See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/326551906

Mechanical properties of ambient cured one-part hybrid OPC-geopolymer

concrete

Preprint in Construction and Building Materials · July 2018

DOI: 10.1016/j.conbuildmat.2018.07.160

CITATIONS READS

77 916

4 authors, including:

Mahya Askarian Professor Zhong Tao

Western Sydney University Western Sydney University
6 PUBLICATIONS 229 CITATIONS 223 PUBLICATIONS 11,782 CITATIONS

SEE PROFILE SEE PROFILE

Bijan Samali
Western Sydney University
376 PUBLICATIONS 8,007 CITATIONS

SEE PROFILE

All content following this page was uploaded by Professor Zhong Tao on 25 July 2018.

The user has requested enhancement of the downloaded file.

 This is the preprint of the following paper submitted to Construction and Building Materials

Askarian M., Tao Z., Adam G., Samali B. (2018). Mechanical properties of ambient cured one-part
hybrid OPC-geopolymer concrete. Construction and Building Materials (accepted for publication).

1 Mechanical properties of ambient cured one-part hybrid OPC-

2 geopolymer concrete

3 Mahya Askarian*, Zhong Tao, Georgius Adam, Bijan Samali

4 Centre for Infrastructure Engineering, Western Sydney University, Penrith, NSW 2751, Australia

5 Abstract

6 To produce geopolymers, a large amount of highly corrosive and hygroscopic alkaline activators, i.e,
7 sodium hydroxide and potassium hydroxide, are commonly used. In this study, one-part hybrid
8 ordinary Portland cement (OPC)-geopolymer concrete mixes were developed, in which solid
9 potassium carbonate (7.5 wt.% of the total geopolymeric raw materials) was used as the main
10 activator and the OPC as a source of silicate and poly silicate was blended with geopolymeric raw
11 materials (fly ash and ground granulated blast furnace slag) in different proportions. The influence
12 of OPC content on the workability, setting time, compressive strength development and
13 microstructure of the concrete mixes was investigated. It was found that the inclusion of OPC in
14 geopolymer mixes reduced the workability and setting time but significantly increased the early age
15 and ultimate compressive strength due to the quick reaction of OPC with alkali activators. By the
16 inclusion of just 10% OPC in the binder system, the compressive strength at 28 days increased from
17 11.4 MPa to 33.4 MPa. Furthermore, control OPC concrete mixes were also prepared and tested. It
18 was found that the percentage of strength increase at 28 days due to alkali activation decreased from
19 82.5% to 24.4% as the OPC content increased from 10% to 60%. Microstructural analysis was also
20 conducted, which confirmed the coexistence of geopolymeric and CSH gels in the one-part hybrid
21 OPC-geopolymer binder system.

22 Keywords: Ambient curing, hybrid one-part OPC-geopolymer concrete, potassium carbonate,

23 calcium hydroxide, sodium silicate, microstructure.

24 * Correspondence author.

25 E-mail address: [email protected] (M. Askarian)

1
26 1. Introduction

27 Geopolymeric materials have emerged as sustainable alternatives to Portland cement for a few

28 decades. The term “geopolymer” was initially introduced by Davidovits [1]. Conventional forms of

29 geopolymers are synthesised from a three-part mix, including high alkaline solutions of sodium

30 hydroxide and sodium silicate with suitable concentration and geological origin materials such as

31 metakaolin or industrial by-products such as fly ash and ground granulated blast furnace slag as a

32 source of aluminosilicate [2]. To obtain suitable mechanical properties, heat treatment is often

33 required during the geopolymerisation process to accelerate the reaction [3]. The outstanding

34 properties of geopolymers such as high resistance to elevated temperatures and fire, high resistance

35 to acids, salts and other corrosive materials, along with their environmentally friendly nature have

36 resulted in considerable research and practical applications [4-6].

37 However, there are also two major drawbacks associated with the production of the

38 conventional geopolymer concrete. Firstly, dealing with high concentration of corrosive and viscous

39 alkaline solutions represents one of the main barriers for commercial applications. Secondly, the

40 need of heat curing prevents the wide application of fly ash based geopolymer concrete since this

41 type of concrete cannot be cured at ambient condition [7]. The geopolymer systems are usually

42 subjected to heat curing for hours or even days at a temperature higher than 50 [8].

43 In order to improve the geopolymerisation at room temperature, researchers incorporated

44 various additives, such as slag, high calcium fly ash, calcium hydroxide and OPC, as additional

45 calcium sources in geopolymer mixes [9-17]. The calcium sources were found to play an important

46 role to shorten the setting time and improve the mechanical properties. For example, Nath and

47 Sarker [9] replaced up to 12% of fly ash by OPC as a source of calcium silicate. In this experiment,

48 alkaline activator solution of combined sodium silicate and sodium hydroxide was used, whilst the

49 curing regime was ambient curing for all specimens. The authors concluded that the optimum

50 mixture was achieved when OPC was used to replace 5% of the total binder and the ratio of liquid

51 activator to the binder was 0.4. The mixture was cured at ambient temperature and set reasonably

2
52 well compared to OPC concrete. Suwan and Fan [16] evaluated the effects of OPC replacement and

53 mixing procedure on properties of geopolymers. The replacement ratio of fly ash with OPC varied

54 from 5% to 70%, and its influence on the setting time, compressive strength and microstructure of

55 the geopolymer was studied. Meanwhile, the sequence of adding fly ash, sodium silicate and

56 sodium hydroxide was changed in the specimen preparation. Results showed that by increasing the

57 amount of OPC from 5% to 70%, the setting time of geopolymer was significantly reduced from 24

58 h to 0.5 h while the early strength at 3 days increased from 0 to 21.91 MPa. It was also reported that

