- Email: [email protected]
A study of the effect of light-burnt dolomite on the hydration of alkali-activated Portland blast-furnace slag cement
Construction and Building Materials 57 (2014) 24–29
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
A study of the effect of light-burnt dolomite on the hydration of alkali-activated Portland blast-furnace slag cement Wan-Hee Yang a,⇑, Dong-Woo Ryu b, Dong-Cheol Park c, Woo-Jae Kim d, Chee-Ho Seo a a
Department of Architectural Engineering, Konkuk University, 1, Hwayang-dong, Gwangjin-gu, Seoul 143-701, Republic of Korea Department of Architectural Engineering, Daejin University, San 11-1, Sundan-dong, Pochon, Gyeonggi-do 487-811, Republic of Korea c R&D Center, Intchem Co., Ltd., 489-3, Maetan-dong, Yeongtong-gu, Suwon, Gyeonggi-do 443-370, Republic of Korea d Research & Engineering Division R&D Center, POSCO E&C, 180-1, Songdo-dong, Yeonsu-Gu, Incheon 406-840, Republic of Korea b
h i g h l i g h t s In Na2SO4-activated Portland blast-furnace slag cement, the hydration heat increased with the addition of LBD. A delay of the initial setting time was observed with the addition of LBD. The compressive strength improved with the addition of LBD. The pore structure of the paste became dense after 28 days of aging.
a r t i c l e
i n f o
Article history: Received 29 April 2013 Received in revised form 16 January 2014 Accepted 22 January 2014 Available online 21 February 2014 Keywords: Alkali-activated cement Blast-furnace slag Light-burnt dolomite Sodium sulfate Heat of hydration
a b s t r a c t This study used the typical powder alkaline material anhydrous sodium sulfate (Na2SO4) as an activator in a system in which Portland cement (10–20%) and ground granulated blast-furnace slag (GGBS, 80–90%) are generally mixed and added to light-burnt dolomite (LBD), one of the economical lime materials. Their effect on the hydration and strength of the cement was then analyzed. According to the analysis, the 1st peak of the minute heat of hydration increased after mixing the LBD. On the other hand, for the 2nd peak, a slight delay was observed. As a result of the mixing, the cumulative heat of hydration increased. Although the rate of setting decreased, the strength of the material improved after the 3rd, 7th and 28th days. In other words, it was confirmed that the pore structure of the paste became dense after 28 days of aging. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Portland cement is the most widely used construction material [1,2], but it accounts for approximately 5% of greenhouse gas emissions [2]. Therefore, there have been many studies on various materials to reduce the use of or substitute Portland cement with the goal of reducing environmental pollution [2]. In particular, alkali-activated slag cement (AASC) is economically advantageous because the industrial by-products can be effectively used. Furthermore, so far, great characteristics substituting Portland cement are reported through various studies in AASC, and some countries have partially used it [3]. However, AASC has lower resistance of carbonation than Portland cement [3–5], and it has greater drying shrinkage [3,6–10]. ⇑ Corresponding author. Tel.: +82 2 3436 7898; fax: +82 2 3436 7897. E-mail address: [email protected] (W.-H. Yang). http://dx.doi.org/10.1016/j.conbuildmat.2014.01.071 0950-0618/Ó 2014 Elsevier Ltd. All rights reserved.
In addition, ground granulated blast-furnace slag (GGBS) which is commonly applied to the AASC, has been used to manufacture concrete along with Portland cement in concrete manufacturing in various countries with advanced steel industries [3,11,12]. On the other hand, there are some problems with AASC that prevent this material from being a practical replacement for Portland cement. For example, this material is not compatible with the existing facilities used to manufacture ready-mix concrete [13]. At present, there are two ways to manufacture AASC: the use of liquid alkaline materials and the use of powdered alkaline materials. In the former method, additional tank to store liquid activator is required to use these liquid materials in the conventional readymixed concrete factories. In the second method the powdered alkaline materials can be easily used in the conventional ready-mixed concrete manufacturing facilities thorough premixing when manufacturing either slag powder or slag cement. Well-known representative powder-phase activators are sodium sulfate and sodium carbonate; sodium sulfate is slightly more
25
W.-H. Yang et al. / Construction and Building Materials 57 (2014) 24–29
economical (for the industrial product, the cost of sodium sulfate is about 50% of the cost of sodium carbonate), and sodium sulfate is known to improve the strength of cementitious materials by generating ettringite because of sulfate included in lime-based cementing materials [3]. The light-burnt dolomite (LBD) is one of economical powderphased lime materials, and it generates Ca(OH)2 from the reaction with water, leading to an applicability as an activator. In the conventional study [14,15], there was a practice of using (10%) Magnesia (MgO) for the purpose of drying shrinkage reduce in AASC using liquid activator (mainly sodium silicate solution), and in this study, the effect mainly on strength development by using LBD (<2%) was to be analyzed. Therefore, this study used the powder alkaline sodium sulfate anhydrous (Na2SO4) as an activator in alkali-activated Portland slag cement (AAPSC), in which Portland cement (10–20%) and GGBS (80–90%) were used and added to light-burnt dolomite (LBD). The effect of these materials on the hydration and strength of the cement was then analyzed. 2. Experimentation 2.1. Raw materials Ground granulated blast-furnace slag (GGBS): GGBS was produced in Incheon, and its chemical composition and physical properties are shown in Table 1; ordinary Portland cement (OPC): OPC was produced in Danyang, and its chemical composition and physical properties are shown in Table 1; activator: sodium sulfate anhydrous with a purity of 99% used in the test; light-burnt dolomite (LBD): LBD was product calcinated under 900–1000 °C that was produced in Jeongseon, and its chemical composition and physical properties are shown in Table 1; the XRD pattern of the LBD is shown in Fig. 1. 2.2. Test method The setting time was tested according to the Korea national standard KS ISO 9597 (ISO 9597) [16]. The strength of the cement mortar was tested according to the Korea national standard KS L 5105-2007 [17] (specimen size: 50 50 50 mm). The heat of hydration was measured using Tokyo-Riko’s threepoint, multipurpose conduction calorimeter with a water-to-solid ratio of 0.5. In addition, XRD was measured (Cu Ka1 radiation with 2h ranging from 5° to 60°) using the apparatus (WDX 200; acceleration voltage: 40 KV; current: 35 mA; scanning speed: 3.00°/min; step width: 0.05°) manufactured by Skyray Instruments. Moreover, the pore structure was measured against the hardened cement paste specimens that were wet-cured for 28 days using Micromeritics’ apparatus (AutoPore IV; applied pressure (measured pore size): 0.5–40,000psia (338 lm4 nm)) with a water-to-solid ratio of 0.45.
