Fuel 184 (2016) 6980

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Physical and chemical changes in coal fly ash during acidic or neutral
wastes treatment, and its effect on the fixation process
Roy Nir Lieberman a,b,c,, Xavier Querol a, Natalia Moreno a, Yitzhak Mastai b, Haim Cohen c,d
a
Institute of Environmental Assessment and Waster Research (IDA), Consejo Superior de Investigaciones Cientficas (CSIC), Barcelona, Spain
b
Bar Ilan University, Ramat-Gan, Israel
c
Ariel University, Ariel, Israel
d
Ben-Gurion University of the Negev, Beer-Sheva, Israel

a r t i c l e i n f o a b s t r a c t

Article history: Coal fly ash (FA) is an abundant commodity in many countries using steam coal for power production. In
Received 14 April 2016 some countries (e.g. Israel) all the FA is utilized whereas in some countries only low percentage is used
Received in revised form 21 June 2016 (e.g. USA). The main mode of utilization is in the civil engineering industries as cement or clinker additive
Accepted 22 June 2016
or for concrete production. In Israel (2014) 1.3 Mt of highly basic type F FA has been produced, mostly
Available online 7 July 2016
from South African and Colombian coals, in the utilities. A potential application for type F FA could be
as an efficient scrubber for hazardous acidic wastes (e.g. the phosphate industry). The scrubbing effi-
Keywords:
ciency of FA stems from interactions of the aluminosilicate surface of the particles with trace elements
Fly ash
Fixation mechanisms
or with fine precipitates of ionic solids (e.g. SrCO3). The process changes the chemical and physical nature
Trace elements of the FA surface. These changes have been studied using SEM, TEM, XRD and particle size measurements
Surface changes of non-treated, compared to treated FA with acidic or neutral aqueous solutions.
The results indicate that an appreciable part of the outer lime layer and part of the glassy aluminosil-
icate material is dissolved, revealing the mullite skeleton as well as Ti, and Fe species. Moreover, forma-
tion of mixed ettringite/thaumasite and halite particles is observed. Thus, this study can lead to a better
understanding of the fixation mechanisms of the trace elements by coal FA.
2016 Elsevier Ltd. All rights reserved.

1. Introduction
Coal fly ash (FA) is produced via firing of coal in pulverized coal
combustion (PCC) facilities worldwide. In Israel, coal is one of the
most significant electricity sources (63% in 2013 [1,2]), consum-
ing 13 Mt of coal annually. Israel has no coal resources and the
bituminous coals are marine imported and stored in large open
piles of 60150 kt in the yards of the 4 major coal fired power
plants. The annual amount of the FA produced is 1.3 Mt of (in
2014) [2,3].
The pulverized coal particle (average size 30 lm) which contain
10% mineral matter is injected into the boiler and undergoes 2
processes during the 15 s residence time. A pyrolysis step, at
800 C, where the light hydrocarbon gases and molecular hydro-
Corresponding author at: Institute of Environmental Assessment and Waster
Research (IDA), Consejo Superior de Investigaciones Cientficas (CSIC), Barcelona,
Spain.
E-mail addresses: 5dk32nl@gmail.com, roy.nir@idaea.csic.es, roynl@ariel.ac.il
(R.N. Lieberman).
gen are released and ignited [4]. During this stage some inorganic
material melts, and solidifies down in the bottom of the boilers, to
yield bottom ash (BA). In the second stage, the coal particle which
is transformed into char and heats up to 1700 C, is burnt and
small solid particles (117 lm)-the FA are formed [5]. The FA
accounts for 90% of the total ash formed during the coal combus-
tion process [1].
Strict air quality regulations in Israel [6] regarding the emis-
sions of pollutants to the atmosphere via the combustion process
impose beneficiation of the imported coals prior to the transport
(via washing with water), in order to reduce inorganic volatile pol-
lutants such as S, Hg, As, and Cd. Moreover, in PCC plants injection
of Ca compounds (e.g. CaO, CaCO3) is used in order to reduce SOx
emissions via the Flue Gas Desulfurization (FGD) process. The
result of the coal beneficiation process is an alkali and alkali-
earth rich FA product, with pozzolanic properties and is defined
as Class F FA [7].
The ashes produced in Israel (BA and FA) are utilized in the con-
struction industry as cement additive or for concrete production,
thus its economic value is poor (615$/ton) [820]. In order to

http://dx.doi.org/10.1016/j.fuel.2016.06.107
0016-2361/ 2016 Elsevier Ltd. All rights reserved.
70 R.N. Lieberman et al. / Fuel 184 (2016) 6980

increase FA economic value, we have evaluated elsewhere the pos- (1) Cation-Exchange Mechanism anionic functional groups
sibility of using it as an effective neutralization and metal fixation of the FA surface aluminates (AOAAlO ) and silicates
reagent for hazardous acidic wastes and low level radioactive (AOASiO2 ) that behave as cation exchange groups.
waste [10,15,16]. The wastes studied were:

(1) Acidic waste from the phosphate industry, which is a bypro-

duct of phosphate rock treatment with either sulfuric or
hydrochloric acid [16] (0.11 M acidity).
(2) Extremely acidic waste (10 M acidity) formed from the
regeneration processes of used motor oil with Oleum [10].
(3) Radioactive wastes (appropriate metal cations were used as
simulants) that are formed from the fission processes of U
[15].
(2) Coordinative Bonding this is the mechanism in which a
The results have indicated that the FA is an efficient acid neu- Lewis base material (the FA) is donating its lone pair elec-
tralization reagent and also a good fixation reagent to trace ele- trons (mainly from O of the aluminates and silicates) to form
ments, thus concluding that FA could be used as a chemical a bond with a metal cation, which is equivalent to metal
scrubber and fixation reagent for these types of wastes. Further- complexes formation.
more, it was shown that the aggregate product formed from the
acidic wastes (items (1) and (2) above) could be used as a partial
substitute for sand and cement in concrete. In all these studies,
the fixation quality of the FA was tested by three types of leaching
procedures: USEPA TCLP 1311 [21], CALWET [22], and the Euro-
pean Directive (EN 12457-2). All leached trace elements concentra-
tions were well under the CALWET, TCLP, EN 12457-2 [23] criteria,
and also Israel drinking limits [24].
Three modes of fixation mechanism for the fixation of the
metal cations by the FA functional groups have been suggested
[15,2529]:

