KSCE Journal of Civil Engineering (2015) 19(3):733-741
Copyright ⓒ2015 Korean Society of Civil Engineers
DOI 10.1007/s12205-015-0629-0
−733−pISSN 1226-7988, eISSN 1976-3808
/12205
Structural Engineering
Effects of Chemical Admixtures and Curing Conditions on some Properties of Alkali-Activated Cementless Slag Mixtures
Cahit Bilim*, Okan Karahan**, Cengiz Duran Ati***, and Serhan lkentapar****
Received October 19, 2013/Revid January 10, 2014/Accepted January 27, 2014/Published Online January 12, 2015··································································································································································································································
告别了
Abstract
This paper reports the results of an investigation on the influences of admixtures and curing conditions on some properties of Alkali-Activated Slag (AAS) mixtures with no cement. In the study, Shrinkage-Reducing (SRA) and superplasticizing and t-retarding (WRRe) admixtures were ud. For the slag activation, sodium metasilicate was ud at two sodium concentrations, 4% and 6% by mass of slag. Setting time, flow loss of fresh mixtures, and shrinkage strain, carbonation, flexural and compressive strength of hardened mixtures were measured. The test results showed that the admixtures generally had no impact on the tting times of AAS pastes. WRRe incread the flow rate of AAS mortars while SRA partially affected the flow values of AAS mortars.
WRRe and SRA did not produce an important difference on the carbonation depths of AAS mortars. However, WRRe and especially SRA admixtures decread the shrinkage values of AAS mortars. Additionally, curing conditions had a significant effect on the mechanical behavior in the hardened state of AAS mortars compared to Normal Portland Cement (NPC) mortars, and the strength development of AAS mortars at early ages was very fast in comparison with NPC mortars when subjected to elevated temperature.
Keywords: alkali-activated slag, chemical admixtures, curing, tting time, shrinkage ··································································································································································································································
1. Introduction
It is widely known that the production of Portland cement consumes sizable energy and emits a large volume of CO2 to the atmosphere. Therefore, industrial by-products have an important place in today’s concrete technology owing to their positive effects on the durability and strength of concrete besides the lower production costs and greenhou gas emissions. One of the mineral admixtures utilized as an ingredient in cement or concrete manufacturing is Ground Granulated Blast Furnace Slag (GGBS) emerging in the cour of the production of pig iron. Due to its high content of silica and alumina in an amorphous state, GGBS shows pozzolanic and binding properties in an alkaline medium (Erdo an, 2003). The usage of GGBS in concrete has some benefits such as improving the workability and strength in addition to reducing the hydration heat, permeability and porosity of concrete (Ayd n, 2008). GGBS has latent hydraulic properties and it can react directly with water, but requires an activator. In concrete, this is the Ca(OH)2 relead from the hydration of Portland cement. However, this situation caus the blended cements to develop strength more slowly at early ages compared to NPC. So, the well-known method to activate the hydraulic properties of slag
is the chemical activation of slag by alkalis such as sodium hydroxide, silicate or carbonate (Wang, 2000).
AAS binders, bad on 100% GGBS plus activator, have recently attracted much attention from the academic area and many rearch findings about the materials have been published to produce cementless mortar or concrete. Several studies indicate that AAS mortar and concretes exhibit some superior properties such as high mechanical strength and excellent durability in aggressive environments like chemical attack, freeze-thaw cycles and high temperatures (Jimenez et al., 1999; Puertas et al., 2002; Bakharev et al., 2003; Douglas et al., 1992). Also, earlier investigations on AAS have shown that the strength of the binders is dependent on the concentration and type of alkaline activator, and that sodium silicate (water glass) is the best activator in terms of the strength development performance (Jimenez et al., 1999; Caijun, 1999; Bakharev et al., 1999a). However, it has been revealed by the previous rearches that slag mortars and concretes activated by water glass exhibit considerable drying shrinkage and workability loss against elapd time (Douglas et al., 1992; Bakharev et al., 1999a; Collins and Sanjayan, 1999a; Collins and Sanjayan, 1999b; Ati et al., 2009). Additionally, it has been reported by Collins and Sanjayan (2000) that microcracks in the materials increa sçI
·
g o
i
sçTECHNICAL NOTE
*Associate Professor, Dept. of Civil Engineering, Mersin University, Mersin 33343, Turkey (Corresponding Author, E-mail: cbilim@)
**Associate Professor, Dept. of Civil Engineering, Erciyes University, Kayri 38039, Turkey (E-mail: okarahan@)
***Professor, Dept. of Civil Engineering, Erciyes University, Kayri 38039, Turkey (E-mail: cdatis@)
****Rearch Assistant, Dept. of Civil Engineering, Erciyes University, Kayri 38039, Turkey (E-mail: rhan@)
骆冰王Cahit Bilim, Okan Karahan, Ce ngiz Duran Ati , and Serhan lkentapar
s çI ·−734−KSCE Journal of Civil Engineering
under the inadequate curing conditions.
