Paste-backfill-of-high-sulphide-mill-tailings-using-alkali-activated-blast-furnace-slag-Effect-of-a

Paste back?ll of high-sulphide mill tailings using alkali-activated blast furnace slag:Effect of activator nature,concentration and slag

properties

Ferdi Cihangir a ,?,Bayram Ercikdi a ,Ayhan Kesimal a ,Haci Deveci a ,Fatih Erdemir b

a Department of Mining Eng.,Karadeniz Technical University,61080Trabzon,Turkey

b

Department of Metallurgical and Material Eng.,Karadeniz Technical University,61080Trabzon,Turkey

a r t i c l e i n f o Article history:

Received 30April 2015Revised 28July 2015

Accepted 31August 2015

Available online 5September 2015Keywords:

A.Alkali-activated slag

B.Activator nature

C.Activator concentration

D.Slag composition

E.Paste back?ll

F.Pozzolanic activity

a b s t r a c t

The effect of activator type,concentration and slag composition on the strength and stability properties of paste back?ll (CPB)of high-sulphide tailings using alkali-activated slag (AAS)as binder (7wt.%)were investigated in this study.Acidic and neutral (AS–NS)slags were activated with liquid sodium silicate (LSS)and sodium hydroxide (SH)at 6–10wt.%concentrations.Ordinary Portland cement (OPC)results were used for comparison.The strength development was found to remarkably improve with increasing the concentration from 6to 8wt.%.Further increase in concentration did not enhance the strength.SH was determined to produce higher early-age strength whilst LSS produced higher long-term strengths as an indication of slag selectivity for activators.More extensive gypsum formation was observed at lower con-centrations in SEM/EDS studies.An increase in Na 2O concentration raised the activator consumption.High concentrations also led to poorly crystallized C–S–H gel,loose structure and drying shrinkage cracks especially in NS–SH samples.A reduction in total porosity up to 20%was obtained in AAS samples compared to OPC.Amorphous structure,chemical modulus ratio and/or basicity index (BI)values were seen to control the pozzolanic reactivity,and therefore,the alkali-activation and hardening process.

ó2015Elsevier Ltd.All rights reserved.

1.Introduction

Paste back?ll is an engineered mixture of mineral processing tailings,binder and water,and used for back?lling of mined out underground voids.Binder is the most critical component of CPB.Earlier studies have pointed out that ordinary Portland cement (OPC),which is widely used as binder in CPB,is not particularly suitable for paste back?ll (PB)of sulphide-rich tailings due to its vulnerability to acid and sulphate attack (Benzaazoua et al.,2002;Fall and Benzaazoua,2005;Fall et al.,2005;Wu and Fall,2012;Cihangir et al.,2013;Ercikdi et al.,2010a,b;Malviya and Chaudhary,2006;Tariq and Nehdi,2007).

Recently,alkali-activated cements (AAC)produced from various pozzolanic materials (industrial by-products such as slag and ?y ash)at low cost have been reported to have superior strength and stability performance in concrete applications even under aggressive environments (Bakharev,2005;Bakharev et al.,2003,2002;Chi,2012;Cincotto et al.,2003;Pacheco-Torgal et al.,2012;Shi et al.,2006).Development of low cost and effective AAC is contingent upon characteristics of materials such as activator type and dosage (concentration),modulus ratio,pozzolan composition,initial pH of activator solution and curing conditions,which have profound effect on hydration properties of alkali-activated binders (Chi,2012;Law et al.,2012).In this regard,most studies have focused particularly on activator characteristics rather than pozzolan composition (Ben Haha et al.,2012).Hydration degree or the dissolution of pozzolans is accelerated when the acti-vator concentration is increased (Acevedo-Martinez et al.,2012;Bondar et al.,2011;Xu and Van Deventer,2000;Zhang et al.,2011).However,the effect of activator concentration is stated to depend on activator type and pozzolan characteristics (Atis et al.,2009;Cincotto et al.,2003;Krizan and Zivanovic,2002;Ravikumar et al.,2010).Low concentrations of alkali activators can lead to a delay in the activation process whilst ef?orescence and brittleness problems as a function of slag composition,activa-tor type and curing temperature occur at excessively high concen-trations (Fernández-Jiménez et al.,1999)which also increase shrinkage (Cincotto et al.,2003).An excess of alkaline concentra-tion in AAC can also cause a decrease in strength (Palomo et al.,1999).Therefore,in addition to alkali activator type and concentra-tion,slag characteristics were shown to affect the properties of alkali-activated slag cement and concrete produced (Shi et al.,2011).

To the authors’knowledge,there are few and very limited detailed studies on the use of alkali-activated binders for safe

https://www.sodocs.net/doc/269657752.html,/10.1016/j.mineng.2015.08.0220892-6875/ó2015Elsevier Ltd.All rights reserved.

?Corresponding author.

E-mail address:cihangir@https://www.sodocs.net/doc/269657752.html,.tr (F.Cihangir).

disposal of mine/mill tailings containing hazardous components. Ahmari et al.(2012)used silicious copper mine tailings(64.8% SiO2)as aggregate and binder phase to study the effect of activator type,concentration,modulus ratio and curing temperature on strength development.On the other hand,although the use of alkali-activated slag as some of the support systems in mining applications in South Africa and Canada was stated(Khater, 2014),no explicit information about the design of materials or the conditions of usage were encountered in open literature of

study area.Therefore,there is still lack of knowledge in design of paste back?ll materials containing alkali-activated slag which is required for?lling of underground mined voids.

Cihangir et al.(2012)studied the effect of activator type and binder content on the mechanical and stability properties of sul-phide rich tailings paste back?ll.They showed that AAS developed consistently higher UCSs(uncon?ned compressive strength values) with remarkably improved durability in the long-term than OPC. The present study is conducted as an extension of above study published by Cihangir et al.(2012)where the effects of type and concentration of alkaline activator and composition of blast fur-nace slag on the performance of AAS as binder in the paste back?ll of high sulphide tailings were investigated.In this study,acidic (AS)and neutral slags(NS)activated with aqueous sodium silicate (LSS)and sodium hydroxide(SH)at different activator concentra-tions(6–10wt.%)were tested.The strength and stability properties of CPB samples prepared with alkali-activated slags(AAS)at7wt.% binder content were examined and compared to OPC samples in the short-and long-term(14–360days).The rationale behind strength and stability problems associated with sulphide-rich tail-ings was studied by the detailed chemical and microstructural characterization of CPB samples.

