Use of crushed clay brick and pumice aggregates in lightweight geopolymer concrete

Construction and Building Materials 188 (2018) 1025–1034 Contents lists available at ScienceDirect Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat Use of crushed clay brick and pumice aggregates in lightweight geopolymer concrete Ampol Wongsa a, Vanchai Sata a,⇑, Peem Nuaklong a, Prinya Chindaprasirt a,b a b Sustainable Infrastructure Research and Development Center, Department of Civil Engineering, Faculty of Engineering, Khon Kaen University, Khon Kaen 40002, Thailand The Academy of Science, The Royal Society of Thailand, Dusit, Bangkok, Thailand h i g h l i g h t s  Lightweight fly ash geopolymer concrete (LWGC) was studied.  Crushed clay brick (CA) and pumice aggregates (PA) were used for LWGC aggregate.  The results of natural aggregate (NA) LWGC were compared. 3  CA and PA LWGC yielded 2.7–18.3 MPa compressive strength with 1011–1749 kg/m .  And theirs exhibit better thermal insulation and fire resistance than that of NA. a r t i c l e i n f o Article history: Received 9 April 2018 Received in revised form 17 August 2018 Accepted 27 August 2018 Keywords: Crushed clay brick Pumice Fly ash Geopolymer Lightweight concrete a b s t r a c t The study aimed to examine the properties of lightweight high-calcium fly ash geopolymer concrete (LWGC) containing crushed clay brick and pumice aggregates. A high-calcium fly ash activated by sodium silicate and sodium hydroxide solutions was used as the geopolymer binder. The properties of LWGCs including workability, compressive strength, splitting tensile strength, surface abrasion resistance, density, thermal conductivity, ultrasonic pulse velocity, and fire resistance were investigated and compared with those of the normal density control geopolymer concrete. Temperatures corresponding to 400 °C, 600 °C, and 800 °C were used to test the fire resistance of the concretes. The results indicated that both crushed clay brick and pumice LWGCs exhibited better thermal insulation and fire resistance characteristics when compared to that of the geopolymer containing natural aggregates (CGCs). The results suggested that the LWGCs produced with crushed clay brick aggregate are suitable for structural lightweight concrete. With respect to LWGCs produced with pumice aggregate, the compressive strength was significantly lower, and is sutiable for the manufacture of concrete blocks. Ó 2018 Elsevier Ltd. All rights reserved. 1. Introduction Lightweight aggregate concrete is available in several parts of the world due to its high strength to weight ratio, good sound insulation characteristics, and low thermal conductivity coefficient given the present of voids in the lightweight aggregate. It can be used in a wide range of densities and suitable strengths for various concrete product applications such as roof decks, wall panels, masonry blocks or brick, and precast concrete units [1]. The use of lightweight aggregate concrete reduces the dead load of structure to produce lighter and smaller structures and also reduces the cost of construction [2,3]. The use of lightweight aggregate ⇑ Corresponding author. E-mail address: vancsa@kku.ac.th (V. Sata). https://doi.org/10.1016/j.conbuildmat.2018.08.176 0950-0618/Ó 2018 Elsevier Ltd. All rights reserved. concrete is also preferred, especially for structures built in seismic zones due to its reduced weight [1]. Several researchers indicated that lightweight aggregate from natural sources (for e.g., perlite, diatomite [4], and pumice [3,5]) or from manufactured and recycled materials (for e.g., coal bottom ash, lightweight waste-based geopolymer aggregates [6], recycled crushed clay brick [7,8], and recycled lightweight concrete aggregate [3]) are used to produce lightweight Portland cement-based concrete. The lightweight aggregate concrete is also investigated with a new binder, namely geopolymer. Geopolymer is an alternative cement binder synthesised by mixing aluminosilicate material and high alkaline solutions [9] and is generally expected to provide good fire resistance due to its ceramic-like properties. It utilizes by-products including coal fly ash [10,11] or bottom ash [12,13], rice husk ash [14,15], and ground granulated blast-furnace slag [16–18] as aluminosilicate sources to react with high alkaline 1026 A. Wongsa et al. / Construction and Building Materials 188 (2018) 1025–1034 solutions such as sodium silicate and sodium hydroxide [19,20]. The utilisation of geopolymer binder in the production of lightweight aggregate concrete offers an alternative environment friendly material and also reduces carbon dioxide emissions due to the elimination of cement. An earlier study [21] indicated that coal bottom ash is used to produce lightweight geopolymer concretes (LWGCs) with compressive strengths of 14.3–18.1 MPa and densities of 1660–1690 kg/m3 that are used as thermal insulation and moderate-strength concrete. Furthermore, Wongsa et al. [21] examined the effect of alkaline solution to fly ash ratios (0.70, 0.75, and 0.80) and SS/SH ratios (0.5, 1.0, and 1.5) on the workability of LWGCs containing bottom ash as fine and coarse aggregate and reported that the lowest alkaline solution to fly ash ratio of 0.70 corresponded to a suitable and sufficient ratio for workability of LWGCs with high SS/SH ratio of 1.5. Zaetang et al. [22] also indicated that pervious high-calcium fly ash geopolymer concrete containing coal bottom ash as aggregate exhibited density of 1470– 1500 kg/m3, compressive strength of 5.7–8.6 MPa, and thermal conductivity of 0.30–0.33 W/m K, and thus it is suitable as an environment friendly concrete. Posi et al. [23] reported that recycled lightweight concrete aggregate from waste lightweight concrete can be used as a lightweight aggregate to produce LWGCs with compressive strengths of 4.5–17.5 MPa and densities of 1200– 1500 kg/m3. Colangelo et al. [24] investigated the use of expanded polystyrene waste as a lightweight aggregate in metakaolin-based geopolymer and reported that the LWGC mixed with expanded polystyrene exhibited lower thermal conductivity and higher strengths when compared to that of Portland cement-based materials. Volcanic pumice is a