- Email: [email protected]
Retrieving clay minerals from stabilised soil compacts
Applied Clay Science 101 (2014) 362–368
Contents lists available at ScienceDirect
Applied Clay Science journal homepage: www.elsevier.com/locate/clay
Research paper
Retrieving clay minerals from stabilised soil compacts B.V. Venkatarama Reddy ⁎, M.S. Latha 1 Department of Civil Engineering, Indian Institute of Science, Bangalore 560012, India
a r t i c l e
i n f o
Article history: Received 6 September 2012 Received in revised form 31 July 2014 Accepted 21 August 2014 Available online 22 September 2014 Keywords: Clay mineral Soil stabilisation Stabilised soil block Rammed earth Compressed earth block Adobe
a b s t r a c t Stabilised soil products such as stabilised soil blocks, rammed earth and stabilised adobe are being used for building construction since the last 6–7 decades. Major advantages of stabilised soil products include low embodied carbon, use of local materials, decentralized production, and easy to adjust the strength, texture, size and shape. Portland cement and lime represent the most commonly used stabilisers for stabilised soil products. The mechanism of strength development in cement and lime stabilised soils is distinctly different. The paper presents results of scientific investigations pertaining to the status of clay minerals in the 28 day cured cement and lime stabilised soil compacts. XRD, SEM imaging, grain size distribution and Atterberg's limits of the ground stabilised soil products and the natural soil were determined. Results reveal that clay minerals can be retrieved from cement stabilised soil products, whereas in lime stabilised soil products clay minerals get consumed in the lime–clay reactions and negligible percentage of clay minerals are left in the stabilised soil compacts. The results of the present investigation clearly demonstrate that cement stabilisation is superior to lime stabilisation in retrieving the clay minerals from the stabilised soil compacts. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Soil is formed due to natural weathering of rocks and it takes millions of years for soil formation. Clay minerals are the essential component of a soil in controlling its engineering characteristics and are essential for supporting the plant growth on soils. Soils are used for the manufacture of construction materials such as burnt clay bricks, which are consumed in bulk quantities. Also, it is used in the form of stabilised soil for the manufacture of blocks (stabilised soil blocks and stabilised adobe) and in-situ construction (rammed earth). Stabilised soils have been successfully used for the construction of road subbases (Chasten, 1952; Lambe, 1962; De, 1964; Ingles and Metcalf, 1972; Robertson and Blight, 1978). The use of stabilised soils for the construction of buildings can be noticed since thirties. Soils blended with stabilisers are generally compacted when used for either building construction or any other application like road sub-base, and retaining structures. Varieties of stabilisers have been explored for stabilised soil products. Inorganic stabilisers such as cement and lime are commonly employed for the production of stabilised soil products. Stabilised soil blocks (also known as stabilised compressed earth blocks), stabilised rammed earth and stabilised adobe represent some of the stabilised soil products being used for building construction since the last 6–7 ⁎ Corresponding author. Tel.: +91 80 2293 3126; fax: +91 80 23600404. E-mail addresses: [email protected] (B.V. Venkatarama Reddy), [email protected] (M.S. Latha). 1 Tel.: +91 80 2293 2327.
http://dx.doi.org/10.1016/j.clay.2014.08.027 0169-1317/© 2014 Elsevier B.V. All rights reserved.
decades. These products are considered energy efficient and have low embodied carbon when compared to the conventional materials like burnt clay bricks (Venkatarama Reddy and Jagadish, 2003; Venkatarama Reddy, 2009; Venkatarama Reddy and Prasanna Kumar, 2010). Generally, soils contain larger amount of clay fraction. Clay content of such soils is adjusted to an optimum level by diluting with sand and then used for stabilised soil block (SSB) production and stabilised rammed earth construction. Ordinary Portland cement (OPC) and lime are the commonly used stabilisers. Soil characteristics, density and stabiliser content play a crucial role in controlling the properties of SSB and stabilised rammed earth construction. Vast amount of knowledge has been accumulated on SSB technology and rammed earth (Olivier and Ali, 1987; Heathcote, 1991; Keable, 1996; Walker and Stace, 1997; Hall, 2002; Houben and Guillaud, 2003; Walker, 2004; Hall and Djerbib, 2004; Venkatarama Reddy and Gupta, 2005; Walker et al., 2005; Venkatarama Reddy et al., 2007; Jayasinghe and Kamaladasa, 2007; Easton, 2007; Venkatarama Reddy and Prasanna Kumar, 2009; Venkatarama Reddy and Prasanna Kumar, 2011a, 2011b, 2011c and many other publications). The mechanism of strength development in cement and lime stabilised soils is distinctly different. In a wetted soil–cement mixture, cement hydrates and cementitious products such as calcium-silicate hydrate (CSH) and calcium aluminate hydrate (CAH) are formed apart from the release of small percentage of calcium hydroxide (lime). These hydration products establish water insoluble bonds and bind the gravel, sand and silt particles. Cementitious products such as CSH
B.V. Venkatarama Reddy, M.S. Latha / Applied Clay Science 101 (2014) 362–368
2. Methodology and the experimental programme X-ray diffraction (XRD) analysis and specific surface estimation were carried out on the ground stabilised soil compacts. Details of XRD analysis and specific surface area estimation are as follows. The XRD patterns were generated from less than 2 μm size samples obtained from the ground samples. The sample passing 75 μm was mixed with distilled water in a beaker, then the mixture was poured into a jar and again agitated. The agitated mixture was allowed to settle down. The silt particles will settle first leaving a suspension containing floating fine clay particles and crystalline CSH, CAH, etc. The suspension with floating particulate matter was decanted and dried. XRD analysis was run on these dried samples which are basically either clay or crystallised CSH and CAH, and calcium hydroxide. The particle size of the aggregated CSH particles in hydrated cement paste will be in the range of 0.8 to 3 μm (Kumar Mehta, 1986). XRD patterns of powdered samples were collected on a fully automated Bruker D8 Advance diffractometer operating at 40 kV and 40 mA using CuKα radiation. Data was collected between 5° and 100° 2θ. The SEM images were obtained using Quanta LV/ESEM with the capacity for operation at high pressures as well as under environments such as water vapour, in addition it is
equipped with a standard secondary (Evehat–Thorley) and solid state scatter detector. The Brunauer, Emmett, and Teller (BET) gas adsorption theory is the foundation for the measurement of surface area in high specific surface materials. Measurements in this study were conducted using gas sorption analyser. Nitrogen was selected as the absorbate. The samples were outgassed at 100 °C. The experimental work involved the preparation of stabilised soil compacts, curing them for 28 days, crushing the cured samples and examining for the presence of original clay minerals in the crushed samples. A natural soil having kaolinite clay mineral was used in the investigations. The natural soil has a specific surface area of 5.40 m2/g and when it was ground in a mixer grinder for 12 min its specific surface increased to 8.1 m2/g. The mixer grinder consists of a metal blade set rotating at 1500 rpm in an enclosed plastic jar. Such grinders are used for grinding solid particles of less than 5 mm to fine powder form in dry condition. Grain size distribution curves for the natural soil and ground soil are shown in Fig. 1. Table 1 gives the details of the characteristics of both the samples. The reasons for grinding the soil sample are discussed in Section 2.2. The grain size curves for the natural soil and the ground soil (ground for 12 min) show the variations in the particle sizes among the two soils. When the soil was ground for 12 min (specific surface = 8.1 m2/g) the percentage of clay and silt size fractions increase whereas the sand size fraction decreases when compared with the grain size fractions of the natural soil. During the grinding process, some of the sand and silt size particles get crushed to clay and silt size, and resulting in higher clay and silt size fractions. The liquid limit and plasticity index values of the ground soil sample are 33.9% and 25.1 respectively. As the particles become much finer their water holding capacity increases as indicated by the increase in the value of liquid limit. Generally, the stabiliser content used in the production of stabilised soil blocks, rammed earth, stabilised adobe, etc. will be in the range of 5–10% by weight. Therefore, lime and cement contents were varied between 4 and 10% representing the wide range of stabiliser percentages. 2.1. Preparation and testing of compacted stabilised soil specimens Ordinary Portland cement and laboratory grade lime (with 95% assay) were used for the preparation of compacted cylindrical specimens. The procedure adopted to prepare the cylindrical specimen of size: 38 mm diameter and 76 mm height is as follows. (a) The soil was oven dried at 60 °C to constant weight and then blended with cement or lime by mixing in a ball mill for 10 min. Ball milling is to ensure uniform mixture. (b) Requisite quantity of potable water was sprayed on to the dry mixture of soil and stabiliser, and then a uniform mixture was
