Durability against wetting-drying cycles for cement-stabilized reclaimed asphalt pavement blended with crushed rock

HOSTED BY

Available online at www.sciencedirect.com

ScienceDirect Soils and Foundations xxx (2018) xxx–xxx www.elsevier.com/locate/sandf

Durability against wetting-drying cycles for cement-stabilized reclaimed asphalt pavement blended with crushed rock Apichat Suddeepong a, Artit Intra b, Suksun Horpibulsuk c,⇑, Cherdsak Suksiripattanapong d, Arul Arulrajah e, Jack Shuilong Shen f a

School of Civil Engineering, Suranaree University of Technology, 111 University Avenue, Muang District, Nakhon Ratchasima 30000, Thailand b School of Civil Engineering, Suranaree University of Technology, Nakhon Ratchasima 30000, Thailand c School of Civil Engineering and Center of Excellence in Innovation for Sustainable Infrastructure Development, Suranaree University of Technology, Thailand d Department of Civil Engineering, Faculty of Engineering and Architecture, Rajamangala University of Technology Isan, 744 Suranarai Road, Muang District, Nakhon Ratchasima 30000, Thailand e Department of Civil and Construction Engineering, Swinburne University of Technology, Melbourne, Australia f Department of Civil and Construction Engineering, Swinburne University of Technology, Hawthorn, Victoria 3122, Australia Received 3 March 2017; received in revised form 12 October 2017; accepted 2 December 2017

Abstract Pavement rehabilitation and reconstruction generate large quantities of reclaimed asphalt pavement (RAP). The improvement of the engineering properties of this RAP is required in order to enable it for use as environmentally friendly alternative construction material in road pavements. The durability of RAP when blended with crushed rock (CR) and stabilized with Portland cement was investigated in this paper. The CR replacement was found to improve the compactibility and durability of the stabilized RAP/CR material. For a particular RAP:CR ratio, the compaction curves of cement-stabilized RAP/CR blends were found to be essentially the same for all cement contents, but different for unstabilized blends; i.e., the maximum dry unit weight of cement-stabilized RAP/CR blends is higher than that of unstabilized RAP-CR blends. The wetting-drying (w-d) cycles led to a loss in weight of the cement-stabilized RCA/CR blends and to a subsequent reduction in strength. The w-d cycle strengths (qu(w-d)) for a state of compaction (dry side, wet side or optimum water content) at any w-d cycle could be approximated from the corresponding initial soaked strength (prior to w-d tests) (qu0). The qu0 of cementstabilized RAP/CR blends increased with an increasing CR replacement and an increasing cement content. Assuming that the CR replacement also results in an increasing cement content, w/[C(1 + kCRc)] was proposed as a critical parameter for developing qu0 and qu(w-d) predictive equations where w is the water content at the optimum water content, C is the cement content, k is the replacement efficiency, and CRc is the CR content. Based on the qu(w-d) predictive equation developed here, a design procedure for the laboratory mixing of cement-stabilized RAP/CR blends was proposed, which would be valuable for an accurate determination of the ingredients (RAP:CR ratio and cement content) required to attain the necessary strength at the design service life. Ó 2018 Production and hosting by Elsevier B.V. on behalf of The Japanese Geotechnical Society.

Keywords: Reclaimed asphalt pavement; Crushed rock; Strength; Durability; Cement stabilization

Peer review under responsibility of The Japanese Geotechnical Society. ⇑ Corresponding author at: School of Civil Engineering, Suranaree University of Technology, 111 University Avenue, Muang District, Nakhon Ratchasima 30000, Thailand. E-mail addresses: [email protected] (A. Suddeepong), [email protected] (S. Horpibulsuk), [email protected], [email protected] (C. Suksiripattanapong), [email protected] (A. Arulrajah), [email protected] (J.S. Shen).

1. Introduction Pavement rehabilitation and reconstruction generate large quantities of reclaimed asphalt pavement (RAP). RAP is defined as the damaged pavement materials that are removed and/or reprocessed from an existing asphalt

https://doi.org/10.1016/j.sandf.2018.02.017 0038-0806/Ó 2018 Production and hosting by Elsevier B.V. on behalf of The Japanese Geotechnical Society.

Please cite this article in press as: Suddeepong, A. et al., Durability against wetting-drying cycles for cement-stabilized reclaimed asphalt pavement blended with crushed rock, Soils Found. (2018), https://doi.org/10.1016/j.sandf.2018.02.017