3. Results and discussion 3.1. Heat of hydration Fig. 2 shows that the heat of hydration after mixing sodium sulfate and LBD in the cementitious material that used GGBS (90%) and OPC (10%). Figs. (a and b) show the results after 2- and 48-h hydrations, respectively. In addition, Fig. (c) shows the cumulated heat of hydration measured every minute. Here, sodium sulfate with weight ratio of 0% and 2% and LBD with weight ratio of 0%, 1% and 2% were used. The samples are marked as follows: N0D0,
Fig. 1. XRD Patterns of the LBD.
N0D1, N0D2, N2D0, N2D1 and N2D2. Unless sodium sulfate was used (N0D0-N0D2), the 1st peak increased with addition of LBD. For the 2nd peak, a time delay was detected. The 1st peak is increased probably because the heat generated from hydration of LBD (MgOCaO + 2H2O >Ca(OH)2 + Mg(OH)2) affected it. With the same condition as in Fig. 2, the result of measuring characteristics of heat of hydration of LBD was shown in Fig. 3. As shown in Fig. 3, the heat of hydration of LBD is very high, but the heat of hydration generated from the use of small amount in the experiment is not high. However, in case where both GGBS and OPC exist as shown in this study, it is regarded that the hydration of LBD influences on activation of hydration of cementitious materials to increase the heat of hydration of GGBS or OPC. Such effect is considered to influence more on increase of heat of hydration of OPC or GGBS if LBD is used with sodium sulfate. The reason the 2nd peak time was delayed due to mixing with LBD is not clear. Yet, LBD generates heat from the reaction with water (H2O) initially, but it is assumed that insoluble Mg(OH)2 is generated due to the reaction between dissolved Mg2+ and OHin solution, leading to decrease in hydroxyl ions in the solution to slightly reduce alkalinity of the paste [1]. The reaction of the cementitious materials was delayed accordingly. When 2% of sodium sulfate was used (N2D0-N2D2), the hydration heat peak was mostly high in both the 1st and 2nd peaks. This result originated from the activated reaction of the cementitious materials by sodium sulfate similar to the results of conventional studies [3,18]. In addition, a delay of the 2nd peak time was observed after the addition of LBD in the N2 series. The cumulative heat of hydration (c) was high depending on the addition of LBD. In addition, it was higher when both sodium sulfate and LBD were used at the same time than when only LBD was added. Fig. 4 shows the hydration heat of OPC and Portland slag cement (PSC, 55% of OPC + 45% of GGBS) under the same conditions. In terms of the 1st peak, the N2D2 was lower than the OPC but similar to the PSC. In addition, the 2nd peak was detected at approximately 12 h of aging in both the OPC and PSC but at approximately 30 h of aging in the N2D2. In terms of the cumulative heat of hydration, the N2D2 was similar to the PSC.
Table 1 Chemical composition and physical properties of the materials. Material
OPC GGBS LBD
Composition (%) SiO2
Al2O3
Fe2O3
CaO
MgO
SO3
22.0 32.7 0.89
5.27 13.2 0.32
3.44 0.41 -
63.4 44.1 52.8
2.13 6.23 32.2
1.96 1.49 –
Alkali
Ig.loss
K2O
Na2O
Total alkali
0.42 0.31 –
0.27 0.20 –
0.55 0.40 –
0.79 0.88 1.85
Density (g/cm3)
Blaine specific area (cm2/g)
3.15 2.88 3.12
3419 7042 3038
26
W.-H. Yang et al. / Construction and Building Materials 57 (2014) 24–29
Fig. 2. Heat of hydration of AAPSC (GGBS 90% + OPC 10%).
