Manufacture of hybrid cements with fly ash and bottom ash from a municipal solid waste incinerator

Construction and Building Materials 105 (2016) 218–226

Contents lists available at ScienceDirect

Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat

Manufacture of hybrid cements with fly ash and bottom ash from a municipal solid waste incinerator I. Garcia-Lodeiro ⇑, V. Carcelen-Taboada, A. Fernández-Jiménez, A. Palomo Instituto de Ciencias de la Construcción Eduardo Torroja, Madrid, Spain

g r a p h i c a l a b s t r a c t


Valorization of high proportions of

2d 28d

60

Mechanical Strength (MPa)

incinerator fly ash and bottom ash via geopolymerization process.
Development of a hybrid cement with useful technical properties.
Immobilization of metal species present un MSW incinerator ashes in the material and hence reduction in their toxicity.

55

Wash off

40 % MSWI

50 45

Fij(c)

h i g h l i g h t s

40 35 30 25 20

40 % MSWI

Diffusion Disolution

15 10 5 0

RS 60/40

CEMENT

FLEXURAL

a r t i c l e

i n f o

Article history: Received 1 July 2015 Received in revised form 11 November 2015 Accepted 11 December 2015

Keywords: MSW incinerator fly ash and bottom ash Alkaline activation Hybrid cements Leaching

CEMENT

COMPRESSIVE

f(t)

a b s t r a c t Concerns about the large volume of fly ash and bottom ash generated by the incineration of municipal solid waste (MSW) have induced the scientific community to seek ways to reduce their environmental impact. One of the proposals that has been researched most intensely is their valorisation as supplementary cementitious materials and aggregates in Portland cement-based pastes, mortars and concretes. The present paper proposes an alternative use for this waste: as a raw material in alkali-activated hybrid cements. A hybrid cement developed for that purpose by blending 60 wt% clinker and 40 wt% incinerator bottom ash and fly ash exhibited good 28-day mechanical strength (upward of 32.5 MPa). The leaching of potentially toxic metals present in the hybrid cement as a result of the inclusion of MSW fly ash and bottom ash was tested with different leaching procedures. The findings and the results of the analysis of the leaching parameters measured (Li, Fij. . .) showed that the hybrid cement proposed can effectively immobilise the potentially hazardous metals present in MSW fly ash and bottom ash. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction Incineration is one of the alternatives for managing municipal solid waste (MSW). Given the steep rise in MSW generation, the number of incinerators in developed countries is expected to grow ⇑ Corresponding author. E-mail address: [email protected] (I. Garcia-Lodeiro). http://dx.doi.org/10.1016/j.conbuildmat.2015.12.079 0950-0618/Ó 2015 Elsevier Ltd. All rights reserved.

RS 60/40

steadily, with a concomitant increase in the amount of bottom and fly ash. Although some 46% of MSW incinerator ashes in Europe is re-used, billions of tonnes remain which must be treated to prevent subsequent environmental problems [1]. There is some potential for using MSW incinerator ashes in construction materials such as concrete fillers, aggregate or admixtures [1–13]. Other industrial by-products, including coal fly ash and blast furnace slags, are routinely used in cement manufacture. Due to its

219

I. Garcia-Lodeiro et al. / Construction and Building Materials 105 (2016) 218–226

significant lime content (24–47%) and the silicates and aluminosilicates in its composition, MSW incinerator fly ash holds promise as a component in cement production. One of the applications for this waste that has been researched with particular interest is the manufacture of calcium sulfoaluminate cements (obtained at relatively kiln temperatures [1–3]. To be so used, however, the ash must be pre-treated to eliminate or control its heavy metal content to avert undesirably high concentrations of such elements [1,2]. Alkaline cements are binders available today whose performance can match or even exceed ordinary Portland cements thanks to their physical–chemical properties, although mineralogically they differ from OPC. They are obtained by alkali-activating amorphous or vitreous aluminosiliceous materials. When added to a highly alkaline medium (normally containing a NaOH or Na2SiO3 solution or certain inorganic salts), such materials undergo an intense structural transformation, ultimately generating compact cementitious skeletons [14–20]. Alkaline cements have also shown good potential in the field of stabilisation and 53 solidification of wastes containing soluble heavy metals [21–23]. Alkaline activation is a technique that has been successfully applied to coalfired steam power plant fly ash [19,24] and blast furnace slag [25–27]. Moreover, research is presently underway to formulate hybrid cements (or blended alkaline cements) by alkali-activating blends of 20–30% Portland clinker and 70–80% natural pozzolanor industrial by-product-based supplementary cementitious materials (SCMs) [28–30]. The present research aimed to assess the application of alkaline activation technology to incinerator bottom ash and fly ash and moderate amounts of Portland clinker to develop a hybrid cement compliant with the quality requirements set out in European standard EN-197 (EN 196-1) [31] and the leaching thresholds laid down by the United States Environmental Protection Agency (USEPA) [32,33]. 2. Experimental 2.1. Starting materials A Portland cement clinker (CK) supplied by a European Union manufacturer and MSW incineration fly ash (R1) and bottom ash (R2) from a facility also in the EU were used in this study. The reference system consisted of a type IV pozzolanic cement (CEM IV) consisting of a blend of 65% clinker and 35% pozzolanic coal fly ash sourced from the same manufacture as the clinker. The clinker (CK) and MSW bottom ash (R2) were dry milled separately until 100% of their particles passed through a 45-lm sieve. The MSW fly ash (R1) was used as produced in the incinerator with no prior grinding, since 85% of the particles in ex-plant sample were already smaller than 45 lm. The ground raw materials were blended to prepare a hybrid cement consisting of 60 wt% clinker (CK) and 40 wt% incinerator waste (R). The incinerator waste (R), in turn, consisted of a blend of 17% fly ash (R1) and 83% bottom ash (R2). Five per cent sulphate (a mix of CaSO4 and Na2SO4) was added to the aforementioned blend to regulate setting and introduce alkalis in the system. The resulting blend was dry milled to obtain a material in which 96% of the particles passed the 45-lm sieve and over 90% the 30-lm sieve. The chemical composition of all these materials is given in Table 1. Elemental composition was determined by X-ray fluorescence, using radiation at an acceleration voltage of 100 kV and an 800-mA current (Philips PW 1404/00/01). The