2

A. Suddeepong et al. / Soils and Foundations xxx (2018) xxx–xxx

pavement. Thus, RAP comprises a mixture of pavement aggregates and aged bituminous additives. Typically, an existing pavement is either removed by the milling of the upper surfaces or by a full-depth removal of the entire pavement section itself. A milling machine is used to remove the top 50 mm of the pavement surface in a single pass, whereas a rhino horn on a bulldozer is used for the full-depth removal of the entire pavement in several broken pieces (Viyanant et al., 2007). These pieces are subsequently subjected to crushing, screening, conveying, and stacking in stockpiles, a process that usually occurs at a central processing plant (Mulheron and O’Mahony, 1990). The sustainable usage of RAP leads to significant economical savings for the construction of new highway pavements (Hajj et al., 2010). RAP can be utilized as a construction material in road bases or subbases, asphalt cement binders, asphalt concrete aggregates, and as embankment or fill material (Arulrajah et al., 2013, 2014; Attia, 2010; Cosentino et al., 2003; Maher et al., 1997; Papp et al., 1998). When used as a total substitute for natural aggregates, however, RAP materials often do not meet the minimum base material requirements specified by international and local state road authority guidelines (Rana, 2004; Hoy et al., 2016a, 2016b), and hence, require some form of improvement. Chemical stabilization by Portland cement is widely used in both the construction of new roads and the rehabilitation of damaged roads, as the engineering properties can be improved rapidly after the stabilization of the RAP materials (Horpibulsuk et al., 2006, 2010, 2011, 2012; Shen et al., 2013, 2017; Du et al., 2014; Yoobanpot et al., 2018). The performance of cementstabilized RAP satisfies the requirements of pavement base/subbase applications (Hoyos et al., 2011; Diefenderfer and Apeagyei, 2014; Taha et al., 1999; 2002; Puppala et al., 2011, 2012; Suebsuk et al., 2014). Suebsuk et al. (2017) reported that bitumen, when adhering to RAP particles, retards the cement hydration and results in the need for high-cement contents. To reduce construction costs and to improve the engineering properties of cement-stabilized RAP bases/subbases, RAP should be replaced with a higher quality material. In Thailand, crushed rock (CR) replacement is a preferred option in terms of lower material and haulage costs as compared to Portland cement (Bureau of Trade and Economic Indices, 2016). In other words, replacement with small quantities of CR can reduce the amount of carbon-intensive Portland cement and can also significantly improve the engineering properties of the cement-stabilized RAP base/subbases. The pavement recycling technique to restore damaged pavement has been widely used in Thailand. It involves about 20 cm of pavement (RAP and some base material) being dug up, mixed with CR and cement, and immediately field compacted with rollers. This technique is economical because cement is readily available at a reasonable cost in Thailand (Horpibulsuk et al., 2014). Horpibulsuk et al. (2006) have studied the development of strength of cement-stabilized coarse-grained soils in the laboratory

and in the field. They reported that the field strengths were 0.5–1.0 times lower than the laboratory strengths under the same dry unit weight, soil-water/cement ratio, and curing time. This was due to several field factors, including nonuniformity in the soil mixed with cement and differences in the compaction methods and curing conditions between the laboratory stabilization and the field stabilization. In practice, many laboratory trial mixes are needed to arrive at the proper strength before the execution of the soilcement pavement. This laboratory strength must be high enough to compensate for conditions which are controllable in the field. Horpibulsuk et al. (2006) also suggested a procedure for the pavement recycling technique and recommended a reduction in field strength of 2.0 for determining the cement content from laboratory tests. Dempsey and Thompson (1967) defined durability as the ability of materials to retain their stability and integrity and to maintain an adequate amount of long-term residual strength so as to provide sufficient resistance to climatic conditions. The durability of cement-stabilized RAP/CR blends under severe climatic conditions is a crucial parameter when used in road construction applications. To date, however, there is little knowledge of the durability of cement-stabilized RAP/CR blends. Cyclic wetting-drying (w-d) tests simulate the changes in weather over a geological age, and are considered capable of simulating critical scenarios that can induce damage to pavement material (Allam and Sridharan, 1981; Sobhan and Das, 2007). A durability study on the w-d cycles of chemically stabilized RAP blended with virgin aggregates has been reported by Ganne (2009). Results of the study show that the loss in strength was approximately 10–15% on average after 14 cycles of w-d tests for all the mixes studied. Kampala et al. (2014) investigated the influence of w-d cycles on the durability of clays stabilized with calcium carbide residue (CCR) and fly ash (FA) in pavement applications, and reported that the optimal CCR and FA contents were found to be 7 and 20%, respectively. However, high FA contents were found to result in a reduction in strength. Furthermore, despite the input of FA enhancing the pozzolanic reaction, the strength of the CCRstabilized clay was found to be significantly reduced by an increasing number of w-d cycles. Al-Obaydi et al. (2010) and Al-Zubaydi (2011) reported that the cyclic w-d cycles caused crack propagation, resulting in severe effects on the engineering properties of the materials, particularly in terms of their residual strength and stability. In general, tropical countries, such as Thailand and parts of Australia, are frequently subjected to changes in weather during the wet (rainy) and dry (summer) seasons. Therefore, an investigation of the service life of cementstabilized RAP/CR blends via w-d cycle tests is significant and is the focus of this research. Tests with a wide range of water contents, RAP:CR ratios, and cement contents for cement-stabilized RAP/CR were undertaken to understand the importance of these parameters on the w-d cycle strengths. Based on an analysis of the test results, a rational

Please cite this article in press as: Suddeepong, A. et al., Durability against wetting-drying cycles for cement-stabilized reclaimed asphalt pavement blended with crushed rock, Soils Found. (2018), https://doi.org/10.1016/j.sandf.2018.02.017

A. Suddeepong et al. / Soils and Foundations xxx (2018) xxx–xxx

cyclic wetting-drying strength (qu(w-d)) predictive equation, in terms of the initial soaked strength both without w-d cycles and with a number of w-d cycles, was proposed. It was found that this relationship can facilitate the determination of a suitable mix proportion of cement-stabilized RAP/CR blends to meet the strength requirement at a target service life. The present research will enable traditional RAP, destined for landfills, to be used in a sustainable manner as pavement base/subbase material, which is significant in terms of engineering, economical, and environmental perspectives. 2. Materials and methods 2.1. Reclaimed asphalt pavement Reclaimed asphalt pavement (RAP) was obtained from a pavement maintenance project for Highway No. 2285, Khon Kaen Province, Thailand. The RAP contained more than 90% coarse aggregates, had no plasticity, and was comprised of approximately 5.5% asphalt content by weight of aggregates. The bulk-specific density of RAP was 2.67. The particle-size distribution curve of RAP met the gradation requirements for cement-stabilized crushed rock base (DOH, 2013), as shown in Fig. 1. The RAP can be classified as a well-graded gravel (GW) according to the Unified Soil Classification System (USCS). The maximum dry unit weight (cd,max) and the optimum water content (OWC) of the compacted RAP under modified Proctor energy were 20.2 kN/m3 and 9.8%, respectively. The California bearing ratio (CBR) value under soaked conditions was 56.4%, which is lower than the requirement for unbound base materials (CBR > 80%) specified by the Department of Highways, Thailand (DOH, 2013). 2.2. Crushed rock Crushed rock (CR) was collected from a quarry in Nakhon Ratchasima Province, Thailand. The particle size distribution curve of the CR is shown in Fig. 1. The CR

Fig. 1. Grain size distribution of CR, RAP, and RAP-CR blends.

