Methods for characterizing the fate and effects of nano zerovalent iron during groundwater remediation.

The emplacement of nano zerovalent iron (nZVI) for groundwater remediation is usually monitored by common measurements such as pH, total iron content, and oxidation-reduction potential (ORP) by potentiometry. However, the interpretation of such measurements can be misleading because of the complex interactions between the target materials (e.g., suspensions of highly reactive and variably aggregated nanoparticles) and aquifer materials (sediments and groundwater), and multiple complications related to sampling and detection methods. This paper reviews current practice for both direct and indirect characterizations of nZVI during groundwater remediation and explores prospects for improving these methods and/or refining the interpretation of these measurements. To support our recommendations, results are presented based on laboratory batch and column studies of nZVI detection using chemical, electrochemical, and geophysical methods. Chemical redox probes appear to be a promising new method for specifically detecting nZVI, based on laboratory tests. The potentiometric and voltammetric detections of iron nanoparticles, using traditional stationary disc electrodes, rotating disc electrodes, and flow-through cell disc electrodes, provide insight for interpreting ORP measurements, which are affected by solution chemistry conditions and the interactions between iron nanoparticles and the electrode surface. The geophysical methods used for characterizing ZVI during groundwater remediation are reviewed and its application for nZVI detection is assessed with results of laboratory column experiments.

Field-scale transport and transformation of carboxymethylcellulose-stabilized nano zero-valent iron.

The fate of nano zerovalent iron (nZVI) during subsurface injection was examined using carboxymethylcellulose (CMC) stabilized nZVI in a very large three-dimensional physical model aquifer with detailed monitoring using multiple, complementary detection methods. A fluorescein tracer test in the aquifer plus laboratory column data suggested that the very-aggressive flow conditions necessary to achieve 2.5 m of nZVI transport could be obtained using a hydraulically constrained flow path between injection and extraction wells. However, total unoxidized nZVI was transported only about 1 m and <2% of the injected nZVI concentration reached that distance. The experimental data also indicated that groundwater flow changed during injection, likely due to hydrogen bubble formation, which diverted the nZVI away from the targeted flow path. The leading edge of the iron plume became fully oxidized during transport. However, within the plume, oxidation of nZVI decreased in a fashion consistent with progressive depletion of aquifer "reductant demand". To directly quantify the extent of nZVI transport, a spectrophotometric method was developed, and the results indicated that deployment of unoxidized nZVI for groundwater remediation will likely be difficult.

Effects of nano zero-valent iron on oxidation-reduction potential.

Oxidation-reduction potential (ORP) measurements have been widely used to assess the results of injection of nano zerovalent iron (nZVI) for groundwater remediation, but the significance of these measurements has never been established. Using rotating disk electrodes (RDE) in suspensions of nZVI, we found the electrode response to be highly complex but also a very sensitive probe for a range of fundamentally significant processes. The time dependence of the electrode response reflects both a primary effect (attachment of nZVI onto the electrode surface) and several secondary effects (esp., oxidation of iron and variations in dissolved H2 concentration). At nZVI concentrations above ∼200 mg/L, attachment of nZVI to the electrode is sufficient to give it the electrochemical characteristics of an Fe(0) electrode, making the electrode relatively insensitive to changes in solution chemistry. Lower nZVI concentrations give a proportional response in ORP, but much of this effect is mediated by the secondary effects noted above. Coating the nZVI with natural organic matter (NOM), or the organic polymers used to make stabile suspensions of nZVI, moderates its effect on ORP measurments. Our results provide the basis for interpretating ORP measurements used to characterize the results of injecting nZVI into groundwater.

Testing in EHS: What is the Current Status of Experimentation?

This paper explores some of the fundamental and practical issues related to the behavior of nanoparticles in the environment and their potential impacts on human health. In our research we have explored the reactive behaviors of nanoparticles with contaminants in the environment, how nanoparticle change in response to their environment and time, and how nanoparticles interact with biological systems of various types. It has become apparent that researchers often underestimate the difficulties of preparing and delivering well characterized nanoparticles for specific types of testing or applications. Difficulties arise in areas that range from not understanding what imparts the "nano" character of a particle to not knowing the impacts of minor species on the properties of high surface area materials. Some of our adventures and misadventures serve as examples of some of these issues as they relate to providing well defined particles for biological studies.

Redox behavior of magnetite: implications for contaminant reduction.

