Bull. Mater. Sci., Vol. 28, No. 2, April 2005, pp. 131–136. © Indian Academy of Sciences.

Some studies on the reaction between fly ash and lime

A BASUMAJUMDAR†, A K DAS†, N BANDYOPADHYAY and S MAITRA*
College of Ceramic Technology, WB University of Technology, Kolkata 700 010, India
†
Department of Chemical Technology, University of Calcutta, Kolkata 700 009, India

MS received 19 April 2004; revised 30 August 2004

Abstract. The reaction between fly ash (FA) and lime is extensively exploited for the manufacture of build-
ing bricks, blocks and aggregates. To get a better idea of this reaction, FA from different sources were mixed
in different ratios with lime and compacted. The compacts were treated both by ordinary water and hydro-
thermal curing to promote lime bearing hydrate bond formation e.g. CaO–SiO2–H2O (C–S–H), CaO–Al2O3–
H2O (C–A–H) etc. The decrease in free lime content in these compacts was measured as a function of curing
time and curing process. This drop in this content was correlated to the chemical composition of the fly ashes.
The mathematical relationships between free lime remaining in the compacts after its maximum decrease in
concentration and lime binding modulus (a ratio between the amount of added lime and the total amount of
lime binding constituents present in FA) for both types of curing were developed. Further, the rate of decrease
in free CaO content under both types of curing conditions was compared from kinetic study. From this study
the orders of the reactions and rate constants were found out.

Keywords. Fly ash; lime; compacts; curing; kinetics.

1. Introduction In the present investigation the reaction between fly

ash and lime in compacted form was studied by measur-
In the presence of moisture, fly ash reacts with lime at ing the free CaO remaining in the mix after different pe-
ordinary temperature and forms a compound possessing riods of curing. The mix compositions were varied and
cementitious properties. The reaction between lime and fly different curing conditions were employed. The relation-
ash produces calcium silicate hydrates, which are respon- ships between free CaO remained in the mixes after opti-
sible for development of strength in fly ash–lime in the mum curing periods and chemical composition of fly
form of bricks and blocks. These bricks/blocks are suit- ashes were developed and the kinetic parameters of these
able for use in masonry just like common burnt clay bricks reactions were also studied.
at lower cost with added advantages.
Barbier (1986) studied on the possible uses of coal fly
2. Experimental
ash in the brick industry with respect to the availability
and characteristics of the fly ash. Song et al (1996) studied
In the present investigation twelve fly ash samples from
the manufacture and properties of coal fly ash–clay bodies.
different power plants of West Bengal (India) were collec-
Tsunematsu et al (1987) studied on the hydrothermal
ted. The ash samples were collected from the same num-
reactivity of fly ash with lime and gypsum with respect to
ber of ash collection hopper for all these units following
the mineral composition. Kumar (2002) made a perspec-
procedures as laid in IS: 1528-1974, Part-VII. Particle
tive study on fly ash–lime–gypsum bricks and hollow
size distributions of the ash samples were measured by
blocks for low cost housing development. Ma and Brown
sieve analyses following specifications as laid in IS:
(1997) also studied the hydrothermal reaction of fly ash
1528-1974, Part-XIV. For the measurement of bulk den-
with Ca(OH)2 and CaSO4⋅2H2O. Muntcan et al (1987)
sity of the ash samples a fixed volume of the sample was
studied on the autoclaved limestone materials with addi-
taken in a measuring cylinder, which was tapped for suf-
tion of fly ash. The physico-mechanical properties of the
ficient number of times for thorough packing and its
resulting siliceous limestone with 10–30% fly ash were
weight was measured. Blain’s air permeability apparatus
found to be superior compared to the limestone obtained
was used for measuring the surface area of the ash sam-
from lime and sand only. Wang et al (1996) studied on
ple. Specific gravity of the ash samples was measured by
the reaction mechanism of fly ash–lime–water system. A
pycnometer following specifications as laid in IS: 1528-
model of the reaction was established.
1974, Part-IX. Chemical analyses of the ash samples
were carried out following conventional technique of
sodium carbonate fusion. Pulverized lime with high per-
*Author for correspondence (maitra_s@vsnl.net) centage of CaO (> 80 wt%) and considerable surface area
131
132 A Basumajumdar et al