100 80
Percent finer
and CAH do not react with clay particles, whereas the small percentage of lime released in the cement hydration process can react with clay minerals forming additional cementitious products. The strength development in cement stabilised soil compacts (CSSC) is mainly attributed to the formation of cementitious products such as CSH and CAH. The clay minerals are pozzolanic in nature and can readily react with lime in the presence of moisture. Therefore, the lime–clay reactions lead to the formation of water insoluble cementitious gel of silicate and silicate–aluminates and this gel finally crystallises into hydrates of calcium silicate, calcium aluminates, etc. with time. The cementitious products formed in lime–clay reactions are similar to those formed during hydration of OPC. The cementitious gel coats the soil particles and establishes bonds between them. The pace of lime–clay reactions is very slow at ambient temperature and curing conditions. There are arguments favouring lime stabilisation of soils instead of cement stabilisation with the assumption that there will be recarbonation of lime in the lime stabilised soil leading to carbon dioxide mitigation. A comparison between the mechanisms of strength development in cement and lime stabilised soils reveals that both the clay minerals and calcium hydroxide get consumed in the lime stabilised soils, whereas clay minerals are nearly intact in the cement stabilised soils. The present investigation attempts to prove this point. Stabilised soil products as well as burnt clay bricks use soil as the main ingredient. The manufacture of burnt clay brick results in irreversible structural changes to clay minerals and permanently changes the soil into a nearly rock form. It is not possible to get back the natural clay minerals from the burnt clay bricks, unless allowed to undergo weathering for millions of years. Whereas there is a possibility of retrieving the clay minerals from some of the stabilised soil products after the end of life of such products without much environmental costs. In the context of sustainability and green buildings there is an emphasis on developing low embodied carbon materials effecting minimum changes to the natural raw materials. In the framework of life-cycle and sustainability analysis of buildings attention has to be paid to the end of life use of a building material. At the end of life if a building material can go back to its native state with minimum environmental costs, then it can be classified under the category of green and sustainable building materials. Considering the Portland cement and lime as soil stabilisers the present investigation attempts to examine the status of clay minerals in the stabilised soil compacts. This is mainly to address the question of recyclability of soil based materials as soil at the end of life of such construction materials. The results of the experiments carried out on the 28 day cured cement and lime stabilised compacted specimens are discussed in the present investigation.
363
Natural soil Ground soil
60 40 20 0 0.001
0.01
0.1
1
Particle size (mm) Fig. 1. Grain size distribution curves of natural soil and ground soil.
10
364
B.V. Venkatarama Reddy, M.S. Latha / Applied Clay Science 101 (2014) 362–368
3. Results and discussion
Table 1 Characteristics of soils. Properties Textural composition, (% by weight) Sand (4.75–0.075 mm) Silt (0.075–0.002 mm) Clay (b0.002 mm) Atterberg's limits Liquid limit, % Plasticity index Unified soil classification Clay mineral pH
Natural soil
Ground soil
72.6 11.6 15.8
44.0 36.4 19.6
26.9 17.5 SC Kaolinite 8.06
33.9 25.1 CL Kaolinite 7.84
obtained by mixing manually. Moulding water content was maintained at 12.5% by weight of the mix. It should be noted here that the cylindrical specimens were prepared employing static compaction process and hence the use of proctor OMC as moulding water content is not valid. More information on static compaction of soils can be found in the studies of Venkatarama Reddy and Jagadish (1993). (c) The wetted mix was fed (known weight) into an open ended cylindrical mould and compaction was carried out from both the ends using a mechanical screw jack set-up. The specimen was extruded from the mould immediately after the compaction. Compacted specimens were kept for curing under wet burlap after 24 h of casting. The weight of the mix fed into the mould was controlled to achieve a dry density of 1800 kg/m3 for the specimen. The strength of compacted stabilised soil specimens increases with the increase in dry density (Venkatarama Reddy and Walker, 2005; Venkatarama Reddy and Prasanna Kumar, 2011b, 2011c) and hence it is kept constant. The compacted stabilised soil specimens were cured under wet burlap for 28 days and then air-dried for 14 days. The air dried specimens were oven dried at 50 °C to constant weight and then tested for compressive strength. Wet compressive strength was determined by testing the saturated specimens (soaked in water for 48 h prior to testing) and dry compressive strength was determined by testing them in oven dry state. The cylindrical specimens were tested in a loading frame at constant piston displacement of 1.25 mm/min.
3.1. Strength of stabilised soil compacts Wet and dry compressive strength of the stabilised soil compacts was determined. Fig. 2 shows a plot of strength versus stabiliser (lime or cement) content. The results shown in the figure represent the mean of six specimens. The figure shows that the strength increases with an increase in the stabiliser content for both the wet and dry cases irrespective of the stabiliser type. The strength (both wet and dry cases) increases by about three times as the cement content increased from 4 to 10%, whereas in case of lime stabilised compacts the strength increase is about 1.5 times. There is a considerable difference in the wet and dry strengths of the stabilised soil compacts. Wet strength to dry strength ratio is in the range of 0.4–0.5 for cement stabilised soil compacts and 0.50–0.55 for lime stabilised soil compacts. More information on the strength of stabilised soil blocks and rammed earth can be found in the literature cited in Section 1. 3.2. XRD patterns, SEM examination and EDS analysis Fig. 3 shows the XRD patterns for both cement and lime stabilised samples and also for the natural soil. The XRD pattern of natural soil shows the presence of only kaolinite clay mineral as the suspended fine particles were only clay minerals. XRD patterns for cement and lime stabilised samples show the presence of kaolinite as well as the cementitious products like CSH and CAH. Also, calcium hydroxide (CH) peaks can be seen in both the lime and cement stabilised samples. CH peaks in cement stabilised samples arise from the lime released during cement hydration process. In case of lime stabilised samples CH peaks indicate the un-reacted lime. There are very few peaks showing kaolinite mineral in the XRD of 7% and 10% lime samples (Fig. 3). The quartz peaks are missing in the XRD patterns shown in Fig. 3. This is attributed to quick settling of heavier silt particles and the decanted materials from the suspension in the jar are finer particles of clay and cement hydration products (CSH, CAH etc.). CAH and CSH products are formed during cement hydration process in the cement stabilised samples. Lime stabilised soil samples also show the formation of CSH and CAH products due to the lime–clay reactions. SEM image shown in Fig. 4(I) for the natural soil does not show the presence of cementitious products (such as CSH and CAH). SEM images and EDS analysis (Fig. 4 II to VI) for lime and cement stabilised soil
