Coal Combustion and Gasification Products is an international, peer-reviewed on-line journal that provides free access to full-text papers, research communications and supplementary data. Submission details and contact information are available at the web site.

© 2011 The University of Kentucky Center for Applied Energy Research and the American Coal Ash Association

Web: www.coalcgp-journal.org

ISSN# 1946-0198

Volume# 2 (2010)

Editor-in-chief: Dr. Jim Hower, University of Kentucky Center for Applied Energy Research CCGP Journal is collaboratively published by the University of Kentucky Center for Applied Energy Research (UK CAER) and the American Coal Ash Association (ACAA). All rights reserved.

The electronic PDF version of this paper is the official archival record for the CCGP journal.

The PDF version of the paper may be printed, photocopied, and/or archived for educational, personal, and/or non-commercial use. Any attempt to circumvent the PDF security is prohibited. Written prior consent must be obtained to use any portion of the paper’s content in other publications, databases, websites, online archives, or similar uses.

Suggested Citation format for this article:

Silva, L., DaBoit, K., Serra, C., Mardon, S.M., Hower, J.C., 2010, Fullerenes and Metallofullerenes in Coal-Fired Stoker Fly Ash. Coal Combustion and Gasification Products 2, 66-79, doi: 10.4177/CCGP-D-10-00007.1

I S S N 1 9 4 6 - 0 1 9 8

j o u r n a l h o m e p a g e : w w w . c o a l c g p - j o u r n a l . o r g

Fullerenes and Metallofullerenes in Coal-Fired Stoker Fly Ash

Luis F.O. Silva1, Katia DaBoit1, Carmen Serra2, Sarah M. Mardon3, James C. Hower4

1 Catarinense Institute of Environmental Research and Human Development – IPADHC, Capivari de Baixo, Santa Catarina, Brazil2 Servicio de Nanotecnologıa y Analisis de Superficies C.A.C.T.I., Universidade de Vigo, Vigo, Spain3 Kentucky Department for Natural Resources, Division of Abandoned Mine Lands, 2521 Lawrenceburg Road, Frankfort, KY 40601, USA4 University of Kentucky, Center for Applied Energy Research, 2540 Research Park Drive, Lexington, KY 40511, USA

A B S T R A C T

A suite of high-As, high-C fly ashes from a university-based stoker-fired coal boiler were analyzed by a number of techniques,

including high-resolution transmission electron microscopy (HR-TEM), time-of-flight secondary ion mass spectrometry (TOF-

SIMS), X-ray photoelectron spectroscopy (XPS), and field-emission scanning electron microscopy (FE-SEM). The sooty carbon

is in the form of nano balls with the major fullerenes at C60+, C70

+, and C80+, with species at C2 increments from C56

+ to C78+.

Arsenic and Hg, among other metals, are found in association with the fullerenes, but, with our techniques, it is not possible to

determine if the metals are encapsulated by the fullerenes or attached to the side of the structure. TOF-SIMS studies suggest an

association of As with the Al-Si glass; an association of Pb with oxides, sulfates, and carbon; Hg with carbon; Se in elemental

form with carbon; and Cr in a variety of forms, including nano carbons, Fe sulfates and oxides, glass, and Cr-oxyhydroxides.

f 2010 The University of Kentucky Center for Applied Energy Research and the American Coal Ash Association

All rights reserved.

A R T I C L E I N F O

Article history: Received 24 September 2010; Received in revised form 1 January 2011; Accepted 14 January 2011

Keywords: fullerene; mercury; arsenic; selenium; coal combustion; fly ash

1. Introduction

The presence of fullerenes has been of great interest since their

discovery in natural systems (Kroto et al., 1985), in part because of

their characteristic structure, described as a cage by Dunsch et al.

(2010) and nicholls et al. (2010), that can retain some potentially

hazardous elements (Utsonomiya et al., 2002; Bai et al., 2003;

Moses et al., 2003; Hower et al., 2008; Dunsch et al., 2010; Nicholls

et al., 2010). Consequently, the questions of natural occurrence of

fullerenes may have substantial environmental and human health

implications should the future for multi-walled nanotubes

(MWNTs) grow as predicted.

The chemical composition of coal-combustion fly ash varies

with the chemical composition of the coal being fired and the

combustion technology (Wang, 2008). In general, coal-combustion

fly ash is comprised of small solid particles, small hollow particles

(cenospheres), and thin-walled hollow spheres (plerospheres)

containing both cenospheres and solid particles (Goodarzi and

Sanei, 2009). Upon combustion in a coal-fired power plant, most

volatile elements in coal, such as zinc and arsenic, will be released

into the flue gas and subsequently captured by the fly ash, with the

concentration of the element on the fly ash increasing as a

function of both the decreasing flue gas temperature and the

decreasing fly ash particle size at the point of capture (Mardon and

Hower, 2004; Hower et al. 2010). In contrast, the capture of

mercury, while also highly dependent upon the flue gas

temperature, is also a function of the amount and form of fly

ash carbon and on the chemistry of the flue gas (summary by

Hower et al., 2010). Selenium, to date, has evaded characterization

into either of the latter categories (Hower et al., 2009).

In this study, we are conducting a detailed study of the fine

structure and the associated chemistry of carbons in the baghouse-

collected fly ash from a coal-fired stoker steam plant. The

petrology and bulk chemistry of the fly ashes had been previously

discussed by Mardon et al. (2008).* Corresponding author: Tel.: 1-859-257-0261. E-mail: hower@caer.uky.edu

doi: 10.4177/CCGP-D-10-00007.1

f 2010 The University of Kentucky Center for Applied Energy Research and the American Coal Ash Association. All rights reserved.

Fig. 1. Examples of the most abundant glassy aluminosilicate spherical particles (FE-SEM and HR-TEM images).

Silva et al. / Coal Combustion and Gasification Products 2 (2010) 67

2. Methods

2.1 Sample Collection

Selected fly ashes were collected from the baghouse hoppers of a

university-based coal-fired stoker boiler (Mardon et al., 2008). The

coal source was a medium-S, Pennsylvanian-age, eastern Kentucky

high volatile A bituminous coal. The whole fly ash, sample 93261,

and fractions wet screened at 100, 200, 325, and 500 mesh were

available for this study.

2.2 X-Ray Diffraction

The mineral composition of coal fly ashes (CFA) was determined

using a Siemens model D5005 X-ray diffraction. The samples were

ground by hand in a ceramic mortar and pestle, dry mounted in

aluminum holders, and scanned at 8–60u 2h with Cu K-a radiation.

