Determination of Nickel Species in Stack Emissions from Eight

Jun 28, 2011 - Department of Chemical and Materials Engineering, University of Kentucky, 103 S. J. Whalen Building, 533 S. Limestone Street, .... util...
2 downloads 9 Views 2MB Size
Download PDF
ARTICLE pubs.acs.org/est

Determination of Nickel Species in Stack Emissions from Eight Residual Oil-Fired Utility Steam-Generating Units Frank E. Huggins,†,* Kevin C. Galbreath,‡ Kurt E. Eylands,‡ Lisa L. Van Loon,|| Jeremy A. Olson,|| Edward J. Zillioux,^ Stephen G. Ward,# Paul A. Lynch,r and Paul ChuO †

)

Department of Chemical and Materials Engineering, University of Kentucky, 103 S. J. Whalen Building, 533 S. Limestone Street, Lexington, Kentucky 40506-0043, United States ‡ Energy & Environmental Research Center, University of North Dakota, Grand Forks, ND, USA Canadian Light Source, Inc., Saskatoon, SK, Canada ^ Zillioux Environmental, LLC, Fort Pierce, FL, USA # Hawaiian Electric Company, Inc., Honolulu, HI, USA r National Grid, Hicksville, NY, USA O Electric Power Research Institute, Inc., Palo Alto, CA, USA

bS Supporting Information ABSTRACT: XAFS spectroscopy has been used to determine the Ni species in particulate matter collected on quartz thimble filters in the stacks of eight residual (No. 6 fuel) oil-burning electric utility steam-generating units. Proper speciation of nickel in emitted particulate matter is necessary to correctly anticipate potential health risks. Analysis of the spectroscopic data using leastsquares linear combination methods and a newly developed method specific for small quantities of Ni sulfide compounds in such emissions show that potentially carcinogenic Ni sulfide compounds are absent within the detection limits of the method (e3% of the total Ni) in the particulate matter samples investigated. In addition to the major nickel sulfate phase (NiSO4 3 6H2O), lesser amounts of (Ni,Mg)O and/or NiFe2O4 were also identified in most emission samples. On the basis of the results from these emission characterization studies, the appropriateness of the U.S. Environmental Protection Agency’s assumption that the Ni compound mixture emitted from residual oil-fired power plants is 50% as carcinogenic as nickel subsulfide (Ni3S2) should be re-evaluated

’ INTRODUCTION The combustion of petroleum fuels for generating steam and electricity is a significant anthropogenic source of Ni released to the atmosphere.1 Petroleum-fired capacity represented about 1% (39 000 000 MWh) of the total U.S. electric utility capacity in 2009.2 Residual (No. 6 fuel) oil, a byproduct from the petroleumrefining industry, is the most widely used for generating steam and electricity because of its relatively low cost compared with that of lighter oils. Residual oil production in the United States has declined since the late 1980s but remains substantial with stocks at about 40 000 000 barrels per month.3 In the United States, the impetus for focusing on individual elements, such as Ni, in air pollution derives from the 1990 Clean Air Act Amendments (CAAA) and the attainment of National Ambient Air Quality Standards.4,5 The U.S. Environmental Protection Agency’s (EPA’s) Integrated Urban Air Toxics Strategy also classifies Ni as an urban hazardous air pollutant (HAP). Many stationary sources report Ni emissions as part of the EPA Toxics Release Inventory (TRI).6 Although TRI and similar reporting r 2011 American Chemical Society

provide estimates of the total Ni released into the environment, they are not an indicator of toxicity because the acute, chronic, and cancer-causing effects vary significantly for the different chemical species of Ni. For example, nickel subsulfide (Ni3S2) is considered the most carcinogenic Ni species on the basis of available human epidemiology and animal studies.7,8 In contrast, inhalation exposure to water-soluble Ni salts alone, such as nickel sulfate hexahydrate (NiSO4 3 6H2O), has not been shown to cause cancer in animal studies.911 Insoluble nickel oxide compounds such as Ni-containing spinel (e.g., NiFe2O4), although still under investigation, are expected to have relatively low cancer potencies compared to Ni3S2 and may not be bioaccessible.7,8,12 Determining the Ni speciation of emission sources is vital for assessing the

