FIELD EVALUATION OF PACKED COLUMN AIR STRIPPING PENSACOLA, FL November 18, 1986 Compound: Tetrachloroethylene by Michael D. Cummins January 1987 U.S. ENVIRONMENTAL PROTECTION AGENCY Office of Drinking Water Technical Support Division Cincinnati, Ohio 45268 ------- TABLE OF CONTENTS Page 1 1 1 2 3 4 5 6 6 8 List of Tables: 1 Field Operating Data 2 Tetrachloroethylene Data Set 3 Data Analysis Results 4 Cost Estimate for 98.5% Removal System 5 " •• •• 99.9% •• " List of Figures: 1 Packed Column Air Stripping Process 2 EPA Pilot System 3 Concentration Profiles 4 System Configuration References: Table of Contents Introduction Packed Column Air Stripping Process Mass Transfer Theory Pilot System Quality Control Field Evaluation Data Analysis Equipment Size Cost Estimate ------- Field Evaluation - Pensacola, FL Tetrachloroethylene Introduction: On November 18, 1986, the Technical Support Division (TSD), Office of Drinking Water, EPA conducted a field evaluation of the removal of tetrachloroethylene from contaminated ground water via packed column air stripping. The field evaluation was conducted at municipal well #8 in Pensacola, FL, which was contaminated with 150 to 250 ug L'1 of tetrachloroethylene. The field evaluation was requested by the State of Florida, Department of Environmental Regulation (DER). The objective of the field evaluation was to determine the feasibility of producing water containing tetrachloroethylene at 3 ug L*1 or less from ground water contaminated at levels of 150 ug L*1 (98.5% removal) and 3000 ug L"1 (99.9% removal), and to estimate the equipment sizes and costs necessary to achieve these objectives. The TSD has developed a mathematical model to aid in evaluating removal of volatile organic compounds via packed column air stripping. The model has been verified using field evaluations in the States of Wisconsin, Pennsylvania, New York, Massachusetts, Missouri, Minnesota, New Jersey, Virginia, Louisiana, and Florida. Prior to this report, tetrachloroethylene had been evaluated in the first 6 states at ground water temperatures ranging from 9 to 16 deg C. The TSD's objective for the Pensacola field evaluation was to verify the model's prediction of tetrachloroethylene removal at 24 deg C. The results of the field evaluation are summarized in this report. Packed Column Air Stripping Process: If water contaminated with volatile organic compounds (VOCs) is brought in contact with uncontaminated air, some of the VOC molecules will transfer to the air. In the packed column air stripping process, this transfer is facilitated as air and water are continuously replenished and mixed together in a countercurrent flow pattern (see Figure 1). (1) Contaminated water is pumped to the top of a column, distributed at the top, and cascades down through a bed of packing material. .. ^ (2) Uncontaminated air is blown in at the bottom of the column and forced up through the same bed of packing material. (3) Packing material provides a combination of a large surface area to provide mixing of air and water, contact time for ------- VOC molecules to transfer from water to air, and a large void volume to minimize energy loss of the air system. (4) As contaminated water cascades down through the column VOC molecules are transferred to the air. (5) Air and VOCs are released to the atmosphere at the top of the column. The concentration of VOC in air released at the top of the column is less than the original concentration of VOC in water due to the large air to water volume ratio. The concentration of VOCs in air is further reduced by dispersion into the atmosphere. (6) The countercurrent flow process provides contact of the most contaminated air and water at the top of the column and the cleanest air and water at the bottom of the column, maximizing efficiency. Mass Transfer Theory: The theory of mass transfer in a packed column has been well developed in the chemical engineering literature (Ref 1). From mass transfer theory, the author developed Equation 1 which can be used to predict the liquid phase concentration at any point along the packing height. In this field evaluation thirty (30) VOC