2.2. Status of clay minerals in the stabilised soil compacts
1. Grain size distribution 2. Atterberg's limits 3. X-ray diffraction (XRD) and 4. SEM imaging and EDS analysis.
Compressive strength (MPa)
Stabilised soil compacts are cemented materials and when exposed to normal weathering process will take a long time (may be several decades) to disintegrate into smaller particles. In order to examine the stabilised soil compacts for the presence of clay minerals the samples have to be crushed to a powder form. Therefore, stabilised soil compacts after the compression test were oven dried at 50 °C and then crushed to a powder form. The crushing/grinding of the samples was carried out in a mixer grinder for 12 min. The specific surface area of the ground samples is 7.04 m2/g and it increases to 12.42 m2/g for the fraction less than 75 μm size. The powdered samples were examined for the presence of clay minerals. The natural soil was also ground for 12 min and the properties of the ground natural soil were compared with those of powdered stabilised soil compacts. The following tests were performed on these powdered samples to ascertain the nature of products formed, presence of clay minerals and the percentage of clay fraction in the stabilised soil compacts.
6 Dry (Cement) Wet (Cement) Dry (Lime) Wet (Lime)
5
4
3
2
1
0 4
5
6
7
8
Stabiliser content (%) Fig. 2. Strength versus stabiliser content.
9
10
B.V. Venkatarama Reddy, M.S. Latha / Applied Clay Science 101 (2014) 362–368
365
Fig. 3. XRD patterns for natural and stabilised soils on fractions less than 2 μm.
Fig. 4. SEM images and EDS analysis of the natural and stabilised soil samples. (I) Natural soil, (II) 7% cement, (III) 10% cement, (IV) 4% lime, (V) 7% lime and (VI) 10% lime; (A: sand/silt particles; B: kaolinite; C: CSH/CAH; D: calcium hydroxide).
B.V. Venkatarama Reddy, M.S. Latha / Applied Clay Science 101 (2014) 362–368
samples indicate the formation of cementitious products (CSH, CAH etc.). This is in tune with the observations of Millogo and Morel (2012) on the SEM and EDS analysis of cement stabilised adobe brick samples, where they detected the presence of cementitious compounds as well as kaolinite clay mineral. Also, Bell (1996) and Millogo et al. (2008) report the formation of CSH and CAH due to the reaction between kaolinite and lime. It is not possible to quantify the clay minerals from the XRD patterns and SEM images. Therefore, grain size analysis of the samples was undertaken on the ground samples. The presence of kaolinite clay mineral in the silt and sand size fractions of the stabilised soil samples was examined through XRD analysis. Fig. 5 shows the typical XRD patterns for silt and sand size fractions in cement (7%) and lime (7%) stabilised soil samples as well as natural soil sample. Kaolinite is absent in the coarser particles (silt and sand size fractions) of natural soil samples, whereas the presence of kaolinite can be seen in the sand and silt size fractions of cement stabilised soil samples. Kaolinite is absent in the silt and sand size fractions of lime stabilised soil samples. It is interesting to note that there is clay mineral in the sand and silt size fractions of cement stabilised soil. Here, the clay minerals are entrapped in the interstices of aggregated coarser soil lumps (i.e. silt and sand size particles) and if such coarser soil particles are allowed for further grinding or natural weathering the clay minerals entrapped in their interstices can get released.
3.3. Grain size and clay fraction Grain size analysis of the ground samples was carried out mainly to assess the clay size fraction in the samples. The grain size distribution curves for the powdered stabilised soil compacts as well as natural soil sample are displayed in Fig. 6 for both the cement stabilised and lime stabilised samples. Fig. 7 shows the plot of clay size fraction versus stabiliser content. The following observations can be made from the results shown in these figures.
Percent finer
366
100 90 80 70 60 50 40 30 20 10 0 0.001
Natural soil 4% cement 7% cement 10% cement 4% lime 7% lime 10% lime
0.01
0.1
1
10
Particle size (mm) Fig. 6. Grain size distribution curves.
The percentage of sand, silt and clay size fractions reduce as the stabiliser content increases. The ground natural soil (specific surface 8.1 m2/g) has a clay size fraction of 19.6%. This value steadily reduces as the stabiliser (cement or lime) content increases (Fig. 7). Cement stabilised samples show a clay size fraction of 10.6%, 8.8% and 6.3% for 4%, 7% and 10% cement contents respectively. The corresponding values for lime stabilised samples are 3.5%, 2.8% and 1.7%. There is a considerable difference in the clay size fractions among cement and lime stabilised samples. Lower percentage of clay size fraction in lime stabilised samples can be attributed to the consumption of clay minerals in the lime–clay reactions. Lime–clay reactions are responsible for the formation of cementitious materials in facilitating the strength development in the lime stabilised compacts. In lime stabilised soils both the clay minerals and the lime get transformed into new products such as CSH and CAH. Therefore, there is not much free lime left for re-carbonation. The samples were ground in a mixer grinder to get a specific surface of 7.04 m2/g. Clay size fraction of the ground samples contains a mixture
Fig. 5. XRD patterns for cement and lime stabilised soils on silt and sand size fractions.
B.V. Venkatarama Reddy, M.S. Latha / Applied Clay Science 101 (2014) 362–368
367
35
Liquid Limit (%)
Cement Lime 30
25
20
15 0
2
4 6 8 Cement or Lime (%)
10
Fig. 9. Variation in liquid limit with stabiliser content. Fig. 7. Clay size fraction versus stabiliser content.
of clay particles as well as ground sand and silt particles. Clay size fraction increases with the grinding duration as illustrated in Fig. 8 for the two types of samples (i.e. 7% cement and 7% lime). In case of lime stabilised sample the increase in clay size fraction with the grinding duration is mainly attributed to the increased percentage of broken clay sized, sand and silt particles, as the clay mineral is absent in the coarser particles (silt and sand) range (Fig. 5). Whereas in case of cement stabilised samples apart from the broken clay sized sand and silt particles, some entrapped clay particles in the aggregated coarser soil lumps (sand and silt size lumps) will also get released and add up to the clay content of the ground sample. This indicates that the clay content of cement stabilised soil as indicated in the grain size curves (Fig. 6) can increase if the sample is ground further or allowed for natural weathering with time. Generally, four weeks of curing (28 days) is carried out for OPC based concrete and cement–soil products, where cement hydration is nearly complete. Continuation of curing beyond 28 days can result in additional strength gain due to the formation of additional cement hydration products. This may not have any influence on the clay mineral content of the soil–cement compact. The pace of lime–clay reactions is slow (at ambient curing conditions) and can continue even beyond 28 days of curing in the presence of hygroscopic moisture. Hence, 7% lime stabilised soil compacts were tested for clay fraction at the age of 60 days. The results indicate that the clay size fraction reduced from
3.5% (at 28 days) to 1.5% (60 days). This result clearly shows that the clay mineral content of the soil further reduces due to continued lime–clay reactions. 3.4. Atterberg's limits Liquid limit (LL) and plasticity index (PI) values of a soil are greatly influenced by the quantity and type of clay mineral. LL and PI values for the ground samples of stabilised as well as natural soil were measured. Fig. 9 shows the relationships between liquid limit and the stabiliser content (lime and cement). Fig. 10 shows the relationship between PI and stabiliser content. Stabilised soil mixes show LL and PI values and hence are not nonplastic. Atterberg's limit values of the natural soil reduce when it is stabilised with cement and lime. There is a drastic reduction (40%) in the LL and PI values for lime stabilised soil as the lime content goes from 0 to 10%, whereas in cement stabilised soil this reduction is marginal (10%). In lime stabilisation the clay fraction in the stabilised soil mix greatly reduces due to lime–clay reactions. This reduction in clay content affects the Atterberg's limits of the resultant stabilised soil. Marginal reduction in Atterberg's limits for cement stabilised soil mixes can be attributed to the presence of large percentage of clay minerals. In cement stabilisation a small percentage of lime is released during cement hydration and this lime can react with clay minerals.