silicocalcareous incinerator fly ash (R1) was high in CaO and low in SiO2 and Al2O3. It also exhibited fairly high Cl and alkali contents. The incinerator bottom ash (R2) was CaO-rich as well but had higher percentages of SiO2, Fe2O3 and Al2O3 than observed in the ash. CaO was the most prominent constituent of the hybrid cement, which had lower silica and alumina contents than found in the CEM IV used as a reference. The trace elements present in the various materials are listed in Table 2. The elements of particular interest, given their toxic and hazardous potential, bear an asterisk (⁄). Other toxic and hazardous elements normally detected in this type of waste [1], such as silver, mercury and cadmium, proved to be below the instrument’s detection threshold and lower than the ceiling concentrations defined by the EPA to constitute toxic and hazardous waste [32,33]. Furthermore, the trace metal content was considerably lower in the hybrid cement (in particular Ba, Cr and Zn) than in the incinerator waste used as a raw material due to the obvious dilution of the latter (Table 2). X-ray diffractograms of powdered samples were recorded with a Phillips PW 1730 CuKa radiation diffractometer. Specimens were step-scanned at 2° min1, with a 2h angle of 2–60°, a 1° divergence slit, a 1° anti-scatter slit and a 0.1-mm receiving slit. The diffractograms for the clinker and commercial cement CEM IV are reproduced in Fig. 1. The clinker comprised a mix of crystalline phases, with alite (C3S) and belite (C2S) as the majority minerals. It also contained other characteristic phases such as tricalcium aluminate (C3A) and C4AF, a ferrite. In addition to the characteristic clinker phases, the reference cement diffraction pattern contained a series of lines attributed to crystalline minerals such as quartz (SiO2) and mullite (3Al2O32SiO2), normally present in the coal combustion fly ash additioned to this cement [24]. Fig. 1 also shows the diffractograms for the incinerator fly ash (R1) and bottom ash (R2). The fly ash consisted primarily of quartz (SiO2), anhydrite (CaSO4), halite (NaCl), periclase (MgO), sylvite (KCl), portlandite (Ca(OH)2), calcite (CaCO3) and calcium hydrochloride (CaClOH). Calcite (CaCO3), portlandite (Ca(OH)2), akermanite (Ca2MgSi2O7), quartz (SiO2), magnetite (Fe2O3), gehlenite (Ca2Al(AlSi)), halite (NaCl), calcium sulphate and an aluminium and magnesium hydroxide hydrate (Mg6Al2(OH)184.5H2O) were identified on the diffractogram for bottom ash. In both cases a hump at 2h 25–40° was observed, normally associated with the presence of vitreous material [21]. It was more visible in the bottom ash, perhaps an indication of a higher percentage of vitreous material than in the ash. 2.2. Hybrid cement mortar and paste preparation As recommended in European standard EN-196-1, the bending and compressive strength values of hybrid cement RS 60/40 were tested on 4  4  16-cm3 prismatic mortar specimens prepared with standardised siliceous sand aggregate at an aggregate/cement ratio of 3:1. De-ionised water was added during mixing at a L/S ratio of 0.5. These specimens were cured for 24 h in a chamber at 21 °C and 99% relative humidity. After removal from the moulds they were stored in the chamber for a further 1 day or 27 days, then tested on an IBERTEST Autotest-200/10-SW frame. The reference specimens consisted in commercial cement type IV mortars cured under the same conditions as the mortars made with hybrid cement RS 60/40. Cement pastes were prepared and cured under the same conditions as the mortars in order to better identify the activation products generated in the hybrid cements. The hydration medium was de-ionised water here as well, at a w/c ratio of 0.4. The specimens were characterised on a JEOL JSM 5400 scanning electron microscope fitted with a LINKS-ISIS energy dispersive microanalysis system and on a PHILIPS diffractometer. 2.3. Leaching tests Hybrid cement immobilisation of the heavy metals present in the incinerator waste (Table 2) was assessed with two test methods [34–39] the toxicity characteristic leaching procedure (TCLP) [37] and ANSI/ANS 16.1 [36]. Whereas the former merely classifies the material as hazardous or otherwise depending on the concentration of the leached element, the latter furnishes valuable information on the mechanisms governing element leaching in the sample analysed and its immobilisation by the matrix.

Table 1 Chemical composition of the raw materials (wt%), as per XRF analysis.

Clinker (CK) CEM IV Inc. fly ash (R1) Inc. bottom ash (R2) RS 60/40d a

CaO

SiO2

Al2O3

Fe2O3

MgO

SO3

P2O5

Na2Oeqa

Cl

Br

Otherb

LoIc

67.96 43.67 37.34 35.01 47.74

20.49 26.78 2.54 16.77 15.20

4.68 12.59 1.15 7.27 4.08

1.12 3.19 0.43 11.97 7.34

1.12 2.29 0.97 3.78 2.06

1.12 3.56 5.10 2.95 5.37

0.12 0.24 0.52 2.45 0.91

0.72 1.56 10.37 3.37 5.27⁄

– 0.04 12.46 1.04 2.05

– – 0.09 0.02 0.01

3.25 1.06 1.25 1.84 1.09

0.46 4.96 27.80 13.60 8.84

Na2O eq.: wt% Na2O + wt% K2O. Others: minority metal oxides. c LoI: loss on ignition (1000 °C). d Hybrid cement: 60 wt% clinker + 40 wt% MSW incinerator waste, where MSW = 17% ash (R1) + 83% slag (R2) (⁄Na2O eq = 5.27% = 1.82% from MSIW + 3.45% from activator). b

220

I. Garcia-Lodeiro et al. / Construction and Building Materials 105 (2016) 218–226

Table 2 Trace elements present in the raw materials and anhydrous hybrid cement RS 60/40. Total metal content (mg/kg)

CK R1 R2 RS 60/40

Ba⁄

Cr⁄

Pb⁄

Zn⁄

Cu⁄

Ni⁄

Mn

Sr

Zr

Sn

Ti

– – 850 278

102 51 277 150

– 785 759 696

321 4587 2747 1944

– – 1697 889

– 19 71 67

1471 132 743 929

507 312 549 524

– 59 481 74

– 600 273 123

899 2127 4249 2097

XRD

p

55 CaO

CaO

c

R1

sc h qh

h s c *c αc

p c

p

Mechanical Strength (MPa)

p g

p α h

c f

k

R2

c/k c pc c η/g c q h g pk p c kp η q cg k A/B A/B

χp f

A CK

C AF 4

A A AA B B C AF 4

AA

C AFm q

C AF CEM IV 4 Am B q

10

20

4

C AF 4

χ

A

A/B

A/B A

C AF 4 BB C AC AF BB

A

3 4 B A/B C AF A A/B 4 mBBAB B BB C4AF m

A C3 A C4 AF 30

2d 28d

60

50

45 40 35 30 25 20 15 10 5 0 RS 60/40

C3 A

40

50

60

2θ Fig. 1. Diffractograms for clinker (CK), reference cement (CEM IV), incinerator fly ash (R1) and bottom ash (R2) (Legend: alite (A = C3S), belite (B = C2S), tricalcium aluminate (C3A), (C4AF), quartz (q = SiO2), mullite (m = 3Al2O32SiO2), anhydrite (g = CaSO4), halite (h = NaCl), periclase (a = MgO), sylvite (s = KCl), portlandite (p = Ca(OH)2), calcite (c = CaCO3), calcium hydrochloride (⁄ = CaClOH)), akermanite (k = Ca2MgSi2O7), magnetite (f = Fe2O3), gehlenite (g = Ca2Al((AlSi)O7), and aluminium and magnesium hydroxide hydrate (v = Mg6Al2(OH)184.5H2O)).

