Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Download
Standard view
Full view
of .
Save to My Library
Look up keyword
Like this
0Activity
0 of .
Results for:
No results containing your search query
P. 1
Wax Appearance Temperatures of Vegetable Oils Determined By

Wax Appearance Temperatures of Vegetable Oils Determined By

Ratings: (0)|Views: 82|Likes:
Published by Shahabuddin Suzan

More info:

Published by: Shahabuddin Suzan on Feb 20, 2012
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

02/20/2012

pdf

text

original

 
Wax appearance temperatures of vegetable oils determined bydifferential scanning calorimetry: effect of triacylglycerolstructure and its modification
$
A. Adhvaryu
, S.Z. Erhan
, J.M. Perez
a
a
 Department of Chemical Engineering, Pennsylvania State University, University Park, PA 16802, USA
b
USDA/NCAUR/ARS, Food and Industrial Oil Research, 1815 N. University Street, Peoria, IL 61604, USA
Received 19 November 2001; received in revised form 6 March 2002; accepted 9 March 2002
Abstract
Crystallization and wax appearance temperatures of a series of vegetable oils (natural, genetically and chemically modified)were studied using differential scanning calorimetry (DSC). The fatty acid chains of a triacylglycerol molecule have a bend‘tuningforkconformationandundergomolecularstackingduringthecoolingprocess.Waxcrystallizationatlowtemperatureiscontrolled by steric and geometrical constrains in these molecules. This study describes an approach to quantify and predict waxappearancetemperatureofvegetableoilsbasedonthestatisticalanalysisofDSCandNMRdata.Amolecularmodelingprogramwas used to design triacylglycerol molecules with different fatty acid (e.g. oleic and linoleic) chains to illustrate their effect onthe crystallization process. Effect of pour point depressant (PPD) additives on vegetable oil crystallization is also discussed.Published by Elsevier Science B.V.
Keywords:
DSC; Vegetable oils; Wax formation; NMR; Molecular modeling; Statistical analysis
1. Introduction
Vegetable oils when subjected to low temperatureenvironmentundergosolidificationthroughcrystalliza-tionandthereforeareamajorhurdleforuseinindustrialapplications. The relatively poor low temperature flowproperties of vegetable oils arise from the appearanceof waxy crystals that rapidly agglomerate resulting inthe solidification of the oil. Vegetable oil is a complexmolecular system and therefore the transition fromliquid to solid state do not occur at a particular tem-perature, but over a wide temperature range involvingseveral polymorphic forms (
a
,
b
0
,
b
)[1–3]contributingto the wax appearance and crystallization process.This deposition of waxy materials from oil results ina rapid viscosity increase leading to poor pumpability,lubrication and rheological behavior.Severalstandardmethods are currentlyavailable fordetermining the low temperature properties of vege-table oils, such as pour point (ASTM D-97) and cloudpoint (ASTM D-2500). These methods are extremelytime consuming and data reproducibility betweenlaboratories is poor. Differential scanning calorimetry(DSC) is a relatively simple and reproducible techni-que capable of providing a direct measurement of the
Thermochimica Acta 395 (2003) 191–200
$
Names are necessary to report factually on available data;however, the USDA neither guarantees nor warrants the standard of the product, and the use of the name by USDA implies no approvalof the product to the exclusion of others that may also be suitable.
*
Corresponding author. Tel.:
þ
1-309-681-6532;fax:
þ
1-309-681-6340.
E-mail address:
erhansz@ncaur.usda.gov (S.Z. Erhan).0040-6031/02/$ – see front matter. Published by Elsevier Science B.V.PII: S0040-6031(02)00180-6
 