30
7% Cement 7% Lime
Cement Lime
25
10 Plasticity Index
Clay size fraction (%)
12
8 6 4
20
15
10
2 0 6
8 10 12 14 Grinding duration (minutes) Fig. 8. Clay size fraction versus grinding duration.
16
5 0
2
4 6 Cement or Lime (%)
8
Fig. 10. Variation in plasticity index with stabiliser content.
10
368
B.V. Venkatarama Reddy, M.S. Latha / Applied Clay Science 101 (2014) 362–368
Marginal reduction in LL and PI values for cement stabilised soil again substantiates the fact that a considerable percentage of clay minerals present in the original soil are intact even after cement stabilisation. 4. Concluding remarks The status of clay minerals in the cement and lime stabilised soil compacts was examined. The XRD patterns, SEM images, grain size curves and Atterberg's limits of the natural and stabilised soils lead to some interesting observations. The results of the investigations indicate that cement stabilised soil products possess considerable amount of the natural clay minerals in nearly intact condition. In case of lime stabilised products the clay minerals get consumed and a very small percentage of the residual clay mineral is left in the sample. Calcium hydroxide used in the lime stabilisation process also gets consumed in the lime–clay reactions. Therefore, the quantity of residual free lime (calcium hydroxide) left in the stabilised soil product is small and hence only a small fraction of the lime added is available for the re-carbonation process. These investigations reveal that it is possible to retrieve a large percentage of natural clay minerals in cement stabilised soil products when subjected to either accelerated or natural weathering process. Whereas in lime-stabilised soil products only a small fraction of the clay minerals which were present in the soil can be retrieved, as the natural clay gets consumed in the lime–clay reactions. Thus cement stabilisation is superior from the point of view of retrieving back the precious clay minerals after the end of life of the building built with cement stabilised soil products such as stabilised soil bricks or rammed earth. References Bell, F.G., 1996. Lime stabilization of clay minerals and soils. Eng. Geol. 42, 223–237. Chasten, F.N., 1952. Soil–cement progress in Australia. Indian Concr. J. 26, 354–356. De, P.L., 1964. Soil–cement for the foundation of buildings. Indian Concr. J. 38, 1–2. Easton, David, 2007. The Rammed Earth House (Revised Ed). Chelsea Green Publishing Company, Vermont USA. Hall, Matthew, 2002. Rammed Earth: Traditional Methods, Modern Techniques, Sustainable Future. Building Engineer, pp. 22–24 (Nov.). Hall, Matthew, Djerbib, Youcef, 2004. Moisture ingress in rammed earth: part 1 — the effect of soil particle size distribution on the rate of capillary suction. Constr. Build. Mater. 18, 269–280. Heathcote, K., 1991. Compressive strength of cement stabilized pressed earth blocks. Build. Res. Inf. 19 (2), 101–105.
Houben, H., Guillaud, H., 2003. Earth Construction — A Comprehensive Guide. Intermediate Technology Publications, Belgium. Ingles, O.G., Metcalf, J.B., 1972. Soil Stabilisation Principles and Practice. Butterworth's publisher, Australia. Jayasinghe, C., Kamaladasa, N., 2007. Compressive strength characteristics of cement stabilised rammed earth walls. Constr. Build. Mater. 21, 1971–1976. Keable, Julian, 1996. Rammed Earth Structures, a Code of Practice. IT publication, U.K. Kumar Mehta, P., 1986. Concrete: Structure, Properties and Materials, 2nd ed. PrenticeHall Inc., New Jersey, USA. Lambe, T.W., 1962. In: Leonard's (Ed.), Soil Stabilization Chapter-4, Foundation Engineering. Mc. Graw Hill, pp. 351–437. Millogo, Younoussa, Morel, Jean-Claude, 2012. Microstructural characterisation and mechanical properties of cement stabilised adobes. Mater. Struct. 45, 1311–1318. Millogo, Younoussa, Hajjaji, Mohamed, Ouedraogo, Raguilnaba, 2008. Microstructure and physical properties of lime-clayey adobe bricks. Constr. Build. Mater. 22, 2386–2392. Olivier, M., Ali, M., 1987. Influence of different parameters on the resistance of earth, used as a building material. Proc. Int. Conf. on Mud Architecture, Trivandrum, India. Robertson, J.A., Blight, G.E., 1978. Stabilized earth fill dams. Proc. Symposium on Soil Reinforcing and Stabilizing Techniques in Engineering Practice. Sydney, Australia, pp. 571–590 (Oct.). Venkatarama Reddy, B.V., 2009. Sustainable materials for low carbon buildings. Int. J. Low Carbon Technol. 4 (3), 175–181. Venkatarama Reddy, B.V., Gupta, A., 2005. Characteristics of soil–cement blocks using highly sandy soils. Mater. Struct. 38 (280), 651–658. Venkatarama Reddy, B.V., Jagadish, K.S., 1993. The static compaction of soils. Geotechnique 43 (2), 337–341. Venkatarama Reddy, B.V., Jagadish, K.S., 2003. Embodied energy of common and alternative building technologies. Energy Build. 35, 129–137. Venkatarama Reddy, B.V., Prasanna Kumar, P., 2009. Compressive strength and elastic properties of stabilised rammed earth and masonry. Masonry Int. 22 (2), 39–46. Venkatarama Reddy, B.V., Prasanna Kumar, P., 2010. Embodied energy in cement stabilised rammed earth walls. Energy Build. 42 (3), 380–385. Venkatarama Reddy, B.V., Prasanna Kumar, P., 2011a. Structural behaviour of story high cement stabilised rammed earth walls under Compression. J. Mater. Civ. Eng. 23 (3), 240–247. Venkatarama Reddy, B.V., Prasanna Kumar, P., 2011b. Cement stabilised rammed earth — part A: compaction characteristics and physical properties of compacted cement stabilised soils. Mater. Struct. 44 (3), 681–694. Venkatarama Reddy, B.V., Prasanna Kumar, P., 2011c. Cement stabilised rammed earth — part B: compressive strength and elastic properties. Mater. Struct. 44 (3), 695–707. Venkatarama Reddy, B.V., Walker, P., 2005. Stabilised mud blocks: problems, prospects. Proc. Int. Earth Building Conf., Sydney Australia, pp. 63–75. Venkatarama Reddy, B.V., Lal, Richardson, Nanjunda Rao, K.S., 2007. Optimum soil grading for the soil–cement blocks. J. Mater. Civ. Eng. 19 (2), 139–148. Walker, P.J., 2004. Strength and erosion characteristics of earth blocks, and earth block masonry. J. Mater. Civ. Eng. 16 (5), 497–506. Walker, P., Stace, T., 1997. Properties of some cement stabilized compressed earth blocks and mortars. Mater. Struct. 30, 545–551. Walker, Peter, Keable, Rowland, Martin, Joe, Maniatidis, Vasilios, 2005. Rammed Earth: Design and Construction Guidelines. BRE Bookshop, U.K.