3

contained more than 90% coarse aggregates, had no plasticity, and can be classified as a well-graded gravel (GW) according to the Unified Soil Classification System (USCS). The bulk-specific gravity values of the coarse particles and the fine particles were 2.74 and 2.78, respectively. The compaction curve of the CR is compared to that of the RAP in Fig. 2. The cd,max and the OWC of the CR were higher and lower, respectively, than those of the RAP. 2.3. RAP/CR blends The studied RAP:CR ratios in the RAP/CR blends were 80:20, 60:40, 40:60, and 20:80. The gradation of all the tested blends met the requirement for base material (DOH, 2013), as shown in Fig. 1. Compaction curves of the RAP, CR, and RAP/CR blends under modified Proctor energy are shown in Fig. 2. The test results indicate that the RAP had the lowest maximum dry unit weight (cd,max), while the CR had the highest value. The RAP was found to have the highest OWC, while the CR had the lowest value. The cd,max of the RAP/CR blends tended to increase as the RAP:CR ratio decreased, while the OWC decreased as the RAP:CR ratio increased. These results indicate that the CR improved the compactibility of the RAP by improving the gradation of the RAP. The flatter compaction curves of the RAP/CR blends suggest their low sensitivity to changes in water content in comparison to the CR; i.e., the RAP/ CR blends exhibited more stable compaction behavior and better workability (Arulrajah et al., 2016). The soaked CBR and Los Angeles abrasion (LA) for different RAP:CR ratios are shown in Table 1. It is evident that the RAP/CR blends at different RAP:CR ratios cannot be used as unbound base material, except for the RAP/CR blend with RAP:CR = 20:80. 2.4. Cement-stabilized RAP/CR blends The studied RAP:CR ratios were 80:20, 60:40, 40:60, and 20:80. The field strength of cement-stabilized base material must be high enough to sustain traffic loads, but not be too high so as to avoid cracks due to changes in weather which can cause damage to the pavement. Little

Fig. 2. Compaction curves of CR, RAP, and CR-RAP blends.

Please cite this article in press as: Suddeepong, A. et al., Durability against wetting-drying cycles for cement-stabilized reclaimed asphalt pavement blended with crushed rock, Soils Found. (2018), https://doi.org/10.1016/j.sandf.2018.02.017

4

A. Suddeepong et al. / Soils and Foundations xxx (2018) xxx–xxx

Table 1 Classification, compaction characteristics, CBR and LA of CR, RAP, and RAP-CR blends. Materials

Classification (USCS)

OWC (%)

cdmax (kN/m3)

CBR (%)

LA (%)

CR RAP RAP:CR = 80:20 RAP:CR = 60:40 RAP:CR = 40:60 RAP:CR = 20:80

GW GW GW SW SW SW

5.4 9.8 8.0 7.0 6.3 6.1

23.90 20.20 21.80 22.07 22.76 23.05

90.4 56.4 60.1 64.7 79.4 85.1

20.8 – 28.54 26.87 25.74 25.72

and Nair (2009) recommend a required 28-day field strength range of 2758–6895 kPa, while the Department of Highways recommends a required 7-day field strength range of 1724–2413 kPa for low traffic volume and high traffic volume, respectively (DRR, 2013; DOH, 2013). Based on the field to laboratory ratio of 0.5, suggested by Horpibulsuk et al. (2006) for Thailand’s road construction, the 7-day laboratory strength should be between 3448 and 4826 kPa to meet the Department of Highways standards, while the 28-day laboratory strength should be between 5516 and 13,790 kPa to meet Little and Nair (2009)’s recommendation. As such, in the research, the cement contents should be between 3 and 7% by weight of the total aggregates in order to cover the range in strength criteria. Compaction tests were performed under modified Proctor energy in a 100-mm standard mold. Unconfined compression (UC) samples were then prepared in a standard cylindrical mold at the OWC and cd,max. After 24 h of curing, the UC samples were dismantled from the mold and wrapped in vinyl bags. All UC samples were stored in a controlled chamber, where the humidity and temperature were kept constant. UC tests were undertaken on the samples with a rate of vertical displacement of 1 mm/min after 7 and 28 days of curing to determine the initial strength (without w-d cycles), qu0. For durability tests, cement-stabilized RAP/CR samples were prepared at OWC of 0.8, 1.0, and 1.2 and cured for 28 days. The method of the wet-dry (w-d) tests, as per ASTM 599-03 (ASTM, 2003), was adopted for the sample preparation. The samples were submerged in deionized water at room temperature for 5 h. They were then dried in an 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 number of w-d cycles, the samples were immersed in deionized water for 2 h at a constant temperature of 25 ± 2 °C. In this study, the UC tests were undertaken on the samples after 0, 1, 3, and 6 w-d cycles. 3. Results and discussion 3.1. Compaction behavior of cement-stabilized RAP/CR blends Compaction curves for cement-stabilized RAP/CR blends at various cement contents and RAP:CR ratios