(10,800 cm2/g) was used in the present investigation. The From the chemical analysis of FA samples it was ob-
test compacts were prepared by mixing fly ash and lime served that the total amount of silica, alumina and iron
in different requisite proportions in a laboratory mixer. oxides varied from 70–90% in the compositions (table 1).
The mixing time was 6 h for all the cases as it was found Glassy phase content of the samples also varied between
to be optimum for homogeneous mixing. The optimality 75 and 85%. Therefore, four different FA samples were
was tested analysing different random proportion of the selected from the lot for carrying out experiment with
mix for SiO2 and lime content. The homogeneous mix total pozzolanic oxide contents of 70%, 80%, 85% and
thus obtained was compacted with 5% moisture using a 90%, respectively (table 2). Particle size distribution of
laboratory hydraulic press at different levels of compac- these four samples is shown in figure 1. These were likely
tion pressure (from 200 kg/cm2 to 325 kg/cm2). The test to represent the trend in compositional variation that
compacts were cured under two different conditions. In would affect the degree of interaction with lime. These
water curing the test specimens were immersed in water different fly ash samples were mixed with lime in different
at ambient temperature. For steam curing the samples were ratios: 90 : 10, 80 : 20, 70 : 30 and 60 : 40, respectively.
kept in a low-pressure autoclave at an average steam
pressure and temperature of 2⋅5 kg/cm2 and 120°C, res-
pectively. The free lime content in the test specimens was 3.1 Effect of compaction pressure
measured chemically using the procedure described by
Hanna et al (1938, 1939). Free lime was determined utiliz- Compaction pressure has a distinct effect on the reaction
ing the solubility of lime in a solvent containing 1 part by between lime and fly ash. To study this effect fly ash and
volume of pure glycerol to 5 volumes of absolute alco- lime mixes were compacted at different pressures and the
hol. The glycerol used was anhydrous containing 99⋅5% compacts were subjected to hydrothermal curing for a
glycerol with a specific gravity of 1⋅251 at 25°C. The period of 10 h. It was observed that a sharp reduction in
alcohol used was 99% ethyl alcohol. The indicator used free CaO content took place at a compaction pressure of
was 0⋅18 g phenolphthalein in 2160 ml of the mixed sol- 200 kg/cm2 and after 300 kg/cm2 of compaction pressure,
vent. The solution used for titrating the lime was pre- no significant change in free lime concentration was ob-
pared by dissolving 16 g dry crystalline ammonium ace- served. Compaction pressure enhanced the inter-granular
tate in 1 litre of alcohol. It was standardized as follows: contacts, thus facilitating the progress of reaction by in-
About 0⋅1 g of freshly ignited CaO was placed in a 200 ml creasing the formation of more interfacial contact areas
Erlenmeyer flask and 60 ml of glycerol–ethyl alcohol between the reactants. But a critical inter-particle gap or
solvent was added to it and the CaO was dispersed by space is required for the formation of different interfaces
shaking. A reflux condenser was fitted and the mixture and migration of the lime bearing hydrated phases for-
was boiled for 5–20 min and the mix was titrated again. med from its origin to other places. Perhaps at higher
This process was continued until the free CaO content pressures this gap is reduced which decreases the rate of
did not increase by more than 0⋅05% after 2 h of boiling. reaction. Therefore, in the present investigation all the
The free lime estimation was carried out by weighing compacts were prepared with a compaction pressure of
1 g of finely ground sample into a flask, adding 60 ml of 200 kg/cm2.
the solvent and proceeding as in the standardization of The depletion in free lime concentrations in the com-
the ammonium acetate solution. The free lime content of pacts was observed to occur at comparatively rapid rate
cement was calculated as follows initially after the onset of curing process. But after a defi-
100 EV nite curing period, lime reaction rate slowed down and no
% CaO = , significant change in free lime concentration took place.
W
where, E is the equivalent of standard ammonium acetate
solution in g CaO per ml, V the ml of ammonium acetate 3.2 Curing conditions and lime binding modulus of fly ashes
solution required and W the weight of the sample taken.
Curing under ambient condition (figure 2) revealed that
3. Results the maximum reduction in the concentration of free CaO
takes place between 50 and 55 days and beyond this pe-
Lime reacts with oxide components like silica, alumina riod the drop in free CaO content in the samples was not
and iron oxide of FA to develop different types of lime significant. In case of hydrothermal curing condition this
bearing phases like calcium silicate, calcium aluminate, period was observed to be 10 h (figure 3). It was also
calcium aluminosilicate etc and these phases are subse- observed that within this period, the extent of reduction
quently hydrated in the presence of water to form different in free CaO content varied from 87–96% for different fly
hydrates. These hydrates are responsible for the deve- ash–lime compacts. In samples with higher proportion of
lopment of strength in the fly ash–lime compacts (Watt fly ash, the extent of reduction in free CaO content was
and Throne 1996). observed to be more. It indicates that the rate of pozzo-
 Some studies on the reaction between fly ash and lime 133

Table 1. Physico-chemical properties of the ash samples.