Received: March 11, 2011 Accepted: June 9, 2011 Revised: May 25, 2011 Published: June 28, 2011 6188

dx.doi.org/10.1021/es200823a | Environ. Sci. Technol. 2011, 45, 6188–6195

Environmental Science & Technology

ARTICLE

Table 1. List of Samples Collected and Information on XAFS Spectroscopy sample code

operator

PTF2 PPE3

ESPa

plant

unit #

# of qtz thimbles

XAFS beamline

FPL group

Turkey Point, FL

2

no

3

SSRL 41

FPL group

Port Everglades, FL

3

yes

1b

SSRL 41

PPE4

FPL group

Port Everglades, FL

4

yes

1b

SSRL 41

PMT1

FPL group

Manatee, FL

1

no

3

SSRL 41

PMR1

FPL group

Martin, FL

1

no

3

SSRL 41

PMR2

FPL group

Martin, FL

2

no

3

SSRL 41

H9

HECO

Honolulu, HI

9

no

2

SSRL 41

W6 K1

HECO HECO

Waiau, HI Kahe, HI

6 1

no no

2 2

NSLS X-18B NSLS X-18B

PJEFF

Nat. Grid

Port Jefferson, NY

4

yes

3

CLS 06ID1

ESP refers to whether or not an electrostatic precipitator was operational at the plant. b Three thimble filters were collected, but only one was examined by XAFS spectroscopy because the amounts of particulate matter collected on filters from this plant were extremely low. a

inhalation health risks associated with airborne Ni-containing particles. The inhalation health risks associated with HAP emissions from oil-fired utility boilers were estimated in 1998 by the U.S. EPA.8 EPA estimated that the potential contribution of Ni to the maximum individual risk exceeded that from all other HAPs by about an order of magnitude because of its relatively high concentration, generally 14 wt % Ni, in residual oil particulate matter and known carcinogenic potency when present as Ni3S2. As highlighted in the EPA’s study and a 2004 proposed Ni emission limit ruling, the speciation of Ni emissions from oil-fired utility boilers is uncertain and requires additional investigation.8,13 Total Ni emissions were measured at 13 of the 149 power plants that burned oil in 1994, and speciation measurements were performed on seven of them. The Ni speciation analyses performed indicated that 3%26% of the total Ni emissions were composed of sulfidic Ni, although it is unknown whether Ni3S2 was present because of the limitations of the indirect (i.e., operationally defined) speciation method employed, sequential Ni extraction.1417 In contrast, Galbreath et al.17,18 used direct and definitive speciation techniques, X-ray diffraction (XRD), and X-ray absorption fine structure (XAFS) spectroscopy, to identify the Ni species occurring in residual oil particulate matter sampled from 385 and 400 MW utility boiler stacks. Direct speciation measurements indicated that >95% of the total Ni in residual oil particulate matter was present as a mixture of NiSO4 3 xH2O and nickel oxide spinel compound, similar in composition to NiFe2O4. The lack of sulfidic Ni emissions from these particular boilers was in sharp contrast to EPA’s risk assessment assumption that the Ni compound mixture emitted from oil-fired power plants is 50% as carcinogenic as Ni3S2.8 As a consequence, the cancer risk from nickel in emissions from residual oil-fired boilers is likely greatly overestimated.18,19 As part of the sampling and analysis activities associated with the EPA’s most recent examination of this issue,20 three complementary methods  XAFS, XRD, and water-soluble Ni extraction followed by XAFS  were used to identify and quantify the Ni species occurring in fly ashes stack-sampled from ten oil-fired units, owned and operated by Florida Power and Light Company (FPL), Hawaiian Electric Company, Inc. (HECO), or National Grid. The XAFS technique is especially well suited for determining Ni speciation because it can directly and nondestructively analyze fly ash filter samples with parts per million sensitivity.18,19,2123 It should also be noted that the three utilities participating in this study account for about 2/3 of the power generated by residual oil combustion in the U.S.

