simplified for m of Eq. [3.112]
gives
[3.114]
or , aft er simplifying and t aking logar it hms,
[3.115]
Using power law fluids wit h differ ent known values of and , a plot
of ver sus is used t o det er mine fr om t he slope
of t he line which is equal t o . If t he plot is a st r aight line, t hen
t he appr oximat ion pr oposed ( ) in Eq. [3.113] is valid.
k
0.127 m
.01 m dia
.008 m dia
15 degree pitch blade
0.04143 m
0.02692 m
Haake Pitched Paddle
Flag
Paddle
k
P
d
3

2
A AK(

a
)
n 1
AK(k)
n 1
log
10

P
K
n + 1
d
3
_

,
log
10
(A) (1 n) log
10
(k)
K n
log
10
(P/[K
n + 1
(d)
3
]) (1 n) k
log
10
(k)

a
k
192 Chapter 3. Rotational Viscometry
One must exer cise car e in using t he slope met hod for calculat ing t he
mixer viscomet er const ant . Small er r or s in det er mining t he slope of t he
line will r esult in lar ge er r or s in because of t he logar it hmic r ela-
t ionship bet ween t he t wo number s. By convent ion, base 10 logar it hms
wer e used in Eq. [3.115]: Ident ical r esult s would be obt ained using base
. Det er minat ion of using t he slope met hod is illust r at ed in Example
Pr oblem 3.8.18.
Matchi ng Vi scosi ty Method. This t echnique involves t he compar ison
of power cur ves for Newt onian and non-Newt onian fluids using t he idea
of mat ching viscosit ies. The phr ase "mat ching viscosit ies" r efer s t o t he
assumpt ion t hat t he aver age shear r at e for a non-Newt onian fluid is
equal t o t he aver age shear r at e for a Newt onian fluid when t he New-
t onian viscosit y equals t he appar ent viscosit y of t he non-Newt onian
fluid. The t echnique is excellent for det er mining t he aver age shear r at e
in a mixer and is also useful t o food pr ocess engineer s in evaluat ing t he
per for mance of commer cial equipment (e.g., a scr aped-sur face heat
exchanger ) having poor ly defined shear fields.
Using Newt onian fluids such as silicone oil or cor n syr up in t he mixer
viscomet er , t he const ant is det er mined fr om t he slope of exper iment al
dat a pr esent ed in t he for m of Eq. [3.111]. Wit h a r out ine viscomet er
(like a concent r ic cylinder or cone and plat e syst em) and t r adit ional
r heological t echniques, t he pr oper t ies ( and ) of a power law fluid ar e
det er mined. This "st andar d" (or r efer ence) power law mat er ial is t hen
placed in t he mixer viscomet er , and mixed at a const ant speed. is
det er mined and t he r esult ing "aver age viscosit y" calculat ed using t he
power number expr ession for t he Newt onian fluids. This equat ion,
found by combining Eq. [3.111] and [3.103], is
[3.116]
Next , t he mat ching viscosit y assumpt ion is applied and an aver age shear
r at e ( ) calculat ed. Fir st , assume,
[3.117]
t hen, subst it ut e Eq. [3.117] int o Eq. [3.116],
[3.118]
k
e k
A
K n
N
Po

(d
2
) (N
Po
)
A

a
K(

a
)
n 1
K(

a
)
n 1

(d
2
) (N
Po
)
A
3.7.1 Mixer Viscometry: Power Law Fluids 193
Figure 3.12. Matching viscosity method to calculate average shear rate.
and solve for t he aver age shear r at e:
[3.119]
The mat ching viscosit y pr ocedur e t o evaluat e t he aver age shear r at e
is summar ized in Fig. 3.12:
St ep 1. Power number r elat ionship is est ablished for Newt onian
fluids and t he numer ical value of is det er mined using Eq.
[3.111];
Po
N
1
2
NNN
Re,I
3
Newtonian
Non-Newtonian
4
=
d
2
5 =
= K
n-1
a
a
=
K
6
K,n
Po
N
NNN
Po
.
.
.
A
1/(n-1)

d
2
(N
Po
)
KA
1
1
]
1/(n 1)
A
194 Chapter 3. Rotational Viscometry
St ep 2. At a st eady impeller speed, t he power number is measur ed
for a non-Newt onian fluid;
St ep 3. Newt onian and non-Newt onian power number s ar e set equal
and t he cor r esponding impeller Reynolds number , fr om t he
Newt onian fluid dat a, is evaluat ed;
St ep 4. "Viscosit y" is calculat ed fr om t he infor mat ion gener at ed in
St ep 3 using Eq. [3.116];
St ep 5. Mat ching viscosit y assumpt ion is applied so t he Newt onian
and appar ent viscosit ies ar e set equal as indicat ed in Eq. [3.117];
St ep 6. Taking t he appar ent viscosit y and power law fluid par am-
et er s, aver age shear r at e is calculat ed using Eq. [3.119] wr it t en
in t er ms of wher e .
is calculat ed as . St eps 2-6 should be r epeat ed at numer ous
impeller speeds t o det er mine how may var y wit h . An aver age value
of may be t aken t o r epr esent t he const ant r equir ed in Eq. [3.113].
Compari son of Calculati on Methods. Ther e ar e var ious
advant ages and disadvant ages t oconsider befor e using t he slope met hod
(SM) or t he mat ching viscosit y met hod (MVM). SM and MVM bot h
r equir e power law fluid "st andar ds" (r efer ence fluids) t o det er mine
t hat , wit h an appr opr iat e select ion of fluids, will cover a suit able r ange
of flow behavior indices. Fewer r efer ence fluids may be needed using
t he MVM. The main advant age of t he SM is t he simplicit y of t he cal-
culat ions; however , pr oblems wit h accur acy may occur due t o t he high
sensit ivit y of wit h t he slope of a logar it hmic plot . In compar ison t o
SM, t he MVM is somewhat labor ious due t o t he lar ger amount of dat a
handling. Result s fr om t he SM and MVM ar e similar . The effect ive
shear r at es of flag and st ar impeller s wer e evaluat ed using bot h met hods
by Rao and Cooley (1984). Result s fr om t hat wor k showed t he t wo
met hods t o be in good agr eement . Br iggs (1995) r eached a similar
conclusion wor king wit h flag impeller s. MVM would be t he met hod of
choice for engineer s at t empt ing t o evaluat e aver age shear r at es in
complex pr ocessing equipment .
Values and Vari ables Influenci ng . Skelland (1983) has sum-
mar ized values of mixer viscomet er const ant s ( ) and power cur ves for
many commer cial mixer s. Using t he aver age shear r at e concept t o
calculat e an appar ent viscosit y value, Nagat a (1975) det er mined power
cor r elat ions of Bingham plast ic mat er ials for t ur bine, anchor , and r ib-
d
2
N
Po
/A
k

a
/
k
k
k
k
k
k k
k
3.7.1 Mixer Viscometry: Power Law Fluids 195
bon mixer s. Bowen (1986) has pr esent ed an infor mat ive discussion
r elat ed t o t he det er minat ion of aver age and maximum shear r at es for
r adial flow t ur bine mixer s. A mixer viscomet er const ant has been
det er mined for t he Br abender Viscogr aph by Wood and Goff (1973), and
for a flag impeller r ot at ed in a number 303 can by Rao (1975).
Using t he slope met hod, St effe and For d (1985) found = 4.47 r ad
-1
for t he Haake MV cup and pit ched paddle impeller (Fig. 3.11). The
pit ched paddle impeller can be par t icular ly useful in maint aining par -
t icle suspension in samples exhibit ing sediment at ion pr oblems. Raw
dat a, and analyt ical pr ocedur es for t he det er minat ion of , used by
St effe and For d (1985) ar e pr esent ed in Example Pr oblem 3.8.18.
Cast ell-Per ez and St effe (1990) st udied shear r at e evaluat ion using
paddle impeller s (Fig. 3.10) for power law fluids. The effect s of numer ous
fact or s on t he det er minat ion of wer e consider ed. Result s showed t hat
was higher wit h less shear -t hinning fluids, incr eases wit h a
decr ease in r ot at ional speed and r eaches an almost const ant value when
oper at ing at speeds gr eat er t hat 0.33 r ev/s (20 r ev/min), and t ends t o
decr ease wit h an incr ease in t he r at io of t he impeller t o cup diamet er
( ). The effor t r esult ed in a number of r ecommendat ions for pr act ical
wor k in mixer viscomet r y involving paddle-t ype impeller s: if possible,
select a syst em wit h a small gap, > 0.709; if t he syst em t o be used
has a lar ge gap, det er mine t he effect of fluid pr oper t ies on ; and use a
minimum impeller speed of 0.33 r ev/s (20 r ev/min).
Br iggs (1995) found values for t he flag impeller descr ibed in Fig.
3.13. Fluid samples wer e held in a Br ookfield (Br ookfield Engineer ing
Labor at or ies, St ought on, MA) small sample cup having a height of 6.48
cm, and an inside diamet er of 1.90 cm making . Dat a wer e
collect ed over speeds r anging fr om 0.5 t o 100 r ev/min (0.5 t o 10.5 r ad/s).
Bot h t he slope and mat ching viscosit y met hods wer e found t o pr oduce
similar r esult s for shear -t hinning fluid foods. The slope analysis pr o-
duced a const ant value of t he mixer viscomet er const ant : r ad
-1
.
This value (and ) is r ecommended for r out ine analysis. The small
sample cup and flag impeller pr ovide a simple, yet power ful, t ool for t he
r heological invest igat ion of fluid foods.
Determi nati on of Non-Newtoni an Flui d Properti es
Evaluati ng versus . The r elat ionship bet ween power
and angular velocit y for a power law fluid was est ablished in Eq. [3.114]:
k
k
k
k k
k
d/D
d/D
k
k
d/D 0.79
k 2.92
A 4.84
log
10
(M) log
10
()
196 Chapter 3. Rotational Viscometry
Figure 3.13. Flag impeller used by Castell-Perez et al. (1993) and Briggs
(1995).
Subst it ut ing , and solving for t he t or que yields
[3.120]
or , in t er ms of logar it hms,
[3.121]
The flow behavior index can be found as t he slope of ver sus
. The const ant ( ) is not r equir ed for t he det er -
minat ion of : It may be used t o est imat e , if and ar e known. The
value of , however , may be found wit hout by applying Eq. [3.120] t o
t wo fluids, one wit h unknown pr oper t ies (fluid indicat ed wit h subscr ipt
) and a r efer ence fluid wit h known pr oper t ies (fluid indicat ed wit h
subscr ipt ):
[3.122]
h
d
blade length
blade shaft diameter
3.0
1.5
0.5
0.5
Flag Impeller
d
e
1.0
[ dimensions in cm ]
0.1
P
d
3

2
AK(k)
n 1
P M
M d
3
AK(k)
n 1

n
log
10
(M) log
10
(d
3
AK(k)
n 1
) + n log
10
()
log
10
(M)
log
10
(d
3
AK(k)
n 1
) log
10
()
n K A k
K A
x
y
M
x
d
3
AK
x
(k)
n
x
1

x
n
x
3.7.1 Mixer Viscometry: Power Law Fluids 197
and
[3.123]
Dividing by , Eq. [3.122] by Eq. [3.123], gives
[3.124]
which may be solved for :
[3.125]
The pr oper t ies of t he known fluid ( and ) ar e det er mined using
convent ional r heological met hods. is found as t he slope of Eq. [3.121]
and is det er mined using t he slope or mat ching viscosit y met hod.
Under some condit ions t he calculat ion of may be gr eat ly simplified:
If and is close t o , t hen is appr oximat ely . Mixer
viscomet r y t echniques discussed above ar e used t o det er mine power law
fluid pr oper t ies in Example Pr oblem 3.8.18.
Applyi ng the Matchi ng Vi scosi ty Assumpti on. Power law fluid
pr oper t ies may also be est imat ed using t he mat ching viscosit y
assumpt ion. An aver age appar ent viscosit y is calculat ed using Eq.
[3.116] at an aver age shear r at e defined by Eq. [3.113], . Then,
fluid pr oper t ies ar e found by r egr ession analysis of wit h an
appr opr iat e set of exper iment al dat a. This t echnique is illust r at ed in
Example Pr oblem 3.8.18.
One advant age of using t he mat ching viscosit y assumpt ion is t hat
it easily allows t he invest igat ion of a wide ar r ay of non-Newt onian
behavior . Alt hough t he aver age shear r at e det er mined using power law
fluids can be consider ed "exact " for t hese mat er ials, it may also be
consider ed a shear r at e appr oximat ion for ot her t ypes of fluids. This is
analogous t o t he way in which Eq. [3.32], t he Newt onian appr oximat ion,
may be used t o est imat e t he shear r at e of a non-Newt onian fluid in a
concent r ic cylinder appar at us.
M
y
d
3
AK
y
(k)
n
y
1

y
n
y
M
x
M
y
M
x
M
y

K
x

x
n
x
(k)
n
x
K
y

y
n
y
(k)
n
y
K
x
K
x
K
y

M
x
M
y
_

y
n
y
(k)
n
y

x
n
x
(k)
n
x
_

,
K
y
n
y
n
x
k
K
x

y

x
n
y
n
x
K
x
K
y
M
x
/M
y

a
k
K(

a
)
n 1
198 Chapter 3. Rotational Viscometry
Usi ng the Concentri c Cyli nder Analogy. Fluid pr oper t ies may be
det er mined fr om shear st r ess and shear r at e appr oximat ions developed
fr om a concent r ic cylinder analogy. Mixing syst ems (ident ical t o t he
one illust r at ed in Fig. 3.10) using differ ent cylindr ical cups and paddle
impeller s wer e consider ed by Cast ell-Per ez et al. (1991) in developing
r heogr ams for power law fluids. The r esear ch assumed t hat mixer s
wer e analogous t o concent r ic cylinder syst ems. This wor k showed t hat
t he aver age shear st r ess and aver age shear r at e could be est imat ed fr om
geomet r ical par amet er s and t he flow behavior index using t he following
equat ions:
[3.126]
and
[3.127]
for and wit h and
. The flow behavior index is found as t he slope of Eq. [3.121].
Cast ell-Per ez et al. (1993) det er mined shear st r ess and shear r at e
est imat es for power law fluids using a flag impeller (Fig. 3.13):
[3.128]
and
[3.129]
wher e , t he equivalent diamet er , is equal t o t he lengt h of t he impeller
blade plus t he diamet er of t he blade shaft . Dimensions of t he impeller
used in t he Cast ell-Per ez et al. (1993) st udy ar e given in Fig. 3.13: The
equivalent diamet er for t hat geomet r y is equal t o 1.0 cm. Blade
t hickness was 0.1 cm. Values of wer e limit ed t o t he r ange of 0.27
t o 0.59; however , t he influence of diamet er was minimal in t hat r ange
making it accept able t o assume in Eq. [3.129]. Wit h t his
assumpt ion, it is possible t o obt ain a r easonable est imat e of t he aver age

d
3
2

h
d
+
1
3
1
1
]
1
1
]
1
M

a
2

(D/d)
(2 n)/n
(D/d)
2n
1
1
1
]

d
h
_

,
n/2
d/D 0.709 0.36 < d/h < 1.80 1.0 cm h 5.0 cm
0.5 n 0.9

a

2M
d
e
2
h

a
2

D
d
_

,
n/2
d
e
d/D
d/D 1
3.7.2 Mixer Viscometry: Bingham Plastic Fluids 199
shear r at e wit hout knowing t he flow behavior index. The gr eat est level
of accur acy, however , is obt ained using t he flow behavior index calcu-
lat ed fr om t he r elat ionship given in Eq. [3.121].
3.7.2. Mi xer Vi scometry: Bi ngham Plasti c Flui ds
Power law fluids have been st udied ext ensively using mixer visco-
met r y, but lit t le wor k has been done wit h mat er ials exhibit ing a yield
st r ess (Cast ell-Per ez and St effe, 1992). Nagat a (1975) invest igat ed
power consumpt ion in mixing Bingham plast ics and some success has
been r epor t ed using mixer viscomet r y met hods on fr esh concr et e
showing Bingham plast ic behavior (Tat t er sall and Banfill, 1983).
Determi nati on of Flui d Properti es. Using t he Newt onian r ela-
t ionship bet ween and , assuming , and using t he definit ion
of appar ent viscosit y for a Bingham plast ic fluid ( ), yields
[3.130]
Subst it ut ing t he definit ion of (Eq. [3.102]) int o Eq. [3.130], r ecog-
nizing t hat , t hen simplifying t he r esult ing equat ion gives an
expr ession r elat ing t or que and speed in a mixer viscomet er :
[3.131]
which, aft er addit ional manipulat ion, becomes
[3.132]
Collect ing mixer viscomet er dat a of t or que ver sus angular velocit y for
a Bingham plast ic and plot t ing t he r esult will pr ovide a slope ( )
and an int er cept ( ) t hat r eflect t he r heological pr oper t ies of t he
fluid. If and ar e known, plast ic viscosit y and t he yield st r ess can
be calculat ed.
Evaluati on of . Theor et ically, can be det er mined wit h t he same
mat ching viscosit y t echnique discussed for power law fluids. Using t he
r elat ionship for Newt onian fluids, Eq. [3.111], is det er mined fr om dat a
involving st andar d Newt onian mat er ials. Taking dat a wit h a t r adi-
N
Po
N
Re, I

a
k

o
/

+
pl
N
Po

A
N
Re, I

A
d
2


A
d
2


A(
o
/

a
+
pl
)
d
2


A(
o
/(k) +
pl
)
d
2

N
Po
P M
M
d
5

o
kd
2

+

pl
d
2

,
M
Ad
3

o
k
+ Ad
3

pl

Ad
3

pl
Ad
3

o
/k
A k
k k
A
200 Chapter 3. Rotational Viscometry
t ional inst r ument , such as a concent r ic cylinder viscomet er , t he pr op-
er t ies of a "st andar d" (or r efer ence) Bingham plast ic fluid ar e evaluat ed.
This mat er ial is placed in a mixer viscomet er and mixed at const ant
speed ( ) allowing det er minat ion of . Using t hat infor mat ion, t he
"viscosit y" is calculat ed fr om Eq. [3.116]:
[3.133]
The mat ching viscosit y assumpt ion ( ) is applied:
[3.134]
and t he aver age shear r at e det er mined:
[3.135]
Using Eq. [3.113], is found as . If t he behavior of differ ent r efer ence
fluids is evaluat ed, t he influence of angular velocit y and fluid pr oper t ies
on can be invest igat ed.
The pr esence of a yield st r ess in a mixer means t her e will be ar eas
in t he syst em wher e . This cr eat es plug flow and "dead spot s" t hat
may adver sely influence r esult s. Wor k by Nagat a et al. (1970) indicat es
t hat may be a funct ion of t he yield st r ess.
3.7.3. Yi eld Stress Calculati on: Vane Method
Yield st r ess can be det er mined using t he same basic equipment
r equir ed in mixer viscomet r y. In t he vane met hod, t he st r ess t o init iat e
flow fr om a vane immer sed in t est mat er ial is measur ed (Nguyen and
Boger , 1985). Vane and vessel dimensions (Fig. 3.14) should st ay wit hin
specified limit s (St effe, 1992): ; ; or
if t he vane is complet ely immer sed in t he sample; wher e is
t he diamet er of t he cont ainer if cir cular , or t he minimum cr ossect ional
dimension if some ot her shape is used. Vanes wit h 4 (Nguyen and Boger ,
1985), 6 and 8 blades (Qiu and Rao, 1988) have been used. All pr oduce
similar r esult s.
N
Po

d
2
N
Po
A


o
/

a
+
pl

d
2
N
Po
A

a

o

d
2
N
Po
A

pl
_

,
1
k

a
/
k
<
o
k
1.5 h/d 4.0 Z
2
/d 0.5 Z
1
0.0 Z
1
/d 1.0
D/d 2.0 D
3.7.3 Yield Stress Calculation: Vane Method 201
If you assume t he t est mat er ial yields along a cylindr ical sur face
(shaft excluded), t hen t he t ot al t or que ( ) t o over come t he yield st r ess
of t he fluid is
[3.136]
wher e is t he shear st r ess on t he end sur faces (t op and bot t om).
Simplificat ion of Eq. [3.136] yields
[3.137]
Assume var ies wit h t he r adius accor ding t o a power r elat ionship:
[3.138]
wher e is a const ant . This assumpt ion is somewhat ar bit r ar y, but
r easonable. Subst it ut ing Eq. [3.138] int o [3.137] yields
[3.139]
which, wit h simplificat ion, gives
Figure 3.14. Vane, with 4 blades, and vessel for yield stress determination.
M
o
M
o
( dh)

d
2
_

o
+ 2

0
d/ 2
2 r
2

e
dr

e
M
o

hd
2
2

o
+ 4

0
d/ 2
r
2

e
dr

e
f (r)

2 r
d
_

,
m

o
m
M
o

hd
2
2

o
+ 4

0
d/ 2
r
2

2r
d
_

,
m

o
dr
d
h
z
z
1
2
D
h
202 Chapter 3. Rotational Viscometry
[3.140]
or , if solved for t he yield st r ess:
[3.141]
Er r or s in using Eq. [3.141] for decr ease wit h lar ger values of
. If , er r or s less t han or equal t o 3.7 per cent may be obt ained
when (St effe, 1992). The assumpt ion t hat is usually
sat isfact or y. It is cer t ainly accept able when making a qualit y cont r ol
compar ison bet ween pr oduct s or when simple, quick solut ions ar e
needed. To eliminat e er r or due t o t he upper end effect , t he t op sur face
of t he vane may be placed even wit h t he fluid ( , Fig. 3.14) giving
(assuming ) t he following t or que equat ion at t he yield point :
[3.142]
This expr ession was used by Yoshimur a et al. (1987) t o evaluat e t he
yield st r esses of var ious non-food emulsions.
Er r or associat ed wit h t he st r ess dist r ibut ion over t he ends can be
avoided if Eq. [3.137] is used t o det er mine t he yield st r ess: A plot of
ver sus is made using dat a collect ed fr om vanes having t he same
diamet er but differ ent lengt hs. The yield st r ess is calculat ed fr om t he
slope of t he r esult ing line which is equal t o . This pr ocedur e
should be used if t her e is any doubt r egar ding t he validit y of t he
assumpt ion t hat or when mor e pr ecise values of t he yield st r ess
ar e needed.
When measur ing t he st at ic yield st r ess (discussed in Sec. 1.6), vanes
should be car efully placed in t he sample t o minimize disr upt ion t o
sur r ounding mat er ial. In qualit y cont r ol wor k, it may be possible t o use
t he food cont ainer it self as t he t est vessel t ominimize sample disr upt ion.
When measur ing t he dynamic yield st r ess, sample disr upt ion is not a
pr oblem because t he weak st r uct ur e is alr eady dest r oyed fr om pr ior
handling oper at ions such as mixing or pumping. This assumpt ion would
be invalid if t he mat er ial was able t o r apidly r ebuild int er nal st r uct ur e.
Rapid r edevelopment of st r uct ur e is unusual in food pr oduct s.
M
o

d
3
2

h
d
+
1
m + 3
_

o

2M
o
d
3

h
d
+
1
m + 3
_

,
1
m > 1
h/d m 1
h/d > 2 m 0
Z
1
0.0
m 0
M
o

d
3
2

h
d
+
1
6
_

o
M
o
h

o
d
2
/2
m 0
3.7.3 Yield Stress Calculation: Vane Method 203
Figure 3.15. Typical torque time response when operating the vane in the con-
trolled rate mode.
Modes of Operati on. The vane may be oper at ed in t wo modes when
collect ing exper iment al dat a: cont r olled r at e or cont r olled st r ess. A
t ypical cur ve for t est ing in t he cont r olled r at e mode (Fig. 3.15) shows a
st eady incr ease in t or que up t o a peak value ( ) followed by a gr adual
decline unt il r eaching an equilibr ium level. In a sample complet ely
br oken down pr ior t o t est ing, t he peak t or que would r epr esent t he
dynamic yield st r ess. In an undist ur bed sample t aken fr om st or age,
t he peak t or que would r epr esent t he st at ic yield st r ess; however , in t his
t ype of cont r olled r at e t est , some weak st r uct ur al bonds may be dis-
r upt ed befor e t he peak t or que is r eached per haps causing t he measur ed
value t o be lower t han t hat found in a cont r olled st r ess exper iment .
In t he cont r olled r at e mode, var ious speeds of oper at ion have been
consider ed. Speeds as low as possible should be used, pr efer ably no
higher t han 1.0 r pm (St effe, 1992). Velocit ies of gr eat er magnit ude will
alt er r esult s for many pr oduct s. The angle of r ot at ion at which yield
occur s, needed t o det er mine t he st r ain at failur e, depends on many
fact or s: t ype of mat er ial being t est ed, size of t he vane, angular velocit y
of t he vane, and t he wind-up char act er ist ics of t he viscomet er . Calcu-
lat ing yield st r ess using t he vane met hod in t he cont r olled r at e mode is
Time at Constant Speed
T
o
r
q
u
e
M
o
M
o
204 Chapter 3. Rotational Viscometry
invest igat ed in Example Pr oblems 3.8.19 and 3.8.20. Wind-up
char act er ist ics of r ot at ional viscomet er s ar e discussed lat t er in t his
sect ion.
Vanes can also be oper at ed in t he cont r olled st r ess mode. In
exper iment at ion, t he t or que (or st r ess) is incr eased unt il flow is
obser ved. Dat a may show t he angle of r ot at ion (r elat ed t o st r ain) as a
funct ion of t ime wit h st epped incr eases in t or que. When t he yield st r ess
is exceeded, t he st r ain will incr ease r apidly wit h t ime. In a t ypical plot ,
t he t or que at yield ( ) is bet ween t or que levels M4 and M5 (Fig. 3.16).
Figure 3.16. Typical response curve when operating the vane in the controlled
stress mode.
Test ing in t he cont r olled st r ess mode has t he implicit assumpt ion
(which is easily t est ed exper iment ally) t hat r esult s ar e not t ime-
dependent so t he amount of t ime spent at each st r ess level has no
influence on t he out come. Fig. 3.16 shows changes in st r ain incr easing
slight ly (but r emaining const ant ) wit h each st epped incr ease in t or que
below t he failur e t or que. In a t ime-dependent mat er ial t he plot may
exhibit a gr adual incr ease in st r ain, in addit ion t o t he st ep incr ease,
dur ing t he holding per iod at t or que levels M1 t hr ough M4 (Fig. 3.16).
M
o
S
t
r
a
i
n
Time at Constant Torque (or Stress)
M1 M2 M3 M4 M5
M1
M2
M3
M4
M5
increasing levels of torque
M4 < M < M5
o
3.7.3 Yield Stress Calculation: Vane Method 205
Cont r olled st r ess r heomet er s may allow t he applicat ion of a r amped
(linear ) incr ease in st r ess. This will pr oduce a st r ain r esponse t hat is
unique for t he mat er ial being t est ed. Result s may be plot t ed t o show
changes in st r ain over t ime, or t he cor r esponding changes in st r ain wit h
t or que or st r ess (Fig. 3.17). Behavior is ver y differ ent depending on
whet her or not t he mat er ial is demonst r at ing solid or fluid-like behavior .
Ther e would be a linear r elat ionship bet ween st r ess and st r ain wit h
Hookean behavior . When t he yield st r ess is exceeded, t he onset of flow
causes a dr amat ic change in st r ain. If t he st r ain r esponse (Fig. 3.17) is
consider ed in t er ms of t wo cur ves, one for t he case wher e and t he
ot her for , t he point of int er sect ion may be defined as t he peak
t or que ( ) r equir ed in t he yield st r ess calculat ion. The act ual yield
st r ess would be found at t he int er sect ion point if st r ess, inst ead of t or que,
is plot t ed. When a br eak point is evident , t his is a convenient met hod
of det er mining t he yield st r ess fr om cont r olled st r ess dat a. Logar it hmic
plot s may be useful in amplifying differ ences.
Figure 3.17. Strain-stress curve showing the location of the yield stress when a
linear increase in stress is applied to the sample.

o
>
o
M
o
S
t
r
a
i
n
Torque or Stress
o
o
o
solid behavior
fluid behavior
M
o
or
206 Chapter 3. Rotational Viscometry
The idea of invest igat ing t he r elat ionship bet ween st r ain and st r ess
t odet er mine a yield st r ess can be applied t o samples held wit hin fixt ur es
ot her t han vanes: concent r ic cylinder , cone and plat e, and par allel plat e.
Dat a for Colgat e Toot hpast e Gel, collect ed at r oom t emper at ur e using
a cone and plat e appar at us, ar e pr esent ed in Fig. 3.18. The infor mat ion
was obt ained by subject ing t he sample t o a linear change in shear st r ess,
fr om 1 Pa t o 5 Pa, over a per iod of 500 s. St r ain is held near zer o unt il
t he applied st r ess exceeds t he yield st r ess (appr oximat ely 2.1 Pa) wher e
a subst ant ial incr ease in st r ain is obser ved (Fig. 3.18). Many samples
donot pr oduce such clear ly defined differ ences. Dat a in t he ear ly por t ion
of t he st r ain - st r ess cur ve (Fig. 3.17) can also be used t o calculat e a
shear modulus.
Figure 3.18. Strain-stress data for presheared Colgate Toothpaste Gel, at 25 C,
collected on a cone (60mm diameter, 4 ) and plate rheometer.
Numer ous means of evaluat ing t he yield st r ess ar e not ed in Table
1.4. Most cannot measur e a st at ic yield st r ess because sample st r uct ur e
is br oken down dur ing t est ing. Also, t he exper iment al difficult ies of
many appr oaches make t hem unat t r act ive. Over all, t he vane met hod
has much t o offer someone in need of a yield st r ess measur ement : The
1 1.5 2 2.5 3 3.5 4
Stress, Pa
S
t
r
a
i
n
,

%
Colgate Toothpaste Gel
Yield Stress = 2.1 Pa
1
0.8
0.6
0.4
0.2
0
-0.2

3.7.3 Yield Stress Calculation: Vane Method 207


t echnique involves a simple t heor y t hat can pr oduce quick, r epr oducible
r esult s. Oper at ing t he vane in t he cont r olled r at e, ver sus cont r olled
st r ess, mode offer s some advant ages because it pr oduces unequivocal
r esult s ( is pr ecisely defined in t he cont r olled r at e mode) which ar e
t he most sensit ive t o differ ences bet ween st at ic and dynamic yield
st r esses (Yoo et al., 1995). In addit ion, cont r olled r at e inst r ument s ar e
gener ally less expensive and mor e dur able t han cont r olled st r ess unit s.
Wi nd-up Characteri sti cs of Rotati onal Vi scometers. It is
impor t ant t o under st and t he physical char act er ist ics of t he t or que
sensor when consider ing yield st r ess measur ement s using a r ot at ional
viscomet er . Two t ypical t or que measur ing concept s, used in cont r olled
r at e inst r ument s, ar e illust r at ed in Fig. 3.19. In one syst em, for ce is
t r ansfer r ed t hr ough t he sample fr om t he bot t om plat e causing t he
deflect ion of a coiled spr ing. Spr ing defor mat ion is measur ed and cor -
r elat ed t o t or que. Full scale wind-up (how much t he upper plat e must
r ot at e befor e t he full scale t or que is measur ed) is high for t his t ype of
syst em, oft en r eaching as much as 80 degr ees.
Figure 3.19. Rotational viscometers with coil spring (high wind-up) and torsion
bar (low wind-up) torque sensing elements.
M
o
High Wind-Up System Low Wind-Up System
coil spring torsion bar
sample
deflection indicator
rotating member
208 Chapter 3. Rotational Viscometry
A syst em wit h low wind-up is also illust r at ed in Fig. 3.19. In t his
syst em t he deflect ion of a t or sion bar is cor r elat ed t o t or que. Full scale
deflect ion is achieved wit h ver y low wind-up values oft en in t he r ange
of 1 t o 2 degr ees. The wind-up char act er ist ics of t wo viscomet er s, one
wit h a spr ing t ype syst em and t he ot her wit h a t or sion bar , ar e con-
sider ed in Example Pr oblem 3.8.21.
Figure 3.20. Parallel plate viscometer with air bearing sensor support.
Cont r olled st r ess inst r ument s use an air bear ing t o pr ovide "fr ic-
t ionless" movement of t he upper fixt ur e (Fig. 3.20) and a dr ag cup mot or
t o gener at e cont r olled t or que on t he shaft . A posit ion indicat or is used
t o measur e displacement . Wind-up in cont r olled st r ess syst ems is
essent ially zer o.
3.7.4. Investi gati ng Rheomalaxi s
Mixer viscomet r y can be a useful, and simple, met hod of subject ively
quant ifying r heomalaxis (ir r ever sible st r uct ur al br eakdown) because
damage is minimized dur ing loading. The impeller of a mixer viscomet er
is placed in an undist ur bed sample and t or que decay dat a ar e collect ed
at var ious speeds (at least t hr ee). A new, undist ur bed, sample is
r equir ed for each t est . Raw dat a ar e plot t ed as shown in Fig. 3.21. A
cr oss plot of t he dat a is made: fir st , one finds t or que at t ime zer o (st ar t
air air
air air
sample
position
air bearing
fixed plate
drag cup
motor
indicator
3.7.4 Investigating Rheomalaxis 209
of t est ) and equilibr ium t ime (aft er t he cur ve becomes hor izont al) for
each angular velocit y, t hen t or que ( ) ver sus angular velocit y ( ) is
plot t ed. The ar ea bet ween t he r esult ing init ial and equilibr ium t ime
cur ves r eflect s t he degr ee of r heomalaxis in t he sample. If t he init ial
and equilibr ium t or que cur ves ar e modeled as funct ions and ,
r espect ively, t he ar ea ( ) bet ween t he cur ves may be calculat ed as
[3.143]
has t he unit s of power , N m s
-1
. If is lar ge, t he power r equir ed t o
mix an undist ur bed sample is high compar ed t o t hat r equir ed for a fully
br oken down sample. If t he mat er ial is not t ime-dependent , . Since
t his is an empir ical t est ing met hod, ident ical equipment and t est ing
pr ocedur es must be followed t o compar e differ ent samples. The above
concept is illust r at ed for st r ained apr icot s in Example Pr oblem 3.8.22.
Figure 3.21. Raw and cross plotted data to quantify rheomalatic behavior.
Tor que decay dat a can pr ovide a useful qualit y cont r ol par amet er
for compar ing t he st r uct ur e of differ ent food pr oduct s by consider ing
ener gy levels r equir ed for mixing. Power ( ) ver sus t ime, at a const ant
value of , is plot t ed so t he ar ea under t he r esult ing cur ve r epr esent s
t he mechanical ener gy input t o t he sample. Samples r equir ing low
ener gy levels for mixing r eflect st r uct ur al br eakdown in t he mat er ial
when compar ed t o similar , but non-t ime-dependent , fluids. One can
M
f
1
() f
2
()

3
(f
1
() f
2
())d

0
M M
Time
1
2
3
1 2 3
Initial
Equilibrium
area between curves
Raw Data Cross Plotted Data
M

210 Chapter 3. Rotational Viscometry


invest igat e r ecover y of sample st r uct ur e by int r oducing a r est per iod,
t hen r epeat ing t he t est pr ocedur e. To make meaningful compar isons,
t he volume of sample subject ed t o agit at ion dur ing t est ing must be
const ant . This ener gy analysis concept is examined in Example Pr oblem
3.8.22.
Mixer viscomet r y dat a can also be used t o evaluat e a st r uct ur al
par amet er for r heomalat ic mat er ials, such as defined by Eq. [1.37]
(For d and St effe, 1986), or t o model t or que decay wit h a simple expo-
nent ial equat ion (St effe and For d, 1985).
3.7.5. Defi ni ng An Effecti ve Vi scosi ty
A simple appr oach can be used t o char act er ize non-t ime-dependent
fluids for t he pur pose of qualit y cont r ol. In t his case absolut e r heological
pr oper t ies ar e not needed or difficult t o obt ain: Only a r epr esent at ive
flow cur ve, which can be compar ed t o a r efer ence flow cur ve, is r equir ed.
The r efer ence flow cur ve is det er mined fr om pr oduct s consider ed t o be
accept able for t he int ended mar ket .
An effect ive viscosit y may be defined in t er ms of mixing par amet er s:
[3.144]
wher e is equal t o t he diamet er of t he impeller . Since effect ive viscosit y
is a r elat ive value, ot her par amet er s, such as t he volume of t he mixing
vessel or t he sample volume swept t hr ough by t he impeller , could be
used in place of . Equipment , sample volume and impeller or ient at ion
should be const ant for all t est s. Eq. [3.144] is similar t o Eq. [3.116] wit h
t he assumpt ion t hat and t he r ecognit ion t hat .
One can assume t hat effect ive viscosit y is dir ect ly pr opor t ional t o
appar ent viscosit y, and angular velocit y is dir ect ly pr opor t ional t o shear
r at e. These values can be plot t ed t o det er mine effect ive r heological
pr oper t ies t o char act er ize samples. Suit able qualit y cont r ol cr it er ia can
be developed using t his concept . The effect ive viscosit y idea is applied
t o st r ained apr icot s in Example Pr oblem 3.8.18.
3.8. Example Problems
3.8.1. Bob Speed for a Bi ngham Plasti c Flui d
Consider t est ing a Bingham plast ic fluid (yield st r ess = 13 Pa ; plast ic
viscosit y = 1.7 Pa s) in a Sear le-t ype (t he bob r ot at es and t he cup is

effective

M
d
3
d
d
3
N
Po
M/(d
5

2
) A 1.0
3.8.1 Bob Speed for a Bingham Plastic Fluid 211
st at ionar y) concent r ic cylinder viscomet er wit h values ( ) of
1.1, 1.3, and 1.5. In each case, est imat e t he minimum angular velocit y
of t he bob r equir ed t o achieve flow at t he wall of t he cup, t hus, assur ing
complet e flow in t he annulus.
Eq. [3.29] st at es t hat
Ther efor e, t he minimum value of t or que t o achieve flow at is
[3.145]
The equat ion r elat ing t or que t o angular velocit y in t his syst em,
assuming flow t hr oughout t he annulus, is given by Eq. [3.28]:
Using Eq. [3.145], t he minimum t or que expr ession, t he minimum
angular velocit y may be calculat ed:
[3.146]
or , given t he above expr ession for minimum t or que,
[3.147]
Simplificat ion, and subst it ut ion of , yields a gener al expr ession
for t he minimum angular velocit y of t he bob:
[3.148]
This equat ion indicat es t he cr it ical impor t ance of gap size in det er -
mining minimum angular velocit y. Using t he specified yield st r ess and
t he plast ic viscosit y gives
[3.149]
R
c
/R
b
M
min
2R
c
2
h
>
o
R
c
M
min
2R
c
2
h
o

M
4h
pl
1
1
]

1
R
b
2

1
R
c
2
1
1
]

pl
ln

R
c
R
b
_

min

M
min
4h
pl
1
1
]

1
R
b
2

1
R
c
2
1
1
]

pl
ln

R
c
R
b
_

min

2R
c
2
h
o
4h
pl
1
1
]

1
R
b
2

1
R
c
2
1
1
]

pl
ln

R
c
R
b
_

,
R
c
/R
b

min


o
2
pl
[
2
1 2ln]

min

13
2(1.7)
[
2
1 2ln]
212 Chapter 3. Rotational Viscometry
Subst it ut ing t he specified values of int o Eq. [3.149] pr oduces t he
following r esult s:
, r ad/s
1.1 0.074 (0.71 r pm)
1.3 0.63 (6.04 r pm)
1.5 1.67 (15.95 r pm)
2.0 6.16 (58.82 r pm)
The minimum bob speed t o maint ain flow of a Bingham plast ic fluid
in t he annulus incr eases r apidly as t he size of t he gap incr eases.
Ther efor e, small gaps, appr opr iat e for given par t icle size limit at ions,
ar e pr efer r ed t o lar ge gaps.
3.8.2. Si mple Shear i n Power Law Flui ds
Det er mine t he er r or involved in using t he simple shear appr oximat ion
for power law fluids at differ ent values of .
To solve t his pr oblem, bot h t he appr oximat e (Eq. [3.30]) and act ual
(Eq. [3.33]) solut ions must be consider ed:
[3.150]
[3.151]
The "% Er r or " may be calculat ed as
[3.152]
Subst it ut ing Eq. [3.150] and Eq. [3.151] int o Eq. [3.152] yields
[3.153]
which descr ibes t he er r or expect ed for differ ent values of and .
Result s ar e best seen in gr aphical for m (Fig. 3.22). To keep er r or t o a
minimum, smaller gaps (r eflect ed in smaller values of ) ar e needed for

R
c
/R
b

min
R
c
/R
b
(

b
)
approximate


1
(

b
)
actual

2
n
_

2/n

2/n
1
1
1
]
% Error

1
(

b
)
approximate
(

b
)
actual
1
1
]
100
% Error

1
n
2( 1)

2/n
1

2/n
_

,
1
1
]
100
n

3.8.3 Newtonian Fluid in a Concentric Cylinder 213


lower values of t he flow behavior index. If and , er r or
involved in using t he simple shear appr oximat ion will be less t han 6%:
At and , t he er r or will exceed 10%.
3.8.3. Newtoni an Flui d i n a Concentri c Cyli nder
Det er mine an expr ession for t he shear r at e as a funct ion of t he r adius
for a Newt onian fluid in t he annulus of a concent r ic cylinder viscomet er .
Using t his expr ession, det er mine t he shear r at e at t he bob and cup.
A Newt onian fluid is defined (Eq. [1.25]) as
and t he equat ion giving t he shear st r ess in t he annulus (Eq. [3.2]) is
Figure 3.22. Error (Eq [3.153]) in shear rate calculation when using the simple
shear approximation for a power law fluid in a concentric cylinder system
( ).
n 0.4 1.0 < < 1.02
n 0.2 > 1.02

f (r)
M
2hr
2
1.01 1.02 1.03 1.04 1.05
0
2
4
6
8
10
%

E
r
r
o
r
n=.2
n=.4
n=.6
n=.8
n=1.0
n=1.2
R
c
/R
b
214 Chapter 3. Rotational Viscometry
for . Combining t he above equat ions, t he shear r at e may be
wr it t en as
[3.154]
Eq. [3.21] st at es t hat
which can be solved for t or que as
[3.155]
wher e, r ecall, . Subst it ut ing Eq. [3.155] int o Eq. [3.154] gives
t he shear r at e as a funct ion of t he r adius for any const ant value of t he
angular velocit y:
[3.156]
wher e . The expr essions for t he shear r at e at t he bob ( , given
pr eviously as Eq. [3.32]) and cup ( ) can be easily calculat ed fr om Eq.
[3.156]:
[3.157]
3.8.4. Representati ve (Average) Shear Rate
Fig. 3.23 shows a popular bob design based on a Ger man st andar d (DIN
53019) developed by t he Ger man Inst it ut e for St andar dizat ion,
Deut sches Inst it ut fr Nor mung, known by t he DIN acr onym. The
following r est r ict ions ar e given for dimensional consider at ions:
; ; ; ; . Pr efer r ed
dimensions ar e also st at ed: ; ; ; ; .
No pr efer r ed dimension is given for . Not e t hat is t he apex angle
of t he cone at t he bot t om of t he inner cylinder , and t he summat ion of
values ( ) is equal t o t he fluid level in t he cup.
R
b
r R
c

f (r)
M
2hr
2


M
4h

1
R
b
2

1
R
c
2
1
1
]
M
4hR
c
2

2
1
R
c
/R
b

f (r)
M
2hr
2

1
2hr
2

1
1
]

4hR
c
2

2
1
1
1
]
2

R
c
2
/r
2

2
1
_

,
R
b
r R
c

b
f (R
b
) 2

2
1
_

c
f (R
c
) 2

2
1
_

,
R
c
/ R
b
1.1 h/R
b
3 h/R
b
1 R
s
/R
b
0.3 90 150
1.0847 h/R
b
3 h/R
b
1 h/R
b
1 120
R
s
/R
b

h
h + h + h
3.8.4 Representative (Average) Shear Rate 215
Figure 3.23. Bob and cup design based on German standard, DIN 53018.
DIN 53019 st at es t hat t he bob and cup combinat ion illust r at ed in
Fig. 3.23 can be used in conjunct ion wit h r epr esent at ive values (aver age
values) of t he shear st r ess and t he shear r at e in det er mining r heological
behavior of Newt onian and non-Newt onian fluids. The r epr esent at ive
shear st r ess, defined as t he aver age shear st r ess bet ween t he bob and
t he cup, was pr esent ed in Eq. [3.31]:
wher e has been added t o account for bot h t he t op and bot t om end
effect s. The DIN st andar d st at es t hat for t he pr efer r ed geo-
met r ical r elat ions allowing t he wor king equat ion for aver age shear
st r ess t o be wr it t en as
[3.158]
R
c
R
s
R
b
h
h
h

a

M(1 +
2
)
4(h + h
o
)R
c
2
h
o
h
o
0.1h

a

M(1 +
2
)
4(h + .1h)R
c
2

M(1 +
2
)
(4.4) hR
c
2
216 Chapter 3. Rotational Viscometry
Questi on: Taking t he Newt onian appr oximat ion for shear r at e (Eq.
[3.32]), det er mine t he r epr esent at ive shear r at e t hat would be appr o-
pr iat e in evaluat ing t he r heogr am.
The r epr esent at ive shear r at e is t he aver age shear r at e ( ) bet ween
t he bob and t he cup:
[3.159]
Using t he values of and (Eq. [3.32] and Eq. [3.157]) given in Example
Pr oblem 3.8.3, Eq. [3.159] can be wr it t en as
[3.160]
Manipulat ion yields a simple equat ion t o calculat e t he r epr esent at ive
shear r at e:
[3.161]
A r heogr am can be developed fr om r epr esent at ive values of shear
st r ess and shear r at e. These equat ions ar e oft en used in t he comput er
soft war e calculat ions found in commer cially available concent r ic cyl-
inder viscomet er s. This idea is applicable t o most concent r ic cylinder
syst ems wit h nar r ow gaps ( ). When using t he aver age shear
r at e equat ion for bobs having an end geomet r y differ ent t han t hat
illust r at ed in Fig. 3.23, t he aver age shear st r ess equat ion would need
t o be adjust ed for differ ent values of . If applied t o t he bob shown in
Fig. 3.1, for example, t he value of would be smaller t han t he value of
specified by t he DIN st andar d. Theor et ical just ificat ion for using
r epr esent at ive shear st r ess and shear r at e values may be found in
Giesekus and Langer (1977).
3.8.5. Concentri c Cyli nder Vi scometer: Power Law Flui d
Develop an expr ession for t he shear r at e at t he bob for a power law fluid
in a concent r ic cylinder viscomet er . Also der ive an equat ion descr ibing
t he fluid velocit y pr ofile found in t his syst em. Assume t he viscomet er
is a Sear le syst em wher e t he bob r ot at es and t he cup is st at ionar y.

b
+

c
2

2
1
_

,
+ 2

2
1
_

,
1
1
]
/ 2

a

(
2
+ 1)

2
1
R
c
/R
b
1.1
h
o
h
o
0.1h
3.8.5 Concentric Cylinder Viscometer: Power Law Fluid 217
This pr oblem can be solved using t he same appr oach t aken in
Example Pr oblem 3.8.3. By consider ing t he definit ion of a power law
fluid ( ) and t he for ce balance on a bob (Eq. [3.2]), t he following
expr ession for t he shear r at e in t he annulus is obt ained:
[3.162]
Then, t he shear r at e at t he bob is
[3.163]
However , fr om Eq. [3.25],
Solving t his equat ion for and subst it ut ing t he r esult int o t he above
expr ession for yields
[3.164]
or , using , Eq. [3.164] becomes
which pr ovides t he final solut ion given ear lier as Eq. [3.33].
To der ive an equat ion for t he velocit y pr ofile of a power law fluid in
a concent r ic cylinder viscomet er , st ar t wit h Eq. [3.15]:
Changing t he limit s of int egr at ion, and subst it ut ing , yields
[3.165]
or , aft er int egr at ion,
K

f (r)

M
2hr
2
K
_

,
1/n

b
f (R
b
)

M
2hR
b
2
K
_

,
1/n

n
2K
1/n

R
b
R
c
_

,
2/n
_

M
2hR
b
2
1
1
]
1/n
M

b

2
n

R
c
2/n
R
c
2/n
R
b
2/n
_

,
R
c
/R
b

b

2
n

2/n

2/n
1
_


0
d
1
2

c
f ()
d

f () (/K)
1/n

0
d
1
2

K
_

,
1/n
d

218 Chapter 3. Rotational Viscometry


[3.166]
Using Eq. [3.2], Eq. [3.166] can be wr it t en as a funct ion of t he r adius:
[3.167]
Simplificat ion gives
[3.168]
which can be used t o gener at e an expr ession for a dimensionless velocit y
pr ofile in t he gap:
[3.169]
wher e . As expect ed, at , and at .
3.8.6. Concentri c Cyli nder Data - Tomato Ketchup
Given r awdat a for t omat oket chup (Table 3.2), det er mine t he r heological
pr oper t ies of t he fluid. Use t he power law appr oximat ion for shear r at e
pr esent ed in Eq. [3.33]:
Also evaluat e t he alt er nat ive equat ion for given by Heywood (1991b).
The fir st st ep in finding t he shear r at e at t he bob is t o calculat e t he
flow behavior index. Regr ession analysis of ver sus yields
meaning, fr om Eq. [3.37], t hat
Wit h t his r esult and t he equat ion for , t he shear r at e at each value of
may be calculat ed. Shear st r ess at t he bob is det er mined fr om Eq.
[3.3]:
f ()
n
2K
1/n
[
1/n

c
1/n
]
f (r)
n
2K
1/n

M
2hr
2
_

,
1/n

M
2hR
c
2
_

,
1/n
1
1
]
f (r)
n
2K
1/n

M
2hR
c
2
_

,
1/n

R
c
r
_

,
2/n
1
1
1
]

f (r)
f (R
b
)

(1/r
2
)
1/n
(1/R
c
2
)
1/n1
]

(1/R
b
2
)
1/n
(1/R
c
2
)
1/n1
]

[(R
c
/r)
2/n
1]
[(R
c
/R
b
)
2/n
1]
R
b
r R
c
r R
b
0 r R
c

2
n
_

2/n

2/n
1
1
1
]

b
ln(M) ln()
ln(M) (.307) ln() 5.0529
n
d ln(M)
d ln()
0.307

3.8.6 Concentric Cylinder Data - Tomato Ketchup 219


Table 3.2. Data and Results for a Rheological Test for Tomato Ketchup at 25 C
Conducted using aConcentric Cylinder Viscometer: =60.00 mm; =21.00 mm;
=20.04 mm; =1050 kg/m
3
.
RPM
(r a d/ s ) (N m) (Pa ) (1/ s )
1 0.105 -2.256 0.00346 -5.666 22.85 2.60
2 0.209 -1.563 0.00398 -5.526 26.29 5.19
4 0.419 -0.870 0.00484 -5.331 31.97 10.39
8 0.838 -0.177 0.00606 -5.106 40.03 20.77
16 1.676 0.516 0.00709 -4.949 46.83 41.55
32 3.351 1.209 0.00848 -4.770 56.01 83.10
64 6.702 1.902 0.01060 -4.547 70.01 166.20
128 13.404 2.596 0.01460 -4.227 96.43 332.39
256 26.808 3.289 0.01970 -3.927 130.12 664.78
Result s of t hese calculat ions ar e pr esent ed in Table 3.2.
Assuming a power law model t o r epr esent t he r heogr am of t his
mat er ial gives an excellent r epr esent at ion of t he dat a and a value for
t he consist ency coefficient ( ) equal t o 15.73 Pa s
n
. Since t his mat er ial
exhibit s power law behavior , t he r esult s ar e almost ident ical t o t hose
found using t he Kr ieger appr oximat ion (Eq. [3.47]) because is
essent ially equal t o . In addit ion, one can obser ve t hat simple and
Newt onian shear r at e equat ions (Eq. [3.30] and [3.32]) give, at best ,
gr oss appr oximat ions. Consider ing t he shear r at e at 256 r pm (26.808
r ad/s) for bot h cases gives
and
Alt hough, t hese values ar e t he same or der of magnit ude as t he power
law appr oximat ion of 664.78 s
-1
, t he er r or would be unaccept able in most
cases.

h R
c
R
b

ln() M ln(M)
b

b

M
2hR
b
2
K
1/s
1/n
(

b
)
simple shear


1

26.808
1.048 1
558.5 s
1
(

b
)
Newtonian
(2)

2
1
1
1
]
2(26.808)

(1.048)
2
(1.048)
2
1
1
1
]
599.0 s
1
220 Chapter 3. Rotational Viscometry
To check for pot ent ial pr oblems due t o secondar y flow Eq. [3.90] must
be evaluat ed:
Recall t hat Taylor vor t ices may be expect ed when t his inequalit y is
sat isfied. Alt hough t he equat ion was est ablished for Newt onian fluids,
a r easonable assessment of t he cur r ent pr oblem can be obt ained by
subst it ut ing t he appar ent viscosit y ( ) for . Taking t he wor st case
(highest speed) t he calculat ion can be made as
which is clear ly much less t han
meaning er r or s due t o t he pr esence of secondar y flow ar e negligible.
Heywood (1991b) offer ed t he following equat ion t o det er mine t he
shear r at e at t he bob in a concent r ic cylinder syst em:
[3.170]
wher e
[3.171]
and
[3.172]
Eq. [3.170] is a simplified for m of an infinit e ser ies solut ion of Eq [3.42]
given by Kr ieger and Mar on (1952). To evaluat e Eq. [3.170] for t he
special case of a power law fluid one must r ecognize, by obser ving Eq.
[3.37], t hat and . Making t hese subst it ut ions, Eq. [3.170]
becomes
[3.173]
In t he case of t omat o ket chup, and .
Subst it ut ing t hese values int o t he Kr ieger and Mar on solut ion yields
R
b
(R
c
R
b
)

