Indian Journal of Engineering & Materials SciencesVol. 4, December 1997, pp.271-281

Physico-chemical characteristics of fly ash samples from thermal powerplants of Orissa

R Mohapatra & S B Kanungo*Regional Research Laboratory, Bhubaneswar 751 013, India

Received I I November 1996; accepted 2 June 1997

Some physico-chemical characteristics of three fly ash samples have been studied with an objectiveto create a data base for different fly ash samples in the country. It has been observed that relevantphysical properties of the samples show systematic variation with particle size distribution. Thus, whileapparent density and optical or geometrical area increase with fmeness of the particles from sample I to3, true density, voidage and void volume exhibit opposite behaviour. Batch leaching studies indicatethat whereas for samples I and 2 most of the trace metals (except Cu) leach out rapidly from the surfaceof the ash particles in the pH range 6.4-7.5, under acidic conditions (pH - 4.5) the rates of leaching ofmost trace metals, except Co and Cd, are relatively slow for sample 3. The pH values of zero charge ofthe three major oxide constituents namely, Si02, Fe203 and Al203 as determined by the solid additionmethod, conform more or less to the litrature values of 2.5, 7.0 and 9.5 respectively. Viscositymeasurements of aqueous suspension indicates that sample 3 is least hydrated in water and thereforeshows low viscosity value even at high concentration of slurry.

Fly ash is the product of combustion of coal inboiler furnace of the thermal power plants. Becauseof the abundant reserve of coal in India, thermalpower contributes more than 75% of the totalpower generation in the country. As more andmore thermal power plants are being set up in thecountry to meet the ever increasing demand ofpower, it is estimated that 115-120 million tonnesof fly ash will be genrated annually in the countryeven before the end of this centuryl. Unless asubstantial portion of this fly ash fmds some use,this will lead to a very serious environmental crisis.

However, during the last several years, the R&Defforts on the utilization of fly ash has gainedconsiderable momentum and several processesleading to the bulk use of fly ash are likely to beput into commercial exploitationl.2• Some of theimportant possible uses of fly ash are in theproduction of building materials (bricks, cement,concrete etc), land and mine fill, construction ofroads and highways, embankments etc. For any ofthe above possible uses of fly ash, the material hasto be characterized and standardized scrupulouslyand all the fly ash produced in the contry must becategorized accordng to tht:;ir physico-chemicalproperties. It is therefore imperative to create a

*For correspondence

detailed data base for the characteristics of fly ashgenerated in all the major thermal power plants inthe country. The present work is a part of the on­going in-house project of the laboratory on thecharacterization and utilization of fly ash fromdifferent thermal power plants in the country. Inthe first instant, the work was initiated witH fly ashsamples from two thermal power plants in Orissa,namely, Talcher and National AluminiumCompany (NALCO), situated in the district ofAngul, about 130 km from Bhubaneswar. Both theplants use coal from Ta1cher Coal Field, but thetechnology of ~ombustion used in the two plants isprobably different. The results of characterizationstudies carried out on fly ash samples of these twothermal power plants are presented here.

Experimental Procedure

Materials

Altogether three samples have been used in thepresent work. Two of these have been collectedfrom Talcher thermal power plant and are:(i) sample collected from hopper below theelectrostatic precipitator (ESP), before coming incontact with water jet injecton·; (ii) ash samplecollected from the settling pond. The third sample

272 INDIAN J. EN&-MATER. SCI., DECEMBER 1997

~ SO

Particle size) /Jm

Fig. I-Particle size distribution of the three fly ash samples

(sample 3) was from the thermal power plant of theNational Aluminium Company (NALCO). Thissample was also collected under dry conditionfrom the hopper below the ESP. The physico­chemical propertis of the ashes obtained from coalcombustion are preserved in such sample, unlikepond ash, where its prolonged contact with watermay lead to variation in the properties.

Voidage(-) 0.5560.5460.383

Void volume,

0.5840.5740.393J ·1 cm gGeometric surface

17544400121862 ·1 area, cm gParticle size distribution(a) 50% finer

3096

than, Ilm (b) 80% finer561710

than, Ilm

Table I-Physical properties of fly ash samples

Properties Sample I Sample 2 Sam~e3

True density, g cm,J 2.09 2.09 1.58Apparent or bulk 0.941 0.950 0.975density, g cm'J

times of gentle and uniform tapping and noting thevolume with an accuracy of ±0.2 cm3. The meanvalue of triplicate measurement was taken. Thesuspension densities for different solid concent­rations in water were determined in a 25 mLpycnometer at 30°C.

