Geopolymers Using Rice Husk Silica and Metakaolin Derivatives; Preparation and Their Characteristics ()
1. Introduction
Geopolymerization is an effective manner to process and reuse silicates, aluminates or aluminosilicates in mineral waste. It is known that geopolymers are alkaline activated aluminosilicates consisting of reactive species of SiO2 and Al2O3 to produce high-strength materials that can effectively immobilize other industrial by-products, and even hazardous waste. Generally, the formula of the geopolymers is Mn(-(Si-O2)z-Al-O)n.wH2O, where M is an alkaline cation (Na, K or Ca), z is generally assigned a value of 1, 2 or 3 and n is the degree of polymerization [1] . Geopolymers are interesting materials formed by polycondensation of the silicon and aluminum species resulting from the dissolution of their raw materials found in fly ash, slag, metakaolin in alkaline solution condition [2] . Usually, the geopolycondenzation depends of SiO2/Al2O3 ratio and alkali concentration [3] [4] . Geopolymerization is affected by the reactive precursor, meaning that OH-concentration and curing conditions finally produce three-dimensional networks that comprise geopolymer matrices at different time and temperature [5] .
Therefore, the advantage is the proven durability and erosion resistance of both ancient mortars and cements and modern geopolymer cements. It was known that a great number of minerals-industrial wastes were studied as raw materials for geopolymer synthesis, including pozzolana [6] , natural aluminosilicate minerals, metakaolin, fly ash, granulated blast furnace slag, fly ash and kaolinite mixture, red mud and metakaolin mixture [3] [7] -[12] .
More importantly, the starting raw materials play a significant role in the geopolymer reaction and affect the mechanical properties and microstructure of the final geopolymericproducts [3] [11] [13] . It would be expected that silica from of rice husk (RH) is an excellent pozzolanic material, but that has not been extensively studied in geopolymerization. RH silica can be generate by the combustion process of RH [14] , itself is biomass and to be an issue for sustainability environment, due to its abundant amount and capability to resist natural degradation. RH silica therefore is an alternative material with good pozzolanic reactivity, to be included in the geopolymer mixture to complete the ratio of modulus of SiO2/Al2O3. In the present study, the effect of the RH silica was determined on the resultant geopolymers as a function of curing time and temperature. It was investigated that the rheometry properties of the geopolymer pastes and properties of resultant geopolymers were compared at different SiO2/Al2O3 ratios at different cure temperature in addition with their bulk characteristics. Since such biomass resource was attractively interesting in point of view of sustainability, this paper focused on the detail in the geopolymerization.
2. Experimental
2.1. Materials
Silica was obtained from by leaching of rice husk in dilute HCI (5 M) and burned at 700˚C to produce pure amorphous white silica. The chemical composition of obtained silica was determined by XRF reveals at 99.5% of SiO2 with a surface area calculated from BET method of 207.5 m2/g. Obtaining Metakaolin was carry out by dehydroxylation of kaolinite at 750˚C by 20 h according to reference [15] [16] . The chemical composition of metakaolin was measured by XRF for 52% of SiO2 and 42.1% of Al2O3. The surface area of metakaolin calculated from BET method was 11.5 m2/g.
2.2. Preparation of Geopolymer
Geopolymers were prepared by changing a molar ratio of SiO2/Al2O3 = 3.0 and 10 for GP-3 and GP-10, respectively. For NaOH and Na2SiO3 molar ratio, Na2O/SiO2 = 0.25 was used. The preparation procedure is as followed as described in Figure 1. The powders of RH silica were mixed with NaOH solution until dissolution. Then, metakaolin (MK) powders were added and mixed together for 15 min in the silica solution. The paste samples were cast in plastic molds with a diameter of 20 mm and height of 40 mm. The curing time for preparation of geopolymers was carried out at 85˚C, 100˚C and 200˚C in a conventional oven at open mold(). Before the curing process, rheological measurements of the paste samples were performed using a controlled stress rheometer (Anton PaarPhysica MCR 301 Rheometer), as operated in controlled strain mode. The geopolymer paste was prepared manually outside the rheometer and then introduced into the rheometer disk after mixing. Transferring the geopolymer to the measurement cell generates residual stresses, when the sample was prior to
Figure 1. Protocol of geopolimerization synthesis.
each rheological measurement. This was done before each test by applying a strain at frequency of 1 Hz. Characterization of the geopolymers was obtained by infrared spectroscopy (FT-IR), X-ray diffraction (XRD) and scanning electron microscope (SEM).
3. Results and Discussion
Figure 1 presents synthesis protocol of RH silica-MK geopolymerization. In this process, the RH silica and MK were produced by sintered rice husk and kaolin at 700˚C and 750˚C, respectively, before the provided.
Figure 2 shows the strain sweep at a frequency of 1 Hz for raw material activated by NaOH and geopolymer paste with a satisfactory repeatability. It was observed in the viscous elastic rheometer that the deformation of the geopolymers pastes (a) was higher than that of RH silica and MK (b). It was understanding that the linear viscoelasticity region was due to nondestructive to the geopolymerization of the silica-alumina component in the
Figure 2. Strain sweep test for (a) geopolymer paste of GP-3and GP-10 at 25˚C, and (b); silica from rice husk and metakaolin activated by NaOH.
activated paste, when a critical strain gC was applied. Strain was generally taken as the strain value at the storage modulus equal to 90% of the plateau modulus. In both cases for GP-3 and GP-10, the elastic modulus (G’) came to be higher than that of viscous modulus (G”). That meant that the paste was considered in the gelation for solid.