2.Materials and methods

2.1.Tailings and binders

The tailings material used in this study was obtained from a copper–zinc ore processing plant located in the northeast of Turkey.The?nes(à20l m)and pyrite contents of the tailings were determined to be56%and76%,respectively.

Ordinary Portland cement(OPC)was used for the comparison of AASs(alkali activated slags)on the mechanical performance of CPB.Aqueous sodium silicate and granular sodium hydroxide were used as activators.In the activation process,SH solution was prepared and mixed with LSS to set the Na2O concentration and modulus ratio(M S:mass ratio of SiO2to Na2O)of LSS to1.0,which was kept constant in this study.Further detailed physical,chemical and mineralogical properties of the tailings and binders used in this study can be found elsewhere Cihangir et al.(2012).

2.2.Reactivity tests of the slags

To assess the reactivity of the slags,the pozzolanic activity test in accordance with ASTM C989/989M-14(2014)and reactive SiO2 test according to TS EN197-1(2002)were carried out.The?neness of the slags was also determined based on TS EN196-6(2010).The basicity index(BI)and chemical modulus ratio([CaO+MgO]/SiO2) of the slags were calculated as a measure of slag reactivity accord-ing to the equations suggested by Baunéet al.(2000)and Wainwright and Rey(2000),respectively.Obtained results considering also the slag composition were given in Table1. X-ray diffraction(XRD)(Philips X’pert PW3040Diffractometer) analyses under2-h range of5–70°with a0.005°step size were performed to investigate the crystalline phases of the slag samples(Fig.1).2.3.Preparation and testing of CPB samples

The CPB samples were prepared by blending the tailings,binder (OPC or AAS)and tap water into a uniform mixture using a Univex SRMF-20Stand model blender equipped with a double spiral.OPC is extensively used in paste back?ll operations at the binder contents of5–9wt.%which depend on the location of back?ll. The primary stopes are?lled at7–9wt.%binder contents whilst secondary and tertiary stopes are?lled at5–7wt.%contents (Kesimal et al.,2012,2010).Therefore,an average binder content of7wt.%was selected for the current study and,the performance of AAS cements were compared with OPC samples.For AAS samples,SiO2and Na2O solid contents of activators were taken into account for the determination of binder content.For this study,the concentration of Na2O in the activator phase was varied in the range of6–10wt.%by weight of the slag.Earlier experimental data published by Cihangir et al.(2012)at8wt.%Na2O activator concentration and of OPC were included for comparison.The consistency of the CPB samples was set to7.5±0.2in.slump which is in line with practical paste back?ll applications in the mining industry.Having been thoroughly blended,the CPB mixtures were ladled into plastic cylinders of100mm in diameter by200mm in length with a perforated bottom for drainage of excess water.The cylinders were then sealed in plastic bags and placed into the curing room maintained at20±1°C and85±1%humidity.A total of234CPB samples(in triplicate for each recipe)were evaluated over a curing period of360days(Table2).

Uncon?ned compressive strength(UCS)tests using a computer-controlled mechanical press with a load capacity of50kN at a dis-placement rate of0.5mm per minute were performed on the cured CPB samples according to ASTM C39/C39M-15a(2015).The mean UCS value of triplicate samples was presented in the results.In the present study,the strength and stability performance of CPB samples were assessed according to the threshold UCS value of

1.0MPa in the short-(28days)and long-term(360days).

2.4.Determination of pH and SO42àin CPB samples

Acid and sulphate generated by the oxidation of pyrite minerals during the hydration and hardening process over360days were monitored to evaluate the performance of CPB samples against acid and sulphate attack.The detailed analytical procedures used for analysis of acid and sulphate can be found elsewhere Cihangir et al.(2012).

2.5.Microstructural studies on CPB samples

Mercury Intrusion Porosimetry(MIP)tests were carried out on the representative samples of fractured CPB specimens at56days to assess the effect of alkali nature,concentration and slag proper-ties on the total porosity of CPB samples.MIP tests were carried out according to ASTM D4404-10(2010)under pressure ranging from 0to414MPa.

Visual inspection over the curing periods and SEM-EDS analysis of CPB samples were carried out to evaluate and interpret their mineralogical and microstructural properties.SEM studies were Table1

Reactivity index properties of the slags.

Type of

slag

Pozzolanic

index(%)

Reactive

SiO2

Blainess;

(cm2/g)

Basicity

index

Chemical

modulus ratio (28days)

NS86.039.104600 1.03 1.11

AS74.339.3846500.92 1.04

118 F.Cihangir et al./Minerals Engineering83(2015)117–127

performed on representative samples of 360-day cured CPB speci-mens collected after UCS tests.The samples were dried in an oven at 50°C for about 60h to remove the pore water and then prepared for SEM studies.The microstructure of the samples was examined under a ZEISS-EVO MA-LS SERIES LEO 1550scanning electron microscope (SEM)operated at 10kV accelerating voltage.Energy dispersive spectrometry (EDS)was also used to identify the mineral phases.

3.Results and discussion

3.1.Evaluation of the reactivity of slags

Chemical and mineral composition,?neness,reactive SiO 2con-tent and pozzolanic activity are among some of the basic parame-ters for the evaluation of the reactivity of pozzolans (i.e.slag).Wainwright and Rey (2000)suggested that the slag reactivity increases with increasing the chemical modulus ratio whilst Shi et al.(2006)reported that slags with higher BI index produced higher strength values.The BI index values of the slags show parallelism with the chemical modulus ratios (Table 1).

Amorphousness of the particles and reactive SiO 2content are the other signi?cant factors determining the slag reactivity (Ganesh Babu and Sree Rama Kumar,2000;Walker and Pavía,2011).Reactive silica content was reported to control the pozzolanic activity and the compressive strength (Papadakis et al.,2002;Ercikdi et al.,2009a;Walker and Pavía,2011).On the other hand,Walker and Pavía (2011)suggested that amorphousness determines the reactivity and the strength development of a paste of any pozzolans to a greater extent than other pozzolan properties such as reactive SiO 2.

Fineness of any pozzolans increases as the particles get ?ner,which also accelerate the reactivity of these materials (Wainwright and Rey,2000).As can be seen from Table 1,there are little differences between ?neness or particle size distribution (Cihangir et al.,2012)and sieve results.Since,the reactive silica contents and the ?neness of both slags are approximately the same,the higher pozzolanic index value for NS can be ascribed to its quasi-amorphous structure (Fig.1b),higher chemical modu-lus ratio and BI index value compared to AS.