Therefore, when 2% sodium sulfate and 2% LBD are used in AAPSC (GGBS 90% + OPC 10%), it appears that the hydration reaction increased up to the level of the PSC (OPC 55% + GGBS 45%). 3.2. Setting time Fig. 5 shows the penetration depth of the needles for measuring the initial setting time of the Vicat apparatus after mixing the sodium sulfate and LBD with the cementitious materials in which GGBS and OPC were mixed. Because sodium sulfate was used, setting time tended to be short. With the mixture of LBD, the setting time of the cementitious paste increased. When 1% and 2% of the LBDs were mixed, however, the setting properties of the paste were similar. There was a significant decrease in the setting time due to the mixture of sodium sulfate was when 90% GGBS was used. This type of pattern appeared probably because the specific area of the GGBS was greater than that of the OPC. Such
Fig. 3. Heat of hydration of LBD.
W.-H. Yang et al. / Construction and Building Materials 57 (2014) 24–29
27
Fig. 4. Heat of hydration of OPC and PSC (OPC 55% + GGBS 45%).
Fig. 5. Penetration of the Vicat test (initial setting time test).
inclination was because the specific area of GGBS (7042 cm2/g) is greater than the specific area of OPC (3419 cm2/g), the case with 90% GGBS receives faster infiltration resistance of Vicat needle at the initial setting time than with 80% GGBS. Similarly, it is regarded that increase in activity due to sodium sulfate is affected quicker as well by using 90% of GGBS instead of 80% of GGBS. The inclination of decrease in the setting time as the fineness of GGBS is higher is similar to conventional results of study [3,19,20].
The initial setting time of the AAPSC is shown in Fig. 6. The setting time was 1 h and 20 min when 2% of sodium sulfate was used from 80% GGBS. However, the setting time was delayed to 4–5 h when 1–2% LBD was added. Generally, considering that the initial setting time of OPC is 3:20–4:00 and that the initial setting time of PSC is 4:10–4:50, and since this type has similar setting time to that of PSC, the result indicates that it is possible to achieve work hours up to the level of OPC when concrete is produced with AAPSC.
28
W.-H. Yang et al. / Construction and Building Materials 57 (2014) 24–29
Fig. 6. Initial setting time of AAPSC.
3.3. XRD analysis Fig. 7 shows the XRD patterns at 1, 3 and 28 days of aging after mixing sodium sulfate and LBD in the cementitious material in which GGBS (90%) and OPC (10%) were used. According to the XRD result, the major hydrate after 28 days of aging was calcium silicate hydrate. At 1–3 day(s) of aging, ettringite (3CaOAl2O3 3CaSO4 32H2O) and calcium silicate hydrate were confirmed as the major hydrate. A slight decrease in the peak of ettringite was then observed at 1 day of aging because of the addition of LBD. Ettringite is the hydrate of a cementitious material at initial aging [21]. The addition of LBD slightly slowed down the hydration of AAPSC by reducing the alkalinity at the initial period. Furthermore, in terms of the N2D2, where 2% sodium sulfate and 2% LBD were mixed, a small amount of calcium sulfate hydrate was generated from the reaction between enough sulfate ions and calcium ions [18]. However, at 3 d, the ettringite peaks of D1 and D2, both of in which LBD was mixed, were similar to that of D0, but there was no calcium sulfate hydrate peak. At 28 days, a sufficient amount of calcium silicate hydrate was observed, and there were no significant differences between samples. 3.4. Compressive strength Fig. 8 shows the compressive strength of the mortar of AAPSC after LBD was mixed in. As the mixture of LBD increased, the compressive strength of AAPSC continuously increased from the 3rd to the 28th day of aging. For the mixture of GGBS (80%) and OPC (20%), the strength increased significantly when sodium sulfate and LBD were used at the same time. Specifically, the addition of sodium sulfate dramatically enhanced the strength at 3 days. In addition, the mixture of LBD greatly increased the strength at the 7th and 28th days. When sodium sulfate and LBD were used at the same time, the strength at days 3, 7 and 28 continuously increased. For N2D1 and N2D2, the compressive strength at 28 days was 42–43 MPa, and it improved by 22–24% compared to the N0D0 in which sodium sulfate and LBD were not used. Therefore, when such strength properties are used effectively, a relatively good strength performance is expected, and the amount of OPC is greatly reduced. When sodium sulfate and LBD were mixed, the AAPSC was continuously hydrated, leading to an improved densification of the microstructure. Through the XRD analysis, there were no found that calcium hydroxide or magnesium hydroxide, but the improvement in the strength is assumed to be related with the generation of expansive hydrates.
Fig. 7. XRD patterns of AAPSC.
3.5. Pore structure Fig. 9 shows the cumulative pore volume at 28 days of aging after mixing LBD in AAPSC (80% of GGBS + 20% of OPC), in which 2% of sodium sulfate was used. In this case, the addition of the LBD decreased the pore volume (0.01–1.0 lm in particular). The capillary voids of cement are detrimental to strength and impermeability in the range over 50 nm, and in the range under 50 nm, the voids are known to be important to drying shrinkage and creep [1,22]. Therefore, the mixture of LBD enhanced the compressive strength of AAPSC and the density pore structure as well. It appears that this type of pore structure enhances the strength. Furthermore, it is predicted that it will affect positively on the reduction of drying shrinkage.
