- Email: [email protected]
Durability against wetting–drying cycles of sustainable Lightweight Cellular Cemented construction material comprising clay and fly ash wastes
Construction and Building Materials 77 (2015) 41–49
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Durability against wetting–drying cycles of sustainable Lightweight Cellular Cemented construction material comprising clay and fly ash wastes Anek Neramitkornburi a, Suksun Horpibulsuk b,c,⇑, Shui Long Shen d, Avirut Chinkulkijniwat a, Arul Arulrajah e, Mahdi Miri Disfani e a
School of Civil Engineering, Suranaree University of Technology, Nakhon Ratchasima 30000, Thailand School of Civil Engineering, Center of Excellence in Civil Engineering, Suranaree University of Technology, Nakhon Ratchasima 30000, Thailand c Centre for Sustainable Infrastructure, Swinburne University of Technology, Australia d Department of Civil Engineering and State Key Laboratory of Ocean Engineering, Shanghai Jiao Tong University, 800 Dongchuan Rd, Shanghai 200240, China e Faculty of Science, Engineering and Technology, Swinburne University of Technology, Melbourne, Australia b
h i g h l i g h t s Waste materials: fly ash and in-situ clay. A green Lightweight Cellular Cemented (LCC) clay. Role of cement, air and FA content on durability. Predictive wetting–drying (w–d) cycled strength equation.
a r t i c l e
i n f o
Article history: Received 5 June 2014 Received in revised form 29 October 2014 Accepted 14 December 2014
Keywords: Air foam Cement Durability Compressive strength Fly ash Wet–dry cycle Lightweight material
a b s t r a c t The viability of using waste materials such as clay and fly ash (FA) for developing a sustainable Lightweight Cellular Cemented (LCC) construction material is investigated in this paper. LCC clay has a wide range of applications in infrastructure rehabilitation as well as in the construction of new facilities. The durability against wetting–drying (w–d) cycles is an important parameter for service life design of LCC clay; however, studies on this aspect to date are very limited. The role of cemented soil structure (fabric and cementation bond) on w–d cycle strength of LCC clay are investigated, analyzed and presented in this paper. The strength reduction with increasing number of w–d is attributed to degradation of the cemented structure. The degradation index, qualifying the rate of degradation with number of w–d cycles, is proposed in term of initial soaked strength (without w–d cycle). Using the degradation index, the predictive w–d cycle strength equation at different number of w–d cycles is furthermore proposed. The applicability of the proposed equation is validated using a separate test data. This approach of predicting w–d cycle strength is beneficial from both engineering and economic points of view. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction There is a myriad of problems associated with the engineering construction in soft clay deposits, particularly in coastal regions in ⇑ Corresponding author at: School of Civil Engineering, Suranaree University of Technology, 111 University Avenue, Muang District, Nakhon Ratchasima 30000, Thailand. Tel.: +66 44 22 4322; fax: +66 44 22 4607. E-mail addresses: [email protected] (A. Neramitkornburi), [email protected]. th, [email protected] (S. Horpibulsuk), [email protected] (S.L. Shen), [email protected]. th (A. Chinkulkijniwat), [email protected] (A. Arulrajah), mmiridisfani@swin. edu.au (M.M. Disfani). http://dx.doi.org/10.1016/j.conbuildmat.2014.12.025 0950-0618/Ó 2014 Elsevier Ltd. All rights reserved.
Southeast Asia such as Chao Phraya Plain in Thailand, Mekong Delta in Vietnam and Cambodia, Central Plains of the Philippines, Coastal Plains of Malaysia, Indonesia, Singapore, Hong Kong, Korea, Japan and Taiwan. This soft soil, located in marine or estuary environments, have low shear strength, low bearing capacity and high natural water content, resulting in high compressibility potential. To mitigate future issues with construction on these soft soil deposits, the deep mixing technique is frequently applied [1–8]. The mechanical behavior of cement admixed clays have been extensively investigated by authors [9–15]. The role of physical properties of soil on the strength development is recently investigated by Goodary et al.
42
A. Neramitkornburi et al. / Construction and Building Materials 77 (2015) 41–49
[16]. They reported that soils with a high specific surface (fine particles) needed less amount of cement to provide the same strength and durability as those with low specific surface (coarse particles). Instead of improving the soft ground (foundation) through the costly deep mixing method, the usage of Lightweight Cellular Cemented (LCC) construction materials is an attractive and economical alternative in construction applications such as in embankment, pavement pipe bedding and backfilling. LCC material is a mixture of aggregate, air foam agent and cementing agent. The usage of recycled waste materials such as Construction and Demolition (C&D) materials incorporating recycled concrete aggregate, crushed brick, and reclaimed asphalt pavement [17–19] has been applied in recent years in a wide range of applications such as embankment fills, pipe-bedding and pavement base/subbase. The usage of recycled waste materials in sustainable manner in the development of LCC materials will further support zero-waste directives currently implemented in many developed and developing countries. Incorporation of waste materials in the development of LCC material will reduce the carbon footprint of our future infrastructures. A sustainable LCC material made from in-situ and waste clay obtained directly from construction sites has been extensively used for highway and port constructions in Southeast Asian countries such as Japan and Thailand [20–26]. To reduce the cost of the LCC clay from an economical and environmental perspective, the replacement of cement by fly ash (FA) is an attractive method. It was evident from the flowability test (undertaken using a flow cone with 117 mm in height and 254 mm diameter at base and 117 mm diameter at top) that FA reduces the plasticity and improves the flowability of the LCC clay mixture before hardening [27,28]. At the same water content and cement content, the LCC clay with higher FA replacement ratio exhibits higher strength than that with lower FA replacement ratio. The use of FA as a supplementary cementitious material in concrete is also well recognized for improving durability of concrete [29,30]. The incorporation of FA results in considerable pore refinement [31], which leads to a low porosity and discontinuous pore structure. Consequently, the permeability of the concrete reduces and the durability of the concrete increases [32]. The in-situ LCC clay as a stabilized engineering fills and pavement materials generally encounters with w–d cycles from the change of weather during wet (rainy) and dry (summer) seasons. This is particular relevant for tropical countries such as Thailand, as well as parts of Australia and China. The w–d cycles result in tension and surface cracks, which can damage the stabilized pavement structure [33–36]. Even though there is available research on the strength development in LCC clay, the investigation of durability against the wetting–drying cycles (w–d cycles), a critical aspect for infrastructure design such as in engineering fills and pavements, is very limited and is the prime focus of this research. The investigation of the service life of the LCC clay via wetting and drying test is significant and is another focus of this research. The strength of LCC clay is dependent upon the cemented soil structure (fabric and cementation bond) [37]. As such, tests with a wide range of water contents, air contents, FA contents and cement contents of the LCC clay are undertaken to understand the role of both fabric and cementation bond on the w–d cycle strengths. Based on the analysis of the test results, a rational empirical relationship between w–d cycle strengths and initial soaked strength (without w–d cycles) is proposed. This equation can facilitate the determination of a suitable mix proportion of LCC materials to meet the strength requirement at a target service life. This research will enable waste excavated soft clay traditionally destined for landfill to be used in a sustainable manner as an aggregate in LCC materials, which is significant in term of engineering, economical and environmental perspectives.
2. Theoretical background For a LCC clay at a water content between 1.5 and 3.0 times the liquid limit, the strength is determined exclusively by the watervoid to cement, wV/C [38]. This parameter is defined as the product of initial clay water content (before mixing with cement and air foam) times V/C, where the water content is expressed in fraction. The parameter V/C is defined as the ratio of volume of voids to the volume of cement in the mix. Strength is independent of water content, air content and cement content in the mix. Based on extensive test results, Horpibulsuk et al. [38] have proposed a predictive strength equation in term of curing time, and wV=C for the LCC Bangkok clay as follows:
(
qðwV=C ÞD
)
qðwV=C Þ28
¼
1:27 ðwV=C Þ28 ð0:027 þ 0:300 ln DÞ ðwV=C ÞD
ð1Þ
where qðwV=C ÞD is the strength of LCC clay to be estimated at watervoid/cement ratio of (wV=C) after D days of curing and qðwV=C Þ28 is the strength of LCC clay at water-void/cement ratio of (wV=C) after 28 days of curing. The unit weight (in kN/m3) is determined in term of water content, cement content and V/C by using Eq. (2) [27,39]:
0 1 Gc Gs c2w ð1 þ wÞ þ Gc cw C BGs c ð1 þ wÞC c¼ ðV=C Þ@ w A Gc cw Gc cw þ1 þ1 C C
ð2Þ
where w is water content (in fraction), Gc and Gs are the specific gravities of cement and soil, respectively, cw is unit weight of water (kN/m3) and C is cement content (kg/m3). Eq. (2) was developed based on the assumption that all air bubbles (air foam) enter into the pore space when mixed with cement and clay. With the variation in water content and cement content, the air content required to attain the required V=C is determined:
Ac ¼ ðV=C Þ
C ð1 þ wGs Þ wGs Gc cw
ð3Þ
3. Materials and methods 3.1. Materials 3.1.1. Soil sample Bangkok clay was collected from Bangkok Noi district, Bangkok, Thailand at a 3 m depth. The clay was composed of 2% sand, 39% silt and 55% clay as shown in Fig. 1. The natural water content was 80% and the specific gravity was 2.64. The liquid and plastic limits were 73% and 31%, respectively. Based on the Unified Soil Classification System (USCS), the clay was classified as inorganic clay of high plasticity (CH). Groundwater was encountered at a depth of approximately 1 m below the surface. The clay was classified as low swelling type with free swell ratio (FSR) of 1.1. The FSR is defined as the ratio of equilibrium sediment volume of 10 g of oven-dried soil passing a 425 mm sieve in distilled water (Vd) to that in kerosene (Vk) [40]. This method was adopted since it is simple and predicts the dominant clay mineralogy of soil satisfactorily [41]. Table 1 summarizes the chemical composition of Bangkok clay using X-ray fluorescence (XRF). Even though Bangkok clay is classified as low swelling types, moderately high clay content of up to 55% may cause swelling and shrinkage on the LCC clay during w–d cycles. This effect will be examined in this paper.
