Professional Documents
Culture Documents
Geotechnical Properties of Fly and Botom Ash Mixtures For Use in Highway Embankments
Geotechnical Properties of Fly and Botom Ash Mixtures For Use in Highway Embankments
Abstract: Class F fly ash and bottom ash are the solid residue byproducts produced by coal-burning electric utilities. They are usually
disposed of together as a waste in utility disposal sites with a typical disposal rate of 80% fly ash and 20% bottom ash. Direct use of these
Downloaded from ascelibrary.org by XAVIER UNIVERSITY on 09/01/16. Copyright ASCE. For personal use only; all rights reserved.
materials in construction projects consuming large volumes of materials, such as highway embankment construction, not only provides a
promising solution to the disposal problem, but also an economic alternative to the use of traditional materials. Representative samples of
class F fly and bottom ash were collected from two utility power plants in Indiana and tested for their mechanical properties 共compaction,
permeability, strength, stiffness, and compressibility兲. Three mixtures of fly and bottom ash with different mixture ratios 共i.e., 50, 75, and
100% fly ash content by weight兲 were prepared for testing. Test results indicated that ash mixtures compare favorably with conventional
granular materials.
DOI: 10.1061/共ASCE兲1090-0241共2005兲131:7共914兲
CE Database subject headings: Fly ash; Bottom ash; Disposal; Mixtures; Highway construction; Embankments.
in a manner similar to that used for bottom ash samples from the
32%, respectively. For Western U.S. Class C fly ash, the values of
Wabash River plant. Since at the A. B. Brown plant fly and bot-
the maximum dry density and the optimum water content varied tom ash are discharged separately through separate pipelines, the
from 13.0 to 18.7 kN/ m3 and from 11 to 19%, respectively. They fly and bottom ash samples were extracted from near their respec-
concluded that, nationwide, the values of the optimum water con- tive discharge locations near the margins of the disposal pond.
tent and maximum dry density spanned extremely wide ranges
and that the large variations in the values were mainly due to the
variation of the fly ash itself, which exhibited different chemical
Testing Methods
and physical characteristics depending on factors such as the
source of coal and the condition of coal combustion.
McLaren and DiGioia 共1987兲 investigated the shear strength of Ash Characterization
fly ash. They compiled a database of shear strength test results for
51 Class F fly ash samples. According to these authors, Class F Grain Size Analysis
fly ash is a frictional material, deriving its shear strength mainly Grain size analysis was performed on fly ash, bottom ash, and
from friction between particles, whereas Class C fly ash can ex- fly/bottom ash mixtures. The fly ash contents of the mixtures were
hibit considerable true cohesive strength due to cementitious re- equal to 50 and 75%. The bottom ash samples obtained from the
Wabash plant contained a small percentage of fines. The bottom
actions.
ash samples were first sieved to exclude the fines portion in the
Seals et al. 共1972兲 and Usmen 共1977兲 presented data obtained
samples prior to testing and only the particles retained on the No.
from West Virginia bottom ash. The standard Proctor maximum
200 sieve 共0.075 mm兲 were used as bottom ash. The gradations of
densities varied between 11.6 and 18.4 kN/ m3; the optimum
the samples, were determined using ASTM D422 共1963兲.
water contents ranged from 12 to 34%. Majidzadeh et al. 共1977兲
reported that the optimum water content of each ash actually oc- Microscopic Examination
curred within a range rather than exhibiting a clear optimum Fly ash and bottom ash samples were examined using a scanning
value. Seals et al. 共1972兲 also performed a series of one- electron microscope 共SEM兲 and a light microscope 共LM兲.
dimensional compression tests on West Virginia bottom ash. They
showed that, at low stress levels, the compressibility of bottom Specific Gravity
ash was comparable to natural granular soils placed at the same The specific gravity of fly and bottom ash was determined using
relative density. ASTM D854 共2000兲 共Method A兲. To prevent the removal of the
Huang 共1990兲 investigated the shear strength of Indiana bot- fly ash particles of lower specific gravity, de-airing was accom-
tom ash and boiler slag compacted to different densities using plished by very slow heating.
direct shear testing. The reported friction angles varied in a wide
range from 35 to 55°, depending on the density.