Since AAS binders can offer a possible solution to deal with industrial by-products such as GGBS, the problems mentioned above need to be solved in order to expand the u of AAS binders as construction materials in commercial practice. On the other hand, organic admixtures, which are developed to be ud in NPC mortars or concretes, have been extensively investigated in the literatu
re until now. But their effects on other binders such as AAS binders have received less attention and a few rearches on AAS binders containing various chemical admixtures have been published (Bakharev et al., 2000; Puertas et al., 2003; W ang et al .,1994). Additionally, since heat treatment provides a significant improvement in volume stability, it can be expected that the curing treatment at elevated temperatures as well as the u of chemical admixtures will have a benefit for the solution of the problems obrved in AAS binders. Accordingly, the aim of this study is to determine how curing temperatures and chemical admixtures influence AAS binders. Furthermore, this paper is a part of rearch work carried out as a contribution to the current state of knowledge on AAS binders, becau the enhancement of the u of AAS mortars and concretes in construction industry, which enables the u of large quantities of by products, is dependent on technical information to be obtained from the rearches.
2. Experimental Study
2.1 Materials
The cement ud in this study was CEM I 42.5 R conforming to TS EN 197-1 (2012). Chemical composition and physical properties of cement and GGBS obtained from OY AK Adana cement facto
ry are given in T able 1. The particle size distributions of the materials, which were obtained using a lar scattering technique, are prented in Fig. 1. CEN standard sand complying with TS EN 196-1 (2009) and municipal water were ud for the preparation of mortar mixtures. While Bilim et al. (2013) utilized liquid sodium silicate activator for the slag activation in previous investigation, GGBS was activated by sodium metasilicate with the dry powdered form in this study. SRA bad on polypropylenglycol
and WRRe bad on modified polymer liquid were ud as chemical admixtures. One percent of each admixture by mass of binder was added to the activator solution.
2.2 Experimental Program
Water/Binder (W/B) ratio of 0.5 was ud to prepare paste and mortar specimens throughout the experimental program. In the ca of mortars, the sand to cementitious binder ratio was 3:1.Sodium concentrations in the mixture proportions containing sodium metasilicate were chon as 4% and 6% by mass of slag.A summary of the experimental program is prented in Table 2.According to TS EN 196-3 (2002), automatic Vicat machine was ud to measure the initial and final tting times of AAS and NPC pastes.
Flow table tests complying with TS EN 1015-3 (2000) were carried out to determine mortar flowability, with and without admixtures, after 0, 15, 30, 45 and 60 min (end of mixing). For each mortar mixture, the diameter was measured in four directions following the flow of mortar onto the table of test apparatus. The results were expresd by taking the average of three specimens.