3.1.2. Cement and air foam agent Type I Portland cement (PC) and air foam agent, Darex AE4, provided by the Grace Construction Products Ltd., were used in this study. The grain size distribution curve, obtained from the laser particle size analysis, and chemical composition of PC are also shown in Fig. 1 and Table 1, respectively. The specific gravity is 3.15 and the D50 is 0.01 mm (10 micron), which is larger than that of the tested clay. The air foam agent is a blend of anionic surfactants and foam stabilizers. It is a liquid air entraining agent used in various types of mortar, concrete and cementitious material.
A. Neramitkornburi et al. / Construction and Building Materials 77 (2015) 41–49
43
mixed with cement for 10 min. The cement content (C) was varied from 10% to 40% by weight of dry soil. The uniform paste was next transferred to cylindrical containers of 50 mm diameter and 100 mm height for w–d strength test. After 24 h, the cylindrical samples were dismantled. The cylindrical samples were wrapped in vinyl bags and stored in a humidity-controlled room of constant temperature (23 ± 2 °C) until 28 days of curing. The method of cyclic wetting and drying test as per ASTM D 559 was adopted for sample preparations. The samples were submerged in deionized water at room temperature for 5 h. They were then dried in the oven at a temperature of 70 °C for 48 h and air-dried at room temperature for at least 3 h. This process is referred to as 1 w–d cycle. After attaining the target w–d cycles, the samples were immersed in deionized water for 2 h at the constant temperature of 25 ± 2 °C. Unconfined Compression (UC) tests were then undertaken with a rate of vertical displacement of 1 mm/min. The 1, 3 and 6 w–d cycles were considered in this study. Based on a critical analysis of the strength data, a rational predictive w–d cycle strength equation is proposed, which facilitates the mix design to attain the strength requirement at a specified service life for civil engineering practitioners. In addition to the above mentioned laboratory tests, the results of the strength tests on separate LCC samples at FA replacement ratios of 40% (w = 198% and 132%, and A = 0%, 25% and 50%) were taken to verify the proposed predictive equation.
4. Results
Fig. 1. Grain size distribution of clay, PC and FA.
Table 1 Chemical composition of Bangkok clay, PC and FA. Chemical composition (%)
Bangkok clay
PC
FA
SiO2 Al2O3 Fe2O3 CaO MgO SO3 Na2O K2O LOI
62.8 21.3 8.4 0.9 1.5 1.2 0.3 2.5 0.8
20.9 4.8 3.4 65.4 1.2 2.7 0.2 0.3 1.1
44.7 23.7 11.0 12.7 2.6 1.3 0.1 2.9 1.0
3.1.3. Fly ash Fly ash (FA) was obtained from the Mae Moh power plant in the north of Thailand. Table 1 summarizes the chemical composition of FA showing that total amount of the major components SiO2, Al2O3 and Fe2O3 in FA are 79.4% and classified as class F according to ASTM C 618. The grain size distribution curve of FA is also shown in Fig. 1.
3.2. Methodology
Fig. 2 shows the typical role of FA on w–d cycle strengths of LCC clay at different number of w–d cycles. All the samples were prepared at different water contents but at the same w/wL of 2.0, C = 10% and A = 25% to have the same flowability of the LCC mixture, where w is water content and wL is liquid limit. It has been proved that at the same w/wL, the LCC mixtures have the same flowability [27]. Fig. 2 shows that FA improves w–d cycle strength, qu(w–d), where significant improvement is clearly observed when FA replacement ratio is greater than 40%. Fig. 3 shows the role of cement content and air content on w–d cycle strength for FA replacement of 20% and 80%. The unit weight decreases with increasing air content and decreasing cement content. Although the air foam helps reduce the unit weight of the LCC clay, it causes strength reduction at a specific cement content. It is evident that the initial soaked (without w–d cycle) strengths (qu0) and w–d cycle strengths (qu(w–d)) decrease with decreasing cement content and increasing air content. At a particular air content, the strength increases significantly with increasing cement content even with a slight increase in unit weight. The unit weight of LCC clay can be approximated from Eqs. (1) and (2) as previously undertaken by Horpibulsuk et al. [28]. The qu(w–d) and N relationship of LCC clay is divided into three different cycles according to its slope. The strength reduction is minimal for the first 0–1 cycle while the dramatic reduction is noted in the 1–3 cycles. The
The clay paste was passed through a 2-mm sieve for removal of shell pieces and other larger size particles, if present. The clay paste was next replaced by FA at replacement ratios of 0%, 20%, 60% and 80% dry weight of clay. Index tests on the mixed soil were subsequently performed. The index properties of the tested clay at different FA replacement ratios are given in Table 2. The water content of the mixed soil was adjusted to 1.5–3 times liquid limit (wL) for the w–d cycle strength tests. The saturated clay was carefully transferred into a mixer and then tamped to minimize air bubbles before mixing with cement and air foam. The lower water content possesses high viscosity and resists the air bubble entry into the pore space [39–42]. The clay–water–FA mixture was mixed with air foam. The air content (Ac) values varied between 0% and 50% by volume of the saturated mixed soil (Vi). The Vi value is the sum of volume of dry soil (Vs) and volume of water (Vw). The Vs value was determined from the dry weight of mixed soil (Ws) and specific gravity values of mixed soil and water. The clay–water–air–FA mixture was then thoroughly
Table 2 Water contents and liquid limits for mixed clay samples. Soil:FA
Liquid limit (%)
Plastic limit (%)
Plasticity index (%)
100:0 80:20 60:40 40:60 20:80
77.1 72.8 66.1 50.2 32.1
32.4 31.2 27.3 24.8 19.8
44.7 41.6 38.8 25.4 12.3
Fig. 2. Influence of FA on w–d cycle strength.
44
A. Neramitkornburi et al. / Construction and Building Materials 77 (2015) 41–49
Fig. 3. Influence of cement content and air content on qu(w–d) for (a) soil:FA = 20:80 and (b) soil:FA = 80:20.
strength reduction in 3–6 cycles is much larger than the 0–1 cycle but is lower than that of 1–3 cycles. For a particular water content, the qu0 and qu(w–d) values are controlled by cement content and air content. It is evident from Fig. 3 that both fabric (water content, air content) and cementation bond (cement content) affect the qu(w–d) of LCC clay. The lower water content and air content and the higher cement content result in the higher qu(w–d). It is therefore pertinent to examine the combination effect of both fabric and cementation bond on qu(w–d) using the structural parameter wV/C. Fig. 4 shows the stress–strain relationship of LCC clay with the same wV/C of 29 but different C and A values at FA replacement ratio of 20. Fig. 5 shows the relationships between qu(w–d) and wV/C for the LCC samples with different C and A values. It is evident that the stress–strain relationships and w–d cycle strengths at the three number of cycles tested are essentially similar as long as wV/C is the same although the cement content and air content vary over a wide range. In other words, the durability against wetting and drying is dependent upon qu0 because the wV/C controls qu0 values of LCC clay. qu0 will be then used as an engineering indicator, indicating the structure strength of LCC clay, to analyze qu(w–d) in the next section. Fig. 4. Stress–strain relationship for samples with the same wV/C of 29.
5. Analysis and discussion Besides unit weight and strength of the LCC material, the flowability of the mixture (before hardening) is also a required
parameter for field construction. The higher flowability results in the lower pump capacity and construction cost. The previous works [27,28] have shown the flowability of the LCC mixture is controlled by w/wL. Since the FA reduces wL of the clay–FA mixture
A. Neramitkornburi et al. / Construction and Building Materials 77 (2015) 41–49
Fig. 5. Relationship between qu(w–d) and wV/C.