59 the mixing sequence could directly affect the properties of geopolymer. A total of three mixing

60 methods were adopted. In the first mixing method, fly ash was initially mixed with sodium

61 hydroxide solution for 90 s before the sodium silicate was added. In the second method, sodium

62 silicate and sodium hydroxide solutions were premixed and left over night and then fly ash was

63 added during mixing. In the last method, fly ash, sodium hydroxide and sodium silicate were dry-

64 mixed before adding water. It was found that only the third mixing sequence ensured the setting of

65 the mix within the first 24 h. However, the mixes prepared using the first and second methods

66 showed higher early strength as a result of enhanced chemical reaction. In another study conducted

67 by Shehab et al. [17], mechanical properties of fly ash based geopolymer concrete mixes with a

68 partial replacement of OPC were investigated by varying the OPC replacement ratio and activator

69 solution concentration. The authors found that the mix with 50% OPC replacement had the highest

70 compressive strength, splitting tensile strength, bond strength and flexural strength at 28 days.

71 One-part “just add water” geopolymer mixes are a newly developed class of geopolymers and

72 their production is almost similar to that of conventional OPC concrete [18]. The development of

73 one-part geopolymers aims to promote the large scale application of geopolymers in construction

74 industry. For most recently developed one-part mixes, aluminosilicate precursors were blended with

75 solid activators and reactions began once water was added to the solid mix. Hajimohammadi et al.

76 [18-20] conducted studies to address the kinetics of geopolymerisation in one-part mixes. In their

77 experiments, geothermal silica was utilised as a source of solid silica and solid sodium aluminate as

3
 78 a source of alkali and alumina. The water content was found to play a major role in the formation of

79 crystallites as a result of rapid dissolution of the alumina and slow dissolution of the silica sources

80 in low water conditions. Furthermore, the early availability of silica and alumina is essential for

81 geopolymer development. However, in their research, the procedure of purifying the raw

82 geothermal silica was rather complicated and also the samples required heat curing at 40 for 2

83 weeks which limits the large scale application of this type of concrete. They also recently evaluated

84 the nanostructure, microstructure and mechanical properties of the one-part fly ash-based binders at

85 different H2O/Na2O molar ratios [21]. The one-part mix with Si/Al molar ratio of 3.6 and

86 H2O/Na2O molar ratio of 12 gave a compressive strength of 65 MPa after 3 weeks. Attempts by

87 other research groups have also been made to synthesise one-part mixes under ambient temperature

88 curing conditions by incorporating commonly used aluminosilicate sources with solid activators.

89 Yang et al. [22] and Yang and Song [23] performed some studies in which pure fly ash and pure

90 slag were activated using either sodium silicate powder or combination of sodium silicate and

91 sodium hydroxide powders at ambient temperature. When 100% slag was used, the 28-day

92 compressive strength could reach 50 MPa; but the corresponding figure was just 9.45 MPa for pure

93 fly ash systems. To avoid the use of highly hygroscopic and corrosive sodium hydroxide powder,

94 Nematollahi et al. [24] attempted to improve the strength of one-part geopolymer mixes by using

95 different grades of sodium silicate activators, whilst different combinations of fly ash, slag,

96 hydrated lime and three different grades of sodium silicate and a combination of sodium silicate and

97 sodium hydroxide powders were used. The authors reported the feasibility of synthesising one-part

98 geopolymers using solid silicate activators, and obtained a compressive strength of 37 MPa under

99 ambient temperature curing conditions.

100 Previous efforts to make one-part geopolymers, however, were mainly focused on geopolymer

101 pastes and mortars. To promote the application of one-part geopolymers in engineering practice,

102 further studies on geopolymer concrete are required since the coarse aggregate strikingly affects the

103 mixture properties. Moreover, the commonly used solid activators are hygroscopic and in case of

4
104 their movement to the surface of concrete during curing or in service, there could be a tendency

105 toward efflorescence, high permeability and water absorption [24]. It has been found that adding

106 OPC in one-part geopolymer systems in the presence of non-hygroscopic alkaline salts such as

107 sodium carbonate and potassium carbonate as activators remarkably enhanced the compressive

108 strength and accelerated the setting time at ambient temperature [25]. This type of blended OPC-

109 geopolymer is relatively cheap and has currently been produced on commercial scale. Typically, it

110 comprises 80% OPC and 20% of geocement, where the calcium silicate and alumina-silicates as

111 geopolymeric materials are activated by potassium carbonate and retarded with citric acid. Also this

112 blended cement-geopolymer has been recognised in the construction industry for its potential to

113 rapidly achieve very high early strength [26]. Moreover, using potassium instead of sodium was

114 reported to reduce efflorescence as potassium is more strongly bound to the aluminosilicate gel

115 framework and potassium carbonate crystals are usually less visually evident than their sodium

116 counterparts [27].

117 The aim of the present study is to develop ambient temperature cured one-part hybrid OPC-

118 geopolymer concrete mixes using solid potassium carbonate as the main activator. It is expected

119 that the one-part concrete mixtures can be directly used to cast structural members. In this regard,

120 one-part hybrid OPC-geopolymer mixes with different replacement ratios of geopolymeric raw

121 materials (fly ash and slag) by OPC ranging from 10% to 60% were activated by solid potassium

122 carbonate. To evaluate the contribution of alkaline activation to strength development, control OPC

123 mixes were also made, which had the same proportions as the hybrid OPC-geopolymer mixes but

124 without any activator. As the mixes with high percentage of OPC (30-60 wt.%) had a high risk of

125 flash setting as a result of rapid chemical reaction between calcium silicates (in the OPC and slag)

126 and alkaline activators [16], citric acid was used as a retarder. The influence of additional calcium

127 hydroxide and sodium silicate in the activator on the strength development is also evaluated.