Fig. 1. Cenospheres and plerospheres of SAFA (A, B) and COFA (C, D), respectively.
 R.N. Lieberman et al. / Fuel 184 (2016) 6980 71

(3) Electrostatic interactions of fine ionic precipitates with FA and the FA surface. Usually in alkaline environment, the FA
surface in alkaline solutions precipitation of several salts has a negative charge, resulting in attraction of these fine
can occur (e.g. SrCO3). It was reported [15] that an effective precipitates to the particles surfaces.
interaction occurs between fine ionic precipitate of SrCO3

Table 1 Table 2
Average chemical composition of SAFA before and after UPDI or 0.1 M HCl treatments. Average chemical composition of COFA before and after UPDI or 0.1 M HCl
treatments.
% SAFA SAFA % Change compared SAFA % Change compared
UPDI to non treated SAFA HCl to non treated SAFA % COFA COFA % Change compared COFA % Change compared
UPDI to non treated COFA HCl to non treated COFA
LOI 5 5.6 12.0 6.1 22.0
SiO2 43.8 44.6 1.8 48.3 10.3 LOI 2.3 1.9 17.4 3.1 34.8
Al2O3 29.9 31.1 4.0 33.3 11.4 SiO2 57.4 57.4 0.0 58.3 1.6
CaO 9.5 8.5 9.7 4.0 57.7 Al2O3 22.2 22.1 0.5 21.5 3.2
Fe2O3 2.9 2.9 1.3 3.2 10.7 CaO 3.4 2.7 20.6 1.3 61.8
K2O 0.8 0.8 0 0.9 9.1 Fe2O3 6.5 6.4 1.5 6.0 7.7
MgO 2.1 2.0 1.7 1.1 46.7 K2O 1.6 1.6 0 1.6 0.0
Na2O 0.2 0.3 50.0 0.4 87.3 MgO 1.7 1.6 5.9 1.3 23.5
P2O5 2.4 1.8 25.0 1.2 48.5 Na2O 2.3 2.1 8.7 2.1 8.7
SO3 0.7 0.5 28.6 0.2 76.6 P2O5 2.3 1.8 21.7 1.1 52.2
TiO2 1.6 1.6 0 1.8 11.7 SO3 0.3 0.3 0.0 0.2 33.3
TiO2 1.0 1.0 0.0 1.1 10.0
mg/kg
mg/kg
Li 387 324.4 16.2 395.2 2.1
Li 102.2 97.4 4.7 92.6 9.4
Ba 2180 1780 18.3 2290 5.0
Ba 3210 3220 0.4 3200 0.3
Be 9.2 7.9 14.1 8.8 4.3
Be 9.8 9.8 0.0 7.5 23.5
B 161 117 27.1 22.2 86.2
B 322 172 46.5 82.5 74.3
Sc 26 21.9 15.8 27.8 6.9
Sc 34.5 36.0 4.3 27.1 21.4
V 143 113 21.3 114 20.6
V 6090 6010 1.4 5660 7.1
Cr 153 132 13.9 138 9.6
Cr 272 255 6.0 183 32.5
Hg 0.42 0.42 0 0.44 +4.8
Hg 0.22 0.22 0 0.25 +13%
Sr 2800 2120 24.3 2430 13.2
Sr 920 849 7.7 536 41.7
Mn 436 340 22.0 318 27.1
Mn 37.6 37.6 0.0 31.8 15.4
Co 30 25.0 16.7 29.1 3.0
Co 99.3 98.4 0.9 85.3 14.1
Ni 48 40.5 15.6 46.0 4.2
Ni 82.5 82.4 0.1 65.7 20.4
Cu 88 74.5 15.3 85.2 3.2
Cu 175 166 5.2 174 0.8
Zn 148 161 8.6 160 8.2
Zn 42.7 40.4 5.4 36.9 13.6
Ga 57 45.6 20.0 50.4 11.6
Ga 26.0 26.0 0.0 11.7 55.0
Ge 11 11.6 5.5 3.7 66.4
Ge 26.5 25.1 5.3 10.5 60.4
As 16 12.6 21.3 7.4 53.8
As 26.8 17.0 36.6 12.5 53.4
Se 10 6.4 36.0 7.1 29.0
Se 62.9 63.1 0.3 61.2 2.7
Rb 34 31.4 7.6 36.8 8.2
Rb 37.6 37.6 0.0 31.8 15.4
Y 85 70.5 17.1 74.2 12.7
Y 53.8 53.8 0.0 36.5 32.2
Zr 279 388.0 39.1 449 60.9
Zr 235 236 0.2 215 8.8
Nb 76 119 55.9 119 56.2
Nb 50.2 50.2 0.0 44.4 11.6
Mo 10 5.9 41.0 6.2 38.0
Mo 25.4 12.1 52.4 11.8 53.5
Cd 0.01 <0.01 <0.01
Cd 2.1 2.2 4.8 2.3 9.5
Sn 6.5 9.6 47.7 7.2 10.8
Sn 4.4 4.4 0.0 3.1 29.5
Sb 0.9 2.2 144.4 0.9 0.0
Sb 15.3 14.2 7.2 11.8 22.9
Cs 4.7 4.3 8.5 5.1 8.5
Cs 5.7 5.7 0.0 5.5 3.5
La 121 104 14.1 122 1.1
La 51.9 51.4 1.0 41.1 20.8
Ce 225 183 18.8 233 3.7
Ce 97.7 101 3.7 75.1 23.1
Pr 27 22.8 15.6 26.5 1.9
Pr 12.2 12.2 0.0 9.4 23.0
Nd 89 75.9 14.7 86.8 2.5
Nd 43.1 42.8 0.7 32.4 24.8
Sm 18 15.4 14.4 17.2 4.4
Sm 9.5 9.4 1.1 6.9 27.4
Eu 2.7 2.3 14.8 2.5 7.4
Eu 2.0 2.0 0.0 1.6 20.0
Gd 17 14.3 15.9 15.7 7.6
Gd 9.4 9.3 1.1 6.8 27.7
Tb 2.5 2.1 16.0 2.2 12.0
Tb 1.4 1.4 0.0 1.0 28.6
Dy 16 13.6 15.0 14.4 10.0
Dy 9.4 9.6 2.1 6.8 27.7
Ho 3.1 2.6 16.1 2.8 9.7
Ho 1.9 1.9 0.0 1.4 26.3
Er 8.6 7.3 15.1 7.9 8.1
Er 5.3 5.3 0.0 3.8 28.3
Tm 1.3 1.1 15.4 1.2 7.7
Tm <0.01 <0.01 N.A. <0.01 N.A.
Yb 8.4 7.2 14.3 7.9 6.0
Yb 5.2 5.3 1.9 3.9 25.0
Lu 1.2 1.0 16.7 1.1 8.3
Lu <0.01 <0.01 N.A. <0.01 N.A.
Hf 6.4 9.9 54.7 10.5 64.1
Hf 5.5 5.6 1.8 5.0 9.1
Ta 10 13.5 35.0 12.8 28.0
Ta 5.0 5.5 10.0 3.6 28.0
W 18 26 44.4 18 0.0
W 14.6 8.9 39.0 7.8 46.6
Tl 0.9 0.9 0.0 0.9 0.0
Tl 1.9 1.8 5.3 1.7 10.5
Pb 80 65 18.8 78 2.5
Pb 43.8 42.9 2.1 43.3 1.1
Bi 3.5 2.6 25.7 3.0 14.3
Bi 1.6 1.6 0.0 1.2 25.0
Th 47 42.5 9.6 52.5 11.7
Th 18.8 18.6 1.1 16.7 11.2
U 17 11.9 30.0 12.8 24.7
U 7.7 7.7 0.0 4.4 42.9
72 R.N. Lieberman et al. / Fuel 184 (2016) 6980