Contents lists available at ScienceDirect
Applied Clay Science journal homepage: www.elsevier.com/locate/clay
Research paper
Retrieving clay minerals from stabilised soil compacts B.V. Venkatarama Reddy ⁎, M.S. Latha 1 Department of Civil Engineering, Indian Institute of Science, Bangalore 560012, India
a r t i c l e
i n f o
Article history: Received 6 September 2012 Received in revised form 31 July 2014 Accepted 21 August 2014 Available online 22 September 2014 Keywords: Clay mineral Soil stabilisation Stabilised soil block Rammed earth Compressed earth block Adobe
a b s t r a c t Stabilised soil products such as stabilised soil blocks, rammed earth and stabilised adobe are being used for building construction since the last 6–7 decades. Major advantages of stabilised soil products include low embodied carbon, use of local materials, decentralized production, and easy to adjust the strength, texture, size and shape. Portland cement and lime represent the most commonly used stabilisers for stabilised soil products. The mechanism of strength development in cement and lime stabilised soils is distinctly different. The paper presents results of scientific investigations pertaining to the status of clay minerals in the 28 day cured cement and lime stabilised soil compacts. XRD, SEM imaging, grain size distribution and Atterberg's limits of the ground stabilised soil products and the natural soil were determined. Results reveal that clay minerals can be retrieved from cement stabilised soil products, whereas in lime stabilised soil products clay minerals get consumed in the lime–clay reactions and negligible percentage of clay minerals are left in the stabilised soil compacts. The results of the present investigation clearly demonstrate that cement stabilisation is superior to lime stabilisation in retrieving the clay minerals from the stabilised soil compacts. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Soil is formed due to natural weathering of rocks and it takes millions of years for soil formation. Clay minerals are the essential component of a soil in controlling its engineering characteristics and are essential for supporting the plant growth on soils. Soils are used for the manufacture of construction materials such as burnt clay bricks, which are consumed in bulk quantities. Also, it is used in the form of stabilised soil for the manufacture of blocks (stabilised soil blocks and stabilised adobe) and in-situ construction (rammed earth). Stabilised soils have been successfully used for the construction of road subbases (Chasten, 1952; Lambe, 1962; De, 1964; Ingles and Metcalf, 1972; Robertson and Blight, 1978). The use of stabilised soils for the construction of buildings can be noticed since thirties. Soils blended with stabilisers are generally compacted when used for either building construction or any other application like road sub-base, and retaining structures. Varieties of stabilisers have been explored for stabilised soil products. Inorganic stabilisers such as cement and lime are commonly employed for the production of stabilised soil products. Stabilised soil blocks (also known as stabilised compressed earth blocks), stabilised rammed earth and stabilised adobe represent some of the stabilised soil products being used for building construction since the last 6–7 ⁎ Corresponding author. Tel.: +91 80 2293 3126; fax: +91 80 23600404. E-mail addresses: [email protected] (B.V. Venkatarama Reddy), [email protected] (M.S. Latha). 1 Tel.: +91 80 2293 2327.
http://dx.doi.org/10.1016/j.clay.2014.08.027 0169-1317/© 2014 Elsevier B.V. All rights reserved.
decades. These products are considered energy efficient and have low embodied carbon when compared to the conventional materials like burnt clay bricks (Venkatarama Reddy and Jagadish, 2003; Venkatarama Reddy, 2009; Venkatarama Reddy and Prasanna Kumar, 2010). Generally, soils contain larger amount of clay fraction. Clay content of such soils is adjusted to an optimum level by diluting with sand and then used for stabilised soil block (SSB) production and stabilised rammed earth construction. Ordinary Portland cement (OPC) and lime are the commonly used stabilisers. Soil characteristics, density and stabiliser content play a crucial role in controlling the properties of SSB and stabilised rammed earth construction. Vast amount of knowledge has been accumulated on SSB technology and rammed earth (Olivier and Ali, 1987; Heathcote, 1991; Keable, 1996; Walker and Stace, 1997; Hall, 2002; Houben and Guillaud, 2003; Walker, 2004; Hall and Djerbib, 2004; Venkatarama Reddy and Gupta, 2005; Walker et al., 2005; Venkatarama Reddy et al., 2007; Jayasinghe and Kamaladasa, 2007; Easton, 2007; Venkatarama Reddy and Prasanna Kumar, 2009; Venkatarama Reddy and Prasanna Kumar, 2011a, 2011b, 2011c and many other publications). The mechanism of strength development in cement and lime stabilised soils is distinctly different. In a wetted soil–cement mixture, cement hydrates and cementitious products such as calcium-silicate hydrate (CSH) and calcium aluminate hydrate (CAH) are formed apart from the release of small percentage of calcium hydroxide (lime). These hydration products establish water insoluble bonds and bind the gravel, sand and silt particles. Cementitious products such as CSH
B.V. Venkatarama Reddy, M.S. Latha / Applied Clay Science 101 (2014) 362–368
2. Methodology and the experimental programme X-ray diffraction (XRD) analysis and specific surface estimation were carried out on the ground stabilised soil compacts. Details of XRD analysis and specific surface area estimation are as follows. The XRD patterns were generated from less than 2 μm size samples obtained from the ground samples. The sample passing 75 μm was mixed with distilled water in a beaker, then the mixture was poured into a jar and again agitated. The agitated mixture was allowed to settle down. The silt particles will settle first leaving a suspension containing floating fine clay particles and crystalline CSH, CAH, etc. The suspension with floating particulate matter was decanted and dried. XRD analysis was run on these dried samples which are basically either clay or crystallised CSH and CAH, and calcium hydroxide. The particle size of the aggregated CSH particles in hydrated cement paste will be in the range of 0.8 to 3 μm (Kumar Mehta, 1986). XRD patterns of powdered samples were collected on a fully automated Bruker D8 Advance diffractometer operating at 40 kV and 40 mA using CuKα radiation. Data was collected between 5° and 100° 2θ. The SEM images were obtained using Quanta LV/ESEM with the capacity for operation at high pressures as well as under environments such as water vapour, in addition it is
equipped with a standard secondary (Evehat–Thorley) and solid state scatter detector. The Brunauer, Emmett, and Teller (BET) gas adsorption theory is the foundation for the measurement of surface area in high specific surface materials. Measurements in this study were conducted using gas sorption analyser. Nitrogen was selected as the absorbate. The samples were outgassed at 100 °C. The experimental work involved the preparation of stabilised soil compacts, curing them for 28 days, crushing the cured samples and examining for the presence of original clay minerals in the crushed samples. A natural soil having kaolinite clay mineral was used in the investigations. The natural soil has a specific surface area of 5.40 m2/g and when it was ground in a mixer grinder for 12 min its specific surface increased to 8.1 m2/g. The mixer grinder consists of a metal blade set rotating at 1500 rpm in an enclosed plastic jar. Such grinders are used for grinding solid particles of less than 5 mm to fine powder form in dry condition. Grain size distribution curves for the natural soil and ground soil are shown in Fig. 1. Table 1 gives the details of the characteristics of both the samples. The reasons for grinding the soil sample are discussed in Section 2.2. The grain size curves for the natural soil and the ground soil (ground for 12 min) show the variations in the particle sizes among the two soils. When the soil was ground for 12 min (specific surface = 8.1 m2/g) the percentage of clay and silt size fractions increase whereas the sand size fraction decreases when compared with the grain size fractions of the natural soil. During the grinding process, some of the sand and silt size particles get crushed to clay and silt size, and resulting in higher clay and silt size fractions. The liquid limit and plasticity index values of the ground soil sample are 33.9% and 25.1 respectively. As the particles become much finer their water holding capacity increases as indicated by the increase in the value of liquid limit. Generally, the stabiliser content used in the production of stabilised soil blocks, rammed earth, stabilised adobe, etc. will be in the range of 5–10% by weight. Therefore, lime and cement contents were varied between 4 and 10% representing the wide range of stabiliser percentages. 2.1. Preparation and testing of compacted stabilised soil specimens Ordinary Portland cement and laboratory grade lime (with 95% assay) were used for the preparation of compacted cylindrical specimens. The procedure adopted to prepare the cylindrical specimen of size: 38 mm diameter and 76 mm height is as follows. (a) The soil was oven dried at 60 °C to constant weight and then blended with cement or lime by mixing in a ball mill for 10 min. Ball milling is to ensure uniform mixture. (b) Requisite quantity of potable water was sprayed on to the dry mixture of soil and stabiliser, and then a uniform mixture was