’ EXPERIMENTAL SECTION (a). Fly Ash Stack Sampling and Analysis Methods. Particulate matter was sampled isokinetically in duplicate or triplicate from the stacks of 10 electric utility steam-generating units located in three U.S. states (Florida, Hawaii, New York) using a modified EPA Method 17 sampling train assembly with the nozzle, probe, and quartz thimble filter maintained at >290 C, well above the sulfuric acid dew point temperature.1315 The nozzle and filter holder were constructed of quartz and glass respectively to prevent contamination of particulate matter by metal surfaces. The sampling train and procedures used in obtaining representative fly ash samples are described in detail by EPA.24 The sampling approach employed is consistent with previous investigations of the Ni speciation of stack emissions from oil-fired utility boilers and sewage sludge incinerators.1419,25 Plant operating and monitoring data and continuous emission monitoring results indicated that the modified EPA Method 17 stack sampling occurred during steady, representative power plant operations. Sample collection times varied from 3 to 5 h for each sample. Such times were sufficient to develop a continuous and cohesive layer of particulate matter on the interior surfaces of the thimble filters (q.v. Figures S1 and S2 of the Supporting Information), except for those collected downstream from electrostatic precipitators (ESPs) at the Port Everglades, FL, power plant. Samples from these units were much less in quantity and merely slightly darkened the filter material. A complete list of samples is provided in Table 1. At least one filter from each unit at the Hawaii and Florida power plants (except for Port Everglades) was also subjected to aqueous leaching to remove the readily soluble Ni species, with the aim of improving the XAFS detection and identification of insoluble Ni species. These experiments were performed in glass vessels using deionized water of >18 MΩ resistance. The filter material was placed under water for 30 min, rinsed, and allowed to air-dry. Selected samples and standards were also examined by XRD. In the case of the Hawaii filter samples, XRD measurements (Supporting Information) provided information regarding speciation that complemented the XAFS Ni speciation results. For standards, such measurements were used to confirm their identity. Previous XRD measurements of commercially available Ni compounds have shown that some Ni sulfides (Ni3S2, NiS, and NiS2) can be mixtures of Ni3S2 and Ni7S6.19,26 Thermogravimetric analysis (TGA) was used to confirm the number of 6189

dx.doi.org/10.1021/es200823a |Environ. Sci. Technol. 2011, 45, 6188–6195

Environmental Science & Technology water molecules associated with the standard, nickel sulfate hexahydrate, and to prepare other nickel sulfates (NiSO4 3 H2O and NiSO4) from it. Further information on the standards used is given in the Supporting Information. (b). XAFS Spectroscopy. Nickel K-edge XAFS measurements were made at three different synchrotrons: (i) at beamline 41 of the Stanford Synchrotron Radiation Lightsource (SSRL) at Stanford University, CA, (ii) at beamline X-18B of the National Synchrotron Light Source (NSLS) at Brookhaven National Laboratory, NY, and (iii) at beamline 06ID-1 of the Canadian Light Source (CLS) in Saskatoon, SK, Canada, through the Industrial Access Program. There were significant differences in beamline hardware (e.g., Si(220) monochromator crystals were used at SSRL, whereas Si(111) crystals were used at the other two beamlines; X-18B at NSLS is a bending magnet line, whereas the CLS and SSRL beamlines are wiggler beamlines; the CLS beamline was focused using a rhodium mirror, whereas the other two beamlines were unfocused) and in detection systems (a Canberra 32-element Ge array detector 27 was used at CLS, a Lytle detector 28 or a Canberra 13-element Ge array detector at SSRL, and a PIPS detector at NSLS). Despite these differences, the actual experimental practice at the three synchrotrons was essentially similar. Nickel XAFS spectra were collected directly from the quartz thimble filters used to collect the samples so as to avoid problems and bias associated with separating the particulate matter from the filter. The thimble filters were either cut in half longitudinally or into smaller rectangular sections (Figures S1 and S2 of the Supporting Information) to expose the particulate matter to the X-ray beam. Nickel XAFS spectra were collected in fluorescence geometry over a range of energy extending at a minimum from 8233 to 8880 eV. Both Soller slits and a Co filter were used to enhance the signal/noise ratio.29 Multiple scans were recorded for each sample and then averaged to provide a single spectrum with a superior signal/noise ratio. The spectra of a number of Ni compounds were collected in transmission geometry at both SSRL and CLS to serve as reference spectra. At all three synchrotrons, energy calibration was based on assigning the energy of the first major peak in the derivative XAFS spectrum of a thin foil of Ni metal to 8333 eV. Conversion of the as-collected Ni XAFS spectra to normalized and background subtracted XAFS spectra and division into separate X-ray absorption near-edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) spectral regions were performed using either SixPack30 or Athena,31 routines that are available in the IFEFFIT XAFS analysis software package. Leastsquares fitting based on linear combinations of spectra of standard Ni compounds was also performed using these software routines.

