&EPA United States Environmental Protection Agency EPA/540/SR-93/509 August 1993 SUPERFUND INNOVATIVE TECHNOLOGY EVALUATION Technology Demonstration Summary Accutech Pneumatic Fracturing Extraction and Hot Gas Injection, Phase I Herbert S. Skovronek The Accutech Pneumatic Fracturing Extractionf™* (PFE)™ process evalu- ated in a SITE Program demonstration improves bedrock permeability for va- por extraction of volatile organic com- pounds by injecting bursts of com- pressed air into wells in the vadose zone, thereby creating new fractures and/or enlarging pre-existing fractures. Based on 4-hr tests, fracturing in- creased the extracted air flow rate 400% to 700%, averaging 600%. With the in- creased air flow rate and improved ac- cessibility, trichloroethene mass re- moval rate increased about 675% over that observed in a 4-hr test before frac- turing. When extracting from radially located monitoring wells, fracturing in- creased the extracted air flow rates 450% to 1,400% at wells 10 ft from the fracture well and 200% to 1,100% in wells at 20 ft, providing a significant improvement in the effective radius for extraction. With monitoring wells open as a passive source of air, even larger increases in air flow rate and TCE mass removal rate were achieved. The pro- cess is particularly useful where the vadose zone permeability is so low that conventional vapor extraction would not be effective. Using data developed in the 4-hr postfracture test, the estimated cost for a hypothetical 1-yr cleanup would be $307/kg ($140/lb) of TCE removed, including capital cost amortization. La- bor was the major cost factor (29%), followed by capital equipment (22%), collection and disposal of extracted VOCs (19%)f«ite preparation (11%), and residuals disposal (10%). Experiments to evaluate the effects of injecting heated air (-200 to 250 °F) into the vadose zone gave inconclu- sive results. This Summary was developed by EPA's Risk Reduction Engineering Laboratory, Cincinnati, OH, to announce key findings of the SITE program dem- onstration that is fully documented in two separate reports (see ordering in- formation at back). Introduction The Superfund Innovative Technology Evaluation (SITE) Program was instituted in 1986 to promote the development and application of innovative technologies to Superfund and other sites contaminated Printed on Recycled Paper ------- with hazardous wastes. This project was carried out under a New Jersey Environ- mental Cleanup Responsibility Act (ECRA) Cleanup Plan for an industrial site in Somerville, NJ where the soil and groundwater were found to be contami- nated with VOCs, primarily trichloroethene (TOE). The Plan calls for decontamination of the vadose zone by vapor extraction, where the formation is a shale with very limited air permeability. Process Description With Pneumatic Fracturing Extraction (PFE), new fractures may be created and existing fractures may be enlarged and/or extended by injecting bursts (10 to 20V sec) of compressed air (up to 500 psig) into narrow (2 ft) intervals of one or more wellbores. Each interval is isolated by a proprietary injector unit equipped with packers during pneumatic injection. The new fractures provide increased connec- tions and an enlarged radius of influence, thus making vapor extraction from the va- dose zone more efficient. Vapor extrac- tion can be carried out from all wells, or with some wells either left open to allow passive air introduction or capped to di- rect the subsurface air flow. The developer, Accutech Remedial Sys- tems, Inc., also is developing a catalytic oxi- dation system for the aboveground destruc- tion of chlorinated hydrocarbons, which may be investigated in a Phase II study. Accutech hypothesizes that further improvements in VOC removal rates can be achieved by in- jecting the hot exhaust gas (600 to 1000 °F) into one or more wells. The catalytic unit is not yet available, consequently a compressor was used in this Phase I study to produce hot air (200 to 250 °F) in order to develop and calibrate a model for hot gas injection and to evaluate the effects of injecting heated air. Test Program^ A series of radially placed, 6-in. diam- eter monitoring wells, and a central 4-in. diameter fracture well (FW), were drilled to a depth of about 20 ft (Figure 1). Each well was cased to about 8 ft be- low land surface (bis) and left open bore for the remaining depth. Approximately 2- ft intervals of the fracture well (FW) were sequentially isolated by packers and frac- tured by injecting a burst of compressed air (<500 psig) into each interval for 10 to 20 seconds. The packer assembly was then moved to the next interval and the fracturing process repeated. Strike BuMng Foundation O FMW5(20-) Dip • O FMW4(10') TMW3(T) TMW2(5') ® © © TMW1 (7') FMW3(10') FW FMW6(7.5') O © • O IW2(18') FMW8 (28') O FMW7(20') TMW4 (S-) O FMW2(W) FMW1 (10') LEGEND: • FW=Fracture Well O FMW = Fracture Monitoring Well © TMW - Thermal Monitoring Well © IW= Injection Well ( -') Distance to Fracture Well Pressures at monitoring wells (capped) and at the extraction well, extracted air flow rates, and TCE mass removal rates were compared before and after fractur- ing. A second prefracture test was carried out after a 24-hr dormant period to docu- ment the recharge effect commonly ob- served when vapor extraction is interrupted and then restarted. Samples of the ex- tracted gas were collected using EPA Method 18 and analyzed onsite by gas chromatography for TCE concentrations. These values were converted to TCE mass removal rates using the air flow rates. Surface heave during each fracture event was estimated with electronic tiltmeters. Pressures and air flow rates also were measured while extracting individually from each monitoring well to determine whether fracturing had established connections between the fracture well and the moni- toring wells and to provide information on the radius of influence created by fractur- ing. The effectiveness of passive air inlet was evaluated by uncapping from one to four monitoring wells while extracting from the fracture well. Pressures, air flow rates, and TCE removal rates were determined. Two experiments were carried out to evaluate the effects of hot gas injection. Compression heated air (~200 to 250 °F) was injected into a central well while ex- tracting from one or more monitoring wells. Temperatures in selected monitoring wells were measured and pressures, air flow rates, and TCE mass removal rates were determined for the extracted air. Results—Pneumatic Fracturing Extraction Tests A comparison of the 4-hr postfracture data with the data from the restart test demonstrated an air flow rate increase of between 400% and 700%, averaging about 600%. Although TCE concentrations after fracturing were only slightly higher than before fracturing (58 ppmv vs. 50 ppmv, avg), when coupled with increased air flow rates, the mass removal rate was in- creased by about 675% (Table 1 and Fig- ure 2). Table 1. Effects of Fracturing, Capped 4-hr Tests Test Pressure, Air Flow, TCE, psia scfm Ib x 1O6/min Prefracture Restart Postfracture Increase, % 11.0 11.1 11.4 — <0.6* <0.6* 4.2 >600* <10.8±1.0 <10.8±1.6 83.S±30.8 >67&> Figure 1. Well location diagram. * Accutech data indicate prefracture air flow rate <0.6 scfm. * % Increase = 100 x (Post - Pre)/Pre. ------- I I e 8 180 170 160 150 140 130 120 100 90 80 70 60 50 40 30 20 10 postfracture prefecture restart prefrai \cture j_ 0 20 40 60 80 100 120 140 160 180 200 220 240 Elapsed time, min Figure 2. TCE mass removal comparison—4-hr test. It was also found that a more complex gas mixture was extracted after fracturing, with higher concentrations of benzene, chloroform, and tetrachloroethene (Table 2). Fracturing may have improved con- nection with pockets of these compounds, making them more accessible for extrac- tion. Extraction at each peripheral monitor- ing well individually before and after frac- turing confirmed that connections were sig- nificantly improved even at wells 20 ft from the fracture well, as shown by ex- tracted air flow rates (Table 3). Attempts were made to determine whether vertical connections existed or were created by fracturing between adja- cent 2-ft intervals, but the data were in- conclusive, probably because of perched water in the vadose zone and the wellbore. Additional tests were carried out with monitoring wells left open as passive air inlets while extracting from the fracture well. In these experiments, even larger increases in air flow rates and TCE mass removal rates were observed after fractur- ing (Table 4). Results—Hot Air Injection Tests During the first hot gas injection test (90-hr), temperatures in the monitoring wells remained essentially constant over the first 10 hr at approximately 58°F. At that time the thermocouples were raised Table 2. VOCs in Extracted Air, Before and After Fracturing Contaminant Concentration, ppmv Before After Fracturing Fracturing Methylene chloride Chloroform c-1 ',2-Dichloroethene Trichloroethene Benzene Tetrachloroethene Toluene Xylene, m/p- Xylene, o- 1.4 3.5 U*(<3) 59.4 5.4 3.3 U (<3.3) U (<2.8) U (<2.8) 26.0 108.5 U (<12.5) 113.4 412.7 220.4 5.2J> U (<11.4) U(<11.4) * U = below detection limit. * J = estimated, below quantitation limit. Table 3. Monitoring Well Extraction Tests from 14 ft bis to 8 ft bis. Elevated tem- peratures were immediately observed and continued to increase over the next 10 hr before stabilizing (Table 5). It is unknown whether one or more thermocouples was immersed in water in the well at the 14 ft depth. In addition, only very low concentra- tions of TCE (~1 ppmv) were found in the extracted air, both before and during hot air injection. Even with the increased air flow rates during injection, the calculated