2.3 Electron Beam Methods

Electron beam methods included Field Emission Scanning

Electron Microscope (FE-SEM) with energy-dispersive X-ray

spectrometer (EDS) capabilities and high-resolution transmission

electron microscope (HR-TEM) with SAED (selected area electron

diffraction) or MBD (microbeam diffraction), and scanning

transmission electron microscopy (STEM). Time of flight secondary

ion mass spectrometry (TOF-SIMS) and Fast Fourier transformation

(FFT) were used to investigate the elemental and molecular

structure of the samples. Surface composition was determined by

X-ray photoelectron spectroscopy (XPS). Procedures for the

electron beam methods are outlined in Appendix A.

2.4 Mineral Extractions

As-received coal fly ashes (CFAs) were also analysed via XRD,

FE-SEM, and HR-TEM. For appropriate geochemical interpretation,

it is crucial to combine the application of XRD, FE-SEM, and HR-

TEM mineralogical analyses of CFAs with sequential extraction

(SE). This provides an improved understanding of the retention

behavior of mobile elements by specific carbonaceous particles,

primary minerals, and secondary soluble minerals. A brief

explanation of the selected experimental conditions and of the

expected information offered by each step of the sequential

extraction are presented in Appendix A.

3. Results and Discussion

Coal-derived fly ash is primarily comprised of ,10-nm- to

.100-mm-diameter glassy aluminosilicate spheres (Figure 1). The

combustion temperature of the coal was in excess of 1300 uC,

slightly lower than pulverized-coal combustion boilers. In the

present work, we detected different structures of metal-agglomer-

ate-bearing carbons, including carbon fibers, graphitic particles,

fullerenes, MWNTs, and amorphous carbon.

3.1 Petrology and X-Ray Diffraction Mineralogy

Details of the petrology of baghouse fly ash sample 93261 and

the wet-screened size fractions from the ash were published by

Mardon et al. (2008). As in most fly ashes, glass and amorphous

carbon comprise the non-crystalline portion of the sample, with

glass comprising the majority of the fly ash. X-ray diffraction

revealed the presence of a number of mineral phases, including

quartz, mullite, maghemite (spinel in Hower et al.’s (1995)

nomenclature), magnesioferrite, wollastonite, microcline, and

albite, not all detected in each fraction. The latter three minerals,

in particular, were rarer than the other minerals.

3.2 Electron Beam Results

HR-TEM/EDS/SAED/MBD and Fast Fourier transformation

(FFT) images show nano particles, some containing hazardous

Fig. 2. Selected HR-TEM image of Al-Si-carbonaceous spheres and marked zones contain encapsulated hazardous volatile elements by carbonaceous phases.

68 Silva et al. / Coal Combustion and Gasification Products 2 (2010)

Fig. 3. Left top - HR-TEM image of CNTs encapsulating fullerenes and Hg; Right top – Basic nanotube structure; Below - Illustration of our ideas related to formation

and metals interaction in CNTs and fullerenes. Nanotubes can encase a C60 sphere which, in turn, can be associated with trace metals. On the right, two possible

mechanisms for trace element association are illustrated. At the scale of this study, it is not possible to resolve the nature of the association.

Silva et al. / Coal Combustion and Gasification Products 2 (2010) 69

elements (Figures 2–5). Most carbon species were found as 10–

200nm carbon nano balls (Figure 2A and 2B), corresponding to

soot or black carbon with concentrically stacked graphitic layers

(as also noted by Chen et al., 2006; Nejar et al., 2007). The fractal-

arranged aggregates vary in size and exhibit a microtexture

consisting of roughly concentrically stacked graphitic layers

(Figure 2B). This is a form of amorphous carbon that gives fly ash

characteristics of a dark and wet powder and is a major

component of smoke from the combustion of carbon-rich organic

fuels in the absence of sufficient oxygen (Chen et al., 2006). Many

hazardous-element/carbon-particle aggregates (containing, for

example, Br, Mo, Se; Figure 2A) form via a vaporization–

condensation mechanism during coal combustion and the cooling

of the flue gas and fly ash in the pollution-control devices. Their

presence in coal fly ash samples and the carbon content detected

by ultimate analysis and HR-TEM/EDS suggests incomplete coal

combustion.

Other carbon structures were observed and characterized as

onion-structured or related concentric, multishell carbon nanopo-

lyhedra (Murr et al., 2005), reminiscent of high-temperature

treatments of soot catalyzed by metals that transform the typical

amorphous soot aggregates to more regular fullerenes (Cobo et al.,

2009). In this process, the precursor fragments are curved around

the metal catalyst, but flat-sheet fragments also form an aggregate

into nanospherules or spherule aggregates.

HR-TEM analysis provides evidence of weak metal/support

interactions, with carbon nanotube (CNTs) tip growth visible

(Figure 3). CNT tip-growth is suggested to occur when a low-strength interaction exists between the metal and the support.

Similar environments were discussed by several studies (Baker,

1989; Kuznetsov et al., 2003; Du et al. 2007; Hower et al. 2008).

While suggested as a possible model, it cannot be known at this

scale of analysis whether the metals are bound within the fullerene

balls or bound to the side of the structures.

The encapsulation of potentially hazardous elements such As,

Cd, Co, Cr, Hg, and Ni inside MWNTs are of considerable

importance in environmental science. A large number of nanotube

ends are visible in the coal fly ashes (particularly after Step 4 from

Appendix A), including both terminations at hazardous metal

particles and metallofullerenes, offering an opportunity to examine

dimensionally confined systems. Encapsulated hazardous ele-

ments (e.g. Hg, Figure 3) are also likely to contribute to element

retention in power plant fly ashes, resisting degradation due to

their surrounding protective carbon shells (Sun et al., 2002).

Studies have been made of the encapsulation of iron-group metals

within the carbon shells, with fewer investigations of the transition

metals (Byeon and Kim, 2010). In the carbon-encapsulated-

hazardous-element nanoparticles (CEHENs) studied here, several

nanometer-thick carbon shells provide chemical stability and

protection from the agglomerates of element nanoparticles (e.g

Figure 4), while the nanosized core metal particles can provide

specific functions that are unavailable in bulk form.