Viscosity-Viscosity values of differentconcentrations of solids in water were determined

by a Brookfield viscometer at 30°C and 60 rpm ofspindle rotation.

Microscopy--The fly ash samples wereexamined under transmitted light in a Leitzpolarizing microscope (Germany). The sampleswere also examined under a scanning electronmicroscope of mOL make (Japan).

X-ray difJraction-X-ray diffraction pattern ofthe samples were taken in a Phillips diffractometerusing CuKa radiation.

Point of zero charge-Points of zero charge(PHpzc) of the samples were determined by thesolid addition method" as follows:

In a series of 100 mL plastic bottles, 45 mL ofNaCI04 solution (10.1 or lO'a M) was transferred.The pH values of the suspensions wereapproximately varied from 3.0 to 10.0 by addingknown volume of either 0.1 N HCI04 or 0.1 NNaOH. The total volume of the suspension in eachbottle was made up to 50 mL by adding the balanceelectrolyte solution. The pH values were againnoted and designated as initial pH or pHi 0.5 g ofaccurately weighed sample was added to eachbottle which was immediately closed to avoid thecontact with atmospheric CO2, The suspensionswere allowed to equilibrate for 72 h withintermittent shaking. The final pH values were then

/ "

/,','jl

I ,. I

I ,, ,, Ii '

I / / 5ampl!! No.Q/:'---~/// ~ ~~ -'-:1 3/ ........L-~.,. 100 200OLIO1

100

•..Cll

C

,eo

(1:>

E:>u

Methods

The::three samples have been characterized bythe following methods.

Chemical analysis-While the major consti­tuents such as Si02, A1203,CaO, MgO, Fe203 etcwere determined by the procedure recommendedby both Indian (IS: 3812) and Japanese standards(nSA 6210), the minor constituents and traceelements were determined by either atomicabsorption spectrophotometry. (AAS) or induc­tively coupled plasma emission spectrophoto­metry (ICPES).

Particle size-The particle size distributionpattern was determined with the help of aMALVERN particle size analyzer (UK, model3600).

Surface area-The particle size analyzer alsoprovides specific surface area per unit volume fromwhich area per unit mass also can be calculatedusing the density data. Surface area determined bythis method is essentially outer or geometric areaand is closely similar to 'Blaine area' obtained byair permeability metho<f.

Density- True densities of the sample weredetermined by the displacement of toluene in a 25mL pycnometer at 30°C. The apparent densityvalues were determined by packing 5 g of bulksample in a 20 mL graduated cylinder after 15-20

MOHAP ATRA & KANUNGO: PHYSICO-CHEMICAL CHARACTERISTICS OF FLY ASH

·-

273

Table 2-Chemical composition of the thtee fly ash samples

Constituents (%)

Sample ISample 2Sample 3

Si02

57.0058.9256.00

Alp)

29.4828.9033.07

Fe2O)

6.205.004.84

CaO1.001.100.27

MgO

0.350.322.10

Ti02

1.902.041.10

Na20

0.150.240.08

Kp

0.410.400.12

P201

0.300.300.09

504 -

0.0640.0000.065Lor

3.00~.151.60Fig. 2-Photomicrograph showing the presence of fine unbumt carbon particles in sample I.

Total99.85499.3799.335

Indian specification (IS-3812) the Blaine area offly ash should be higher than 2800 cm2/g.Therefore, both particle size distribution andsurface area data corroborate each other on thefineness of the samples.

Density measurement-The above observationsare also corroborated from density measurements.While the true or particle density value of samples1 and 2 is same, i.e. 2.09 g cm-3, sample 3 showsvery low value of about 1.58 g cm-3.

Both particle size and microscopic obvservation(see below) reveal that sample 3 consists of veryfine particles, most of which tend to float whensprinkled on water. All these suggest that thepresence of hollow cenospheres is morewidespread in this sample than in the samples 1and 2, thereby showing low value of particledensity. The bulk density values of the threesamples follow the similar pattern of variation aswith particle size distribution, the sample ofcoarsest size showing lowest value of bulk orapparent density because of larger interparticlevoid space. The results are shown in Table 1.