TCE mass removal rate actually decreased during hot air injection (Table 6), possibly due to changes in the subsurface air flow directions when the system configuration was changed from extraction only to ex- traction and injection. A second experiment, lasting 24 hr, was carried out in another area of the site where higher TCE concentrations were anticipated. A new hot gas injection well and an additional extraction well were in- stalled (IW2, FMW8). Air was extracted from two wells (FMW6 and FMW8), each 10 ft from the injection well (see Figure 1). Although initial temperatures in the ex- traction wells were somewhat higher (~65 to 75 °F) than in the first hot gas injection test, well temperatures did not increase further in this case. Compared to an ex- traction-only pretest, TCE mass removal rate did increase about 50%, reflecting both increased airflow rates and increased TCE concentration in the extracted air (Table 7). Perched water in the wells may explain some of the inconsistencies in the air flow rate and temperature results from the two hot air injection experiments. Costs Operating and capital equipment costs provided by Accutech were coupled with the demonstration study to estimate the cost for remediation of a hypothetical site comparable to the demonstration site, with an estimated area of 100 ft by 150 ft or 15,000 ft2. With an effective extraction ra- Distance from FW.ft 10s* 10o/s 10 d 10s 20s 7.5 d 20 d Well No. FMW1 FMW2 FMW3 FMW4 FMW5 FMW6 FMW7 Air Flow, scfm pre- <.62* <.62-.88 <.62 <.62 <.62 <.88 <.62 post- 5.15-6.36 6.99-5.22 5.11-9.35 5.7-8.11 5.48-7.46 4.83-7.1 1.94-1.96 Increase, (post-pre) pre >7.3 - 9.2 >4.9 -10.3 >7.2-14.1 >8.2-12.1 >7.8-11.0 >4.5-7.1 >2.1 -2.2 * s = strike; d = dip; o/s = off strike and dip. " Some prefracture air flows are based on Accutech data. ------- Table 4. Passive Air Inlet Tests Test Prefecture Postfractura Increase, % Pressure, psia avg 10.8 14.6 Air Flow, TCE Mass Removed, scfm avg Ib x 10^/min 0.39tO.04 76.4±4.8 19,500 4.9*1.4 116±91.0 2,270 Table 5. Hot Air Injection Well Temperatures, 90 hr Well TMW2 TMW4 TMW1 TMW3 FMW4 FMW2 Distance from Injection Well, ft 5 5 7 7 10 10 Monitorinp Well Temperature. °F Initial to 10 hr @14ft 59 58 57 58 57 56 After 11 hr, @8ft 72 71 75 75 65 64 After 21 hr, @8ft 77 74 75 76 70 68 After 89 hr, ,@8ft 75 73 72 74 71 69 , dlus of 25 ft and 15% to 20% overlap, about 15 wells would be needed. Although serious limitations are recog- nized with such an approach to a cost analysis, the results of the 4-hr tests were extrapolated to a 1-yr cleanup effort. It was also assumed that the observed TCE concentration and mass removal rate would continue unchanged for a full year, even though a decrease in the rate of removal as the concentration in the for- mation decreases is more realistic. Accutech estimates that a fracturing sys- tem consisting of two injector/packer as- semblies, a bank of 12 air cylinders, a 12- hp compressor to recharge cylinders be- tween fractures, and auxiliary equipment would cost about $7,131/wkfor 2 wk while the 15 wells were fractured. A monitoring and analysis system (including a field GC, tiltmeters, datalogger, and supporting mi- crocomputers) would add an additional $6,656/Wk. The vacuum extraction system for si- multaneous extraction from 15 wells con- sists of a 40-hp vacuum blower capable of 500 scfm, associated piping, instrumen- tation, and a water knock-out vessel. The estimated cost for this system is $1,0907 wk, for the entire year of operation. The cost for pumping perched water out of the formation was included, but the cost of disposal was not since it was as- sumed that it would be air stripped to- gether with contaminated groundwater and the incremental cost would be very small. Well cuttings and a small amount of pro- tective gear will require disposal. The total cost for dewatering of the vadose zone and disposal of these materials was esti- mated at $37,200. Since Accutech's proposed catalytic oxi- dation unit is not yet available and has not been evaluated, carbon adsorption was selected for the extracted VOCs. The cost to adsorb the calculated 1209 kg (2,660 Ib) of TCE to be extracted over the year of operation and to remove and replace the carbon is estimated at about $70,000/yr. Personnel will be required throughout the year (24 hr/wk for 49 wk) to oversee the ongoing extraction and a more inten- sive effort (120 hr/wk) will be required during the 2 wk of fracturing. At an aver- Table 6. Hot Air Injection Test, 90 hr age rate of $65/hr, the total labor cost is estimated at $107,640/yr. The resulting cost for 1 yr of operation, during which 1209 kg (2,660 Ib) of TCE would be removed, is estimated at $371,364, equivalent to about $307/kg or $140/lb of TCE removed. Table 8 pro- vides a summary of the costs and the percent each subcategory contributes to total cost. No cost estimate was