3.2.Effect of LSS concentration on the strength and stability properties of CPB

Fig.2demonstrates the short-and long-term strength and sta-bility performance of CPB samples prepared with OPC and sodium silicate activated acidic (AS–LSS)and neutral slags (NS–LSS)at 6,8and 10wt.%activator concentrations.OPC samples were observed to maintain the target strength value of 1.0MPa in the short-and long-term.However,the strength gain appeared to be arrested after 56days and a strength loss of 5.52%occurred in the long-term after 224days.The strength loss in OPC samples was reported to stem from acid and sulphate attack which was discussed in detail by Cihangir et al.(2012).LSS samples prepared at 6wt.%Na 2O concentration failed to achieve the threshold strength of 1.0MPa at 28days.However,these samples produced signi?cantly higher strengths than OPC samples after 56days.

The development of strength of NS–LSS samples substantially improved with increasing the Na 2O concentration from 6to 8wt.%at which the highest rate of strength gain was achieved (Fig.2).Although a similar trend for strength gain was observed over an initial period at 8and 10wt.%Na 2O,CPB samples produced discernibly lower strengths at 10wt.%Na 2O than 8wt.%.A loss

of

Fig.1.XRD pro?les of acidic (a)and neutral (b)slags.

Table 2

A summary of the experimental conditions for CP

B samples of OP

C and AAS.Binder Cont.(wt.%)Cement/slag type Activator type/Na 2O conc.(wt.%)Solids Cont.(wt.%)Water cont.(wt.%)Water/Binder ratio (w /c )Binder cont.per m 3(kg)LSS cont.per m 3kg SH Cont.per m 3kg Slag cont.per m 3kg Slump (in.)7OPC –76.9523.05 4.28128.03–

––

7.57NS LSS/677.8522.15 4.06126.5529.17 4.16107.367.37NS LSS/879.0220.98 3.79129.5938.43 5.48106.087.47NS LSS/1079.0620.94 3.78132.9547.62 6.78105.147.57NS SH/677.4822.52 4.15124.42–8.94115.487.67NS SH/877.6022.40 4.12124.81–11.68113.137.57NS SH/1077.7422.26 4.09125.05–

14.29110.767.57AS LSS/678.6021.40 3.89129.8431.26 4.45115.037.57AS LSS/879.0420.96 3.79131.7740.75 5.80112.477.37AS LSS/1079.1320.87 3.77132.8249.537.06109.367.47AS SH/677.5822.42 4.13125.05–8.99116.077.57AS SH/877.6222.38 4.12125.26–11.72113.547.67

AS

SH/10

77.78

22.22

4.08

125.95

–

14.39

111.56

7.5

strength by5.3%and6.6%after224days appeared to occur at6and 10wt.%Na2O,respectively,whilst no loss of strength was recorded at8wt.%Na2O.

Fig.2also shows the effect of slag type on the strength perfor-mance of CPB samples.In this regard,AS–LSS samples were observed to harden at a slower rate than NS–LSS samples produc-ing consistently lower UCSs at the corresponding Na2O levels over the curing periods up to360days.These?ndings can be attributed to the lower pozzolanic properties of AS compared to NS(Table1 and Fig.1).However,AS–LSS samples showed higher stability than NS–LSS samples as no strength loss was discerned for AS–LSS sam-ples at all the levels of Na2O tested.A similar trend for strength development was also noted for AS–LSS samples,which produced the highest strengths at a Na2O dosage of8wt.%.The360-day strength of the samples at10wt.%Na2O was even lower than that at6wt.%Na2O.It can be inferred from these?ndings that the optimum concentration of Na2O is8wt.%for the strength and stability of CPB samples irrespective of slag type.

3.3.Effect of SH concentration on the strength and stability properties of CPB

The strength and stability behaviour of SH samples were similar to that of LSS samples although the latter samples produced slightly higher strengths.This could be attributed to the improved binder structure owing to the polimerization of the silicate ions of sodium silicate activator(LSS).Cincotto et al.(2003)found that silica has an important effect on gel densi?cation.The silicate polimerization was also reported to bridge the slag particles and ?ll the voids between slag grains(Shi et al.,2006),which enhances the binder structure.These two cases explain higher strength development rate when slag is activated with LSS compared to other potential activators.Only the AS–SH samples at6wt.% Na2O were unable to reach the threshold value of strength at 28-days(Fig.3).The losses in the stability of SH samples particu-larly at10wt.%was also found to occur in the long-term(Fig.3). It was observed that SH samples at10wt.%Na2O were softer and more brittle(Fig.4b)than those samples at lower concentrations. It was found that AS–SH samples produced higher UCSs after 224-days than those of NS–SH(Fig.3)in contrast to LSS samples (Fig.2).This indicates the importance of the interactions/coher-ence between slag and activator for the binding performance of activated slag cements.

When slag particles are subjected to alkali-activation process, the covalent bonds of Si–O–Si,Al–O–Al,Ca–O etc.on the slag grains begin to breakdown under the polarization effect of OHàions. Thereafter,ions passed into the dispersion medium/matrix accu-mulate during the induction period at the second step.After the condensation and crystallization of the accumulated products at the third step,a reorganization of the structure forming the hydration products occurs.Consequently,mechanical strength development of the structure begins via hardening process.

It is well agreed that the nature and concentration of activators and the properties of slags determine the hydration characteristics and strength development of alkali-activated slag cements.The current study suggests that amorphous structure,chemical modu-lus ratio and BI values determine the pozzolanic reactivity and therefore,control the alkali-activation and hardening characteris-tics of the slags.Additionally,effect of activators in alkali-activation process may differ for slags from different origins (Shi et al.,2006).It means that the selectivity of activators depends on slag properties,which causes different hydration product formation and different strength development rates(Shi and Day, 1996)in the short-and long-term as in the current study.