Since fly ash is the product of high temperaturemolten phase of alumino silicate compounds(glass), the internal porosity and pore volume ~revery small in such materials. Therefore, it is moreappropriate to use the terms voidage (&) and voidvolume ('1/), instead of porosity and pore volume.The '1/ and & values can be estimated from therelations5•

1 1'1/=--- ...(1)

Ph Ptrue

noted and designated hereafter as pHf• pHpzc

corresponds to the condition when pHj=pHf.Leaching in aqueous medium-In a series of

100 mL plastic bottles, 50 mL of distilled waterwas placed. Two grams of accurately weighedsample were added to each bottle which were thenclosed immediately. The suspensions were allowedto equilibrate for different periods of time rangingfrom 24 h to 30 days with intermittent shaking.After the desired periods of equilibration, thesuspensions were filtered and the liberated ionicspecies were determined by either AAS or IePES.

Results and Discussion

Physical CharacteristicsParticle size distribution-Particle size dis­

tribution or fineness is probably one of the mostimportant characteristics which influence theactivity of fly ash more than any other physicalproperties. Fig. 1 shows the size distributionpatterns of three samples of fly ash. Sample 1shows a wide range of distribution from 85-6 I-lm,

whereas sample 2 shows a narrow range from 28-4I-lm. The finer size of the pond ash indicate thatlarger particles had settled to a depth above whichthe present sample was collected. Sample 3 is ofextremely fine variety with a distribution rangefrom 15 to 1 I-lm or even less. The results aresummarized in Table 1.

The optical surface area obained from theMAL \TERN size analyzer is essentially theexternal area of the material of assorted size range.The values for the three samples are 1754, 4400

and 12186 cm2jg respectively. According tolhe&=1-~

Ptrue... (2)

274 INDIAN 1. ENG. MATER.SCl., DECEMBER 1997

Fig. 3-SEM photographs of sample 1 showing-(a) A typicalspherical particle of glassy material. (b) Agglomeration ofseveral spheres to form "pleospheres" or "cenospheres". (c),(d) Relatively low melting (temperature) phase surrounded bypartially molten high temperature phase. Fine particles ofrefractory oxides or iron oxides are seen adhered to the outersurface of large spheres.

where A and Ptrue are the bulk and true (or particle)densities respectively in g cm-). The results givenin Table I indicate that If and G values decreasewith decrease in average particle size from sampleI to 3.

Microscopic observation-Extensive micro­copic observation under the transmitted light showthat all the fly ash samples contain fine unbumtcarbon particles of irregular shape and occurs tothe extent of about 3-4% by weight (Fig. 2). Thenon-carbon ash consists of non-crystalline glassymaterials, generally spherical in shape due tosurface tension and other hydrodynamic effect of

the environment. Scan~ing electron micrographsshow that all the three samples commonly consist

of glassy spheres of assorted size range with verysmall quantity of crystalline materials. Fig. 3aillustrates the SEM photograph of a typicalspherical particle constituting the fly ash sample Ifrom Talcher thermal power plant. The size rangeof such spheres, however, varies widely from I to­100 J.l.m.The two important features that can benoted from such spheres are--(i) the presence ofcertain granular deposits possibly of somerefractory oxides or iron oxide or even unbumtcarbon, (ii) the presence of holes or vesicles asidentified by black spots on the surface6• Thesevesicles or holes are believed to be formed by thedifferential cooling when a low melting phase(NazO/KzO/CaO-SiOz-AlzO) solidifies earlier

MOHAP ATRA & KANUNGO: PHYSICo.CHEMICAL CHARACTERISTICS OF FLY ASH 275

Fig. 5-X-ray diffraction patterns of the three fly ash samplesalong with the diffraction patterns of some typical hightemperature minerals.

SAMPLE

(3 ,--S.A.MPl,.-E..(2 )---SAMPl E(I)

FAYALIT E(20- 1139)---MAGNETITE(19 - 629)---HERCYNITE

'~ NE.PHF. LINE

{ll -19131---ARNEGIEITt(11- 220)---MUUITE.(15-176)----SllllMAN;TE(10- 3691

C'ORUNO'UM(<L-AlzO,)~

~rRIOYMITE

{l4 - 260)10

U

..>

6-0 5- 0 4-0

E..0::

Fig. 4-SEM photographs of sample 3 showing-(a) La: redstructure of a large aluminosilicate mineral pO&v.blycontaining high iron, (b) Lathe-like structure of a decomposedpyrite grain dislodged from coal

than the overall molten droplets (SiOrAlzO)). Fig.3b shows a typical agglomerated spheres (ceno­spheres). Fig. 3c and 3d demonstrates the partialmelting of a high temperature agglomerated phase.Relatively low temperature phase therein cools andsolidifies earlier forming spheroidal particles andholes in the major matrix.6 ..