> 41.3

R
c
R
c
R
b

b
/

b

(26.81) (.02004) (.02100 .02004) (1050)
(130.12/664.78)
2.77
41.3

.02100
.02100 .02004
193.16

b


ln

1 +
ln
a
+
1
3

ln
a
_

,
2
(1 a) + .
1
1
]
a
d(lnM)
d(ln)
a
d(a)
d(ln)
a n a 0

b


ln

1 +
ln
n
+
1
3

ln
n
_

,
2
1
1
]
n 0.307 21.00/20.04 1.048
3.8.7 Infinite Cup - Single Point Test 221
The exact power law solut ion is
meaning t he above equat ions t o calculat e shear r at e ar e equivalent t o
t he second decimal place.
3.8.7. Infi ni te Cup - Si ngle Poi nt Test
A viscomet er wit h a single bob, and no cup, is t o be used in developing
a single point qualit y cont r ol t est for t omat o ket chup (at 25 C) when
held in a lar ge vat . If t he inst r ument is r un at 50 r pm, what appar ent
viscosit y would be expect ed?
Pr evious wor k (Example Pr oblem 3.8.6) wit h a concent r ic cylinder
inst r ument yielded power law par amet er s for t he fluid: = 15.73 Pa s
n
and = 0.307 . The shear r at e at t he bob for t his mat er ial in a concent r ic
cylinder syst em (Eq. [3.33]) is
In a vat , which equals is lar ge so t he t er m in br acket s is effect ively
one. Ther efor e, t he shear r at e may be expr essed as
allowing t he shear r at e at 50 r pm t o be calculat ed:
Wit h t his value of t he shear r at e, t he expect ed appar ent viscosit y may
be det er mined:
3.8.8. Infi ni te Cup Approxi mati on - Power Law Flui d
Det er mine t he er r or involved in using t he infinit e cup appr oximat ion
for power law fluids at differ ent values of .

b


ln(1.048)

1 +
ln(1.048)
.307
+
1
3

ln(1.048)
.307
_

,
2
1
1
]
24.75

2
.307
_

1.048
2/.307
1.048
2/.307
1
1
1
]
24.75

K
n

2
n
_

2/n

2/n
1
1
1
]
R
c
/R
b

b

2
n

b

2(50) (2)
(.307)60
34.11 s
1
K(

b
)
n 1
15.73(34.11)
.307 1
1.36 Pa s
R
c
/R
b
222 Chapter 3. Rotational Viscometry
Figure 3.24. Error (Eq. [3.177]) in using the infinite cup solution for power law
fluids in a concentric cylinder system ( ).
The appr oximat e solut ion is found fr om Eq. [3.49] by not ing (Eq.
[3.36]) t hat for a power law fluid:
[3.174]
The act ual solut ion, pr ovided Eq. [3.33], is
[3.175]
The "% Er r or " may be calculat ed as
[3.176]
Subst it ut ing Eq. [3.174] and Eq. [3.175] int o Eq. [3.176] yields
[3.177]
2 4 6 8 10 12
0
1
2
3
4
5
%

E
r
r
o
r
n=.2
n=.4
n=.6
n=.8
n=1.0
n=1.2
R
c
/R
b
d(ln)/d(ln
b
) 1/n
(

b
)
approximate

2
n
(

b
)
actual

2
n
_

2/n

2/n
1
1
1
]
% Error

1
(

b
)
approximate
(

b
)
actual
1
1
]
100
% Error

2/n
1

2/n
_

,
1
1
]
100
3.8.9 Infinite Cup - Salad Dressing 223
which gives t he er r or expect ed for differ ent values of and . Plot t ing
t he r esult s (Fig. 3.24) makes it clear t hat lar ger cont ainer s ar e r equir ed
(lar ger ) t o keep er r or t o accept able levels as t he flow behavior index
incr eases. Wit h , must be gr eat er t han 3 t o keep t he er r or below
5%. The t echnique of er r or det er minat ion used above (and in Example
Pr oblem 3.8.2) can be ver y helpful in evaluat ing t he assumpt ions made
in t he "aut omat ic" calculat ions of commer cial inst r ument s.
3.8.9. Infi ni te Cup - Salad Dressi ng
Rheological dat a (Table 3.3) for Kr aft Fr ench salad dr essing wer e col-
lect ed using a concent r ic cylinder geomet r y wit h t he following dimen-
sions: bob r adius = 20.04 mm; cup r adius = 73.00 mm; bob height = 60.00
mm. The bob has a r ecessed t op and bot t om like t he one illust r at ed in
Fig. 3.1 so we will assume end effect s ar e negligible. Using t he infinit e
cup assumpt ion, det er mine t he r heological pr oper t ies of t his t ime-
independent mat er ial.
Table 3.3. Rheological Data for Kraft French Salad Dressing at 22 C [Concentric
Cylinder System: =20.04 mm, =73.00 mm, =60.00 mm]
(r a d/ s ) (N m) (Pa ) (1/ s )
0.146 0.000609 4.02 0.79
0.512 0.000998 6.59 2.77
1.036 0.001264 8.35 5.60
2.087 0.001623 10.72 11.28
4.163 0.002033 13.43 22.50
6.276 0.002430 16.05 33.92
8.359 0.002708 17.89 45.18
10.490 0.002970 19.62 56.70
12.590 0.003149 20.80 68.05
14.680 0.003335 22.03 79.35
16.770 0.003509 23.18 90.65
Shear st r ess at t he bob is calculat ed fr om Eq. [3.3] as
The shear r at e may be det er mined fr om Eq. [3.49] as
n

n 0.6

R
b
R
c
h
M
b

b

M
2hR
b
2

b
(2)
d(ln)
d(ln
b
)
224 Chapter 3. Rotational Viscometry
Figure 3.25. Logarithmic plot of angular velocity versus shear stress for Kraft
French salad dressing.
Figure 3.26. Rheogram for Kraft French salad dressing at 22 C.
3 5 10 20 30
0.1
0.2
0.5
1
2
5
10
20
Shear Stress, Pa
A
n
g
u
l
a
r

V
e
l
o
c
i
t
y
,

r
a
d
/
s
Kraft French Dressing
0 20 40 60 80 100
0
5
10
15
20
25
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Kraft French Dressing

3.8.10 Infinite Cup - Yield Stress Materials 225


Regr ession of t he dat a indicat es a good fit of t he logar it hmic plot (Fig.
3.25) wit h a slope of 2.73:
Knowing t he slope, t he shear r at e can be calculat ed as
Wit h t he shear st r ess and shear r at e dat a comput ed (Table 3.3), t he
r heogr am may be plot t ed (Fig. 3.26), power law const ant s det er mined
(mor e r egr ession analysis) and t he final model pr esent ed:
Not e, t he slope of t he logar it hmic plot is t he r ecipr ocal of t he flow
behavior index ( ) because, as st at ed in t he pr evious
example pr oblem, t he infinit e cup solut ion for a power law fluid is
Also, in consult ing Fig. 3.24, one can see t he er r or in using t he infinit e
cup appr oximat ion for and should be less
t han 0.5%.
3.8.10. Infi ni te Cup - Yi eld Stress Materi als
Under what condit ions could Eq. [3.49], an expr ession t o det er mine t he
shear r at e at t he bob in an infinit e cup, be used for a vat cont aining a
mat er ial wit h a yield st r ess?
Knowing t he yield st r ess, t he cr it ical r adius ( ) can be calculat ed
by consider ing Eq. [3.2]:
[3.178]
wher e is t he maximum t or que gener at ed by t he inst r ument dur ing
t est ing. defines t he shear ed ar ea of t he sample in t he vat when fluid
mot ion is due only t o bob r ot at ion.
d(ln)
d(ln
b
)
2.73

b
(2)2.73
4.43(

)
.37
1/n 1/0.37 2.7

b

2
n
73.00/20.04 3.64 n 0.37
R
o
R
o

M
max
2h
o
M
max
R
o
226 Chapter 3. Rotational Viscometry
Figure 3.27. Bob placement, showing sheared area, in a vat full of fluid having a
yield stress.
The inst r ument should be posit ioned (Fig. 3.27) so t he cent er of t he
bob is, at a minimum, a dist ance equal t o fr om t he wall of t he vat .
Under t hese condit ions, t he physical sit uat ion in t he vat is analogous
t ohaving a concent r ic cylinder syst em, wit h a par t ially shear ed annulus;
hence, Eq. [3.49] should give sat isfact or y r esult s.
3.8.11. Cone and Plate - Speed and Torque Range
Assume t omat o ket chup will be t est ed in a cone and plat e viscomet er
so t he r esult s can be compar ed wit h t hose found in t he concent r ic cyl-
inder t est s. Using a cone wit h an angle of 0.0524 r ad (3 degr ees) and a
diamet er of 5.0 cm, what speed and t or que r esponse would be needed
t o cover an exper iment al shear r at e r ange of 1 t o 100 s
-1
?
In a cone and plat e syst em, t he r elat ionship bet ween angular velocit y
and shear r at e is given by r ear r anging Eq. [3.51]:
Using t his expr ession, t he minimum and maximum velocit ies may be
calculat ed as
R
o
vat wall
bob
sheared area
R
o


(tan)
3.8.12 Cone and Plate - Salad Dressing 227
and
To cover t he shear r at e r ange of 1 t o 100 s
-1
t he inst r ument should have
a minimum speed r ange of 0.5 t o 50 r pm. The r equir ed t or que capabilit y
of t he inst r ument may be consider ed fr om Eq. [3.55]:
Rear r anging t his expr ession, and incor por at ing t he definit ion of t he
shear st r ess for a power law fluid yields
Subst it ut ing t he appr opr iat e number s (fr om Example 3.8.6:
Pa s
n
, ) and calculat ing t he r esult s gives
and
To t est samples over t he shear r at e r ange of 1 t o 100 s
-1
t he inst r ument
t or que r esponse must include t he r ange fr om 5,150 t o 21,2000 dyne cm.
3.8.12. Cone and Plate - Salad Dressi ng
Rheological dat a (Table 3.4) wer e collect ed for Kr aft Fr ench salad
dr essing using a cone and plat e syst em. Det er mine if a power law model
will adequat ely descr ibe t he behavior of t his mat er ial.
Using Eq. [3.51] and [3.55], t he shear r at e and shear st r ess can be
calculat ed dir ect ly fr om t he r aw dat a:
and

min
1(tan(0.0524)) 0.0524 rad/s(0.50 rpm)

max
100(tan(0.0524)) 0.524 rad/s(5.00 rpm)

3M
2R
3
M
2R
3

3

2R
3
K(

)
n
3
K 15.73
n 0.307
M
min

2(.05/2)
3
(15.73) (1)
.307
3
0.000515 N m 5, 150.0 dyne cm
M
max

2(.05/2)
3
(15.73) (100)
.307
3
0.00212 N m 21, 000 dyne cm



tan

3M
2R
3
228 Chapter 3. Rotational Viscometry
Table 3.4. Cone and Plate Data ( =25mm; =0.02 rad) for Kraft French Salad
Dressing at 22 C
r a d/ s N m 1/ s Pa
0.002 1.34E-4 0.10 4.09
0.005 1.65E-4 0.25 5.04
0.013 2.16E-4 0.63 6.60
0.025 2.59E-4 1.25 7.91
0.040 2.98E-4 1.99 9.11
0.063 3.42E-4 3.16 10.45
0.100 3.97E-4 5.01 12.13
0.159 4.58E-4 7.94 14.00
0.252 5.28E-4 12.59 16.13
0.399 6.24E-4 19.95 19.07
0.632 7.33E-4 31.62 22.40
1.002 8.70E-4 50.11 26.59
Figure 3.28. Rheogram for Kraft French salad dressing (22 C) determined from
cone and plate data.
R

M


0 10 20 30 40 50 60
0
5
10
15
20
25
30
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
French Dressing

3.8.13 Parallel Plate - Methylcellulose Solution 229


Regr ession analysis of t he r esult s (Table 3.4 and Fig. 3.28) yields a
st at ist ically accept able fit t o t he exper iment al dat a as
wher e = 7.64 Pa s
n
and = 0.303. These r esult s ar e differ ent t han
t hose found in Example Pr oblem 3.8.9 when evaluat ing Kr aft Fr ench
salad dr essing using t he infinit e cup assumpt ion wit h a concent r ic
cylinder viscomet er . Differ ences may be due t o var ious fact or s such as
analyt ical assumpt ions, nat ur al biological var iabilit y, and differ ences
in shear r at e cover age. Mor e exper iment al dat a would be r equir ed t o
r esolve t hese issues.
3.8.13. Parallel Plate - Methylcellulose Soluti on
Dat a for a 3% hydr oxypr opyl met hylcellulose solut ion have been col-
lect ed using a par allel plat e viscomet er (Table 3.5). Gener at e a r heo-
gr am for t his mat er ial.
Table 3.5. Parallel Plate Data ( =25mm; =0.70mm) for a3% Aqueous Solution
of Hydroxypropyl Methylcellulose (Methocel K4M, Dow Chemical Co.) at 24.2
C
ln ln
N m 1/ s Pa
0.116E-4 0.0127 -11.36 -4.37 0.5
0.211E-4 0.0198 -10.77 -3.92 0.8
0.334E-4 0.0317 -10.31 -3.45 1.3
0.442E-4 0.0503 -10.03 -2.99 1.7
0.807E-4 0.0797 -9.42 -2.53 3.2
1.259E-4 0.1262 -8.98 -2.07 4.9
2.029E-4 0.1999 -8.50 -1.61 7.9
2.979E-4 0.3166 -8.12 -1.15 11.7
4.536E-4 0.5016 -7.70 -0.69 17.8
6.687E-4 0.7945 -7.31 -0.23 26.2
9.343E-4 1.258 -6.98 0.23 36.6
12.900E-4 1.994 -6.65 0.69 50.5
17.270E-4 3.158 -6.36 1.15 67.6
22.700E-4 5.003 -6.09 1.61 88.8
29.260E-4 7.925 -5.83 2.07 114.5
37.320E-4 12.55 -5.59 2.53 146.0
The shear r at e at t he r im of t he plat e was det er mined, by consider ing
angular velocit y and geomet r y, fr om Eq. [3.59] as
7.64(

)
.303
K n
R h

R
M

R

R
230 Chapter 3. Rotational Viscometry
and shear st r ess was calculat ed wit h Eq. [3.66]:
Using linear r egr ession, t he following r elat ionship was det er mined fr om
t he t or que and shear r at e dat a (Fig. 3.29):
which ident ifies t he slope t er m as
Figure 3.29. Torque versus shear rate for 3% aqueous solution of hydroxypropyl
methylcellulose at 24.2 C.
This r esult can be used t o evaluat e t he shear st r ess at t he r im for each
t or que value:

R

R
h

R
f (

R
)
M
2R
3

3 +
d ln(M)
d ln(

R
)
1
1
]
ln(M) 7.12 + .934ln(

R
)
d ln(M)
d ln(

R
)
.934
-6 -4 -2 0 2 4
-12
-11
-10
-9
-8
-7
-6
-5
ln
l
n

(
M
)
R

3.8.14 End Effect Calculation for a Cylindrical Bob 231


Shear st r ess values calculat ed fr om t his equat ion ar e pr esent ed in Table
3.5 and plot t ed in Fig. 3.30. Using t he power law model, values of t he
consist ency coefficient and t he flow behavior index wer e det er mined:
= 25.3 Pa s
n
and = 0.83.
Figure 3.30. Flow behavior of a 3% aqueous solution of hydroxypropyl methyl-
cellulose (Methocel K4M, Dow Chemical Co.) at 24.2 C.
3.8.14. End Effect Calculati on for a Cyli ndri cal Bob
Det er mine t he bot t om end effect ( ) for t he bob ( = 1.95 cm) and cup
( = 2.00 cm) combinat ion illust r at ed in Fig. 3.31.

R

M
2R
3
[3 + .934]
K
n
0 5 10 15 20 25 30 35
0
50
100
150
200
250
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a

h
o
R
b
R
c
232 Chapter 3. Rotational Viscometry
Figure 3.31. Bob and cup combination tested for end effect correction.
Figure 3.32. End effect data for the "A-bob" of the Hercules high-shear
viscometer.
5.00 cm
0.60 cm
1.50 cm
4.00 cm
3.90 cm
0 1 2 3 4 5
0
100
200
300
400
500
600
h, cm
M
,

k
d
y
n
e

c
m
600 rpm 900 rpm 1100 rpm
h
17 degrees
3.8.15 Bob Angle for a Mooney-Couette Viscometer 233
The end cor r ect ion for t he bot t om of t he bob was evaluat ed fr om
t or que dat a t aken at t hr ee fixed bob speeds wit h differ ent values of
(Fig. 3.32) using t he t echnique pr esent ed in Sec. 3.5. Not e t hat is
measur ed fr om t he t op of t he cone shaped pr oject ion of t he bob. All t est s
wer e r un at r oom t emper at ur e wit h a Dow Cor ning 200 fluid which is
a dimet hyl-silicone mat er ial having t he following char act er ist ics at
25 C: Newt onian behavior wit h = 95.63 cP , and = 0.965 g cm
-3
.
Tor que dat a wer e plot t ed (Fig. 3.32) as at t hr ee speeds:
600 r pm, 900 r pm, and 1100 r pm. values wer e det er mined (Table
3.6) fr om linear r egr ession coefficient s: at , . Cor r elat ion
coefficient s wer e 0.99 for each cur ve. Result s showed t he magnit ude of
t he end effect incr easing wit h incr easing speed going fr om 0.141 cm at
600 r pm t o 0.182 cm at 1100 r pm. An aver age value of = 0.158 cm is
r ecommended for making pr act ical end effect cor r ect ions over t he r pm
r ange consider ed.
Table 3.6. Linear Regression Parameters ( ) Relating Torque
( ) and Length ( ) of Bob Immersed in Fluid to Evaluate End Effect
Correction ( )
600 r pm 900 r pm 1100 r pm
, dyne 61.74 92.48 113.72
, dyne cm 8.68 14.04 20.67
, cm 0.141 0.152 0.182
3.8.15. Bob Angle for a Mooney-Couette Vi scometer
The Mooney-Couet t e viscomet er is a concent r ic cylinder syst em con-
st r uct ed sot he bob has a conical bot t om which almost t ouches t he bot t om
sur face of t he cup (Fig. 3.8). If t he syst em is designed so t he shear r at e
in t he bot t om gap is equal t o t he shear r at e in t he annulus, t he end effect
can be significant ly r educed. Det er mine t he bot t om angle r equir ed t o
h
h

M ah + b
h
o
M 0 h
o
b/a
h
o
M ah + b
M h
h
o
a
b
h
o
234 Chapter 3. Rotational Viscometry
r edesign a concent r ic cylinder bob so t he shear r at e in t he conical and
annular gaps ar e equal. Also explain how t he end effect can be incor -
por at ed int o t he shear st r ess calculat ion. Assume t he bob under con-
sider at ion has t he following dimensions: , ,
.
The Newt onian appr oximat ion (Eq. [3.32]) can pr ovide a r easonable
est imat e of t he shear r at e in t he gap:
wher e:
One can solve for t he r equir ed cone angle by equat ing t he shear r at es
in t he conical sect ion, given by Eq. [3.51], t o t he shear r at e in t he gap:
Solving for yields
so
Ther efor e, making a bob wit h a bot t om angle (Fig. 3.8) of 2.56 degr ees
will r esult in an appar at us wher e t he shear r at e in t he annulus and
bot t om gap ar e appr oximat ely equal. If t he simple shear appr oximat ion
(Eq. [3.30]) is used in t he det er minat ion, a slight ly lar ger angle of 2.75
degr ees is calculat ed.
The end effect can be incor por at ed int o t he calculat ion of t he shear
st r ess at t he bob by finding a numer ical value for . When ignor ing t he
end effect , shear st r ess at bob is given by Eq. [3.3]:
and t he shear st r ess in t he conical sect ion is t aken fr om Eq. [3.55] as
R
c
21.00 mm R
b
20.04 mm
h 60.00 mm

b
2

2
1
_

,

R
c
R
b

21.00
20.04
1.048

tan()
2

2
1
_

,
tan()
tan()
1
2

2
1

2
_

,

1
2

1.048
2
1
1.048
2
_

,
0.0447
.0447 radians .0447

360
2
_

,
2.56 degrees
h
o

b

M
2hR
b
2
3.8.16 Viscous Heating 235
Wit h t he Mooney-Couet t e sensor t he r adius of t he cone and t he bob ar e
equal, , so t he above equat ion may be wr it t en as
[3.179]
Assuming t he shear st r ess cont r ibut ion in t he conical sect ion can be
calculat ed in t er ms of , t he shear st r esses for t he bob and t he cup can
be equat ed:
[3.180]
Solving for t he effect ive height yields
[3.181]
Using Eq. ]3.181], t he over all shear st r ess in t he Mooney-Couet t e syst em
can be calculat ed in t er ms of t he effect ive height :
[3.182]
3.8.16. Vi scous Heati ng
Consider ing a pr evious pr oblem (Example Pr oblem 3.8.6) dealing wit h
concent r ic cylinder dat a for t omat o ket chup, est imat e t he ext ent of
viscous heat ing t hat may occur dur ing dat a collect ion.
To consider t he wor st case, t ake t he highest shear r at e given, 664.78
s
-1
. Also, assume t he bob sur face is adiabat ic and t he cup wall is
maint ained at a const ant t emper at ur e equal t o . Then, t he maximum
t emper at ur e differ ence found at t he sur face of t he bob may be est imat ed
as (Eq. [3.80])
This expr ession can be evaluat ed aft er subst it ut ing an appar ent vis-
cosit y funct ion for t he Newt onian viscosit y and t he size of t he annulus
for :

3M
2R
3
R R
b

3M
2R
b
3
h
o
M
2h
o
R
b
2

3M
2R
b
3
h
o

R
b
3

b

M
2(h + h
o
)R
b
2

M
2(h + R
b
/3)R
b
2
T
o
T
max
T
o



()
2
s
2
2k
s
236 Chapter 3. Rotational Viscometry
Assuming a t her mal conduct ivit y of 0.516 W/(m K) and, using r heo-
logical pr oper t ies det er mined in Example Pr oblem 3.8.6 ( = 15.73 Pa
s
n
and = 0.307), yields
At emper at ur e incr ease of t his magnit ude will usually have a negligible
influence on exper iment al r esult s.
3.8.17. Cavi tati on i n Concentri c Cyli nder Systems
Given a concent r ic cylinder syst em like t he one illust r at ed in Fig. 3.31
( , , ), what bob speed is r equir ed t o
pr oduce cavit at ion in 40 C wat er ?
At at mospher ic pr essur e (101.35 kPa), wat er at 40 C has t he fol-
lowing pr oper t ies: = 7.38 kPa, = 992.2 kg m
-3
, . Con-
sider ing t he cavit at ion cr it er ion given in Sec 3.5 yields
When > 13.76 m/s, bob speed is 6738 r pm and cavit at ion may be
pr esent .
In checking t he t her modynamic pr oper t ies of wat er , one finds t hat
t he vapor pr essur e of wat er var ies fr om 1.4 kPa at 12 C t o 19.9 kPa at
60 C. Ther efor e, t he speed r equir ed t o achieve cavit at ion in wat er is
higher at lower t emper at ur es. Minimum speeds r equir ed t o pr oduce
cavit at ion can be incr eased in pr essur ized syst ems: They will decr ease
in syst ems held under a vacuum.
It is int er est ing t o compar e t he speed r equir ed for cavit at ion and t he
for mat ion of Taylor vor t ices. Vor t ex for mat ion is expect ed when t he
inequalit y expr essed by Eq. [3.90] is sat isfied:
or
T
max
T
o

K(

)
n 1
(

)
2
(R
c
R
b
)
2
2k
K
n
T
max
T
o

15.73(664.78)
.307 1
(664.78)
2
(.02100 .02004)
2
2(.516)
.069 C
R
b
1.95 cm R
c
2.00 cm h 5.00 cm

P
vap
0.6529 cP
u >

2(P
atm
P
vap
)

2(101.35 7.38) (1000)


992.2
13.76 m/s
u

R
b
(R
c
R
b
)

> 41.3

R
c
R
c
R
b
3.8.18 Mixer Viscometry 237
Meaning t hat Taylor vor t ices can be expect ed when bob speeds exceed
168.3 r pm. In t his example, t he minimum bob speed r equir ed for vor t ex
for mat ion is much lower t han t he minimum speed needed for t he onset
of cavit at ion. It would be unusual t o develop cavit at ion pr oblems when
t est ing food pr oduct s. In gener al, t he laminar flow assumpt ion would
be violat ed befor e t he onset of cavit at ion.
3.8.18. Mi xer Vi scometry
A mixer viscomet er was used t o det er mine t he r heological pr oper t ies of
a st ar ch-t hickened baby food pr oduct , st r ained apr icot s at 25 C. The
mat er ial was mixed sufficient ly t o r emove t ime-dependent effect s.
Mixer viscomet er dat a wer e collect ed using a r ot at ional viscomet er
equipped wit h a cup (inside r adius = 0.021 m), and Haake pit ched-paddle
impeller (blade diamet er = 0.04143 m, blade height = 0.02692 m)
illust r at ed in Fig. 3.11. Rheological pr oper t ies ( and ) st at ed in Table
3.7 wer e det er mined using concent r ic cylinder dat a, wit h Eq. [3.3] t o
calculat e shear st r ess, t he Kr ieger met hod t o calculat e shear r at e (Eq.
[3.47]), and st andar d st at ist ical met hods t o evaluat e t he r heogr am.
Questi ons: Part a. Taking t he r aw dat a for cor n syr up and var ious
aqueous solut ions found in Table 3.7, det er mine t he mixer viscomet er
const ant ( ) using t he slope met hod. Part b. Using t he dat a for st r ained
apr icot s given in Table 3.8, calculat e t he flow behavior index and t he
consist ency coefficient of t he pr oduct . Compar e differ ent met hods of
evaluat ion. Part c. Using t he dat a for st r ained apr icot s (Table 3.8),
det er mine an effect ive viscosit y flow cur ve for t his mat er ial which may
be used for qualit y cont r ol pur poses.
Part a. Fir st , t he r aw dat a ar e manipulat ed int o t he appr opr iat e for m
(Table 3.7) and plot t ed (Fig. 3.33). Regr ession analysis of
ver sus ( ) as specified in Eq. [3.115],
(.0195) (.02 .0195)992.2
.0006529
> 41.3

.02
.02 .0195
> 17.63 rad/s or 168.3 rpm

K n
k
log
10
(P/(K
n + 1
d
3
)) 1 n
238 Chapter 3. Rotational Viscometry
Table 3.7. Raw (first threecolumns) and Manipulated Dataused toDetermine the
Mixer Viscometer Constant. [ ] Data from: Steffe and
Ford (1985)
Flu id
(Pa s
n
) (-) (N m/ s ) (-)
2.5% 16.55 0.513 0.0585 0.487 3.0882 0.4897
Hydr oxypr opyl 15.56 0.520 0.0573 0.480 3.1762 0.5019
Met h ylcellu los e 16.61 0.505 0.0577 0.495 3.0799 0.4885
15.90 0.513 0.0562 0.487 3.0881 0.4897
2.0% 7.29 0.574 0.0300 0.426 3.2141 0.5071
Hydr oxypr opyl 6.94 0.586 0.0311 0.414 3.4236 0.5345
Met h ylcellu los e 6.72 0.588 0.0306 0.412 3.4661 0.5398
7.06 0.576 0.0308 0.424 3.3948 0.5308
1.5% 2.10 0.676 0.0129 0.324 3.9776 0.5996
Hydr oxypr opyl 2.16 0.672 0.0127 0.328 3.8352 0.5838
Met h ylcellu los e 2.16 0.675 0.0130 0.325 3.9042 0.5915
2.10 0.680 0.0129 0.320 3.9484 0.5964
1.0% 1.84 0.560 0.0072 0.440 3.1532 0.4988
Hydr oxypr opyl 1.26 0.605 0.0058 0.395 3.4136 0.5332
Met h ylcellu los e 0.85 0.683 0.0054 0.317 4.0685 0.6094
1.83 0.504 0.0063 0.496 3.0385 0.4827
Cor n Syr u p 2.70 1.000 0.0488 0.000 6.4520 0.8097
2.84 0.992 0.0474 0.008 6.0462 0.7815
2.64 1.000 0.0482 0.000 6.5175 0.8141
1.5% Gu a r Gu m 30.98 0.158 0.0328 0.842 1.7762 0.2495
31.65 0.159 0.0339 0.841 1.7936 0.2537
27.39 0.169 0.0316 0.831 1.8968 0.2780
1.0% Gu a r Gu m 7.28 0.275 0.0113 0.725 2.1002 0.3223
4.93 0.292 0.0083 0.708 2.2159 0.3455
7.99 0.266 0.0121 0.734 2.0832 0.3187
8.62 0.237 0.0123 0.763 2.0704 0.3160
10.76 0.224 0.0143 0.776 1.9749 0.2955
gives a good st r aight line fit ( ) wit h , and a slope
equal t o -0.6503 meaning t hat
so
60 rpm 6.283 rad/s
K n P M 1 n P
K
n +1
d
3
log
10

P
K
n +1
d
3
_

,
log
10

P
K
n + 1
d
3
_

,
log
10
(A) (1 n) log
10
(k)
r
2
0.99 log
10
A 0.803
log
10
(k) 0.6503
3.8.18 Mixer Viscometry 239
Also since , r ad
-1
.
Figure 3.33. Plot of data needed to determine the mixer viscometer constant ( )
using the slope method.
Part b. Raw dat a ar e manipulat ed int o t he appr opr iat e for m (Table
3.8). Evaluat ion of Eq. [3.121],
using linear r egr ession analysis yields and
wit h . may be found wit hout using t he numer ical
value of , as , fr om Eq. [3.125],
Taking dat a at 60 r pm (mid-r ange value) and t he aver age values of t he
1%hydr oxypr opyl met hylcellulose as t he known fluid gives: = 0.00533
k 4.47 rad
1
log
10
0.803 A 6.35
1-n
Mixer Viscometer Data
0.8
0.9
0.7
0.6
0.5
0.4
0.3
0.2
0 0.2 0.4 0.6 0.8 1.0
l
o
g
1
0
[
P
/
(
K





d


)
n
+
1
3
k
log
10
(M) log
10
(d
3
AK(k)
n 1
) + n log
10
()
log
10
[d
3
AK(k)
n 1
] 2.597
r
2
0.99 n 0.378 K
A K
x
K
x
K
y

M
x
M
y
_

y
n
y
(k)
n
y

x
n
x
(k)
n
x
_

,
M
x
240 Chapter 3. Rotational Viscometry
Table 3.8. Raw (first two columns) and Manipulated Mixer Viscometer Data for
Strained Apricots at 25 C
Speed
(r pm) (r a d/ s ) (N m) (Pa s ) (1/ s ) (Pa s )
20 2.094 0.00332 0.321 -2.479 3.511 9.36 22.30
40 4.189 0.00429 0.622 -2.368 2.268 18.72 14.40
60 6.283 0.00533 0.798 -2.273 1.879 28.09 11.93
80 8.378 0.00556 0.923 -2.255 1.470 37.45 9.33
100 10.47 0.00607 1.020 -2.217 1.284 46.81 8.15
120 12.57 0.00655 1.099 -2.184 1.154 56.17 7.33
N m; = 6.283 r ad/s; = 0.378; = 4.47 r ad
-1
; = 0.588; =
0.000984 N m; = 1.45 Pa s
n
. Subst it ut ing t hese values int o t he above
equat ion yields
Each "st andar d" or r efer ence fluid used in det er mining gives a slight ly
differ ent value of . Var iat ions of 5 t o 10% have been obser ved wit h
t his met hod of calculat ion (St effe and For d, 1985).
The consist ency coefficient may also be found using t he known value
of and t he int er cept det er mined fr om our evaluat ion of Eq. [3.121]:
Subst it ut ing known values gives
which can be solved for t he consist ency coefficient :
This compar es r easonably well wit h t he value of 15.77 Pa s
n
found
pr eviously, and pr obably gives a mor e accur at e indicat ion of sample
behavior .
Wit h a known value of , power law fluid pr oper t ies can also be
det er mined by applying t he mat ching viscosit y assumpt ion t o t he r aw
dat a. An aver age appar ent viscosit y may be found using Eq. [3.116]
wr it t en as

M log
10
log
10
M

a
M
1
d
3

x

y
n
x
k n
y
M
y
K
y
K
x

(.00533) (1.45) (6.28)
.588
(4.47)
.588
(.000984) (6.28)
.378
(4.47)
.378
15.77 Pa s
n
k
K
x
A
log
10
[d
3
AK(k)
n 1
] 2.597
log
10
[(.04143)
3
(6.35)K(4.47)
.378 1
] 2.597
K 14.22 Pa s
n
k
3.8.18 Mixer Viscometry 241
Since was found t o be 6.35 r ad
-1
and t he diamet er was given ( =
0.04143 m), appar ent viscosit y can be dir ect ly calculat ed fr om t he
t or que:
The mixer viscomet er const ant was det er mined t o be 4.47 r ad
-1
so t he
aver age shear r at e is
and , evaluat ed using t he above equat ions, ar e given in Table 3.8.
Assuming power law fluid behavior , t hese t er ms ar e r elat ed as
or
Regr ession analysis of t he dat a, using t his r elat ionship, yields
and Pa s
n
. Since and ar e dir ect ly pr opor t ional t o and ,
r espect ively, r esult s ar e t he same as t hose found pr eviously. Not e t hat
aver age shear st r ess could also be calculat ed as , and r egr ession
analysis could be pr efor med on ver sus .
Part c. An effect ive viscosit y can be calculat ed on t he basis of Eq.
[3.144]:
Numer ical values of t his t er m ar e given in Table 3.8 and plot t ed as a
funct ion of angular velocit y in Fig. 3.34. Assuming effect ive viscosit y
is dir ect ly pr opor t ional t o appar ent viscosit y and angular velocit y is
dir ect ly pr opor t ional t o shear r at e, t he viscosit y may be modeled as a
power law funct ion:

d
2

A
(N
Po
)
d
2

M
d
5

,

M
Ad
3

A d

M
6.35(.04143)
3

a
k 4.47

a
K(

a
)
n 1
ln ln K + (n 1) ln

a
n 0.378
K 14.22

a
M
/

effective

M
d
3
242 Chapter 3. Rotational Viscometry
Figure 3.34. Effective viscosity of strained apricots at 25 C determined from
mixer viscometry data.
[3.183]
wher e may be defined as an effect ive consist ency coefficient (Pa )
and as an effect ive flow behavior index (dimensionless). Per for ming
a r egr ession analysis on t he dat a gave
meaning = 35.59 Pa and = 0.378. This expr ession is plot t ed as
t he line displayed in Fig. 3.34. Due t o t he definit ion of effect ive viscosit y,
t he value of is unique t o t he fluid in quest ion wit h r espect t o t he
exper iment al equipment employed; , however , is numer ically t he same
as t he value of t he flow behavior index found in Par t b. These r esult s
could be compar ed t o t hose fr om a r efer ence fluid (pr oduct found
accept able in t he mar ket place) and used in making qualit y cont r ol
decisions in a commer cial food pr ocessing oper at ion.
0 2 4 6 8 10 12
0
5
10
15
20
25
Angular Velocity, rad/s
E
f
f
e
c
t
i
v
e

V
i
s
c
o
s
i
t
y
,

P
a

s

effective
K( )
n 1
s
n

K
n

effective
35.59( )
.623
s
n

K n
K
n
3.8.19 Vane Method - Sizing the Viscometer 243
3.8.19. Vane Method - Si zi ng the Vi scometer
The vane met hod will be used t o invest igat e var ious food pr oduct s t hat
have yield st r esses r anging fr om 10 t o 150 Pa. Vane speed is 0.1 r pm
(or as low as possible) and t he dimensions of t he vane ar e = 26.15 mm
and = 1.92. It is pr oposed t hat one of t he following Br ookfield vis-
comet er s be used for t his t ask (Not e, 1 N m = 10
7
dyne cm):
Inst r ument Model Full Scale Tor que Range (dyne cm)
LV Ser ies 673.7
RV Ser ies 7,187
HAT Ser ies 14,374
HBT Ser ies 57,496
Part a. Det er mine t he feasibilit y of using one of t he above viscomet er s
for t he pr oposed yield st r ess measur ement . Part b. Design an appr o-
pr iat e sample cont ainer for yield st r ess measur ement and specify t he
minimum sample size r equir ed for conduct ing each exper iment .
Part a. The t or que r ange r equir ed t o det er mine t he specified yield
st r esses must be evaluat ed. Assuming t he vane is fully immer sed in
t he sample, and , t he peak t or que is evaluat ed fr om Eq. [3.141] as
Ther efor e t he r ange of maximum t or que r esponse will be
or
or
Accor ding t o t he t or que capabilit ies of t he inst r ument s, none ar e suf-
ficient t o handle t he job. The HBT, which has t he maximum t or que
capacit y, can measur e a yield st r ess wit h t he following maximum value:
d
h/d
m 0
M
o

o

d
3
2
_

h
d
+
1
3
_

,
1
1
]

o

(.02615)
3
2
_

1.92 +
1
3
_

,
1
1
]
6.329(10
5
)
o
6.329(10
5
) (10) < M
o
< 6.329(10
5
) (150)
6.329(10
4
) N m < M
o
< 9.494(10
3
) N m
6, 329 dyne cm < M
o
< 94, 940 dyne cm

57496
6.329(10
5
)
_

,
10
7
90.85 Pa
244 Chapter 3. Rotational Viscometry
One solut ion t o t his pr oblem is t o wor k wit h Br ookfield t o "special or der "
an inst r ument wit h a lar ger spr ing const ant . Alt er nat ively, a smaller
vane could be ut ilized t o r educe t he t or que r equir ement s of t he
inst r ument .
Part b. Typical cont ainer dimensions, suggest ed in Sec. [3.7.3] and
illust r at ed in Fig. 3.14, may be calculat ed fr om t he size of t he vane. If
m and , t hen m. The minimum cont ainer
size can be calculat ed fr om t his infor mat ion:
m
m
m
Wit h t hese r esult s, t he minimum sample volume may be det er mined:
A minimum sample volume of 192.1 cm
3
is needed t o per for m t he t est .
3.8.20. Vane Method to Fi nd Yi eld Stresses
Calculat e t he st at ic and dynamic yield st r esses, using bot h t he single
point and slope met hods, wit h t he dat a for t omat o ket chup found in
Table 3.9. Dat a for calculat ing st at ic yield st r esses wer e collect ed using
undist ur bed samples. Dynamic st r ess dat a wer e t aken fr om st ir r ed
samples wher e gent le agit at ion eliminat ed t he t ime-dependent st r uc-
t ur e of t he mat er ial. Vanes wer e complet ely immer sed in t he sample
following t he guidelines for vane and vessel dimensions given in Sec.
3.3.7 and summar ized in t he pr evious example pr oblem.
Eq. [3.141] is needed t o calculat e t he yield st r ess using t he single
point met hod:
d 0.02615 h/d 1.92 h 0.05021
D/d 2.0 so D 2.0(d) 0.0523
Z
1
/d 1.0 so Z
1
1.0(d) 0.02615
Z
2
/d 0.5 so Z
1
0.5(d) 0.01308
min. sample

D
2
_

,
2
(z
1
+ h + z
2
)
min. sample

0.0523
2
_

,
2
(0.02615 + 0.05021 + 0.01308) 1.921(10
4
) m
3
192.1 cm
3
3.8.20 Vane Method to Find Yield Stresses 245
Table 3.9. Raw Data, and Yield Stresses, for Tomato Ketchup at 21 C Taken with
aBrookfield Viscometer Having aFull ScaleTorque of 0.00575 N m Using aVane
with 4 Blades having a Diameter of 2.5 cm.
St r es s Tor qu e Tor qu e Va n e Heigh t Yield St r es s
1
(% t ot a l) (N m) (cm) (Pa )
St a t ic 30.00 0.00173 5.0 30.12
St a t ic 29.25 0.00168 5.0 29.37
St a t ic 27.50 0.00158 5.0 27.61
St a t ic 26.00 0.00150 5.0 26.11
St a t ic 34.00 0.00196 6.6 26.79
St a t ic 37.25 0.00214 6.6 29.35
St a t ic 38.00 0.00219 6.6 29.94
St a t ic 47.50 0.00273 9.0 28.29
St a t ic 48.00 0.00276 9.0 28.59
St a t ic 48.75 0.00280 9.0 29.04
Aver a ge 28.52
SDEV 1.33
Dyn a mic 22.50 0.00129 5.0 22.59
Dyn a mic 22.50 0.00129 5.0 22.59
Dyn a mic 21.00 0.00121 5.0 21.08
Dyn a mic 20.00 0.00115 5.0 20.08
Dyn a mic 25.33 0.00146 6.6 19.96
Dyn a mic 26.50 0.00152 6.6 20.87
Dyn a mic 27.50 0.00158 6.6 21.67
Dyn a mic 37.00 0.00213 9.0 22.04
Dyn a mic 35.00 0.00201 9.0 20.85
Dyn a mic 36.00 0.00207 9.0 21.44
Dyn a mic 34.75 0.00200 9.0 20.70
Aver a ge 21.26
SDEV 0.90
1
Yield St r es s ca lcu la t ed by t h e s in gle poin t met h od, Eq. [3.141].
Assuming , t his expr ession can be r educed t o

o

2M
o
d
3

h
d
+
1
m + 3
_

,
1
m 0

o

6M
o
d
2
(3h + d)
246 Chapter 3. Rotational Viscometry
At ypical calculat ion can be used t o illust r at e t he met hod. If, for example
(see fir st dat a line, Table 3.9), , ,
and , t hen . Result s for each t r ial, as well as
aver age values, ar e given in Table 3.9.
Eq. [3.137] is r equir ed t o calculat e t he yield st r ess using t he slope
met hod:
Not ing t hat t he int egr al t er m is const ant , ver sus may be plot t ed
as
[3.184]
The slope ( ) of t he best fit t ing line t hr ough t he dat a point s will include
t he yield st r ess:
[3.185]
t her efor e,
[3.186]
This t echnique must be per for med separ at ely for st at ic and dynamic
dat a set s. Dat a ar e plot t ed (Fig. 3.35), and slopes ar e found fr om a
st andar d linear r egr ession pr ogr am, allowing t he st at ic and dynamic
yield st r esses t o be calculat ed fr om Eq. [3.186]:
and
Reasonably good agr eement was found bet ween yield st r esses calculat ed
using t he slope and single point met hods (Table 3.10). Single point yield
st r esses ar e aver aged values t aken fr om Table 3.9.
M
o
0.3(0.00575) 0.00173 N m d 0.025 m
h 0.050 m
o
30.12 Pa
M
o

hd
2
2

o
+ 4

0
d/ 2
r
2

e
dr
M
o
h
M
o
a h + constant
a
a
d
2

o
2

o

2a
d
2
(
o
)
static

2a
d
2

2(0.0285)
(0.025)
2
29.03 Pa
(
o
)
dynamic

2a
d
2

2(0.0205)
(0.025)
2
20.88 Pa
3.8.21 Vane Rotation in Yield Stress Calculation 247
Figure 3.35. Plot of the raw data needed to determine the yield stress using the
slope method.
Table 3.10. Comparison of Yield Stresses Calculated for Tomato Ketchup at 21 C.
Met h od St a t ic Yield St r es s Dyn a mic Yield St r es s
(Pa ) (Pa )
Sin gle Poin t 28.52 21.26
Slope 29.03 20.88
3.8.21. Vane Rotati on i n Yi eld Stress Calculati on
Typical dat a, for undist ur bed t omat o ket chup at 23 C, collect ed t o
det er mine t he yield st r ess using t he vane met hod, ar e illust r at ed in Fig.
3.36 and 3.37. The dat a in each figur e ar e t aken wit h t wo inst r ument s
having differ ent wind-up char act er ist ics: t he Br ookfield uses a coiled
spr ing and t he Haake inst r ument uses a t or sion bar t ype syst em (Fig.
3.19). Bot h syst ems have r ot at ing vanes and st at ionar y sample holder s.
Given t he char act er ist ics of each inst r ument (Table 3.11), det er mine
t he amount of vane r ot at ion (Fig. 3.38) which has occur r ed bet ween t he
st ar t of t he t est and t he t ime t he peak t or que value ( ) is obser ved.
Also, compar e yield st r esses calculat ed fr om each cur ve (Fig. 3.36 and
Fig. 3.37) using t he single point met hod.
0
0.0005
0.001
0.0015
0.002
0.0025
0.003
h (m)
M

(
N

m
)
Static Dynamic
M = 0.0285 h + 0.000199
M = 0.0205 h + 0.000192
o
o
o
.04 .05 .06 .07 .08 .09 0.1

M
o
248 Chapter 3. Rotational Viscometry
Figure 3.36. Torque response for 23 C ketchup obtained using the Brookfield
HBDT viscometer and a vane with 4 blades.
Figure 3.37. Torque response for 23 C ketchup obtained using the Haake VT500
viscometer and a vane with 4 blades.
0 5 10 15 20 25 30 35
0
0.05
0.1
0.15
0.2
Time, s
T
o
r
q
u
e
,

N

c
m
Instrument: Brookfield HBDT
Speed = 0.5 rpm
Vane: h=6.6 cm, d=2.5 cm

0 1 2 3 4 5 6 7
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Time, s
T
o
r
q
u
e
,

N

c
m
Instrument: Haake VT500
Speed = 2.0 rpm
Vane: h=6.0 cm, d=4.0 cm

3.8.21 Vane Rotation in Yield Stress Calculation 249


Figure 3.38. Movement of a 4-bladed vane during yield stress testing.
Table 3.11. Vane Rotation Data and Instrument Characteristics for Two Vis-
cometers.
Ha a ke Br ookfield
VT500 HBDT
Va n e h eigh t ( ), cm 6.0 6.6
Va n e dia met er ( ), cm 4.0 2.5
Fu ll Sca le Tor qu e, N cm 2.0 0.575
Fu ll Sca le Win d-u p, degr ee 1.0 80
Time t o Pea k Tor qu e, s 1.3 18
Pea k Tor qu e ( ), N cm 0.605 0.184
Tes t Speed, r ev/ min (degr ee/ s ec) 2 (12) 0.5 (3.0)
Zer o-Loa d Rot a t ion a t Pea k Tor qu e, degr ee 15.6 54.0
Win d-u p a t Pea k Tor qu e, degr ee 0.3 25.6
Act u a l Rot a t ion a t Pea k Tor qu e, degr ee 15.3 28.4
Yield St r es s ( ) fr om Eq. [3.141], Pa 32.8 28.4
Initial Vane Position
Vane Position at Peak Torque
Angle of Rotation
h
d
M
o

o
250 Chapter 3. Rotational Viscometry
The maximum t or que measur ed in t he VT500 t est was 0.605 N cm.
Since t he full scale t or que of t his inst r ument is 2 N cm and t he full scale
wind-up is 1 degr ee, t he wind-up in t his exper iment was appr oximat ely
0.605 (1) / 2 = 0.3 degr ees. The t est was r un wit h an impeller speed of
2 r pm or 12 degr ees /s. Taking t he t ime t o achieve peak t or que as 1.3
s means t hat an unr est r ict ed impeller would t r avel 1.3 (12) = 15.6
degr ees. This is r efer r ed t o as t he zer o-load r ot at ion. Aft er account ing
for wind-up, t he act ual VT500 vane r ot at ion can be calculat ed as 15.6-0.3
= 15.3 degr ees. The compar able value for t he Br ookfield is 28.4 degr ees
(Table 3.11).
St at ic yield st r esses wer e calculat ed using Eq. [3.141] wit h t he
assumpt ion t hat = 0. The value measur ed wit h t he Br ookfield ( =
25.1 Pa) is lower t han t he value measur ed by t he VT500 ( = 32.8 Pa).
This is due, in par t , t o t he differ ent wind-up char act er ist ics of each
viscomet er : At full scale t or que, t he VT500 wind-up is 1 degr ee and t he
Br ookfield wind-up is appr oximat ely 80 degr ees. The combinat ion of
vane speed and wind-up caused a gr eat er amount of vane r ot at ion in
t he Br ookfield dur ing t est ing. This pr oduced gr eat er fluid mot ion which
subsequent ly r educed peak t or que values. Differ ences in t he wind-up
char act er ist ics of r ot at ional viscomet er s should always be t aken int o
consider at ion when evaluat ing yield st r esses det er mined using t he vane
met hod. These differ ences may be par t icular ly impor t ant when t he
yield st r esses under consider at ion exhibit st r ong t ime-dependent
behavior .
3.8.22. Rheomalaxi s from Mi xer Vi scometer Data
Tor que decay dat a for st ar ch t hickened st r ained apr icot s wer e collect ed
at var ious angular velocit ies (Table 3.12, Fig. 3.39). The mat er ial was
r heomalact ic, i.e., no st r uct ur al r ecover y was evident in a r easonable
per iod of t ime (t hr ee hour s). Evaluat e t he r elat ive degr ee of ir r ever sible
t hixot r opy in t his mat er ial by calculat ing t he ar ea bet ween t he init ial
and equilibr ium t or que cur ves. Also det er mine t he ener gy input t o t he
sample when t he impeller is r ot at ed at 3.14 r ad/s for appr oximat ely 10
minut es. (Not e: dat a in Table 3.12 could also be used t o evaluat e t he
par amet er s in t he t hixot r opy model given by Eq. [1.36].)
Init ial and equilibr ium t or que values, obt ained fr om t he r aw dat a,
ar e plot t ed on Fig. 3.40 and summar ized in Table 3.13. Fit t ing t hese
cur ves t o a power equat ion (any equat ion which adequat ely fit s t he dat a
would be accept able) yields:
m
o

o
3.8.22 Rheomalaxis from Mixer Viscometer Data 251
Table 3.12. Torque Decay Data for Starch Thickened, Strained Apricots at 22 C
Collected with a Rotational Viscometer using a Cup ( =21.00 mm) and Pitched
Paddle Impeller (Fig. 3.11)
Tor qu e, N m
Time (min .) 1.05 r a d/ s 1.57 r a d/ s 2.09 r a d/ s 2.62 r a d/ s 3.14 r a d/ s
0.42 0.00301 0.00324 0.00369 0.00404 0.00424
1.223 0.00259 0.00292 0.00324 0.00348 0.00369
2.023 0.00240 0.00280 0.00310 0.00330 0.00354
2.821 0.00238 0.00272 0.00299 0.00322 0.00340
3.623 0.00230 0.00266 0.00293 0.00312 0.00332
4.421 0.00224 0.00260 0.00287 0.00306 0.00326
5.223 0.00219 0.00255 0.00284 0.00303 0.00321
6.023 0.00214 0.00251 0.00279 0.00298 0.00315
6.821 0.00210 0.00245 0.00273 0.00294 0.00311
7.623 0.00205 0.00242 0.00270 0.00292 0.00309
8.421 0.00201 0.00238 0.00267 0.00287 0.00304
9.223 0.00198 0.00237 0.00265 0.00286 0.00301
10.02 0.00194 0.00234 0.00262 0.00282 0.00298
10.83 0.00192 0.00230 0.00260 0.00280 0.00297
11.63 0.00190 0.00228 0.00258 0.00277 0.00295
12.43 0.00188 0.00227 0.00255 0.00275 0.00293
13.23 0.00186 0.00224 0.00254 0.00273 0.00289
14.03 0.00184 0.00223 0.00253 0.00270 0.00288
14.83 0.00182 0.00220 0.00250 0.00269 0.00286
15.63 0.00180 0.00218 0.00248 0.00267 0.00286
and
The ar ea bet ween t he t wo cur ves is found fr om Eq. [3.143]:

R
c
M
initial
.00290()
.33
M
equilibrium
.00179()
.42

low

high
[M
initial
M
equlibrium
]d
252 Chapter 3. Rotational Viscometry
Figure 3.39. Torque decay data for strained apricots at 22 C.
Figure 3.40. "Initial" and "equilibrium" torque curves for apricots at 22 C.
0 5 10 15
0.0015
0.002
0.0025
0.003
0.0035
0.004
0.0045
Time, min.
T
o
r
q
u
e
,

N

m
3.14 2.62 2.09 1.57 1.05
Angular Velocity, rad/s
Strained Apricots

1 1.5 2 2.5 3 3.5


0.0015
0.002
0.0025
0.003
0.0035
0.004
0.0045
Angular Velocity, rad/s
T
o
r
q
u
e
,

N

m
.42 min. 15.6 min.
Strained Apricots

3.8.22 Rheomalaxis from Mixer Viscometer Data 253


Evaluat ing t he int egr al wit h t he appr opr iat e t or que cur ves and limit s
yields
is a r elat ive measur e of t hixot r opy; when none is pr esent
Table 3.13. "Initial" and "Equilibrium" Torque values , for Strained Apricots , at Different
Angular Velocities
Tor qu e, N m
"In it ia l" "Equ ilibr iu m"
(r a d/ s ) (a ft er 0.42 min .) (a ft er 15.6 min .)
1.05 .00301 .00180
1.57 .00324 .00218
2.09 .00369 .00248
2.62 .00404 .00267
3.14 .00424 .00286
Tor que decay cur ves also pr ovide a pr act ical way of evaluat ing
t hixot r opy by consider ing t he ener gy input t o t he sample. Dat a (Table
3.12) t aken at an impeller angular velocit y of 3.14 r ad/s may be r eplot t ed
in t er ms of power ( ) ver sus t ime. The ar ea under t he r esult ing
power decay cur ve (Fig. 3.41) r epr esent s t he mechanical ener gy input
t o t he sample.
The power decay dat a wer e fit t o a simple mat hemat ical model
yielding
which is t he line plot t ed on Fig. 3.41. The exponent in t his equat ion is
an index of t hixot r opy: Mor e negat ive values indicat e a r apid r at e of
st r uct ur al br eakdown, and a value of zer o means t he sample is not
t ime-dependent . Using t he above equat ion, t he ener gy input over t he
per iod of t ime (Table 3.12) under consider at ion (t = 0.42 min = 25.2 s t o
t = 10.02 min = 601.2 s) is easily calculat ed:

1.05
3.14
(.00290()
.33
.00179()
.42
)d .00261 N m s
1
0.

