Hydration characteristics and environmental friendly performance of a cementitious material composed of calcium silicate slag

Journal of Hazardous Materials 306 (2016) 67–76

Contents lists available at ScienceDirect

Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat

Hydration characteristics and environmental friendly performance of a cementitious material composed of calcium silicate slag Na Zhang a,b , Hongxu Li a,b , Yazhao Zhao a , Xiaoming Liu a,b,∗ a b

School of Metallurgical and Ecological Engineering, University of Science and Technology Beijing, Beijing 100083, China Beijing Key Laboratory of Rare and Precious Metals Green Recycling and Extraction, University of Science and Technology Beijing, Beijing 100083, China

h i g h l i g h t s • • • •

Cementitious material was designed according to [SiO4 ] polymerization degree of raw materials. The cementitious material composed of calcium silicate slag yields excellent physical and mechanical properties. Amorphous C–A–S–H gel and rod-like ettringite are predominantly responsible for the strength development. Leaching toxicity and radioactivity tests show the cementitious material is environmentally acceptable.

a r t i c l e

i n f o

Article history: Received 14 August 2015 Received in revised form 27 November 2015 Accepted 28 November 2015 Available online 2 December 2015 Keywords: Calcium silicate slag High-alumina fly ash Cementitious material Hydration products Environmental friendly performance

a b s t r a c t Calcium silicate slag is an alkali leaching waste generated during the process of extracting Al2 O3 from high-alumina fly ash. In this research, a cementitious material composed of calcium silicate slag was developed, and its mechanical and physical properties, hydration characteristics and environmental friendly performance were investigated. The results show that an optimal design for the cementitious material composed of calcium silicate slag was determined by the specimen CFSC7 containing 30% calcium silicate slag, 5% high-alumina fly ash, 24% blast furnace slag, 35% clinker and 6% FGD gypsum. This blended system yields excellent physical and mechanical properties, confirming the usefulness of CFSC7. The hydration products of CFSC7 are mostly amorphous C–A–S–H gel, rod-like ettringite and hexagonalsheet Ca(OH)2 with small amount of zeolite-like minerals such as CaAl2 Si2 O8 ·4H2 O and Na2 Al2 Si2 O8 ·H2 O. As the predominant hydration products, rod-like ettringite and amorphous C–A–S–H gel play a positive role in promoting densification of the paste structure, resulting in strength development of CFSC7 in the early hydration process. The leaching toxicity and radioactivity tests results indicate that the developed cementitious material composed of calcium silicate slag is environmentally acceptable. This study points out a promising direction for the proper utilization of calcium silicate slag in large quantities. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Fly ash is a typical industrial solid waste obtained from thermal power plant. In Inner Mongolia of China, large quantities of fly ash are disposed every year, and most of them are high-alumina fly ash with alumina content as high as 50% [1–3]. It was reported that over 12 million tons of high-alumina fly ash is generated annually in Inner Mongolia, and its stockpile amount has exceeded 100 million tons [4]. Extracting Al2 O3 from this high-alumina fly

∗ Corresponding author at: Room 810, School of Metallurgical and Ecological Engineering, University of Science and Technology Beijing, Beijing 100083, China. Fax: +86 10 62332786. E-mail address: [email protected] (X. Liu). http://dx.doi.org/10.1016/j.jhazmat.2015.11.055 0304-3894/© 2015 Elsevier B.V. All rights reserved.

ash is a good solution for China to relieve the shortage of bauxite resource and develop circular economy. In order to utilize this kind of fly ash resource, Inner Mongolia Datang International Renewable Resources Development Co. Ltd., developed pre-desilication and alkali-lime-calcination process to extract Al2 O3 from highalumina fly ash [4,5], and a production line with annual capacity of 0.2 million tons of alumina has been built beside Inner Mongolia Datang Togtoh power plant. However, an insoluble residue named calcium silicate slag is generated during the alumina extraction from fly ash. Usually consuming 1 t of high-alumina fly ash generates 1 t of calcium silicate slag and 0.4 t of Al2 O3 . Due to NaOH and Na2 CO3 are added during the process of extracting Al2 O3 from highalumina fly ash by the pre-desilication and alkali-lime-calcination method, calcium silicate slag is an alkali leaching waste containing as high as 5% Na2 O. As a corrosively hazardous material, the

68

N. Zhang et al. / Journal of Hazardous Materials 306 (2016) 67–76

Fig. 1. Particle size distribution of dried calcium silicate slag after milling.

discharge of this alkali residue will cause serious environmental problems such as soil contamination and water pollution due to lye leaching. If calcium silicate slag cannot be utilized in an effective way, it will not only seriously influence the local ecological environment of Inner Mongolia of China, but also restrict the application and popularization of extracting Al2 O3 from high-alumina fly ash. As calcium silicate slag is mainly composed of CaO, SiO2 , Al2 O3 and Na2 O, it is thought that using calcium silicate slag in cement production is an efficient method for large-scale recycling of this solid waste. In the past few years, using calcium silicate slag in cement clinker has been widely carried out in China [6–9]. It was reported that calcium silicate slag can improve burnability of raw meal during the preparation of cement clinker [8,9]. However, it is noticed that the calcium silicate slag used in the production of cement clinker contains small amount of alkali (Na2 O + 0.658K2 O) [10]. Due to the alkali content of raw meal needs to be controlled in a low level, the calcium silicate slag is required to be pretreated by dealkalization and after that it can be used for producing cement clinker. Generally, calcium silicate slag is dealkalized through adding lime milk, washing and filtering process. It was reported that the optimal dealkalization process for the calcium silicate slag derived from Inner Mongolia Datang International Renewable Resources Development Co. Ltd., was lime milk addition of 10%, temperature of 85 ◦ C for 3 h and washing twice, and subsequently the alkali content of dealkalized calcium silicate slag was 0.83% (comparing to the original alkali content of 4.12%) in experimental investigation [11]. Moreover, preparation of cement clinker requires a high temperature between 1350 ◦ C and 1450 ◦ C consuming lots of energy. Yang et al. [5] conducted an investigation on using raw calcium silicate slag and dealkalized calcium silicate slag respectively as an admixture in Portland cement, and it showed that the early-age compressive strength of cement composed of raw calcium silicate slag is higher than that of cement composed of dealkalized calcium silicate slag, but the dealkalized calcium silicate slag is more beneficial to maintain the long-term strength, lower the hydration heat and reduce the dry shrinkage of the cement comparing with the raw calcium silicate slag. As a matter of fact, the high amount of Na2 O in calcium silicate slag without dealkalization restricts its large-scale utilization in cement production. Red mud is a typical industrial waste obtained from alumina production with high amount of Na2 O (2.0–6.0%) [12]. Many efforts have been made in our research team to find effective ways of utilizing red mud. We have carried out some investigations on using bauxite-calcination-method red mud, coal gangue, blast furnace slag, clinker and gypsum as raw materials to produce cementitious materials [12–14]. It has been demonstrated that the developed red mud-coal gangue based cementitious material has good phys-