On the other hand the fly ash sample 3 consistsof very fine spherical particles ranging from I to30 /lm in size. These spheres generally revealsrelatively smooth surfaces, though, some alsocontain vesicles or holes. Fig. 4a is an unusuallylarge particle having a layered structure. Thisprobably represents a partially convertedaluminosilicate mineral rich in iron. The lathe like

structure in Fig. 4b possibly represents a partiallydecomposed pyrite grain dislodged from coal. But,the occurrence of such material is rather rare in flyash.

X-ray diffraction-Mineral matters present incoal directly control the mineralogy of fly ash.However, during the combustion of coal, itsmineral matters undergo considerable change.Though XRD is the most frequently used methodto determine the mineralogical characteristics of flyash, quite a number of reflections are difficult to beassigned unambiguously to any distinct mineral.This is because during the ash fusion a variety ofaluminosilicate compounds are forme<!_for whichstandard literature references are not avail~b~el.Theproblem becomes further complicated as many ofthese compounds are non-crystalline (glass) forwhich no XRD pattern can be obtained. Fig. 5suggests that the major minerals present in Talcherfly ash are corrundum (a-AlzO)), mullite,;illimanite, tridymite and small amounts ofmagnetite and fayalite. But those minerals whichare present in 5% or less by weight in the sampleare difficult to be identified by XRD method. Foridentifying such minerals chemical analysis datacombined with careful microscopic observationsare employed.

276 INDIAN J. ENG. MATER. seI., DECEMBER 1997

5.0

Fig. 6--Variation of pH values ofleachates of the three fly ashsamples with time ofleaching.

'" Sa"", •• 1

OJ·' 2o IJ )

8 12 16 20 l' 28 12Periodof leaching, days

8.0

Constituents Sample ISample 2Sample 3ppm Ni

40(18-76)*4249Cu

212026Mn

757736Co

82 (23)8849Ph

75(63.7)*7086Zn

485053Cd

549Cr

67(45-160)*13562Hg

0.10.060.08*Values reported in ref. 9.

Table 3-Analysis of trace elements in the three fly ashsamples .

Chemical Analysis

Initially spectrographic measurements werecarried out to detect the presence of various traceelements in the samples. The results indicate thatwhile Cr, Ni, Co and V occur in appreciableamount (~ 80-100 ppm), Be, B, Cu, Ga, Se etc arepresent in trace quantities (~ 20 ppm). Besidesthese, there are many elements which are present inminute traces.

Tables 2 and 3 depict the major and minorconstituents of the fly ash samples respectively. Itcan be seen from the data that the three majorconstituents namely, Si02, Al203 and Fe203 areclosely similar within the acceptable (IS-3812)range and are typical of the ash samples collectedfrom different parts of Indial•2. This is becausealmost all the thermal power plants use high ash(35-40%) coals in their boiler. However, sample 3contains higher AI203 content but little CaOcontent, suggesting that the ash sample is more ofrefractory and abrasive in nature. Such type of ashmay create erosion problem not only in the plantcomponents, but also during its transport throughpipe line.

From the building material point of view thesethree major constituents are represented as"modulus of silica" or SiOAAI203+Fe203) ratio.The value of this ratio should lie between 1.3 and2.5 (Ref. 7). For the three present samples thevalues are 1.60, 1.74 and 1.47 respectivelysuggesting that chemical composition-wise thesesamples·can be used for making building materials.An'other interesting aspect for the sample 3 is thatwhil.e i(s total alkali content is rather low, its MgO~ontent is higher. Possibly, excess Al and Mg

.-contents oocur in the form of spinel ferrite8 i.e.

. ;(Mg{Fe) (F@,AI)204'

. '>..-:, .~~~:~. "

As far as trace elements are concerned the threesamples have almost similar composition, excepthigh chromium content in sample 2. Some of thetrace metal contents are within the range of valuesgiven by Sahu9 for Talcher fly ash. However, Hgand Cd contents are low in all the samples.

Leaching Characteristics

Leaching of trace elements from fly ash inaqueous medium is an important aspect from thepoint of view of environmental contamination.Recently, this aspect of flyash disposal problemhas been excellently reviewed by Prasad et al.10.There are various methods available in literature todetermine the leachability of various trace elementsfrom fly ash and their comparative evaluation hasalso been reviewed 10. In the present work, thebatch leaching method has been used because ofthe simplicity of its operation and minimuminterference of extraneous factors.

