Langmuir 2004, 20, 11509-11516
11509
Oxygen, Carbon, and Sulfur Segregation in Annealed and Unannealed Zerovalent Iron Substrates Efthimia Papastavros,† Patrick J. Shea,‡ and Marjorie A. Langell*,† Department of Chemistry, University of Nebraska, Lincoln, Nebraska 68588-0304 and School of Natural Resources, University of Nebraska, Lincoln, Nebraska 68583-0915 Received July 8, 2004. In Final Form: September 24, 2004
Finely ground and pretreated iron substrates known as “zerovalent iron” or “Fe0” are used as reductants in the environmental remediation of halogenated hydrocarbons, and the composition of their surfaces significantly affects their reactivity. Samples of unannealed and annealed (heat-treated under H2/N2) zerovalent iron were analyzed using X-ray photoelectron spectroscopy (XPS) and Auger electron spectroscopy (AES). Surface concentration of the iron and of the impurities observed by XPS and AES, carbon, chlorine, sulfur, and oxygen, were measured before and after soaking in trichloroethylene (TCE) and in water saturated with TCE (H2O/TCE) to simulate chlorocarbon remediation conditions. Samples pretreated by annealing at high temperature under H2 contained less iron carbide. The carbide contaminant was evident in both iron and carbon XPS spectra, with binding energies of 709.0 and 283.3 eV for the Fe 2p3/2 and C 1s, respectively. The annealed Fe0 surface also contained more sulfur. The carbide concentration was essentially unchanged by TCE and H2O/TCE exposure, whereas the sulfur decreased in proportion to chlorine adsorption following the dechlorination reaction. While oxygen concentration is initially lower on the annealed substrate surface, it rapidly increased during the model TCE remediative treatment process and thus does not represent a significant effect of the annealing process on surface reactivity.
1. Introduction Zerovalent iron (Fe0) is an economical, widely used remediant for contaminated water and soil environments.1-4 Fe0 is commonly used as finely divided iron filings that are packed into trenches and berms through which the polluted water flows or is flushed. The iron effectively reduces halogenated organics such as trichloroethylene (TCE),5,6 as well as nitrogenated organics,7 nitrate,8 and highly oxidized metal pollutants, including UO24+ 9-11 and Cr6+.9,10,12 In the process, the iron metal is oxidized13,14 and can be considered a reagent that is stoichiometrically consumed in the remediation process. However, the surface chemistry of the active remediant is considerably more complex and may also serve a catalytic role.15,16 The working surface is a complicated mixture17 and through * To whom correspondence should be addressed. E-mail:
[email protected]. † Department of Chemistry. ‡ School of Natural Resources. (1) Gillham, R. W.; O’Hannesin, S. F. Groundwater 1994, 32, 958. (2) Miehr, R.; Tratnyek, P. G.; Bandstra, J. Z.; Scherer, M. M.; Alowitz, M. J.; Bylaska, E. J. Environ. Sci. Technol. 2004, 38, 139. (3) Wilkin, R. T.; Puls, R. W.; Sewell, G. W. Groundwater 2003, 41, 493. (4) Bigg, T.; Judd, S. J. Environ. Technol. 2000, 21, 661. (5) Kenneke, J. F.; McCutcheon, S. C. Environ. Sci. Technol. 2003, 37, 2829. (6) Farrell, J.; Kason, M.; Melitas, N.; Li, T. Environ. Sci. Technol. 2000, 34, 514. (7) Hundal, L. S.; Singh, J.; Bier, E. L.; Shea, P. J.; Comfort, S. D.; Powers, W. L. Environ. Pollut. 1997, 97, 55. (8) Huang, C. P.; Wang, H. W.; Chiu, P. C. Water Res. 1998, 32, 2357. (9) Qiu, S. R.; Lai, H.-F.; Roberson, M. J.; Hunt, M. L.; Amrhein, C.; Giancarlo, L. C.; Flynn, G. W.; Yarmoff, J. A. Langmuir 2000, 16, 2230. (10) Blowes, D. W.; Ptacek, C. J.; Benner, S. G.; McRae, C. W. T.; Bennett, T. A.; Puls, R. W. J. Contam. Hydrol. 2000, 45, 123. (11) Amrhein, C.; Hunt, M. L.; Roberson, M. J.; Yarmoff, J. A.; Qiu, S. R.; Lai, H.-F. Mineral. Magazine 1998, 62A, 51. (12) Astrup, T.; Stipp, S. L. S.; Christensen, T. H. Environ. Sci. Technol. 2000, 34, 4163. (13) Roh, Y.; Lee, S. Y.; Elless, M. P. Environ. Geol. 2000, 40, 184. (14) Scherer, M. M.; Balko, B. A.; Tratnyek, P. G. Mineral-Water Reactions: Kinetics and Mechanisms; ACS Symposium Series 715; American Chemical Society: Wachington, DC, 1998; p 301.