ical and mechanical properties [12], and fibrous C–A–S–H gels and needle-shaped/rod-like ettringite are mainly formed in the early hydration period, and then the fibrous intertwined C–A–S–H gels gradually grow into amorphous phase and further into network shape with increasing of the hydration time [13]. Feng found out that C–A–S–H gel with low Ca/Si ratio was formed in the hydrated paste of silica-alumina based cementitious material composed of red mud [15], and due to Na+ balanced the negative charges which were caused by the substitution of Al for Si, this kind of hydration product had high capability of Na+ solidification. The usefulness of red mud based cementitious materials has been confirmed through these investigations [12–17], which provides important guidance for using calcium silicate slag to prepare cementitious materials. In this paper, according to [SiO4 ] polymerization degree of raw materials, calcium silicate slag was blended with high degree of [SiO4 ] polymerization material (high-alumina fly ash), middle degree of [SiO4 ] polymerization material (blast furnace slag), and low degree of [SiO4 ] polymerization material (clinker) as main compositions to produce cementitious material. The aim of the present study is to investigate the feasibility of the developed cementitious material composed of calcium silicate slag, including its mechanical and physical properties, hydration products, hydration kinetics and environmental friendly performance. Instrumental techniques such as X-ray diffraction (XRD), Fourier transform infrared (FTIR) and scanning electron microscopy (SEM) were applied to obtain useful information on the hydration products of cementitious material composed of calcium silicate slag. 2. Experimental 2.1. Materials Calcium silicate slag, high-alumina fly ash and FGD gypsum were obtained from Inner Mongolia Datang International Renewable Resources Development Co. Ltd. The calcium silicate slag used in this work was original without dealkalization. Granulated blast furnace slag (BFS) was supplied by Tangshan steel refining plant, and clinker was provided by Inner Mongolia Mengxi cement plant. The chemical composition (analyzed using XRF-1800 sequential X-ray fluorescence spectrometer) and specific surface area of raw materials are presented in Table 1. The mineralogical phases (determined by XRD) of raw materials are presented in Table 2. 2.2. Experimental procedure Calcium silicate slag was dried at 105 ◦ C in an oven and ground in a laboratory ball mill for 20 min. The size distribution, SEM observation and thermal analysis of the dried calcium silicate slag after milling were performed. The calcium silicate slag was blended with

N. Zhang et al. / Journal of Hazardous Materials 306 (2016) 67–76

69

Table 1 Chemical composition and specific surface area of raw materials. Oxides (%)

Calcium silicate slag

Fly ash

BFS

Clinker

FGD gypsum

SiO2 Al2 O3 CaO Fe2 O3 Na2 O K2 O MgO TiO2 SO3 LOI Specific surface area (m2 /kg)

22.92 8.17 40.01 1.78 4.11 0.31 2.02 0.91 0.73 17.70 475

49.87 42.38 2.46 2.44 0.17 0.31 0.22 0.82 0.32 – 505

34.19 12.53 37.25 1.22 0.22 0.51 9.33 0.41 2.13 – 405

22.19 4.21 63.65 3.58 0.30 1.03 3.57 – 0.88 – 400

3.06 1.41 38.10 0.58 0.19 0.34 1.58 – 45.77 7.20 410

Table 2 Mineralogical phases of raw materials. Phase constitution

Calcium silicate slag

Fly ash

BFS

Clinker

FGD gypsum

Dicalcium silicate, C2 S Tricalcium aluminate, C3 A Calcite Aragonite Gibbsite Mullite Quartz Gehlenite Akermanite Tricalcium silicate, C3 S Tetracalcium aluminoferrite, C4 AF Ca2 SO4 ·2H2 O

** ** ** ** ** X X X X X X X

X X X X X ** ** X X X X X

X X X X X X X ** ** X X X

** ** X X X X X X X ** ** X

X X X X X X X X X X X **

**: Phase existing; X: phase absent.

Table 3 Designed proportions of cementitious materials (%). Symbol

Calcium silicate slag

Fly ash

BFS

Clinker

FGD gypsum

CFSC1 CFSC2 CFSC3 CFSC4 CFSC5 CFSC6 CFSC7

0 19.5 39 0 22 44 30

39 19.5 0 44 22 0 5

25 25 25 20 20 20 24

30 30 30 30 30 30 35

6 6 6 6 6 6 6

granulated blast furnace slag, fly ash, clinker and FGD gypsum in appropriate proportions to produce cementitious materials. The designed proportions are listed in Table 3, in which seven batches were prepared for the mechanical test with purpose of finding the best mix proportion of cementitious material composed of calcium silicate slag. All of these seven batches were designed with the Ca/(Si + Al) ratio less than 1.5 in the chemical composition of the blended system. Mechanical tests were performed according to Chinese Standard GB/T 17671-1999 [18]. Mortar specimens in size of 40 mm × 40 mm × 160 mm were prepared with a water/cement ratio of 0.50 and cement/sand ratio of 1:3. Subsequently, they were cured in a moist cabinet at 95% humidity and 20 ◦ C for 24 h, and then demoulded and transferred to an isothermal curing cabinet at the previously mentioned humidity and temperature. Fluidity of mortars was tested according to Chinese Standard GB/T 2419-2005 [19]. Setting time and soundness of the cementitious material were determined according to Chinese Standard GB/T 1346-2011 [20]. The pastes of cementitious material composed of calcium silicate slag were prepared with water to solid ratio of 0.35 and molded in size of 20 mm × 20 mm × 20 mm. They were cured in a moist cabinet at 95% humidity and 20 ◦ C for 24 h, and then demoulded and placed in the isothermal curing cabinet at the previously mentioned humidity and temperature. The hydration of the pulverized and sieved specimens was terminated by alcohol drenching at the