natural lightweight aggregate with a sponge-like structure and is observed in the granulated form that is produced from the rapid cooling of molten lava [1]. It is used as an aggregate in the production of lightweight concrete in several countries [5]. Specifically, it is found in United States, Italy, Turkey, Greece, and Spain [1,25]. With respect to Thailand, it is found in Rayong, Buriram, and Lopburi provinces. Crushed bricks as aggregates are of particular interest since their use reduces the problem of waste storage and also aids in the preservation of natural aggregate resources [7]. Recent successful studies on the use of crushed clay bricks as aggregates in concrete are reported in several parts of the world including Algeria [7], Egypt [8], Hong Kong [26,27], United States [28,29], United Kingdom [30], and China [31]. In Thailand, several old constructions and buildings typically consume large quantities of clay brick/block and reach the end of their service life facing demolition. This results in large quantities of wasted clay bricks in several parts of country. Waste bricks generated from construction and demolition sites are generally transported to landfills for disposal. With the limited landfill space in Thailand, it is necessary to discover the possible use of crushed clay brick as a new civil engineering material. There are many published studies on the use of recycled crushed clay brick aggregate (CA) and pumice aggregate (PA) with geopolymer-based material. Previous studies [32,33] reported that CA can be used as coarse aggregate to produce pervious geopolymer concrete. However, there is a paucity of studies that reported on the use of CA and PA as fine and coarse aggregates in fly ash-based geopolymer binder to produce LWGC. This study aims to fill the aforementioned knowledge gap. Therefore, we focus on the use of CA and PA as fine and coarse aggregates in lightweight high-calcium fly ash geopolymer concrete. In the experiment part, the effects of the sodium silicate to sodium hydroxide solution (SS/SH) ratio and elevated temperature exposure on the properties of LWGCs containing CA and PA as fine and coarse aggregates including workability, compressive strength, splitting tensile strength, surface abrasion resistance, density, thermal conductivity, and ultrasonic pulse velocity were investigated. The results were also compared to theose of control geopolymer concrete (CGC) produced with crushed limestone and river sand. 2. Materials and preparation of sample 2.1. Materials The materials used in the study are crushed clay brick aggregate (CA), pumice aggregate (PA), crushed natural limestone aggregate and river sand (NA), and lignite coal fly ash. The CA was obtained from crushed broken clay brick from a construction site in Khon Kaen province, north-eastern Thailand. Commercially available acidic pumice in Bangkok, Thailand was used as the PA. The coarse aggregates were crushed and/or sieved into similar particles in the size range of 4.8 to 9.5 mm while fine aggregates ranged from 75 mm to 4.8 mm. The fineness modulus of fine CA (3.73), fine PA (3.72), and river sand (3.30) slightly exceeded the range (2.3–3.1) based on ASTM C33/C33M-18 [34]. The CA and river sand were sieved to achieve as-received PA grain size and to avoid any effects due to the aggregate size. The grain size distributions of the aggregates relative to the upper and lower limits based on ASTM C33/C33M-18 [34] are shown in Fig. 1. The visual observation of all aggregates clearly indicated that the particle shape of coarse PA was round while those of coarse CA and crushed limestone were angular. The surface texture of coarse PA was slightly rougher and more porous than those of coarse CA and crushed limestone. Table 1 shows the physical properties of all fine and coarse aggregates used in the study. The bulk density and specific Fig. 1. Grain size distributions of aggregates. 1027 A. Wongsa et al. / Construction and Building Materials 188 (2018) 1025–1034 Table 1 Physical properties of coarse and fine aggregates. Properties 3 Bulk density (kg/m ) Specific gravity (SSD) Water absorption (%) Fineness modulus Los Angeles abrasion loss (%) Crushed limestone Coarse CA Coarse PA River sand Fine CA Fine PA Testing standard 1576 2.70 0.54 6.00 26.0 1008 2.03 15.34 6.13 45.3 395 1.07 66.11 6.08 71.3 1671 2.63 0.35 3.30 – 1040 2.09 13.04 3.73 – 601 1.40 34.88 3.72 – ASTM ASTM ASTM ASTM ASTM Table 2 Chemical composition of fly ash, CA, and PA. Oxides (%) Fly ash CA PA SiO2 Al2O3 Fe2O3 CaO K2O MgO MnO Na2O P2O5 TiO2 SO3 Loss on Ignition (LOI) 39.4 20.8 11.5 14.5 2.4 2.2 – 1.4 0.2 0.5 4.2 1.5 67.9 15.2 5.1 0.6 1.5 1.2 0.1 0.8 – 0.8 – – 60.4 16.7 4.0 2.4 3.4 0.9 0.1 4.0 0.2 0.5 – – gravity in saturated surface dry (SSD) condition of PA and CA were lower than those of crushed limestone and river sand while the water absorptions were higher as expected. The aggregates were fabricated in the SSD condition prior to the mixing of concrete. The Los Angeles abrasion test of coarse aggregates was presented in terms of the percent weight loss. The weight losses of CA and PA exceeded those of crushed limestone. The results indicated that CA and PA exhibited lower abrasion resistance than crushed limestone and were consistent in terms of bulk density and specific gravity. Although, the Los Angeles abrasion loss value and water absorption of PA exceeded the tolerated threshold. However, PA can be used as a lightweight aggregate prepared by processing natural materials (pumice, scoria or tuff) for structural concrete based on ACI Committee 213 [35]. Fly ash with specific gravity of 2.17 and 44% (by weight) retained on sieve no. 325 (45 mm) was obtained from Lampang province in the North of Thailand and used as the primary source material to prepare a geopolymer binder. The chemical compositions of the materials are given in Table 2. Specifically, 10 mol/L (Molar) sodium hydroxide (SH) and sodium silicate (SS) solution were used as alkaline activators to produce geopolymer binder. Additionally, 10 M SH was prepared by dissolving 400 gm of sodium hydroxide pellets with distilled water in a final volume of 1 L and left for 24 h prior to use. The commercial grade SS with 12.53% Na2O, 30.24% SiO2, and 57.23% H2O by weight was used without any modification. 