are shown in Fig. 3. These curves are typical for compacted soils in that the dry unit weight increases with an increasing water content until the maximum dry unit weight is reached at an OWC. Beyond this OWC, the dry unit weight decreases as the water content increases. For a given RAP: CR ratio, the cement-stabilized RAP/CR blends exhibit higher dry unit weight than the unstabilized RAP/CR blends. The cd,max and OWC relationship is essentially the same for various cement contents, i.e., increasing the cement content does not affect the compaction curve. This result is consistent with the studies reported previously by Horpibulsuk et al. (2006) for coarse-grained soils. For a given cement content, the cd,max increases with an increasing CR replacement ratio. 3.2. Unconfined compressive strength of cement-stabilized RAP/CR blends Fig. 4 shows the relationship between the 7-day qu and the RAP:CR ratio of cement-stabilized RAP/CR blends at the OWC. The 7-day qu increases as the RAP:CR ratio decreases at the same cement content. For a particular RAP:CR ratio, the value of the 7-day qu increases as the cement content increases, which is consistent with the results of cement-stabilized coarse-grained soils reported by Horpibulsuk et al. (2006). 3.3. Durability of cement-stabilized RAP/CR blends Figs. 5 and 6 show the relationships between the wetting-drying (w-d) cycle strength, qu(w-d), versus the number of w-d cycles, N, and the weight loss versus the number of w-d cycles, N, respectively, of the cement-stabilized RAP/CR blends at OWC of 0.8, 1.0, and 1.2, various cement contents, and various RAP:CR ratios for 28 days of curing. Fig. 5 indicates that for a particular cement content, water content, and number of w-d cycles, the qu(w-d) of the cement-stabilized RAP/CR blends increases with a decreasing RAP:CR ratio. The OWC provides the highest strengths for all CR replacement contents and cement contents, and is followed by 0.8OWC and 1.2OWC, respectively. The strengths decrease significantly with the number of cycles, which is similar to the results reported by Neramitkornburi et al. (2015) for lightweight cellularcemented clay and by Kampala et al. (2014) for CCRstabilized clay. Fig. 6 indicates that the weight loss of the cement-stabilized RAP/CR blends increases with an increase in N, which is associated with the decrease in compressive strength of the cement-stabilized RAP/CR blends. The weight loss increases sharply within the 1st w-d cycle, and thereafter, gradually increases with an increase in N. The weight loss during the wetting-drying cycles is caused by crack propagation. In the wetting process, the available pore space is filled with water, thus increasing the volume of samples, similar to what was reported in the results by Aldaood et al. (2014) for lime-stabilized soil. At the drying

Please cite this article in press as: Suddeepong, A. et al., Durability against wetting-drying cycles for cement-stabilized reclaimed asphalt pavement blended with crushed rock, Soils Found. (2018), https://doi.org/10.1016/j.sandf.2018.02.017

A. Suddeepong et al. / Soils and Foundations xxx (2018) xxx–xxx

5

Fig. 3. Compaction curves of cement stabilized RAP-CR blends.

stage, the samples exhibit cracks as a result of shrinkage. As the water content decreases due to drying, the suction forces increase until the tensile stress is equal to the cohesion force. At this stage, the formation of cracks results in further shrinkage. With further water loss, the primary cracks grow in size and new cracks are formed. The external surface of the soil samples will be more affected as a faster water loss occurs at the surface rather than inside the sample. The cracking during the wetting-drying cycles is an irreversible process. Macro-cracks, developing with an increasing N, lead to loss in weight and a reduction in strength of the stabilized materials (Al-Obaydi et al., 2010; Al-Zubaydi, 2011; Hoy et al., 2017; Du et al., 2016). It is evident from Fig. 6 that the weight loss at a particular w-d cycle decreases as the RAP:CR ratio decreases

and the cement content increases. For a particular cement content and RAP:CR ratio, the weight loss at the OWC is the lowest, which is associated with the highest compressive strength when compared to that at 0.8 and 1.2 times OWC. The weight loss at 6 w-d cycles, for all the samples, is lower than 14%, which is the limit value for the cement-stabilized coarse-grained soils recommended by the Portland Cement Association (PCA, 1992). It is evident from the durability test results that the qu(wd) value at different N is dependent upon the initial soaked (without w-d cycles) strength (qu0) value. As such, normalized strength qu(w-d)/qu0 is used as a variable for analyzing the reduction in strength due to the w-d cycles, as was previously done by Kampala et al. (2014) for clay stabilized with calcium carbide residue (CCR), by Neramitkornburi

Please cite this article in press as: Suddeepong, A. et al., Durability against wetting-drying cycles for cement-stabilized reclaimed asphalt pavement blended with crushed rock, Soils Found. (2018), https://doi.org/10.1016/j.sandf.2018.02.017

6

A. Suddeepong et al. / Soils and Foundations xxx (2018) xxx–xxx

shown in Fig. 7. This relationship is the same for different cement contents and RAP:CR ratios, and is expressed by the following logarithm function: quðwdÞ =qu0 ¼ a  b lnðN Þ

Fig. 4. Relationship between 7-days unconfined compressive strength (qu,7 days) and RAP:CR ratio.

et al. (2015) for lightweight cellular-cemented clay, and by Horpibulsuk et al. (2015) for water treatment sludge-fly ash geopolymer. The relationship between qu(w-d)/qu0 and N is

for 1 6 N 6 6

ð1Þ

where a and b are constants, which are dependent upon the water content. From an analysis of the test data, the values for a and b are 0.927 and 0.066 for 0.8OWC, 0.948 and 0.061 for OWC, and 0.928 and 0.073 for 1.2OWC. The variation in parameter b is dependent on the state of compaction (dry side or wet side or OWC), while the a value is almost constant and can be taken as 0.930. The b value for the wet side is the highest; i.e., the reduction in strength of the blends on the wet side of optimum is the highest, which is in agreement with the highest weight loss. By using Eq. (1), the qu(w-d) of cement-stabilized RAP/CR blends can be approximated at various RAP:CR ratios and cement contents once the corresponding qu0 is known. The qu0 predictive equation was developed at the OWC, which provides the highest compressive strength and the