desired testing time, and then dried at 60 ◦ C in a vacuum oven for further characterization. The size distribution was performed using LMS-30 size distribution analyzer. SEM observation was carried out on JSM-6480LV scanning electron microscope. TG-DSC analysis was conducted using Netzsch STA 409C comprehensive thermal analyzer. XRD analysis was carried out using M21X X-ray diffractometer with CuK␣ radiation, a voltage of 40 kV, a current of 200 mA and 2 scanning ranging between 5◦ and 70◦ . FTIR analysis was performed using a Spectrum GX PerkinElmer Fourier transform infrared spectrometer. The hydration heat evolution rate and cumulative hydration heat of cementitious material composed of calcium silicate slag were measured using an isothermal calorimeter (TAMAIR), and this test was performed under a constant temperature of 20 ◦ C within 180 h. The leaching toxicity test of the developed cementitious material were carried out according to the solid waste – extraction procedure for leaching toxicity – horizontal vibration method (Chinese Standard HJ 557-2010) [21]. The concentrations of heavy metals were analyzed using OPTIMA 7000DV ICP optical emission spectrometer. 3. Results and discussion 3.1. Characterization of the calcium silicate slag As shown in Table 1, the main oxides of calcium silicate slag used in this study are CaO, SiO2 , Al2 O3 and Na2 O. With Na2 O content of 4.11%, Na+ is mainly occurring in the pore solution of calcium silicate slag. It can be seen from Table 2 that the major mineral components of the calcium silicate slag are Ca2 SiO4 , calcite (CaCO3 ), aragonite (CaCO3 ), gibbsite (Al(OH)3 ) and Ca3 Al2 O6 . The particle size distribution of the dried calcium silicate slag after milling is shown in Fig. 1. It can be seen that 80% of the ground calcium silicate slag particles are within the range of 1.46–9.54 ␮m with a mean diameter of 4.59 ␮m. Combining the SEM image in Fig. 2, it is

70

N. Zhang et al. / Journal of Hazardous Materials 306 (2016) 67–76

Fig. 2. SEM image of dried calcium silicate slag after milling.

observed that the shape of the ground calcium silicate slag particles are irregular, and fine particles with diameter less than 1 ␮m are also present in the dried calcium silicate slag after milling. The TG-DSC diagram of dried calcium silicate slag is given in Fig. 3. The TG curve displays two major steps for mass loss with the change of temperature. The first step at a temperature range of 35–400 ◦ C corresponds to the evaporation of physically absorbed water and chemical bonding water. As the calcium silicate slag has been dried at 105 ◦ C before using for TG-DSC analysis, there is a small amount of physically absorbed water occurring in the dried calcium silicate slag. Therefore, the mass loss of 7.87% at the first step is mainly related to the evaporation of chemical bonding water. The second step with a mass loss of 9.81% can be detected in the temperature range of 400–1000 ◦ C mostly corresponding to the release of CO2 from the decomposition of aragonite and calcite. The DSC curve shows mainly five endothermic peaks. The peak located at 110 ◦ C is attributed to evaporation of the left physically absorbed water. The peak around 260 ◦ C is associated with the loss of chem-

ical bonding water from the decomposition of gibbsite (Al(OH)3 ). The endothermic peaks around 647 ◦ C and 742 ◦ C correspond to the decomposition of aragonite and calcite, respectively. Combining the chemical composition of calcium silicate slag with the above TG-DSC analysis, more valuable information can be obtained. As shown in Table 1, the content of Al2 O3 is 8.17% in the calcium silicate slag. If considering that all amount of Al2 O3 (8.17%) was fixed as Al(OH)3 , the mass loss from the decomposition of Al(OH)3 would be 4.33%. However, the mass loss at the first step is 7.87% according to the TG analysis, which confirms that there is another phase in the calcium silicate slag losing chemical bonding water with an endothermic peak around 153 ◦ C presented in the DSC curve. Based on the CO2 loss (9.81%) at the second step recorded by TG analysis, the amount of CaO fixed as CaCO3 was calculated to be 12.49%. As Table 1 shows 40.01% CaO is composed in the calcium silicate slag, the content of CaO not fixed as CaCO3 would be 27.52%. If considering that the left amount of CaO (27.52%) was fixed as Ca2 SiO4 , the amount of SiO2 fixed as Ca2 SiO4 would be 14.74%. Whereas the content of SiO2 is 22.92% as shown in Table 1, this suggests an amorphous phase composed of SiO2 could be present in the calcium silicate slag. Certain amounts of amorphous aluminosilicates have been found in the bauxite-calcination-method red mud [22,23]. Thus, it is thought that amorphous aluminosilicate phase could be also occurring in the calcium silicate slag, and the endothermic peak at 153 ◦ C displayed in the DSC curve is most likely to be associated with the decomposition of amorphous aluminosilicate phase.

3.2. Physical and mechanical properties The general physical and mechanical properties of cementitious materials corresponding to Table 3 are presented in Table 4. It can be seen that all the designed cementitious materials have good fluidity higher than 180 mm and qualified soundness. The 3-day flexural and compressive strength values of blended cementitious materials composed of calcium silicate slag (CFSC2, CFSC3, CFSC5, CFSC6 and CFSC7) are higher than those of the cementitious materials without calcium silicate slag (CFSC1 and CFSC4). It is noted that with

Fig. 3. TG-DSC diagram of dried calcium silicate slag.

N. Zhang et al. / Journal of Hazardous Materials 306 (2016) 67–76

71

Table 4 Physical and mechanical properties of cementitious materials. Sample

Setting time (min)

CFSC1 CFSC2 CFSC3 CFSC4 CFSC5 CFSC6 CFSC7

Soundness

Initial

Final

200 95 40 195 85 35 60

260 160 60 280 140 55 125

1

Fluidity (mm)

Good Good Good Good Good Good Good

205 190 185 210 200 190 195

Flexural strength (MPa)

Compressive strength (MPa)

3 days

28 days

3 days

28 days

4.8 5.9 6.8 4.7 5.4 6.7 7

8.2 8.6 8.8 8 8.4 8.7 7.9

17.2 24.3 26.5 16.8 23.6 25.3 22.9

38.9 38.6 38.4 36.6 37.0 37.3 37.5

1

1. C-S-H

2. Ca(OH)2

3. AFt

4. CaAl2 Si2O8 . 4H2O

5. Na2Al2Si2 O8 . H2O 6. C3S

1 5 5

7 4 6 3 3

2

5

10

15

20

4

5

6

30

35

40

45

50

55

60

65

70

5

10

15

20

25

30

2 3 4

35

40

45

2 4

50

55

60

65

70

2 theta (degree )

2 theta (degree)

Fig. 5. XRD pattern of CFSC7 paste hydrated for 90 days.