concentrations were measured at five (5) vertical locations and six (6) air loadings. Equation 1 was fitted to the field data by adjusting the top of packing concentration (Xt), mass transfer coefficient (Kla), and Henry's coefficient (H) such that the deviation between Eq. 1 and the field data was minimized. There are a total of nine (9) terms in Equation 1 required to predict the liquid phase concentration (X). Concentration Profile with Unstrippable Fraction: X = Xt * ( Fu * (1-C) + C ) Eq. 1 Where : C-( (R*B)-1)/ ( (R*A) -1) A = EXP(( Zb ) * (Kla/L) * ((R-l)/R) ) B - EXP((Zb-Z) * (Kla/L) * ((R-l)/R) ) R=(G/L)*(H/Pt) Xt: VOC concentration at top of packing (ug L"1). The concentration at the top of the packing material (Xt) is approximately the raw water concentration; however, there may be some VOC loss as water travels through the well pump, the piping between the well and the air stripping system, and the liquid distribution system at the top of the packed column. This VOC loss is collectively referred to as the influent end effect. In this field evaluation Xt was determined by regression analysis. It ------- was not necessary to evaluate the raw water concentration or the influent end effect. Fu: VOC unstrippable fraction (fraction of Xt). The unstrippable fraction (Fu) is used to account for anomalies that are observed in real systems. The unstrippable fraction may be due to short circuiting of water through the packed column, background VOC contamination in the influent air, a physical or chemical complex holding the VOC in the liquid phase, cross contamination between sample bottles, or other unknown phenomenon. In general the unstrippable fraction is a fraction of Xt that, for some reason or another, cannot be removed by the air stripping process at even the highest air to water ratios or the tallest packing heights. In this field evaluation no unstrippable fraction was observed. Thus, Fu was set to 0. G: Air loading (m3 m"2 sec"1). The air loading term (G) is the total air flow through the column per unit of cross- sectional area of the column. L: Liquid loading (m3 m"2 sec'1). The liquid loading term (L) is the total liquid flow through the column per unit of cross-sectional area of the column. H: Henry's coefficient (atm m3 water m'3 air). Henry's coefficient (H) is a physical-chemical property that expresses the volatility of 'the particular VOC. Henry's coefficient is dependent only on the temperature and molecular properties of the VOC and not dependent on the other eight (8) terms in Equation 1. In this field evaluation Henry's coefficient for tetrachloroethylene was determined by regression analysis. Pt: Operating pressure (atm). The operating pressure (Pt) is generally 1 atmosphere. Zb: Packing height (m). The term Zb is the total height of the packing material. Z: Vertical location within column (m). The term Z is the vertical location within the column measured from the top of the packing material. Kla: Mass transfer coefficient (sec'1). The mass transfer coefficient (Kla) expresses the overall rate of VOC transfer from the liquid phase to the air phase. The mass transfer coefficient is dependent on the VOC molecular properties, packing material properties, liquid loading, and air loading. The mass transfer coefficient is not dependent on the concentration terms Xt and Fu or the packing height terms Zb and Z. In this field ------- evaluation the mass transfer coefficient was determined by regression analysis. X: Predicted VOC concentration (ug L"1). The predicted VOC concentration at location Z is a function of the other nine parameters. Pilot System: The pilot system, shown in Figure 2, consists of a 24 ft tall, 2 ft diameter aluminum column packed with 18 ft of 1 inch plastic saddles. The pilot system was constructed in 3 ft tall sections to facilitate transportation to the field. At Pensacola a 2 inch fire hose was used to connect well #8 to the influent line of the pilot system. Mounted on the influent line of the pilot system was a control valve, orifice plate, and mercury manometer to control and monitor the liquid flow. Water was pumped through the liquid