3.3 TOF-SIMS Characterization of Fullerenes

Positive-ion TOF-SIMS analysis, maintaining a primary ion dose

less than 1012 ions/cm2 to ensure the static SIMS conditions, results

in minimum surface destruction. The positive spectrum was

calibrated to H+, H2+, H3+, C+, CH+, CH2+, CH3+, C2H3+, C3H5+,

C4H7+, C5H9+ and C6H11+. The negative TOF-SIMS spectrum was

not considered in this study due its minimal information content

and the absence of molecular fullerene peaks.

The TOF-SIMS peaks between the mass 650 amu and 970 amu,

the region where fullerene ions should appear, is shown on

Fig. 4. Carbon-encapsulated hazardous-element nanoparticles (CEHENs) contain fullerenes and amorphous As-Pb-Se-Br-Si-O-carbonaceous nanoparticle.

70 Silva et al. / Coal Combustion and Gasification Products 2 (2010)

Fig. 5. (A) As-bearing jarosite following water extraction (fly ash1003500). STEM image and mapping to As (in red) and Fe (in green); (B) Jarosite crystals contain

mixed As-O-Pb amorphous; (C) As-bearing carbonaceous material; (D) Spherical Al-Si-Pb particle and carbonaceous matrix with Pb-oxide nanoparticle.

Silva et al. / Coal Combustion and Gasification Products 2 (2010) 71

Figure 6. In this region, we have detected the fullerene peaks

corresponding to C60+ at 720 amu, C70

+ at 840 amu, and C80+ at

960 amu but we have also detected intermediate fragments

which correspond to the consecutive loss of two units of carbon

(C2 loss) from each of the principal fullerene molecules. Therefore,

from C60+ and due to carbon depletion phenomena, we have

detected C58+ and C56

+. From C70+ we have also detected C68

+, C66+,

C64+, and C62

+ and from C80+ we have detected C78

+, C76+, C74

+, and

C72+. This phenomenon of loss or depletion of two carbons in

fullerenes has been reported by Foltin et al. (1993) and Kato et al.

(2008).

3.4 Volatile Element Behavior

Hazardous-element-bearing anthropogenic nanoparticles and

secondary nanominerals are attracting attention due to their

unique role as agents of elemental transport and their enhanced

reactivity in thermal systems (Hochella et al., 2008; Hower et al.,

2008; Graham et al., 2008; Silva et al., 2009; Silva and DaBoit,

2010). The behavior of nanoparticles and nanominerals in

anthropogenic systems within coal-fired power plants is still not

well understood due to a lack of experimental evidence about their

physical-chemical properties. While a decrease in particle size to

the nanoscale level can promote phase instability due to the

increase in surface energy (Reich et al., 2006), information is

needed on how nanoscale effects influence the occurrence of

nanoparticles at high-temperature conditions.

In general, the combustion of coal results in a redistribution of

the solid byproducts into approximately 20–25% bottom ash, 75–

80% fly ash (Stultz and Kitto, 2005), and ,0.5% fine particles

emitted from the stack (Goodarzi, 2006; Goodarzi et al., 2008). In

oxidizing environments in coal-fired power plants, metals tend to

be converted into (or trapped by) less-volatile compounds, such as

oxides, sulfates, amorphous aluminosilicates, and carbonaceous

phases. The volatile trace element behavior of As, Hg, Se, Pb, and

Cr were observed mainly in individual particles (usually occurring

as individual ,100 nm-size particles) and will be discussed below

with respect to their fate in the environment and impact on health.

3.4.1 Arsenic

Arsenic exposure in humans is associated with increased risks of

leukemia and skin, lung, liver, breast, bladder, and bone cancers

(Lyman et al., 1985), and is the result of chronic ingestion or

chronic inhalation. The dose-response curve is dependent on

location, sources, and population susceptibility and/or tolerance

(Fergusson, 1990; Ding et al., 2001; Mandal and Suzuki, 2002).

Coal-fired power plants are a significant anthropogenic source of

fine- and nano-As particles (Figures 4, 5 and 7), and power station

emissions have received considerable attention as they are

Fig. 6. TOF-SIMS spectra of C60, C70, and C80 fullerenes and their fragment molecules.

Fig. 7. XPS spectra of the binding energy of As at the fly ash surface.

72 Silva et al. / Coal Combustion and Gasification Products 2 (2010)

associated with a number of types of fine-particle pollution which

may contribute to health effects. However, As-CFA interactions are

complex.

HR-TEM and FE-SEM investigations demonstrate that As-CFA is

associated with amorphous nanoparticles (Figure 4), sulfates (e.g.

jarosite, Figure 5), CEHENs-bearing fullerenes, Fe, and Al-oxide

deposits on the particles of CFA. However, it was difficult to

estimate the total number of nanoparticles due to their short

lifetime under the HR-TEM and FE-SEM beam, particularly for

hydrated sulfate and nitrate salts species. The combination of FE-

SEM/EDS, STEM/EDS, hazardous/volatile elements mapping, and

HR-TEM/SAED/FFT was important in investigations of nanopar-

ticles and nanominerals in CFAs samples (e.g. Figure 5A of As-

bearing jarosite). In addition, the high resolution spectrum of As3d

obtained by XPS showed the presence of As at the sample surface.

The position in binding energy of the peak As3d5/2 at 45.85 eV can

be attributed to As2O5 - As(V) (Figure 7).

The surface chemical maps obtained by TOF-SIMS correspond-

ing to As and Al distribution are shown in Figure 8. The aluminum

distribution is ubiquitous across the surface, but is most intensive

in the center of the image. The arsenic (AsO) signal is detected in

much lower intensity than the Al but its distribution shows

concordance with the Al distribution, suggesting that the As

detected is on the surface of the Al-Si glassy fly ash particles.

3.4.2 Lead

Human exposure to Pb occurs through various intermediate

routes, but there are four main pathways of Pb from the

environment to humans: inhaled aerosol particles, dusts, food,

and drinking water (US EPA, 2008). High concentrations of Pb

cause serious problems in the central nervous system, peripheral

nervous system, and the vascular system of humans (Mejıa et al.,

1997). Both PbO and Pb-arsenate are listed in the Hazardous

Substance Data Bank (NLM, 2008). In addition, PbO is classified as

a potential carcinogenic compound, and Pb-arsenates are known to

be carcinogenic. Pb sulfate is listed in the Toxicology Data Network

(NLM, 2008). In general, Pb sulfate appears to be more toxic than

elemental Pb upon ingestion; overall, Pb compounds are more

toxic when inhaled than when administered via other routes.