Mechanical Properties of Ash Mixtures
Standard compaction, hydraulic conductivity, one-dimensional
Testing Materials compression, and drained triaxial tests were performed on the
fly/bottom ash mixtures with fly ash contents of 50, 75, and
100%. The particles coarser than the No. 200 sieve 共0.075 mm兲
Ash Sources and Sampling were used as bottom ash when preparing the mixtures.
Ash Sources Compaction
The ash samples used in this study were extracted from two Standard compaction tests were performed following ASTM
power plants in Indiana, United States: The Wabash River plant D698 共2000兲. Premeasured quantities of fly ash and bottom ash
and the A. B. Brown plant. Both power plants produce Class F fly were mixed slowly by hand at first, and then water was sprayed
ash and bottom ash with a typical production ratio of 80% fly ash on gradually while the mixing was continued in a mortar mixer.
and 20% bottom ash. At the Wabash River power plant, fly ash Samples were then compacted in a 4 in. diameter mold.
and bottom ash are co-disposed, existing in the form of mixtures
at a disposal pond. At the A. B. Brown plant, fly and bottom ash Hydraulic Conductivity
are conveyed through separate pipelines and discharged into sepa- The hydraulic conductivity of the ash mixtures was measured by
rate locations within a disposal pond. As a result, the disposal falling head tests using a rigid-wall compaction-mold permeame-
One-Dimensional Compression
One-dimensional compression tests were performed on samples
prepared at their optimum water content, and compacted using a
standard manual rammer to R = 95% in a split mold with a diam-
eter of 72 mm 共2.8 in.兲 and a height of 55 mm 共2.2 in.兲. The tests
were conducted according to the normal consolidation test proce-
Downloaded from ascelibrary.org by XAVIER UNIVERSITY on 09/01/16. Copyright ASCE. For personal use only; all rights reserved.
ing a large variation between ash sources. The wide range in viously, a fly ash containing a large percentage of hollow particles
specific gravity can be attributed to two factors: 共1兲 Chemical would have a lower apparent specific gravity than one with
composition, and 共2兲 presence of hollow fly ash particles or par- mostly solid particles. In fact, the two factors affecting the spe-
ticles of bottom ash with porous or vesicular textures. The low cific gravity of fly ash may be related. Guo et al. 共1996兲 examined
specific gravities of Wabash plant fly and bottom ash are ex- the chemical compositions of hollow and solid fly ash particles
plained by their low iron oxide contents and, conversely, the high separately, and the data revealed that hollow-particle fly ash had a
specific gravities of Brown plant fly and bottom ash by their high significantly lower iron content 共4.5%兲 than solid-particle fly ash
iron oxide contents. Different amounts of hollow particles present 共25.1%兲.
in fly ash also cause a variation in apparent specific gravity. Ob- The apparent specific gravity of bottom ash is also affected by
Fig. 2. Scanning electron microscopy photomicrographs of fly ash from 共a兲 and 共b兲 Wabash River plant; 共c兲 and 共d兲 A. B. Brown plant
Compaction Behavior
Fig. 4 shows the compacted dry unit weight versus the water
content curves of the ash mixtures. The results show that, as the
fly ash content 共F兲 increased from 50 to 100%, the maximum dry
unit weight 共␥d,max兲 decreased, while the optimum water content
共wopt兲 increased. The studies of silty sands have revealed that, for
low nonplastic silt contents ranging from zero to about 25%, both
the ␥d, max and ␥d, min of a silty sand increase with increasing fines
content because the fines occupy the voids between sand par-
ticles, whereas further increases in the fines content, exceeding
about 25%, cause the fines to begin to separate adjacent sand
Fig. 3. Light microscope photomicrographs of bottom ash from 共a兲 particles, resulting in a decrease in ␥d, max and ␥d, min 共Kuerbis et
and 共b兲 Wabash River plant; 共c兲 A. B. Brown plant al. 1988; Lade and Yamamuro 1997; Salgado et al. 2000兲. Simi-
larly, in the ash mixtures with high fly ash content 共i.e., F
⬎ 50%兲, the bottom ash particles are not, on average, in contact.