128 Furthermore, the influence of OPC content and the use of solid potassium carbonate on the

129 workability, setting time, compressive strength development and microstructure of the concrete

5
130 mixes are investigated. Microstructural analyses were also performed to explain the observed

131 behaviour.

132
133 2. Experimental procedure

134 2.1. Materials

135 In this study, the cement used meets the requirement of ASTM C150 Type 1 cement. The class

136 F fly ash used in this study is mainly glassy with some mullite and quartz crystalline inclusion. The

137 granulated blast furnace is supplied with pre-blended gypsum. The results of X-ray Fluorescence

138 (XRF) analyses of the raw materials are presented in Table 1. Natural river sand with a specific

139 gravity of 2.6 and water absorption of 3.5% was used as fine aggregate, whereas crushed basalt with

140 a maximum nominal size of 10 mm, a specific gravity of 2.8 and water absorption of 1.6% was used

141 as coarse aggregate. Prior to mixing, all aggregates were dried in an oven at 105 for a period of

142 48 h to remove any moisture.

143 Table 1 Chemical composition of OPC, fly ash and slag determined by XRF.
a
Material

OPC (%) 20.2 4.9 2.6 62.7 2.0 0.2 0.4 2.2  2.0

Fly ash (%) 50.3 27.56 12.68 2.7 1.3 0.6 0.7 0.36 1.42 0.67

Slag (%) 32.5 13.0 0.55 43.2 4.7 0.2 0.3 4.3 0.51 0.08

144 a
Loss of ignition

145 This study has used three types of solid activators, including potassium carbonate, calcium

146 hydroxide and sodium silicate. Powder form potassium carbonate denoted as “K2CO3” with 99.99%

147 purity, an industrial grade of calcium hydroxide Ca(OH)2 commonly used in the construction

148 industry and sodium silicate-based activator denoted as “Na2SiO3-Anhydrous” were used. The

149 supplied sodium silicate has the chemical composition: 51% Na2O, 46% SiO2 and 3% H2O with a

150 modulus (Ms = SiO2/Na2O) of 0.9.

151 2.2. Concrete mix proportions and specimen preparation

152 The proportions of one-part hybrid OPC-geopolymer concrete mixes together with their control

6
153 mixes (with no activator) are illustrated in Table 2. All the hybrid OPC-geopolymer mixes included

154 solid activator of potassium carbonate with a weight of 7.5% of the total geopolymeric raw

155 materials; this amount was determined based on several trial tests. The hybrid OPC-geopolymer

156 mixes were designated with the percentage of OPC in the binder. Accordingly, for mixes GP60,

157 GP40, GP30, GP20, GP10 and GP0, the amounts of OPC were 60 wt.%, 40 wt.%, 30 wt.%, 20

158 wt.%,15 wt.%, 10 wt.% and 0% of the total binder, respectively. For a corresponding control mix,

159 “GP” in the designation is simply replaced by “C”. For example, C60 without activator is the

160 control mix of GP60. It should be noted that the replacement of fly ash and slag by OPC was based

161 on the weight of the materials and the difference in material densities was not considered in this

162 study. It is estimated that the material replacement caused a variation in the total mixture volume up

163 to 0.9%, which is negligible. This simplification was also adopted in [28, 29] to adjust their mix

164 design. Furthermore, three additional concrete mixes A13 were prepared to evaluate the influence

165 of combining calcium hydroxide or sodium silicate with potassium carbonate as activator on the

166 strength development. The OPC percentage was kept constant at 10% by weight of the total binder

167 in these concrete mixes. Apart from the use of potassium carbonate (27 kg/m3), additional sodium

168 silicate and calcium hydroxide at a weight of 2.5 wt.%, 5 wt.% and 2.5 wt.% of the total

169 geopolymeric raw materials were added to mixes A1, A2 and A3, respectively. The amount of

170 sodium silicate in mix A2 was 18 kg/m3, which is two times that in mix A1.

171 All the concrete mixes in this study had the same amount of binder and aggregate and a

172 constant water to binder (w/b) ratio of 0.3. The aggregates proportions shown in Table 2 were based

173 on the saturated surface dry (SSD) condition. Furthermore, the slag content of all mixes was kept

174 constant at 10 wt.% of the total geopolymeric raw materials, although future research might be

175 conducted to investigate the variation of slag content as well.

176 All the mixes were prepared using a vertical pan mixer. Solid ingredients, including the OPC,

177 fly ash, slag, solid activator, citric acid, coarse aggregate and sand were added to the mixer and dry-

178 mixed thoroughly for 3 min before adding water. The water and superplasticiser were then added

7
179 gradually and mixing was continued for another 4-6 min until a uniform mix was achieved. The

180 freshly mixed concrete was then cast in 100 × 200 mm cylindrical moulds in two layers and each

181 layer was compacted on a vibrating table to achieve proper consolidation. The samples were

182 demoulded 24 h after casting and were wrapped with plastic sheet. Then they were stored in a

183 controlled temperature of 2023 and relative humidity of 65±10% until testing.

184 In addition to the concrete mixes in Table 2, one-part hybrid OPC-geopolymer pastes were also

185 prepared with the same proportions of binder and activator to measure the initial and final setting

186 times and to be used for microstructural analysis. The pastes were mixed manually in a laboratory

187 bowl until a uniform mixture was achieved.