flow control system to the top of the pilot system where four upturned 2 inch elbows distributed the liquid onto the packing material. Water cascaded down through the packing material and was collected in an effluent tank at the bottom of the column. A 4 inch fire hose was used to discharge the effluent water to a storm sewer. Influent air was drawn into the system at the bottom of the column, passed up through the packing material to the top, and returned to ground level through a 6 inch air duct. Mounted inside the air duct was a pitot tube, an orifice plate, and a control damper to monitor and control the air flow. Following the control damper the air duct was connected to the intake side of a blower. Air was discharged at ground level 30 ft downwind from the air intake. Water manometers were installed above the top of packing material and near the pitot tube to measure air pressure. Eighteen sample ports were installed at 1 ft intervals along the column height to sample liquid from the center 1 ft of the column. The sample ports were designed such that air was not withdrawn with the sample. This sampling system permitted monitoring the concentration profile of VOCs predicted by Eq. 1 along the column height. Five sample ports, selected at 3 ft intervals, were used in this evaluation. The pilot system was operated at steady state for 30 minutes before samples were collected. Quality Control: Prior to conducting the field evaluation at Pensacola, a contract was established with Southwest Research Institute (SwRI), San Antonio, Texas, for the bulk of the sample analyses. Resources were also reserved at the EPA, TSD laboratory in Cincinnati, OH, for analysis of duplicate samples. Laboratory sample numbers were assigned to field run numbers/sample port locations. The ------- laboratory sample numbers were used by the laboratories and the run numbers /sample port locations were used by the field crew. Information on the sample labels included the sampling date, city name, preservative used, project officer's name, and laboratory that would receive the sample bottle. All of the labels were prepared by a computer plotter using waterproof ink. All sample bottles were spiked with mercuric chloride as a preservative. The operation and sampling of the packed column air stripping system were conducted by the EPA project engineer and EPA equipment mechanic who together designed and constructed the pilot system. All samples were collected by only one individual. At the end of the field operation all samples to be sent to SwRI laboratory were repacked in shipping containers, preserved with ice, and sent via overnight Federal Express delivery. Samples were analyzed within three weeks using EPA Method 501.2, liquid-liquid extraction with electron capture detector. Twelve (12) bottles were filled with laboratory organics-free water, labeled "Blank", and packed with the sample bottles so that cross contamination between sample bottles could be evaluated. Six (6) blank samples were analyzed by SwRI for tetrachloroethylene and none indicated tetrachloroethylene. Thus cross contamination was not observed. Six samples were collected for blind duplicate analysis to evaluate the relative standard error associated with sampling, shipping, and analysis. These samples were given a fictitious sample port location, packed with the other samples, and sent to SwRI for analysis. SwRI analyzed all of the blind duplicate samples. The relative standard error was 18% which is considered to be acceptable for EPA Method 501.2. A different set of six duplicate samples were collected to evaluate the accuracy of the laboratory analysis. This set of duplicate samples were sent to TSD for analysis and the original samples sent to SwRI for analysis. All pairs of duplicate samples were analyzed. The percent difference between the TSD analysis and the SwRI analysis indicated that on the average the SwRI analysis were lower than the TSD analysis by 15%. This indicates that the tetrachloroethylene concentration may be higher than reported by SwRI. However, this does not impact the removal efficiency observed in the pilot system. A