Final Report: Mechanistic Investigations of Fe(0) Reactions with Organohalides

EPA Grant Number: R828164
Title: Mechanistic Investigations of Fe(0) Reactions with Organohalides
Investigators: Roberts, A. Lynn , Fairbrother, D. Howard
Institution: The Johns Hopkins University
EPA Project Officer: Klieforth, Barbara I
Project Period: September 1, 2000 through August 31, 2002
Project Amount: $225,000
RFA: Exploratory Research - Engineering, Chemistry, and Physics) (1999) RFA Text |  Recipients Lists
Research Category: Water , Land and Waste Management , Air , Safer Chemicals

Objective:

The overall objective of this research project was to systematically investigate several poorly understood aspects of alkyl and vinyl halide reactions with zero-valent iron (Fe0). The specific objectives of this research project were to: (1) investigate how experimental variables, such as iron mass and surface area loading, buffer composition, initial pH, and mixing method affect the rate of organohalide removal in Fe0 batch systems; (2) determine reaction rate constants, product distributions, and predominant reaction pathways for the reduction of alkyl polyhalides by Fe0 in batch systems, while examining the effect of organohalide structure on Fe0 reactivity; (3) develop reactivity correlations (e.g., linear free energy relationships) for the reactions of alkyl halides with Fe0 by correlating kinetic data with alkyl halide molecular descriptors, such as one-electron reduction potentials (E1), carbon-halogen bond dissociation energies (DRX), and lowest unoccupied molecular orbital energy level (ELUMO); (4) examine the potential for inter- and intraspecies competitive effects among alkyl and vinyl halides in column studies with commercial grade granular iron; and (5) design a liquid cell coupled to a surface analysis chamber and determine the influence of Fe surface characteristics on the remediation of organohalides.

Each of these areas has been explored in this research, providing an increased understanding of the chemical processes fundamental to Fe0 permeable reactive barriers (PRBs). Furthermore, the results obtained from this research project will reduce uncertainty regarding the likelihood of success for the application of this emerging remediation technology to previously unstudied alkyl halide contaminants.

Summary/Accomplishments (Outputs/Outcomes):

Effect of Experimental Variables on Rates of Organohalide Reduction

In granular iron batch systems, it commonly is assumed that the rate of organohalide removal will vary in a linear and first-order manner with respect to the granular iron surface area present in the system. This assumption is the basis for the surface-area-normalized kinetic model that most often is applied to reactions with granular iron. In this model, experimentally measured pseudo-first-order rate constants for organohalide removal (kobs) are normalized by granular iron surface area to yield a surface-area-normalized rate constant (kSA) for organohalide reduction. Despite the widespread use of this kinetic model, this assumption upon which it is based previously has never been rigorously tested.

To test the validity of the surface-area-normalized kinetic model, batch systems for a variety of alkyl halide species were conducted over a range of iron mass and surface area loadings. Initial batch experiments investigating the reaction of 1,1,1-trichloroethane (1,1,1-TCA) with high purity electrolytic iron powder revealed that the relationship between kobs for 1,1,1-TCA reduction and iron surface area loading (rhoa in m2/L) was in fact nonlinear, as kobs was found to be proportional to rhoa0.5. As a result, kSA values for 1,1,1-TCA reduction varied by as much as one order of magnitude over the range of rhoa investigated. This half-order relation also was observed in batch systems in which buffer composition, initial system pH, mixing method, and mixing rate were systematically varied, suggesting that these experimental parameters were not responsible for the nonlinear result. Furthermore, similar nonlinear, nonfirst-order correlations between kobs and rhoa were obtained for several other alkyl halide species investigated.

This result is of great importance in the iron PRB field, as it calls into question the universal application of kSA values to granular iron systems. kSA values obtained from laboratory batch and column studies can be used in PRB modeling and design, as well as in the development of reactivity correlations and intersystem reactivity comparisons. The application of kSA values obtained at low values of rhoa in laboratory studies to larger, field-scale Fe0 systems, in which rhoa is much higher, could compromise the efficiency of contaminant removal, potentially increasing the possibility of contaminant breakthroughs at PRB sites. Therefore, it is necessary that either rate constants other than kSA values are used to characterize organohalide reactivity in Fe0 systems, or that kSA values are applied only to systems in which a linear, first-order correlation between kobs and rhoa has been observed experimentally.

This result also can be used to gain insight into the mechanism of alkyl polyhalide reduction by Fe0. For reactions with 1,1,1-TCA, it was possible to obtain a linear, first-order relationship if kobs was correlated with the aqueous phase FeII concentrations ([FeII(aq)]) present in batch systems. A linear, first-order correlation also was observed when kobs was related to the amount of protons reduced in batch systems ([H+red]), a value that was estimated by monitoring the change in system pH as a function of time. As [FeII(aq)] and [H+red] are measures of the extent of Fe0 corrosion in batch systems (as both are products of Fe0 corrosion), the linear correlations obtained with these variables suggest that the reduction of alkyl polyhalides is occurring directly as a result of Fe0 corrosion, most likely within a region where an oxide overlayer on the iron particle surface is relatively thin. Identification of the surface site responsible for electron transfer and alkyl halide reduction may result in further advances of PRB technologies that will increase the overall efficiency of contaminant removal afforded by iron PRBs.