For the TCLP test, cubic mortar specimens were ground to a particle size of 4.5– 9.9 mm. The test was conducted with 100 g of sample and 2 L of fluid 2, consisting of 5.7 mL of glacial acetic acid in up to 1 L of distilled water (pH = 2.88 ± 0.05) [37]. The leachant containing the sample was stirred for 18 h at 30 rpm and a temperature of 23 °C. The liquid phase was then poured off and its pH measured, while the concentration of the elements studied was determined with ICP. For the ANSI/ANS test, 3-cm cubic RS 60/40 hybrid cement mortar specimens were cured for 28 days in a chamber at 21 °C and 99% RH. They were subsequently placed in individual leaching tanks containing de-ionised water (conductivity 5 lh, TOC < 3 ppm, 20–25 °C). The leachant was refreshed after the following standard test times: 30 s, 2 h, 7 h, 1 d, 2 d, 3 d, 4 d, 5 d, 18 d, 28 d, 48 d and 90 d [36]. The pH was found and the concentration of the elements determined with ICP after each of the aforementioned times. Three replicas were performed. The pH values were measured on a TOLEDO pH-meter and the elemental concentrations were found on a Perkin-Elmer Spectromass 2000 inductively coupled plasma-mass spectrometer.

3. Results 3.1. Mechanical strength and characterisation of reaction products Fig. 2 shows the bending and compressive strengths of the mortars made with hybrid cement RS 60/40 and the CEM IV reference material. The mortars made with the hybrid cement developed good bending strength, with values slightly lower than those observed for the reference. In cement RS 60/40, both bending and compressive strength rose with hydration time, the latter reaching 33 MPa in the 28-day specimens, slightly higher than required for commercial cements (EN 196-1). This finding served as an initial indication that the alkaline activation applied to hybrid cement RS

CEMENT

FLEXURAL

RS 60/40

CEMENT

COMPRESSIVE

Fig. 2. Flexural and compressive strength values for hybrid cement RS 60/40 and reference cement CEM IV.

60/40 induced the development of reaction products with suitable cementitious properties. The diffractograms for anhydrous hybrid cement RS 60/40 and its 2- and 28-day alkali-activated pastes are reproduced in Fig. 3. The diffractogram for the anhydrous cement exhibited a series of lines attributed to the crystalline phases in the original anhydrous clinker (alite, belite, C4AF and C3A) (Fig. 1). The less intense lines observed were associated with crystalline phases detected in the incinerator waste (see Fig. 1): portlandite, calcite, quartz, gehlenite, magnetite and halite (NaCl). Due to the dilution involved in blending the starting ash and bottom ash with the clinker, the other crystalline phases identified in the waste fractions (Fig. 1) when analysed separately were not detected in the newly designed cement. A series of lines on this XRD pattern and not observed previously were generated by the CaSO4 added as a setting regulator. The diffractogram also exhibited an amorphous hump at 2h 24–40° attributed to the vitreous component present in the incinerator bottom ash. The patterns for the activated pastes differed substantially from the traces for the anhydrous cement. The lines for alite and belite declined in intensity with hydration time, nearly disappearing after 28 days, an indication of normal clinker hydration. Highly soluble inorganic salts, NaCl and sulphates, disappeared completely after hydration. The gehlenite present in the bottom ash in the initial anhydrous blend also disappeared. Phases such as quartz and magnetite remained essentially unaltered, an indication that they were not involved in alkaline activation [24]. The diffractograms for the hydrated pastes also exhibited a series of diffraction lines associated with products normally detected in Portland cement hydration, primarily ettringite and portlandite [40,41]. While Ca(OH)2 was identified as a minority compound in the anhydrous hybrid cement (present in the incinerator ash and bottom ash), the intensity of its lines rose after alkaline activation, particularly in the 2-day materials, an unequivocal sign of normal

221

I. Garcia-Lodeiro et al. / Construction and Building Materials 105 (2016) 218–226

RS 60/40

c

c

*

*

p

p

10

20

30

40

50

60

p

q

ε

ε

28d

A/B A/B

f p q e e A/g/c 2d q

η

g

η

f

A c

A/B A/B A/B

A/f η h g

A e

c/f

A/B A/B

A/B e

c/f

A/B A

A/B B f

c/f

B/c

C3A C4AF

Anh.

26

c

er

p

A/c

c/f

pc p ε c pA p Af c/f er c Ae e/c eq ε p e re e p A/c 28d A/B A/B A/p c pA A/B p Af c p f η A/Bf Ac c/f p q c e e ee A/B e e e A/g/c BA/B e e A A/B h A η h A/B pc 2d A A/c q A B f f C4AF g A p hA/f g/η C3A B/c C4AF Anh.

28

30

32

34

36

38

40

42

44

2θ

2θ

Fig. 3. Diffractograms for anhydrous RS 60/40 and its 2- and 28-day hydrated pastes (Legend: alite (A = C3S), belite (B = C2S), tricalcium aluminate (C3A), ferrite (C4AF), quartz (q = SiO2), halite (h = NaCl), portlandite (p = Ca(OH)2), calcite (c = CaCO3), (gypsum g = Ca2SO4), magnetite (f = Fe2O3), gehlenite (g = Ca2Al((AlSi)O7), ettringite (e = Ca6Al2(SO4)3(OH)1226H2O), r: AFm-SO4-CO Ca4Al2O6(CO3)0.67(SO3)0.3311H2O), e: Ca8Al6Si24O8018.9H2O; ⁄ (Mg0.92Ca0.08CO33H2O).