2.2. Mix proportions Given the low bulk densities of CA and PA, they were used as complete replacement (100%) of natural aggregate (NA) to produce lightweight aggregate geopolymer concrete. The proportions in all mixtures were maintained as constant and consisted of 51%, 15%, and 34% by volume of bulk coarse aggregate, fine aggregate, and geopolymer paste, respectively. Their mix proportions are shown in Table 3. The parameters of the study corresponded to the types of fine and coarse aggregates (NA, CA, and PA) and the SS/SH ratios by mass (0.5, 1.0, and 1.5). The alkaline C29/C29M-17 [36] C127-15 [37] C127-15 [37] C136/C136M-14 [38] C131/C131M-14 [39] solution to fly ash ratio by mass and concentration of SH were maintained as constant at 0.70 and 10 M, respectively [21]. The proportions in the mixture depended on volumetric bases. The ratio of volume of bulk coarse aggregate to fine aggregate and geopolymer paste volume was carefully maintained. The names of the mixtures were based on the aggregate type and SS/SH ratio. For example, PA-1.0 denotes concrete with fine and coarse pumice aggregates and an SS/SH ratio of 1.0. 2.3. Mixing, casting, and curing The mixing of LWGCs and CGCs was performed in a controlled room at 25 °C. The fly ash and SH were mixed in a pan-type mixer for 5 min, and the fine and coarse aggregates were then added and mixed for 2 min. Finally, SS was added to the mixture and mixed for 1 min. After mixing, fresh mixtures were placed in moulds and compacted on a vibrating table for 10 s. Two sizes of the samples were cast. Cylindrical samples with a diameter of 10 cm and height of 20 cm were prepared for compressive and splitting tensile strength tests. Additionally, 10-cm cube samples were used for surface abrasion resistance, density, ultrasonic pulse velocity, and thermal conductivity tests. The specimens were covered with a plastic sheet to prevent moisture loss and were allowed to stand for 1 h at 25 °C. Subsequently, the specimens were cured at 60 °C for 48 h. Following the heat curing, the specimens were placed in the laboratory for cooling, and they were demoulded on the following day. The specimens were then covered with a plastic sheet and stored in a controlled room at 25 °C until testing commenced. 3. Testing The workability of fresh LWGCs and CGCs mixes was examined by conducting slump tests as per ASTM C143/C143M-15a [40]. The compressive strength, splitting tensile strength, and surface abrasion resistance were tested at the age of 7 d as per ASTM C39/ C39M [41], ASTM C496/C496M-17 [42], and ASTM C944/C944M [43], respectively. The reported results corresponded to the average of three samples. The surface abrasion resistance was tested by the rotating–cutting method. The samples were abraded with a load of 98 N on the surface for 2 min following contact between the cutter and sample surfaces. The abrasion resistance of each sample was tested on the three sides of the sample. The densities of the samples were measured as per ASTM C1754/C1754M-12 [44]. The ultrasonic pulse velocity on concretes was determined as per ASTM C597-16 [45]. The transducers that were used exhibited a diameter of 50 mm with a frequency of 50 kHz. The thermal conductivities were measured using a direct measuring instrument with a surface probe (ISOMET2114). The measurement ranges of the device corresponded to 0.04–6.0 W/m K. Table 3 Mix proportions per m3 of LWGCs and CGCs. Mix NA-0.5 NA-1.0 NA-1.5 CA-0.5 CA-1.0 CA-1.5 PA-0.5 PA-1.0 PA-1.5 Aggregate Geopolymer binder Coarse (kg) Fine (kg) Fly ash (kg) SS (kg) SH (kg) 1385 1385 1385 1049 1049 1049 536 536 536 383 383 383 354 354 354 215 215 215 350 354 357 350 354 357 350 354 357 82 124 150 82 124 150 82 124 150 163 124 100 163 124 100 163 124 100 Note: NA = crushed limestone and river sand, CA = crushed clay brick, PA = pumice aggregate. 1028 A. Wongsa et al. / Construction and Building Materials 188 (2018) 1025–1034 In order to examine the fire resistance properties of LWGCs, only the SS/SH = 1.0 mixture was selected to expose at temperatures of 400 °C, 600 °C, and 800 °C. At the age of 7 d, the specimens were subjected to target temperatures at a gradual incremental rate of 3–5 °C/min. Immediately after the target temperature was attained, it was maintained for 1 h before the furnace was shut down to allow the specimens in the furnace to cool to room temperature. Following the temperature exposure, the compressive strength, density, ultrasonic pulse velocity, and thermal conductivity were tested. Figs. 3–5. The mechanical properties of LWGC containing CA and PA were lower than those of NA due to its looser texture that was also indicated by the decreases in the density and Los Angeles abrasion resistance. Similar results were reported in previous studies [5,7] with respect to the negative effect of using crushed brick and volcanic pumice as an aggregate on the mechanical properties of the Portland cement-based concrete. Based on Khandaker and Anwar [5], the compressive strength and modulus of elasticity decreased with increases in the volcanic PA content due to the 4. Results and discussion 4.1. Slump value As shown in Fig. 2, the slump value of lightweight geopolymer concretes (LWGCs) containing PA is in the range of 201–263 mm, and this exceeds that of LWGCs containing recycled CA by approximately 15–24% and by 5–22% for control geopolymer concrete (CGCs) containing crushed limestone and river sand (NA). This was reasonable due to the particle shapes of aggregates, i.e. PA was round, while CA and NA were angular. Additionally, the high water absorption of fine and coarse PA was potentially because the volume of water at the SSD condition of PA particles exceeded those of CA and NA particles. Part of the water inside PA particles was available and improved the workability of the mixture and led to increases in slump values of concrete. However, the slump value of LWGCs containing CA was slightly lower than that of CGCs containing NA. This was due to the rounder particle shape of river sand than that of fine CA although the crushed limestone and coarse CA exhibited similar particle shapes. Additionally, the results clearly indicated that the slump value of LWGCs also depended on the SS/SH ratio. The increases in the SS/SH ratio decreased the slump value of LWGCs. This was due to the high viscosity of SS that reduced the flow of mixtures. For example, slump values of 263, 235, and 201 mm were obtained with PA mixes with SS/SH ratios of 0.5, 1.0, and 1.5, respectively. A previous study [21] reported a similar finding wherein the slump values of fresh lightweight geopolymer concrete containing bottom ash as aggregates and high-calcium fly ash as geopolymer binder decreased with increases in the SS/SH ratio. Fig. 3. Compressive strength and SS/SH ratios for various types of aggregates. 