Home

User's Guide for Polyethylene-Based Passive Diffusion Bag

1. amp 1 5 2 euaujeoJ0 U2IQ L L 014210 0002 0 WHO gwp dund Sursn 5 y pue uonejnuenb ogroods o dures st 11501 jou uonejnuenb eonATeue sem INPA N sem o duues q 12d swego 811 0002 Menuer eiuoJi eo pue s pue UOISNYIP 2 pejoeles JO 9 Diffusion Sampler Testing at Naval Air Station North Island B 12 San Diego County California November 1999 to January 2000 ns ns 5 ns ns ns ns ns ns ns ns ns 0 61 V9 MIA CS N N T N N N ns N N ns ns CLI V9O MIN TS ns ns ns ns ns ns ns ns ns ns ns V9 MIA CS ns ns ns ns ns ns ns ns ns ns ns Lyi VO MIN CS ns ns S I ns ns ns ns ns ns ns ns Cel VO MIN CS ns ns 4 ns ns N S ns ns ns ns ns OCI VO MIN CS 4 c S N NS ns N N ns ns 9 01 8 ns ns ns ns ns T6 V
2. f uonejguenb eonAqeue oq sem npea N Pamp sem q 8 penunuo2 o00z Menuer eiuoJ eo pue s YON uoneig Burdues Moj Mo pue uoisnyjlp ui orueBJo ejnejoA peyeuuo uo Jo iscussion B 13 Results and D Table 6 Concentrations of benzene ethylbenzene toluene and total xylenes in water from diffusion low flow sampling Naval Air Station North Island California January 2000 ug L micrograms per liter U value was below the analytical quantitation limit J estimated value NA not applicable sample collected by using bladder pump data from OHM Remediation Services Corporation 2000 Benzene Ethylbenzene Toluene Total xylenes Well pan ug L ug L ug L ug L Diffusion Low flow Diffusion Low flow Diffusion Low flow Diffusion Low flow MW 5D 50 8 50 U 25 U 50 U 25 U 50 U 25 U 50 U 25 U MW 5D 52 3 250 U 50 U 250 U 50 U 250 U 50 U 250 U 50 U MW 5D 54 2 50 U 100 U 50 U 100 U 50 U 100 U 50 U 100 U MW 5D 55 8 5 U 25 5 U 25 U 5 U 25 U 5 U 25 U MW 5D 574 50 U 5 U 50 U 5 0 50 U 5U 50 U 5 U MW 5D 59 0 50 U 5 U 50 U 5 0 50 U 5U 50 U 5 U MW 9 27 6 250 U 250 U 250 U 250 U 250 U 250 U 250 U 250 U MW 10 7 8 5U 5U 5 U 5U 5 U 5U 5U 5 U MW 10 9 2 5 U 5U 5 U 5U 5 U 5U 5U 5 U MW 10 11 2 5U 5 U 5 U 5 U 5 U 5
3. ea 27 4 E A n a 28 po 1 11111 po 1 10 100 1 000 10 000 CONCENTRATION IN MICROGRAMS PER LITER EXPLANATION LOW FLOW LOW FLOW D SIT PERISTALTIC BLADDER PUMP SAMPLE PUMP SAMPLE cis 1 2 DICHLOROETHENE y ege WV TOLUENE 1 NOT DETECTED TOTAL XYLENES A NOT DETECTED VINYL CHLORIDE Figure 3 Comparison of diffusion and low flow samples in ground water at well MW 13B Naval Air Station North Island California January 2000 using peristaltic pumps or if mixing in the well screens occurred during pumping The vertical concentration distribution between the two methods implies that the VOC concentrations measured in water from diffusion samplers reflected the vertical distribution of contami nants in the aquifer adjacent to the screened interval more accurately than the peristaltic pump sampling Further comparison of TCE concentration data from the two sampling methods indicates that diffu sion sampling provides a point sample whereas sequential low flow sampling of multiple horizons within a single well screen can induce mixing In gen eral the vertical sequence of low flow sampling in the wells began with the shallowest depth interval and ended with the deepest interval In well PW 66 TCE data show that concentrations in water collected with a diffusion sampler were highest in the shallowest sam pled depth and then decreased sharply over the 5 ft dept
4. 779679 0016 779679 0017 779679 0018 779679 0019 779679 0027 779679 0028 779679 0117 Diffusion sampler laboratory identifier 79679 0099 79679 0100 719679 0101 79679 0102 79679 0103 79679 0104 79679 0105 NA 779679 0042 779679 0043 779679 0044 779679 0045 779679 0046 NA 779679 0047 779679 0048 779679 0049 779679 0054 779679 0055 779679 0056 779679 0025 779679 0026 779679 0020 779679 0021 NA 779679 0022 779679 0050 779679 005 1 779679 0116 Depth to diffusion sampler center ft bls 7 75 9 15 11 1 13 1 15 1 17 1 18 8 18 8 6 50 7 95 9 35 10 9 12 4 NA 24 8 26 1 27 5 NA 45 4 46 6 46 6 48 0 NA 21 7 23 0 34 5 37 0 37 0 38 5 56 0 3455 59 0 Date installed 11 12 99 11 12 99 11 12 99 11 12 99 11 12 99 11 12 99 11 12 99 11 12 99 11 10 99 11 10 99 11 10 99 11 10 99 11 10 99 NA 11 10 99 11 10 99 11 10 99 NA 11 10 99 11 10 99 11 10 99 11 10 99 NA 11 9 99 11 9 99 11 9 99 11 9 99 11 9 99 11 9 99 11 9 99 11 9 99 11 9 99 Date recovered 1 18 00 1 18 00 1 18 00 1 18 00 1 18 00 1 18 00 1 18 00 1 18 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 17 00 1 19 00 Number of days diffusion samplers were in wells 67 67 67 6
5. 472 MW 68C2 K 779679 0172 779679 0192 55 5 11 11 99 1 20 00 70 472 MW 68C2 L 779679 0170 779679 0193 24 5 11 11 99 1 20 00 70 472 MW 68C2 M 779679 0168 779679 0194 59 5 11 11 99 1 20 00 70 472 MW 68C2 M repl NA 79679 0195 59 5 11 11 99 1 20 00 70 472 MW 68C2 N 779679 0167 779679 0196 61 5 11 11 99 1 20 00 70 379 PW 15 A 779679 0083 779679 0089 25 4 11 13 99 1 19 00 67 379 PW 15 B 79679 0084 779679 0118 27 1 11 13 99 1 19 00 67 379 PW 15 B repl NA 779679 0119 27 1 11 13 99 1 19 00 67 379 PW 15 C 779679 0085 779679 0156 28 5 11 13 99 1 19 00 67 379 PW 15 D 779679 0086 779679 0157 30 2 11 13 99 1 19 00 67 379 PW 15 E 779679 0087 779679 0158 31 7 11 13 99 1 19 00 67 379 PW 15 E repl NA 19679 0159 31 7 11 13 99 1 19 00 67 379 PW 15 F 779679 0088 779679 0164 33 1 11 13 99 1 19 00 67 379 PW 55 A 779679 0077 779679 0109 27 1 11 13 99 1 19 00 67 379 PW 55 B 779679 0078 779679 0110 28 9 11 13 99 1 19 00 67 379 PW 55 B repl NA 779679 0114 28 9 11 13 99 1 19 00 67 379 PW 55 C 779679 0079 779679 0111 30 6 11 13 99 1 19 00 67 379 PW 55 D 779679 0080 779679 0112 31 9 11 13 99 1 19 00 67 379 PW 55 E 779679 0081 779679 0113 33 1 11 13 99 1 19 00 67 379 PW 55 E repl 779679 0082 779679 0115 33 1 11 13 99 1 19 00 67 379 PW 66 A 779679 0106 779679 0145 25 5 11 10 99 1 18 00 69 379 PW 66 B 779679 0107 779679 0146 21 3 11 10 99 1 18 00 69 379 PW 66 B repl NA 779679 0151 27 3 11 10 99 1 18 00 69 6 Diffusion Sampler Testing at Naval Air Station North Island San Diego C
6. Army Corps of Engineers USACE Johnette Shockley US Army Corps of Engineers Federal Remediation Technologies Roundtable Interstate Technology and Regulatory Cooperation Work Group ITRC George H Nicholas New Jersey Team Lead Paul M Bergstrand South Carolina Chris A Guerre California David Randolph Tennessee INTERSTATE 5 g A E 5 2 Bi 2 5 2 a Funding for this Guide was provided by the U S AIR FORCE and NAVFAC Southern and Southwest Divisions Additionally the following persons are recognized for their leadership and support to this project Marty Faile Joe Dunkle Kay Wishkaemper Vince Malott and the Passive Diffusion Bag Sampler PDBS Work Group User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 1 Deployment Recovery Data Interpretation and Quality Control and Assurance By Don A Vroblesky U S Geological Survey Water Resources Investigations Report 01 4060 Prepared in cooperation with the U S AIR FORCE U S NAVAL FACILITIES ENGINEERING COMMAND U S ENVIRONMENTAL PROTECTION AGENCY FEDERAL REMEDIATION TECHNOLOGIES ROUNDTABLE DEFENSE LOGISTICS AGENCY U S ARMY CORPS OF ENGINEERS and INTERSTATE TECHNOLOGY AND REGULATORY COOPERATION WORK GROUP a USGS science for a changing world Columbia South Carolina U S DEPARTMENT OF THE
7. J estimated value U value was below the analytical quantitation limit 11DCA 1 1 dichloroethane 11DCE 1 1 dichloroethene cDCE cis 1 2 dichloroethene TCE trichloroethene Well Depth to dif i coa TIDCA cDCE eee e Yok ds PE contar tion ug L ug L ug L ug L ug L L ug L ft bls MW 12 D 35 1 Site 11 86 J 1 500 100 100 U 1 800 100 U 100 U MW 12 D repl 35 1 Site 11 89J 1 500 110 100 U 1 700 100 U 100 U MW 5D C 542 Site 11 170 2 800 E 61 500 930 50 U 50 U MW 5D C repl 54 2 Site 11 170 2 900 E 61 50 U 930 50 U 50 U MW 68C2 39 1 472 2 500 U 4 100 1 000 J 2 500 U 47 000 2 500 U 2 500 U MW 68C2 B repl 39 1 472 5 000 U 4 300 J 960 J 5 000 U 52 000 5 000 U 5 000 U MW 68C2 M 59 5 472 500 U 350 J 500 U 500 U 7 000 500 U 500 U MW 68C2 M repl 59 5 472 500 U 360 J 500 U 500 U 6 800 500 U 500 U PW 15 B 27 1 379 52 8 130 15 7 72 52 PW 15 B repl 27 1 379 54 8 130 17 5 75 57 PW 15 E 31 7 379 500 U 500 U 1 900 500 U 5 500 500 U 500 U PW 15 E repl 31 7 379 500 U 500 U 1 900 500 U 5 600 500 U 500 U PW 55 B 28 9 379 2 500 U 2 500 U 6 500 2 500 U 39 000 2 500 U 2 500 U PW 55 B repl 28 9 379 2 500 U 2 500 U 6 700 2 500 U 36 000 2 500 U 2 500 U PW 55 E 33 1 379 2 500 U 2 500 U 6 300 2 500 U 33 000 2 500 U 2 500 U PW 55 E repl 33 1 379 2 500 U 2 500 U 6 100 2 500 U 31 000 2 500 U 2 500 U PW 66 B 28 9 379 500 U 500 U 3 400 500 U 9 000 500 U 500 U PW 66 B r
8. equilibration time of 2 weeks is suggested When applying PDB samplers in waters colder than previ ously tested 10 C or for compounds without suffi cient corroborating field data a side by side com parison with conventional sampling methodology is advisable to justify the field equilibration time In less permeable formations longer equilibra tion times may be required It is probable that water in the well bore eventually will equilibrate with the pore water chemistry however if the rate of chemical change or volatilization loss in the well bore exceeds the rate of exchange between the pore water and the well bore water then the PDB samplers may under Passive Diffusion Bag Sampler and Sample Recovery 9 estimate pore water concentrations Guidelines for equilibration times and applicability of PDB samplers in low permeability formations have not yet been established Therefore in such situations a side by side comparison of PDB samplers and conventional sampling methodology is advisable to ensure that the PDB samplers do not underestimate concentrations obtained by the conventional method A detailed discussion of diffusion rates relevant to diffusion sampler equilibrium in slow moving ground water systems can be found in Harrington and others 2000 Following the initial equilibration period the samplers maintain equilibrium concentrations with the ambient water until recovery Thus there is no speci fied maximum tim
9. repl 26 653 5U 5U 3 200 5U 5U 1 400 5U MW 13C B 46 6 653 5U 5U 3 5U 5U 5U 5U MW 13C 46 6 653 5U 5U 2 5U 5U 5U 5U MW 5D D 55 7 Site 11 51 760 23 25U 320 25U 25U MW SD D repl 55 7 Site 11 44 670 221 2500 280 250 25 U MW 68C2 J 53 9 472 2 500 U 2 500 J 2 500 U 2 500 U 38 000 2 500 U 2 500 U MW 68C2 J repl 53 9 472 2 500 U 2 600 2 500 U 2 500 U 38 000 2 500 U 2 500 U MW 68B B 37 0 472 5 000 U 4 400 J 5 000 U 5 000 U 34 000 5 000 U 5 000 U MW 68B B repl 37 0 472 5 000 U 4 900 J 5 000 U 5 000 U 33 000 5 000 U 5 000 U PW 55 E 33 1 379 2 500 U 2 500 U 5 500 2 500 U 29 000 2 500 U 2 500 U PW 55 E repl 33 1 379 2 500 U 2 500 U 5 700 2 500 U 29 000 2 500 U 2 500 U S2 MW 6A J 18 9 Site 2 5U 5U 5U 5U 5U 5U 5U S2 MW 6A J repl 18 9 Site 2 5U 5U 5U 5U 5U 5U 5U following low flow sampling from three overlying depths using a peristaltic pump thus the concentra tion interface potentially shifted upward toward the bladder pump intake It is possible that in well mixing was more pronounced for cDCE than for TCE because there was a greater percentage of change in concentra tions with depth for cDCE than for TCE The cDCE concentration increased by a factor of 26 100 to 2 600 ug L over a depth of 3 4 feet whereas TCE increased by only a factor of 4 6 over the same depth interval 1 700 to 7 800 ug L fig 2 The con centration data indicate that in well MW 12 the diffu sion samplers collected po
10. 00 1 fc N 001 LS f 96 015 00571 cS f I6 TCE CI MIN ns N 001 00t T 00Z T fv N 001 LL 66 046 00 1 L f v6 9 N ns ns ns f 0 N S N N 6 81 OI MIA N ns 6 ns N S ns N ns ns ns TLI OI MIA ns ns OI ns N S N N ns fo ns ns ns TSI OI MIA N ns I ns ns ns ns ns fl ns ns N ctl ns ns LI ns N S N ns I fl N ns ns CLL OI MIA ns ns 81 ns ns ns ns fl ns ns 5 T6 OIL MIA ns ns oE N ns ns ns ns c ns ns ns SL OAN N OST OST 00c 00 097 N OST OST 00 000r OLT OLT 91 x67MN ns N 05 9 cs ns 0 9 f 01 v6 061 9 N os 06S ns N 05 OcI 09I ns N 05 05 09 097 0c N 05 YLS N ST N 015 sc N S fte LG 09L 0051 15 59 S ec N 001 N 05 065 056 N 001 05 f 9t 19 0025 f 56 0 1 TYS 05 OST 001 1 00c T 05 N OST 9 f 071 ooer 009 012 09c ETS N ST N 0s 099 069 N ST N OS T9 99 a 0061 00571 007 097 8 05 uoisnyig uoisnyig UON uorsnyig uoinyig uoisnyla TT 9 61 7 61 1 61 1 61 1 61 7 61 VEM AulA 5 2 11 eueujeoJo u2IQ L 000c WHO gwp S
11. 245 266 Pitkin S E Ingleton R A and Cherry J A 1999 Field demonstrations using the Waterloo Ground Water Profiler Ground Water Monitoring and Remediation v 19 no 2 Spring 1999 p 122 131 Powell R M and Puls R W 1993 Passive sampling of ground water monitoring wells without purging Multi level well chemistry and tracer disappearance Journal of Contaminant Hydrology v 12 p 51 77 Reilly T E Frank O L and Bennet G D 1989 Bias in ground water samples caused by wellbore flow ASCE Journal of Hydraulic Engineering v 115 p 270 276 Reilly T E and Gibs J 1993 Effects of physical and chemical heterogeneity of water quality samples obtained from wells Ground Water v 31 no 5 p 805 813 Reilly T E and LeBlanc D R 1998 Experimental evalua tion of factors affecting temporal variability of water samples obtained from long screened wells Ground Water v 36 no 4 p 566 576 Robin M J L and Gillham R W 1987 Field evaluation on well purging procedures Ground Water Monitoring Review v 7 no 4 p 85 93 Robbins GA 1989 Influence of using purged and partially penetrating monitoring wells on contaminant detection mapping and modeling Ground Water v 27 no 2 p 155 162 Robbins G A and Martin Hayden J M 1991 Mass balance evaluation of monitoring well purging Part I Theoretical models and implications for representative sampling Journal of Contaminant Hy
12. Comparison of diffusion and low flow samples in ground water well MW 13A Naval Air Station North Island California January 2000 diffusion samplers did not contribute contaminants to the water Diffusion Samplers in Free Phase Fuel The diffusion samplers deployed in buckets con taining free phase JP 5 and Stoddard solution from wells MW 11 and PW 17 did not show evidence of structural integrity loss during the 2 months of equili bration The VOCs detected in the free phase fuel also were detected in the water from the diffusion samplers table 8 The VOC concentrations in water from the diffusion samplers were lower than the VOC concen trations in the fuel however this is to be expected because the first is an aqueous solution and the second is an organic solvent concentration The diffusion sam plers provided an alternative method for showing that the free phase fuel in ground water from well MW 11 also contained TCE table 5 Contaminant Stratification in Well Screens The data from this investigation show that sub stantial stratification of VOCs can be present within a 10 ft well screen At four observation wells MW 12 MW 5 PW 66 and PW 15 the data showed a sharp layering of VOCs within the screened interval figs 2 5 7 and 8 The diffusion sampler data show that the vertical change in TCE concentrations over a distance of about 5 ft was approximately 17 500 ug L in well PW 66 approximately 7 300 ug L in well P
13. I8 ns sc L 116 V89 MIN ns VN N S ns VN I VN N S VN ns VN oor SxOtI MIA ns ns ns ns ns ns fc ns ns ns ns ns 187 O L MN ns ns ns ns ns t ns ns ns ns ns 1797 OE I MIN ns ns ns ns ns ns c ns ns ns ns OE L MN 00v I VN ns VN ns VN 00 ns VN ns VN 0 9c 9 001 007 L 8 ns ns 006 c 004 s cs ns ns OLE SET MIN 0091 0007 9 rv ns ns 009 c 009 c N S rv ns ns TOIT SEC I MIN 0061 006 rv ns ns 0097 00 v ns ns 6 vc SEL MIN L VN N S ns VN 19 N S N 5 OCI E lt ns ns ns 66 9y ns ns ns ns vc VELMAN L 8 ns ns ns ES vL N S ns ns ns 6 0I VELMAN 9 6 ns ns ns ns 17 LL ns ns ns ns v6 VEL AN 9 6 0s ns ns ns 9v 8L N S ns ns ns 08 VELMAN 9 fv ns ns ns ns Ly 6 N S ns N 5 ns 59 VELMAN MO uonya MO J MO uona MO UOISNyIG Uoisnylg Uuoisnyiq uona 161 16r NEM jAulA eueujeoJo uoIp e 512 1 210 0002 uone10d10o se rA1eg uonerpouros WHO 4 dund Sursn Aq paaro o duues y uonoojop oy pue uonejnuenb 5 uooAgoq 51 1 1 q jou WN onea poreumse uonej
14. INTERIOR GALE A NORTON Secretary U S GEOLOGICAL SURVEY Charles G Groat Director Use of trade product or firm names in this publication is for descriptive purposes only and does not imply endorsement by the U S Geological Survey Copies of this report can be obtained from U S Environmental Protection Agency USEPA National Service Center for Environmental Publications NSCEP Box 42419 Cincinnati OH 45242 0419 and U S Geological Survey Branch of Information Services Box 25286 Denver CO 80225 Phone 888 ASK USGS For additional information write to District Chief U S Geological Survey Stephenson Center Suite 129 720 Gracern Road Columbia SC 29210 7651 Additional information about water resources in South Carolina is available on the World Wide Web at http sc water usgs gov Report can be downloaded from http www itrcweb org and http www frtr gov 5 Executive Summary tede eite etum ue 1 3 Summary of Passive Diffusion Bag Sampler Advantages and Limitations essere e 5 AAV ANA SES Rx 5 t ore aeu en etae it nd pe ee e d IBN DD HER 5 Passive Diffusion Bag Sampler 1 6 Passive Dif
15. PDB samplers are to determine whether contaminant stratification is present and to locate the zone of highest concentration The midpoint of each sampler should be positioned at the midpoint of the interval to be sampled For 1 5 ft long samplers at each sampling depth in the screened interval make two attachment points on the weighted line at a distance of about 1 5 ft apart The attachment points should be positioned along the weighted line at a distance from the bottom end of the weight such that the midpoint between the knots will be at the desired sampling depth along the well screen Sampler intervals are variable but a simple approach is to use the top knot loop of one sampler interval as the bottom knot loop for the overlying sampler interval Passive Diffusion Bag Sampler Deployment samplers should not be used in wells having screened or open intervals longer than 10 ft unless used in conjunction with borehole flow meters or other techniques to characterize vertical variability in hydraulic conductivity and contaminant distribution or used strictly for qualitative reconnaissance purposes This is because of the increased potential for cross con tamination of water bearing zones and hydrauli cally driven mixing effects that may cause the contaminant stratification in the well to differ from the contaminant stratification in the adja cent aquifer material If it is necessary to sample such wells then multiple PD
16. U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U MW 68C2 53 9 1 000 U 2 500 U 1 000 U 2 500 U 1 000 U 2 500 U 1 000 U 2 500 U MW 68C2 55 6 500 U 1 000 U 500 U 1 000 U 500 U 1 000 U 500 U 1 000 U MW 68C2 57 9 500 U 500 U 500 U 500 U 500 U 500 U 500 U 500 U MW 68C2 59 5 500 U 500 U 500 U 500 U 500 U 500 U 500 U 500 U MW 68C2 61 5 500 U 500 U 500 U 500 U 500 U 500 U 500 U 500 U PW 15 25 4 4 2J 16 9 7 3J 75 28 PW 15 27 1 3J 3J 15 3 5J 5 U 52 7 PW 15 28 5 100 U 250 U 100 U 250 U 100 U 250 U 100 U 250 U PW 15 30 2 250 U 500 U 250 U 500 U 250 U 500 U 250 U 500 U PW 15 31 8 500 U 500 U 500 U 500 U 500 U 500 U 500 U 500 U PW 15 33 2 500 U 500 U 500 U 500 U 500 U 500 U 500 U 500 U PW 55 27 1 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U PW 55 28 9 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U PW 55 30 6 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U Results and Discussion B 15 Table 6 Concentrations of benzene ethylbenzene toluene and total xylenes in water from diffusion and low flow sampling Naval Air Station North Island California January 2000 Continued ug L micrograms per liter U value was below the analytical quantitation limit J estimated value NA not applicable sample collected by using bladder pump data from OHM Remediation Services Corporation 2000 Benzene Ethylbenzene Toluene Total xylenes Well psi ug L ug L ug L ug L Diffusion Low
17. flow sampling progressed toward the center of the screened interval the correla tion between concentrations obtained from the diffu sion samples began to differ substantially from those obtained by low flow sampling fig 6D Between the depths of approximately 40 to 50 ft TCE concentra tions from low flow sampling were approximately 47 to 84 percent lower than TCE concentrations from dif fusion samplers additionally the low flow sampling data did not indicate a TCE peak concentration at a depth of 42 ft as shown by the diffusion sampling data A probable explanation for the concentration differ ences between the two methods is that initially rela tively uncontaminated water was pumped into the screened interval thus mixing the ground water in the well and diluting concentrations of TCE As a result San Diego County California November 1999 to January 2000 DEPTH BELOW OF CASING IN FEET DEPTH BELOW TOP OF CASING IN FEET 50 T T T T 50 T T _ ea _ 52 Z 5 E Z N 54 5 jam A 56 2 J E Q zi 58 E T 60 60 1 0 50 100 150 200 250 300 0 2 000 4 000 6 000 1 1 DICHLOROETHANE 1 1 DICHLOROETHENE CONCENTRATION IN CONCENTRATION IN MICROGRAMS PER LITER MICROGRAMS PER LITER 50 T T T 50 T T 52 54 56 58 DEPTH BELOW TOP OF CASING IN FEET 60 60 0 20 40 60 80 100 1