In the present study, the Pb phases identified by FE-SEM, HR-

TEM, EDS, FFT, SAED, and MBD include Pb-oxide [e.g. massicot

(PbO), Figure 5D], Pb-sulfate (e.g anglesite as isolated particles

,50–300nm in length), Pb encapsulated in carbonaceous matter

(e.g. Figure 5D), abundant amorphous minerals, and a Pb-As phase

(in a ,25-nm particle). Almost all of the identified Pb-phases are

in the respirable particle matter category. Our results emphasize the

need for detailed single-nanoparticle characterization of anthro-

pogenic nanoparticle and secondary nanominerals emitted from

coal-fired power stations.

Fig. 8. TOF-SIMS image of overlapping Al and AsO (in blue) showing the concordance between both signals.

Fig. 9. XPS spectra of the binding energy of Pb at the fly ash surface.

Silva et al. / Coal Combustion and Gasification Products 2 (2010) 73

The high resolution spectrum of Pb 4f obtained by XPS showed

the presence of Pb at the sample surface (Figure 9). The position in

binding energy of the Pb4f7/2 at 139.9 eV can be attributed to

PbSO422 - Pb (II), in excellent concordance with anglesite phase

detected by FE-SEM, HR-TEM. TOF-SIMS results also demonstrate

the presence of Pb in the fly ash samples. Lead was detected in

elemental form in conformity with the expected isotopic distribu-

tion for this element (Figure 10).

3.4.3 Mercury and Selenium

Coal power plants are the largest point sources of Hg release to

the atmosphere. The high Hg and Se concentration in the CFA

Fig. 10. TOF-SIMS spectra of Pb isotopes in fly ash.

Fig. 11. Amorphous Al-Cr-Fe-Mg-Si-Ti spherical nanoparticles.

74 Silva et al. / Coal Combustion and Gasification Products 2 (2010)

investigated here was reported by Mardon et al. (2008). There were

some trends among groups of samples, especially between carbon

and the Hg captured by the mechanical ashes, but there was wide

scatter in the data for some of the ashes (Mardon et al., 2008).

The relationship between carbon nanoparticles, Se (Figure 2A),

and Hg (e.g. Figure 3) suggests that there were other (metalloful-

lerene, elemental metals, and metal-bearing organic compounds)

factors involved in the relationship between carbon nanoparticles

and these elements. FE-SEM/EDS and HR-TEM/EDS results indicate

that the carbon (e.g. fullerene, amorphous carbon, and MWNTs) in

CFA is responsible for capturing some of the Hg and Se, suggesting

higher carbon content would yield a higher Hg and, perhaps, Se

content in the CFA. The relationship between fly ash carbon and

Hg capture is well established for pulverized-fuel (pf) fly ashes, in

general (Hower et al., 2010), and for fullerenes in pf ashes (Hower

et al., 2008). For the stoker fly ash, a possible explanation for this

might be that the lower combustion temperature in the boiler

(compared to pulverized fuel boilers) may benefit Hg and Se

condensation/absorption on, and inclusion within, the nanocarbon

particles, thus influencing element trace capture and/or encapsu-

lation.Fig. 12. XPS spectra of the binding energy of Cr at the fly ash surface.

Fig. 13. A/ TOF-SIMS images of Al, Fe, and Cr distribution on fly ash. B/ Overlap of Al and Fe TOF-SIMS images. C/ Overlap of Al and Cr TOF-SIMS images.

Silva et al. / Coal Combustion and Gasification Products 2 (2010) 75

Despite the low Se-concentrations detected by EDS, nanoparti-

cles of amorphous elemental Se have been observed in the CFA by

SAED, MBD, and FFT. In general, in our FE-SEM and HR-TEM

images, Se appears in dark, rounded nanoparticles (e.g. Figure 2A

and Figure 4). EDS spectra of Se-rich regions do not display an

oxygen peak. In general, these particles consist of Se (93.8 wt%),

with minor Al (2.2wt%), Si (2.7wt%), and S (1.3wt%). Based on

these observations and the Se reaction path, it is reasonable to

speculate that the Se-particles observed in our CFA consist of

elemental Se encapsulated by carbonaceous nanoparticles.

3.4.4 Chromium

In the present study, the Cr phases identified by FE-SEM, HR-

TEM, EDS, FFT, SAED, and MBD include Cr-oxide (e.g. chromite);

Cr encapsulated in carbonaceous matter containing amorphous Al-

Si (e.g. Figure 11); absorbed in Fe-sulfates (e.g. schwertmannite,

paracoquimbite, and jarosite) and /or Fe-oxides (e.g. hematite and

magnetite); in association with aluminosilicate phases (such as

glass); and poorly crystallized chromium oxyhydroxide (CrOOH).

The high resolution XPS spectrum of Cr2p shows the presence of a

very low intensity Cr signal at the sample surface (Figure 12). The

position in binding energy of the Cr2p3/2 is around 576.7 eV, which

is normally attributed to chromium oxide Cr(III). TOF-SIMS

illustrates the coincidence of Al-, Fe-, and Cr-bearing areas

(Figure 13), which would suggest a spinel association, but does not

provide added clues to the dominance of any single Cr association.

4. Summary

Fly ashes collected from the baghouse of a university-based

coal-fired stoker boiler were analyzed by XRD and a number of

electron beam techniques, including HR-TEM, TOF-SIMS, XPS, and

FE-SEM.

The fly ashes, derived from an eastern Kentucky high volatile A

bituminous coal, were dominated by an Al-Si glass and fine

carbon. XRD results showed that the mineral phases included

quartz, mullite, maghemite, and magnesioferrite; with lesser

amounts of wollastonite, microcline, and albite in some of the

samples.

The original ash, as discussed by Mardon et al. (2008), had

significant concentrations of some trace elements, including As

and Hg. Some of the sooty carbon is in the form of nano-carbon

balls. TOF-SIMS peaks between 650–970 amu show the major

fullerenes at C60+, C70

+, and C80+ and species from C56

+ to C78+

reflecting C2 loss from the major species. HR-TEM indicated the

association of metal, including As and Hg, associations with

fullerenes and multi-walled nanotubes. At the level of resolution

available with HR-TEM, it is not possible to determine if the metals

are bound within the fullerene balls or to the side of the structure.

In either case, the association with the carbons contributes to the

retention of trace metals in fly ashes.