Ashes may be alkalinell and acidic I I dependingupon its chemical constituents. As illustrated inFig. 6 the pH values of the aqueous suspensionafter different periods of equilibration exhibit widevariation for these three ash samples. Whereas forsample 1pH value increases gradually from 7.0 to7.5 in about 4 d~ys, for sample 2 pH valueincreases steadily from 5.5 to 6.5 in about 8-10days and then remain constant up to 30 days. Incontrast, the pH value of sample 3 increases veryslowly from 4.5 to 5.0 over the entire period of 30days. Thus, while sample 1 is distinctly an alkalineash, sample 3 is acidic and 2 is intermediatebetween the two.

Since the amorphous glassy phase in fly ash isthermodynamically metastable6, it tends to reactwith water releasing alkali and alkaline earth metal

MOHAPA TRA & KANUNGO: PHYSICO-CHEMICAL CHARACTERISTICS OF FLY ASH 277

Fig. 7-Leaching b.ehaviourof some selected metal ions fromsample I in aqueous suspension.

Ul

~co

.2~C••uco

0'9 0u

1'3

10

Sample 2~

60,," 0 Pb

• Cu

• Ni

501-. Hgli CrI

0.40<-co

g 30ecg 20ou

0'78 12 16 20 24 28 32

Period of Leaching, days

I.l

11.2 g'"

-!...

.• ~0.8

1.1 C7I

~g

10 ~

-,:) I -E

10.9 ~ou

07g ~ W ~ n n

lpac.hing P~riod,days,

Sompl. 1° Co

'O~ 0 Pb• Cu

SO~ • Ni•• HgA Cr

01 40'~co~ 3011:..g 201u

ions in solution with the resultant rise12inpH.

M20+H20~2 M++2OH­

MeO+H20~M+++2 OH-

where M=Na or K and Me=Ca or Mg.In the case of pond ash (sample 2) these metal

ions probably occur within the matrix as most ofthe surface concentration possibly had gotdepleted, thereby showing slower rate of releasecompared to that of sample 1. But other factorssuch as long-term hydrolysis of dissolved iron andaluminium may release H+ thereby restricting the.. HI 112rIse mp eve .

The acidic pH of sample 3, is, however, due tothe release of the oxides of sulphur (S02 or S03)adsorbed on highly reactive and extremely fine ashparticles, producng H2S03 or H2S04 in solutionl3.The low alkali and alkaline earth metal contents,

but high dissolved SO~- content provide adequateevidences for the acidic character of the sample.The released H+ slowly attacks iron oxides andeven some alumino-silicate minerals, resulting inthe slow increase in pH value was demonstrated inFig. 6.

On the basis of pH variation with time, theleaching chracteristics of trace metals may bediscussed. The leaching behaviour of six tracemetals from samples 1 and 2 (Figs 7 and 8) areessentially similar. The initial rates of leaching ofNi, Co_d Pb are more rapid than those of Hg andCr. While Ni and Co concentrations /attain theirconstant values after 7-8 days of leaching, theconcentrations of Hg and Pb continue to increaseslowly up to 30 days and possibly beyond this

Fig. 8-Leaching behaviour of some selected metal ions fromsample 2 in aqueous suspension.

period. From environmental point of view, thesteady release of these two toxic metal ions is amatter of concern, as this may lead to possiblecontamination of both ground and surface water.Figs 7 and 8 also indicate that leaching of Cu fromsamples 1 and 2 is very slow. It is interesting tonote that no Cd was found in measurable quantityin the leachates of these two samples.

The leaching characteristics of various tracemetals (except Ni and Cu) from sample 3 aredifferent from those of samples 1 and 2 asdemonstrated in Fig. 9. After an inital rapiddissolution up to 4 days, though the concentrationof Ni tends to attain an almost constant value, theconcentration of Co continues to increase steadilyup to 30 days. Another notable difference is thatCd begins to leach out from this samples right fromthe beginning. Acidic medium accelerates14,15 theleaching of Co and Cd. Fig. 9, however, shows thatthe rate of leaching of Pb, Hg and Cr isconsiderably slowed down in acidic medium.

Mechanism of Leaching-It is now well­established that most of the trace elements occur onthe surface of fine particles of fly ash, some ofwhich are in soluble forms.

