nd
2 International Conference on Concrete & Development th nd April 30 –May2 , 2005, Tehran, Iran
CD1-001
Alkali Activation of Natural Pozzolan for Geopolymer Cement Production D. Bondar 1, C. J. Lynsdale 1, A. A. Ramezanianpour 2 1. Dept. of Civil and Structural Engineering, University of Sheffield, Sheffield, UK 2. Dept. of Civil Engineering, Amir Kabir University, Tehran
Abstract: The challenge for the civil engineering community in the near future will be to realize structures in harmony with the concept of sustainable development, through the use of high performance materials of low environmental impact that are produced at reasonable cost. Geo-polymeric materials provide a route towards this objective. The main benefit of geo-polymeric cement is the reduction in environmental impact. Using lesser amounts of calcium-based raw materials, lower manufacturing temperature and lower amounts of fuel, result in reduced carbon dioxide emissions for geopolymer cement manufacture by up to 80%-90%, in comparison with Portland cement. This paper reviews geopolymer cement technology and presents preliminary results using activated Iranian natural pozzolans namely, taftan and Shahindej.
Keywords: Natural pozzolan, zeolite, alkali activated pozzolan, kaolinite, geopolymer
1
1. Introduction: The challenge for the civil engineering community in the near future will be to realize structures in harmony with the concept of sustainable development, and this involves the use of high performance materials produced at reasonable cost with the lowest possible environmental impact. Unfortunately, the production of Portland cement, a major component material of concrete worldwide, releases large amounts of CO2 into the atmosphere. It is estimated that the production of 1 tone of OPC results in the release of 1 tone of CO2, a major contributor to the greenhouse effect and the global warming of the planet [1]. Given the huge amounts of concrete used worldwide (one cubic meter of concrete per person per year), cement production is estimated to contribute around 7% of the global CO2 emissions. Geopolymer cement, which is based on the polymerisation of Al-Si minerals using alkalis, offers a route for the production of concrete with low environmental impact. The production of geopolymer cement requires much lower manufacture temperature (<750 C) in comparison to that required for OPC (1400 C) and the manufacture also does not involve the calcinations of materials high in calcium content (e.g. calcium carbonate in OPC). Therefore, overall fuel consumption is reduced to a third and the CO2 emissions are lowered by 80-90% in comparison with OPC production. In addition, materials based on aluminosilicates are abundant naturally worldwide and in many wastes and by-products and geopolymer cement may be manufactured using existing cement works, therefore no new expenditure is necessary. °
°
Geopolymer concrete is a concrete with a geopolymer cement matrix binding fine and coarse aggregates. It is also referred to in the literature [2] as rock-concrete since the finished product is almost identical to natural rock in appearance. With geopolymer cement it is possible to produce a mixture that can be poured, moulded, worked, and yet set far quicker and harder than normal Portland cement concrete [3]. This paper reviews current literature on geopolymer cement and concrete and describes initial results produced with the activation of Iranian natural pozzolan, namely, taftan and Shahindej. 2. Background and forecasting: The idea came out of the Great Pyramid Mystery. Davidovits state stone of the pyramid was not quarried, as block as, directly from the quarries of Turah and Mokattam. The major difference he noticed: The geological content of the 87% limestone is apparently identical, but the stone has a 10% content of synthetic zeolite based on a mix of soluble silica, alumina, and caustic soda. The stone is harder and lighter than the natural quarry stone. The experience shows when bubbles have been present in natural stone, they have always been round, not flattened [4]. In view of the global sustainable development, it is imperative that supplementary cementing materials be used to replace large proportions p roportions of cement in the concrete conc rete industry [1]. These supplementary cementing materials are coal and lignite fly ash, rice husk ash, palm oil fuel ash, other ashes, blast furnace and steel slag, silica fume, limestone, metakaolin, natural pozzolan, and geo-polymer with mineral and metal resources. The greater durability imparted by the fly ash activated with a suitable activator like lime a nd maximum replacement of (55-60)% by the weight of Portland cement to the final concrete element but the tendency in world electricity