Additionally,an increase in Na2O concentration leads to the increase in activator content and causes high pH levels.High amount of OHàions are introduced into the dispersion medium by this way.As a result,the alkalinity of the mixture increases which also increases the dissolution level of alumina-silicates (i.e.slag)(Gasteiger et al.,1992)and,higher amount of hydration products is generated.In this case,the increase of activator concen-tration decreases the induction period.However,the quick precip-itation of hydration products onto the slag particles hinders the further reaction where the excess alkali(OHà)does not react with slag particles(Komnitsas et al.,2009).Therefore,increasing the activator content produces mechanical strength to an extend beyond which there is no further substantial strength acquisition (Allahverdi et al.,2010;Bondar et al.,2011;Krizan and Zivanovic, 2002;Law et al.,2012;Panagiotopoulou et al.,2007; Ruiz-Santaquiteria et al.,2012;Zhang et al.,2011).However,an excess of OHàconcentration in AAC system was reported to lower the strength of the system(Palomo et al.,1999)and the durability (Yang et al.,2012).Komnitsas and Zaharaki(2007)reported that especially the alkali cations affect and control the gel hardening and structure formation of the geopolymers.High amount of alkali metals in the case of high Na2O concentration was also reported to be undesirable since such cations such as Na+lead to subsequent stresses during the geopolimerization(Xu and Van Deventer, 2000).Besides,an increase in free alkali content in the case of high activator content was stated to cause brittleness having a

120 F.Cihangir et al./Minerals Engineering83(2015)117–127

detrimental effect on the binding products based on the slag prop-erties and activator nature(Bondar et al.,2011).These are consis-tent with the current?nding(Figs.3and4)in that increasing the concentration of Na2O up to10%led to comparatively low long-term strengths with stability losses after224days.

On the other hand,low activator concentration results in low pH values which slows down the dissolution of slag particles and produces weakly bound polymers resulting in poor mechanical strength(Lloyd et al.,2010).Atis et al.(2009)reported that LSS activated slag mortar samples at4–6-and8wt.%Na2O gave higher UCSs compared with SH-activated slag and OPC samples up to 90days curing time.They also obtained the highest strength per-formance at8.0wt.%Na2O.

According to Ben Haha et al.(2011),as BI increased for slags (they did not evaluate BI index in their study)whose BI indexes are in the range of1.05–1.17,the strength gain was observed to improve for SS(sodium silicate)and SH where the early strength is higher in the latter.Ben Haha et al.(2012)also examined the in?uence of the characteristics of slags having BI indexes(they did not evaluate this index in their study,either)in the range of 1.07–0.96on hydration properties of AAS.Strength gain of the samples was observed to slightly increase in the case of decrease in BI index when SH was used.However,the rate of strength gain was seen to show a decreasing trend as BI index decreased for SS. These observations depicts the importance of the selectivity of slags.The rate of hardening of the samples were also seen to slow down as BI decreased according to above two studies.Shi et al. (2006)reported that slags with higher BI index produced higher strength values.These are in good agreement with the current ?ndings(Figs.2and3)for the effect of activator type and slag properties in terms of slag selectivity for the short-and long-term strengths of CPB samples.

Low early UCSs of paste back?ll samples in the present study can be attributed to relatively slow development of activation pro-cess for both slags at6wt.%Na2O.On the other hand,high gain of early strength at10wt.%Na2O was presumed to be concomitant with the high amount of mineral dissolution and faster reaction at this level of Na2O(Al-Otaibi,2008;Cincotto et al.,2003; Krizˇan et al.,2005;Ravikumar et al.,2010;Xu and Van Deventer, 2000;Zˇivica,2007;Kashani et al.,2014).

Pacheco-Torgal et al.(2012)stated that alkali-activated binders show high chemical resistance owing to the low content of soluble calcium compounds.In the present study,the stability characteris-tics of CPB samples were observed to be different for each activator type,concentration and slag.Stability problems observed for CPB samples of10wt.%Na2O in the long-term could be ascribed to the interaction between activator type and slag composition,and, in particular,excessive concentration of alkali hydroxide(Shi and Day,1996;Shi et al.,2011,2006;Yang et al.,2012).Therefore, the?ndings of the current study indicated that optimum activator concentration signi?cantly depends on the slag characteristics and the alkali activators used.

A gradual increase(from6wt.%to10wt.%)in Na2O concentra-tion led to increase the total binder content by1.0–2.5%in LSS mix-tures.This can be ascribed to the repulsion effect of silicates between?ne clay minerals(SiO2+Al2O3)which also prevents the agglomeration of the slimes(Ayadi et al.,2013;Ersoy et al., 2014;Kashani et al.,2014)and thus,reduce the w/c ratio in CPB

Fig.4.Images of NS–SH/8sample(a),and NS–SH/10sample(b)after UCS tests at360-days.

F.Cihangir et al./Minerals Engineering83(2015)117–127121

mixtures.On the other hand,the gradual increase of Na2O concentration reduced the amount of slag by1.0–3.0%per cubic meter in all CPB mixtures and increased the amount of activator by22.0–32.0%per cubic meter of CPB mixtures(Table2).There-fore,the greatest effect was seen to occur on activator consump-tion via an increase in Na2O concentration which also directly increases the binder costs(Law et al.,2012)for a given under-ground void(stope to be?lled)size.

3.4.Evaluation of acid and sulphate based on activator concentration for CPB

Fig.5demonstrates the evolution of pH in the samples con-taining LSS activated slags over the curing period.Increasing the activator concentration increased pH for both slags.Up to 14-days,an increase in the pH of all CPB samples was recorded due to the alkaline hydration products.For OPC samples, portlandite(Ca(OH)2–CH)is thought to be the reason for the highest pH at the beginning of hydration(Lloyd et al.,2010). After14-days,a trend of decrease in pH was observed for all CPB samples.pH of OPC samples was the highest up to 112-days;thereafter a severe decrease was noted to occur producing the lowest pH(i.e.pH10.28)at360days.

Higher pH levels were recorded for SH samples(data not shown)than LSS samples.NS–LSS samples gave slightly higher pH values for all samples than those of AS–LSS,which is consistent with the alkaline characteristics of these slag samples used. The?ndings also suggest that AAS samples have higher overall buffering capacity than OPC.

Fig.6demonstrates the change in sulphate(SO42à)content of CPB samples of LSS and OPC over the curing period.OPC samples consumed free SO42à(water-extractable sulphate)at the early cur-ing ages as the concentration of SO42àincreased after56-days.This could be attributed to the consumption of SO42àby hydration prod-ucts i.e.ettringite and gypsum.In effect,the OPC samples had higher acidity than AAS samples suggesting more extensive forma-tion of sulphate.Contrary to OPC samples,AS and NS samples had consistently higher water extractable SO42àlevels over the curing period.At the corresponding Na2O dosages,NS samples appeared to release somewhat higher SO42àthan AS samples(Fig.6).It is also pertinent to note that comparatively low levels of SO42à(6000–7500ppm)were recorded when SH was used as the activator(data not shown).These?ndings suggest no apparent correlation between water extractable SO42àand mechanical performance.This could also be due to the formation of some sulphate-consuming hydration products in case of SH(Shi et al.,2006)and different acid neutralizing capacities of AAS samples(Cihangir,2011;Shi and Day,1996).