Fig. 5. Relationship between w-d cycle strength and number of w-d cycles at various water content, cement content, and RAP:CR ratio. Please cite this article in press as: Suddeepong, A. et al., Durability against wetting-drying cycles for cement-stabilized reclaimed asphalt pavement blended with crushed rock, Soils Found. (2018), https://doi.org/10.1016/j.sandf.2018.02.017

A. Suddeepong et al. / Soils and Foundations xxx (2018) xxx–xxx

7

Fig. 6. Relationship between weight loss and number of w-d cycles at various water content, cement content, and RAP:CR ratio.

lowest weight loss. Miura et al. (2001) and Horpibulsuk et al. (2005) proposed a parameter for governing the strength development in cement-stabilized clay at high water contents based on Abrams’s law (1918). This was designated as the water/cement ratio, w/C, which is the ratio of the initial water content of the material (%) to the cement content (%). The cement content, C, is the ratio of the dry weight of the cement to the dry weight of the material. Horpibulsuk et al. (2006) and Chinkulkijniwat and Horpibulsuk (2012) extended this parameter to analyze the strength development in the cement-stabilized coarsegrained soils compacted at OWC and on the wet side of optimum under various compaction energies. As shown in Fig. 4, the 7-day qu increases linearly when the CR replacement content increases (RAP:CR ratio decreases) at the same cement content. The same tendency was observed for all the cement contents tested. These

results indicate that the unconfined compressive strength of cement-stabilized RAP/CR blends is controlled by the CR replacement content and the cement content. Therefore, the contribution of the CR replacement is regarded as being akin to the addition of cement. The CR content, CRc is herein proposed for representing the weight of CR (WCR) with respect to the total dry weight of the blend (WCR+WRAP). CRc ¼ W CR =ðW CR þ W RAP Þ

ð2Þ

By assuming that the CR replacement also results in an increase in cement content, parameter w/C is thus modified to be w/[C(1 + kCRc)], where k is an empirical material constant. The results in Fig. 4, which are the relationship between qu0 and the RAP:CR ratio at 7 days of curing at the OWC, were reanalyzed using w/[C(1 + kCRc)] and are shown in Fig. 8. It is noted that the effect of the cement

Please cite this article in press as: Suddeepong, A. et al., Durability against wetting-drying cycles for cement-stabilized reclaimed asphalt pavement blended with crushed rock, Soils Found. (2018), https://doi.org/10.1016/j.sandf.2018.02.017

8

A. Suddeepong et al. / Soils and Foundations xxx (2018) xxx–xxx

Fig. 8. Relationship between qu and w/[C(1 + kCRc)].

qðw=CÞ1 A=½w=½Cð1 þ kCRc ÞB1 ¼ qðw=CÞ2 A=½w=½Cð1 þ kCRc ÞB2  0:69 w=½Cð1 þ 5:5CRc Þ2 ¼ w=½Cð1 þ 5:5CRc Þ1 Fig. 7. Relationship between qu(w-d)/qu0 and number of cycles.

content, CR replacement, and water content can be handled by w/[C(1 + kCRc)]. The relationship between qu0 and w/[C(1 + kCRr)] is expressed as qu0 ¼ A=½w=½Cð1 þ kCRc Þ

B

ð3Þ

where A and B are empirical material constants and w is the water content at the OWC of different mixtures. qu0 and A are expressed in the same unit. B and k are dimensionless constants. Parameter k is identified as a replacement efficiency for improving the strength of the cementstabilized RAP/CR blends. The three constants, A, B, and k, can be determined using a multi-regression analysis (MRA). It is noted that Eq. (3) is similar to the equation developed for cement-stabilized coarse-grained soils by Horpibulsuk et al. (2006) when CRc = 0 (no CR replacement). From the analysis of the test data, the value of parameter A is dependent on the curing time, while the value of parameter B can be taken as 0.69. This B value is close to the range between 0.65 and 0.66 reported by Horpibulsuk et al. (2006) for cement-stabilized coarsegrained soils. The k value is 5.5 for the tested cementstabilized RAP/CR blends. Parameter A in Eq. (3) can be eliminated by taking the ratio of the strength developing at two w/[C(1 + kCRc)] values. This is based on the fact that parameter A has a specific value at the same curing time. Taking the analyzed parameters, B = 0.69 and k = 5.5, results in the following relation:

ð4Þ

where qðw=CÞ1 is the strength to be estimated at the [w/[C(1 + kCRc)]]1 value and qðw=CÞ2 is the strength to be estimated at the [w/[C(1 + kCRc)]]2 value. From this equation, the strength development at various w/[C(1 + kCRr)] values of the cement-stabilized RAP/CR blends at the OWC can be estimated from a single laboratory strength test value determined at a specific curing time and at any w/[C(1 + kCRr)] value. To validate the applicability of Eq. (4), the strength of the cement-stabilized RAP/CR blends at 28 days of curing was predicted using Eq. (4) and is presented in Table 2 and Fig. 9. It is found that the predicted strengths are very close to the laboratory strengths and that all the strength ratios (predicted to measured strength, qup/ qul) are between 0.8 and 1.2, as shown in Fig. 9. This error from the strength prediction is found to be within acceptable limits for road construction practice. Even though the formulation for the model is based on particular RAP and CR, the development is founded on sound principles and can possibly be applied for other sources of materials. The role of the material source on the model parameters and the determination of the generalized model parameters are extensive enough for a separate future research project. 4. Proposed laboratory mix design procedure Recently, the use of cement stabilization has been successfully applied in full-depth reclamation (FDR) processes, which consist of pulverizing existing asphalt pavement and underlying base layers (Diefenderfer and Apeagyei, 2014). However, those studies have not considered the effect of w-d cycles on the target strength of the cement stabilization. Based on an analysis of the laboratory test results, a mix