A. Chemical composition of the ash samples (A–H)

A B C D E F G H
Wt% Wt% Wt% Wt% Wt% Wt% Wt% Wt%

SiO2 53⋅20 62⋅20 54⋅30 61⋅11 57⋅48 60⋅11 60⋅20 64⋅31

Al2O3 24⋅50 22⋅50 23⋅30 23⋅80 32⋅60 26⋅55 23⋅90 25⋅60
Fe2O3 6⋅20 7⋅60 8⋅30 7⋅20 3⋅48 5⋅84 7⋅80 4⋅45
CaO 1⋅70 1⋅56 1⋅42 1⋅35 1⋅98 4⋅83 2⋅06 1⋅03
MgO 1⋅30 1⋅01 0⋅68 1⋅35 1⋅06 1⋅25 0⋅70 0⋅97
LOI 8⋅40 2⋅40 5⋅80 1⋅14 1⋅22 0⋅63 0⋅35 0⋅58
Na2O 0⋅30 0⋅21 1⋅43 0⋅30 0⋅28 0⋅22 0⋅51 0⋅60
K2O 2⋅15 0⋅92 1⋅38 1⋅10 0⋅46 0⋅75 0⋅73 1⋅02
SO3 0⋅20 0⋅11 0⋅23 0⋅25 0⋅19 0⋅35 1⋅97 0⋅16
TiO2 1⋅03 1⋅40 1⋅45 2⋅10 0⋅94 0⋅78 0⋅42 –
Cl 0⋅01 – 0⋅02 – 0⋅01 – – –
P 2O5 0⋅86 – – – 0⋅24 – – –

B. Physical properties
Bulk density (g/cm3) 0⋅97 0⋅70 0⋅91 0⋅79 0⋅86 0⋅90 0⋅82 0⋅72
Specific gravity 2⋅50 1⋅98 2⋅21 2⋅11 2⋅14 2⋅43 2⋅12 1⋅98
Surface area (Blain’s, cm2/g) 5515 3960 3775 3358 6280 3025 4120 5210

A. Chemical composition of the ash samples (I–L)

I J K L
Wt% Wt% Wt% Wt%

SiO2 62⋅72 61⋅57 52⋅00 56⋅37

Al2O3 29⋅93 26⋅33 23⋅00 26⋅80
Fe2O3 2⋅13 6⋅35 2⋅30 6⋅10
CaO 2⋅33 0⋅94 10⋅50 2⋅65
MgO 0⋅72 0⋅84 3⋅25 1⋅15
LOI 0⋅52 1⋅40 2⋅50 2⋅14
Na2O 0⋅19 0⋅15 1⋅33 0⋅36
K2O 0⋅25 0⋅70 1⋅88 1⋅17
SO3 0⋅19 0⋅30 0⋅63 0⋅35
TiO2 0⋅75 2⋅09 1⋅95 1⋅10
Cl – – 0⋅02 –
P 2O5 – 0⋅13 – 0⋅31
B. Physical properties
Bulk density (g/cm3) 0⋅77 0⋅82 0⋅84 0⋅91
Specific gravity 2⋅10 2⋅12 2⋅12 2⋅41
Surface area (Blain’s, cm2/g) 4510 2990 5172 3635

lanic reaction between fly ash and CaO is the maximum Table 2. Pozzolanicity of the fly ash samples.
up to a critical curing period. The anomalies observed in Pozzolanicity [total per cent of
the intermediate regions might be due to the difference in Fly ash (SiO2 + Al2O3 + Fe2O3) constituents]
reactivity of the FA samples, which depends on the parti- A 83⋅90
cle size, chemical constituents, phase compositions, and B 92⋅30
the nature of the glassy phases present. Therefore, it may C 85⋅90 (FA2)
be inferred that the rate of formation of lime bearing hy- D 92⋅11
drated phases such as C–S–H, C–A–H was maximum up E 93⋅56
F 92⋅50
to this period. G 91⋅90 (FA3)
The reaction between fly ash and lime can be consi- H 94⋅36
dered mostly as interface controlled. In presence of water I 94⋅78 (FA4)
vapour the lime is converted to calcium hydroxide and J 94⋅25
the formed calcium hydroxide reacts with components K 77⋅30 (FA1)
L 89⋅27
like silica, alumina, iron oxide and titania constituents of
134 A Basumajumdar et al