TOF-SIMS analysis indicated a correspondence between the

location of As and Al on the surface, indicating some As

association with the Al-Si glass component of the fly ash. Lead

phases included oxides, sulfates, and Pb associated with the

carbon. As anticipated, mercury was noted to be in association

with carbon. Selenium, although in low concentrations, appears to

be present in elemental form in association with the nano carbons.

Chromium is encapsulated by nano carbons, adsorbed in Fe-

sulfates or Fe-oxides, in association with glass, and in the form of

Cr-oxyhydroxide. While there is some indication of an association

of Cr with spinel minerals, TOF-SIMS did not provide an indication

of a dominant Cr association.

Acknowledgements

The work performed by the group from Brazil (FE-SEM, HR-

TEM, XRD, and mineral extractions) was carried out with support

from the Catarinense Institute of Environmental Research and

Human Development – IPADHC.

References

Baker, R. T. K., 1989. Catalytic growth of carbon filaments. Carbon 27, 315–323.Bai, J., Virovets, A. V., Scheer, M., 2003. Synthesis of inorganic fullerene-like

molecules. Science 300, 781–783.Byeon, J.H., Kim, J., 2010. Carbon-Encapsulated Metal Nanoparticles from

Various Ambient Metal-Carbon Spark Discharges. Applied Materials &Interfaces 2, 947–951.

Chen Y., Shah N., Huggins F.E., Huffman G.P., 2004. Investigation of themicrocharacteristics of PM2.5 in residual oil fly ash by analyticaltransmission electron microscopy. Environmental Science and Technology38, 6553–6560.

Chen Y., Shah N., Huggins F.E., Huffman G.P., 2006. Microanalysis of ambientparticles from Lexington, KY, by electron microscopy. AtmosphericEnvironment 40, 651–663.

Chen Y., Shah N., Huggins F.E., Huffman, G. P., Dozier, A., 2005. Characterizationof ultrafine coal fly ash particles by energy-filtered TEM. Journal ofMicroscopy 217, 225–234.

Cobo M., Galvez A., Conesa J.A., Correa C.M., 2009. Characterization of fly ashfrom a hazardous waste incinerator in Medellin, Colombia. Journal ofHazardous Materials 168, 1223–1232.

Ding, Z., Zheng, B., Long, J., Belkin, H.E., Finkelman, R.B., Chen, C., Zhou, D.,Zhou, Y., 2001. Geological and geochemical characteristics of high arseniccoals from endemic arsenosis areas in southwestern Guizhou Province, China.Applied Geochemistry 16, 1353–1360.

Du, A.-B., Liu, X.-G., Fu, D.-J., Han, P.-D., Xu, B.-S., 2007. Onion-like fullerenessynthesis from coal. Fuel 86, 294–298.

Dunsch, L., Yang, S., Zhang, L., Svitova, A., Oswald, S., Popov, A. A., 2010. Metalsulfide in a C82 fullerene cage: A new form of endohedral clusterfullerenes.Journal of the American Chemical Society 132, 5413–5421.

Fergusson, J. E., 1990. The heavy elements, chemistry, environmental impact andhealth effects; Pergamon, NY.

Foltin, M., Lezius, M., Scheier, P., Mark, T. D., 1993. On the unimolecularfragmentation of C60

+ fullerene ions: The comparison of measured andcalculated breakdown patterns. Journal of Chemical Physics 98, 9624–9634.

Goodarzi F., Sanei H., 2009. Plerosphere and its role in reduction of emitted finefly ash particles from pulverized coal-fired power plants. Fuel 88, 382–386.

Goodarzi, F., 2006. Morphology and chemistry of fine particles emitted fromsome Canadian coal-fired power plants. Fuel 85, 273–280.

Goodarzi, F., Huggins, F.E., Sanei, H., 2008. Assessment of elements, speciation ofAs, Cr, Ni and emitted Hg for a Canadian power plant burning bituminouscoal, International Journal of Coal Geology 74, 1–12.

Graham, U.M., Dozier, A., Khatri, R.A., Tseng, M., Hower, J.C., Davis, B.H., 2008.Ultra-fine PM derived from fullerene-like carbon in electrostatic precipitatorfly ash. American Institute of Chemical Engineers, 2008 Annual Meeting, 16–21 November, Philadelphia, PA.

Hochella, M., Lower, S., Maurice, P., Penn, L., Sahai, N., Sparks, D., Twining, B.,2008. Nanominerals, Mineral Nanoparticles, and Earth Systems. Science 319,1631–1635.

Hower, J. C., Graham, U. M., Dozier, A., Tseng, M. T., Khatri, R. A., 2008.Association of the sites of heavy metals with nanoscale carbon in a Kentuckyelectrostatic precipitator fly ash. Environmental Science and Technology 42,8471–8477.

Hower, J.C., Rathbone, R.F., Graham, U.M., Groppo, J.G., Brooks, S.M., Robl, T.L.,Medina, S.S., 1995. Approaches to the petrographic characterization of flyash. International Coal Testing Conference, 11th, Lexington, KY, May 10–12,1995, 49–54.

Hower, J.C., Robl, T.L., Thomas, G.A., Hopps, S.D., Grider, M., 2009. Chemistry ofcoal and coal combustion products from Kentucky power plants: Results fromthe 2007 sampling, with emphasis on selenium: Coal Combustion &Gasification Products 1, 50–62. http://www.coalcgp-journal.org/papers/2009/CCGP-D-09-00013-Hower-supp.pdf

76 Silva et al. / Coal Combustion and Gasification Products 2 (2010)

Hower, J.C., Senior, C.L., Suuberg, E.M., Hurt, R.H., Wilcox, J.L., Olson, E.S., 2010.Mercury capture by native fly ash carbons in coal-fired power plants.Progress in Energy and Combustion Science 36, 510–529.

Kato, N., Yamashita, Y., Iida, S., Sanada, N., Kudo, M., 2008. Analysis of TOF-SIMS spectra from fullerene compounds. Applied Surface Science 255, 938–940.

Kroto, H. W., Health, J. B., O’Brien, S.C., Curl, R. F., Smalley, R. E., 1985. C60:Buckminsterfullerene. Nature 318, 162–163.

Kuznetsov, V. L., Usoltseva, A. N., Butenko, Y. V., 2003. Mechanism of coking onmetal catalyst surfaces: I. Thermodynamic analysis of nucleation. ReactionKinetics and Catalysis Letters 44, 726–734.