clinker hydration (C3S + H2O ? Ca(OH)2 + C-S-H gel) [40,41]. The AFm-SO4-CO3-like phase identified here was also found by other authors working with cement mortars containing incinerator ash and bottom ash [1,42]. Carbonates, essentially CaCO3 but also calcium–magnesium compounds, were observed at both ages, especially after 28 days. The presence of calcium carbonate had a dual explanation. It was both present in the incinerator waste fraction of the anhydrous cement (primarily the bottom ash, Fig. 1) and generated in the hydrated pastes as a result of portlandite carbonation by atmospheric CO2 (Ca(OH)2 + CO2 ? CaCO3) [40,41]. That, in turn, translated into a decline in the intensity of its diffraction lines on the 28-day pattern and a rise in the lines associated with carbonates. In addition to carbonating, portlandite may react with the silica present in the ash and bottom ash to generate C-S-H gel-like products (pozzolanic reaction) [40,41]. The calcium aluminosilicate (Ca8Al6Si24O8018.9H2O) that appeared in the 28-day XRD pattern may have been a secondary product of that reaction (Fig. 3). The 2-day SEM analysis run to obtain a fuller understanding of the gel or gels generated in this hybrid cement (Fig. 4) revealed that the pastes had a compact matrix containing anhydrous clinker phases, crystalline phases from the incinerator waste (Fig. 4(a)) and magnetite (Fig. 4(b)). A C-A-S-H-like gel comprising essentially calcium, silicon and aluminium, as well as lesser concentrations of elements such as sulphur (S) (Ka = 2.3075), chlorine (Cl) (Ka = 2.6219) and sodium (Na) (Ka = 1.0410), precipitated heavily throughout the matrix (see EDX microanalysis, Fig. 4(a)). Formation of this gel was associated with the uptake of these elements (especially aluminium) by a C-S-H gel precipitating during Portland clinker hydration. Generation of part of these C-A-S-H gels by the activated incinerator waste, particularly the bottom ash, cannot be ruled out, however, for similar behaviour has been observed in alkali-activated blast furnace slag [26,27]. Nonetheless, the chemical composition of the bottom ash and the low proportion of vitreous waste in MSW (normally under 15%) [1] would limit that source of the gel to fairly small amounts. The low concentration of the heavy elements in the cementitious matrix hinders their detection. That notwithstanding, some of the EDX analyses of points in the cement matrix where the gel phase was present contained diffraction lines attributable to metals such as Pb (Ka = 2.3426; La = 10.5517) and Zn (La = 1.0118; Ka = 3.6313) (see EDX, Fig. 4(a)), suggesting that the matrix was

able to fix the heavy elements initially present in the incineration waste. Fig. 5(a) reproduces a 28-day paste micrograph by way of example. Here also the matrix was compact with barely distinguishable anhydrous clinker phases (confirming the XRD findings). Quartz particles from the incinerator waste were likewise identified, along with Ca deposits associated with carbonate formation. An analysis of different points on the cement matrix (gel) revealed the presence of gels with similar morphology but different compositions. One series of C-A-S-H-like gels was found to be high in calcium, silica and alumina, while others contained those three elements as well as the sulphur and chlorine (see EDX, Fig. 5(a)) detected in the 2-day materials (see Fig. 4(a)). As in the 2-day analysis, the EDX for the cement matrix (gel phase) exhibited diffraction lines associated with trace metals, primarily Zn and Pb. Fig. 5(b) shows both the main reaction product, a C-A-S-H-like gel with variable amounts of S, and crystals precipitating in a pore which, further to their EDX-determined chemical composition, could be chlorine-bearing AFm-like phases (AFm-Cl phases). The literature contains reports of AFm formation as a secondary reaction product in systems containing incinerator waste. This chlorine-bearing phase in particular was detected by Zhu et al. [43] in materials containing incinerator ash. 3.2. Leaching tests 3.2.1. TCLP The toxic characteristic leaching procedure (TCLP) is a leaching test that simply indicates whether the concentration of the toxic elements leached from the matrix studied lies within EPAdefined limits [2,33]. Table 3 lists the post-leaching concentrations of Ba, Cr and Pb, elements deemed by the US EPA to be toxic, and of other metals present in the hybrid cement studied here (Zn, Cu, Ni, Mn, Sr, Zr, Sn and Ti) for which the EPA has established no hazard threshold. Given the high concentration of Cl in the anhydrous hybrid cement (>2%), primarily from R1, the amount of Cl in the eluate was also analysed. Further to the findings for hybrid cement RS 60/40, the concentration of Ba and Cr in the eluates was much lower than the EPA ceilings. The values for Pb, Zr and Sn were below the instrument’s detection threshold, which were much lower than those ceilings. The inference is that the cement designed was able to effectively immobilise the three elements in question.

222

I. Garcia-Lodeiro et al. / Construction and Building Materials 105 (2016) 218–226

Fig. 4. SEM micrographs of 2-day paste RS 60/40, showing (a) clinker (C3S) and gel phase particles; and (b) matrix with magnetite crystals (Fe2O3).

Fig. 5. SEM micrograph of 28-day paste RS 60/40 showing (a) quartz (SiO2) particles and Ca deposits, along with (N,C)-A-S-H- and C-A-S-H-like gels with S and Cl; and (b) AFm-Cl crystals precipitating in a pore embedded in the matrix.

The rest of the trace metals present in the original matrix (see Table 2) were also detected in the eluates. While the EPA establishes no ceilings above which Ni, Cu and Zn in particular would be deemed toxic or hazardous, European standard EN-12457 (where the leachant is water) [44] does regard them as hazardous if they exceed certain threshold concentrations. The ceilings established in that standard are 40, 100 and 200 ppm for Ni, Cu and Zn, respectively.

Metal ion mobility depends heavily on the leachant pH: metals are normally more soluble in acid media. In other words, even though the TCLP test was more aggressive because of the acidic leachant used, the metal concentrations measured in the eluates were below the thresholds set by European standard EN-12457, which prescribes less demanding test conditions. Further to the data in Table 3, the element that leached least was Ti. In light of the high concentration of this element in the

223

I. Garcia-Lodeiro et al. / Construction and Building Materials 105 (2016) 218–226 Table 3 Concentration (ppm) of metals in the eluate determined by the TCLP.