(Table 2), the water content, w of the mixture to attain the same w/wL is reduced after adding FA. At the same w/wL (Fig. 2), the LCC mixture with higher FA replacement ratio exhibits higher qu0 and qu(w–d) due to the pozzolanic reaction between cement and FA and the reduction in water to cement ratio. Consequently, FA improves not only the flowability of the LCC mixture but also the durability of LCC material. The strength reduction with number of w–d cycles is due to cracking effect. The wetting causes the swelling of the clay particles due to the expansion of diffusion double layer while the drying causes the shrinkage of the clay particles due to the loss of water [43]. Thus, the swelling and shrinkage for each w–d cycle leads to the tension cracks on the LCC sample. The cracking effect can be depicted by the increase in water content after the end of each w–d cycle (Fig. 6). Due to the cracks, the pore space in the samples increases and carries more water content. It is evident from the test results (Figs. 4 and 5) that the qu(w–d) values at different N is dependent upon qu0 value. As such, qu0 is used as a variable in analyzing the relationships between qu(w–d) versus N as shown in Fig. 7. The relationships are for samples with various air contents and cement contents but with the same qu0 (same wV/C values) at FA replacement ratios of 80 and 60. For a particular FA replacement ratio, the qu(w–d) versus N relationship
45
of LCC clay are of similar pattern as long as the qu0 value is the same, even though the air content and cement content are varied over a wide range. The qu(w–d) versus N relationships can be represented by two functions: linear and logarithm. The linear and logarithmic functions fit very well for 0–1 cycle and 1–6 cycles, respectively. To understand the fundamental role of qu0 on qu(w–d), the normalized strength qu(w–d)/qu0 is plotted versus N as shown in Fig. 8 as previously undertaken by Kampala et al. [43] for Calcium Carbide Residue (CCR) stabilized clay. The qu(w–d)/qu0 for CCR stabilized clay at a particular N is essentially the same for different CCR contents and FA contents. Subsequently, the unique relationship between qu(w–d)/qu0 and N was proposed and is useful for mix design purposes. The same is not true for LCC clay, which possesses very high air contents. For 1 w–d cycle, the qu(1 w–d)/qu0 is independent of qu0; i.e., qu(1 w–d)/qu0 is constant for all mix properties, where qu(1 w–d) is the 1 w–d cycle strength. Beyond 1 w–d cycle, the qu(w–d)/qu0 is dependent upon the qu0 value. The lower qu0 is associated with the larger qu(w–d)/qu0. This implies that the durability is controlled by qu0; i.e., the samples with the same qu0 exhibit the same qu(w–d) even though they were prepared at different mix proportions of water content, cement content and air content. The samples with higher qu0 exhibit higher qu(w–d). This further establishes the fact that durability is lower (strength reduction with increasing N is larger) for a lower structure strength. Based on the results shown in Figs. 7 and 8, the relationship between qu(w–d) and N for different FA replacement ratios, water contents, cement contents and air contents can be represented by a logarithmic function as follows:
quðw—dÞ ¼ quð1
w—dÞ
b ln N
ð4Þ
where b is the degradation index, quantifying the rate of degradation of structure strength due to w–d cycles. As seen in Fig. 8, the qu(1 w–d) is slightly lower than qu0 and essentially the same for all mix proportions. The qu(1 w–d) and qu0 relationship can then be developed based on a linear regression analysis of the strength data (Fig. 9):
quð1
w—dÞ
¼ 0:95qu0
ð5Þ
with a high degree of correlation of 0.997. The b value can be approximated in term of qu0 in a power function (Fig. 10):
b ¼ 0:65ðqu0 Þ5=6 with a high degree of correlation of 0.950.
Fig. 6. Relationship between w and N for (a) soil:FA = 20:80 and (b) soil:FA = 80:20.
ð6Þ
46
A. Neramitkornburi et al. / Construction and Building Materials 77 (2015) 41–49
Fig. 7. qu(w–d) versus N relationship for (a) soil:FA = 20:80 and (b) soil:FA = 80:20.
quðw—dÞ ¼ 0:95qu0 0:65ðqu0 Þ5=6 lnN for 150 kPa < qu0 < 3500 kPa ð7Þ
Fig. 8. Normalized strength qu(w–d)/qu0 versus N relationship.
By combining Eqs. (4)–(6), a predictive w–d cycle strength equation in term of qu0 is shown as follows:
Using Eq. (7), the w–d cycle strengths of samples with various mix proportions at a target number of w–d cycle can be approximated once the corresponding qu0 is known. The qu0 is simply determined directly from laboratory UC test or approximated by available strength equations such as those proposed by Horpibulsuk et al. [38]. Eq. (7) is thus useful for civil engineering practitioners since the durability test is a time-consuming. Besides strength, the modulus of the LCC clay subjected to different number of w–d cycles is a required parameter for deformation analysis. It is logical to relate modulus with UC strength because both are generally dependent on the cemented structure. Fig. 11 presents the relationship between modulus of deformation at 50% strength, E50 and qu(w–d) of LCC samples under various number of w–d cycles. E50 varies between 120 and 200 times qu(w–d) for all number of w–d cycles tested, which is in agreement with the test data for LCC clay without w–d cycles [27,39]. As such, E50 at any number of w–d cycle can be estimated after qu(w–d) is predicted (using Eq. (7)).
A. Neramitkornburi et al. / Construction and Building Materials 77 (2015) 41–49
Fig. 9. Relationship between qu(1
w–d)
47
and qu0 for different soil:FA.
Fig. 10. Relationship between b and qu0 for different soil:FA.
Table 3 shows a prediction of qu(w–d) of the separate LCC samples with FA replacement ratio of 40% (w = 198% and 132% and A = 0–50%). The qu0 was obtained from the UC test after 28 days of curing. The qu(w–d) for various water contents, cement contents, and air contents were predicted by Eq. (7). It is found that the predicted and measured qu(w–d) values are in a good agreement with a small absolute percent error of 6.3. This reinforces the application of the proposed equation. Even though Eq. (7) was developed from a specific soil, the formulation of the proposed equation is on sound principles and can be used as fundamental for other soils. The empirical equation can be further refined with the analysis of more data. Eqs. (1)–(3) and (7) can be used for mix design purpose to meet both unit weight and strength requirement at a target service life. The strength requirement for stabilized pavement material at the target service life is different for different countries. For instance, the strength requirement is 2068 kPa, 1471 kPa, and 2403 kPa for the U.S. Army Corps of Engineers, the Department of Rural Road of Thailand and the Department of Highways of Thailand, respectively. Last but not least, low quality FA that may marginally pass ASTM C-618 standards, or even not even pass, is of interest for
Fig. 11. Relationship between E50 and qu(w–d) for (a) 1, (b) 3 and (c) 6 w–d cycles.
future sustainable research. This FA is available in abundance and has a little market value. The main stumbling block in construction uses has been the variability in FA. FA coming out of power plants varies significantly according to the source of coal, method of burning, and other factors. FA produced by a given power plant can even considerably change for a number of reasons, such as a change in coal source. In recent years, many power plants have modified their operation to create less pollution and as a result, produce FA with carbon content greater than 4% and a high level of dead burn CaO, which will interfere with hydration process. As such, strength and durability test results of the construction and building materials with different FA from variety of sources needs to be investigated and analyzed based on the chemical composition of the fly ashes. Moreover, in North America and other temperate countries, the durability against freeze and thaw testing in accordance with ASTM C-666 standards is also significant. The compressive and flexural strength testing after every 50 freeze–thaw cycles is recommended.
48
A. Neramitkornburi et al. / Construction and Building Materials 77 (2015) 41–49
Table 3 Predicted qu(w–d) for Soil:FA = 60:40. w (%) A (%) C (%) N quo (kPa) qu(1
198
132
0 25 50 0 25 50 0 25 50
50.2 34.5 21.3 50.2 34.5 21.3 50.2 34.5 21.3
1
0 25 50 0 25 50 0 25 50
39.0 27.0 16.8 39.0 27.0 16.8 39.0 27.0 16.8
1
3
6
1012.0 604.0 112.0 1012.0 604.0 112.0 1012.0 604.0 112.0
w–d)
(kPa) Eq. (5) b (kPa) (Eq. (6)) Predicted qu(w–d)P (kPa) (Eq. (7)) Measured qu(w–d)M (kPa) jquðw—dÞM quðw—dÞP j 100 ð%Þ quðw—dÞM
961.4 573.8 106.4 961.4 573.8 106.4 961.4 573.8 106.4
1467.4 1394.0 842.7 800.6 327.5 311.1 3 1467.4 1394.0 842.7 800.6 327.5 311.1 6 1467.4 1394.0 842.7 800.6 327.5 311.1 Mean Absolute Percent Error, MAPE
207.6 135.0 33.2 207.6 135.0 33.2 207.6 135.0 33.2
961.4 573.8 106.4 733.3 425.4 70.0 589.4 331.9 47.0
998.6 590.3 104.2 722.1 385.8 88.4 665.3 281.8 38.0
3.7 2.8 2.1 1.6 10.3 20.8 11.4 17.8 23.7
282.9 178.2 81.1 282.9 178.2 81.1 282.9 178.2 81.1
1394.0 800.6 311.1 1083.2 604.8 222.0 887.1 481.2 165.8
1442.7 831.4 298.7 1126.4 627.1 267.4 907.3 462.8 197.2
3.4 3.7 4.2 3.8 3.6 17.0 2.2 4.0 15.9 6.3
MAPE ¼ 1n
Pn
i¼1
! jquðw—dÞM quðw—dÞP j 100 quðw—dÞM
6. Conclusions
Acknowledgements
This research investigates the viability of using waste materials (clay and FA) for developing sustainable construction LCC materials. Results of this study suggest that the initial soaked strength is critical for analysis of wet–dry cycle strength of LCC clay. The following conclusions can be drawn from this research study.
This work was financially supported by the Thailand Research Fund under the TRF Senior Research Scholar program Grant No. RTA5680002, the Ph.D. Royal Jubilee program, Suranaree University of Technology and the Office of Higher Education Commission under NRU project of Thailand. The authors acknowledge the reviewer for a valuable suggestion on future sustainable research.