the porosity of its particles. Comparing the specific gravities re- At a certain level of fly ash content, the bottom ash particles may
ported in Table 2 for fly ash and bottom ash from the Brown be completely separated, floating in a fly ash matrix. The behavior
plant, the bottom ash has a lower specific gravity than the fly ash, of a material with a floating fabric 共Salgado et al. 2000兲 may be
although Table 3 indicates that a slightly higher iron content may quite different from one in which the bottom ash particles are in
be present in the bottom ash. This may be due to the presence of contact.
highly porous popcorn-like bottom ash particles. From another perspective, the gradations of the ash mixtures
varying with different mixture ratios also explain the change in
dry unit weight. The addition of bottom ash to fly ash leads to
Table 2. Specific Gravity of Fly and Bottom Ash increasingly more well-graded size distributions, which allows
Ash source Fly ash Bottom ash the fly and bottom ash particles to pack more closely, resulting in
the increase in ␥d, max. The higher wopt associated with higher F
Wabash River plant 2.30 2.32
values follows from the need to release the capillary tension from
A. B. Brown plant 2.81 2.62
the greater exposed surface of the finer fly ash particles.
Although the same trends were observed for wopt and ␥d, max The larger specific surface of fly ash causes more resistance to
with respect to F, the values of ␥d, max of the ash mixtures were flow of water through the voids. Huang 共1990兲 performed a series
very different between the two ash sources studied. The differ- of hydraulic conductivity tests on Indiana bottom ashes. He ob-
ences appeared to be primarily due to the large variations in the served that the fines included in bottom ash had a predominant
specific gravities. The Brown plant ash mixtures, whose specific effect on the permeability, and thus the hydraulic conductivity
gravities were much higher than those of the Wabash plant ash decreased as the fine contents increased.
mixtures, had higher ␥d, max values. It was also often observed,
especially in the Brown plant ash mixtures, that some weak large
bottom ash particles were broken down into finer particles by Compressibility
compaction. Some bottom ash particle crushing during compac- Fig. 5 shows the one-dimensional compression curves of the com-
tion may have contributed to the increase in the ␥d, max of ash pacted ash mixture samples with F = 50, 75, and 100% from the
mixtures. two ash sources. Regardless of the ash source, a general observed
Compared with the ␥d, max of compacted soils, the ␥d, max val- trend was that as F decreased from 100 to 50% 共i.e., as B
ues of ash mixtures tended to be lower than those of soils, which increased from zero to 50%兲, the ash samples became slightly
range typically from 17 to 20 kN/ m3 共U.S. Navy 1986兲. more compressible. Two possible mechanisms may explain the
increasing compressibility with decreasing F 共i.e., with increasing
Hydraulic Conductivity B兲: 共1兲 angularity and porous surface texture of bottom ash par-
ticles, and 共2兲 particle crushing. Bottom ash particles are mostly
Table 4 shows the values of hydraulic conductivity for compacted
angular and irregular in shape. Granular materials with angular
ash mixtures with F = 50, 75, and 100%. The measured values
particles are typically more compressible than those with well-
varied from 1 ⫻ 10−7 to 3 ⫻ 10−8 m / s, indicating a range similar
rounded particles because the sharp edges in the angular particles
to those of fine sand/silt mixtures or silt. As F increased from 50
tend to be overstressed during increases in confining stress 共and
to 100%, the values of hydraulic conductivity decreased gradu-
ally, and the decreasing rate with increasing F was essentially the shear stress as well兲, and thus may break during compression
same for both the Wabash plant and the Brown plant ash mix- 共Roberts and DeSouza 1958; Schultze and Moussa 1961兲. Particu-
tures. larly weak bottom ash particles, which in most cases have internal
pores, can break at relatively low stress levels, as discussed by
Huang 共1990兲. Furthermore, it is possible that fine fly ash par-
ticles adjacent to or filling the external pores of the bottom ash
Table 4. Hydraulic Conductivity of Ash Mixtures
particles may be squeezed out with increasing stress, resulting in
Hydraulic conductivity an increase in deformation.