developed for the effect of hot gas injection; the planned catalytic oxidation unit was not used as the source of hot gas. Applicability to Other Sites Based on the demonstration and other information, the PFE process would ap- pear to be attractive for VOC-contami- nated formations with low permeability, such as most clays and shales. Studies have suggested that improvements in VOC extraction rates are also obtained with more permeable formations such as sands and silts, but the effects are not as great. There is no a priori reason to expect a fracture necessarily to intersect a pocket of contamination. And, as observed in this demonstration, fractures do not always propagate in the direction or to the dis- tances expected. Natural and man-made obstacles such as boulders, building foun- dations, buried pipelines, etc., can affect fracture propagation, provide undesirable paths, or decrease resistance to fractur- ing. This can result in surface failures during fracturing or the unexpected es- cape of vapors, as occurred during the SITE demonstration. Perched water also may hamper fracturing and/or interfere with Test Pre-hot air Hot air injection (one well extraction) Hot air injection (four wells extraction) Air Flow, Inject. 69.3*5.4 73.043.4 scfm avg Extract 11.6±1.S 55.8±3.4 82.6±7.1 TCE Mass Removed, Ib x 10*/min 172±18 20.4±32.0 31.2±10.3 Table 7. Second Hot Air Injection Test, 24 hr Test Air Flow, Inject. scfm avg Extract. TCE Mass Removed, Ib x 10*/min Pre-hot air inject Hot air inject (2 wells) Increase, % to26.r 3.7±1.8 9.2±4.7 150 63±27 97133 54 ' Some data lost due to leak in manifold; measured values ranged from 10.9 to 26,1. ------- Table 8. Operating Cost of Full Scale Pneumatic Fracturing Extraction Cost Item Site preparation Permitting/regulatory Capital equipment (1.5yr) Startup Labor salary Consumables/supplies Utilities Emission control Residuals (water, etc.) Analytical services Repair, replacement Demobilization Total Total Cost, $ 42,000 1,750 82,074 8,200 107,640 4,000 17,000 70,000 37,200 N/A N/A 1,500 $371,364 Cost/lb TCE, $/lb 15.79 0.66 30.85 3.08 40.47 1.50 6.39 26.32 13.98 _ — 0.56 139.60 % Of Total 11.3 0.5 22.1 2.2 29.0 1.1 4.6 18.8 10.0 — 0.4 100.0 air flow through the formation, thereby de- creasing VOC removal rates while increas- ing cost. Conclusions For properly selected formations, PFE can significantly improve vapor extraction effectiveness. The nature of the forma- tion, moisture content, air permeability, uniformity, water table, and the presence of obstacles or potential sources of short circuits must all be considered when evalu- ating PFE as a remediation option. In the demonstration, Accutech's claims were far exceeded: fracturing increased extracted air flow rates 400% to 700% and TCE mass removal rates by almost 700% when operating with a single frac- ture/extraction well and no air inlet sources. With passive air inlets, the air flow rate and the TCE mass removal rate after frac- turing are increased even more, 19,000% and 2,300%, respectively, when compared with the prefracture results. The radius of influence can be increased significantly by fracturing, with 1,100% to 1,400% increases in extracted air flow rates in wells at distances of 10 ft and 200% to 1,100% even in wells at 20 ft. The estimated cost for PFE remediation of a site such as that in Somerville, NJ, is $307/kg or $140/lb of TCE removed. La- bor, capital equipment, and emissions con- trol are the three major cost factors. The effects of hot air injection are in- conclusive. Increases in the temperature of the formation may be produced if suffi- cient heat is introduced, but this does not necessarily increase the TCE mass re- moval rate. •&U.S. GOVERNMENT PRINTING OFFICE: 1993 - 750-071/80050 ------- ------- ------- Herbert S. Skovronek is with Science Applications International Corporation, Hackensack, NJ 07601 Uwa Frank is the EPA Project Officer (see below). The complete report, entitled 'Technology Evaluation Report: SITE Program Demonstration Test—Accutech Pneumatic Fracturing Extraction and Hot Gas Injection—Phase I," (Order No. PB93-216596/AS; Cost: $17.50, subject to change) will be available only from: National Technical Information Service 5285 Port Royal Road Springfield, VA 22161 Telephone: 703-487-4650 A related report discusses the applications of the demonstrated technology: "SITE Program Applications Analysis Report: Accutech Pneumatic Fracturing Extraction and Hot Gas Injection—Phase I" (EPA/540/AR-93/509). The EPA Project Officer can be contacted at: Risk Reduction Engineering Laboratory U.S. Environmental Protection Agency (MS-106) Edison, NJ 08837-3679 United States Environmental Protection Agency Center for Environmental Research Information Cincinnati, OH 45268 Official Business Penalty for Private Use $300 BULK RATE POSTAGE & FEES PAID EPA PERMIT No. G-35 EPA/540/SR-93/509 ------- |