RS 60/40 % Lch.a TCLP thresholdb Instr. DTc a b c

Ba (ppm)

Cr (ppm)

Pb (ppm)

Zn (ppm)

Cu (ppm)

Ni (ppm)

Mn (ppm)

Zr (ppm)

Sn (ppm)

Ti (ppm)

Cl (ppm)

0.340 4.90 100a 0.07

0.084 2.23 5a 0.012


4.379 4.38 –c 0.008

0.765 3.43 –c 0.005

0.130 6.70 –c 0.023

1.37 5.89 –c 0.02

0.020 0.04 –c 0.012

70 13.65 –c 5

% Leached relative to the total content of this metal in the sample. EPA threshold for deeming waste to be toxic or hazardous. ICP instrument detection threshold; italics: no PA threshold for this element.

hybrid cement, the conclusion drawn is that thanks to alkaline activation, cement RS 60/40 was able to immobilise this element very effectively. The concentrations found for the other elements (Ni > Mn > Ba > Zn > Cu > Cr) never exceeded the EPA toxic/hazardous material threshold. The element with the highest concentration in the eluate was chlorine. From the standpoint of toxicity, however, this element poses less of a problem than Cr, Pb or Cu. Nonetheless, in view of the severe durability concerns (reinforcement corrosion) prompted by its presence, the existing European legislation (EN-197-1; EN 14.216:2005 [44]) limits the Cl content in high-strength cements to 0.1%. Based on the percentage of Cl leached and the initial content in the RS 60/40 blend, the hybrid cement proposed would not meet the chemical requirements in place for high-strength cements. Nonetheless, the same legislation allows over 0.1% chlorides in blast furnace slag cements (type III or VLH III), subject only to specifying the maximum Cl content on bags and delivery slips. The hybrid cement studied could be likened to such cements in view of its high percentage of additions, although its use with steel reinforcement would be precluded. 3.2.2. ANSI/ANS As noted earlier, the ANSI/ANS leaching test is routinely used to assess the potential of a matrix to immobilise/stabilise pollutants over time. The values of the eluate pH at different run times shown in Fig. 6(a) flattened after an initial rapid rise to about 11.5, induced by the release of ionic species weakly bonded to the matrix in the early hours of analysis. Element leaching was analysed using the Godbee et al. [34] equations (as adapted by Conner [35] to calculate the Fij (parameter used to determine the leaching mechanism (cm)) and the Li (leaching index):

Fij ¼ V l 

X

icij=Wj  qp  Sp

where Vl: eluate volume; cij: [Ai], element concentration in the eluate; qp: specimen density; Wj: fraction f Ai in the sample; Sp: specimen area and

Li ¼ Log Dij Dij ¼ s  ð%Aj=DsiÞ  ðVp=SpÞ2  Ti 2

where

p: the number p; %Aj, ratio of Aj leached; Dsi: time interval in

p p seconds; Ti: [( ti + ti  1)/2]2, total cumulative time; Vp: specimen volume Li, the leaching index, is a parameter whose value characterises waste resistance to leaching. The higher the value, then, the greater is the degree of immobilisation, i.e., the greater the capacity of the matrix to prevent release of the pollutant into the environment.

The established ceiling is 6 cm2 s1 [35]: values of over 6 mean that the matrix is able to effectively immobilise the element in question. Fig. 6(a) plots the Fij values for the elements against t1/2 for the matrix studied, RS 60/40. It also shows the type of mechanism that governs the system, based on the slope of the leached element concentration versus time curve (F(c) vs. F/t)). Where dissolution of the surface material is the sole governing mechanism, the slope is zero. Where, on the contrary, the relationship between concentration and time is linear with a slope of 0.5, the sole mechanism governing leaching is diffusion. Where the slope is one, the mechanism involved is simple surface wash-off (Fig. 6(b)). As the graph in Fig. 6(a) shows, the curves for all the elements except Pb and Ba had two slopes. During the first 5–7 h both the metals and the chlorine diffused swiftly, to dissolve more slowly thereafter. In the early leaching stages, Mn and Ni exhibited very steep slopes associated with surface wash-off. In Pb and Ba, surface material dissolution was the sole mechanism from the outset. As in the TCLP analysis, the Zr and Sn concentrations were below the instrumental detection threshold. No Ti was observed to leach. Fig. 7 plots the Li values for the elements against leaching time. Note that as the threshold value of 6 that defines the capacity of a matrix to immobilise waste was exceeded in all cases (except for Mn during the first few minutes), the hybrid cement designed could effectively stabilise nearly all the elements analysed. The sole (and only briefly) non-compliant element, Mn, although a metal, is not a heavy metal whose presence might pose a risk of toxicity. 4. Discussion The present paper addresses a subject of utmost social, environmental and technological importance, namely the immobilisation of toxic or hazardous waste by adding it to cement. The study deployed an innovative technology apt for designing not only new cement-based immobilising matrices, but commercial cements containing a sizeable fraction of urban incinerator (ash + bottom ash) waste that has no adverse effect on their cementitious properties. The compositional and mineralogical analysis of MSW incinerator fly ash (R1) and bottom ash (R2) revealed that this type of material differs substantially from the materials conventionally used in alkaline activation (coal fly ash and blast furnace slag). Nonetheless, the compositional and mineralogical properties of incinerator bottom ash (high Ca, Si and Al contents) make it a promising, potentially reactive material. The alkaline activation applied to produce cement RS 60/40 led to the development of binders with acceptable mechanical strength. The 28-day compressive strength attained by its mortars, 33 MPa, qualified the blend designed here for European cement category 32.5 (EN 196-1). The reaction products proved to be very similar to the products characteristic of normal Portland cement hydration. The calcium silicates in the clinker hydrated normally, disappearing in the 28-day

224

I. Garcia-Lodeiro et al. / Construction and Building Materials 105 (2016) 218–226 12

(a)

11

pH

10 9 8 7

Fij (cm)*10

-3

F(c)

(b) Cl

200 100

-3

Fij (cm)*10

Surfacewash-off effect Slope=0.5

High leaching

Diffusion controlled release

0 60 50

Slope=0

Mn

40

Disolution of material from surface Medium leaching

30 20

Cr

10

Zn

0 10

F(t)

Ni

-3

Fij (cm)*10

Slope=1

Low leaching

5

Cu Pb

0

Ba

0

50

100

150

1/2

(Time)

1/2

(s)

200

*10

250

300

-2

Fig. 6. (a) Increase in pH in eluates and parameter Fij (cm) vs. t1/2 for the elements leaching out of the matrix and (b) graph illustrating the leaching mechanism: concentration function (F(c)) against time function F(t) (adapted from Heasman et al., 1998).