total of forty-two (42) samples were sent to SwRI for analysis. All were successfully analyzed. Field Evaluation: The liquid and air loadings for the Pensacola field evaluation were selected using the results from prior field evaluations. Based on prior field evaluations and the physical/chemical properties of tetrachloroethylene, the Henry's coefficient in this ------- situation should be in the range 0.3 to 0.6 atm m3 water m"3 air. Using the lower value as an estimate for Henry's coefficient and the Onda correlation (Ref 2) to predict the mass transfer coefficient, the model indicated that 99% tetrachloroethylene removal should be achieved using a liquid loading of 0.020 m3 m'2 sec"1 (30 gpm ft'2), an air to water ratio of 18 to 1, and a packing height of 5.3 m (17.5 ft). This liquid loading and six air to water ratios were selected for evaluation. The air to water ratio for the first run was selected so a high removal efficiency would be obtained. The air to water ratios for the sequential runs were stepwise reduced such that the air flow in the sixth run would result in low removal efficiency. The air to water ratios were 35:1, 18:1, 10:1, 5:1, 2.5:1, and 1.5:1, respectively. The field operating data are shown in Table 1. The concentration data for tetrachloroethylene are shown in Table 2. The first sample port is located 0.15 m (0.5 ft) from the top of the packing material. The sample ports are located at about 3 ft intervals along the column height with the last sample port (5.34 m) located 0.15 m (0.5 ft) from the the bottom of the packing. The tetrachloroethylene concentration decreased as water passed through the column (Table 2 and Figure 3). Data Analysis: The top of packing concentration (Xt), mass transfer coefficient (Kla), and Henry's coefficient (H) were determined by adjusting their values such that the deviations between Eq 1 and the concentration data were minimized. The solid lines in Figure 3 are the best fit concentration profiles. The overall relative standard error between the best fit concentration profiles and the data was 19% which was within the range of the relative standard error associated with sampling, shipping, and laboratory analysis. A relative standard error of 19% represents a good fit between a math model and data as observed in Figure 3. The results of the data analysis are presented in Table 3. The best fit value for Xt, Kla, and H are 146 ug L'1, 0.0185 sec'1, and 0.39 atm m3 water m"3 air. respectively. The 95% confidence intervals are 133 to 160 ug L , 0.0172 to 0.0196 sec'1, and 0.32 to 0.46 atm m3 water m*3 air respectively. These confidence intervals represent tolerances of +/- 9.2, 6.5, and 18 percent of the respective best fit values. The best fit removal efficiencies are 99.15%, 98.9%, 98.3% 95.6%, 82%, and 58% for the six air to water ratios, respectively. The best fit effluent tetrachloroethylene concentrations from the first three runs were 1.2, 1.7, and 2.5 ug L'1, respectively. Thus, the pilot air stripping system did reduce tetrachloroethylene to less than 3 ug L'1 in the first three runs. ------- The Onda correlation (Ref 2), in general, provides a reasonable estimation of the mass transfer coefficient. The Kla values predicted for runs 1 through 6 by the Onda correlation using the best fit Henry's coefficient are 0.0235, 0.0230, 0.0225, 0.0215, 0.0200, and 0.0186 sec"1, respectively. The Kla value measured in the field evaluation was 16% lower than the average Kla value predicted by the Onda correlation. Thus, the packing height necessary to achieve a treatment objective will be sightly greater than determined in the preliminary analysis using the Onda correlation. Equipment Size: The result of the field evaluation/data analysis can be used to size a packed column air stripping system. Applying safety factors to uncertain parameters such as the mass transfer coefficient and Henry's coefficient is prudent engineering practice. In general, the lower 95% confidence limit offers a reasonable safety margin; however, other confidence limits may be preferred depending on the application. The lower 95% confidence limits for the mass transfer