4.2. Compressive strength, splitting tensile strength and weight losses from surface abrasion test A summary of mechanical properties of LWGCs and CGCs including compressive strengths, splitting tensile strengths, and weight losses from the surface abrasion test at 7 d are given in Fig. 2. Slump value and SS/SH ratios for various types of aggregates. Fig. 4. Splitting tensile strength and SS/SH ratios for various types of aggregates. Fig. 5. Weight losses from surface abrasion test and SS/SH ratios for various types of aggregates. A. Wongsa et al. / Construction and Building Materials 188 (2018) 1025–1034 replacement of normal strong crushed gravel aggregate with relatively weak PA. Debieb and Kenai [7] also reported that the compressive strength, flexural strength, and modulus of elasticity of Portland cement-based concrete containing 100% fine and coarse crushed brick were observed as lower than those of comparable ordinary concrete. However, the maximum compressive strength of LWGCs containing CA (18.3 MPa) and PA (7.0 MPa) in the study was in the range of the compressive strength requirement for structural lightweight concrete (17.0–41.0 MPa) and lightweight moderate-strength concrete (2.0–14.0 MPa), respectively, based on ACI Committee 213 [35]. Furthermore, the average ratios of the splitting tensile strengths to compressive strengths of LWGCs containing CA of 7.7% and PA of 10.3% slightly exceeded those of CGCs containing crushed limestone and river sand (6.6%). It should be noted that that the use of CA and PA as aggregates to produce LWGCs resulted in significant improvements in the splitting tensile strength for geopolymer concrete with similar compressive strength. This is because the surface roughness and angular shape of the crushed material were advantageous in terms of a good bond between the geopolymer paste and the aggregates, and this increased the splitting tensile strength performances [7]. 1029 With respect to the effect of SS/SH ratio, the mechanical properties including compressive strength, splitting tensile strength, and surface abrasion resistance of LWGCs containing CA and PA tended to increase with increases in the SS/SH ratios as shown in Figs. 3–5. For example, the SS/SH ratios of LWGC containing CA increased from 0.5 to 1.5, compressive strength increased from 8.2 to 18.3 MPa, splitting tensile strength increased from 0.6 to 1.6 MPa, and weight losses from the abrasion surface test decreased from 4.17 to 2.63 g. Visual observation of the fracture surfaces of LWGCs specimens after the splitting tensile strength test are shown in Figs. 6 and 7 and clearly indicated that the majority of failure of LWGCs with SS/SH ratio of 0.5 occurred at the interface between aggregate and geopolymer paste. Conversely, the failure surface of LWGCs with SS/SH ratio of 1.0 and 1.5 exhibited several fractures through the aggregate particles. It should be noted that that the increases in the SS/SH ratio increased the geopolymer paste and bonding between aggregate and geopolymer paste. When the LWGC approached the ultimate load, the cracks passed though the zones as opposed to the interface between aggregate and geopolymer paste. The result confirmed the findings in a previous study by Eiamwijit et al. [46] who investigated the properties of Fig. 6. Fracture surface of LWGCs containing CA as aggregates. Fig. 7. Fracture surface of LWGCs containing PA as aggregates. 1030 A. Wongsa et al. / Construction and Building Materials 188 (2018) 1025–1034 fluidised bed coal combustion fly ash geopolymers with different SS/SH mass ratios (1.0, 1.5, 2.0, and 2.5) and indicated that the high SS/SH ratio mixes with increased water glass content led to dense and homogeneous composites. Abdullah et al. [47] and Risdanareni et al. [48] investigated the effect of SS/SH ratios (0.5, 1.0, 1.5, 2.0, 2.5, and 3.0) on the properties and morphology of geopolymer synthesis using low-calcium fly ash and indicated that geopolymers gel with optimum SS/SH ratios exhibited a continuous matrix and homogeneous and less porous microstructure. This was due to the increases in the sodium oxide content in the system, which was mainly required for the geopolymerisation reaction since sodium oxide was used to balance the charges and formed the aluminosilicate network [49]. However, previous studies [21,50] also reported that the compressive strength decreased when increased silicate was added into the system because an excess of sodium silicate hindered water evaporation and structure formation. Furthermore, increases in SS/SH ratio also decreased the dosage of SH. At a high SS/SH ratio, the leaching of alumina and silica decreased, and this resulted in a low level of geopolymerisation, and thereby decreased the strength. A similar effect was reported by El-Sayed et al. [51] wherein geopolymer-based mortar with a dosage of 6% SH achieved higher gains in compressive strength when compared with a dosage of 2% SH, thereby clearly indicating the effect of the SH dosage on the early compressive strength gain and on the degree of geopolymerisation. As shown in the study, the compressive strength and splitting tensile strength of CGCs tended to decrease with increases in the SS/SH ratios as shown in Figs. 3 and 4. 