188 Table 2 Mix proportions of one-part geopolymer concrete (kg/m3).

Mix No. Designation Coarse Sand OPC Fly ash Slag Activator Citric acid Water Superplasticiser
aggregate
1 GP60a 1220 635 240 144 16 12 0.96 120 6

2 C60 1220 635 240 144 16 - 0.96 120 6

3 GP40a 1220 635 160 216 24 18 0.64 120 6

4 C40 1220 635 160 216 24 - 0.64 120 6

5 GP30a 1220 635 120 252 28 21 0.48 120 6

6 C30 1220 635 120 252 28 - 0.48 120 6

7 GP20a 1220 635 80 288 32 24  120 6

8 C20 1220 635 80 288 32 - - 120 6

9 GP10a 1220 635 40 324 36 27  120 6

10 C10 1220 635 40 324 36 - - 120 6

11 GP0a 1220 635  360 40 30  120 6

12 A1b 1220 635 40 324 36 36  120 6

13 A2b 1220 635 40 324 36 45  120 6

14 A3c 1220 635 40 324 36 36  120 6

189 a
Potassium carbonate (powder form) was used as activator

190 b
Potassium carbonate and sodium silicate (powder form) were combinedly used as activator

191 c
Potassium carbonate and calcium hydroxide (powder form) were combinedly used as activator

192
193
194

8
195 2.3. Testing procedure

196 The workability of fresh concrete mixes was tested by slump test immediately after mixing

197 [30]. The setting time of the paste was measured at a temperature of 2123 using a Vicat

198 apparatus in accordance with ASTM C 191-08 [31]. Compressive strength tests were conducted at

199 3, 7 and 28 days of age using a universal testing machine. Cylindrical specimens were tested at a

200 loading rate of 20±2 MPa/min according to AS1012.9 [32]. All the cylinders were ground flat on

201 both ends to ensure having smooth and parallel top and bottom faces prior to performing the tests.

202 To observe the microstructure of the hybrid OPC-geopolymer paste samples, a JEOL 6510LV

203 scanning electron microscope (SEM) was utilised. The elemental composition of samples was

204 observed using a Moran Scientific microanalysis EDS (energy dispersive spectroscopy) system.

205 XRD patterns were achieved implying a Bruker D8 Advance Power Diffractometer. Diffraction

206 analysis were made from 15 to 60 2 using copper K radiation. The excitation voltage was 40 kV at

207 40 mA, counting was 5 s.

208 3. Results and discussion

209 3.1. Behaviour of fresh mixes

210 Slump tests were conducted for the one-part hybrid OPC-geopolymer concrete mixes and the

211 results are presented in Fig. 1. It can be observed that OPC content had a significant effect on the

212 workability of the one-part mixes. The geopolymer concrete mix (GP0) that contains no OPC

213 exhibited the highest workability as indicated by the slump value of 122 mm. A decrease in slump

214 was found with increasing OPC content when the ratio of activator to geopolymeric raw materials

215 was kept at 7.5%. The decrease was 16% and 22% for mixes with 10 wt.% and 20 wt.% OPC

216 addition compared to GP0 (without OPC addition). The slump dropped significantly for GP30 with

217 30 wt.% OPC. A very low slump of 30 mm was observed for mix GP60 with 60 wt.% OPC. The

218 decrease in slump value with increasing OPC could be due to the rapid reaction of OPC with the

219 alkaline activator. Addition of superplasticiser might be helpful to improve the workability of

220 hybrid mixes with high OPC content.

9
 140

120

100

Slump (mm) 80

60

40

20

0
GP0 GP10 GP20 GP30 GP40 GP60
221

222 Fig. 1. Effect of OPC inclusion on the slump.

223

224 Initial and final setting times were measured for the corresponding pastes of mixes GP60,

225 GP40, GP30, GP20, GP10, and GP0, and the obtained results are shown in Fig. 2. For the

226 convenience of description, the mix designations presented in Table 2 are also used for paste

227 samples in this section. Mix GP0 which contained only fly ash and slag as binder did not set within

228 the first 24 h as a result of slow rate of chemical reaction at ambient temperature. By incorporating

229 OPC in the mixes, both the initial and final setting times were significantly decreased which is in

230 agreement with previous studies [15, 16]. It has been reported that the initial setting times were only

231 22 and 11 min when the OPC to the total binder ratios were increased to 50% and 70%, respectively

232 [16]. The quick setting of the OPC-rich mixes would prevent transportation and proper casting of

233 the concrete. In this study, different amounts of citric acid were added to mixes GP60, GP40 and

234 GP30. The OPC hydration in these mixes was effectively retarded and the setting time was

235 increased because of the use of citric acid. For mixes with 60% (GP60), 40% (GP40) and 30%

236 (GP30) of OPC in the binder, the initial setting times are 30, 39, and 44 min and the final setting

237 times are 53, 75, and 107 min, respectively. In contrast, the mixes with a lower content of OPC

238 generally required longer setting times although no citric acid was used. The initial setting times are

239 61 and 86 min and the final setting times are 102 and 139 min for mixes with 20% (GP20) and 10%

240 (GP10) of OPC in the binder, respectively. It can be concluded that the incorporation of OPC in the
10
241 mixture can accelerate the setting of one-part geopolymer concrete. If the OPC content in the binder

242 is 30% or above, a suitable amount of citric acid can be added to retard the OPC hydration.