2
production is the opposite of implementing more and more coal-fuelled power plants. Even if power plants successfully tackling the quality issue, until year 2015 a maximum amount of 290 million tones of fly ash about 8% of the worlds cement market would be available for cement applications [2]. Slag is by-product of the metallurgical industry. Slag is either crystalline stable solid used as aggregates or glassy material used as hydraulic binder (Regourd, 1986). Slag cements can contain more than 80% slag activated chemically with lime or sulfate and NaOH activators. The cementing potential of granulated blast furnace slag is to a large extent dependent on its glass content. The quantities of blast-furnace slag available for blending with Portland cement are declining in industrial countries due to the changes occurring in metallurgical processes. Due to special cooling process, the problem that slag becomes highly reactive in melting and damages significantly to refractory material during adding it to cement kiln, also being useful as aggregate in concrete it shouldn’t be thought as a significant impact in production of cement. Quenched vitreous slag could represent the availability of respectively 290 to 560 million tones of blast slag in the year 2015 for cement production [2]. Metakaolin is a highly pozzolanic material produced by calcinic China clay at a temperature of 700-900C (EN 206/BS 5328). It is formed heating Kaolin (Al2Si2O5 (OH) 4). A calcination temperature in the range of 700-800c has generally given the best activation [4]. The alkali activation of metakaolin (MK) leads to the production of high-mechanical-performance networkstructure materials. Results show the effectiveness of MK in preventing damage due to alkalisilica reaction, prolonged the initial and final setting times, reduced the porosity, and improved the microstructure by re-crystallizing hydrated calcium silicates together with the forming hexagonal calcium aluminate hydrate [5]. Metakaolin is a product, which is manufactured for use rather than by-product, and the process of producing this material seems similar to geo-polymers. Natural pozzolan are vitreous pyroclastic produced by violent eruptive volcanic action. The ancient Romans used natural pozzolan for producing their famous Roman cement, obtaining by blending lime and pozzolan [2]. A geo-polymeric cement process can transform a wide range of alumina- silica natural or wastes such as fly ash, blast furnace slag, metakaolin, a nd natural pozzolan into building bu ilding products.
Fig.1.Distribution of BaU world cement market for the year 2015 , total 3500 million tonnes
3
3. Properties of Natural pozzolan (as raw material): In this research natural pozzolan are going to use as raw materials for being alkali activated to prepare geo-polymer, so it would be useful to review the literatures for properties of this and different way of using it as building material: Within the building industry, the term “pozzolana” covers all the materials react on limes at presence of water, giving calcium silicate and aluminate hydrates having cementing properties. As a consequence, all pozzolan have to be rich in reactive silica or alumina plus silica. Natural pozzolan can be classified according to their origin and their essential active constituent and therefore, natural pozzolanas are usually classified as follows [6]: Volcanic, incoherent, rich in unaltered or partially altered glass (A characteristic feature of these pozzolanas is the high alkali content which can exceed 10 percent. L.O.I. generally ranges between three percent and six percent but it can be higher in some altered materials. ) Tuffs, where volcanic glass has been transformed, entirely or partially, into zeolite compounds (In these pozzolanas, the L.O.I. value has been considered to be an index of the intensity of the transformation that the original volcanic material underwent owing to weathering.) Sedimentary, rich in opaline diatoms (A high alumina percentage means that the main component of pozzolana, opal, is associated with clay minerals. The presence of clay creates some problems.) Diagenetic, rich