FA to form calcium silicate, calcium aluminate, calcium tionship between lime binding modulus and free lime
ferrate and calcium titanate. Formation of complex phases remaining at these mentioned periods was attempted to
like calcium aluminosilicate, calcium iron aluminium develop. Such relationship may be useful in formulating
silicate cannot be ruled out. These lime-bearing phases lime–fly ash mix for optimum performances. The rela-
are ultimately converted into their hydrates in presence of tionship between lime binding modulus and free lime
water vapour. content for hydrothermal curing follows an exponential
As lime only reacts with the neutral and acidic com- fit and the equation takes the following formula,
ponents of FA for the development of lime bearing hy-
y = ae(bx), (1)
drated phases on curing, major constituents like silica,
alumina and iron oxide may be considered responsible where, y is the free CaO remaining in the compacts after
for binding of lime. The ratio of concentration of initial 10 h and x the lime binding modulus of fly ash. The values
added lime and the sum total of silica, alumina and iron of the coefficient are a = 0⋅19 and b = 7⋅36, respectively.
oxide, i.e. CaO/(SiO2 + Al2O3 + Fe2O3), was designated Standard error is 0⋅11 and correlation coefficient is 0⋅9937.
as the lime binding modulus of fly ash. The relationship between lime binding modulus and free
The concentration of free CaO remaining after ~ 55 lime content for water curing follows 4th degree polyno-
days for ordinary curing and after ~ 10 h for hydrother- mial fit
mal curing may be taken as more or less constant as the y = a + bx + cx2, (2)
rate of decrease in free CaO beyond these periods was
found to be practically insignificant. Therefore, a rela- y = free CaO remaining in the compacts after 55 days and
x the lime binding modulus of fly ash. The values of the
coefficient are a = 1⋅28, b = – 13⋅90, c = 54⋅29, standard
error is 0⋅24 and correlation coefficient is 0⋅9848, respec-
tively.
The equations thus developed were tested on different
fly ash–lime compacts with different lime binding modu-
lus (appendix 1). The variations in all the cases were ob-
served within ± 5–6%. This might be related to the
difference in mineralogical compositions, particularly the
glassy phases in the fly ash samples.
From the relationship it is evident that the drop in free
lime concentration under steam curing takes place at a
faster rate compared to the natural curing process. This
might be ascribed to the relatively more activation of the
fly ash framework at elevated temperature and steam
pressure.
To understand the nature of reaction of fly ash with
Figure 1. Particle size distribution of different fly ash samples. lime, the order of the reaction was found out from the

Figure 2. Variation in free CaO of fly ash–lime compacts Figure 3. Variation in free CaO content of fly ash–lime com-
(80 : 20) with water curing. pacts (80 : 20) with steam curing.
 Some studies on the reaction between fly ash and lime 135

previous experiment. For the sake of simplicity in the 4. Conclusions

experimental procedures here only the fly ash with the
maximum lime binding modulus was taken with four From the present investigation, it appeared that
different compositions of fly ash and lime, ca., 90 : 10, (I) The optimum compaction pressure for making class F
80 : 20, 70 : 30 and 60 : 40. fly ash–lime compacts by water/steam curing process is
From the standard error and correlation coefficient of 200–250 kg/cm2.
linearity of all the plots for different orders it might be (II) For water curing process the maximum reduction in
concluded that for the water curing process the rate of free CaO content takes place after 50–55 days, whereas,
reaction between fly ash and lime is most likely to follow for steam curing process the same takes place after 10 h.
first order kinetics with respect to the lime concentration (III) The relationship between lime binding modulus of
while in steam curing the reaction between lime and fly fly ash (CaO/SiO2 + Al2O3 + Fe2O3) and free lime re-
ash in the compacted form is most likely to follow sec- maining after 10 h of steam curing is y = a⋅ebx, where y is
ond order reaction kinetics with respect to the lime con- the free CaO content and x the lime binding modulus of
centration. Probably steam pressure aids in the formation fly ash. The values of the coefficients are a = 0⋅19 and
of more interfaces between fly ash and lime and the b = 7⋅36, respectively. Standard error is 0⋅11 and correla-
thermal energy associated with the steam helps in over- tion coefficient is 0⋅9937.
coming the energy barrier of the reaction. It was also (IV) The relationship between lime binding modulus of
observed that with the increase in lime concentration the fly ash and free lime remaining after 55 days of water
values of the rate constants increased. curing is y = a + bx + cx2, where y = free CaO remaining
The rate of the reaction between fly ash and lime pro- in the compacts after 55 days and x the lime binding
gressively decreased with curing time. This might be re- modulus of fly ash. The values of the coefficient are a =
lated to the formation of different lime bearing hydrated 1⋅28, b = – 13⋅90, c = 54⋅29, standard error is 0⋅24 and
phases, such as calcium silicate hydrate, calcium alumi- correlation coefficient is 0⋅9848, respectively.
nate hydrate etc on the surface of the ash particles. These (V) The decrease in lime concentration with fly ash with
hydrated phases cover the surface of the ash particles and curing time follows mostly first order kinetics and in case
reduce the contact of lime and particles further, thereby of steam curing it follows mostly second order kinetics
reducing the rate of reaction. with respect to the lime concentration in the mix.