100 80
Percent finer
and CAH do not react with clay particles, whereas the small percentage of lime released in the cement hydration process can react with clay minerals forming additional cementitious products. The strength development in cement stabilised soil compacts (CSSC) is mainly attributed to the formation of cementitious products such as CSH and CAH. The clay minerals are pozzolanic in nature and can readily react with lime in the presence of moisture. Therefore, the lime–clay reactions lead to the formation of water insoluble cementitious gel of silicate and silicate–aluminates and this gel finally crystallises into hydrates of calcium silicate, calcium aluminates, etc. with time. The cementitious products formed in lime–clay reactions are similar to those formed during hydration of OPC. The cementitious gel coats the soil particles and establishes bonds between them. The pace of lime–clay reactions is very slow at ambient temperature and curing conditions. There are arguments favouring lime stabilisation of soils instead of cement stabilisation with the assumption that there will be recarbonation of lime in the lime stabilised soil leading to carbon dioxide mitigation. A comparison between the mechanisms of strength development in cement and lime stabilised soils reveals that both the clay minerals and calcium hydroxide get consumed in the lime stabilised soils, whereas clay minerals are nearly intact in the cement stabilised soils. The present investigation attempts to prove this point. Stabilised soil products as well as burnt clay bricks use soil as the main ingredient. The manufacture of burnt clay brick results in irreversible structural changes to clay minerals and permanently changes the soil into a nearly rock form. It is not possible to get back the natural clay minerals from the burnt clay bricks, unless allowed to undergo weathering for millions of years. Whereas there is a possibility of retrieving the clay minerals from some of the stabilised soil products after the end of life of such products without much environmental costs. In the context of sustainability and green buildings there is an emphasis on developing low embodied carbon materials effecting minimum changes to the natural raw materials. In the framework of life-cycle and sustainability analysis of buildings attention has to be paid to the end of life use of a building material. At the end of life if a building material can go back to its native state with minimum environmental costs, then it can be classified under the category of green and sustainable building materials. Considering the Portland cement and lime as soil stabilisers the present investigation attempts to examine the status of clay minerals in the stabilised soil compacts. This is mainly to address the question of recyclability of soil based materials as soil at the end of life of such construction materials. The results of the experiments carried out on the 28 day cured cement and lime stabilised compacted specimens are discussed in the present investigation.
363
Natural soil Ground soil
60 40 20 0 0.001
0.01
0.1
1
Particle size (mm) Fig. 1. Grain size distribution curves of natural soil and ground soil.
10
364
B.V. Venkatarama Reddy, M.S. Latha / Applied Clay Science 101 (2014) 362–368
3. Results and discussion
Table 1 Characteristics of soils. Properties Textural composition, (% by weight) Sand (4.75–0.075 mm) Silt (0.075–0.002 mm) Clay (b0.002 mm) Atterberg's limits Liquid limit, % Plasticity index Unified soil classification Clay mineral pH
Natural soil
Ground soil
72.6 11.6 15.8
44.0 36.4 19.6
26.9 17.5 SC Kaolinite 8.06
33.9 25.1 CL Kaolinite 7.84
obtained by mixing manually. Moulding water content was maintained at 12.5% by weight of the mix. It should be noted here that the cylindrical specimens were prepared employing static compaction process and hence the use of proctor OMC as moulding water content is not valid. More information on static compaction of soils can be found in the studies of Venkatarama Reddy and Jagadish (1993). (c) The wetted mix was fed (known weight) into an open ended cylindrical mould and compaction was carried out from both the ends using a mechanical screw jack set-up. The specimen was extruded from the mould immediately after the compaction. Compacted specimens were kept for curing under wet burlap after 24 h of casting. The weight of the mix fed into the mould was controlled to achieve a dry density of 1800 kg/m3 for the specimen. The strength of compacted stabilised soil specimens increases with the increase in dry density (Venkatarama Reddy and Walker, 2005; Venkatarama Reddy and Prasanna Kumar, 2011b, 2011c) and hence it is kept constant. The compacted stabilised soil specimens were cured under wet burlap for 28 days and then air-dried for 14 days. The air dried specimens were oven dried at 50 °C to constant weight and then tested for compressive strength. Wet compressive strength was determined by testing the saturated specimens (soaked in water for 48 h prior to testing) and dry compressive strength was determined by testing them in oven dry state. The cylindrical specimens were tested in a loading frame at constant piston displacement of 1.25 mm/min.
3.1. Strength of stabilised soil compacts Wet and dry compressive strength of the stabilised soil compacts was determined. Fig. 2 shows a plot of strength versus stabiliser (lime or cement) content. The results shown in the figure represent the mean of six specimens. The figure shows that the strength increases with an increase in the stabiliser content for both the wet and dry cases irrespective of the stabiliser type. The strength (both wet and dry cases) increases by about three times as the cement content increased from 4 to 10%, whereas in case of lime stabilised compacts the strength increase is about 1.5 times. There is a considerable difference in the wet and dry strengths of the stabilised soil compacts. Wet strength to dry strength ratio is in the range of 0.4–0.5 for cement stabilised soil compacts and 0.50–0.55 for lime stabilised soil compacts. More information on the strength of stabilised soil blocks and rammed earth can be found in the literature cited in Section 1. 3.2. XRD patterns, SEM examination and EDS analysis Fig. 3 shows the XRD patterns for both cement and lime stabilised samples and also for the natural soil. The XRD pattern of natural soil shows the presence of only kaolinite clay mineral as the suspended fine particles were only clay minerals. XRD patterns for cement and lime stabilised samples show the presence of kaolinite as well as the cementitious products like CSH and CAH. Also, calcium hydroxide (CH) peaks can be seen in both the lime and cement stabilised samples. CH peaks in cement stabilised samples arise from the lime released during cement hydration process. In case of lime stabilised samples CH peaks indicate the un-reacted lime. There are very few peaks showing kaolinite mineral in the XRD of 7% and 10% lime samples (Fig. 3). The quartz peaks are missing in the XRD patterns shown in Fig. 3. This is attributed to quick settling of heavier silt particles and the decanted materials from the suspension in the jar are finer particles of clay and cement hydration products (CSH, CAH etc.). CAH and CSH products are formed during cement hydration process in the cement stabilised samples. Lime stabilised soil samples also show the formation of CSH and CAH products due to the lime–clay reactions. SEM image shown in Fig. 4(I) for the natural soil does not show the presence of cementitious products (such as CSH and CAH). SEM images and EDS analysis (Fig. 4 II to VI) for lime and cement stabilised soil