in amorphous silica, resulting from the weathering process of siliceous rocks (In some countries, pozzolanic materials with very high silica content occur. This can be due to the transformation of the original minerals into silica gel by the action of hot spring. Silica prevails with percentages reaching 90 percent.) The pozzolanic activity includes two parameters, that is, the maximal amount of lime that a pozzolana can combine, and the rate at which such combination occurs and these depends on % of dissolved SiO2 and Al2O3+SiO2.The main factors affecting the pozzolanic poz zolanic reaction are [6]: Nature and composition of the active phases and their content in pozzolana. Their specific surface The lime/pozzolana ratio of the mix The water/mix ratio The curing time and temperature Methods for the activation of pozzolanic reactivity of natural pozolan [7,8,9]: Acid Treatments is only suitable for low-calcium pozzolan. The prolonged grinding of the natural pozzolan accelerates the pozzolanic reaction during the first 3 days. The thermal activation decreased the ultimate strength. Chemical activation is the most efficient and feasible method for the activation of natural pozzolan. So adding of 4% Na2SO4 or1% CaCl2 .2H2O based on the mass of lime pozzolan blends increased the ultimate strength of lime-pozzolan mixtures significantly. The addition of Na2SO4 can improve both the early and the later strength of hardened LPC paste. CaCl2 isn’t helpful for the early strength of cement pastes, but increases the later strength greatly. The later age strength of the hardened cement pastes with CaCl2 is about 2.2 to 2.6 times higher than that of control pastes, while the strength of pastes with Na2SO4 is only about 1.5 to 1.7 times higher. Pozzolanas significantly increase the chemical resistance of cement. The reasons for the high chemical resistance of pozzolanic cements are many but these can be reduced to two: Lower content of portlandite in the paste Lower permeability of the paste
4
4. Geo-polymers: The term “geo-polymer” describes a family of mineral binders that have a polymeric siliconoxygen-aluminum framework structure similar to that found in zeolites. Geo-polymers are often viewed as the amorphous equivalents of zeolites because they have approximately the same Al:Si ratio as a comparable zeolite but without the crystal structure. In general, depending on starting material, when solid to liquid ratio (S/L) is less than about 0.1, a zeolite z eolite forms, whereas when S/L is greater than about 0.1, geo-polymer forms [10]. Therefore this phenomenon involves a chemical reaction between various alumina-silicate oxides with silicates under highly alkaline conditions, yielding a three-dimensional polymeric chain and ring structure consisting of Si-OAl-O bonds. A high ratio Si:Al, higher than 15, provides polymeric character the geo-polymeric material. As a complex set of fast reactions, it consists of four steps, i.e. dissolution, gelling, condensation and precipitation [10]. The three main steps in the process are similar to those for more zeolites [10]: Dissolution, with the formation of mobile precursors through the complex action of hydroxide ions. Partial orientation of mobile precursors as well as the partial internal restructuring of the alkali poly-silicates. Re-precipitation where the whole system hardens to form an inorganic polymeric structure. These materials can poly-condense just like organic polymers, at temperatures lower than 100C.Geo-polymerization involves the chemical reaction of alumino-silicate oxides with alkali poly-silicates yielding polymeric Si-O-Al bonds; the amorphous or semi-crystalline three dimensional silico-aluminate structures (Davidovits, 1991). A geo-polymer can take one of the three basic forms [10]: Poly (sialite) (-Si-O-AL-O-) (-Si-O-Al-O-Si-O-) Poly (sialate-siloxo) (-Si-O-Al-O-Si-O-) Poly (sialate-disiloxo) (-Si-O-Al-O-Si-O-Si-O-) The distribution and relative amounts of each of the different Al and Si building blocks affect on chemical and physical properties of the final product [14]. [14]. For the chemical designation designation of geo polymers based on silico-aluminates, poly (sialate) was suggested. The sialate network consists + + of SiO2 and AlO4 tetrahedral linked alternately by sharing all the oxygen. Positive ions (Na , K , + ++ ++ + + Li , Ca , Ba , NH4 , H3O ) must be present in the framework cavities to balance the negative + charge of Al3 in IV-fold coordination. Poly (sialates) has this empirical formula: Mn {-(SiO2) z-AlO2} n, wH2O Wherein M is a cation such as potassium, sodium or calcium, and «n» is a degree of poly+ + condensation; «z» is 1, 2, 3. Poly (sialate) is chain and ring polymer with Si 4 and Al3 in IV-fold coordination with oxygen and range from amorphous to semi-crystalline. Geo-polymeric compounds involved in materials developed for industrial applications are either crystalline or non-crystalline (amorphous or glassy structure). Crystalline Poly (sialate) Mn - (-Si-O-Al-O-)n and Poly(sialate-siloxo) Mn-(-Si-O-Al-O-Si-O-)n result from hydrothermal setting conditions, whereas hardening at ambient temperature induces amorphous or glassy structures[11]. Geo-polymerization is a general term to describe all the chemical processes that are involved in reacting alumina-silicates with aqueous alkaline solutions to produce a new class of inorganic binders called geo-polymers. Properties of alumina-silicate gel, and hence the macroscopic characteristics of the geo-polymer, obviously are affected by the nature of the solid raw materials and activating solutions as well as the curing temperature, humidity, pressure, and possible contamination. In general greater physical strengths are achieved if amorphous solids, or solids
5
containing high proportions of amorphous phases, are used as raw materials, when compared with crystalline materials [12]. One hardening mechanism involves the chemical reaction of alumino-silicate oxide (Si2O5, Al2O3) with alkali poly-silicates yielding polymeric Si-O-Al bonds. The fabrication of (Si2O5, Al2O3) is carried out by calcination alumino silicate hydroxides (Si2O4, Al2 (OH)4): 2(Si2O5, Al2 (OH) 4) 2(Si2O5, Al2O2) + 4H2O Poly-silicates are generally sodium or potassium silicates supplied by the chemical industry or manufactured fine silica powder, a by-product of the ferro-silicon metallurgy (condensed silica fume). Polymerization by the alkali can be shown as follows: Na OH, KOH n (Si2O5,Al2O2) +2nSiO2+4nH2O n(OH)3-Si-O-Al-O-Si-(OH)3 l (OH ) 2
Na OH, KOH l l l (Na, K)-(-Si-O-Al-O-Si-O-) K)-(-Si-O-Al-O-Si-O-) + 4nH2O l l l O O O
n(OH)3-Si-O-Al-O-Si-(OH)3 l (OH )2 5. Properties of geo-polymers: This cement hardens rapidly at room temperature and provides compressive strength in the range of 20Mpa, after 4 hours at 20c, when tested in accordance with the standards applied to hydraulic binder mortars. The final 28-days compression strength is in the range of 70-100Mpa [2]. Shrinkage during setting time is less than 0.05%. Investigation show that different natural Al-Si minerals are to some degree soluble in concentrated alkaline solution, with in general a higher extent of dissolution in NaOH than KOH medium. Statistical analysis revealed that framework silicates show a higher extent of dissolution in alkaline solution than the chain, sheet and ring structures. MASNMR spectroscopy can be use to show the shape of silicates. It is significant that all minerals demonstrate higher compressive strengths after geo-polymerization in KOH than NaOH, despite of the higher extent of dissolution in NaOH than in KOH. It is proposed that the mechanism of mineral dissolution as well as the mechanism geo-polymerization can be explained by ion-pair ion-pair theory. Increasing the M2O/H2O ratio (where M=Na and/or K) results in an increase in the dissolution. The SiO2/M2O ratio affects significantly the degree to which polymerization occurs [13]. As stated before, the process of geo-polymerization starts with the dissolution of Al and Si from Al-Si materials in alkaline solution as hydrated reaction products with NaOH or KOH, hence forming the {Mx (AlO2)y(SiO 2)z.nMOH.mH2O} gel. Subsequently, after a short time setting proceeds, with the gel hardening into geo-polymers, consequently an understanding of the extent of dissolution of natural Al-Si minerals is imperative for an understanding of geo-polymerization reaction. Dissolution data of Al-Si minerals in alkaline solution shows the following general trends [13]: Minerals have a higher extent of dissolution with increase concentrations of alkaline solution. o f dissolution in the NaOH than in the KOH solution. Minerals show a higher extent of The concentrations of Si are higher than the corresponding Al, which could be caused partly by higher content of Si than Al in the minerals, but also by the higher intrinsic extent of dissolution of Si than Al.