160

140 Initial

120 Final
Setting time (Min)

100

80

60

40

> 24 h
20

0
GP60 GP40 GP30 GP20 GP10 GP0
243

244 Fig. 2. Setting times of one-part pastes.

245 3.2. Compressive strength

246 The measured compressive strengths at 1, 7 and 28 days are presented in Table 3 for all of the

247 hybrid OPC-geopolymer concrete and their control mixes. Mix GP0 without OPC was found to be

248 very weak and only achieved a low strength of 11.4 MPa at 28 days. In contrast, the increase in

249 OPC percentage significantly improved the corresponding compressive strength at various ages for

250 both hybrid OPC-geopolymer and control mixes. In general, the compressive strengths of the hybrid

251 OPC-geopolymer concrete mixes are higher than those of the control mixes without activator

252 regardless of their OPC content. This strength enhancement is even more significant at early ages of

253 1 day and 7 days. At 28 days, mixes GP10, GP20, GP30, GP40 and GP60 achieved compressive

254 strengths of 33.4, 39.7, 50.3, 52.5 and 55.0 MPa, respectively. In contrast, the control mixes C10,

255 C20, C30, C40 and C60 achieved lower compressive strengths of 18.3, 24.3, 33.0, 38.8 and 44.2

256 MPa, respectively. This comparison confirms the contribution of alkali activation to strength

257 development of the hybrid OPC-geopolymer concrete. Meanwhile, it is found that the percentage of

11
258 strength increase at 28 days due to alkali activation decreased from 82.5% to 24.4% as the OPC

259 content increased from 10% to 60%.

260 The quick hardening of the geopolymer mixes in the presence of calcium compounds with

261 respect to the control mixes could be attributed to the rapid reaction of OPC with the alkaline

262 activator, which consumes the expelled water from the geopolymerisation process and consequently

263 promote this process [13, 15]. The hydration heat of OPC might also accelerate the alkali activation,

264 leading to increased strength development. Clearly, when the OPC content in the binder reaches

265 10% or higher, the 28-day compressive strength of the corresponding concrete exceeds 32 MPa,

266 ensuring the structural use of this type of concrete.

267 Table 3 Compressive strengths of geopolymer concrete mixes at 1, 7, 28 and 56 days (MPa)

Mix No. Designation 1 day 7 days 28 days

1 GP60 27.3 0.1 45.3 0.3 55.0 1.8

2 C60 7.3 0.9 28.4 1.4 44.2 1.5

3 GP40 23.8 1.4 44 2.6 52.5 0.4

4 C40 6.6 0.8 25.6 0.2 38.8 1.2

5 GP30 25.3 0.3 45.6 0.5 50.3 1.3

6 C30 5.8 1.0 24.2 0.5 33 0.6

7 GP20 17.8 1.1 31.3 0.8 39.7 0.8

8 C20 4.1 1.3 17.8 1.2 24.3 0.6

9 GP10 13.2 1.9 26.7 0.4 33.4 0.6

10 C10 3.2 0.4 12.8 1.2 18.3 2.8

11 GP0 0 0 2.3 0.7 11.4 0.9

268

269 The current research also indicates that, when the OPC content in the binder reaches 10% or

270 higher, the compressive strength of a hybrid OPC-geopolymer mix is comparable to that of

271 conventional geopolymer using alkaline solution of sodium silicate, sodium hydroxide or their

272 combination [33, 34]. For example, mix GP30 with 30% of OPC in the binder achieved a

273 compressive strength of 25.3 MPa at 1 day and 50.3 MPa at 28 days. It is worth noting that this mix

274 only used potassium carbonate powder as activator at 7.5 wt.% of the total geopolymeric raw
12
275 materials. In contrast, the corresponding percentage in conventional geopolymer normally ranges

276 from 13 wt.% to 16 wt.% [35]. As can be seen, the current one-part hybrid OPC-geopolymer

277 concrete used much less activator compared with the conventional geopolymer. Therefore, the

278 developed hybrid mixture is more feasible and economical for on-site applications.

279 The influence of additional calcium hydroxide and sodium silicate in the activator on the

280 strength development is demonstrated in Fig. 3. Compared with mix GP10, the addition of sodium

281 silicate in mixes A1 or A2 has only moderate influence on the concrete strength. Although the

282 amount of sodium silicate was doubled in A2 as compared to A1, the strength increase of 4.2% at

283 28 days was marginal. It seems that sodium silicate has no obvious influence on the alkali reaction

284 in the hybrid OPC-geopolymer concrete. However, the strength increase is more significant when

285 additional calcium hydroxide (2.5 wt.% of the total geopolymeric raw materials) was added to mix

286 A3, which is attributable to the increased pH value of the solution to promote geopolymerisation.

287 Compared with the 28-day strength of GP10 (33.4 MPa), the corresponding strength of mix A3 was

288 increased to 37.6 MPa.

289
40

35
Compressive strength (MPa)

30

25

20
A1
15
A2
10 A3
GP10
5

0
0 5 10 15 20 25 30
Age (day)
290
291 Fig. 3. Influence of calcium hydroxide and sodium silicate on strength development

292

13
293 3.3. Microstructural analysis

294 The effect of OPC content on the microstructure of the one-part mixes was studied. Fig. 4

295 presents the microstructural images of the samples at 28 days of age using backscattered electron

296 (BSE). As can be seen in the SEM micrographs, increasing the OPC content in the mix leads to a

297 more compact and less porous structure. Meanwhile, there are less unreacted spherical fly ash

298 particles in the matrix. Because of the inclusion of a large amount of OPC, mix GP60 shown in Fig.

299 4 also comprises both unreacted and partially reacted OPC particles as confirmed by the EDS

300 results. Komnitsas [36] also reported that, when the OPC concentration is high, microstructure

301 porosity decreases for the formation of amorphous Ca-Al-Si gels, leading to increased strength. It

302 can also be seen in Fig. 4 that unreacted or partially reacted OPC particles are rarely seen in the

303 paste having 30% (GP30) or lower amount of OPC, which shows their participation in the

304 formation of either geopolymer gel or a new gel. The microstructure of ambient cured GP0 contains

305 both unreacted fly ash particles and geopolymeric matrix, demonstrating a loose and amorphous

306 structure. The large amount of unreacted fly ash particles explain the very low compressive strength

307 as confirmed by the compression tests. GP30 has a more compact microstructure with less

308 unreacted fly ash particles with respect to GP0, GP10 and GP20, although it has a more porous

309 microstructure than GP60 and GP40. For GP30, the major product of this low calcium content

310 structure is geopolymeric gel surrounded by partial CSH gel in the entire network as was also

311 observed previously [16].