2. Material and methods

2.1. Materials & characterization

Two types of FAs from the combustion of South African

During the fixation process, the FA is exposed to different con-
ditions, where it undergoes physical and chemical changes. Zeta
potential studies of these FA have shown that the surface charge
is changing, due to the exposure of the FA to different environmen-
tal conditions.
As shown by previous studies the FA investigated here has an
average particle size of 517 lm [28]. Two types of aqueous solu-
tion have been used for treatment of the FA: ultra pure deionized
water (UPDI), and 0.1 M hydrochloric acid (HCl), in order to simu-
late the neutral and acidic environments that the FA is exposed to
during the scrubbing process.
This study aims to determine the physical and chemical
changes occurring during the exposure of the FA to different aque-
ous/organic solutions. The objectives are: (i) to evaluate the fixa-
tion capacity of the FA for metal ions from waste, and (ii) to
study the changes occurring at particle surfaces upon contact with
neutral or acidic environments.
coal (denoted as SAFA) or Colombian coal (denoted as COFA) were
supplied by the Israeli Electricity Company (IEC) and were
used in this study. These coals contain ash (SA 13.9 wt%, and
CO 8.7 wt%), the density of SAFA is 0.98 g/cm3, and of COFA is
0.85 g/cm3.
The FA is composed mainly of two types of spherical
particles, cenospheres and plerospheres (Fig. 1A/B SAFA, and C/D
COFA). Both types are glass bubbles, however, cenospheres
are hollow and the plerospheres are filled with smaller glassy
particles [30]. Also some other particle types are present, in
much minor proportion such as spinel, magnetite, and unburnt
carbon.
The FAs are considered as a non-porous material, with a rela-
tively large surface area (2.41 0.01 m2/g for SAFA and
1.58 0.01 for COFA). The chemical content of the SAFA and COFA
that were used in this study is given in Tables 1 and 2 respectively.
Also, XRF analysis of both FAs is given is Table 3.
Hydrochloric acid (HCl), nitric acid (HNO3), and Treion deion-
ized water columns (UPDI) with a resistance >18 MOhms/cm were
used in this study. All other chemicals were of analytical grade
(AR), from Sigma Aldrich.

Table 3
XRF analysis of SAFA and COFA before and after treatment.

wt% SiO2 Al2O3 TiO2 Fe2O3 MgO MnO CaO Na2O K2O P2O5 LOI Total
SAFA 43.9 33.7 1.7 3.2 2.2 0.1 10.1 <0.4 0.9 2.7 1.2 99.6
SAFA UPDI 44.5 34.1 1.7 3.2 2.2 0.1 9.2 <0.4 0.9 2.6 1.1 99.6
SAFA HCl 48.3 37.2 1.9 3.5 1.2 0.1 4.5 <0.4 1.1 1.4 0.4 99.6
COFA 57.4 23.1 1.1 6.6 1.7 0.04 3.3 2.4 1.6 0.3 2.4 100.0
COFA UPDI 57.8 23.8 1.1 6.6 1.7 0.04 2.9 2.3 1.7 0.3 1.3 99.5
COFA HCl 61.1 22.8 1.1 6.3 1.3 0.04 1.4 2.4 1.8 0.2 1.6 100.0

Fig. 2. Scheme of the work procedure.