Fig. 4. XRD pattern of CFSC7 paste hydrated for 3 days.

increasing the content of calcium silicate slag designed in Table 3, both of the initial and final setting time of the cementitious materials decrease significantly. The initial setting time of CFSC3 and CFSC6 composed of 39% and 44% calcium silicate slag is 40 min and 35 min, respectively, which can not satisfy the requirement in Chinese Standard GB 175-2007 for common Portland cement (initial setting time should not be less than 45 min) [10]. Samples of CFSC2, CFSC5, and CFSC7 possess good physical and mechanical properties, and their strength can meet the requirements of P.C 32.5R in Chinese Standard GB 175-2007 (3 days compressive strength ≥15 MPa; 28 days compressive strength ≥32.5 Mpa; 3 days flexural strength ≥3.5 MPa; 28 days flexural strength ≥5.5 MPa) [10]. CFSC7 is an optimal design for the blended cementitious material when taking into account the above excellent properties and the large utilization ratio of calcium silicate slag comprehensively. Therefore, the proper mix ratio for the cementitious material composed of calcium silicate slag can be determined by the specimen CFSC7 containing 30% calcium silicate slag, 5% high-alumina fly ash, 24% blast furnace slag, 35% clinker and 6% FGD gypsum. With high 3-day flexural strength of 7 MPa, the compressive strength of this developed cementitious material composed of calcium silicate slag at 3 and 28 days can achieve to 22.9 and 37.5 MPa. Subsequently, characterization of the hydrated CFSC7 paste was further carried out to investigate the hydration products of the developed cementitious material composed of calcium silicate slag.

3

5 2 3

3

2

25

4

5

3

4. CaAl2Si2O 8 . 4H2O

4 3 5

6 2

4

2. Ca(OH) 2

3. AFt

5. Na2 Al2Si 2O8 . H2O

1

7. C2S

1. C-S-H

cate hydrate (C–S–H) gel, portlandite (Ca(OH)2 ) and ettringite (AFt, Ca6 Al2 (SO4 )3 (OH)12 ·26H2 O) are occurring in the hydrated pastes of CFSC7. Besides, both of the XRD patterns show some diffraction peaks of zeolite-like minerals such as CaAl2 Si2 O8 ·4H2 O and Na2 Al2 Si2 O8 ·H2 O. It is notable that both patterns show a broad diffuse halo in the background between 2 of 25◦ and 38◦ , indicating that amorphous C–A–S–H gel is present in the hydrated pastes of CFSC7. Comparing with the XRD pattern in Fig. 5, the diffraction peaks of Ca2 SiO4 (C2 S) and Ca3 SiO5 (C3 S) can be detected in the XRD pattern of CFSC7 paste hydrated for 3 days in Fig. 4. Ca2 SiO4 is obtained from calcium silicate slag and clinker, while Ca3 SiO5 is obtained from clinker. Both of them are hydrated gradually to form C–S–H gel and they cannot be detected in the XRD pattern of CFSC7 paste hydrated for 90 days. Ca(OH)2 is a crystalline product mainly from the hydration of clinker, and it is consumed by the pozzolanic reaction of calcium silicate slag, fly ash and blast furnace slag to form C–A–S–H gel, which can be expressed as the following: SiO2 + OH− + H2 O → [H3 SiO4 ]− −

−

(1)

AlO2 + OH + H2 O → [H3 AlO4 ] −

[H3 SiO4 ] + [H3 AlO4 ]

2−

+ Ca

2+

2−

→C−A−S−H

(2) (3)

where SiO2 represents reactive siliceous substances occurring in the calcium silicate slag, fly ash and blast furnace slag, and AlO2 − represents reactive aluminous substances dissolved from the calcium silicate slag, fly ash and blast furnace slag.

3.3. Hydration products 3.3.1. XRD analysis Figs. 4 and 5 display XRD patterns of the CFSC7 pastes hydrated for 3 and 90 days, respectively. It can be seen that calcium sili-

3.3.2. FTIR analysis Fig. 6 shows FTIR spectra of the CFSC7 pastes hydrated for 3 and 90 days. Both spectra show bands at 3435, 1630, 1440, 980 and 875 cm−1 except the band at 3644 cm−1 in the spectrum of CFSC7

72

N. Zhang et al. / Journal of Hazardous Materials 306 (2016) 67–76

70

90d

Transmittance (%)

60

50

3d 1630

40

3644 30

3435 1440

20

10 4000

875 980 3500

3000

2500

2000

1500

1000

-1

Wavenumbers Ocm P Fig. 6. FTIR spectra of CFSC7 pastes hydrated for 3 and 90 days.

paste hydrated for 3 days. The band around 3644 cm−1 is related to Ca OH stretching vibration in Ca(OH)2 [24]. This band disappears in the paste hydrated for 90 days, indicating that Ca(OH)2 gradually participates into the hydration due to the pozzolanic reaction of calcium silicate slag, fly ash and blast furnace slag. The wide band around 3435 cm−1 is associated with Al OH stretching vibration in the [Al(OH)6 ] octahedral structure of ettringite. The band at 1630 cm−1 corresponds to bending vibration of H O H band for interlayer water [17]. With increasing of the hydration period, free water participates into the hydration and gradually transforms into crystal water. It is noted that the bands at 1440, 980 and 875 cm−1 are indicative of the presence of C–S–H gel. With the increase of hydration time from 3 to 90 days, these three bands associated with C–S–H gel tend to sharpen and their transmittances decrease correspondingly, suggesting that more and more C–S–H gel forms and the amorphous C–S–H gel tends to crystallize along with the hydration going on. The band around 875 cm−1 is due to bending vibration of Si OH band in the C–S–H gel. The band at 980 cm−1 is attributed to anti-symmetric stretching vibration of Si–O–Si(Al) in the [SiO4 ] tetrahedral structure of C–S–H gel [25]. It is reported that C–S–H gels formed in the hydrated pastes of Portland cement generally have a typical infrared absorption band around 960–970 cm−1 [25,26], whereas here the C–S–H gel formed in the hydrated pastes of CFSC7 has a higher wavenumbers at 980 cm−1 , supporting a more polymerized structure. According to Si O stretching vibrations for the SiQn units displaying infrared absorption bands around 850, 900, 950, 1100 and 1200 cm−1 for n = 0, 1, 2, 3 and 4, respectively [25,27], the band at 980 cm−1 illustrates the SiQn units distributed around SiQ2 and SiQ3 for the CFSC7 pastes hydrated for 3 and 90 days. It suggests that the C–S–H gel formed in the developed cementitious material composed of calcium silicate slag is similar to the aluminosilicate gels found in the alkali-activated systems [25,26,28,29]. Therefore, it is believed that C–A–S–H gel with a more polymerized structure of SiQ2 and SiQ3 units is present in the hydrated pastes of cementitious material composed of calcium silicate slag.