3.5.Microstructural and mineralogical properties of CPB samples

The total porosities of CPB samples were found to be46.12%,in the range of36.75–38.13%and39.53–41.13%for OPC,LSS and SH samples,respectively.LSS and SH activated slags reduced the total porosity by17–20%and11–14%,respectively,compared to those of OPC.Pore structure re?nement in samples reducing the total porosity was observed with increasing the concentration of activa-tor.Lower porosity results were also obtained in AS samples (Fig.7).

The average moisture content of CPB samples was monitored to be17–21%.As LSS and higher activator concentrations were seen to reduce the moisture content up to7%,AS samples were observed to retain more moisture than those of NS(Fig.7).

Based on the tailings characteristics(i.e.?neness content),bin-der type,curing conditions,etc.,total porosity and water content were reported to be30–50%and15–25%for CPB,respectively (Ercikdi et al.,2013;Ouellet et al.,2007;Yilmaz et al.,2011).The use of high sulphide tailings in CPB leads to production of acid and sulphate(Figs.5and6)due to the porous structure and high water retaining capacity of CPB(Benzaazoua et al.,2004).The loss of strength encountered in the long-term(Figs.2and3)due acid and sulphate attack for the present study is in good agreement with those reported in literature(Benzaazoua et al.,2002; Cihangir et al.,2012;Ercikdi et al.,2010b,2009a,b).

Fig.8depicts the NS–LSS/8and OPC samples showing the pro-gress of oxidation in the latter.In both samples,some macroscopic secondary gypsum mineral formations were observed.The OPC samples were seen to have loose microstructure with the increased susceptibility of its pyrite content to oxidation.On the other hand, compared with the OPC samples,the AAS samples were observed to have denser structure,mitigating the in?ltration of air and hence the oxidation of pyrite.Fast precipitation of AAS hydration products rich in Si gel on/around the surface of sulphide minerals setting a physical barrier and dense structure with reduced porosi-ties were(Fig.7)assumed to inhibit the progress of oxygen (Fig.8a.)into the CPB matrix.

Fig.9illustrates the SEM images and EDS spectrums of AS–LSS samples at6–8and10wt.%Na2O concentrations at360-days. Secondary gypsum was rarely detected in the samples prepared at10wt.%Na2O compared with those at6-and8wt.%Na2O.In

122 F.Cihangir et al./Minerals Engineering83(2015)117–127

other words,the formation of gypsum tended to increase with decreasing the concentration of Na2O(Fig.9a and b).Low activator concentration has been reported to slow down the alkali-activation and hydration process of pozzolanic materials(Fernández-Jiménez et al.,1999),which is in good agreement with strength develop-ment in AAS samples(Figs.2and3).Low concentration level was also reported to cause more Ca and less Si release into the pore solution during hydration process,resulting in higher Ca/Si ratio in C–S–H gel structure(Song et al.,2000)which accounts for more gypsum formation in case of high sulphate medium.

In OPC samples,secondary gypsum was observed at112-and 360-days curing times(Fig.10a).Therefore,the strength loss in OPC samples was ascribed to the acid and sulphate(Fig.10a) attack,which was discussed in detail by Cihangir et al.(2012).

Fig.7.Total porosities and moisture contents of CPB samples at360-days.

Secondary gypsum minerals

(a)(b)

Fig.8.Visual appearance of a NS-LSS/6sample(a)and an OPC sample at360-days(b).

F.Cihangir et al./Minerals Engineering83(2015)117–127123

Compared with OPC samples,more abundant occurrence of mostly crystalline and denser C–S–H gel phase was observed in AAS sam-ples (Figs.9and 10).It was also noted that C–S–H structure had higher crystallinity in LSS samples than SH samples.In addition,C–S–H gel structure was loose and poorly crystallized in NS–SH/10samples.This could be attributed to the inhibitory effect of excess OH àions on polymerization process (Ruiz-Santaquiteria et al.,2012).Poorly polymerized C–S–H product,engendered by high activator concentration was reported to lead to loose struc-ture and the brittleness of the samples (Ahmari et al.,2012;Shi et al.,2006;Song et al.,2000).Furthermore,Na atoms bonded into C–S–H gel structure was also found to cause sample brittleness (-Fernández-Jiménez et al.,1999;Hong and Glasser,1999;Shi et al.,2006).Extensive drying shrinkage cracks especially in NS–SH/10sam-ples (Fig.11)were observed during the SEM studies.These cracks can be attributed to the development of pore structure and decreasing the moisture content in the long-term at high activator concentrations (Fig.7).Shrinkage cracks were reported to be asso-ciated with the loss of water from microstructure of the samples leading to pore contraction (Chang et al.,2005).Cincotto et al.(2003)stated that drying shrinkage was caused by the loss of water from pores.Atis et al.(2009)reported that increasing the Na 2O concentration increases the drying shrinkage and suggested that sample brittleness increases as the drying shrinkage gets higher.These are in line with the lowest water contents (wt.%)observed in CPB samples at 10wt.%Na 2O concentrations with lower porosi-ties for the current study.Therefore,the strength losses in the

(f)

(e)(d)

(b)

(c)

Gypsum Pyrite

C–S–H

(a)

Gypsum

9.Microstructural SEM images and EDS pro?les of AS-LSS samples at 6-(a,b),8-(c,d)and 10(e,f)wt.%Na 2O concentrations at 360-days.

124 F.Cihangir et al./Minerals Engineering 83(2015)117–127

long-term can be ascribed to these cracks which are concomitant with the brittleness and ready mechanical disintegration of360-day NS–SH/10samples under UCS test(Fig.4b).

4.Conclusions

AASs in different Na2O concentrations produced1.5–3.5-fold strengths and stability performance than those of OPC.An increase in alkaline concentration from6to8wt.%Na2O exerted great in?u-ence on the short-and long-term strength development in CPB samples.However,further increase in Na2O dosage to10wt.%pro-duced no further substantial strength gain and SH samples were adversely affected with strength losses after224days.Optimum concentration of Na2O was found to be8wt.%

and stability of CPB samples irrespective of slag

were obtained with SH in the short-term and with

term.Better interactions were obtained between

‘‘AS–SH”couples as an indication of slag selectivity

of alkaline activators.Pozzolanic activity and

activation and hardening process were seen to

amorphous structure,chemical modulus ratio and

values of the slags.