P M
P .0186(t )
.108
Energy Input

t 25.2
t 601.2
.0186(t )
.108
dt 5.91 N m
254 Chapter 3. Rotational Viscometry
In cont r ast , if t her e was no t hixot r opy, t he ener gy r equir ed t o maint ain
= 3.14 r ad/s over t he same t ime would be appr oximat ely
0.00424(3.14)(601.2 - 25.2) = 7.67 N m. This calculat ion assumes t he
power level is const ant and equal t o t he init ial value when no t ime-
dependent behavior is pr esent .
Figure 3.41. Power decay curves obtained for strained apricots at 22 C with a
constant impeller velocity of 3.14 rad/s.
Tor que or power decay dat a cur ves can pr ovide a useful measur e of
t hixot r opy. Result s depend on t he par t icular mixing syst em and sample
volume used in t he st udy. These fact or s must be kept const ant t o make
meaningful sample t o sample compar isons. Temper at ur e is also a
significant fact or which should be car efully cont r olled dur ing t est ing.

0 100 200 300 400 500 600


0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
Time, s
P
o
w
e
r
,

N

m

/

s
Strained Apricots
Angular Velocity = 3.14 rad/s

Chapt e r 4 . Ext e ns ional Flow


4.1. Introducti on
A basic knowledge of ext ensional viscosit y is essent ial for under -
st anding many flow sit uat ions found in t he food indust r y. The t hr ee
basic t ypes of ext ensional flow - uniaxial, biaxial, and planar - along
wit h numer ous applicat ions of ext ensional defor mat ion ar e descr ibed
in Sect ion 1.7. Exper iment al met hods t odet er mine ext ensional viscosit y
ar e discussed in t his chapt er . Techniques involving t ension (st r et ching
or pulling), squeezing flow (flow bet ween par allel plat es being pushed
t oget her ), conver ging flow int o a die, flow int o opposing jet s, spinning,
and t he t ubeless siphon phenomenon (Fano flow) ar e pr esent ed. In
addit ion, alt er nat ive int er pr et at ions of squeezing flow dat a ar e con-
sider ed because of t he r elat ive ease wit h which t his exper iment can be
conduct ed in a labor at or y. Biaxial ext ension, pr oduced by lubr icat ed
squeezing flow, is consider ed in t er ms of st eady shear pr oper t ies (Sec.
4.8.1). Similar consider at ions ar e made in t he case of non-lubr icat ed
squeezing flow (Sec. 4.8.2) which gener at es a combined shear and
ext ensional defor mat ion.
4.2. Uni axi al Extensi on
Consider t he uniaxial ext ension (Fig. 1.24) of a mat er ial wher e one
end is st at ionar y and t he ot her end is moving at some velocit y, . This
can be accomplished exper iment ally if one end is fixed (by gluing or
clamping) and t he ot her end is pulled by a moving clamp, wr apping it
ar ound a r od, or pulling it bet ween r ot at ing gear s. The differ ent ial
Hencky st r ain (Eq. [1.2]: ) descr ibing displacement is
[4.1]
and t he st r ain r at e is
[4.2]
Since is t he velocit y at t he end of t he sample, t he st r ain r at e can
be expr essed as
[4.3]
u
1

h
ln(L/L
o
)
d
h

dL
L

h

d
h
dt

1
L
dL
dt
dL/dt

h

u
1
L
256 Chapter 4. Extensional Flow
If t he velocit y is held const ant ( ) dur ing exper iment at ion, t hen t he
st r ain r at e would cont inually decr ease dur ing elongat ion due t o
incr easing values of . At ver y low speeds, a const ant velocit y t est may
appr oximat e a const ant st r ain r at e; however , t o maint ain a t r uly
const ant value of t he st r ain r at e, t he velocit y of t he moving end of t he
sample must be incr eased dur ing t est ing.
Int egr at ing Eq. [4.2] for a const ant st r ain r at e ( ), using as t he
init ial sample lengt h, gives
[4.4]
r esult ing in t he following expr ession for as a funct ion of t ime:
[4.5]
To pr oduce a const ant st r ain r at e, Eq. [4.3] shows t hat .
Combining t his expr ession and Eq. [4.5] demonst r at es t hat t he velocit y
at t he moving end of t he sample must exponent ially incr ease over t ime
t o maint ain a const ant st r ain r at e:
[4.6]
Assuming t he mat er ial is incompr essible, t he volume is const ant so
[4.7]
wher e is t he init ial cr ossect ional ar ea. Combining Eq. [4.5] and [4.7]
shows t he cr ossect ional ar ea of t he sample must decr ease exponent ially
dur ing t est ing as a consequence of t he exponent ial change in lengt h:
[4.8]
The nor mal st r ess differ ence over t he mat er ial defines t he net t ensile
st r ess:
[4.9]
is t he for ce r equir ed t o st r et ch t he sample. Using Eq. [4.7], t he
st r et ching st r ess can be expr essed as
[4.10]
In an ideal st r et ching exper iment t he st r ain is incr eased t o a
const ant value ( ) inst ant aneously:
[4.11]
u
1
u
o
L

ho
L
o

0
t

ho
dt

L
o
L
dL
L
L
L f (t ) L
o
exp(

ho
t )
L u
1
/

ho
u
1
f (t )

ho
L
o
exp(

ho
t )
AL A
o
L
o
A
o
A f (t ) A
o
exp(

ho
t )

E

11

22

F
A
F

E

FL
A
o
L
o

ho

ho
for t 0
4.2 Uniaxial Extension 257
which est ablishes t he velocit y funct ion descr ibed by Eq. [4.6]. The for ce
( ) r equir ed t o maint ain is exper iment ally det er mined dur ing
t est ing. Result s may be pr esent ed in t er ms of t he t ensile gr owt h funct ion
defined by combining Eq. [4.5] and [4.10]:
[4.12]
The st r ess gr owt h funct ion becomes equal t o t he t ensile viscosit y as t ime
goes t o infinit y and a const ant value of t he st r ess ( ) is obt ained:
[4.13]
wher e and . Bagley and Chr ist ianson (1988) have
not ed t hat t his limit may not be obser ved over exper iment ally pr act ical
t ime per iods.
Meissner (1972) devised a met hod of achieving a const ant st r ain r at e
in polymer melt s using a device (Fig. 4.1) consist ing of t wo set s of count er
r ot at ing gear s moving at const ant speeds. The st r ain r at e is r elat ed t o
t he const ant sample velocit ies, in opposit e dir ect ions ( and ), locat ed
a const ant dist ance ( ) apar t :
[4.14]
Const ant velocit ies can be used because t he sample lengt h is const ant .
The t ensile for ce, r equir ed t o maint ain a const ant st r ain r at e, is
measur ed dur ing t est ing and used t o calculat e t he st r et ching st r ess.
Ext ensional viscosit y is det er mined fr om Eq. [4.12]. In t he or iginal
device (Meissner , 1972), samples wer e suppor t ed in a silicone oil bat h.
Rheomet r ics, Inc. (Piscat away, NJ ), int r oduced a commer cial ver sion of
t his inst r ument , int ended for polymer melt s, which uses an iner t gas t o
suppor t t he sample.
F(t )

ho

E
+
f (

ho
, t )

E

ho

FL
A
o
L
o

ho

F exp(

ho
t )
A
o

ho

E
lim
t

E
+

E
+
f (

ho
, t )
E
f (

ho
)
u
a
u
b
L
o

ho

u
a
+ u
b
L
o
258 Chapter 4. Extensional Flow
Figure 4.1. Counter rotating clamps, based on the design of Meissner (1972), to
achieve a constant strain rate in uniaxial extension.
4.3. Bi axi al Extensi on
Biaxial ext ension can be achieved in a lubr icat ed squeezing flow
bet ween par allel plat es. Fig. 4.2 illust r at es t he nor mal case wher e t he
lower plat e is fixed and t he upper plat e moves ver t ically downwar d.
Dur ing squeezing, t he cylindr ical shape is maint ained while t he ar ea
in cont act wit h t he plat e incr eases as t he height of t he sample decr eases.
Test samples ar e lubr icat ed wit h a lower viscosit y liquid.
The impor t ance of pr oper lubr icat ion cannot be under est imat ed.
Unlubr icat ed samples r equir e mor e for ce t o achieve defor mat ion t han
t hat r equir ed for lubr icat ed samples (Chr ist ianson et al., 1985). This
is even t r ue in food mat er ials, such as but t er (Fig. 4.3), t hat have a high
fat cont ent and may appear t o be self-lubr icat ing. The ext r a for ce is
needed t o over come t he fr ict ion int r oduced by t he shear defor mat ion
t hat occur s in unlubr icat ed t est ing. Per fect ly lubr icat ed samples
exper ience only ext ensional defor mat ion.
L
o
sample
4.3 Biaxial Extension 259
Figure 4.2. Flow between lubricated parallel plates to create sample deformation
in biaxial extension.
Figure 4.3. Force versus compression for lubricated and unlubricated squeezing
flow of butter at 15 C: = 15 mm, = 14mm, = 100 mm/min. (Data from
Rohm, 1993))
t = 0 t > 0
h
h(t)
o
Lubricated
Surface
r
o
r(t)
0 2 4 6 8 10
0
20
40
60
80
100
Compression, mm
F
o
r
c
e
,

N
lubricated plates unlubricated plates
Butter
F
compression
h
o
R
o
u
z
260 Chapter 4. Extensional Flow
The velocit y dist r ibut ion in biaxial ext ension can be expr essed in
t er ms of Hencky st r ain as
[4.15]
[4.16]
[4.17]
wher e . These equat ions ar e a slight ly modified ver sion of Eq.
[1.67], [1.68] and [1.69]. Compr essive st r ain in t he ver t ical dir ect ion is
[4.18]
wher e , t he height separ at ing t he plat es. The st r ain r at e is
[4.19]
To pr oduce a const ant st r ain r at e dur ing exper iment at ion, t he
velocit y of t he moving plat e must decr ease as t he plat es appr oach each
ot her . Following t he same pr ocedur e given for uniaxial ext ension, t he
st r ain r at e equat ion descr ibing ver t ical mot ion (Eq. [4.19]) can be
int egr at ed t o give
[4.20]
or
[4.21]
wher e is t he init ial sample t hickness. Eq. [4.21] shows t hat t he
dist ance separ at ing t he plat es must exponent ially decr ease over t ime.
Assuming t he mat er ial is incompr essible and t he volume is const ant
yields:
[4.22]
wher e is t he init ial ar ea of t he sample in cont act wit h one plat e.
Biaxial st r ain is inst ant aneously incr eased t o a const ant value ( )
in an ideal exper iment :
[4.23]
which det er mines t he plat e spacing r equir ement s specified by Eq. [4.21].
Biaxial st r ess (r adial net st r et ching st r ess) is det er mined fr om t he
squeezing for ce ( ) which is measur ed dur ing t est ing:
u
z
2

B
z

h
z
u
r

B
r

h
r
2
u

h
/2
d
h

dh
h
h f (t )

h

d
h
dt

1
h
_

,
dh
dt

0
t

ho
dt

h
o
h
dh
h
h f (t ) h
o
exp(

ho
t )
h
o
Ah A
o
h
o
A
o

Bo

Bo
for t 0
F f (t )
4.3 Biaxial Extension 261
[4.24]
Abiaxial ext ensional gr owt h funct ion is defined by combining Eq. [4.21]
and Eq. [4.24]
[4.25]
The st r ess gr owt h funct ion becomes equal t o t he biaxial ext ensional
viscosit y as t ime goes t o infinit y and a const ant value of t he net
st r et ching st r ess ( ) is obt ained:
[4.26]
wher e and . Obt aining st eady-st at e condit ions
dur ing exper iment at ion may be difficult . Compar ing gr owt h funct ions
(Eq. [4.25]) alone may be sufficient t o dist inguish food samples.
If t he lubr icat ed squeezing flow exper iment is oper at ed so t he
downwar d velocit y of t he upper plat e is const ant and t he bot t om plat e
is fixed (t he common mode of oper at ion in t est ing equipment such as
t he Inst r on Univer sal Test ing Machine), t hen sample height decr eases
linear ly:
[4.27]
The biaxial ext ensional st r ain r at e (also called t he r adial ext ension r at e)
is equal t o one-half t he ver t ical Hencky st r ain r at e:
[4.28]
Ext ensional viscosit y is calculat ed fr om t he net st r et ching st r ess and
t he st r ain r at e:
[4.29]
is obt ained fr om one of t he following equat ions depending on t he
degr ee of fill bet ween t he plat es dur ing t est ing:
[4.30]
or

B

rr

zz

F
A

Fh
A
o
h
o

B
+
f (

Bo
, t )

B

Bo

2F exp(

ho
t )
A
o

ho

F exp(2

Bo
t )
A
o

Bo

B
lim
t

B
+

B
+
f (

Bo
, t )
B
f (

Bo
)
h f (t ) h
o
u
z
t

1
2
_

1
2h
_

,
dh
dt

u
z
2(h
o
u
z
t )

B
f (t )

B


B
2(h
o
u
z
t )
u
z

B

F
r
2
(partially full gap where r R)
262 Chapter 4. Extensional Flow
[4.31]
wher e is t he r adius of t he sample, is t he r adius of t he plat e, and
is t he exper iment ally det er mined for ce pushing down on t he upper plat e.
When t he gap is filling wit h an incompr essible mat er ial, t he sample
volume is const ant making . This expr ession can be solved
for allowing Eq. [4.30] t o be calculat ed in t er ms of :
[4.32]
wher e is t he init ial r adius of t he sample.
High levels of st r ain can r esult in loss of lubr icat ion bet ween plat es,
and subsequent int r oduct ion of shear flow int o t he exper iment . The
maximum st r ain can be calculat ed as
[4.33]
Accor ding t o Macosko (1994), loss of lubr icat ion is t ypical when t he
maximum st r ain is near 1.0. Macosko (1994) also not es t hat t he decr ease
in lubr icant t hickness ( ) is appr oximat ely equal t o t he squar e of r oot
of t he gap:
[4.34]
wher e is t he init ial t hickness of t he lubr icant . Fur t her mor e, he
r ecommends (based on Secor , 1988) t he following cr it er ion for lubr ica-
t ion:
[4.35]
wher e and ar e t he Newt onian shear viscosit ies of t he lubr icant and
t est sample, r espect ively. These r esult s give us valuable insight int o
t he lubr icat ion pr oblem. Unfor t unat ely, t hey ar e not dir ect ly applicable
t o t he biaxial ext ension of non-Newt onian foods. Exper iment al condi-
t ions used in lubr icat ed squeezing flow of some food mat er ials ar e
summar ized in Table 4.1.

B

F
R
2
(full gap)
r R F
R
o
2
h
o
r
2
h
r
2
h

B

Fh
R
o
2
h
o
for r R (partially full gap)
R
o
(
B
)
max

1
2
ln

h
h
o
_

h
h
o

o
2
h
<

L
R
2

2
< 20

L

4.4 Flow Through a Converging Die 263
Typical dat a analysis of biaxial ext ensional flow, for t he case wher e
t he bot t om plat e is fixed and t he upper plat e moves downwar d at a
const ant velocit y, is discussed in Example Pr oblems 4.9.1. Result s show
a t ypical change (shar p incr ease followed by a gr adual decr ease) in
ext ensional viscosit y wit h st r ain r at e. Example Pr oblem 4.9.2 illus-
t r at es how r esult s fr om individual t est s, such as t he one given in
Example 4.9.1, can be combined t o examine t he t ension-t hinning (or
t ension-t hickening) char act er ist ics of a sample.
Table 4.1. Product, Initial Sample Size, Lubricant, and Test Conditions used to
Determine Biaxial Extensional Viscosity in Lubricated Squeezing Flow
Pr odu ct lu br ica n t Refer en ce
(cm) (cm) (cm/ min )
bu t t er 1.5 2.80 pa r a ffin oil
1
0.1,1.0,10 Roh m, 1993
bu t t er 1.0 3.67 cookin g oil
2
0.05 Sh u kla et a l.,
1995
s t a r ch 1.2-4.0 3.85 pa r a ffin oil
3
0.5 Ch r is t ia n s on
gels et a l., 1985
ch ees e 2.0-4.0 2.85 pa r a ffin oil
3
0.5 - 5.0 Ca s ir a gh i et
a l., 1985
ch ees e 3.18 s ilicon e oil
4
1.0 Ca mpa n ella et
a l., 1987
wh ea t 2.0 3.1 pa r a ffin oil
3
0.2 - 5.0 Ba gley a n d
dou gh Ch r is t ia n s on ,
1986
wh ea t 0.54-1.92 5.72-8.0 Su per Lu be
5
con s t a n t Hu a n g a n d
dou gh s t r es s Kokin i, 1993
1
Mer ck 7174; 70 mPa s a t 20 C;
2
PAM, Amer ica n Home Foods ;
3
USP/ TCC wh it e,
ligh t . Fis h er Scien t ific Co.;
4
Dow Cor n in g 500;
5
Su per Lu be, In c., Boh emia , NY
4.4. Flow Through a Convergi ng Di e
Flow int o a conver gence involves an ener gy loss due t o shear and an
addit ional loss due t o t he ext ension (st r et ching) of fluid st r eamlines.
Conver ging flow is somet imes called uncont r olled flow because fluid
st r eamlines ar e a funct ion of fluid pr oper t ies. It is also not pur e
ext ensional flow because it involves a combinat ion of bot h shear and
ext ensional defor mat ion. Analyses descr ibed her e ar e based on sepa-
r at ing t he ent r ance pr essur e dr op int o t wo component s: one due t o shear
and t he ot her due t o ext ension. An alt er nat ive met hod, called t he sink
flowanalysis based on t he wor k of Met zner and Met zner (1970), assumes
h
o
R
o
u
z

264 Chapter 4. Extensional Flow
pur e (no shear ) ext ensional flow. This met hod r equir es knowledge of
an angle of conver gence t ypically found fr om flowvisualizat ion. Anot her
t echnique, involving ener gy pr inciples, has been developed by Binding
(1988).
Figure 4.4. Shear (showing velocity profile) and extensional flow components
causing pressure drop in a convergence.
4.4.1. Cogswells Equati ons
Cogswell (1972) assumed t he ent r y pr essur e dr op over an ar ea of
conver ging flow, fr om a cir cular bar r el int o a capillar y die, was made
up of t wo component s, one r elat ed t o shear flow and one r elat ed t o
ext ensional flow (Fig. 4.4):
[4.36]
This flow sit uat ion could be r efer r ed t o as unlubr icat ed, const r ained
conver gence (Cogswell, 1978). It is assumed t hat no slip occur s at t he
wall.
The pr essur e dr op due t o shear can be der ived by consider ing t he
differ ent ial pr essur e dr op (based on t he power law for m of t he flow r at e
equat ion, Eq. [2.31]) over t he lengt h at r adius (Fig. 4.5):
Shear Flow Extensional Flow
P
en
P
en, S
+ P
en, E
dl r
4.4.1 Cogswells Equations 265
Figure 4.5. Geometry for converging flow from a barrel of radius into a capil-
lary die of radius .
[4.37]
Since , . Subst it ut ing t his equat ion int o Eq. [4.37]
yields
[4.38]
which must be int egr at ed bet ween t he die and t he bar r el:
[4.39]
Evaluat ing t he int egr al and simplifying t he r esult yields t he component
of t he pr essur e dr op in t he conver gence due t o shear flow:
[4.40]
wher e , t he appar ent wall shear r at e in t he die.
The volumet r ic aver age velocit y at any par t icular cr ossect ion of t he
die is
R
b
r
dl
l
R
R
b
R
d(P
en, S
)
Q
n
r
3n + 1

3n + 1
n
_

,
n
2Kdl
r/l tan dl dr/ tan
d(P
en, S
)
Q
n
r
3n + 1

3n + 1
n
_

,
n
2K
tan
dr

0
P
en, S
d(P
en, S
)

R
R
b
Q
n
r
3n + 1

3n + 1
n
_

,
n
2K
tan
dr
P
en, S

n

3n + 1
4n
_

,
n
2K
3n tan

R
R
b
_

,
3n
_

,
4Q/(R
3
)
266 Chapter 4. Extensional Flow
[4.41]
Eq. [4.41] may be differ ent iat ed t o give t he aver age ext ensional st r ain
r at e at each cr ossect ion:
[4.42]
Unlike t he shear r at e, t his expr ession does not depend on t he for m of
t he velocit y pr ofile. The differ ent ial pr essur e dr op, due t ot he dissipat ion
of ext ensional ener gy, may be wr it t en in t er ms of an aver age ext ensional
st r ess act ing on an annulus:
[4.43]
or , r ecognizing t hat is negligible, Eq. [4.43] is simply
[4.44]
Assuming a power law r elat ionship bet ween t he aver age st r ess and
t he aver age st r ain r at e,
[4.45]
allows Eq. [4.44] t o be r ewr it t en as
[4.46]
Subst it ut ing Eq. [4.42], for t he aver age st r ain r at e, int o Eq. [4.46] yields
[4.47]
Eq. [4.47] must be int egr at ed bet ween t he r adius of t he capillar y ( )
and t he r adius of t he bar r el ( ) t o det er mine t he component of t he
ent r ance pr essur e loss due t o ext ensional flow:
[4.48]
Evaluat ing t he int egr als and simplifying t he r esult gives
[4.49]
u
Q
r
2

Q
l
2
tan
2

d u
dl
_

,

(2Q)
l
3
tan
2


2Q tan
r
3
(dP
en, E
) r
2

E
((r + dr)
2
r
2
)
(dr)
2
d(P
en, E
) 2
E
dr
r

E
K
E
(

E
)
m
d(P
en, E
) 2K
E
(

E
)
m
dr
r
d(P
en, E
) 2K
E

2Q tan
r
3
_

,
m
dr
r
R
R
b

0
P
en, E
d(P
en, E
)

R
R
b
2K
E

2Q tan
r
3
_

,
m
dr
r
P
en, E

m

2K
E
3m
_

tan
2
_

,
m

R
R
b
_

,
3m
_

,
4.4.1 Cogswells Equations 267
Taking t he logar it hm of t his equat ion yields
[4.50]
which can be used in r egr ession analysis t o calculat e fr om t he slope
of t he line, and fr om t he int er cept . Values of and may be used
t o calculat e t he aver age ext ensional viscosit y:
[4.51]
wher e is t he aver age ext ensional st r ain r at e at t he die (wher e )
comput ed fr om Eq. [4.42] as
[4.52]
Using t he above equat ions, t he following pr ocedur e can be ut ilized
t o calculat e ext ensional viscosit y fr om die ent r y pr essur e dat a:
1. Det er mine t he t ot al ent r ance pr essur e loss ( ) using t he Bagley
pr ocedur e discussed in Sec. 2.5. Also, det er mine t he shear flow
r heological par amet er s ( ) using st andar d met hods in capillar y
viscomet r y pr esent ed in Chapt er 2. Alt hough it may be convenient
t o find shear pr oper t ies fr om capillar y dat a, any st andar d r heo-
logical t echnique could be used t o det er mine and .
2. Using Eq. [4.40], find t he pr essur e dr op in t he conver gence due t o
shear flow, .
3. Subt r act t his fr om , found in St ep 1, t o yield t he component of
t he pr essur e dr op in t he ent r ance due t o ext ensional flow, .
4. Use Eq. [4.50] in a r egr ession analysis pr ocedur e t o evaluat e and
. Calculat e t he st r ain r at e and t he ext ensional viscosit y fr om
Eq. [4.52] and [4.51], r espect ively.
The pr eceding equat ions descr ibe behavior accept ably up t o die
angles of appr oximat ely 45 degr ees (Gibson, 1988). When , shear
flow is dominant ; when , mat er ials may for m t heir own conver -
gence pat t er n (Cogswell, 1981) r esult ing in an unknown ent r y angle.
For mat ion of a pr oduct conver gence pat t er n will be a funct ion of t he
ln(P
en, E
) m ln+ ln

2K
E
3m
_

tan
2
_

,
m

R
R
b
_

,
3m
_

,
1
1
]
m
K
E
K
E
m

E
K
E
(

E, R
)
m 1

E, R
r R

E, R

tan
2
P
en
K, n
K n
P
en, S
P
en
P
en, E
m
K
E
< 10
> 45
268 Chapter 4. Extensional Flow
r heological pr oper t ies. It may be a par t icular ly significant pr oblem wit h
foods having a high yield st r ess. Cogswells equat ions ar e used t o
invest igat e soy dough in Example Pr oblem 4.9.3.
Cogswell (1972) for mulat ed expr essions for t he net aver age ext en-
sional st r ess and net aver age ext ensional st r ain for flat ent r y dies (Fig.
4.5 wit h ) as
[4.53]
and
[4.54]
wher e is t he appar ent shear viscosit y based on a power law r ela-
t ionship: . The aver age ext ensional viscosit y is easily calcu-
lat ed fr om Eq. [4.53] and [4.54]:
[4.55]
Eq. [4.55] is ver y convenient for making a r apid compar ison bet ween
fluids.
An alt er nat ive analysis for conver ging flow has been pr oposed by
Binding (1988) and used t o evaluat e cor n meal dough (Padmanabhan
and Bhat t achar ya, 1993) and polymer melt s (Padmanabhan and
Bhat t achar ya, 1994). The t echnique involves axis-symmet r ic flow but
emphasizes planar cont r act ion which involves t he conver gence fr om a
r ect angular channel int o a r ect angular or ifice. Also, a r efined ver sion
of Cogswells met hod has been pr oposed, and successfully used for
molt en polymer s, by Ber st ed (1993). This analysis does not r equir e a
const ant ext ensional viscosit y in t he conver gence, but an it er at ive
pr ocedur e is needed t o find a solut ion t o t he pr essur e dr op equat ions.
4.4.2. Gi bsons Equati ons
Using spher ical coor dinat es, Gibson (1988) developed a met hod of
det er mining ext ensional viscosit y for t he full r ange of die angles up t o
90 degr ees. The analysis included a power lawmodel (Eq. [4.45]) r elat ing
aver age ext ensional st r ess and st r ain. Component s of t he pr essur e dr op
due t o shear flow and ext ensional flow wer e found t o be
90

E

3
8
(n + 1) P
en

E

4
2
3(n + 1) P
en

K()
n 1

E


E

9(n + 1)
2
(P
en
)
2
32
2
4.4.2 Gibsons Equations 269
[4.56]
wher e is expr essed in r adians and
[4.57]
r espect ively. The die exit effect int egr al( ) was given as a funct ion of
and t he angle of conver gence:
[4.58]
can be int egr at ed dir ect ly at but a numer ical solut ion is r equir ed
at ot her values of . Solut ions t o Eq. [4.58] cover ing most pr act ical
sit uat ions ar e given in Table 4.2 and Fig. 4.6. Int er polat ion bet ween
differ ent values of and die ent r y angles (Table 4.2) does not int r oduce
a significant level of er r or . Equat ions given in Fig. 4.6 wer e gener at ed
fr om linear r egr ession analyses of t he infor mat ion pr ovided in Table
4.2.
The aver age die exit elongat ional st r ain r at e is
[4.59]
Pr ocedur es for finding , , and ar e t he same as t hose out lined,
in t he pr eceding sect ion, for Cogswells met hod. Values of and ar e
det er mined fr om r aw dat a of ver sus . Taking t he logar it hm of
Eq. [4.57] yields
[4.60]
showing t hat is t he slope of t he r esult ing line. The numer ical value
of is det er mined fr om t he int er cept . Ext ensional viscosit y is
calculat ed fr om Eq. [4.51] wit h defined by Eq. [4.59]. The last t er m
in Eq. [4.60] (equal t o ) is small and oft en negligible for die angles
less t han 45 degr ees. This analysis, like t hat of Cogswell, may be
complicat ed by t he t endency of foods wit h a high yield st r ess t o for m a
nat ur al angle of conver gence. Gibsons equat ions ar e used t o invest igat e
soy dough in Example Pr oblem 4.9.4.
P
en, S

2K(sin
3n
)
3n
1 + 3n

1 + 3n
4n
_

,
n

R
R
b
_

,
3n
_

P
en, E
K
E

2
3m

(sin) (1 + cos )
4
_

,
m

R
R
b
_

,
3m
+

4
m
_

,
1
1
]

m
f (m, )

(sin
m + 1
) (1 + cos )
m 1
d
m 1
m
m

E, R

(sin) (1 + cos )
4
P
en, E
K n
K
E
m
P
en, E

ln(P
en, E
) m ln() + ln

K
E

2
3m

(sin) (1 + cos )
4
_

,
m

R
R
b
_

,
3m
+

4
m
_

,
1
1
]
1
1
]
m
K
E

E, R
/4
m
270 Chapter 4. Extensional Flow
Table4.2. Numerical Solutions of theDieExit Effect Integral (fromGibson, 1988):
0.1 0.0660 0.1558 0.2882 0.4677 0.7003
0.2 0.0629 0.1535 0.2892 0.4730 0.7080
0.3 0.0601 0.1515 0.2904 0.4787 0.7162
0.4 0.0575 0.1497 0.2920 0.4848 0.7249
0.5 0.0550 0.1481 0.2939 0.4913 0.7339
0.6 0.0528 0.1467 0.2961 0.4983 0.7434
0.7 0.0507 0.1455 0.2985 0.5055 0.7533
0.8 0.0488 0.1444 0.3011 0.5132 0.7636
0.9 0.0470 0.1435 0.3040 0.5212 0.7743
1.0 0.0453 0.1427 0.3071 0.5295 0.7854
1.2 0.0422 0.1414 0.3139 0.5472 0.8087
1.4 0.0394 0.1406 0.3215 0.5662 0.8334
1.6 0.0370 0.1401 0.3299 0.5865 0.8597
1.8 0.0348 0.1399 0.3391 0.6082 0.8874
2.0 0.0328 0.1399 0.3490 0.6313 0.9167
Figure 4.6. Numerical solutions of the die exit effect integral for different values
of the extensional flow behavior index ( ) and die entry angles.
f (m, )

(sin
m + 1
) (1 + cos )
m 1
d
m 30 45 60 75 90
0 0.5 1 1.5 2
0
0.2
0.4
0.6
0.8
1
Extensional Flow Behavior Index
D
i
e

E
x
i
t

E
f
f
e
c
t

I
n
t
e
g
r
a
l
30 degrees
90 degrees
75 degrees
60 degrees
45 degrees
= .6787 + .1133 m
= .4498 + .0856 m
= .2789 + .0319 m
= .1529 - .0081 m = .0643 - .0172 m
m
4.4.3 Empirical Method 271
4.4.3. Empi ri cal Method
Ext ensional viscosit y may be est imat ed using a st andar d mat er ial.
Assuming making (fr om Eq. [4.51]) , and t hat t he shear
cont r ibut ion t o t he pr essur e loss is small, t hen Eq. [4.49] shows t hat an
aver age ext ensional viscosit y is pr opor t ional t o t he ent r ance pr essur e
dr op divided by t he appar ent wall shear r at e in t he die:
[4.61]
wher e is a dimensionless const ant assumed t o be a funct ion of t he
syst em geomet r y, not st r ain r at e or t he r heological pr oper t ies of t he
sample. The numer ical value of could be est imat ed using a st andar d
Newt onian mat er ial wit h a known value of .
Eq. [4.61] can also be used wit h a zer o lengt h die wher e it is assumed
t he ent ir e pr essur e dr op is t he ent r ance loss. This idea is illust r at ed in
Fig. 4.7 wher e a plunger , moving downwar d at a const ant velocit y ( ),
is for cing mat er ial t hr ough an or ifice wit h a 90 degr ee ent r y angle. The
ent r ance pr essur e loss is calculat ed fr om t he for ce on t he plunger and
t he cr ossect ional ar ea of t he bar r el: . Assuming t he t est
mat er ial is incompr essible, t he volumet r ic flow t hr ough t he or ifice is a
funct ion of t he plunger velocit y ( ). Given Eq. [4.61] and t he
above definit ions, ext ensional viscosit y may be est imat ed as
[4.62]
Eq. [4.62] can be used, for bot h t aper ed or flat ent r y dies, as t he basis
of a qualit y cont r ol t est if making a r elat ive compar ison bet ween t he
ext ensional viscosit ies of similar foods. A commer cially manufact ur ed
on-line syst em (Rheomet r ics, Inc., Piscat awa, NJ ), based on t his pr in-
ciple, is available t o t he food indust r y. The unit has been used
successfully wit h cookie and cr acker dough. It should be not ed t hat
could be a funct ion of t he st r ain r at e and t he r heological pr oper t ies of
t he t est subst ance. This may be par t icular ly impor t ant for high yield
st r ess mat er ials t est ed in inst r ument s const r uct ed wit h lar ge ent r y
angles. Menjivar et al. (1992) used dat a fr om a zer o lengt h die t o
calculat e Tr out on r at ios which cor r elat ed well wit h t he ext r udat e swell
behavior of wheat flour doughs.
m 1
E
K
E

E
C
P
en

C
C

E
u
z
P
en
F/(R
b
2
)
Q u
z
R
b
2

E
C
P
en

C
4
_

FR
3
u
z
R
b
4
_

,
C
272 Chapter 4. Extensional Flow
Figure 4.7. Flow through an orifice (zero-length die) controlled by the constant
downward velocity of the plunger.
4.5. Opposi ng Jets
In t he met hod of opposing jet s, opposing nozzles ar e complet ely
immer sed in a t est fluid (Fig. 4.8). Using a vacuum, fluid is sucked int o
t he nozzles cr eat ing uniaxial ext ensional flow in t he r egion bet ween
t hem. This flow causes a t ensile st r ess which, if unr est r ained, would
cause t he nozzles t oappr oach each ot her . In exper iment at ion, one nozzle
is fixed and t he ot her nozzle is r est r ained fr om movement but allowed
t ot r ansmit a r esult ant for ce t oan appr opr iat e t r ansducer . Amoment um
balance indicat es t hat pr essur e and moment um for ces cancel, so t he
for ce ( ) measur ed at t he t r ansducer r epr esent s t he t ensile st r ess (Fuller
et al., 1987). An opposing jet device could also be used t o cr eat e uniaxial
compr essive flow if fluid was for ced t o move out of t he nozzles inst ead
of being sucked int o t hem.
R
R
b
F
Side View
Bottom View
bottom plate
barrel
orifice
plunger
u = constant
z
F
4.5 Opposing Jets 273
Figure 4.8. Opposing jets configuration to achieve uniaxial extensional flow by
sucking fluid into the nozzles.
The aver age ext ensional st r ain r at e may be t aken as t he mean
velocit y at t he nozzle divided by one-half t he dist ance separ at ing t he
nozzles:
[4.63]
wher e is t he dist ance separ at ing t he nozzles, is t he t ot al volumet r ic
flow r at e int o bot h nozzles, and is t he cr oss-sect ional flow ar ea of t he
nozzle. Ext ensional viscosit y is calculat ed by dividing t he mean t ensile
st r ess ( ) by t he st r ain r at e:
[4.64]
Since t he st r ain r at e is an aver age value and t he r esidence t ime of fluid
element s ent er ing t he or ifice is non-unifor m, t he value calculat ed using
Eq. [4.64] should be consider ed an aver age ext ensional viscosit y.
Using t he opposing jet met hod, Fuller et al. (1987) wer e able t oobt ain
good r esult s (appr oximat ely cor r ect Tr out on r at ios) for Newt onian fluids
consist ing of glycer in-wat er mixt ur es. Only qualit at ive differ ences,
however , wer e obser ved for non-Newt onian fluids made fr om polymer s
(Xant han gum and polyacr ylamide) dissolved in mixt ur es of glycer in
and wat er . In t he Fuller et al. (1987) exper iment s, a nozzle diamet er
of 1mm was used and, in most t est ing, t he separ at ion dist ance bet ween
b

Q
A
_

1
b/2
_

,

2Q
Ab
b Q
A
F/A

E


11

22

F/A
2Q/(Ab)

Fb
2Q
274 Chapter 4. Extensional Flow
nozzles was 1mm. A commer cial inst r ument based on t he opposing jet
pr inciple is available (Rheomet r ics, Inc., Piscat awa, NJ ). This met hod
of measur ement may yield valuable r esult s for numer ous fluid food
syst ems such as pancake syr up wher e st r inginess (for mat ion of t hin
t hr eads) is an impor t ant fact or in evaluat ing qualit y. Similar consid-
er at ions may be associat ed wit h t hickened dr inks and or al dr ugs using
a fluid car r ier (cough syr up, pain dr ugs, and ant ibiot ics).
4.6. Spi nni ng
Figure 4.9. Spinning apparatus to evaluate extensional viscosity.
Spinning is a met hod of evaluat ing ext ensional viscosit y of moder at e
viscosit y fluids by subject ing t he sample t o a uniaxial elongat ion.
Ext r udat e dr awing, melt spinning, and fiber spinning ar e synonyms for
t he pr ocedur e. The t est is conduct ed by ext r uding a sample fr om a small
diamet er t ube on t o a r ot at ing dr um or wheel (Fig. 4.9). Ext ension r at es
exper ienced by t he mat er ial may be var ied by changing t he wind-up
speed of t he dr um. St r ess on t he sample is det er mined fr om t he for ce
measur ed on t he dr um.
L
1
2
F
4.6 Spinning 275
Exper iment al dat a fr om a spinning t est is or dinar ily evaluat ed using
a simplified analysis pr oducing aver age values. Assuming t he ext ension
r at e is const ant over t he lengt h of t he sample, t he aver age r at e of
ext ension may be calculat ed as
[4.65]
wher e and ar e t he mean velocit ies at point s 1 and 2 separ at ed by
a dist ance equal t o (Fig. 4.9). Taking as t he volumet r ic flow r at e
t hr ough t he ext r usion t ube, Eq. [4.65] may be wr it t en as
[4.66]
wher e and ar e t he r adii at point s 1 and 2, r espect ively. These r adii
ar e usually det er mined by phot ogr aphic met hods. St r ain imposed on
t he sample dur ing t est ing is calculat ed as
[4.67]
An aver age ext ensional st r ess is det er mined fr om t he t ensile for ce
on t he sample:
[4.68]
wher e is t aken as t he aver age r adius over t he lengt h of t he sample:
. Using Eq. [4.66] and [4.68], an aver age ext ensional
viscosit y may be calculat ed:
[4.69]
Spinning t est s have commonly been conduct ed on polymer melt s and
solut ions. The met hod has also been used as a means of evaluat ing t he
st r et chabilit y of Mozzar ella cheese (Cavella et al., 1992). This r esear ch
showed t hat cheese r eached a maximum st r et chabilit y over a well
defined t emper at ur e r ange. Exper iment at ion also allowed an accur at e
evaluat ion of maximum elongat ion and sample st r engt h. Infor mat ion
of t his t ype may be ver y useful in compar ing and scr eening differ ent
Mozzar ella cheeses for use wit h pizza and past a pr oduct s.

E

u
2
u
1
L
u
1
u
2
L Q

E

1
L

Q
R
2
2

Q
R
1
2
_

,

Q
L

1
R
2
2

1
R
1
2
_

,
R
1
R
2

E
ln

u
2
u
1
_

,
ln

R
1
2
R
2
2
_

E

F
R
2
R
R (R
1
+ R
2
)/2

E


E

FL
R
2
Q

R
2
2
1

R
2
R
1
_
,
2
_

,
276 Chapter 4. Extensional Flow
4.7. Tubeless Si phon (Fano Flow)
Figure 4.10. Apparatus to evaluate extensional viscosity using the tubeless
siphon phenomenon.
Dat a collect ed employing t he t ubeless siphon phenomenon (also
called open siphon or Fano flow) ar e useful in evaluat ing ext ensional
behavior . Test ing is car r ied out using a low pr essur e r eser voir wit h a
pr ot r uding capillar y t ube (Fig. 4.10). The upper cont ainer is lower ed
unt il t he t ube t ouches t he fluid cont ained in t he bot t om vessel. Then,
t he vacuum is applied and t he t ube is slowly r aised allowing a fr ee
st anding column of fluid t o be for med. An advant age of t his met hod,
over t he spinning t echnique, is t hat t he sample r eceives less sever e
t r eat ment pr ior t o t est ing. St r ain r at es and ext ensional viscosit ies can
be calculat ed using t he same consider at ions discussed for spinning.
Compar at ive dat a can be easily collect ed mer ely by det er mining t he
maximum column height t hat can be obt ained wit h a const ant vacuum.
4.8. Steady Shear Properti es from Squeezi ng Flow Data
Biaxial ext ensional viscosit y can be det er mined fr om squeezing flow
dat a (Sec. 4.3). Alt er nat ive int er pr et at ions of t hese dat a ar e pr esent ed
in t his sect ion. Power law fluid pr oper t ies, for example, can also be
vacuum
4.8.1 Lubricated Squeezing Flow 277
est imat ed fr om squeezing flow dat a. In addit ion, par allel plat e equip-
ment can be used t o pr oduce a combined shear and ext ensional flow
when t he int er face bet ween t he plat e and t he fluid is not lubr icat ed so
t he sample adher es fir mly t o t he plat e. In t his case, Newt onian and
power law fluid pr oper t ies as well as yield st r esses may be calculat ed.
Squeezing flow dat a may also pr ovide empir ical infor mat ion t hat can
for m t he basis of a useful qualit y cont r ol t est .
4.8.1. Lubri cated Squeezi ng Flow
Squeezing flow bet ween par allel plat es can be achieved in many food
r heology labor at or ies. When t his defor mat ion is execut ed bet ween
lubr icat ed plat es, biaxial ext ensional flow is achieved and an ext en-
sional viscosit y can be calculat ed. Dat a fr om lubr icat ed squeezing flow
may also be evaluat ed in a manner t hat pr oduces st eady shear fluid
par amet er s. This is advant ageous in some cases. Food mat er ials, for
example, t hat have a high fat cont ent (r aw meat emulsions, but t er ,
mar gar ine, soft cheese, et c.) may violat e t he no slip boundar y condit ion
r equir ed in a t r adit ional viscomet er . Tr ue slip, however , is a necessar y
condit ion in lubr icat ed squeezing flow. Campanella and Peleg (1987c)
have t aken advant age of t his idea in evaluat ing t he power law flow
behavior of peanut but t er . Squeezing flow can also be useful for ver y
t hick fluids which ar e difficult t o load int o convent ional viscomet er s.
Figure 4.11. Lubricated squeezing flow with sample completely filling the gap
between two parallel plates.
R
F(t)
h(t)
278 Chapter 4. Extensional Flow
In lubr icat ed squeezing flow of a power law fluid wit h a syst em
having a fixed bot t om plat e, and const ant sample ar ea in cont act wit h
t he plat es (meaning t he gap bet ween t he plat es is complet ely full dur ing
t est ing, Fig. 4.11), t he for ce ( ) r equir ed t o maint ain a const ant down-
war d velocit y ( ) of t he upper plat e is (Campanella and Peleg, 1987c),
for a power law fluid,
[4.70]
wher e . Taking t he logar it hm of each side yields
[4.71]
wher e:
[4.72]
Power law fluid par amet er s ( ) may be det er mined fr om linear
r egr ession of exper iment al dat a using Eq. [4.71]. Ear ly dat a point s
should be neglect ed: Only t he linear por t ion of t he cur ve, wher e st eady
flow is pr esent , should be used in t he analysis. This t echnique is
demonst r at ed t o evaluat e t he behavior of peanut but t er in Example
Pr oblem 4.9.5.
The squeezing flow solut ion for a power law fluid in t he case of flow
under a const ant for ce, pr oducing a const ant st r ess when t he gap is
complet ely full (Fig. 4.11) dur ing t est ing, is (Campanella and Peleg,
1987c)
[4.73]
wher e:
[4.74]
and is a const ant for ce applied t o t he sample t hr ough t he upper plat e.
Since cont ains bot h unknown pr oper t ies, mult iple exper iment s wit h
at least t wo differ ent loads ar e r equir ed t o det er mine t he numer ical
values of and .
F
u
z
F
3
(n + 1)/2
R
2
Ku
z
n
h
n
h f (t ) h
o
u
z
t
ln(F) C
1
+ n ln

1
h
_

,
C
1
ln(3
(n + 1)/2
R
2
Ku
z
n
)
K, n
ln

h
h
o
_

,
C
2
t
C
2

W
3
(n + 1)/2
R
2
K
_

,
1/n
W
C
2
K n
4.8.2 Nonlubricated Squeezing Flow 279
The pot ent ial er r or int r oduced by t he exist ence of a yield st r ess
dur ing lubr icat ed squeezing flow should not be over looked. For t unat ely,
t his pr oblem is not ser ious r egar dless of whet her or not a yield st r ess is
pr esent : In const ant defor mat ion exper iment s flow will occur ir r e-
spect ive of t he yield st r ess, and in const ant st r ess exper iment s, applied
st r esses must exceed t he yield st r ess for flow t o occur (Campanella and
Peleg, 1987a).
It is inst r uct ive t o consider t he t heor et ical r elat ionship bet ween t he
power law fluid par amet er s det er mined by t he above pr ocedur es and
ext ensional viscosit y. Consider , for example, dat a obt ained in t he
const ant displacement mode. Eq. [4.70], involving st eady shear
par amet er s for power law fluids ( ), may be r ewr it t en as
[4.75]
which, by consider ing Eq. [4.28] and [4.31], is equivalent t o
[4.76]
Since , Eq. [4.76] can be expr essed as
[4.77]
or , in t er ms of biaxial ext ensional viscosit y as,
[4.78]
A compar able expr ession for t ensile ext ensional viscosit y was given
ear lier as Eq. [4.51]. For t he special case of a Newt onian fluid ( and
), Eq. [4.78] shows t he ext ensional viscosit y t o be six t imes t he shear
viscosit y as pr edict ed by Eq. [1.79].
4.8.2. Nonlubri cated Squeezi ng Flow
Nonlubr icat ed squeezing flow pr oduces a bar r eling effect (Fig. 4.12)
due t o shear flow caused by adhesion of t he sample t o t he plat es. The
r esult ing complex flow is not pur ely viscomet r ic or ext ensional, but some
combinat ion of each. Equat ions r elat ing t he for ce r equir ed t o move t he
plat es and plat e separ at ion dist ance can be der ived by assuming specific
const it ut ive r elat ionships. The device or iginally made t o pr oduce
nonlubr icat ed squeezing flow was called a par allel plat e plast omet er
K(

)
n
F
R
2
3
(n + 1)/2
K

u
z
h
_

,
n

B
3
(n + 1)/2
K(

h
)
n

h
2

B
3
(n + 1)/2
2
n
K(

B
)
n

B


B

B
3
(n + 1)/2
2
n
K(

B
)
n 1
n 1
K
280 Chapter 4. Extensional Flow
(Dienes and Klemm, 1946). That t er minology is st ill common t oday, but
t he published lit er at ur e cont ains numer ous synonyms for t he par allel
plat e plast omet er (Bir d and Leider , 1974): par allel plat e viscomet er ,
compr ession plast omet er , t r ansver se flow viscomet er , and par al-
lel-plat e plast imet er . Mor e r ecent ly (Covey and St anmor e, 1981), it has
been called t he squeeze film viscomet er .
Figure 4.12. Flow between lubricated and nonlubricated parallel plates.
It is impor t ant t o r emember t hat t he equat ions pr esent ed in t his
sect ion ar e based on t he no-slip pr emise which may be enhanced when
t est ing wit h met al fixt ur es by using var ious met hods such as pit t ed
plat es, an adhesive like cyanoacr ylat e est er (Super glue), emer y paper
at t ached wit h double adhesive t ape (Rhom and Weidinger , 1993), or by
coat ing t he plat es wit h spr ay lacquer and or dinar y sand (Nolan et al.,
1989; Navackis and Bagley, 1983). Unless ot her wise not ed, t he solut ions
pr esent ed in t his sect ion ar e for t he case of a fully loaded gap. Init ial
sample size is impor t ant : A sufficient amount of mat er ial t o pr oduce a
t = 0
t = 0
t > 0
t > 0
Lubricated
Plates
Nonlubricated
Plates
lubricant
no slip
barreling
4.8.2 Nonlubricated Squeezing Flow 281
value of , wher e equals t he r adius of t he plat e and is t he
init ial height of t he sample (also t he init ial dist ance bet ween t he plat es),
is r ecommended (Dienes and Klemm, 1946). Development of t he fol-
lowing equat ions also ignor es elast ic effect s which could be significant
at high squeezing r at es.
Newtoni an Flui ds. Der ivat ions and solut ions for Newt onian fluids
and power law fluids ar e given in Bir d et al. (1987) and Leider and Bir d
(1974). The nonlubr icat ed squeezing flow of a Newt onian fluid, in an
exper iment al syst em using an immobile bot t om plat e wit h a fully loaded
gap, is (Wint her et al., 1991)
[4.79]
wher e and , t he for ce r equir ed t o maint ain a
const ant downwar d velocit y of t he upper plat e. Eq. [4.79] is known as
t he St efan equat ion. The viscosit y can be calculat ed as t he slope of
ver sus det er mined fr om r egr ession analysis.
The shear r at e, evaluat ed fr om t he velocit y pr ofile in t he gap, is
(Wint her et al., 1991; Chur chill, 1988)
[4.80]
wher e is t he dist ance fr om t he cent er line and is t he ver t ical dist ance
fr om t he hor izont al midplane of t he sample. A maximum shear r at e is
found at and :
[4.81]
The aver age shear r at e is equal t o of t he maximum shear r at e.
The case for squeezing flow of a Newt onian fluid wit h a const ant
for ce, full gap, can be det er mined by int egr at ing t he St efan equat ion.
Fir st , it must be r ecognized t hat , so Eq. [4.79] may be wr it t en
as
[4.82]
R > 10h
o
R h
o
F
3R
4
u
z

2h
3
h f (t ) h
o
u
z
t F f (t )
F
1/h
3

f (r, z)
6rzu
z
h
3
r z
r R z h/2

max

3Ru
z
h
2
2/3
u
z
dh/dt
h
3
dh
2W
3 R
4

dt
282 Chapter 4. Extensional Flow
wher e is t he const ant for ce applied t o t he sample. This equat ion is
easily int egr at ed t o give
[4.83]
The const ant of int egr at ion is evaluat ed fr om t he init ial condit ion t hat
at making . Subst it ut ing t his int o Eq. [4.83], and
simplifying t he r esult , pr oduces t he final solut ion:
[4.84]
Viscosit y can be calculat ed fr om t he slope of ver sus . Also, if t he
viscosit y was known, could be det er mined by solving for .
Power Law Flui ds. The nonlubr icat ed squeezing flow of a power law
fluid, in an exper iment al syst em using an immobile bot t om plat e wit h
a full gap, is (Wint her et al., 1991)
[4.85]
wher e . Eq. [4.85] is known as t he Scot t equat ion and it
r educes t o t he St efan equat ion when . Wint her et al. (1991) suggest
evaluat ing Eq. [4.85] by mult iplying bot h sides by giving
[4.86]
t hen, t aking t he logar it hm of each side of t he equat ion, t o get
[4.87]
Using Eq. [4.87], may be found fr om t he slope of ver sus .
The value of is det er mined fr om t he int er cept . Since t he iner t ia t er ms
in t he equat ions of mot ion ar e neglect ed in developing Eq. [4.87], t he
ear liest dat a point s may not lie on t he line and should be neglect ed. The
aver age shear r at e for a power law fluid in nonlubr icat ed squeezing flow
is (Wint her et al., 1991)
W
1
2h
2

2Wt
3 R
4

+ C
C 1/(2h
o
2
) h h
o
t 0

h
o
h
_

,
2

4Wh
o
2
t
3 R
4

+ 1
(h
o
/h)
2
t
h f (t ) h
F

2n + 1
n
_

,
n

2KR
n + 3
n + 3
_

u
z
n
h
2n + 1
_

,
h f (t ) h
o
u
z
t
n 1
h
Fh

2n + 1
n
_

,
n

2KR
n + 3
n + 3
_

u
z
h
2
_

,
n
ln(Fh) ln

2n + 1
n
_

,
n

2KR
n + 3
n + 3
_

,
1
1
]
+ n ln

u
z
h
2
_

,
ln(u
z
/h
2
) n ln(Fh)
K
4.9.1 Biaxial Extension of Processed Cheese Spread 283
[4.88]
Int egr at ing Eq. [4.85] can pr oduce a solut ion for descr ibing t he case
wher e a const ant weight is used t o defor m a power law fluid. This
pr oblem has been solved by Leider and Bir d (1974).
Yi eld Stress Evaluati on. Nonlubr icat ed squeezing flow has been used
t o evaluat e t he behavior of Bingham plast ic and Her schel-Bulkley fluids
(Covey and St anmor e, 1981). Gener al solut ions ar e somewhat cum-
ber some but t he wor k ver ified a simple pr ocedur e t o det er mine t he yield
st r ess of semi-solid mat er ials. In t est ing, a const ant for ce ( ) is placed
on samples which complet ely fill t he gap bet ween par allel plat es. The
yield st r ess is calculat ed on t he basis of t he asympt ot ic or r esidual
t hickness ( ) of t he sample:
[4.89]
The t echnique wor ks r easonably well when t he mat er ials t est ed have a
high yield st r ess and t he r at e of defor mat ion pr oduced by t he const ant
for ce is small. Campanella and Peleg (1987a) used t his met hod t o
evaluat e t he yield st r esses of t omat o ket chup, must ar d, and mayon-
naise. Also, values of t he yield st r ess, for a /t r eacle past e, det er mined
wit h Eq. [4.89] compar ed well t o t hose found by ext r apolat ion of a
r heogr am t o zer o shear r at e (Covey and St anmor e, 1981).
4.9. Example Problems
4.9.1. Bi axi al Extensi on of Processed Cheese Spread
Given t he dat a in Table 4.3, det er mine t he biaxial ext ensional st r ain
r at e and t he ext ensional viscosit y of pr ocessed cheese spr ead. Also
calculat e t he maximum st r ain found dur ing t est ing. Dat a ar e for t he
lubr icat ed squeezing flow bet ween par allel plat es wher e t he gap is filling
dur ing exper iment at ion (Fig. 4.12). The lower plat e is fixed and t he
upper plat e is moving downwar d wit h a const ant velocit y. Par affin oil
was used as t he lubr icant .