4.3. Density, thermal conductivity, and ultrasonic pulse velocity The measured density, thermal conductivity, and ultrasonic pulse velocity of LWGCs and CGCs are shown in Table 4. The results indicated that the use of CA and PA as fine and coarse aggregates in LWGC reduced the density and thermal conductivity. This was due to the high porosity and low density of CA and PA and low density of LWGCs. The densities of LWGCs containing CA (1685–1749 kg/m3) and PA (1011–1111 kg/m3) were approximately 25% and 50% lower than those of CGCs containing crushed limestone and river sand (2258–2299 kg/m3). Furthermore, the densities of LWGCs containing CA and PA were within the range of 1440–1850 and 1000–1400 kg/m3 for structural lightweight concrete and lightweight moderate-strength concrete as per ACI Committee 213 [35]. Previous studies reported that the ultrasonic pulse velocity and the thermal conductivity are affected by the density of concrete [52,53]. The thermal conductivity of LWGCs containing CA (0.62– 0.65 W/m K) and PA (0.20–0.22 W/m K) were approximately 3 and 7 times lower than those of CGCs containing crushed limestone and river sand due to the low density of aggregates and concretes. It was noted that the use of CA and PA as fine and coarse aggregate to produce LWGCs exhibited good density and excellent thermal insulation and were suitable for insulating purposes. The Fig. 8. Relationship between thermal conductivity and density of LWGCs and CGCs. relationship between the thermal conductivity and the density of LWGCs and CGCs is plotted in Fig. 8. The correlation coefficient of the results was high with R2 = 0.9972. As shown in the figure, the thermal conductivity increased with increases in the concrete density. The density and thermal conductivity are expressed as follows: T ¼ 0:037e0:0016D ð1Þ where T denotes the thermal conductivity (W/m K) D denotes the dry density (kg/m3) With respect to ultrasonic pulse velocity, the ultrasonic pulse velocity is also directly related to porosity of concrete since discontinuities of pores in aggregate and concrete cause a delay in the wave propagation [54]. A previous study [55] also reported that the ultrasonic pulse velocity in concrete is significantly influenced by the types and unit weight of aggregate and density of concrete. In the study, the ultrasonic pulse velocities of the LWGCs containing CA (1985–2285 m/s) and PA (1586–1858 m/s) were significantly lower than that of CGCs containing NA (3089–3419 m/s). This was due to the lower densities of CA and PA when compared with that of NA as shown in Tables 1. The relationship between the density and the ultrasonic pulse velocity of LWGCs and CGCs is shown in Fig. 9. In a manner similar to the thermal conductivity results, the ultrasonic pulse velocity increased with increases in the concrete density and is expressed as follows: U ¼ 957:16e0:0005D ð2Þ where U denotes the ultrasonic pulse velocity (m/s) D denotes the dry density (kg/m3) With respect to the ratio of alkali activators, the SS/SH ratios slightly affected the density, thermal conductivity, and ultrasonic pulse velocity of LWGCs and CGCs. For example, with respect to the densities of 1111, 1011, and 1079 kg/m3, thermal conductivities Table 4 Density, thermal conductivity, and ultrasonic pulse velocity of LWGCs and CGCs. Mix Density (kg/m3) Thermal conductivity (W/m K) Ultrasonic pulse velocity (m/s) NA-0.5 NA-1.0 NA-1.5 CA-0.5 CA-1.0 CA-1.5 PA-0.5 PA-1.0 PA-1.5 2258 2299 2272 1731 1685 1749 1111 1011 1079 1.41 1.58 1.58 0.62 0.62 0.65 0.21 0.20 0.22 3089 3419 3303 1985 2285 2173 1586 1745 1858 Fig. 9. Relationship between ultrasonic pulse velocity and density of LWGCs and CGCs. A. Wongsa et al. / Construction and Building Materials 188 (2018) 1025–1034 of 0.21, 0.20, and 0.22 W/m K and ultrasonic pulse velocities of 1586, 1745, and 1858 m/s, respectively, were obtained for LWGC containing PA with SS/SH ratios of 0.5, 1.0, and 1.5, respectively. A similar effect was reported in an earlier study [21] wherein the effects of SS/SH ratios on the density, thermal conductivity, and ultrasonic pulse velocity of high-calcium fly ash lightweight geopolymer concrete containing bottom ash as aggregate were not significant. 4.4. Change in appearance of the specimens after temperature exposure The changes in the physical appearance of CGCs and LWGCs containing CA and PA cylinders at different temperature exposures are shown in Figs. 10–12. The original colour of CGCs containing NA and LWGCs containing PA was dark grey while that of CA exhibited a red colour. All the concrete cylinders did not exhibit a significant change in colour when exposed to various temperatures. The only visible difference was that the original colour became lighter after exposure to higher temperatures at 400 °C, 600 °C, and 800 °C due to the high moisture loss at high temperatures. 1031 Physical observations after the fire exposures also indicated that the LWGCs suffered less damage in terms of spalling and cracking when compared to the CGC. As shown in Figs. 10–12, the CGC specimens suffer severe spalling and cracking at 600 °C and 800 °C exposures, while the LWGCs containing CA and PA at 800 °C exposure contained a number of small cracks at the surface. Based on Sarker et al. [56], the surface spalling and cracking of geopolymer concrete occurred due to the temperature differential between the surface and centre of the specimens. This also occurred as a result of the differential strain that is caused by a temperature gradient through the cross section of the concrete [56]. Thus, the results indicated that the LWGCs containing CA and PA in the study exhibited excellent fire resistance that is essential for its use as a building material. 