interaction with ambient oxygen and water forms oxyhydroxides which are important in the remediation mechanism and effectiveness of iron-promoted treatment systems. Impurities and defects in the iron can significantly impact reactivity18 and thus both short- and longterm remediation performance. Because the source of the iron filings is often recycled scrap iron or stainless steel, pretreatment is used to minimize contaminants and prepare a more reactive surface. In particular, Fe0 is often prepared by annealing the finely divided metal under reducing conditions. The annealing pretreatment decreases surface oxide concentration but brings other contaminants to the substrate surface. Sulfur and chlorine are common segregants that appear upon annealing iron,19 and they are observed here as well. We present evidence with the surface analytical techniques of Auger electron spectroscopy (AES) and X-ray photoelectron spectroscopy (XPS) that while surface oxides are initially reduced on the annealed iron filings, these species rapidly reform under model TCE remediation conditions to give comparable near-surface oxide stoichiometry to that of the unannealed iron prior to TCE exposure. We also show that a major compositional difference between annealed and unannealed Fe0 substrates is that the annealed iron contains more sulfur but substantially less surface carbide. While sulfur pretreated iron and iron sulfide substrates are well investigated for their surface reactivity in chlorocarbon remediation, iron carbide has not received as much emphasis. Its prevalence on the Fe0 substrate and its variable concentration with (15) Venkatapathy, R.; Bessingpas, D. G.; Canonica, S.; Perlinger, J. A. Appl. Catal., B 2002, 37, 139. (16) Bergendahl, J. A.; Thies, T. P. Water Res. 2004, 38, 327. (17) Liang, L.; Korte, B.; Gu, B.; Puls, R.; Reeter, C. Adv. Environ. Res. 2000, 4, 273. (18) Bonin, P. m. L.; Jedral, W.; Odziemkowski, M. S.; Gillham, R. W. Corros. Sci. 2000, 42, 1921. (19) Ueda, K.; Shimizu, R. Surf. Sci. 1973, 36, 789.
10.1021/la048288j CCC: $27.50 © 2004 American Chemical Society Published on Web 11/17/2004
11510
Langmuir, Vol. 20, No. 26, 2004
Papastavros et al.
Figure 1. AES spectra of unannealed and annealed Fe0 samples.
pretreatment conditions indicate that the influence of carbide on surface reactivity should be further pursued. 2. Experimental Section Zerovalent iron substrates were obtained from Peerless Metal Powders (Detroit, MI) in both annealed and unannealed forms. Materials from the same lot of these samples were previously used to model the environmental remediation of halocarbons and nitrocarbons/nitramines, results from which are published elsewhere.20-22 Prior to insertion in the surface analysis chamber, the samples were ground with an agate mortar and pestle to present a fresh surface, pressed into a clean piece of indium foil to form a thick layer of the iron, and mounted on a vacuumcompatible sample platen that allowed rapid transfer into the ultrahigh vacuum (UHV) surface analysis chamber. Two additional sets of samples were created from the same lot of unannealed and annealed Fe0 by gently stirring the finely divided powders in either trichloroethylene (TCE) or water saturated with trichloroethylene (H2O/TCE), each at room temperature for 48 h. Excess liquid was decanted and the samples were mounted on indium foil, as described above, and immediately transferred to the UHV chamber. The UHV chamber was operated at a base pressure of approximately 1 × 10-8 Pa, and all volatile adsorbates were immediately lost to the vacuum. The chamber was equipped with a Physical Electronics (Eden Prairie, MN) model 15-255G double pass cylindrical mirror analyzer, which could be operated in lockin or pulse count mode. Auger electron spectroscopy (AES) was measured as the dN(E)/dE spectrum with lock-in detection at a modulation energy of 2 eV, a primary Auger beam voltage of 3 kV in 1 eV intervals, and a scan rate of 1 eV/s. The beam size is approximately 1 mm in diameter, several times larger than the averaged Fe0 grain size. Unannealed samples were obtained from the manufacturer sieved from 50 mesh to dust (