average

2(2n + 1)
3n
_

u
z
R
h
2
_

,
W
h
a

o

3Wh
a
2R
3
TiO
2
284 Chapter 4. Extensional Flow
Table 4.3. Data for the Lubricated Squeezing Flow ( =0.04 m, =0.057 m,
=0.0005 m/min) of Processed Cheese Spread at 7 C (Data from Casiraghi et al.,
1985)
compr es s ion
(N) (m) (m) (1/ s ) (1/ s ) (MPa s )
5.1 0.0014 0.0386 0.000108 0.000216 4.5
13.8 0.0027 0.0373 0.000112 0.000227 11.3
22.9 0.0048 0.0352 0.000118 0.000236 16.7
29.6 0.0067 0.0333 0.000125 0.000250 19.3
39.1 0.0094 0.0306 0.000136 0.000272 21.5
50.0 0.0120 0.0279 0.000149 0.000298 22.9
59.5 0.0141 0.0258 0.000161 0.000322 23.3
69.7 0.0160 0.0240 0.000174 0.000348 23.5
80.8 0.0180 0.0220 0.000189 0.000378 23.0
91.5 0.0200 0.0199 0.000209 0.000418 21.3
106.3 0.0227 0.0173 0.000241 0.000482 18.7
137.1 0.0260 0.0140 0.000298 0.000596 15.8
192.3 0.0294 0.0106 0.000392 0.000784 12.7
Figure 4.13. Raw data of force versus compression for the lubricated squeezing
flow of processed cheese spread at 7 C.
h
o
R
o
u
z

F h

h

B
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035
0
50
100
150
200
250
Compression, m
F
o
r
c
e
,

N
Processed Cheese Spread
F
compression
h

4.9.1 Biaxial Extension of Processed Cheese Spread 285


Plot t ing t he r aw dat a (Fig. 4.13) clear ly illust r at es t he change in for ce
dur ing compr ession. Result s ar e summar ized in Table 4.3 and t he
following equat ions show t he comput at ions needed t o gener at e t he fir st
r ow of infor mat ion:
Biaxial ext ensional viscosit y, plot t ed in t er ms of t he Hencky st r ain r at e,
is illust r at ed in Fig. 4.14. The maximum st r ain found dur ing t est ing is
calculat ed fr om Eq. [4.33]:
Since t his value is less t han 1.0, it is unlikely t hat exper iment al er r or s
associat ed wit h lubr icant loss occur r ed dur ing t est ing.
Figure 4.14. Biaxial extensional viscosity of processed cheese spread at 7 C.
h h
o
- compression 0.0400 0.0014 0.0386 m

B
0.5

u
z
h
_

,
0.5

.0005
(60).0386
_

,
0.000108 s
1

h
2

B
2(0.000108) 0.000216 s
1

B

Fh
R
o
2
h
o

5.14(.0386)
(.057)
2
(.040) (.000108)
4.5(10)
6
Pa s 4.5 MPa s
(
B
)
max

1
2
ln

h
h
o
_

,

1
2
ln

.0106
.04
_

,
0.66
0.0002 0.0004 0.0006 0.0008 0.001
0
5
10
15
20
25
h
.
, 1/s
B
,

M
P
a

s
Processed Cheese Spread

286 Chapter 4. Extensional Flow


4.9.2. Bi axi al Extensi on of Butter
Biaxial ext ension dat a ar e available for but t er at 15 C and 20 C (Fig.
4.15). These t est s wer e conduct ed using par allel plat es wit h a fixed
lower plat e. The upper plat e was lower ed at differ ent downwar d velo-
cit ies indicat ed by t he cr osshead speeds. All samples had an init ial
diamet er of 28 mm and an init ial height of 15 mm. Evaluat e t he behavior
of t his mat er ial.
Figure 4.15. Biaxial extensional viscosity of butter at 15 C and 20 C (Data from
Rohm, 1993).
Higher cr osshead speeds pr oduced lar ger st r ain r at es. In each t est ,
a shar p init ial incr ease was followed by a gr adual decline, similar t o t he
r esult s found for pr ocessed cheese spr ead (Fig. 4.14) in t he pr evious
example. Plot t ing a line t hr ough t he downwar d slope of each dat a set
(dashed line, Fig. 4.15) suggest a power law r elat ionship bet ween vis-
cosit y and st r ain r at e. Fit t ing t he line t hr ough r epr esent at ive point s
yields

0.001 0.002 0.005 0.01 0.02 0.05 0.1 0.2 0.5 1
15 C
20 C
Butter
100
30
10
3
1
0.3
0.1
0.03
0.01
B
,

M
P
a

s
h
, 1/s
.
A
B
C
A : 1 mm / min
B : 10 mm / min
C : 100 mm / min
Crosshead Speed

4.9.3 45 Converging Die, Cogswells Method 287
and
The slope t er ms given above ar e analogous t o used t o expr ess t he
r elat ionship (Eq. [1.28]) bet ween appar ent viscosit y and shear r at e in
a st andar d power law fluid. Hence, t he r esult s could be int er pr et ed as
ext ensional-t hinning behavior . Similar r esult s have been obser ved for
pr ocessed cheese spr ead. Ot her mat er ials, such as mozzar ella cheese,
showa leveling off of ext ensional viscosit y aft er t he init ial shar p incr ease
(Casir aghi et al., 1985) making t he int er pr et at ion of dat a much mor e
difficult .
4.9.3. 45 Convergi ng Di e, Cogswells Method
Taking t he capillar y viscomet er dat a for soy dough pr esent ed in
Example 2.12.2, est imat e t he ext ensional viscosit y assuming t hat t he
ent r ance pr essur e loss can be divided int o t wo separ at e component s:
one r elat ed t o shear and one r elat ed t o ext ensional flow. Dat a fr om
Table 2.8 ar e summar ized in Table 4.4. Assume t he mat er ial for ms a
nat ur al conver gence angle of in t he die. Solve t he
pr oblem using Cogswells equat ions pr esent ed in Sec. 4.4.1.
Analysis of t he dat a conduct ed in Example 2.12.2 gener at ed values
of t he flow behavior index and t he consist ency coefficient : = 0.29, =
23,300 Pa s
n
. Eq. [4.40] must be evaluat ed t o det er mine t he component
of t he ent r ance pr essur e dr op due t o shear :
Given, = 0.00159 m and = 0.0075 m, t he equat ion may be simplified:
Calculat ing for each value of shows t hat t he pr essur e dr op,
due t o shear flow, is a small por t ion of t he t ot al pr essur e dr op at t he
ent r ance (Table 4.4). The component of t he ent r ance pr essur e dr op due
t o ext ensional flow is found fr om Eq. [4.36]:

B
9.55(10
4
) (

h
)
.08 1
at 15C

B
3.55(10
4
) (

h
)
.10 1
at 20C
n 1

/4 rad 45
n K
P
en, S
()
n

3n + 1
4n
_

,
n
2K
3n tan

R
R
b
_

,
3n
_

,
R R
b
P
en, S
()
.29

3(.29) + 1
4(.29)
_

,
.29
2(23, 300)
3(.29) tan(/4)

.00159
.0075
_

,
3(.29)
_

,
()
.29
(45, 560)
P
en, S

P
en, E
P
en
P
en, S
288 Chapter 4. Extensional Flow
Result s ar e summar ized in Table 4.4.
Ext ensional flow par amet er s may be est imat ed fr om r egr ession of
Eq. [4.50]:
yielding t he slope,
and int er cept ,
Subst it ut ing appr opr iat e values int o t he int er cept equat ion gives
which can be solved for t he ext ensional consist ency coefficient :
Aver age ext ensional viscosit y is det er mined fr om Eq. [4.51]:
wit h t he aver age ext ensional st r ain r at e at t he die evaluat ed fr om Eq.
[4.52]:
The Tr out on number ( ) is calculat ed fr om Eq. [1.82] wher e is
evaluat ed, as uniaxial ext ension, at a shear r at e numer ically equal t o
:
Result s ar e summar ized in Table 4.4.
ln(P
en, E
) m ln+ ln

2K
E
3m
_

tan
2
_

,
m

R
R
b
_

,
3m
_

,
1
1
]
m
d(lnP
en
)
d(ln)
0.159
ln

2K
E
3m
_

tan
2
_

,
m

R
R
b
_

,
3m
_

,
1
1
]
14.479
K
E

2
3(.159)
_

tan(/4)
2
_

,
(.159)

.00159
.0075
_

,
3(.159)
_

,
K
E
(1.96) exp(14.479)
K
E
991, 000 Pa s
m

E
K
E
(

E, R
)
m 1
991, 000(

E, R
)
.159 1

E, R

tan(/4)
2
(.5)

E
/

E, R
K(

E, R
)
n 1
23, 300(

E, R
)
.29 1
4.9.4 45 Converging Die, Gibsons Method 289
Table 4.4. Capillary Viscometer Data for Defatted Soy Flour Dough (34.7%) at
Room Temperature (Data from Morgan, 1979)
(1/ s ) (MPa ) (k Pa ) (MPa ) (1/ s ) (kPa s
m
) (kPa s
n
) (-)
47.4 3.58 139.5 3.58 23.7 65.1 1.65 39
47.4 3.58 139.5 3.58 23.7 65.1 1.65 39
47.4 3.58 139.5 3.58 23.7 65.1 1.65 39
94.8 4.63 170.6 4.63 47.4 38.2 1.01 38
94.8 4.63 170.6 4.63 47.4 38.2 1.01 38
94.8 4.63 170.6 4.63 47.4 38.2 1.01 38
190.0 4.38 208.7 4.38 95.0 21.2 0.614 35
190.0 4.38 208.7 4.38 95.0 21.2 0.614 35
190.0 4.38 208.7 4.38 95.0 21.2 0.614 35
948.0 6.17 332.6 6.17 474 5.44 0.196 28
948.0 6.17 332.6 6.17 474 5.44 0.196 28
948.0 6.17 332.6 6.17 474 5.44 0.196 28
4.9.4. 45 Convergi ng Di e, Gi bsons Method
Reexamine t he soy dough dat a given in t he pr evious example pr oblem
using Gibsons equat ions (Sec. 4.4.2) t o separ at e t he pr essur e dr op int o
t he shear and ext ensional component s. Assume t he nat ur al angle of
conver gence ( ) pr oduced by t he mat er ial is r adians or 45 degr ees.
Analysis of t he dat a conduct ed in Example 2.12.2 gave values of t he
flow behavior index and t he consist ency coefficient : = 0.29, = 23,300
Pa s
n
. Eq. [4.56] must be evaluat ed t o det er mine t he component of t he
ent r ance pr essur e dr op due t o shear :
Given, = 0.00159 m and = 0.0075 m, t he equat ion may be simplified:
Ent r ance pr essur e loss due t o ext ensional flow is found, fr om Eq. [4.36],
as
Result s of t he pr essur e loss calculat ions ar e summar ized in Table 4.5.
Using t hese dat a, ext ensional flow par amet er s ar e est imat ed fr om
r egr ession of Eq. [4.60]:
(P)
en
(P)
en, S
(P)
en, E

E, R

E
N
Tr

/4
n K
P
en, S

2K(sin
3n
)
3n
1 + 3n

1 + 3n
4n
_

,
n

R
R
b
_

,
3n
_

,
R R
b
P
en, S

2(23, 300) (sin
3(.29)
(/4))
3(.29) (/4)
1 + 3(.29)

1 + 3(.29)
4(.29)
_

,
.29

.29

.00159
.0075
_

,
3(.29)
_

,
52, 995( )
.29
P
en, E
P
en
P
en, S
290 Chapter 4. Extensional Flow
Table 4.5. Capillary Viscometer Data for Defatted Soy Flour Dough (34.7%) at
Room Temperature (Data from Morgan, 1979)
(1/ s ) (MPa ) (k Pa ) (MPa ) (1/ s ) (kPa s
m
) (kPa s
n
) (-)
47.4 3.58 162.3 3.42 14.3 92.1 2.38 39
47.4 3.58 162.3 3.42 14.3 92.1 2.38 39
47.4 3.58 162.3 3.42 14.3 92.1 2.38 39
94.8 4.63 198.4 4.43 28.6 51.4 1.46 35
94.8 4.63 198.4 4.43 28.6 51.4 1.46 35
94.8 4.63 198.4 4.43 28.6 51.4 1.46 35
190.0 4.38 242.7 4.14 57.4 28.6 0.889 32
190.0 4.38 242.7 4.14 57.4 28.6 0.889 32
190.0 4.38 242.7 4.14 57.4 28.6 0.889 32
948.0 6.17 386.8 5.78 286 7.40 0.284 26
948.0 6.17 386.8 5.78 286 7.40 0.284 26
948.0 6.17 386.8 5.78 286 7.40 0.284 26
yielding t he slope,
and int er cept ,
Subst it ut ing known values int o t he int er cept equat ion gives
which can be solved for t he ext ensional consist ency coefficient :
Hence, t he aver age ext ensional viscosit y can be expr essed as
wher e t he aver age ext ensional st r ain r at e at t he die is det er mined fr om
Eq. [4.59]:
(P)
en
(P)
en, S
(P)
en, E

E, R

E
N
Tr
ln(P
en, E
) m ln() + ln

K
E

2
3m

(sin) (1 + cos )
4
_

,
m

R
R
b
_

,
3m
+

4
m
_

,
1
1
]
1
1
]
m
d(lnP
en, E
)
d(ln)
0.158
ln

K
E

2
3m

(sin) (1 + cos )
4
_

,
m

R
R
b
_

,
3m
+

4
m
_

,
1
1
]
1
1
]
14.478
K
E

2
3(.156)

(sin(/4)) (1 + cos(/4))
4
_

,
.156

.00159
.0075
_

,
3(.156)
+
.1516
4
.156
_

,
1
1
]
K
E
(2.24) exp(14.478)
K
E
865, 467 Pa s
m

E
K
E
(

E, R
)
m 1
865, 467(

E, R
)
.158 1
4.9.5 Lubricated Squeezing Flow of Peanut Butter 291
The Tr out on number ( ) is calculat ed fr om Eq. [1.82] wher e is
evaluat ed at a shear r at e of :
Result s ar e given in Table 4.5.
Compar ing solut ions fr om t he Cogswell (Table 4.4) and Gibson
(Table 4.5) met hods shows pr act ically ident ical Tr out on Number s.
Values of t he ext ensional and st eady shear viscosit ies ar e t he same or der
of magnit ude but significant ly differ ent . Since fluid mot ion in a con-
ver gence has bot h shear and ext ensional component s, it is not consid-
er ed pur e flowmaking it difficult t osay t hat one met hod is quant it at ively
super ior t o t he ot her . Bot h ar e r easonable and pr ovide a good basis for
invest igat ing ext ensional flow of food in a conver gence.
4.9.5. Lubri cated Squeezi ng Flow of Peanut Butter
Dat a for t he lubr icat ed squeezing flow of peanut but t er wer e collect ed
for t he case of const ant displacement and const ant ar ea (Fig. 4.11, Table
4.6). Assume power law fluid behavior , and det er mine t he st eady shear
r heological pr oper t ies ( and ) of t he mat er ial. Also, comput e t he
biaxial ext ensional st r ain r at e and t he compar able shear r at e.
Using t he linear por t ion of t he cur ve (Fig. 4.16: ) wher e
st eady flow has been achieved, r egr ession of Eq. [4.71],
wher e, fr om Eq. [4.72],
yields = 0.79 and = -2.118. Solving for gives
The biaxial ext ensional st r ain r at e is

E, R

(sin) (1 + cos )
4

(sin(/4)) (1 + cos(/4))
4
(.302)

E
/

E, R
K(

E, R
)
n 1
23, 300(

E, R
)
.29 1
K n
1/h(t ) > 200 1/m
ln(F) C
1
+ n ln

1
h
_

,
C
1
ln(3
(n + 1)/2
R
2
Ku
z
n
)
n C
1
K
2.118 ln[3
(.79 + 1)/2
(.0318)
2
K

.05
(100) (60)
_

,
.79
]
K 146 kPa s
n
292 Chapter 4. Extensional Flow
Figure 4.16. Raw data of force versus reciprocal height for the lubricated squeez-
ing flow of peanut butter at 23 C.
and a compar able shear r at e, based on our consider at ion of t he Tr out on
number (Eq. [1.84]), is
Sample calculat ions for t he fir st dat a point ( = 230 m
-1
) included in
t he st r aight line r elat ionship bet ween and , ar e
and
Result s ar e summar ized in Table 4.6. The appr oximat e shear r at e r ange
cover ed in det er mining and was 0.00332 t o 0.00938 s
-1
.
100 200 300 500 700 1,000
1
2
3
5
10
20
30
50
100
1/h(t), 1/m
F
,

N
Peanut Butter

B

u
z
2(h
o
u
z
t )

u
z
2

1
h
_

12

B
1/h
F 1/h

B

u
z
2

1
h
_

,

.05(230)
2(60) (100)
9.58(10)
4
s
1

12

12 (9.58) (10)
4
3.32(10)
3
s
1
K n
4.9.5 Lubricated Squeezing Flow of Peanut Butter 293
Table 4.6. Lubricated Squeezing Flow ( = 7.14 mm, = 0.0318 m, = 0.05
cm/min) Datafor Peanut Butter at 23 C(DatafromCampanellaand Peleg, 1987c)
F h (t ) 1/ h (t ) ln (F) ln (1/ h (t ))
N mm 1/ m 1/ s 1/ s
2.25 6.67 150 0.811 5.012 - -
3.16 6.25 160 1.151 5.075 - -
4.50 5.88 170 1.504 5.136 - -
5.75 5.56 180 1.749 5.193 - -
6.90 5.26 190 1.932 5.247 - -
7.50 5.00 200 2.015 5.298 - -
8.50 4.44 230 2.140 5.438 9.58E-4 3.32E-3
9.50 3.85 260 2.251 5.561 1.08E-3 3.75E-3
10.9 3.33 300 2.389 5.704 1.25E-3 4.33E-3
12.3 3.03 330 2.509 5.799 1.38E-3 4.76E-3
14.0 2.50 400 2.639 5.991 1.67E-3 5.77E-3
15.3 2.27 440 2.728 6.087 1.83E-3 6.35E-3
16.7 1.96 510 2.815 6.234 2.13E-3 7.36E-3
18.2 1.67 600 2.901 6.397 2.50E-3 8.66E-3
19.5 1.54 650 2.970 6.477 2.71E-3 9.38E-3
h
o
R u
z

12

B
Chapt e r 5 . Vis c oe las t ic it y
5.1. Introducti on
Rheology is t he science of t he defor mat ion and flow of mat t er . Ther e
ar e t hr ee ways t o defor m a subst ance: shear , ext ension, and bulk
compr ession. Shear and ext ensional defor mat ion have been t hor oughly
discussed in ear lier chapt er s. Bulk compr ession, wher e a unifor m
change in ext er nal pr essur e pr oduces a volumet r ic change in t he
mat er ial, was br iefly discussed in defining t he bulk modulus (Sec. 1.4).
It is possible t o conduct t est s in all t hr ee modes of defor mat ion, under
st eady st at e or dynamic condit ions, and compar e t he r esult ing moduli
and compliances (Fer r y, 1980). This chapt er will focus on viscoelast ic
mat er ial funct ions det er mined fr om shear and ext ensional defor mat ion
dat a. Bulk compr ession t est ing is not consider ed her e because t ech-
nology in t he ar ea is not well developed. Fut ur e r esear ch may show t he
concept t o be valuable in defining a pr essur e dependent viscosit y
funct ion t o examine high pr essur e pr ocesses such as food ext r usion.
Vi scoelasti c Materi al Functi ons. In Chapt er s 2 (Tube Viscomet r y)
and 3 (Rot at ional Viscomet r y) t he emphasis was on measur ement
met hods t o det er mine flow cur ves for non-Newt onian mat er ials under
st eady shear condit ions. All foods have unique flow cur ves and t his
infor mat ion is ver y useful in a lar ge number of indust r ial applicat ions.
Clear ly, fr om an engineer ing st andpoint , t he st eady flow cur ve is t he
most valuable way t o char act er ize t he r heological behavior of fluid foods.
St eady shear viscosit y is a pr oper t y of all fluids r egar dless of whet her
or not t hey exhibit elast ic behavior ; however , many phenomena cannot
be descr ibed by t he viscosit y funct ion alone and elast ic behavior must
be t aken int o consider at ion. This chapt er will invest igat e exper iment al
met hods t o gener at e dat a t hat r eflect t he combined viscous and elast ic
char act er of mat er ials.
In st eady shear , viscoelast ic fluids exhibit nor mal st r esses and
measur ing t hem pr ovides one way of char act er izing elast icit y. Nor mal
st r ess differ ences (Eq. [1.23] and Eq. [1.24]) can be measur ed on r ot a-
t ional r heomet er s pr oduced by var ious inst r ument companies. Com-
put at ions (Eq. [3.57]) r equir e an evaluat ion of axial for ce under st eady
shear condit ions. Unst eady st at e shear measur ement s pr ovide a
dynamic means of evaluat ing viscoelast icit y. The t wo major cat egor ies
of unst eady shear t est ing ar e t r ansient and oscillat or y.
5.1 Introduction 295
Table 5.1. Material Functions Determined in Transient Shear Flow Experiments
St art -up flow: Ma t er ia l a t r es t is s u dden ly s u bject ed t o a con s t a n t s h ea r r a t e.
Sh ea r s t r es s gr owt h fu n ct ion
Fir s t n or ma l s t r es s gr owt h fu n ct ion
Secon d n or ma l s t r es s gr owt h fu n ct ion
Ce s s at ion of s t e ady s he ar flow: Ma t er ia l u n der goin g s t ea dy s t a t e s h ea r flow is
s u dden ly br ou gh t t o r es t .
Sh ea r s t r es s deca y fu n ct ion
Fir s t n or ma l s t r es s deca y fu n ct ion
Secon d n or ma l s t r es s deca y fu n ct ion
St e p s t rain: Ma t er ia l a t r es t is given a s u dden s t ep in cr ea s e in s t r a in .
Sh ea r s t r es s r ela xa t ion fu n ct ion
Fir s t n or ma l s t r es s r ela xa t ion fu n ct ion
Secon d n or ma l s t r es s r ela xa t ion fu n ct ion
Cre e p: Ma t er ia l a t r es t is given a s u dden s t ep in cr ea s e in s t r es s .
Sh ea r cr eep complia n ce
St ea dy-s t a t e complia n ce
Re c oil: St r es s , in a flu id in s t ea dy s t a t e s h ea r flow, is s u dden ly br ou gh t t o zer o.
Ma t er ia l is con s t r a in ed in on e dir ect ion a n d r ecoil (t h e degr ee of r et r a ct ion ) is mea -
s u r ed in t h e s econ d dir ect ion .
Recoil fu n ct ion
Ult ima t e r ecoil fu n ct ion
Tr ansient shear t est ing (Table 5.1) is a cat egor y t hat includes
numer ous measur ement concept s: st ar t -up flow, cessat ion of st eady
shear flow, st ep st r ain, cr eep, and r ecoil. Dat a gener at ed using t hese
met hods may lead t o numer ous mat er ial funct ions (Table 5.1) such as
t he shear st r ess gr owt h funct ion, shear st r ess decay funct ion, shear
st r ess r elaxat ion funct ion, shear cr eep compliance, and t he r ecoil
funct ion. In oscillat or y t est ing, a sample is subject ed t o har monically
var ying (usually sinusoidal) small amplit ude defor mat ions in a simple
shear field. Var ious companies make inst r ument s t o accomplish
oscillat or y t est s which have pr oved t heir usefulness in addr essing
numer ous food indust r y pr oblems. Many funct ions (Table 5.2) can be
gener at ed fr om oscillat or y exper iment s. A compr ehensive list of vis-
coelast ic mat er ial funct ions and coefficient s is available in Dealy (1994).
Ext ensional flow was invest igat ed in Chapt er 4. What has been
discussed above, r elat ive t o shear flow, is also t r ue for ext ensional flow.
The ext ensional viscosit y funct ion is det er mined in st eady-st at e
ext ensional flow, but t r ansient exper iment s gener at e dat a which r eflect
296 Chapter 5. Viscoelasticity
t he viscoelast ic char act er of t he mat er ial. The following t est s, for
example, could be conduct ed in t ensile ext ension: t ensile st ar t -up, ces-
sat ion of st eady t ensile ext ension, t ensile st ep st r ain, t ensile cr eep, and
t ensile r ecoil. This t ype of exper iment al t est ing leads t o numer ous
viscoelast ic mat er ial funct ions (Table 5.3). Similar funct ions could be
gener at ed for exper iment s involving biaxial or planar ext ensional flow.
Table 5.2. Material Functions Determined in Oscillatory Shear Testing (Har-
monically Varying Simple Shear) Experiments
Complex vis cos it y
Dyn a mic vis cos it y
Ou t -of-ph a s e compon en t of t h e complex vis cos it y
Complex s h ea r modu lu s
Sh ea r s t or a ge modu lu s
Sh ea r los s modu lu s
Complex s h ea r complia n ce
Sh ea r s t or a ge complia n ce
Sh ea r los s complia n ce
Li near Versus Non-li near Vi scoelasti ci ty. In pr ocess engineer ing,
dat a on viscoelast icit y may be ver y helpful in under st anding var ious
pr oblems. Shear cr eep dat a, for example, ar e useful in examining
gr avit y dr iven phenomena such as coat ing and sagging. This infor -
mat ion can also be an invaluable t ool in pr oduct development . Means
of evaluat ing linear viscoelast ic behavior ar e t he pr imar y foci of t his
chapt er . When mat er ials ar e t est ed in t he linear r ange, mat er ial
funct ions do not depend on t he magnit ude of t he st r ess, t he magnit ude
of t he defor ming st r ain, or t he r at e of applicat ion of t he st r ain. If linear ,
an applied st r ess will pr oduce a pr opor t ional st r ain r esponse. Doubling
t he st r ess, for example, will double t he st r ain r esponse. The linear r ange
of t est ing is det er mined fr om exper iment al dat a. Test ing can easily
ent er t he non-linear r ange by applying excessive st r ain (usually gr eat er
t han 1%) or high defor mat ion r at es t o a sample.
The impor t ance of lar ge defor mat ion (non-linear ) behavior in food
r heology should not be over looked. Many pr ocesses, such as mast icat ion
and swallowing, ar e only accomplished wit h ver y lar ge defor mat ions.
Collect ing viscoelast ic dat a r elevant t o t his t ype of pr oblem involves
5.2 Transient Tests for Viscoelasticity 297
t est ing in t he non-linear r ange of behavior . These dat a may be useful
in at t acking pr act ical pr oblems; however , fr om a fundament al st and-
point , t hey can only be used for compar at ive pur poses because t he
t heor et ical complexit y of non-linear viscoelast icit y makes it impr act ical
for most applicat ions. Mor e r esear ch is needed in t his ar ea.
Table 5.3. Experimental Tests and Material functions Determined in Transient
Tensile Extension
Te ns ile s t art -up: Ma t er ia l a t r es t is s u dden ly s u bject ed t o a con s t a n t ext en s ion a l
s t r a in r a t e.
Ten s ile s t r es s gr owt h fu n ct ion
Ce s s at ion of s t e ady t e ns ile e xt e ns ion: Ma t er ia l s u bject ed t o a s t ea dy s t a t e
ext en s ion a l s t r a in r a t e is s u dden ly br ou gh t t o r es t .
Ten s ile s t r es s deca y coefficien t
Te ns ile s t e p s t rain: Ma t er ia l a t r es t is given a s u dden s t ep in cr ea s e in s t r a in .
Ten s ile r ela xa t ion modu lu s
Te ns ile c re e p: Ma t er ia l a t r es t is given a s u dden s t ep in cr ea s e in t en s ile s t r es s .
Ten s ile cr eep complia n ce
Ten s ile cr eep r a t e deca y fu n ct ion
Te ns ile re c oil: Ma t er ia l s u bject t o s t ea dy s t a t e s t r es s a n d s t r a in h a s t h e s t r es s
s u dden ly r edu ced t o zer o.
Ten s ile r ecoil fu n ct ion
Ult ima t e t en s ile r ecoil fu n ct ion
5.2. Transi ent Tests for Vi scoelasti ci ty
In t his sect ion, t hr ee t ypical t r ansient t est s of viscoelast icit y ar e
pr esent ed: st ep st r ain (st r ess r elaxat ion), cr eep, and st ar t -up flow.
These t est s involve small st r ains and can be conduct ed wit h commer -
cially available or easily const r uct ed inst r ument s. Alt hough t her e ar e
numer ous pot ent ial exper iment al met hods (Table 5.1 and 5.3) t o
elucidat e viscoelast icit y, many ar e difficult t o per for m and quest ionable
for use on food pr oduct s. Pr act ical r heology is applicat ion dr iven so t he
most appr opr iat e t est for a par t icular mat er ial depends on t he pr oblem
under consider at ion. Test s descr ibed in t his sect ion have all been
conduct ed on food, and pr oduced useful r esult s.
298 Chapter 5. Viscoelasticity
5.2.1. Mechani cal Analogues
Massless mechanical models, composed of spr ings and dashpot s, ar e
useful in concept ualizing r heological behavior . The spr ing is consider ed
an ideal solid element obeying Hookes law:
[5.1]
and t he dashpot is consider ed an ideal fluid element obeying Newt ons
law:
[5.2]
Spr ings and dashpot s can be connect ed in var ious ways t o por t r ay t he
behavior of viscoelast ic mat er ials; however , a par t icular combinat ion of
element s is not unique because many differ ent combinat ions can be used
t o model t he same set of exper iment al dat a. The most common
mechanical analogs of r heological behavior ar e t he Maxwell and Kelvin
(somet imes called Kelvin-Voigt ) models depict ed in Fig. 5.1.
Figure 5.1. Maxwell and Kelvin models.
Mechanical analogues pr ovide a useful means of invest igat ing cr eep
and st ep st r ain dat a. These dat a may also be pr esent ed in t er ms of
var ious compliance and modulus dist r ibut ion funct ions (or spect r a) as
well as elect r ical models (Mohsenin, 1986; Polakowski and Ripling,
1966; Sher man, 1970; Bar nes et al., 1989; Whor low, 1992; Fer r y, 1980).
In addit ion, st ep st r ain and cr eep cur ves can be nor malized and pr es-
ent ed in linear for m (Peleg, 1980). This t echnique can be quit e useful
in biological mat er ials wher e it is oft en difficult t o achieve equilibr ium
condit ions.
G

Maxwell
G
Kelvin
G
5.2.2 Step Strain (Stress Relaxation) 299
5.2.2. Step Strai n (Stress Relaxati on)
In a st ep st r ain t est t he sample is given an inst ant aneous st r ain and
t he st r ess r equir ed t omaint ain t he defor mat ion is obser ved as a funct ion
of t ime. This exper iment is commonly known as a "st r ess r elaxat ion"
t est and it may be conduct ed in shear , uniaxial t ension, or uniaxial
compr ession. St r ess r elaxat ion dat a may also be obt ained by subject ing
a fluid t o a const ant r at e of st r ain (in, for example, a concent r ic cylinder
viscomet er ), t hen suddenly st opping t he defor mat ion and obser ving t he
change in st r ess over t ime.
Figure 5.2. Stress relaxation curves.
A wide r ange of behavior may be obser ved in st r ess r elaxat ion t est s
(Fig. 5.2). No r elaxat ion would be obser ved in ideal elast ic mat er ials
while ideal viscous subst ances would r elax inst ant aneously. Viscoe-
last ic mat er ials would r elax gr adually wit h t he end point depending on
t he molecular st r uct ur e of t he mat er ial being t est ed: st r ess in
viscoelast ic solids would decay t o an equilibr ium st r ess ( ), but t he
r esidual st r ess in viscoelast ic liquids would be zer o.
St r ess r elaxat ion dat a ar e commonly pr esent ed in t er ms of a st r ess
r elaxat ion modulus:
o
0
S
t
r
a
i
n
S
t
r
e
s
s
t=0
0
Ideal Elastic Material
Viscoelastic Solid
Viscoelastic Liquid
Ideal Viscous Material
Time
e

e
> 0
300 Chapter 5. Viscoelasticity
[5.3]
If a mat er ial is per fect ly elast ic, t he r elaxat ion modulus is equal t o t he
shear modulus defined by Eq. [1.10]: . is a similar , but
t ime-dependent , quant it y det er mined fr om exper iment al dat a. Cur ves
of t he st r ess r elaxat ion modulus ver sus t ime, gener at ed at differ ent
levels of st r ain, over lap if dat a ar e collect ed in t he linear viscoelast ic
r egion of mat er ial behavior . Relat ed funct ions (Fer r y, 1980) can be
found in t ension ( ) and bulk compr ession ( ).
The Maxwell model, one which cont ains a Hookean spr ing in ser ies
wit h a Newt onian dashpot , has fr equent ly been used t o int er pr et st r ess
r elaxat ion dat a for viscoelast ic liquids, par t icular ly polymer ic liquids.
The t ot al shear st r ain in a Maxwell fluid element (Fig. 5.1) is equal t o
t he sum of t he st r ain in t he spr ing and t he dashpot :
[5.4]
Differ ent iat ing Eq. [5.4] wit h r espect t o t ime, and using t he definit ions
pr ovided by Eq. [5.1] and [5.2], yields
[5.5]
or
[5.6]
wher e t he r elaxat ion t ime (also called t he char act er ist ic t ime of a
Maxwell fluid) is defined as
[5.7]
Alt hough an exact definit ion of is difficult , it can be t hought of as t he
t ime it t akes a macr omolecule t o be st r et ched out when defor med.
Relaxat ion t imes for common fluids var y a gr eat deal as shown by t he
infor mat ion in Table 5.4. The above equat ions ar e pr esent ed in t er ms
of shear defor mat ion. If t est ing is conduct ed in uniaxial t ension or
compr ession, t hen t he r elaxat ion t ime can be t hought of in t er ms of an
ext ensional viscosit y ( ) and Youngs modulus ( ).
G f (t )

constant
G / G(t )
E(t ) K(t )
()
spring
+ ()
dashpot
d
dt



1
G

d
dt
_

,
+

+
rel

d
dt
_

rel


G

rel

E
E
5.2.2 Step Strain (Stress Relaxation) 301
Table 5.4. Relaxation Time and Limiting Viscosity at Zero Shear for various
Viscoelastic Fluids (Source: Tanner, 1985).
Flu id T
( C) (s ) (Pa s )
Wa t er 20 1E-12 0.001
Min er a l oil 30 7E-10 0.5
Poly-dimet h yls iloxa n e 30 1E-6 0.3
125 1.7E-4 100
Low-den s it y polyet h ylen e 115 10 2E5
240 0.1 3000
High -den s it y polyet h ylen e 180 0.07 2000
220 0.05 1000
High -impa ct polys t yr en e 170 7 2E5
210 3 1E5
0.5% Hydr oxyet h yl-cellu los e 27 0.1 1.3
in wa t er
Gla s s 27 > 1E5 > 1E18
The Maxwell model is useful in under st anding st r ess r elaxat ion
dat a. Consider a st ep st r ain (st r ess r elaxat ion) exper iment wher e t her e
is a sudden applicat ion of a const ant shear st r ain, . When t he st r ain
is const ant , t he shear r at e is equal t o zer o ( ) and Eq. [5.6] becomes
[5.8]
This equat ion may be int egr at ed using t he init ial condit ion t hat
at :
[5.9]
or , aft er evaluat ing t he int egr al,
[5.10]
Eq. [5.10] descr ibes t he gr adual r elaxat ion of st r ess (fr om t o zer o)
aft er t he applicat ion of a sudden st r ain. The r elat ionship pr ovides a
means of det er mining t he r elaxat ion t ime: is t he t ime it t akes for t he
st r ess t o decay t o 1/e (appr oximat ely 36.8%) of it s init ial value.

rel

o

0.0
+
rel

d
dt
_

,
0

o
t 0

0
t

dt

rel
f (t )
o
exp

rel
_

rel
302 Chapter 5. Viscoelasticity
Exper iment al dat a show t hat t he Maxwell model does not account
for t he st r ess r elaxat ion behavior of many viscoelast ic mat er ials because
it does not include an equilibr ium st r ess ( ). This pr oblem may be
addr essed for numer ous foods by const r uct ing a model which has a single
Maxwell element connect ed in par allel wit h a spr ing. The st r ess
r elaxat ion equat ion descr ibed by t his mechanical model (Fig. 5.3a) is
[5.11]
wit h t he fr ee spr ing (wher e ) account ing for t he equilibr ium
st r ess (Fig. 5.4). The r elaxat ion t ime is defined in t er ms of t he st andar d
Maxwell por t ion of t he model: . Applicat ion of Eq. [5.11] is
illust r at ed in Example Pr oblem 5.8.1.
Figure 5.3. Maxwell elements in parallel with a spring: a) one Maxwell element
and a free spring, b) three Maxwell elements and a free spring.
Added complexit y can be obt ained by const r uct ing a mor e gener al-
ized Maxwell model consist ing of sever al Maxwell element s in par allel
wit h an independent spr ing. If t he syst em is subject ed t o a const ant
st r ain, t he t ot al st r ess is t he sum of t he individual st r esses in each
element . In a four element model, a model cont aining t hr ee Maxwell
element s and a spr ing (Fig. 5.3b), t he solut ion for st r ess as a funct ion
of t ime is
[5.12]

e
f (t )
e
+ (
o

e
) exp

rel
_

e

o
G
0

rel

1
/G
1
G
0
1
G
1
G
1 2 3
G G
G
0
1 2 3
a
b
f (t ) A
1
exp

t G
1

1
_

,
+ A
2
exp

t G
2

2
_

,
+ A
3
exp

t G
3

3
_

,
+
o
G
0
5.2.2 Step Strain (Stress Relaxation) 303
or
[5.13]
wher e t he subscr ipt s r efer t o differ ent mechanical element s in t he
syst em. Each Maxwell element may have a differ ent r elaxat ion t ime.
This concept can be gener alized t o det er mine a r elaxat ion spect r a for a
viscoelast ic mat er ial (Fer r y, 1980). Eq. [5.10], [5.11], [5.12], and [5.13]
can be expr essed in t er ms of t he r elaxat ion modulus, defined by Eq.
[5.3], by dividing each equat ion by t he applied st r ain.
Figure 5.4. Typical stress relaxation curve modeled with Eq. [5.11] describing a
single Maxwell element in parallel with a spring.
Peleg and Nor mand (1983) not ed t wo major pr oblems in collect ing
st r ess r elaxat ion dat a for foods: 1) When subject ed t o lar ge defor mat ion
t hey usually exhibit non-linear viscoelast ic behavior ; 2) Nat ur al
inst abilit y or biological act ivit y make it difficult t o det er mine equilib-
r ium mechanical par amet er s. To over come t hese difficult ies, t hey
f (t ) A
1
exp

t
(
rel
)
1
_

,
+ A
2
exp

t
(
rel
)
2
_

,
+ A
3
exp

t
(
rel
)
3
_

,
+
e
Time
S
t
r
e
s
s
=
e
+ .368 (
o
-
e
) =
e
+ (
-
e
)
exp (
-t
_
)
e
o
o
rel
rel
304 Chapter 5. Viscoelasticity
suggest st r ess r elaxat ion dat a be calculat ed as a nor malized st r ess (a
nor malized for ce t er m is also accept able) and fit t o t he following linear
equat ion:
[5.14]
wher e is t he init ial st r ess, is t he decr easing st r ess at t ime , and
and ar e const ant s. The r ecipr ocal of depict s t he init ial decay r at e
and is a hypot het ical value of t he asympt ot ic nor malized for ce. Fit t ing
exper iment al dat a t o Eq. [5.14] is a quick and effect ive way t o handle
st r ess r elaxat ion dat a for many foods. Typical values of and ar e
summar ized in Table 5.5. Also, t he t echnique is illust r at ed for apple
t issue in Example Pr oblem 5.8.2.
Table 5.5. Stress Relaxation Parameters of Eq. [5.14] for Various Biological
Materials (Source: Peleg and Normand, 1983)
Ma t er ia l
(s or min ) -
Ch edda r Ch ees e 3.23 min 1.11
Cor n Gr a in s 10.9 min 5.18
Low Met h oxyl Pect in 68.2 s 1.21
Gel (No.31)
Pea Bea n s 2.41 s 2.26
Pot a t o Fles h 4.40 s 1.56
5.2.3. Creep and Recovery
In a cr eep t est , an inst ant aneous st r ess is applied t o t he sample and
t he change in st r ain (called t he cr eep) is obser ved over t ime. When t he
st r ess is r eleased, some r ecover y may be obser ved as t he mat er ial
at t empt s a r et ur n t o t he or iginal shape. Incr eased availabilit y of con-
t r olled st r ess r heomet er s has gr eat ly impr oved our abilit y t o conduct
shear cr eep and r ecover y exper iment s on a wide var iet y of mat er ials.
These t est s can be par t icular ly useful in st udying t he behavior exist ing
in const ant st r ess envir onment s such as t hose found in leveling, sedi-
ment at ion, and coat ing applicat ions wher e gr avit y is t he dr iving for ce.

o
t

o

k
1
+ k
2
t

o
t k
1
k
2
k
1
k
2
k
1
k
2
k
1
k
2
5.2.3 Creep and Recovery 305
Cr eep exper iment s can also be conduct ed in uniaxial t ension or
compr ession. The analyt ical met hods pr esent ed in t his sect ion have
been used t o st udy t ypical fluid foods like salad dr essing (Par edes et al.,
1989), and ver y complex bodies such as whole or anges (Chuma et al.,
1978).
Figure 5.5. Creep and recovery curves.
Idealized cr eep and r ecover y cur ves ar e illust r at ed in Fig. 5.5.
Subject ed t o a const ant st r ess, st r ain in an ideal elast ic mat er ial would
be const ant due t o t he lack of flow, and t he mat er ial would r et ur n t o t he
or iginal shape upon r emoval of st r ess. An ideal viscous mat er ial would
showst eady flow, pr oducing a linear r esponse t o st r ess wit h t he inabilit y
t o r ecover any of t he imposed defor mat ion. Viscoelast ic mat er ials (e.g.,
br ead dough) would exhibit a nonlinear r esponse t o st r ain and, due t o
t heir abilit y t o r ecover some st r uct ur e by st or ing ener gy, show a per -
manent defor mat ion less t han t he t ot al defor mat ion applied t o t he
sample. This st r ain r ecover y, or cr eep r ecover y, is also called r ecoil and
may be invest igat ed in t er ms of a r ecoil funct ion (Dealy, 1994).
Cr eep dat a may be descr ibed in t er ms of a cr eep compliance funct ion:
0
S
t
r
a
i
n
S
t
r
e
s
s
t=0
0
Time
o
Ideal Elastic Material
Viscoelastic Material
Ideal Viscous Material
Permanent
Deformation
(flow, partial recovery)
(no flow, complete recovery)
(steady flow, no recovery)
306 Chapter 5. Viscoelasticity
[5.15]
Compliance cur ves gener at ed at differ ent st r ess levels over lap when
dat a ar e collect ed in t he r ange of linear viscoelast ic behavior . Wit h a
per fect ly elast ic solid, , t he r ecipr ocal of t he shear modulus;
however , differ ent t ime pat t er ns in exper iment al t est ing mean t hat
. Eq. [5.15] is pr esent ed in t er ms of shear defor mat ion.
Similar mat er ial funct ions (Fer r y, 1980) can be det er mined fr om cr eep
dat a gener at ed in t ension ( ) and bulk compr ession ( ) st udies.
To develop a mechanical analog descr ibing cr eep behavior , t he
st ar t ing point is t he Kelvin model (Fig. 5.1) which cont ains a spr ing
connect ed in par allel wit h a dashpot . When t his syst em is subject ed t o
shear st r ain, t he spr ing and dashpot ar e st r ained equally:
[5.16]
The t ot al shear st r ess ( ) caused by t he defor mat ion is t he sum of t he
individual shear st r esses which, using Eq. [5.1] and Eq. [5.2], can be
wr it t en as
[5.17]
Differ ent iat ing Eq. [5.17] wit h r espect t o t ime yields
[5.18]
wher e t he r et ar dat ion t ime ( ) is unique for any subst ance. If a
mat er ial was a Hookean solid, t he r et ar dat ion t ime would be zer o and
t he maximum st r ain would be obt ained immediat ely wit h t he applica-
t ion of st r ess: Time t o achieve maximum st r ain in viscoelast ic mat er ials
is delayed (or r et ar ded). The r et ar dat ion t ime can be t hought of in t er ms
of ext ensional viscosit y ( ) and Youngs modulus ( ) if t est ing is
conduct ed in uniaxial t ension or compr ession.
In cr eep, wher e t he mat er ial is allowed t o flow aft er being subject ed
t o a const ant shear st r ess ( ), t he change in st r ess wit h t ime is zer o
( ) and t he solut ion t o Eq. [5.18] is
[5.19]
J f (t )

constant
J 1/G
J(t ) 1/G(t )
D(t ) B(t )
()
spring
()
dashpot

G +

1
G
d
dt


+ (
ret
)
d

dt

ret
/G

E
E

o
d/dt 0
f (t )

o
G

1 exp

ret
_

,
_

,
5.2.3 Creep and Recovery 307
showing t hat t he init ial st r ain is zer o ( at ). Eq. [5.19] pr edict s
a st r ain t hat asympt ot ically appr oaches t he maximum st r ain ( )
associat ed wit h t he spr ing. is t he t ime t aken for t he delayed st r ain
t o r each appr oximat ely 63.2% (1-1/e) of t he final value. Mat er ials wit h
a lar ge r et ar dat ion t ime r each full defor mat ion slowly.
The Kelvin model (Fig. 5.1) shows excellent elast ic r et ar dat ion but
is not gener al enough t o model cr eep in many biological mat er ials. The
solut ion t o t his pr oblem is t o use a Bur ger s model (Fig. 5.6) which is a
Kelvin and a Maxwell model placed in ser ies. Dat a following t his
mechanical analog show an init ial elast ic r esponse due t ot he fr ee spr ing,
r et ar ded elast ic behavior r elat ed t o t he par allel spr ing-dashpot
combinat ion, and Newt onian t ype flow aft er long per iods of t ime due t o
t he fr ee dashpot (Fig. 5.7):
[5.20]
wher e , t he r et ar dat ion t ime of t he Kelvin por t ion of t he model.
Figure 5.6. Four element Burgers model.
The Bur ger s model can also be expr essed in t er ms of cr eep com-
pliance by dividing Eq. [5.20] by t he const ant st r ess:
[5.21]
Wr it ing t he r esult as a cr eep compliance funct ion (Eq. [5.15]) yields
0 t 0

o
/G

ret
f (t )

o
G
0
+

o
G
1

1 exp

ret
_

,
_

,
+

o
t

ret

1
/G
1
G
0
1
G
1
0

o
f (t )
1
G
0
+
1
G
1

1 exp

ret
_

,
_

,
+
t

0
308 Chapter 5. Viscoelasticity
[5.22]
wher e is t he inst ant aneous compliance, is t he r et ar ded compliance,
is t he r et ar dat ion t ime ( ) of t he Kelvin component , and is t he
Newt onian viscosit y of t he fr ee dashpot . The sum of and is called
t he st eady st at e compliance. Using t he same pr ocedur e, Eq. [5.19] could
also be expr essed in t er m of t he cr eep compliance funct ion. The Bur ger s
model (Fig. 5.6), less t he fr ee spr ing ( ), is somet imes called t he J effr eys
model. Eq. [5.22] is applied t o skim milk cur d in Example Pr oblem 5.8.3.
Par amet er s of Eq. [5.22], for cr eamy salad dr essing, ar e given in
Appendix 6.19.
Figure 5.7. Typical creep curve showing where various elements of the Burgers
model (Fig. 5.6 and Eq. [5.20]) describe flow behavior.
When conduct ing cr eep exper iment s, cont r olled st r ess r heomet er s
allow one t o measur e t he st r ain r ecover ed when t he const ant st r ess is
r emoved. The complet e cr eep and r ecover y cur ve may be expr essed
using t he Bur ger s model (Fig. 5.6). When calculat ed as compliance, t he
cr eep is given by Eq. [5.22] for wher e is t he t ime when t he
const ant st r ess is r emoved. At t he beginning of cr eep, t her e is an
J f (t ) J
0
+ J
1

1 exp

ret
_

,
_

,
+
t

0
J
0
J
1

ret

1
/G
1

0
J
o
J
1
G
o
Instantaneous Response
Time
S
t
r
a
i
n
Retarded Elastic Behavior
Long Time Viscous Flow
G
0
1
G
1
0
0 < t < t
1
t
1
5.2.3 Creep and Recovery 309
inst ant aneous change in compliance ( ) due t o t he spr ing in t he
Maxwell por t ion of t he model (Fig. 5.8). Then, t he Kelvin component
pr oduces an exponent ial change in compliance r elat ed t ot he r et ar dat ion
t ime. Aft er sufficient t ime has passed, t he independent dashpot (Fig.
5.6) gener at es a pur ely viscous r esponse. Dat a fr om t he linear por t ion
of t he cr eep cur ve (Fig. 5.8) ar e r elat ed t o t wo par amet er s: t he slope is
equal t o ; and t he int er cept , somet imes called t he st eady st at e
compliance, is equal t o .
Figure 5.8. Compliance and recovery (or recoil) curves showing compliance
parameters for the Burgers model (Fig. 5.6 and Eq. [5.22]).
At , t he st r ess is r emoved ( ) and t her e is an inst ant aneous
change in compliance (Fig. 5.8) equal t o . The fr ee dashpot causes
per manent defor mat ion in t he mat er ial r elat ed t o a compliance of .
This fact or is dir ect ly r elat ed t o t he non-r ecover able sample st r ain of
. If a subst ance obeying t he Bur ger s model is t est ed in t he linear
J
0
1/
0
J
0
+ J
1
Time
C
o
m
p
l
i
a
n
c
e
Creep Recovery
t
1
0
0
J
1
J
0
J
0
0
= =
0
t
1 0
/
0
Slope = 1/
J
1
t t
1
0
J
0
t
1
/
0

0
t
1
/
0
310 Chapter 5. Viscoelasticity
viscoelast ic r egion of mat er ial behavior , t hen t he values of and
det er mined fr om t he cr eep cur ve will be equal t o t he values of and
det er mined fr om t he r ecover y cur ve.
If necessar y, addit ional Kelvin element s can be added t o t he Bur ger s
model t o bet t er r epr esent exper iment al dat a. Mat hemat ically, t his idea
can be descr ibed wit h t he following equat ion:
[5.23]
wher e is t he t ot al number of Kelvin element s in t he model, each having
a unique r et ar ded compliance and r et ar dat ion t ime. A syst em wit h t wo
Kelvin element s ( ) was used by Halim and Shoemaker (1990) t o
model skim milk cur d. This model is explor ed in Example Pr oblem 5.8.3.
The same equat ion wor ked well for Pur kayast ha et al. (1985) in
evaluat ing t he compr essive cr eep behavior of pot at o flesh and cheddar
cheese. Balaban et al. (1988) pr esent ed syst emat ic pr ocedur es t o
det er mine t he const ant s involved in Eq. [5.23]. Some advanced
r heomet er s pr ovide comput er soft war e t o gener at e appr opr iat e con-
st ant s. Asimple linear ized model (pr esent ed in Example Pr oblem 5.8.3),
similar t o Eq. [5.14], has been suggest ed by Peleg (1980) t o char act er ize
t he cr eep of biological mat er ials.
5.2.4. Start-Up Flow (Stress Overshoot)
Dur ing st ar t -up flow, a shear r at e is suddenly imposed on a vis-
coelast ic fluid held pr eviously at r est . Shear st r ess pr oduced by t his
t r ansient defor mat ion displays an init ial over shoot befor e r eaching a
st eady-st at e value; hence, t he phenomenon is commonly r efer r ed t o as
st r ess over shoot . Result s can be used t o pr oduce a shear st r ess gr owt h
funct ion. The gener al behavior can be modeled using an empir ical
equat ion developed by Leider and Bir d (1974) which includes r heological
pr oper t ies r elat ed t o t he fir st nor mal st r ess differ ence, and t he shear
r at e. This equat ion, usually called t he Bir d-Leider equat ion, is
expr essed as a funct ion of t ime and t he applied shear r at e:
[5.24]
wher e:
[5.25]
J
0
J
1
J
0
J
1
J f (t ) J
0
+
i 1
m

J
i

1 exp

t
(
ret
)
i
_

,
_

,
1
1
]
+
t

0
m
m 2

21
+
f (

, t ) K(

)
n

1 + (b

t 1) exp

t
an
_

,
_

K
2K
_

,
1/(n n)
5.2.4 Start-Up Flow (Stress Overshoot) 311
[5.26]
[5.27]
is t he t ime const ant while and ar e adjust able par amet er s. Over
shor t per iods of t ime, t he equat ion models t he elast ic r esponse (sudden
over shoot ) of t he mat er ial. Once t he peak t or que is r eached, exponent ial
decay is simulat ed. Aft er long per iods of t ime, t he Bir d-Leider equat ion
collapses t o t he st andar d power law equat ion:
[5.28]
Typical cur ves of Eq. [5.24], modeling t he behavior of mayonnaise, ar e
illust r at ed in Fig. 5.9. The infor mat ion is plot t ed in t er ms of a
dimensionless shear st r ess:
[5.29]
Figure 5.9. Typical stress overshoot curves at various shear rates for mayonnaise
at 25 C (based on data from Kokini and Dickie, 1981).