4.5. Residual strength of geopolymer concretes after temperature exposure A summary of results of compressive strengths before and after temperature exposure at 400 °C, 600 °C, and 800 °C are given in Table 5. The LWGCs containing CA and PA gained some strength with a residual strength of 146% and 104%, respectively, after exposure at Fig. 10. CGCs containing NA as aggregate after exposure at 25 °C, 400 °C, 600 °C, and 800 °C. Fig. 11. LWGCs containing CA as aggregate after exposure at 25 °C, 400 °C, 600 °C, and 800 °C. 1032 A. Wongsa et al. / Construction and Building Materials 188 (2018) 1025–1034 Fig. 12. LWGCs containing PA as aggregate after exposure at 25 °C, 400 °C, 600 °C, and 800 °C. Table 5 Compressive strength and percentage residual strength of geopolymer concretes. Temperature (°C) 25 400 600 800 NA-1.0 CA-1.0 PA-1.0 Comp. strength (MPa) (STD. dev) Residual strength (%) Comp. strength (MPa) (STD. dev) Residual strength (%) Comp. strength (MPa) (STD. dev) Residual strength (%) 36.0 (1.3) 13.7 (0.9) 4.1 (0.4) 3.0 (0.6) 100 38 12 8 17.5 (0.1) 25.6 (0.6) 14.5 (0.4) 6.9 (0.1) 100 146 83 39 7.0 7.4 3.3 2.0 100 104 47 29 400 °C. This was potentially due to the loss of water in the system due to sintering at the aforementioned temperature and a combination of geopolymer matrix and aggregates (CA and PA). This was the main factor that contributed to the higher bonding and residual strength following exposure at 400 °C. This is consistent with the visual observations of the failure characteristics after the compressive strength test as shown in Fig. 13. The CGCs specimen showed complete disintegration, and the LWGCs specimens containing CA and PA exhibited large fragments of specimens. After the exposure at 600 °C and 800 °C, the LWGCs exhibited residual strengths of 83% and 39% for CA-1.0 and 47% and 29% for PA-1.0, respectively. (0.3) (0.3) (0.1) (0.0) With respect to the residual strength of CGC, CGCs containing NA suffered from significant strength loss following exposure to high temperature. The residual strength of CGCs after temperature exposure at 400 °C, 600 °C, and 800 °C decreased by 38%, 12%, and 8%, respectively. The trends of strength reduction in the CGC specimens were similar to those observed by Sarker et al. [56] and Abdulkareem et al. [57]. Based on Sarker et al. [56], the strength loss in the geopolymer concrete specimens after exposure at 400 °C, 650 °C, 800 °C, and 1000 °C was attributed to the difference between the thermal expansions of geopolymer matrix and aggregates [57]. However, the results also indicated that the LWGCs Fig. 13. Failure characteristics of compressive strength test specimens after exposure at 400 °C. A. Wongsa et al. / Construction and Building Materials 188 (2018) 1025–1034 containing CA and PA exhibited higher residual strength relative to that of CGCs at all three exposure temperatures. When compared with CGCs, LWGC exhibited less cracking and spalling and higher residual compressive strength, and thus the performance of LWGCs exceeded that of CGCs. 4.6. Density, thermal conductivity, and ultrasonic pulse velocity after temperature exposure Table 6 summarises the results of density, thermal conductivity, and ultrasonic pulse velocity of LWGCs and CGCs after the temperature exposure at 400 °C, 600 °C, and 800 °C. The results indicated that the densities of LWGCs and CGCs decreased with increases in the exposure at elevated temperatures. This was due to the loss of moisture in aggregates and alkaline solution contents in concrete and surface spalling of specimens. The density losses of the LWGCs containing PA significantly exceeded those of CA and NA. Following the temperature exposure at 400 °C, 600 °C, and 800 °C, the densities of the PA-1.0 decreased by 24%, 29%, and 31%, the densities of the CA-1.0 decreased by 14%, 17%, and 18%, and the densities of the NA-1.0 decreased by 5%, 6%, and 7%, respectively. This was because the water absorption of PA exceeded those of CA and NA. There is generally a complete loss of mass due to elevated temperature exposure in concrete that significantly impacts the compressive strength of the sample [56]. However, the mass losses at 400 °C temperature exposure in the study did not adversely affect the compressive strength of the LWGCs containing CA and PA. The thermal conductivity measured after the temperature exposure was also directly related to the density of concrete as shown in Table 6. The increase in high temperature exposure contributed to increases in the loss of moisture and alkali solution and specimen porosity, thereby leading to decreases in the density and thermal conductivity of the concrete. For example, when the exposure to elevated temperatures increased from 25 °C to 800 °C, the density decreased from 2339 to 2099 kg/m3 for CGCs containing NA, 1961 to 1638 kg/m3 for LWGCs containing CA, and 1357 to 1016 kg/m3 for LWGCs containing PA. Similarly, the thermal conductivity decreased from 1.75 to 0.91 W/m K for CGCs containing NA, 0.70 to 0.57 W/m K for LWGCs containing CA, and 0.26 to 0.23 W/m K for LWGCs containing PA. With respect to the ultrasonic pulse velocity results after the temperature exposure, the ultrasonic pulse velocity values were consistent with the density and the residual strength results of concrete. This is consistent with the results obtained in a study by Ghosh et al. [58]. The study explained the ultrasonic pulse velocity in situ test for assuring the quality of concrete and reported that the fly ash-based geopolymer concrete ultrasonic pulse velocity value increased with increases in the compressive Table 6 Density, thermal conductivity, and ultrasonic pulse velocity of geopolymer concretes after temperature exposure. Mix Density (kg/m3) Thermal conductivity (W/m K) Ultrasonic pulse velocity (m/s) NA-1.0-25 NA-1.0-400 NA-1.0-600 NA-1.0-800 CA-1.0-25 CA-1.0-400 CA-1.0-600 CA-1.0-800 PA-1.0-25 PA-1.0-400 PA-1.0-600 PA-1.0-800 2339 2212 2162 2099 1961 1700 1652 1638 1357 1079 1023 1016 1.75 1.32 1.04 0.91 0.70 0.58 0.57 0.57 0.26 0.23 0.22 0.23 3419 1480 393 – 2285 2306 1299 1296 1745 1832 1108 1081 1033 strength [59]. Furthermore, Mohammed and Rahman [55] also observed that the ultrasonic pulse travelled faster in dense concretes when compared to that in poor quality concretes. In the study, the compressive strength and the ultrasonic pulse velocity of LWGCs containing CA and PA with exposure at 400 °C exhibited higher values than those exposed to 25 °C, 600 °C, and 800 °C, respectively. It should be noted that that a high temperature exposure up to 400 °C can be used to improve the quality of LWGCs containing CA and PA. However, the precise values of ultrasonic pulse velocity in the NA-1.0–800 specimens with the lowest residual strength are not determined due the surface spalling and cracking of the specimens as shown in Fig. 10(d). 