18. i qIdWVS 09 n NOISQHdId N F T fs 09 NOISQ4HId L c pws 7 s T F 7 LE L 4 J 85 os T QE w 07 510 N L 4 se IES Pa xi Ig 4 95 pug e so dureg Im b H J TAS M SN CO89 MIA 89 MIN 889 AATN V89 MI 89 A iy g v 0 l vS TE i Ic Lad NI ONISVO dOL MOTH H LdHd Diffusion Sampler Testing at Naval Air Station North Island B 22 San Diego County California November 1999 to January 2000 DEPTH BELOW TOP OF CASING IN FEET 24 f 26 28 30 32 34 DEPTH BELOW OF CASING IN FEET 36 l DIFFUSION SAMPLE Q LOW FLOW PERISTALTIC PUMP SAMPLE 0 2 000 4 000 6 000 1 8 000 10 000 12 000 14 000 16 000 18 000 TRICHLOROETHENE CONCENTRATION IN MICROGRAMS PER LITER Figure 7 Comparison of trichloroethene concentrations in diffusion and low flow samples in ground water at well PW 66 Naval Air Station North Island California 2000 24 34 1 1 0 500 1 000 1 500 2 000 2 500 3 000 3 500 cis 1 2 DICHLOROETHENB CONCENTRATION IN MICROGRAMS PER LITER DEPTH BELOW TOP OF CASING IN FEET 24 26 28 30 32 34 0 2 000 4000 6 000 8 000 TRICHLOROETHENE CONCENTRATI
19. is a potential for each low flow sampling event to disturb the equilibrated water column If the pumping rate during low flow sampling is low enough to prevent drawdown in the well then all of the pumped water is replaced by ground water from the aquifer however the zone of influence contributing water to the well may not be adjacent to the pump Thus in a chemi cally stratified screened interval where multiple depth intervals are sequentially sampled water entering the well screen from early low flow samplings may influ ence concentrations obtained in later samplings as a result of vertical transport and mixing in the well screen Despite these uncertainties the use of multiple level low flow sampling methods using peristaltic pumps sometimes can provide an estimate of contaminant vertical distribution in the screened interval which can be used as a comparison for the diffusion samplers In most of the observation wells the vertical concentration gradients obtained using the diffusion sampler and low flow sampler methods were similar However in several cases the concentrations in water obtained by using the peristaltic pump were lower than the concentrations in water obtained by using the dif fusion samplers figs 4 5 and 6 An example of this is TCE concentrations measured in water from wells MW 68A MW 68B and MW 68C TCE concentra tions were approximately 43 to 73 percent lower in water samples collected by using low f
20. methods was poorer for tetrachloroethene PCE concentrations The PCE concentration in water from the diffusion sampler was 2 percent lower than the concentration in water San Diego County California November 1999 to January 2000 Table 2 Sampler deployment and recovery information Naval Air Station North Island California November 1999 to January 2000 repl replicate sample NA not applicable low flow bladder pump sample data from OHM Remediation Services Corporation 2000 Site or building designation 653 653 653 653 653 653 653 653 653 653 653 653 653 653 653 653 653 653 653 653 653 653 653 472 472 472 472 472 472 472 472 472 Well identifier MW 10 MW 10 MW 10 MW 10 MW 10 MW 10 MW 10 MW 10 MW 13A MW 13A MW 13A MW 13A MW 13A MW 13A MW 13B MW 13B MW 13B MW 13B MW 13C MW 13C MW 13C MW 13C MW 13C MW 68A MW 68A MW 68B MW 68B MW 68B MW 68B MW 68C MW 68C MW 68C Sampling interval identifier OD O W Q G repl wo aw gt NA w NA B repl NA w Low flow sample laboratory identifier 779679 0091 779679 0092 779679 0093 779679 0094 719679 0095 79679 0096 779679 0097 779679 0098 779679 0030 779679 003 1 779679 0032 779679 0033 779679 0034 NA 779679 0035 779679 0036 779679 0037 NA 779679 0038 779679 0039 779679 0041 779679 0040 NA 719679 0023 779679 0024
21. plug from the sampler bottom insert the short funnel into the sampler and pour laboratory grade deionized water into the sampler The sampler should be filled until water rises and stands at least half way into the funnel Remove excess bubbles from the sampler Remove the funnel and reattach the plug A small air bubble from the plug is of no concern 6 User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 1 Deployment Recovery Data Interpretation and Quality Control and Assurance The following steps should be used for deploying PDB samplers in wells 1 Measure the well depth and compare the measured depth with the reported depth to the bottom of the well screen from well construction records This is to check on whether sediment has accumulated in the bottom of the well whether there is a nonscreened section of pipe sediment sump below the well screen and on the accuracy of well construction records If there is an uncertainty regarding length or placement of the well screen then an independent method such as video imaging of the well bore is strongly suggested 2 Attach a stainless steel weight to the end of the line Sufficient weight should be added to counterbal ance the buoyancy of the PDB samplers This is particularly important when multiple PDB samplers are deployed One approach discussed in the following paragraphs is to have the we
22. sam ples Ground Water v 35 no 2 p 201 208 Gillham R W II Robin M J L Barker J F and Cherry J A 1985 Field evaluation of well flushing proce dures Washington D C American Petroleum Institute Publication 4405 Grisak G E Merritt W F and Williams D W 1977 A fluoride borehole dilution apparatus for ground water velocity measurements Canadian Geotechnical Journal v 14 p 554 561 Halevy E Moser H Zellhofer O and Zuber A 1967 Borehole dilution techniques A critical review Isotopes in Hydrology Vienna Austria International Atomic Energy Agency p 531 564 Hare P W 2000 Passive diffusion bag samplers for moni toring chlorinated solvents in ground water The Second International Conference on Remediation of Chlorinated and Recalcitrant Compounds Battelle May 22 25 2000 Monterey California Harrington G A Cook P G and Robinson N I 2000 Equilibration times of gas filled diffusion samplers in slow moving ground water systems Ground Water Monitoring and Remediation Spring 2000 p 60 65 Hess A E 1982 A heat pulse flowmeter for measuring low velocities in boreholes U S Geological Survey Open File Report 82 699 44 p 1984 Use of a low velocity flowmeter in the study of hydraulic conductivity of fractured rock Proceedings of National Water Well Association Conference on Surface and Borehole Geophysics San Antonio Texas p 812 831 1986 Identifying hydraulic
23. samplers in a well where chemical stratification is suspected within the screened interval multiple diffusion samplers can B 26 Diffusion Sampler Testing at Naval Air Station North Island be used to at least initially delineate the stratification Analytical costs during such an investigation can be minimized by using field gas chromatography to delineate the stratification and to select particular sam ples for more detailed laboratory analyses SUMMARY The ground water VOC concentrations obtained by using water filled polyethylene diffusion samplers were compared to the ground water VOC concentra tions obtained by using low flow sampling methods with a peristaltic pump and dedicated bladder pumps in observation wells at Naval Air Station North Island California Comparisons of VOC concentrations obtained by using bladder pumps and diffusion sam plers showed a generally good correlation Concentra tions of 1 1 dichloroethene 1 1 DCE and trichloroethene TCE in ground water obtained from well MW 9 obtained using the diffusion sampler agreed well 12 and 3 percent difference respectively with those samples obtained using the bladder pump At well MW 5 the TCE concentration in water from the diffusion sampler was higher than in water from the bladder pump implying that the sample collected by the bladder pump may have underestimated actual concentrations as a result of mixing Similarly the San Diego County California N
24. sampling method while quarterly monitoring of VOCs can be accomplished using diffusion sampling technology REFERENCES United States Environmental Protection Agency USEPA 1994 Agency National Functional Guidelines for Organic Data Review PB 94 963502 Washington D C Vroblesky D A and Campbell T R 1999 Protocol for Use of Low Density Water Filled Polyethylene Diffusion Samplers for Volatile Organic Compounds in Wells Draft March Diffusion Sampler Testing at Naval Air Station North Island San Diego County California November 1999 to January 2000 by Vroblesky D A and Peters B C Not included as part of this package This is a published Water Resources Investigations Report Investigation of Polyethylene Passive Diffusion Samplers for Sampling Volatile Organic Compounds in Ground Water at Davis Global Communications Sacramento California August 1998 to February 1999 by Vroblesky D A Borchers J W Campbell T R and Kinsey W Not included as part of this package This is a published Water Resources Inves tigations Report Naval Industrial Reserve Ordnance Plant Fridley Minnesota by Vroblesky D A and Petkewich M D Not included as part of this package This is a published Water Resources Investigations Report Evaluation of a Diffusion Sampling Method for Determining Concentrations of Volatile Organic Compounds in Ground Water Hanscom Air Force Base Massachusetts by Church P E Not included as
25. semivolatile compounds in contact with PDB samplers showed a higher concentration of phthalates inside the PDB sampler than outside the PDB sampler suggesting that the polyethylene may contribute phthalates to the enclosed water Thus the samplers should not be used to sample for phthalates VOC concentrations in PDB samplers represent concentrations in the vicinity of the sampler within the well screen or open interval This may be a limitation for PDB samplers and some other types of sampling such as low flow sampling if the ground water contamination is above or below the screen or not in the sample intervals providing water movement to the PDB samplers If there is a vertical hydraulic gradient in the well then the concentrations in the sampler may represent the concentrations in the water flowing verti cally past the sampler rather than in the formation directly adjacent to the sampler Vertically spaced multiple PDB samplers may be needed in chemically stratified wells or where flow patterns through the screen change as a result of ground water pumping or seasonal water level fluctuations The purposes of this document are to present methods for PDB sampler deployment and recovery to discuss approaches to determine the applicability of passive diffusion samplers and to discuss various factors influencing interpretation of the data The intended audience for the methodology sections of this report is managers and field personnel
26. the U S Geological Survey Technology Enterprise Office Mail Stop 211 National Center 12201 Sunrise Valley Drive Reston Virginia telephone 703 648 4450 fax 703 648 4408 Diffusion samplers were attached to intakes of bladder pumps by means of plastic cable ties Attached to each remaining diffusion sampler was a Tygon tube extending from the sampler to land surface The tubing was secured to the diffusion sampler and to a weighted line at approximately 10 foot intervals by using plastic cable ties The purpose of the tubing was to allow ground water to be collected adjacent to each diffusion sampler by using low flow methodology with a peri staltic pump The diffusion samplers were deployed in 15 wells at NAS North Island during November 11 1999 table 1 All wells were constructed of 4 inch diameter casing The samplers were attached by plas tic cable ties to either a weighted line or a 1 2 inch outside diameter PVC pipe When multiple sections of PVC pipe were required to reach the top of the cas ing the sections were joined using stainless steel screws The PVC pipe was secured to the top of the well casing to prevent the diffusion samplers from shifting during the equilibration period Two of the sampled wells PW 15 and PW 55 contained floating nonaqueous phase liquid LNAPL consisting of free phase petroleum and Stoddard sol vent To install diffusion samplers in these wells a Table 1 Summary of well informatio
27. the ground water samples Sevee and others 2000 The authors suggest that in such cases a more appropriate sampling methodology may be to collect a slug or passive sample from the well screen under the assumption that the water in the well screen is in equilibrium with the surrounding aquifer Isolating a particular contributing fracture zone with straddle packers in an uncased borehole allows depth discrete samples to be collected from the target horizon Hsieh and others 1993 Kaminsky and Wylie 1995 Strategically placed straddle packers often can minimize or eliminate the impact of vertical gradients in the sampled interval However even within a packed interval isolating inflowing fracture zones deviations between VOC concentrations in water from PDB samplers and water sampled by conventional methods still may occur if the conventional method mixes chemically stratified water outside the borehole or if the packed interval straddles chemically heteroge neous zones The use of multilevel PDB samplers and other types of multilevel samplers Ronen and others 1987 Kaplan and others 1991 Schirmer and others 1995 Gefell and others 1999 Jones and others 1999 poten tially can delineate some of the chemical stratification Diffusion sampling and other sampling methodologies however can be influenced by vertical hydraulic gradi ents within the well screen or the sand pack When vertical hydraulic gradients are present within t
28. the well immediately following low flow sampling by using a bladder pump from the depth at which the diffusion sampler had equilibrated well MW 9 only A second method involved using a peristaltic pump to low flow sample ground water adjacent to each of the diffusion sam plers by means of the dedicated Tygon tubing attached to each diffusion sampler The depths were low flow sampled beginning with the shallowest and proceeding to the deepest In some wells MW 5 and MW 12 one of the depths was sampled using a dedicated blad der pump while the remaining depths were sampled using a bladder pump attached to dedicated tubing Two wells MW 13A and MW 13B were low flow sampled by using a peristaltic pump the diffusion samplers were recovered a bladder pump was inserted into each well and the wells were then immediately B 4 Diffusion Sampler Testing at Naval Air Station North Island resampled by low flow methodology using the bladder pump The diffusion samplers were recovered from the wells by means of the attached weighted line or PVC pipe The samplers were cut open and the water was slowly decanted into glass vials pretreated with hydro chloric acid The water samples were sent to a contract laboratory for analysis by using Environmental Pro tection Agency Method 8260B U S Environmental Protection Agency 1999 Replicate samples were col lected from approximately 10 percent of the sampling sites In general both diffusion sa
29. where a nondetectable concentration of an analyte was reported for a sample a value of zero was assigned for the purposes of the ANOVA testing only For the conventional purging each of the three depth intervals evaluated was assigned the same analytical value reported for the one sample collected from that well Results and Discussion A 5 Table 1 Analytical results for samples ug L migrograms per liter First Mobilization Second Mobilization Well ID USGS Micropurge Conventional DMLS Conventional TCE pg L MWII 8 to 23 24 29 8to 10 21 Mw241 3 8 to 40 27 to 33 41 27 to 33 32 Mw242 3 4 to 6 2 8 to 3 5 4 3 3 to 5 3 3 1 trans 1 2 DCE ug L MWII ND ND ND ND ND Mw241 ND to 1 2 0 90 to 0 98 1 0 77 to 1 4 0 99 Mw242 ND ND ND ND ND cis 1 2 DCE ug L MWII 0 95 to 2 3 3 4 3 8 1 1 to 1 4 3 3 Mw241 0 63 to 9 2 6 5 to 7 2 72 6to 11 6 8 Mw242 ND ND ND ND ND 1 1 DCE pg L MWII 34 to 89 170 220 58 to 77 170 Mw241 2 1 to 22 15 to 19 23 19 to 21 18 Mw242 44 to 9 3 8 to 6 3 54 5 2 to 10 3 1 1 1 DCA pg L MWII 0 66 to 1 6 1 6 17 0 54 to 0 69 1 5 Mw241 0 36 to 4 4 3 5 to 3 6 3 6 2 9 to 4 3 3 4 Mw242 ND ND ND ND to 0 22 ND 1 1 2 TCA pg L MWII 0 58 to 1 6 1 3 1 6 0 47 to 0 68 1 5 Mw241 0 32 0 23 to 0 28 0 32 0 22 to 0 27 0 27 Mw242 ND ND ND ND ND 1 2 DCA pg L MWII 0 95 to 2 2 2 2 0 74 to 0 83 1 9 Mw241 1 8 to 16 14 to 16 15 12 to 15 15 Mw242 0 43 to 1 6 0 98 to 3 5 5 3 0 78 to 1 4 3 6 Notes 8 to 23 Range of concentr
30. 0 samples were considered in this study These analytes include trichloroethene TCE trans 1 2 dichloroethene DCE cis 1 2 DCE 1 1 DCE 1 1 dichloroethane DCA 1 2 DCA and 1 1 2 trichloroethane TCA A summary of analytical results for these analytes is presented in table 1 The different methods of sample collection were evaluated using the following criteria accuracy or compa rability of data other method specific criteria and cost These criteria are described in the following sections Accuracy Comparability of Data The analysis of variance ANOVA test was used to compare analytical data collected using the different sampling techniques The limited number of samples available as few as 3 per sampling method precluded the use of linear statistical models in a quantitative manner Therefore the ANOVA was used in a qualitative manner to provide weight of evidence support for data accuracy and similarity The ANOVA test returns a p value between zero and one indicating a pass or fail condition A p value of 0 05 or greater represents a pass indicating that the distributions are similar at the 95 percent confidence level ANOVA is a parametric test and it is common practice to verify that the data fit a parametric distribution prior to applying the tests However due to the limited number of samples in the data set normality tests were not performed on the data sets before performing the ANOVA In instances
31. 00 L N 005 N 005 00S N 005 OIL f oer 005 005 CLS TO89 MIN N 000 1 N 005 000 1 008 8 N 0001 N OOS N 000 T N 00 000 T 009 0001 N 00 9 cc C2089 MIN N 005 N 000 1 000 8 000 1 N 005 N 000 1 N 00 c N 0001 005 001 1 N 00 c 0001 6 CO089 MIN N OOST N 00ST 000 ct 000 ct N 00 N 00 N oos c N 005 0087 000 N 00 c N 005 0765 N OOST N 000 0I 000 000 001 N oos c N 000 01 N oos c N 000 01 00t t 00c L N 00 c N 000701 660 N 000 5 N 000 01 000 SS 000 011 N 000 5 000 01 N 000 000 01 f 00 f 00t L N 000 5 000701 6 N 000 1 N 000 0I 000 LT 000 011 N 000 1 N 000 01 N 0001 N 000 01 0021 f 00 N 0001 N 000701 Lor CO89 MIN N 005 N 000 01 000 vS 000 011 N 005 N 000 01 N 00 c N 000 01 009 008 L N 00 c N 000 0I 289 N 005 N 00001 000 9 000 00c N 00 N 000 01 N oos c N 000 01 008 c 000 vT N 00 c N 000 01 N OOST N 000 000 6 000 r8 N OOST 000 N oos c N 000 S 000 007 N 00 c N 000 Sor C 089 MIN N OOST N OOST 000 000 Lt N OOST N o0 c oos c f 000I 00t oort N 00 c N OOST Loc CO89 MIN N OOS N O00 T 001 6 000 61 N 005 N 000 1 f 0956 f 06r 00c T 00t c N 00 N 0001 CLE C2089 MIN MOT uoisnyig Moj Uuoisnyig uoisnyig uona uona uona 6r 1 61 idet ISM J UIA
32. 000 112 1 100 J 350 100 000 from well PW 17 Free product 10 U 5 000 U 10 U 5 000 U 13 5 700 10 U 5 000 U 120 43 000 from well MW 11 1 145 ug L over a vertical distance of about 7 ft fig 5 The concentrations decreased with depth at some wells MW 5 and PW 66 figs 5 and 7 respec tively and increased with depth at others MW 12 and PW 15 figs 2 and 8 respectively The presence of contaminant stratification in well screens has importance for ground water sam pling In an environment with a sharp concentration gradient small disturbances in the water column can obscure the stratification Thus small amounts of mix ing during low flow sampling can result in large varia tions in VOC concentrations from pumped samples In addition the potential for stratification is an important consideration when selecting a sampling depth For example the data indicate that if the dedi cated bladder pump at well MW 12 had been set about 3 ft deeper the pump would have been in contact with water containing approximately 6 000 ug L more TCE than was present at the original sampling depth If the dedicated bladder pump at well MW 5 had been set about 3 ft shallower the pump would have been in contact with water containing approximately 690 ug L higher concentrations of TCE This consideration is even more important for diffusion samplers which sample only the water in the immediate vicinity of the sampler Therefore when using diffusion
33. 2 a science for a changing world USER S GUIDE FOR POLYETHYLENE BASED PASSIVE DIFFUSION BAG SAMPLERS TO OBTAIN VOLATILE ORGANIC COMPOUND CONCENTRATIONS IN WELLS PART 1 DEPLOYMENT RECOVERY DATA INTERPRETATION AND QUALITY CONTROL AND ASSURANCE Water Resources Investigations Report 01 4060 PART 2 FIELD TESTS Water Resources Investigations Report 01 4061 Prepared in cooperation with the U S AIR FORCE U S NAVAL FACILITIES ENGINEERING COMMAND U S ENVIRONMENTAL PROTECTION AGENCY FEDERAL REMEDIATION TECHNOLOGIES ROUNDTABLE DEFENSE LOGISTICS AGENCY U S ARMY CORPS OF ENGINEERS and INTERSTATE TECHNOLOGY AND REGULATORY COOPERATION WORK GROUP U S Department of the Interior U S Geological Survey Acknowledgments Technical Advisory Board U S Air Force Base Conversion Agency AFBCA U S Air Force Center For Environmental Excellence AFCEE Maj Jeff Cornell AFCEE Mario Ierardi AFBCA Dr Javier Santillan AFCEE Defense Logistics Agency DLA Lt Col Daniel L Welch U S Environmental Protection Agency EPA Steve Schmelling Office of Research and Development ORD Dick Willey EPA Region 1 Kathy Davies EPA Region 3 Richard Steimle Technology Innovation Office TIO U S Naval Facilities Engineering Command NAVFAC C Casey Southern Division Dennis Howe Engineering Service Center Richard G Mach Jr Southwest Division Nick Ta Engineering Service Center U S
