Environ. Sci. Technol. 2005, 39, 3639-3645
Persistence and Biodegradation of Monoethanolamine and 2-Propanolamine at an Abandoned Industrial Site STEVEN B. HAWTHORNE,* A L E N A K U B AÄ T O V AÄ , J O H N R . G A L L A G H E R , JAMES A. SORENSEN, AND DAVID J. MILLER Energy and Environmental Research Center, University of North Dakota, Campus Box 9018, Grand Forks, North Dakota 58201-9018
Soil and groundwater samples were collected at the site of a former chemical processing plant in areas impacted by accidental releases of MEA (monoethanolamine) and IPA (2-propanolamine or isopropanolamine). Although their use had ceased ca. 10 years before sample collection, soils collected at contamination sites had MEA concentrations ranging from ca. 400 to 3000 mg/kg and IPA concentrations from ca. 30 to 120 mg/kg. Even though alkanolamines are miscible in water, transport to groundwater was slow, apparently because they are present in soil as bound cations. Only one groundwater sample (near the most highly contaminated soil) from wells directly adjacent to and downgradient from the contaminated soils had detectable MEA, and none had detectable IPA. However, ammonia was found in the soil samples collected in the MEAcontaminated areas (ca. 500-1400 mg/kg) and the groundwater (80-120 mg/L), as would be consistent with bacterial degradation of MEA to ammonia, followed by transport of ammonia into the groundwater. Counts for bacteria capable of using MEA or IPA as a sole carbon source were ca. 5 × 106 and 1 × 106 (respectively) per gram in uncontaminated site soil, but no such organisms were found in highly contaminated soils. Similarly, bacterial degradation of MEA in slurries of highly contaminated soils was slow, with ca. 8-20 days required for half of the initial concentrations of MEA to be degraded at 20 °C and 30-60 days at 10 °C. In contrast, bacterial degradation studies using uncontaminated site soils spiked with ca. 1300 mg/L either MEA or IPA showed very rapid degradation of both compounds, with more than 99% degradation occurring in less than 3 days with quantitative conversion to ammonia, followed by slower conversion to nitrite and nitrate. The results obtained in the site soils, the groundwater samples, and from the biodegradation studies demonstrate that MEA and IPA can persist for decades on soil at high (hundreds of mg/kg) concentrations without significant migration into groundwater, despite the fact that they are miscible in water. Since MEA and IPA exist primarily as cations at the pH of site soils, their persistence apparently results from strong binding to soil, as well as inhibition of natural bioremediation in highly contaminated field soils. * Corresponding author phone: (701)777-5256; fax: (701)777-5181; e-mail:
[email protected]. 10.1021/es040556c CCC: $30.25 Published on Web 04/13/2005
2005 American Chemical Society
Introduction Alkanolamines are used in the production of a variety of chemical products including surfactants, cosmetics, metalworking fluids, textiles, and agricultural chemicals. An estimated 1.5 billion kg of ethanolamines was used worldwide in 2001, with about one-half of the production in the United States (1). Alkanolamines are also commonly used for the removal of acid (or sour) gases (hydrogen sulfide and carbon dioxide) from natural gas, a process known as natural gas sweetening (2, 3). In 1993, approximately 20 million kg of alkanolamines was used for natural gas sweetening in the United States (3). This volume has likely increased in recent years as more sour natural gas reservoirs have been exploited to meet the growing demand for natural gas in North America. Geographically, the use of alkanolamines is widespread, with an estimated 500 alkanolamine-based acid gas removal plants located throughout the gas producing regions of North America (4). The use of alkanolamines can be expected to increase significantly as natural gas production increases, and if proposed, processes to remove carbon dioxide from fossil fuel combustion flue gases for underground disposal and stabilization are implemented (4). Alkanolamines consist of ethanol-, 2-propanol-, and butanol-substituted amines, all of which include both amino and alcohol functional groups. Because of the amine functionality, they are basic compounds with acid dissociation constant (pKa) values in the higher pH range. For example, monoethanolamine (MEA) and 2-propanolamine (isopropanolamine or IPA) have pKa values of 9.68 and 9.66, respectively (2), and therefore exist primarily as protonated cations at pH values lower than 9. In the past, waste alkanolamines have been introduced into the subsurface environment through accidental releases or disposal into unlined pits (5). However, little data exist on the fate of alkanolamines in the environment. Their Henry’s Law and vapor pressure values are very low; therefore, alkanolamines released to the environment would likely partition between soil and water rather than vaporize (1). Since alkanolamines are miscible in water, partitioning models have predicted their preferential partitioning from soil to water, and therefore, they might be expected to transport to groundwater (2) as has been reported for diisopropanolamine (DIPA) at a natural gas processing site (6). However, such models do not consider the fact that alkanolamines are moderately strong bases and therefore will be protonated cations under most environmental pH conditions. Since soils have cation exchange capabilities, such models may underpredict the extent of sorption to soil that actually occurs in the environment (7). For example, measured Koc values for MEA spiked on soils as high as 4000 have been reported (8), as compared to a Koc of 5 predicted by partitioning models (2). DIPA mobility has been shown to vary widely on different soils, with KD values for natural soils ranging from 0.5 to 4 and increasing to 40 for montmorillinite clay (6). Partitioning of DIPA to wetland plants has also been reported (9). Biodegradation of alkanolamines has been reported by a number of researchers under laboratory conditions (2, 4, 8, 10-17), as well as in the field in a 500 m3 engineered bio-pile (18). The field bio-pile study was conducted with nutrient addition and moisture control in an effort to maximize degradation. Although no degradation rates have been reported for alkanolamines under field conditions where no action has been taken to stimulate biodegradation, high concentrations of alkanolamines have been reported to VOL. 39, NO. 10, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3639
FIGURE 1. Relative locations of soil boring (+) and groundwater sampling wells (O). S-1 (etc.) refers to soil boring locations, W-1d (etc.) to deep wells, and W-2s (etc.) to shallow wells. persist in subsurface soils for several years at a natural gas sweetening plant in Alberta (19). In an effort to better understand the fate of alkanolamines at historically contaminated field sites, soil and groundwater samples were collected from a former chemical storage and mixing site that had used MEA and IPA until its abandonment approximately 10 years before sample collection. MEA, IPA, and ammonia concentrations were measured in the soil and water samples, and the biodegradation behavior of MEA and IPA in field soil/water slurries was determined.
Materials and Methods Site Samples. Soil and groundwater samples were collected at a chemical handling facility in the northeast United States that had not been used for approximately 10 years. Soil borings were performed at (and adjacent to) locations where MEA and IPA had previously been stored in aboveground tanks or transported and mixed into product formulations (Figure 1). Soils were collected from each boring at ca. 0.5 m and at ca. 1.5-2.5 m from the surface. Background soils were collected in adjacent areas where no alkanolamine contamination was expected. Groundwater samples were also collected from monitoring wells located in areas suspected to be contaminated with alkanolamines and in areas not expected to be impacted. Soil samples were primarily silty clay and sand; however, some samples also included gravel and general construction residue. Depth to groundwater was approximately 3-3.5 m throughout the sampling locations. The average groundwater velocity is estimated to be less than 1 m per year, which indicates that 3640
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 10, 2005
rapid dilution and removal of any alkanolamine contaminants via groundwater flow is likely minimal. Average precipitation in the area is 110 cm/year. All soils and water samples were immediately cooled upon collection with ice and shipped overnight to the laboratory where they were stored in the dark at 4 °C for a maximum of 14 days before analysis. Soils were sieved to