coefficient and Henry's coefficient are 0.0172 sec'1 and 0.32 atm m3 m'3, respectively. Applying safety factors to both the mass transfer coefficient and Henry's coefficient will increase the confidence above 95%. Equation 1 can be manipulated to obtain the packing height necessary to achieve a desired effluent concentration. The result is Equation 2. Packing height Equation: (Xt-Xu) * (R—l) + 1 7h s " L R * In (Xb-Xu) Kla (R-i; _ R Where: Zb = Packing height (m) Xb = VOC Concentration at bottom of packing (ug L"1) Xu = Unstrippable concentration (ug L'1) (Assumed to be zero) Other terms are defined in Eq 1 , The design requirement at Pensacola is to reduce 200 ug L'1 tetrachloroethylene to 3 ug L'1 in well #8 which has a design flow of 0.13 m3 sec'1 (2000 gpm). Using the liquid and air loadings from run number 2, 0.020 m3 sec'1 (30 gpm ft'2) and 18 to l, respectively, the required packing height will be 5.7 m (18.7 ft), the column diameter will be 2.9 m (9.5 ft), the air flow will be 2.3 m3 sec'1 (4900 SCFM), and the air pressure drop through the column will be ------- 180 N m"2 (0.7 inch water column) . The equipment configuration is presented in Figure 4. It should be noted that the safety factors increase the system size thus the most probable effluent concentration will be lower than 3 ug L"1. The most probable effluent concentration is computed, using the best fit Kla and H (0.0185 sec'1 and 0.39 atm m3 m'3 respectively) and Eg. 1, to be 1.9 ug L'1, a reasonable margin of safety. If the design influent concentration were increased to 3000 ug L'1 then the required removal efficiency will be increased to 99.9%, the packing height will be increased to 9.5 m (31 ft), and the air pressure drop will be increased to 300 N m'2 (1.2 inch water column). The column diameter and air flow will remain the same. The most probable effluent concentration will be 1.4 ug L"1. Cost Estimate: The estimated capital and operating costs of the 98.5% removal system are $270,000 and $21,000 per year, respectively, whereas the estimated capital and operating costs of the 99.9% removal system are $340,000 and $27,000 per year, respectively. The cost estimates for the two systems are shown in Table 4 and 5, respectively. The process equipment shown in Table 4 and 5 include the column shell, column internals (i.e., liquid distributor, liquid redistributor, and packing material support plate), packing material, one blower, and one pump. The cost estimate for the process equipment is based on vendor quotes of the individual items and includes delivery to the site but does not include assembly or installation. The cost estimates for the column shell and internals are based on 304 stainless steel (SS); however, other materials of construction are also available and may reduce or increase the cost. The support equipment includes assembly and installation of the above process equipment, a 4.2 by 4.2 by 4.0 m (14 x 14 x 13 ft) concrete air well which is a foundation for the packed column and a 70 m3 (18,000 gallon) liquid reservoir, 60 m (200 ft) of piping, instrumentation, air duct, and electrical connections. The electrical connection cost estimate is based on 25% of the blower and pump capital cost. The other support equipment cost estimates are based on material quantities. The support equipment cost estimate should provide a reasonable estimate of cost for these items; however, these items will vary from site to site and should be reviewed by a design engineer familiar with the Pensacola site. The total direct cost includes all equipment installed at the site and is the sum of the process and support equipment. ------- The indirect cost includes all non-physical items required for the air stripping system. This includes sitework, design engineering, contractor overhead and profit, legal and financal, interest during construction, and contingencies. The cost estimate for each of these items is based on percentage of the total direct cost. The precentages were selected by a committee of engineers such that the Office of Drinking Water's cost estimating procedures for various water treatment technologies can be