W.-H. Yang et al. / Construction and Building Materials 57 (2014) 24–29
29
(d) The compressive strength improved with the addition of LBD. This type of pattern got stronger when mixed with sodium sulfate. It continuously improved until the 28th day. In addition, this type of result was confirmed in a constant manner through the analysis on the pore structure.
Acknowledgement This subject is supported by Korea Ministry of Environment as ‘‘The Eco-technopia 21 project’’ (405-111-006). References
Fig. 8. Effect of LBD on compressive strength of AAPSC.
Fig. 9. Effect of LBD on cumulative pore volume of AAPSC (GGBS 80% + OPC 20%).
4. Conclusions This study used the powder alkaline sodium sulfate anhydrous (Na2SO4) as an activator in alkali-activated Portland slag cement (AAPSC), in which Portland cement (10–20%) and GGBS (80–90%) were used and added to light-burnt dolomite (LBD), and one of the economical lime materials. The result of effect by these materials on the hydration and strength of the AAPSC was analyzed as follows. (a) In Na2SO4-activated PSC (80–90% of GGBS + 10–20% of OPC), the 1st peak of the minute hydration heat increased with the addition of LBD. On the contrary, the 2nd peak was slightly delayed, and the cumulative heat of hydration increased. (b) A delay of the initial setting time was observed with the addition of LBD. (c) The major hydrates of AAPSC were calcium silicate hydrate and ettringite. It was confirmed that the mixture of LBD slightly delayed the creation of ettringite during the initial period.
[1] Mehta PK, Monteiro PJM. Concrete Microstructure, Properties, and Materials. 3rd ed. New York: McGraw-Hill Companies; 2006. [2] International Energy Agency (IEA), World Business Council for Sustainable Development (WBCSD). Cement Technology Roadmap 2009. Paris (FR): Corlet; 2010. [3] Shi C, Krivenko PV, Roy D. Alkali activated cement and concretes. New York: Taylor & Francis; 2006. [4] Bakharev T, Sanjayan JG, Cheng YB. Resistance of alkali-activated slag concrete to carbonation. Cem Concr Res 2001;31(9):1277–83. [5] Shi C. Corrosion resistance of alkali-activated slag cement. Adv Cem Res 2003;15(2):77–81. [6] Collins FG, Sanjayan JG. Strength and shrinkage properties of alkali-activated slag concrete containing porous coarse aggregate. Cem Concr Res 1999;29(4): 607–10. [7] Bakharev T, Sanjayan JG, Cheng YB. Effect of admixtures on properties of alkaliactivated slag concrete. Cem Concr Res 2000;30(9):1367–74. [8] Bakharev T, Sanjayan JG, Cheng YB. Effect of elevated temperature curing on properties of alkali-activated concrete. Cem Concr Res 1999;29(10):1619–26. [9] Collins FG, Sanjayan JG. Strength and shrinkage properties of alkali-activated slag concrete placed into a large column. Cem Concr Res 1999;29(5):659–66. [10] Collins FG, Sanjayan JG. Effect of pore size distribution on drying shrinkage of alkali-activated slag concrete. Cem Concr Res 2000;30(9):1401–6. [11] Ryu DW, Kim WJ, Yang WH, You JH, Ko JW. An experimental study on the freezing–thawing and chloride resistance of concrete using high volumes of GGBS. J Korea Inst Build Constr 2012;12(3):315–22. [12] Ryu DW, Kim WJ, Yang WH, Park DC. An experimental study on the carbonation and drying shrinkage of concrete using high volumes of ground granulated blast-furnace slag. J Korea Inst of Build Constr 2012;12(4): 393–400. [13] Yang WH, Ryu DW, Kim WJ, Park DC, Seo CH. An experimental study on early strength and drying shrinkage of high strength concrete using high volumes of ground granulated blast-furnace slag(GGBS). J Korea Inst Build Constr 2013;13(4):391–9. [14] Shen W, Wang Y, Zhang T, Zhou M, Li J, Cui X. Magnesia modification of alkaliactivated slag fly ash cement. J Wuhan Univ Technol-Mater 2011;126(1): 121–5. [15] Fang YH, Liu JF, Chen YG. Effect of magnesia on properties and microstructure of alkali-activated slag cement. Water Sci Eng 2011;4(4):463–9. [16] KS ISO 9597. Specifies the methods for determining standard consistence, setting times and soundness of cements. International Organization for Standardization, 2008. [17] KS L 5105. Testing method for compressive strength of hydraulic cement mortars. Korean Industrial Standards, 2007. [18] Shi C, Day RL. Pozzolanic reaction in the presence of chemical activators:Part II. Reaction products and mechanism. Cem Concr Res 2000;30(4):607–13. [19] Shi C, Li Y. Investigation on some factors affecting the characteristics of alkaliphosphorus slag cement. Cem Concr Res 1989;19(4):527–33. [20] Andersson R, Gram HE. Properties of alkali-activated slag concrete. Nordic Concr Res 1987;6:7–18. [21] Gartner EM, Young JF, Damidot DA, Jawed I. Hydration Portland Cem. In: Bensted J, Barnes P, editors. Structure and performance of cements. New York: Taylor & Francis; 2002. p. 57–139. [22] Mindess S, Young JF. Concrete. Englewood Cliffs NJ (USA): Prentice-Hall International Inc., 1981.