Appendix:
Table A. Theoretical and experimental free lime content in samples after 10
hours of steam curing.

Predicted free Determined free

Lime binding lime after 10 h lime after 10 h Variation
modulus of steam curing of steam curing (%)

0⋅217 0⋅938 0⋅987 5⋅3 (+)

0⋅32 2⋅00 1⋅932 3⋅4 (–)
0⋅224 0⋅987 1⋅05 6⋅4 (+)
0⋅357 2⋅64 2⋅79 5⋅9 (+)
0⋅212 0⋅903 0⋅927 2⋅7 (+)
0⋅424 4⋅3 4⋅08 4⋅9 (+)
0⋅112 0⋅43 0⋅447 4⋅1 (+)

Table B. Theoretical and experimental free CaO content in samples after 55

days of water curing.

Predicted free Determined free

Lime binding lime after 55 days of lime after 55 days of Variation
modulus water curing water curing (%)

0⋅326 2⋅51 2⋅66 6⋅1 (+)

0⋅448 5⋅95 6⋅27 5⋅5 (+)
0⋅214 0⋅79 0⋅83 4⋅7 (+)
0⋅238 1⋅05 1⋅02 2⋅8 (–)
0⋅119 0⋅39 0⋅38 1⋅9 (–)
0⋅163 0⋅45 0⋅479 6⋅5 (+)
0⋅265 1⋅416 1⋅474 4⋅1 (+)
136 A Basumajumdar et al

Table C. Standard error (S), correlation coefficient values (R) of linearity for plots of free lime concentration
against curing period of different fly ash and lime ratios for different assumed orders and reaction rate constants (k)
from the order.

A: Water curing
Fly ash: lime Order S R k

9:1 1 0⋅2530 0⋅9607 3⋅22 × 10 day–1

–2

2 1⋅6352 0⋅9059
4:1 1 0⋅2530 0⋅9611 2⋅6 × 10–2 day–1
2 0⋅5332 0⋅8903
7:3 1 0⋅1957 0⋅9833 2⋅19 × 10–2 day–1
2 0⋅1225 0⋅9616
3:2 1 0⋅2148 0⋅9778 2⋅5 × 10–2 day–1
2 0⋅0774 0⋅9638

B: Steam curing
Fly ash: lime Order S R k

9:1 1 0⋅6279 0⋅9511

2 1⋅9093 0⋅9700 1⋅0076 per cent–1 hour–1
3 0⋅9093 0⋅9134
4:1 1 0⋅9093 0⋅9134
2 0⋅3006 0⋅9783 20⋅29 × 10–2 per cent–1 hour–1
3 0⋅7031 0⋅9418
7:3 1 0⋅5029 0⋅9427
2 0⋅1619 0⋅9726 9⋅68 × 10–2 per cent–1 hour-1
3 0⋅3183 0⋅9613
3:2 1 0⋅4802 0⋅9471
2 0⋅1153 0⋅9768 7⋅52 × 10–2 per cent–1 hour–1
3 0⋅1991 0⋅9586

References Ma W and Brown P W 1997 Cem. Concr. Res. 27 1237

Muntcan M, Paul F and Haret B 1987 Mater. Constr. 17 226
Barbier U 1986 Ind. Ital. Laterizi 40 53 Song J T, Yun S D, Ryan D W and Han K S 1996 J. Korean
Hanna W C, Hicks T A and Saeger G A 1938 Bull. Am. Ceram. Ceram. Soc. 33 154
Soc. Test. Mater. 94 47 Tsunematsu S, Inow N and Hara N 1987 Gypsum and Lime 211
Hanna W C, Hicks T A and Saeger G A 1939 Proc. Am. Soc. 362
Test. Mater. 39 314 Wang X, Yang N and Zhang B 1996 J. Chin. Ceram. Soc. 24 137
Kumar S 2002 Construction and Building Materials 16 519 Watt J D and Thorne D J 1996 J. Appl. Chem. 15 585