The key objective of the project was to determine whether the application of gaseous ammonia (NH3) to unsaturated soils can be effective for increasing soil potential of hydrogen (pH) and subsequently for treating 1,2,3-trichloropropane (TCP), 1,2-dichloropropane (1,2-DCP), 1,3-dichloropropane (1,3-DCP), dibromochloropropane (DBCP) and other priority contaminants subject to alkaline hydrolysis. A secondary objective was to determine if addition of NH3 to soils promotes a cometabolic polishing effect due to induction of the enzyme ammonia monooxygenase (AMO) within microorganisms present near the treatment zone. If this effect could be documented, it would represent a separate mechanism of contaminant destruction, and possibly a separate treatment approach using lower NH3 concentrations.
he reactive gas process entails injection of a blend of air and NH3 in order to raise the pH of soil and promote the destruction of contaminants via alkaline hydrolysis. When NH3 is added to soil, it combines with H2O to produce ammonium ion (NH4+) and hydroxide ion (OH-), subsequently increasing soil pH. The pKa of this reaction is 9.25 at 25oC. The process may also stimulate cometabolic biodegradation reactions mediated by the enzyme AMO as a secondary effect as pH declines or at the edge of the reactive zone. Thus, this approach may have utility for treatment of any contaminant subject to alkaline hydrolysis and may also promote cometabolic treatment as a polishing step.
Five column studies were conducted with Brown and Bryant Superfund Site (B&B) site soils to evaluate the potential to increase soil pH by adding different concentrations of NH3 (5% or 9.5% in air) to an inlet port under continuous flow. After 8-14 days (depending on the study; 5-6 g total NH3 added to each column), columns were sectioned, and subsamples were collected to quantify the soil pH and concentration of NH3-N along the column length. The effect of soil moisture on pH was also evaluated. In all studies, pH increases from ~8.2 to 10 were observed as ammonia levels increased from ~1,500 to 2,000 mg NH3-N/kg (baseline values at B&B Site) to ~4,000 mg NH3-N/kg. The pH increased very little thereafter, with a measured soil pH of <10.5 at ammonia levels exceeding 10,000 mg-N/kg. The biphasic curve is consistent with the pKa of the NH3/NH4++ reaction. The data suggest that it will be easy to increase soil pH to 10 in the field but difficult to bring soil pH much above this value. Based on the tests, the conditions under which the greatest quantity of soil (at typical field moisture) was impacted by the gaseous ammonia was the addition of 5% NH3 at a flow rate of 10 standard cubic centimeters per minute (sccm). These conditions were subsequently selected to evaluate the effectiveness of the NH3 addition process on treatment of TCP, 1,2-DCP, and 1,3-DCP in B&B soils.
For the contaminated soil studies, B&B core material was spiked with TCP, 1,2-DCP, 1,3-DCP and DBCP. Baseline soil samples showed mean soil concentrations of 5,700 μg/L for TCP, 322,000 μg/L for 1,2-DCP, 1,600 μg/L for 1,3-DCP, and 20,000 μg/L for DBCP. The “Treated Column” then received 5 % NH3 in air at a flow rate of 10 sccm for 15 days, after which time the soil was stored for ~ 30 days to provide additional time for hydrolysis of chlorinated propanes to occur. A second “Control Column” was constructed with the same design except that it was supplied with nitrogen gas (N2) (no NH3) at 10 sccm for 15 days as a means to evaluate volatile rather than destructive losses of the contaminants. The soil was also incubated for an additional 30 days and then sampled as described for the Treated Column. In the NH3-treated column, the final soil pH was ~ 10.2 over the entire length of the soil column. Conversely, in the N2-treated soil column, the soil pH was ~8.6 in those extracted with deionized (DI) water over the entire length of the soil column.
The data from the Treated Column indicated nearly a complete loss (99.6 % – 100%) of the four different halogenated propanes added to the soil columns compared to the initial concentrations. However, the data from the Control Column showed a similar loss percentage for each of the compounds, ranging from 92.3 % to 99.7%. Thus, the data indicate that the passage of gas (NH3 or N2) through the soil columns likely resulted in significant physical stripping/removal from the soil phase. Only a small percentage of the contaminants were trapped by adsorbent tubes placed at the end of the columns, so it was not possible to obtain a reasonable mass balance.
Batch microcosms were prepared to evaluate the potential for cometabolic degradation of chlorinated propanes at the B&B site via ammonia oxidation. The initial objective of this study was to assess whether nitrification could be stimulated in B&B soils with the addition of different quantities of NH3. However, the exceedingly high NH3 in the soil during baseline conditions (i.e., ~2,500 mg-N/kg soil) prevented the execution of the study as planned. Rather, the project team evaluated whether nitrification was ongoing in the soils under in situ conditions with the high NH3 present. The initial data, particularly the nitrite data (an intermediate product in nitrification), suggested the potential for activity, but nitrite only increased at the first sample time, and decreased thereafter. Moreover, the high levels of nitrate evident in the soil (which may have resulted from nitrification of ammonia or discharge of nitrate-containing agricultural products) largely prevented an accurate assessment of the occurrence of nitrification via measuring increases in soil nitrate. If appreciable nitrification activity had been observed over the 28-week study in the Live (but not the Killed) microcosms, soil samples were to be collected and reanalyzed for AMO and ammonia oxidizing archaea to evaluate increases in the relevant organisms, and then the microcosms were to be spiked with 1,2,3-TCP, 1,2-DCP, and 1,3-DCP to evaluate cometabolic biodegradation of the compounds. However, the microcosm data did not show definitive nitrification activity (other than the small initial increase in nitrite) when Live and Killed samples were compared, and the contaminated soil column results indicate that the gas addition process is likely to strip the halogenated propanes from the soil matrix, so the study was terminated at 28 weeks.
The nearly complete loss of the halogenated propanes in the column treated with N2 gas for 15 days is problematic for the field implementation of this approach because the data suggest that it may not be possible to discern losses of these contaminants to hydrolysis from losses due to simple stripping from the soil phase. Modeling suggests that several months of NH3 gas addition at 5% would be required to increase soil pH over a 10 – 15 ft radius from the gas injection wells in the field at relatively high flow rates (~10,000 to 30,000 sccm). Extrapolating from the column tests, this addition is likely to increase soil pH to >10 over the treatment area as desired, but it is also likely to strip a high percentage of the chlorinated propanes in the process. Thus, the technology will most likely be effective at removing most of the contaminants within the treatment radius of influence, but much of this removal may reflect volatilization rather than hydrolysis. At a minimum, the influence of the two processes would not be independently quantifiable. As a result, the technology was not scaled for field implementation at the B&B site.