RS 60/40

16 14

2

Li (cm /s)

12 10 8 6 4

Ba Pb

2

Cr Zn

Cu Mn

Ni Cl

0 0

2

4

6

8 10 12 14 40

60

80

100 120 140

Leaching time (h) Fig. 7. Li (cm2/s) measured for the elements detected against leaching time in the RS 60/40 cement matrix.

specimens. Neither the quartz nor the magnetite in the anhydrous waste were affected by activation, whereas the diffraction lines for gehlenite were observed to disappear in the activated materials. The hydraulic gehlenite present in incinerator bottom ash forms octahedral aluminosilicates of the type Ca2Al2SiO78H2O. That reaction takes place fairly rapidly in vitreous gehlenite and more slowly in its crystalline form, although it is hastened at rising pH values [45]. This phase practically disappeared after activation, generating a crystalline calcium aluminosilicate, Ca8Al6Si24O8018.9H2O, detected on the 28-day XRD traces (see Fig. 3). That finding is consistent with results reported by Qiao et al. [45] for hydrothermally treated, alkali-activated ash.

As a result of alkaline activation and the high (>47%) calcium content in the system, the main reaction product was a mix of essentially C-A-S-H- and C-S-H-like gels containing aluminium, alkalis and other elements such as Cl and sulphur, the latter present not only in the calcium sulphate added as a setting regulator, but in the anhydrite identified in the incinerator fly ash and bottom ash. The inclusion of S in the cementitious gels generated is consistent with the findings observed by Donatello et al. [46] in the alkaline activation of hybrid cements in which sodium sulphate was used as an activator. The affinity of C-S-H gels for sulphate ions has been proven in a number of studies [46–48]. These gels take up not only the aforementioned elements, however, but also heavy metals such as Zn or Pb (detected with EDX microanalysis, see Figs. 4 and 5). Heavy elements are known to be immobilised as a result of physical encapsulation or chemical interaction with the cementitious gel in the host matrix [49–51]. The impact of the chloride ion on hydrated Portland cement mineralogy, in turn, has been amply addressed in the literature [52–58]. Chloride ions may interact with hydrated phases in the cement, forming chloraluminates such as Friedel’s or Kuzel’s salts or solid solutions with AFm-like phases [53,56,58] or they may be chemically adsorbed by C-S-H gel [53,54]. In the hybrid cement designed, chlorine appeared both in the composition of the gels generated during activation (see Fig. 5(a)) and in the formation of AFm-Cl-like crystals (see Fig. 6(b)). Rises in the sulphate content in cement lower the binding capacity of chloride. Since sulphates bind more strongly to C3A than chlorides, the presence of sulphate ions de-stabilises Friedel’s salt, converting it into ettringite [57], a phase identified as a secondary product in the cementitious system studied here (see Fig. 3). The high sulphate concentration initially added to the system was depleted within 2 days of hydration (see XRD pattern in Fig. 3) as a result of both the uptake of this species by the gel generated and ettringite precipitation [46–48],

I. Garcia-Lodeiro et al. / Construction and Building Materials 105 (2016) 218–226

although it also participated in the formation of phases such as AFm-SO4-CO3, observed both here and previously in pastes containing incinerator bottom ash [43]. The cementitious matrix generated after cement activation was compact and exhibited no significant macroporosity (see Figs. 5 and 6). One of the concerns around the use of incinerator waste is that residual Al and other amphoteric metals may react with hydroxyl ions at high pH values, producing hydrogen gas further to the following equation:

225


The concentration of chloride ions in the hybrid cement proposed (RS 60/40) exceeded the ceiling laid down in the respective standard. Nonetheless, a fair share of these chlorides can be retained in the cementitious matrix (chemically adsorbed onto the surface of the C-A-S-H gel formed or forming secondary AFm-Cl phases). If these cements are to be used to manufacture structural concrete, preliminary studies are recommended to prevent possible reinforcement corrosion.



2Al þ 2OH þ 6H2 O ! 2½AlðOHÞ4 ðaqÞ þ 3H2ðgasÞ ½1; 50; 51 If the aluminium is in the form of Al(OH)3, however, the alkaline conditions prevailing in the system favour its dissolution without releasing hydrogen: 

AlðOHÞ3 þ OH ! ½AlðOHÞ4 ðaqÞ ½1; 50; 51 Electron microscopic (SEM/EDX) analysis consistently showed that the matrix was compact with barely any pores, which would largely explain its good mechanical performance. An aluminium hydroxide phase that also contained magnesium (Mg6Al2(OH)184.5H2O) was detected on the diffractogram for the anhydrous bottom ash (see Fig. 1(b)). The absence of any substantial porosity would appear to indicate that after alkaline activation this phase, not detected in the anhydrous cement (as a result of dilution), dissolved and aluminium of the form [Al(OH)4] ( aq) participated in the formation of calcium aluminates and the C-A-S-H gels detected by SEM (see Fig. 6(b)). Leaching tests showed that the alkaline activation of hybrid cements lowers the toxicity that might be expected due to the leaching of the metal species initially present in incinerator waste. Essentially three factors were involved in lowering metal species mobility in the cementitious matrix: (1) uptake of the elements in cementitious gels; (2) their precipitation in the form of insoluble inorganic salts (hydroxyls and sulphates); and (3) their dilution due to mixing with the aggregate, binder and hydration liquid [59–62]. While the uptake of certain metals such as Pb or Cr may decline in the presence of Cl ions (which interacts with these elements to form the respective soluble chlorides), none of the elements leached out of the matrix in amounts that exceeded the established toxicity ceilings. This finding merits study in greater depth in future research. The presence of sulphate ions may help reduce the leaching of certain elements such as Ba via the formation of insoluble inorganic salts [59–62]. The present research, then, yielded cement which, despite its 40% MSWI ash content, retained acceptable mechanical strength and proved able to immobilise heavy metals originally present in the ashes. 5. Conclusions
Activating incinerator waste-containing hybrid cements with moderately alkaline solids yielded binders with 28-day compressive strength of 33 MPa, i.e., higher than required for commercial cements.
The study of the reaction products generated after activation revealed the presence of several types of gels: C-(A)-S-H and C-A-S-H gels containing S, Cl and heavy metals such as Pb. Portlandite and ettringite precipitated as secondary reaction products, along with AFm- and AFt-like phases.
The alkaline activation of hybrid cements containing incinerator fly ash and bottom ash lowered the amounts of metal species possibly leaching out of the material and hence their toxicity. The resulting matrices solidified and stabilised these elements, acting not only as a physical but also a chemical barrier via uptake by the precipitating gel or by favouring the precipitation of their insoluble salts.