(1) FA improves flowability of LCC mixture (before hardening) and durability of LCC material. For the same flowability, the initial soaked strength and w–d cycle strength increase with increasing FA. Significant improvement is found when FA replacement ratio is greater than 40%. (2) The strength reduction with number of w–d cycles is caused by the swelling and shrinkage of clay particles, which in turn leads to crack development (degradation of cemented structure). Due to the cracks, the pore space in the samples increases and carries more water content. (3) The degradation of cemented soil structure is controlled by the initial structure strength. The degradation index (b) is proposed to qualify the strength reduction with number of w–d cycles. (4) The w–d cycle strength and number of w–d cycle relationship is represented by linear function for 0–1 cycle and logarithmic function for 1–6 cycles. Based on the proposed functions and the degradation index, the predictive w–d strength equation is proposed and verified. This equation facilitates mix design to attain the required strength at a target service life, which is very useful for civil engineering practitioners since the durability test is time-consuming. The formulation of the proposed equation is based on sound engineering principles and can be used for a fundamental understanding of soft soils. The empirical equation can be further refined with the analysis of more data. (5) Because the cemented soil structure controls the engineering properties at different w–d cycles, it is logical to relate E50 in term of qu(w–d). The E50 and qu(w–d) relationship is found to be essentially independent of cement content, water content, air content, FA replacement ratio and number of w–d cycle. The relationship is similar to that of LCC clay without w–d cycle. Using the proposed equation, E50 at any number of w– d cycle can be estimated once qu(w–d) is predicted.
References [1] Arulrajah A, Bo MW. Characteristics of Singapore marine clay at Changi. Geotech Geol Eng 2008;26(4):431–41. [2] Chai JC, Pongsivasathit S. A method for predicting consolidation settlements of floating column improved clayey subsoil. Front Archit Civ Eng 2010;4(2):241–51. [3] Shen SL, Wang ZF, Horpibulsuk S, Kim YH. Jet grouting with a newly developed technology: the twin-jet method. Eng Geol 2013;152(1):87–95. [4] Shen SL, Wang ZF, Sun WJ, Wang LB, Horpibulsuk S. A field trial of horizontal jet grouting using the composite-pipe method in soft deposit of Shanghai. Tunn Undergr Space Technol 2013;35:142–51. [5] Horpibulsuk S, Rachan R, Suddeepong A. State of art in strength development of soil–cement columns. Ground Improv 2012;165(4):201–15. [6] Horpibulsuk S, Rachan R, Suddeepong A. Assessment of strength development in blended cement admixed Bangkok clay. Constr Build Mater 2011;25(4):1521–31. [7] Horpibulsuk S, Rachan R, Suddeepong A, Chinkulkijniwat A. Strength development in cement admixed Bangkok clay: laboratory and field investigations. Soils Found 2011;51(2):239–51. [8] Horpibulsuk S, Chinkulkijniwat A, Cholphatsorn A, Suebsuk J, Liu MD. Consolidation behavior of soil cement column improved ground. Comput Geotech 2012;43:37–50. [9] Kasama K, Ochiai H, Yasufuku N. On the stress–strain behaviour of lightly cemented clay based on an extended critical state concept. Soils Found 2000;40(5):37–47. [10] Kavvadas M, Amorosi A. A constitutive model for structured soils. Géotechnique 2000;50(3):263–73. [11] Horpibulsuk S, Bergado DT, Lorenzo GA. Compressibility of cement admixed clays at high water content. Geotechnique 2004;54(2):151–4. [12] Horpibulsuk S, Miura N, Bergado DT. Undrained shear behavior of cement admixed clay at high water content. J Geotech Geoenviron Eng ASCE 2004;130(10):1096–105. [13] Horpibulsuk S, Liu MD, Liyanapathirana DS, Suebsuk J. Behavior of cemented clay simulated via the theoretical framework of the structured cam clay model. Comput Geotech 2010;37:1–9. [14] Suebsuk J, Horpibulsuk S, Liu MD. Modified structured cam clay: a constitutive model for destructured, naturally structured and artificially structured clays. Comput Geotech 2010;37:956–68. [15] Suebsuk J, Horpibulsuk S, Liu MD. A critical state model for overconsolidated structured clays. Comput Geotech 2011;38:648–58.
A. Neramitkornburi et al. / Construction and Building Materials 77 (2015) 41–49 [16] Goodary R, Lecomte-Nana GL, Petit C, Smith DS. Investigation of the strength development in cement-stabilised soils of volcanic origin. Constr Build Mater 2012:592–8. [17] Arulrajah A, Piratheepan J, Disfani MM, Bo MW. Geotechnical and geoenvironmental properties of recycled construction and demolition materials in pavement subbase applications. J Mater Civ Eng 2009;25(8): 1077–88. [18] Arulrajah A, Disfani MM, Horpibulsuk S, Suksiripattanapong C, Prongmanee N. Physical properties and shear strength response of recycled construction and demolition materials in unbound pavement base/subbase pavement. Constr Build Mater 2014;58:245–57. [19] Rahman MA, Arulrajah A, Piratheepan J, Bo MW, Imteaz MA. Resilient modulus and permanent deformation responses of geogrid-reinforced construction and demolition materials. J Mater Civ Eng 2009;26(3):512–9. [20] Tsuchida T, Porbaha A, Yamane N. Development of a geomaterial from dredge bay mud. J Mater Civ Eng ASCE 2001;13(2):152–60. [21] Satoh T, Tsuchida T, Mitsukuri K, Hong Z. Field placing test of lightweight treated soil under seawater in Kumamoto port. Soils Found 2001;41(4):145–54. [22] Hayashi Y, Suzuki A, Matsuo A. Mechanical properties of air-cement-treated soils. Ground Improv 2002;6(1):69–78. [23] Otani J, Mukunoki T, Kikuchi Y. Visualization for engineering property of insitu lightweight soils with air foams. Soils Found 2002;42(3):93–105. [24] Jamnongpipatkul P, Dechasakulsom M, Sukolrat J. Application of air-foam stabilized soil for bridge-embankment transition zone in Thailand. In: GeoHuman international conference. Geotechnical Special Publication No.190. 2009. p. 181–193. [25] Kikuchi Y, Nagatome T. Durability of cement treated clay with air foam used in water front structures. In: Proceedings of Geo-Shanghai 2010, ASCE Special Publication No.207, June 2–5, 2010, China, 2011. p. 1–30. [26] Kikuchi Y, Nagatome T, Mizutani T, Yoshio H. The effect of air foam inclusion on the permeability and absorption of light weight soil. Soils Found 2011;51(1):151–65. [27] Neramitkornburi A, Horpibulsuk S, Suksiripattanapong C, Arulrajah A, Disfani MM. Engineering properties of lightweight cellular cemented clay. Soils Found 2015;55(2) (accepted for publication on December 12, 2014). [28] Horpibulsuk S, Wijitchot A, Neramitkornburee A, Shen SL. Factors influencing unit weight and strength of lightweight cemented clay. Q J Eng Geol Hydrogeol 2014;47:101–8.
49
[29] Concrete Institute of Australia. Fly ash and its use in concrete, current practice note No. 25, ISBN 0 909375 61 5, November 2003, 8p. [30] Bayat O. Characterization of Turkish fly ashes. Fuel 1998;77(9/10):1059–66. [31] Joshi RC, Lohtia RP. Fly ash in concrete – production, properties and uses. Advances in concrete technology, vol. 2. Gordon and Breach Science Publishers; 1997. [32] Estakhri CK, Saylak D. Potential for reduced greenhouse gas emissions in Texas through the use of high volume fly ash concrete. Transp Res Board 2004. [33] Dif AE, Bluemel WF. Expansive soil under cyclic drying and wetting. Geotech Test J 1991;14:96–102. [34] Khattab SAA, Al-Mukhtar M, Fleureau JM. Long-term stability characteristics of a lime-treated plastic soil. J Mater Civil Eng ASCE 2007;19(4):358–66. [35] Rao SM, Reddy BVV, Muttharam M. The impact of cyclic wetting and drying on the swelling behavior of stabilized expansive soil. Eng Geol 2001;60:223–33. [36] Sazzad BS, Rahman K, Mustafa Y, Ireen A. The long-term performance of two fly ash stabilized find-grained soil subbase. Resour Conserv Recycl 2010;54:666–72. [37] Mitchell JK. Fundamentals of soil behavior. New York: John Wiley & Sons, Inc.; 1993. [38] Horpibulsuk S, Suddeepong A, Suksiripattanapong C, Chinkulkijniwat A, Arulrajah A, Disfani MM. Water-void/cement ratio identity of lightweight cellular cemented material. J Mater Civil Eng ASCE 2014;26(10). 06014021(1– 10). [39] Horpibulsuk S, Suddeepong A, Chinkulkijniwat A, Liu MD. Strength and compressibility of lightweight cemented clays. Appl Clay Sci 2014;69:11–21. [40] Prakash K, Sridharan A. Free swell ratio and clay mineralogy of fine-grained soils. Geotech Test J ASTM 2004;27(2):220–5. [41] Horpibulsuk S, Shibuya S, Fuenkajorn K, Katkan W. Assessment of engineering properties of Bangkok clay. Can Geotech J 2007;44(2):173–87. [42] Horpibulsuk S, Rachan R, Suddeepong A, Liu MD, Du YJ. Compressibility of lightweight cemented clays. Eng Geol 2013;159:59–66. [43] Kampala A, Horpibulsuk S, Prongmanee N, Chinkulkijniwat A. Influence of wet–dry cycles on compressive strength of calcium carbide residue-fly ash stabilized clay. J Mater Civ Eng ASCE 2014;24(1):633–43.