Determination of Reaction Rate Constants and Product Distributions for Alkyl Halides

Batch experiments exploring the reduction of several C1 (containing one carbon center) and C2 (containing two carbon centers) alkyl halides with electrolytic iron powder revealed the following reactivity trends. Rates of alkyl halide removal were observed to increase with an increasing degree of halogenation. Therefore, highly halogenated alkyl polyhalides (e.g., hexachloroethane), were far more reactive than lesser halogenated species (e.g., 1,2-dichloroethane). Furthermore, alkyl bromides and mixed halogenated species containing bromine substituents were several orders of magnitude more reactive than their chlorinated counterparts.

The predominant reaction pathway observed for C2 alkyl polyhalides was reductive ß-elimination. For species that did not possess the necessary vicinal halide substituents (i.e., halogens located on neighboring carbon centers) for ß-elimination, hydrogenolysis and reductive ß-elimination were typically observed. Ethylene and ethane were commonly observed as the major reduction products for lesser halogenated alkyl halides. Highly substituted C2 alkyl halides, on the other hand, often resulted in the production of far less-reactive halogenated alkene intermediates, allowing for the possibility that toxic, more persistent reduction byproducts may be formed as a result of reaction with Fe0. Product species and reaction pathways were not extensively investigated for C1 alkyl halides.

These experimental results provide valuable rate and product distribution information for a class of organic contaminants that thus far has remained largely uncharacterized in the literature. The availability of reliable data developed in this study can be used by engineers to assess the effectiveness of iron PRBs toward the remediation of an alkyl halide contaminant plume, and could result in increased application of field scale Fe0 PRBs. Additionally, information obtained regarding alkyl halide reduction pathways and product formation will aid in determining if hazardous, persistent degradation byproducts will be generated as a result of reduction by Fe0. In such cases, the net reduction of human health risk afforded by PRB implementation would be compromised if deleterious product formation were a possibility.

Development of Linear Free Energy Relationships

Linear free energy relationships for both C1 and C2 alkyl polyhalides revealed strong correlations of kobs for alkyl halide reduction with the following alkyl halide molecular descriptors: one electron reduction potential (E1), carbon-halogen bond dissociation energy (DRX), and lowest unoccupied molecular orbital energy level (ELUMO). Strong correlations also were observed with rate constants for alkyl halide reduction by CrII(aq), FeII porphyrin, and juglone, all of which are model, one- or two-electron aqueous phase reductants.

The strength of the correlations obtained with E1 and ELUMO suggest that the rate-limiting step in alkyl halide reduction by Fe0 is electron transfer. Furthermore, the presence of a strong correlation with DRX suggests that the rate-limiting electron transfer is coupled with a concerted dissociation of the weakest carbon-halogen bond on the alkyl halide contaminant. Finally, the strong correlation of kobs with rate constants for alkyl halide reduction by CrII(aq), a model, aqueous phase one-electron reductant, suggests that the electron transfer in the rate-limiting step is a one-electron transfer. This claim is supported by the fact that a strong correlation was observed between kobs and E1 values, although no such correlation was observed between kobs and the two-electron reduction potential (E2), which would be expected if a two-electron transfer process were occurring. Therefore, based on these correlations, the rate-limiting step in alkyl halide reduction is a dissociative, one-electron transfer that results in the formation of a carbon-centered radical (Equation 1).

R - X + e- — Rº + X-     (1)

These correlations not only lend valuable insights into the reaction mechanism of alkyl halide reduction by Fe0, but they also represent powerful tools for predicting alkyl halide reactivity in iron systems. These reactivity correlations can be used by engineers and regulators to predict the reactivity in granular iron systems of new compounds not yet investigated in laboratory studies.

Column Studies of Inter- and Intraspecies Competitive Effects

Column studies, conducted with commercial grade Connelly granular iron, were used to investigate the intra- and interspecies competitive effects among alkyl and vinyl halide species. Such competition could be observed in Fe0 PRBs employed to treat groundwater plumes comprised of complex contaminant mixtures at relatively high aqueous phase concentrations. The vinyl halide species investigated in these column studies were trichloroethylene (TCE), cis- and trans-dichloroethylene (DCE), and 1,1-DCE. The alkyl halides 1,1,2-trichloroethane (TCA) and 1-bromo-2-chloroethane (BrClEta) also were studied.

Alkyl halides were found to be more reactive than vinyl halide species in column reactors, as the following reactivity sequence was observed: BrClEta > 1,1,2-TCA > trans-DCE > TCE > cis-DCE >> 1,1-DCE. The major reduction products in most cases were ethane and ethylene, although TCE was reduced primarily to cis-DCE, while vinyl chloride was the major product of 1,1,2-TCA reduction.