’ RESULTS AND DISCUSSION (a). Nickel XAFS Spectra. Examples of the as-collected Ni K-edge XAFS spectral data for the emission samples are shown in Figure 1. The samples examined from the seven units not equipped with electrostatic precipitators (ESPs) exhibited spectra of similar high quality to those shown in part a of Figure 1 and were recorded using either a Lytle or PIPS detector. The samples from the two units at Port Everglades, FL, and the unit at Port Jefferson, NY, showed significantly weaker spectra because of the presence of ESPs, which greatly reduced the amount of particulate matter emissions at these units. Even when collected using the more sensitive Ge detector, the spectra for the Port Everglades samples shown in part b of Figure 1 are clearly much

ARTICLE

Figure 1. Comparison of nickel XAFS spectra of residual oil emissions collected on quartz thimble filters from (a) Unit 2, Turkey Point, FL, measured using a Lytle detector, and (b) Units 3 and 4, Port Everglades, FL, measured using a multielement Ge detector. Note also the featureless spectrum of a blank filter shown at the bottom of part a of Figure 1.

weaker than those shown in part a of Figure 1. The lack of a significant Ni signal is reflective of the very low particulate loadings downstream from the ESPs at the Port Everglades units. Consequently, no further analysis of the spectra from these samples was attempted. The unit from the Port Jefferson, NY, power plant is also equipped with an ESP, but one of older design. Here, enough material avoided capture by the ESPs to provide a reasonable Ni XAFS spectrum when measured by the high sensitivity Canberra 32-element Ge array detection system at CLS (q.v. Figure S12 of the Supporting Information). Also shown in part a of Figure 1 is an example of the spectrum of a blank quartz filter collected under identical conditions to those of the spectra for the particulate matter samples. Although there is a very small Ni edge exhibited by the blank filter (not discernible at the scale of the plot in part a of Figure 1), it is negligible in comparison to the edge exhibited by Ni in the particulate matter samples and no correction of the spectral data for this slight Ni contamination was necessary. Representative normalized nickel XANES and derivative XANES spectra are shown in parts a and b of Figure 2 respectively for various particulate matter samples. Normalized spectra for all samples are shown in the associated Supporting Information 6190

dx.doi.org/10.1021/es200823a |Environ. Sci. Technol. 2011, 45, 6188–6195

Environmental Science & Technology

ARTICLE

Figure 2. Representative nickel XANES spectra (a) and derivative XANES spectra (b) obtained from as-received material collected on quartz thimble filters. Spectra for PMR11, PTF21, H9Ni-2 were measured at SSRL, W6Ni-2 at NSLS, and PJEFF-1 at CLS. Samples PMR11 and PTF21 were collected at Florida power plants, W6Ni-1 and H9Ni-2 at Hawaii power plants, and PJEFF-1 from a power plant in New York.