3.3.3. SEM analysis The microstructure of the CFSC7 pastes hydrated for 3 and 90 days is shown in Fig. 7. The SEM image of 3-day hydrated CFSC7 paste shows the presence of rod-like ettringite and amorphous C–A–S–H gel with hexagonal-sheet Ca(OH)2 crystals being evident in localized areas. It is observed that unhydrated particles are coated by a layer of amorphous C–A–S–H gel, and the rod-like ettringite crystals are dispersed through the hydrated paste. The hydration products and unhydrated substances are well connected to a cementitious matrix, leading to a compacted structure with low porosity. With increasing of the hydration time, the structure of the hydrated paste is continually densified. It can be seen that the unhydrated particles are further connected to a whole matrix with a thicker and denser layer of amorphous C–A–S–H gel coating on them from the SEM image of 90-day hydrated CFSC7 paste. While few hexagonal-sheet Ca(OH)2 crystals can be found in the hardened paste at the hydration time of 90 days, suggesting that Ca(OH)2 was depleted gradually by the pozzolanic reaction of calcium silicate slag, fly ash and blast furnace slag. This agrees well with the above FTIR analysis. It is thought that as the predominant hydration products, rod-like ettringite crystals and amorphous C–A–S–H gel play a positive role in promoting the densification of the paste structure, and also play an important role in the strength development of the cementitious material composed of calcium silicate slag in the early hydration process. In our previous work [12,17], we also found that amorphous aluminous C–S–H gel and ettringite are principally responsible for the strength development of bauxite-calcination-method red mud based cementitious materials (such as red mud-coal gangue based or red mud-fly ash based). The Ca/(Si + Al) ratio of these cementitious materials is designed less than 1.5 in common, the hydration products of which are generally C-A-S-H gel and ettringite. From the above analyses of hydration products, it is thought that in this blended cementitous material, Ca(OH)2 from the hydration of low degree of [SiO4 ] polymerization material—clinker is essential for the pozzolanic reaction of calcium silicate slag, blast furnace slag and high-alumina fly ash. Appropriate amounts of clinker and calcium silicate slag provide an alkaline environment for the hydration reaction of middle degree of [SiO4 ] polymerization material – blast

N. Zhang et al. / Journal of Hazardous Materials 306 (2016) 67–76

73

Fig. 7. Microstructure of CFSC7 pastes hydrated for (a) 3 days and (b) 90 days.

furnace slag and high degree of [SiO4 ] polymerization material – high-alumina fly ash. In this case, ions of [H3 SiO4 ]− , [H3 AlO4 ]2− and [Al(OH)6 ]3− are dissolved from the reactive siliceous and aluminous substances occurring in the calcium silicate slag, blast furnace slag and high-alumina fly ash. When the hydration starts, clinker and FGD gypsum react quickly with water to release Ca2+ , OH− , [H3 SiO4 ]− , [Al(OH)6 ]3− , and SO4 2− . In this complicated liquid environment, [Al(OH)6 ]3− will combine with Ca2+ and SO4 2− to form ettringite, and [H3 SiO4 ]− , [H3 AlO4 ]2− will react with Ca2+ to form C–A–S–H gel gradually. 3.4. Hydration kinetics In order to deeply understand the hydration reaction of this cementitious material composed of calcium silicate slag, the hydration kinetics was investigated based on the hydration heat analysis. Fig. 8 displays the hydration heat evolution rate and cumulative hydration heat of specimen CFSC7 at 20 ◦ C. It can be seen from Fig. 8(a) that the curve of hydration heat evolution rate shows the first exothermic peak in the initial hydration time up to 1 h which is due to the release of surface energy and the fast reaction of aluminates and sulphates to form AFt when the cementitious material contact with water [30,31]. According to the characteristics of hydration heat evolution rate, the hydration process of cement based materials is mostly divided into five periods: quick reaction period, induction period, acceleration period, deceleration period and decay period. As shown in Fig. 8, the acceleration period of CFSC7 started from 3.32 h with cumulative hydration heat of 16.23 J/g, and it finished at about 9.6 h along with the appearance of second exothermic peak. Krstulovic´ and Dabic´ [32] proposed a kinetics model of the cement hydration process which includes three basic processes: nucleation and crystal growth (NG), interactions at phase bound´ aries (I) and diffusion (D). The Krstulovic–Dabi c´ model has been successfully used to study the hydration reaction of Portland cement, blended cement composed of fly ash or blast furnace ´ slag [33,34]. In this paper, the Krstulovic–Dabi c´ model was also adopted to investigate the hydration kinetics of specimen CFSC7. ´ The kinetics equations in the Krstulovic–Dabi c´ model are expressed as follows.NG process: [−ln (1 − ˛)]

1/n

=

ln [−ln (1 − ˛)] =

K1

(t − t0 )

n × lnK1

+ n × ln (t − t0 )

d␣ = F1 (˛) = K1 n (1 − ˛) [−ln (1 − ˛)](n−1)/n dt

Fig. 8. Hydration heat evolution rate (a) and cumulative hydration heat (b) of CFSC7 at 20 ◦ C.