Lyman, G. H., Lyman, C. G., Johnson, W., 1985. Association of leukemia withradium groundwater contamination. Journal of the American MedicalAssociation 254, 621–626.

Mandal, B.K., Suzuki, K.T. 2002. Arsenic around the world: a review. Talanta 58,201–235.

Mardon, S.M., Hower, J.C., 2004. Impact of coal properties on coal combustionby-product quality: Examples from a Kentucky power plant. InternationalJournal of Coal Geology 59, 153–169.

Mardon, S.M., Hower, J.C., O’Keefe, J.M.K., Marks, M.N., Hedges, D.H., 2008. Coalcombustion by-product quality at two stoker boilers: Coal source vs. fly ashcollection system design. International Journal of Coal Geology 75 48–254.

Mejıa, J. J., Dıaz-Barriga, F., Calderon, J., Rıos, C., Jimenez-Capdeville, M. E.,1997. Neurotoxicol. Teratol. 19, 489–497.

Moses, M. J., Fettinger, J. C., Eichhorn, B. W., 2003. Interpenetrating As20

fullerene and Ni12 icosahedra in the onion-skin [As@Ni12@As20]32 ion.

Science 300, 778–780.Murr L.E., Soto K.F., 2005. A TEM study of soot, carbon nanotubes, and related

fullerene nanopolyhedra in common fuel–gas combustion sources. MaterialsCharacterization 55, 50–65.

Nejar N., Makkee M., Illan-Gomez M.J., 2007. Catalytic removal of NOx and sootfrom diesel exhaust: oxidation behaviour of carbon materials used as modelsoot, Applied Catalysis B 75, 11–16.

Nicholls, R.J., Sader, K., Warner, J.H., Plant, S.R., Porfyrakis, K., Nellist, P.D.,Briggs, G.A.D., Cockayne, D.J.H., 2010. Direct imaging and chemicalidentification of the encapsulated metal atoms in bimetallic endofullerenepeapods. ACS Nano 4, 3943–3948.

NLM. TOXNET. 2008. National Library of Medicine. http://www.toxnet.nlm.nih.gov .

Reich, M., Utsunomiya, S., Kesler, S., Wang, L., Ewing, R.C., Becker, U., 2006.Thermal behavior of metal nanoparticles in geologic materials, Geology 34,1033–1036.

Silva, L.F.O., Moreno, T., Querol, X., 2009. An introductory TEM study of Fe-nanominerals within coal fly ash. Science of the Total Environment 407,4972–4974.

Silva, L.F.O., DaBoit, K.M., 2010. Nanominerals and nanoparticles in feed coal andbottom ash: implications for human health effects. Environmental Monitor-ing and Assessment, DOI 10.1007/s10661-010-1449-9.

Stultz, C.R., Kitto, J.B., 2005. Steam - Its Generation and Use, 41st edition.Babcock and Wilcox Company, Barberton, Ohio.

Sun, X.Ch., Reyes-Gasga, J. Y., Dong, X.L. 2002. Formation and microstructure ofcarbon encapsulated super-paramagnetic Cu nanoparticles. Molecular Physics100, 3147–3150.

U.S. Environmental Protective Agency, 2006. Special report on lead pollution.http://www.epa.gov/air/airtrends/lead.html (accessed December 10, 2008).

Utsonomiya, S., Jensen, K. A., Keeler G. J., Ewing, R. C., 2002. Uraninite andFullerene in Atmospheric Particulates, Environmental Science & Technology36, 4943–4947.

Wang, S., 2008. Application of solid ash based catalysts in heterogeneouscatalysis. Environmental Science & Technology 42, 7055–7063.

Appendix A

A.1 Field Emission Scanning Electron Microscope (FE-SEM) and

High-Resolution Transmission Electron Microscope (HR-TEM)

Procedures

Field Emission Scanning Electron Microscope (FE-SEM) and

high-resolution transmission electron microscope (HR-TEM) allow

the direct (real space) visualization of nanominerals and ultra-fine

particles. In this investigation morphology, structure, and chemical

composition of ultrafine particles and minerals were investigated

using a FE-SEM Zeiss Model ULTRA plus with charge compensa-

tion for all applications on conductive as well as non-conductive

samples and a 200-keV JEOL-2010F HR-TEM equipped with an

Oxford energy dispersive X-ray detector, and a scanning (STEM)

unit (Silva and DaBoit 2010). The FE-SEM was equipped with an

energy-dispersive X-ray spectrometer (EDS) and the mineral

identifications were made on the basis of morphology and grain

composition using both secondary electron and back-scattered

electron modes (Silva et al., 2009a, b). Geometrical aberrations

were measured by HR-TEM and controlled to provide less than a p/

4 phase shift of the incoming electron wave over the probe-

defining aperture of 14.5 mrad (Silva et al., 2010c). The scanning

acquisition was synchronized to the AC electrical power to

minimize 60-Hz noise, and a pixel dwell time of 32 ms was

chosen. EDS spectra were recorded in FE-SEM and HR-TEM images

mode and then quantified using ES Vision software that uses the

thin-foil method to convert X-ray counts of each element into

atomic or weight percentages. Electron diffraction patterns of the

crystalline phases were recorded in SAED (selected area electron

diffraction) or MBD (microbeam diffraction) mode, and the d

spacings were compared to the International Center for Diffraction

Data (ICDD, 2010) inorganic compound powder diffraction file

(PDF) database to identify the crystalline phases. Hexane, acetone,

dichloromethane, and methanol suspensions were utilized to

prevent possible mineralogical changes in individual solvents.

The suspension was pipetted onto lacycarbon films supported by

Cu grids (200 mesh) and left to evaporate before inserting the

sample into the FE-SEM and HR-TEM. This method may have led

to agglomeration but is a widely used standard procedure for most

minerals, including metal sulphates (Giere et al., 2006; Silva et al.,

2010b, c). Before FE-SEM and STEM analysis, the HR-TEM

specimen holder was cleaned with an Advanced Plasma System-

APS (Gatan Model 950) to minimize contamination. A drift

correction system was used for the STEM-EDS mapping.

A.2 X-Ray Photoelectron Spectroscopy (XPS) and Time-of-Flight

Secondary Ion Mass Spectrometry (TOF-SIMS) Procedures

Detailed information on both the surface and internal structures

of CFA micro and nanoparticles, plays an important role in the

development of new analytical techniques for toxic compounds on

and beneath the surface (Sakamoto et al., 2003; Silva et al., 2009a).