34. 2 25 2000 Monterey California the field test Cost savings from more standard well monitoring activities have been reported to range from 25 to 70 percent Alexander and Lammons 1999 Hare 2000 U S Army Corps of Engineers 2000 Brian Peters OHM Remediation Services Corp written commun 2000 Cost savings of PDB sampling over conventional three casing purge sampling are described in the McClellan AFB Envi ronmental Management Directorate report 2000 however calculation errors obscure the actual amount of the savings Due to the availability of reports at the time of publication the case studies included herein are limited to applications at sites where chlorinated aliphatic hydrocarbons are the primary contaminants The case studies present data suggesting that PDB samplers can provide representative concentrations of the target compounds in a variety of environments The method is a cost effective simple alternative to traditional sampling methodologies McClellan Air Force Base Environmental Management Directorate 2000 Passive diffusion membrane sam plers Final August 7 2000 Technology application analysis report McClellan Air Force Base Environmen tal Management Sacramento California Tunks J Guest P and Santillan J 2000 Diffusion sam pler testing of chlorinated VOCs in ground water The Second International Conference on Remediation of Chlorinated and Recalcitrant Compounds May 22 25 2000 M
35. 20 140 0 400 800 1 200 cis 1 2 DICHLOROETHENE TRICHLOROETHENE CONCENTRATION IN CONCENTRATION MICROGRAMS PER LITER IN MICROGRAMS PER LITER EXPLANATION DIFFUSION SAMPLE LOW FLOW PERISTALTIC PUMP SAMPLE WV LOW FLOW BLADDER PUMP SAMPLE U VALUE WAS BELOW THE ANALYTICAL QUANTITATION LIMIT Figure 5 Comparison of diffusion and low flow samples in ground water at well MW 5 Naval Air Station North Island California January 2000 Results and Discussion B 21 0002 Menuer eiuoji eo 5 uomels 289 21 0 sjam sejdures uoisnyip jo uosueduioo 3 pue 2989 pue 089 889 9 v89 MIN v punojJ sejdures pue jo 9 auna Wd SNVADOAIN NI NOLLV3LLNHONOO 1 1 00000Z 000001 0 000 007 000001 0 000 0 000 021 00009 0 0001 00 0 T T OL T T 0L T T c9 T Ir T vt NS annd O89MIN a MOTEMOT as qaivoraad A A 4 A SMNVS 0 C9 qalvorqad L 4 V89MIN N i Dee ATdNVS MOTIMOT 09 MOTH MOT SHIdNVS 1 ec e NOISNsdIG 9 se
36. 6 Denver CO 80225 Phone 888 ASK USGS For additional information write to District Chief U S Geological Survey Stephenson Center Suite 129 720 Gracern Road Columbia SC 29210 7651 Additional information about water resources in South Carolina is available on the World Wide Web at http sc water usgs gov Report can be downloaded from http www itrcweb org and http www frtr gov 5 Em 1 ER TNI EE SE EATS E AA 2 Case Studies Diffusion Sampler Evaluation of Chlorinated VOCs in Groundwater By John Tunks Peter Guest and Javier Santillan A 1 to 8 Diffusion Sampler Testing at Naval Air Station North Island San Diego County California November 1999 to January 2000 U S Geological Survey Water Resources Investigations Report 00 4182 Don Vroblesky and Brian C Peters essere tenete B 1 to B 27 Investigation of Polyethylene Passive Diffusion Samplers for Sampling Volatile Organic Compounds in Ground Water at Davis Global Communication Sacramento California August 1998 to February 1999 U S Geological Survey Open File Report 00 307 By Don A Vroblesky James W Borchers Ted R Campbell and Willey Kinsey C 1 to C 13 Field Testing of Passive Diffusion Bag Samplers for Collection of Ground Water Volatile Organic Compound Concentratio
37. 7 67 67 67 67 68 68 68 68 68 NA 68 68 68 NA 68 68 68 68 NA 69 69 69 69 69 69 69 69 71 Results and Discussion B 5 Table 2 Sampler deployment and recovery information Naval Air Station North Island California November 1999 to January 2000 Continued repl replicate sample NA not applicable low flow bladder pump sample data from OHM Remediation Services Corporation 2000 s Depth to Number Site or Sampling Diffusion diffusion ol days building Well interval sample sampler sampler Date Date diffusion designation identifier identifier laboratory laboratory center installed recovered samplers identifier identifier ft bls were in wells 472 MW 68C2 A 779679 0166 779679 0181 37 2 11 11 99 1 20 00 70 472 MW 68C2 B 779679 0169 779679 0182 39 1 11 11 99 1 20 00 70 472 MW 68C2 B repl NA 19679 0183 39 1 11 11 99 1 20 00 70 472 MW 68C2 C 779679 0171 779679 0184 40 5 11 11 99 1 20 00 70 472 MW 68C2 D 779679 0173 779679 0185 42 0 11 11 99 1 20 00 70 472 MW 68C2 E 779679 0176 779679 0186 44 1 11 11 99 1 20 00 70 472 MW 68C2 F 779679 0178 779679 0187 46 0 11 11 99 1 20 00 70 472 MW 68C2 G 79679 0180 779679 0188 47 8 11 11 99 1 20 00 70 472 MW 68C2 H 779679 0179 779679 0189 49 8 11 11 99 1 20 00 70 472 MW 68C2 I 779679 0177 779679 0190 51 9 11 11 99 1 20 00 70 472 MW 68C2 J 779679 0174 779679 0191 53 9 11 11 99 1 20 00 70 472 MW 68C2 J repl 779679 0175 53 9 11 11 99 1 20 00 70
38. 71 43 p Vroblesky D A Rhodes L C and Robertson J F 1996 Locating VOC contamination in a fractured rock aqui fer at the ground water surface water interface using passive vapor collectors Ground Water v 34 no 2 p 223 230 Vroblesky D A and Robertson J F 1996 Temporal changes in VOC discharge to surface water from a fractured rock aquifer during well installation and operation Greenville South Carolina Ground Water Monitoring and Remediation v 16 no 3 p 196 201 References B 27
39. 779679 0129 523 11 12 99 1 18 00 67 Site 11 MW 5D C 779679 0123 779679 0130 54 2 11 12 99 1 18 00 67 Site 11 MW 5D C repl NA 779679 0134 542 11 12 99 1 18 00 67 Site 11 MW 5D D 779679 0124 779679 0131 55 75 11 12 99 1 18 00 67 Site 11 MW 5D D repl 779679 0125 55 75 11 12 99 1 18 00 67 Site 11 MW 5D E 779679 0126 779679 0132 57 4 11 12 99 1 18 00 67 Site 11 MW 5D F 779679 0127 779679 0133 59 0 11 12 99 1 18 00 67 Site 11 MWO9 NA 779679 0154 779679 0155 31 11 12 99 1 19 00 68 Site 2 S2 MW6A A 79679 0062 779679 0135 6 5 11 13 99 1 18 00 66 Site 2 S2 MW6A B 779679 0063 779679 0136 7 85 11 13 99 1 18 00 66 Site 2 S2 MW6A C 779679 0064 779679 0137 9 2 11 13 99 1 18 00 66 Site 2 S2 MW6A D 779679 0065 779679 0138 10 6 11 13 99 1 18 00 66 Site 2 S2 MW6A E 779679 0066 779679 0139 11 95 11 13 99 1 18 00 66 Site 2 S2 MW6A F 779679 0067 779679 0140 13 3 11 13 99 1 18 00 66 Site 2 S2 MW6A G 779679 0068 779679 0141 14 65 11 13 99 1 18 00 66 Site 2 S2 MW6A H 779679 0069 779679 0142 16 05 11 13 99 1 18 00 66 Site 2 S2 MW6A I 79679 0070 779679 0143 17 5 11 13 99 1 18 00 66 Site 2 S2 MW6A J 779679 0071 779679 0144 18 95 11 13 99 1 18 00 66 Site 2 S2 MW6A J repl 779679 0072 18 95 11 13 99 1 18 00 66 Results and Discussion B 7 Table 3 Comparison of replicate samples collected by diffusion sampler methodology Naval Air Station North Island California January 2000 repl replicate sample ft bls feet below land surface ug L micrograms per liter
40. 9 MIA CS ns ns 4 ns ns ns ns NS ns ns ns ns 6L VO MIN CS ns ns S ns ns ns ns ns ns ns ns ns VO MIN CS N 005 N 000 9 I 00 I N 00S ns N 005 051 N 005 ns 6 99 Md N 005 ns 00c 9 8v N 00S ns N 00 ns f1 005 cL N 00 ns ETE 99 Md N 005 N ST 009 081 005 sc N 00S N ST N 005 005 N ST 8 0 99 Md N 005 N 0 008 6 OLL N 00 N 0 005 TET N 005 8t N 00 05 167 99 Md N 005 N 005 00t 9 000 6 005 005 0001 00t N 005 N 005 N 005 N 005 ELT 99 Md N 005 000 1 000 ET 000 LT N 665 N 0001 0091 000 8 0001 005 0001 OST 99 Md 005 005 000 6c 000 ooc c 006 005 00 9 N ooc c N ooc z 005 005 Dec SS Md 00ST OOST 000 6c 0008 00 c OOST 00 c 008 9 005 005 N OOST OOST ole SS Md Moj uona Uoisnyiq MO Uoisnyiq uona MO uoisnyiq MO Uoisnylq PM epuo uo 0 01 210 2 L S 9 90101 210 1 2 0 1 000c uone10d10 WHO gwp dund 1oppe q Sursn Aq ardues pue ST 10591 poj2ojop q you YN
41. B samplers should be installed vertically across the screened or open interval to determine the zone of highest concentration and whether contaminant stratifi cation is present 4 The samplers should be attached to the weights or weighted line at the time of deployment For samplers utilizing the hanger and weight assembly Figure 2 Example of multiple PDB samplers prepared for deployment the line can be attached directly to the top of the sampler PDB samplers utilizing an outer protective mesh can be attached to a weighted line by using the following procedure a Insert cable ties through the attachment points in the weighted line b At each end of the PDB sampler weave the ends of the cable ties or clamp through the LPDE mesh surrounding the sampler and tighten the cable ties Thus each end of the PDB sampler will be attached to a knot loop in the weighted line by means of a cable tie or clamp The cable ties or clamps should be positioned through the polyethylene mesh in a way that prevents the PDB sampler from sliding out of the mesh c Trim the excess from the cable tie before placing the sampler down the well Caution should be exercised to prevent sharp edges on the trimmed cable ties that may puncture the LDPE 8 User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 1 Deployment Recovery Data Interpretation and Quality Contro
42. Clellan AFB Tunks and others 2000 McClellan AFB Environmental Management Directorate 2000 Because of the length of the McClellan AFB Environmental Management Director ate 2000 study only a summarization of the report is included herein The detailed report is available from McClellan AFB Environmental Management Director ate 5050 Dudley Boulevard Suite 3 McClellan AFB California 95652 1389 Most of the case studies are previously published reports or summaries of previously published reports some of which are authored by non U S Geological Survey personnel Therefore the formatting of the individual reports varies and not all formats are stan dard for the U S Geological Survey Moreover the methods used for these investigations preceded publi cation of standardized approaches for using PDB samplers in wells Therefore investigators should refer to Part 1 of this document for guidance on recom mended methodology for PDB sampler applications rather than to the case studies presented here PDB sampler methodology was compared to conventional purging methods purging at least three casing volumes used at McClellan AFB and Davis Global Communications and to low flow methods used at NAS North Island and Hanscom AFB Both conventional purging and low flow purging were compared with using PDB samplers at NIROP Fridley The study by Tunks and others at McClellan AFB compared the PDB samplers to conventional and low flow t
43. Environmental Protection Agency U S Geological Survey Volatile organic analysis Volatile organic compound Contents User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 1 Deployment Recovery Data Interpretation and Quality Control and Assurance By Don A Vroblesky EXECUTIVE SUMMARY Water filled passive diffusion bag PDB samplers described in this report are suitable for obtaining con centrations of a variety of volatile organic compounds VOCs in ground water at monitoring wells The sug gested application of the method is for long term moni toring of VOCs in ground water wells at well characterized sites The effectiveness of the use of a single PDB sampler in a well is dependent on the assumption that there is horizontal flow through the well screen and that the quality of the water is representative of the ground water in the aquifer directly adjacent to the screen If there are vertical components of intra bore hole flow multiple intervals of the formation contributing to flow or varying concentrations of VOCs vertically within the screened or open interval then a multiple deployment of PDB samplers within a well may be more appropriate for sampling the well A typical PDB sampler consists of a low density polyethylene LDPE lay flat tube closed at both ends and containing deionized water The sampler is posi tioned
44. High equipment in each well Decontamination required Minimal High Moderate Moderate if dedicated equipment is not used Immediacy of sample Slow Slow Rapid Rapid availability Can analytes other than No No Yes Yes VOCs be monitored Can vertical distribution of Possible Possible Partial No contaminants be evaluated Suitable for natural attenu No No Yes Partial ation monitoring Supplemental to the criteria shown in table 3 concerns specific to the USGS and DMLS samplers were noted Being placed in a well for potentially long periods these samplers are susceptible to the effects of fluctuating groundwater elevations If groundwater elevations decrease such that a portion of the diffusion sampler is exposed to air the potential exists for volatilization of VOCs which would compromise the samples collected from these devices A second concern was identified with the DMLS sampling device in that the sample volume of each dialysis cell is only 38 mL When collecting samples for VOC analysis the typical sample container is a 40 mL VOA which will require more than one dialysis cell to fill As shown in table 3 many benefits can be realized through the use of diffusion samplers however these devices also present limitations which may preclude their use in certain groundwater sampling applications Cost Cost estimates per sample for each of the four sampling methods evaluated are presented in table 4 The following expe
45. ON IN MICROGRAMS PER LITER EXPLANATION DIFFUSION SAMPLE LOW FLOW SAMPLE Figure 8 Comparison of diffusion and low flow samples in ground water at well PW 15 Naval Air Station North Island California January 2000 Results and Discussion B 23 concentrations were lowered in ground water col lected from subsequently sampled depths Additional VOC losses by degassing during the use of peristaltic pumps probably resulted in further concentration dif ferences between the two sampling methods TCE concentration data in diffusion samplers collected from wells MW 68B and MW 68C and con centration data in diffusion samplers collected from adjacent well MW 68C2 support the vertical distribu tion indicated by the diffusion samplers in well MW 68C2 figs 6B 6C and 6D Diffusion samplers from well MW 68C2 indicate that the lowest concen trations in the screened interval are below a depth of approximately 55 ft and the detected concentrations are similar to those from the same depth in the adja cent well MW 68C fig 6E Similarly diffusion samplers from wells MW 68C2 and MW 68B both indicate concen trations increasing with depth between approximately 35 and 40 ft fig 6E The TCE concentrations in dif fusion samples from well MW 68B are higher than those from the corresponding depth in well MW 68C2 fig 6E The reasons for the concentration difference between wells MW 68C2 and MW 68B are not kno
46. Station North Island California January 2000 stratified contamination the low flow sample concen trations generally increased higher than the diffusion sample concentrations which is to be expected if the zone of influence for the low flow pumping captured the more contaminated ground water in the well In general the data suggest that diffusion sampling pro vides a more precise delineation of the contaminant stratification within the screened interval than low flow sampling Insight into the use of diffusion samplers in a chemically stratified screened interval can be observed in the data from wells at the MW 68 cluster figs 6D and 6E Unlike the other wells two peristaltic pumps were used to low flow sample well MW 68C2 Start ing simultaneously from both the uppermost and the lowermost sample depths sampling progressed sequentially toward the center of the 25 ft screened interval Results from both the diffusion samples and the low flow samples showed that the uppermost and lowermost parts of the screened interval were rela tively uncontaminated Concentration data from the B 20 Diffusion Sampler Testing at Naval Air Station North Island diffusion samples show that substantially higher TCE concentrations occurred between depths of approxi mately 40 to 50 ft with a sharp peak at about 42 ft fig 6D Thus the first water pulled into the well screen from both ends of the screen was relatively uncontaminated As the low
47. U 5 U 5 U MW 10 13 2 5U 5U 5 U 5U 5 U 5U 1J 5 U MW 10 15 2 5 U 5 U 5 U 5 U 5 U 5 U 5 U 5 U MW 10 17 2 5U 5U 5 U 5U 5 U 5U 5U 5 U MW 10 18 9 5U 5U 5 U 5 U 5 U 5 0 5 0 25 MW 12 30 5 100 U 5U 100 U 5U 100 U 5U 100 U 5 U MW 12 32 2 100 U 5 U 100 U 5 0 100 U 5 0 100 U 5 0 MW 12 33 7 100 U 5 U 100 U 5 0 100 U 5 0 100 U 5 U MW 12 35 1 100 U 100 U 100 U 100 U 100 U 100 U 100 U 100 U MW 12 37 0 100 U 500 U 100 U 120 U 100 U 500 U 100 U 360 U MW 12 38 5 500 U 500 U 500 U 500 U 500 U 500 U 500 U 500 U MW 13A 6 5 5U 5U 5 U 5U 5 U 5U 5U 5 U MW 13A 8 0 5U 5U 5 U 5U 5 U 5U 5U 5 U MW 13A 9 4 5 U 5 U 5 U 5 U 5 U 5 U 5 U 5 0 MW 13A 10 9 5U 5 U 5 U 5U 5 U 5U 5U 5 U MW 13A 12 4 5 U 5 U 5 U 5 U 5 U 5 U 5 U 5 U MW 13A 12 0 NA 5 U NA 5 U NA 5U NA MW 13B 24 9 9 5 5U 5 0 5U 5U 5 5 U MW 13B 26 2 5 4 5 5U 5 5U 5 5 U MW 13B 27 6 1J 41 5 U 5 0 1 J 5 U 5 U 5 U MW 13B 26 0 NA 5 U NA 5 U NA 5 U NA 5 U MW 13C 45 5 5U 5U 5U 5U 5U 5U 5U 5 U MW 13C 46 7 5U 5 U 5U 5 U 5 U 5 U 5U 5 U B 14 Diffusion Sampler Testing at Naval Air Station North Island San Diego County California November 1999 to January 2000 Table 6 Concentrations of benzene ethylbenzene toluene and total xylenes in water from diffusion and low flow sampling Naval Air Station North Island California January 2000 Continued ug L micrograms per liter U value was below the analytical quantitation limit J estimated value NA not applicable sample collected by using bladder pu
48. W 15 and approximately 5 900 ug L in well MW 12 At well MW 5 the 1 1 DCE concentration changed by 3 410 ug L and the TCE concentration changed by Results and Discussion B 25 Table 8 Concentrations of selected volatile organic compounds in free phase fuel JP 5 removed from ground water in water from diffusion samplers deployed in a bucket containing the free phase fuel Naval Air Station North Island California January 2000 ug L micrograms per liter J estimated value U value was below the analytical quantitation limit Sample source 1 1 Dichloroethane 1 1 Dichloroethene cis 1 2 Dichloroethene Tetrachloroethene Trichloroethene ug L ug L ug L ug L ug L Diffusion Free Diffusion Free Diffusion Diffusion Diffusion Free phase Free phase Free phase sampler phase sampler phase sampler sampler sampler fuel fuel fuel water fuel water fuel water water water Free product 4 J 5 000 U 5 U 5 000 U 3 9 J 5 000 U 5 U 5 000 U 2J 5 000 U from well PW 17 Free product 10 U 5 000 U 10 U 5 000 U 10 5 000 U 7 4 300 65 5 200 from well MW 11 Vinyl chloride Benzene Ethylbenzene Toluene Total xylenes ug L ug L ug L ug L ug L Diffusion Free Diffusion Diffusion Diffusion Diffusion Free Free phase Free phase Free phase sampler phase sampler sampler sampler sampler phase fuel fuel fuel fuel wate fuel water water water water Free product 5 U 5 000 U 5 000 U 70 21
49. a D A Vroblesky U S Geological Survey written commun 1998 of semi volatile compounds in contact with PDB samplers showed a higher concentration of phthalates inside the PDB sampler than outside the PDB sampler suggesting that the polyethylene may contribute phthalates to the enclosed water Thus the samplers should not be used to sample for phthalates 3 PDB samplers rely on the free movement of water through the well screen In situations where ground water flows horizontally through the well screen the VOC concentrations in the open interval of the well probably are representative of the aquifer water in the adjacent formation Gillham and others 1985 Robin and Gillham 1987 Kearl and others 1992 Powell and Puls 1993 Vroblesky and Hyde 1997 In these situa tions the VOC concentration of the water in contact with the PDB samplers and therefore the water within the diffusion samplers probably represents local condi tions in the adjacent aquifer However if the well screen is less permeable than the aquifer or the sandpack then under ambient conditions flowlines may be diverted around the screen Such a situation may arise from inad equate well development or from iron bacterial fouling of the well screen In this case the VOC concentrations in the PDB samplers may not represent concentrations in Introduction 5 the formation water because of inadequate exchange across the well screen PDB samplers have not yet b