When comparison made between the paste of GP-3 and GP-10, high values of G’ and G” for geopolymers pastes were observed, indicating that the geopolymerization in the activator condition was easily occur at the point of G”/G’ = 1, GP-3 containing higher alumina and higher strain. This means that the high MK content was better for the polycondensation reaction.
Figure 3 show time curves of G’ and G” at different temperature. It was also observed at fixed temperature of 60˚C, 85˚C and 100˚C for GP-3 and GP-10. The both G’ and G” were increased with increasing time and then
Figure 3. The G’ and G” as a function of time and maintaining a constant temperature at 60˚C, 85˚C and 100˚C for GP-3 and GP-10 [®for elastic modulus (G’) and ¯ for viscous modulus (G’’)].
became to be constant at about 250 seconds. The comparison between GP-3 and GP-10 meant that the cure ability of the activated paste was higher in GP-10. The time curves suggested that the curing was finished at about 2 min heating. But, the GP-3 increased the G’ and G” values near 200-250 seconds and then was cured in the geopolymer solidified. Due to that the polycondensation reaction occurs, the geopolymer paste loses viscosity, while the paste became a solid material, the elastic modulus became constant at the longer time. This change was exatly observed after 250 s for GP-3 at 60˚C and 210 s for GP-3 at 85˚C and 100˚C. At a high amount of Al2O3 for GP-3, it was less to be on geopolymerization. However, the GP-10 registered a constant modulus G’ and G” for all the cases due to with a high content of RH silica on the geopolymer pastes which made a very elastic and viscous material even after 250 s. Figure 4 shows viscosity of the pastes of GP-3 and GP-10 at different temperature. It was noticed that the viscoelasticity of the paste decreased, when the temperature was high.
The resultant viscosity of the GP-10 was higher than that of GP-3, due to the increment of the SiO2 in geopolymer composition resulted in a decrease of the viscosity of the pastes. After polycondensation reaction for GP-3 and GP-10, the microstructures of geopolymers at different curing temperature were observed. Figure 5 shows
Figure 4. Viscosity test for silica from rice husk, metakaolin activated by NaOH and geopolymer pasteof GP-3 and GP-10 from 20˚C to 60˚C.
Figure 5. SEM imagines of (a) GP-3 at 85˚C; (b) GP-3 at 100˚C; (c) GP-3 at 200˚C; (d) GP-10 at 85˚C; (e) GP-10 at 100˚C and (f) GP-10 at 200˚C.
microscope picture for the resultant geopolymers. The pictures showed the formation of the heterogeneous matrix after the polycondensation of raw material. It was seen that the geopolymers of the GP-3 became to be more porosity matrix. Their porosities of the geopolymers decreased at the GP-10 containing higher RH silica contents. It was seen that with the increment of the temperature, changing into a homogeneous phase was found in the geopolymer matrix, wich was condensed at 200˚C of the geopolymer matrix.
The XRD patters of the geopolymers of GP-3 and GP-10 at different curing temperature are shown in Figure 6. It was seen that the peak shift toward higher 2q side was significant in the cases of GP-3. This meant that the
Figure 6. X-ray diffraction of geopolymers synthetized by alkali activation of silica from rice husk and metakaolin different curing time.
reconstruction of silica-alumina order in the paste was highly observed at the polycondensed GP-3. This was strongly demonstrated geopolymerization with the amorphous nature of the samples. In addition with the XRD results, the alteration and restructuring of materials during geopolymerization were also suggested in the diffraction pattern by the marked shift in the scattering diffraction peak for each samples. When the starting materials were activated with an alkaline solution, the scattering diffraction peak shifted from ~20˚ to ~25˚ - 28˚ in 2q. This suggested that the local bonding environment was changed during the geopolymerization process [17] . Apparently, an increase of the temperature from 85˚C to 200˚C, no influence of the amorphous structure of the geopolymers was observed.
The data suggested that the silica/alumina component was affected to the reconstruction. In order to confirm this from chemical reaction of RH silica and MK we measured FT-IR spectra of the resultant GP-3 and GP-10. Figure 7 shows the infrared spectra of the geopolymer matrixes. The band around 460 cm−1 was related to Al?O and Si?O in plane and bending modes, 730 cm−1 with octahedral Al and 820 cm−1 with tetrahedral Al?O stretching and, the peak 1030 cm−1 was assigned with asymmetric Si-O-Al stretching [18] [19] . The change in the intensity of the IR peaks was associated with the structural reorganization by the geopolymerization of RH silica and MK. In the geopolymer samples prepared using the RH silica, there was an increase and broaden in the IR intensity of IR peaks indicating higher polymerization, when the content of the RH silica was high. The IR peak around 1030 cm−1 meant that overlapping of the IR peaks related to SiQn (n = 0 ? 4) structural units [20] -[24] was caused and became broaden, especially a higher temperature. In addition, the spectral data has broadening peak at 3400 cm−1 for OH stretching of Si-OH and Al-OH. It was apparent that the GP-10 system contained broadening tendency on the OH stretching peak. This meant that higher RH silica content easily occurred in the geopolymerization with MK.
4. Conclusion
Based on the result of these analyses, it could be concluded that silica obtained from rice husk combustion at 700˚C became to be raw material for geopolymers and significantly contributed to the formation of geopolymer
Figure 7. Infrared spectra of geopolymers synthetized by alkali activation of silica from rice husk and metakaolin at different curing time.
matrix with MK. The presence of the RH silica was an important during the geopolymerization reaction due tosodium hydroxide activation. The evolution of the viscoelastic parameters of the geopolymers GP-3 and GP-10 were performed by dynamic rheology method for pastes. In general, the elastic modulus quickly exceeded the viscous modulus regardless of the geopolymer studied, indicating that both materials was on the process of solidified geopolimerization having reconstructed Si and Al species.