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
A study of the effect of light-burnt dolomite on the hydration of alkali-activated Portland blast-furnace slag cement Wan-Hee Yang a,⇑, Dong-Woo Ryu b, Dong-Cheol Park c, Woo-Jae Kim d, Chee-Ho Seo a a
Department of Architectural Engineering, Konkuk University, 1, Hwayang-dong, Gwangjin-gu, Seoul 143-701, Republic of Korea Department of Architectural Engineering, Daejin University, San 11-1, Sundan-dong, Pochon, Gyeonggi-do 487-811, Republic of Korea c R&D Center, Intchem Co., Ltd., 489-3, Maetan-dong, Yeongtong-gu, Suwon, Gyeonggi-do 443-370, Republic of Korea d Research & Engineering Division R&D Center, POSCO E&C, 180-1, Songdo-dong, Yeonsu-Gu, Incheon 406-840, Republic of Korea b
h i g h l i g h t s In Na2SO4-activated Portland blast-furnace slag cement, the hydration heat increased with the addition of LBD. A delay of the initial setting time was observed with the addition of LBD. The compressive strength improved with the addition of LBD. The pore structure of the paste became dense after 28 days of aging.
a r t i c l e
i n f o
Article history: Received 29 April 2013 Received in revised form 16 January 2014 Accepted 22 January 2014 Available online 21 February 2014 Keywords: Alkali-activated cement Blast-furnace slag Light-burnt dolomite Sodium sulfate Heat of hydration
a b s t r a c t This study used the typical powder alkaline material anhydrous sodium sulfate (Na2SO4) as an activator in a system in which Portland cement (10–20%) and ground granulated blast-furnace slag (GGBS, 80–90%) are generally mixed and added to light-burnt dolomite (LBD), one of the economical lime materials. Their effect on the hydration and strength of the cement was then analyzed. According to the analysis, the 1st peak of the minute heat of hydration increased after mixing the LBD. On the other hand, for the 2nd peak, a slight delay was observed. As a result of the mixing, the cumulative heat of hydration increased. Although the rate of setting decreased, the strength of the material improved after the 3rd, 7th and 28th days. In other words, it was confirmed that the pore structure of the paste became dense after 28 days of aging. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Portland cement is the most widely used construction material [1,2], but it accounts for approximately 5% of greenhouse gas emissions [2]. Therefore, there have been many studies on various materials to reduce the use of or substitute Portland cement with the goal of reducing environmental pollution [2]. In particular, alkali-activated slag cement (AASC) is economically advantageous because the industrial by-products can be effectively used. Furthermore, so far, great characteristics substituting Portland cement are reported through various studies in AASC, and some countries have partially used it [3]. However, AASC has lower resistance of carbonation than Portland cement [3–5], and it has greater drying shrinkage [3,6–10]. ⇑ Corresponding author. Tel.: +82 2 3436 7898; fax: +82 2 3436 7897. E-mail address: [email protected] (W.-H. Yang). http://dx.doi.org/10.1016/j.conbuildmat.2014.01.071 0950-0618/Ó 2014 Elsevier Ltd. All rights reserved.
In addition, ground granulated blast-furnace slag (GGBS) which is commonly applied to the AASC, has been used to manufacture concrete along with Portland cement in concrete manufacturing in various countries with advanced steel industries [3,11,12]. On the other hand, there are some problems with AASC that prevent this material from being a practical replacement for Portland cement. For example, this material is not compatible with the existing facilities used to manufacture ready-mix concrete [13]. At present, there are two ways to manufacture AASC: the use of liquid alkaline materials and the use of powdered alkaline materials. In the former method, additional tank to store liquid activator is required to use these liquid materials in the conventional readymixed concrete factories. In the second method the powdered alkaline materials can be easily used in the conventional ready-mixed concrete manufacturing facilities thorough premixing when manufacturing either slag powder or slag cement. Well-known representative powder-phase activators are sodium sulfate and sodium carbonate; sodium sulfate is slightly more
25
W.-H. Yang et al. / Construction and Building Materials 57 (2014) 24–29
economical (for the industrial product, the cost of sodium sulfate is about 50% of the cost of sodium carbonate), and sodium sulfate is known to improve the strength of cementitious materials by generating ettringite because of sulfate included in lime-based cementing materials [3]. The light-burnt dolomite (LBD) is one of economical powderphased lime materials, and it generates Ca(OH)2 from the reaction with water, leading to an applicability as an activator. In the conventional study [14,15], there was a practice of using (10%) Magnesia (MgO) for the purpose of drying shrinkage reduce in AASC using liquid activator (mainly sodium silicate solution), and in this study, the effect mainly on strength development by using LBD (<2%) was to be analyzed. Therefore, this study used the powder alkaline sodium sulfate anhydrous (Na2SO4) as an activator in alkali-activated Portland slag cement (AAPSC), in which Portland cement (10–20%) and GGBS (80–90%) were used and added to light-burnt dolomite (LBD). The effect of these materials on the hydration and strength of the cement was then analyzed. 2. Experimentation 2.1. Raw materials Ground granulated blast-furnace slag (GGBS): GGBS was produced in Incheon, and its chemical composition and physical properties are shown in Table 1; ordinary Portland cement (OPC): OPC was produced in Danyang, and its chemical composition and physical properties are shown in Table 1; activator: sodium sulfate anhydrous with a purity of 99% used in the test; light-burnt dolomite (LBD): LBD was product calcinated under 900–1000 °C that was produced in Jeongseon, and its chemical composition and physical properties are shown in Table 1; the XRD pattern of the LBD is shown in Fig. 1. 2.2. Test method The setting time was tested according to the Korea national standard KS ISO 9597 (ISO 9597) [16]. The strength of the cement mortar was tested according to the Korea national standard KS L 5105-2007 [17] (specimen size: 50 50 50 mm). The heat of hydration was measured using Tokyo-Riko’s threepoint, multipurpose conduction calorimeter with a water-to-solid ratio of 0.5. In addition, XRD was measured (Cu Ka1 radiation with 2h ranging from 5° to 60°) using the apparatus (WDX 200; acceleration voltage: 40 KV; current: 35 mA; scanning speed: 3.00°/min; step width: 0.05°) manufactured by Skyray Instruments. Moreover, the pore structure was measured against the hardened cement paste specimens that were wet-cured for 28 days using Micromeritics’ apparatus (AutoPore IV; applied pressure (measured pore size): 0.5–40,000psia (338 lm4 nm)) with a water-to-solid ratio of 0.45.