21
K(

)
n

11

22
K(

)
n
a b

21
+
f (

, t ) K(

)
n

21
+

21
+
1 + (b

t 1) exp

t
an
_

,
0 2 4 6 8 10
0
0.5
1
1.5
Time, s
100 1/s
10 1/s
1.0 1/s
0.1 1/s
2
1
+
2
1
+
8
Mayonnaise

312 Chapter 5. Viscoelasticity


The Bir d-Leider equat ion was used by Kokini and Dickie (1981) and
Dickie and Kokini (1982) t o evaluat e st r ess over shoot dat a for var ious
foods including ket chup, must ar d, mayonnaise, apple but t er , but t er ,
mar gar ine, and canned fr ost ing. The aut hor s concluded t hat t he r ela-
t ionship pr ovided a moder at ely good pr edict ion of peak shear st r ess and
peak t imes but gave only a cr ude pr edict ion of st r ess decay. Campanella
and Peleg (1987b) wer e able t o get a bet t er fit of st r ess gr owt h dat a for
mayonnaise using a mor e complex model pr oposed by Lar son (1985).
The Bir d-Leider equat ion, in st r ict ly empir ical for m ( not
r equir ed) has also been used t o int er pr et st r ess over shoot behavior in
cr eam (Pr ent ice, 1992). St r ess over shoot dat a wer e found useful in
modeling t he human per cept ion of fluid t hickness in t he mout h (Dickie
and Kokini, 1983).
5.3. Osci llatory Testi ng
In oscillat or y inst r ument s, samples ar e subject ed t o har monically
var ying st r ess or st r ain. This t est ing pr ocedur e is t he most common
dynamic met hod for st udying t he viscoelast ic behavior of food. Result s
ar e ver y sensit ive t o chemical composit ion and physical st r uct ur e so
t hey ar e useful in a var iet y of applicat ions including gel st r engt h
evaluat ion, monit or ing st ar ch gelat inizat ion, st udying t he glass
t r ansit ion phenomenon, obser ving pr ot ein coagulat ion or denat ur at ion,
evaluat ing cur d for mat ion in dair y pr oduct s, cheese melt ing, t ext ur e
development in baker y and meat pr oduct s, shelf-life t est ing, and cor -
r elat ion of r heological pr oper t ies t o human sensor y per cept ion. Food
scient ist s have found oscillat or y t est ing inst r ument s t o be par t icular ly
valuable t ools for pr oduct development wor k.
Oscillat or y t est ing may be conduct ed in t ension, bulk compr ession,
or shear . Typical commer cial inst r ument s oper at e in t he shear defor -
mat ion mode and t his is t he pr edominant t est ing met hod used for food.
Shear st r ain may be gener at ed using par allel plat e, cone and plat e, or
concent r ic cylinder fixt ur es. Dynamic t est ing inst r ument s may be
divided int o t wo gener al cat egor ies: cont r olled r at e inst r ument s wher e
t he defor mat ion (st r ain) is fixed and st r ess is measur ed, and cont r olled
st r ess inst r ument s wher e t he st r ess amplit ude is fixed and t he defor -
mat ion is measur ed. Bot h pr oduce similar r esult s. The emphasis in
t his sect ion is on fluid and semi-solid foods. Dynamic t est ing of solid
foods has been r eviewed by Rao and Skinner (1986).
K, n, K, n
5.3 Oscillatory Testing 313
Appli cati on of Stress and Strai n. A number of assumpt ions ar e
made in developing t he mat hemat ical equat ions t o descr ibe oscillat or y
t est ing: st r ain is t he same at all point s in t he sample, sample iner t ia
may be neglect ed, and t he mat er ial behaves as a linear viscoelast ic
subst ance. When t hese assumpt ions ar e violat ed, mor e complex ana-
lyt ical consider at ions ent er t he pr oblem (Whor low, 1992).
In oscillat or y t est s, mat er ials ar e subject ed t o defor mat ion (in con-
t r olled r at e inst r ument s) or st r ess (in cont r olled st r ess inst r ument s)
which var ies har monically wit h t ime. Sinusoidial, simple shear is
t ypical. To illust r at e t he concept , consider t wo r ect angular plat es or i-
ent ed par allel t o each ot her (Fig. 5.10). The lower plat e is fixed and t he
upper plat e is allowed t o move back and for t h in a hor izont al dir ect ion.
Assume t he sample being t est ed is locat ed bet ween t he plat es of a
cont r olled r at e device. Suppose t he st r ain in t he mat er ial bet ween t he
plat es is a funct ion of t ime defined as
[5.30]
wher e is t he amplit ude of t he st r ain equal t o when t he mot ion of
t he upper plat e is . is t he fr equency expr essed in r ad/s which
is equivalent t o her t z. The per iod of t ime r equir ed t o complet e
one cycle is equal t o . If t he t wo plat es (Fig. 5.10) wer e separ at ed
by a dist ance of 1.5 mm and t he upper moved 0.3 mm fr om t he cent er
line, t hen t he maximum st r ain amplit ude may be calculat ed as 0.2 or
20%: . A 10% st r ain could be achieved by main-
t aining mm and moving t he plat e 0.15 mm.
Figure 5.10. Oscillatory strain between rectangular plates.

o
sin(t )

o
L/h
L sin(t )
/(2)
2/

o
L/h 0.3/1.5 0.2
h 1.5
Stationary Plate
L
L sin ( t)
h
Oscillating Plate
SAMPLE
314 Chapter 5. Viscoelasticity
Effect s of changing t he amplit ude and fr equency on t he input st r ain
funct ion ar e illust r at ed in Fig. 5.11. Doubling t he amplit ude fr om 0.1
(Cur ve 2) t o 0.2 (Cur ve 1 or Cur ve 3) doubles t he height of t he cur ve.
Doubling t he fr equency fr om 1 r ad/s (Cur ve 1) t o 2 r ad/s (Cur ve 2 or
Cur ve 3) cut s, by one-half, t he t ime bet ween peaks of t he t wo cur ves.
Using a sine wave for st r ain input r esult s in a per iodic shear r at e
found by t aking t he der ivat ive of Eq. [5.30]:
[5.31]
which can be evaluat ed as
[5.32]
Wit h a small st r ain amplit ude (so t he mat er ial will behave in a linear
viscoelast ic manner ), t he following shear st r ess is pr oduced by t he st r ain
input :
[5.33]
wher e is t he amplit ude of t he shear st r ess (not t o be confused wit h
t he yield st r ess symbolized by in ear lier chapt er s) and is t he phase
lag or phase shift (also called t he mechanical loss angle) r elat ive t o t he
st r ain. The t ime per iod associat ed wit h t he phase lag is equal t o .
can be t hought of as t he peak for ce per unit ar ea r eceived by t he
st at ionar y plat e (Fig. 5.10). Dividing bot h sides of Eq. [5.33] by yields
[5.34]
The complet e r esult s of small amplit ude oscillat or y t est s can be
descr ibed by plot s of t he amplit ude r at io ( ) and t he phase shift ( )
as fr equency dependent funct ions. These par amet er s alone, however ,
ar e not commonly used t o descr ibe r esult s and ot her mat er ial funct ions
(which may all be wr it t en in t er ms of and ) have been defined.
The shear st r ess out put pr oduced by a sinusoidal st r ain input may
be wr it t en as
d
dt



d(
o
sin(t ))
dt


o
cos(t )

o
sin(t + )

o

/

o
_

,
sin(t + )

o
/
o

o
/
o

5.3 Oscillatory Testing 315
Figure 5.11. Strain input functions showing variations in frequency and strain
amplitude: Curve 1) , ; Curve 2) , ; Curve 3)
, .
[5.35]
(called t he shear st or age modulus) and (called t he shear loss
modulus) ar e bot h funct ions of fr equency and can be expr essed in t er ms
of t he amplit ude r at io and t he phase shift :
[5.36]
and
[5.37]
may be int er pr et ed as t he component of t he st r ess in phase wit h
t he st r ain: may be int er pr et ed as t he component of t he st r ess 90
out of phase wit h t he st r ain. Addit ional fr equency dependent mat er ial
funct ions (Table 5.2) include t he complex modulus ( ), complex vis-
Time, s
S
t
r
a
i
n

I
n
p
u
t
0.3
0.2
0.1
0
-0.1
-0.2
-0.3
Curve 1
Curve 2
Curve 3
0 2.0 4.0 6.0 8.0 10.0 12.0

o
0.2 1 rad/s
o
0.1 2 rad/s

o
0.2 2 rad/s
G + (G/)

G G
G

o
_

,
cos()
G

o
_

,
sin()
G
o
G
o

G
*
316 Chapter 5. Viscoelasticity
cosit y ( ), dynamic viscosit y ( ), out of phase component of t he complex
viscosit y ( ), complex compliance ( ), st or age compliance ( ), and t he
loss compliance ( ):
[5.38]
[5.39]
[5.40]
[5.41]
[5.42]
[5.43]
[5.44]
Alt hough , it is impor t ant t o not e t hat and .
Oscillat or y dat a for var ious food pr oduct s may be found in Appendices
[6.20], [6.21], and [6.22].
Using Eq. [5.40], Eq. [5.35] can be expr essed as
[5.45]
which is an excellent equat ion t o r epr esent mat er ial behavior because
it clear ly indicat es t he elast ic ( ) and viscous ( ) nat ur e of a
subst ance. This idea is expanded in Sec. 5.5 t o explain t he behavior of
"silly put t y" using t he Debor ah number concept .
Anot her popular mat er ial funct ion used t o descr ibe viscoelast ic
behavior is t he t angent of t he phase shift or phase angle (called t an
delt a) which is also a funct ion of fr equency:
[5.46]
This par amet er is dir ect ly r elat ed t o t he ener gy lost per cycle divided
by t he ener gy st or ed per cycle. Since , can var y fr om zer o
t o infinit y. Obser vat ions of polymer syst ems give t he following
numer ical r anges for : ver y high for dilut e solut ions, 0.2 t o 0.3 for

J
*
J
J
G
*

o
(G)
2
+ (G)
2

G
*

()
2
+ ()
2

G

J
*

1
G
*
J
G
(G)
2
+ (G)
2
J
G
(G)
2
+ (G)
2
J
*
1/G
*
J 1/G J 1/G
G +

tan()
G
G
0 /2 tan
tan
5.3 Oscillatory Testing 317
amor phous polymer s, low (near 0.01) for glassy cr yst alline polymer s and
gels. Values of for t ypical food syst ems (dilut e solut ion, concen-
t r at ed solut ion, and gel) ar e consider ed in Sec. 5.4.
To bet t er under st and t he viscoelast ic par amet er s defined above, it
is helpful t o look at behavior which is solely Hookean or solely Newt o-
nian. If a mat er ial is a Hookean solid, t he st r ess and st r ain ar e in phase
and . Hence, and ar e also equal t o 0 because t her e is no viscous
dissipat ion of ener gy. In t his case, is a const ant equal t o t he shear
modulus ( ). If a mat er ial behaves as a Newt onian fluid, t he st r ess (Eq.
[5.33]) and st r ain (Eq. [5.30]) ar e 90 degr ees out of phase ( ); hence,
t he shear r at e (Eq. [5.32]) is also 90 degr ees out of phase wit h t he shear
st r ess. In t his case, and ar e zer o because t he mat er ial does not
st or e ener gy. Then, is const ant and equal t o t he Newt onian viscosit y
( ). Similar behavior is oft en obser ved for non-Newt onian fluids as t he
fr equency appr oaches zer o.
Fur t her assessment of t he phase lag concept can help clar ify t he
meaning of given as par t of t he amplit ude r at io in Eq. [5.34]. As t he
phase lag appr oaches zer o, for ce is t r ansmit t ed t hr ough t he sample (Fig.
5.10) quickly and t he changes in st r ess ar e obser ved at near ly t he same
t ime as t he applied defor mat ion pr oduces st r ain. In solids, r apid for ce
t r ansmission is due t o t he cr yst alline nat ur e of t he mat er ial. The
amount of for ce t r ansmit t ed for a given st r ain depends on t he mat er ial
modulus. In a Hookean solid t he maximum for ce per unit ar ea t r ans-
mit t ed t hr ough t he sample ( ) is equal t o t he shear modulus t imes t he
maximum st r ain ( ). Viscous heat ing absor bs some ener gy in
viscoelast ic mat er ials r esult ing in smaller values of .
The phase lag appr oaches t he maximum value of wit h fluids
exhibit ing a high degr ee of Newt onian behavior . The maximum for ce
per unit ar ea ( ) t r ansmit t ed t hr ough a Newt onian fluid depends on
t he maximum shear r at e induced dur ing defor mat ion. Consider ing Eq.
[5.32], t he maximum shear r at e is fr equency dependent and may be
calculat ed as . Consequent ly, is equal t o t he maximum shear r at e
t imes t he viscosit y ( ) for a Newt onian fluid. is lar ger in vis-
coelast ic mat er ials t hat have less t endency t o flow: In t hese subst ances,
gr eat er for ce is t r ansmit t ed t hr ough t he sample because t he viscous
dissipat ion of ener gy is smaller .
tan
0 G
G
G
/2
G

o
G
o

o
/2

o

o

o

o
318 Chapter 5. Viscoelasticity
The Maxwell fluid model (Fig. 5.1) is oft en used t o int er pr et dat a
fr om dynamic t est ing of polymer ic liquids. If t he st r ain input is har -
monic, , t hen . This r elat ionship (Eq. [5.32])
can be subst it ut ed int o Eq. [5.6] and t he r esult ing differ ent ial equat ion
solved t opr oduce a number of fr equency dependent r heological funct ions
for Maxwell fluids:
[5.47]
[5.48]
[5.49]
[5.50]
wher e , t he r elaxat ion t ime of a Maxwell fluid, is equal t o . Looking
at exper iment al dat a may allow mat er ial const ant s of t he Maxwell
model t o be evaluat ed fr om t he asympt ot es: as goes t o zer o, goes t o
; and as goes t o infinit y, goes t o . These ideas ar e used in Sec.
5.5 t o examine t he behavior of "silly put t y."
Compari son of Moduli and Compli ances. The mat hemat ical
r elat ionships pr esent ed in t his sect ion ar e for shear defor mat ion.
Analogous moduli and compliances can be defined for t he t ension (or
compr ession) and bulk compr ession modes of defor mat ion (Table 5.6).
These funct ions ar e gener ally mor e difficult t o measur e t han shear
funct ions and few dat a ar e available in t he published lit er at ur e.
Analogues for t he r elaxat ion modulus and cr eep compliance ar e also
pr esent ed in Table 5.6.
Typi cal Operati ng Modes of an Osci llatory Testi ng Instrument.
Commer cially available oscillat or y inst r ument s will oper at e in
numer ous modes. A st r ain or st r ess sweep, conduct ed by var ying t he
amplit ude of t he input signal at a const ant fr equency (Fig. 5.12), is used
t o det er mine t he limit s of linear viscoelast ic behavior by ident ifying a
cr it ical value of t he sweep par amet er . In t he linear r egion (Fig. 5.13),
r heological pr oper t ies ar e not st r ain or st r ess dependent . St or age and

o
sin(t )


o
cos(t )
G
G
2

rel
2
1 +
2

rel
2
G
G
rel
1 +
2

rel
2


1 +
2

rel
2
tan
G
G

1

rel

rel
/G

G G
5.3 Oscillatory Testing 319
loss moduli ver sus t he sweep par amet er ar e plot t ed in Fig. 5.13. Some
exper iment er s pr efer t o plot combined mat er ial funct ions such as t he
complex modulus or t he complex viscosit y.
Table 5.6. Comparison of Moduli and Compliances Determined in Oscillatory
Testing Using Three Modes of Deformation: Shear, Bulk Compression, and Ten-
sion (Ferry, 1980).
Sh ea r Bu lk Compr es s ion Ten s ion
Complex Modu lu s
St or a ge Modu lu s
Los s Modu lu s
Complex Complia n ce
St or a ge Complia n ce
Los s Complia n ce
Rela xa t ion Modu lu s
Cr eep Complia n ce
Figure 5.12. Strain or stress sweep mode in oscillatory testing.
G
*
() K
*
() E
*
()
G() K() E()
G() K() E()
J
*
() B
*
() D
*
()
J() B() D()
J() B() D()
G( t ) K( t ) E( t )
J( t ) B( t ) D( t )
Time
A
m
p
l
i
t
u
d
e
320 Chapter 5. Viscoelasticity
Figure 5.13. Typical response to a strain or stress sweep showing the linear vis-
coelastic region defined by the critical value of the sweep parameter.
St r ain and st r ess sweeps ar e conduct ed on cont r olled r at e and
cont r olled st r ess inst r ument s, r espect ively. It has been suggest ed t hat
st r ess sweeps pr oduce super ior r esult s (Holland, 1994); however , bot h
st r ain and st r ess sweeps ar e known t o pr ovide an excellent basis for
compar ing t he viscoelast ic nat ur e food pr oduct s. In addit ion t o est a-
blishing t he linear viscoelast ic r ange of t he sweep par amet er , st r ain and
st r ess sweeps have been used t o differ ent iat e weak and st r ong gels:
St r ong gels may r emain in t he linear viscoelast ic r egion over gr eat er
st r ains t han weak gels.
The fr equency sweep is pr obably t he most common mode of oscil-
lat or y t est ing because it shows how t he viscous and elast ic behavior of
t he mat er ial changes wit h t he r at e of applicat ion of st r ain or st r ess. In
t his t est t he fr equency is incr eased while t he amplit ude of t he input
signal (st r ess or st r ain) is held const ant (Fig. 5.14). Fr equency sweeps
ar e ver y useful in compar ing, somet imes called "finger pr int ing," dif-
fer ent food pr oduct s or in compar ing t he effect s of var ious ingr edient s
and pr ocessing t r eat ment s on viscoelast icit y. Mat er ials usually exhibit
mor e solid like char act er ist ics at higher fr equencies.
Strain or Stress
M
o
d
u
l
u
s
Storage Modulus
Critical Value
Linear Viscoelastic Region
Loss Modulus
5.3 Oscillatory Testing 321
Figure 5.14. Frequency sweep mode in oscillatory testing.
An isot her mal t ime sweep, wher e fr equency and amplit ude ar e
const ant over t ime, can indicat e t ime-dependent st r uct ur al changes
such as t hose associat ed wit h fir ming of cheese cur d or yogur t . A t ime
sweep (Fig. 5.15) may be conduct ed in conjunct ion wit h a cont r olled
change in t emper at ur e (Fig. 5.16). This t ype of t est ing is ver y useful in
st udying pr oblems t hat involve t emper at ur e induced changes in r heo-
logical behavior . Typical examples associat ed wit h heat ing would
include t he soft ening of chocolat e or cheese due t o t he melt ing of fat ,
gelat ion t o for m pect in gels, t hickening of solut ions fr om st ar ch gela-
t inizat ion, and fir ming of meat or egg pr oduct s caused by pr ot ein
denat ur at ion. The st udy of t ime-dependent flowbehavior r esult ing fr om
chemical r eact ions such as t hese is called chemor heology.
Strai n i n Rotati onal-Type Fi xtures. Par allel plat e, cone and plat e,
and concent r ic cylinder fixt ur es ar e t he pr efer r ed geomet r ies for sub-
ject ing fluid and semi-solid foods t o an oscillat or y st r ain. In a par allel
plat e appar at us, such as t hat used in a t r adit ional par allel plat e
viscomet er (Fig. 5.17), t he shear st r ain is a funct ion of t he r adius. It
var ies fr om zer o at t he cent er of t he sample ( ) t o a maximum at t he
out er edge of t he plat e ( ). Maximum st r ain ( ) is calculat ed as t he
dist ance t r aveled at t he r im of t he upper plat e ( ) divided by t he dist ance
bet ween t he plat es ( ):
Time
A
m
p
l
i
t
u
d
e
r 0
r R
o
r
h
322 Chapter 5. Viscoelasticity
Figure 5.15. Time sweep mode in oscillatory testing.
Figure 5.16. Controlled temperature changes in oscillatory testing.
Time
A
m
p
l
i
t
u
d
e
T
e
m
p
e
r
a
t
u
r
e
Time
Ramp Change
Step Change
5.3 Oscillatory Testing 323
[5.51]
Assume exper iment at ion st ar t s wit h t he syst em at r est , t hen t he sample
is subject ed t o a sine wave st r ain funct ion descr ibed by Eq. [5.30]. The
plat e will move t hr ough a posit ive sweep angle ( ), t hen pass t hr ough
t he st ar t ing posit ion while r ot at ing t o a negat ive value of t he sweep
angle ( ). (in r adians) is t he amount of angular t r avel of t he upper
plat e fr om t he st ar t ing posit ion t o t he point of maximum r ot at ion. At
t his point t he maximum st r ain is induced in t he sample. To achieve,
for example, a 10% maximum st r ain in a par allel plat e appar at us wit h
50 mm diamet er plat es separ at ed by a dist ance of 2.0 mm will r equir e
t he following sweep angle:
[5.52]
A sweep angle r esult ing in a st r ain of 10% or less is usually r equir ed t o
st ay in t he r egion of linear viscoelast ic mat er ial behavior .
Figure 5.17. Sweep angle in a parallel plate apparatus.

o

R

r
h
2R

2
_

1
h
_

,

R
h
+


(
o
)h
R

.1(2)
50/2
0.008 rad or 0.458 degrees
R
h
r
r
+
-
324 Chapter 5. Viscoelasticity
Shear st r ain in a cone and plat e fixt ur e (Fig. 3.3) is unifor m
t hr oughout t he gap and equal t o t he sweep angle divided by t he t angent
of t he cone angle:
[5.53]
Recall t hat for t he pr efer r ed cone angles, r ad (4 degr ees).
Shear st r ain in a concent r ic cylinder fixt ur e (Fig. 3.1) is a funct ion of
t he r adius. In t he case wher e t he bob r ot at es and t he cup is fixed, t he
maximum st r ain occur s at t he bob:
[5.54]
wher e, r ecall, . When t he bob is fixed and t he cup r ot at es, t he
maximum st r ain occur s at t he cup:
[5.55]
Shear st r esses for par allel plat e, cone and plat e, and concent r ic cylinder
fixt ur es ar e calculat ed using t he appr opr iat e expr essions developed in
Chapt er 3: Eq. [3.71] wit h , Eq. [3.55], and Eq. [3.3], r espect ively.
5.4. Typi cal Osci llatory Data
Si ngle Frequency Tests. Typical input and out put for a dilut e solut ion
showing Newt onian t ype behavior is illust r at ed in Fig. 5.18 wit h
accompanying dat a in Table 5.7. In t his case, an input st r ain amplit ude
of 10% ( ) wit h a const ant fr equency of 10 r ad/s gives an input
st r ain funct ion of . The out put st r ess funct ion is shift ed by
1.48 r ad (85 ) and shows an amplit ude r at io of 0.1 ( ):
[5.56]
Typical out put dat a for a gel showing ver y elast ic (solid like) behavior
and a concent r at ed solut ion showing viscoelast ic behavior ar e seen in
Fig. 5.19 wit h companion dat a in Table 5.7. The input st r ain funct ion
is t he same as t hat shown in Fig. 5.18: .

o


tan
tan 0.07

o

bob
2

2
1
_

,
R
c
/R
b

o

cup
2

2
1
_

,
n 1

o
0.1
0.1sin(10t )
G
*

o
/
o
0.1
G
*

o
sin(t + ) .01sin(10t + 1.48)
0.1sin(10t )
5.4 Typical Oscillatory Data 325
Figure 5.18. Typical input and output for a dilute solution showing Newtonian
like behavior at a constant frequency of 10 rad/s.
Table5.7. Typical Output for aDiluteSolution (Newtonian Behavior), Gel (Elastic
Behavior) and aConcentrated Solution (ViscoelasticBehavior) from aStrain Input
Function having a Frequency of 10 rad/s and a Strain Amplitude of 10 Percent:
Dilu t e Solu t ion Gel Con cen t r a t ed
Solu t ion
, r a d (degr ees ) 1.48 (85 ) 0.0698 (4 ) 0.873 (50 )
, Pa 0.100 5200 200
, Pa 0.00907 5187 129
, Pa 0.0996 363 153
, Pa s 0.00996 36.3 0.140
, Pa s 0.000907 518.7 12.9
, Pa s 0.010 520.0 20.0
10.9 0.0699 1.19
0 0.2 0.4 0.6 0.8 1 1.2
Time, s
S
t
r
a
i
n

I
n
p
u
t

(
-
)

o
r

S
t
r
e
s
s

O
u
t
p
u
t

(
P
a
)
0.15
0.05
0
-0.05
-0.10
-0.15
0.10
G
*
= .1 Pa ; phase shift = 85 degrees
Strain Input: 10 % strain ; 10 rad/s
Output:
0.1sin(10t )

o
/
o
G
*
G

G
*
cos
G

G
*
sin

*
G
*
/
tan G

/G

326 Chapter 5. Viscoelasticity


Figure 5.19. Typical output (Input: ) for a gel showing elastic
type behavior and a concentrated solution showing viscoelastic behavior.
Out put st r ess for t he gel (Fig. 5.19) has a small phase shift ,
r ad (4 ), but a lar ge amplit ude r at io ( Pa) pr oducing t he
following shear st r ess funct ion:
[5.57]
The out put st r ess for t he concent r at ed solut ion has a gr eat er phase shift ,
r ad (50 ), and an amplit ude r at io bet ween t he dilut e solut ion
and t he gel ( Pa):
[5.58]
Normali zed Strai n and Stress. Input and out put at a par t icular
fr equency may also be visualized in t er ms of a nor malized st r ain,
[5.59]
and a nor malized st r ess,
[5.60]
0 0.2 0.4 0.6 0.8 1 1.2
Time, s
S
t
r
e
s
s

O
u
t
p
u
t
,

P
a
G
*
= 5200 Pa : phase shift = 4 degrees
G
*
= 200 Pa : phase shift = 50 degrees
600
400
200
0
-200
-400
-600
Gel
Concentrated Solution
0.1sin(10t )
0.0698
G
*

o
/
o
5213
G
*

o
sin(t + ) 521.3sin(10t + 0.0698)
0.873
G
*
200
G
*

o
sin(t + ) 20sin(10t + 0.873)

o
5.4 Typical Oscillatory Data 327
Values of and r ange fr om -1.0 t o 1.0.
The nor malized st r ain input funct ion is t he same for t he t hr ee
mat er ials cur r ent ly under consider at ion:
[5.61]
Nor malized st r ess out put funct ions ar e
[5.62]
for t he dilut e solut ion,
[5.63]
for t he concent r at ed solut ion, and
[5.64]
for t he gel. Eq. [5.61] t hr ough [5.64] ar e plot t ed in Fig. 5.20. Phase
shift s for t he dilut e solut ion (1.48 r ad) and t he concent r at ed solut ion
(0.873 r ad) ar e indicat ed on t he figur e. The phase shift for t he gel (0.0698
r ad) is small but st ill visible. Advanced r heomet er s may pr ovide a
cont inuously updat ed display of t hese (or equivalent ) wavefor ms dur ing
oscillat or y t est ing. This infor mat ion can be ver y useful in det er mining
whet her or not t he mat er ial being t est ed is displaying viscoelast ic
behavior . It can also be used t o det ect dist or t ed wavefor ms t hat may
pr oduce misleading r esult s.
Fur t her consider at ion of Fig. 5.20 can pr ovide added physical
meaning t o t he phase lag phenomenon. Fr equency of t he st r ain input
is 10 r ad/s so t he fr equency of t he out put signals for each mat er ial (dilut e
solut ion, concent r at ed solut ion, and gel) is also 10 r ad/s. The per iod of
t ime r equir ed t o complet e a sine wave st r ain cycle is . In t his
exper iment , t he per iod is equal t o or 0.628 s. The maximum phase
lag ( ) which can occur is meaning t he st r ess signal, cor r esponding
t o a par t icular st r ain input , could be obser ved at a maximum of
s aft er t he st r ain is applied. St r ess is obser ved
1.48/10 = 0.148 s aft er t he st r ain in t he dilut e solut ion because is 1.48
r ad. This is close t o maximum phase lag expect ed for a Newt onian fluid
at t he 10 r ad/s t est fr equency. A value of 0.0698 r ad (t he minimum
possible value of is zer o) for t he gel means t he st r ess signal is 0.0698/10
= 0.00698 s behind t he st r ain input . This Hookean-like behavior is not

o
0.1sin(10 t )/0.1

o
0.01sin(10 t + 1.148)/0.01

o
20sin(10 t + 0.873)/20

o
521.3sin(10 t + 0.0698)/521.3
2/
2/10
/2
/(2) /(2) (10) 0.157

328 Chapter 5. Viscoelasticity


Figure 5.20. Normalized curves for input strain ( ) and output stresses
( ) showing the phase shift for a dilute solution (1.48 rad or 85 ) and a con-
centrated solution (0.873 rad or 50 ).
unusual due t o t he high level of molecular st r uct ur e pr esent in a gel.
The lag per iod of 0.0873 s found for t he concent r at ed solut ion r epr esent s
a signal t r ansmission t ime bet ween t he dilut e solut ion and t he gel.
Clear ly, t he concent r at ed solut ion has mor e st r uct ur e t han t he dilut e
solut ion but significant ly less t han t he gel.
Frequency Sweeps. Fig. 5.18, 5.19, 5.20 illust r at e r heological
behavior at a single fr equency of 10 r ad/s. To effect ively char act er ize
t he r heological behavior of t hese subst ances, mat er ial funct ions should
be det er mined over a wide r ange of fr equencies using t he fr equency
sweep t est ing concept illust r at ed in Fig. 5.14. Dat a may be plot t ed wit h
t he fr equency given as r adians per second or her t z by r ecognizing t hat
1 hz = 1 cycle/s = 2 r ad/s. The choice is essent ially a mat t er of individual
pr efer ence.
0 0.2 0.4 0.6 0.8 1 1.2
Time, s
N
o
r
m
a
l
i
z
e
d

S
t
r
a
i
n

o
r

S
t
r
e
s
s

(
-
)
1.5
1.0
0.5
0
-0.5
-1.0
-1.5
input
strain
dilute
solution
concentrated
solution
gel
1.48 rad 0.873 rad
/
o
/
o

5.4 Typical Oscillatory Data 329


Adilut e solut ion, a concent r at ed solut ion, and a gel show dist inct ive
behavior (Fig. 5.21, 5.22 and 5.23) when subject ed t o a fr equency sweep.
Wit h a dilut e solut ion (Fig. 5.21), is lar ger t han over t he ent ir e
fr equency r ange but appr oach each ot her at higher fr equencies. and
cur ves int er sect at t he middle of t he fr equency r ange for t he con-
cent r at ed solut ion (Fig. 5.22) showing a clear t endency for mor e solid-
like behavior at higher fr equencies. The cr ossover fr equency is
somet imes a useful cr it er ion for pr oduct evaluat ion. It occur s when
, t he point wher e t he phase lag ( ) equals . is significant ly
higher t han t hr oughout t he fr equency r ange for t he gel (Fig. 5.23).
It is meaningful t oobser ve t hat moduli ar e a st r ong funct ion of fr equency
in t he dilut e and concent r at ed solut ions, but pr act ically const ant wit h
t he gel. Alt er nat e ways of plot t ing oscillat or y dat a ar e discussed in
Example Pr oblem 5.8.4.
Figure 5.21. Mechanical spectra for a dilute solution made from 5% dextrin
(Data from Ross-Murphy, 1988).
G G
G
G
G G /4 G
G
0.1 0.2 0.5 1 2 5 10 20 50 100
0.001
0.002
0.005
0.01
0.02
0.05
0.1
0.2
0.5
1
G

(
P
a
)

o
r

G

(
P
a
)
G
Dilute Solution
G
(rad/s)
330 Chapter 5. Viscoelasticity
Figure 5.22. Mechanical spectra for a concentrated solution made from 5%
lambda carrageenan (Data from Ross-Murphy, 1988).
Figure 5.23. Mechanical spectra for a gel made from 1% agar (Data from Ross-
Murphy, 1988).
0.01 0.03 0.1 0.3 1 3 10 30 100
0.1
0.3
1
3
10
30
100
300
1,000
G

(
P
a
)

o
r

G

(
P
a
)
(rad/s)
Concentrated Solution
G
G
0.01 0.03 0.1 0.3 1 3 10 30 100
10
30
100
300
1,000
3,000
10,000
30,000
100,000
G

(
P
a
)

o
r

G

(
P
a
)
(rad/s)
Gel
G
G
5.4 Typical Oscillatory Data 331
Analysi s of the Phase Lag ( ). The t endency of t he dilut e solut ion
and t he concent r at ed solut ion t o exhibit mor e fluid or solid-like behavior
wit h incr easing fr equency can be examined in mor e det ail by consider ing
t he fr equency dependence of t he phase lag. Fir st , t he dat a shown in
Fig. 5.21, 5.22, and 5.23 ar e summar ized in t he for m of power law
equat ions (Table 5.8). Using t his infor mat ion t he phase lag may be
calculat ed fr om t he loss and st or age moduli: . Result s
ar e illust r at ed in Fig. 5.24. The maximum phase lag which can be
obser ved is found wit h a Newt onian fluid: The minimum value is
zer o found wit h a Hookean solid. High values of at low fr equencies
indicat e a t endency t owar d mor e fluid-like behavior for bot h t he dilut e
and concent r at ed solut ions at low defor mat ion r at es. Mor e solid-like
behavior is obser ved for t hese solut ions at t he high defor mat ion r at es
associat ed wit h high fr equencies. The phase lag for t he gel (Fig. 5.24)
is pr act ically const ant indicat ing consist ent solid-like behavior over t he
ent ir e fr equency r ange.
Table 5.8. Values of the storage modulus ( ) and the loss
modulus ( ) for typical materials: dilute solution (Fig. 5.21),
concentrated solution (Fig. 5.22), and gel (Fig. 5.23).
Ma t er ia l
(Pa s
b
) ( - ) (Pa s
d
) ( - ) (r a d/ s )
Dilu t e Solu t ion .00028 1.66 .01186 .934 .1 - 100
Con cen t r a t ed Solu t ion 16.26 .840 27.78 .520 .01 - 100
Gel 5626 .0371 344.7 .0145 .01 - 100
Ti me to Complete an Osci llatory Test. Oscillat or y dat a for a par -
t icular fr equency must be collect ed over one complet e sine wave cycle,
and t wo or t hr ee cycles may be needed t o obt ain equilibr ium values. It
is impor t ant t o consider t he t ime r equir ed t o complet e low fr equency
t est s when select ing a fr equency r ange for t he exper iment at ion. If, for
example, a t est was conduct ed at a low fr equency of = 0.01 r ad s
-1
,
t hen t he t ime r equir ed t o complet e one cycle can be calculat ed
( ) as 10.5 minut es. Ther efor e, it would t ake at least 31.5
minut es t o obt ain a single pair of and values if t hr ee cycles wer e
needed! It is usually unnecessar y t o collect dynamic dat a at fr equencies

arctan(G/G)
/2

G a
b
G c
d
a b c d

2/(.01) (60)
G G
332 Chapter 5. Viscoelasticity
Figure 5.24. Variation of the phase lag ( ) with frequency ( ) for typical mate-
rials. The upper limit is represented by a Newtonian fluid ( ) and the
lower limit by a Hooke solid ( ).
less t han 0.01 r ad s
-1
and a lower limit of = 0.1 r ad s
-1
is oft en adequat e.
In cont r ast , t he t ime r equir ed t o complet e one cycle at a fr equency of
100 r ad s
-1
, a suit able upper limit for many exper iment s, is 0.063 s
meaning t hat a t hr ee cycle t est could be complet ed in less t han 0.19 s.
5.5. Deborah Number
Mar cus Reiner (Reiner , 1964) pr oposed t he Debor ah number as a
means of dist inguishing bet ween solids and liquids. The number , named
aft er t he pr ophet ess Debor ah, is based on his int er pr et at ion of a biblical
passage found in J udges 5:5. He st at ed t hat Debor ahs song, given aft er
t he vict or y over t he Philist ines, could be t r anslat ed as "The mount ains
flowed befor e t he Lor d." His point was t hat mount ains could flow befor e
God, not befor e man, because t he t ime of obser vat ion for God is infinit e.
(A similar ar gument could be made fr om Psalm 97:5: "The mount ains
melt like wax befor e t he Lor d.") Thus, Pr of. Reiner r ecognized t he need
t o consider t he t ime of t he pr ocess as a fact or when char act er izing a
0.1 0.2 0.5 1 2 5 10 20 50 100
0
0.5
1
1.5
Newtonian Fluid
Dilute Solution
Concentrated Solution
Hooke Solid
Gel
(
r
a
d
)
(rad/s)

/2
0

5.5 Deborah Number 333


subst ance as a solid or a liquid. This idea has not been ext ensively
applied t o food pr oduct s but st ill pr ovides a ver y useful means for
under st anding t he behavior of viscoelast ic mat er ials.
The Debor ah number is defined as
[5.65]
wher e is t he char act er ist ic t ime of t he mat er ial and is t he
char act er ist ic t ime of t he pr ocess. for a Maxwell fluid and
for a Kelvin solid (Fig. 5.1). Giving a pr ecise definit ion of
is mor e difficult . It is r elat ed t o t he t ime scale of t he defor mat ion and
is r oughly equal t o t he lengt h scale in t he flow dir ect ion divided by t he
mean velocit y. It might be t he r ecipr ocal of fr equency for an oscillat ing
sur face, or t he t ime for a par t icle t o pass t hr ough a conver ging die and
exper ience t he subsequent incr ease in velocit y. Any pr ocess must be
invest igat ed individually but Debor ah number s t end t o be high for
pr ocesses such as fiber spinning and plast ic moulding but low for
ext r usion.
All t he viscomet r ic flow sit uat ions descr ibed in Chapt er s 2, 3 and 4
involve low values of t he Debor ah Number . Cr eep and st r ess r elaxat ion
discussed in t his chapt er also involve low Debor ah Number values. The
ot her main t opic of t his chapt er , oscillat or y t est ing, involves bot h low
and moder at e values of t he Debor ah Number . Changes in t he phase
lag wit h incr easing fr equency for t he dilut e and concent r at ed solut ions
illust r at ed in Fig. 5.24 demonst r at e t he concept . Alt hough t he char -
act er ist ic t ime of t he mat er ial is unaffect ed by oscillat or y t est ing, t he
char act er ist ic t ime of t he pr ocess is inver sely pr opor t ional t o t he
fr equency; t her efor e, as t he fr equency incr eases, t he Debor ah Number
incr eases causing t he solut ions t o exhibit t he mor e solid-like char ac-
t er ist ics associat ed wit h lower values of t he phase lag (Fig. 5.24).
It is impor t ant t o r emember t hat all mat er ials ar e viscoelast ic, but
t he viscous or t he elast ic char act er may dominat e in cer t ain sit uat ions.
Pipkin (1986), suggest s t hat t he char act er ist ic t ime of a mat er ial may
be pr ovisionally consider ed as an or der -of-magnit ude est imat e for how
long it t akes t he subst ance t o complet e a st r ess r elaxat ion pr ocess. If
a mat er ial is an ideal elast ic solid, , and no r elaxat ion occur s.
Wit h an ideal viscous mat er ial, , meaning r elaxat ion is imme-
N
De

t
material
t
process
t
material
t
process

rel
t
material

ret
t
material
t
process
t
material

t
material
0
334 Chapter 5. Viscoelasticity
diat e. The char act er ist ic t ime of wat er is on t he or der of 10
-12
s and t he
char act er ist ic t ime for window glass (a super cooled liquid at r oom
t emper at ur e) is gr eat er t han 10
5
s. Hence, when wat er is defor med,
elast ic effect s ar e difficult t o obser ve because t he mat er ial r elaxes so
quickly. Conver sely, t he viscous behavior of glass is not easily obser ved
because t he r esponse t o t he defor ming for ce (gr avit y) on a ver t ical pane
is ver y slow; consequent ly, decades ar e needed for obser vable changes
t o occur . The char act er ist ic t ime of t he mat er ial and t he pr ocess must
be examined t oget her t o det er mine if viscous or elast ic t ype behavior
will dominat e a par t icular flow pr oblem. Mat er ials exhibit pr onounced
viscoelast ic behavior when t hese t imes ar e similar in magnit ude.
The pr evious discussion pr ovides insight concer ning how t he
numer ical value of can be used as a measur e of t he degr ee of vis-
coelast icit y. If , t he st r ess is pr opor t ional t o t he viscosit y t imes
t he shear r at e and t he mat er ial behaves as a viscous liquid. On t he
ot her hand, if , t hen st r ess is pr opor t ional t o t he modulus of
r igidit y t imes t he st r ain and t he mat er ial behaves like an elast ic solid.
If t he is on t he or der of one, mat er ials will show viscoelast ic behavior .
When is ver y high, t he model for a Hookean solid may best descr ibe
mat er ial behavior . The Newt onian model may be t he best choice when
ver y small ar e pr esent .
To fur t her expand t he concept of t he Debor ah number , consider dat a
fr om a common silicone polymer known as "bouncing put t y" or "silly
put t y". This subst ance, sold as a t oy for childr en, is int er est ing because
it has a r elaxat ion t ime well wit hin t he limit s of human per cept ion. If
t he mat er ial is assumed t o behave as a Maxwell fluid, const ant s may
be det er mined fr om t he asympt ot es found in Fig. 5.25. Taking t he limit
(as goes t o infinit y) of Eq. [5.47] shows t hat Pa and
t aking t he limit (as goes t o zer o) of Eq. [5.49] indicat es t hat
Pa s. Using t hese values, t he r elaxat ion t ime can be
calculat ed: s.
N
De
N
De
1
N
De
1
N
De
N
De
N
De
G G 260, 000

80, 000

rel
/G 0.31
5.5 Deborah Number 335
Figure 5.25. Oscillatory shear measurements on a silicone polymer (Data from
Denn, 1980).
Cont inued analysis of silly put t y shows t hat Pa and
Pa s at r ad/s. Expr essing t his in t er ms of t he shear st r ess
given by Eq. [5.45] yields
[5.66]
Clear ly, t he elast ic por t ion of t he equat ion ( ) dominat es t he
mat hemat ical r elat ionship. Assuming , t he Debor ah number for
t his case can be is appr oximat ed as
[5.67]
which is a numer ical value of sufficient magnit ude t o indicat e t hat
significant elast ic effect s would be pr esent . The char act er ist ic t ime of
t he pr ocess would be much lower , causing t he Debor ah number t o be
much higher , in sit uat ions wher e silly put t y bounces when dr opped and
shows br it t le fr act ur e when pulled quickly.
0.1 0.3 1 3 10 30 100 300 1,000
100
300
1,000
3,000
10,000
30,000
100,000
300,000
1,000,000
dynamic viscosity storage modulus
(
P
a

s
)

o
r

G

(
P
a
)
(rad/s)
Silicone Polymer
G 260, 000
100 100
G +

260, 000 + 100


260, 000

o
0.1
N
De


rel
t
process


rel
(
o
)
1
0.31(.1) (100) 3.1
336 Chapter 5. Viscoelasticity
Viewing t he pr oblem at a low fr equency ( r ad/s), Pa
and Pa s, gives a shear st r ess equat ion (Eq. [5.45]) of
[5.68]
Test ing under t hese condit ions r esult in a low value of t he Debor ah
number :
[5.69]
At low fr equency (long char act er ist ic t ime of t he pr ocess), t he Debor ah
number is ver y small and t he viscous component ( ) dominat es
mat er ial behavior explaining why t he subst ance flows, as a liquid, when
pulled slowly. Given t he above discussion, one can conclude t hat
adver t ising silly put t y as t he "r eal solid-liquid" is t echnically accept able.
5.6. Experi mental Di ffi culti es i n Osci llatory Testi ng of Food
Ther e ar e var ious sour ces of er r or which may be par t icular ly pr ob-
lemat ic when t est ing food pr oduct s using oscillat or y met hods. Many
foods may not exhibit a well defined st r ain or st r ess r ange wher e t he
pr inciple of linear viscoelast icit y applies. This may be due t o non-
homogeneous samples, t he pr esence of wall slip, or t ime-dependent
mat er ial behavior . Pr eshear ing wit h st eady r ot at ion, befor e conduct ing
oscillat or y t est s, can be useful in cont r olling (or eliminat ing) t hixot r opy
in some mat er ials. Ser r at ed or r oughened sur faces may be effect ive in
dealing wit h wall slip. Special sur faces have been used, for example,
in t est ing cr acker and cookie doughs (Menjivar , 1994), and cheddar
cheese (Rosenber g et al., 1995).
The pr esence of a yield st r ess may also cause ser ious difficult ies in
oscillat or y t est ing of food. This pr oblem was obser ved many year s ago
by Elliot and Ganz (1971, 1977) in st udies involving mayonnaise, salad
dr essing, and mar gar ine. If t he applied st r ain causes t he r esult ing
st r ess t o exceed t he yield st r ess, t hen t he out put sin wave will be
t r uncat ed due t o par t ial flow of t he mat er ial. This deviant behavior is
illust r at ed in Fig. 5.26 wher e t he peak values of t he nor malized st r ess
ar e equal t o 0.7: . Since t he yield st r ess is equal t o 70% of
t he maximum shear st r ess ( ), cannot exceed 0.7.
0.1 G 200
80, 000
G +

200 + 80, 000


N
De


rel
t
process


rel
(
o
)
1
0.31(.1) (.1) 0.003
80, 000

*
/
o
0.7

o
5.6 Experimental Difficulties in Oscillatory Testing of Food 337
The t r ansit ion fr om solid-like t o fluid-like behavior may be t hought
of in t er ms of a cr it ical st r ain or st r ess, below which t he mat er ial behaves
as an elast ic solid. Applied st r esses which exceed t he yield st r ess cause
a deviat ion fr om solid-like behavior r esult ing in a non-sinusoidal out put
wave; consequent ly, measur ed values of t he st or age and loss moduli do
not fall wit hin t he scope of linear viscoelast icit y. These moduli, alt hough
quit e useful in compar ing differ ent food pr oduct s, will be a funct ion of
bot h t he st r ain (or st r ess) amplit ude and fr equency (Yoshimur a and
Pur dhomme, 1987). The pr oblem can be fur t her complicat ed by t he fact
t hat t he yield st r ess it self may be a t ime-dependent par amet er for many
food pr oduct s (St effe, 1992).
Figure 5.26. Stress output when the applied strain takes the material outside the
regime where the material exhibits solid-like behavior.
The above explanat ion of t he yield st r ess issue in oscillat or y t est ing
illust r at es t he dilemma, but may over simplify t he pr oblem. An alt er -
nat ive appr oach is t o assume t hat t he yield st r ess does not exist and
ever yt hing flows. This means t hat even mat er ials appear ing t o have a
ver y solid-like nat ur e would, given sufficient t ime, flow. Bar nes et al.
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Time
N
o
r
m
a
l
i
z
e
d

S
t
r
a
i
n

o
r

S
t
r
e
s
s
1.5
1.0
0.5
0
-0.5
-1.0
-1.5
Strain
Input
Stress
Output
yield stress exceeded
338 Chapter 5. Viscoelasticity
(1987) ment ion gr owing evidence showing t hat concent r at ed syst ems
(e.g., gels, mar gar ine, ice cr eam, and st iff past es) flow at ver y low
st r esses. They explain t hat t he hypot hesis of a yield st r ess, and t he
subsequent obser vat ion t hat a mat er ial does not flow, could be int er -
pr et ed t o mean t hat t he mat er ial has a ver y high zer o shear viscosit y.
Nor mal window glass, for example, has a zer o shear viscosit y gr eat er
t han 10
18
Pa s. This mat er ial appear s t o be solid but given sufficient
t ime (decades) shows obser vable changes due t o flow. Lat t er wor k
(Bar nes, 1992) has acknowledged scient ific evidence suggest ing t he
pr esence of a yield st r ess in non-int er act ing and flocculat ed suspensions.
Regar dless of t he explanat ion for deviant behavior , oscillat or y infor -
mat ion is of limit ed usefulness if dat a ar e not collect ed wit hin t he linear
viscoelast ic r ange of t he mat er ial behavior .
5.7. Vi scometri c and Li near Vi scoelasti c Functi ons
Nor mal st r ess funct ions ar ise fr om t he viscoelast ic nat ur e of
mat er ials; t her efor e, it is r easonable t o expect linear viscoelast ic
mat er ial funct ions det er mined fr om oscillat or y t est ing t o be r elat ed t o
st eady shear behavior . Exact r elat ionships can be det er mined in t he
lower limit s of shear r at e and fr equency (Walt er s, 1975):
[5.70]
and
[5.71]
wher e, r ecall (Eq. [1.23]), . Alt hough t heir use-
fullness is limit ed because t hey only apply at low fr equencies and shear
r at es, t he above equat ions have pr oven t o be valid for numer ous polymer
melt s and solut ions.
Cox and Mer z (1958) obser ved t hat t he complex viscosit y is near ly
equal t o t he st eady shear viscosit y when t he shear r at e and fr equency
ar e equal:
[5.72]
This empir ical r elat ionship, now r efer r ed t o as t he "Cox-Mer z r ule," may
be useful for mat er ials t hat ar e mor e easily t est ed under oscillat or y
inst ead of st eady shear condit ions. The r ule may, for example, apply t o

0

0
G

2
1
1
1 0


1
2
1
1
1
0

1
f (

) (
11

22
)/(

)
2

5.7 Viscometric and Linear Viscoelastic Functions 339


polymer ic fluids t hat have a lar ge nor mal st r ess differ ence, leading t o
a Weissenber g effect (r od climbing), t hat complicat es st eady shear
t est ing.
The Cox-Mer z r ule seems t o wor k well wit h many synt het ic and
biopolymer disper sions (da Silva and Rao, 1992). Rao and Cooley (1992)
found t hat Eq. [5.72] could be applied t o t omat o past e by int r oducing a
simple shift fact or int o t he comput at ion: Complex viscosit y was evalu-
at ed at a fr equency of wher e , t he shift fact or , was equal t o
appr oximat ely 0.0074. Bist any and Kokini (1983) found t he Cox-Mer z
r ule t o be inapplicable t o var ious foods. In t he same wor k, however , t he
following r elat ionships wer e able t o effect ively r elat e st eady shear and
dynamic r heological pr oper t ies:
[5.73]
and
[5.74]
wher e , , , and ar e empir ical const ant s det er mined fr om
exper iment al dat a. Values of t hese const ant s for var ious foods ar e given
in Table 5.9. The cor r esponding nor mal st r ess differ ences and power
law fluid pr oper t ies (needed t o calculat e appar ent viscosit y) for t he foods
list ed in Table 5.9 ar e given in Appendix [6.6].
Dor aiswamy et al. (1991) pr oposed an ext ended Cox-Mer z r ule
r elat ing t he complex viscosit y t o t he st eady shear viscosit y. They found
t hat a plot of t he complex viscosit y ver sus was equivalent t o a plot
of t he appar ent viscosit y ver sus shear r at e. was defined as t he
"effect ive shear r at e" wher e is t he st r ain amplit ude given in Eq. [5.30].
The equivalence of complex and st eady shear viscosit ies was found t o
be valid over a wide r ange of par amet er s involving concent r at ed sus-
pensions having yield st r esses.