D
 H 
(change in enthalpy) for a system undergoingphysical and chemical changes during heating orcooling[4
6]. A number of studies have been carriedout on the cooling behavior of different wax bearingcrudes and mineral base oils using DSC[7
12].Reports are also available on the application of DSC for identi
cation of various stable polymorphsin pure single acid triacylglycerol structure by heatingand cooling experiments ([13
15]and referencestherein). However the situation in an unmodi
ed orgeneticallymodi
edvegetable oilisfarmore complex.This is primarily due to the presence of different fattyacid moieties, their chain length, abundance, structuraldifference and con
guration in the triacylglycerol thatmakes vegetable oils signi
cantly different from otherpure molecules. All the above factors individually andcollectively are capable of in
uencing the physicalproperties of vegetable oils.In this paper we describe the application of DSC todetermine wax appearance and crystallization tem-perature in different vegetable oils. The effect of fattyacid composition, structure and spatial orientation oncooling process is discussed. A molecular modelingprogram was used to design different triacylglycerolmolecules using an energy minimization approach.The results were used to explain the stacking processof the
tuning for
structure and wax appearancetemperature of vegetable oils. The usefulness of statistical models based on structural data derivedfrom proton nuclear magnetic resonance (
1
H NMR),for the prediction of low temperature crystallization of vegetable oils is also discussed.
2. Experimental
2.1. Materials
The vegetable oils used in the study were cottonseed, corn (local grocery store), sun
ower (sun
r),saf 
ower (sa
r) (Liberty Vegetable Oil, CA), soybeanoil (SBO) (Pioneer Hi Bred Intl., IA). The geneticallymodi
ed vegetable oils were high oleic sun
ower(HO sun
r) (Intl. Flora Tech, Gilbert, AZ), high oleicsaf 
ower (HO sa
r) and high linoleic sa
ower(HL sa
r) oil (Oil Seeds Intl., CA). The chemicallymodi
ed oils were epoxidized soybean (ESBO) oil(Elf Atochem Intl., PA), CMSBO #1, #2 and heatmodi
ed soybean oil (HMSBO; M
N viscosity).HMSBO was prepared by thermal treatment of SBO under N
2
atmosphere in a closed reactor underconstant stirring condition. Viscosity was carefullycontrolled to M
N grade according to Gardner scale.Chemically modi
ed soybean oils (CMSBO #1and 2) were synthesized as part of a separate researchproject. In this procedure, the vegetable oil molecularstructure was chemically modi
ed in order to enhanceitslowtemperature
uidity.Alltheseoilswerederivedfrom SBO and maintained the basic triacylglycerolbackbone. The pour point depressant (PPD) used wasa mixture with the following composition: sun
oweroil, mineral oil and undisclosed compounds (LubrizolCorp., TM 7670).
2.2. Sub-ambient DSC procedures
DSC experiments were done on a TA Instruments(New Castle, DE) 2910 DSC-model 2100 with acomputer-based controller. Before each experiment,the DSC cell was purged with low-pressure nitrogengas. Typically 10 mg of the oil sample was accuratelyweighed in an open aluminum pan and placed in theDSC module with a similar empty pan as reference.The procedure involved rapidly heating the sample to50
8
C and then held under isothermal condition for10 min. This helps in dissolving and homogenizingany waxy material present in the oil, which mayinadvertently act as seed to accelerate wax crystalformation during cooling. The system was then cooledto
À
100
8
C at a steady rate of 10
8
C/min using liquidnitrogen as the cooling medium. The heat
ow (W/g)versus temperature for each experiment was analyzedto determine wax appearance temperature (
C1
,
8
C)and onset temperature (
C2
,
8
C) corresponding to thesecond freezing peak (Pk-2). Average value of threeindependent measurements was taken in each case.The temperature repeatability was
Æ
0.7
8
C.
2.3. NMR experiments
Over the years, quantitative NMR spectroscopy hasproved quite useful in elucidating the complex mole-cular structure of mineral[16
18]and vegetable oils[19,20].
1
H NMR data of thevegetable oils were recor-ded in Fourier Transform mode on an AMX 400 MHzBRUKER machine at an observing frequency of 400 MHz using 5 mm dual nuclei probe. Deuterated
192
A. Adhvaryu et al./Thermochimica Acta 395 (2003) 191–200
 