5. Conclusions In the study, the results indicated that the use of crushed clay brick (CA) and PA to produce lightweight geopolymer concretes (LWGCs) exhibited lower compressive strength, splitting tensile strength, surface abrasion resistance, density, thermal conductivity, and ultrasonic pulse velocity when compared to those of geopolymer concrete containing natural aggregate (CGCs). However, the compressive strength and density of LWGCs containing CA (8.2–18.3 MPa and 1685–1749 kg/m3) and PA (2.7–7.0 MPa and 1011–1111 kg/m3) revealed that the mixes with highest compressive strength was in the range of the required values for structural lightweight concrete and lightweight moderate-strength concrete as per ACI Committee 213. Furthermore, the LWGCs containing CA and PA exhibited thermal insulation and residual strength exceeding those of CGCs at exposure temperatures corresponding to 400 °C, 600 °C, and 800 °C. Conflict of interest None declared. Acknowledgements The authors would like to acknowledge the financial supports from the Post-Doctoral Training Program from Research Affairs and Graduate School, Khon Kaen University, Thailand (Grant No. 59254); the Thailand Research Fund (TRF) and Khon Kaen University under the Royal Golden Jubilee Ph.D. Program (Grant No. PHD 0083/2556), the TRF Basic Research Grant (Grant No. BRG6080010) and the TRF Distinguished Research Professor (Grant No. DPG6180002). References [1] D. Sari, A.G. Pasamehmetoglu, The effects of gradation and admixture on the pumice lightweight aggregate concrete, Cem. Concr. Res. 35 (5) (2005) 936– 942. [2] P. Posi, C. Teerachanwit, C. Tanutong, S. Limkamoltip, S. Lertnimoolchai, V. Sata, et al., Lightweight geopolymer concrete containing aggregate from recycle lightweight block, Mater. Des. 52 (2013) 580–586. [3] Y. Zaetang, A. Wongsa, V. Sata, P. Chindaprasirt, Use of lightweight aggregates in pervious concrete, Constr. Build. Mater. 48 (2013) 585–591. [4] P. Posi, S. Lertnimoolchai, V. Sata, P. Chindaprasirt, Pressed lightweight concrete containing calcined diatomite aggregate, Constr. Build. Mater. 47 (2013) 896–901. [5] K.M. Anwar Hossain, Properties of volcanic pumice based cement and lightweight concrete, Cem. Concr. Res. 34 (2) (2004) 283–291. [6] Yliniemi, Paiva, Ferreira, Tiainen, Illikainen, Development and incorporation of lightweight waste-based geopolymer aggregates in mortar and concrete, Constr. Build. Mater. 131 (Supplement C) (2017) 784–792. [7] F. Debieb, S. Kenai, The use of coarse and fine crushed bricks as aggregate in concrete, Constr. Build. Mater. 22 (5) (2008) 886–893. [8] A.A. Aliabdo, A.-E.M. Abd-Elmoaty, H.H. Hassan, Utilization of crushed clay brick in concrete industry, Alexandria Eng. J. 53 (1) (2014) 151–168. [9] J. Davidovits, Geopolymer chemistry and applications. in: 3rd Edition Institut Geopolymere, editor, 2011. 1034 A. Wongsa et al. / Construction and Building Materials 188 (2018) 1025–1034 [10] P. Chindaprasirt, T. Chareerat, V. Sirivivatnanon, Workability and strength of coarse high calcium fly ash geopolymer, Cem. Concr. Compos. 29 (3) (2007) 224–229. [11] F. Fan, Z. Liu, G. Xu, H. Peng, C.S. Cai, Mechanical and thermal properties of fly ash based geopolymers, Constr. Build. Mater. 160 (2018) 66–81. [12] V. Sata, A. Sathonsaowaphak, P. Chindaprasirt, Resistance of lignite bottom ash geopolymer mortar to sulfate and sulfuric acid attack, Cem. Concr. Compos. 34 (5) (2012) 700–708. [13] A. Sathonsaowaphak, P. Chindaprasirt, K. Pimraksa, Workability and strength of lignite bottom ash geopolymer mortar, J. Hazard. Mater. 168 (1) (2009) 44– 50. [14] K. Kaur, J. Singh, M. Kaur, Compressive strength of rice husk ash based geopolymer: the effect of alkaline activator, Constr. Build. Mater. 169 (2018) 188–192. [15] P. Sturm, G.J.G. Gluth, H.J.H. Brouwers, H.C. Kühne, Synthesizing one-part geopolymers from rice husk ash, Constr. Build. Mater. 124 (2016) 961– 966. [16] A. Karthik, K. Sudalaimani, C.T. Vijaya Kumar, Investigation on mechanical properties of fly ash-ground granulated blast furnace slag based self curing bio-geopolymer concrete, Constr. Build. Mater. 149 (2017) 338–349. [17] S. Saha, C. Rajasekaran, Enhancement of the properties of fly ash based geopolymer paste by incorporating ground granulated blast furnace slag, Constr. Build. Mater. 146 (2017) 615–620. [18] M.Z.N. Khan, Fu.A. Shaikh, Y. Hao, H. Hao, Synthesis of high strength ambient cured geopolymer composite by using low calcium fly ash, Constr. Build. Mater. 125 (2016) 809–820. [19] S. Pangdaeng, T. Phoo-ngernkham, V. Sata, P. Chindaprasirt, Influence of curing conditions on properties of high calcium fly ash geopolymer containing Portland cement as additive, Mater. Des. 53 (2014) 269–274. [20] P. Pavithra, M. Srinivasula Reddy, P. Dinakar, B. Hanumantha Rao, B.K. Satpathy, A.N. Mohanty, A mix design procedure for geopolymer concrete with fly ash, J. Cleaner Prod. 133 (2016) 117–125. [21] A. Wongsa, Y. Zaetang, V. Sata, P. Chindaprasirt, Properties of lightweight fly ash geopolymer concrete containing bottom ash as aggregates, Constr. Build. Mater. 111 (2016) 637–643. [22] Y. Zaetang, A. Wongsa, V. Sata, P. Chindaprasirt, Use of coal ash as geopolymer binder and coarse aggregate in pervious concrete, Constr. Build. Mater. 96 (2015) 289–295. [23] P. Posi, P. Thongjapo, N. Thamultree, P. Boontee, P. Kasemsiri, P. Chindaprasirt, Pressed lightweight fly ash-OPC geopolymer concrete containing recycled lightweight concrete aggregate., Constr. Build. Mater. 127 (Supplement C) (2016) 450–456. [24] F. Colangelo, G. Roviello, L. Ricciotti, V. Ferrándiz-Mas, F. Messina, C. Ferone, et al., Mechanical and thermal properties of lightweight geopolymer composites, Cem. Concr. Compos. 86 (2018) 266–272. [25] G. Amato, G. Campione, L. Cavaleri, G. Minafò, N. Miraglia, The use of pumice lightweight concrete for masonry applications, Mater. Struct. 45 (5) (2012) 679–693. [26] C.S. Poon, D. Chan, Feasible use of recycled concrete aggregates and crushed clay brick as unbound road sub-base, Constr. Build. Mater. 20 (8) (2006) 578– 585. [27] C.S. Poon, D. Chan, Paving blocks made with recycled concrete aggregate and crushed clay brick, Constr. Build. Mater. 20 (2006) 569–577. [28] M. Adamson, A. Razmjoo, A. Poursaee, Durability of concrete incorporating crushed brick as coarse aggregate, Constr. Build. Mater. 