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Durability against wetting–drying cycles of sustainable Lightweight Cellular Cemented construction material comprising clay and fly ash wastes Anek Neramitkornburi a, Suksun Horpibulsuk b,c,⇑, Shui Long Shen d, Avirut Chinkulkijniwat a, Arul Arulrajah e, Mahdi Miri Disfani e a
School of Civil Engineering, Suranaree University of Technology, Nakhon Ratchasima 30000, Thailand School of Civil Engineering, Center of Excellence in Civil Engineering, Suranaree University of Technology, Nakhon Ratchasima 30000, Thailand c Centre for Sustainable Infrastructure, Swinburne University of Technology, Australia d Department of Civil Engineering and State Key Laboratory of Ocean Engineering, Shanghai Jiao Tong University, 800 Dongchuan Rd, Shanghai 200240, China e Faculty of Science, Engineering and Technology, Swinburne University of Technology, Melbourne, Australia b
h i g h l i g h t s Waste materials: fly ash and in-situ clay. A green Lightweight Cellular Cemented (LCC) clay. Role of cement, air and FA content on durability. Predictive wetting–drying (w–d) cycled strength equation.
a r t i c l e
i n f o
Article history: Received 5 June 2014 Received in revised form 29 October 2014 Accepted 14 December 2014
Keywords: Air foam Cement Durability Compressive strength Fly ash Wet–dry cycle Lightweight material
a b s t r a c t The viability of using waste materials such as clay and fly ash (FA) for developing a sustainable Lightweight Cellular Cemented (LCC) construction material is investigated in this paper. LCC clay has a wide range of applications in infrastructure rehabilitation as well as in the construction of new facilities. The durability against wetting–drying (w–d) cycles is an important parameter for service life design of LCC clay; however, studies on this aspect to date are very limited. The role of cemented soil structure (fabric and cementation bond) on w–d cycle strength of LCC clay are investigated, analyzed and presented in this paper. The strength reduction with increasing number of w–d is attributed to degradation of the cemented structure. The degradation index, qualifying the rate of degradation with number of w–d cycles, is proposed in term of initial soaked strength (without w–d cycle). Using the degradation index, the predictive w–d cycle strength equation at different number of w–d cycles is furthermore proposed. The applicability of the proposed equation is validated using a separate test data. This approach of predicting w–d cycle strength is beneficial from both engineering and economic points of view. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction There is a myriad of problems associated with the engineering construction in soft clay deposits, particularly in coastal regions in ⇑ Corresponding author at: School of Civil Engineering, Suranaree University of Technology, 111 University Avenue, Muang District, Nakhon Ratchasima 30000, Thailand. Tel.: +66 44 22 4322; fax: +66 44 22 4607. E-mail addresses: [email protected] (A. Neramitkornburi), [email protected]. th, [email protected] (S. Horpibulsuk), [email protected] (S.L. Shen), [email protected]. th (A. Chinkulkijniwat), [email protected] (A. Arulrajah), mmiridisfani@swin. edu.au (M.M. Disfani). http://dx.doi.org/10.1016/j.conbuildmat.2014.12.025 0950-0618/Ó 2014 Elsevier Ltd. All rights reserved.
Southeast Asia such as Chao Phraya Plain in Thailand, Mekong Delta in Vietnam and Cambodia, Central Plains of the Philippines, Coastal Plains of Malaysia, Indonesia, Singapore, Hong Kong, Korea, Japan and Taiwan. This soft soil, located in marine or estuary environments, have low shear strength, low bearing capacity and high natural water content, resulting in high compressibility potential. To mitigate future issues with construction on these soft soil deposits, the deep mixing technique is frequently applied [1–8]. The mechanical behavior of cement admixed clays have been extensively investigated by authors [9–15]. The role of physical properties of soil on the strength development is recently investigated by Goodary et al.
42
A. Neramitkornburi et al. / Construction and Building Materials 77 (2015) 41–49
[16]. They reported that soils with a high specific surface (fine particles) needed less amount of cement to provide the same strength and durability as those with low specific surface (coarse particles). Instead of improving the soft ground (foundation) through the costly deep mixing method, the usage of Lightweight Cellular Cemented (LCC) construction materials is an attractive and economical alternative in construction applications such as in embankment, pavement pipe bedding and backfilling. LCC material is a mixture of aggregate, air foam agent and cementing agent. The usage of recycled waste materials such as Construction and Demolition (C&D) materials incorporating recycled concrete aggregate, crushed brick, and reclaimed asphalt pavement [17–19] has been applied in recent years in a wide range of applications such as embankment fills, pipe-bedding and pavement base/subbase. The usage of recycled waste materials in sustainable manner in the development of LCC materials will further support zero-waste directives currently implemented in many developed and developing countries. Incorporation of waste materials in the development of LCC material will reduce the carbon footprint of our future infrastructures. A sustainable LCC material made from in-situ and waste clay obtained directly from construction sites has been extensively used for highway and port constructions in Southeast Asian countries such as Japan and Thailand [20–26]. To reduce the cost of the LCC clay from an economical and environmental perspective, the replacement of cement by fly ash (FA) is an attractive method. It was evident from the flowability test (undertaken using a flow cone with 117 mm in height and 254 mm diameter at base and 117 mm diameter at top) that FA reduces the plasticity and improves the flowability of the LCC clay mixture before hardening [27,28]. At the same water content and cement content, the LCC clay with higher FA replacement ratio exhibits higher strength than that with lower FA replacement ratio. The use of FA as a supplementary cementitious material in concrete is also well recognized for improving durability of concrete [29,30]. The incorporation of FA results in considerable pore refinement [31], which leads to a low porosity and discontinuous pore structure. Consequently, the permeability of the concrete reduces and the durability of the concrete increases [32]. The in-situ LCC clay as a stabilized engineering fills and pavement materials generally encounters with w–d cycles from the change of weather during wet (rainy) and dry (summer) seasons. This is particular relevant for tropical countries such as Thailand, as well as parts of Australia and China. The w–d cycles result in tension and surface cracks, which can damage the stabilized pavement structure [33–36]. Even though there is available research on the strength development in LCC clay, the investigation of durability against the wetting–drying cycles (w–d cycles), a critical aspect for infrastructure design such as in engineering fills and pavements, is very limited and is the prime focus of this research. The investigation of the service life of the LCC clay via wetting and drying test is significant and is another focus of this research. The strength of LCC clay is dependent upon the cemented soil structure (fabric and cementation bond) [37]. As such, tests with a wide range of water contents, air contents, FA contents and cement contents of the LCC clay are undertaken to understand the role of both fabric and cementation bond on the w–d cycle strengths. Based on the analysis of the test results, a rational empirical relationship between w–d cycle strengths and initial soaked strength (without w–d cycles) is proposed. This equation can facilitate the determination of a suitable mix proportion of LCC materials to meet the strength requirement at a target service life. This research will enable waste excavated soft clay traditionally destined for landfill to be used in a sustainable manner as an aggregate in LCC materials, which is significant in term of engineering, economical and environmental perspectives.
2. Theoretical background For a LCC clay at a water content between 1.5 and 3.0 times the liquid limit, the strength is determined exclusively by the watervoid to cement, wV/C [38]. This parameter is defined as the product of initial clay water content (before mixing with cement and air foam) times V/C, where the water content is expressed in fraction. The parameter V/C is defined as the ratio of volume of voids to the volume of cement in the mix. Strength is independent of water content, air content and cement content in the mix. Based on extensive test results, Horpibulsuk et al. [38] have proposed a predictive strength equation in term of curing time, and wV=C for the LCC Bangkok clay as follows:
(
qðwV=C ÞD
)
qðwV=C Þ28
¼
1:27 ðwV=C Þ28 ð0:027 þ 0:300 ln DÞ ðwV=C ÞD
ð1Þ
where qðwV=C ÞD is the strength of LCC clay to be estimated at watervoid/cement ratio of (wV=C) after D days of curing and qðwV=C Þ28 is the strength of LCC clay at water-void/cement ratio of (wV=C) after 28 days of curing. The unit weight (in kN/m3) is determined in term of water content, cement content and V/C by using Eq. (2) [27,39]:
0 1 Gc Gs c2w ð1 þ wÞ þ Gc cw C BGs c ð1 þ wÞC c¼ ðV=C Þ@ w A Gc cw Gc cw þ1 þ1 C C
ð2Þ
where w is water content (in fraction), Gc and Gs are the specific gravities of cement and soil, respectively, cw is unit weight of water (kN/m3) and C is cement content (kg/m3). Eq. (2) was developed based on the assumption that all air bubbles (air foam) enter into the pore space when mixed with cement and clay. With the variation in water content and cement content, the air content required to attain the required V=C is determined:
Ac ¼ ðV=C Þ
C ð1 þ wGs Þ wGs Gc cw
ð3Þ
3. Materials and methods 3.1. Materials 3.1.1. Soil sample Bangkok clay was collected from Bangkok Noi district, Bangkok, Thailand at a 3 m depth. The clay was composed of 2% sand, 39% silt and 55% clay as shown in Fig. 1. The natural water content was 80% and the specific gravity was 2.64. The liquid and plastic limits were 73% and 31%, respectively. Based on the Unified Soil Classification System (USCS), the clay was classified as inorganic clay of high plasticity (CH). Groundwater was encountered at a depth of approximately 1 m below the surface. The clay was classified as low swelling type with free swell ratio (FSR) of 1.1. The FSR is defined as the ratio of equilibrium sediment volume of 10 g of oven-dried soil passing a 425 mm sieve in distilled water (Vd) to that in kerosene (Vk) [40]. This method was adopted since it is simple and predicts the dominant clay mineralogy of soil satisfactorily [41]. Table 1 summarizes the chemical composition of Bangkok clay using X-ray fluorescence (XRF). Even though Bangkok clay is classified as low swelling types, moderately high clay content of up to 55% may cause swelling and shrinkage on the LCC clay during w–d cycles. This effect will be examined in this paper.