3. Results and discussion 3.1. Heat of hydration Fig. 2 shows that the heat of hydration after mixing sodium sulfate and LBD in the cementitious material that used GGBS (90%) and OPC (10%). Figs. (a and b) show the results after 2- and 48-h hydrations, respectively. In addition, Fig. (c) shows the cumulated heat of hydration measured every minute. Here, sodium sulfate with weight ratio of 0% and 2% and LBD with weight ratio of 0%, 1% and 2% were used. The samples are marked as follows: N0D0,
Fig. 1. XRD Patterns of the LBD.
N0D1, N0D2, N2D0, N2D1 and N2D2. Unless sodium sulfate was used (N0D0-N0D2), the 1st peak increased with addition of LBD. For the 2nd peak, a time delay was detected. The 1st peak is increased probably because the heat generated from hydration of LBD (MgOCaO + 2H2O >Ca(OH)2 + Mg(OH)2) affected it. With the same condition as in Fig. 2, the result of measuring characteristics of heat of hydration of LBD was shown in Fig. 3. As shown in Fig. 3, the heat of hydration of LBD is very high, but the heat of hydration generated from the use of small amount in the experiment is not high. However, in case where both GGBS and OPC exist as shown in this study, it is regarded that the hydration of LBD influences on activation of hydration of cementitious materials to increase the heat of hydration of GGBS or OPC. Such effect is considered to influence more on increase of heat of hydration of OPC or GGBS if LBD is used with sodium sulfate. The reason the 2nd peak time was delayed due to mixing with LBD is not clear. Yet, LBD generates heat from the reaction with water (H2O) initially, but it is assumed that insoluble Mg(OH)2 is generated due to the reaction between dissolved Mg2+ and OHin solution, leading to decrease in hydroxyl ions in the solution to slightly reduce alkalinity of the paste [1]. The reaction of the cementitious materials was delayed accordingly. When 2% of sodium sulfate was used (N2D0-N2D2), the hydration heat peak was mostly high in both the 1st and 2nd peaks. This result originated from the activated reaction of the cementitious materials by sodium sulfate similar to the results of conventional studies [3,18]. In addition, a delay of the 2nd peak time was observed after the addition of LBD in the N2 series. The cumulative heat of hydration (c) was high depending on the addition of LBD. In addition, it was higher when both sodium sulfate and LBD were used at the same time than when only LBD was added. Fig. 4 shows the hydration heat of OPC and Portland slag cement (PSC, 55% of OPC + 45% of GGBS) under the same conditions. In terms of the 1st peak, the N2D2 was lower than the OPC but similar to the PSC. In addition, the 2nd peak was detected at approximately 12 h of aging in both the OPC and PSC but at approximately 30 h of aging in the N2D2. In terms of the cumulative heat of hydration, the N2D2 was similar to the PSC.
Table 1 Chemical composition and physical properties of the materials. Material
OPC GGBS LBD
Composition (%) SiO2
Al2O3
Fe2O3
CaO
MgO
SO3
22.0 32.7 0.89
5.27 13.2 0.32
3.44 0.41 -
63.4 44.1 52.8
2.13 6.23 32.2
1.96 1.49 –
Alkali
Ig.loss
K2O
Na2O
Total alkali
0.42 0.31 –
0.27 0.20 –
0.55 0.40 –
0.79 0.88 1.85
Density (g/cm3)
Blaine specific area (cm2/g)
3.15 2.88 3.12
3419 7042 3038
26
W.-H. Yang et al. / Construction and Building Materials 57 (2014) 24–29
Fig. 2. Heat of hydration of AAPSC (GGBS 90% + OPC 10%).
