Indian Journal of Engineering & Materials Sciences Vol. 17, August 2010, pp. 282-288
Effect of basalt on the burnability of raw meal of Portland cement clinker H El-Didamonya, A Abdel Rahmanb*, F Nassarb & M Sarayab a
Faculty of science, Zagazig University, Zagazig, Egypt Faculty of Science, Al-Azhar University, Cairo, 11884, Egypt
b
Received 17 February 2009; accepted 24 May 2010 This study presents the effect of basalt as a flux and mineralizer on the formation of Portland cement clinker fired at different temperatures. Five mixtures are prepared by addition of different amounts of basalt such as 0.0, 2.5, 5.0, 7.5 and 10.0 mass % to the kiln feed prepared for the ordinary Portland cement clinker manufacture. Each mixture from the above mixes is molded in one inch cylindrical bar under 300 kg/cm2 and fired in muffle furnace at different firing temperatures such as 1200ºC, 1300ºC and 1350ºC for 2 h as soaking time. The fired clinker is quenched in the air to prevent the dissociation of the formed clinker. The degree of clinkerization is studied by the determination of insoluble residue, free lime contents and the identification of mineralogical phases such as C3S, β-C2S, C3A and C4AF by XRD analysis. The potential phase composition of clinker is also calculated by the modified Bogue formulae. The hydration characteristics are investigated by the determination of free lime, combined water contents, total porosity, bulk density and compressive strength for all clinkers fired at different temperatures. In addition, IR spectra of hydrated cement pastes of some selected samples are studied. Keywords: Basalt, Burnability, Portland cement clinker
The addition of small amounts of some substances that are not normal clinker raw meals may distinct alter the progress of the clinkerization processes. These substances that have the capacity to intensify the rate of clinker formation are called fluxes or mineralizers. The use of fluxes and mineralizers to facilitate of clinkerization was studied by adding some compounds in the ordinary cement raw meal. Among these compounds, fluorides and silicofluorides have been studied more, although they have rarely commercially survived, owing to the rapid deterioration of refractory kiln lining and the effects on setting time and strength1-3. A fluxing agent or fluxes are substances that accelerate the formation of clinker by decreasing the temperature at which a melt starts to be formed in the system and by an increase of the amount of the liquid phase at any temperature above. Under these conditions, tricalcium silicate (C3S) starts to form at a lower temperature; the reaction progress faster and the clinkerization process may be completed at a temperature much lower than that need in the absence of the fluxing agent. Unlike fluxing agents, mineralizers are substances that accelerate the rate of solid-state reaction or reactions that take place within the liquid phase or at liquid-solid
interface without significantly altering the temperature of melt formation and the amount of melt present. In the presence of mineralizers/fluxes, the viscosity and surface tension of the formed melt may be also altered. Generally, the rate of clinkerization accelerates as viscosity declines and surface tension increases4,5. There are many studies to use basalt as an aggregate for concrete. The physico-mechanical properties of basalt can be summarized as6: compressive strength (MPa) 158, tensile strength (MPa) 12.2, specific gravity 2.8 and absorption (%) 1.4. Basalt is basic rock, poor in silica and alkalis and rich in magnesium, iron and calcium7,8. The results show that basalt is by far the most reactive of aggregate studies. Ions were both released and absorbed by this aggregate; possibly indicating that hydration of the surface layer had taken place9.
