United States Environmental Protection Agency Water Engineering Research Laboratory Cincinnati OH 45268 Research and Development EPA/600/S2-88/020 Apr. 1988 <>EPA Project Summary Control of Volatile Organic Contaminants in Groundwater by In-Well Aeration Judith A. Coyle, Harry J. Borchers, Jr., and Richard J. Miltner At a 0.1 million gallon per day well contaminated with several volatile organic compounds (VOC's), principally trichloroethylene (TCE), several in-well aeration schemes were evaluated as control technologies. The well was logged by the USGS to define possible zones of VOC entry. A straddle packer and pump apparatus were utilized to isolate those zones and define their yield and level of VOC concentration. The technical literature together with this knowledge of the well were used to design an air lift pump. Operation of the air lift pump confirmed literature prediction of its low wire- to-water efficiency. Removal of TCE did not exceed 65%. Mass transfer occurred in the pump's eductor. Air lift pumping coupled with in-well diffused aeration increased TCE removal to 78%. When in-well diffused aeration was used with an electric submersible pump, TCE removal averaged 83%. In the latter two schemes, mass transfer occurred utilizing the well as a countercurrent stripper. These technologies are limited by the volume of air that can be transferred to the well (air-to-water ratios below 12:1) and the cost of compressing air under high head. Thus, these technologies are not cost-effective compared to packed tower aeration. They are, however, quickly put on-line, easy to operate, and can serve as short-term remedies while above-ground technologies are under design and construction. This Project Summary was developed by EPA's Water Engineering Research Laboratory, Cincinnati, OH, to announce key findings of the research project that is fully documented in a separate report of the same title (see Project Report ordering information at back). Introduction Contamination of groundwater with volatile organic compounds (VOC's) has become common throughout the United States. Southeastern Pennsylvania and the North Penn Water Authority (NPWA) have not escaped this problem. In 1979, a large amount of trichloroethylene (TCE) was spilled in nearby Collegeville, PA. The Authority sampled all 34 of their operating wells and found 8 to be contaminated with TCE and other VOC's. These wells were shut down, resulting in a loss of approximately one-third of the system's total pumping capacity. Since that time NPWA has been actively pursuing various methods of dealing with the VOC problem. Surface water was purchased from a neighboring water supplier, a granular activated carbon treatment plant was installed at one well, and packed tower aerators were installed at others. This investigation was ------- carried out to evaluate in-well aeration techniques. This investigation covered the design and operation of various in-well aeration configurations examined by NPWA during the time period of January 1982 to May 1985. The configurations included air-lift pumping with and without in- well diffused aeration, i.e., sparging, as well as electric submersible pumping with in-well diffused aeration. The well selected for this study was Lansdale number 8 (well L-8). This well was heavily contaminated with VOC's and was being pumped to waste in an attempt to control the contamination plume. In addition to TCE, well L-8 contained vinyl chloride; carbon tetra- chloride; tetrachloroethylene; cis-1,2- dichloroethylene; 1,1 -dichloroethylene; and 1,1,1-trichloroethane. It is 286 ft deep and in a mixed resi- dential/commercial area, with homes in close proximity to the well house. Shortly after the discovery of VOC pollution at NPWA, a series of preliminary tests were performed using an air lift pump and an electric submersible pump with a sparger. No attempt was made to record air-to- water ratios or operating conditions; however, even under these uncontrolled conditions, nearly 80% removal was observed for TCE. This provided the incentive for further study of in-well aeration. In-Well Aeration The air lift pump used for this study was similar in design to pumps used by the Lartsdale Municipal Authority (predecessor to NPWA) in the 1920's. Compressed air was introduced by an air line into an open-ended pipe in the well called an eductor. The aerated water in the eductor was less dense than the surrounding water in the well and was, therefore, forced up the eductor and out of the well as a result of the density gradient. Mass transfer of VOC's occurred in the eductor. Air and stripped VOC's were removed in an open tank at the surface called a separator. When sparging, a pipe was used to introduce compressed air into the well at the