Prismatic specimens with 40×40×160 mm dimensions and shrinkage specimens measuring 25×25×285 mm were prepared from both fresh NPC and AAS mortar mixtures for the tests.After 24 h, the specimens were demoulded and cured in three
T able 1. Physical, Chemical and Mechanical Properties of Cement and Slag
Chemical Composition (%) Cement
三角粽子的包法
Slag
Physical properties of Portland cement
Physical properties of slag
SiO 2
18.6933.78Specific gravity 3.12Specific gravity
2.78Al 2O 3
5.619.55Initial tting time (min)190Specific surface (Blaine) (cm 2/g)5200Fe 2O 3
2.520.88Final tting time (min)225Pozzolanic activity index (%) of slag
CaO 62.6839.80Soundness (mm) 1.0
7 days 62MgO 2.63
6.80Specific surface (Blaine) (cm 2
/g)
320028 days
94
Na 2O
0.130.32Compressive strength (MPa) of cement
K 2O
0.770.88 2 days 27.2SO 3
2.73 1.667 days 41.0Cl
- 0.010.0328 days 51.2
LOI 2.88 2.89Insoluble residue 0.96-Free CaO 0.93
-Fig. 1. Particle Size Distributions of Cement and Slag
Effects of Chemical Admixtures and Curing Conditions on some Properties of Alkali-Activated Cementless Slag Mixtures
Vol. 19, No. 3 / March 2015
−735−
ways until the time of testing; One group of specimens were placed in a humidity cabinet at 23±2°C and 95% relative humidity (Cure 1) while cond group of specimens were cured in a humidity cabinet at 23±2°C and 50% relative humidity (Cure 2). In the third method, the specimens were immerd in water and the heater was turned on. The water temperature reached at 65°C in 2 h and
the water temperature was maintained at 65°C for 5 h. Subquently, the heater was turned off and, after the cooling periods of the specimens, they were placed in a humidity cabinet at 23 ± 2°C and 50% relative humidity (Cure 3).
The strength tests of the specimens were conducted at 2, 7 and 28 days of age according to TS EN 1015-11 (2000). The compressive strength test was performed on six broken pieces left from flexural strength test. The flexural strength was determined by taking the average of three test results while the compressive strength was determined by taking the average of six test results. A phenolphthalein method was ud to monitor the pH of mortar specimens in the carbonation experiments, and the tests were conducted at 2, 7 and 28 days of age. At the time of measurement, a 1% phenolphthalein solution in alcohol was sprayed on a broken surface of the mortar prism after flexural strength test, and the depth of neutralization was measured. The values were expresd by taking the average of three specimens.For shrinkage measurements, two prismatic specimens with 25× 25 × 285 mm dimensions were prepared from both fresh NPC and AAS mortar mixtures and demoulded the day after. The initial lengths of the shrinkage specimens were measured before they were subjected to three different curing environments. The shrinkage values of mortars were measured until 180 days according to ASTM C596 (2005) and the results were given by taking the average of two specimens for each mixture.
3. Results and Discussion
3.1 Flow T ests
The flow values of mortar mixtures are prented in Fig. 2. As en in Fig. 2, the flow values of mortars decread in time due to some reasons such as the evaporation of mixing water and the tting of binder paste depending on the environmental conditions. It was also obrved that the workability of AAS
binders was nsitive to the alterations in Na concentration as reported by other rearchers (Collins and Sanjayan, 1999b;Douglas et al., 1991). Accordingly, even though an increa in the Na dosage incread the flow values in the first quarter for the slag mortars, the fluidity after 15 min. was badly affected by the dosage increment due to quick reaction and hardening of AAS (Bakharev et al ., 1999a).
The results showed that the flow values obtained from AAS mortars with or without admixture were lower than tho of NPC control mortar. However, WRRe admixture incread the flow rate of AAS mortars, showing a positive effect on mortar workability when compared to the slag mortars with no admixtures.Also, SRA admixture slightly improved the workability of AAS mortars. On the other hand,
except for SRA, WRRe addition into the mixture enhanced the flow values of NPC mortar as expected.
3.2 T ests on Setting Times
Table 3 shows the initial and final tting times obtained from Vicat measurements on AAS and NPC pastes.