The factors controlling rates of contaminant reduction by magnetite (Fe3O4) are poorly understood. Here, we measured the reduction rates of three ArNO2 compounds by magnetite particles ranging from highly oxidized (x = Fe2+/Fe3+ = 0.31) to fully stoichiometric (x = 0.50). Rates of ArNO2 reduction became almost 5 orders of magnitude faster as the particle stoichiometry increased from x = 0.31 to 0.50. To evaluate what was controlling the rate of ArNO2 reduction, we measured apparent 15N kinetic isotope effects ((15)N-AKIE) values for nitrobenzene and magnetite open-circuit potentials (E(OCP)). 15N-AKIE values were greater than unity for all magnetite stoichiometries investigated, indicating that mass transfer processes are not controlling the rate of ArNO2 reduction by magnetite. E(OCP) measurements showed that the E(OCP) for magnetite was linearly related to the stoichiometry, with more stoichiometric magnetite having a lower potential. Based on these results, we propose that conceptual models that incorporate both redox and Fe2+ diffusion processes, rather than those that rely solely on diffusion of Fe2+, are more appropriate for understanding contaminant reduction by magnetite. Our work indicates that particle stoichiometry should be considered when evaluating rates of contaminant reduction by magnetite.

Degradation of 1,2,3-trichloropropane (TCP): hydrolysis, elimination, and reduction by iron and zinc.

1,2,3-Trichloropropane (TCP) is an emerging contaminant because of increased recognition of its occurrence in groundwater, potential carcinogenicity, and resistance to natural attenuation. The physical and chemical properties of TCP make it difficult to remediate, with all conventional options being relatively slow or inefficient. Treatments that result in alkaline conditions (e.g., permeable reactive barriers containing zerovalent iron) favor base-catalyzed hydrolysis of TCP, but high temperature (e.g., conditions of in situ thermal remediation) is necessary for this reaction to be significant. Common reductants (sulfide, ferrous iron adsorbed to iron oxides, and most forms of construction-grade or nano-Fe(0)) produce insignificant rates of reductive dechlorination of TCP. Quantifiable rates of TCP reduction were obtained with several types of activated nano-Fe(0), but the surface area normalized rate contants (k(SA)) for these reactions were lower than is generally considered useful for in situ remediation applications (10(-4) L m(-2) h(-1)). Much faster rates of degradation of TCP were obtained with granular Zn(0), (k(SA) = 10(-3) - 10(-2) L m(-2) h(-1)) and potentially problematic dechlorination intermediates (1,2- or 1,3-dichloropropane, 3-chloro-1-propene) were not detected. The advantages of Zn(0) over Fe(0) are somewhat peculiar to TCP and may suggest a practical application for Zn(0) even though it has not found favor for remediation of contamination with other chlorinated solvents.

Natural organic matter enhanced mobility of nano zerovalent iron.

Column studies showed that the mobility of nanometer-sized zerovalent iron (nZVI) through granular media is greatly increased in the presence of natural organic matter (NOM). At NOM concentrations of 20 mg/L or greater, the nZVI was highly mobile during transport experiments in 0.15-m long columns packed with medium sand. Below 20 mg/L NOM, mobility of the nZVI was less; however, even at 2 mg/L the nZVI showed significantly increased mobility compared to the no-NOM case. Spectrophotometric and aggregation studies of nZVI suspensions in the presence of NOM suggest that sorption of the NOM onto the nZVI, resulting in a reduced sticking coefficient, may be the primary mechanism of enhanced mobility. Modeling the mobility of nZVI in porous media with filtration theory is challenging, but calibration of a simple model with experimental results from the column experiments reported here allows simulation of transport distances during injection. The simulation results show that the increased mobility due to NOM combined with the decrease in mobility due to decreased velocity with distance from an injection well could produce an injection zone that is wide enough to be useful for remediation but small enough to avoid reaching unwanted receptors.

Oxidation of chlorinated ethenes by heat-activated persulfate: kinetics and products.

In situ chemical oxidation (ISCO) and in situ thermal remediation (ISTR) are applicable to treatment of groundwater contaminated with chlorinated ethenes. ISCO with persulfate (S2O8(2-)) requires activation, and this can be achieved with the heat from ISTR, so there may be advantages to combining these technologies. To explore this possibility, we determined the kinetics and products of chlorinated ethene oxidation with heat-activated persulfate and compared them to the temperature dependence of other degradation pathways. The kinetics of chlorinated ethene disappearance were pseudo-first-order for 1-2 half-lives, and the resulting rate constants-measured from 30 to 70 degrees C--fit the Arrhenius equation, yielding apparent activation energies of 101 +/- 4 kJ mol(-1) for tetrachloroethene (PCE), 108 +/- 3 kJ mol(-1) for trichloroethene (TCE), 144 +/- 5 kJ mol(-1) for cis-1,2-dichloroethene (cis-DCE), and 141 +/- 2 kJ mol(-1) for trans-1,2-dichloroethene (trans-DCE). Chlorinated byproducts were observed, but most of the parent material was completely dechlorinated. Arrhenius parameters for hydrolysis and oxidation by persulfate or permanganate were used to calculate rates of chlorinated ethene degradation by these processes over the range of temperatures relevant to ISTR and the range of oxidant concentrations and pH relevant to ISCO.