 R.N. Lieberman et al. / Fuel 184 (2016) 6980 73

2.2. Analytical methods Table 4

Results of the leaching test of SAFA anl COFA before anl after treatment.
A three-step procedure was used (Fig. 2) to treat SAFA and COFA mg/kg SAFA SAFA COFA COFA SAFA COFA Non
in order to study the effect of neutral and acidic environments (as UPDI HCl UPDI HCl hazarlous
described in Lieberman et al. [28]). Al 52.2 3.7 32.0 1165 7.2 285.5
The elemental composition was analyzed via means of Induc- Ca 672 643 203 2007 3442 2939
tively Coupled Plasma Atomic Emission Spectrometry (ICP-AES) Fe <0.01 <0.01 <0.01 77.0 <0.01 <0.01
K <0.01 14.7 <0.01 120 34.3 68.5
and Inductively Coupled Plasma Mass Spectrometry (ICP-MS).
Mg 1.2 86.8 42.0 415 <0.01 <0.01
XRF of PANalyticl model PerlX3 and ICP-MS analysis were used Na 8.4 60.7 35.3 337 57.0 1102
to determine the major compounds and trace elements in the FA, P <0.01 3.2 <0.01 12.4 <0.01 <0.01
especially Si, as the digestion carrier for ICP uses HF, which SO4 491 161 327 215 225 7680 20,000
destroys the Si layer. Si 62.9 31.2 11.8 58.3 15.8 <0.01
Sr 10.7 10.5 5.0 59.4 80.3 48.1
A two-step special digestion method for the analysis of poten-
Li 1.0 0.6 0.4 1.8 15.8 5.2
tially volatile elements in coals and FA by Querol et al. [31], was Be <0.01 <0.01 <0.01 0.2 <0.01 <0.01
used to dissolve the samples prior to analysis. The international B 6.7 1.5 9.9 18.8 0.3 116.1
reference material NBS1633b underwent the same treatment and Sc <0.01 <0.01 <0.01 0.1 <0.01 <0.01
Ti <0.01 <0.01 <0.01 8.8 <0.01 <0.01
analytical procedures to determine the accuracy of the methods.
V 0.7 0.7 2.2 1.4 <0.01 3.4
Mercury was analyzed using LECO AMA 254 Gold-Amalgam Cr 0.5 <0.01 0.4 0.3 1.0 4.4 10.0
Atomic-Absorption Spectrometer. The mineralogy of samples was Mn <0.01 2.3 <0.01 10.4 <0.01 <0.01
determined by X-ray powder diffraction (XRD) using a Bruker D8 Co <0.01 <0.01 <0.01 0.9 <0.01 <0.01
Advance diffractometer (monochromatic Cu Ka1,2 radiation; Ni <0.01 0.1 <0.01 2.4 <0.01 <0.01 10.0
Cu 0.3 0.1 0.1 0.5 <0.01 <0.01
k = 15,405 at 40 kV; and 40 mA). The primary parallel X-ray beam
Zn 0.3 8.5 0.1 4.5 <0.01 <0.01 50.0
was generated by a Gbbel mirror and the scattered beam was Ga 0.3 <0.01 0.1 0.2 0.2 1.5
analyzed by a Sol-X detector (scanning parameters: from 4 at 60 Ge <0.01 0.1 <0.01 0.3 <0.01 0.1
of 2h, a step size of 0.05, 3 s time per step). As 2.6 1.3 1.0 0.9 <0.01 0.2 2.0
Se 0.4 0.1 1.9 0.5 0.1 10.7 0.5
The particle composition and morphology of samples were
Rb <0.01 <0.01 <0.01 0.4 0.1 0.2
investigated using a MK2 Quanta 200 SEM with energy dispersive Y <0.01 <0.01 <0.01 3.9 <0.01 <0.01
X-ray analyzer (SEM-EDX). Grain size distribution of <63 lm frac- Mo 0.4 <0.01 0.8 0.3 2.0 15.3 10.0
tion was determined by a laser light-scattering-based particle Cl <0.01 <0.01 <0.01 1.2 <0.01 <0.01 1.0
sizer, MALVERN Hydro 2000MU, working range from 0.1 to Sb 0.3 0.2 0.4 0.1 <0.01 0.6 0.7
Ba 1.3 4.6 3.2 32.1 20.1 10.0 100
1000 lm.
La <0.01 <0.01 <0.01 2.1 <0.01 <0.01
The European Standard leaching test EN-12457 (Council deci- Ce <0.01 <0.01 <0.01 4.3 <0.01 <0.01
sion 2003/33/EC) was applied to treated and non-treated FA sam- Pr <0.01 <0.01 <0.01 0.6 <0.01 <0.01
ples in order to determine the leaching potential of major and trace Nl <0.01 <0.01 <0.01 2.1 <0.01 <0.01
elements using an overhead shaker, Heidolph model REAX 2. Sm <0.01 <0.01 <0.01 0.5 <0.01 <0.01
Eu <0.01 <0.01 <0.01 0.1 <0.01 <0.01
The pH and ionic conductivity were determined by conven- Gl <0.01 <0.01 <0.01 0.5 <0.01 <0.01
tional methods. Tb <0.01 <0.01 <0.01 0.1 <0.01 <0.01
Brunauer-Emmett-Teller (BET) model Micromeritics Gemini V Dy <0.01 <0.01 <0.01 0.6 <0.01 <0.01
surface area and pore size analyzer was used to study the change Ho <0.01 <0.01 <0.01 0.1 <0.01 <0.01
Er <0.01 <0.01 <0.01 0.4 <0.01 <0.01
of surface area of the fly ashes after treatments. The fly ashes were
Yb <0.01 <0.01 <0.01 0.3 <0.01 <0.01
heated to 50 C and bubbled (with N2 and He gases in order to W 0.8 <0.01 0.5 0.3 1.3 6.9
eliminate O2) for 2 h. During measurement the sample was Pb 0.7 0.1 <0.01 0.1 <0.01 <0.01 10.0
inserted into liquid N2. Bi 0.2 0.1 0.1 0.1 <0.01 <0.01

3. Results and discussion

Table 5
Particle size measurements before anl after treatments of the FA.
3.1. Mobility of major and trace elements during fly ash treatments
COFA Particle size (lm) SAFA Particle size (lm)
3.1.1. Chemical characterization of fly ash samples Before treatment 17.4 Before treatment 20.2
The FA samples in this study are the by-products of the combus- UPDI 16.2 UPDI 17.9
tion process of pulverized bituminous coals in the Israeli utilities. HCI 17.5 HCI 21.2