312 Overall, the SEM images depict the coexistence of geopolymeric gel and CSH gel phases,

313 showing a hybrid product like calcium alumino-silicate hydrate (CASH) as suggested by Garcia-

314 Lodeiro et al. [37]. This will be further confirmed from the C/S ratios measured as follows.

14
315

316 Fig. 4. BSE images of hybrid OPC-geopolymer mixes

317 Quantitative EDS analysis was performed on the hybrid OPC-geopolymer paste samples at 28

318 days to evaluate the reaction products of mixes. The oxide molar ratios presented in Table 4 are the

319 averages measured at 5 different spots. The EDS analysis indicates that the CaO-to-SiO2 (C/S) ratio

320 decreases with decreasing OPC content in the hybrid OPC-geopolymer mixes. The C/S ratio is 1.1

321 in GP60 and decreases to 0.23 for GP10, whereas the corresponding ratio of GP0 is only 0.06. By

322 adding calcium hydroxide into the mix, the C/S ratio slightly increases from 0.23 in GP10 to 0.29 in

323 A3. Obviously, a mix with a high C/S ratio has a high potential for the formation of CSH, leading to
15
324 increased early strength. For neat OPC paste, the C/S ratio was reported to vary in the range of 1.2-

325 2.3 [38], which is higher than those found in this study (0.18-1.1). It can be inferred that the nature

326 of CSH formed in the hybrid mixes is different from that of normal CSH formed in OPC.

327 Table 4 Oxide molar ratios in different mixes

Mix No. Designation SiO2/Al2O3 CaO/SiO2 CaO/Al2O3 K2O/SiO2 K2O/Al2O3

1 GP60 2.51 1.10 2.78 0.16 0.41

3 GP40 2.50 0.72 1.81 0.20 0.50

5 GP30 2.47 0.53 1.32 0.22 0.54

7 GP20 2.41 0.37 0.89 0.23 0.55

9 GP10 2.39 0.23 0.56 0.23 0.55

10 GP0 2.40 0.06 0.14 0.24 0.57

12 A1 2.52 0.19 0.46 0.23 0.57

13 A2 2.53 0.18 0.47 0.24 0.60

14 A3 2.47 0.29 0.72 0.23 0.56

328

329 The XRD graphs of hybrid OPC-geopolymer pastes are plotted in Fig. 5. In general, the

330 hydrated OPC contains major crystalline phases of CSH while geopolymer concrete is a mixture of

331 crystalline phases of Quartz and Mullite and amorphous phases observed in the region of

332 . Previous studies showed that CSH and NASH (main reaction product of class F fly ash

333 activation) gels are compatible and precipitation of a mix of gels is responsible for the setting and

334 hardening of this type of hybrid concrete [39]. The OH- concentration and the mix composition

335 were reported as controlling factors to form CSH/semi–crystalline phase or NASH and CASH

336 (main product of fly ash activation reaction) [39]. Peaks due to the crystalline components of Quartz

337 and Mullite from the fly ash were observed in all the pastes and the intensity of their peaks

338 decreases with increasing OPC content in the mix. On the other hand, CSH, CASH, Nepheline,

339 Hatrurite, Calcite and Portlandite appear and their intensity consistently increase with increasing

340 OPC content. In all of the mixes, a broad and amorphous hump was observed, which indicates the

341 coexistence of geopolymeric and hydrated OPC products in the binder system. The XRD results for

342 mixes GP30, GP20 and GP10 differ from the results achieved for mixes GP60 and GP40 as
16
343 Portlandite does not appear as a hydration product in GP30, GP20 and GP10. In general, it can be

344 concluded that the one-part hybrid OPC-geopolymer mixes had different reaction products with

345 respect to OPC and conventional geopolymers. This is consistent with previous research findings

346 [15, 16]. The initial reaction of calcium compounds with activator resulted in the quick formation of

347 both amorphous CSH and semi-crystalline geopolymeric network. In the later stage, the CSH gel

348 increasingly formed due to the decrease of alkalinity in the mix. The results of SEM, EDS and XRD

349 confirm that the hybrid OPC-geopolymer mix produces a mixture of cross-linked network of

350 geopolymeric and CSH gels in the binder system.

351

352
353 Fig. 5. XRD analysis of various hybrid OPC-geopolymer systems at the age of 28 days.

354

355 4. Conclusions

356 This study aimed to develop ambient temperature cured one-part hybrid OPC-geopolymer

357 concrete mixes using ordinary Portland cement (1060 wt.% of the total binder) as initial binder in

17
358 combination with solid potassium carbonate (7.5 wt.% of the total geopolymetric raw materials) as

359 alkaline activator. The following conclusions can be drawn within the scope of this study:

360 1. Increasing OPC content decreased workability of one-part geopolymer concrete mixes. The

361 geopolymer concrete mix (GP0) obtained the highest slump (122 mm) while the mix

362 containing 60% OPC (GP60) achieved the lowest value of 30 mm. The decrease in slump

363 value with increasing OPC could be due to the rapid reaction of OPC with the alkaline

364 activator.