Please cite this article in press as: Suddeepong, A. et al., Durability against wetting-drying cycles for cement-stabilized reclaimed asphalt pavement blended with crushed rock, Soils Found. (2018), https://doi.org/10.1016/j.sandf.2018.02.017

A. Suddeepong et al. / Soils and Foundations xxx (2018) xxx–xxx

9

Table 2 Strength prediction of cement stabilized RAP-CR blends. Curing time, D (days)

Water content, w (%)

Cement content, C (%)

RAP content (%)

CR replacement content, CRr (%)

w/[C(1 + kCRr)]

Laboratory strength, qul (kPa)

Predicted strength, qup (kPa)

28 28 28 28 28 28 28 28 28 28 28 28

7.28 6.61 6.34 5.71 6.98 6.7 6.23 5.54 6.74 6.62 6.08 5.43

3 3 3 3 5 5 5 5 7 7 7 7

80 60 40 20 80 60 40 20 80 60 40 20

20 40 60 80 20 40 60 80 20 40 60 80

1.16 0.69 0.49 0.35 0.66 0.42 0.29 0.21 0.46 0.30 0.20 0.14

3752 4839 5912 7622 5455 6919 8377 9913 6954 8472 10,589 12,878

3434 4909 6195 7793 5030 Reference 8920 11,319 6499 8800 11,442 14,476

(4) Using Eq. (4), develop relationships between qu0 and the RAP:CR ratio at various cement contents, similar to Fig. 4. (5) From the required laboratory strength in Step (3), determine the cost-effective RAP:CR ratio and the cement content based on the transportation and material costs of RAP, CR, and cement and on the relationships developed in Step (4). (6) Conduct the compaction and UC tests (e.g., ASTM D5102 (ASTM, 2009)) on the selected cement content and RAP:CR ratio to determine the actual value.

Fig. 9. Comparison between laboratory and predicted strengths.

design procedure to arrive at the target qu(w-d) of the cement-stabilized RAP/CR blends at the OWC is proposed and presented by the following steps:

The RAP should be mixed with water, CR, and cement at the OWC using the pavement recycling machine based on the laboratory mix design outcomes. The mixture must be compacted by rollers to attain cd,max of at least 95%. The cored samples from the construction site should be taken to check the uniformity of the mixing, and unconfined compressive strength tests should be undertaken every 500 m after 28 days of curing according to the specifications of the Department of Highways. 5. Conclusions

(1) Perform the compaction test on cement-stabilized RAP/CR blends at any cement content and RAP: CR ratio. Determine the OWC and cd,max from the compaction curve. Conduct the compaction test in accordance with ASTM D1557 (ASTM, 2012). (2) Perform the cyclic wetting–drying test on the cementstabilized RAP-CR blends at the cement content and RAP:CR ratio at the OWC which is obtained from Step (1) to determine the values for parameters a and b in Eq. (1). Cure the samples for 28 days as recommended by ASTM 599-03 (ASTM, 2003). (The a and b values can also be taken as 0.930 and 0.061 when there is no test data.) (3) From Eq. (1), determine qu0 for the target qu(w-d) at the required N. Non-uniformity in mixing the RAP/ CR with the cement in the field can be considered using the field strength to laboratory strength ratio of 0.5 as recommended by Horpibulsuk et al. (2006).

This research was undertaken to investigate the durability against wetting and drying (w-d) cycles of cementstabilized RAP-CR blends at various RAP:CR ratios and cement contents, and also to develop a w-d cycle strength (qu(w-d) predictive equation. The following conclusions can be drawn from this study: (1) The compactibility of RAP was improved by CR replacement. The compaction curves of the cementstabilized RAP/CR blends at a particular RAP:CR ratio were similar for all cement contents. The maximum dry unit weight of the stabilized RAP/CR blends was higher than that of the unstabilized RAP-CR blends at the same RAP:CR ratio. (2) The weight loss of all the cement-stabilized RAP/CR samples increased as the number of w-d cycles increased, which was associated with the reduction

Please cite this article in press as: Suddeepong, A. et al., Durability against wetting-drying cycles for cement-stabilized reclaimed asphalt pavement blended with crushed rock, Soils Found. (2018), https://doi.org/10.1016/j.sandf.2018.02.017

10

A. Suddeepong et al. / Soils and Foundations xxx (2018) xxx–xxx

in unconfined compressive strength with the w-d cycles. The logarithmic relationship between the w-d strength, qu,w-d, and the number of w-d cycles, N, was proposed. The initial strength (without w-d cycles), qu0, of the cement-stabilized RAP/CR blends was found to control qu(w-d). The CR replacement also resulted in an increase in cement content. Parameter w/[C (1 + kCRc)] was thus proposed to take into account the effect of the CR replacement on the strength development for developing the qu0 predictive equation. (3) From the critical analysis, a laboratory mix design procedure was proposed to ascertain the serviceability of cement-stabilized RAP/CR blends. The proposed relationship facilitates the determination of the proper mix design in order to attain the required strength at a target service life. This is very useful for civil engineering practitioners and researchers since the durability test is time-consuming.