Therefore, when 2% sodium sulfate and 2% LBD are used in AAPSC (GGBS 90% + OPC 10%), it appears that the hydration reaction increased up to the level of the PSC (OPC 55% + GGBS 45%). 3.2. Setting time Fig. 5 shows the penetration depth of the needles for measuring the initial setting time of the Vicat apparatus after mixing the sodium sulfate and LBD with the cementitious materials in which GGBS and OPC were mixed. Because sodium sulfate was used, setting time tended to be short. With the mixture of LBD, the setting time of the cementitious paste increased. When 1% and 2% of the LBDs were mixed, however, the setting properties of the paste were similar. There was a significant decrease in the setting time due to the mixture of sodium sulfate was when 90% GGBS was used. This type of pattern appeared probably because the specific area of the GGBS was greater than that of the OPC. Such
Fig. 3. Heat of hydration of LBD.
W.-H. Yang et al. / Construction and Building Materials 57 (2014) 24–29
27
Fig. 4. Heat of hydration of OPC and PSC (OPC 55% + GGBS 45%).
Fig. 5. Penetration of the Vicat test (initial setting time test).
inclination was because the specific area of GGBS (7042 cm2/g) is greater than the specific area of OPC (3419 cm2/g), the case with 90% GGBS receives faster infiltration resistance of Vicat needle at the initial setting time than with 80% GGBS. Similarly, it is regarded that increase in activity due to sodium sulfate is affected quicker as well by using 90% of GGBS instead of 80% of GGBS. The inclination of decrease in the setting time as the fineness of GGBS is higher is similar to conventional results of study [3,19,20].
The initial setting time of the AAPSC is shown in Fig. 6. The setting time was 1 h and 20 min when 2% of sodium sulfate was used from 80% GGBS. However, the setting time was delayed to 4–5 h when 1–2% LBD was added. Generally, considering that the initial setting time of OPC is 3:20–4:00 and that the initial setting time of PSC is 4:10–4:50, and since this type has similar setting time to that of PSC, the result indicates that it is possible to achieve work hours up to the level of OPC when concrete is produced with AAPSC.
28
W.-H. Yang et al. / Construction and Building Materials 57 (2014) 24–29
Fig. 6. Initial setting time of AAPSC.
3.3. XRD analysis Fig. 7 shows the XRD patterns at 1, 3 and 28 days of aging after mixing sodium sulfate and LBD in the cementitious material in which GGBS (90%) and OPC (10%) were used. According to the XRD result, the major hydrate after 28 days of aging was calcium silicate hydrate. At 1–3 day(s) of aging, ettringite (3CaOAl2O3 3CaSO4 32H2O) and calcium silicate hydrate were confirmed as the major hydrate. A slight decrease in the peak of ettringite was then observed at 1 day of aging because of the addition of LBD. Ettringite is the hydrate of a cementitious material at initial aging [21]. The addition of LBD slightly slowed down the hydration of AAPSC by reducing the alkalinity at the initial period. Furthermore, in terms of the N2D2, where 2% sodium sulfate and 2% LBD were mixed, a small amount of calcium sulfate hydrate was generated from the reaction between enough sulfate ions and calcium ions [18]. However, at 3 d, the ettringite peaks of D1 and D2, both of in which LBD was mixed, were similar to that of D0, but there was no calcium sulfate hydrate peak. At 28 days, a sufficient amount of calcium silicate hydrate was observed, and there were no significant differences between samples. 3.4. Compressive strength Fig. 8 shows the compressive strength of the mortar of AAPSC after LBD was mixed in. As the mixture of LBD increased, the compressive strength of AAPSC continuously increased from the 3rd to the 28th day of aging. For the mixture of GGBS (80%) and OPC (20%), the strength increased significantly when sodium sulfate and LBD were used at the same time. Specifically, the addition of sodium sulfate dramatically enhanced the strength at 3 days. In addition, the mixture of LBD greatly increased the strength at the 7th and 28th days. When sodium sulfate and LBD were used at the same time, the strength at days 3, 7 and 28 continuously increased. For N2D1 and N2D2, the compressive strength at 28 days was 42–43 MPa, and it improved by 22–24% compared to the N0D0 in which sodium sulfate and LBD were not used. Therefore, when such strength properties are used effectively, a relatively good strength performance is expected, and the amount of OPC is greatly reduced. When sodium sulfate and LBD were mixed, the AAPSC was continuously hydrated, leading to an improved densification of the microstructure. Through the XRD analysis, there were no found that calcium hydroxide or magnesium hydroxide, but the improvement in the strength is assumed to be related with the generation of expansive hydrates.
Fig. 7. XRD patterns of AAPSC.
3.5. Pore structure Fig. 9 shows the cumulative pore volume at 28 days of aging after mixing LBD in AAPSC (80% of GGBS + 20% of OPC), in which 2% of sodium sulfate was used. In this case, the addition of the LBD decreased the pore volume (0.01–1.0 lm in particular). The capillary voids of cement are detrimental to strength and impermeability in the range over 50 nm, and in the range under 50 nm, the voids are known to be important to drying shrinkage and creep [1,22]. Therefore, the mixture of LBD enhanced the compressive strength of AAPSC and the density pore structure as well. It appears that this type of pore structure enhances the strength. Furthermore, it is predicted that it will affect positively on the reduction of drying shrinkage.