I process: 1

(4)

[1 − (1 − ˛)1/3 ] = K2 (t − t0 )

(7)

(5)

ln[1 − (1 − ˛)1/3 ] = lnK2 + ln (t − t0 )

(8)

(6)

d˛ = F2 (˛) = 3 × K2 (1 − ˛)2/3 dt

(9)

74

N. Zhang et al. / Journal of Hazardous Materials 306 (2016) 67–76 -1

51.6

51.5

-2

1/Q=0.004133+0.09692/(t-t0)

ln[-ln(1-a)]= -4.02803+1.23358ln(t-t0) R=0.99732

R=0.99997

51.4

ln[-ln(1-a)]

-3

-1

(1/Q)×10 / (J ·g)

(a)

4

51.3

51.2

-4

-5

51.1 -6

51.0 -2.5

10.0

10.1

10.2

10.3 3

10.4

-2.0

-1.5

-1.0

-0.5

10.5

0.0

0.5

1.0

1.5

2.0

2.5

ln(t-t0)

-1

[1/(t-t0)]×10 / h

Fig. 9. Determination of maximum hydration heat (Qmax ) from linear regression.

-4

D process:

-5

] =

K3

(t − t0 )

1/3

d˛ (1 − ˛)2/3 = F3 (˛) = 3 × K3   dt 2 − 2(1 − ˛)1/3

(12)

1/3

(11)

where ␣ is the degree of hydration reaction; K1  , K2  and K3  is the reaction rate constant; t0 is the hydration time when the induction period ends (also the time when the acceleration period starts); n is the reaction exponent; d˛/dt is the reaction rate. In order to use the hydration heat data presented in Fig. 8 to calculate the degree of hydration reaction (˛) and the reaction rate (d˛/dt), the following equations are employed:



dQ/dt

-7

-8

-9

-10 -5.0

-4.5

-4.0

-3.5

-3.0

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

ln(t-t0)

(13)



Qmax

t50 1 1 + = Q (t) Qmax [Qmax × (t − t0 )]

(14)

-1.0

(15)

where Qmax is the ultimate total hydration heat; (t − t0 ) is the hydration time starting from the acceleration period; Q(t) is the cumulative hydration heat at reaction time of (t − t0 ) which needs to be calculated from the beginning of acceleration period; dQ/dt is the hydration heat evolution rate; t50 is the hydration reaction time when the cumulative hydration heat is half of Qmax . Eq. (15) is the Knudson kinetics formula which determines Qmax through linear regression. The relationship between 1/Q(t) and 1/(t − t0 ) for the specimen CFSC7 is shown in Fig. 9. From the linear regression presented in Fig. 9, Qmax of CFSC7 can be obtained to be 242 J/g. Based on the Eqs. (13), (5), (8), and (11), the hydration kinetics factors of n, K1  , K2  and K3  can be obtained through linear regression. Fig. 10 presents determination of kinetics factors of NG process, I process and D process from linear regression, respectively, and the hydration kinetics factors of CFSC7 at 20 ◦ C are obtained and listed in Table 7. Subsequently, based on the Eqs. (6), (9), (12), and (14), the curves of relationship between the reaction rate (F1 (˛), F2 (˛), F3 (˛) and d˛/dt) and the hydration reaction degree (˛) are obtained and shown in Fig. 11. It can be seen from Fig. 11 that the curves of F1 (␣) and F3 (␣) have been simulated well in the former part (when ˛ ≤ 0.06) and latter part (when

(c) 1/3

2ln[1-(1-a) ]= -6.428+1.00055ln(t-t0)

-1.5

R=0.98066 -2.0 1/3

d˛ = dt

Q (t) Qmax

R=0.99964

-6

2ln[1 − (1 − ˛)1/3 ] = lnK3 + ln (t − t0 )

˛ (t) =

ln[1-(1-a) ]= -5.19212+1.0005ln(t-t0)

(10)

ln[1-(1-a) ]

1/3 2

2ln[1-(1-a) ]

[1 − (1 − ˛)

(b)

-2.5

-3.0

-3.5

-4.0 2.5

3.0

3.5

4.0

4.5

5.0

5.5

ln(t-t0) Fig. 10. Determination of kinetics factors of NG process, I process and D process from linear regression: (a) NG process, (b) I process, (C) D process.

N. Zhang et al. / Journal of Hazardous Materials 306 (2016) 67–76 Table 7 Hydration kinetic factors of CFSC7 at 20 ◦ C.

0.06

0.05

da/dt

0.04

75

F1(a) F2(a)

aD=0.205

aNG-D=0.06

n

K1 

1.23358

0.03818

K2 

K3 

˛NG-D

˛D

0.00162

0.06

0.205

˛NG-D is the hydration reaction degree from which both NG and D dominate the hydration reaction; ˛D is the hydration reaction degree from which D controls the hydration reaction.

F3(a) da/dt

0.03

0.02

0.01

0.00 0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

a Fig. 11. Hydration rate curves of CFSC7 at 20 ◦ C.

Table 5 Leaching toxicity test results. Element

CFSC7 (ppm)

GB 5749-2006 limits (ppm)

Arsenic Antimony Cadmium Chromium Copper Lead Manganese Nickel Zinc

<0.001 <0.001 <0.001 <0.001 0.182 <0.001 <0.001 <0.001 0.000

0.010 0.005 0.005 0.050 1.000 0.010 0.100 0.020 1.000

˛ ≥ 0.205) of the actual hydration rate (d˛/dt) curve. However, the curve of F2 (˛) has deviated from the d␣/dt curve, suggesting that the interactions at phase boundaries (I) process is excluded in the hydration reaction of CFSC7. It is thought that the hydration reaction of CFSC7 involves two basic processes which are nucleation and crystal growth (NG) and diffusion (D). NG dominates at the early hydration stage within the hydration reaction degree (˛) of 0.06, and D controls the later hydration stage with ˛ larger than 0.205. The stage with ˛ between 0.06 and 0.205 is a transition stage which is controlled by both NG and D processes, and with increasing of ˛, nucleation and crystal growth process gradually transforms into diffusion controlling. 3.5. Environmental friendly performance As the developed cementitious material composed of calcium silicate slag can be utilized mainly as building materials and stabilization/solidification materials, it is necessary to evaluate its environmental impact. Here the leaching toxicity and radioactivity of CFSC7 sample have been tested to investigate the environmental friendly performance of the produced cementitious material composed of calcium silicate slag, and the results are shown in Tables 5 and 6, respectively. Because the main environmental concern about utilization of cementitious materials composed of industrial solid wastes is