desired depth. VOC mass transfer occurred utilizing the well as a counter- or cocurrent stripper. The in-well aeration equipment used in this study was easily constructed from materials already in NPWA stock. Objectives When a well is found to be contaminated it is common practice to pump to waste in order to prevent the contamination plume from spreading throughout the aquifer. In-well aeration can treat the water as it is being pumped. This may reduce the amount of pollutants being discharged into the sewer system as at well L-8 or, in some cases, may treat the water to potability. Treating the water while pumping eliminates the need for construction of above-ground treatment devices. This investigation was undertaken to evaluate the cost effectiveness of in-well aeration as an alternative to above-ground technologies. Well Characterization The first stage of the in-well aeration system design in this investigation was characterization of well L-8.This was done by the U.S. Geological Survey (USGS) with a series of well loggings. These tests included caliper, conductivity, temperature, radiological, and brine trace logs, among others. The USGS study determined possible water entry zones. Inflatable straddle packers and a pump were placed in the well to isolate these zones. Each zone was analyzed for VOC's and specific capacity. Three different depths for in-well aeration equipment were evaluated at well L-8, based on the results of the well characterization and air lift pump theory. Scope of Work Parameters measured in the field during in-well aeration testing included air pressure, air temperature, air flow rate, water flow rate, and water level in the well. These parameters allowed calculations of pumping efficiencies and air-to-water ratios. The in-well aeration systems tested were evaluated based on these findings, as well as on VOC removal and cost. Certain secondary effects of in-well aeration treatment techniques were also examined. Off-gases in the well house were tested to determine whether hazardous conditions were present. The air outside the well house in the adjacent residential area was also tested for VOC's. Bacteriological changes as well as corrosion related factors (changes in pH or dissolved oxygen with aeration) were examined. Results Well Characterization The USGS well logging identified t major and five minor potential water en zones into the well. The packer test accounted for 81% of the well's spec capacity. That capacity was observed the upper 200 ft of the well. T remaining 19% of the specific capac was contributed by zones not isolated packer testing. Seventy-four percent the well's specific capacity was in i upper 130 ft. Differences in V( concentrations were observed in t isolated zones. The two most heav VOC-contaminated zones were abc 130 ft and were also the largest wa producing zones. An open borehole pumping t( showed that VOC concentration chang considerably with time. Over short-te tests, such as the in-well aeration te performed, large concentration variatic could be expected. Footpiece Tests Two air lift pump footpieces w« tested. In one, the air line was op ended and produced large bubbles in 1 eductor; in the other, the end of the line was coupled to a diffusing dev and produced small bubbles. The t footpieces showed no significe difference in air lift pump efficieni There was no difference in VOC remo brought about by changing from a lar bubble to small bubble footpie suggesting that small bubbles coalesc above the footpiece. The small bubl footpiece caused greater operati pressure. The pressure difference w greatest at higher air-to-water rati (5:1 to 12:1). The most efficient operati of an air lift pump was found to be agreement with the literature at a mi lower air-to-water ratio (1.5:1) wh« the pressure differences between t footpieces were very small. If the air pump was operated at its maximi pumping efficiency, there would be 111 difference in operating pressure betwe the two footpieces. If, however, the pur were operated at a higher air-to-wa ratio in order to obtain better V( removal, the small bubble footpie would have greater operating pressi and greater operating cost. Since th( was no difference in VOC remo1 between the two footpieces, and sir the small bubble footpiece would potentially more expensive to opera the small bubble configuration w abandoned and the large bubt ------- footpiece was used for all in-well aeration testing. When sparging in the well, large and small bubble air lines, identical to the air lines used for air lift pumping, were compared. The small bubble sparger had a higher operating pressure, and therefore, a higher operating cost than the large bubble sparger. There was no differences in VOC removal between the large