2.2. Status of clay minerals in the stabilised soil compacts
1. Grain size distribution 2. Atterberg's limits 3. X-ray diffraction (XRD) and 4. SEM imaging and EDS analysis.
Compressive strength (MPa)
Stabilised soil compacts are cemented materials and when exposed to normal weathering process will take a long time (may be several decades) to disintegrate into smaller particles. In order to examine the stabilised soil compacts for the presence of clay minerals the samples have to be crushed to a powder form. Therefore, stabilised soil compacts after the compression test were oven dried at 50 °C and then crushed to a powder form. The crushing/grinding of the samples was carried out in a mixer grinder for 12 min. The specific surface area of the ground samples is 7.04 m2/g and it increases to 12.42 m2/g for the fraction less than 75 μm size. The powdered samples were examined for the presence of clay minerals. The natural soil was also ground for 12 min and the properties of the ground natural soil were compared with those of powdered stabilised soil compacts. The following tests were performed on these powdered samples to ascertain the nature of products formed, presence of clay minerals and the percentage of clay fraction in the stabilised soil compacts.
6 Dry (Cement) Wet (Cement) Dry (Lime) Wet (Lime)
5
4
3
2
1
0 4
5
6
7
8
Stabiliser content (%) Fig. 2. Strength versus stabiliser content.
9
10
B.V. Venkatarama Reddy, M.S. Latha / Applied Clay Science 101 (2014) 362–368
365
Fig. 3. XRD patterns for natural and stabilised soils on fractions less than 2 μm.
Fig. 4. SEM images and EDS analysis of the natural and stabilised soil samples. (I) Natural soil, (II) 7% cement, (III) 10% cement, (IV) 4% lime, (V) 7% lime and (VI) 10% lime; (A: sand/silt particles; B: kaolinite; C: CSH/CAH; D: calcium hydroxide).
B.V. Venkatarama Reddy, M.S. Latha / Applied Clay Science 101 (2014) 362–368
samples indicate the formation of cementitious products (CSH, CAH etc.). This is in tune with the observations of Millogo and Morel (2012) on the SEM and EDS analysis of cement stabilised adobe brick samples, where they detected the presence of cementitious compounds as well as kaolinite clay mineral. Also, Bell (1996) and Millogo et al. (2008) report the formation of CSH and CAH due to the reaction between kaolinite and lime. It is not possible to quantify the clay minerals from the XRD patterns and SEM images. Therefore, grain size analysis of the samples was undertaken on the ground samples. The presence of kaolinite clay mineral in the silt and sand size fractions of the stabilised soil samples was examined through XRD analysis. Fig. 5 shows the typical XRD patterns for silt and sand size fractions in cement (7%) and lime (7%) stabilised soil samples as well as natural soil sample. Kaolinite is absent in the coarser particles (silt and sand size fractions) of natural soil samples, whereas the presence of kaolinite can be seen in the sand and silt size fractions of cement stabilised soil samples. Kaolinite is absent in the silt and sand size fractions of lime stabilised soil samples. It is interesting to note that there is clay mineral in the sand and silt size fractions of cement stabilised soil. Here, the clay minerals are entrapped in the interstices of aggregated coarser soil lumps (i.e. silt and sand size particles) and if such coarser soil particles are allowed for further grinding or natural weathering the clay minerals entrapped in their interstices can get released.
3.3. Grain size and clay fraction Grain size analysis of the ground samples was carried out mainly to assess the clay size fraction in the samples. The grain size distribution curves for the powdered stabilised soil compacts as well as natural soil sample are displayed in Fig. 6 for both the cement stabilised and lime stabilised samples. Fig. 7 shows the plot of clay size fraction versus stabiliser content. The following observations can be made from the results shown in these figures.
Percent finer
366
100 90 80 70 60 50 40 30 20 10 0 0.001
Natural soil 4% cement 7% cement 10% cement 4% lime 7% lime 10% lime
0.01
0.1
1
10
Particle size (mm) Fig. 6. Grain size distribution curves.
The percentage of sand, silt and clay size fractions reduce as the stabiliser content increases. The ground natural soil (specific surface 8.1 m2/g) has a clay size fraction of 19.6%. This value steadily reduces as the stabiliser (cement or lime) content increases (Fig. 7). Cement stabilised samples show a clay size fraction of 10.6%, 8.8% and 6.3% for 4%, 7% and 10% cement contents respectively. The corresponding values for lime stabilised samples are 3.5%, 2.8% and 1.7%. There is a considerable difference in the clay size fractions among cement and lime stabilised samples. Lower percentage of clay size fraction in lime stabilised samples can be attributed to the consumption of clay minerals in the lime–clay reactions. Lime–clay reactions are responsible for the formation of cementitious materials in facilitating the strength development in the lime stabilised compacts. In lime stabilised soils both the clay minerals and the lime get transformed into new products such as CSH and CAH. Therefore, there is not much free lime left for re-carbonation. The samples were ground in a mixer grinder to get a specific surface of 7.04 m2/g. Clay size fraction of the ground samples contains a mixture
Fig. 5. XRD patterns for cement and lime stabilised soils on silt and sand size fractions.
B.V. Venkatarama Reddy, M.S. Latha / Applied Clay Science 101 (2014) 362–368
367
35
Liquid Limit (%)
Cement Lime 30
25
20
15 0
2
4 6 8 Cement or Lime (%)
10
Fig. 9. Variation in liquid limit with stabiliser content. Fig. 7. Clay size fraction versus stabiliser content.
of clay particles as well as ground sand and silt particles. Clay size fraction increases with the grinding duration as illustrated in Fig. 8 for the two types of samples (i.e. 7% cement and 7% lime). In case of lime stabilised sample the increase in clay size fraction with the grinding duration is mainly attributed to the increased percentage of broken clay sized, sand and silt particles, as the clay mineral is absent in the coarser particles (silt and sand) range (Fig. 5). Whereas in case of cement stabilised samples apart from the broken clay sized sand and silt particles, some entrapped clay particles in the aggregated coarser soil lumps (sand and silt size lumps) will also get released and add up to the clay content of the ground sample. This indicates that the clay content of cement stabilised soil as indicated in the grain size curves (Fig. 6) can increase if the sample is ground further or allowed for natural weathering with time. Generally, four weeks of curing (28 days) is carried out for OPC based concrete and cement–soil products, where cement hydration is nearly complete. Continuation of curing beyond 28 days can result in additional strength gain due to the formation of additional cement hydration products. This may not have any influence on the clay mineral content of the soil–cement compact. The pace of lime–clay reactions is slow (at ambient curing conditions) and can continue even beyond 28 days of curing in the presence of hygroscopic moisture. Hence, 7% lime stabilised soil compacts were tested for clay fraction at the age of 60 days. The results indicate that the clay size fraction reduced from
3.5% (at 28 days) to 1.5% (60 days). This result clearly shows that the clay mineral content of the soil further reduces due to continued lime–clay reactions. 3.4. Atterberg's limits Liquid limit (LL) and plasticity index (PI) values of a soil are greatly influenced by the quantity and type of clay mineral. LL and PI values for the ground samples of stabilised as well as natural soil were measured. Fig. 9 shows the relationships between liquid limit and the stabiliser content (lime and cement). Fig. 10 shows the relationship between PI and stabiliser content. Stabilised soil mixes show LL and PI values and hence are not nonplastic. Atterberg's limit values of the natural soil reduce when it is stabilised with cement and lime. There is a drastic reduction (40%) in the LL and PI values for lime stabilised soil as the lime content goes from 0 to 10%, whereas in cement stabilised soil this reduction is marginal (10%). In lime stabilisation the clay fraction in the stabilised soil mix greatly reduces due to lime–clay reactions. This reduction in clay content affects the Atterberg's limits of the resultant stabilised soil. Marginal reduction in Atterberg's limits for cement stabilised soil mixes can be attributed to the presence of large percentage of clay minerals. In cement stabilisation a small percentage of lime is released during cement hydration and this lime can react with clay minerals.