Acknowledgments This research was funded by the Spanish Ministry of the Economy and Competitiveness under projects BIA2013-43293-R and RTC-2014-2351-5. One of the authors worked under a Postgraduate Studies Council grant co-funded by the Spanish National Research Council and the European Social Fund (ESF). The authors wish to thank A. Gil for his aid with the laboratory trials. References [1] M. Tyrer, Municipal solid waste incinerator (MSWI) concrete, in: F. PachecoTorgal, S. Jalali, J. Labrincha, V.M. John (Eds.), Eco-efficient Concrete, Woodhead Publishing, Cambridge, UK, 2013. [2] C.C. Wiles, Municipal waste combustion ash: state of knowledge, J. Hazard. Mater. 47 (1996) 325–344. [3] P.J. Wainwright, I. Hadzinakis, P. Robery, A review of the methods of utilization of incinerator residues as a construction material, in: Proceedings of the International Conference of Low Cost and Energy Saving Construction Materials, Rio de Janeriro, Brazil, 1984. [4] L. Bertolini, M. Carcana, D. Cassago, M. Collepardi, Q. Curzio, MSWI ashes as mineral additions in concrete, Cem. Concr. Res. 34 (2004) 1899–1906. [5] C.R. Cheeseman, A. Makinde, S. Bethenis, Properties of lightweight aggregate produced by rapid sintering of incinerator bottom ash, Resour. Conserv. Recycl. 43 (2) (2005) 147–162. [6] M. Ferraris, M. Salvo, A. Ventrella, L. Buzzi, M. Veglia, Use of vitrified MSWI bottom ashes for concrete production, Waste Manage. 29 (2009) 1041–1047. [7] O. Ginés, J.M. Chimenosa, A. Vizcarro, J. Formosa, J.R. Rosell, Combined use of MSWI bottom ash and fly ash as aggregate in concrete formulation: environmental and mechanical considerations, J. Hazard. Mater. 169 (2009) 643–650. [8] C. Jaturapitakkul, R. Cheerarot, Development of bottom ash as pozzolanic material, J. Mater. Civ. Eng. 151 (2003) 48–53. [9] B. Juric, L. Hanzic, R. Ilic, N. Samec, Utilization of municipal solid waste bottom ash and recycled aggregate in concrete, Waste Manage. 26 (2006) 239–259. [10] L.L. Lin, K.S. Wang, B.Y. Tzeng, C.Y. Lin, The reuse of municipal solid waste incinerator fly ash slag as a cement substitute, Resour. Conserv. Recycl. 39 (2003) 315–324. [11] U. Müller, K. Rübner, The microstructure of concrete made with municipal waste incinerator bottom ash as an aggregate component, Cem. Concr. Res. 36 (2006) 1434–1443. [12] Z. Pavlík, M. Jerman, M. Keppert, M. Pavlíková, P. Reiterman, R. Cerny, Use of municipal solid waste incineration waste materials as admixtures in concrete, in: Coventry University and The University of Wisconsin Milwaukee Centre for By-products Utilization, Second International Conference on Sustainable Construction Materials and Technologies, 28–30 June, Univesita Politecnica delle Marche, Ancona, Italy, 2010. [13] J. Pera, L. Coutaz, J. Ambroise, M. Chababbet, Use of incinerator bottom ash in concrete, Cem. Concr. Res. 27 (1) (1997) 1–5. [14] Z. Zhang, J.L. Provis, A. Reid, H. Wang, Geopolymer foam concrete. An emerging material for sustainable construction, Constr. Build. Mater. 56 (2014) 113–127. [15] V. Glukhovsky, Ancient, modern and future concretes, in: First Inter. Conf. Alkaline Cements and Concretes, vol. 1, Kiev, Ukraine, 1994, pp. 1–8. [16] A. Palomo, M.W. Grutzeck, M.T. Blanco, Alkali-activated fly ashes – a cement for the future, Cem. Concr. Res. 29 (8) (1999) 1323–1329. [17] C. Shi, P.V. Krivenko, D. Roy (Eds.), Alkali-Activated Cements and Concretes, Taylor & Francis, 2006. [18] C. Shi, Fernández Jiménez, A. Palomo, New cements for the 21st century, the pursuit of an alternative to Portland cement, Cem. Concr. Res. 41 (2011) 750– 763. [19] W.K. Part, M. Ramli, C.B. Cheah, An overview on the influence of various factors on the properties of geopolymer concrete derived from industrial by-products, Constr. Build. Mater. 77 (2015) 370–395. [20] P. Duxson, G.C. Lukey, J.S.J. Van De Venter, S.W. Mallicoat, W.M. Kriven, Microstructural characterisation of metakaolin-based geopolymers, Ceram. Trans. 165 (2005) 71–85. [21] J.G.S. Van Jaarseveld, J.S.J. Van Deventer, A. Schwartzman, The potential use of geopolymeric materials to immobilise toxic metals: Part II. Material and leaching characteristics, Miner. Eng. 12 (1999) 75–91.