3.1.2. Cement and air foam agent Type I Portland cement (PC) and air foam agent, Darex AE4, provided by the Grace Construction Products Ltd., were used in this study. The grain size distribution curve, obtained from the laser particle size analysis, and chemical composition of PC are also shown in Fig. 1 and Table 1, respectively. The specific gravity is 3.15 and the D50 is 0.01 mm (10 micron), which is larger than that of the tested clay. The air foam agent is a blend of anionic surfactants and foam stabilizers. It is a liquid air entraining agent used in various types of mortar, concrete and cementitious material.
A. Neramitkornburi et al. / Construction and Building Materials 77 (2015) 41–49
43
mixed with cement for 10 min. The cement content (C) was varied from 10% to 40% by weight of dry soil. The uniform paste was next transferred to cylindrical containers of 50 mm diameter and 100 mm height for w–d strength test. After 24 h, the cylindrical samples were dismantled. The cylindrical samples were wrapped in vinyl bags and stored in a humidity-controlled room of constant temperature (23 ± 2 °C) until 28 days of curing. The method of cyclic wetting and drying test as per ASTM D 559 was adopted for sample preparations. The samples were submerged in deionized water at room temperature for 5 h. They were then dried in the oven at a temperature of 70 °C for 48 h and air-dried at room temperature for at least 3 h. This process is referred to as 1 w–d cycle. After attaining the target w–d cycles, the samples were immersed in deionized water for 2 h at the constant temperature of 25 ± 2 °C. Unconfined Compression (UC) tests were then undertaken with a rate of vertical displacement of 1 mm/min. The 1, 3 and 6 w–d cycles were considered in this study. Based on a critical analysis of the strength data, a rational predictive w–d cycle strength equation is proposed, which facilitates the mix design to attain the strength requirement at a specified service life for civil engineering practitioners. In addition to the above mentioned laboratory tests, the results of the strength tests on separate LCC samples at FA replacement ratios of 40% (w = 198% and 132%, and A = 0%, 25% and 50%) were taken to verify the proposed predictive equation.
4. Results
Fig. 1. Grain size distribution of clay, PC and FA.
Table 1 Chemical composition of Bangkok clay, PC and FA. Chemical composition (%)
Bangkok clay
PC
FA
SiO2 Al2O3 Fe2O3 CaO MgO SO3 Na2O K2O LOI
62.8 21.3 8.4 0.9 1.5 1.2 0.3 2.5 0.8
20.9 4.8 3.4 65.4 1.2 2.7 0.2 0.3 1.1
44.7 23.7 11.0 12.7 2.6 1.3 0.1 2.9 1.0
3.1.3. Fly ash Fly ash (FA) was obtained from the Mae Moh power plant in the north of Thailand. Table 1 summarizes the chemical composition of FA showing that total amount of the major components SiO2, Al2O3 and Fe2O3 in FA are 79.4% and classified as class F according to ASTM C 618. The grain size distribution curve of FA is also shown in Fig. 1.
3.2. Methodology
Fig. 2 shows the typical role of FA on w–d cycle strengths of LCC clay at different number of w–d cycles. All the samples were prepared at different water contents but at the same w/wL of 2.0, C = 10% and A = 25% to have the same flowability of the LCC mixture, where w is water content and wL is liquid limit. It has been proved that at the same w/wL, the LCC mixtures have the same flowability [27]. Fig. 2 shows that FA improves w–d cycle strength, qu(w–d), where significant improvement is clearly observed when FA replacement ratio is greater than 40%. Fig. 3 shows the role of cement content and air content on w–d cycle strength for FA replacement of 20% and 80%. The unit weight decreases with increasing air content and decreasing cement content. Although the air foam helps reduce the unit weight of the LCC clay, it causes strength reduction at a specific cement content. It is evident that the initial soaked (without w–d cycle) strengths (qu0) and w–d cycle strengths (qu(w–d)) decrease with decreasing cement content and increasing air content. At a particular air content, the strength increases significantly with increasing cement content even with a slight increase in unit weight. The unit weight of LCC clay can be approximated from Eqs. (1) and (2) as previously undertaken by Horpibulsuk et al. [28]. The qu(w–d) and N relationship of LCC clay is divided into three different cycles according to its slope. The strength reduction is minimal for the first 0–1 cycle while the dramatic reduction is noted in the 1–3 cycles. The
The clay paste was passed through a 2-mm sieve for removal of shell pieces and other larger size particles, if present. The clay paste was next replaced by FA at replacement ratios of 0%, 20%, 60% and 80% dry weight of clay. Index tests on the mixed soil were subsequently performed. The index properties of the tested clay at different FA replacement ratios are given in Table 2. The water content of the mixed soil was adjusted to 1.5–3 times liquid limit (wL) for the w–d cycle strength tests. The saturated clay was carefully transferred into a mixer and then tamped to minimize air bubbles before mixing with cement and air foam. The lower water content possesses high viscosity and resists the air bubble entry into the pore space [39–42]. The clay–water–FA mixture was mixed with air foam. The air content (Ac) values varied between 0% and 50% by volume of the saturated mixed soil (Vi). The Vi value is the sum of volume of dry soil (Vs) and volume of water (Vw). The Vs value was determined from the dry weight of mixed soil (Ws) and specific gravity values of mixed soil and water. The clay–water–air–FA mixture was then thoroughly
Table 2 Water contents and liquid limits for mixed clay samples. Soil:FA
Liquid limit (%)
Plastic limit (%)
Plasticity index (%)
100:0 80:20 60:40 40:60 20:80
77.1 72.8 66.1 50.2 32.1
32.4 31.2 27.3 24.8 19.8
44.7 41.6 38.8 25.4 12.3
Fig. 2. Influence of FA on w–d cycle strength.
44
A. Neramitkornburi et al. / Construction and Building Materials 77 (2015) 41–49
Fig. 3. Influence of cement content and air content on qu(w–d) for (a) soil:FA = 20:80 and (b) soil:FA = 80:20.
strength reduction in 3–6 cycles is much larger than the 0–1 cycle but is lower than that of 1–3 cycles. For a particular water content, the qu0 and qu(w–d) values are controlled by cement content and air content. It is evident from Fig. 3 that both fabric (water content, air content) and cementation bond (cement content) affect the qu(w–d) of LCC clay. The lower water content and air content and the higher cement content result in the higher qu(w–d). It is therefore pertinent to examine the combination effect of both fabric and cementation bond on qu(w–d) using the structural parameter wV/C. Fig. 4 shows the stress–strain relationship of LCC clay with the same wV/C of 29 but different C and A values at FA replacement ratio of 20. Fig. 5 shows the relationships between qu(w–d) and wV/C for the LCC samples with different C and A values. It is evident that the stress–strain relationships and w–d cycle strengths at the three number of cycles tested are essentially similar as long as wV/C is the same although the cement content and air content vary over a wide range. In other words, the durability against wetting and drying is dependent upon qu0 because the wV/C controls qu0 values of LCC clay. qu0 will be then used as an engineering indicator, indicating the structure strength of LCC clay, to analyze qu(w–d) in the next section. Fig. 4. Stress–strain relationship for samples with the same wV/C of 29.
5. Analysis and discussion Besides unit weight and strength of the LCC material, the flowability of the mixture (before hardening) is also a required
parameter for field construction. The higher flowability results in the lower pump capacity and construction cost. The previous works [27,28] have shown the flowability of the LCC mixture is controlled by w/wL. Since the FA reduces wL of the clay–FA mixture
A. Neramitkornburi et al. / Construction and Building Materials 77 (2015) 41–49
Fig. 5. Relationship between qu(w–d) and wV/C.