(Figures S3S12). Spectra of replicate samples from the same unit were generally similar. However, minor variations among the samples from different power plants were discerned, although the spectra have many features in common. As shown in part a of Figure 2, as well as those in the Supporting Information, the XANES spectra all exhibit a small but distinct pre-edge peak at about 8333 eV, a rising absorption edge centered at about 8345 eV, and a maximum at about 8352 eV. These spectral features are more similar to those of Ni sulfate and oxide compounds than Ni sulfides. The EXAFS and RSF spectra for the particulate matter samples from FL and HI are shown in Figures S3S12 of the Supporting Information. A major peak is observed in the RSF (uncorrected for phase shift) at about 1.5 Å with a second, usually smaller, peak at about 2.6 Å. Comparison of these peaks to those exhibited by the standard compounds (Figures S13S16 of the Supporting Information) indicates compatibility with Ni sulfates and oxides but not with Ni sulfides. The aqueous-leached samples, especially those from Hawaii, generally show a much more prominent peak at 2.6 Å, consistent with a higher content of one or other Ni oxide species. (b). Least-Squares Fitting of Ni XANES. Various approaches were considered for applying least-squares fitting to both the XANES and EXAFS spectra of the particulate matter samples based on linear combinations of the corresponding spectra for various Ni standards. The method found best applicable to all samples, including the aqueous-leached samples, consisted of fitting the Ni XANES spectra to a linear combination of the spectra for NiSO4 3 6H2O, NiFe2O4, and (Ni0.5Mg0.5)O. As is well documented in the literature,32,33 the Ni XANES spectra of (Nix,Mg1x)O solid solutions vary significantly with the compositional index, x, and replacement of the spectrum of NiO by that of (Ni0.5Mg0.5)O much improved the least-squares fitting of the spectra of samples from the Florida power plants. The oxide is a mixed (Ni, Mg) oxide because Mg(OH)2 and other Mg-containing additives are added to the fuel oil to minimize deposit formation and corrosion at the Florida units. The exact form of the NiO had less impact for the spectra from the samples from the Hawaii or

New York power plants because it was less significant in these samples. Fits were also attempted that included one or other Ni sulfide spectrum in the linear combination, but the fits were always worse statistically or indicated a negative amount of the sulfide in the fit. Examples of the least-squares linear combination fitting of the Ni XANES spectra are shown in Figure 3. The deviations of the least-squares fit from the data are attributed principally to the composition-dependent variability of the Ni XANES spectrum of the (Nix.Mg1x)O phase. Figure 4 summarizes the results of the least-squares fitting of the Ni XANES for all samples investigated in this study. The contributions from the two oxide components are shown as separate components; however, there is a large uncertainty, perhaps as much as (10%, in quantifying the separate contributions from the two oxides because their spectra are similar and the exact composition of the (Nix,Mg1x)O phase is unknown. As indicated in Figure 4, NiSO4 3 6H2O is more prevalent in the Hawaii and New York samples, whereas the (Ni,Mg) oxide component is more significant in the Florida samples. Results are also indicated for the leached samples. The aqueous leaching of the quartz thimble filters significantly reduced the NiSO4 3 6 H2O contents of all samples, but did not completely remove it. Even so, the contributions from the oxide phases could be more clearly established. (c). Nickel XANES Spectra and Systematics. As indicated by the least-squares fitting and general systematics of the Ni XAFS spectra, it is clear that the contents of Ni sulfide and subsulfide in the particulate matter samples are small, if present at all. However, the estimated detection limit for a minor component in the least-squares linear combination fitting method is at least 5%. To improve the precision of this result, a detailed examination of the pre-edge region (83308340 eV) of the Ni XANES spectra was undertaken. Examination of the spectra of the standard compounds (q.v. Figures S13S16 of the Supporting Information) over this region shows that the Ni sulfides exhibit significant absorption peaks in their derivative XANES spectra, whereas Ni oxides and Ni sulfates show much less, almost flat absorption over the same region. This also can be seen from 6191

dx.doi.org/10.1021/es200823a |Environ. Sci. Technol. 2011, 45, 6188–6195

Environmental Science & Technology

ARTICLE

Figure 3. Examples of least-squares fitting of Ni XANES spectra to linear combinations of the spectra for NiSO4 3 6H2O, NiFe2O4, and (Ni0.5, Mg0.5)O. (a) As-received emission sample K2Ni-2, and (b) leached sample K2Ni-2 L.

Figure 4. Summary of nickel speciation results for as-received and leached emission samples collected on quartz thimble filters. Top, samples from Florida power plants; bottom, samples from Hawaii and New York power plants.