based on the same assumptions. The percentages are 15, 15, 12, 2.5, 6, and 15%, respectively. The actual percentages will be site specific and should be reviewed by an engineer familiar with the Pensacola site. The total capital cost is the sum of the direct and indirect costs. The amortized cost is the total capital cost amortized over a 20-year time period at 10% interest rate. The operating cost is intended to reflect the additional cost of the air stripping system and does not include existing operating costs. The operating cost is based on the projected volume of water treated per year, the electrical power consumed by the air stripping system, and the maintenance of the mechanical and non- mechanical process equipment. The electrical power for the blower was estimated based on the air flow, air pressure drop through the column plus 500 N m"2 (2 inch water column) of pipe friction, 50% fan efficiency, 70% motor efficiency, and 25% motor size-up. The electrical power for the pump was estimated based on the liquid flow, liquid head loss from the top of the packing material to the bottom of the air well plus 12 ft of pipe friction, 80% pump efficiency, 80% motor efficiency, and 25% motor size-up. The electrical power rate was assumed to be 10 cents per Kw Hr and the volume of water treated per year assumed to be 50% of the design flow. The maintenance cost is based on 10% and 4% of the mechanical and non-mechanical process equipment cost, respectively. The labor operating cost is based on a flat rate of $0.0008 per m3 (0.3 cent per 1000 gallon) treated and the volume of liquid treated per year. The administrative cost is based on 20% and 25% of the labor and maintenance cost, respectively. The total annual cost is the sum of the amortized capital and operating cost. Finally, the total production cost is the total annual cost divided by the volume of water treated per year. The cost model was developed to give a general indication of the economics of the packed column air stripping process. These costs are presented only to evaluate the economic feasibility of using a packed column air stripping system to remove tetrachloroethylene from ground water. The specific site requirements at Pensacola, FL, will differ from the assumptions of the general cost model. The use of a large scale, on-site pilot system lends confidence in the accuracy of equipment size necessary to obtain tetrachloroethylene removal; however, it is emphasized ------- that these costs should be reviewed by an engineer familiar with the Pensacola site. This field evaluation was conducted using 1 inch plastic saddles as packing material. The 1 inch plastic saddle packing material was selected due to availability from a number of suppliers and the existence of packing characteristics in the technical literature. Other packing materials are also available; however, the removal efficiencies will differ from the packing material used in this field evaluation. The fate of volatile organic compounds (VOCs) when discharged into the atmosphere is uncertain. The compounds will disperse in the wind currents and may break down when exposed to sunlight. In most cases the impact of VOC discharge to the atmosphere from a packed column treating contaminated drinking water will be minimal. The information contained in this report should not be interpreted as requirements or recommendations from EPA. The field evaluation, laboratory analysis, data analysis, and calculations are believed to be correct; however, neither the author nor EPA can be responsible for any errors resulting directly or indirectly from the use of the information contained in this report. The author expresses special thanks to Robert Kneipp, Ken Evans, Mary Ann Feige, and SwRI analytical laboratory. Without their professional assistance in setting up and operating the pilot system and analyzing the resulting samples this field evaluation could not have been accomplished. REFERENCES 1. Treybal R. E., Mass-Transfer Operations, 3rd Ed., McGraw Hill (1980). 