50. a well The suggested application is for long term monitoring of VOCs in ground water wells Where the screened interval is greater than 10 feet ft the potential for contaminant stratification and or intra borehole flow within the screened interval is greater than in screened intervals shorter than 10 ft It is impor tant that the vertical distribution of contaminants be determined in wells having 10 ft long well screens and that both the vertical distribution of contaminants and the potential for intra borehole flow be determined in wells having screens longer than 10 ft For many VOCS of environmental interest table 1 the VOC concentration in water within the sampler approaches the VOC concentration in water outside of the PDB sampler over an equilibration period The resulting concentrations represent an integration of chemical changes over the most recent part of the equilibration period approximately 48 to 166 hours depending on the water temperature and the type of compound being sampled The approach is inexpensive and has the potential to eliminate or substantially reduce the amount of purge water removed from the well A variety of PDB samplers have been utilized in well applications fig 1 Although the samplers vary in specific construction details a typical PDB sampler consists of a 1 to 2 ft long LDPE tube closed at both ends and containing laboratory grade deionized water fig 1 The typical diameter for PDB sampler
51. ability of passive diffusion samplers and to discuss various factors influencing interpretation of the data The intended audience for the methodology sections of this report is managers and field personnel involved in using PDB samplers The discussion of PDB sampler applicability and interpretation of the data is suited for project managers technical personnel and the regula tory community Part 2 of this report presents case studies of PDB sampler field applications Introduction 3 x XSAN NW AM UE COSA SN Figure 1 Typical water filled passive diffusion bag samplers used in wells including A diffusion bag with polyethylene mesh B diffusion bag without mesh and C bag and mesh attached to bailer bottom Table 1 Compounds tested under laboratory conditions for use with passive diffusion bag samplers From Vroblesky and Campbell 2001 Tested compounds showing good correlation average differences in concentration of 11 percent or less between diffusion sampler water and test vessel water in laboratory tests Benzene 2 Chlorovinyl ether cis 1 2 Dichloroethene 1 1 1 Trichloroethane Bromodichloromethane Dibromochloromethane trans 1 2 Dichloroethene 1 1 2 Trichloroethane Bromoform Dibromomethane 1 2 Dichloropropane Trichloroethene Chlorobenzene 1 2 Dichlorobenzene cis Dichloropropene Trichlorofluoromethane Carbon tetrachloride 1 3 Dichlorobenzene 1 2 Dibromoethane 1 2 3 Trichloropropane Chl
52. ally conductive fractures with a slow velocity borehole flowmeter Canadian Geotechnical Journal v 23 no 1 p 69 78 1990 A thermal flowmeter for the measurement of slow velocities in boreholes U S Geological Survey Open File Report 87 121 Hess K M Wolf S H and Celia M A 1992 Large scale natural gradient tracer test in sand and gravel Cape Cod Massachusetts 3 Hydraulic conductivity variabil ity and calculated macrodispersivities Water Resources Research v 28 no 8 p 2011 2027 Hsieh P A Shapiro A M Barton C C Haeni F P Johnson C D Martin C W Paillet F L Winter T C and Wright D L 1993 Methods of characterizing fluid movement and chemical transport in fractured rocks in Field Trip Guidebook for the Northeastern United States 1993 Boston GSA Cheney J T and Hepburn J C eds RI R30 Amherst University of Massachu setts Department of Geology and Geography 16 User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 1 Deployment Recovery Data Interpretation and Quality Control and Assurance Huckins J N Petty J D Lebo Orazio C E Prest H F Tillitt D E Ellis G S Johnson B T and Manu weera G K 1996 Semipermeable membrane devices SPMDs for the concentration and assessment of bio available organic contaminants in aquatic environments in Ostander G K ed Technique
53. and in water from low flow purging using a bladder pump at the same depth Naval Air Station North Island California January 2000 average percent difference concentration measured in diffusion sampler was lower than concentration measured in low flow sample Diffusion samples Low flow bladder pump samples Constituent Depth in feet Concentration Depth in feet Concentration Percent below land in micrograms below land in micrograms difference surface per liter surface per liter Well MW 9 1 1 Dichloroethene 31 4 000 31 3 500 2 0 1 1 Tetrachloroethene PCE 31 260 31 330 21 0 Trichloroethene 31 3 300 31 3 200 3 0 Well MW 5 Trichloroethene TCE 35 75 360 53 75 280 320 17 Well MW 12 cis 1 2 Dichloroethene 35 1 100 35 1 450 78 cDCE Trichloroethene TCE 35 1 1 800 35 1 2 100 14 Well MW 13A cis 10 9 12 4 46 74 12 61 Within range Vinyl chloride 10 9 12 4 5 8 12 7 4 Within range Well MW 13B cis 1 2 Dichloroethene 24 85 26 15 3 100 2 600 26 3 100 Within range cDCE Toluene 24 85 26 15 9 26 lt 5 Not applicable Total xylenes 24 85 26 15 111 110 26 lt 5 Not applicable Vinyl chloride 24 85 26 15 1 900 2 000 26 1 600 18 pumps may representative of concentrations in ground water at some wells but may underestimate actual concentrations in ground water at other wells Moreover when multiple depths within a screened interval are purged using low flow methods there
54. ane may require between 93 and 166 hours to equilibrate at 10 C T M Sivavec and S S Baghel General Electric Company written commun 2000 The initial equili bration under field conditions may be longer to allow well water contaminant distribution and flow dynamics to restabilize following sampler deployment 2 Water filled polyethylene PDB samplers are not appropriate for all compounds For example although methyl tert butyl ether and acetone Vroblesky 2000 Paul Hare General Electric Company oral commun 2000 and most semivolatile compounds are transmitted through the polyethylene bag laboratory tests have shown that the resulting concentrations were lower than in ambient water A variety of factors influence the ability of compounds to diffuse through the polyethylene membrane These factors include the molecular size and shape and the hydrophobic nature of the compound Compounds having a cross sectional diameter of about 10 angstroms or larger such as humic acids do not pass through the polyethylene because the largest transient pores in polyethylene do not exceed about 10 angstroms in diameter Flynn and Yalkowsky 1972 Hwang and Kammermeyer 1975 Comyn 1985 The samplers are not appropriate for hydrophilic polar molecules such as inorganic ions A detailed discussion of the relation between hydrophobicity and compound transport through polyethylene can be found in Gale 1998 Unpublished laboratory test dat
55. at 166 hours Different equilibra tion times may exist for other compounds Differences in equilibration times if any between single solute or mixed VOC solutions have not yet been thoroughly examined The samplers should be left in place long enough for the well water contaminant distribution and flow dynamics to restabilize following sampler deployment Laboratory and field data suggest that 2 weeks of equili bration probably is adequate for many applications therefore a minimum equilibration time of 2 weeks is suggested In less permeable formations longer equili bration times may be required When applying PDB samplers in waters colder than previously tested 10 C or for compounds without sufficient corrobo rating data a side by side comparison with conven tional methodology is advisable to justify the field equilibration time Executive Summary 1 Following the initial equilibration period the samplers maintain equilibrium concentrations with the ambient water until recovery Thus there is no specified time for sampler recovery after initial equilibration PDB samplers routinely have been left in ground waters having concentrations of greater than 500 parts per million ppm of trichloroethene for 3 months at a time with no loss of bag integrity and at one site the PDB samplers have been left in place in VOC contaminated ground water for 1 year with no reported loss of sampler integrity The effects of long term grea
56. at the target horizon of the well by attachment to a weighted line or fixed pipe The amount of time that the sampler should be left in the well prior to recovery depends on the time required by the PDB sampler to equilibrate with ambi ent water and the time required for the environmental disturbance caused by sampler deployment to return to ambient conditions The rate that the water within the PDB sampler equilibrates with ambient water depends on multiple factors including the type of compound being sampled and the water temperature The concentrations of benzene cis 1 2 dichloroethene tetrachlorethene trichloroethene toluene naphthalene 1 2 dibromoethane and total xylenes within the PDB samplers equilibrated with the concentrations in an aqueous mixture of those compounds surrounding the samplers under laboratory conditions within approximately 48 hours at 21 degrees Celsius A subsequent laboratory study of mixed VOCs at 10 C showed that tetrachloroethene and trichloroethene were equilibrated by about 52 hours but other compounds required longer equilabration times Chloroethane cis 1 2 dichloroethene trans 1 2 dichloroethene and 1 1 dichloroethene were not equilibrated at 52 hours but appeared to be equilibrated by the next sampling point at 93 hours Vinyl chloride 1 1 1 trichloroethane 1 2 dichloroethane and 1 1 dichloroethane were not equilibrated at 93 hours but were equilibrated by the next sampling point
57. ations measured over sampled depth intervals ND Not detected Data validation qualifiers did not affect the usability of the data for this evaluation and are therefore not included in table 1 A 6 User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 2 Field Tests As presented in table 2 in all instances the p values calculated for the populations of results for the different sampling methods exceeded Table 2 ANOVA results 0 05 These ANOVA results indicate that there are no statistically signif icant differences among analytical results obtained using the four groundwater sampling techniques Given that the evaluated diffusion samplers provide comparable accuracy with traditional sampling tech niques other criteria must be considered in evaluating the suitability of one sampling technique over another Analyte p value 1 1 2 TCA 0 74 1 1 DCA 0 99 1 1 DCE 0 47 1 2 DCA 0 88 cis 1 2 DCE 0 96 TCE 0 59 trans 1 2 DCE 0 99 Other Method Specific Criteria Additional qualitative and semi quantitative criteria were considered in this evaluation and are sumarized in table 3 Table 3 Summary of other method specific criteria results Criteria USGS DMLS Micropurge Conventional Ease of use Excellent Fair Poor Fair Labor hours required per 0 66 1 Deals 3 66 sample Generation of IDW liters I 1 100 500 Cost to provide dedicated Low High Low
58. btained from Johnson Screens New Brighton Minnesota in August 1999 and a sampler currently being developed and used by the US Geological Survey USGS The standard sampling methods used for comparison to the diffusion sampling results were 1 Groundwater sampling following conventional purging of at least 3 casing volumes of water and stabilization of water quality parameters i e conventional sampling and 2 Sampling following low flow minimal drawdown purging i e micropurging The groundwater samples were analyzed for total VOCs using US Environmental Protection Agency USEPA Method SW8260B 5030 USEPA 1994 Introduction A 1 FIGURE 1 SITE LOCATION SACRAMENTO McCLELLAN AFB lan North y 89 Rio Linda f_ Sacramento BUS d N 2 0 4 SCALE IN MILES A 2 User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 2 Field Tests MATERIALS AND METHODS Diffusion sampling is a relatively new technology designed to use passive sampling techniques that elimi nate the need for well purging A diffusive membrane capsule is filled with deionized distilled water sealed mounted in a suspension device and lowered to a specified depth in a monitoring well Over time no less than 72 hours VOCs in the groundwater diffuse across the capsule membrane and co
59. ction with pump sampling as appropriate can be used to locate zone s of high est concentration in the well Multiple PDB samplers also may be needed to track the zone of maximum concentration in wells where flow patterns through the screened interval change as a result of ground water pumping or seasonal water table fluctuations PASSIVE DIFFUSION BAG SAMPLER DEPLOYMENT A variety of approaches can be used to deploy the PDB samplers in wells A typical deployment approach described in this section is to attach the PDB samplers to a weighted line It also is acceptable to attach the weights directly to the PDB sampler if the attachment point is of sufficient strength to support the weight The weights attached to the bottom of the line are stainless steel and can be reused but must be thoroughly decontaminated with a detergent before the first use or before using in a different well Rope such as 90 pound 3 16 inch braided polyester can be used as the line for single use applications if it is of suffi cient strength to support the weight and sampler is nonbuoyant and is subject to minimal stretch how ever the rope should not be reused because of the high potential for cross contamination Stainless steel or Teflon coated stainless steel wire is preferable The weighted lines should not be reused in different wells to prevent carryover of contaminants A possible exception is coated stainless steel wire which can be reused aft
60. drology v 8 no 3 4 p 203 224 Ronen Daniel Magaritz Mordeckai and Levy Itzhak 1987 An in situ multilevel sampler for preventive monitoring and study of hydrochemical profiles in aquifers Ground Water Monitoring and Remediation Fall p 69 74 Schirmer M Jones I Teutsch G and Lerner D N 1995 Development and testing of multiport sock samplers for ground water Journal of Hydrology v 171 p 239 257 References 17 Shanklin D E Sidle W C and Ferguson M E 1995 Micro purge low flow sampling of uranium contami nated ground water at the Fernald Environmental Management Project Ground Water Monitoring and Remediation v 15 no 3 p 168 176 Sevee J E White C A and Maher D J 2000 An analy sis of low flow ground water sampling methodology Ground Water Monitoring and Remediation Spring 2000 v 20 no 2 p 87 93 Smith R L Harvey R W and LeBlanc D R 1991 Importance of closely spaced vertical sampling in delineating chemical and microbial gradients in ground water studies Journal of Contaminant Hydrology v 7 p 285 300 Vroblesky D A 2000 Simple inexpensive diffusion sam plers for monitoring VOCs in ground water The Second International Conference on Remediation of Chlorinated and Recalcitrant Compounds May 22 25 2000 Monterey California Vroblesky D A and Campbell T R 2001 Equilibration times stability and compound selectivity of diffusion samplers f
61. e deemed acceptable by local state and Federal regulatory agencies and meet the site specific data quality objectives then a PDB sampler may be approved for use in that well to moni tor ambient VOC concentrations If concentrations from the PDB sampler are higher than concentrations from the conventional method it is probable that concentrations from the PDB sampler adequately represent ambient conditions because there usually is a greater potential for dilution from mixing during sampling using conventional methods than during sampling using PDB samplers If however the conventional method produces concentrations that are significantly higher than those obtained using the PDB sampler then it is uncertain whether the PDB sampler concentrations represent local ambient conditions In this case further testing can be done to determine whether contaminant stratifi cation and or intra borehole flow is present Multiple sampling devices can be used to determine the pres ence of contaminant stratification and borehole flow meters can be used to determine whether intra borehole flow is present When using flowmeters to measure vertical flow in screened boreholes however the data should be considered qualitative because of the potential for water movement through the sand pack Borehole dilution tests Halevy and others 1967 Drost and others 1968 Grisak and others 1977 Palmer 1993 can be used to determine whether water is freely
62. e for sampler recovery PDB samplers have routinely been left in ground waters having concentrations of greater than 500 ppm of TCE for 3 months at a time with no loss of bag integrity and at one site the PDB samplers have been left in place in VOC contaminated ground water for 1 year with no reported loss of sampler integrity Paul Hare General Electric Company oral commun 2000 The effects of long term greater than 1 month PDB sampler deploy ment on sampler and sample integrity have not yet been thoroughly tested for a broad range of compounds and concentrations Moreover in some environments development of a biofilm on the polyethylene may be a consequence of long term deployment Investigations of semipermeable membrane devices SPMDs have shown that the transfer of some compounds may be reduced but not stopped across a heavily biofouled polyethylene membrane Ellis and others 1995 Huckins and others 1996 Huckins and others in press If a heavy organic coating is observed on a PDB sampler it is advisable to determine the integrity of the sample by comparing contaminant concentra tions from the PDB sampler to concentrations from a conventional sampling method before continuing to use PDB samplers for long term deployment in that well Recovery of PDB samplers is accomplished by using the following approach 1 Remove the samplers from the well by using the attached line The PDB samplers should not be exposed t
63. e with no loss of bag integrity and at one site the PDB samplers were left in place in VOC contaminated ground water for 1 year with no reported loss of sampler integrity The effects of long term greater than 1 month PDB sampler deployment on sampler and sample integrity have not yet been thoroughly tested for a broad range of compounds and concentrations In some environ ments development of a biofilm on the polyethylene may be a consequence of long term deployment Investigations of semipermeable membrane devices SPMDs have shown that the transfer of some compounds across a heavily biofouled polyethylene membrane may be reduced but not stopped If a heavy organic coating is observed on a PDB sampler it is advisable to determine the integrity of the sample by comparing sampler results to a conventional sampling method concentrations before continuing to use PDB samplers for long term deployment in that well PDB methodology is suitable for a broad variety of VOCs including chlorinated aliphatic compounds and petroleum hydrocarbons The samplers however are not suitable for inorganic ions and have a limited applicability for non VOCs and for some VOCs For example although methyl tert butyl ether and acetone and most semivolatile compounds are transmitted through the polyethylene bag laboratory tests have shown that the resulting concentrations were lower than in ambient water The samplers should not be used to sample for ph
64. echniques as well as another type of diffusion device the DMLS sampler The sites showed close correspondence between concentrations obtained by the PDB samplers and concentrations obtained by using other techniques at most tested locations Most of the field studies also reported some disagreement between results from the PDB samplers and results from the comparative method at a few wells The places where disagreements between results were observed are of interest because they illustrate differences between the sources of water for each type of sampling method For example in a well at Davis Global Communications where concen trations from the PDB samplers were lower than from the conventional purge heat pulse flowmeter testing was used to show that the water from the purged sampling probably was transported downward from a shallower contaminated aquifer during the well purge When the well was not being pumped however the greatest amount of water entering the screen was from the sand layer adjacent to the screen The data suggest that the PDB samplers provided concentrations characteristic of the aquifer under normal circum stances whereas the pumped sample represented a mixture of water from the near vicinity of the well Introduction 1 screen as well as contaminated water from a shallower horizon Although the two methods did not agree it appears that the PDB samplers provided results more characteristic of the aquifer adjacen