(Table 2), the water content, w of the mixture to attain the same w/wL is reduced after adding FA. At the same w/wL (Fig. 2), the LCC mixture with higher FA replacement ratio exhibits higher qu0 and qu(w–d) due to the pozzolanic reaction between cement and FA and the reduction in water to cement ratio. Consequently, FA improves not only the flowability of the LCC mixture but also the durability of LCC material. The strength reduction with number of w–d cycles is due to cracking effect. The wetting causes the swelling of the clay particles due to the expansion of diffusion double layer while the drying causes the shrinkage of the clay particles due to the loss of water [43]. Thus, the swelling and shrinkage for each w–d cycle leads to the tension cracks on the LCC sample. The cracking effect can be depicted by the increase in water content after the end of each w–d cycle (Fig. 6). Due to the cracks, the pore space in the samples increases and carries more water content. It is evident from the test results (Figs. 4 and 5) that the qu(w–d) values at different N is dependent upon qu0 value. As such, qu0 is used as a variable in analyzing the relationships between qu(w–d) versus N as shown in Fig. 7. The relationships are for samples with various air contents and cement contents but with the same qu0 (same wV/C values) at FA replacement ratios of 80 and 60. For a particular FA replacement ratio, the qu(w–d) versus N relationship
45
of LCC clay are of similar pattern as long as the qu0 value is the same, even though the air content and cement content are varied over a wide range. The qu(w–d) versus N relationships can be represented by two functions: linear and logarithm. The linear and logarithmic functions fit very well for 0–1 cycle and 1–6 cycles, respectively. To understand the fundamental role of qu0 on qu(w–d), the normalized strength qu(w–d)/qu0 is plotted versus N as shown in Fig. 8 as previously undertaken by Kampala et al. [43] for Calcium Carbide Residue (CCR) stabilized clay. The qu(w–d)/qu0 for CCR stabilized clay at a particular N is essentially the same for different CCR contents and FA contents. Subsequently, the unique relationship between qu(w–d)/qu0 and N was proposed and is useful for mix design purposes. The same is not true for LCC clay, which possesses very high air contents. For 1 w–d cycle, the qu(1 w–d)/qu0 is independent of qu0; i.e., qu(1 w–d)/qu0 is constant for all mix properties, where qu(1 w–d) is the 1 w–d cycle strength. Beyond 1 w–d cycle, the qu(w–d)/qu0 is dependent upon the qu0 value. The lower qu0 is associated with the larger qu(w–d)/qu0. This implies that the durability is controlled by qu0; i.e., the samples with the same qu0 exhibit the same qu(w–d) even though they were prepared at different mix proportions of water content, cement content and air content. The samples with higher qu0 exhibit higher qu(w–d). This further establishes the fact that durability is lower (strength reduction with increasing N is larger) for a lower structure strength. Based on the results shown in Figs. 7 and 8, the relationship between qu(w–d) and N for different FA replacement ratios, water contents, cement contents and air contents can be represented by a logarithmic function as follows:
quðw—dÞ ¼ quð1
w—dÞ
b ln N
ð4Þ
where b is the degradation index, quantifying the rate of degradation of structure strength due to w–d cycles. As seen in Fig. 8, the qu(1 w–d) is slightly lower than qu0 and essentially the same for all mix proportions. The qu(1 w–d) and qu0 relationship can then be developed based on a linear regression analysis of the strength data (Fig. 9):
quð1
w—dÞ
¼ 0:95qu0
ð5Þ
with a high degree of correlation of 0.997. The b value can be approximated in term of qu0 in a power function (Fig. 10):
b ¼ 0:65ðqu0 Þ5=6 with a high degree of correlation of 0.950.
Fig. 6. Relationship between w and N for (a) soil:FA = 20:80 and (b) soil:FA = 80:20.
ð6Þ
46
A. Neramitkornburi et al. / Construction and Building Materials 77 (2015) 41–49
Fig. 7. qu(w–d) versus N relationship for (a) soil:FA = 20:80 and (b) soil:FA = 80:20.
quðw—dÞ ¼ 0:95qu0 0:65ðqu0 Þ5=6 lnN for 150 kPa < qu0 < 3500 kPa ð7Þ
Fig. 8. Normalized strength qu(w–d)/qu0 versus N relationship.
By combining Eqs. (4)–(6), a predictive w–d cycle strength equation in term of qu0 is shown as follows:
Using Eq. (7), the w–d cycle strengths of samples with various mix proportions at a target number of w–d cycle can be approximated once the corresponding qu0 is known. The qu0 is simply determined directly from laboratory UC test or approximated by available strength equations such as those proposed by Horpibulsuk et al. [38]. Eq. (7) is thus useful for civil engineering practitioners since the durability test is a time-consuming. Besides strength, the modulus of the LCC clay subjected to different number of w–d cycles is a required parameter for deformation analysis. It is logical to relate modulus with UC strength because both are generally dependent on the cemented structure. Fig. 11 presents the relationship between modulus of deformation at 50% strength, E50 and qu(w–d) of LCC samples under various number of w–d cycles. E50 varies between 120 and 200 times qu(w–d) for all number of w–d cycles tested, which is in agreement with the test data for LCC clay without w–d cycles [27,39]. As such, E50 at any number of w–d cycle can be estimated after qu(w–d) is predicted (using Eq. (7)).
A. Neramitkornburi et al. / Construction and Building Materials 77 (2015) 41–49
Fig. 9. Relationship between qu(1
w–d)
47
and qu0 for different soil:FA.
Fig. 10. Relationship between b and qu0 for different soil:FA.
Table 3 shows a prediction of qu(w–d) of the separate LCC samples with FA replacement ratio of 40% (w = 198% and 132% and A = 0–50%). The qu0 was obtained from the UC test after 28 days of curing. The qu(w–d) for various water contents, cement contents, and air contents were predicted by Eq. (7). It is found that the predicted and measured qu(w–d) values are in a good agreement with a small absolute percent error of 6.3. This reinforces the application of the proposed equation. Even though Eq. (7) was developed from a specific soil, the formulation of the proposed equation is on sound principles and can be used as fundamental for other soils. The empirical equation can be further refined with the analysis of more data. Eqs. (1)–(3) and (7) can be used for mix design purpose to meet both unit weight and strength requirement at a target service life. The strength requirement for stabilized pavement material at the target service life is different for different countries. For instance, the strength requirement is 2068 kPa, 1471 kPa, and 2403 kPa for the U.S. Army Corps of Engineers, the Department of Rural Road of Thailand and the Department of Highways of Thailand, respectively. Last but not least, low quality FA that may marginally pass ASTM C-618 standards, or even not even pass, is of interest for
Fig. 11. Relationship between E50 and qu(w–d) for (a) 1, (b) 3 and (c) 6 w–d cycles.
future sustainable research. This FA is available in abundance and has a little market value. The main stumbling block in construction uses has been the variability in FA. FA coming out of power plants varies significantly according to the source of coal, method of burning, and other factors. FA produced by a given power plant can even considerably change for a number of reasons, such as a change in coal source. In recent years, many power plants have modified their operation to create less pollution and as a result, produce FA with carbon content greater than 4% and a high level of dead burn CaO, which will interfere with hydration process. As such, strength and durability test results of the construction and building materials with different FA from variety of sources needs to be investigated and analyzed based on the chemical composition of the fly ashes. Moreover, in North America and other temperate countries, the durability against freeze and thaw testing in accordance with ASTM C-666 standards is also significant. The compressive and flexural strength testing after every 50 freeze–thaw cycles is recommended.
48
A. Neramitkornburi et al. / Construction and Building Materials 77 (2015) 41–49
Table 3 Predicted qu(w–d) for Soil:FA = 60:40. w (%) A (%) C (%) N quo (kPa) qu(1
198
132
0 25 50 0 25 50 0 25 50
50.2 34.5 21.3 50.2 34.5 21.3 50.2 34.5 21.3
1
0 25 50 0 25 50 0 25 50
39.0 27.0 16.8 39.0 27.0 16.8 39.0 27.0 16.8
1
3
6
1012.0 604.0 112.0 1012.0 604.0 112.0 1012.0 604.0 112.0
w–d)
(kPa) Eq. (5) b (kPa) (Eq. (6)) Predicted qu(w–d)P (kPa) (Eq. (7)) Measured qu(w–d)M (kPa) jquðw—dÞM quðw—dÞP j 100 ð%Þ quðw—dÞM
961.4 573.8 106.4 961.4 573.8 106.4 961.4 573.8 106.4
1467.4 1394.0 842.7 800.6 327.5 311.1 3 1467.4 1394.0 842.7 800.6 327.5 311.1 6 1467.4 1394.0 842.7 800.6 327.5 311.1 Mean Absolute Percent Error, MAPE
207.6 135.0 33.2 207.6 135.0 33.2 207.6 135.0 33.2
961.4 573.8 106.4 733.3 425.4 70.0 589.4 331.9 47.0
998.6 590.3 104.2 722.1 385.8 88.4 665.3 281.8 38.0
3.7 2.8 2.1 1.6 10.3 20.8 11.4 17.8 23.7
282.9 178.2 81.1 282.9 178.2 81.1 282.9 178.2 81.1
1394.0 800.6 311.1 1083.2 604.8 222.0 887.1 481.2 165.8
1442.7 831.4 298.7 1126.4 627.1 267.4 907.3 462.8 197.2
3.4 3.7 4.2 3.8 3.6 17.0 2.2 4.0 15.9 6.3
MAPE ¼ 1n
Pn
i¼1
! jquðw—dÞM quðw—dÞP j 100 quðw—dÞM
6. Conclusions
Acknowledgements
This research investigates the viability of using waste materials (clay and FA) for developing sustainable construction LCC materials. Results of this study suggest that the initial soaked strength is critical for analysis of wet–dry cycle strength of LCC clay. The following conclusions can be drawn from this research study.
This work was financially supported by the Thailand Research Fund under the TRF Senior Research Scholar program Grant No. RTA5680002, the Ph.D. Royal Jubilee program, Suranaree University of Technology and the Office of Higher Education Commission under NRU project of Thailand. The authors acknowledge the reviewer for a valuable suggestion on future sustainable research.