chloroform (CDCl
3
) was used to prepare the samplesolution (15%, w/v) containing 1% tetramethylsilane(TMS)asaninternalstandard.Apulsedelay(D1)of1 s,acquisition time (AQ) 6.8 s and pulsewidth (P1) 8.5
m
swas used for 16 repeated scans (NS). Further details of the experiments are reported in an earlier publication[21].
2.4. Molecular modeling
A molecular modeling program (Spartan Pro) wasused to build representative triacylglycerol moleculesusing a minimum energy calculation approach[22].Modi
cations were made in the fatty acid chains bychanging the oleic or linoleic acid moiety. Differenttriacylglycerol molecules were designed having botholeic chains, both linoleic chains and one each of oleic and linoleic chain on the two end side of 
tuningfork 
structure, while the stem side had either oleic orlinoleic chain. Geometrical and energy calculationson bond angle, planer distance between hydrogensin the same and adjacent chains, molecular energyassociated with a sterically hindered geometry andconformation etc. were done on individual molecules.
2.5. Statistical metho
Statistical analysis was used to develop correlationsbased on vegetable oil structural data obtained from
1
H NMR analysis and
C
determined from the DSCcooling curve. A professional version of Minitab
1
12statistical program was used in the current study. Theprogram analyses individual NMR derived predictorvariables one at a time and determines its in
uence onthe
C
. The
p
-values were calculated and
-tests weredone to identify variables that imparted maximumin
uence on
C
. The
nal model included those vari-ables that showed the best correlation on the basisof their individual and collective in
uence on waxformation temperature.
3. Results and discussion
3.1. Analysis of sub-ambient DSC data
Fig. 1shows the conventional DSC trace for HOsa
r obtained by cooling the sample from
þ
50 to
À
100
8
C at the rate of 10
8
C/min. Previous studieshaveshownthatcoolingratecanin
uencetheshapeof the curve as well as the corresponding temperatureobtained[12].Table 1presents the wax appearance temperatures (
C1
,
8
C) of SBO corresponding to dif-ferent cooling rates. It was observed that at a lowercooling rate (1
5
8
C/min) the signal-to-noise ratio islow and the onset temperature is hard to estimate.ThesensitivityofDSCmeasurementisdirectlyrelatedto heating and cooling rate used[23]. Thereforesensitivity of the instrument at a lower cooling ratecan make the identi
cation of wax appearance tem-perature dif 
cult. The wax appearance onset tempera-ture is relatively well de
ned from the base line at ahigher cooling rate. With an increase in the rate of cooling,
C1
progressively shows lower value and thedelta change is small compared to low cooling rate.
Fig. 1. Cooling curve of HO sa
r oil at a cooling rate of 10
8
C/minfrom
þ
50 to
À
100
8
C showing the exothermic peaks Pk-1 and Pk-2and corresponding
C1
and
C2
.Table 1Wax appearance temperature of SBO at different cooling ratesrange
¼ þ
50 to
À
50
8
C, sample size: 9.9646 mg
a
BasefluidCooling rate(
8
C/min)Wax appearancetemperature
C1
(
8
C)Delta changein
C1
(
8
C)SBO 1.0
À
8.34
SBO 2.0
À
9.29 11.39SBO 3.0
À
10.26 10.44SBO 5.0
À
11.07 7.89SBO 7.0
À
12.13 9.57SBO 10.0
À
13.16 8.49SBO 12.0
À
14.14 7.44SBO 15.0
À
14.52 2.68
a
DSC data reported is the average value of duplicateexperiments.
 A. Adhvaryu et al./Thermochimica Acta 395 (2003) 191
 – 
200
193

You're Reading a Free Preview

Download
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->