365 2. By incorporating OPC in the mixes with potassium carbonate, both the initial and final

366 setting times were significantly decreased as the presence of OPC accelerated the

367 geopolymerisation reaction. In contrast, the mix containing only fly ash and slag as binder

368 did not set within the first 24 h as a result of slow rate of chemical reaction at ambient

369 temperature. If the OPC content in the binder is 30% or above, a suitable amount of citric

370 acid can be added to retard the OPC hydration.

371 3. The compressive strengths of the hybrid OPC-geopolymer concrete mixes are higher than

372 those of the control mixes without activator regardless of their OPC content. This strength

373 enhancement is even more significant at early ages of 1 day and 7 days. The percentage of

374 strength increase at 28 days due to alkali activation decreased from 82.5% to 24.4% as the

375 OPC content increased from 10% to 60%. When the OPC content in the binder reaches 10%

376 or higher, the 28-day compressive strength of the corresponding concrete exceeds 32 MPa,

377 ensuring the structural use of this type of concrete.

378 4. The type of activator has some influence on the concrete strength. An increase was found

379 when additional calcium hydroxide (2.5 wt.% of the total geopolymeric raw materials) was

380 used. However, while additional sodium silicate (2.5 wt.% and 5 wt.% of the total

381 geopolymeric raw materials) was added, the increase in compressive strength was moderate.

382 Further research is needed to find out the optimised amount and combinations of activators

383 to be used in the mix of the one-part hybrid OPC-geopolymer concrete.

18
384 5. Microstructural investigation showed that hybrid OPC-geopolymer mixes had different final

385 product compare to OPC and geopolymers. The results of SEM, EDS and XRD confirm that

386 the hybrid OPC-geopolymer mix produces a mixture of cross-linked network of

387 geopolymeric and CSH gels in the binder system, which allows the concrete to develop high

388 early and ultimate strength.

389 Overall, the developed one-part hybrid OPC-geopolymer concrete could be a suitable material

390 for on-site operation as the need for heat curing and the use of highly alkaline solutions have been

391 eliminated. The setting time, workability, compressive strength and microstructure properties can be

392 controlled by adjusting the OPC content and activator concentrations. Further investigations could

393 be conducted to study other properties of the hybrid OPC-geopolymer concrete, such as shrinkage,

394 creep, thermal properties and durability, etc.

395 Acknowledgements

396 The authors would like to acknowledge the contributions from the laboratory staff, Mr Robert

397 Marshal and Mr Zac White. The authors also would like to thank the WSU Advanced Materials

398 Characterisation Facility for the support of its staff.

399 References

400 [1] Davidovits, J., Geopolymers: inorganic polymeric new materials. Journal of Thermal Analysis and calorimetry,

401 1991. 37(8): p. 1633-1656.

402 [2] Feng, D., J.L. Provis, and J.S. Deventer, Thermal activation of albite for the synthesis of one-part mix

403 geopolymers. Journal of the American Ceramic Society, 2012. 95(2): p. 565-572.
404 [3] Noushini, A., M. Babaee, and A. Castel, Suitability of heat-cured low-calcium fly ash-based geopolymer

405 concrete for precast applications. Magazine of Concrete Research, 2016. 68(4): p. 163-177.
406 [4] van Deventer, J.S., Provis, J. L., Duxson, P., and Brice, D. G., Chemical research and climate change as drivers

407 in the commercial adoption of alkali activated materials. Waste and Biomass Valorization, 2010. 1(1): p. 145-
408 155.
409 [5] Vickers, L., Pan, Z., Tao, Z., and Van Riessen, A., In situ elevated temperature testing of fly ash based

410 geopolymer composites. Materials, 2016. 9(6): p. 445.

411 [6] Pan, Z., Tao, Z., Cao, Y. F., Wuhrer, R., and Murphy, T., Compressive strength and microstructure of alkali-

412 activated fly ash/slag binders at high temperature. Cement and Concrete Composites, 2018. 86: p. 9-18.
19
413 [7] Hardjito, D., C.C. Cheak, and C.H.L. Ing, Strength and setting times of low calcium fly ash-based geopolymer

414 mortar. Modern applied science, 2008. 2(4): p. 3.

415 [8] Duxson, P., Fernández-Jiménez, A., Provis, J. L., Lukey, G. C., Palomo, A., and Van Deventer, J. S. J.,

416 Geopolymer technology: the current state of the art. Journal of Materials Science, 2007. 42(9): p. 2917-2933.
417 [9] Nath P, Sarker PK., Geopolymer concrete for ambient curing condition, the past, present and future of

418 structural engineering. In: Proceedings of the Australasian Structural Engineering Conference, Perth,
419 Western Australia; July 11-13, 2012.
420 [10] Palomo, A., Fernández-Jiménez, A., Kovalchuk, G., Ordoñez, L. M., and Naranjo, M. C., OPC-fly ash

421 cementitious systems: study of gel binders produced during alkaline hydration. Journal of Materials Science,
422 2007. 42(9): p. 2958-2966.
423 [11] Chindaprasirt, P., Chareerat, T., Hatanaka, S., and Cao, T., High-strength geopolymer using fine high-calcium

424 fly ash. Journal of Materials in Civil Engineering, 2010. 23(3): p. 264-270.
425 [12] Canfield, G.M., Eichler, J., Griffith, K., and Hearn, J. D., The role of calcium in blended fly ash geopolymers.