As en in Table 3, SRA and especially WRRe admixture lengthened the tting times of pastes containing NPC. However,the chemical admixtures were not effective enough on AAS pastes although the influence of WRRe admixture was more than that of SRA in terms of increasing the tting times of AAS
T able 2. Summary of Experimental Program
Mix no.Binder Activator
Concentration
Admixture
1NPC (control)
---2AAS Sodium metasilicate 4% Na (5.4% Na 2O)-3AAS Sodium metasilicate
6% Na (8.1% Na 2O)
-4NPC --WRRe 5AAS Sodium metasilicate 4% Na (5.4% Na 2O)WRRe 6AAS Sodium metasilicate
6% Na (8.1% Na 2O)
WRRe 7NPC --SRA 8AAS Sodium metasilicate 4% Na (5.4% Na 2O)SRA 9
AAS
Sodium metasilicate
6% Na (8.1% Na 2O)
SRA
Fig. 2. Flow Values of AAS and NPC Mortars (mm)
Cahit Bilim, Okan Karahan, Ce ngiz Duran Ati , and Serhan lkentapar
s çI ·−736−
KSCE Journal of Civil Engineering
pastes. Additionally, the activator dosage increasing from 4% to 6% reduced the initial and final tting times of AAS pastes since an increa in the sodium concentration accelerated the activation of slag. Accordingly, the admixtures lost their efficiency when high dosages of Na was ud. For example, the initial and final tting times of AAS paste with 6% Na, in the abnce of admixtures, were measured as 101 and 157 min, respectively.But 1% SRA addition into the mixture had no impa
ct on the tting times of AAS paste with 6% Na, and the initial and final tting times for this paste became 98 and 161 min. This situation may be resulting from the modifications that high alkaline media induced in the chemical structures of the admixtures ud (Palacios and Puertas, 2005). Conquently, the tting times of AAS pastes were much shorter than tho of NPC control paste. The results agree with tho reported elwhere (J imenez et al., 1999; Ati et al., 2009; Caijun and Yinyu, 1989; Wang et al ., 1995).
3.3 Compressive Strengths
The compressive strength values measured at the age of 2, 7and 28 days of AAS and NPC mortars are prented in Fig. 3,Fig. 4 and Fig. 5 for three different curing conditions, respectively. The results indicated that the compressive strengths of AAS mortars incread with an increa in the Na dosage of activator.This finding is valid for each curing age and curing regimes. For example, as en in Fig. 3, the compressive strength at the age of 28 days of AAS mortar with 4% Na subjected to 23 ± 2°C and 50% relative humidity was 24.88 MPa, while the compressive strength at the age of 28 days of AAS mortar with 6% Na under the same curing conditions was 37.38 MPa. This is attributed to the chemical reaction that takes place between slag and activator.The strength development of AAS binders is greatly influenced by the synthesis of the anions of GGBS and the ca
4月22日是什么星座
tions of the alkaline activators. The chemical reaction occurring by ion exchange between the silicate anions of slag and cations of alkaline activators leads to the formation of silica gel. This silica gel may turn into silica-rich calcium silicate hydrates gel (C-S-H) by further reacting with calcium ion of GGBS (Y ang et al .,
入党的动机2008).
工程预算表As en in all figures, WRRe and SRA chemical admixtures did not produce a remarkable difference in the compressive strength of NPC mortars. Similarly, WRRe and SRA did not show any negative effect on the compressive strengths of AAS mortars.