6
The correlation coefficient between the extents of dissolution of Al and Si is 0.93.So Si and Al could dissolve from the mineral surface in some linked form. Mineral with framework structure possess a higher extent of dissolution than di-ortho, ring, chain, and sheet structures in both NaOH and KOH solutions. Results indicate that K contained in the gel phase, either derived from KOH solution or from the dissolution of K-feldspar, is observed to increase the extent of disorder of gel phase as well as mechanical strength of the formed geo-polymers. The reason for this effect is the higher polymerizing activity between silicate and alumino-silicate species caused by the presence of K [14]. [14 ]. It has been found that for an Al-Si mineral structured with alkali metal Na or K, a higher compressive strength of the geo-polymer could be expected when the alkali metal in the alkaline solution differs from the alkali metal structured in the Al-Si mineral [14]. The compressive strength depends significantly on Si/Al and Na/Al mortar ratios. The microstructure changes with the chemical composition. Some guidance to interpretation of this complex system due to previous literature has been be en shown as below [13]: Factors have a positive correlation with compressive strength are: %CaO in the original mineral Molar Si-Al in the original mineral Use of KOH The extent of dissolution of Si and the molar Si/Al ratio in solution The harness of the mineral Factors have a negative correlation with compressive strength are: %K 2O in the original mineral Use of NaOH The following three factors were identified as having a significant effect on strength: The type of alkali %K2O in the mineral ppm Si in solution When the gel (kao) / gel (Al-Si) ratio becomes very low, it has been observed that the resulting geo-polymers appear cracked and on the other hand if the gel (Kao) / gel (Al-Si) ratio is very high, the resulting geo-polymer is with low gel formation and demonstrates low compressive strength [13]. The Si/Al and Na/Al molar ratios are the nominal compositions based on the starting chemistry and the compressive strengths in (Mpa) vary due to this starting chemistry as below (Matthew Rowles and Brain O’Connor, 2003):
7
Table1. Relation between geo-polymer chemistry and compressive strength Na/Al
Si/Al
0.51
0.72
1.00
1.08 1.5 2.0 2.5 3.0
0.42
2.2 6.2
4.4 23.4 51.3
1.29
1.53
2.00
19.8 53.1 64
11.8 49. 2.6
19.9
7. Experimental investigations: Related binders are known as geo-polymers, alkali activated natural pozzolans are considered in this research. In all cases the activation of the solid is made with alkaline solutions containing alkali hydroxides and alkali silicates. Two different materials were used in our investigations as base materials: Taftan Natural Pozzolan and Shahindej Natural Pozzolan. These two were used lonely and in combination with Kaolinite. The chemical compositions of these materials are shown in Table and Figure 2.
Table 2.Chemical composition of investigated materials
Taftan Pozzolan Shahindej Pozzolan Kaolinite
LOI 1.64 10.62 6.84
SiO2 58.35 71.43 72.88
Al2O3 12.98 8.29 16.65
Fe2O3 4.86 1.49 0.53
CaO 15.01 3.72 2.21
MgO 1.69 1.04 0.15
TiO2 0.58 0.23 0.00
K2O 1.61 1.88 0.45
Na2O 2.70 1.03 0.10
LOI
Kaolinite
SiO2 Al2O3 Fe2O3
Shahindej Pozzolan
CaO MgO TiO2 K2O
Taftan Pozzolan
Na2O
0%
20%
40%
60%
80%
100%
Fig.2. Chemical composition of investigated materials
8