Figure 5, which compares the derivative XANES spectra for NiSO4 3 6H2O, (Ni,Mg)O, Ni3S2, and NiS. On the basis of these

differences, various determinative curves for hypothetical mixtures of Ni sulfide or subsulfide and Ni sulfate or Ni oxide compounds 6192

dx.doi.org/10.1021/es200823a |Environ. Sci. Technol. 2011, 45, 6188–6195

Environmental Science & Technology

ARTICLE

were derived. The equations for mixtures of Ni3S2 and NiSO4 3 6H2O and of NiS and NiSO4 3 6H2O are as follows: %NiðNi3 S2 Þ ¼ 813hð8336Þ  4:3

ð1Þ

%NiðNiSÞ ¼ 979hð8337Þ  10:1

ð2Þ

and

where h(8336) and h(8337) eV are the absorption heights at the energies corresponding to the pre-edge maxima in the derivative XANES spectra for Ni3S2 and NiS, respectively (Figure 5). The determinative curves for NiS and Ni3S2 with NiSO4 3 6H2O are closely similar, especially at the low sulfide end of the simulated mixtures (Figure S19 of the Supporting Information). Furthermore, the two peak maxima differ by only about 1 eV, so that if both sulfides were present there would be extensive overlap between the two contributions. Hence, by averaging the trends for the two sulfides, a single equation can be derived for estimating the percentage of Ni in the forms of Ni sulfide and subsulfide (% sulfidic Ni) in the presence of major Ni sulfate: %sulfidic Ni ¼ 444ðhð8336Þ + hð8337ÞÞ  6:9

ð3Þ

The application of this equation is limited to situations where both Ni sulfides are minor ( 2.5 μm) fraction consisting of much unburned carbon but were much less abundant or absent in the fine particulate matter (PM < 2.5 μm) fraction. Conversely, nickel ferrite was found to be more prevalent in the fine particulate matter fraction. A synthesis of the results of the current study and those of the laboratory study indicates that nickel sulfides are likely to be encountered only under oxygen-poor conditions and are converted to oxide forms under high stoichiometry conditions. The latter conditions are generally typical of commercial residual-oil combustion for power generation. The absence of Ni sulfides in particulate matter samples from the various commercial oil-burning power plants investigated in this study is significant because sulfidic Ni compounds are generally considered to be the most highly carcinogenic Ni compounds.8,35 The assumption made by the U.S. EPA that the Ni compound mixture emitted from U.S. oil-fired power plants is 50% as carcinogenic as Ni3S2 would appear overly conservative with respect to the findings of this study and should be reassessed.

’ ASSOCIATED CONTENT

bS

Supporting Information. Photographs of deposits, nickel XAFS spectral data (consisting of XANES, derivative XANES, k3chi EXAFS, and RSFs) for each filter and standard, additional details on the development of calibration curves for estimating % sulfidic Ni, and examples of XRD diffractograms for selected filters. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Tel.: +1 859 257 4045; fax: +1 859 257-7215; E-mail: Frank. [email protected].

’ ACKNOWLEDGMENT This investigation was supported by Florida Power & Light Company, Hawaiian Electric Company, Inc., National Grid, and

ARTICLE

the Electric Power Research Institute, Inc. We acknowledge the U.S. Department of Energy and Canadian Light Source, Inc., for their support of national synchrotron facilities at Stanford University (SSRL), Brookhaven National Laboratory (NSLS), and the University of Saskatchewan (CLS). The authors also acknowledge Air Hygiene International, Inc., AirKinetrics, Inc., and Air Nova, Inc., for sampling the residual oil particulate matter from the stacks of Florida Power & Light Company, Hawaiian Electric Company, Inc., and National Grid power plants, respectively.