Previous studies [28], have shown that most of the alkaline groups
of the FA treated with HCl are neutralized, resulting in an acidic fly
ashes (pH  4.5). This is supported by the decrease of Ca content
after each treatment. While the UPDI treatment decreases the Ca
content by 2034% (Tables 1 and 2), the 0.1 M HCl treatment
indicates a higher Ca dissolution of 5862% as a result of the
acidic conditions. The results of the bulk chemical analysis of FA
after treatments (Tables 1 and 2) indicate that the dissolved frac-
tion was composed of Ca, P, and S after UPDI treatment. Acidic
environment results in an increased extraction of these com-
pounds as well as MgO content, which has very low solubility in
water but is soluble under acidic conditions [32]. Moreover, it is
clearly shown (Tables 1 and 2) that several trace elements have
high solubility [33,34]. Boron, V, Mn, As, Se, Ge, Mo, Bi, and U dis-
solve in considerably high percentages during the treatment
(2141% due to UPDI treatment, and 2186% due to HCl treatment,
respectively) [35,36]. There is a correlation between the leaching of
U, As, V, and Se to the decrease of P content in the treated FA.
Although, many toxic trace elements were washed during treat-
ment, Hg doesnt leach as a result of the treatment. Thus, the Hg
is trapped within the inner matrix of the FA. Furthermore, most
of the elements that condense as oxy-anionic species are much
more mobile in the alkaline pH than in the acidic one, except for
arsenate (after treatment) [34,37,38].
In order to assess the leaching ability of the treated FA, the
European Directive leaching test (EN 12457-2) was applied to trea-
74 R.N. Lieberman et al. / Fuel 184 (2016) 6980

Fig. 3. Mineral identification of SAFA and COFA treated with UPDI or 0.1 M HCl.
 R.N. Lieberman et al. / Fuel 184 (2016) 6980
Fig. 4. Surface transformations, due to exposure of the SAFA to different environments (acidic (79), and neutral (46)) as compared to the non-treated FA (13).
ted FA samples. The results were compared to leaching tests of the
non-treated FA. As it was observed in previous studies [16,39,40],
COFA has high leaching rate of Mo and Se. Although after treat-
ments of the FA, the leaching of trace elements is within the
non-hazardous limit, the leaching behavior for the COFA treated
with 0.1 M HCl is changed (Table 4). The leaching of Al and a num-
ber of trace elements (such as some Rare Earth Elements (REE), Se,
Cu and Cd) increase, probably as a result of unraveling of the top
layer of the FA and the weathering and partial dissolution of the
AlASi rich glassy fraction and to the increase of dissolution of iron
and oxyanions during the HCl treatment.
3.1.2. Fixation mechanism changes upon fly ash treatment
As previously reported, prior studies [15,16,39,40] indicated
that FA interact with trace elements and cations in solution or slur-
ries via three types of mechanisms:
(a) Cation exchange mechanism,
(b) Coordination bonding, and
(c) Electrostatic interaction of fine ionic precipitates with the FA
surface.
75
Zeta potential measurements [15] have shown that the FA
undergoes changes upon treatments with UPDI or HCl solutions.
Both in COFA and SAFA the zeta potential is positive in acidic
pHs. Also, both the treated and non-treated FA samples show that
the FA changes from the positive charge (acidic) to zero charge (at
neutral pHs) and to negative charge (at alkaline pHs). Hence, it is
deduced that the treated surfaces in UPDI and HCl reveal more
negative functional surface groups, which means that the concen-
tration of AAlO2 and ASiO3 at the FA surface increases. This is
supported by the above-demonstrated dissolution of CaO.
3.2. Physical changes of the fly ash
3.2.1. Particle size
Non-treated and treated FA were suspended in ethanol solution
(to avoid dissolution of specific salts) in order to study the size dis-
tribution of the FA samples. The results, Table 5, clearly indicate
that the particle size is changed, probably as a result of CaO disso-
lution occurring at the surface particles. Non treated SAFA and
COFA show a relatively fine particle size (1720 lm). Measure-
ments of the UPDI treated FA samples indicate that the dissolution
76 R.N. Lieberman et al. / Fuel 184 (2016) 6980
of CaO has decreased the particle size to 1618 lm. Also, the
0.1 M HCl treatment lowers the pH of the FA (1/10 Solid/Liq-
uid kg/L ratio in UPDI), due to increased dissolution, namely of sur-
face CaO and glassy material.
In order to check the validity of these observations, BET mea-
surements of the surface areas of the treated vs. the untreated FA
samples were carried out. The results are summed up in Table S1.
The results clearly support the particle size measurements,
namely the UPDI treated FA particle surface area have been
increased as the average size is decreased, due to dissolution of
the outer layer of CaO, and AlASi glassy matrix of the particles.
However, 0.1 M HCl treatment has dissolved appreciable number
of the small particles, and thus the average size of FA particles
measured has increased. Thus, the surface area of BET measure-
ments of the 0.1 M HCl particles has been reduced.
3.2.2. Mineralogy of untreated/treated fly ash
Glassy particles containing aluminosilicates matrix are the
major constituents of the FA (>70%) both for SAFA and COFA. A pre-
vious study [15] has shown via the XRD measurements that quartz
(SiO2) and mullite (3Al2O3A2SiO2) are the major crystalline phases
present in the SAFA and COFA, with traces of calcite (CaCO3), lime
(CaO), hematite (Fe2O3) and magnetite (Fe3O4). Although after
Fig. 5. Surface transformations, due to exposure of the COFA to different environments (acidic (79), and neutral (46)) as compared to the non-treated FA (13).
UPDI or 0.1 M HCl treatments these structures remain as the main
mineral phases, changes do occur (Fig. 3). The bulk chemical anal-
yses show that S and Mg content in SAFA are 100 and 26% higher
than in the COFA. Moreover, SAFA treatment with UPDI reveals for-
mation of gypsum (CaSO4), and dolomite (CaMg(CO3)2). On the
other hand, COFA is enriched with more than 100% Fe compared
to SAFA, and no changes of the mineral composition (quartz, mul-
lite, and magnetite) were detected after the UPDI treatment.
The mineral composition of the FA changed appreciably as a
result of the 0.1 M HCl acid treatment. The higher solubility of P
in acidic solution, in addition to the lower pH (4.5) of the SAFA
caused the formation of trace amounts of P bearing species: alu-
minum phosphate (AlPO4) and tetra-calcium diphosphide oxide
(Ca4P2O). Moreover, acidic treatment (pH  4) with the much
higher content of Na in the COFA (0.2 wt% in SAFA compared with
2.3 wt% in COFA), resulted in formation of halite (NaCl) crystals.
3.2.3. Surface transformations
The surface structures of the 6 samples of FA (non-treated and
UPDI or 0.1 M HCl treated SAFA and COFA) were investigated via
SEM, TEM, and EDX (Figs. 4 and 5). The results show that there
are several changes upon treatment in neutral or acidic environ-
ment. With the SAFA (Fig. 4) and COFA (Fig. 5) post UPDI treat-
 R.N. Lieberman et al. / Fuel 184 (2016) 6980 77