65. ed at both ends and containing deionized water The sampler is positioned at the target horizon by attachment to a weighted line or fixed pipe The amount of time that the samplers should be left in the well prior to recovery depends on the time required by the PDB sampler to equilibrate with ambient water and the time required for environmental disturbances caused by sampler deployment to return to ambient conditions The rate that water within the PDB sampler equilibrates with ambient water depends on multiple factors including the type of compound being sampled and the water temperature Concen trations of benzene cis 1 2 dichloroethene tetra chlorethene trichloroethene toluene naphthalene 1 2 dibromoethane and total xylenes within the PDB samplers equilibrated with the concentrations in an 14 User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 1 Deployment Recovery Data Interpretation and Quality Control and Assurance aqueous mixture of those compounds surrounding the samplers under laboratory conditions within approxi mately 48 hours at 21 C A subsequent laboratory study of mixed VOCs at 10 C showed that tetrachloro ethene and trichloroethene were equilibrated by about 52 hours but other compounds required longer equila bration times Chloroethane cis 1 2 dichloroethene trans 1 2 dichloroethene and 1 1 dichloroethene were not equilib
66. ed to degrees Fahrenheit F by the following equation F 9 5 C 32 Sea level refers to the National Geodetic Vertical Datum of 1929 NGVD of 1929 a geodetic datum derived from a general adjustment of the first order level nets of the United States and Canada formerly called Sea Level Datum of 1929 Chemical concentration in water is expressed in metric units as milligrams per liter mg L or micrograms per liter ug L Additional Abbreviations ft d ft mg 8 L Hg um uL mg mL mL min cubic feet per day cubic feet per milligram gram liter microgram micrometer microliter milligram milliliter milliliter per minute Contents User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 2 Field Tests By Don A Vroblesky editor INTRODUCTION This report presents six case studies where passive diffusion bag PDB samplers were tested under field conditions The sites represent two U S Naval facilities Naval Air Station NAS North Island California and Naval Industrial Reserve Ordnance Plant NIROP Fridley Minnesota and three U S Air Force facilities Davis Global Communications Cali fornia Hanscom Air Force Base AFB Massachu setts and McClellan California The primary ground water contaminants of interest were chlorinated hydrocarbons Two independent studies included herein were done at Mc
67. een adequately tested to determine their response under such conditions 4 VOC concentrations in PDB samplers represent ground water concentrations in the vicinity of the screened or open well interval that move to the sampler under ambient flow conditions This is a limitation if the ground water contamination lies above or below the well screen or open interval and requires the operation of a pump to conduct contaminants into the well for sampling 5 In cases where the well screen or open inter val transects zones of differing hydraulic head and variable contaminant concentrations VOC concentra tions obtained using a PDB sampler may not reflect the concentrations in the aquifer directly adjacent to the sampler because of vertical transport in the well However a vertical array of PDB samplers used in conjunction with borehole flow meter testing can provide insight on the movement of contaminants into or out of the well This information then can be used to help determine if the use of PDB samplers is appropri ate for the well and to select the optimal vertical location s for the sampler deployment 6 In wells with screens or open intervals with stratified chemical concentrations the use of a single PDB sampler set at an arbitrary by convention depth may not provide accurate concentration values for the most contaminated zone However multiple PDB samplers distributed vertically along the screened or open interval in conjun
68. entration in water from the bladder pump table 7 and fig 2 This difference may be due to in well mixing by low flow sampling in a chemically stratified part of the screened interval Data from diffusion samplers show that the VOC concentrations substantially increased with depth over a distance of only 3 4 ft and that the bladder pump was positioned at a transition zone between two depths of differing concentrations table 5 and fig 2 The bladder pump was sampled San Diego County California November 1999 to January 2000 Table 4 Comparison of replicate samples collected by low flow methodology Naval Air Station North Island California January 2000 repl replicate sample sample collected by using bladder pump low flow samples without were collected by using a peristaltic pump data from OHM Remediation Services Corp 2000 ft bls feet below land surface Ug L micrograms per liter J estimated value U value was below the analytical quantitation limit 1 1 dichloroethane 11DCE 1 1 dichloroethene cDCE cis 1 2 dichloroethene TCE trichloroethene Depth to Well diffusion Site or build Ethyl Vinyl Total Identifier and depth sampler ing 1 Bn benzene en chloride xylenes code center designation ug ug L ug ug L ug ug L ug L ft bls MW 10 G 18 8 653 5U 0J 5U 5U 5U 3J MW 10 G repl 18 8 653 5U 5U 1J 5U 5U 5U MW 13B 26 653 5U 5U 3 100 5U 5U 1 600 5U MW 13B 4t
69. entrations while diffusion samplers are capable of producing representative samples even at low less than 20 ug L concentrations According to historical data OHM Remediation Services Corpora tion 2000 TCE has never previously been detected in well MW 10 sampling dates July 1998 March 1999 June 1999 and September 1999 Furthermore a resa mpling of the well using low flow methodology at multiple horizons in February 2000 also showed that TCE was not present Thus it seems that the diffusion samplers accurately reflected VOC concentrations in ground water the source of low TCE concentrations found in water obtained from low flow peristaltic pump sampling is unknown but may represent a cross contamination source not related to local ground water Wells S2 MW 06A and MW 13C contained no detectable VOCs less than 5 ug L in water from either the diffusion samples or from the low flow samples Thus the construction materials used in the San Diego County California November 1999 to January 2000 A E Diffusion sample jaa 4 not completely gt submerged 5 B z 8 lt U 9 2 10 4 Q c3 11 4 amp 12 4 13 1 1 0 20 40 60 80 100 CONCENTRATION IN MICROGRAMS PER LITER EXPLANATION LOW FLOW LOW FLOW DIFFUSION SAMPLE PERISTALTIC BLADDER cis 1 2 DICHLOROETHENE VINYL CHLORIDE A PUMP SAMPLE PUMP SAMPLE Figure 9
70. epl 28 9 379 500 U 500 U 3 200 500 U 9 200 500 U 500 U PW 66 F 33 1 379 5U 130 5U 5U 13 5U 50 PW 66 F repl 33 1 379 5U 120 5U 5U 18 5U 50 obtained using the bladder pump table 7 The reason for the difference in tetrachloroethene concentrations is not known The data from well MW 5 site 11 show that the diffusion samplers performed favorably At well MW 5 where a bladder pump was used to obtain water adjacent to a diffusion sampler and where peri staltic pumps were used at the other depths the differ ence between the TCE concentration in water from the adjacent diffusion sampler and the average concentra tion 300 ug L in water from the bladder pump was relatively small 17 percent difference table 7 Moreover the higher TCE concentration in water from the diffusion sampler compared to the concentration in water from the bladder pump implies that the sample collected by the diffusion method was more discrete than the sample collected by using the bladder pump B 8 Diffusion Sampler Testing at Naval Air Station North Island A comparison between diffusion samples and a bladder pump sample at well MW 12 showed that the TCE concentration in water from the diffusion sampler was similar to the TCE concentration in water from the bladder pump 1 800 and 2 100 ug L respec tively however the cis 1 2 dichloroethene cDCE concentration in water from the diffusion sampler was substantially lower 78 percent than the conc
71. er sufficient decontamination An alternative deployment approach not discussed in this section is to attach the PDB samplers to a fixed pipe in the well Vroblesky and Peters 2000 p 3 also included in Part 2 of this publication The PDB samplers should not con tact non aqueous phase liquid NAPL during deploy ment or retrieval to prevent cross contamination An approach that can be utilized to deploy diffusion sam plers through a layer of floating NAPL is described in the field test at Naval Station North Island California Vroblesky and Peters 2000 p 3 4 also included in Part 2 of this publication If the PDB sampler is to be compared with a conventional pumping approach to sampling then it is suggested that both the pump and the PDB sampler be deployed at the same time with the sampler attached near such as directly below the pump inlet This approach eliminates potential concentration differences between the two methods that may result from well disturbance during equipment removal and deploy ment at the time of sampling An alternative method is to deploy the PDB samplers independently of the pumps and recover the samplers immediately prior to placing the pump down the well PDB samplers are available either prefilled field ready with laboratory grade deionized water or unfilled The unfilled samplers are equipped with a plug and funnel to allow for field filling and sample recovery To fill these samplers remove the
72. exchanged between the aquifer and the well screen Once the source of the difference between the two methods is determined a decision can be made regarding the well specific utility of the PDB samplers Tests may show that VOC concentrations from the PDB samplers adequately represent local ambient conditions within the screened interval despite the higher VOC concentration obtained from the conven tional method This may be because the pumped samples incorporated water containing higher concen trations either from other water bearing zones induced along inadequate well seals or through fractured clay Vroblesky and others 2000 from other water bear ing zones not directly adjacent to the well screen as a result of well purging prior to sampling Vroblesky and Petkewich 2000 or from mixing of chemically strati fied zones in the vicinity of the screened interval Vroblesky and Peters 2000 The mixing of waters from chemically stratified zones adjacent to the screened interval during pumping probably is one of the more important sources of apparent differences between the results obtained from PDB sampling and conventional sampling because such stratification probably is common Vertical strati fication of VOCS over distances of a few feet has been observed in aquifer sediments by using multilevel sampling devices Dean and others 1999 Pitkin and others 1999 and considerable variation in hydraulic conductivity and water chemistr
73. f aquifer concentrations near the outflow horizon In areas where vertical stratification of VOC concentrations is anticipated using multiple PDB samplers may more fully characterize the contami nated horizon than using a single PDB sampler This is particularly true in wells having screens 10 ft or longer however significant VOC stratification has been observed over intervals of less than 5 ft Vroblesky and Peters 2000 Because of the increased probability of vertical concentration or hydraulic gradients within the open interval of long screened greater than 10 ft wells it is advisable to determine the zones of inflow and outflow within the screened or open interval of these wells using borehole flowmeter analysis Hess 1982 1984 1986 1990 Young and others 1998 Comparison of Passive Diffusion Bag Sampling Methodology to Conventional Methodologies Traditional sampling methodologies such as the purge and sample or conventional purging method low flow or low volume sampling and using straddle packers and multilevel samplers produce VOC 12 User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 1 Deployment Recovery Data Interpretation and Quality Control and Assurance concentrations that may differ from concentra tions obtained from PDB samplers because the meth odologies sometimes are influenced in different ways by aquifer h
74. f structural integrity loss during the 2 months of equilibration The VOCs detected in the free phase fuel also were detected in water from the diffusion samplers REFERENCES Gefell M J Hamilton L A and Stout D J 1999 A com parison between low flow and passive diffusion bag sampling results for dissolved volatile organics in frac tured sedimentary bedrock Proceedings of the Petro leum and Organic Chemicals in Ground Water Prevention Detection and Remediation Conference November 17 19 1999 Houston Texas p 304 315 OHM Remediation Services Corporation 2000 Quarterly ground water report Table 1 Summary of ground water VOC analytical results UST 653 December 1999 January 2000 Consultant s Report to U S Naval Facilities Command Southwest Division 9 p U S Environmental Protection Agency 1999 On line SW 846 methods accessed December 21 1999 at URL tap search epa gov epaoswer hazwaste test tx8xxx htm 8 XXX Vroblesky D A and Hyde W T 1997 Diffusion samplers as an inexpensive approach to monitoring VOCs in ground water Ground Water Monitoring and Remedia tion v 17 no 3 p 177 184 Vroblesky D A Nietch C T Robertson J F Bradley P M Coates John and Morris J T 1999 Natural attenuation potential of chlorinated volatile organic compounds in ground water TNX flood plain Savan nah River Site South Carolina U S Geological Sur vey Water Resources Investigations Report 99 40
75. field conditions the samplers should be left in place long enough for the well water contami nant distribution and flow dynamics to restabilize fol lowing sampler deployment The results of borehole dilution studies show that wells can recover to 90 per cent of the predisturbance conditions within minutes to several hours for permeable to highly permeable geo logic formations but may require 100 to 1 000 hours 4 to 40 days in muds very fine grained loamy sands and fractured rock and may take even longer in frac tured shales recent loams clays and slightly fractured solid igneous rocks Halevy and others 1967 In general where the rate of ground water movement past a diffusion sampler is high equilibra tion times through various membranes commonly range from a few hours to a few days Mayer 1976 Harrington and others 2000 One field investigation showed adequate equilibration of PDB samplers to aquifer trichloroethene TCE and carbon tetrachloride CT concentrations within 2 days in a highly perme able aquifer Vroblesky and others 1999 n other investigations PDB samplers recovered after 14 days were found to be adequately equilibrated to chlorinated VOCs Obrien amp Gere Engineers Inc 1997a 1997b Hare 2000 therefore the equilibration period was less than or equal to 14 days for those field conditions Because it appears that 2 weeks of equilibration proba bly is adequate for many applications a minimum
76. flow Diffusion Low flow Diffusion Low flow Diffusion Low flow PW 55 31 9 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U PW 55 33 1 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U PW 66 255 1 000 U 500 U 1 000 U 500 U 1 000 U 500 U 1 000 U 500 U PW 66 27 3 500 U 500 U 500 U 500 U 500 U 500 U 500 U 500 U PW 66 29 1 50 U 500 U 50 U 500 U 50 U 500 U 50 U 500 U PW 66 30 8 25 U 500 U 25 U 500 U 25 U 500 U 25 U 500 U PW 66 32 3 5U 500 U 5U 500 U 5U 500 U 5U 500 U PW 66 33 9 5U 500 U 5U 500 U 5U 500 U 5U 500 U S2 MW 6A 6 5 5U 5U 5U 5U 5U 5U 5U 5U S2 MW 6A 7 9 5U 5U 5U 5U 5U 5U 5U 5U S2 MW 6A 9 2 5U 5U 5U 5U 5U 5U 5U 5U S2 MW 6A 10 6 5U 5U 5U 5U 5U 5U 5U 5U S2 MW 6A 12 0 5U 5U 5U 5U 5U 5U 5U 5U S2 MW 6A 13 3 5U 5U 5U 5U 5U 5U 5U 5U S2 MW 6A 14 7 5U 5U 5U 5U 5U 5U 5U 5U S2 MW 6A 16 1 5U 5U 5U 5U 5U 5U 5U 5U S2 MW 6A 17 5 5U 5U 5U 5U 5U 5U 5U 5U S2 MW 6A 19 0 5U 5U 5U 5U 5U 5U 5U 5U 20 percent the concentrations obtained using the dif fusion samplers were slightly higher than those con centrations obtained using the bladder pump The concentrations obtained using the diffusion samplers in well MW 13B were slightly higher but similar to the concentrations obtained using the peristaltic pump fig 3 Concentrations of toluene and total xylenes were present in water obtained from both the diffusion samplers and the peristaltic pump fig 3 toluene and total xylenes were not detectable less tha
77. fusion Bag Sampler and Sample Recovery sess eene eene enne tenent retener etre nnne 9 Determining Applicability of Passive Diffusion Bag Samplers and Interpretation of Data see 11 Influences of Hydraulic and Chemical Heterogeneity on Sample Quality in Long Screened 15 12 Comparison of Passive Diffusion Bag Sampling Methodology to Conventional Methodologies 12 Quality Control and Assurance iced teet eet ieu ete oe e PP ERR 14 f Ses oben id ca EEA 14 d e a E A E EE ERU a oes iss eG DO dog ete 16 Figures 1 Photo showing typical water filled passive diffusion bag samplers used in wells including diffusion bag with polyethylene mesh diffusion bag without mesh and bag and mesh attached to bailer bottom 4 2 Photo showing example of multiple passive diffusion bag samplers prepared for deployment sss 8 Table 1 Compounds tested under laboratory conditions for use with passive diffusion bag samplers sees 4 Contents Conversion Factors Vertical Datum Acronyms and Abbreviations Multiply By To obtain Length inch in 25 4 millimeter foot ft 0 3048 meter mile mi 1 609 kilometer Area square mile mi 2 590 sq
78. fusion sampler Each set of VOC vials should be analyzed for comparison Approximately 10 percent of the samplers should be replicated The length of the PDB sampler can be adjusted to accommodate the data quality objectives for the sampling event The length can be increased if addi tional volume is required for collection of replicate and matrix spike matrix spike duplicate samples Trip blanks are used to determine whether exter nal VOCs are contaminating the sample due to bottle handling and or analytical processes not associated with field processing Trip blanks are water filled VOA vials prepared offsite stored and transported with the other bottles used for collecting the environ mental sample and then submitted for analysis with the environmental sample Consideration also should be given to the collection of a predeployment PDB trip blank to determine if the PDB samplers are exposed to extraneous VOCs prior to deployment The predeploy ment trip blank should be a PDB sampler that is stored and transported with the field PDB samplers from the time of sampler construction to the time of deploy ment in the wells An aliquot of the predeployment blank water should be collected from the PDB sampler ina VOA vial and submitted for analysis at the time of sampler deployment Water used to construct the diffusion samplers should be analyzed to determine the presence of back ground VOCs Although many VOCs accidentally introduced int
79. h interval below this shallowest depth fig 7 Although the highest TCE concentration obtained by low flow sampling also was at the shallowest horizon it was approximately 24 percent lower than the con centration obtained from the corresponding diffusion sampler and the vertical stratification was less sharply defined These data suggest that as low flow sampling with a peristaltic pump progressed vertically down ward the pumping gradually mixed the TCE contami nated water from the shallowest sampling depth with water from deeper intervals thus obscuring the origi nal contaminant stratification fig 7 A similar effect can be seen in the data from wells MW 12 and PW 15 figs 2 and 8 At these wells the shallowest interval was relatively uncontam inated The comparison between diffusion samples and low flow samples at this shallowest depth showed a relatively close match between cDCE and TCE con centrations However as sampling progressed vertically downward toward the interface of the Results and Discussion B 19 26 27 29 30 31 32 DEPTH BELOW OF CASING IN FEET 33 34 1 1 0 10 000 20 000 30 000 40 000 50 000 CONCENTRATION IN MICROGRAMS PER LITER EXPLANATION DIFFUSION LOW FLOW SAMPLE SAMPLE cis 1 2 DICHLOROETHENE 9 Ov TRICHLOROETHENE A sees Figure 4 Comparison of diffusion and low flow samples in ground water at well PW 55 Naval Air
80. he PDB samplers An approach using rope as a weighted line with knots tied at the appropriate sampler attachment points is discussed below a For 5 ft long or shorter well screens the center point of the PDB sampler should be the vertical midpoint of the saturated well screen length For example if the well screen is at a depth of 55 to 60 ft below the top of casing and the measured depth of the well is 59 ft then the bottom of the well probably has filled with sedi ment In this case the midpoint of the sampler between the attachment points on the line will be midway between 55 and 59 ft or at 57 ft Thus for a 1 5 ft long sampler the attachment points on a weighted line should be tied at distances of 1 25 ft 2 ft 0 75 ft and 2 75 ft 2 ft 0 75 ft from the top of the sediment in the well or the bottom of the well making adjustments for the length of the attached weight When the PDB sampler is attached to the line and installed in the well the center of the sampler will be at 57 ft depth If however independent evidence is available showing that the highest concentration of contaminants enters the well from a specific zone within the screened interval then the PDB sampler should be positioned at that interval b For 5 to 10 ft long well screens it is advisable to utilize multiple PDB samplers verti cally along the length of the well screen for at least the initial sampling fig 2 The purposes of the multiple