_________________ *Corresponding author (E-mail:
[email protected])
Kiln feed Basalt
Experimental Procedure The materials used in this study were raw meal (kiln feed), and basalt which dry ground in an automatic agate mortar. Table 1 shows the grain size distribution of these materials. Table 1 Sieve analysis of starting materials (wt% w) Sieve > 150 µm 150 µm- 106 µm- 90 µm- 75 µm- 90 µm 18.50 %
Mix 2
Mix 3
Mix 4
Mix 5
97.50 95.00 92.50 90.00 2.50 5.00 7.50 10.00 0.88 0.85 0.78 0.75 14.51 14.90 15.71 15.96 4.04 4.29 4.48 4.66 2.39 2.62 2.87 3.02 41.91 40.01 39.62 38.91 1.52 1.59 1.61 1.64 0.36 0.35 0.35 0.34 0.41 0.45 0.49 0.52 0.30 0.31 0.31 0.32 0.23 0.23 0.23 0.23 34.99 34.35 34.33 34.32 18.00 % 13.50 % 13.00 % 13.00 %
Table 2 Chemical analysis of starting materials (wt%)
Kiln feed Basalt
SiO2
Al2O3
Fe2O3
CaO
MgO
SO3
Na2O
K2O
Cl-
I.L950°C
Total
13.58 50.35
3.77 14.43
2.21 12.75
41.88 10.18
1.35 6.05
0.38 0.11
0.36 1.88
0.29 1.04
0.24 0.15
35.94 1.90
100 99.71
284
INDIAN J. ENG. MATER. SCI., AUGUST 2010
chemical composition of raw mixes changes due to the increase in the A2O3, Fe2O3, MgO, alkalis and traces of transition elements content, therefore the values of LSF decrease to 0.88 and 0.85 respectively. SM and AM lead to improve the burnability of raw mixes, reduce the melt point of liquid phase and maximum amount of liquid phase at lower temperature. These additions reduce the free lime content in the produced clinker and increase the clinker phases. Addition of 7.5% and 10% basalt on the kiln feed to prepare raw mixes (M4 and M5) tends to decrease the value of LSF to 0.78 and 0.75 respectively, as well as SM and AM. However these high amounts of basalt reduce the lime content and increase the Al2O3 and Fe2O3 contents in the raw mixes, which lead to the consumption of high amount of lime to produce C3A and C4AF. Therefore, these raw mixes become not suitable to produce clinker phases, especially alite C3S (OPC) or to produce belite cement. The mineralizing effect of basalt on raw mixes M1M5 which fired at different temperatures such as 1200°C, 1300°C and 1350°C for 2 h soaking time to show the rate of clinker formation. This was investigated by determination of free lime and insoluble residue as well as the formed cement phases using XRD technique. Free lime contents and insoluble residue contents
The free lime contents of the fired raw mixes (M1-M2) containing different amounts of basalt such as 0.0, 2.5, 5.0 , 7.5 and 10.0% w/w and fired at different temperatures, are graphically plotted as a function of firing temperature and mix composition as shown in Fig. 2.
Fig. 2 Free lime contents of raw mixes fired at different temperatures
Figure 2 shows that the free lime contents in the samples fired at different temperatures, which are an indication of the progress of clinkerization, decrease with firing temperature. Also, the free lime contents decrease with the amount of basalt in the raw mixes. When the raw mixes fired at 1200ºC, the free lime contents are still high. This possibly means that the firing temperature is insufficient to complete the clinkerization process. But, at the same temperature when the amount of basalt increases in the raw mixes the amount of free lime contents decreases. This result shows the effect of the basalt as mineralizing materials as well as decreases the LSF value. Therefore, the free lime content decreases. Also, at firing temperature 1300ºC, the free lime contents are also high especially in raw mix M1 (0% basalt), but lower than at 1200ºC. This indicates the progress of clinkerization process with the firing temperature as well as the increase of the amount of basalt decreases the LSF value. When the firing temperature is 1350ºC, the free lime contents in the fired raw mixes decrease to lower limits, especially in the raw mixes containing high percentages of basalt. But, the free lime is still high in the raw mix M1 (0 % basalt). This means that the raw mix M1 needs high temperature, to produce PC clinker with suitable free lime content, than the raw mixes with different percentages of basalt. The free lime decreases with the content in the raw mix due to the mineralizing effect of basalt, which increases the amount of liquid phase and improves the clinker phase formation. The insoluble residue decreases with the firing temperature as shown in Fig. 3. Also, the insoluble
Fig. 3 Insoluble residue of cement raw mixes fired at different temperatures
EL-DIDAMONY et al.: PORTLAND CEMENT CLINKER