ment, it is clear that a large inside surface is exposed due to disso- turn yielded to the exposure of hematite, magnetite and the mul-
lution of free lime and other species from the surface of the FA par- lite skeleton embedded to the glassy matrix. Also, the SEM-EDX
ticles. The COFA measurements show that the inner part of the results of FA particle surface revealed a decrease in the total sur-
outer sphere is composed from a hematite structure, which is face composition of S, Mg, Na, and P. This significant decrease of
uncovered. Also, surfaces (Fig. 4, image 5) of the SAFA particles Mg after HCl treatment is a result of the much higher solubility
become rough because of the treatment. of MgO in acidic solutions. Furthermore, SEM-EDX analysis of
Furthermore, SAFA particles treated with UPDI show a glassy particles (Fig. 7) clearly indicates the increase of Ti content
formation of a unique structure (Fig. 4, images 4 and 6). Although after UPDI or HCl treatments. Also a decrease of Ca, S, Na (after
Ca is considered quite soluble in water, only 2034% of the content both treatments) and also P and Mg (after HCl treatment) content
is dissolved. These results indicate that some of the lime is in the SAFA is observed. COFA analysis shows similar behavior with
entrapped within the inner matrix of the glassy structure, and a high Ca, S, and Na dissolution from the surface, and also the
also that precipitation of Ca-hydrated phases such as a the increase of the residual Ti content. Phosphorus content decreases
formation of ettringite [Ca6Al2(SO4)3(OH)12
26H2O], or thuamasite in both the UPDI and HCl FA treatments due to dissolution of sol-
[CaSiO3
CaCO3
CaSO4
15H2O] occurs (as has been shown also in uble P compounds at the surface of the FA. As indicated above,
previous studies [28,39]). This formation is probably also, due to the Ca, Na, P, and S content is decreased significantly, but still an
the decrease in the pH (from 10.5 to 12.5 in the non-treated to appreciable amount is trapped within the inner matrix of the indi-
910 in the UPDI treated FA) or due to sulfate attack, favoring for- vidual FA treated particles as only 50, 19, 8, and 22% were
mation of mixed minerals, ettringite and thaumasite [39]. Also, the leached respectively.
occurrence of some gypsum crystals in the thaumasite is in line In order to have a better understanding of the surface transfor-
with the XRD results of the UPDI treated SAFA. The crystallization mations, TEM-EDX analyses were also carried out (Fig. 8) and these
of thaumasite as a result of sulfate attack requires several yielded quite similar results to those obtained by the SEM-EDX: a
conditions [39,41]: (1) a source of calcium silicate, (2) availability decrease of Ca, Mg, and S, and increase of the residual Ti, Si con-
of sulfates and carbonate, and (3) high humidity. All these tents at the surface of the FA particles.
conditions are found in the SAFA during the treatment with UPDI.
Furthermore, SEM-EDX analysis (Fig. S1) of the string-like struc-
ture shows that the major elements are: 20% Al, 18% Si, 45%,
Ca, and 14% S. Previous studies of the thaumasite formation
under the experimental conditions showed that the formation of
this mineral can be via neutral sulfates ions added to concrete
(source of S, which is found in high amount in the SAFA) [42,43].
Also, it was found [44] that at low temperature (0100 C) mixed
aggregate of ettringite/thaumasite/gypsum does occur. This is in
line with the results indicating formation of a mixture of these
two mineral compositions.
Post UPDI treated COFA changes are much lighter than in SAFA.
However, SEM-EDX results (Fig. S2) comparing the surface compo-
sition of the FA particles show a depleted (80%) Ca content (Fig. 5,
images 3 and 6) in the treated FA with respect to the original one.
SEM images of the acid treated FA samples are shown in Fig. 4
(images 79) and 5 (images 79). 0.1 M HCl treatment of the SAFA
resulted in a more aggressive surface dissolution, uncovering inner
layers and the destruction of the particle structure (Fig. 4, images 8
and 9). COFA treated with 0.1 M HCl followed the same patterns
(Fig. 5, image 7), and the formation of cubic halite. As also con-
firmed via the XRD results. Zeta potential analysis in a previous
study [15] indicated that after UPDI/HCl treatment the surface of
the FA becomes more positive, as a result of partial destruction
of the surface glassy aluminosilicate matrix.
SEM-EDX results (Figs. 6 and 7) helped to determine the mor-
phological and elemental composition occurred following the
water and acidic treatments.
With both SAFA and COFA the particle surface Si, P, S, Ca, Ti, Al,
Mg, and Na contents were modified, most probably due to partial
dissolution of the glassy matrix and of free lime and sulfate phases.
Heavier metals such as Ti, and Fe are segregated to the surface
at high temperature (during the formation of the FA particles), due
to the low solubility of Si, Al, Fe, and Ti melts and crystallization as
Ti, and Fe oxides (magnetite Fig. 5 image 9, and hematite Fig. 5
images 3 and 6). A small increase in the concentrations of Si, Al,
and Ti after UPDI or HCl treatment (which are buried underneath
the dissolved Ca layers in the FA [45]) is observed by the SEM-
EDX results. This is corroborated by the increase of the Si, Al, and
Ti concentrations determined in the digestion results, and may