81. he pipe extended to land surface A smaller diameter pipe then was used to pound out the rubber cap which was recovered from the well along the outside of the 2 inch diameter pipe by means of a rope attached to the cap The diffusion samplers were lowered into the well through the 2 inch diameter pipe thereby avoid ing direct contact with the LNAPL The pipe was secured in place to allow the diffusion sampler to be recovered without contact with the LNAPL Collection of Pumped Ground Water Samples The diffusion samplers were allowed to remain undisturbed in the well water for 65 to 71 days table 2 The wells were sampled at the time of sam pler recovery using low flow techniques Low flow sampling consisted of purging the well by means of using a dedicated bladder pump or a peristaltic pump connected to the Tygon tubing that had been attached to each of the diffusion samplers prior to deployment Purging was done at a rate of 120 milliliters per minute until measurements of pH water temperature and spe cific conductance stabilized In general purging involved about 20 minutes of pumping and removal of less than 1 gallon of water from each sampling port Decontamination of equipment was not required because each sampling interval had dedicated tubing A variety of methods were used to retrieve the diffusion samplers and to low flow sample the well The first method of sample retrieval involved recover ing the diffusion sampler from
82. he well water contacting the PDB sampler may not be from a horizon adjacent to the PDB sampler Rather the water may represent a mixing of water from other contribut ing intervals within the borehole In a screened well even multilevel samplers with baffles to limit vertical flow in the well cannot prevent influences from Determining Applicability of Passive Diffusion Bag Samplers and Interpretation of Data 13 vertical flow in the gravel pack outside the well screen Such vertical flow can result from small vertical differences in head with depth A field test conducted by Church and Granato 1996 found that vertical head differences ranging from undetectable to 0 49 ft were sufficient to cause substantial flows as much as 0 5 liters minute in the well bore QUALITY CONTROL AND ASSURANCE The sources of variability and bias introduced during sample collection can affect the interpretation of the results To reduce data variability caused during sampling a series of quality control samples should be utilized Replicate samples are important for the quality control of diffusion sampler data Sample replicates provide information needed to estimate the precision of concentration values determined from the combined sample processing and analytical method and to evaluate the consistency of quantifying target VOCs A replicate sample for water filled diffusion samplers consists of two separate sets of VOC vials filled from the same dif
83. ight resting on the bottom of the well with the line taut above the weight Alterna tively the PDB sampler and weight may be suspended above the bottom but caution should be exercised to ensure that the sampler does not shift location Such shifting can result from stretching or slipping of the line or if multiple samplers are attached end to end rather than to a weighted line stretching of the samplers 3 Calculate the distance from the bottom of the well or top of the sediment in the well up to the point where the PDB sampler is to be placed A variety of approaches can be used to attach the PDB sampler to the weight or weighted line at the target horizon The field fillable type of PDB sampler is equipped with a hanger assembly and weight that can be slid over the sampler body until it rests securely near the bottom of the sampler When this approach is used with multiple PDB samplers down the same borehole the weight should only be attached to the lowermost sampler An additional option is to use coated stainless steel wire as a weighted line making loops at appropriate points to attach the upper and lower ends of PDB samplers Where the PDB sampler position varies between sampling events movable clamps with rings can be used When using rope as a weighted line a simple approach is to tie knots or attach clasps at the appropriate depths Nylon cable ties or stainless steel clips inserted through the knots can be used to attach t
84. in shallow homogeneous aquifers have been observed to bias sampling using conventional purging because the majority of the pumped water may come from a particular horizon not related to the contami nated zone and because the intra well flow that intruded the aquifer may not be adequately removed during purging Hutchins and Acree 2000 Thus differences may be observed between concentrations obtained from a pumped sample and from a PDB sample in a chemically stratified interval if the pumped sample represents an integration of water collected from multiple horizons and the PDB sampler repre sents water collected from a single horizon Low flow purging and sampling Barcelona and others 1994 Shanklin and others 1995 disturbs the local ground water less than conventional purge and sample methods Thus samples obtained by PDB samplers are likely to be more similar to samples obtained by using low flow purging than to those obtained by using conventional purge and sample methods Even under low flow conditions however purging still can integrate water within the radius of pumping influence potentially resulting in a deviation from VOC concentrations obtained by PDB sampling One investigation found that in low hydraulic conduc tivity formations low flow sampling methodology caused excessive drawdown which dewatered the screened interval increased local ground water veloci ties and caused unwanted colloid and soil transport into
85. int samples of ground water whereas the bladder pump either collected water from a greater radius of influence or from water induced up the well bore by low flow sampling at shallower depths Wells MW 13A and MW 13B were tested using diffusion samplers and low flow sampling with a peri staltic pump Following sample collection with the peristaltic pump the diffusion samplers were recov ered and the wells were sampled by using a bladder pump bladder pump data from OHM Remediation Services Corporation 2000 The data show that at well MW 13A the cDCE and vinyl chloride concen trations in water obtained using the bladder pump were within the concentration ranges for water obtained from diffusion samplers that bracketed the depth interval of the bladder pump intake table 7 At well MW 13B the cDCE concentration also was within the range measured in those diffusion samplers bracketing the depth of the bladder pump intake table 7 and fig 3 Although vinyl chloride concentra tions differed between the two methods by 16 to Results and Discussion B 9 N 005 N 005 009 L 008 L N 00 N OOS 000 009 c 006 L 008 8 008 1 009 1 005 001 008 00t c 1 00 00I 00 1 005 00L 009 c 068 0tc OLE 001 001 OOS 008 1 N 001 N 001 ost 001 00c c 00571 09c f 98 TSE xt IMAN ns N 001 096 006 1 fc N 001 66 08 00571 18 f 16 L N 001 008
86. involved in using PDB samplers The discussion of passive diffusion sampler applicability and interpretation of the data is 2 User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 1 Deployment Recovery Data Interpretation and Quality Control and Assurance suited for project managers technical personnel and the regulatory community Part 2 of this report presents case studies of PDB sampler field applications INTRODUCTION The use of PDB samplers for collecting ground water samples from wells offers a cost effective approach to long term monitoring of VOCSs at well characterized sites Vroblesky and Hyde 1997 Gefell and others 1999 The effectiveness of the use of a single PDB sampler in a well is dependent on the assumption that there is horizontal flow through the well screen and that the quality of the water is repre sentative of the ground water in the aquifer directly adjacent to the screen If there are vertical components of intra borehole flow multiple intervals of the forma tion contributing to flow or varying concentrations of VOCS vertically within the screened or open interval then deployment of multiple PDB samplers within a well may be more appropriate for sampling the well The samplers consist of deionized water enclosed in a LDPE sleeve fig 1 and are deployed adjacent to a target horizon within a screened or open interval of
87. l and Assurance 5 When using samplers without the protec tive outer mesh the holes punched at the ends of the bag outside the sealed portion can be used to attach the samplers to the weighted line Stainless steel spring clips have been found to be more reliable than cable ties in this instance but cable ties also work well 6 Lower the weight and weighted line down the well until the weight rests on the bottom of the well and the line above the weight is taut The PDB samplers should now be positioned at the expected depth A check on the depth can be done by placing a knot or mark on the line at the correct distance from the top knot loop of the PDB sampler to the top of the well casing and checking to make sure that the mark aligns with the lip of the casing after deployment 7 Secure the assembly in this position sug gested method is to attach the weighted line to a hook on the inside of the well cap Reattach the well cap The well should be sealed in such a way as to prevent surface water invasion This is particularly important in flush mounted well vaults that are prone to flooding 8 Allow the system to remain undisturbed as the PDB samplers equilibrate PASSIVE DIFFUSION BAG SAMPLER AND SAMPLE RECOVERY The amount of time that the samplers should be left in the well prior to recovery depends on the time required by the PDB sampler to equilibrate with ambi ent water and the time required for environmental di
88. l holes pre drilled through a 152 cm long polyvinyl chloride PVC rod and are separated by viton spacers or well seals that fit the inner diameter of the well The PVC rod can accommodate as many as 12 sampling cells pre drilled cylindrical hole spacing is 12 7 cm and a string of up to 5 rods can be connected together for sampling over long screened well intervals Once loaded with the prepared dialysis cells the PVC rods are lowered into a well to the desired depth within the screened interval and are secured with a rope to the top of the well casing A stainless steel weight is attached to the bottom of the deepest PVC rod to ensure that the samplers are positioned at the correct depth in the well and that the PVC rods do not float through the water column Upon retrieval of the PVC rods the dialysis cells are removed from the PVC rod emptied into a decon taminated container for compositing and then transferred to 40 mL VOA containers The samples are preserved and sent to a laboratory for analysis Conventional Sampling Groundwater sampling using conventional well purging involves removing a large volume of water 3 to 5 well casing volumes from the well over a short time The objective of conventional purg ing is to remove all water present within the well casing as well as groundwater present in the surrounding well filter pack Theoretically by removing this water quickly the stagnant water that resided in the well and fil
89. l observation wells MW 12 MW 5 PW 15 and PW 66 The diffusion sampler data show that the vertical change in TCE concentrations over a distance of about 5 ft was approximately 17 500 ug L in well PW 66 approxi mately 7 300 ug L in well PW 15 and approximately 5 900 ug L in well MW 12 At well MW 5 the 1 1 DCE concentration changed by 3 410 ug L and the TCE concentration changed by 1 145 ug L over a vertical distance of about 7 ft Concentrations decreased with depth at some wells PW 66 and MW 5 and increased with depth at others MW 12 and PW 15 The presence of stratification in well screens is important for ground water sampling because small disturbances in the water column can mix the stratification resulting in large variations in VOC concentrations from pumped samples The data imply that care must be exercised when selecting a sampling depth When using diffusion samplers in a well where chemical stratification is suspected within the screened interval multiple diffusion samplers can be used to at least initially delineate the stratification Analytical costs during such an investigation can be minimized by using field gas chromatography or indicator tube technology to delineate the stratification and to select particular samples for more detailed labo ratory analyses The diffusion samplers deployed in buckets con taining free phase JP 5 and Stoddard solution col lected from observation wells did not show evidence o
90. lectromagnetic borehole flowmeter U S Environmental Protection EPA 600 R 98 058 56 p 18 User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 1 Deployment Recovery Data Interpretation and Quality Control and Assurance User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 2 Field Tests By Don A Vroblesky editor U S Geological Survey Water Resources Investigations Report 01 4061 Prepared in cooperation with the U S AIR FORCE U S NAVAL FACILITIES ENGINEERING COMMAND U S ENVIRONMENTAL PROTECTION AGENCY FEDERAL REMEDIATION TECHNOLOGIES ROUNDTABLE DEFENSE LOGISTICS AGENCY U S ARMY CORPS OF ENGINEERS and INTERSTATE TECHNOLOGY AND REGULATORY COOPERATION WORK GROUP 2156 science for a changing world Columbia South Carolina U S DEPARTMENT OF THE INTERIOR GALE A NORTON Secretary U S GEOLOGICAL SURVEY Charles G Groat Director Use of trade product or firm names in this publication is for descriptive purposes only and does not imply endorsement by the U S Geological Survey Copies of this report can be obtained from U S Environmental Protection Agency USEPA National Service Center for Environmental Publications NSCEP Box 42419 Cincinnati OH 45242 0419 and U S Geological Survey Branch of Information Services Box 2528
91. logies the second section examines the differences in sample quality between these meth odologies in situations of hydraulic and chemical heterogeneity Influences of Hydraulic and Chemical Heterogeneity on Sample Quality in Long Screened Wells Sampling biases and chemical variability in long screened wells which can be loosely defined as wells having significant physical and chemical hetero geneity within the screened interval and in the adja cent aquifer Reilly and Leblanc 1998 have been the subject of numerous investigations Sources of chemi cal variability in such wells include non uniform flow into wells Robbins and Martin Hayden 1991 Reilly and Gibs 1993 Chiang and others 1995 Church and Granato 1996 Reilly and LeBlanc 1998 lithologic heterogeneity Reilly and others 1989 Robbins 1989 Martin Hayden and others 1991 Gibs and others 1993 Reilly and Gibs 1993 and in well mixing In a well open across a chemically or hydraulically heterogeneous section of the aquifer differences in the sampling methodology can produce significant differences in the sampling results Long screened wells have the potential to redistribute chemical constituents in the aquifer where there are vertical hydraulic gradients within the screened interval Water can move into the well from one horizon and exit the well at a different horizon Church and Granato 1996 Reilly and LeBlanc 1998 If there is vertical flow in the scree
92. low sampling methods and peristaltic pumps than in samples col lected by the diffusion samplers This substantial dif ference in concentrations between the two methods is expected if VOCs were lost by degassing as a result of Results and Discussion B 17 DEPTH BELOW TOP OF CASING IN FEET 30 32 34 36 38 40 0 400 800 1 200 1 600 2 000 DEPTH BELOW TOP OF CASING IN FEET 40 0 2 000 4 000 6 000 8 000 10 000 1 1 DICHLOROETHANE CONCENTRATION 1 1 DICHLOROETHENE CONCENTRATION 30 32 34 36 IN MICROGRAMS PER LITER 30 32 34 36 IN MICROGRAMS PER LITER DEPTH BELOW TOP OF CASING IN FEET DEPTH BELOW TOP OF CASING IN FEET 38 38 L 40 40 1 0 1 000 2 000 3 000 2 000 4 000 6 000 8 000 10 000 cis 1 2 DICHLOROETHENE CONCENTRATION TRICHLOROETHENE CONCENTRATION IN MICROGRAMS PER LITER IN MICROGRAMS PER LITER EXPLANATION 9 DIFFUSION SAMPLE LOW FLOW PERISTALTIC PUMP SAMPLE V LOW FLOW BLADDER PUMP SAMPLE Figure 2 Comparison of diffusion and low flow samples in ground water at well MW 12 Naval Air Station North Island California January 2000 B 18 Diffusion Sampler Testing at Naval Air Station North Island San Diego County California November 1999 to January 2000 24 m a fy e 25 5 Z N x oO a 26r d
93. mp data from OHM Remediation Services Corporation 2000 Benzene Ethylbenzene Toluene Total xylenes Well un ug L ug L ug L ug L Diffusion Low flow Diffusion Low flow Diffusion Low flow Diffusion Low flow MW 13C 48 1 5U 5U 5U 5U 5U 5U 5U 5 U MW 13C 46 0 NA 5 U NA 5 U NA 5 U NA 5 U MW 68A 21 7 25 U 5 25 U 5 25 U 5 25 U 5 U MW 68A 23 0 50 U 5 U 50 U 5U 50 U 5U 50 U 5U MW 68B 34 5 250 U 5 000 U 250 U 5 000 U 250 U 5 000 U 250 U 5 000 U MW 68B 37 0 5 000 U 5 000 U 5 000 U 5 000 U 5 000 U 5 000 U 5 000 U 5 000 U MW 68B 38 5 5 000 U 5 000 U 5 000 U 5 000 U 5 000 U 5 000 U 5 000 U 5 000 U MW 68C 56 0 50 U 50 U 50 U 50 U 50 U 50 U 50 U 50 U MW 68C 57 9 100 U 50 U 100 U 50 U 100 U 50 U 100 U 50 U MW 68C 59 0 250 U 12 J 250 U 50 U 250 U 50 U 250 U 50 U MW 68C2 37 3 1 000 U 500 U 1 000 U 500 U 1 000 U 500 U 1 000 U 500 U MW 68C2 39 1 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U 2 500 U MW 68C2 40 5 5 000 U 2 500 U 5 000 U 2 500 U 5 000 U 2 500 U 5 000 U 2 500 U MW 68C2 42 1 10 000 U 2 500 U 10 000 U 2 500 U 10 000 U 2 500 U 10 000 U 2 500 U MW 68C2 44 2 10 000 U 2 500 U 10 000 U 2 500 U 10 000 U 2 500 U 10 000 U 2 500 U MW 68C2 46 1 10 000 U 1 000 U 10 000 U 1 000 U 10 000 U 1 000 U 10 000 U 1 000 U MW 68C2 47 9 10 000 U 5 000 U 10 000 U 5 000 U 10 000 U 5 000 U 10 000 U 5 000 U MW 68C2 49 9 10 000 U 2 500 U 10 000 U 2 500 U 10 000 U 2 500 U 10 000 U 2 500 U MW 68C2 52 0 2 500
94. mplers appear to have been successful in approxi mately locating the zone of highest concentrations between the depths of 37 to 52 ft fig 6E VOC concentrations in water collected from well MW 13A varied less and generally were lower for peristaltic pump sampling compared to diffusion sampling fig 9 Following low flow sampling using a peristaltic pump well MW 13A was immediately resampled by low flow sampling using a bladder pump Although subject to the same mixing potential as the peristaltic pump the bladder pump has less potential for volatilization loss than the peristaltic pump and thus probably provides a more representa tive sample than the peristaltic pump The concentra tions of cDCE and TCE in water obtained using low flow sampling methods with a bladder pump approxi mated the average of concentrations obtained in water from the diffusion samplers directly above and below the bladder pump fig 8 These findings suggest that data obtained by using the diffusion samplers provided depth specific VOC concentrations while the data from low flow sampling represented a mixing of waters in well MW 13A In well MW 10 low flow peristaltic pump sam pling detected low concentrations 30 ug L or less of whereas diffusion sampling detected none table 5 This difference in concentrations 15 unusual because the potential for volatilization loss using the peristaltic pump usually results in underestimating ambient conc
95. mples and low flow samples compared well with their respective replicate samples tables 3 and 4 RESULTS AND DISCUSSION VOC concentrations in water obtained from dif fusion samplers were similar to concentrations obtained by using low flow sampling methods for most of the tested wells tables 5 and 6 respectively As will be shown most concentration differences between the two sampling methods probably can be attributed to VOC degassing during peristaltic pump sampling or to in well mixing Comparison of Diffusion Sampler Results to Bladder Pump Results Tests showing the most direct comparison between diffusion sampling and low flow sampling were in wells where a bladder pump was used to low flow sample The test producing the least amount of well water disturbance was in well MW 9 where a dif fusion sampler was recovered immediately following low flow sampling using a bladder pump from the same depth Concentrations of 1 1 dichloroethene 1 1 DCE and trichloroethene TCE obtained using the diffusion sampler agreed well 12 and 3 percent difference respectively with those obtained using the bladder pump table 7 The difference is about the same as the differences approximately 12 percent in 1 1 DCE and TCE concentrations measured in repli cate samples collected by using a dedicated bladder pump at well MW 5D table 4 Thus 12 percent is within the sample collection variability for 1 1 DCE and TCE Agreement between the