residue increases with the amount of basalt in the raw mixes, especially M4 and M5. It is clear that the insoluble residue of the fired clinkers decreases with basalt content up to 5.0% and then increases up to 10% at all firing temperatures. Also the rate of the decrease of insoluble residue is the same at 0.0, 2.5 and 5.0% basalt and it changes to higher values of basalt such as 7.5 and 10%. Raw mixes with 0.0, 2.5 and 5.0% basalt show a sharp decrease of insoluble residue from 1200 up to 1300ºC. At 1350ºC, the insoluble residue content is nearly disappeared in the above clinkers up to 5.0% basalt. This shows the effect of basalt as mineralizer in the clinker formation especially with lower value up to 5.0%. As the amount of basalt increases such as 7.5 and 10.0% the insoluble residue increased with basalt content at all firing temperatures. The decrease of LSF tends to increase the insoluble residue in the fired clinkers. Also, the rate of clinker formation in these raw mixes is slower than those with lower values of basalt from 1200ºC up to 1300ºC. As the firing temperature increases the rate of the decrease of insoluble residue is sharply diminished due to the increase of the melt content. As the basalt increases the Al2O3 and Fe2O3 increases, therefore the melt formation enhances. From these findings it can be concluded that the addition of basalt up to 5% to the raw mix increases the formation of clinker at lower temperature than the original mix. The increase of insoluble residue with basalt content is due to the decrease of lime content which reacts with silica to form C3S and β-C2S.
285
(alite) and β-C2S (belite) increase. This mainly due to the mineralizing effect of basalt as well as the decrease of LSF. Also, the silica in basalt is found in the combined state. Therefore, the reactivity of the formation of clinker in the presence of basalt enhances. Figure 5 shows the XRD pattern of the clinkers fired at 1350°C for 2 h. This figure shows increasing of belite phase with basalt content and the alite decreases, especially at 7.5 and 10.0% basalt. The belite is the major phase in the clinkers M4 and M5. These clinkers are of low heat Portland cement.
Fig. 4 XRD pattern of cement raw mixes fired at 1200°C for 2 h
Phase composition of the fired raw mixes (M1-M5)
Figure 4 illustrate the XRD patterns of the fired clinkers (M1-M5) for 2 h at 1200°C. It illustrate a single peak at 32.22 and 37.40 (2θ), indicating the presence of free lime in the all mixes in a decreasing order M1>M2>M3>M4>M5. Also Fig. 4 shows the presence of a single peak at 29.4 (3.03Å), 32.3 and 34.2 2θ indicating alite phase in a small amount in M4 and M5. It is clear that the formation of calcium hydroxide at 18.2, 2θ is mainly due to the hydration of free lime. The free lime is still present after firing at 1200°C as well as Ca(OH)2 in all mixes. This means that the firing temperature at 1200°C is insufficient for firing of Portland cement clinker. The β-C2S is the major phase at this temperature. As the amount of basalt increases the free lime decreases and the clinker phases such as C3S
Fig. 5 XRD pattern of cement raw mixes fired at 1350°C for 2 h
286
INDIAN J. ENG. MATER. SCI., AUGUST 2010
temperature the clinkers gave lowest free lime content and highest main clinker phase formation.
Calculated phase composition in the fired clinkers
The calculated phases of the clinkers fired at different temperatures with various amounts of basalt (M1-M5) are seen in Table 4. It is clear that the ferrite and aluminate phases are nearly the same at 1200, 1300 and 1350°C. This is mainly due to that the formation of these phases appears at low temperatures from 850°C up to 1200°C. On the other side, as the amount of basalt increases the amount of aluminate and ferrite phases increase due to the increase of Al2O3 and Fe2O3 in basalt than in clay. The alite phase starts to form at 1200°C in mixes M1, M2, M3 and M4. Sample M5 has no amount of alite due to the decrease of LSF. Also, as the basalt content increases the alite content decreases. As the firing temperature increases the alite content increases but also decreases with basalt content in the M1-M5. Mix sample M2 with 2% basalt shows the higher value of alite at 1350°C. This means that 2.5% basalt acts as a good mineralizer in the formation of Portland cement clinker with higher values of alite. The higher values of belite are formed in all clinker fired at 1200°C. Also, the belite content increases with the amount of basalt. As the firing temperature increases up to 1300°C the belite decreases due to the increase of alite content on the expense of belite. Samples fired at 1350°C show nearly the same amount of belite. Only 2.5% basalt increases the amount of alite on the expense of belite at 1350°C.