The results showed that the curing conditions were very important in terms of the strength development of AAS mortars.In comparison with other curing conditions given in Fig. 4 and Fig. 5, when subjected to 23 ± 2°C and 50% relative humidity, all mixtures with and without chemical admixtures exhibited the
s çT able 3.The Initial and Final Setting Times obtained from AAS and
NPC Pastes Mix no.Binder Admixture
Initial tting time (min)Final tting time (min)
1NPC (control)-
3904752AAS, 4% Na -1252633
AAS, 6% Na
-1011574NPC WRRe 84511505AAS, 4% Na WRRe 2504506AAS, 6% Na怎么卖股票
年的成语WRRe 1262207NPC SRA 4685888AAS, 4% Na SRA 1552759
AAS, 6% Na
SRA
98
161
Fig. 3.Compressive Strengths of AAS and NPC Mortars Expod
to Dry Curing (MPa)
Fig. 4.Compressive Strengths of AAS and NPC Mortars Expod
to Heat Curing (MPa)
Fig. 5.Compressive Strengths of AAS and NPC Mortars Expod
to Moist Curing (MPa)
Effects of Chemical Admixtures and Curing Conditions on some Properties of Alkali-Activated Cementless Slag Mixtures
Vol. 19, No. 3 / March 2015
−737−
lowest compressive strengths as shown in Fig. 3, and the mixtures did not show a sufficient strength development after the age of 7 days. For instance, the compressive strength at the age of 7 days of AAS mortar containing 4% Na and 1% WRRe expod to dry curing was 23.46 MPa whereas the compressive strength value at 28 days for the same mortar was 24.15 MPa.This situation aris from the loss of water in mixture depending on dry curing conditions (Neville, 1981). Additionally, under dry curing conditions, the Na dosage of 4% could not provide a sufficient activation for GGBS mortars with and without chemical admixtures. This is attributed to prence of water that provides ion exchange between materials. Since dry curing caus quick drying of sample, abnce of water stops ion exchange.Accordingly, the strengths of AAS mortars with 4% Na was lower than the values obtained from NPC mortars at all ages of curing. But, under the same dry curing conditions, the Na dosage increasing from 4% to 6% became effective in terms of the reach to the higher compressive strengths than tho of NPC mortars as of the venth day. On the other hand, the curing treatment performed at 65°C in water for 5 h significantly incread the strength values at the early ages of AAS mortars by accelerating the activation of slag. Accordingly, the highe
st compressive strengths at 2 and 7 days were obtained from AAS mortars subjected to heat curing as en in Fig. 4. Also, it was obrved that when cured at high temperatures, the strength development at the early stage of AAS mortars was much faster than that of NPC mortars. The results conform to tho of another study (Bakharev et al., 1999b). However, after the hot curing, since the specimens were kept under dry curing conditions at 23 ± 2°C and 50% relative humidity, this situation badly influenced the strength development at later ages. Thus, the maximum strength development after 7 days were obtained from AAS mortars continuously expod to 23 ± 2°C and 95% relative humidity as en from Fig. 5. The findings are in accordance with tho of other rearchers (Collins and Sanjayan, 1999a; Collins and Sanjayan, 2001; Kutti et al ., 1992; Chi, 2012) reporting that the strength development of AAS binder was susceptible to the type of curing environment.
3.4 Flexural Strengths
The flexural strengths at the age of 2, 7 and 28 days of AAS and NPC mortars are prented in Fig. 6, Fig. 7 and Fig. 8 for three different curing conditions, respectively.
As en in all figures, when compared to the mortars without admixture, SRA or WRRe addition into the mixture did not produce a negative effect on the flexural strengths of AAS and NPC mortars.
On the other hand, the lowest flexural strength values at 2, 7and 28 days were obtained from AAS mortars subjected to 23 ±2°C and 50% relative humidity as en from Fig. 6. It was also obrved that the flexural strengths of AAS mortars were affected from the dry curing conditions more than the compressive strength values, and that the flexural strengths of AAS mortars containing the high Na concentration decread with time.
Namely, for the mixtures without admixtures, the flexural strength values at the age of 2, 7 and 28 days of AAS mortar with 6% Na cured at 23 ± 2°C and 50% relative humidity were 1.99MPa, 2.93 MPa and 2.70 MPa respectively. The possible reason for such behavior is the expansion of cracks (Collins and Sanjayan, 2000; Collins and Sanjayan, 2001). As well as the dry curing conditions, the micro-cracks which are developed due to the higher magnitude of shrinkage cau a decrea in the flexural strength by reducing a valid area of cross-ctions.
Furthermore, since the heat treatment in water at 65°C for 5 h
Fig. 6.
Flexural Strengths of AAS and NPC Mortars Expod to
Dry Curing (MPa)
Fig. 7.Flexural Strengths of AAS and NPC Mortars Expod to
Heat Curing (MPa)
Fig. 8.Flexural Strengths of AAS and NPC Mortars Expod to
Moist Curing (MPa)