W.-H. Yang et al. / Construction and Building Materials 57 (2014) 24–29
29
(d) The compressive strength improved with the addition of LBD. This type of pattern got stronger when mixed with sodium sulfate. It continuously improved until the 28th day. In addition, this type of result was confirmed in a constant manner through the analysis on the pore structure.
Acknowledgement This subject is supported by Korea Ministry of Environment as ‘‘The Eco-technopia 21 project’’ (405-111-006). References
Fig. 8. Effect of LBD on compressive strength of AAPSC.
Fig. 9. Effect of LBD on cumulative pore volume of AAPSC (GGBS 80% + OPC 20%).
4. Conclusions This study used the powder alkaline sodium sulfate anhydrous (Na2SO4) as an activator in alkali-activated Portland slag cement (AAPSC), in which Portland cement (10–20%) and GGBS (80–90%) were used and added to light-burnt dolomite (LBD), and one of the economical lime materials. The result of effect by these materials on the hydration and strength of the AAPSC was analyzed as follows. (a) In Na2SO4-activated PSC (80–90% of GGBS + 10–20% of OPC), the 1st peak of the minute hydration heat increased with the addition of LBD. On the contrary, the 2nd peak was slightly delayed, and the cumulative heat of hydration increased. (b) A delay of the initial setting time was observed with the addition of LBD. (c) The major hydrates of AAPSC were calcium silicate hydrate and ettringite. It was confirmed that the mixture of LBD slightly delayed the creation of ettringite during the initial period.
[1] Mehta PK, Monteiro PJM. Concrete Microstructure, Properties, and Materials. 3rd ed. New York: McGraw-Hill Companies; 2006. [2] International Energy Agency (IEA), World Business Council for Sustainable Development (WBCSD). Cement Technology Roadmap 2009. Paris (FR): Corlet; 2010. [3] Shi C, Krivenko PV, Roy D. Alkali activated cement and concretes. New York: Taylor & Francis; 2006. [4] Bakharev T, Sanjayan JG, Cheng YB. Resistance of alkali-activated slag concrete to carbonation. Cem Concr Res 2001;31(9):1277–83. [5] Shi C. Corrosion resistance of alkali-activated slag cement. Adv Cem Res 2003;15(2):77–81. [6] Collins FG, Sanjayan JG. Strength and shrinkage properties of alkali-activated slag concrete containing porous coarse aggregate. Cem Concr Res 1999;29(4): 607–10. [7] Bakharev T, Sanjayan JG, Cheng YB. Effect of admixtures on properties of alkaliactivated slag concrete. Cem Concr Res 2000;30(9):1367–74. [8] Bakharev T, Sanjayan JG, Cheng YB. Effect of elevated temperature curing on properties of alkali-activated concrete. Cem Concr Res 1999;29(10):1619–26. [9] Collins FG, Sanjayan JG. Strength and shrinkage properties of alkali-activated slag concrete placed into a large column. Cem Concr Res 1999;29(5):659–66. [10] Collins FG, Sanjayan JG. Effect of pore size distribution on drying shrinkage of alkali-activated slag concrete. Cem Concr Res 2000;30(9):1401–6. [11] Ryu DW, Kim WJ, Yang WH, You JH, Ko JW. An experimental study on the freezing–thawing and chloride resistance of concrete using high volumes of GGBS. J Korea Inst Build Constr 2012;12(3):315–22. [12] Ryu DW, Kim WJ, Yang WH, Park DC. An experimental study on the carbonation and drying shrinkage of concrete using high volumes of ground granulated blast-furnace slag. J Korea Inst of Build Constr 2012;12(4): 393–400. [13] Yang WH, Ryu DW, Kim WJ, Park DC, Seo CH. An experimental study on early strength and drying shrinkage of high strength concrete using high volumes of ground granulated blast-furnace slag(GGBS). J Korea Inst Build Constr 2013;13(4):391–9. [14] Shen W, Wang Y, Zhang T, Zhou M, Li J, Cui X. Magnesia modification of alkaliactivated slag fly ash cement. J Wuhan Univ Technol-Mater 2011;126(1): 121–5. [15] Fang YH, Liu JF, Chen YG. Effect of magnesia on properties and microstructure of alkali-activated slag cement. Water Sci Eng 2011;4(4):463–9. [16] KS ISO 9597. Specifies the methods for determining standard consistence, setting times and soundness of cements. International Organization for Standardization, 2008. [17] KS L 5105. Testing method for compressive strength of hydraulic cement mortars. Korean Industrial Standards, 2007. [18] Shi C, Day RL. Pozzolanic reaction in the presence of chemical activators:Part II. Reaction products and mechanism. Cem Concr Res 2000;30(4):607–13. [19] Shi C, Li Y. Investigation on some factors affecting the characteristics of alkaliphosphorus slag cement. Cem Concr Res 1989;19(4):527–33. [20] Andersson R, Gram HE. Properties of alkali-activated slag concrete. Nordic Concr Res 1987;6:7–18. [21] Gartner EM, Young JF, Damidot DA, Jawed I. Hydration Portland Cem. In: Bensted J, Barnes P, editors. Structure and performance of cements. New York: Taylor & Francis; 2002. p. 57–139. [22] Mindess S, Young JF. Concrete. Englewood Cliffs NJ (USA): Prentice-Hall International Inc., 1981.