Acknowledgements The authors are grateful to the Thailand Research Fund under the TRF Senior Research Scholar Program Grant No. RTA5980005 and Suranaree University of Technology. References Abrams, D.A., 1918. Design of Concrete Mixtures. Structural Materials Research Laboratory, Lewis Institute, Chicago. Aldaood, A., Bouasker, M., Al-Mukhtar, M., 2014. Impact of wettingdrying cycles on the microstructure and mechanical properties of limestabilized gypseous soils. Eng. Geol. 174, 11–21. Allam, M.M., Sridharan, A., 1981. Effect of wetting and drying on shear strength. J. Geotech. Eng. Div. 107, 421–438. Al-Obaydi, M., Al-Kiki, I., Al-Zubaydi, A., 2010. Strength and durability of gypseous soil treated with waste lime and cement. J. Al-Rafidain Eng. 18, 28–42. Al-Zubaydi, A., 2011. Effect of static soaking under different temperatures on the lime stabilized gypseous soil. Tikrit J. Eng. Sci. 18, 42–51. Arulrajah, A., Piratheepan, J., Disfani, M.M., 2013. Reclaimed asphalt pavement and recycled concrete aggregate blends in pavement subbases: laboratory and field evaluation. J. Mater. Civ. Eng. 26, 349–357. Arulrajah, A., Disfani, M.M., Horpibulsuk, S., Suksiripattanapong, C., Prongmanee, N., 2014. Physical properties and shear strength responses of recycled construction and demolition materials in unbound pavement base/subbase applications. Constr. Build. Mater. 58, 245–257. Arulrajah, A., Horpibulsuk, S., Maghool, F., 2016. Recycled construction and demolition materials in pavement and footpath bases. In: The 6th International Symposium on Rural Roads, Bangkok, Thailand. ASTM C599-03, 2003 Standard test methods for wetting and drying compacted soil-cement mixtures, West Conshohcken, PA. ASTM D1557-12e1, 2012. Standard Test Methods for Laboratory Compaction Characteristics of Soil Using Modified Effort (56000 ftlbf/ft 3 (2700 kN-m/m3)), West Conshohcken, PA. ASTM D5102-09, 2009. Standard Test Method for Unconfined Compressive Strength of Compacted Soil-Lime Mixtures, West Conshohcken, PA.

Attia, M., 2010. Characterization of the structural behavior of reclaimed asphalt pavement as pavement base layer. Doctoral Dissertation. North Dakota State University of Agriculture and Applied Science. Bureau of Trade and Economic Indices, 2016. Construction Material Prices in Bangkok, Thailand (February 06, 2016). Chinkulkijniwat, A., Horpibulsuk, S., 2012. Field strength development of repaired pavement using the recycling technique. Quart. J. Eng. Geol. Hydrogeol. 45 (2), 221–229. Cosentino, P., Kalajian, E., Shieh, C., Mathurin, W., Gomez, F., Cleary, E., 2003. Developing specifications for using recycled asphalt pavement as base, subbase or general fill materials, phase II. Dempsey, B.J., Thompson, M.R., 1967. Durability properties of lime-soil mixtures. Highway Research Record No. 235. HRB. National Research Council, Washington D.C. 61–75. Diefenderfer, B.K., Apeagyei, A.K., , 2014. I-81 In-Place Pavement Recycling Project, Report No. FHWA/VCTIR 15–R1, Virginia Department of Transportation, Richmond, VA. DOH, DH-S 203, 2013. Standard of cement stabilized crushed rock base. Department of Highways, Thailand. DRR, DRR244-2013, 2013. Standard of Soil Cement Base. Department of Rural Roads, Thailand. Du, Y.J., Horpibulsuk, S., Wei, M.L., Suksiripattanapong, C., Liu, M.D., 2014. Modeling compression behavior of cement treated zinc contaminated clayey soils. Soils Found. 54 (5), 1018–1026. Du, Y.J., Bo, Y.L., Jin, F., Liu, C.Y., 2016. Durability of reactive magnesia-activated slag-stabilized low plasticity clay subjected to drying–wetting cycle. Eur. J. Environ. Civ. Eng. 20, 215–230. Ganne, V.K., 2009. Long-Term Durability Studies On Chemically Treated Reclaimed Asphalt Pavement (RAP) Materials MS Thesis. The University of Texas, Arlington. Hajj, E., Sebaaly, P., Kandiah, P., 2010. Evaluation of the use of reclaimed asphalt pavement in airfield HMA pavements. J. Transp. Eng. 136, 181–189. Horpibulsuk, S., Miura, N., Nagaraj, T., 2005. Clay-water/cement ratio identity for cement admixed soft clays. J. Geotech. Geoenviron. Eng. 131, 187–192. Horpibulsuk, S., Katkan, W., Sirilerdwattana, W., Rachan, R., 2006. Strength development in cement stabilized low plasticity and coarse grained soils: laboratory and field study. Soils Found. 46 (3), 351– 366. Horpibulsuk, S., Rachan, R., Chinkulkijniwat, A., Suddeepong, A., 2010. Analysis of strength development in cement-stabilized silty clay based on microstructural considerations. Constr. Build. Mater. 24 (10), 2011–2021. Horpibulsuk, S., Rachan, R., Suddeepong, A., 2011. Assessment of strength development in blended cement admixed Bangkok clay. Constr. Build. Mater. 25 (4), 1521–1531. Horpibulsuk, S., Phetchuay, C., Chinkulkijniwat, A., 2012. Soil stabilization by calcium carbide residue and fly ash. J. Mater. Civ. Eng., DOI: http://doi.org/10.1061/(ASCE)MT.1943-5533.0000370, 184–193. Horpibulsuk, S., Chinkulkijniwat, A., Suddeepong, A., Neramitkornburee, A., Suksiripattanapong, C., 2014. Cement stabilization for pavement material in Thailand. Geotech. Eng. J. SEAGS & AGSSEA 45 (1), 95–102. Horpibulsuk, S., Suksiripattanapong, C., Samingthong, W., Rachan, R., Arulrajah, A., 2015. Durability against wetting-drying cycles of water treatment sludge-cement and silty clay-cement system. J. Mater. Civ. Eng. 04015078. Hoy, M., Horpibulsuk, S., Arulrajah, A., 2016a. Strength development of recycled asphalt pavement-fly ash geopolymer as a road pavement material. Constr. Build. Mater. 117, 209–219. Hoy, M., Horpibulsuk, S., Rachan, R., Chinkulkijniwat, A., Arulrajah, A., 2016b. Recycled asphalt pavement-fly ash geopolymers as a sustainable pavement base material: strength and toxic investigations. Sci. Total Environ. 573, 19–26.