’ REFERENCES (1) U.S. Environmental Protection Agency. Locating and Estimating Air Emissions from Sources of Nickel; EPA-450/484007f; Office of Air Quality Planning and Standards: Research Triangle Park, NC, March 1984. (2) U.S. Energy Information Administration. Electric Power Industry 2009: Year in Review; January 4, 2011. www.eia.doe.gov/cneaf/ electricity/epa/epa_sum.html. (3) U.S. Energy Information Administration. Petroleum Navigator; www.eia.gov/dnav/pet/pet_stoc_typ_d_nus_SAE_mbbl_m.htm. (4) Provisions for Attainment and Maintenance of National Ambient Air Quality Standards; U.S. Public Law, 101549, 1990. (5) U.S. Environmental Protection Agency. National Ambient Air Quality Standards (NAAQS); Washington, DC, Nov 2002. http://www. epa.gov/air/criteria.html. (6) U.S. Environmental Protection Agency. 1996 Toxics Release Inventory, Public Data Release  Ten Years of Right-to-Know; EPA 745R-08005; U.S. Environmental Protection Agency: Washington, DC, May 1998. (7) National Toxicology Program. Toxicology and Carcinogenesis Studies of Nickel Subsulfide (CAS #12035722) in F344/N Rats and B6C3F1Mice (Inhalation Studies); Technical Report Series No. 453; National Toxicology Program: Research Triangle Park, NC, 1996. (8) U.S. Environmental Protection Agency. Study of Hazardous Air Pollutant Emissions from Electric Utility Steam Generating Units—Final Report to Congress: Vol. 1; EPA453/R98004a; U.S. Environmental Protection Agency: Washington, DC, Feb 1998. (9) Toxicology Excellence for Risk Assessment. Toxicological Review of Soluble Nickel Salts; Prepared for the Metal Finishing Association of Southern California, Inc., U.S. Environmental Protection Agency, and Health Canada under Subcontract in Part with Science Applications International Corporation, EPA Contract No. 68-C70011; March 1999. (10) National Toxicology Program. Toxicology and Carcinogenesis Studies of Nickel Sulfate Hexahydrate (CAS #10101970) in F344/N Rats and B6C3F1Mice (Inhalation Studies); Technical Report Series No. 454; National Toxicology Program: Research Triangle Park, NC, 1996. (11) Oller, A. R. Respiratory carcinogenicity assessment of soluble nickel compounds. Environ. Health Perspectives 2002, 110 (Supplement 5), 841–844. (12) Heaney, P. J.; Banfield, J. A. Structure and chemistry of silica, metal oxides, and phosphates. In Health Effects of Mineral Dusts; Guthrie, G. D. Jr.; Mossman, B. T., Eds.; Mineralogical Society of America, Reviews in Mineralogy, 1993; Vol. 28, Chapt. 5, pp 185233. (13) U.S. Environmental Protection Agency. Proposed National Emission Standards for Hazardous Air Pollutants; and, in the Alternative, Proposed Standards of Performance for New and Existing Stationary Sources: Electric Utility Steam Generating Units; Proposed Rule, Federal Register, Vol. 69, No. 20, Jan 31, 2004; pp 46524752. (14) Hone, J. R. Speciation of nickel emissions from oil-fired utility boilers. Proceedings, 87th Annual Meeting - Air & Waste Management Association, Vol. 4B: Hazardous Air Pollution Emissions & Control, 1994; pp 94/RP130.01. (15) Goldstein, L. S. Chemical speciation and bioavailability of nickel from coal and oil ash. Presented at the EPRIDOE International 6194