and small bubble spargers. As with the air lift pump, the large bubble sparger was used for all testing. Reproducibility Raw water VOC concentration varied within a given test, which confirmed the findings of the pumping test conducted during well characterization. Even though VOC concentrations varied over time in the short term, test results were found to be reproducible from one day to the next, thereby giving confidence to the procedures employed. In the short term, static water levels were consistent. Test results generally were not reproducible when conducted months apart. Raw water VOC concentrations varied from one test to another over time. Over the long term, static water levels changed. It is possible that with changes in static water level, the yield within a water entry zone changed slightly and made reproducibility difficult. Changes in static water level .cause changes in submergence which, in turn, cause changes in pump efficiency. A given air flow rate produced different water flow rates and air-to-water ratios over the long term. Air Lift Pump Tests Based on well characterization, the air lift pump was studied at 130, 200, and 280 ft depths. The 130 ft depth coincided with 65% submergence, which is reported as optimum for air lift pump efficiency. Operation of the air lift pump confirmed its highest efficiency at 65% submergence. The maximum efficiency was found to be 30% to 35%. The efficiency decreased as submergence increased, also confirming the predictions of the literature. VOC removal was poorer at the 280 ft setting than it was at 130 ft or 200 ft. Best VOC control for the air lift pump ranged from 90% for vinyl chloride (VC) with the highest Henry's Law constant to 47% for cis- 1,2-dichloroethylene (cis-1,2-DCE) with the lowest constant. Percent removal of the other VOC's was consistent with their Henry's Law constants. TCE was 65% removed by air lift pumping. This level of control occurred at the higher, more expensive, air-to-water ratios. Tests with Sparging and Air Lift Pumping In these tests, the air lift pump was located at 130, 200, and 280 ft depths. While the air lift pump was fixed in the well, the air sparger was located at 130, 200, or 280 ft depths. Sparging air into the well decreased the pumping efficiency of the air lift pump because the density gradient between the well and the eductor was diminished. However, the efficiency of the combined devices was higher than if all of the air had been delivered to the air lift pump alone. Therefore, in terms of efficiency and cost, it was better to operate an air lift pump and sparger combination than the air lift pump alone. The air lift pump and sparger combination yielded VOC removal percentages ranging from 99% for VC to 65% for cis-1,2-DCE, with TCE having 78% removal. A higher air-to-water ratio was obtained by using the air lift pump and sparger combination than by using the air lift pump alone. This accounted for the higher VOC removal. The highest air-to-water ratios obtained were 10.6:1 for the air lift pump alone and 17:1 for the air lift pump and sparger combination. When sparging, the air-to-water ratio was limited to the point at which water actually bubbled out of the well. In a well with a wider bore, the air-to-water ratio might be higher because water would not be forced out of the well as readily. At well L-8, some of the cross section was taken up by test equipment, e.g., sample pump and water level probe, which would not be in the well during regular operation. No significant removal differences were observed when the equipment was operated at different depths. This was attributed to poor reproducibility of sparger tests over long periods of time. Tests with Sparging and Electric Pumping An electric submersible pump was operated at 200 ft with sparger testing being conducted at 130, 200, and 280 ft. The best VOC control was observed with the sparger at 130 ft. Sparging at the 200 ft depth gave the poorest control. VOC control was consistent with what was expected from well characterization. Sparging at 130 ft caused counter- current stripping as water from the most contaminated zone waspulled past the bubbles on its way to the pump. A counter- and cocurrent stripper would have been created by the 280 ft sparger, with at least some of the air being pulled into the pump before it reached the most heavily contaminated zone. With the sparge directly adjacent to the pump, most of the air could have been pulled into the pump before any stripping occurred in the well. The VOC control obtained during electric pump and sparger tests averaged 83% for TCE, 80% for cis-1,2-DCE, and 93% for VC. These removals were better than those achieved by air lift pumping with or without a sparger. The air-to-water ratio