Please cite this article in press as: Suddeepong, A. et al., Durability against wetting-drying cycles for cement-stabilized reclaimed asphalt pavement blended with crushed rock, Soils Found. (2018), https://doi.org/10.1016/j.sandf.2018.02.017

A. Suddeepong et al. / Soils and Foundations xxx (2018) xxx–xxx Hoy, M., Rachan, R., Horpibulsuk, S., Arulrajah, A., Mirzababaei, M., 2017. Effect of wetting-drying cycles on compressive strength and microstructure of recycled asphalt pavement-fly ash geopolymer. Constr. Build. Mater. 144, 624–634. Hoyos, L., Puppala, A., Ordonez, C., 2011. Characterization of cementfiber-treated reclaimed asphalt pavement aggregates: preliminary investigation. J. Mater. Civ. Eng. 23, 977–989. Kampala, A., Horpibulsuk, S., Prongmanee, N., Chinkulkijniwat, A., 2014. Influence of wet-dry cycles on compressive strength of calcium carbine residue-fly ash stabilized clay. J. Mater. Civ. Eng. 26, 633–643. Little, D., Nair, S., 2009. Recommended Practice for Stabilization of Subgrade Soils and Base Materials, NCHRP Web-Only Document 144. Transportation Research Board of the National Academies, Washington, DC. Maher, M.H., Gucunski, N., Papp, W., 1997. Recycled asphalt pavement as a base and sub-base material. ASTM Spec. Tech. Publ. 1275, 42–53. Miura, N., Horpibulsuk, S., Nagaraj, T.S., 2001. Engineering behavior of cement stabilized clay at high water content. Soils Found. 41 (5), 33– 45. Mulheron, M., O’Mahony, M.M., 1990. Properties and performance of recycled aggregates. Highways Transport. 37 (2), 35–37. Neramitkornburi, A., Horpibulsuk, S., Shen, S.L., Chinkulkijniwat, A., Arulrajah, A., Disfani, M.M., 2015. Durability against wetting-drying cycles of sustainable lightweight cellular cemented construction material comprising clay and fly ash wastes. Constr. Build. Mater. 77, 41– 49. Papp, W.J., Maher, M.H., Bennet, T.A., Gucunski, N., 1998. Behavior of construction and demolition debris in base and subbase applications, Recycled Materials in Geotechnical Applications, ASCE Geotechnical Special Publication. Portland Cement Association (PCA), 1992. Soil-cement laboratory handbook, Portland Cement Association, Skokie, IL. Puppala, A., Hoyos, L., Potturi, A., 2011. Resilient moduli response of moderately cement-treated reclaimed asphalt pavement aggregates. J. Mater. Civ. Eng. 23, 990–998.

11

Puppala, A., Saride, S., Williammee, R., 2012. Sustainable reuse of limestone quarry fines and RAP in pavement base/subbase layers. J. Mater. Civ. Eng. 24, 418–429. Rana, A.S.M.A., 2004. Evaluation of recycled material performance in highway applications and optimization of their use Ph.D. dissertation. Texas Tech Univ., Lubbock, TX, p. 170. Shen, S.L., Wang, Z.F., Horpibulsuk, S., Kim, Y.H., 2013. Jet grouting with a newly developed technology: the twin-jet method. Eng. Geol. 152 (1), 87–95. Shen, S.L., Wamg, Z.F., Cheng, W.F., 2017. Estimation of lateral displacement induced by jet grouting in clayey soil. Geotechniques 67 (7), 621–630. Sobhan, K., Das, B.M., 2007. Durability of soil–cements against fatigue fracture. J. Mater. Civil Eng. 19, 26–32. Suebsuk, J., Suksan, A., Horpibulsuk, S., 2014. Strength assessment of cement treated soil-reclaimed asphalt pavement (RAP) mixture. Int. J. GEOMATE 6, 878–884. Suebsuk, J., Horpibulsuk, S., Suksan, A., Suksiripattanapong, C., Phoongernkham, T., Arulrajah, A., 2017. Strength prediction of cement stabilised reclaimed asphalt pavement and lateritic soil blends. Int. J. Pavement Eng. https://doi.org/10.1080/10298436.2017.1293265. Taha, R., Ali, G., Basma, A., Al-Turk, O., 1999. Evaluation of reclaimed asphalt pavement aggregate in road bases and subbases. Transp. Res. Rec. J. Transp. Res. Board 1652, 264–269. Taha, R., Ai-Harthy, A., Ai-Shamsi, K., 2002. Cement stabilization of reclaimed asphalt pavement aggregate for road bases and subbases. J. Mater. Civ. Eng. 14, 239–245. Viyanant, C., Rathje, E.M., Rauch, A.F., 2007. Creep of compacted recycled asphalt pavement. Can. Geotech. J. Natil Res. Council Canada 44 (6), 687–697. Yoobanpot, N., Jamsawang, P., Horpibulsuk, S., 2018. Strength behavior and microstructural characteristics of soft clay stabilized with cement kiln dust and fly ash residue. Appl. Clay. Sci (6).

Please cite this article in press as: Suddeepong, A. et al., Durability against wetting-drying cycles for cement-stabilized reclaimed asphalt pavement blended with crushed rock, Soils Found. (2018), https://doi.org/10.1016/j.sandf.2018.02.017