Free lime contents
The free lime contents of cement pastes of clinkers (M1-M5) fired at 1350°C and hydrated for 3, 7, 28 and 90 days are graphically represented as a function of curing time up to 90 days in Fig. 6. The results show that, the free lime contents of the hardened cement pastes increase with curing time up to 90 days. This is due to the continuous hydration of the main cement phases such as C3S and β-C2S that liberating free lime during the hydration. Also, the free lime contents of cement pastes of clinkers M2 and M3 is higher than those of other cement pastes. This is mainly due to the increase of the C3S and β-C2S in these clinkers as a result of addition of basalt as mineralizer. On the other side the cement pastes of clinkers M2 which contains 2.5% basalt, liberates high amount of Ca(OH)2 at all curing times. This fact indicates that the 2.5% basalt is the optimum amount of basalt as mineralizer. Also, the free lime content decreases with the basalt content and the LSF decreases which is proportion to the amount of alite. Alite gives higher values of free lime than belite.
Hydration of Portland cement clinkers (M1-M5)
The kinetics of the hydration of the prepared cements fired at different temperatures are studied by the determination of the liberated free lime and chemically combined water contents of cement pastes up to 90 days. The physico-mechnical properties such as bulk density, total porosity and compressive strength of hardened cement pastes are also determined as a function of curing time and basalt content for all cement paste samples especially for clinkers fired at 1350°C because at this firing
Fig. 6 Free lime contents of hardened cement pastes of clinkers (M1-M5 fired at 1350°C as a function of curing time
Table 4– Phase composition of Portland cement clinkers (M1-M5) fired at different temperatures (mass %) clinker M1 M2 M3 M4 M5
1200°C Alite
Belite
C3A
22.60 23.39 19.14 13.63 0.00
47.87 51.68 56.23 63.00 67.10
10.43 11.42 12.46 12.68 13.43
1300°C C4AF 7.45 7.85 8.49 9.76 9.76
1350°C
Alite
Belite
C3A
46.04 42.82 34.73 22.37 13.22
30.20 36.29 42.86 56.14 65.55
10.43 11.42 12.46 12.68 13.43
C4AF 7.46 7.85 8.49 9.32 9.76
Alite
Belite
C3A
C4AF
41.20 53.10 36.88 24.05 14.87
33.79 28.63 42.88 63.94 63.35
10.46 11.54 12.46 13.00 13.44
8.18 8.45 9.25 10.03 10.47
EL-DIDAMONY et al.: PORTLAND CEMENT CLINKER
The chemically combined water of the hydrated cement pastes up to 90 days of clinkers (M1-M5) fired at 1350°C are graphically plotted as a function of curing time in Fig. 7. It is clear that the cement pastes of clinkers M2 and M3 give higher values of combined water. This is due to the high amount of C3S in the fired clinkers. On the other side as the amount of basalt increases the combined water content decreases due to the decreases of LSF and then C3S. Bulk density and total porosity
The bulk density and total porosity of hardened cement pastes of clinkers (M1-M5) fired at 1350°C and hydrated to 90 days are graphically presented as a function of curing time in Figs 8 and 9. From these figures, it is clear that the bulk density increases and the total porosity decreases with curing
Fig. 7 Chemically combined water contents pastes of clinkers (M1-M5 fired at 1350°C as a function of curing time
Fig. 8 Bulk density of hardened cement pastes of clinkers (M1-M5) fired 1350°C as a function of curing time
287