Results of the column studies suggest that vinyl halides (e.g., TCEs and DCEs) are capable of inhibiting the reduction of both alkyl and vinyl halide species via interspecies competitive effects. TCE was found to be the most potent interspecies competitor, while cis- and trans-DCE showed only moderate competitive ability. The alkyl halides BrClEta and 1,1,2-TCA, on the other hand, were not observed to inhibit the reduction of either alkyl or vinyl halide species. They were, however, very susceptible to inhibition by vinyl halide species. Similarly, intraspecies competitive effects were most pronounced for vinyl halides, while no such competitive effects were observed with alkyl halides.

The insights into intra- and interspecies competition obtained from these column studies will allow for improved design and more accurate modeling of granular iron PRBs. In particular, for sites in which PRBs are employed to treat groundwater plumes comprised of complex contaminant mixtures at high aqueous phase concentrations, such information will be crucial to properly characterizing Fe0 reactivity. Ideally, the higher quality design and modeling afforded by these column studies will help to minimize the number of ineffective field-scale applications of granular iron PRBs, while optimizing organohalide removal efficiency.

Surface Chemical Investigations

Experimental evidence indicates that the elementary steps in Fe-based reductive transformations of organohalides are mediated by surface reactions. Understanding the molecular level events responsible for Fe-based dehalogenation is, therefore, important in developing the mechanistic insight needed to anticipate reaction rates and product distributions. As a result of the importance of these interfacial processes, there has been an increasing interest in using surface analytical techniques to study the reaction of organohalides with iron. One limitation of this approach is the ex situ nature of the surface measurements, which exposes the sample to the atmosphere and consequently adds unwanted uncertainty because of the potential for surface oxidation and contamination from background species.

To overcome these problems, we have designed and implemented a liquid cell coupled to a surface analysis chamber in an approach that minimizes sample exposure to the atmosphere. A glass reaction vessel is coupled to an ultra-high vacuum (UHV) chamber with capabilities for x-ray photoelectron spectroscopy (XPS) and surface modification. Reactions at the liquid/solid interface were initiated by bringing an iron surface into contact with a deoxygenated solution containing an organohalide, while an overpressure of nitrogen in the reaction vessel maintains an anoxic environment throughout the course of the reaction. The sample then can be transferred into the UHV environment for surface analysis without exposure to air. This approach also can be used to prepare model surfaces in the analysis chamber and study their subsequent reactivity towards halocarbons in solution.

Through analysis of the Fe(2p) XPS region on iron foils, initial studies in the cell showed that acid washing is responsible for a significant reduction in the thickness of the native oxide overlayer present on iron. Thinning of the iron oxide overlayer may represent an important component for improving the reactivity of iron particles because the initial step in the reduction process is believed to involve electron transfer through the oxide overlayer. Reaction studies involving iron foils, however, were hampered by a number of significant experimental difficulties. These included the requirement to synthesize a nonvolatile organohalide that would: (1) not partition into the gas phase; and (2) yield product species that could be detected by available solution phase analytical techniques, specifically high performance liquid chromatography (HPLC). In addition, a significant pump-down time was required after the sample had been removed from the liquid cell, typically on the order of several hours. This greatly reduced the effective experimental duty cycle and also meant that unwanted surface oxidation and contamination, while lessened when compared to traditional methods of surface analysis, was still evident. Furthermore, the limited surface area of iron foil used, reduced the reaction rate to a level that required several days of contact time to produce a measurable change in the solution phase composition. As a result of these problems, the cell proved ineffective for studying the influence of controlled modifications to the iron surface on reaction rates. The cell has proved to be effective in improving our molecular-level understanding of environmental redox reactions involving iron pyrite, for which issues of unwanted surface oxidation in particular could be eliminated.


Journal Articles on this Report : 1 Displayed | Download in RIS Format

Other project views: All 3 publications 2 publications in selected types All 2 journal articles
Type Citation Project Document Sources
Journal Article McGuire MM, Carlson DL, Vikesland PJ, Kohn T, Grenier AC, Langley LA, Roberts AL, Fairbrother DH. Applications of surface analysis in the environmental sciences: dehalogenation of chlorocarbons with zero-valent iron and iron-containing mineral surfaces. Analytica Chimica Acta. 2003;496(1-2):301-313.
full text available
R828164 (Final)
R828771C006 (2003)
not available

Supplemental Keywords:

zero-valent iron, permeable reactive barriers, groundwater remediation, chlorinated organic solvents, environmental chemistry, and engineering., RFA, Scientific Discipline, Toxics, Waste, Sustainable Industry/Business, National Recommended Water Quality, cleaner production/pollution prevention, Sustainable Environment, Chemistry, Technology for Sustainable Environment, Hazardous Waste, Engineering, Groundwater remediation, Hazardous, Environmental Engineering, reaction engineering, halogens, environmental chemistry, reactivity, Volatile Organic Compounds (VOCs)

Relevant Websites:

http://www.jhu.edu/~dogee/roberts/ Exit

Progress and Final Reports:

Original Abstract
  • 2001 Progress Report