The initial rapid dissolution of certain quantityof Ni and Co in the first few days followed by analmost plateau region up to 30 days suggest thatcertain fractions of these two elements occur insoluble form, possibly as sulphates. Indeed, manymetal oxides including nickel and cobalt oxides getsulphatized at temperature around 800°C by S02present in the flue gasl6. A small portion of the

278 INDIAN 1. ENG. MATER. SCI., DECEMBER 1997

Fig. 9-Leaching behaviour of some selected metal ions fromsample 3 in aqueous suspension.

so

-36 Sample 1 (10_3101 NaCl041o Sample 2 00 M NaCl041

• Sample 2 (10-1101 NaCI04.13 ... / "

o Sample 3 (1(j M NaCl041

o

-o·{,

-1'2

-0,8

C1>

~c.Q1?-C :t:

:';a.

rI" u :t:a.II:t:a.<1

1.0

.&i

L.-~0.728 32

/ 11.2<~~ ~-- .;~ __ 1.1

'--.J20 24

Period of Leaching, days

Sample 360~ 0 Co

a Pb• Cu

• Ni

• Hg

li Cr

~ Cd

c.2

•Fig, IO--Oetermination of the points of zero charges of thethree fly ash samples by the solid addition method.

different major constituents of fly ash (Fig. 10). Itcan be seen that there is no change in ~pH in thepH range 3-4. Therefore, in this pH range theinterfacial behaviour of the material is pre­dominantly controlled by silica or aluminosilicateminerals.

As the pH value is increased from 4.0 to 5.5 thefinal pH value of the suspension becomes higherthan that of the initial value as the slicate mineral

surface becomes more and more negativelycharged. In such a situation H+ will beconcentrated close to the surface forming thecounter-ion layer and OH- will tend to form diffusedouble layer resulting in the increase in pHf. Thesuspensions are expected to remain in dispersedstate at the minimum value of ~pH i.e., maximumvalue of final pH when the fly ash particles arelikely to experience mutual repulsion due to highnegative surface charge.

As pHi value is increased above 5.5, pHf valuedecreases gradually up to a pHi value of 6-7 when~pH becomes zero. Here the oxides of ironprobably play an important role in controlling theinterfacial behaviour of fly ash. Since the amountof iron oxide phase is much lower than that ofeither silicate or alumina, a mutual interaction of

the three major constituents lower the cross-overpoint in some cases below the usual pHpzc of ironoxide. At pHi> 7.0 the equilibrium or final pH

sulphates of Ni and Co might have escapeddecomposition at higher temperature when gottrapped within the glassy/siliceous materials. Onthe other hand, Hg, Pb etc. are probably leachedout from the surface of the glassy spheres, possiblyas hydroxy species below their solubilityprQducts17. The formation of the hydroxy speciesof these metals including Cu is favoured withincrease inpH. AtpH below 5.5 Cu does not occuras soluble hydroxy species, but above 6.0 not only

+ +Cu but also Pb and Hg occur as CuOH , PbOHand HgOH+ respectivelyJ7.18. However, above pH7.0, CuOH+ tend to get precipitated as CU(OH)2,while the other two positively charged hydroxyspecies are stable even up to pH 8.5. Therefore, therelease of Pb and Hg increases with the period ofleaching but CU tends to attain a plateau region.

Interfacial Properties in Aqueous SuspensionMany of the uses of fly ash involve processing

of the material in presence of water. In aqueoussuspension the major constituents such as silica,aluminosilicates, iron oxides etc. in fly ash undergosurface hydration, hydroxylation and finallyacquire surface charge by releasing either H+ orOH- ions. Silica and alumino silicate minerals have

points of zero charge (PZC) at pH 2.2-3.5 (ref. 19),whereas oxides of iron have PZC value at pHaround 7.0 (ref. 4, 19). The PZC value of AI20), onthe other hand, lies at pH 9.5-10.0 (ref. 19). Thechange in pH value (~H=pHi-pHf) observed bythe addition of fly ash into aqueous solution of anindifferent 1-1 electrolyte clearly demonstrates thechanges that take place on the surfaces of the three

-1"3D. l,·o SO 6·0 7-0 8=0 9,0 10·0

pH,

MOHAPA TRA & KANUNGO: PHYSICO-CHEMICAL CHARACTERISTICS OF FLY ASH 279

... (4)

... (3)

o

Curve 1 0 Sample 1

II lIt:. II 2" m. II 3

oI I ..•••••••....•••••••.'-

010 0 15 0 20 025 0 30 035

¢ (SCALE FOR CURVES J ,Ill

J I I I -L-------..J0·2 0,3 0·4 05 06 07

¢Sea Ie for Curves 1,11

!0'1

o OS

~~KI 2·8 12·1

11 3,0 23·811l 1·5 26·0

1o

0·01

oo

0-06

O-oS

0'02

a.III

~ 0'03-e-

Fig. 12-Plot of (AI" versus ,pc according to Eq. (5).