be probably due to (i) dissolution of free lime and gypsum aggre-
gates from the surface of the FA particles and (ii) partial dissolution Fig. 6. Surface composition changes of the SAFA (A) and COFA (B), due to treatment
of the glassy matrix of the walls of spheres and cenospheres, that in in neutral and acidic environments.
78 R.N. Lieberman et al. / Fuel 184 (2016) 6980
Fig. 7. Average surface composition of glassy particles in the SAFA ash (A), and COFA (B).
TEM images of the non-treated/treated FA are given in Fig. S3.
Image 2 in Fig. S3 clearly shows that acicular particles typically
of mullite shown in SEM are also detected pointing to the dissolu-
tion of the glassy matrix that embedded the mullite skeleton of the
walls of the particles.
4. Conclusions
The physical and chemical changes that occur in particles of
Israeli class F fly ashes (SAFA and COFA) upon exposure to neutral
or acidic (0.1 M HCl) environments have been investigated and cor-
related to the reported high fixation efficiency for elements when
applied as scrubber of trace elements pollutants from hazardous
waste liquids.
The results clearly demonstrated that indeed the FA particle
surface and external structure are changed. Treatment via UPDI
or HCl causes dissolution of the outer layer of lime (CaO), thus,
exposing large number of the anionic species (AAlO2 and ASiO3 )
at the surface of the particles. Also, the high AlASi glassy matrix
of the cenospheres is partially dissolved and exposes the mullite
internal skeleton of cenospheres walls (Fig. 5 images 3 and 6).
These results show high potential efficiency for fixation of trace
elements and fine precipitates by three modes: cation exchange,
coordination bonding, and electrostatic interaction with ionic fine
precipitates. Furthermore, the average particle size after the UPDI
treatment is decreased, whereas because of the dissolution of top
CaO layers, many small FA particles are destroyed during 0.1 M
HCl treatment resulting in increase of the average particle size.
Concerning the chemical properties of the treated FA sample
particle surface, there is an increase of the Si, Al, Ti, and Fe surface
content, which is attributed to the low solubility of these elements
due to their partial occurrence in very resistant mineral phases.
Furthermore, SEM and TEM analyses evidenced the formation of
fiber-like species upon the water treatment, and also an increase of
the Ca, S, Si, Al contents of fiber particles, pointing to the formation
of Thaumasite/Ettringite/gypsum mixture as deduced by the TEM.
Also, cubic halite particles are formed as a result of the high Na and
Cl content of the COFA during the 0.1 M HCl treatment.
The results of this study can contribute appreciably to the
understanding of the chemical, and electrostatic interactions
between the FA particles and ions during fixation processes.
Acknowledgements
This project was supported by the Israeli Ministry of Environ-
mental Preservation, the Israeli Coal Ash Authority and the Gener-
alitat de Catalunya (AGAUR 2015 SGR33). We would like to thank
the team of Analytical Ctr. in the IDAEA/CSIC institute in Barcelona
for the ICP-MS, ICP-OES, and XRD analysis. Also the Department
Chemical Sciences of Ariel University and the IDAEA&CSIC insti-
tutes are acknowledged for providing the laboratory space for
the studies carried out.
 R.N. Lieberman et al. / Fuel 184 (2016) 6980
Fig. 8. Chemical changes in the surface of FA particle deduced from TEM-EDX of SAFA and COFA upon treatment.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.fuel.2016.06.107.
References
[1] The National Coal Ash Board. General infoAvailable from: <http://www.coal-
ash.co.il/english/info.html>2003 [Accessed 10.01.16].
[2] The National Coal Supply Corporation LTD. <http://ncsc.co.il;2014> [Accessed
10.01.16].
[3] The Israel Electric Corp.; 2014. <http://www.iec.co.il> [Accessed 10.01.16].
[4] Mimran A. Comparison of inorganic parameters of different coals and the
combustion products in hadera utility. A.D.M.; 1997 [in hebrew].
[5] Lederman E. Coal fly ash as a chemical reagent for scrubbing industrial acidic
waste. In: Chemistry. Ben Gurion University; 2008.
[6] Israel Ministry of Environmental Protection. Abatement of nuisances
regulations (air quality), 5752-1992. In: Israel ministry of environmental
protection. p. 5.
[7] Wesche K. Fly ash in concrete: properties and performance. CRC Press; 2004.
[8] Ahmaruzzaman M. A review on the utilization of fly ash. Prog Energy Combust
Sci 2010;36:32763.
[9] Arivazhagan K, Ravichandran M, Dube S, Mathur V, Khandakar R,
Yagnanarayana K, et al. Effect of coal fly ash on agricultural crops: showcase
project on use of fly ash in agriculture in and around thermal power station
areas of National Thermal Power Corporation Ltd., India. In: World coal ash
conference May 2011. p. 912.
[10] Cohen H. Fly ash: a potential excellent scrubber for acidic wastes in Israel. In:
International ash utilization symposium, center for applied energy
research. University of Kentucky; 2003.
79
[11] European Coal Combustion Products Association e.V. (ECOBA). What are CCPs,
utilisationAvailable from: <http://www.ecoba.com/ecobaccputil.html>2014
[Accessed 10.01.16].
[12] Fatih T, Umit A. Utilization of fly ash in manufacturing of building bricks. In:
International ash utilization symposium. Center for applied energy
research. University of Kentucky Paper; 2001.