426 Journal of Materials Science, 2014. 49(17): p. 5922-5933.

427 [13] Assi, L., Ghahari, S., Deaver, E. E., Leaphart, D., and Ziehl, P., Improvement of the early and final compressive

428 strength of fly ash-based geopolymer concrete at ambient conditions. Construction and Building Materials,
429 2016. 123: p. 806-813.
430 [14] Aliabdo, A.A., A.E.M.A. Elmoaty, and H.A. Salem, Effect of cement addition, solution resting time and curing

431 characteristics on fly ash based geopolymer concrete performance. Construction and Building Materials,
432 2016. 123: p. 581-593.
433 [15] Nath, P. and P.K. Sarker, Use of OPC to improve setting and early strength properties of low calcium fly ash

434 geopolymer concrete cured at room temperature. Cement and Concrete Composites, 2015. 55: p. 205-214.
435 [16] Suwan, T. and M. Fan, Influence of OPC replacement and manufacturing procedures on the properties of self-

436 cured geopolymer. Construction and Building Materials, 2014. 73: p. 551-561.
437 [17] Shehab, H.K., A.S. Eisa, and A.M. Wahba, Mechanical properties of fly ash based geopolymer concrete with

438 full and partial cement replacement. Construction and Building Materials, 2016. 126: p. 560-565.
439 [18] Hajimohammadi, A., J.L. Provis, and J.S. Van Deventer, One-part geopolymer mixes from geothermal silica

440 and sodium aluminate. Industrial & Engineering Chemistry Research, 2008. 47(23): p. 9396-9405.
441 [19] Hajimohammadi, A., J.L. Provis, and J.S. van Deventer, Effect of alumina release rate on the mechanism of

442 geopolymer gel formation. Chemistry of Materials, 2010. 22(18): p. 5199-5208.

443 [20] Hajimohammadi, A., J.L. Provis, and J.S. van Deventer, The effect of silica availability on the mechanism of

444 geopolymerisation. Cement and Concrete Research, 2011. 41(3): p. 210-216.

445 [21] Hajimohammadi, A. and J.S. van Deventer, Characterisation of one-part geopolymer binders made from fly

446 ash. Waste and Biomass Valorization, 2017. 8(1): p. 225-233.

447 [22] Yang, K.-H., Song, J. K., Ashour, A. F., and Lee, E. T., Properties of cementless mortars activated by sodium

448 silicate. Construction and Building Materials, 2008. 22(9): p. 1981-1989.

20
449 [23] Yang, K.-H. and J.-K. Song, Workability loss and compressive strength development of cementless mortars

450 activated by combination of sodium silicate and sodium hydroxide. Journal of materials in Civil Engineering,
451 2009. 21(3): p. 119-127.
452 [24] Nematollahi, B., J. Sanjayan, and F.U.A. Shaikh, Synthesis of heat and ambient cured one-part geopolymer

453 mixes with different grades of sodium silicate. Ceramics International, 2015. 41(4): p. 5696-5704.
454 [25] Heitzmann, R.F., Gravitt, B.B. and Sawyer, J.L., Cement Composition Curable at Low Temperature. US Patent

455 4,842,649. 1989 June 27.

456 [26] Davidovits, J. 30 years of successes and failures in geopolymer applications. Market trends and potential

457 breakthroughs, in: G.C. Lukey (Ed.), Geopolymers 2002 Turn Potential into Profit, Siloxo Pty. Ltd. Melbourne,
458 Australia, 2002,. CDROM Proceedings.
459 [27] Kani, E.N., A. Allahverdi, and J.L. Provis, Efflorescence control in geopolymer binders based on natural

460 pozzolan. Cement and Concrete Composites, 2012. 34(1): p. 25-33.

461 [28] Noushini, A., et al., Compressive stress-strain model for low-calcium fly ash-based geopolymer and heat-

462 cured Portland cement concrete. Cement and Concrete Composites, 2016. 73: p. 136-146.
463 [29] Deb, P.S., P. Nath, and P.K. Sarker, The effects of ground granulated blast-furnace slag blending with fly

464 ash and activator content on the workability and strength properties of geopolymer concrete cured at
465 ambient temperature. Materials & Design (1980-2015), 2014. 62: p. 32-39.
466 [30] Standards Australia. Method of testing concrete: determination of properties related to the consistency of

467 concrete – slump test, AS 1012.3.1-1998.

468 [31] ASTM C 191–08. Standard test methods for time of setting of hydraulic cement by Vicat Needle. West

469 Conshohocken, PA: ASTM International; 2008.

470 [32] Standards Australia. Method of testing concrete: determination of the compressive strength of concrete

471 specimen, AS 1012.9-1999, Sydney, Australia; 1999.

472 [33] Part, W.K., M. Ramli, and C.B. Cheah, An overview on the influence of various factors on the properties of

473 geopolymer concrete derived from industrial by-products. Construction and Building Materials, 2015. 77: p.
474 370-395.
475 [34] Singh, B., Ishwarya, G., Gupta, M., and Bhattacharyya, S. K., Geopolymer concrete: A review of some recent

476 developments. Construction and building materials, 2015. 85: p. 78-90.

477 [35] Rangan, B. V., Fly ash-based geopolymer concrete. Research Report GC4, Curtin University of Technology,

478 2008, 44 pp.”

479 [36] Komnitsas, K. and D. Zaharaki, Geopolymerisation: A review and prospects for the minerals industry.

480 Minerals engineering, 2007. 20(14): p. 1261-1277

481 [37] Garcia-Lodeiro, I., et al., Compatibility studies between NASH and CASH gels. Study in the ternary

482 diagram Na2O–CaO–Al2O3–SiO2–H2O. Cement and Concrete Research, 2011. 41(9): p. 923-931.
483 [38] Yip, C. and J. Van Deventer, Microanalysis of calcium silicate hydrate gel formed within a geopolymeric

484 binder. Journal of Materials Science, 2003. 38(18): p. 3851-3860.

21
485 [39] Palomo, A., et al., OPC-fly ash cementitious systems: study of gel binders produced during alkaline

486 hydration. Journal of Materials Science, 2007. 42(9): p. 2958-2966.

22

View publication stats