However, conventional analytical techniques such as XRD, FE-

SEM, HR-TEM, ICP-AES and ICP-MS evaluate the morphology and

average concentration of trace elements in bulk samples, but do

not provide information on their chemical state or molecular

structure at the outermost surface where many chemical changes

occurs. Therefore we have employed surface sensitive techniques

to afford the chemical analysis of the surface.

Time of flight secondary ion mass spectrometry (TOF-SIMS IV

instrument from Ion-Tof GmbH) was applied to investigate the

elemental and molecular structure of the samples giving the

possibility of a better understanding of the chemical composition,

localization and relative quantity of the different present species at

the surface. The secondary ions collected and represented at the

mass spectrum can be attributed to complete molecules, large

fragments of molecules that only have lost functional groups or

parts of repeating units, small molecular fragments as end groups

and elements.

In the TOF-SIMS experiment, the sample was bombarded with a

pulsed Bismuth ion beam. The secondary ions generated were

Silva et al. / Coal Combustion and Gasification Products 2 (2010) 77

extracted with a 10 KV voltage and their time of flight from the

sample to the detector was measured in a reflection mass

spectrometer. Typical analysis conditions for this work were: (1)

25 keV pulsed Bi3+beam at 45u incidence, rastered over

2003200 um2 for spectroscopy and 4643464 um2; for mapping

(2) Low energy Flood Gun was applied compensate the surface

charge during the experiments. As a result, the surface excitation

by particle bombardment leads to the emission of secondary ions

characteristic of the chemical composition in the uppermost

monolayer. For each type of molecule, secondary ion emission

results in a characteristic set of ionized molecular fragments with

well-defined mass spectra. These secondary ions are analyzed with

respect to their mass to charge ratio. In a time-of-flight (TOF) mass

spectrometer the separation is based on the fact that ions of the

same energy that have different masses have different flight times

towards the detector.

The chemical maps produced by TOF-SIMS represent the ions

that reached the detector rather than the ions that were present on

the surface, and each solid has its own characteristic ability to

release ions, so intensities cannot be used to derive absolute

surface concentrations. However, chemical maps are very good at

indicating relative surface concentration and its change as a

function of sample treatment or time. This technique provides

information on the elemental distributions not only at the surface

but also in the interior of a CFA particle (Hayashi et al., 2010).

In the case of coal fly ash, the presence of a wide range of

organic and inorganic compounds was determined. In general

terms the composition of this kind of samples is very complex, and

we have detected light elements, heavy metals, and hydrocarbons.

In general terms, we will emphasize the regions of the spectrum

that have the maximum interest: polluting heavy metals and

fullerenes.

For high resolution chemical state investigation we have applied

XPS which provides information about surface composition with

spatial resolution of a millimeter or less, with information depth

limited to 10 nanometers. This allows the possibility to investigate

the average chemical nature of the solid. Chemical identity,

bonding structure, and redox state can be determined. XPS

analysis of the CFA samples was performed using a Thermo

Scientific K-Alpha ESCA instrument equipped with aluminum

Ka1,2 monochromatized radiation at 1486.6 eV X-ray source. Due

the non conductor nature of samples it was necessary to use an

electron flood gun to minimize surface charging. Neutralization of

the surface charge was performed by using both a low energy flood

gun (electrons in the range 0 to 14 eV) and a low energy Argon

ions gun. The XPS spectra for CFA were obtained using X-ray

photoelectron spectrometer with a monochromatized Al Ka

(1486.6 eV) source. The pressure in the analytical chamber was

below 10-9 mbar. The survey and narrow scans were recorded with

pass energy of 100 and 20 eV, respectively. In order to obtain

oxidation status of surface contaminants, narrow scan spectra of

the C 1s, Cr 2p, S 2p, O 1s, As 3d, and Pb 4f were acquired. Due to

the nature of the sample, it would be expected that charge effects

could occur, resulting in the shifts of spectra. To prevent problems

we have used a dual beam (electrons and Argon ions) to

compensate the surface charge. To calibrate in binding energies,

the peak at the lower binding energy in the C1s spectrum was fixed

at 285 eV and was used as a binding energy reference. The narrow

scan spectra were fitted using an Avantage fitting program with

Gaussian Lorentzian function through background-subtraction

corrections using a Shirley-type optimization. The atomic con-

centrations were determined from the XPS peak areas using the

Shirley background subtraction technique and the Scofield

sensitivity factors. The High Resolution spectra were recorded,

acquiring several scans (40 scans) in order to improve the signal/

noise ratio, for the minor elements including Cr, Pb, and As.

Sample preparation for TOF-SIMS and XPS examination:

Special preparation of the sample was done to prevent – in both

instruments- that any particle can be accidentally detached from

the sample holder inside the analysis chamber. The sample

preparation consists only of a process for powders compaction

(molded), which was done using a pneumatic press and by

obtaining tablets.

A.3 Sequential Extraction Procedures

1) For the water-soluble fraction, a 1-mg CFAs sample (five

replicates) was mixed with Millipore-system water (1mL) with

electric conductivity of 0.1–0.5mS/cm. Samples were shaken in

the dark for 4 h, then centrifuged (3000 rpm, 10 min) and

filtered (, 22 mm). This extraction dissolves salts (e.g.

epsomite, albite, and pickeringite), jarosite, alunogen, gypsum

(partially soluble in water), and other minerals soluble in

water. For the secondary sulfates and salts formed in CFAs

environments, such an initial water-extraction step is essential

for a complete understanding of materials involved. The solid

fraction was cold dried, suspended in acetone, pipetted on to

separated lacy carbon films supported by Cu grids, and left to

evaporate before inserting the sample into the FE-SEM and

HR-TEM. The liquid fraction was air dried/crystallized before

inserting the sample into the XRD, and FE-SEM.

2) Exchangeable fraction, 1M NH4-acetate pH 4.5, samples were

shaken in the dark for 2 hours at room temperature, then

centrifuged (3000 rpm, 10 min) and filtered (,22 mm). This

extraction dissolves vermiculite-type mixed-layer, absorbed,

and exchangeable ions (Dold, 2003).

3) For poorly ordered Fe and Al oxyhydroxides/sulphates

(Gagliano et al., 2004; Regenspurg et al., 2004), a 1-mg

mineral sample (five replicates) was mixed with ammonium

oxalate (1mL) reagent (28 g/L ammonium oxalate + 15 g/L

oxalic acid solution, pH , 2.7). The samples were then

shaken in the dark for 4 hours, centrifuged (3000 rpm,

10 min) and filtered (, 22 mm) (Peretyazhko et al., 2009).