2. Perry, R. H. and C. H. Chilton, Chemical Engineers' Handbook, 5th Ed., McGraw Hill (1973). ------- Table 1 Field Evaluation - Tetrachloroethylene Operation Data Packing Height = 5.5 (m) Temperature = 24.0 (Deg C) Liquid Loading - 0.020 (in3 m"2 sec'1) Run Air Air Air Pressure Water Loading Drop Gradient Ratio (m3 m'2 sec'1) (N m*2 m'1) 1 35. 0.70 140. 2 18. 0.35 32. 3 10. 0.20 14. 4 5.0 0.10 4.6 5 2.5 0.050 2.3 6 1.5 0.030 * * Below limit of measuring device. Table 2 Sample Port Location Concentration Profile Data Set Tetrachloroethylene Concentration (ug L"1) Run (ro) 1 2 3 4 5 0.15 130. 120. 120. 130. 150. 160 1.68 38. 39. 47. 62. 100. 150 2.90 12. 17. 20. 43. 81. 120 4.12 3.6 6.9 12. 22. 59. 120 5.34 0.88 1.8 2.9 7.6 22. 55 ------- Table 3 Data Analysis Results Pensacola, F1 Compound : Tetrachloroethylene Ground Water Temperature : 24 (Deg C) Number of Samples : 30 Relative Standard Error : 19% Best Fit 95% Confidence Interval Xt (ug L"1) 146 133 to 160 Kla (sec1) 0.0185 0.0172 to 0.0196 Henry (atm m3 m'3) 0.39 0.32 to 0.46 Air to Water Ratio 35 18 10 5 2.5 1.5 Best fit Effluent (ug L'1) 1.2 1.7 2.5 6.4 26. 62. Best fit Removal (%) 99.15 98.9 98.3 95.6 82. 58. Kla predicted by Onda 0.0235 0.0230 0.0225 0.0215 0.0200 0.0186 ------- Table 4 98.5% Removal System Cost Study Estimate Interest Rate= 10. % per year Power Cost= 10. c KW*1 hr*1 Financing Period= 20. Year Use Rate= 50. % Design Flow Construction Cost Index= 405.00 Process Equipment: Capital Cost (K$) 304 SS Column Shells 13.8 Column Internals 13.1 Packing 18.7 t Blower 2.7 Pumps 12.1 Total Process Equipment 60.3 Support Equipment: Installation 32.5 Air Well 35.3 Piping 24.8 Instrumentation 5.1 Air Duct 0.5 Electrical 3.7 Total Support Equipment 101.9 Total Direct Cost 162.2 Indirect Cost: Sitework 24.3 Engineering 24.3 Contractor 19.5 Legal & Financial 4.1 Int. During Const. 9.7 Contingencies 24.3 Total Indirect Cost Total Capital Cost Amortized Capital Cost Operating Cost: Liquid Pumping Blower Labor Maintenance Adminstrative Total Operating Cost 21.3 14.2 0.6 1.9 3.3 1.2 106.2 268.4 K$ 31.5 (K$ per Year) Total Annual Cost Total Production Cost treated ------- Table 5 99.9% Removal System Cost Study Estimate Interest Rate= 10. % per year Power Cost= 10. c KW"1 hr"1 Financing Period= 20. Year Use Rate= 50. % Design Flow Construction Cost Index= 405.00 Process Equipment: Capital Cost (K$) 304 SS Column Shells 21.1 Column Internals 19.4 Packing 30.6 Blower 2.7 Pumps 12.1 Total Process Equipment 86.0 Support Equipment: Installation 49.7 Air Well 35.3 Piping 26.1 Instrumentation 5.1 Air Duct 0.5 Electrical 3.7 Total Support Equipment 120.5 Total Direct Cost 206.4 Indirect Cost: Sitework 31.0 Engineering 31.0 Contractor 24.8 Legal & Financial 5.2 Int. During Const. 12.4 -Contingencies 31.0 Total Indirect Cost 135.2 Total Capital Cost 341.7 K$ Amortized Calital Cost 40.1 (K$ per Year) Operating Cost: Liquid Pumping 18.4 Blower l.l Labor 1.9 Maintenance 4.3 Adminstrative 1.5 Total Operating Cost 27.2 Total Annual Cost Total Production Cost treated 67.3 K$ per year 3.3 c per m3 ------- Figure 1 Packed Column Air Stripping Process ( 1) Watir Flow (6) VOC (3) Packing Material \\ liilltt ((f 111 Kit lllllilill SSSfSfllS lllllilill inn in HKfmn mFQuTT? uiiTTrrti iiiiiiini f f(f(If(IS mi wmmt HMiiuTi siiiiiiiii f f f f f f t f M (it niniiiil (4) VOC Transfer (2) Air Flow (6) 9 9.0% Removal Possible ------- Figure 2 Packed Column Aii Strip.pina Pilot System Influent Watar from Wall Pvmp Efflaant Watar to Drain ------- Figure 4 ——WmB— Cost Model Configuration Column Dlametor 7A7 S»l»ct Fill Not to Scale ------- Figure 3 Pensacola, FL - Tetrach1oroethy1ene ^ ® Run 1 Run 2 Run 3 _j g Ai 111111 m | it 111111111 ii 11 m >» A A1111111111111111» 111111111111 Am 11111 it 1111111111111 ti 11111A 1111111111111111 ii 11111111 ii i 11111111111111111111111111111 Run 4 i }i i < 111111111111111111 ii 11111 ij 1111111 ii11111111111111111111 Run 5 Run 6 il 111111 111 I 111 11111 11 ii 11III i A Ai it 11111 1111 111 1111111111111 ij j ! : S ! 1 I! !! !! ® © Q 0- !! j! I I I I | I I I I | I I I I 1 I I I I i I I I I | I I I I I | I I I I |l I I I | I I I I 1 I I M | I I II | I I I I 0 1 2 3 4 5 Location (Z) 6 0 1 2 3 4 5 Location (Z) 6 0 1 2 3 4 5 6 Location (Z) ------- |