Increasing the activator concentration increased

capacity of CPB samples with concomitantly

decrease in pHs due to the oxidation of pyrite minerals

for AASs over curing whilst it was severe for OPC

112days.Notwithstanding this,water-extractable

tent of OPC samples was substantially lower

was consistent with the abundance of gypsum in

Lower porosities up to20%were obtained based

of AAS compared to OPC.SEM observations revealed

mation of gypsum tended to decrease with increasing

dosage.A loose and poorly crystallized C–S–H gel structure and drying shrinkage cracks responsible from strength losses were observed in NS–SH/10samples.These?ndings highlight the prime importance of slag characteristics,type and concentration of acti-vator for the effectiveness of AAS binders in CPB of sulphide tailings.

Acknowledgements

The authors would like to express their sincere thanks and appreciation to the Research Foundation of Karadeniz Technical University(Project no.2010.112.008.1),to Cayeli Bakir Isletmeleri A.S.,for the material and?nancial support,to Karcimsa A.S.,to Unye Cement A.S.and?nally to Ege Chemicals Ltd.

(b)

Gypsum

C–S–H(c)

Gypsum Feldspar (d)

Gypsum

Gypsum(a)

Fig.10.Microstructural SEM images of OPC(a)and NS–SH samples at6-(b),8-(c)and10(d)wt.%Na2O concentrations at360-days.

Fig.11.Drying shrinkage cracks in NS–SH/10samples.

F.Cihangir et al./Minerals Engineering83(2015)117–127125

References

Acevedo-Martinez, E.,Gomez-Zamorano,L.Y.,Escalante-Garcia,J.I.,2012.

Portland cement-blast furnace slag mortars activated using waterglass:–Part 1:Effect of slag replacement and alkali concentration.Constr.Build.Mater.37, 462–469.

Ahmari,S.,Zhang,L.,Zhang,J.,2012.Effects of activator type/concentration and curing temperature on alkali-activated binder based on copper mine tailings.J.Mater.Sci.47,5933–5945.https://www.sodocs.net/doc/269657752.html,/10.1007/s10853-012-6497-9.

Allahverdi,A.,Shaverdi,B.,Kani,E.N.,2010.In?uence of sodium oxide on properties of fresh and hardened paste of alkali-activated blast-furnace slag.Int.J.Civ.Eng.

8,304–314.

Al-Otaibi,S.,2008.Durability of concrete incorporating GGBS activated by water-glass.Constr.Build.Mater.22,2059–2067.https://www.sodocs.net/doc/269657752.html,/10.1016/ j.conbuildmat.2007.07.023.

ASTM C39/C39M-15a,2015.Standard Test Method for Compressive Strength of Cylindrical Concrete Specimens.ASTM International.

ASTM C989/989M-14.,2014.Standard Speci?cation for Slag Cement for Use in Concrete and Mortars.

ASTM D4404-10.,2010.Standard Test Method for Determination of Pore Volume and Pore Volume Distribution of Soil and Rock by Mercury Intrusion Porosimetry.

Atis,C.D.,Bilim,C.,Celik,O.,Karahan,O.,2009.In?uence of activator on the strength and drying shrinkage of alkali-activated slag mortar.Constr.Build.Mater.23, 548–555.

Ayadi,A.J.,Soro,J.,Kamoun,A.,Baklouti,S.,2013.Study of clay’s mineralogy effect on rheological behavior of ceramic suspensions using an experimental design.

Int.J.Res.Rev.Appl.Sci.14,374–384.

Bakharev,T.,2005.Resistance of geopolymer materials to acid attack.Cem.Concr.

Res.35,658–670.https://www.sodocs.net/doc/269657752.html,/10.1016/j.cemconres.2004.06.005. Bakharev,T.,Sanjayan,J.G.,Cheng,Y.B.,2003.Resistance of alkali-activated slag concrete to acid attack.Cem.Concr.Res.33,1607–1611.https://www.sodocs.net/doc/269657752.html,/

10.1016/S0008-8846(03)00125-X.

Bakharev,T.,Sanjayan,J.G.,Cheng,Y.B.,2002.Sulfate attack on alkali-activated slag concrete.Cem.Concr.Res.32,211–216.https://www.sodocs.net/doc/269657752.html,/10.1016/S0008-8846

(01)00659-7.

Bauné,E.,Bonnet,C.,Liu,S.,2000.Reconsidering the basicity of a FCAW consumable —Part1:Solidi?ed slag composition of a FCAW consumable as a basicity indicator.Weld.Res.Suppl.,57–65

Ben Haha,M.,Lothenbach,B.,Le Saout,G.,Winnefeld,F.,2012.In?uence of slag chemistry on the hydration of alkali-activated blast-furnace slag—Part II:Effect of Al2O3.Cem.Concr.Res.42,74–83.https://www.sodocs.net/doc/269657752.html,/10.1016/j.cemconres.

2011.08.005.

Ben Haha,M.,Lothenbach,B.,Le Saout,G.,Winnefeld,F.,2011.In?uence of slag chemistry on the hydration of alkali-activated blast-furnace slag—Part I:Effect of MgO.Cem.Concr.Res.41,955–963.https://www.sodocs.net/doc/269657752.html,/10.1016/j.cemconres.

2011.05.002.

Benzaazoua,M.,Belem,T.,Bussiere,B.,2002.Chemical factors that in?uence the performance of mine sulphidic paste back?ll.Cem.Concr.Res.32,1133–1144. Benzaazoua,M.,Marion,P.,Picquet,I.,Bussière,B.,2004.The use of paste?ll as a solidi?cation and stabilization process for the control of acid mine drainage.

Miner.Eng.17,233–243.https://www.sodocs.net/doc/269657752.html,/10.1016/j.mineng.2003.10.027. Bondar,D.,Lynsdale,C.J.,Milestone,N.B.,Hassani,N.,Ramezanianpour,A.A.,2011.

Effect of type,form,and dosage of activators on strength of alkali-activated natural https://www.sodocs.net/doc/269657752.html,pos.33,251–260.https://www.sodocs.net/doc/269657752.html,/

10.1016/j.cemconcomp.2010.10.021.

Chang,J.J.,Yeih,W.,Hung,C.C.,2005.Effects of gypsum and phosphoric acid on the properties of sodium silicate-based alkali-activated slag pastes.Cem.Concr.