96. n Naval Air Station North Island California ft feet ft bls feet below land surface ft msl feet relative to mean sea level Elev elevation NM not measured NA not avail able TOC top of casing A bladder pump attached to the diffusion sampler B peristaltic pump using tubing attached to individual diffusion samplers C same as B except one depth was sampled using a bladder pump attached to a diffusion sampler D same as B except the well was resampled using a bladder pump following removal of the diffusion samplers Depth to Site or Depth to Saturated Depth to Elev of Low flow building i Pete screen Screen water water sampling designation top ft bls ft length ft ft bls ft msl method 653 MW 10 5 20 0 13 0 7 01 2 65 B 653 MW 13A 4 14 0 8 18 6 01 1 81 D 653 MW 13B 24 3 29 2 5 00 6 15 1 53 D 653 MW 13C 44 8 49 8 5 00 6 00 1 61 D 472 MW 68 C2 37 63 0 25 0 NM NA B 472 MW 68A 14 24 0 2 76 21 38 2 34 B 472 MW 68B 33 40 0 5 00 21 42 2 33 B 472 MW 68C 64 3 70 5 5 00 21 6 1 99 B 379 PW 15 20 35 0 9 94 23 34 2 61 B 379 PW 55 20 35 0 9 33 24 32 2 34 B 379 PW 66 20 35 0 10 0 25 10 2 40 B Site 11 MW 12 30 39 7 13 7 NM NA C Site 11 MW 5D NA 60 0 35 5 NM NA C Site 11 MW 9 23 31 9 4 10 28 18 5 64 A Site 2 S2 MW 6A 5 20 0 14 3 5 64 2 35 B Methods B 3 rubber cap was placed on the lower end of a section of 2 inch diameter PVC pipe and lowered into the well to a depth below the LNAPL The top end of t
97. n 5 ug L in water from the bladder pump The data suggest that the diffusion samplers performed equally well with the bladder pump in wells MW 13A and MW 13B for cDCE The higher concentrations of vinyl chloride toluene and total xylenes in water from the diffusion samplers relative to water from the bladder pump indi cate that the diffusion samplers obtained more discrete samples from these wells however disturbing the well water by using the peristaltic pump and removing the B 16 diffusion samplers prior to sampling with the bladder pump may have induced mixing and affected the qual ity of the water sampled by the bladder pump Comparison of Diffusion Sampler Results to Peristaltic Pump Results The remaining comparisons between diffusion sam pler and low flow sampler methods utilized multiple diffusion sampling and low flow sampling points within screened intervals At most depths low flow sampling was conducted by using peristaltic pumps In contrast to bladder pumps using peristaltic pumps in some wells potentially could cause degassing of sam ples during recovery which could result in underesti mating actual VOC concentrations Thus VOC concentrations in water obtained using peristaltic Diffusion Sampler Testing at Naval Air Station North Island San Diego County California November 1999 to January 2000 Table 7 Comparison of concentrations of selected volatile organic compounds in water from a diffusion sampler
98. ned or open inter val and the zone of low hydraulic head outflow from the well 15 within the contaminated horizon then the PDB samplers or any standard sampling methodol ogy can underestimate or not detect the contamina tion The reason is that in this case the contaminated horizon does not contribute water to the well under static conditions Instead water from other horizons with higher hydraulic head will invade the contami nated horizon by way of the well screen Under pumped conditions the majority of the extracted water will be from the most permeable interval which may not be the contaminated zone Even when pumping induces inflow from the contaminated interval much of that inflow will be a reflection of the residual invaded water from other horizons In this situation a substantial amount of purging would be required before water representative of the aquifer could be obtained Jones and Lerner 1995 Such sampling is not likely to reflect a significant contribution from the contaminated zone and concentrations in the contami nated zone probably will be underestimated Similarly if VOC contaminated water is flow ing into the well and is exiting the well at a different horizon then VOCs will be present along the screened interval between the two horizons In this case VOC concentrations in the screened interval may be repre sentative of aquifer concentrations at the inflow horizon but may not be representative o
99. ness districts or busy streets where vehicle traffic control is a concern 7 Multiple PDB samplers distributed vertically along the screened or open interval may be used in conjunction with borehole flow meter testing to gain insight on the movement of contaminants into and out of the well screen or open interval or to locate the zone of highest concentration in the well Analytical costs when using multiple PDB samplers sometimes can be reduced by selecting a limited number of the samplers for labora tory analysis based on screening by using field gas chro matography at the time of sample collection 8 Because the pore size of LDPE is only about 10 angstroms or less sediment does not pass through the membrane into the bag Thus PDB samplers are not subject to interferences from turbidity In addition none of the data collected suggest that VOCs leach from the LDPE material or that there is a detrimental effect from the PDB material on the VOC sample Limitations 1 PDB samplers integrate concentrations over time This may be a limitation if the goal of sampling is to collect a representative sample at a point in time in an aquifer where VOC concentrations substantially change more rapidly than the samplers equilibrate Laboratory results obtained indicate that a variety of compounds equilibrated within 48 hours at 21 Vroblesky and Campbell 2001 Vinyl chloride 1 1 1 trichloroethane 1 2 dichloroethane and 1 1 dichloroeth
100. nologies were applied to large scale monitoring programs a reduction in the per sample cost would probably be realized due in part to reusable equipment that is associated with some of the sampling methods As shown in table 4 the cost per sample using the USGS diffusion sampler was substantially less than using any other methods Conversely the DMLS sampler per sample cost was substantially more that any other method CONCLUSIONS The Air Force groundwater diffusion sampler evaluation indicates that diffusive sampling technology can be a cost effective and accurate method for environmental groundwater monitoring of VOCs However use of diffusion samplers may not be appropriate for all applications Of the diffusion sampling technologies evaluated the USGS sampler is the recommended device based on the evaluation criteria presented herein Additional comparisons between the different sampling technologies should be performed to develop a more robust data set upon which to base analytical result comparisons Particularly varying hydrogeologic settings e g low perme ability to high permeability aquifers and increasing the number of wells in the evaluation would allow for more thorough evaluation of the comparability of the analytical data If natural attenuation monitoring is required a combination of sampling techniques should be considered For instance annual monitoring of natural attenuation parameters can be performed using a traditional
101. ns at Naval Industrial Reserve Ordnance Plant Fridley Minnesota November 1999 to May 2000 U S Geological Survey Water Resources Investigations Report 00 4246 By Don A Vroblesky and Matthew D Petkewich eee eere rennen nnne tne D 1 to D 10 Evaluation of a Diffusion Sampling Method for Determining Concentrations of Volatile Organic Compounds in Ground Water Hanscom Air Force Base Massachusetts U S Geological Survey Water Resources Investigations Report 00 4242 By NDS VR E 1 to E 20 Summarization by Walter Berger MitreTech Inc of Passive Diffusion Membrane Samplers Final August 7 2000 Technology Application Analysis Report McClellan Air Force Base Environmental Management Sacramento California By McClellan Air Force Base Environmental Management Directorate F 1 to F 1 Contents Conversion Factors Vertical Datum Acronyms and Abbreviations Multiply inch in foot ft mile mi square mile mi foot per day ft d foot squared per day ft d gallons per minute gal min gallons per day gal d inches per year in yr gallon gal By Length 25 4 0 3048 1 609 Area 2 590 Flow 0 3048 0 09294 0 06308 0 003785 25 4 Volume 3 785 To obtain millimeter meter kilometer square kilometer meter per day meter squared per day liter per second cubic meter per day millimeters per year liter Temperature is given in degrees Celsius C which can convert
102. nses were considered in the development of a cost analysis for each different sampling method labor equipment and disposal or management of investigation derived waste IDW Some of the costs involved in these activities are one time expenses that are not incurred each time a sample is collected e g PVC rods for use with the DMLS samplers and stainless steel weights Furthermore labor and material costs can vary depending on the scope of the sampling event e g it is less expensive on a unit cost basis to collect 100 samples than to collect 5 samples However to present the most accurate estimate of costs associated with Table 4 Cost summary this evaluation only the costs incurred during this field study were considered in the cost analysis Labor costs were based on actual hours expended as docu Sampling Cost per mented in the field notes and the burdened labor rate for a typical field scientist LES SESS EE Equipment costs were taken directly from invoices when available or were esti USGS 65 mated from vendor quotes Costs associated with disposal or management of DMLSTM 555 Micropurge 308 IDW can vary widely depending on the approach used For this analysis the only CREE saad costs considered in the management of IDW are those dealing with containerizing the waste Results and Discussion A 7 As noted these costs are approximated based on the limited scope of this investigation If these sampling tech
103. ntaminant concentrations in the water inside the sampler attain equilibrium with the ambient groundwater The sampler is subsequently removed from the well and the water within the diffusion sampler is transferred to a sample container and submitted for analysis The diffusive membranes evaluated in this study are rated for VOCs only These membranes are not appropriate for monitoring larger or more electrically charged molecules Once a diffusion sampler is placed in a well it remains undisturbed until equilibrium is achieved between the water in the well casing and the water in the diffusion sampler Depending on the hydrogeologic characteristics of the aquifer the diffusion samplers can reach equilibrium within 3 to 4 days Vroblesky and Campbell 1999 however for this evaluation a minimum 14 day equilibrium period was used Groundwater samples collected using the diffusion samplers are thought to be representative of water present within the well during the previous 24 to 72 hours USGS Sampler The standard USGS diffusion sampler shown in figure 2 consists of water filled low density polyethylene tubing which acts as a semi permeable membrane The USGS sampler typically is constructed of a 45 centimeter cm long section of 5 08 cm diameter 4 mil polyethylene tubing that is heat sealed on both ends The sampler holds approximately 300 milliliters mL of deionized distilled water A longer 7 62 cm diameter sam pler that holds approximatel
104. o heat or agitated 2 Examine the surface of the PDB sampler for evidence of algae iron or other coatings and for tears in the membrane Note the observations in a sampling field book If there are tears in the membrane the sample should be rejected If there is evidence that the PDB sampler exhibits a coating then this should be noted in the validated concentration data 3 Detach and remove the PDB sampler from the weighted line Remove the excess liquid from the exte rior of the bag to minimize the potential for cross contamination 4 A variety of approaches may be used to trans fer the water from the PDB samplers to 40 mL volatile organic analysis VOA vials One type of commer cially available PDB sampler provides a discharge device that can be inserted into the sampler If discharge devices are used the diameter of the opening should be kept to less than about 0 15 inches to reduce volatilization loss Two options are presently available to recover water from the sample using discharge devices One option involves removing the hanger and weight assembly from the sampler inverting the sampler so that the fill plug is pointed upward and removing the plug The water can be recovered by directly pouring in a manner that minimizes agitation or by pouring through a VOC discharge accessory inserted in place of the plug The second approach involves piercing the sampler near the bottom with a small diameter discharge tube and allo
105. o the diffusion sampler water probably will reequilibrate with surrounding water once the diffusion samplers are deployed some VOCs may become trapped within the diffusion sampler water For example acetone which is a common laboratory contaminant does not easily move through the poly ethylene diffusion samplers Paul Hare General Elec tric Company oral commun 1999 Thus acetone inadvertently introduced into the diffusion sample water during sampler construction may persist in the samplers resulting in a false positive for acetone after sampler recovery and analysis SUMMARY Water filled passive diffusion bag PDB sam plers described in this report are suitable for obtaining a variety of VOCs in ground water at monitoring wells The suggested application for PDB samplers is for long term monitoring of VOCs in ground water wells at well characterized sites Where the screened interval is greater than 10 ft the potential for contaminant stratification and or intra borehole flow within the screened interval is greater than in screened intervals shorter than 10 ft It is suggested that the vertical distri bution of contaminants be determined in wells having 10 ft long well screens and that both the vertical dis tribution of contaminants and the potential for intra borehole flow be determined in wells having screens longer than 10 ft A typical PDB sampler consists of a 1 to 2 ft long low density polyethylene lay flat tube clos
106. od of sampling VOCs in Acknowledgments ground water at the base VOC concentrations in water obtained from diffusion samplers set at multiple levels in wells are compared to VOC concentrations in water obtained from low flow sampling Diffusion samplers were placed in 15 observation wells and 2 samplers were placed in buckets of free phase JP 5 and Stod dard solvent The Department of the Navy Southwestern Division Naval Facilities Command funded this work The fieldwork was a cooperative effort between the U S Geological Survey and OHM Remediation Services Corp B 2 Diffusion Sampler Testing at Naval Air Station North Island San Diego County California November 1999 to January 2000 METHODS Diffusion samplers were tested in 15 wells at NAS North Island California VOC concentrations in water from the diffusion samplers were compared to VOC concentrations in water from low flow sampling ports open adjacent to each diffusion sampler Low flow sampling was accomplished by using a peristaltic pump at most sites and a bladder pump at selected sites Diffusion Sampler Construction and Deployment Each diffusion sampler consisted of a 2 inch diameter low density polyethylene LDPE tube heat sealed at both ends and containing deionized water On the outside of each sampler an LDPE mesh provided abrasion protection This sampling methodology is patented patent number 5 804 743 and is available for non exclusive licensing from
107. onterey California 10 p U S Army Corps of Engineers and Dames amp Moore 2000 Diffusive sampler study Dover Air Force Base Dover Delaware Consultant s report prepared for the U S Department of the Air Force Dover Air Force Base 45 p 2 User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 2 Field Tests Diffusion Sampler Evaluation of Chlorinated VOCs in Groundwater By John Tunks Peter Guest and Javier Santillan Diffusion Sampler Evaluation of Chlorinated VOCs in Groundwater By John Tunks and Peter Guest Parsons Engineering Science Inc Denver Colorado USA and Javier Santillan Air Force Center for Environmental Excellence San Antonio Texas USA ABSTRACT Groundwater sample collection using diffusion samplers represents a relatively new technology that utilizes passive sampling methods for monitoring volatile organic compounds VOCs in groundwater The potential benefits and cost savings of diffusion sampler use as an instrument for long term monitoring are significant as no purge waters are generated and labor requirements for sampler installation and retrieval are minimal The efficacy of diffusion samplers for evaluating chlorinated VOCs in groundwater was assessed Using two types of diffusion samplers groundwater samples were collected at discrete depths to assess vertical contamination profiles Groundwater samples also we
108. or collection of ground water VOC concen trations Advances in Environmental Research v 5 no 1 p 1 12 Vroblesky D A and Hyde W T 1997 Diffusion samplers as an inexpensive approach to monitoring VOCs in ground water Ground Water Monitoring and Remedia tion v 17 no 3 p 177 184 Vroblesky D A Nietch C T Robertson J F Bradley P M Coates John and Morris J T 1999 Natural attenuation potential of chlorinated volatile organic compounds in ground water TNX flood plain Savan nah River Site South Carolina U S Geological Survey Water Resources Investigations Report 99 4071 43 p Vroblesky D A and Peters B C 2000 Diffusion sampler testing at Naval Air Station North Island San Diego County California November 1999 to January 2000 U S Geological Survey Water Resources Investigations Report 00 4812 27 p Vroblesky D A and Petkewich M D 2000 Field testing of passive diffusion bag samplers for volatile organic compound concentrations in ground water Naval Industrial Reserve Ordnance Plant Fridley Minnesota November 1999 and May 2000 U S Geological Survey Water Resources Investigations Report 00 4246 10 p Wolf S H Celia M A and Hess K M 1991 Evaluation of hydraulic conductivities calculated from multiport permeameter measurements Ground Water v 29 no 4 p 516 552 Young S C Julian H E Pearson H S Molz and Boman 1998 Application of the e
109. oroethane 1 4 Dichlorobenzene trans 1 3 Dichloropropene 1 1 2 2 Tetrachloroethane Chloroform Dichlorodifluoromethane Ethyl benzene Tetrachloroethene Chloromethane 1 2 Dichloroethane Naphthalene Vinyl chloride 1 1 Dichloroethene Toluene Total xylenes Tested compounds showing poor correlation average differences in concentration greater than 20 percent between diffusion sampler water and test vessel water in laboratory tests Acetone Methyl rert butyl ether Styrene T M Sivavec and S S Baghel General Electric Company written commun 2000 User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 1 Deployment Recovery Data Interpretation and Quality Control and Assurance Summary of Passive Diffusion Bag Sampler Advantages and Limitations Advantages 1 samplers have the potential to eliminate or substantially reduce the amount of purge water asso ciated with sampling 2 PDB samplers are inexpensive 3 The samplers are easy to deploy and recover 4 Because PDB samplers are disposable there is no downhole equipment to be decontaminated between wells 5 A minimal amount of field equipment is required 6 Sampler recovery is rapid Because of the small amount of time and equipment required for the sampling event the method is practical for use where access is a problem or where discretion is desirable that is residential communities busi
110. ounty California November 1999 to January 2000 Table 2 Sampler deployment and recovery information Naval Air Station North Island California November 1999 to January 2000 Continued repl replicate sample NA not applicable low flow bladder pump sample data from OHM Remediation Services Corporation 2000 Depth to Number Site or Sampling Low tlow Diffusion diffusion i days building Well interval sample sampler sampler Date Date diffusion designation identifier identifier laboratory laboratory center installed recovered samplers identifier identifier ft bls were in wells 379 PW 66 C 779679 0073 779679 0147 29 1 11 10 99 1 18 00 69 379 PW 66 D 779679 0108 779679 0148 30 8 11 10 99 1 18 00 69 379 PW 66 E 779679 0074 779679 0149 32 3 11 10 99 1 18 00 69 379 PW 66 F 779679 0075 779679 0150 33 9 11 10 99 1 18 00 69 379 PW 66 F NA 779679 0152 33 9 11 10 99 1 18 00 69 Site 11 MW 12 A 779679 0006 779679 0012 30 5 11 13 99 1 17 00 65 Site 11 MW 12 B 779679 0007 779679 0013 32 1 11 13 99 1 17 00 65 Site 11 MW 12 C 779679 0008 779679 0014 33 7 11 13 99 1 17 00 65 Site 11 MW 12 D 779679 0009 779679 0057 35 1 11 13 99 1 17 00 65 Site 11 MW 12 D repl NA 779679 0060 35 1 11 13 99 1 17 00 65 Site 11 MW 12 E 79679 0010 779679 0058 36 9 11 13 99 1 17 00 65 Site 11 MW 12 F 779679 0011 779679 0059 38 5 11 13 99 1 17 00 65 Site 11 MW 5D A 779679 0121 779679 0128 50 8 11 12 99 1 18 00 67 Site 11 MW 5D B 779679 0122