the possibility of certain hazardous constitutes leaching into the groundwater with concentrations determined to be potentially harmful to human health [35,36], standards for drinking water quality in Chinese Standard GB 5749-2006 [37] was used to evaluate the leaching toxicity results listed in Table 5. It can be seen from Table 5 that none of the metal leaching concentrations exceeds the limits specified in standards for drinking water quality (Chinese Standard GB 5749-2006), indicating that the developed cementitious material composed of calcium silicate slag is friendly to groundwater. Table 6 reveals that as a main material for building, the internal exposure index and external exposure index of the produced cementitious material composed of calcium silicate slag are less than 1.0, which can satisfy the requirement in Chinese Standard GB 6566-2010 [38]. These results indicate that the developed cementitious material composed of calcium silicate slag is environmentally acceptable. This cementitious material composed of calcium silicate slag has been successfully applied for 70 m long road construction in the plant of Inner Mongolia Datang International Renewable Resources Development Co. Ltd. Moreover, it is found that concrete prepared by the cementitious material composed of calcium silicate slag has excellent durable performances including resistances to permeability, chloride ion penetration, sulfate erosion and alkali aggregate reaction [39]. 4. Conclusions The major mineral components of the calcium silicate slag studied in this paper are Ca2 SiO4 , calcite, aragonite, gibbsite and Ca3 Al2 O6 . Besides, amorphous aluminosilicate phase is thought to be occurring in the calcium silicate slag from the TG-DSC analysis. An optimal design for the cementitious material composed of calcium silicate slag was determined by the specimen CFSC7 containing 30% calcium silicate slag, 5% high-alumina fly ash, 24% blast furnace slag, 35% clinker and 6% FGD gypsum. This blended system yields excellent physical and mechanical properties. With high 3day flexural strength of 7 MPa, the compressive strength of this developed cementitious material at 3 and 28 days can achieve to 22.9 and 37.5 MPa, which can meet the requirements of P.C 32.5R in Chinese Standard GB 175-2007. The produced cementitious material composed of calcium silicate slag can be utilized mainly as building materials and stabilization/solidification materials. The hydration products of the cementitious material composed of calcium silicate slag are mostly amorphous C–A–S–H gel, rod-like ettringite and hexagonal-sheet Ca(OH)2 with small amount of zeolite-like minerals such as CaAl2 Si2 O8 ·4H2 O and Na2 Al2 Si2 O8 ·H2 O. With increasing of the hydration time, Ca(OH)2 was depleted gradually by the pozzolanic reaction of calcium silicate slag, fly ash and blast furnace slag to form C–A–S–H gel. As the predominant hydration products, rod-like ettringite and

Table 6 Radioactivity test results. Sample

Specific activity of radionuclide (Bq/kg) 226

CFSC7

Ra

67.3

232

Th

54.2

Main material for building 40

K

233.2

Internal exposure index

External exposure index

0.3

0.4

76

N. Zhang et al. / Journal of Hazardous Materials 306 (2016) 67–76

amorphous C–A–S–H gel play a positive role in promoting the densification of the paste structure, and also play an important role in the strength development of the cementitious material composed of calcium silicate slag in the early hydration process. The hydration reaction of this cementitious material involves two basic processes which are nucleation and crystal growth (NG) and diffusion (D). NG dominates at the early hydration stage within the hydration reaction degree (˛) of 0.06, and D controls the later hydration stage with ˛ larger than 0.205. While both NG and D processes dominate the hydration reaction with ˛ between 0.06 and 0.205. The present research evaluated the leaching toxicity and radioactivity of the cementitious material composed of calcium silicate slag. The results show that none of the tested elements in the produced cementitious material exceeds the limits specified in standards for drinking water quality (Chinese Standard GB 5749-2006), and both of the internal and external exposure indexes of the produced cementitious material are within the limits of radionuclides in building materials (Chinese Standard GB 6566-2010), indicating that the developed cementitious material composed of calcium silicate slag is environmentally acceptable. This study provides an effective way for the use of calcium silicate slag to produce cementitious materials, which points out a promising direction for the proper utilization of calcium silicate slag in large quantities. Possessing important environmental significances, the developed cementitious material composed of calcium silicate slag can not only alleviate environment pollution, but also promote the sustainable development of extracting Al2 O3 from high-alumina fly ash industry.

Acknowledgements This work was financially supported by the National Natural Science Foundation of China (No. 51302012 and No. 51234008) and Fundamental Research Funds for the Central Universities (FRF-TP14-111A2).. The authors would like to acknowledge Professor H. Sun (University of the Pacific, Stockton, USA) for his helpful suggestions. Furthermore, the authors gratefully acknowledge the Inner Mongolia Datang International Renewable Resources Development Co. Ltd., for providing the raw materials and related support, and also acknowledge Green Construction Materials and Circulation Economy Center of Architectural Design and Research Institute of Tsinghua University Co. Ltd., for supplying facilities to carry out this research work. The Metalllurgical Experimental Center at University of Science & Technology Beijing and China Building Material Test Center are also acknowledged for their assistance to perform microanalysis and radioactivity tests, respectively.

References [1] Z. Zhang, J. Sun, Y. He, G. Liu, Y. Wang, H. Cao, Distribution of some major and trace elements in high aluminum fly ash, Geochimica 35 (2006) 660–666 (in Chinese). [2] S. Dai, L. Zhao, S. Peng, C. Chou, X. Wang, Y. Zhang, D. Li, Y. Sun, Abundances and distribution of minerals and elements in high-alumina coal fly ash from the Jungar Power Plant, Inner Mongolia, China, Int. J. Coal Geol. 81 (2010) 320–332. [3] C. Feng, Y. Yao, Y. Li, X. Liu, H. Sun, Thermal activation on calcium silicate slag from high-alumina fly ash: a technical report, Clean Technol. Environ. Policy 16 (2014) 667–672. [4] J. Sun, B. Wang, Z. Zhang, Resource utilization of high aluminum fly ash and circular economy, Light Metals 10 (2012) 1–5 (in Chinese). [5] Z. Yang, J. Sun, Z. Zhang, J. Ye, R. Miao, Using silicate-calcium slag generated in process of extracting alumina from fly ash as cement admixture, Chin. J. Environ. Eng. 8 (2014) 3989–3995 (in Chinese). [6] Y. Zhu, Investigation of calcium silicate slag used as cement raw material, Miner. Metall. Eng. 3 (1983) 44–47 (in Chinese). [7] Y. Xu, A study of silicium-calcium slag as raw material for cement, J. Huazhong Univ. Sci. Technol. 20 (1992) 147–152 (in Chinese).