G

2
C(
1
)

*
C()

C C

o
340 Chapter 5. Viscoelasticity
Table 5.9. Empirical Constants
*
for Eq. [5.73] and [5.74] (Source: Kokini, 1992)
Food
Wh ipped cr ea m ch ees e 93.21 0.750 13.87 1.146
Cool Wh ip 50.13 1.400 6.16 1.098
St ick Bu t t er 49.64 0.986 0.79 1.204
Wh ipped bu t t er 43.26 0.948 33.42 1.255
St ick ma r ga r in e 35.48 0.934 1.28 1.140
Ket ch u p 13.97 0.940 14.45 1.069
Pea n u t bu t t er 13.18 1.266 1.66 1.124
Squ eeze ma r ga r in e 11.12 1.084 52.48 1.022
Ca n n ed fr os t in g 4.40 1.208 4.89 1.098
Ma r s h ma llow flu ff 3.53 0.988 1.26 0.810
* Da t a collect ed a t r oom t emper a t u r e over a s h ea r r a t e a n d fr equ en cy r a n ge of 0.1
t o 100 s
-1
. Un it s of a n d u s ed in Eq. [5.73] a n d [5.74] a r e r a d s
-1
, s
-1
,
Pa , Pa s
2
, Pa s a n d Pa s , r es pect ively.
Table 5.10. Stress Relaxation Data from Compression Testing of Apple Tissue at
Room Temperature (Cylindrical Sample: Length =2 cm, Diameter =2 cm)
(s ) (Pa ) (s )
0.0 754
0.6 601 3.0
1.2 572 5.0
1.8 562 7.1
2.4 549 8.8
3.0 536 10.4
3.6 526 11.9
6.0 504 18.1
9.0 484 25.1
12.0 468 31.6
15.0 461 38.6
18.0 452 44.9
24.0 435 56.7
30.0 426 69.0
42.0 409 91.8
54.0 406 117.0
C C
,

, G,
1
,
*

t
o
t

o

5.8.1 Generalized Maxwell Model of Stress Relaxation 341
5.8. Example Problems
5.8.1. Generali zed Maxwell Model of Stress Relaxati on
Using t he gener alized Maxwell model (Eq. [5.11]), est imat e t he r elax-
at ion t ime ( ) of t he apple t issue descr ibed by t he dat a pr esent ed in
Table 5.10.
Recall Eq. [5.11], wit h r efer r ing t o t he st r ess in compr ession:
Assume t he mat er ial has r eached equilibr ium at t he complet ion of t he
t est so = 406 Pa. Then, r egr ession of t he dat a (Fig. 5.27) using t he
logar it hmic t r ansfor mat ion of Eq. [5.11],
yields s
-1
or s, and Pa making
= 606.75 Pa. Not e t hat t he value of , det er mined fr om t he r egr ession
par amet er , is lower t han t he exper iment al value of 754 Pa given for t he
init ial st r ess in Table 5.10.
Figure 5.27. Stress relaxation data for apple tissue modeled using Eq. [5.11].

rel

f (t )
e
+ (
o

e
) exp

rel
_

e
ln( 406) ln(
o

e
) t /
rel
1/
rel
0.0906
rel
11.04
o
406 200.75

o

o
0 10 20 30 40 50 60
0
100
200
300
400
Time, s
e
,

P
a
e
Apple Tissue
= 406 Pa
342 Chapter 5. Viscoelasticity
5.8.2. Li neari zed Stress Relaxati on
St r ess r elaxat ion dat a for apple t issue ar e given in Table 5.10 wit h t he
r aw dat a plot t ed in Fig. 5.28. Det er mine t he coefficient s for t he nor -
malized st r ess equat ion pr oposed by Peleg and Nor mand (1983), Eq.
[5.14]:
wher e t he value of is t aken fr om Table 5.10 as 754 Pa.
Linear r egr ession of t his equat ion (an excellent fit ) yields t he fol-
lowing: = 4.52 s, = 2.12. Linear ized dat a ar e plot t ed in Fig. 5.29.
Result s ar e compar able t o t hose r epor t ed for pot at o flesh (Table 5.5:
= 4.40 s, = 1.56).
Figure 5.28. Stress relaxation data for apple tissue at room temperature.

o
t

o

k
1
+ k
2
t

o
k
1
k
2
k
1
k
2
0 10 20 30 40 50 60
300
400
500
600
700
800
Time, s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Apple Tissue
5.8.3 Analysis of Creep Compliance Data 343
Figure 5.29. Normalized stress relaxation data for apple tissue.
5.8.3. Analysi s of Creep Compli ance Data
Est imat e t he par amet er s of t he four par amet er Bur ger s model (Fig. 5.6)
t o r epr esent t he shear cr eep compliance dat a given in Table 5.11.
Compar e t he r esult ing cur ve t o t he 6 par amet er model (Fig. 5.30) for -
mulat ed by adding an addit ional Kelvin element t o t he Bur ger s model.
Also, det er mine t he par amet er s for t he simple linear ized cr eep
compliance model suggest ed by Peleg (1980).
The Bur ger s model, expr essed in t er ms of t he cr eep compliance
funct ion, is descr ibed by Eq. [5.22]:
Model par amet er s may be est imat ed using t he following pr ocedur e:
1. The inst ant aneous compliance, , is det er mined fr om t he or iginal
dat a (Cur ve 1 on Fig. 5.31 or Table 5.11) at as 0.00220
cm
2
/dyne.
0 10 20 30 40 50 60
0
20
40
60
80
100
120
140
Time, s
Apple Tissue
,


s
o
t
o
-
J f (t ) J
0
+ J
1

1 exp

ret
_

,
_

,
+
t

0
J
0
t 0
344 Chapter 5. Viscoelasticity
Table 5.11. Shear Creep Compliance of Skim Milk Curd Fortified with -casein
(Data from Halim and Shoemaker, 1990)
Da t a Time J
a
J
b
J
c
t / J
d
Poin t (s ) (cm
2
/ dyn e) (cm
2
/ dyn e) (cm
2
/ dyn e) (s dyn e/ cm
2
)
1 0 0.00220 0.00220 0.00220 0
2 0.89 0.00249 0.00237 0.00241 357
3 2.01 0.00264 0.00257 0.00262 761
4 3.79 0.00293 0.00284 0.00287 1,293
5 8.93 0.00331 0.00337 0.00328 2,698
6 14.06 0.00355 0.00369 0.00352 3,961
7 18.97 0.00372 0.00389 0.00369 5,099
8 23.44 0.00386 0.00401 0.00382 6,073
9 28.71 0.00398 0.00411 0.00395 7,241
10 34.02 0.00408 0.00419 0.00406 8,338
11 39.06 0.00418 0.00426 0.00416 9,344
12 44.20 0.00426 0.00432 0.00424 10,376
13 48.88 0.00434 0.00437 0.00432 11,263
14 58.48 0.00447 0.00447 0.00445 13,083
15 68.97 0.00458 0.00458 0.00458 15,059
16 78.57 0.00469 0.00468 0.00469 16,752
17 93.75 0.00485 0.00484 0.00485 19,330
18 109.25 0.00499 0.00500 0.00502 21,894
19 118.97 0.00509 0.00509 0.00512 23,373
a
r a w da t a ;
b
pr edict ed u s in g 4 pa r a met er Bu r ger s model;
c
pr edict ed u s in g 6 pa r a met er
model;
d
ca lcu la t ed fr om r a w da t a
2. Subt r act fr om Cur ve 1 t o gener at e Cur ve 2 (Fig. 5.31). Using
t he st r aight por t ion of t he cur ve (t he last six dat a point s, ),
linear r egr ession analysis yields and fr om t he int er cept and
slope of
[5.75]
as = 0.00168 cm
2
/dyne and = 97,847 P. Not e t hat r eflect s
t he fully ext ended (equlibr at ed) Kelvin element making t he
exponent ial t er m in t he or iginal equat ion equal t o zer o.

J
0
t 58.48
J
1

0
( J J
0
) J
1
+
t

0
J
1

0
J
1
5.8.3 Analysis of Creep Compliance Data 345
3. Using t he exponent ial por t ion of t he dat a (t he fir st 8 dat a point s,
s), t he r et ar dat ion t ime is det er mined fr om linear r egr es-
sion analysis (including a logar it hmic t r ansfor mat ion of t he dat a)
over of
[5.76]
as = 8.635 s.
Figure 5.30. Six parameter model for creep compliance.
Subst it ut ing t he const ant s found above, t he complet e Bur ger s model
may be expr essed as
t 28.71
J <J
1
+J
0
ln

1
J J
0
J
1
_

,

t

ret

ret
G
1
G
1
0
0
G
2 2
J 0.0022 + 0.00168

1 exp

t
8.635
_

,
_

,
+
t
97847
346 Chapter 5. Viscoelasticity

Figure 5.31. Shear creep compliance of skim milk curd fortified with -casein.
Acompar ison of t he cur ve pr edict ed wit h t he above equat ion and act ual
dat a r eveals (Fig. 5.31, Table 5.11) t hat t he Bur ger s model adequat ely
r epr esent s t he dat a; however , bet t er accur acy is obt ained using a mor e
complex 6 par amet er model (Fig. 5.30) descr ibed by Eq. [5.23], wit h
, as
[5.77]
wher e:
[5.78]
and
[5.79]
0 20 40 60 80 100 120
0
0.001
0.002
0.003
0.004
0.005
0.006
Time, s
C
r
e
e
p

C
o
m
p
l
i
a
n
c
e
,

c
m

/
d
y
n
e
2
6 parameter model
4 parameter Burgers model
Curve 1, raw data
Curve 2, adjusted raw data
Skim Milk Curd

m 2
J f (t ) J
0
+ J
1

1 exp

t
(
ret
)
1
_

,
_

,
+ J
2

1 exp

t
(
ret
)
2
_

,
_

,
+
t

0
(
ret
)
1


1
G
1
(
ret
)
2


2
G
2
5.8.3 Analysis of Creep Compliance Data 347
The const ant s in Eq. [5.77] (for t he same -casein for t ified skin milk
cur d under consider at ion) wer e det er mined by Halim and Shoemaker
(1990), using nonlinear cur ve fit t ing pr ocedur es, wit h t he following
r esult :
Compar ing t his pr edict ion equat ion t o t he r aw dat a shows t hat it does
an excellent job modeling t he shear cr eep compliance (Fig. 5.31, Table
5.11).
Figure 5.32. Linearized shear creep compliance of skim milk curd fortified with
-casein.
Peleg (1980) suggest ed t hat cr eep dat a could be modeled wit h t he
following linear equat ion:
[5.80]

J 0.0022 + 0.00064

1 exp

t
3.14
_

,
_

,
+ 0.001068

1 exp

t
19.5
_

,
_

,
+
t
97707
0 20 40 60 80 100 120 140
0
5,000
10,000
15,000
20,000
25,000
Time, s
t

/

J
,

s

d
y
n
e
/
c
m
2
Skim Milk Curd

t
J
k
1
+ k
2
t
348 Chapter 5. Viscoelasticity
Regr ession analysis of t he dat a, using Eq. [5.80], yielded dyne
cm
-2
s and dyne cm
-2
wit h . Raw dat a and t he r esult ing
cur ve ar e plot t ed in Fig. 5.32. The 6 par amet er model involves complex
cur ve fit t ing, but gives a ver y accur at e r epr esent at ion of exper iment al
dat a. The simple linear equat ion is less pr ecise, but oft en adequat e for
invest igat ing pr act ical cr eep pr oblems involving biological mat er ials.
5.8.4. Plotti ng Osci llatory Data
Develop alt er nat ive plot s showing t he dynamic behavior of t he con-
cent r at ed solut ion (5% lambda car r ageenan solut ion) illust r at ed in Fig.
5.21.
The dat a given in Fig. 5.22 wer e fit t o a power law model r esult ing
in t he following equat ions (Table 5.8):
and
Using t hese equat ions ot her r heological par amet er s can be easily cal-
culat ed:
Result s ar e pr esent ed in Fig. 5.33.
Two ver y useful met hods of r epr esent ing mechanical spect r a ar e t o
plot t he st or age modulus and t he loss modulus, or t he st or age modulus
and t he dynamic viscosit y. Cr ossover point s (Fig. 5.33), such as t he one
pr ovided by t he int er sect ion of and , may pr ovide useful bench-
mar ks for compar ing pr oduct s or t r eat ment s. In t his case t he cr ossover
occur s at a fr equency of 5.61 r ad/s. Since at t hat point , it is also
t he fr equency wher e t he t angent of t he phase angle ( ) is equal t o 1
k
1
1050.8
r
2
0.99 k
2
196.6
G 16.26()
.84
G 27.78()
.52

G

27.78()
.48

G

16.26()
.16
tan
G
G
1.71()
.32
G
*
(G)
2
+ (G)
2
(16.26()
.84
)
2
+ (27.78()
.52
)
2

*
()
2
+ ()
2
(27.78()
.48
)
2
+ (16.26()
.16
)
2
G G
G G
tan
5.8.4 Plotting Oscillatory Data 349
Figure 5.33. Oscillatory testing results for a 5% lambda carrageenan solution.
because . Obser ving changes in is a useful means of evalu-
at ing t he t r ansit ion fr om liquid-like t o solid-like behavior : t an delt a
decr eases wit h mor e solid-like behavior . Combined par amet er s, and
, ar e also useful in looking at over all r heological behavior . A popular
combinat ion is t o plot t he var iat ion of , , and wit h fr equency. In
some cases, single point measur ement s ar e useful: , measur ed at 50
r ad/s, cor r elat es well wit h t he human per cept ion of t hickness or viscosit y
(Hill, 1991; Hill et al., 1995).
0.01 0.03 0.1 0.3 1 3 10 30 100
0.1
0.3
1
3
10
30
100
300
1,000
P
a

o
r

P
a

s

o
r

D
i
m
e
n
s
i
o
n
l
e
s
s
(rad/s)
G
tan
G
*
*
G
Concentrated Solution
/4 tan

*
G
*

*
G G

*
6 . Appe ndic e s
6 . 1 . Conve rs ion Fac t ors and SI Pre fixe s
Density
1 g cm
-3
= 1000 kg m
-3
= 62.428 lbm ft
-3
= 0.0361 lbm in
-3
1 lbm ft
-3
= 16.0185 kg m
-3
Force
1 N = 1 kg m s
-2
= 10
5
dyne = 0.22481 lbf = 0.102 kgf
1 lbf = 4.448 N = 0.4536 kgf
1 dyne = 1 g cm s
-2
= 10
-5
N
Length
1 m = 100 cm = 1000 mm = 10
6
m = 3.2808 ft = 39.37 in = 1.0936 yd
1 in = 2.54 cm = 25.40 mm = 0.0254 m = 0.0833 ft = 0.02778 yd
Power
1 hp = 550 ft lbf s
-1
= 745.70 W = 0.7457 kW = 0.7068 Btu s
-1
1 Btu hr
-1
= 0.2931 W = 0.2931 J s
-1
Pressure and Stress
1 bar = 10
5
N m
-2
= 10
5
Pa = 14.5038 lbf in
-2
= 0.987 atm = 10.2 m at 4 C
1 Pa = 1 N m
-2
= 10 dyne cm
-2
= 9.8692 (10
-6
) atm = 0.1020 kgf m
-2
1 lbf in
-2
= 6895 Pa = 6.804 (10
-2
)atm = 6.895 kPa = 2.307 ft at 4 C
1 dyne cm
-2
= 0.10 Pa = 10
-6
bar = 0.987 (10
-6
) atm
1 atm = 1.01325 (10
5
) N m
-2
= 101.325 kPa = 14.696 psi = 1.013 bar = 760 torr
1 atm = 760 mm Hg at 0 C = 33.90 ft at 4 C = 1.013 (10
6
) dyne cm
-2
Revolution and Rotational Speed
1 rev = 1 cycle = 2 rad
1 Hz = 1 cycle/s = 1 rev/s = 2 rad/s = 6.283 rad/s
1 rev/min = 0.1047 rad/s
Temperature
T
Kelvin
= T
Celsius
+ 273.15
T
Kelvin
= (T
Fahrenheit
+ 459.67) / 1.8
T
Fahrenheit
= 1.8 T
Celsius
+ 32
T
Celsius
= (T
Fahrenheit
- 32) / 1.8
Torque, Energy, and Work
1 N m = 100 N cm = 1 J = 1 kg m
2
s
-2
= 10
7
dyne cm = 8.85 in lbf
1 dyne cm = 10
-7
N m = 10
-5
N cm
Viscosity (Absolute or Dynamic, followed by Kinematic)
1 P = 1 dyne s cm
-2
= 0.1 Pa s = 100 cP = 100 mPa s
1 Pa s = 1000 cP = 10 P = 1 kg m
-1
s
-1
= 1 N s m
-2
= 0.67197 lbm ft
-1
s
-1
1 cP = 1 mPa s = 0.001 Pa s = 0.01 P
1 lbm ft
-1
s
-1
= 1.4882 kg m
-1
s
-1
= 1488.2 cP = 2.0885 (10
-2
) lbf s ft
-2
kinematic viscosity (cSt) = absolute viscosity (cP) / density (g cm
-3
)
1 cSt = 0.000001 m
2
s
-1
= 1 mm
2
s
-1
= 5.58001 in
2
hr
-1
= 0.00155 in
2
s
-1
1 St = 100 cSt = 0.0001 m
2
s
-1
1 m
2
s
-1
= 10
-5
cSt = 10.7639 ft
2
s
-1
Volume
1 m
3
= 10
6
cm
3
= 10
3
L (liter) = 264.17 gal (US) = 35.315 ft
3
= 219.97 gal (UK)
1 ft
3
= 0.028317 m
3
= 7.481 gal (US) = 28.317 L = 6.2288 gal(UK)
1 gal (US) = 4 qt = 3.7854 L = 3785.4 cm
3
= 0.8327 gal (UK) = 0.003785 m
3

H
2
O
H
2
O
H
2
O

351
SI Prefixes
Fa ct or Pr efix Symbol Fa ct or Pr efix Symbol
10
18
exa E 10
-1
deci d
10
15
pet a P 10
-2
cen t i c
10
12
t er a T 10
-3
milli m
10
9
giga G 10
-6
micr o
10
6
mega M 10
-9
n a n o n
10
3
kilo k 10
-12
pico p
10
2
h ect o h 10
-15
femt o f
10
1
deka da 10
-18
a t t o a
6 . 2 . Gre e k Alphabe t
Alph a Nu
Bet a Xi
Ga mma Omicr on
Delt a Pi
Eps ilon Rh o
Zet a Sigma
Et a Ta u
Th et a Ups ilon
Iot a Ph i
Ka ppa Ch i
La mbda Ps i
Mu Omega













352 Appendices
6 . 3 . Mat he mat ic s : Root s , Powe rs , and Logarit hms
Root s a n d Power s
*
Loga r it h ms
*
If , t h en is t h e loga r it h m of t o t h e ba s e wh er e is a
fin it e pos it ive n u mber ot h er t h a n 1. Als o, if t h en . In
t h is t ext , va lu es of equ a l t o 10 a n d a r e u s ed mea n in g
a n d
Th e followin g pr oper t ies of loga r it h ms , expr es s ed in t er ms of ba s e
, a r e t r u e for a n y ba s e:
*
Based on a summary given in Hudson, R.G. 1939. The Engineers Manual (second edition, 25th
printing). John Wiley & Sons, NY.
a
n
= a a a a to n factors
a
n
=
1
a
n
a
m
a
n
= a
m + n
a
m
a
n
= a
m n
(ab)
n
= a
n
b
n

a
b

n
=
a
n
b
n
(a
m
)
n
= (a
n
)
m
= a
mn
(
n

a )
n
= a
a
1
n
=
n

a a
m
n
=
n
a
m
n

ab = (
n

a ) (
n

b ) n

a
b
=
n

a
n

b
log
b
N = x x N b b
b
x
= N log
b
N = x
b e
log
10
N = x or 10
x
= N
log
e
N = lnN = x or e
x
= exp( x) = N
e
ln(MN) = ln(M) + ln(N)
ln

M
N

= ln(M) ln(N)
ln(M
a
) = a ln(M)
ln

a
M
c

=
c
a
ln(M)
353
6 . 4 . Line ar Re gre s s ion Analys is of Two Variable s
Lin ea r r egr es s ion is a s t a t is t ica l met h od of fit t in g t wo va r ia bles
t o a lin ea r equ a t ion :
Th e con s t a n t s , (t h e in t er cept ) a n d (t h e s lope), a r e ca lcu la t ed wit h
t h e followin g for mu la s :
wh er e:
= n u mber of obs er va t ion s in t h e s a mple
= mea n of x va lu es
= mea n of y va lu es
Th e met h od of lea s t s qu a r es min imizes t h e s u m of t h e s qu a r ed
er r or bet ween t h e a ct u a l a n d es t ima t ed va lu es : a n d a r e fou n d
u s in g a pr ocedu r e wh ich min imizes
wh er e is t h e va lu e t o be es t ima t ed a n d is t h e cor r es pon din g
es t ima t e of a t . Compu t a t ion s a r e ea s ily don e u s in g compu t er
s pr ea d-s h eet pr ogr a ms or h a n d ca lcu la t or s .
A mea s u r e of cor r ela t ion bet ween t wo va r ia bles ca n be defin ed in
t er ms of t h e s a mple cor r ela t ion coefficien t ( ) wh ich ca n t a ke on
va lu es bet ween -1 a n d 1. A cor r ela t ion va lu e of zer o mea n s t h er e
is n o a s s ocia t ion . A va lu e of -1 mea n s t h er e is a per fect n ega t ive
cor r ela t ion a n d a va lu e of +1 mea n s t h er e is a per fect pos it ive cor -
r ela t ion (Fig. 6.1). Cor r ela t ion ma y a ls o be expr es s ed in t er ms of
t h e coefficien t of det er min a t ion , . Wh en , a ll poin t s fa ll on t h e
pr edict ed lin e. If , t h er e is a s mu ch va r ia t ion in t h e es t ima t e
of a s t h er e is in t h e va r ia ble mea n in g t h a t a n y obs er ved flu c-
t u a t ion s a r e du e t o r a n dom va r ia t ion s in .
Th e s t a n da r d er r or of t h e es t ima t e ca n a ls o be a u s efu l mea s u r e
of h ow well t h e r egr es s ion lin e fit s t h e da t a . Th is pa r a met er is
ca lcu la t ed a s
y = a + b x
a b
b =
n
i = 1
n
x
i
y
i

i = 1
n
x
i

i = 1
n
y
i
n
i = 1
n
( x
i
)
2


i = 1
n
x
i

2
a = y bx
n
x
y
a b
n
i = 1
n
( y
i
y
x
)
2
y
i
y
x
y x
r
r
2
r
2
= 1
r
2
= 0
y y
y
354 Appendices
wh er e:
= t h e s t a n da r d er r or of t h e es t ima t e of on
Pr oper t ies of t h e s t a n da r d er r or of t h e es t ima t e a r e a n a logou s t o
t h os e of s t a n da r d devia t ion . As s u min g a s u fficien t ly la r ge s a mple,
pa ir s of lin es con s t r u ct ed pa r a llel t o t h e r egr es s ion lin e of on , a t
ver t ica l dis t a n ces of , 2 a n d 3 , s h ou ld in clu de 68%, 95%
a n d 99.7% (r es pect ively) of t h e s a mple da t a poin t s .
Figure 6.1. Relationship between and for a perfect positive correlation (r =
1), a perfect negative correlation (r = -1), and no correlation (r = 0).
Example Proble m. Det er min e t h e power la w, Bin gh a m pla s t ic,
Ca s s on , a n d Her s ch el-Bu lkley model pa r a met er s fr om t h e da t a for
t h e ca r r a geen a n gu m s olu t ion given in Ta ble 6.1.
Lin ea r r egr es s ion ca n on ly be per for med on a lin ea r equ a t ion of
t h e for m . Th e Bin gh a m pla s t ic model a lr ea dy exis t s in t h is
for m; h owever , t h e ot h er models ma y r equ ir e a t r a n s for ma t ion of
t h e da t a befor e t h ey ca n be pr es en t ed in a lin ea r for m. A s u mma r y
s
y, x
=

i = 1
n
(y
i
y
x
)
2
n 2
s
y, x
y x
y x
s
y, x
s
y, x
s
y, x
r = -1
r = 1
r = 0
x
y
y x
y = a + bx
355
Table 6.1. Raw and Transformed Data of a 1% aqueous solution of carrageenan
gum at 25 C (Data from Prentice and Huber, 1983)
2.61 9.88 0.96 2.29 1.62 3.14 0.412
2.97 11.4 1.09 2.43 1.72 3.38 0.626
2.81 12.9 1.03 2.56 1.68 3.59 0.536
3.44 14.1 1.24 2.65 1.85 3.75 0.850
3.80 17.6 1.34 2.87 1.95 4.20 0.993
4.85 26.3 1.58 3.27 2.20 5.13 1.322
6.61 42.0 1.89 3.74 2.57 6.48 1.707
6.19 48.6 1.82 3.88 2.49 6.97 1.627
5.89 49.3 1.77 3.90 2.43 7.02 1.567
7.22 55.5 1.98 4.02 2.69 7.45 1.812
8.20 58.8 2.10 4.07 2.86 7.67 1.960
9.08 75.4 2.21 4.32 3.01 8.68 2.076
11.63 104.1 2.45 4.65 3.41 10.20 2.354
10.65 110.4 2.37 4.70 3.26 10.51 2.257
12.75 120.5 2.55 4.79 3.57 10.98 2.455
13.10 136.5 2.57 4.92 3.62 11.68 2.484
14.90 145.8 2.70 4.98 3.86 12.07 2.624
15.85 187.1 2.76 5.23 3.98 13.68 2.691
12.70 210.2 2.54 5.35 3.56 14.50 2.451
20.50 270.0 3.02 5.60 4.53 16.43 2.965
of t h is t r a n s for ma t ion , a n d r egr es s ion r es u lt s , a r e given in Ta ble
6.2. Not e t h a t t h e yield s t r es s for t h e Her s ch el-Bu lkley model mu s t
be es t ima t ed befor e lin ea r r egr es s ion ca n pr oceed. Es t ima t in g ,
, a n d s imu lt a n eou s ly wou ld r equ ir e a n on lin ea r es t ima t ion
pr ocedu r e.
In for ma t ion pr es en t ed in Ta ble 6.2 ca n be pr es en t ed in t er ms of t h e
models u n der in ves t iga t ion :
Bin gh a m, :
Pa , Pa s
Her s ch el-Bu lkley, :
Pa , Pa s
n
,
Ca s s on , :
Pa , Pa
.5
s
.5
Power La w, :
Pa s
n
,

()
.5
(

)
.5


ln ln

ln( 1.1)

o
K n
=
pl
(

) +
o

o
= 3.20
pl
= 0.066
= K(

)
n
+
o

o
= 1.10 K = 0.306 n = 0.74
()
.5
= K
1
(

)
.5
+ (
o
)
.5

o
= 1.16 K
1
= 0.211
= K(

)
n
K = 0.66 n = 0.60
356 Appendices
Coefficien t s of det er min a t ion (Ta ble 6.2) in dica t e t h a t t h e power la w
a n d Her s ch el-Bu lkley equ a t ion s pr ovide t h e bes t model of t h e
exper imen t a l da t a .
Table 6.2. Linear regression parameters for different rheological models.
Model
Bin gh a m 0.92 1.44
Power La w 0.98 0.094
Ca s s on 0.96 0.179
*
Her s ch el- 0.98 0.119
Bu lkley
*
Pa wa s det er min ed fr om a gr a ph ica l a n a lys is of t h e da t a .
r
2
y x a b s
y, x



o
= 3.20
pl
= 0.066
ln ln

lnK = 0.411 n = 0.60


()
.5
(

)
.5
(
o
)
.5
= 1.078 K
1
= 0.211
ln(
o
) ln

lnK = 1.185 n = 0.74

o
= 1.1
357
6 . 5 . Hooke an Prope rt ie s
Material Material
(N m
-2
) (N m
-2
)
Soft foam rubber 10
2
Carrots (2-4) x 10
7
Rubber 8 x 10
5
Pears (1.2-3) x 10
7
Dry Spaghetti 0.3 x 10
10
Potatoes (0.6-1.4) x 10
7
Lead 1.0 x 10
10
Apples raw (0.6-1.4) x 10
7
Concrete 1.7 x 10
10
Gelatin (gel) 0.02 x 10
7
Glass 7 x 10
10
Peach (0.2-2) x 10
7
Iron 8 x 10
10
Banana (0.08-0.3) x 10
7
Steel 25 x 10
10
Material Material
( - ) ( - )
Cheddar cheese 0.50 Copper 0.33
Potato tissue 0.49 Steel 0.30
Rubber 0.49 Glass 0.24
Apple tissue 0.37 Bread-crumbs 0.00
Apple 0.21-0.34 Cork 0.00
Material Material
(N m
-2
) (N m
-2
)
Dough 1.4 x 10
6
Silver 10
11
Rubber 1.9 x 10
7
Steel 1.6 x 10
11
Granite 3 x 10
10
Glass 3.9 x 10
10
Source: Lewis, 1987.
E E

K K
358 Appendices
6 . 6 . St e ady She ar and Normal St re s s Diffe re nc e
Product
*
( ) (-) ( ) (-)
Apple butter 222.90 0.145 156.03 0.566
Canned frosting 355.84 0.117 816.11 0.244
Honey 15.39 0.989 - -
Ketchup 29.10 0.136 39.47 0.258
Marshmallow cream 563.10 0.379 185.45 0.127
Mayonnaise 100.13 0.131 256.40 -0.048
Mustard 35.05 0.196 65.69 0.136
Peanut butter 501.13 0.065 3785.00 0.175
Stick butter 199.29 0.085 3403.00 0.393
Stick margarine 297.58 0.074 3010.13 0.299
Squeeze margarine 8.68 0.124 15.70 0.168
Tub margarine 106.68 0.077 177.20 0.358
Whipped butter 312.30 0.057 110.76 0.476
Whipped cream cheese 422.30 0.058 363.70 0.418
Whipped desert topping 35.98 0.120 138.00 0.309
*
Data were collected over a shear rate of 0.1 to 100 s
-1
at 30.5 C
Source: Dickie and Kokini, 1983.
K n K n
Pa s
n
Pa s
n

= K(

)
n
N
1
=
11

22
= K(

)
n
359
6 . 7 . Yie ld St re s s of Fluid Foods
Pr odu ct (Pa ) Mea s u r emen t Met h od Sou r ce
Ket ch u p 22.8 ext r a pola t ion Ofoli et a l. (1987)
Mu s t a r d 34.0 ext r a pola t ion Ofoli et a l. (1987)
Mir a cle Wh ip 54.3 ext r a pola t ion Ofoli et a l. (1987)
Apr icot Pu r ee 17.4 ext r a pola t ion Ofoli et a l. (1987)
Milk Ch ocola t e 10.9 ext r a pola t ion Ofoli et a l. (1987)
Min ced Fis h Pa s t e 1600-2300 ext r a pola t ion Na ka ya ma et
a l.(1980)
Ma yon n a is e 24.8-26.9 s t r es s t o in it ia t e flow De Kee et a l.(1980)
Ket ch u p 15.4-16.0 s t r es s t o in it ia t e flow De Kee et a l.(1980)
Toma t o Pa s t e 83.9-84.9 s t r es s t o in it ia t e flow De Kee et a l.(1980)
Ra w Mea t Ba t t er 17.9 ext r a pola t ion Toledo et a l.(1977)
Toma t o Pu r ee 23.0 s t r es s deca y Ch a r m(1962)
Apples a u ce 58.6 s t r es s deca y Ch a r m(1962)
Toma t o Pa s t e 107-135 s qu eezin g flow Ca mpa n ella a n d
Peleg(1987)
Ket ch u p 18-30 s qu eezin g flow Ca mpa n ella a n d
Peleg(1987)
Mu s t a r d 52-78 s qu eezin g flow Ca mpa n ella a n d
Peleg(1987)
Ma yon n a is e 81-91 s qu eezin g flow Ca mpa n ella a n d
Peleg(1987)
Apples a u ce 45-87 s qu eezin g flow Ca mpa n ella a n d
Peleg(1987)
Apples a u ce 46-82 va n e met h od Qiu a n d Ra o(1988)
Ket ch u p 26-30 va n e met h od Mis s a ir e et a l.
(1990)
Spa gh et t i Sa u ce 24-28 va n e met h od Mis s a ir e et a l.
(1990)
Toma t o Pu r ee 25-34 va n e met h od Mis s a ir e et a l.
(1990)
Pu mpkin Fillin g 20 va n e met h od Mis s a ir e et a l.
(1990)
Apples a u ce 38-46 va n e met h od Mis s a ir e et a l.
(1990)
Ba by food, pea r s 49 va n e met h od Mis s a ir e et a l.
(1990)
Ba by food, 25 va n e met h od Mis s a ir e et a l.
pea ch es (1990)
Ba by food, ca r r ot s 71 va n e met h od Mis s a ir e et a l.
(1990)

o
360 Appendices
Static and Dynamic Yield Stresses of Food Products
Pr odu ct
*
St a t ic Yield Dyn a mic Yield
St r es s St r es s
(Pa ) (Pa )
Toma t o pu r ee, Br a n d A 34.4 3.7 28.1 4.2
Toma t o pu r ee, Br a n d B 30.0 4.2 27.6 3.4
Apple s a u ce, Br a n d A 77.3 0.0 48.2 4.7
Apple s a u ce, Br a n d B 48.2 4.7 38.0 4.7
Toma t o ket ch u p, Br a n d A 51.3 5.0 40.6 4.5
Toma t o ket ch u p, Br a n d A 43.2 3.4 39.6 3.4
Spa gh et t i s a u ce, Br a n d A 26.3 4.5 18.3 0.0
Spa gh et t i s a u ce, Br a n d B 24.8 3.4 21.9 0.0
Ba by food, pea r s 31.8 5.0 24.8 4.0
Ba by food, ca r r ot s 64.0 4.0 35.7 5.0
Ba by food, pea ch es 22.9 3.4 20.1 3.4
Ma yon n a is e, Br a n d A 204.4 5.0 112.6 4.0
Ma yon n a is e, Br a n d B 163.8 4.2 99.4 4.0
Mu s t a r d, Br a n d A 82.5 5.0 53.1 5.3
Mu s t a r d, Br a n d B 103.8 5.0 53.4 5.0
*
Da t a wer e collect ed a t 25 C by u s in g t h e va n e met h od oper a t in g
in t h e con t r olled r a t e mode. Sou r ce: Yoo et a l., 1995.
Br iggs a n d St effe (1996) det er min ed t h e yield s t r es s of fr ozen ice
cr ea m u s in g t h e va n e met h od. Res u lt s wer e s u mma r ized wit h t h e
followin g equ a t ion s :
wh er e is t h e yield s t r es s in Pa , a n d is t h e t emper a t u r e in .
Th es e equ a t ion s a r e va lid over t h e r a n ge of t emper a t u r es t ypica lly
ma in t a in ed for h a n d-dipped ice cr ea m: -12 t o -16 . Yield
s t r es s es , r a n gin g fr om 2.5 t o 8.0 kPa , decr ea s ed wit h r edu ct ion s in
t emper a t u r e. Ch ocola t e h a d s igh t ly h igh er va lu es t h a n va n illa .














o
= 22.33exp(.37T) for chocolate ice cream

o
= 39.32exp(.33T) for vanilla ice cream

o
T C
C C
361
6 . 8 . Ne wt onian Fluids
Product T
( C) (cP)
Acetic acid 15 1.31
18 1.30
25.2 1.155
30 1.04
41 1.00
59 0.70
70 0.60
100 0.43
Acetone -42.5 0.695
-30.0 0.575
-20.9 0.510
-13.0 0.470
-10.0 0.450
0 0.399
15 0.337
25 0.316
30 0.295
41 0.280
Ethanol -32.01 3.84
-17.59 2.68
-0.30 1.80
0 1.773
10 1.466
20 1.200
30 1.003
40 0.834
50 0.702
60 0.592
70 0.504
Glycol 20 19.9
40 9.13
60 4.95
80 3.02
100 1.99
Glycerin -42 6.71E6
-36 2.05E6
-25 2.62E5
-20 1.34E5
-15.4 6.65E4
-10.8 3.55E4
-4.2 1.49E4

362 Appendices
0 1.21E4
6 6260
15 2330
20 1490
25 954
30 629
Mercury -20 1.855
-10 1.764
0 1.685
10 1.615
20 1.554
30 1.499
40 1.450
50 1.407
60 1.367
70 1.331
80 1.298
90 1.268
100 1.240
150 1.300
200 1.052
250 0.995
300 0.950
340 0.921
Water 0 1.787
5 1.519
10 1.307
15 1.139
20 1.002
25 0.8904
30 0.7975
35 0.7194
40 0.6529
45 0.5960
50 0.5468
55 0.5040
60 0.4665
65 0.4335
70 0.4042
75 0.3781
80 0.3547
85 0.3337
90 0.3147
95 0.2975
100 0.2818
Source: Weast et al., 1984.
363
Vis c os it y Pre dic t ion Equat ions for Various Liquids
Equation 1
Equation 2
Equation 3
where is in units of Pa s and T in degrees Kelvin.
Material Eq. # T range(K)
Acetic Acid 2 1.2106x10
6
-3.6612 270-390
Corn Oil 3 -3.5581 -263.32 183.60 290-340
Corn Syrup: DE=100, 3 -4.4500 -234.80 159.91 280-360
20% dry substance
Corn Syrup: DE=42.9, 3 -4.0137 -129.07 201.23 280-360
20% dry substance
Corn Syrup: DE=35.4, 1 -5.5478 868.35 280-360
20% dry substance
Corn Syrup: DE=35.4, 3 -3.8025 -258.10 186.88 280-360
50% dry substance
Corn Syrup: DE=42.9, 3 -3.9975 -267.36 182.96 280-360
50% dry substance
Corn Syrup: DE=75.4, 1 -5.8508 924.84 201.23 280-360
20% dry substance
Ethanol 3 -5.5972 -846.95 -24.124 210-350
Ethylene Glycol 3 -4.5448 -417.05 146.53 280-420
Groundnut Oil 3 -3.9621 -407.46 151.23 290-340
Rapseed Oil 3 -4.4802 -597.20 1119.99 290-340
Soybean Oil 3 -4.4977 -581.28 115.71 290-340
Sunflower Oil 3 -3.6505 -304.27 168.98 290-340
Water 3 -4.5318 -220.57 149.39 270-380
Source: Viswanath and Natarajan, 1989.
log() = A + (B/T)
= AT
B
log() = A + B/(C T)

A B C
364 Appendices
Vis c os it y (c e nt ipois e ) of Suc ros e Solut ions
% sucrose by weight
T (
o
C) 20 40 60
0 3.818 14.82 -
5 3.166 11.60 -
10 2.662 9.830 113.90
15 2.275 7.496 74.90
20 1.967 6.223 56.70
25 1.710 5.206 44.02
35 1.336 3.776 26.62
40 1.197 3.261 21.30
45 1.074 2.858 17.24
50 0.974 2.506 14.06
55 0.887 2.227 11.71
60 0.811 1.989 9.87
65 0.745 1.785 8.37
70 0.688 1.614 7.18
80 0.592 1.339 5.42
85 0.552 1.226 4.75
90 - 1.127 4.17
95 - 1.041 3.73
Source: International Critical Tables 5:23 (1917).
Vis c os it y (c e nt ipois e ) of Et hanol-Wat e r Mixt ure s
Concentration of ethanol, % by weight
T (
o
C) 10 20 30 40 50 60 70 80 90 100
0 3.311 5.32 6.94 7.14 6.58 5.75 4.76 3.69 2.73 1.77
5 2.58 4.06 5.29 5.59 5.26 4.63 3.91 3.12 2.31 1.62
10 2.18 3.16 4.05 4.39 4.18 3.77 3.27 2.71 2.10 1.47
15 1.79 2.62 3.26 3.53 3.44 3.14 2.77 2.31 1.80 1.33
20 1.54 2.18 2.71 2.91 2.87 2.67 2.37 2.01 1.61 1.20
25 1.32 1.82 2.18 2.35 2.40 2.24 2.04 1.75 1.42 1.10
30 1.16 1.55 1.87 2.02 2.02 1.93 1.77 1.53 1.28 1.00
35 1.01 1.33 1.58 1.72 1.72 1.66 1.53 1.36 1.15 0.91
60 0.91 1.16 1.37 1.48 1.50 1.45 1.34 1.20 1.04 0.83
Source: Matz, 1962.
365
Vis c os it y (c e nt ipois e ) of Corn Syrup
1
DS T ( F)
2
DE=35.4 DE=42.9 DE=53.7 DE=75.4 DE=92.4
85 60 0 0 457000 -
80 7080000 1410000 537000 83200 -
100 1000000 227000 85200 17000 -
120 188000 50100 20000 4270 -
140 44900 13000 6310 1660 -
160 13000 5190 2290 589 -
180 4420 1760 944 275 -
80 60 266000 89100 24000 -
80 126000 59600 17800 4570 -
100 29900 15000 5010 1550 -
120 9810 4840 1800 603 -
140 3350 1860 785 282 -
160 1410 851 367 141 -
180 687 386 196 75.9 -
75 60 39800 18200 7590 6030 -
80 10000 5390 2140 741 501
100 3020 1880 807 331 211
120 1260 817 372 159 106
140 620 389 191 83.2 56.2
160 325 197 103 47.9 32.4
180 180 110 62 28.8 19.5
65 60 1060 560 389 178 -
80 398 237 159 77.6 56.2
100 182 119 83.2 45.7 30.7
120 108 69.2 47.3 26.3 18.6
140 67.9 43.4 29.0 16.2 12.2
160 43.2 26.6 18.6 10.7 8.41
180 29.0 17.8 12.6 7.76 5.96
50 60 54.6 34.0 33.5 18.6 -
80 30.0 19.5 18.4 11.8 9.66
100 18.5 12.0 11.6 7.94 6.31
120 12.9 8.51 7.71 5.50 4.39
140 9.44 6.31 5.43 3.24 3.20
160 6.92 4.49 4.03 3.02 2.50
180 5.27 3.55 3.16 2.19 1.97
35 60 9.16 7.16 7.5 5.13 -
80 6.31 4.75 4.95 3.22 3.35
100 4.52 3.29 3.35 2.74 2.41
120 3.35 2.51 2.51 2.07 1.78
140 2.66 1.99 1.88 1.57 1.41
160 2.20 1.59 1.56 1.27 1.12
180 1.68 1.33 1.26 1.00 0.944
20 60 2.90 2.82 2.63 2.26 -
80 2.24 2.04 1.94 1.70 1.72
100 1.78 1.45 1.45 1.33 1.27
120 1.35 1.14 1.12 1.06 1.00
140 1.12 0.922 0.908 0.832 0.804
160 1.00 0.759 0.794 0.692 0.689
180 0.773 0.673 0.668 0.562 0.576
1
DS = Per cen t Dr y Su bs t a n ce;
2
DE = Dext r os e Equ iva len t
Sou r ce: Er icks on et a l., 1966.

366 Appendices
6 . 9 . Dairy, Fis h and Me at Produc t s
Product T
( C) (-) (Pa s
n
) (Pa) (s
-1
)
Cream, 10% fat 40 1.0 .00148 - -
60 1.0 .00107 - -
80 1.0 .00083 - -
Cream, 20% fat 40 1.0 .00238 - -
60 1.0 .00171 - -
80 1.0 .00129 - -
Cream, 30% fat 40 1.0 .00395 - -
60 1.0 .00289 - -
80 1.0 .00220 - -
Cream, 40% fat 40 1.0 .00690 - -
60 1.0 .00510 - -
80 1.0 .00395 - -
Minced fish paste 3-6 .91 8.55 1600.0 67-238
Raw, Meat Batters
15
1
13
2
68.8
3
15 .156 639.3 1.53 300-500
18.7 12.9 65.9 15 .104 858.0 .28 300-500
22.5 12.1 63.2 15 .209 429.5 0 300-500
30.0 10.4 57.5 15 .341 160.2 27.8 300-500
33.8 9.5 54.5 15 .390 103.3 17.9 300-500
45.0 6.9 45.9 15 .723 14.0 2.3 300-500
45.0 6.9 45.9 15 .685 17.9 27.6 300-500
67.3 28.9 1.8 15 .205 306.8 0 300-500
Milk, homogenized 20 1.0 .002000 - -
30 1.0 .001500 - -
40 1.0 .001100 - -
50 1.0 .000950 - -
60 1.0 .000775 - -
70 1.0 .00070 - -
80 1.0 .00060 - -
Milk, raw 0 1.0 .00344 - -
5 1.0 .00305 - -
10 1.0 .00264 - -
20 1.0 .00199 - -
25 1.0 .00170 - -
30 1.0 .00149 - -
35 1.0 .00134 - -
40 1.0 .00123 - -
1
%Fat;
2
%Protein;
3
%MC
Source: Steffe et al., 1986.
n K
o

367
6 . 1 0 . Oils and Mis c e llane ous Produc t s
Product % Total T
Solids ( C) (-) (Pa s
n
) (Pa) (s
-1
)
Chocolate, melted - 46.1 .574 .57 1.16 -
Honey - -
Buckwheat 18.6 24.8 1.0 3.86 - -
Golden Rod 19.4 24.3 1.0 2.93 - -
Sage 18.6 25.9 1.0 8.88 - -
Sweet Clover 17.0 24.7 1.0 7.20 - -
White Clover 18.2 25.2 1.0 4.80 - -
Mayonnaise - 25 .55 6.4 - 30-1300
- 25 .60 4.2 - 40-1100
Mustard - 25 .39 18.5 - 30-1300
- 25 .34 27.0 - 40-1100
Oils
Castor - 10 1.0 2.42 - -
- 30 1.0 .451 - -
- 40 1.0 .231 - -
- 100 1.0 .0169 - -
Corn - 38 1.0 .0317 - -
- 25 1.0 .0565 - -
Cottonseed - 20 1.0 .0704 - -
- 38 1.0 .0306 - -
Linseed - 50 1.0 .0176 - -
- 90 1.0 .0071 - -
Olive - 10 1.0 .1380 - -
- 40 1.0 .0363 - -
- 70 1.0 .0124 - -
Peanut - 25.5 1.0 .0656 - -
- 38.0 1.0 .0251 - -
- 21.1 1.0 .0647 - .32-64
- 37.8 1.0 .0387 - .32-64
- 54.4 1.0 .0268 - .32-64
Rapeseed - 0.0 1.0 2.530 - -
- 20.0 1.0 .163 - -
- 30.0 1.0 .096 - -
Safflower - 38.0 1.0 .0286 - -
- 25.0 1.0 .0522 - -
Sesame - 38.0 1.0 .0324 - -
Soybean - 30.0 1.0 .0406 - -
- 50.0 1.0 .0206 - -
- 90.0 1.0 .0078 - -
Sunflower - 38.0 1.0 .0311 - -
Source: Steffe et al., 1986.
n K
o

368 Appendices
6 . 1 1 . Fruit and Ve ge t able Produc t s
Product % Total T
Solids ( C) (-) (Pa s
n
) (s
-1
)
Apple
Pulp - 25.0 .084 65.03 -
Sauce 11.6 27 .28 12.7 160-340
11.0 30 .30 11.6 5-50
11.0 82.2 .30 9.0 5-50
Sauce 10.5 26 .45 7.32 .78-1260
9.6 26 .45 5.63 .78-1260
8.5 26 .44 4.18 .78-1260
Apricots
Puree 17.7 26.6 .29 5.4 -
23.4 26.6 .35 11.2 -
41.4 26.6 .35 54.0 -
44.3 26.6 .37 56.0 .5-80
51.4 26.6 .36 108.0 .5-80
55.2 26.6 .34 152.0 .5-80
59.3 26.6 .32 300.0 .5-80
Reliable,
Conc., green 27.0 4.4 .25 170.0 3.3-137
27.0 25 .22 141.0 3.3-137
" ripe 24.1 4.4 .25 67.0 3.3-137
24.1 25 .22 54.0 3.3-137
" ripened 25.6 4.4 .24 85.0 3.3-137
25.6 25 .26 71.0 3.3-137
" overripe 26.0 4.4 .27 90.0 3.3-137
26.0 25 .30 67.0 3.3-137
Banana
Puree A - 23.8 .458 6.5 -
Puree B - 23.8 .333 10.7 -
Puree - 22 .283 107.3 28-200
(17.7 Brix)
Blueberry
Pie Filling - 20 .426 6.08 3.3-530
Carrot
Puree - 25 .228 24.16 -
Green Bean
Puree - 25 .246 16.91 -
Guava
Puree - 23.4 .494 39.98 15-400
(10.3 Brix)
Mango
Puree - 24.2 .334 20.58 15-1000
(9.3 Brix)
n K

369
Orange Juice
Concentrate
Hamlin, early - 25 .585 4.121 0-500
42-5 Brix - 15 .602 5.973 0-500
- 0 .676 9.157 0-500
- -10 .705 14.25 0-500
Hamlin, late - 25 .725 1.930 0.500
41.1 Brix - 15 .560 8.118 0.500
- 0 .620 1.754 0-500
- -10 .708 13.87 0-500
Pineapple, - 25 .643 2.613 0-500
early
40.3 Brix - 15 .587 5.887 0-500
- 0 .681 8.938 0-500
- -10 .713 12.18 0-500
Pineapple, late - 25 .532 8.564 0-500
41.8 Brix - 15 .538 13.43 0-500
- 0 .636 18.58 0-500
- -10 .629 36.41 0-500
Valencia, early - 25 .583 5.059 0-500
43.0 Brix - 15 .609 6.714 0-500
- -10 .619 27.16 0-500
Valencia, late - 25 .538 8.417 0-500
- 15 .568 11.80 0-500
41.9 Brix - 0 .644 18.75 0-500
- -10 .628 41.41 0-500
Naval
65.1 Brix - -18.5 .71 29.2 -
- -14.1 .76 14.6 -
- -9.3 .74 10.8 -
- -5.0 .72 7.9 -
- -0.7 .71 5.9 -
- 10.1 .73 2.7 -
- 19.9 .72 1.6 -
- 29.5 .74 .9 -
Papaya
Puree - 26.0 .528 9.09 20-450
(7.3 Brix)
Peach
Pie Filling - 20.0 .46 20.22 1.-140
Puree 10.9 26.6 .44 .94 -
17.0 26.6 .55 1.38 -
21.9 26.6 .55 2.11 -
26.0 26.6 .40 13.4 80-1000
29.6 26.6 .40 18.0 80-1000
37.5 26.6 .38 44.0 -
40.1 26.6 .35 58.5 2-300
49.8 26.6 .34 85.5 2-300
58.4 26.6 .34 440.0 -
Puree 11.7 30.0 .28 7.2 5-50
370 Appendices
11.7 82.2 .27 5.8 5-50
10.0 27.0 .34 4.5 160-3200
Pear
Puree 15.2 26.6 .35 4.25 -
24.3 26.6 .39 5.75 -
33.4 26.6 .38 38.5 80-1000
37.6 26.6 .38 49.7 -
39.5 26.6 .38 64.8 2-300
47.6 26.6 .33 120.0 .5-1000
49.3 26.6 .34 170.0 -
51.3 26.6 .34 205.0 -
45.8 32.2 .479 35.5 -
45.8 48.8 .477 26.0 -
45.8 65.5 .484 20.0 -
45.8 82.2 .481 16.0 -
14.0 30.0 .35 5.6 5-50
14.0 82.2 .35 4.6 5-50
Plum
Puree 14.0 30.0 .34 2.2 5-50
14.0 82.2 .34 2.0 5-50
Squash
Puree A - 25 .149 20.65 -
Puree B - 25 .281 11.42 -
Tomato
Juice Conc. 5.8 32.2 .59 .223 500-800
5.8 38.8 .54 .27 500-800
5.8 65.5 .47 .37 500-800
12.8 32.2 .43 2.0 500-800
12.8 48.8 .43 2.28 500-800
12.8 65.5 .34 2.28 500-800
12.8 82.2 .35 2.12 500-800
16.0 32.2 .45 3.16 500-800
16.0 48.8 .45 2.77 500-800
16.0 65.5 .40 3.18 500-800
16.0 82.2 .38 3.27 500-800
25.0 32.2 .41 12.9 500-800
25.0 48.8 .42 10.5 500-800
25.0 65.5 .43 8.0 500-800
25.0 82.2 .43 6.1 500-800
30.0 32.2 .40 18.7 500-800
30.0 48.8 .42 15.1 500-800
30.0 65.5 .43 11.7 500-800
30.0 82.2 .45 7.9 500-800
Source: Steffe et al., 1986.
371
6 . 1 2 . Polyme r Me lt s
Polymer T
( C) (-) (Pa s
n
) (Pa s
n
) (s
-1
)
High impa ct 170 0.20 7.58 x 10
4
2.1 x 10
5
100-7000
polys t yr en e 190 0.21 4.57 x 10
4
1.48 x 10
5
100-7000
210 0.19 3.61 x 10
4
1.05 x 10
5
100-7000
Polys t yr en e 190 0.22 4.47 x 10
4
1.4 x 10
4
100-4500
210 0.25 2.38 x 10
4
9.2 x 10
3
100-4000
225 0.28 1.56 x 10
4
6.6 x 10
3
100-5000
St yr en e Acr ylon it r il 190 0.21 9.0 x 10
4
2.2 x 10
4
100-9000
220 0.27 3.22 x 10
4
9.0 x 10
3
100-8000
250 0.35 1.11 x 10
4
4.2 x 10
3
100-8000
Th er mopla s t ic 200 0.27 2.75 x 10
4
3.6 x 10
4
100-5000
olefin 220 0.30 1.83 x 10
4
2.15 x 10
4
100-4000
240 0.28 1.99 x 10
4
1.35 x 10
4
100-3000
Acr ylon it r ile bu t a - 170 0.25 1.19 x 10
5
7.95 x 10
4
100-5500
dien e s t yr en e 190 0.25 6.29 x 10
4
4.4 x 10
4
100-6000
210 0.25 3.93 x 10
4
2.6 x 10
4
100-7000
Polypr opylen e 180 0.37 6.79 x 10
3
4.21 x 10
3
100-4000
190 0.41 4.89 x 10
3
3.02 x 10
3
100-3500
200 0.41 4.35 x 10
3
205 x 10
3
100-4000
Et h ylen e et h yl 170 0.38 1.21 x 10
4
5.4 x 10
3
100-6000
a cr yla t e 190 0.43 6.91 x 10
3
3.5 x 10
3
100-4000
210 0.48 3.77 x 10
3
2.3 x 10
3
100-6000
High den s it y 180 0.56 6.19 x 10
3
2.1 x 10
3
100-1000
polyet h ylen e 200 0.59 4.68 x 10
3
1.52 x 10
3
100-1000
220 0.61 3.73 x 10
3
1.17 x 10
3
100-1000
Low den s it y 160 0.41 9.36 x 10
3
6.3 x 10
3
100-4000
polyet h ylen e 180 0.46 5.21 x 10
3
3.2 x 10
3
100-6500
200 0.47 4.31 x 10
3
1.7 x 10
3
100-6000
Nylon 220 0.63 2.62 x 10
3
1.6 x 10
3
100-2500
230 0.66 1.95 x 10
3
1.3 x 10
3
100-2000
235 0.66 1.81 x 10
3
1.1 x 10
3
100-2300
Polymet h ylmet h a - 220 0.19 8.83 x 10
4
1.3 x 10
3
100-6000
cr yla t e 240 0.25 4.27 x 10
4
6.0 x 10
3
100-6000
260 0.27 2.62 x 10
4
2.9 x 10
3
100-7000
Polyca r bon a t e 280 0.64 8.39 x 10
3
1.54 x 10
3
100-1000
300 0.67 4.31 x 10
3
8.0 x 10
2
100-1000
320 0.80 1.08 x 10
3
4.2 x 10
2
100-1000
Sou r ce: Ta dmor , Z. a n d C.G. Gogos . 1979.
n K
o

372 Appendices
6 . 1 3 . Cos me t ic and Toile t ry Produc t s
Appr oxima t e pa r a met er s , a t 25 C, for t h e a ppa r en t vis cos it y fu n c-
t ion ba s ed on t h e power la w for m of t h e model:
(Cu r ve #): Pr odu ct
( - ) (Pa s ) (1/ s )
1: Sh in e On La s t in g Color Na il En a mel
*
0.86 754 3.0 - 99
2: Cover Gir l Ma r a t h on Ma s ca r a
*
0.24 191 0.1 - 75
3: Mu ppet s Toot h pa s t e, Bu bble Gu m Gel 0.28 117 0.1 - 90
4: J oh n s on s Ba by Su n block Lot ion
*
0.28 75 0.1 - 50
5: Pon ds Cold Cr ea m
*
0.45 25 0.3 - 90
6: Oil of Ola y Bea u t y Flu id 0.22 26 0.1 - 90
7: Hea d & Sh ou lder s Sh a mpoo 1.0 4 0.1 - 6
8: J er gen s Lot ion En r ich ed Liqu id Soa p 1.0 4 0.1 - 6
*
Aver a ge va lu es , or igin a l da t a in dica t e s ome t ime-depen den t beh a vior .
0.1 0.2 0.5 1 2 5 10 20 50 100
0.1
1
10
100
1,000
10,000
Shear Rate, 1/s
A
p
p
a
r
e
n
t

V
i
s
c
o
s
i
t
y
,

P
a

s
1
2
3
4
5
6
7
8
Approximate Temperature = 25 C
Data from Laba, 1993a

= K(

)
n 1
n K

373
Appr oxima t e pa r a met er s , a t 25 C, for t h e a ppa r en t vis cos it y fu n c-
t ion ba s ed on t h e power la w for m of t h e model:
(Cu r ve #): Pr odu ct
( - ) (Pa s ) (1/ s )
1: Colga t e Toot h pa s t e
*
0.22 121.7 3.0 - 700
2: Su a ve Ba ls a m & Pr ot ein Sh a mpoo 0.95 11.1 20.0 - 700
3: Va s elin e In t en s ive Ca r e Lot ion 0.34 13.7 0.09 - 750
4: Mois t u r izin g Su n block Lot ion
*
0.32 9.1 0.05 - 700
5: Ba by Ma gic Ba by Lot ion
*
0.38 6.7 0.02 - 700
6: BAN An t iper s pir a n t / Deodor a n t Roll-on 0.52 3.3 1.5 - 750
7: Dr y Idea Dr y Roll-on Deodor a n t 0.13 9.5 12.0 750
*
Aver a ge va lu es , or igin a l da t a in dica t e s ome t ime-depen den t beh a vior .
0.1 0.3 1 3 10 30 100 300 1,000
0.01
0.1
1
10
100
1,000
Shear Rate, 1/s
A
p
p
a
r
e
n
t

V
i
s
c
o
s
i
t
y
,

P
a

s
1
2
3
4
5
6
7
Approximate Temperature = 25 C
Data from Laba, 1993a

= K(

)
n 1
n K

374 Appendices
6 . 1 4 . Ene rgy of Ac t ivat ion for Flow for Fluid Foods
Flu id food Con cen t r a t ion a t 50 C
1
(-) (kca l/ g (mPa s
n
)
mol)
Depect in ized a pple 75 Br ix 1.0 14.2 150.0
ju ice
a
50 Br ix 1.0 8.4 4.0
30 Br ix 1.0 6.3 1.6
15 Br ix 1.0 5.3 0.7
Clou dy a pple ju ice
a
40 Br ix 1.0 5.8 4.9
30 Br ix 1.0 5.1 2.0
Con cor d gr a pe ju ice
a
50 Br ix 1.0 6.9 15.0
30 Br ix 1.0 6.2 1.8
Clou dy a pple ju ice
a
65.5 Br ix 0.65 9.1 258.5
50.0 Br ix 0.85 6.1 25.0
Apple s a u ce
a
11.0 Br ix 0.30 1.2 730.0
Pea ch pu r ee
a
11.7 Br ix 0.30 1.7 190.0
Pea r pu r ee
a
16.0 Br ix 0.30 1.9 375.0
Filt er ed or a n ge ju ice
b
18.0 Br ix 1.0 5.8 1.5
10 Br ix 1.0 5.3 0.8
Wh ole egg
c
75 Wt .% wa t er 1.0 5.9 3.7
St a bilized egg wh it e
c
88 Wt .% wa t er 1.0 5.9 1.9
Pla in yolk
c
55 Wt .% wa t er 1.0 6.4 48.2
Sa lt ed yolk
c
10 Wt .% s a lt 1.0 8.3 153.0
Su ga r ed yolk
c
10 Wt .% s u ga r 1.0 6.9 32.1
1
= a ppa r en t vis cos it y a t 100 1/ s .
Sou r ce:
a
Ra o (1986);
b
Sa r a va cos (1970);
c
Sca lzo et a l. (1970)
n E
a

375
6 . 1 5 . Ext e ns ional Vis c os it ie s of Ne wt onian Fluids
Trouton Ratio
Liquid (Pa s) (s
-1
) (-)
Silicone Oil 102.5 0.5-10 2.4-3.6
Polybutene 24 1.6-5 2.7-3.3
Polybutene 23 2-10 2.6-3.8
Corn Syrup 25 - 4
Oil 750 0.08-0.14 3.1-3.5
Maltose Sryup 104 2-5 5
Glycerol-water 0.357 4000-9000 1.7-3.4
Glycerol-water 0.4-1.7 40-180 2.7-3.3
Glycerol-water 0.12-0.25 200-4000 2.4-3.9
Viscasil 30 1-30 2.4-3.9
Source: Gupta and Sridhar, 1987.

h
376 Appendices
6 . 1 6 . Ext e ns ional Vis c os it ie s of Non-Ne wt onian Fluids
Solute Solvent Conc. Trouton
(%) (s
-1
) ( ) Ratio
(-)
PAA Glycerol 1.5 0.1-10 2-8E3 -
PAA Water 1.0 8-230 2.4-250 -
PAA Glycerol 1.0 1-10 - 0.1-6.0
PAA Glycerol 0.175-0.5 0.17-1.4 - 20-1000
PAA Glycerol 0.5 0.1-0.2 1-5E4 -
PAA Maltose 0.1 0.5-5.0 - 70
PAA Water 0.1 100-750 - 3000
PAA Water 0.1 40-80 - 250-400
PAA Water 0.01-0.5 33-19000 - 500-29000
PAA Corn Syrup 0.05 - - 300
PAA Glycerol 0.01-0.03 4-17 30-100 -
PAA Glycerol 0.005-0.05 50-1400 1-100 -
PAA Water 0.0051 50-800 0.5-25 -
PEO Sucrose 3.0 1-20 20-500 -
PEO Glycerol 3.0 0.4-2.0 0.4-7.0E6 -
PEO Water 0.1 100-750 8-11.4 -
PEO Water 0.1 40-80 - 1500-2400
PIB Decalin 6.4-11.6 2-100 0.5-7 -
PIB Kerosene 3.0-4.0 3-100 - 1.2-40
PIB Polybutene 0.18 0.4-3.0 - 3-30
XG Glycerol 0.03-0.05 20-40 1-10 -
XG Glycerol 0.005-0.01 50-1400 1-8 -
HPC Water 2.0 0.1-1.0 - 2.0-5.0
HPC Acetic Acid 40.0 0.01-10.0 - 10
HC Jet Fuel 0.4-1.0 3-40 - 80-2000
PU DMF 18-30 - 8-30 -
PAA, polyacrylamide; PEO, polyethylene oxide; PIB, poly-isobutylene; XG,
xanthan gum; HPC, hydroxy-propylcellulose; HC, hydrocarbon; PU,
polyurethane; DMF, dimethylformamide.
Source: Gupta and Sridhar, 1987.

h

E
Pa s
3
7
7
6
.
1
7
.