(1) FA improves flowability of LCC mixture (before hardening) and durability of LCC material. For the same flowability, the initial soaked strength and w–d cycle strength increase with increasing FA. Significant improvement is found when FA replacement ratio is greater than 40%. (2) The strength reduction with number of w–d cycles is caused by the swelling and shrinkage of clay particles, which in turn leads to crack development (degradation of cemented structure). Due to the cracks, the pore space in the samples increases and carries more water content. (3) The degradation of cemented soil structure is controlled by the initial structure strength. The degradation index (b) is proposed to qualify the strength reduction with number of w–d cycles. (4) The w–d cycle strength and number of w–d cycle relationship is represented by linear function for 0–1 cycle and logarithmic function for 1–6 cycles. Based on the proposed functions and the degradation index, the predictive w–d strength equation is proposed and verified. This equation facilitates mix design to attain the required strength at a target service life, which is very useful for civil engineering practitioners since the durability test is time-consuming. The formulation of the proposed equation is based on sound engineering principles and can be used for a fundamental understanding of soft soils. The empirical equation can be further refined with the analysis of more data. (5) Because the cemented soil structure controls the engineering properties at different w–d cycles, it is logical to relate E50 in term of qu(w–d). The E50 and qu(w–d) relationship is found to be essentially independent of cement content, water content, air content, FA replacement ratio and number of w–d cycle. The relationship is similar to that of LCC clay without w–d cycle. Using the proposed equation, E50 at any number of w– d cycle can be estimated once qu(w–d) is predicted.
References [1] Arulrajah A, Bo MW. Characteristics of Singapore marine clay at Changi. Geotech Geol Eng 2008;26(4):431–41. [2] Chai JC, Pongsivasathit S. A method for predicting consolidation settlements of floating column improved clayey subsoil. Front Archit Civ Eng 2010;4(2):241–51. [3] Shen SL, Wang ZF, Horpibulsuk S, Kim YH. Jet grouting with a newly developed technology: the twin-jet method. Eng Geol 2013;152(1):87–95. [4] Shen SL, Wang ZF, Sun WJ, Wang LB, Horpibulsuk S. A field trial of horizontal jet grouting using the composite-pipe method in soft deposit of Shanghai. Tunn Undergr Space Technol 2013;35:142–51. [5] Horpibulsuk S, Rachan R, Suddeepong A. State of art in strength development of soil–cement columns. Ground Improv 2012;165(4):201–15. [6] Horpibulsuk S, Rachan R, Suddeepong A. Assessment of strength development in blended cement admixed Bangkok clay. Constr Build Mater 2011;25(4):1521–31. [7] Horpibulsuk S, Rachan R, Suddeepong A, Chinkulkijniwat A. Strength development in cement admixed Bangkok clay: laboratory and field investigations. Soils Found 2011;51(2):239–51. [8] Horpibulsuk S, Chinkulkijniwat A, Cholphatsorn A, Suebsuk J, Liu MD. Consolidation behavior of soil cement column improved ground. Comput Geotech 2012;43:37–50. [9] Kasama K, Ochiai H, Yasufuku N. On the stress–strain behaviour of lightly cemented clay based on an extended critical state concept. Soils Found 2000;40(5):37–47. [10] Kavvadas M, Amorosi A. A constitutive model for structured soils. Géotechnique 2000;50(3):263–73. [11] Horpibulsuk S, Bergado DT, Lorenzo GA. Compressibility of cement admixed clays at high water content. Geotechnique 2004;54(2):151–4. [12] Horpibulsuk S, Miura N, Bergado DT. Undrained shear behavior of cement admixed clay at high water content. J Geotech Geoenviron Eng ASCE 2004;130(10):1096–105. [13] Horpibulsuk S, Liu MD, Liyanapathirana DS, Suebsuk J. Behavior of cemented clay simulated via the theoretical framework of the structured cam clay model. Comput Geotech 2010;37:1–9. [14] Suebsuk J, Horpibulsuk S, Liu MD. Modified structured cam clay: a constitutive model for destructured, naturally structured and artificially structured clays. Comput Geotech 2010;37:956–68. [15] Suebsuk J, Horpibulsuk S, Liu MD. A critical state model for overconsolidated structured clays. Comput Geotech 2011;38:648–58.
A. Neramitkornburi et al. / Construction and Building Materials 77 (2015) 41–49 [16] Goodary R, Lecomte-Nana GL, Petit C, Smith DS. Investigation of the strength development in cement-stabilised soils of volcanic origin. Constr Build Mater 2012:592–8. [17] Arulrajah A, Piratheepan J, Disfani MM, Bo MW. Geotechnical and geoenvironmental properties of recycled construction and demolition materials in pavement subbase applications. J Mater Civ Eng 2009;25(8): 1077–88. [18] Arulrajah A, Disfani MM, Horpibulsuk S, Suksiripattanapong C, Prongmanee N. Physical properties and shear strength response of recycled construction and demolition materials in unbound pavement base/subbase pavement. Constr Build Mater 2014;58:245–57. [19] Rahman MA, Arulrajah A, Piratheepan J, Bo MW, Imteaz MA. Resilient modulus and permanent deformation responses of geogrid-reinforced construction and demolition materials. J Mater Civ Eng 2009;26(3):512–9. [20] Tsuchida T, Porbaha A, Yamane N. Development of a geomaterial from dredge bay mud. J Mater Civ Eng ASCE 2001;13(2):152–60. [21] Satoh T, Tsuchida T, Mitsukuri K, Hong Z. Field placing test of lightweight treated soil under seawater in Kumamoto port. Soils Found 2001;41(4):145–54. [22] Hayashi Y, Suzuki A, Matsuo A. Mechanical properties of air-cement-treated soils. Ground Improv 2002;6(1):69–78. [23] Otani J, Mukunoki T, Kikuchi Y. Visualization for engineering property of insitu lightweight soils with air foams. Soils Found 2002;42(3):93–105. [24] Jamnongpipatkul P, Dechasakulsom M, Sukolrat J. Application of air-foam stabilized soil for bridge-embankment transition zone in Thailand. In: GeoHuman international conference. Geotechnical Special Publication No.190. 2009. p. 181–193. [25] Kikuchi Y, Nagatome T. Durability of cement treated clay with air foam used in water front structures. In: Proceedings of Geo-Shanghai 2010, ASCE Special Publication No.207, June 2–5, 2010, China, 2011. p. 1–30. [26] Kikuchi Y, Nagatome T, Mizutani T, Yoshio H. The effect of air foam inclusion on the permeability and absorption of light weight soil. Soils Found 2011;51(1):151–65. [27] Neramitkornburi A, Horpibulsuk S, Suksiripattanapong C, Arulrajah A, Disfani MM. Engineering properties of lightweight cellular cemented clay. Soils Found 2015;55(2) (accepted for publication on December 12, 2014). [28] Horpibulsuk S, Wijitchot A, Neramitkornburee A, Shen SL. Factors influencing unit weight and strength of lightweight cemented clay. Q J Eng Geol Hydrogeol 2014;47:101–8.
49
[29] Concrete Institute of Australia. Fly ash and its use in concrete, current practice note No. 25, ISBN 0 909375 61 5, November 2003, 8p. [30] Bayat O. Characterization of Turkish fly ashes. Fuel 1998;77(9/10):1059–66. [31] Joshi RC, Lohtia RP. Fly ash in concrete – production, properties and uses. Advances in concrete technology, vol. 2. Gordon and Breach Science Publishers; 1997. [32] Estakhri CK, Saylak D. Potential for reduced greenhouse gas emissions in Texas through the use of high volume fly ash concrete. Transp Res Board 2004. [33] Dif AE, Bluemel WF. Expansive soil under cyclic drying and wetting. Geotech Test J 1991;14:96–102. [34] Khattab SAA, Al-Mukhtar M, Fleureau JM. Long-term stability characteristics of a lime-treated plastic soil. J Mater Civil Eng ASCE 2007;19(4):358–66. [35] Rao SM, Reddy BVV, Muttharam M. The impact of cyclic wetting and drying on the swelling behavior of stabilized expansive soil. Eng Geol 2001;60:223–33. [36] Sazzad BS, Rahman K, Mustafa Y, Ireen A. The long-term performance of two fly ash stabilized find-grained soil subbase. Resour Conserv Recycl 2010;54:666–72. [37] Mitchell JK. Fundamentals of soil behavior. New York: John Wiley & Sons, Inc.; 1993. [38] Horpibulsuk S, Suddeepong A, Suksiripattanapong C, Chinkulkijniwat A, Arulrajah A, Disfani MM. Water-void/cement ratio identity of lightweight cellular cemented material. J Mater Civil Eng ASCE 2014;26(10). 06014021(1– 10). [39] Horpibulsuk S, Suddeepong A, Chinkulkijniwat A, Liu MD. Strength and compressibility of lightweight cemented clays. Appl Clay Sci 2014;69:11–21. [40] Prakash K, Sridharan A. Free swell ratio and clay mineralogy of fine-grained soils. Geotech Test J ASTM 2004;27(2):220–5. [41] Horpibulsuk S, Shibuya S, Fuenkajorn K, Katkan W. Assessment of engineering properties of Bangkok clay. Can Geotech J 2007;44(2):173–87. [42] Horpibulsuk S, Rachan R, Suddeepong A, Liu MD, Du YJ. Compressibility of lightweight cemented clays. Eng Geol 2013;159:59–66. [43] Kampala A, Horpibulsuk S, Prongmanee N, Chinkulkijniwat A. Influence of wet–dry cycles on compressive strength of calcium carbide residue-fly ash stabilized clay. J Mater Civ Eng ASCE 2014;24(1):633–43.