used to achieve this level of control was 8.2:1, which is lower by half than the maximum air-to-water ratio used in the tests with the air lift pump and sparger. Better control resulted from directing available air to the in-well sparger than by directing all or a portion of it to the air lift pump. Secondary Effects All configurations of in-well aeration increased the pH by an average of 0.4 pH units as carbon dioxide was stripped. Dissolved oxygen (DO) was raised to saturation by all of the in-well aeration methods tests as a result of air introduced under high head. Water entering the separator was bubbly in appearance and actually milky white when sparging, but all of the bubbles were released by the time the water passed from the separator. Bacteriological testing of raw and treated water was inconclusive, with large variations in bacterial counts masking any trends. The R2A method provided consistently higher recovery of organisms than the heterotrophic plate count. Air sampling showed that in-well aeration would probably not cause air quality problems of industrial hygiene concern; however, it may be considered an air pollution source and require the appropriate permits. Venting of VOC off gases from the separator, and from the well bore when sparging, may be prudent. Conclusions In-well aeration is limited by the amount of air that can be transferred to the well and the cost of compressing air under high head. With limited air-to- water ratios, removal will not reach that achievable with above-ground technologies. In-well aeration can be a useful treatment technique for VOC removal on a short-term emergency basis. Electric ------- submersible pumping with the use of a sparger is particularly well suited to this application. The addition of an air compressor and the installation of an air sparger was completed for this study in a matter of a few days with readily available materials. The sparger should not be placed directly adjacent to the pump intake as this will draw the bubbles into the pump and VOC stripping in the well will be minimized. Air should be added in slowly-increasing amounts until the foaming water is just visible below the well head. This will produce the greatest possible air-to-water ratio. Both an air and water separator and repumping to the distribution system are necessary. A chlorine contact chamber might be easily modified for this purpose. The time required to build or put an off-the-shelf tank in place as a separator may negate the usefulness of in-well aeration as a short-term emergency technology. While the cost to compress air may reach 25C/1.000 gallons depending on the depth of the sparger, the total cost, assuming 3 mo emergency service, may reach $1.90/1,000 gal at 0.1 MGD under NPWA conditions. While well characterization was useful during this project, both for experimental design and data interpretation, it would not be a prudent investment in an emergency situation. Optimum location of the sparger could be more cost- effectively determined by trial and error. As with any aeration technology, air quality must be considered. The necessity to treat off gases to remove VOC's or to vent off gases to limit human exposure could negate its advantages as a quickly-installed emergency technology. Finally, water saturated with DO may be corrosive to some distribution system materials, even in the short term. This too could negate the advantages of in- well aeration for emergency treatment. The full report was submitted in fulfillment of CR 809758 by the North Penn Water Authority under the sponsorship of the U.S. Environmental Protection Agency. U. S. GOVERNMENT PRINTING OFFICE: 1988/548-158/67110 ------- usciJUUI Judith A. Coyle and Harry J. Borchers, Jr., are with the North Penn Water Authority, Lansdale, PA 19446; the EPA author Richard J. Miltner (also the EPA Project Officer, see below) is with the Water Engineering Research Laboratory, Cincinnati, OH 45268. The complete report, entitled "Control of Volatile Organic Contaminants in Groundwater by In-Well Aeration," (Order No. PB 88-180 1121 AS; Cost: $19.95, subject to change) will be available only from: National Technical Information Service 5285 Port Royal Road Springfield, VA 22161 Telephone: 703-487-4650 The EPA Project Officer can be contacted at: Water Engineering Research Laboratory U.S. Environmental Protection Agency Cincinnati, OH 45268 ^JJ .;} United States Environmental Protection Agency Center for Environmental Research Information Cincinnati OH 45268 BULK RATE POSTAGE & FEES PAID EPA PERMIT No. G-35 Official Business Penalty for Private Use $300 EPA/600/S2-88/020 0000329 PS ll S EMVIR MOTECTXOII *Gi«CY — ^ j*. «L« tf ft V B D A tUP V P IMV IVH J L. Jt P T\ W « * __ clicieo1** it 60604 ------- |