time. This is mainly due to the progress of hydration process. The hydration products fill a part of the pores therefore the density increases and the total porosity decreases. The hardened cement pastes of clinker M2 gives higher values of bulk density and lower total porosity up to 90 days than those of the other clinkers. As amount of basalt increases the density decreases and the total porosity increases due to the decrease of LSF, which increases the amount of belite on the expense of alite. Belite has lower rate of hydration than alite phase. The compressive strength of hardened cement pastes of clinkers (M1-M5) fired at 1350°C is shown in Fig. 10. It is clear that the compressive strength of cement paste increases with curing time for all cement pastes. As the firing temperature increases the clinker
Fig. 9 Total porosity of hardened cement pastes of clinkers (M1-M5) fired 1350°C as a function of curing time
Fig. 10 Compressive strength of hardened clinkers (M1-M5) fired 1350°C as a function of curing time
288
INDIAN J. ENG. MATER. SCI., AUGUST 2010
curing time. Also, the absorption band at 1458.3 cm-1 is due to presence of carbonate CO32-. Conclusions The results show that the burnability of raw mixes increases with the basalt content and firing temperature. The physico-mechanical properties of raw mix containing 2.5% basalt gives the best hydration characteristics than the other cement pastes. This is mainly due to that the amount of 2.5% basalt is considered the optimum amount of basalt as mineralizer. Fig. 11 IR spectra of hydrated M2 (2.5% BS) as a function of curing time
phases, especially β-C2S and C3S content increase in clinker. Therefore it gives high amount of hydration products, which fill the pores of cement paste and give high strength. Mix M3 with 5.0% basalt gives the higher values at all curing times than other mixes. This is mainly due to the effect of 5.0% basalt as mineralizer for the formation of Portland cement clinker, i.e., the silicate phases are well formed. These phases are the main source of strength. As the amount of basalt increases than 5.0% the strength decreases due to the decrease of alite and the increase of β-C2S, C3A and C4AF. The alite is the major phase for the contribution of compressive strength. Figure 11 illustrates the IR spectra corresponding to hydrated cement paste M2 (fired at 1350°C) cured at 3, 7, 28 and 90 days. The broad band near at 3431.2 cm-1 and small band at 1654.1 cm-1 related to H2O. However, the 3644.6 cm-1 band of OH- in Ca(OH)2 liberated during the hydration of Portland cement13-15. The intensity of the 3431.2 and 3644.6 cm-1 bands increases with curing time due to the progress of hydration process. The absorption band 987.4 cm-1 indicates the formation of CSH, which increase with
References 1 Kolovos K G, Cem Concr Compos, 28 (2006) 133-143. 2 Palomo M T A & Vázquez F T, Cem Concr Res, 14 (1984) 397-406. 3 KlemmW A, Jawed I & Holub K J, Cem Concr Res, 9 (1979) 489-496. 4 Osokin A P & Potapova E N, Silikattechnik, 37 (1986) 79-80. 5 Bordoloi D, Baruah A CH, Barkakati P & Borthakur P CH, Cem Concr Res, 28 (1998) 329-333. 6 Tasong W A, Cripps J C & Lynsdale C J, Cem Concr Res, 28 (1998) 1453-1465. 7 Batic O, Maiza P & Sota J, Cem Concr Res, 24 (1994) 13171326. 8 Tasong W A, Cripps J C & Lynsdale C J, Cem Concr Res, 28 (1998) 1037-1048. 9 Xiaocun Liu, Yanjun Li & Ning Zhang, Cem Concr Res, 32 (2002) 1125-1129. 10 Barros A M, Espinosa D C R & Tenório J A S, Cem Concr Res, 34 (2004) 1795-1801. 11 Barros A M, Tenório J A S & Espinosa D C R, J Hazard Mater, 112 (2004) 71-78. 12 Ract P G, Espinosa D C R & Tenório J A S, Waste Manage, 23 (2003) 281-285. 13 Alexander Mc Birney R, Igneous Petrology, Freeman Ooper and company, 1984. 14 Abd El-Maksoud M A, Petrograpgical geological and physical properties of Egypt basaltic rock, M.Sc. Thesis, Cairo University, 1969. 15 Alons S & Palomo A, Cem Concr Res, 31 (1) (2001) 25-30.