226

I. Garcia-Lodeiro et al. / Construction and Building Materials 105 (2016) 218–226

[22] C. Shi, A. Fernández-Jiménez, Stabilization/solidification of hazardous and radioactive wastes with alkali-activated cements, J. Hazard. Mater. B 137 (2006) 1656–1663. [23] S. Donatello, A. Fernandez-Jimenez, A. Palomo, An assessment of mercury immobilisation in alkali activated fly ash (AAFA) cements, J. Hazard. Mater. 213–214 (2012) 207–215. [24] A. Fernández-Jiménez, A. Palomo, Characterization of fly ashes. Potential reactivity as alkaline cements, Fuel 82 (18) (2003) 2259–2265. [25] S.D. Wang, K.L. Scrivener, Hydration products of alkali activated slag cement, Cem. Concr. Res. 25 (3) (1995) 561–571. [26] F. Puertas, Cementos de escorias activadas alcalinamente: Situación actual y perspectivas de futuro, Materiales de Construcción 45 (239) (1995) 53–92. [27] A. Fernández-Jiménez, Cementos de escorias activadas alcalinamente: influencia de las variables y modelización del proceso (Ph.D. thesis), Universidad Autónoma, Madrid, Spain, 2000. [28] I. Garcia-Lodeiro, Fernández-Jiménez, A. Palomo, Variation in hybrid cements over time. Alkaline activation of fly ash–Portland cement blends, Cem. Concr. Res. 52 (2013) 112–122. [29] A. Fernández-Jiménez, F. Zibouche, N. Boudissa, I. García-Lodeiro, M.T. Abadlia, A. Palomo, ‘‘Metakaolin-slag-clinker blends”, the role of Na+ or K+ as alkaline activators of theses ternary blends, J. Am. Ceram. Soc. 96 (6) (2013) 1991– 1998. [30] I. Garcia-Lodeiro, O. Maltseva, Fernández-Jiménez, A. Palomo, Hybrid alkaline cements: Part I. Fundamentals, Rom. J. Mater. 42 (4) (2012) 330–335. [31] EN 196-1, Methods of Testing Cement. Determination of Strength. [32] U.S.E.P.A. Federal Register, 51 (114), 21686-June 13, 1986. [33] Environment Agency, Interpretation of the Definition and Classification of Hazardous Waste, second ed., version 2.2, 2008. [34] H. Godbee et al., Application of mass transport theory to the leaching of radionuclides from solid state waste, Nucl. Chem. Waste Manage. 1 (1980) 29– 31. [35] J.R. Conner, Chemical Fixation and Solidification of Hazardous Wastes, Van Nostrand Reinhold, New York, 1990. [36] ANSI-ANS-16.1.1986, American National Standards Institute, Measurement of the Leachability of Solidified Low-Level Radioactive Waste by a Shotty-Term Test Procedure, American Nuclear Society, 555 North Kensington Avenue, Labrange Park, IL 605525, 1986. [37] U.S.E.P.A Method 1311, Toxicity Characteristic Leaching Procedure, Code of Federal Regulations, 40 CFR part 261, Appendix II, July 1991. [38] J.E. Enrique, E. Monfort, Leaching kinetic of toxic elements immobilized in cement matrices (I), Mater. Eng. 8 (2) (1997) 151–169. [39] L. Heasman, H.A. van der Sloot, Ph. Quevauviller, Harmonization of Leaching/ Extraction Tests, Elsevier, Amsterdam, The Netherlands, 1998 (Chapter 2). [40] F.M. Lea, The Chemistry of Cement and Concrete, third ed., Edward Arnold, Glasgow, UK, 1974. [41] H.F.W. Taylor, Cement Chemistry, second ed., Thomas Telford, London, 1997. [42] X.C. Qiao, M. Tyrer, C.S. Poon, C.R. Cheeseman, Novel cementitious materials produced from incinerator bottom ash, Resour. Conserv. Recycl. 52 (2008) 496–510. [43] F. Zhu, M. Takaota, K. Oshitaa, Y. Kitajima, Y. Inada, S. Mosrisawa, H. Tsunoa, Chlorides behavior in raw fly ash washing experiments, J. Hazard. Mater. 178 (2010) 547–552. [44] EN 12457-2:2002, Characterisation of Waste – Leaching – Compliance Test for Leaching of Granular Waste Materials and Sludges – Part 2: One Stage Batch

[45]

[46]

[47]

[48] [49]

[50]

[51]

[52] [53] [54]

[55] [56]

[57]

[58]

[59]

[60]

[61]

[62]

Test at a Liquid to Solid Ratio of 10 l/kg for Materials with Particle Size Below 4 mm (Without or With Size Reduction). X.C. Qiao, M. Tyrer, C.S. Poon, C.R. Cheeseman, Characterization of alkaliactivated thermally treated incinerator bottom ash, Waste Manage. 28 (2008) 1955–1962. S. Donatello, O. Maltseva, A. Fernandez-Jimenez, A. Palomo, The early age hydration reactions of hybrid cement containing a very high content of coal bottom ash, J. Am. Ceram. Soc. 97 (3) (2013) 929–937. S. Donatello, A. Fernandez-Jimenez, A. Palomo, Very high volume of fly ash cements. Early age hydration study using Na2SO4 as an activator, J. Am. Ceram. Soc. 96 (3) (2013) 900–906. R. Skapa, Optimum Sulphate Content of Portland Cement (Ph.D. thesis), Aberdeen, UK, 2009. A. Roy, H.C. Eaton, F.K. Cartledge, M.E. Tittlebaum, Solidification/ stabilization of heavy metal sludge by a Portland cement/fly ash binding mixture hazard, Waste Hazard. Mater. 8 (1992) 33. J.E. Aubert, B. Husson, A. Vaquier, Metallic aluminium in MSWI fly ash: quantification and influence on the properties of cement-base products, Waste Manage. 24 (2004) 589–596. U. Müller, K. Rübner, The microstructure of concrete made with municipal waste incinerator bottom ash as an aggregate component, Cem. Concr. Res. 36 (2006) 1434–1443. T. Luping, L.O. Nilsson, Chloride binding capacity and binding isotherms of OPC pastes and mortars, Cem. Concr. Res. 23 (2) (1993) 247–253. Y. Xu, The influence of sulphates on chloride binding and pore solution chemistry, Cem. Concr. Res. 27 (12) (1997) 1841–1850. F.P. Glasser, A. Kindness, S.A. Stronach, Stability and solubility relationships in AFm phases Part I. Chloride, sulfate and hydroxide, Cem. Concr. Res. 29 (1999) 861–866. E.P. Nielsen, D. Herfort, M.R. Geiker, Binding of chloride and alkalis in Portland cement systems, Cem. Concr. Res. 35 (2005) 117–123. E. Nielsen, D. Herfort, M. Geiker, D. Hooton, Effect of solid solution of AFm phases on chloride binding, in: Proceedings, 11th Int. Congress on the Chemistry of Cement, South Africa, 2003. H. Hirao, K. Yamada, H. Takahashi, H. Zibara, Chloride binding of cement estimated by binding isotherms of hydrates, J. Adv. Concr. Technol. 3 (1) (2005) 77–84. M. Balonis, The Influence of Inorganic Chemical Accelerators and Corrosion Inhibitors on the Mineralogy of Hydrated Portland Cement Systems (Ph.D. thesis), Aberdeen, UK, 2010. I. Lancellotti, E. Kamseu, M. Michelazzi, L. Barbieri, A. Corradi, C. Leonelli, Chemical stability of geopolymers containing municipal waste incinerator fly ash, Waste Manage. 30 (2010) 673–679. Y. Luna-Galiano, C. Fernández-Pereira, J. Vale, Stabilization/solidification of a municipal solid waste incinerator residue using fly ash-based geopolymers, J. Hazard. Mater. 18 (2011) 5373–5381. E.I. Diaz-Loya, Toxicity mitigation and solidification of municipal solid waste incinerator fly ash using alkaline activated coal ash, Waste Manage. 32 (2012) 1521–1527. C. Ferone, F. Colangelo, F. Messina, L. Santoro, R. Cioffi, Recycling of pre-washed municipal solid waste incinerator fly ash in the manufacturing of low temperature setting geopolymer materials, Materials 6 (2013) 3420–3437, http://dx.doi.org/10.3390/ma6083420.