Compos.27,85–91.https://www.sodocs.net/doc/269657752.html,/10.1016/j.cemconcomp.2003.12.001. Chi,M.,2012.Effects of dosage of alkali-activated solution and curing conditions on the properties and durability of alkali-activated slag concrete.Constr.Build.

Mater.35,240–245.https://www.sodocs.net/doc/269657752.html,/10.1016/j.conbuildmat.2012.04.005. Cihangir,F.,2011.Investigation of utilization of alkali-activated blast furnace slag as binder in paste back?ll(Ph.D.thesis).Karadeniz Technical University,Turkey, 187p.

Cihangir,F.,Ercikdi,B.,Kesimal,A.,Deveci,H.,Akyol,Y.,Ocak,S.,Kurtulus?,M.,2013.

The effect of geopolymers on the strength and workability characteristics of paste back?ll.In:Proceedings of the23rd Int.Mining Congress&Exhibition.

Antalya,Turkey,pp.475–487.

Cihangir,F.,Ercikdi,B.,Kesimal,A.,Turan,A.,Deveci,H.,2012.Utilisation of alkali-activated blast furnace slag in paste back?ll of high-sulphide mill tailings:effect of binder type and dosage.Miner.Eng.30,33–43.

Cincotto,M.A.,Melo,A.A.,Repette,W.L.,2003.Effect of different activators type and dosages and relation to autogenous shrinkage of activated blast furnace slag cement.In:Proceedings of the11th International Congress on the Chemistry of Cement.Durban,South Africa,pp.1878–1887.

Ercikdi,B.,Baki,H.,Izki,M.,2013.Effect of desliming of sulphide-rich mill tailings on the long-term strength of cemented paste back?ll.J.Environ.Manage.115, 5–13.

Ercikdi,B.,Cihangir,F.,Kesimal,A.,Deveci,H.,Alp,I.,2010a.Utilization of water-reducing admixtures in cemented paste back?ll of sulphide-rich mill tailings.J.

Hazard.Mater.179,940–946.

Ercikdi,B.,Cihangir,F.,Kesimal,A.,Deveci,H.,Alp,I.,2010b.Effect of natural pozzolans as mineral admixture on the performance of cemented-paste back?ll of sulphide-rich tailings.Waste Manage.Res.28,430–435.Ercikdi,B.,Cihangir,F.,Kesimal,A.,Deveci,H.,Alp,I.,2009a.Utilization of industrial waste products as pozzolanic material in cemented paste back?ll of high sulphide mill tailings.J.Hazard.Mater.168,848–856.

Ercikdi,B.,Kesimal,A.,Cihangir,F.,Deveci,H.,Alp,I.,2009b.Cemented paste back?ll of sulphide-rich tailings:importance of binder type and dosage.Cem.Concr.

Compos.31,268–274.

Ersoy,B.,Evcin,A.,Uygunoglu,T.,Akdemir,Z.B.,Brostow,W.,Wahrmund,J.,2014.

Zeta potential-viscosity relationship in kaolinite slurry in the presence of dispersants.Arab.J.Sci.Eng.39,5451–5457.https://www.sodocs.net/doc/269657752.html,/10.1007/s13369-014-1146-z.

Fall,M.,Benzaazoua,M.,2005.Modeling the effect of sulphate on strength development of paste back?ll and binder mixture optimization.Cem.Concr.

Res.35,301–314.

Fall,M.,Benzaazoua,M.,Ouellet,S.,2005.Experimental characterization of the in?uence of tailings?neness and density on the quality of cemented paste back?ll.Miner.Eng.18,41–44.https://www.sodocs.net/doc/269657752.html,/10.1016/j.mineng.2004.05.012. Fernández-Jiménez,A.,Palomo,J.G.,Puertas,F.,1999.Alkali-activated slag mortars.

Cem.Concr.Res.29,1313–1321.https://www.sodocs.net/doc/269657752.html,/10.1016/S0008-8846(99) 00154-4.

Ganesh Babu,K.,Sree Rama Kumar,V.,2000.Ef?ciency of GGBS in concrete.Cem.

Concr.Res.30,1031–1036.https://www.sodocs.net/doc/269657752.html,/10.1016/S0008-8846(00)00271-4. Gasteiger,H.A.,Frederick,W.J.,Streisel,R.C.,1992.Solubility of aluminosilicates in alkaline solutions and a thermodynamic equelibrium model.Ind.Eng.Chem.

Res.31,1183–1190.

Hong,S.,Glasser,F.P.,1999.Alkali binding in cement pastes.Cem.Concr.Res.29, 1893–1903.

Kashani,A.,Provis,J.L.,Qiao,G.G.,Van Deventer,J.S.J.,2014.The interrelationship between surface chemistry and rheology in alkali activated slag paste.Constr.

Build.Mater.65,583–591.https://www.sodocs.net/doc/269657752.html,/10.1016/j.conbuildmat.

2014.04.127.

Kesimal,A.,Cihangir,F.,Ercikdi,B.,Deveci,H.,Alp,I.,2010.Optimization of Paste Back?ll Performance for Different Ore Types in Cayeli Copper Mine.Karadeniz Technical University,Revolving Fond Project,Turkey.

Kesimal,A.,Deveci,H.,Ercikdi,B.,Cihangir,F.,2012.Evaluation of Paste Back?ll Performance of Different Mill Tailings in Kure Copper Mine.Karadeniz Technical University,Revolving Fond Project,Turkey.

Khater,H.M.,2014.Studying the effect of thermal and acid exposure on alkali-activated slag geopolymer.Adv.Cem.Res.26,1–9.https://www.sodocs.net/doc/269657752.html,/10.1680/ adcr.11.00052.

Komnitsas,K.,Zaharaki,D.,2007.Geopolymerisation:a review and prospects for the minerals industry.Miner.Eng.20,1261–1277.https://www.sodocs.net/doc/269657752.html,/10.1016/j.

mineng.2007.07.011.

Komnitsas,K.,Zaharaki,D.,Perdikatsis,V.,2009.Effect of synthesis parameters on the compressive strength of low-calcium ferronickel slag inorganic polymers.J.

Hazard.Mater.161,760–768.

Krizˇan, D.,Komljenovic′,M.,Zˇivanovic′, B.,2005.The in?uence of different parameters on the hydration process of binders based on alkali activated slag.