Ash source Mixture composition 共m / s兲 The Brown plant ash mixture samples exhibited greater com-
Wabash River plant F100 3 ⫻ 10−8 pressibility than the Wabash plant ash mixture samples. The dif-
F75 B25 6 ⫻ 10−8 ference in the compressibility appears to be mainly due to differ-
F50 B50 1 ⫻ 10−7 ent compressibility of the fly ash rather than the bottom ash,
because the increasing rate in the compressibility with increasing
A. B. Brown plant F100 6 ⫻ 10−8
B was similar between the two ash sources. Relatively high com-
F75 B25 9 ⫻ 10−8
pressibility in the Brown plant fly ash may be attributed to the
F50 B50 1 ⫻ 10−7
breaking of particle agglomerations, which were more abundant
Shear Strength
Fig. 7. Stress-strain and volume change results from consolidated-drained triaxial tests completed on ash mixtures from 共a兲 and 共b兲 Wabash River
plant; and 共c兲 and 共d兲 A. B. Brown plant
deviatoric stress ⬘d were associated with a slight initial volumet- 100 kPa confining stress. At ⬘3 = 200 kPa, the volumetric behav-
ric contraction, followed by a gradually increasing rate of volume ior became predominantly 共for the Wabash River plant ash兲 and
expansion 共dilation兲. The peak strength occurred when the rate of entirely 共for the Brown plant ash兲 contractive.
change of the volumetric strain with respect to the axial strain
dv / da reached its maximum value. The postpeak reduction in Peak Friction Angle and Critical State Friction Angle
⬘d was associated with a decreased rate of dilation until the stress Ash materials are, in general, frictional materials. The peak fric-
state reached the critical state with constant stress and volume. tion angle 共⬘p兲 can be expressed in terms of the principal effec-
The specimens at R = 90% behaved similar to a sandy soil in a tive stresses at peak state based on the Mohr–Coulomb failure
loose state. The d⬘ increased gradually up to a peak level and then criterion with zero-cohesion intercept as follows:
stayed practically unchanged with increasing a. The volumetric
冢 冣
strains 共v兲 were contractive throughout shearing. When compar- ⬘1
−1
ing the response for F = 100% and F = 75%, at a compaction ratio ⬘3
R = 95%, as F decreased from 100 to 75%, the peak deviatoric sin ⬘p =
1⬘
⬘
stress d共peak兲 increased slightly while both the maximum rate of +1
⬘3 peak
dilation and the maximum dilation decreased slightly. For R
= 90%, v increased with decreasing F. where 1⬘ / 3⬘⫽effective principal stress ratio or stress obliquity.
Fig. 8 shows the effect of the confining stress ⬘3 共50 to 200 For dilative behavior, ⬘p is associated with the maximum rate of
kPa兲 on the volumetric response of the R = 95% , F = 75% ash dilation, which normally develops at relatively small strains. For
mixtures. Dilation decreased gradually as ⬘3 increased from 50 to ideal contractive behavior, ⬘p coincides with the critical state
Notation
Table 5. Critical State Friction Angles of Ash Mixtures
The following symbols are used in this paper:
Critical state
B ⫽ bottom ash content;
friction angle,
F ⫽ fly ash content;
Mixture ⬘
critical
Ash source composition 共degrees兲 R ⫽ relative compaction;
wopt ⫽ optimum water content;
Wabash River plant F100 28 ␥d, max ⫽ maximum dry unit weight;
F75 B25 30 ␥d, min ⫽ minimum dry unit weight;
F50 B50 32 a ⫽ axial strain;
A.B. Brown plant F100 32 v ⫽ volumetric strain;
F75 B25 33 ⬘3 ⫽ confining stress;
F50 B50 35 ⬘d ⫽ deviatoric stress;
Note: Ottawa sand: 29° 共Salgado et al. 2000兲; Berlin sand: 33° 共Bolton ⬘c ⫽ critical state friction angle.