It has been observed by several investigators22-24

that both relative and specific viscosity ofsuspension of solid particles in water is related notonly to the true volume fractipn of solid but also tothe volume fraction of "free liquid" which isdefined as the difference between apparent(measured) and true volume fractions of solid «(A)

in a suspension. The true volume fraction of a solidis the volume fraction of solid in a closest packedsediment and is obtained from voidage (&) valuegiven in Table 1. The apparent volume fraction (¢).may be estimated experimentally from the

I· h' 25re atIons Ip

¢= Ps -PmPc -Pm

where Pc, Pm Ps are the densitIes of solid (true),medium and suspension respectively. Since theratio (S') between apparent and true volumefraction of soiid gives the relative. volume of solidin a suspension, the apparent volume fraction ofsolid is S¢ the volume fraction of free liquid is I­S',¢. The physical significance of the value of S· is,that it represents the degree of hydration or thevolume of bound water per unit volume of solid.

As stated -above, the specific viscosity (1]sp) isdirectly proportional to the volume fraction of solidand inversely to the volume fraction of 'freeliquid'. Combining these two variables Robinson24

obtained Eq.(4)

1 K¢1]r - = 1]sp = I-S'¢

A

0'65b

0:1o

oI I I

0'2 0,3 0:4 0'5

Volume Fraction of Solid

Fig. Il-Change in the relative viscosity with the volumefraction ofsolid in aqueous suspensions of the three samples.

(PHf) becomes lower than that of pHi as OIr forma counter ion layer for positively charged silicatesand iron oxides with consequent release of H+ ionin the bulk solution. The difference i.e., ~Hbecomes maximum at the initial pH of 8.5 to 9.0where also the suspension is likely to be in highlydispersed state due to mutual repulsion at highlypositive surface charge. At pHi=9.5 all the three

samples exhibit ~H=O which is the usual pHpzc ofAh03'

20

Viscosity of SuspensionsViscosity or rheology of fly ash suspension is an

important parameter from the point of view of itshydraulic transportation through pipe line. Theengineering design of transport through pipe linerequires precise determination of rheologicalproperties which, in turn, are governed by thewater bound to fly ash particles in aqueous slurry20.While some interparticle bound water is necessaryto impart requisite flow at high concentration(>65% solid) of slurry. Large amount of such waterbrings about considerable resistance to flow. In thepresent work, an attempt has been made to estimatethe bound water from viscosity measurement.

The interfacial properties of a suspension exertconsiderable influence on its rheological or flowbehaviour21. Viscosity measurement at pH close topzc will tend to show a value higher than thatmeasured away from it, i.e. closer to pH value ofhighest dispersion.

120

0Sample I

V""1001- b

))2

0)13

I >-

80

- 'iii0u 60III:;:

QI

.~ 40"0 -.;a:

280 INDIAN J. ENG. MATER. SCI., DECEMBER 1997

... (5)

2"I 0Sample 1

'1

t>»2

•")

1·6 ~

•l7'l

0-'

bOL- _ . ~_

o 0·1 0·2 0,3 0,' 0·5 0,6 0'7 0·9

-Log (1- (/J/¢C)

Fig. 13-Log-Iog plot of TJr versus (J-+'+c) for three fly ashsamples according to Eq. (6).

where, K is the overall constantRearranging,

L=l..._ S17sp K K

Some authors22,23 have, however, preferred touse logarithmic plot of relative viscosity versusvolume fraction of solid or free liquid to explainthe exponential rise in viscosity with concentrationof solid. A typical form of equation is given23•

... (6)

when m is related to the shape factor. For sphericalparticles its value is 2.5

The results of viscosity measurement of theaqueous suspension of the three fly ash samples'may be explained on the basis of abovecorrelations.