Characterization and properties of metallic iron nanoparticles: spectroscopy, electrochemistry, and kinetics.

There are reports that nano-sized zero-valent iron (Fe0) exhibits greater reactivity than micro-sized particles of Fe0, and it has been suggested that the higher reactivity of nano-Fe0 may impart advantages for groundwater remediation or other environmental applications. However, most of these reports are preliminary in that they leave a hostof potentiallysignificant(and often challenging) material or process variables either uncontrolled or unresolved. In an effort to better understand the reactivity of nano-Fe0, we have used a variety of complementary techniques to characterize two widely studied nano-Fe0 preparations: one synthesized by reduction of goethite with heat and H2 (Fe(H2)) and the other by reductive precipitation with borohydride (Fe(BH)). Fe(H2) is a two-phase material consisting of 40 nm alpha-Fe0 (made up of crystals approximately the size of the particles) and Fe3O4 particles of similar size or larger containing reduced sulfur; whereas Fe(BH) is mostly 20-80 nm metallic Fe particles (aggregates of <1.5 nm grains) with an oxide shell/coating that is high in oxidized boron. The FeBH particles further aggregate into chains. Both materials exhibit corrosion potentials that are more negative than nano-sized Fe2O3, Fe3O4, micro-sized Fe0, or a solid Fe0 disk, which is consistent with their rapid reduction of oxygen, benzoquinone, and carbon tetrachloride. Benzoquinone-which presumably probes inner-sphere surface reactions-reacts more rapidly with FeBH than Fe(H2), whereas carbon tetrachloride reacts at similar rates with FeBH and Fe(H2), presumably by outer-sphere electron transfer. Both types of nano-Fe0 react more rapidlythan micro-sized Fe0 based on mass-normalized rate constants, but surface area-normalized rate constants do not show a significant nano-size effect. The distribution of products from reduction of carbon tetrachloride is more favorable with Fe(H2), which produces less chloroform than reaction with Fe(BH).

Electrochemical properties of natural organic matter (NOM), fractions of NOM, and model biogeochemical electron shuttles.

In this study, cyclic voltammetry was used to characterize the redox properties of natural organic matter (NOM). Using a stationary platinum working electrode, minimal concentrations of electrolyte, and dimethyl sulfoxide (DMSO) as the solvent, we were able to resolve two pairs of oxidation and reduction peaks for a fraction of Georgetown NOM that is enriched in polyphenolic moieties (NOM-PP). Applying our method to other fractions of Georgetown NOM, and to samples of NOM from a wide range of other sources, gave cyclic voltammograms (CVs) that generally contained fewer distinguishing features than those obtained with NOM-PP. For comparison, CVs were also obtained using our method on six quinone model compounds: anthraquinone-2,6-disulfonate (AQDS), lawsone, juglone, menadione, menaquinone-4, and ubiquinone-5. The CVs of these quinones were similar in shape to the CV of NOM-PP, consistent with the notion that quinones are the dominant redox-active moieties associated with NOM. Quantitative analysis of the peaks in these CVs showed that the peak potentials (Ep) were separated by more than 0.059 V and that the peak currents (i(p)) were linearly related to the square root of the scan rate (v0.5) and concentration (C) for both NOM-PP and the model quinones. Equivalent results were obtained with a rotating Pt disk electrode. From this we conclude that NOM-PP and the model quinones undergo similar sequences of two one-electron, quasi-reversible, diffusion controlled, electron transfers at the Pt electrode surface in DMSO. Although it is difficult to relate these results to Nernstian standard potentials vs the standard hydrogen electrode (SHE) under aqueous conditions, it is clear that the apparent formal potential for NOM-PP lies between the corresponding potentials for menadione and juglone and well above that of AQDS. Attempts to derive correlations between Ep and i(p) for the NOMs with quantifiable electrode response and other measurable properties of NOM (including trace metal content and UV-vis absorbance) did not yield any strong relationships.