94 (2015) 426–432. [29] J. Hu, K. Wang, J.A. Gaunt, Behavior and mix design development of concrete made with recycled aggregate from deconstructed lead-contaminated masonry materials, Constr. Build. Mater. 40 (2013) 1184–1192. [30] J. Yang, Q. Du, Y. Bao, Concrete with recycled concrete aggregate and crushed clay bricks, Constr. Build. Mater. 25 (4) (2011) 1935–1945. [31] L. Hu, J. Hao, L. Wang, Laboratory evaluation of cement treated aggregate containing crushed clay brick, J. Traffic Transp. Eng. (English Edition) 1 (5) (2014) 371–382. [32] V. Sata, A. Wongsa, K. Somna, P. Chindaprasirt, Interfacial transition zone of pervious cement and geopolymer concrete containing crushed clay brick, J. Sci. Faculty Chiang Mai Univ. 43 (2) (2015) 402–408. [33] V. Sata, A. Wongsa, P. Chindaprasirt, Properties of pervious geopolymer concrete using recycled aggregates, Constr. Build. Mater. 42 (2013) 33–39. [34] American Society for Testing and Materials, Standard Specification for Concrete Aggregates, ASTM C33/C33M-18. Book of Standards Volume: 04.02: Developed by Subcommittee: C09.20, 2018. [35] ACI Committee 213, Guide for Structural Lightweight Aggregate Concrete, American Concrete Institute, Detroit, USA, 2000. [36] American Society for Testing and Materials, Standard Test Method for Bulk Density (‘‘Unit Weight”) and Voids in Aggregate. ASTM C29/C29M – 17. Book of Standards Volume: 04.02: Developed by Subcommittee: C09.20, 2017. [37] American Society for Testing and Materials, Standard Test Method for Relative Density (Specific Gravity) and Absorption of Coarse Aggregate, ASTM C127-15, Book of Standards Volume: 04.02: Developed by Subcommittee: C09.20, 2015. [38] American Society for Testing and Materials, Standard Test Method for Sieve Analysis of Fine and Coarse Aggregates, ASTM C136/C136M– 14, Book of Standards Volume: 04.02: Developed by Subcommittee: C09.20, 2014. [39] American Society for Testing and Materials, Standard Test Method for Resistance to Degradation of Small-Size Coarse Aggregate by Abrasion and Impact in the Los Angeles Machine, ASTM C131/C131M – 14, Book of Standards Volume: 04.02: Developed by Subcommittee: C09.20, 2014. [40] American Society for Testing and Materials, Standard Test Method for Slump of Hydraulic-Cement Concrete. ASTM C143/C143M – 15a, Book of Standards Volume: 04.02: Developed by Subcommittee: C09.60, 2015. [41] American Society for Testing and Materials, Standard Test Method for Compressive Strength of Cylindrical Concrete Specimens, ASTM C39/C39M. Book of Standards Volume: 04.02: Developed by Subcommittee: C09.61, 2015. [42] American Society for Testing and Materials, Standard Test Method for Splitting Tensile Strength of Cylindrical Concrete Specimens. ASTM C496/C496M-17. Book of Standards Volume: 04.02: Developed by Subcommittee: C09.61, 2017. [43] American Society for Testing and Materials. Standard Test Method for Abrasion Resistance of Concrete or Mortar Surfaces by the Rotating-Cutter Method. ASTM C944/C944M - 12. Book of Standards Volume: 04.02: Developed by Subcommittee: C09.62, 2012. [44] American Society for Testing and Materials. Standard test method for density, absorption, and voids in hardened concrete. ASTM C642-13. Book of Standards Volume: 04.02: Developed by Subcommittee: C09.49, 2012. [45] American Society for Testing and Materials. Standard Test Method for Pulse Velocity Through Concrete. ASTM C597-16. Book of Standards Volume: 04.02: Developed by Subcommittee C09.64, 2016. [46] M. Eiamwijit, K. Pachana, S. Kaewpirom, U. Rattanasak, P. Chindaprasirt, Comparative study on morphology of ground sub-bituminus FBC fly ash geopolymeric material, Adv. Powder Technol. 26 (4) (2015) 1053–1057. [47] M. Abdullah, H. Kamarudin, M. Binhussain, R. Razak, Z. Yahya, Effect of Na2SiO3/NaOH ratios and NaOH molarities on compressive strength of fly-ashbased geopolymer, ACI Mater. J. (2012). Title no. 109-M48. [48] P. Risdanareni, J. Ekaputri, T. Triwulan, The influence of alkali activator concentration to mechanical properties of geopolymer concrete with trass as a filler, Mater. Sci. Forum 803 (2014) 125–134. [49] M. Morsy, S. Alsayed, Y. Al-Salloum, T. Almusallam, Effect of sodium silicate to sodium hydroxide ratios on strength and microstructure of fly ash geopolymer binder, Arabian J. Sci. Eng. 39 (6) (2014) 4333–4339. [50] F. Skvara, L. Kopecky, J. Nimeeek, Z. Bittnar, Microstructure of geopolymer materials based on fly ash, Ceram.-Silik. 50 (4) (2006) 208–215. [51] H. El-Sayed, E.-E.S. Abo, H. Khater, S. Hasanein, Resistance of alkali activated water-cooled slag geopolymer to sulphate attack, Ceram.-Silik. 55 (2) (2011) 153–160. [52] R. Arellano Aguilar, O. Burciaga Díaz, J.I. Escalante García, Lightweight concretes of activated metakaolin-fly ash binders, with blast furnace slag aggregates, Constr. Build. Mater. 24 (7) (2010) 1166–1175. [53] M.Y.J. Liu, U.J. Alengaram, M.Z. Jumaat, K.H. Mo, Evaluation of thermal conductivity, mechanical and transport properties of lightweight aggregate foamed geopolymer concrete, Energy Build. 72 (2014) 238–245. [54] B. Muñoz-Sánchez, M.J. Arévalo-Caballero, M.C. Pacheco-Menor, Influence of acetic acid and calcium hydroxide treatments of rubber waste on the properties of rubberized mortars, Mater. Struct. 50 (1) (2016) 75. [55] T.U. Mohammed, M.N. Rahman, Effect of types of aggregate and sand-toaggregate volume ratio on UPV in concrete, Constr. Build. Mater. 125 (2016) 832–841. [56] P.K. Sarker, S. Kelly, Z. Yao, Effect of fire exposure on cracking, spalling and residual strength of fly ash geopolymer concrete, Mater. Des. 63 (2014) 584– 592. [57] O.A. Abdulkareem Mustafa, A.M. Al Bakri, H. Kamarudin, I. Khairul Nizar, A.E.A. Saif, Effects of elevated temperatures on the thermal behavior and mechanical performance of fly ash geopolymer paste, mortar and lightweight concrete, Constr. Build. Mater. 50 (2014) 377–387. [58] R. Ghosh, S.P. Sagar, A. Kumar, S.K. Gupta, S. Kumar, Estimation of geopolymer concrete strength from ultrasonic pulse velocity (UPV) using high power pulser, J. Build. Eng. 16 (2018) 39–44. [59] S. Shankar, H.R. Joshi, Comparison of concrete properties determined by destructive and non-destructive tests, J. Inst. Eng. 10 (1) (2014) 130–139.