A total of 950 g of natural pozzolan with and without kaolinite (were dry mixed at a specified mass ratio 2.5:1 for 5 minute when is used with kaolinite before the addition of mixed KOH and Na2SiO3 solutions) the ratio of KOH (ml)/ Na2SiO3 (ml) and total dry mix (g)/total solution (ml) were 7.7 and 3.2, respectively. The mixture was then blended by the Hobart Canada N-501425rpm for another 3 min. The resulting paste was transferred to a polyvinyl chloride cylinder (PVC) mould with 50 mm diameter and 100 mm in height, and left in an oven for setting at 40C for 24 hr. After being removed from the mould, the sample was left in the oven for further setting and hardening at 40C for 27 days. Then, the compressive strength of each sample was tested correspond to ASTM C39. 8. Results: Three samples of each condition were tested with average compressive strength values reported as the results in Table 3. For measuring setting time Vicat Needle method method has been used and the results are presented in Table 3. Table 3.Primary results (Setting time & Compressive Strength) Setting time Sample Code
Ingredients
TaKS2.5
Taftan Pozzolan +KOH(2.5Molar)+Na2Sio3
TaKS5
Taftan Pozzolan +KOH(5Molar)+Na2Sio3
TaKS7.5
TaKaKS2.5
TaKaKS5
TaKaKS7.5 ShKS5 ShKS7.5
Taftan Pozzolan +KOH(7.5Molar)+Na2Sio3 Taftan Pozzolan +Kaolinite+KOH(2.5Molar) +Na2Sio3 Taftan Pozzolan +Kaolinite+KOH(5Molar) +Na2Sio3 Taftan Pozzolan +Kaolinite+KOH(7.5Mola)+N a2Sio3 Shahindej Pozzolan +KOH(5Molar)+Na2Sio3 Shahindej Pozzolan +KOH(7.5Molar)+Na2Si3
Method of blending 120 second 60rpm & 120 second 120rpm & 60 second 180rpm 45 second 60rpm & 60 second 120rpm &75 second 180rpm 45 second 60rpm & 60 second120rpm &75 second 180rpm 120 second 60rpm &120 second 120rpm & 60 second 180rpm
Compressive strength 7days 28days (mpa) (mpa)
6hr +45min
17. 6
18.2
5.5hr
15.07
15.13
75min
10.2
15.13
6hr + 15min
13.12
11.14
8.86
13.62
50min
7.0
16.83
180 second 60rpm
-
-
4.33
180 second 60rpm
-
-
9.15
45 second 60rpm &135 second 120rpm 45 second 60rpm & 60 second 120rpm &75 second 180rpm
4.5hr
9. Conclusion: 1. The primary results show the best concentration for KOH solution is 5molar. 2. Existence of Kaolinite in the mixture reduces the strength. 3. Humidity control during curing should be considered in the future experiments. 4. There is only about 15 minutes between beginning and end of setting.
9
10-References: 1. Bilodeau and Mohan Malhotra, “High-Volume Fly Ash System: Concrete Solution for Sustainable Development”, ACI Material Journal, Vol.1, No.1, 20003. st 2. Davidovits, “High-Alkali Cements for 21 Century Concretes”, Concrete Technology Past, Present, Future, P.383-397, 1994 3. Taylor H., Cement Chemistry, Second edition, Thomas Telford, London , 1997 4. Davidovits, Building materials, McMullen, 2003 5. Tixier and Mobasher, “Blended Cements”, Journal of Cement & Concrete, Cement Research Progress, Chapter 5, 1997 6. Massazza, Structures and Performance of Cements, J. Bensted, P. Barnes, University of Sheffield, 2003 7. Shi and Day, “Acceleration of strength gain of lime-pozzolan cements by thermal activation”, Cement and Concrete Research, Vol.23, No. 4, P.824-832, 1993 8. Shi and Day, “Chemical activation of blended cements made with lime and natural pozzolans”, Cement and Concrete Research, Vol.23, No. 6, P.1389-1396, 1993 9. Shi and Day, Comparison of different methods for enhan cing reactivity of pozzolans”, Cement and Concrete Research, Vol.31, No. 5, P.813-818, 2001 10. www.cepmagazine.org www.cepmagazine.org,, 2001 11. Davidovits, “Geo-polymers: Inorganic Polymeric New Materials”, Journal of Thermal Analysis, Vol. 37, P. 1633-1656, 1991 12. Lee & Deventer, “The interface between natural siliceous aggregates and geo- polymers”, Cement and Concrete Research 34, Issue 2, P.195-206, 2003 13. Xu & Deventer, “The geo-polymerization of alumino-silicate minerals”, International Journal of Mineral Processing, Vol. 59, No.3, P.247-266, 2000 14. Xu & Deventer, “The effect of alkali metals on the formation of geo-polymeric gels from alkali-feldspars”, Colloids and Surfaces Surfaces A: Physicochem. Eng. Aspects, Vol. 216, P2744, 2003
10