[8] P. Wen, Feasibility researching for silicate-calcium slag make as cement raw material, World Build. Mater. 33 (2012) 5–7 (in Chinese). [9] Z. Yang, J. Sun, J. Ye, Z. Zhang, R. Miao, The effect of silicate-calcium slag on cement clinker burn-ability, Mater. Sci. Technol. 22 (2014) 41–47 (in Chinese). [10] GB 175-2007, Common Portland Cement, Standards Press of China, Beijing, 2008 (in Chinese). [11] J. Liu, J. Zhang, J. Sun, H. Wang, J. Ye, D. Shi, Dealkalizaiton of calcium silicate slag and study of using it as cement admixture, New Build. Mater. 12 (2012) 37–39 (in Chinese). [12] N. Zhang, H. Sun, X. Liu, J. Zhang, Early-age characteristics of red mud-coal gangue cementitious material, J. Hazard. Mater. 167 (2009) 927–932. [13] X. Liu, N. Zhang, Y. Yao, H. Sun, H. Feng, Micro-structural characterization of the hydration products of bauxite-calcination-method red mud-coal gangue based cementitious materials, J. Hazard. Mater. 262 (2013) 428–438. [14] N. Zhang, X. Liu, H. Sun, Hydration characteristics of intermediate-calcium based cementitious materials from red mud and coal gangue, Chin. J. Mater. Res. 28 (2014) 325–332 (in Chinese). [15] X. Feng, Research on the solidification mechanism for Na+ during the hydration process of red mud-cementitious materials, in: Dissertation of Doctoral Degree, Tsinghua University, Beijing, 2007. [16] N. Zhang, X. Liu, H. Sun, XPS analysis on hydration process of red mud-coal gangue based cementitious materials, Metal Mine 3 (2014) 171–176 (in Chinese). [17] X. N. Zhang, H. Liu, L. Li Sun, Evaluation of blends bauxite-calcination-method red mud with other industrial wastes as a cementitious material: properties and hydration characteristics, J. Hazard. Mater. 185 (2011) 329–335. [18] GB/T 17671-1999, Method of Testing Cements—Determination of Strength, Standards Press of China, Beijing, 1999 (in Chinese). [19] GB/T 2419-2005, Test Method for Fluidity of Cement Mortar, Standards Press of China, Beijing, 2005 (in Chinese). [20] GB/T 1346-2011, Test Methods for Water Requirement of Normal Consistency, Setting Time and Soundness of the Portland Cement, Standards Press of China, Beijing, 2012 (in Chinese). [21] HJ 557-2010, Solid Waste-extraction Procedure for Leaching Toxicity—Horizontal Vibration Method, China Environmental Science Press, Beijing, 2010 (in Chinese). [22] Y. Zhang, Z. Pan, Characterization of red mud thermally treated at different temperatures, J. Jinan Univ. Sci. Technol. 19 (2005) 293–297 (in Chinese). [23] X. Liu, N. Zhang, H. Sun, J. Zhang, L. Li, Structural investigation relating to the cementitious activity of bauxite residue—red mud, Cem. Concr. Res. 41 (2011) 847–853. [24] N. Zhang, X. Liu, H. Sun, L. Li, Pozzolanic behaviour of compound-activated red mud-coal gangue mixture, Cem. Concr. Res. 41 (2011) 270–278. [25] I. Lecomte, C. Henrist, M. Liegeois, F. Maseri, A. Rulmont, R. Cloots, (Micro)-structure comparision between geopolymers, alkali-activated slag cement and Portland cement, J. Eur. Ceram. Soc. 26 (2006) 3789–3797. [26] A. Palomo, A. Fernandez-Jimenez, G. Kovalchuk, L.M. Ordonez, M.C. Naranjo, Opc-fly ash cementitious systems: study of gel binders produced during alkaline hydration, J. Mater. Sci. 42 (2007) 2958–2966. [27] N.J. Clayden, S. Esposito, A. Aronne, P. Pernice, Solid state 27 Al NMR and FTIR study of lanthanum aluminosilicate glasses, J. Non-Cryst. Solids 258 (1999) 11–19. [28] A. Fernandez-Jimenez, F. Puertas, I. Sobrados, J. Sanz, Structure of calcium silicate hydrates formed in alkaline-activated slag: influence of the type of alkaline activator, J. Am. Ceram. Soc. 86 (2003) 1389–1394. [29] A. Fernandez-Jimenez, A. Palomo, T. Vazquez, R. Vallepu, T. Terai, K. Ikeda, Alkaline activation of blends of metakaolin and calcium aluminate, J. Am. Ceram. Soc. 91 (2008) 1231–1236. ´ Effect of w/c and temperature on the [30] V. Tydlitát, T. Matas, R. Cerny, early-stage hydration heat development in Portland-limestone cement, Constr. Build. Mater. 50 (2014) 140–147. [31] F. Han, Z. Zhang, D. Wang, P. Yan, Hydration heat evolution and kinetics of blended cement containing steel slag at different temperatures, Thermochim. Acta 605 (2015) 43–51. ´ P. Dabic, ´ A conceptual model of the cement hydration process, [32] R. Krstulovic, Cem. Concr. Res. 30 (2000) 693–698. [33] P. Yan, F. Zheng, Kinetics model for the hydration mechanism of cementitious materials, J. Chin. Ceram. Soc. 34 (2006) 555–559 (in Chinese). [34] F. Han, D. Wang, P. Yan, Hydration kinetics of composite binder containing different content of slag of fly ash, J. Chin. Ceram. Soc. 42 (2014) 613–620 (in Chinese). [35] S. Turkel, Long-term compressive strength and some other properties of controlled low strength materials made with pozzolanic cement and Class C fly ash, J. Hazard. Mater. 137 (2006) 261–266. [36] Y. Yao, H. Sun, A novel silica alumina-based backfill material composed of coal refuse and fly ash, J. Hazard. Mater. 213–214 (2012) 71–82. [37] GB 5749-2006, Standards for Drinking Water Quality, Standards Press of China, Beijing, 2007, 2015 (in Chinese). [38] GB 6566-2010, Limits of Radionuclides in Building Materials, Standards Press of China, Beijing, 2011 (in Chinese). [39] Y. Zhao, X. Liu, H. Li, N. Zhang, H. Zhang, Study on the performance of eco-concrete with calcium-silicate slag from alumina metallurgy, Metal Mine 11 (2014) 175–180 (in Chinese).