Laserfiche WebLink
dient experiment conducted+Mackay et al. (1986). create "04oncrete." The concrete mixture was <br /> For information on the general geology and hydrogeo- formed or broken into small cubical briquet-shaped <br /> logical characteristics of the site, the reader is referred forms, approximately 2 inches by 2 inches by 2 inches <br /> to Mackay et al.(1986),Sudicky(1986),and MacFarlane in dimension. As described below, both types of ORC <br /> (1983).For purposes of this paper,it is sufficient to note formulations were tested in this experiment. <br /> that the relatively uniform sandy aquifer is unconfined; According to the manufacturer, the total available <br /> water table elevation at this site varies between ground oxygen in the ORC powder used in this work was 6.5 <br /> surface and 2 in below ground surface, and hydraulic percent (t0.3 percent) by weight; after incorporation <br /> gradient in the area varies between approximately into the concrete,the available oxygen content remain- <br /> 0.0035 and 0.0054 throughout the year in response to ing in the ORC was 5.2 percent(±0.3 percent)by weight <br /> snowmelt and precipitation. (Plant Research Laboratories, unpublished data). PRL <br /> In previous studies of the fate and transport of BTEX reports that pure ORC powder, when activated by <br /> in the Borden Aquifer,it has been shown that biodegra- hydration, has a two-phase oxygen release profile. In <br /> dation occurs most rapidly when oxygen is the electron the first phase, after a brief lag time, the ORC begins <br /> acceptor. These studies also demonstrated that BTEX releasing oxygen at a rate approximating first order <br /> biodegradation during uncontrolled migration is limited kinetics. The duration of the first phase, as described <br /> primarily by oxygen availability (Barker et al. 1987). below, depends on environmental conditions. The sec- <br /> When sufficient oxygen and nutrients are available,bio- and phase involves a slower release rate that approxi- <br /> degradation is controlled by growth of naturally occur- mates zeroth order kinetics. Placement of ORC in con- <br /> ring BTEX-degrading microorganisms (Allen 1991). trete tends to truncate the rapid first order release phase <br /> However, when excess oxygen was supplied under and results in a more constant oxygen release through- <br /> natural conditions, nitrogen availability limited the out the entire lifetime of the ORC. <br /> BTEX biodegradation rate. This rate limitation was Laboratory studies conducted by PRL indicated that <br /> removed when nitrogen,supplied as nitrate,was added the release rate was increased by certain environmental <br /> (Allen 1991). Nitrogen-limited aerobic degradation of factors, including microbial activity, lower pH, and <br /> aromatic compounds in ground water samples has also higher temperature. Therefore, it was expected that <br /> been observed by Aelion and Bradley (1991) and by actual field oxygen release rates (as determined by dis- <br /> Armstrong et al. (1991). When anaerobic, denitrifying solved oxygen measurements) would differ somewhat <br /> conditions were induced in the Borden Aquifer by from laboratory results. Laboratory oxygen release <br /> nitrate addition, benzene was recalcitrant, and toluene, experiments with samples of the ORC concrete used in <br /> ethylbenzene, and xylenes were shown to be incom- this work suggested that oxygen would be produced for <br /> pletely biodegraded (Barbaro et al. 1993). approximately 13 months in a clay matrix and for <br /> Since the BTEX-degrading capabilities of the approximately 30 months in a loam matrix.These times <br /> microbes indigenous to the Borden Aquifer have are about four to five times longer than those estimated <br /> already been shown to be limited primarily by the availa- for the pure ORC powder. <br /> bility of dissolved oxygen, the Borden site serves as the Based on the laboratory studies conducted by PRL <br /> simplest case for field testing the ORC. If the use of using the ORC concrete and ORC powder that were <br /> ORC did not lead to an increase in the available oxygen used for this study, it was expected that the oxygen <br /> and therefore a decrease in benzene and toluene concen- release rates were approximately the following (in mg <br /> trations at this site,it would seem unlikely that it could of 02 per day per gram of ORC): 0.21 for the first 25 <br /> offer promise elsewhere. days and 0.12 for the next 375 days (ORC concrete), <br /> Because of knowledge gained from previous experi- and 1.6 for the first 4 days and 0.4 for the next 150 days <br /> ments conducted at the site (e.g.. Barker et al. 1987), no (ORC powder). The behavior of the ORC at the end <br /> effort was made in this study to confirm that the disappear- of the second phase is presently unknown, but presum- <br /> ance of benzene and toluene was in fact due tobiodegrada- ably the oxygen production diminished relatively <br /> tion. Thus, throughout the following discussion, it is quickly as the last of the available oxygen was released. <br /> assumed that the transformation process that is acting on Experimental Design <br /> these compounds is biologically mediated. <br /> The experimental design was chosen to maximize <br /> the information that could be obtained from a short <br /> ORC Characteristics duration test (Figure 1). A 5-m-wide source was used <br /> ORC is currently provided by the manufacturer to inject benzene,toluene,potassium nitrate(as a nutri- <br /> Plant Research Laboratories (Corona del Mar. Cali- ent). and sodium bromide (as a ground water tracer) <br /> fornia) as a fine-grained powder. Because the powder into the saturated zone. For this initial demonstration, <br /> itself is relatively impermeable to water, an alternate nitrate was added to decrease both the cost and duration <br /> formulation was desired which would allow water to of the experiment. The target injection concentrations <br /> flow through a porous solid source. One such formu- for benzene and toluene were 4 mg/L for each com- <br /> lation was previously co-developed by Dr. S. Koenigs- pound. The target injection concentrations of the inor- <br /> berg from Plant Research Laboratories(PRL)and Pro- ganic salts were 25 mg/L nitrate and 300 mg/L bromide. <br /> fessor R. Borden (North Carolina State University). In The source was made of 16 5-foot sections of <br /> that method. ORC powder was mixed with cement to 10-inch O.D. PVC well screen jetted to approximately <br /> WINTER 1994 GWMR ■ 121 <br />