This extraction dissolves poorly-crystalline Fe (III) oxides

(e.g. ferrihydrite, schwertmannite) in preference to more

insoluble crystalline Fe (III) oxides (e.g. goethite, hematite)

(Cornell and Schwertmann, 2003; Silva and Da Boit, 2010).

More than 85% of the total iron was released in this step. The

solid fraction was cold dried before inserting the sample into

the Raman, XRD, FE-SEM, and HR-TEM.

4) Highly ordered Fe3+ hydroxides and oxides (hematite and

goethite) were partially dissolved by acid ammonium oxalate

(Kumpulainen et al., 2007) using a SE step described by Dold

(2003). As in step 2, the extractant used was 0.2 M NH4-

oxalate for 2h, but in this case the samples in batch system

were exposed to light and heated to 80 uC in a water bath.

The solid fraction was cold dried, suspended in toluene,

pipetted onto separated lacy carbon films supported by Cu

grids, and left to evaporate before inserting the sample into

the FE-SEM and HR-TEM. The liquid fraction was air dried/

78 Silva et al. / Coal Combustion and Gasification Products 2 (2010)

crystallized before inserting the sample into the XRD and FE-

SEM.

5) For organic matter (after steps 1–4) a 0.1-mg resultant CFAs

sample (five replicates) was mixed with 1 mL of dimethyl

sulfoxide (DMSO) reagent. The samples were shaken in the

dark for 12 hours, then centrifuged and filtered (, 22 mm).

This extraction dissolves organic anions absorbed on Al-OH/

Fe-OH surface of minerals, due to electrostatic interactions,

and can lead to increased repulsion between uniformly and

negative charged particles;

6) Residual organic and secondary metal-sulfides were mixed

with 35% H2O2 heat in water batch for 1hour, then

centrifuged and filtered (, 22 mm). This extraction dissolves

sulfides (chalcopyrite, cinnabar, galena, orpiment, pyrite,

sphalerite, stibnite, and tennantite-tetrahedrite). The solid

fraction was hot dried (100 uC), suspended in toluene, and

pipetted onto lacy carbon films supported by Cu grids (400

mesh). Before FE-SEM analysis, the HR-TEM specimen holder

was cleaned with an APS to minimize contamination.

References

Cornell, R.M., Schwertmann, U., 2003. The Iron Oxides: Structure, Properties,Reactions, Occurrence and Uses, second, completely revised and extendededition. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.

Dold, B., 2003. Speciation of the most soluble phases in a sequential extractionprocedure adapted for geochemical studies of copper sulfide mine waste.Journal of Geochemical Exploration, 80, 55–68.

Gagliano, W.B., Brill, M.R., Bigham, J.M., Jones, F.S., Traina, S.J., 2004. Chemistryand mineralogy of ochreous sediments in a constructed mine drainagewetland. Geochimica et Cosmochimica Acta 68, 2119–2128.

Giere R, Blackford M, Smith K., 2006. TEM Study of PM2.5 Emitted from Coal andTire Combustion in a Thermal Power Station. Environmental Science &Technology 40, 6235–40.

Hayashi, S., Takahashi, T., Kanehashi, K., Kubota, N., Mizuno, K., Kashiwakura, S.,Sakamoto, T., Nagasaka, T., 2010. Chemical state of boron in coal fly ashinvestigated by focused-ion-beam time-of-flight secondary ion massspectrometry (FIB-TOF-SIMS) and satellite-transition magic angle spinningnuclear magnetic resonance (STMAS NMR). Chemosphere 80, 881–7.

International Center for Diffraction Data (ICDD), 2010. http://www.icdd.com[accessed: 20 July 2010].

Kumpulainen, S., Carlson, L., Raisanen, M.L., 2007. Seasonal variations ofochreous precipitates in mine effluents in Finland. Applied Geochemistry 22,760–777.

Peretyazhko, T., Zachara, J. M., Boily, J.-F., Xia, Y., Gassman, P. L., Arey, B. W.Burgos, W. D., 2009. Mineralogical transformations controlling acid minedrainage chemistry. Chemical Geology 262, 169–178.

Regenspurg, S., Brand, A., Peiffer, S., 2004. Formation and stability ofschwertmannite in acidic mining lakes. Geochimica et Cosmochimica Acta68, 1185–1197.

Sakamoto, T., Shibata, K, Takanashi, K., Owari, M., Nihei, Y., 2003. Analysis ofsurface composition and surface of fly ash particlesunsing an ion ans electronmultibeam microanalyser. Applied Surface Science, 203, 762–6.

Silva, LFO, Moreno, T, Querol, X., 2009a. An introductory TEM study of Fe-nanominerals within coal fly ash. Science of the Total Environment 407,4972–4974.

Silva, L.F.O., Oliveira, M.L.S., da Boit, K.M., Finkelman, R.B., 2009b. Character-ization of Santa Catarina (Brazil) coal with respect to Human Health andEnvironmental Concerns. Environ Geochem Health 31, 475–485.

Silva, L.F.O., DaBoit, K.M., 2010. Nanominerals and nanoparticles in feed coal andbottom ash: implications for human health effects. Environ Monit Assess, DOI10.1007/s10661-010-1449-9

Silva, L. F. O., Macias, F., Oliveira, M. L. S., Da Boit, K. M., Waanders, F., 2010a.Coal Cleaning Residues and Fe-minerals Implications. EnvironmentalMonitoring and Assessment, DOI: 10.1007/s10661-010-1340-8

Silva, L. F. O., Izquierdo, M., Querol, X., Finkelman, R.B., Oliveira M.L.S.,Wollenschlager M., Towler M., Perez-Lopez R., Macias F., 2010b. Leaching ofpotential hazardous elements of coal cleaning rejects. EnvironmentalMonitoring and Assessment, DOI: 10.1007/s10661-010-1497-1

Silva, L. F. O., Hower, J. C., Izquierdo M., Querol X., 2010c. Complexnanominerals and ultrafine particles assemblages in phosphogypsum of thefertilizer industry and implications on human exposure. Science of the TotalEnvironment, DOI: 10.1016/j.scitotenv.2010.07.023

Silva et al. / Coal Combustion and Gasification Products 2 (2010) 79