[13] Feuerborn H-J. Coal ash utilisation over the world and in Europe. In: Workshop
on environmental and health aspects of coal ash utilization. p. 15.
[14] Joshi RC, Lohita R. Fly ash in concrete: production, properties and uses. 1st
ed. Amsterdam CRC Press; 1997.
[15] Lieberman RN, Green U, Segev G, Polat M, Mastai Y, Cohen H. Coal fly ash as a
potential fixation reagent for radioactive wastes. Fuel 2015;153:43744.
[16] Lieberman RN, Lederman E, Pelly I, Cohen H. Fly ash as a potential scrubber of
acidic wastes from the phosphate industries in Israel. Coal Combust Gasif Prod
2014;6:715.
[17] Moreno N, Querol X, Ayora C, Pereira CF, Janssen-Jurkovicov M. Utilization of
zeolites synthesized from coal fly ash for the purification of acid mine waters.
Environ Sci Technol 2001;35:352634.
[18] Prez-Lpez R, Nieto JM, de Almodvar GR. Utilization of fly ash to improve the
quality of the acid mine drainage generated by oxidation of a sulphide-rich
mining waste: column experiments. Chemosphere 2007;67:163746.
[19] Rios CA, Williams CD, Roberts CL. Removal of heavy metals from acid mine
drainage (AMD) using coal fly ash, natural clinker and synthetic zeolites. J
Hazard Mater 2008;156:2335.
[20] Sheng G, Zhai J, Li Q, Li F. Utilization of fly ash coming from a CFBC boiler co-
firing coal and petroleum coke in Portland cement. Fuel 2007;86:262531.
[21] US Environmental Protection Agency (EPA). Toxcitiy Characteristic Leaching
Procedure (TCLP) method 1311; 1992.
[22] California Department of Toxic substances Control. California waste extraction
test (CAL-WET): environmental health standards hazardous waste,
66261.126 appendix II; 2005.
[23] European Committee for Standardization. Compliance test for leaching of
granular waste materials and sludges (EN12457-2); 2003.
80 R.N. Lieberman et al. / Fuel 184 (2016) 6980
[24] Minsitry of Health. Drinking water regulationsAvailable from: <http://www.
health.gov.il/Subjects/Environmental_Health/drinking_water/Documents/
Briut47-Eng.pdf>2014 [Accessed 10.01.16].
[25] Bayat B. Combined removal of zinc (II) and cadmium (II) from aqueous
solutions by adsorption onto high-calcium Turkish fly ash. Water Air Soil
Pollut 2002;136:6992.
[26] Hequet V, Ricou P, Lecuyer I, Le Cloirec P. Removal of Cu2+ and Zn2+ in aqueous
solutions by sorption onto mixed fly ash. Fuel 2001;80:8516.
[27] Kanungo S, Mohapatra R. Leaching behavior of various trace metals in aqueous
medium from two fly ash samples. J Environ Qual 2000;29:18896.
[28] Lieberman RN, Teutsch N, Cohen H. Chemical and surface transformations of
bituminous coal fly ash used in Israel following treatments with acidic and
neutral aqueous solutions. Energy Fuels 2014;28:465765.
[29] Wang S, Viraraghavan T. Wastewater sludge conditioning by fly ash. Waste
Manage 1998;17:44350.
[30] Grossman SL, Mathews A. The mineralogy and chemistry of coal ash generated
by the Hadera M.D. power station. J Coal Qual 1988;7:226.
[31] Querol X, Fernndez-Turiel J, Lpez-Soler A. Trace elements in coal and their
behaviour during combustion in a large power station. Fuel 1995;74:33143.
[32] Dean JA. Langes handbook of chemistry. 16th ed. New York: McGraw-Hill;
2005.
[33] Izquierdo M, Querol X. Leaching behaviour of elements from coal combustion
fly ash: an overview. Int J Coal Geol 2012;94:5466.
[34] Kim AG, Kazonich G, Dahlberg M. Relative solubility of cations in class F fly
ash. Environ Sci Technol 2003;37:450711.
[35] Dreesen DR, Gladney ES, Owens JW, Perkins BL, Wienke CL, Wangen LE.
Comparison of levels of trace elements extracted from fly ash and levels found
in effluent waters from a coal-fired power plant. Environ Sci Technol
1977;11:10179.
[36] Jankowski J, Ward CR, French D, Groves S. Mobility of trace elements from
selected Australian fly ashes and its potential impact on aquatic ecosystems.
Fuel 2006;85:24356.
[37] Izquierdo M, Font O, Moreno N, Querol X, Huggins FE, Alvarez E, et al. Influence
of a modification of the petcoke/coal ratio on the leachability of fly ash and
slag produced from a large PCC power plant. Environ Sci Technol
2007;41:53305.
[38] Izquierdo M, Moreno N, Font O, Querol X, Alvarez E, Antenucci D, et al.
Influence of the co-firing on the leaching of trace pollutants from coal fly ash.
Fuel 2008;87:195866.
[39] Lieberman RN, Anker Y, Font O, Querol X, Mastai Y, Knop Y, et al. Potential of
hazardous waste encapsulation in concrete compound combination with coal
ash and quarry fine additives. Environ Sci Technol 2015;49:1414655.
[40] Lieberman RN, Cohen H, Font O, Querol X. Morphology changes in bituminous
coal fly ash via aqueous solutions treatments (acidic and neutral). Geological
Society of America 125th, paper no. 1; 2013.
[41] Myneni SC, Traina SJ, Logan TJ, Waychunas GA. Oxyanion behavior in alkaline
environments: sorption and desorption of arsenate in ettringite. Environ Sci
Technol 1997;31:17618.
[42] Crammond N, Halliwell M. The thaumasite form of sulfate attack in concretes
containing a source of carbonate ions a microstructural overview. ACI Spec
Publ 1995;154.
[43] Oberholster R, Van Aardt J, Brandt M. Durability of cementitious systems.
Bames, P. Structure and performance of cements. London; 1983. p. 365.
[44] Hewlett P. Leas chemistry of cement and concrete. 4th ed. Oxford
Butterworth-Heinemann; 2004.
[45] Font O. Trace elements in integrated gasification combined cycle fly ash:
extraction of potentially valuable elements. Universitat POlitecnica de
Catalunya; 2007.