30
7% Cement 7% Lime
Cement Lime
25
10 Plasticity Index
Clay size fraction (%)
12
8 6 4
20
15
10
2 0 6
8 10 12 14 Grinding duration (minutes) Fig. 8. Clay size fraction versus grinding duration.
16
5 0
2
4 6 Cement or Lime (%)
8
Fig. 10. Variation in plasticity index with stabiliser content.
10
368
B.V. Venkatarama Reddy, M.S. Latha / Applied Clay Science 101 (2014) 362–368
Marginal reduction in LL and PI values for cement stabilised soil again substantiates the fact that a considerable percentage of clay minerals present in the original soil are intact even after cement stabilisation. 4. Concluding remarks The status of clay minerals in the cement and lime stabilised soil compacts was examined. The XRD patterns, SEM images, grain size curves and Atterberg's limits of the natural and stabilised soils lead to some interesting observations. The results of the investigations indicate that cement stabilised soil products possess considerable amount of the natural clay minerals in nearly intact condition. In case of lime stabilised products the clay minerals get consumed and a very small percentage of the residual clay mineral is left in the sample. Calcium hydroxide used in the lime stabilisation process also gets consumed in the lime–clay reactions. Therefore, the quantity of residual free lime (calcium hydroxide) left in the stabilised soil product is small and hence only a small fraction of the lime added is available for the re-carbonation process. These investigations reveal that it is possible to retrieve a large percentage of natural clay minerals in cement stabilised soil products when subjected to either accelerated or natural weathering process. Whereas in lime-stabilised soil products only a small fraction of the clay minerals which were present in the soil can be retrieved, as the natural clay gets consumed in the lime–clay reactions. Thus cement stabilisation is superior from the point of view of retrieving back the precious clay minerals after the end of life of the building built with cement stabilised soil products such as stabilised soil bricks or rammed earth. References Bell, F.G., 1996. Lime stabilization of clay minerals and soils. Eng. Geol. 42, 223–237. Chasten, F.N., 1952. Soil–cement progress in Australia. Indian Concr. J. 26, 354–356. De, P.L., 1964. Soil–cement for the foundation of buildings. Indian Concr. J. 38, 1–2. Easton, David, 2007. The Rammed Earth House (Revised Ed). Chelsea Green Publishing Company, Vermont USA. Hall, Matthew, 2002. Rammed Earth: Traditional Methods, Modern Techniques, Sustainable Future. Building Engineer, pp. 22–24 (Nov.). Hall, Matthew, Djerbib, Youcef, 2004. Moisture ingress in rammed earth: part 1 — the effect of soil particle size distribution on the rate of capillary suction. Constr. Build. Mater. 18, 269–280. Heathcote, K., 1991. Compressive strength of cement stabilized pressed earth blocks. Build. Res. Inf. 19 (2), 101–105.
Houben, H., Guillaud, H., 2003. Earth Construction — A Comprehensive Guide. Intermediate Technology Publications, Belgium. Ingles, O.G., Metcalf, J.B., 1972. Soil Stabilisation Principles and Practice. Butterworth's publisher, Australia. Jayasinghe, C., Kamaladasa, N., 2007. Compressive strength characteristics of cement stabilised rammed earth walls. Constr. Build. Mater. 21, 1971–1976. Keable, Julian, 1996. Rammed Earth Structures, a Code of Practice. IT publication, U.K. Kumar Mehta, P., 1986. Concrete: Structure, Properties and Materials, 2nd ed. PrenticeHall Inc., New Jersey, USA. Lambe, T.W., 1962. In: Leonard's (Ed.), Soil Stabilization Chapter-4, Foundation Engineering. Mc. Graw Hill, pp. 351–437. Millogo, Younoussa, Morel, Jean-Claude, 2012. Microstructural characterisation and mechanical properties of cement stabilised adobes. Mater. Struct. 45, 1311–1318. Millogo, Younoussa, Hajjaji, Mohamed, Ouedraogo, Raguilnaba, 2008. Microstructure and physical properties of lime-clayey adobe bricks. Constr. Build. Mater. 22, 2386–2392. Olivier, M., Ali, M., 1987. Influence of different parameters on the resistance of earth, used as a building material. Proc. Int. Conf. on Mud Architecture, Trivandrum, India. Robertson, J.A., Blight, G.E., 1978. Stabilized earth fill dams. Proc. Symposium on Soil Reinforcing and Stabilizing Techniques in Engineering Practice. Sydney, Australia, pp. 571–590 (Oct.). Venkatarama Reddy, B.V., 2009. Sustainable materials for low carbon buildings. Int. J. Low Carbon Technol. 4 (3), 175–181. Venkatarama Reddy, B.V., Gupta, A., 2005. Characteristics of soil–cement blocks using highly sandy soils. Mater. Struct. 38 (280), 651–658. Venkatarama Reddy, B.V., Jagadish, K.S., 1993. The static compaction of soils. Geotechnique 43 (2), 337–341. Venkatarama Reddy, B.V., Jagadish, K.S., 2003. Embodied energy of common and alternative building technologies. Energy Build. 35, 129–137. Venkatarama Reddy, B.V., Prasanna Kumar, P., 2009. Compressive strength and elastic properties of stabilised rammed earth and masonry. Masonry Int. 22 (2), 39–46. Venkatarama Reddy, B.V., Prasanna Kumar, P., 2010. Embodied energy in cement stabilised rammed earth walls. Energy Build. 42 (3), 380–385. Venkatarama Reddy, B.V., Prasanna Kumar, P., 2011a. Structural behaviour of story high cement stabilised rammed earth walls under Compression. J. Mater. Civ. Eng. 23 (3), 240–247. Venkatarama Reddy, B.V., Prasanna Kumar, P., 2011b. Cement stabilised rammed earth — part A: compaction characteristics and physical properties of compacted cement stabilised soils. Mater. Struct. 44 (3), 681–694. Venkatarama Reddy, B.V., Prasanna Kumar, P., 2011c. Cement stabilised rammed earth — part B: compressive strength and elastic properties. Mater. Struct. 44 (3), 695–707. Venkatarama Reddy, B.V., Walker, P., 2005. Stabilised mud blocks: problems, prospects. Proc. Int. Earth Building Conf., Sydney Australia, pp. 63–75. Venkatarama Reddy, B.V., Lal, Richardson, Nanjunda Rao, K.S., 2007. Optimum soil grading for the soil–cement blocks. J. Mater. Civ. Eng. 19 (2), 139–148. Walker, P.J., 2004. Strength and erosion characteristics of earth blocks, and earth block masonry. J. Mater. Civ. Eng. 16 (5), 497–506. Walker, P., Stace, T., 1997. Properties of some cement stabilized compressed earth blocks and mortars. Mater. Struct. 30, 545–551. Walker, Peter, Keable, Rowland, Martin, Joe, Maniatidis, Vasilios, 2005. Rammed Earth: Design and Construction Guidelines. BRE Bookshop, U.K.