F
a
n
n
i
n
g

F
r
i
c
t
i
o
n

F
a
c
t
o
r
s
:

B
i
n
g
h
a
m

P
l
a
s
t
i
c
s

100 200 300 500 1,000 2,000 3,000 5,000 10,000
0.001
0.002
0.005
0.01
0.02
0.05
0.1
0.2
0.5
1
F
a
n
n
i
n
g

F
r
i
c
t
i
o
n

F
a
c
t
o
r
100 1,000 5,000 10,000 25,000 50,000 100,000 200,000
N
He
N
Re,B
Bingham Plastics
in
Laminar Flow
378 Appendices
6 . 1 8 . Fanning Fric t ion Fac t ors : Powe r Law Fluids

2
,
0
0
0
4
,
0
0
0
6
,
0
0
0
8
,
0
0
0
1
0
,
0
0
0
0
.
0
0
2
0
.
0
0
4
0
.
0
0
6
0
.
0
0
8
0
.
0
1
0
.
0
1
2
0
.
0
1
4
F a n n i n g F r i c t i o n F a c t o r
N
R
e
,
P
L
n

=

1
.
2
n

=

1
.
0
n

=

0
.
8
n

=

0
.
6
n

=

0
.
4
n

=

0
.
3
n

=

0
.
2
P
o
w
e
r

L
a
w

F
l
u
i
d
s

i
n

T
u
r
b
u
l
e
n
t

F
l
o
w
379
6 . 1 9 . Cre e p (Burge rs Mode l) of Salad Dre s s ing
Creamy-style Salad
Dressing at 2.8 C (cm
2
dyne
-1
) (cm
2
dyne
-1
) (s) (P)
Regular, bottled
1
0.000704 0.00100 16.2 79800
Reduced Calorie, bottled
1
0.00182 0.00192 43.2 148800
Regular, bottled
1
0.000490 0.000481 18.5 587400
Reduced Calorie, bottled
1
0.000870 0.00370 26.1 672000
Regular, dry mix
2
0.00161 0.000980 8.92 19800
Regular, dry mix
2
0.00164 0.00208 2.42 3600
1
Con s t a n t Applied Sh ea r St r es s : = 55.2 dyn e cm
2
2
Con s t a n t Applied Sh ea r St r es s : = 22.8 dyn e cm
2
Sou r ce: Pa r edes et a l., 1989
Th e a bove con s t a n t s wer e det er min ed for t h e Bu ger s model (Eq.
[5.22]) wr it t en in t er ms of t h e s h ea r cr eep complia n ce fu n ct ion :
Res u lt s a r e t ypica l of cr ea my-s t yle s a la d dr es s in g pu r ch a s ed in a
bot t le or ma de fr om dr y mix. Cr eep t es t in g ma y be u s efu l in con -
s ider in g t h e s t a bilit y of s a la d dr es s in g.
J
o
J
1

ret

o

o
J = f (t ) = J
0
+ J
1

1 exp

ret

+
t

0
380 Appendices
6 . 2 0 . Os c illat ory Dat a for But t e r
Appr oxima t e ma t h ema t ica l pa r a met er s des cr ibin g t h e a bove cu r ves :
a n d
T
( C) (Pa s
1+b
) ( - ) (Pa s
d
) ( - )
5 1,073,300 -1.0 4,395,900 0.18
10 350,100 -1.0 1,339,500 0.16
15 140,400 -1.0 561,400 0.16
20 35,000 -1.0 151,700 0.17
0.001 0.01 0.1 1 10 100
Frequency, rad/s
T( C)
5
10
15
20
5
10
15
20
storage modulus
dynamic viscosity
10
9
10
8
10
7
10
6
10
5
10
4
10
3
Butter
Data from Rohm and Weidinger, 1993
S
t
o
r
a
g
e

M
o
d
u
l
u
s

(
P
a
)

o
r

D
y
n
a
m
i
c

V
i
s
c
o
s
i
t
y

(
P
a

s
)
= a()
b
G = c()
d
a b c d

381
6 . 2 1 . Os c illat ory Dat a Iot a-Carrage e nan Ge l
Th is figu r e illu s t r a t es t h e effect of t emper a t u r e in t r a n s for min g a
gel in t o a s olu t ion . Iot a -ca r r a geen a n (1%) wa s pr epa r ed by dis -
per s in g t h e gu m in cold wa t er a n d h ea t in g for 15 min . Th e fr equ en cy
s weep wa s con du ct ed a t a s t r a in a mplit u de of 8% u s in g a coa xia l
cylin der geomet r y ( ). A s t r u ct u r ed gel is obs er ved a t 25 C
wh er e t h e s t or a ge modu lu s is on ly s ligh t ly depen den t on fr equ en cy.
Th e s t or a ge modu lu s a n d t h e los s modu lu s a r e a lmos t t h e s a me a t
35 C. Beh a vior t ypica l of a ma cr omolecu la r s olu t ion (t h e s t or a ge
modu lu s is les s t h a n t h e los s modu lu s ) is obs er ved a t 39 C.
0.01 0.03 0.1 0.3 1 3 10 30 100
0.001
0.01
0.1
1
10
100
Frequency, rad/s
S
t
o
r
a
g
e

o
r

L
o
s
s

M
o
d
u
l
u
s
,

P
a
Iota-Carragenan-Water (1%)
Data from Cuvelier et al., 1990
25 C
Storage Modulus
Loss Modulus
35 C
39 C
R
b
/R
c
= 0.96

382 Appendices
6 . 2 2 . St orage and Los s Moduli of Fluid Foods

Product
( C) (Pa s
b
) ( - ) (Pa s
d
) ( - ) (rad/ s)
Mustard
1
fine 25 800.6 0.120 149.5 0.177 0.01 - 100
standard 25 897.5 0.130 185.9 0.171 0.01 - 100
course 25 1160.0 0.135 265.0 0.139 0.01 - 100
Tomato Paste
2
40 8434.0 0.117 2101.0 0.153 5.0 - 55.0
Blueberry Pie 40 278.3 0.17 64.2 0.26 1 - 100
Filling
3
85 237.9 0.13 45.5 0.26 1-100
Cookie Dough
4
RT
*
4.66E7 1.49 1.20E7 1.30 0.01-100
Cracker Dough
4
RT
*
6.50E6 1.24 2.10E6 1.12 0.01-100
Mozzarella
Cheese
5
natural 70 22,700 0.17 10,300 0.19 0.01 - 100
with 1% Ca 70 59,200 0.20 19,800 0.14 0.01 - 100
caseinate
with 2% Ca 70 15,900 0.21 19,800 0.16 0.01 - 100
caseinate
*
RT = Room Temperature
1
from Aguilar et al., 1991.
2
from Rao and Cooley, 1992.
3
from Steffe et al., 1989.
4
from Menjivar and Faridi, 1994.
5
from Nolan et al., 1989.
G = a
b
G = c
d
T a b c d

383
0.01 0.03 0.1 0.3 1 3 10 30 100
Frequency, rad/s
S
t
o
r
a
g
e

M
o
d
u
l
u
s
,

P
a
Mustard, Standard
25 C
Tomato Paste
40 C
Blueberry Pie Filling
40 C
Mozarella Cheese, Natural
70 C
Cookie Dough
Room Temperature
Cracker Dough
Room Temperature
1E+13
1E+11
1E+09
1E+07
1E+05
1E+03
1E+01
384 Appendices
0.01 0.03 0.1 0.3 1 3 10 30 100
Frequency, rad/s
L
o
s
s

M
o
d
u
l
u
s
,

P
a
Mustard, Standard
25 C
Tomato Paste
40 C
Blueberry Pie Filling
40 C
Mozarella Cheese, Natural
70 C
Cookie Dough
Room Temperature
Cracker Dough
Room Temperature
1E+12
1E+10
1E+08
1E+06
1E+04
1E+02
1E+00
Nome nc lat ure
shift fact or , dimensionless
ar ea, m
2
init ial sample ar ea, m
2
height of mixer blade, m
bulk compr ession cr eep compliance, Pa
-1
bulk compr ession complex compliance, Pa
-1
bulk compr ession st or age compliance, Pa
-1
bulk compr ession loss compliance, Pa
-1
, dimensionless
specific heat , J kg
-1
K
-1
mass concent r at ion, g/dl or g/100 ml
impeller blade or vane diamet er , m
equivalent diamet er , m
diamet er , m
t ensile cr eep compliance, Pa
-1
t ensile complex compliance, Pa
-1
t ensile st or age compliance, Pa
-1
t ensile loss compliance, Pa
-1
diamet er of ext r udat e, m
Youngs modulus or modulus of elast icit y, Pa
t ensile r elaxat ion modulus, Pa
ener gy of act ivat ion for flow, cal/g mole
t ensile complex modulus, Pa
t ensile st or age modulus, Pa
t ensile loss modulus, Pa
Fanning fr ict ion fact or , dimensionless
for ce, N
acceler at ion due t o gr avit y, 9.81 m s
-2
shear modulus, Pa
shear r elaxat ion modulus, Pa
shear complex modulus, Pa
a
T
A
A
o
b
B
B
*
B
B
c
o
/
w
c
p
C
d
d
e
D
D
D
*
D
D
D
e
E
E
E
a
E
*
E
E
f
F
g
G
G
G
*
386 Nomenclature
shear st or age modulus, Pa
shear loss modulus, Pa
height , m
asympot ic or r esidual t hickness, m
effect ive height , m
init ial sample height , m
shear cr eep compliance, Pa
-1
inst ant aneous compliance, Pa
-1
r et ar ded compliance, Pa
-1
shear complex compliance, Pa
-1
shear st or age compliance, Pa
-1
shear loss compliance, Pa
-1
falling ball viscomet er const ant [ ], m
2
s
-2
glass capillar y viscomet er const ant [ ], m
2
s
-2
t her mal conduct ivit y, W m
-1
K
-1
fr ict ion loss coefficient , dimensionless
mixer viscomet er const ant , r ad
-1
bulk modulus, Pa
consist ency coefficient , Pa s
n
bulk compr ession r elaxat ion modulus, Pa
bulk compr ession complex modulus, Pa
bulk compr ession st or age modulus, Pa
bulk compr ession loss modulus, Pa
ext ensional consist ency coefficient , Pa s
m
t emper at ur e dependent const ant
t emper at ur e and concent r at ion dependent const ant
shear r at e, t emper at ur e and conc. dependent const ant
var iable lengt h, m
lengt h, m
init ial lengt h or lengt h of undefor med sample, m
ext ensional flow behavior index, dimensionless
t or que, N m
G
G
h
h
a
h
o
h
o
J
J
0
J
1
J
*
J
J
2R
2
g/(9L) k
ghR
4
/(8LV) k
k
k
f
k
K
K
K
K
*
K
K
K
E
K
T
K
T, C
K
, T, C
l
L
L
o
m
M
387
t or que t o over come yield st r ess, N m
end effect t or que on a bob, N m
flow behavior index, dimensionless
aver age flow behavior index, dimensionless
fir st nor mal st r ess differ ence [ ], Pa
second nor mal st r ess differ ence [ ], Pa
Boussinesq number , dimensionless
Debor ah number , dimensionless
Fr oude number [ ], dimensionless
Hedst r om number [ ], dimensionless
power number [ ], dimensionless
Newt onian fluid Reynolds number [ ], dimensionless
Bingham fluid Reynolds number [ ], dimensionless
impeller Reynolds number [ ], dimensionless
Power law fluid Reynolds number , dimensionless
Tr out on r at io, dimensionless
Weber number [ ], dimensionless
Weissenber g number [ ], dimensionless
power input t o a mixer [ ], N m s
-1
pr essur e, Pa
st at ic pr essur e (absolut e) in t he undist ur bed flow, Pa
exit pr essur e (absolut e) of a slit viscomet er , Pa
exit pr essur e of a slit viscomet er , Pa
liquid vapor pr essur e (absolut e), Pa
volumet r ic flow r at e in a pipe, m
3
s
-1
measur ed volumet r ic flow r at e, m
3
s
-1
volumet r ic flow r at e wit hout slip, m
3
s
-1
r adial coor dinat e, m
univer sal gas const ant , 1.987 cal/(g-mole K)
r adius, m
M
o
M
e
n
n
N
1

11

22
N
2

22

33
N
Bo
N
De

2
d/g N
Fr
D
2

o
/(
pl
)
2
N
He
P/(d
5

3
) N
Po
N
Re
Du/
N
Re, B
Du/
pl
d
2
/ N
Re, I
N
Re, PL
N
Tr
N
2
d
3
/
st
N
We
N
Wi

1
/
P M
P
P
atm
P
ex
P
ex
P
vap
Q
Q
m
Q
ws
r
R
R
388 Nomenclature
bob r adius, m
r adius of bar r el in conver ging die, m
cup r adius, m
r adius of undefor med cylinder , m
init ial r adius of sample, m
cr it ical r adius, m
r adius of shaft , m
r adius of t r uncat ed por t ion of cone, m
t ime, s
t emper at ur e, K or C
r efer ence t emper at ur e, K or C
velocit y, m s
-1
t ur bulent velocit y [ ], dimensionless
fr ict ion velocit y [ ], m s
-1
velocit y in t he and dir ect ions, m s
-1
velocit y in t he , and dir ect ions, m s
-1
effect ive slip velocit y, m s
-1
t er minal velocit y, m s
-1
volumet r ic aver age velocit y [ ], m s
-1
volume of mixing vessel, m
3
volume of a glass capillar y viscomet er bulb, m
3
widt h of slit or blade, m
wor k out put per unit mass, J kg
-1
const ant for ce, N
Car t esian coor dinat es, m
pipe lengt h r equir ed for fully developed flow, m
dist ance fr om pipe wall int o fluid [ ], m
dist ance fr om t he t ube wall [ ], dimensionless
height above a r efer ence plane, m
dist ance bet ween t wo point s (mixer viscomet er ), m
axial coor dinat e, m
R
b
R
b
R
c
R
o
R
o
R
o
R
s
R
T
t
T
T
r

u
u
+
u/u
*

w
/ = u

f /2 u
*
u
1
, u
2
, u
3
x
1
, x
2
x
3
u
r
, u
z
, u

r z
u
s
u
t
Q/(R
2
) u
V
V
w
W
W
x
1
, x
2
, x
3
X
E
y R r
y
+
u
*
y/
z
z
z
389
, dimensionless
, dimensionless
kinet ic ener gy cor r ect ion fact or , dimensionless
effect ive slip coefficient [ ], m Pa
-1
s
-1
cor r ect ed slip coefficient [ ], m
2
Pa
-1
s
-1
angle of shear , r ad
shear st r ain, dimensionless
amplit ude of st r ain funct ion [ ], dimensionless
const ant shear st r ain, dimensionless
st r ain at t he r im of a par allel plat e, dimensionless
nor malized shear st r ain ( ), dimensionless
shear r at e, s
-1
aver age shear r at e, s
-1
shear r at e at t he bob, s
-1
shear r at e at t he r im of a par allel plat e, s
-1
shear r at e at t he wall, s
-1
appar ent wall shear r at e [ ], s
-1
phase shift or phase angle, r ad
change in height , m
linear displacement , Pa
pr essur e dr op, Pa
ent r ance pr essur e loss, Pa
ent r ance pr essur e loss due t o ext ensional flow, Pa
ent r ance pr essur e loss due t o shear flow, Pa
r adial displacement , Pa
change in t ime, s
Cauchy st r ain [ ], dimensionless
Hencky st r ain [ ], dimensionless
Hencky st r ain r at e, s
-1
const ant Hencky st r ain r at e, s
-1
biaxial ext ensional st r ain r at e, s
-1
R
c
/R
b
d/D

u
s
/
w

c
R

o
=
o
sin(t )

*
/
o

w
4Q/(R
3
)

h
L
P
P
en
P
en, E
P
en, S
R
t

c
(L L
o
)/L
o

h
ln(L/L
o
)

ho

B
390 Nomenclature
const ant biaxial ext ensional st r ain r at e, s
-1
aver age ext ensional st r ain r at e in conver ging flow, s
-1
aver age ext ensional st r ain r at e at t he die, s
-1
appar ent viscosit y, Pa s
biaxial ext ensional viscosit y, Pa s
biaxial gr owt h funct ion, Pa s
equilibr ium appar ent viscosit y, Pa s
t ensile ext ensional viscosit y, Pa s
t ensile gr owt h funct ion, Pa s
inher ent viscosit y, dl g
-1
int r insic viscosit y, dl g
-1
limit ing viscosit y at zer o shear r at e, Pa s
planar ext ensional viscosit y, Pa s
r efer ence appar ent viscosit y, Pa s
r educed viscosit y, dl g
-1
r elat ive viscosit y, dimensionless
specific viscosit y, dimensionless
limit ing viscosit y at infinit e shear r at e, Pa s
complex viscosit y, Pa s
dynamic viscosit y, Pa s
out of phase component of , Pa s
angle, r ad
apex angle of cone at bot t om of bob, degr ees
half angle of conver ging die, degr ees or r ad
cir cumfer ent ial coor dinat e, r ad
st r uct ur al par amet er , dimensionless
t ime const ant in Bir d-Leider equat ion, dimensionless
equilibr ium value of st r uct ur al par amet er , , dimensionless
r elaxat ion t ime, s
r et ar dat ion t ime, s
Newt onian viscosit y, Pa s

Bo

E, R

B
+

E
+

inh

int

red

rel

sp

rel

ret

391
cor r ect ed viscosit y, Pa s
plast ic viscosit y of a Bingham fluid, Pa s
r efer ence Newt onian viscosit y, Pa s
kinemat ic viscosit y ( ), m
2
s
-1
Poissons r at io, dimensionless
densit y, kg m
-3
liquid densit y, kg m
-3
spher e densit y, kg m
-3
shear st r ess, Pa
aver age shear st r ess, Pa
shear st r ess at bob, Pa
biaxial (r adial) st r et ching st r ess, Pa
shear st r ess at cup, Pa
shear st r ess on ends of vane, Pa
equilibr ium shear st r ess, Pa
t ensile st r et ching st r ess, Pa
const ant t ensile st r et ching st r ess, Pa
st r ess on plane per pendicular t o i in t he dir ect ion of j, Pa
yield st r ess, Pa
init ial shear st r ess, Pa
const ant shear st r ess, Pa
amplit ude of t he st r ess funct ion [ ], Pa
shear st r ess at t he r im of a par allel plat e, Pa
sur face t ension, N m
-1
shear st r ess at t he wall of t ube or slit , Pa
nor malized shear st r ess ( ), dimensionless
shear st r ess gr owt h funct ion, Pa
die exit effect int egr al, dimensionless
sweep angle, r ad
fir st nor mal st r ess coefficient , Pa s
2
second nor mal st r ess coefficient , Pa s
2

pl

r
/

Eo

ij

o
=
o
sin(t )

st

*
/
o

21
+

2
392 Nomenclature
angular velocit y at r , r ad s
-1
angular velocit y [2 (r pm)/60], r ad s
-1


Bibliography
Aguilar , C., S.S.H. Rizva, J .F. Ramir ez and A. Inda. 1991. Rheological
behavior of pr ocessed must ar d. I: effect of milling t r eat ment . J .
Text ur e St ud. 22: 59-84.
Ak, M.M. and S. Gunasekar an. 1995. Measur ing elongat ional pr oper t ies
of Mozzar ella cheese. J . Text ur e St ud. 26: 147-160.
Alt omar e, R.E., M. Anelich and R. Rakos. 1992. An exper iment al
invest igat ion of t he r heology of r ice flour dough wit h an ext r uder -
coupled slit die r heomet er . In: Kokini, J .L., C-T. Ho and M.V. Kar we
(edit or s). Food Ext r usion Science and Technology. Mar cel Dekker ,
Inc., New Yor k. pg. 233-254.
Bagley, E.B. 1957. End cor r ect ions in t he capillar y flow of polyet hylene.
J . Appl. Phys. 28: 624-627.
Bagley, E.B. and D.D. Chr ist ianson. 1986. Response of chemically
leavened dough t o uniaxial compr ession. In: Far idi, H. and J .M.
Faubion (edit or s). Fundament als of Dough Rheology. Amer . Assoc.
of Cer eal Chemist s, St . Paul, MN.
Bagley, E.B. and D.D. Chr ist ianson. 1988. Behavior of foods in lar ge
defor mat ions. In: Blanshar d, J .M.V. and J .R. Mit chel (edit or s). Food
St r uct ur e - It s Cr eat ion and Evaluat ion. But t er wor t hs, London. pg
401-416.
Bagley, E.B., D.D. Chr ist ianson and W.J . Wolf. 1985. Fr ict ional effect s
in compr essional defor mat ion of gelat in and st ar ch gels and com-
par ison of mat er ial r esponse in simple shear , t or sion, and lubr icat ed
compr ession. J . Rheol. 29: 103-108.
Balaban, M., A.R. Car r illo and J .L. Kokini. 1988. A comput er ized
met hod t o analyze t he cr eep behavior of viscoelast ic foods. J . Text ur e
St ud. 19: 171-183.
Baner jee, T.K., D. Manas and S.K. Das. 1994. Non-Newt onian liquid
flow t hr ough globe and gat e valves. Can. J . Chem. Engn. 72: 207-211.
Bar bosa Canovas, G.V. and M. Peleg. 1983. Flow par amet er s of select ed
commer cial semi-liquid food pr oduct s. J . Text ur e St ud. 14: 213-234.
Bar nes, H.A. 1992. The yield st r ess myt h? r evisit ed. In: Moldenaer s,
P. and R. Keunings (edit or s). Theor et ical and Applied Rheology,
Volume 2. Pr oc. XIt h Int . Congr . on Rheology, Br ussels, Belgium,
Aug. 17-21, 1992. Elsevier Science Publisher s, New Yor k. pg.
576-578.
Bar nes, H.A., J .F. Hut t on and K. Walt er s. 1989. An Int r oduct ion t o
Rheology. Elsevier Science Publishing Company Inc., New Yor k.
Bar nes, H.A. and K. Walt er s. 1985. The yield st r ess myt h. Rheol. Act a
24: 323-326.
Ber st ed, B.H. 1993. Refinement of t he conver ging flow met hod of
measur ing ext ensional viscosit y in polymer s. Polymer Engn. Sci.
33(16): 1079-1083.
Bhat t achar ya, M. 1993. Slit r heomet er st udies of wheat flour dough. J .
Text ur e St ud. 24:391-409.
394 Bibliography
Bhat t achar ya, M., and M. Padmanabhan. 1992. On-line r heological
measur ement s of food dough dur ing ext r usion cooking. In: Kokini,
J .L., C-T. Ho and M.V. Kar we (edit or s). Food Ext r usion Science and
Technology. Mar cel Dekker , Inc., New Yor k. pg. 213-232.
Bhat t achar ya, M., M. Padmanabhan and K. Seet hamr aju. 1994. Uni-
axial ext ensional viscosit y dur ing ext r usion cooking fr om ent r ance
pr essur e dr op met hod. J . Food Sci. 59: 221-226, 230.
Binding, D.M. 1988. A appr oximat e analysis for cont r act ion and con-
ver ging flows. J . Non-Newt onian Fluid Mech. 27: 173-189.
Bir d, R.B., R.C. Ar mst r ong and O. Hassager . 1987. Dynamics of Poly-
mer ic Liquids, Volume 1. J ohn Wiley and Sons, New Yor k.
Bir d, R.B., W.E. St ewar t and E.N. Light foot . 1960. Tr anspor t Phe-
nomena. J ohn Wiley and Sons, Inc., New Yor k.
Bist any, K.L. and J .L. Kokini. 1983. Compar ison of st eady shear r he-
ological pr oper t ies and small amplit ude dynamic viscoelast ic pr op-
er t ies of fluid food mat er ials. J . Text ur e St ud. 14: 113-124.
Block, H. and J .P. Kelly. 1986. Elect r o-r heology. J . Phys. D: Appl. Phys.
21: 1661-1677.
Boger , D.V. 1982. Cir cular ent r y flows of inelast ic and viscoelast ic fluids.
In: Mujumdar , A.S. and R.A. Mashelkar (edit or s). Advances in
Tr anspor t Pr ocesses, Vol. II. Halst ed Pr ess. J ohn Wiley and Sons,
New Yor k. pg. 43-104.
Boger , D.V. and K. Walt er s. 1993. Rheological Phenomena in Focus.
Elsevier Science Publisher s, New Yor k.
Bogue, D.C. and J .L. Whit e. 1970. Engineer ing Analysis of Non-
Newt onian Fluids. AGARDogr aph No. 144. Nor t h At lant ic Tr eat y
Or ganizat ion. Advisor Gr oup for Aer ospace Resear ch &
Development . Et at Major For ces Ar iennes, Caser ne Pr ince Bau-
douin, Place Dailly, Br uxelles, 3, Belgium.
Bour ne, M.C. 1968. Text ur e pr ofile of r ipening pear s. J . Food Sci. 33:
223-226.
Bour ne, M.C. 1974. Text ur e changes in r ipening peaches. J . Can Inst .
Food Sci. Technol. 7: 11-15.
Bour ne, M.C. 1978. Text ur e pr ofile analysis. Food Technol. 32(7): 62-66,
72.
Bour ne, M.C. 1982. Food Text ur e and Viscosit y: Concept and Mea-
sur ement . Academic Pr ess, New Yor k.
Bour ne, M.C., J .F. Kenny and J . Bar nar d. 1978. Comput er -assist ed
r eadout of dat a fr om t ext ur e pr ofile analysis cur ves. J . Text ur e
St udies 9: 481-494.
Bowen, R.L. 1986. Unr aveling t he myst er ies of shear -sensit ive mixing
syst ems. Chem. Engn. J une 9: 55-63.
Br ennan, J .G. 1980. Food t ext ur e measur ement . In: King, R.D. (edit or ).
Development s in Food Analysis Techniques - 2. Applied Sci. Pub.
Lt d., London. pg. 1-78.
Br iggs, J .L. 1995. Met hods t o Char act er ize Power Fluids Using a
Br ookfield Viscomet er . MS Thesis. Dept . of Agr icult ur al Engi-
neer ing. Michigan St at e Univer sit y.
395
Br iggs, J .L. and J .F. St effe. 1995. Kinet ic ener gy cor r ect ion fact or of a
Her schel-Bulkley fluid. J . Food Pr oc. Engn. 18: 115-118.
Br iggs, J .L. and J .F. St effe. 1996. Vane met hod t o evaluat e t he yield
st r ess of fr ozen ice cr eam. J . Dair y Sci. 79: 527-531.
But ler , F. and P. McNult y. 1995. Time-dependent r heological char ac-
t er izat ion of but t er milk at 5 C. J . Food Pr oc. Engn. 25: 569-580.
Br odkey, R.S. and H.C. Her shey. 1988. Tr anspor t Phenomena.
McGr aw-Hill Book Company. New Yor k.
Callist er , W.D. 1991. Mat er ial Science and Engineer ing (Second Edi-
t ion). J ohn Wiley & Sons, Inc., New Yor k.
Campanella, O.H. and M. Peleg. 1987a. Det er minat ion of t he yield st r ess
of semi-liquid foods fr om squeezing flow dat a. J . Food Sci. 52: 214-217.
Campanella, O.H. and M. Peleg. 1987b. Analysis of t r ansient flow of
mayonnaise in a coaxial cylinder viscomet er . J . Rheol. 31: 439-452.
Campanella, O.H. and M. Peleg. 1987c. Squeezing flow viscosimet r y of
peanut but t er . J . Food Sci. 52: 180-184.
Campanella, O.H., L.M. Popplewell, J .R. Rosenau and M. Peleg. 1987.
Elongat ional viscosit y measur ement s of melt ing Amer ican pr ocess
cheese. J . Food Sci. 52: 1249-1251.
Campos, D.T., J .F. St effe and R.Y. Ofoli. 1994. St at ist ical met hod t o
evaluat e t he cr it ical Reynolds number for pseudoplast ic fluids in
t ubes. J . Food Engn. 23: 21-32.
Car r eau, P.J . 1968. Ph.D. Disser t at ion. Dept . of Chemical Engineer ing,
Univer sit y of Wisconsin, Madison.
Casir aghi, E.M., E.B. Bagley and D.D. Chr ist ianson. 1985. Behavior of
Mozzar ella, Cheddar and pr ocess cheese spr ead in lubr icat ed and
bonded uniaxial compr ession. J . Text ur e St ud. 16: 281-301.
Casson, N. 1959. A flow equat ion for pigment ed-oil suspension of t he
pr int ing ink t ype. In: Mill, C.C. (edit or ). Rheology of Disper sed
Syst ems. Per gamon Pr ess, New Yor k. pg. 84-104.
Cast aldo, D., L. Palmier i, A. Lo Voi and P. Cost abile. 1990. Flow
pr oper t ies of Babaco (Car ica Pent agona) pur ees and concent r at es. J .
Text ur e St ud. 21: 253-264.
Cast ell-Per ez, M.E., R.G. Mor eir a and J .F. St effe. 1993. Mt odo sim-
plificado par a car act er izar fluidos que siguen la ley de pot encia con
un viscosmet r o de mezcla t ipo bander a. Revist a Espaola de Ciencia
y Tecnologa de Aliment os 33: 529-547.
Cast ell-Per ez, M.E. and J .F. St effe. 1990. Evaluat ing shear r at es for
power law fluids in mixer viscomet r y. J . Text ur e St ud. 21: 439-453.
Cast ell-Per ez, E.M. and J .F. St effe. 1992. Using mixing t o evaluat e
r heological pr oper t ies. In: Rao, A.M. and J .F. St effe (edit or s). Vis-
coelast ic Pr oper t ies of Food. Elsevier Applied Sci. Pub., Lt d., Bar king,
England. pg. 247-284.
Cast ell-Per ez, M.E., J .F. St effe and R.G. Mor eir a. 1991. Simple
det er minat ion of power law flow cur ves using a paddle t ype mixer
viscomet er . J . Text ur e St ud. 22: 303-316.
Cavella, S., S. Chemin and P. Masi. 1992. Object ive measur ement of
t he st r et chabilit y of Mozzar ella cheese. J . Text ur e St ud. 23: 185-194.

396 Bibliography
Char m, S.E. 1962. Det er minat ion of shear st r ess shear r at e behavior in
foods in t he pr esence of a yield st r ess. J . Food Sci. 28: 107-113.
Chat r aei, SH., C.W. Macosko. 1981. Lubr icat ed squeezing flow: a new
biaxial ext ensional r heomet er . J . Rheol. 25: 433-443.
Cheng, D.C-H. 1970. A design pr ocedur e for pipeline flow of non-
Newt onian disper sed syst ems. Pr oceedings of t he Fir st Int er nat ional
Confer ence in t he Hydr aulic Tr anspor t of Solids in Pipes
(Hydr ot r anspor t I), Sept . 1-4. Br it ish Hydr omechanics Resear ch
Associat ion, Cr anfield, Bedfor d, England. pg. J 5-77 t o J 5-95.
Cheng, D.C-H. 1973. A differ ent ial for m of const it ut ive r elat ion for
t hixot r opy. Rheol. Act a 12: 228-233.
Cheng, D.C-H. 1986. Yield st r ess: a t ime-dependent pr oper t y and how
t o measur e it . Rheol. Act a 25: 542-554.
Cheng, D.C-H. and B.R. Par ker . 1976. The det er minat ion of wall-slip
velocit y in t he coaxial cylinder viscomet er . Pr oceedings of t he
Sevent h Int er nat ional Congr ess on Rheology. Got enber g, Sweden.
pg. 518-519.
Chhabr a, R.P. 1992. Bubbles, Dr ops and Par t icles in Non-Newt onian
Fluids. CRC Pr ess, Boca Rat on, FL.
Chr ist ensen, C.M. 1987. Per cept ion of solut ion viscosit y. In: Moskowit z,
H.R. (edit or ). Food Text ur e. Mar cel Dekker , New Yor k.
Chr ist ianson, D.D., Casir aghi, E.M. and E.B. Bagley. 1985. Uniaxial
compr ession of bonded and lubr icat ed gels. J . Rheol. 29: 671-684.
Chuma, Y., T. Shiga and M. Iwamot o. 1978. Mechanical pr oper t ies of
Sat suma or ange as r elat ed t o t he design of a cont ainer for bulk
shipment . J . Text ur e St ud. 9: 461-479.
Chur chill, S.W. 1988. Viscous Flows: The Pr act ical Use of Theor y.
But t er wor t hs, Bost on.
Clapp, R.M. 1961. Tur bulent heat t r ansfer in pseudoplast ic non-
Newt onian fluids. Int er nat ional Development in Heat t r ansfer .
ASME, Par t III, Sec. A., pg. 652-661.
Cogswell, F.N. 1972. Conver ging flow of polymer melt s in ext r usion
dies. Poly. Engn. and Sci. 12(1): 64-73.
Cogswell, F.N. 1978. Conver ging flow and st r et ching flow: a compila-
t ion. J . Non-Newt onian Fluid Mech. 4: 23-38.
Cogswell, F.N. 1981. Polymer Melt Rheology. Halst ed Pr ess, New Yor k.
Collins, M. and W.R. Schowalt er . 1963. Behavior of non-Newt onian
fluids in t he ent r y r egion of a pipe. AIChE J . 9: 804-809.
Covey, G.H. and B.R. St anmor e. 1981. Use of t he par allel plat e plas-
t omet er for t he char act er izat ion of viscous fluids wit h a yield st r ess.
J . Non-Newt onian Fluid Mech. 8: 249-260.
Cox, W.P. and E.H. Mer z. 1958. Cor r elat ion of dynamic and st eady flow
viscosit ies. J . Polymer Sci. 28: 619-622.
Cr ane Co. 1982. Flow of fluids t hr ough valves, fit t ings and pipe.
Technical Paper No. 410M, t went y-fir st pr int ing. Cr ane Co. 300 Par k
Ave., New Yor k.
Cr ist escu, N. 1989. Rock Rheology. Kluwer Academic Pub., Bost on.
397
Cr oss, M.M. 1965. Rheology of non-Newt onian flow: equat ion for
pseudoplast ic syst ems. J . Colloid Sci. 20: 417-437.
Cuvelier , G., C. Peigney-Nour y and B. Lunay. 1990. Viscoelast ic pr op-
er t ies of physical gels: cr it ical behavior at t he gel point . In: Phillips,
G.O., P.A. Williams and D.J . Wedlock (edit or s). Gums and St abilizer s
for t he Food Indust r y 5, IRL Pr ess, New Yor k.
Dail, R.V. and J .F. St effe. 1990a. Rheological char act er izat ion of
cr osslinked waxy maize st ar ch solut ions under low acid asept ic
pr ocessing condit ions using t ube viscomet r y t echniques. J . Food Sci.
53: 1660-1665.
Dail, R.V. and J .F. St effe. 1990b. Dilat ancy in st ar ch solut ions under
low acid asept ic pr ocessing condit ions. J . Food Sci. 53: 1764-1765.
Dally, J .W. and W.F. Riley. 1965. Exper iment al St r ess Analysis.
McGr aw-Hill Book Company, New Yor k.
DAppolonia, B.L. and W.H. Kuner t h (edit or s). 1984. The Far inogr aph
Handbook, Thir d Edit ion. The Amer ican Associat ion of Cer eal Che-
mist s, St . Paul, MN.
Dar by, R. 1976. Viscoelast ic Fluids: An Int r oduct ion t o Pr oper t ies and
Behavior . Mar cel Dekker , New Yor k.
Dar by, R. 1985. Couet t e viscomet er dat a r educt ion for mat er ials wit h
a yield st r ess. J . Rheol. 29: 369-378.
da Silva, J .A.L., M.P. Gonalves and M.A. Rao. 1994. Influence of
t emper at ur e on t he dynamic and st eady-shear r heology of pect in
disper sions. Car bohydr at e Polymer s 23: 77-87.
da Silva, J .A.L. and M.A. Rao. 1992. Viscoelast ic pr oper t ies of food
hydr ocolloid disper sions. In: Rao, A.M. and J .F. St effe (edit or s).
Viscoelast ic Pr oper t ies of Food. Elsevier Applied Sci. Pub., Lt d.,
Bar king, England. pg. 247-284.
Dauber t , C.R. and J .F. St effe. 1995. Elect r or heology of milk chocolat e.
J . Text ur e St ud. 27: 93-108.
Dealy, J .M. 1982. Rheomet er s for Molt en Plast ics. Van Nost r and
Reinhold Company. New Yor k.
Dealy, J .M. 1994. Official nomenclat ur e for mat er ial funct ions
descr ibing t he r esponse t o a viscoelast ic fluid t o var ious shear ing and
ext ensional defor mat ions. J . Rheol. 38: 179-191.
Dealy, J .D. and A.J . Giacomin. 1988. Sliding plat e and sliding cylinder
r heomet er s. In: Collyer , A.A. and D.W. Clegg (edit or s). Rheological
Measur ement s. Elsevier Applied Science, New Yor k. pg. 383-404.
Dealy, J .M. and K.F. Wissbur n. 1990. Melt Rheology and it s Role in
Plast ics Pr ocessing. Van Nost r and Reinhold, New Yor k.
de Br uijne, D.W., J . de Looff and A. van Eulem. 1990. The r heological
pr oper t ies of br ead dough and t heir r elat ion t o baking. In: Car t er ,
R.E. (edit or ). Rheology of Food, Phar maceut ical and Biological
Mat er ials wit h Gener al Rheology. Elsevier Applied Science, New
Yor k.
De Kee, D., G. Tur cot t e and K. Fildey. 1980. New met hod for t he
det er minat ion of yield st r ess. J . Text ur e St ud. 10: 281-288.
398 Bibliography
Denn, M.M. 1980. Pr ocess Fluid Mechanics. Pr ent ice-Hall, Inc.,
Englewood Cliffs, NJ .
Der visoglu, M. and J .L. Kokini. 1986. St eady shear r heology and fluid
mechanics of four semi-solid foods. J . Food Sci. 51: 541-545, 625.
Dickie, A.M. and J .L. Kokini. 1982. Use of t he Bir d-Leider equat ion in
food r heology. J . Food Pr oc. Engn. 5: 157-174.
Dickie, A.M. and J .L Kokini. 1983. An impr oved model for food t hickness
fr om non-Newt onian fluid mechanics in t he mout h. J . Food Sci. 48:
57-61, 65.
Dienes, G.J . and H.F. Klemm. 1946. Theor y and applicat ion of par allel
plat e plast omet er . J . Appl. Phys. 17: 458-471.
Dint enfass, L. 1985. Blood Viscosit y, Hyper viscosit y & Hyper visco-
saemia. MTP Pr ess Limit ed. Bost on, MA.
Dodge, D.W. and A.B. Met zner . 1959. Tur bulent flow of non-Newt onian
syst ems. AIChE J . 5: 189-204.
Dolan, K.D. and J .F. St effe. 1990. Modeling t he r heological behavior of
gelat inizing st ar ch solut ions using mixer viscomet r y dat a. J . Text ur e
St ud. 21: 265-294.
Dolan, K.D., J .F. St effe and R.G. Mor gan. 1989. Back ext r usion and
simulat ion of viscosit y development dur ing st ar ch gelat inizat ion. J .
Food Pr oc. Engn. 11: 79-101.
Dor aiswamy, D., A.N. Mujumdar , I. Tsao, A.N. Ber is, S.C. Danfor t h and
A.B. Met zner . 1991. The Cox-Mer t z r ule ext ended: A r heological
model for concent r at ed suspensions and ot her mat er ials wit h a yield
st r ess. J . Rheol. 35: 647-685.
Dr ake, B. 1987. Food psychor heology. In: Moskowit z, H.R. (edit or ). Food
Text ur e. Mar cel Dekker , New Yor k.
Dur gueil, E.J . 1987. Det er minat ion of t he consist ency of non-Newt onian
fluids using a Br ookfield HBT viscomet er . Pr oc. Sout h Afr ican Sugar
Technol. Assoc., J une: 32-39.
Edwar ds, D.A., H. Br enner and D.T. Wasan. 1991. Int er facial Tr anspor t
Pr ocesses and Rheology. But t er wor t h-Heinemam, St onheam, MA.
Elliot t , J .H. and A.J . Ganz. 1971. Modificat ion of food char act er ist ics
wit h cellulose hydr ocolloids, I. Rheological char act er izat ion of an
or ganolept ic pr oper t y. J . Text . St ud. 2: 220-229.
Elliot t , J .H. and A.J . Ganz. 1977. Salad Dr essings - pr eliminar y r he-
ological char act er izat ion. J . Text . St ud. 8: 359-371.
Ellis, S.B. 1927. Thesis, Lafayet t e College, Pa. Cit ed in: Whor low, R.W.
1980. Rheological Techniques. Halst ed Pr ess (Div. of J ohn Wiley &
Sons), New Yor k.
Er ickson, E.R., R.A. Ber nt sen and M.A. Eliason. 1966. Viscosit y of cor n
syr up. J . of Chem. and Engn. Dat a 11: 485-488.
Far idi, H. and V.F. Rasper . 1987. The Alveogr aph Handbook. Amer .
Assoc. of Cer eal Chemist s. St . Paul, MN.
Fer guson, J . and Z. Kemblowski. 1991. Applied Fluid Rheology. Elsevier
Applied Science, New Yor k.
Fer r y, J .D. 1977. Oscillat ion viscomet r y - effect s of shear r at e and fr e-
quency. Measur ement & Cont r ol, Sept ember - Oct ober : 89-91.
399
Fer r y, J .D. 1980. Viscoelast ic Pr oper t ies of Polymer s (Thir d Edit ion).
J ohn Wiley and Sons, New Yor k.
For d, E.W. and J .F. St effe. 1986. Quant ifying t hixot r opy in st ar ch-
t hickened, st r ained apr icot s using mixer viscomet r y t echniques. J .
Text ur e St ud. 17: 75-85.
Fr iedman, H.H., J .E. Whit ney and A.S. Szczesnaik. 1963. The Text u-
r omet er - a new inst r ument for t ext ur e measur ement . J . Food Sci.
28: 390-396.
Fuller , G.G., C.A. Cat hey, B. Hubbar d and B.E. Zebr owski. 1987.
Ext ensional viscosit y measur ement for low-viscosit y fluids. J . Rheol.
31: 235-249.
Gar cia, E.J . and J .F. St effe. 1986. Opt imum economic pipe diamet er
for pumping Her schel-Bulkley fluids in laminar flow. J . Food Pr oc.
Engn. 8: 117-136.
Gar cia, E.J . and J .F. St effe. 1987. Compar ison of fr ict ion fact or equa-
t ions for non-Newt onian fluids in t ube flow. J . Food Pr oc. Engn. 9:
93-120.
Gibson, A.G. 1988. Conver ging dies. In: Collyer , A.A. and D.W. Clegg
(edit or s). Rheological Measur ement s. Elsevier Applied Science, New
Yor k. pg. 49-92.
Giesekus, H. and G. Langer . 1977. Die Best immung der wahr en Flie-
kur ven nicht -newt onscher Flssigkeit en und plast ischer St offe mit
der Met hode der r epr sent at iven Viskosit t . Rheol. Act a 16: 1-22.
Gogos, C.G. and S. Bhakuni. 1992. An on-line slit r heomet er for mea-
sur ement of t he r heological pr oper t ies of doughs. In: Kokini, J .L.,
C-T. Ho and M.V. Kar we (edit or s). Food Ext r usion Science and
Technology. Mar cel Dekker , Inc., New Yor k. pg. 255-262.
Goodr ich, K., A. Yoshimur a and R.K. Pr udhomme. 1989. Measur ement
of t he modulus and yield st r engt h of soft gels: exper iment s and
numer ical simulat ion. J . Rheol. 33: 317-327.
Gr iskey, R.G. and R.G. Gr een. 1971. Flow of dilat ant (shear -t hickening)
fluids. AIChE J . 17: 725-728.
Gr ovier , G.W. and K. Aziz. 1972. The Flowof Complex Mixt ur es in Pipes.
R.E. Kr ieger Publishing Co., Malabar , Flor ida.
Gr ulke, E.A. 1994. Polymer Pr ocess Engineer ing. PTR Pr ent ice Hall,
Englewood Cliffs, NJ .
Gupt a, B.P. 1984. Dimensional analysis. In: Hill, H.E. and J .W. Pr ane
(edit or s). Applied St at ist ical Techniques for Select ed Indust r ies.
J ohn Wiley and Sons, New Yor k. pg. 551-581.
Gupt a, R.K. and T. Sr idhar . 1988. Elongat ional r heomet er s. In: Collyer ,
A.A. and D.W. Clegg (edit or s). Rheological Measur ement s. Elsevier
Applied Science, New Yor k. pg. 211-246.
Haghighi, K., A.K. Sr ivast ava and J .F. St effe. 1987. The r heological
appr oach t o soil modeling. Tr ans. ASAE 30: 1661-1672.
Haight on, A.J . 1959. The measur ement of har dness of mar gar ine and
fat s wit h cone penet r omet er s. J . Amer . Oil Chem. Soc. 36: 345-348.