1986兲; Monterey No. 0 sand: 37° 共Bolton 1986兲. ⬘p ⫽ peak friction angle; and
test methods for laboratory compaction characteristics of soil using liquefaction of sands.” Can. Geotech. J. 34共6兲, 918–928.
standard effort 关关12, 400 ft lbf/ ft3 共600 kN m / m3兲兴.” Designation Majidzadeh, K., El-Mitiny, R. N., and Bokowski, G. 共1977兲. “Power
D698-00a, Philadelphia. plant bottom ash in black base and bituminous surfacing.” Vol. 2,
American Society for Testing and Materials 共ASTM兲. 共1995兲. “Standard User’s Manual, Federal Highway Administration, Report No. FHWA-
test method for measurement of hydraulic conductivity of porous ma- RD-78-148, Washington, D. C.
terial using a rigid-wall, compaction-mold permeameter.” Designation McLaren, R. J., and DiGioia, A. M. 共1987兲. “The typical engineering
D5856-95, Philadelphia. properties of fly ash.” Proc., Geotechnical Practice for Waste Dis-
American Society for Testing and Materials 共ASTM兲. 共1996兲. “Standard posal ’87, Geotechnical Special Publication No. 13, E. Wood, ed.,
test method for one-dimensional consolidation properties of soils.”
ASCE, New York, 683–697.
Designation D2435-96, Philadelphia.
Roberts, J. E., and DeSouza, J. M. 共1958兲. “The compressibility of
Bolton, M. D. 共1986兲. “The strength and dilatancy of sands.” Geotech-
sands.” Proc., American Society for Testing and Materials, 58, 1269–
nique, 36共1兲, 65–78.
Carrier, W. D., III. 共2000兲. “Compressibility of a compacted sand.” J. 1277.
Geotech. Geoenviron. Eng. 126共3兲, 273–275. Salgado, R., Bandini, P., and Karim, A. 共2000兲. “Shear strength and stiff-
Diamond, S. 共1985兲. “Selection and use of fly ash for highway concrete.” ness of silty sand.” J. Geotech. Geoenviron. Eng. 126共5兲, 451–462.
Report No. FHWA/IN/JHRP-85/8, Final Report, Purdue Univ., W. Schultze, E., and Moussa, A. 共1961兲. “Factors affecting the compressibil-
Lafayette, Ind. ity of sand.” Proc., 5th Int. Conf. Soil Mechanics and Foundation
DiGioia, A. M., McLaren, R. J., Burns, D. L., and Miller, D. E. 共1986兲. Eng., Vol. 1, 335–340.
“Fly ash design manual for road and site application, Vol. 1: Dry or Seals, R. K., Moulton, L. K., and Ruth, B. E. 共1972兲. “Bottom ash: An
conditioned placement.” Manual Prepared for EPRI, CS-4419, Re- engineering material.” J. Soil Mech. Found. Div., 98共4兲, 311–325.
search Project 2422-2, Interim Report, Electric Power Research Insti- Selvig, W. A., and Gibson, F. H. 共1956兲. “Analysis of ash from United
tute, Palo Alto, Calif. States coals.” Bulletin 567, Bureau of Mines.
Dry, C. M. 共1995兲. “The potential use of waste 共ash兲 materials for EIFS U.S. Navy. 共1986兲. “Design manual—Soil mechanics, foundations, and
insulation and related components.” Development, Use, and Perfor- earth structures.” NAVFAC DM-7, Dept. of the Navy, Washington,
mance of Exterior Insulation and Finish Systems 共EIFS兲, ASTM STP D.C.
1187, Mark F. Williams and Richard G. Lampo, eds., ASTM, Phila- Usmen, M. A. 共1977兲. “A critical review of the applicability of conven-
delphia. tional test methods and materials specifications to the use of coal-
Engineering News Record 共ENR兲. 共1980兲. “Billions at stake in coal associated wastes in pavement construction.” PhD dissertation, West
fight.” Eng. News-Rec., 204, 2. Virginia Univ., Morgantown, W.Va.