dx.doi.org/10.1021/es200823a |Environ. Sci. Technol. 2011, 45, 6188–6195

Environmental Science & Technology

ARTICLE

Conference on Managing Hazardous and Particulate Air Pollutants, Sheraton Centre Toronto, Toronto, ON, Canada, Aug 1517, 1995, 5 pp. (16) Bell, A. C.; Wong, J.; Chu, P. Speciation of nickel emissions from oil-fired power plants. Proceedings, 89th Annual Meeting - Air & Waste Management Association, 1996; pp tp5604/1-tp5604/14. (17) Wong, J. L.; Qian, J.; Chen, C. H. Speciation of fly ash by phase extraction. Anal. Chim. Acta 1997, 349, 121–129. (18) Galbreath, K. C.; Toman, D. L.; Zygarlicke, C. J.; Huggins, F. E.; Huffman, G. P.; Wong, J. Nickel speciation of residual oil fly ash and ambient particulate matter using X-ray absorption spectroscopy. J. Air Waste Manage. Assoc. 2000, 50, 1876–1886. (19) Galbreath, K. C.; Schulz, R. L.; Toman, D. L.; Nyberg, C. M.; Huggins, F. E.; Huffman, G. P.; Zillioux, E. J. Nickel and sulfur speciation of residual oil fly ashes from two electric utility steam-generating units. J. Air Waste Manage. Assoc. 2005, 55, 309–318. (20) U.S. Environmental Protection Agency. Information Collection Effort for New and Existing Coal- and Oil-Fired Electric Utility Steam Generating Units; EPA ICR No. 2362.01, OMB Control No. 2060New, Federal Register, Vol. 74, No. 216, Nov 10, 2009; pp 5801258013. (21) Galbreath, K. C.; Crocker, C. R.; Nyberg, C. M.; Huggins, F. E.; Huffman, G. P.; Larson, K. P. Nickel speciation measurements of urban particulate matter: method evaluation and relevance to risk assessment. J. Environ. Monitor. 2003, 5, 56N–61N. (22) Huggins, F. E.; Shah, N.; Huffman, G. P.; Robertson, J. D. XAFS spectroscopic characterization of elements in combustion ash and fine particulate matter. Fuel Process. Technol. 2000, 6566, 203–218. (23) Huggins, F. E; Huffman, G. P.; Robertson, J. D. Speciation of elements in NIST particulate matter SRMs 1648 and 1650. J. Hazard. Mater. 2000, 74, 1–23. (24) U.S. Environmental Protection Agency. Emission Measurement Technical Information Center, Method 17  Determination of Particulate Emissions from Stationary Sources; March 2003, Washington, DC. www.epa.gov/ttnemc01/methods/method17.html. (25) Segal, R. R.; DeWees, W. G.; Bostian, H. E. Emission of Metals, Chromium and Nickel Species, and Organics from Municipal Wastewater Sludge Incinerators; Vol. I: Summary Report; EPA 600/R-92/003a; U.S. Environmental Protection Agency, March 1992. (26) Tirez, K.; Silversmit, G.; Vincze, L.; Servaes, K.; Vanhoof, C.; Mertens, M.; Bleux, N.; Berghmans, P. Speciation and fractionation of nickel in airborne particulate matter: comparison between selective leaching and XAS spectroscopy. J. Anal. At. Spectrom. 2011, 26, 514–527. (27) Cramer, S. P.; Tench, O.; Yocum, M.; George, G. N. A 13element Ge detector for fluorescence EXAFS. Nucl. Instrum. Methods Phys. Res., Sect. A 1988, 266, 586–591. (28) Lytle, F. W.; Greegor, R. B.; Sandstrom, D. R.; Marques, E. C.; Wong, J.; Spiro, C. L.; Huffman, G. P.; Huggins, F. E. Measurements of soft X-ray absorption spectra with a fluorescent ion chamber detector. Nucl. Instrum. Methods Phys. Res., Sect. A 1984, 226, 542–548. (29) Stern, E. A.; Heald, S. M. An X-ray filter assembly for fluorescence EXAFS measurements. Rev. Sci. Instrum. 1979, 50, 1579–1582. (30) Webb, S. M. SIXPack: A graphical user interface for XAS analysis using IFEFFIT. Phys. Scr. 2005, T115, 1011–1014. (31) Ravel, B.; Newville, M. ATHENA, ARTEMIS, HEPHAESTUS: data analysis for X-ray absorption spectroscopy using IFEFFIT. J. Synchrotron Rad. 2005, 12, 537–541. (32) Kuzmin, A.; Mironova, N.; Purans, J.; Rodionov, A. X-ray absorption spectroscopy study of NicMglcO solid solutions on the Ni K edge. J. Phys.: Condens. Matter 1995, 7, 9357–9368. (33) Yoshida, T.; Tanaka, T.; Yoshida, H.; Funabiki, T.; Yoshida, S. Study on the dispersion of nickel ions in the NiO-MgO system by X-ray absorption fine structure. J. Phys. Chem. 1996, 100, 2302–2309. (34) Huggins, F. E.; Huffman, G. P.; Linak, W. P.; Miller, C. A. Quantifying hazardous species in particulate matter derived from fossilfuel combustion. Environ. Sci. Technol. 2004, 38, 1836–1842. (35) Sunderam, F. W., Jr. Physicochemical and biological attributes of nickel compounds in relationship to carcinogenic activities. In Toxicology of Metals; Brown, S. S., Kodama, Y., Eds.; Ellis Horwood, Ltd.: Chichester, U.K., 1987; pp 355365. 6195

dx.doi.org/10.1021/es200823a |Environ. Sci. Technol. 2011, 45, 6188–6195