400 Bibliography
Halim, H.K. and C.F. Shoemaker . 1990. Effect of addit ion of -, -, and
-casein, and Na-caseinat e on viscoelast ic pr oper t ies of skim milk
cur d. J . Text ur e St ud. 21: 323-337.
Halliday, P.J . and A.C. Smit h. 1995. Est imat ion of t he wall slip velocit y
in t he capillar y flow of pot at o gr anule past es. J . Rheol. 39: 139-149.
Halmos, A.L. and C. Tiu. 1981. Liquid foodst uffs exhibit ing yield st r ess
and shear -degr adabilit y. J . Text ur e St ud. 12: 39-46.
Hamann, D.D. 1983. St r uct ur al failur e in solid foods. In: Peleg, M. and
E.B. Bagley (edit or s). Physical Pr oper t ies of Foods. AVI Publishing
Co., West por t , CT. pg. 351-384.
Hamann, D.D. and T.C. Lanier . 1987. Inst r ument al met hods for pr e-
dict ing seafood sensor y t ext ur e qualit y. In: Kr amer , D.E. and J .
List on (edit or s). Seafood Qualit y Det er minat ion. Elsevier Applied
Sci. Publisher s, Bar king, Essex, England. pg. 123-136.
Han, C.D. 1988. Slit r heomet r y. In: Collyer , A.A. and D.W. Clegg (edi-
t or s). Rheological Measur ement s. Elsevier Applied Science, New
Yor k. pg. 25-48.
Hanks, R.W. 1963. Laminar -t ur bulent t r ansit ion of fluids wit h a yield
st r ess. AIChE J . 9: 306-309.
Har nby, N., M.F. Edwar ds and A.W. Nienow (edit or s). 1985. Mixing in
t he Pr ocess Indust r ies. But t er wor t h, Bost on.
Har per , J .C. and A.F. El Sahr igi. 1965. Viscomet r ic behavior of t omat o
concent r at es. J . Food Sci. 30: 470-476.
Har t net t , J .P. and R.Y.Z. Hu. 1989. The yield st r ess-an engineer ing
r ealit y. J . Rheol. 33: 671-679.
Hayashi, N., I. Hayakawa and Y. Fujio. 1991. Ent r ance effect cor r ect ion
on t he flow of moist ur ized soy pr ot ein isolat e melt in an ext r usion
r heomet er . Int . J . Food Sci. Technol. 26: 567-574.
Heywood, N. 1991a. Pipeline design for non-set t ling slur r ies. In: Br own,
N.P. and N.I. Heywood (edit or s). Slur r y Handling: Design of Solid
Liquid Syst ems. Elsevier Applied Science, New Yor k. pg. 125-165.
Heywood, N.I. 1991b. Rheological char act er izat ion of non-set t ling
slur r ies. In: Br own, N.P. and N.I. Heywood (edit or s). Slur r y Han-
dling: Design of Solid-Liquid Syst ems. Elsevier Applied Science, New
Yor k.
Hill, M.A. 1991. Dynamic r heology: applicat ion t o t he for mulat ion of a
r eal food pr oduct . Food Sci. Technol. Today 5(1): 29-31.
Hill, M.A., J .R. Mit chel and P.A. Sher man. 1995. The r elat ionship
bet ween t he r heological and sensor y pr oper t ies of lemon pie filling.
J . Text ur e St ud. 26: 457-470.
Holdswor t h, S.D. 1993. Rheological models used for t he pr edict ion of
t he flow pr oper t ies of food pr oduct s: a lit er at ur e r eview. Tr ans. I.
Chem. E. 71 (Par t C): 139-179.
Holland, D. 1994. Let t er t o t he edit or : which is mor e cr it ical - t he st r ess
or t he st r ain. J . Rheol. 38: 1941-1943.
Holland, F.A. and F.S. Chapman. 1966. Liquid Mixing and Pr ocessing
in St ir r ed Tanks. Reinholt Publishing Cor p., New Yor k.

401
Houska, M., J . Sesr k, J . J eschke, M. Adam and J . Pr idal. 1988. In: H.
Giesekus (edit or ). Pr ogr ess and Tr ends in Rheology II. Pr oceedings
of t he Second Confer ence of Eur opean Rheologist s, Pr ague, J une
17-20, 1986. Spr inger -Ver lag, New Yor k. pg 460-463.
Huang, H. and J .L. Kokini. 1993. Measur ement of biaxial ext ensional
viscosit y of wheat flour doughs. J . Rheol. 37: 879-891.
J ames, A.E., D.J .A. Williams and P.R. Williams. 1987. Dir ect mea-
sur ement of st at ic yield pr oper t ies of cohesive suspensions. Rheol.
Act a 26: 437-446.
J ames, D.F. and K. Walt er s. 1993. A cr it ical appr aisal of available
met hods for t he measur ement of ext ensional pr oper t ies of mobile
syst ems. In: Collyer , A.A. (edit or ). Techniques in Rheological Mea-
sur ement . Chapman & Hall, New Yor k.
J ast r zebski, Z.D. 1967. Ent r ance effect s and wall effect s in an ext r usion
r heomet er dur ing t he flow of concent r at ed suspensions. Ind. Engn.
Chem. Fund. 6: 445-454.
J ones, D.M., K. Walt er s and P.R. Williams. 1987. On t he ext ensional
viscosit y of mobile polymer syst ems. Rheol. Act a 26: 20-30.
Kalet unc-Gencer , G. and M. Peleg. 1984. Digit izer aided det er minat ion
of yield st r ess in semi-liquid foods. J . Food Sci. 49: 1620-1621.
Kawanar i, M., D.D. Hamann, K.R. Swar t zel and A.P. Hansen. 1981.
Rheological and t ext ur e st udies of but t er . J . Text ur e St ud. 12:
483-505.
Kawat a, M.K. Kur ase, A. Nagashima and K.Yoshida. 1991. Capillar y
viscomet er s. In: Wakeham, W.A., A. Nagashima and J .V. Senger s
(edit or s). Measur ement of t he Tr anspor t Pr oper t ies of Liquids, Vol.
III. Blackwell Scient ific, London.
Keent ok, H. 1982. The measur ement of yield st r ess of liquids. Rheol.
Act a 21: 325-332.
Kiljanski, T. 1989. A met hod for t he cor r ect ion of wall-slip effect in a
Couet t e r heomet er . Rheol. Act a 28: 61-64.
Kit t r edge, C.P. and D.S. Rowley. 1957. Resist ance coefficient s for
laminar and t ur bulent flow t hr ough one-half inch valves and fit t ings.
Tr ans. ASME 79: 1759-1766.
Kokini, J .L. 1992. Rheological pr oper t ies of food. In: Heldman, D.R. and
D.B. Lund (edit or s). Handbook of Food Engineer ing. Mar cel Dekker ,
New Yor k. pg. 1-38.
Kokini, J .L. and E.L. Cussler . 1987. The psychophysics of fluid t ext ur e.
In: Moskowit z, H.R. (edit or ). Food Text ur e. Mar cel Dekker , New
Yor k.
Kokini, J .L. and A. Dickie. 1981. An at t empt t o ident ify and model
t r ansient viscoelast ic flow in foods. J . Text ur e St ud. 12: 539-557.
Kokini, J .L. and G.L. Plut chok. 1987. Viscoelast ic pr oper t ies of semisolid
foods and t heir biopolymer ic component s. Food Technol. 41: 89-95.
Kout sakos, E. and A.W. Nienow. 1990. Effect s of r heological pr oper t ies
of simulat ed fer ment at ion br ot hs on flowin st ir r ed bior eact or s: a laser
402 Bibliography
anemomet r y st udy. In: R.E. Car t er (edit or ). Rheology of Food,
Phar maceut ical and Biological Mat er ials wit h Gener al Rheology.
Elsevier Applied Science, New Yor k. pg. 284-303.
Kr aynik, A.M., J .H. Auber t , R.N. Chapman and D.C. Gyur e. 1984. The
helical scr ew r heomet er : a new concept in r ot at ional r heomet r y. Pr oc.
Ann. Tech. Conf. Soc. Plast ics Engr s. pg. 405-408.
Kr ieger , I.M. 1968. Shear r at e in couet t e viscomet er . Tr ans. Soc. Rheol.
12: 5-11.
Kr ieger , I.M. and H. Elr od. 1953. Dir ect det er minat ion of t he flow cur ves
of non-Newt onian Fluids. II. Shear r at e in t he concent r ic cylinder
viscomet er . J . Appl. Physics. 24: 134-136.
Kr ieger , I.M. and S.H. Mar on. 1952. Dir ect det er minat ion of t he flow
cur ves of non-Newt onian fluids. J . Appl. Physics 23: 147-149.
Kr ieger , I.M. and S.H. Mar on. 1954. Dir ect det er minat ion of t he flow
cur ves of non-Newt onian Fluids. III. St andar dized t r eat ment of
viscomet r ic dat a. J . Appl. Physics. 25: 72-75.
Laba, D. 1993a. Rheological pr ofiles. In: Laba, D. (edit or ). 1993. Rhe-
ological Pr oper t ies of Cosmet ics and Toilet r ies. Mar cel Dekker , Inc.
New Yor k. pg. 403-420
Laba, D. (edit or ). 1993b. Rheological Pr oper t ies of Cosmet ics and Toi-
let r ies. Mar cel Dekker , Inc. New Yor k.
Lang, E.R. and C. Rha. 1981. Det er minat ion of yield st r esses of
hydr ocolloid disper sions. J . Text ur e St ud. 12: 47-62.
Langhaar , H.L. 1980. Dimensional Analysis and Theor y of Models.
Rober t E. Kr ieger Publishing Company, Hunt ingt on, New Yor k.
Lar son, R.G. 1985. Const it ut ive r elat ionships for polymer ic mat er ials
wit h power -law of r elaxat ion t imes. Rheol. Act a 24: 327-334.
Lar son, R.G. 1992. Inst abilit ies in viscoelast ic flows. Rheol. Act a 31:
213-263.
Launay, B. and J . Bur . 1977. Use of t he Chopin Alveogr aphe as a
r heological t ool. II. Dough pr oper t ies in biaxial ext ension. Cer eal
Chem. 54: 1152-1158.
Leider , P.L. and R.B. Bir d. 1974. Squeezing flow bet ween par allel
plat es. I. Theor et ical analysis. Ind. Engn. Chem. Fundam. 13:
336-341.
Leight on, A., A. Levit on and O.E. Williams. 1934. The appar ent vis-
cosit y of ice cr eam. J . Dair y Sci. 17: 639-650.
Lewicki, P.P. and K. Skier kowski. 1988. Flow of fr uit and veget able
pur ees t hr ough pipelines. In: H. Giesekus (edit or ). Pr ogr ess and
Tr ends in Rheology II. Pr oceedings of t he Second Confer ence of
Eur opean Rheologist s, Pr ague, J une 17-20, 1986. Spr inger -Ver lag,
New Yor k. pg 443-445.
Lewis, M.J . 1987. Physical Pr oper t ies of Foods and Food Pr ocessing
Syst ems. Ellis Hor wood. Chichest er , England.
Lodge, A.S. 1988. Nor mal st r ess differ ences fr om hole pr essur e mea-
sur ement s. In: Collyer , A.A. and D.W. Clegg (edit or s). Rheological
Measur ement s. Elsevier Applied Science, New Yor k. pg. 345-382.
403
Lor d, D.L., B.W. Hulsey and L.L. Melt on. 1967. Gener al t ur bulent pipe
flow scale-up cor r elat ion for r heologically complex fluids. Soc. Pet r ol.
Engr s. J . 7: 252-258.
Mackey, K.L., R.Y. Ofoli, R.G. Mor gan and J .F. St effe. 1989. Rheological
modeling of pot at o flour dur ing ext r usion cooking. J . Food Pr oc. Engn.
12: 1-11.
Macosko, C. W. 1994. Rheology: Pr inciples, Measur ement s and
Applicat ions. VCH Publisher s. New Yor k.
Mat z, S.A. 1962. Food Text ur e. AVI Publishing Co., Inc., West por t ,
Connect icut .
McCar t hy, K.L. and J .D. Seymour . 1994. Gr avit y cur r ent analysis of
t he Bost wick Consist omet er for power law fluids. J . Text . St ud. 25:
207-220.
Meissner , J . 1972. Development of a univer sal ext ensional r heomet er
for t he uniaxial ext ension of polymer melt s. Tr ans. Soc. Rheol. 16:
405-420.
Menjivar , J .A., C.N. Chang and L. Maceil. 1992. Ext r udat e swell
behavior of wheat flour doughs. In: P. Moldenaer s and R. Keunings
(edit or s). Theor et ical and Applied Rheology, Volume 2. Pr oceedings
of t he XIt h Int er nat ional Congr ess on Rheology. Br ussels, Belgium,
Aug. 17-21, 1992. Elsevier Science Publisher s, New Yor k. pg.
714-716.
Menjivar , J .A. and H. Far idi. 1994. Rheological pr oper t ies of cookie and
cr acker doughs. In: Far idi, H. (edit or ). The Science of Cookie and
Cr acker Pr oduct ion. Chapman & Hall, New Yor k. pg. 283-322.
Met zner , A.B. 1956. Non-Newt onian t echnology: fluid mechanics,
mixing, heat t r ansfer . In: Dr ew, T.B. and J .W. Hoopes (edit or s).
Advances in Chemical Engineer ing, Vol. 1. Academic Pr ess, Inc., New
Yor k. pg. 77-153.
Met zner , A.B. 1961. Flow of non-Newt onian fluids. In: Handbook of
Fluid Dynamics, Sect ion 7, McGr aw-Hill Book Co., Inc. pg. 7-1 t o 7-30.
Met zner , A.B. and A.P. Met zner . 1970. St r ess levels in r apid ext ensional
flows of polymer ic fluids. Rheol. Act a 9: 174-181.
Mewis, J . 1979. Thixot r opy - a gener al r eview. J . Non-Newt onian Fluid
Mech. 6: 1-20.
Michiyosi, L., K. Mizuno and Y. Hoshiai. 1966. St udies on t he flow of
slur r ies t hr ough pipe. I. Ent r ance Region of laminar flow. Int . Chem.
Engn. 6: 373-381
Middleman, S. 1968. The Flow of High Polymer s. Int er science Pub-
lisher s, New Yor k.
Middleman, S. 1977. Fundament als of Polymer Pr ocessing. McGr aw-
Hill Book Co., New Yor k.
Missair e, F., C-G. Qiu and M.A. Rao. 1990. Yield st r ess of st r uct ur ed
and unst r uct ur ed food suspensions. J . Text ur e St ud. 21: 479-490.
Mit schka, P. 1982. Simple conver sion of Br ookfield R.V.T. r eadings int o
viscosit y funct ions. Rheol. Act a 21: 207-209.
Mizr ahi, S. and Z. Ber k. 1972. Flow behavior of concent r at ed or ange
juice: mat hemat ical t r eat ment . J . Text ur e St ud. 3: 69-79.
404 Bibliography
Mohsenin, H. 1986. Physical Pr oper t ies of Plant and Animal Mat er ials,
Vol. 1 (Second Edit ion). Gor don and Br each Science Publisher s, New
Yor k.
Mont ejano, J .G., D.D. Hamann and T.C. Lanier . 1985. Compar ison of
t wo inst r ument al met hods wit h sensor y t ext ur e of pr ot ein gels. J .
Text ur e St ud. 16: 403-424.
Mooney, M. 1931. Explicit for mulas for slip and fluidit y. J . Rheol. 30:
337-357.
Mor gan, R.G. 1979. Modeling t he effect s of t emper at ur e-t ime hist or y,
t emper at ur e, shear r at e and moist ur e on t he viscosit y of defat t ed soy
flour . Ph.D. Disser t at ion. Texas A&M Univer sit y, College St at ion,
TX.
Mor gan, R.G., J .F. St effe and R.Y. Ofoli. 1989. Agener alized r heological
model for ext r usion of pr ot ein doughs. J . Food Pr oc. Engn. 11: 55-78.
Mor t on-J ones, D.H. 1989. Polymer Pr ocessing. Chapman and Hall. New
Yor k.
Muoz, A.M., A.S. Szczesnaik, M.A. Einst ein and N.O. Schwar t z, 1992.
The t ext ur e pr ofile. In: Hoot man, R.C. (edit or ). Manual on
Descr ipt ive Analysis Test ing for Sensor y Evaluat ion. ASTM Manual
Ser ies: MNL 13. Amer . Soc. for Test ing and Mat er ials, 1916 Race
St r eet , Philadelphia, PA.
Mur at a, T. and S. Oka. 1968. Det er minat ion of par amet er s of a mat er ial
obeying Cassons equat ion by some r heomet er s. Bior heology 5:
95-102.
Mur phy, C.E. 1950. Similit ude in Engineer ing. The Ronald Pr ess
Company, New Yor k.
Nagat a, S. 1975. Mixing: Pr inciples and Applicat ions. J ohn Wiley and
Sons, New Yor k.
Nagat a, S., N. Masabuma, T. Hisayuki, H. Hir abayashi and S. Got oh.
1970. Power consumpt ion of mixing impeller s in Bingham plast ic
liquids. J . Chem. Engn. of J apan 3: 237-243.
Nakayama, T., E. Niwa and I. Hamada. 1980. Pipe t r anspor t at ion of
minced fish past e. J . Food Sci. 45: 844-847.
Navickis, L.L. and E.B. Bagley. 1983. Yield st r esses in concent r at ed
disper sions of closely packed, defor mable gel par t icles. J . Rheol. 27:
519-536.
Nielsen, L.E. and R.F. Landel. 1994. Mechanical Pr oper t ies of Polymer s
and Composit es (Second Edit ion). Mar cel Dekker , Inc. New Yor k.
Nguyen, Q.D. and D.V. Boger . 1985. Dir ect yield st r ess measur ement
wit h t he vane met hod. J . Rheol. 29: 335-347.
Nguyen, Q.D. and D.V. Boger . 1987. Char act er izat ion of yield st r ess
wit h concent r ic cylinder viscomet er s. Rheol. Act a 26: 508-515.
Nolan, E.J ., V.H. Holsinger and J .J . Shieh. 1989. Dynamic r heological
pr oper t ies of nat ur al and imit at ion mozzar ella cheese. J . Text ur e
St ud. 20: 179-189.
Ofoli, R.Y., R.G. Mor gan and J .F. St effe. 1987. Agener alized r heological
model for inelast ic fluid foods. J . Text ur e St ud. 18: 213-230.
405
Ofoli, R.Y. and J .F. St effe. 1992. Some obser vat ions on t he use of slit
r heomet r y for char act er izing t he pr imar y nor mal st r ess differ ence of
ext r udat es. J . Food Engn. 18: 145-157.
Oldr oyd, J .G. 1956. Non-Newt onian flow of solids and liquids. In: Eir ich,
F.R. (edit or ). Rheology: Theor y and Applicat ions, Vol. 1. Academic
Pr ess Inc., New Yor k. pg 653-682.
Oldshue, J .Y. 1983 Fluid Mixing Technology. Chemical Engineer ing,
McGr aw-Hill Publicat ions Co., New Yor k
Osor io, F.A. and J .F. St effe. 1984. Kinet ic ener gy calculat ions for non-
Newt onian fluids in cir cular t ubes. J . Food Sci. 49: 1295-1296, 1315.
Osor io, F.A. and J .F. St effe. 1986. Back ext r usion of power law fluids.
J . Text ur e St ud. 18: 43-63
Padden, F.J . and T.W. Dewit t . 1954. Some r heological pr oper t ies of
concent r at ed polyisobut ylene solut ions. J . Appl. Physics 28:
1086-1091.
Padmanabhan, M. and M. Bhat t achar ya. 1993. Planar ext ensional
viscosit y of cor n meal dough. J . Food Engn. 18: 389-411.
Padmanabhan, M. and M. Bhat t achar ya. 1994. In-line measur ement s
of r heological pr oper t ies of polymer melt s. Rheol. Act a 33: 71-87.
Par edes, M.D.C., M.A. Rao and M.C. Bour ne. 1989. Rheological char -
act er ist ics of salad dr essings 2: effect of st or age. J . Text ur e St ud. 20:
235-250.
Par zonka, W. and J . Vocadlo. 1968. Mt hode de la car act r ist ique du
compor t ement r hologique des subst ances viscoplast iques dapr s les
mesur es au viscosimt r e de Couet t e (modle nouveau t r ois
par amt er s). Rheol. Act a. 7: 260-265.
Pat t on, T.C. 1964. Paint Flow and Pigment Disper sion. Int er science
Publisher s, New Yor k.
Peleg, M. 1980. Linear izat ion of r elaxat ion and cr eep cur ves of solid
biological mat er ials. J . Rheol. 24: 451-463.
Peleg, M. and M.D. Nor mand. 1983. Compar ison of t wo met hods for
st r ess r elaxat ion dat a pr esent at ion of solid foods. Rheol. Act a 22:
108-113.
Peleg, M. and M.D. Nor mand. 1995. St iffness assessment fr om jagged
for ce-defor mat ion r elat ionships. J . Text ur e St ud. 26: 353-370.
Pet r ie, C.J .S. 1979. Ext ensional Flows. Pit t man, London.
Philippoff, W. 1935. Kolloid-Zeit . 71:1-16. Cit ed in: Br odkey, R.S. 1967.
The Phenomena of Fluid Mot ions. Addison-Wesley Publishing Co.,
Menlo Par k, Califor nia.
Pipkin, A.C. 1986. Lect ur es on Viscoelast icit y Theor y (Second Edit ion).
Spr inger -Ver lag, New Yor k.
Pokor ny, J ., J . Davidek, S. Dobiasova, P. St er n and J . Cmolik. 1984.
Compar ison of sensor y analysis and r ot at ional r heomet r y in t he
t ext ur e evaluat ion of mar gar ines. J . Text ur e St ud. 15: 395-406.
Polakowski, N.H. and E.J . Ripling. 1966. St r engt h and St r uct ur e of
Engineer ing Mat er ials. Pr ent ice-Hall, Inc., Englewood Cliffs, NJ
Powell, R.E. and H.J . Eyr ing. 1944. Mechanisms for t he r elaxat ion
t heor y of viscosit y. Nat ur e 154: 427-428.
406 Bibliography
Powell, R.L. 1988. Rot at ional viscomet r y. In: Collyer , A.A. and D.W.
Clegg (edit or s). Rheological Measur ement s. Elsevier Applied Science,
New Yor k. pg. 247-296.
Pr ent ice, J .H. 1992. Dair y Rheology: A Concise Guide. VCH Publisher s,
Inc., New Yor k.
Pr ent ice, J .H. and D. Huber . 1983. Result s of a collabor at ive st udy on
measur ing r heological pr oper t ies of foodst uffs. In: J owit t R., F.
Escher , B. Hallst r m, H.F.Th. Meffer t , W.E.L. Spies and G. Vos
(edit or s). Physical Pr oper t ies of Foods. Applied Science Publisher s.
pg. 123-184.
Pr incen, H.M. 1986. A novel design t o eliminat e end effect s in a con-
cent r ic cylinder viscomet er . J . Rheol. 30: 271-283.
Pr ogelhof, R.C. and J .L. Thr one. 1993. Polymer Engineer ing Pr inciples.
Hansen Publisher s, New Yor k.
Pur kayast ha, S., M. Peleg, E.A. J ohnson and M.D. Nor mand. 1985. A
comput er aided char act er izat ion of t he compr essive cr eep behavior
of pot at o and cheddar cheese. J . Food Sci. 50: 45-50, 55.
Qiu, C-G. and M.A. Rao. 1988. Role of pulp cont ent and par t icle size in
yield st r ess of apple sauce. J . Food Sci. 53: 1165-1170.
Qiu, C-G. and M.A. Rao. 1989. Effect of disper sed phase on t he slip
coefficient of applesauce in a concent r ic cylinder viscomet er . J . Tex-
t ur e St ud. 20: 57-70.
Rao, M.A. 1975. Measur ement of flow of a food suspension wit h a mixer .
J . Text ur e St ud. 6: 533-539.
Rao, M.A. 1986. Rheological pr oper t ies of fluid foods. In: Rao, M.A. and
S.S.H. Rizvi (edit or s). Engineer ing Pr oper t ies of Foods. Mar cel
Dekker , Inc., New Yor k. pg. 1-48.
Rao, M.A. and H.J . Cooley. 1984. Det er minat ion of effect ive shear r at es
in r ot at ional viscomet er s wit h complex geomet r ies. J . Text ur e St ud.
15: 327-335.
Rao, M.A. and H.J . Cooley. 1992. Rheological behavior of t omat o past es
in st eady and dynamic shear . J . Text ur e St ud. 23: 415-425.
Rao, M.A. and J .F. St effe (edit or s). 1992. Viscoelast ic Pr oper t ies of
Foods. Elsevier Applied Science, New Yor k.
Rao, V.N.M. and G.E. Skinner . 1986. Rheological pr oper t ies of solid
foods. In: Rao, M.A. and S.S.H. Rizvi (edit or s). Engineer ing Pr oper t ies
of Foods. Mar cel Dekker , Inc., New Yor k. pg. 215-254.
Rasper , V.F. and K.R. Pr est on (edit or s). 1991. The Ext ensigr aph
Handbook. The Amer ican Associat ion of Cer eal Chemist s, St . Paul,
MN.
Reiner , M. 1964. The Debor ah Number . Physics Today. J anuar y: 62.
Rha, C. and P. Pr adipasena. 1986. Viscosit y of pr ot eins. In: Mit chell,
J .R. and D.A. Lewar d (edit or s). Funct ional Pr oper t ies of Food
Macr omolecules. Elsevier Science Publishing Co., Inc., New Yor k.
Rodr iguez, F. 1982. Pr inciples of Polymer Syst ems (Second Edit ion).
Hemispher e Publishing Cor p., New Yor k.
Rohm, H. 1993. Rheological behaviour of but t er at lar ge defor mat ions.
J . Text ur e St ud. 24: 139-155.
407
Rohm, H. and K.-H. Weidinger . 1993. Rheological behaviour of but t er
at small defor mat ions. J . Text ur e St ud. 24: 157-172.
Roos, Y.H. 1992. Phase t r ansit ions in food syst ems. In: Heldman, D.R.
and D.B. Lund (edit or s). Handbook of Food Engineer ing. Mar cel
Dekker , Inc., New Yor k.
Roos, Y.H. 1995. Glass t r ansit ion - r elat ed chemical changes in foods.
Food Technol. 49: 97-102.
Rosenber g, M., Z. Wang, S.L. Chuang and C.F. Shoemaker . 1995.
Viscoelast ic pr oper t y changes in Cheddar cheese dur ing r ipening. J .
Food Sci. 60: 640-644.
Ross-Mur phy, S.B. 1988. Small defor mat ion measur ement s. In: Blan-
shar d, J .M.V. and J .R. Mit chel (edit or s). Food St r uct ur e - It s Cr eat ion
and Evaluat ion. But t er wor t hs, London. pg 387-400.
Ryan, N.W. and M.M. J ohnson. 1959. Tr ansit ion fr om laminar t o t ur -
bulent flow in pipes. AIChE J . 5: 433-435.
Sakiadis, B.C. 1984. Fluid and par t icle mechanics. In: Per r y, R.H., D.W.
Gr een and J .O. Maloney (edit or s). Per r ys Chemical Engineer s
Handbook (Sixt h Edit ion), Sect ion 5. McGr aw-Hill Book Company,
New Yor k. pg. 5-4 t o 5-68.
Sar avacos, G.D. 1970. Effect of t emper at ur e on viscosit y of fr uit juices
and pur ees. J . Food Sci. 35: 122-125.
Saunder s, S.R., D.D. Hamann and D.R. Lineback. 1992. A syst ems
appr oach t o food mat er ial adhesion. Lebensm.-Wiss. u.-Technol. 25:
309-315.
Scalzo, A.M., R.W. Dicker son, J .T. Peeler and R.B. Read. 1970. The
viscosit y of egg pr oduct s. Food Tech. 24 (Nov.): 113-119.
Secor , R.B. 1988. Oper abilit y of Ext ensional Rheomet r y by St agnat ion,
Squeezing, and Fiber Dr awing Flows: Comput er -Aided-Analysis,
Viscoelast ic Char act er izat ion, and Exper iment al Analysis. Ph.D.
Disser t at ion. Univ. of Minnesot a.
Seet hamr aju, K. and M. Bhat t achar ya. 1994. Effect of ingr edient s on
t he r heological pr oper t ies of ext r uded cor n meal. J . Rheol. 38:
1029-1044.
Senouci, A. and A.C. Smit h. 1988a. An exper iment al st udy of food melt
r heology I. Shear viscosit y using a slit die viscomet er and a capillar y
r heomet er . Rheol. Act a 27: 546-554.
Senouci, A. and A.C. Smit h. 1988b. An exper iment al st udy of food melt
r heology II. End pr essur e effect s. Rheol. Act a 27: 649-655.
Sever s. E.T. 1962. Rheology of Polymer s. Reinhold Publishing Company,
New Yor k.
Shar ma, S.K., A.R. Hill, H.D. Goff and R. Yada. 1989. Measur ement of
coagulat ion t ime and cur d fir mness of r ennet ed milk using a Namet er
viscomet er . Milchwissenschaft 44: 682-685.
Sher man, P. 1970. Indust r ial Rheology. Academic Pr ess. New Yor k.
Sher man, P. 1988. The sensor y - r heology int er face. In: Blanshar d,
J .M.V. and J .R. Mit chel (edit or s). Food St r uct ur e - It s Cr eat ion and
Evaluat ion. But t er wor t hs, London. pg 417-432.
408 Bibliography
Shuey, W.C. and K.H. Tipples (edit or s). 1980. The Amylogr aph
Handbook. The Amer ican Associat ion of Cer eal Chemist s, St . Paul,
MN.
Shukla, A. and S.S.H. Rizvi. 1995. Measur ement s of flowabilit y of but t er
by capillar y r heomet r y. J Text ur e St ud. 26: 299-311.
Shukla, A., S.S.H. Rizvi and J .A. Bar t sch. 1995. Rheological char ac-
t er izat ion of but t er using lubr icat ed squeezing flow. J Text ur e St ud.
26: 313-323.
Silvest er , R.S. 1985. Mixing of Non-Newt onian Media, A Technical
Review. (Br it ish Hyr domechanics Resear ch Associat ion. Fluid
Engineer ing. Fluid Engineer ing Ser ies; Vol. 12.) BHRA, Cr anfield,
Bedfor d, England
Skalak, R. and S. Chien (edit or s). 1987. Handbook of Bioengineer ing.
McGr aw-Hill Book Company, New Yor k.
Skelland, A.P.H. 1967. Non-Newt onian Flow and Heat Tr ansfer . J ohn
Wiley and Sons, Inc. New Yor k.
Skelland, A.P.H. 1983. Mixing and agit at ion of non-Newt onian fluids.
In: Cher emisinoff, N.P. and R. Gupt a (edit or s). Handbook of Fluids
in Mot ion. Ann Ar bor Science Publisher s, Ann Ar bor , Michigan. pg.
179-209.
Sone, T. 1972. Consist ency of Foodst uffs. D. Reidel Publishing Co.,
Dor dr echt , Holland.
St effe, J .F. 1984. Pr oblems in using appar ent viscosit y t o select pumps
for pseudoplast ic fluids. Tr ans. ASAE 27: 629-634.
St effe, J .F. 1992. Yield st r ess: phenomena and measur ement . In: Singh,
R.P. and A. Wir akar t akusumah (edit or s). Advances in Food Engi-
neer ing. CRC Pr ess, Boca Rat on, FL. pg. 363-376.
St effe, J .F., E.M. Cast ell-Per ez, K.J . Rose and M.E. Zabik. 1989. Rapid
t est ing met hod for char act er izing t he r heological behavior of gela-
t inizing cor n st ar ch slur r ies. Cer eal Chem. 66: 65-68.
St effe, J .F. and E.W. For d. 1985. Rheological t echniques t o evaluat e t he
shelf-st abilit y of st ar ch-t hickened, st r ained apr icot s. J . Text ur e St ud.
16: 179-192.
St effe, J .F., I.O. Mohamed and E.W. For d. 1986. Rheological pr oper t ies
of fluid foods: dat a compilat ion. In: Okos, M.R. (edit or ). Physical and
Chemical Pr oper t ies of Foods. ASAE. St . J oseph, Michigan. pg. 1-13.
St effe, J .F., I.O. Mohamed and E.W. For d. 1984. Pr essur e dr op acr oss
valves and fit t ings for pseudoplast ic fluids in laminar flow. Tr ans.
ASAE 27: 616-619.
St effe, J .F. and R.G. Mor gan. 1986. Pipeline design and pump select ion
for non-Newt onian fluid foods. Food Technol. 40: 78-85. [Addendum:
Food Technol. 41: 32].
St effe, J .F. and F.A. Osor io. 1987. Back ext r usion of non-Newt onian
fluids. Food Technol. 41: 72-77.
St effe, J .F., F.A. Osor io and R.Y. Ofoli. 1989. Dynamic and st eady
r heological pr oper t ies of st ar ch based pie filling. In: Ang, H.G. (edi-
409
t or ). Tr ends in Food Science. Par t Six, Rheological St udies," pg.
128-130. The Singapor e Inst . Food Sci. and Technol., Republic of
Singapor e.
Sweeney, E.T. 1978. An Int r oduct ion and Lit er at ur e Guide t o Mixing.
- (Br it ish Hyr domechanics Resear ch Associat ion. Fluid Engineer ing.
Fluid Engineer ing Ser ies; Vol. 5.) BHRA, Cr anfield, Bedfor d, England
Szczesnaik, A.S. 1995. Text ur e pr ofile analysis-met hodology int er -
pr et at ion clar ified. J . Food Sci. (Let t er s) 60: vii.
Szczesnaik, A.S., M.A. Br andt and H.H. Fr iedman. 1963. Development
of st andar d r at ing scales for mechanical par amet er s of t ext ur e and
cor r elat ion bet ween t he object ive and sensor y met hods of t ext ur e
evaluat ion.. J . Food Sci. 28: 397-403.
Tadmor , Z. and C.G. Gogos. 1979. Pr inciples of Polymer Pr ocessing.
J ohn Wiley and Sons, New Yor k.
Tamur a, M.S., J .M. Hender son, R.L. Powell and C.F. Shoemaker . 1989.
Evaluat ion of t he helical scr ew r heomet er as an on-line viscomet er .
J . Food Sci. 54: 483-484.
Tanaka, M., J .M. De Man and P.W. Voisey. 1971. Measur ement of
t ext ur al pr oper t ies of foods wit h const ant speed cone penet r omet er .
J . Text ur e St ud. 2: 306-315.
Tanner , R.I. 1985. Engineer ing Rheology. Oxfor d Univer sit y Pr ess,
Oxfor d.
Tanner , R.I. 1988. Recover able elast ic st r ain and swelling r at io. In:
Collyer , A.A. and D.W. Clegg (edit or s). Rheological Measur ement s.
Elsevier Applied Science, New Yor k. pg. 93-118.
Tat t er sall, G.H. and P.F.G. Banfill. 1983. The Rheology of Fr esh
Concr et e. Pit man Publishing Inc., Bost on.
Tir t aat madja, V. and T. Sr idhar . 1993. A filament st r et ching device for
measur ement of ext ensional viscosit y. J . Rheol. 37: 1081-1102.
Tiu, C. and D.V. Boger . 1974. Complet e r heological char act er izat ion of
t ime-dependent food pr oduct s. J . Text ur e St ud. 5: 329-338.
Toledo, R., J . Cabot and D. Br own. 1977. Relat ionship bet ween com-
posit ion, st abilit y and r heological pr oper t ies of r awcomminut ed meat
bat t er s. J . Food Sci. 42: 725-727.
Toms, B.A. 1958. Fundament al t echniques: fluids. In: Eir ich, F.R.
(edit or ). Rheology: Theor y and Applicat ions, Volume II. Academic
Pr ess, New Yor k. pg. 475-500.
Tr out on, F.T. 1906. On t he coefficient of viscous t r act ion and it s r elat ion
t o t hat of viscosit y. Pr oceedings of t he Royal Societ y of London. Ser ies
A, Vol. LXXVII: 426-440.
Uhl, V.W. and J .B. Gr ay (edit or s). 1986. Mixing, Theor y and Pr act ice,
Vol. III. Academic Pr ess, Inc., New Yor k.
Ulbr echt , J .J . and P. Car r eau. 1985. Mixing of viscous non-Newt onian
liquids. In: Ulbr echt , J .J . and G.K. Pat t er son (edit or s). Mixing of
Liquids by Mechanical Agit at ion. Gor don and Br each Science Pub-
lisher s, New Yor k. pg. 93-138.
410 Bibliography
Ulbr echt , J .J . and G.K. Pat t er son (edit or s). 1985. Mixing of Liquids by
Mechanical Agit at ion. Gor don and Br each Science Publisher s, New
Yor k.
Ulbr icht , D. M.D. Nor mand and M. Peleg. 1995. Cr eat ing t ypical jagged
for ce-defor mat ion r elat ionships fr om t he ir r egular and ir r epr oducible
compr ession dat a of cr unchy foods. J . Sci. Food Agr ic. 67: 453-459.
Ur y, J .F. 1966. Resist ance t o st eady laminar flow of pseudoplast ic fluids
t hr ough pipes and valves. J . Mech. Engn. Sci. 8: 226-233.
Van Wazer , J .R. 1963. Viscosit y and Flow Measur ement . Int er science
Publisher s, New Yor k.
Ver cr uysse, M.C.M. and J .F. St effe. 1989. On-line viscomet r y for pur eed
baby food: cor r elat ion of Bost wick Consist omet er r eadings and
appar ent viscosit y dat a. J . Food Pr oc. Engn. 11: 193-202.
Vocadlo, J .J . and M.E. Char les. 1971. Measur ement of yield st r ess of
fluid-like viscoplast ic subst ances. Canadian J . Chem. Engn. 49:
576-582.
Viswanat h, D.S. and G. Nat ar ajan. 1989. Dat a Book on t he Viscosit y of
Liquids. Hemispher e Publishing Cor p., New Yor k.
Vit ali, A.A. and M.A. Rao. 1984. Flowpr oper t ies of low pulp concent r at ed
or ange juice: effect of t emper at ur e and concent r at ion. J . Food Sci.
49: 882-888.
Vyalov, S.S. 1986. Rheological Fundament als of Soil Mechanics.
Elsevier Science Publisher s, New Yor k.
Walt er s, K. 1975. Rheomet r y. Chapman and Hall, London.
War r en, R.C. 1988. Viscous heat ing. In: Collyer , A.A. and D.W. Clegg
(edit or s). Rheological Measur ement s. Elsevier Applied Science, New
Yor k. pg. 119-150.
Weast , R.C., M.J . Ast le and W.H. Beyer . 1985. CRC Handbook of
Chemist r y and Physics. CRC Pr ess, Inc.. Boca Rat on, FL
Weiper t , D., H.-D. Tscheuschner and E. Windhab (edit or s). 1993.
Rheologie der Lebensmit t el. B. Behr s Ver lag GmbH & Co., Hamber g,
Ger many.
Whor low, R.W. 1992. Rheological Techniques (Second Edit ion). Ellis
Hor wood, New Yor k.
Wikst r m, K., L. Bohlin and A. Eliasson. 1994. Biaxial ext ension of
wheat flour doughs. Rheology 94, 4: 192-197.
Williams, M.L., R.F. Landel and J .D. Fer r y. 1955. The t emper at ur e
dependence of r elaxat ion mechanisms in amor phous polymer s and
ot her glass for ming liquids. J . Am. Chem Soc. 77: 3701-3707.
Williams, R.W. 1979. Det er minat ion of viscomet r ic dat a fr om t he
Br ookfield R.V.T. viscomet er . Rheol. Act a 18: 345-359.
Wint her , G., K. Almdal and O. Kr amer . 1991. Det er minat ion of t he
polymer melt viscosit y by squeezing flow wit h const ant plat e vis-
cosit y. J . Non-Newt onian Fluid Mech. 39: 119-136.
Wood, F.W. and T.C. Goff. 1973. The det er minat ion of t he effect ive
shear r at e in t he Br abender Viscogr aph and in ot her syst ems of
complex geomet r y. Die St ar ke 25: 89-91.
411
Yang, T.M.T. and I.M. Kr ieger . 1978. Compar ison of met hods for cal-
culat ing shear r at es in coaxial viscomet er s. J . Rheol. 22: 413-421.
Yanovsky, Y.G. 1993. Polymer Rheology: Theor y and Pr act ice. Chap-
man and Hall, New Yor k.
Yoo, B., M.A. Rao and J .F. St effe. 1995. Yield st r ess of food disper sions
wit h t he vane met hod at cont r olled shear r at e and shear st r ess. J .
Text ur e St ud. 26: 1-10.
Yoshimur a, A.S. and R.K. Pr udhomme. 1987. Response of and elast ic
Bingham fluid t o oscillat or y shear . Rheol. Act a 26: 428-436.
Yoshimur a, A.S. and R.K. Pr udhomme. 1988. Wall slip cor r ect ions for
couet t e and par allel disk viscomet er s. J . Rheol. 32: 53-67.
Yoshimur a, A.S. R.K. Pr udhomme, H.M. Pr incen and A.D. Kiss. 1987.
A compar ison of t echniques for measur ing yield st r ess. J . Rheol. 31:
699-710.
Zukoski, C.F. 1993. Mat er ial pr oper t ies and t he elect r or heological
r esponse. Annu. Rev. Mat er . Sci. 23: 45-78.
Index
Activation energy, 33, 85
Adams Consistometer, 67
Adhesiveness, 75
Agar gel, 330
Alveograph, 66
Amplitude ratio, 314
Anelasticity, 8
Angle of shear, 5
Apparent viscosity, 24
Apparent wall shear rate, 100
Apple tissue, 340, 343
Apricots, 240, 250
Arrhenius equation, 33, 85
Average shear rate, 165, 216
Average shear stress, 215
Bagley plot, 112
Biaxial extension, 45
Bingham plastic fluid
definition, 20
in concentric cylinder viscometer,
163
in mixer viscometer, 199
minimum bob speed, 210
tube flow rate, 103
tube velocity profile, 105
Bioyield point, 71
Bird-Leider equation, 310
Bostwick Consistometer, 67
Boussinesq number, 53
Brabender-FMC Consistometer, 70
Brinkman number, 183
Brookfield Viscometer, 69
Bubble growth, 40
Buckingham pi theorem, 186
Buckingham-Reiner equation, 103
Bulk compression, 294
Bulk modulus, 11
Bulk viscosity, 53
Burgers model, 50, 307, 343
Butter, 258
Calendering, 40
Cannon-Fenske viscometer, 3, 94
Cantilever beam, 10
Capillary viscometer, 94
errors in operating, 110
Carrageenan gum solution, 77
Casson equation, 24
with milk chocolate, 82
Casson fluid, 149
Cauchy strain, 4
Cavitation, 182, 236
Cessation of steady shear flow, 48,
295
Cessation of steady tensile extension,
297
Characteristic time
of material, 333
of process, 333
Cheese spread, 283
Chemorheology, 321
Chewiness, 75
CMC solution, 146
Coating, 91
Coefficient of viscosity, 19
Cogswells equations, 264, 287
Cohesiveiness, 75
Complex compliance, 316
Complex modulus, 49, 315
Complex viscosity, 49, 315
Compression-extrusion cell, 70, 71
Compressive stress, 6
Concentric cylinder viscometry, 158
Bingham plastic fluids, 162
cavitation, 236
end correction, 174
Krieger method, 165
Newtonian approximation, 164
Newtonian fluids, 161
413
power law approximation, 165
power law fluids, 162
secondary flow, 182
shear rate calculations, 163
simple shear approximation, 164
viscous heating, 177
wall effects (slip), 181
Cone and plate viscometry, 3, 169,
226, 227
truncated cone, 183
Cone penetrometer, 66
Conservation of momentum equa-
tions, 141
Constitutive equation, 1, 7
Converging die, 263, 287, 289
Corn starch, 79
Corn syrup, 150
Couette system, 158
Cox-Merz rule, 48, 338
Creep, 48, 295, 304
Creep compliance, 48, 305, 343
Critical radius
from bob, 225
tube flow, 105
Cross equation, 24
Darcy friction factor, 129
Data corrections
concentric cylinder viscometer, 174
cone and plate viscometer, 182
parallel plate viscometer, 182
tube viscometer, 110
Deborah number, 12, 332
Dextrin solution, 329
Die exit effect integral, 269
Die swell
see jet expansion
Dilatent behavior, 20
Dimensional analysis, 186
DIN standard, 176, 214
Disk surface viscometer, 54
Dough testing equipment, 65
Dynamic viscosity, 19, 49, 316, 348
Edge failure, 183
Elastic behavior, 49
Elastic solids, 8
Elastoplastic material, 11
Electrorheology, 55
Ellis equation, 24
Elongational viscosity, 39
End correction
concentric cylinder viscometer, 174
tube viscometer, 111
Energy of activation, 33, 85
Entrance length in tube, 114
Equation of state, 7
Extensigraph, 65
Extensional-thickening, 45
Extensional-thinning, 45
Extensional flow, 39
biaxial, 43, 258, 283
planar, 43
tensile, 43
uniaxial, 255
Extensional viscosity, 39
biaxial, 45
planar, 46
tensile, 45
Extrudate diameter, 122
Extrudate drawing, 40, 274
Extrudate swell
see jet expansion
Falling ball viscometer, 69, 82
Fanning friction factor, 128, 131
Fano flow, 276
Farinograph, 65
Fiber spinning, 274
414 Index
Finite bob in infinite cup, 168
First normal stress difference
definition, 16
from cone and plate data, 171
in jet expansion, 122
Flexural testing, 9
Fluidity, 19
FMC Consistometer, 70
Fracturability, 75
Frequency sweep, 320
Friction loss coefficient, 133, 135,
136
Froude number, 189
Gibsons equations, 268, 289
Glass capillary viscometer, 2, 94, 125
kinetic energy correction, 127
Glass transition temperature, 34
Gleisselss mirror relation, 48
Gumminess, 75
Hardness, 75
Hedstrom number, 109
Helical screw rheometer, 185
Helipath Stand, 69
Hencky strain, 5
Herschel-Bulkley fluid, 20
tube flow rate, 103
tube velocity profile, 106
Hoeppler Viscometer, 70
Hole pressure error, 120
Honey, 82
Hookes law, 8, 298
Hookean behavior, 49
Hystersis loop, 29
Impeller Reynolds number, 188
Inelastic fluids, 19
Infinite cup solution, 168, 221, 223,
225
Inherent viscosity, 27
Interfacial rheology, 53
Interfacial viscometer, 54
Intrinsic viscosity, 27
Jet expansion, 16, 47, 121
Kelvin model, 50, 298, 306
Kinematic viscosity, 19, 127
Kinetic energy correction, 110, 127
Kinetic energy correction factor, 131,
133
Kramer Shear Cell, 70, 71
Krieger method, 165
Lambda carrageenan solution, 330
Laminar flow criteria
concentric cylinder viscometer, 182
tube viscometer, 107
Launs rule, 48
Leibnitz rule, 99, 173
Limiting viscosity
at infinite shear, 23
at zero shear, 23
Linear elastic material, 11
Linear viscoelastic behavior, 47, 49,
318
Linear viscoelasticity, 296
Loss compliance, 316
Loss modulus, 49, 315, 348
Lubricated squeezing flow, 277
Magnetorheological fluids, 55
Margules equation, 162
Matching viscosity method, 192
Maxwell model, 50, 298, 318, 341
Mechanical analogues, 298
Mechanical energy balance, 128, 152
Melt flow index, 77
Melt Flow Indexer, 76
Melt fracture, 16
Melt spinning, 274
Milk chocolate, 81
Mixer viscometer, 3
Mixer viscometer constant
definition, 190
415
matching viscosity method, 193,
199
slope method, 191
variables influencing, 194
Mixer viscometry, 185
average shear rate, 190
Bingham plastic fluids, 199
matching viscosity method, 192,
194
power law fluids, 190, 195, 237
rheomalaxis, 208
slope method, 191, 194
Mixing
commercial, 185
impellers, 186
power consumption, 187
Mixograph, 65
Modeling rheological behavior, 32
concentration effects, 34
shear effects, 86
temperature effects, 34, 86
Modulus of elasticity
see Youngs modulus
Mooney-Couette bob, 176, 233
Nahme number, 119
Newtonian behavior, 49
Newtonian fluid
definition, 19
in concentric cylinder viscometer,
162, 213
in glass capillary viscometer, 125
in parallel plate viscometer, 173
in slit viscometer, 122
tube flow rate, 101
tube velocity profile, 103
turbulent pipe flow, 138, 155
Non-linear elastic material, 11
Non-linear viscoelasticity, 296
Normal stress coefficient, 15
Normal stress difference, 16
On-line viscometer
capillary flow type, 58
concentric cylinder type, 59
falling piston type, 62
off-set rotating sensor, 60
vibrating rod type, 60
vibrating sphere type, 61
Opposing jets, 272
Orange juice concentrate, 86
Orifice viscometer, 68
Oscillatory testing, 48, 336
application of stress and strain, 313
frequency sweep, 321
operating modes, 318
strain sweep, 320
temperature changes, 322
time sweep, 322
typical data, 324
Ostwald viscometer, 94
Parallel plate plastometer, 280
Parallel plate viscometer, 3
Parallel plate viscometry, 172, 229
Peanut butter, 291
Phase shift, 314
Pipe viscometer, 3, 94, 96
Pipeline design calculations, 128
Planar extension, 45
Poiseuille-Hagen equation, 101, 125
Poissons ratio, 9
Powell-Eyring equation, 24
Power law fluid, 20
in concentric cylinder viscometer,
162, 212, 216
in cone and plate viscometer, 171
in mixer viscometer, 195
in parallel plate viscometer, 173
tube flow rate, 102
tube velocity profile, 104
turbulent pipe flow, 140, 156
Power number, 188
Pseudoplastic behavior, 20
416 Index
Rabinowitsch-Mooney equation, 97,
100, 145, 147
Rapid Visco Analyser, 68
Recoil, 17, 47, 48, 295, 305
Recoverable shear, 122
Reduced viscosity, 27
Reiner-Philippoff equation, 24
Reiner-Riwlin equation, 163
Relative viscosity, 27
Relaxation time, 301, 341
Representative shear rate, 165, 216
Representative shear stress, 215
Reynolds number
Bingham, 108
Newtonian, 107
power law, 107
Rheodestruction, 28
Rheogoniometer, 18
Rheogram, 19
Rheological equation of state, 1, 7
Rheological instruments, 2
Rheomalaxis, 28
evaluation by mixer viscometry,
208, 250
Rheopectic material, 28
Rising bubble viscometer, 69
Rod climbing, 47
Rolling ball viscometer, 70
Rupture point, 71
Sagging, 40
Salad dressing, 223, 227
Scott equation, 282
Searle system, 158
Second normal stress difference
definition, 16
from parallel plate data, 174
Secondary flow, 182
Shear-thickening, 20
Shear-thinning, 20, 23
Shear modulus, 8, 300
Shear rate
definition, 13
in mixing, 51
in spreading or brushing, 51
tube flow, 52
typical values, 15
Shear stress, 6, 13
Sheet stretching, 40
Silicone polymer, 335
Silly putty, 334
Simple compression, 71
Sink flow analysis, 263
Skim milk curd, 343
Slip
concentric cylinder viscometer, 181
oscillatory testing, 336
tube viscometer, 116
Slit viscometry, 122, 150
Slope method, 191
Small amplitude oscillatory testing,
49
Sodium carboxymethylcellulose solu-
tion, 146
Sol-gel transition, 28
Solid behavior
elastoplastic, 12
linear elastic, 12
non-linear elastic, 12
Solution viscosities, 27
Soy dough, 143, 287, 289
Specific viscosity, 27
Spinning, 274
Spreadability, 66
Springiness, 75
Squeeze film viscometer, 280
Squeezing flow, 276
lubricated, 277, 291
nonlubricated, 279
Start-up flow, 48, 295, 310
Steady shear flow, 13
Stefan equation, 281
Step strain, 48, 295, 299
Stokes law, 82
417
Storage compliance, 316
Storage modulus, 49, 315, 348
Strain, 4, 313
axial, 9
Cauchy, 4
concentric cylinders, 324
cone and plate, 324
engineering, 4
Hencky, 5
lateral, 9
rotational, parallel plates, 321
true, 5
volumetric, 11
Strain sweep, 318
Stress, 4, 313
compressive, 7
tensile, 7
Stress overshoot, 310
mayonnaise, 311
Stress relaxation, 48, 299, 342
Stress relaxation modulus, 48, 299
Stress sweep, 318
Stress tensor, 14
Stringiness, 75
Structural parameter, 30
Surface loading, 4
T-bars, 69
Tan delta, 316, 348
Taylor vortices, 182
Tensile creep, 297
Tensile recoil, 297
Tensile start-up, 297
Tensile step strain, 297
Tensile stress, 6
Tension-thickening, 45
Tension-thinning, 45
Texture Profile Analysis, 72, 73
Texture profile curve, 74
Thixotropic material, 28
structure, 37
Three-point bending, 10
Time-dependent functions, 27, 50
Time-dependent thickening, 28
Time-dependent thinning, 28
Time-independent functions, 13, 49
Time sweep, 321
Tomato ketchup, 218, 226, 235
Torsional flow
see parallel plate viscometry
Trouton number, 47, 288, 291
Trouton viscosity, 39
Truncated cone, 183
Tubeless siphon, 16, 47, 276
Turbulent pipe flow, 138
Uniaxial compression, 8
Uniaxial extension, 43
Unlubricated squeezing flow, 279
Unsteady shear testing
oscillatory, 294
transient, 294
U-tube viscometer, 94
Vane method
see yield stress
Velocity profile in tube flow
Bingham plastic fluid, 105
Herschel-Bulkley fluid, 106
Newtonian fluid, 103
power law fluid, 104
turbulent flow, 138
Vibrational viscometer, 4
Visco-Amylograph, 68
Viscoelastic behavior
mechanical analogues, 298
Viscoelastic functions, 338
Viscometric flow, 15
Viscometric functions, 13, 15, 338
Viscosity
apparent, 24
extensional, 39
inherent, 26
intrinsic, 26
418 Index
limiting at infinite shear, 23
limiting at zero shear, 23, 301
reduced, 26
relative, 26
solution, 26
specific, 26
Viscous behavior, 49
Viscous heating
concentric cylinder viscometer, 177,
235
tube flow, 118
Volume loaded viscometer, 4
Wall effects
see slip
Warner-Bratzler Shear, 67
Weber number, 189
Weissenberg number, 189
Weissenberg effect, 16, 47
Williams-Landel-Ferry equation, 34
Wind-up characteristics, 207, 250
Yield number, 38
Yield stress
concentric cylinder viscometer, 163
definition, 35
dynamic, 37, 202
in Bingham plastic fluid, 21
in Herschel-Bulkley fluid, 21
measurement methods, 36
oscillatory testing, 336
static, 37, 202
tube viscometer, 121
vane method, 200, 243, 244, 247
Youngs modulus, 9
Zahn viscometer, 68