J.Serbian Chem.Soc.70,97–105.https://www.sodocs.net/doc/269657752.html,/10.2298/JSC0501097K. Krizan,D.,Zivanovic,B.,2002.Effects of dosage and modulus of water glass on early hydration of alkali–slag cements.Cem.Concr.Res.32,1181–1188.http://dx.doi.

org/10.1016/S0008-8846(01)00717-7.

Law,D.W.,Adam,A.A.,Molyneaux,T.K.,Patnaikuni,I.,2012.Durability assessment of alkali activated slag(AAS)concrete.Mater.Struct.45,1425–1437.http://dx.

https://www.sodocs.net/doc/269657752.html,/10.1617/s11527-012-9842-1.

Lloyd,R.R.,Provis,J.L.,Van Deventer,J.S.J.,2010.Pore solution composition and alkali diffusion in inorganic polymer cement.Cem.Concr.Res.40,1386–1392. Malviya,R.,Chaudhary,R.,2006.Factors affecting hazardous waste solidi?cation/ stabilization:a review.J.Hazard.Mater.137,267–276.

Ouellet,S.,Bussière, B.,Aubertin,M.,Benzaazoua,M.,2007.Microstructural evolution of cemented paste back?ll:mercury intrusion porosimetry test results.Cem.Concr.Res.37,1654–1665.https://www.sodocs.net/doc/269657752.html,/10.1016/j.cemconres.

2007.08.016.

Pacheco-Torgal,F.,Abdollahnejad,Z.,Cam?es,A.F.,Jamshidi,M.,Ding,Y.,2012.

Durability of alkali-activated binders:a clear advantage over Portland cement or an unproven issue?Constr.Build.Mater.30,400–405.

Palomo,A.,Grutzeck,M.W.,Blanco,M.T.,1999.Alkali-activated?y ashes.Cem.

Concr.Res.29,1323–1329.https://www.sodocs.net/doc/269657752.html,/10.1016/S0008-8846(98)00243-9. Panagiotopoulou, C.,Kontori, E.,Perraki,T.,Kakali,G.,2007.Dissolution of aluminosilicate minerals and by-products in alkaline media.J.Mater.Sci.42, 2967–2973.

Papadakis,V.G.,Antiohos,S.,Tsimas,S.,2002.Supplementary cementing materials in concrete.Part II:A fundamental estimation of the ef?ciency factor.Cem.

Concr.Res.32,1533–1538.https://www.sodocs.net/doc/269657752.html,/10.1016/S0008-8846(02)00829-3. Ravikumar,D.,Peethamparan,S.,Neithalath,N.,2010.Structure and strength of NaOH activated concretes containing?y ash or GGBFS as the sole binder.Cem.

https://www.sodocs.net/doc/269657752.html,pos.32,399–410.https://www.sodocs.net/doc/269657752.html,/10.1016/j.cemconcomp.

2010.03.007.

Ruiz-Santaquiteria,C.,Skibsted,J.,Fernández-Jiménez,A.,Palomo,A.,2012.Alkaline solution/binder ratio as a determining factor in the alkaline activation of aluminosilicates.Cem.Concr.Res.42,1242–1251.https://www.sodocs.net/doc/269657752.html,/10.1016/j.

cemconres.2012.05.019.

Shi,C.,Day,R.L.,1996.Selectivity of alkaline activators for the activation of slags.

Cem.Concr.Aggr.18,8–14.

Shi,C.,Fernández-Jiménez,A.,Palomo,A.,2011.New cements for the21st century: the pursuit of an alternative to Portland cement.Cem.Concr.Res.41,750–763.

https://www.sodocs.net/doc/269657752.html,/10.1016/j.cemconres.2011.03.016.

126 F.Cihangir et al./Minerals Engineering83(2015)117–127

Shi,C.,Krivenko,P.V.,Roy,D.,2006.Alkali-Activated Cements and Concretes.Taylor and Francis,London,New York.

Song,S.,Sohn,D.,Jennings,H.M.,Mason,T.O.,2000.Hydration of alkali-activated ground granulated blast furnace slag.J.Mater.Sci.35,249–257.http://dx.doi.

org/10.1023/A:1004742027117.

Tariq,A.,Nehdi,M.,2007.Developing durable paste back?ll from sulphidic tailings.

Waste Resour.Manage.160,155–166.https://www.sodocs.net/doc/269657752.html,/10.1680/warm.2007.

160.4.155.

TS EN196-6,2010.Methods of Testing Cement–Part6:Determination of Fineness. TS EN197-1,2002.Cement.Part https://www.sodocs.net/doc/269657752.html,position and Conformity Criteria for Common Cements,TSE-Turkish Standards Institution,Ankara.

Wainwright,P.J.,Rey,N.,2000.In?uence of ground granulated blastfurnace slag (GGBS)additions and time delay on the bleeding of concrete.Cem.Concr.

Compos.22,253–257.https://www.sodocs.net/doc/269657752.html,/10.1016/S0958-9465(00)00024-X. Walker,R.,Pavía,S.,2011.Physical properties and reactivity of pozzolans,and their in?uence on the properties of lime–pozzolan pastes.Mater.Struct.44,1139–1150.Wu,D.,Fall,M.,2012.Coupled modeling of temperature distribution and evolution in cemented tailings back?ll structures that contain mineral admixtures.

Geotech.Geol.Eng.30,935–961.

Xu,H.,Van Deventer,J.S.J.,2000.The geopolymerisation of alumino-silicate minerals.Int.J.Miner.Process.59,247–266.https://www.sodocs.net/doc/269657752.html,/10.1016/S0301-7516(99)00074-5.

Yang,T.,Chang,T.,Chen,B.,Shih,J.,Lin,W.,2012.Effect of alkaline solutions on engineering properties of alkali-activated GGBFS paste.J.Mar.Sci.Technol.20, 311–318.

Yilmaz, E.,Belem,T.,Bussiere, B.,Benzaazou,M.,2011.Relationships between microstructural properties and compressive strength of consolidated and unconsolidated cemented paste back?https://www.sodocs.net/doc/269657752.html,pos.33,702–715. Zhang,L.,Ahmari,S.,Zhang,J.,2011.Synthesis and characterization of?y ash modi?ed mine tailings-based geopolymers.Constr.Build.Mater.25,3773–3781.

Zˇivica,V.,2007.Effects of type and dosage of alkaline activator and temperature on the properties of alkali-activated slag mixtures.Constr.Build.Mater.21,1463–1469.

F.Cihangir et al./Minerals Engineering83(2015)117–127127