Fig. 11 depicts the plot of relative· viscosityversus volume fraction of fly ash slurry at30±0.5°C. It can be seen that Newtonian behaviour

follows up to 40% slurry, above which viscosityincreases almost exponentially for all the three

samples. Fig. 12 shows the plot of ¢/17sp versus t/J

for aqueous suspension of the samples up to 65%solid (w/v). It can be seen that the linear behaviourfollows better at high solid concentration. Thevalues of S' and K were determined from the slopeand intercept values respectively. It can be seenfrom the figure that while the bound water for

samples 1 and 2 is about 3 times the true volume ofsolid, for sample 3 the value is 1.5 suggesting thatthe suspension of sample 3 is less hydrated,possibly due to low alkali or alkaline earth metalcontents. The physical significance of K isgenerally attributed to the frictional coefficientwhich is influenced by factors such as particleroughness and the adsorbed layer of water aroundh . 1 24t e partIc es .

Fig. 13 shows the plot of log 17r against log (1­cj>/cI>c)' Unlike Fig. 12, a common straight line with adistinct intercept value, can describe the Eq. (6) forall the three fly ash samples. The intercept value isrelated to yield stress of the suspensions. The slopevalue of -2.6 is closely similar to -2.5 as suggestedby Roscoe23 for suspensions of spherical particles.

AcknowledgementsThe authors are thankful to Dr C R Panda, Dr B

D Nayak, for providing us with fly ash samplescollected by them from different power plants. Theauthors are also thankful to their colleagues forgenerous help in providing XRD, OES, SEM, AASand ICP facilities. Thanks are also due to the

Director, RRL, Bhubaneswar for his kindpermission to publish the work.

References1 Palit A, Gopal R, Dubey S K & Mandai P K, Proc of the

International Conference on Environmental Impact ofCoal Utilization (liT, Bombay), 1991.

2 Sengupta J, Proc of the International Conference onEnvironmental Impact of Coal Utilization (lIT, Bombay),1991.

3 Indian Standard Institution, Method of Testing forPozzolanic Materials, ISI 1727-1976, New Delhi, 1968.

4 Balitrieri, L S & Murray J W, Am J&i, 281 (1981) 788;Kanungo S B, J Colloid Interface Sci, 162 (1994) 86.

5 Thomas J & Thomas W, Introduction to the Principles ofHeterogeneous Catalysis (Academic Press, London),1962.

6 Xu A & Sarkar S L, in Mineral admixtures in cement andconcrete, Vol. 4, edited by Sarkar S L & Ghosh S N (ABIBooks Pvt Ltd, New Delhi), 1994, 174.

7 Jensen S A, in Fly ash and coal conversion by-products:characterization, utilization and disposal II, edited byMcCarthy C J, Glasser F P & Roy D M (MaterialsResearch Society, Pittsburgh, USA), 1986,27.

8 Stevenson R J & McCarthy C J, in Fly ash and coalconversion by-products : characterization, utilization anddisposal II. edited by McCarthy C J, Glasser F P & Roy DM (Materials Research Society, Pittsburgh, USA), 1986,77.

9 Sahu K C, in Proc International Conferences on theEnvironmental Impact of Coal Utilization (lIT, Bombay),1991,12.

MOHAPATRA & KANUNGO: PHYSICO-CHEMICAL CHARACTERISTICS OF FLY ASH 281

10 Prasad B, Banerjee N N & Dhar B B, J Sci [nd Res ([ndia),55 (I 996) 772.

II Roy W R & Griffin R A,J Environ Qual, II (1983) 563.12 Roy W R & Griffin R A, Environ Sci Technol, 18 (1984)

739.

13 Heming R T, BerryE E,· Cornelius. B J & Sheetz B E,Mater Res Sac Symp Proc, 86 (1987) 81.

14 Eisenberg S H, Tittlebaum M E, Eaton H C & Soroczak MM,J Environ Sci Health A. 21 (1986) 383.

15 Aushin D E & Newland L W, Chemosphere,14 (1985) 41.16 Ingraham T R & Kerby R. Can Metall Q, 6 (1967) 89.

17 Turner D R, Whitefield M & Dickson A G, GeochimCosmochim Acta, 45 (1981) 855.

18 Ruppert H, Chem Erde, 39 (1980) 97.19 Davis J A, James R 0 & Leckie J 0, J Colloid Interface

Sci,63 (1978) 480.20 Richter B & Falbe J, J Chem Ing T~h, 57 (1985) 478.21 Napper D H, [nd Eng Chem Prod Res Dev, 8 (1970) 467.22 Mooney M, J Colloid Sci, 6 (I 951) 162.23 RoScocR. JAppl Phys, 3 (1952) 267.24 Robinson J V, J Phys Chem, 53 (1949) 1042.25 Kanungo S B & De P K, Indian J Technol, 8 (1970) 180.