111. ovember 1999 to January 2000 higher concentrations of vinyl chloride toluene and total xylenes in water from the diffusion samplers in wells MW 13A and MW 13B compared to water from the bladder pump imply that the concentrations obtained by the bladder pump may have underesti mated actual concentration as a result of mixing in these wells Concentration differences between the diffusion sampling and bladder pump sampling meth ods were noted in samples from well MW 12 and probably are related to mixing in a chemically strati fied part of the screened interval The findings of this investigation suggest that diffusion samplers provide a viable sampling alternative for VOCs in ground water in most tested wells at NAS North Island Comparisons of volatile organic compound VOC concentrations in water obtained by using dif fusion samplers to concentrations obtained by low flow sampling using a peristaltic pump were used to gain information on the vertical distribution of con tamination in the wells In several wells the probable effects of mixing or volatization during pumping resulted in lower VOC concentrations in water obtained by using the peristaltic pump compared to concentrations obtained by using the diffusion sam plers however the data from the low flow sampling supported the vertical VOC stratification identified by using the diffusion samplers Substantial VOC stratification was observed in the screened intervals of severa
112. part of this package This is a Water Resources Investigations Report with USGS Director s approval A 8 User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 2 Field Tests Diffusion Sampler Testing at Naval Air Station North Island San Diego County California November 1999 to January 2000 USGS Water Resources Investigations Report 00 4182 By Don A Vroblesky and Brian C Peters Diffusion Sampler Testing at Naval Air Station North Island San Diego County California November 1999 to January 2000 By Don A Vroblesky and Brian C Peters ABSTRACT Volatile organic compound concentrations in water from diffusion samplers were compared to concentrations in water obtained by low flow purging at 15 observation wells at the Naval Air Station North Island San Diego California Mul tiple diffusion samplers were installed in the wells In general comparisons using bladder pumps and diffusion samplers showed similar volatile organic carbon concentrations In some wells sharp concentration gradients were observed such as an increase in cis 1 2 dichloro ethene concentration from 100 to 2 600 micro grams per liter over a vertical distance of only 3 4 feet In areas where such sharp gradients were observed concentrations in water obtained by low flow sampling at times reflected an average concentration over the area of influence however concen
113. rated at 52 hours but appeared to be equili brated by the next sampling point at 93 hours Vinyl chloride 1 1 1 trichloroethane 1 2 dichloroethane and 1 1 dichloroethane were not equilibrated at 93 hours but were equilibrated by the next sampling point at 166 hours Different equilibration times may exist for other compounds Differences in equilibration times if any between single solute or mixed VOC solutions have not yet been thoroughly examined The samplers should be left in place long enough for the well water contaminant distribution and flow dynamics to restabilize following sampler deployment Laboratory and field data suggest that 2 weeks of equilibration probably is adequate for many applica tions Therefore a minimum equilibration time of 2 weeks is suggested In less permeable formations longer equilibration times may be required When deploying PDB samplers in waters colder than previously tested 10 or for compounds without sufficient corroborating data a side by side compari son with conventional methodology is advisable to justify the field equilibration time Following the initial equilibration period the samplers maintain equilibrium concentrations with the ambient water until recovery Thus there is no speci fied maximum time for sampler recovery after initial equilibration PDB samplers have routinely been left in ground waters having concentrations of greater than 500 ppm of TCE for 3 months at a tim
114. re collected following low flow minimal drawdown purging and conventional purging techniques Results obtained using the various sampling techniques suggest that the diffusion samplers provide comparable accuracy with and can be significantly less expensive than traditional sampling techniques INTRODUCTION Parsons Engineering Science Inc Parsons ES was retained by the US Air Force Center for Environmental Excellence Technology Transfer Division AFCEE ERT to perform an evaluation of passive groundwater diffu sion sampling technology The diffusion sampler evaluation is part of the AFCEE ERT Remedial Process Optimi zation RPO demonstration project being performed at six Air Force bases AFBs nationwide One of these bases McClellan AFB California figure 1 was selected as the site for this evaluation A field study was performed in August 1999 at a site on McClellan AFB where deep groundwater more than 30 meters below ground surface is contaminated with various chlorinated VOCs as a result of solvent disposal into burn pits during the 1940s through 1970s The objective of the diffusion sampler evaluation was to evaluate the efficacy of this groundwater sampling method relative to standard sampling methods Field sampling was conducted using two types of diffusion samplers to collect groundwater samples from varying depths at selected monitoring wells The diffusion samplers evaluated included the commercially available DMLS sampler o
115. rogeneity of the aquifer within the screened or open interval of the well and the relative permeability of the well screen Summary 15 REFERENCES Barber C and Davis G B 1987 Representative sampling of ground water from short screened boreholes Ground Water v 25 no 5 p 581 587 Barcelona M Wehrmann H A and Varljen M D 1994 Reproducible well purging procedures and VOC stabi lization criteria for ground water sampling Ground Water v 32 p 12 22 Chiang C C Raven Gary and Dawson Clint 1995 The relationship between monitoring well and aquifer solute concentrations Ground Water v 32 no 5 p 718 126 Church P E and Granato GE 1996 Effects of well design and sampling methods on bias of water quality samples Ground Water v 34 no 2 p 262 273 Cohen R M and Rabold R R 1988 Simulation of sam pling and hydraulic tests to assess a hybrid monitoring well design Ground Water Monitoring Review v 8 no 1 p 55 59 Comyn J 1985 Polymer Permeability New York Elsevier Applied Science Publishers Ltd 383 p Dean S M Lendvay J M Barcelona M J Adriaens P and Katopodes N D 1999 Installing multilevel sam pling arrays to monitor ground water and contaminant discharge to a surface water body Ground Water Monitoring and Remediation Fall 1999 p 90 96 Drost W Klotz D Koch A Moser H Neurnaier F and Rauert W 1968 Point dilutions methods of inves
116. s in Aquatic Toxicol ogy Boca Raton Fla CRC Lewis Publishers p 625 655 Huckins J N Petty J D Prest H F Clark R C Alverez D A Orazio C E Lebo J A Cranor W L and Johnson B T in press A guide for the use of semiper meable membrane devices SPMDs as samplers of waterborne hydrophobic organic contaminants Report for the American Petroleum Institute API Washing ton DC API publication number 4690 Hutchins S R and Acree S D 2000 Ground water sam pling bias observed in shallow conventional wells Ground Water Monitoring and Remediation Winter 2000 p 86 93 Hwang S T and Kammermeyer K 1975 Membranes in Separations Malabar Fla Robert E Krieger Publish ing Company Inc 559 p Jones Ian and Lerner D N 1995 Level determined sam pling in an uncased borehole Journal of Hydrology v 171 p 291 317 Jones Ian Lerner D N and Baines O P 1999 Multiport sock samplers Alow cost technology for effective multilevel ground water sampling Ground Water Monitoring and Remediation v 19 no 1 p 134 142 Kaminsky J F and Wylie A H 1995 Vertical contaminant profiling of volatile organics in a deep fractured basalt aquifer Ground Water Monitoring and Remediation v 15 no 2 p 97 103 Kaplan Edward Banerjee Sujit Ronen Daniel Margaritz Mordeckai Machlin Alber Sosnow Michael and Koglin Eric 1991 Multilayer sampling in the water table region of a
117. s used in a 2 inch diameter well is approximately 1 2 inches however other dimensions may be used to match the well diameter Equilibration times may be longer for larger diameter PDB samplers On the outside of the PDB sampler a low density polyethylene mesh some times is used for protection against abrasion in open boreholes and as a means of attachment at the pre scribed depth The PDB sampler can be positioned at the target horizon by attachment to a weighted line or by attachment to a fixed pipe PDB samplers for use in wells are available commercially Authorized distributors as of March 2001 are Columbia Analytical Services 800 695 7222 www caslab com and Eon Products 800 474 2490 www eonpro com A current list of vendors and PDB sampler construction details can be obtained from the U S Geological Survey Technology Transfer Enterprise Office Mail Stop 211 National Center 12201 Sunrise Valley Drive Reston Virginia 20192 telephone 703 648 4344 fax 703 648 4408 PDB samplers employ patented technology U S patent number 5 804 743 and therefore require that the user purchase commercially produced samplers from a licensed manufacturer or purchase a nonexclusive license for sampler construction from the U S Geological Survey Technology Enterprise Office at the above address The purposes of this document are to present methods for PDB sampler deployment and recovery to discuss approaches for determining the applic
118. sandy aquifer Ground Water v 29 no 2 p 191 198 Kearl P Korte N and Cronk T 1992 Suggested modifi cations to ground water sampling procedures based on observations from the colloidal borescope Ground Water Monitoring Review v 12 no 2 p 155 166 Keely J F and Boateng K 1987 Monitoring well installa tion purging and sampling techniques Part 1 Conceptualizations Ground Water v 25 no 3 p 3300 313 Martin Hayden J M 2000 Sample concentration response to laminar wellbore flow Implications to ground water data variability Ground Water v 38 no 1 p 12 19 Martin Hayden J M Robbins G A and Bristol R D 1991 Mass balance evaluation of monitoring well purg ing Part II Field tests at a gasoline contamination site Journal of Contaminant Hydrology v 8 no 3 4 p 225 241 Mayer L M 1976 Chemical water sampling in lakes and sediments with dialysis bags Limnology and Ocean ography v 21 p 909 912 Obrien amp Gere Engineers Inc 19972 Passive bag sampling results JMT Facility Brockport New York Consult ant s report to General Electric Company Albany New York October 10 1997 10 p 1997b Passive bag sampling results JMT Facility Brockport New York Consultant s report to General Electric Company Albany New York December 12 1997 10 p Palmer C D 1993 Borehole dilution tests in the vicinity of an extraction well Journal of Hydrology v 146 p
119. sturbances caused by sampler deployment to return to ambient conditions The rate that the water within the PDB sampler equilibrates with ambient water depends on multiple factors including the type of compound being sampled and the water temperature The concen trations of benzene cis 1 2 dichloroethene cDCE tetrachlorethene PCE trichloroethene tolu ene naphthalene 1 2 dibromoethane EDB and total xylenes within the PDB samplers equilibrated with the concentrations in an aqueous mixture of those compounds surrounding the samplers under laboratory conditions within approximately 48 hours at 21 C Vroblesky and Campbell 2001 A subsequent labora tory study of mixed VOCs at 10 C showed that PCE and TCE were equilibrated by about 52 hours but other compounds required longer equilibration times T M Sivavec and S S Baghel General Electric Company written commun 2000 Chloroethane cDCE trans 1 2 dichloroethene and 1 1 dichloroethene were not equilibrated at 52 hours but appeared to be equilibrated by the next sampling point at 93 hours Vinyl chloride 1 1 1 trichloroethane 1 2 dichloroethane and 1 1 dichloroethane were not equilibrated at 93 hours but were equilibrated by the next sampling point at 166 hours Different equilibration times may exist for other compounds Differences in equilibration times if any between single solute or mixed VOC solutions have not yet been thoroughly examined Under
120. t to the screened interval Typically other field studies also found that concentration differences between the PDB samplers and the pumping methodology used for comparison often could be attributed to an obscuring of the contaminant stratification by the mixing of water during pumping Field evidence to support this hypothesis is shown in the reports on NAS North Island Hanscom AFB McClellan AFB and NIROP Fridley In general the data show that even when the results of the PDB sampling and the conventional or low flow purging approaches disagree the results of the PDB sampling often appear to accurately reflect the local concentrations whereas those of the pumped sampling method reflect a mixing The investigation at McClellan AFB by Tunks and others included in this report shows a cost comparison for various sampling methods however some of these costs include a one time investment for References Alexander A W and Lammons T L 1999 Measurement of volatile organic compounds DO pH and ORP by diffusive ground water sampling technology in Pro ceedings of the Petroleum and Organic Chemicals in Ground Water Prevention Detection and Remedia tion Conference November 17 19 1999 Houston Texas p 41 45 Hare P W 2000 Passive diffusion bag samplers for moni toring chlorinated solvents in ground water The Second International Conference on Remediation of Chlori nated and Recalcitrant Compounds Batelle May 2
121. ted to minimize drawdown Because micropurging relies on a pumping rate that does not exceed the natural groundwater recharge rate the water elevation in the well must be monitored to ensure that drawdown does not Field Activities Three monitoring wells were selected for use in this evaluation In each of the wells a maximum of three depth intervals spaced equally across the well screen were monitored using the different sampling methods Using the two types of diffusion samplers as designed it was necessary to perform the diffusion sampling consec utively as samples from the two types of diffusion samplers could not be collected concurrently from the same interval within a well To evaluate the potential changes in groundwater concentrations over the sampling periods approximately 14 days between diffusion sampler collection events conventional groundwater sampling was per formed following completion of each diffusion sampling event Significant differences in groundwater chemistry measured between the two sampling events could be normalized using the two sets of conventional groundwater data RESULTS AND DISCUSSION Of the 67 analytes included in the SW8260B analysis 17 were reported to have detectable concentrations in at least one of the samples submitted for analysis For the purposes of comparing the analytical accuracy or comparability using the different sampling methods only those analytes that were detected in at least 1
122. ter FIGURE 3 DMLS DIFFUSION SAMPLER Connecting hook PVC rod Polypropylene Viton well cylinder seal Dialysis cell Cellulose acetate filter Dialysis cell Monitoring well Assembled DMLS sampler prior to being lowered into monitoring well A 4 User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 2 Field Tests pack will be replaced with fresh groundwater from the surrounding formation with minimal mixing The fresh groundwater that is then sampled is considered to be representative of the local groundwater Rapid drawdown of the water level in a well is not uncommon and often wells are purged dry using this method Conventional purging is frequently performed using a bailer or a high flow submersible pump e g Grundfos Redi Flo2 pump Micropurging The objective of micropurging is to remove a small volume of water at a low flow rate from a small portion of the screened interval of a well without mixing water among vertical zones Ideally by placing the inflow port of a pump at a prescribed depth within the screened interval of a well and by withdrawing water at a slow rate groundwater will be drawn from the aquifer into the well only in the immediate vicinity of the pump This discrete depth sampling allows for vertical definition of contamination in the aquifer The pumping rate is adjus
123. ter than 1 month PDB sampler deployment on sampler and sample integ rity have not yet been thoroughly tested for a broad range of compounds and concentrations however Moreover in some environments development of a biofilm on the polyethylene may be a consequence of long term deployment Investigations of semipermeable membrane devices SPMDs have shown that the trans fer of some compounds across a heavily biofouled poly ethylene membrane may be reduced but not stopped If a heavy organic coating is observed on a PDB sampler it is advisable to determine the integrity of the sample by comparison to a conventional sampling method before continuing to use PDB samplers for long term deployment in that well Recovery consists of removing the samplers from the well and immediately transferring the enclosed water to 40 milliliter sampling vials for anal ysis The resulting concentrations represent an integra tion of chemical changes over the most recent portion of the equilibration period approximately 48 to 166 hours depending on the water temperature and the type of compound The method has both advantages and limitations when compared to other sampling methods Advan tages include the potential for PDB samplers to elimi nate or substantially reduce the amount of purge water associated with sampling The samplers are relatively inexpensive and easy to deploy and recover Because PDB samplers are disposable there is no downhole eq
124. thalates because of the potential for the LDPE to contribute phthalates to the water sample When attempting to determine whether the use of PDB samplers is appropriate at a particular well a common approach is to do a side by side comparison with a conventional sampling method This approach is strongly suggested in wells having temporal concentra tion variability In a well having relatively low tempo ral concentration variability comparison of the PDB sampler results to historical concentrations may pro vide enough information to determine whether the PDB samplers are appropriate for the well In general if the two approaches produce concentrations that agree within a range deemed acceptable by the local state and Federal regulatory agencies then use of a PDB sampler in that well will provide VOC concentra tions consistent with the historical record If concentra tions from the PDB sampler are higher than concentra tions from the conventional method then it is probable that the concentrations from the PDB sampler are an adequate representation of ambient conditions If how ever the conventional method produces concentrations that are substantially higher than the concentrations found by using the PDB sampler then the PDB sam pler may or may not adequately represent local ambi ent conditions In this case the difference may be due to a variety of factors including mixing or transloca tion due to hydraulic and chemical hete
125. tigating ground water flow by means of radioisotopes Water Resources Research v 4 no 1 p 125 146 Ellis G S Huckins J N Rostad C E Schmitt C J Petty J D and MacCarthy Patrick 1995 Evaluation of lipid containing semipermeable membrane devices for monitoring organochlorine contaminants in the upper Mississippi River Environmental Science and Tech nology v 14 no 11 p 1875 1884 Flynn G L and Yalkowsky S H 1972 Correlation and prediction of mass transport across membrane I Influ ence of alkyl chain length on flux determining proper ties of barrier and diffusant Journal of Pharmaceutical Science v 61 p 838 852 Gale R W 1998 Three compartment model for contami nant accumulation by semipermeable membrane devices Environmental Science and Technology v 32 p 2292 2300 Gefell M J Hamilton L A and Stout D J 1999 A com parison between low flow and passive diffusion bag sampling results for dissolved volatile organics in frac tured sedimentary bedrock in Proceedings of the Petro leum and Organic Chemicals in Ground Water Prevention Detection and Remediation Conference November 17 19 1999 Houston Texas 304 315 Gibs Jacob Brown G A Turner K S MacLeod C L Jelinski J C and Koehnlein S A 1993 Effects of small scale vertical variations in well screen inflow rates and concentrations of organic compounds on the collection of representative ground water quality
126. trations obtained by using the diffusion sampler seemed to represent the immediate vicin ity of the sampler When peristaltic pumps were used to collect ground water samples by low flow purging the volatile organic compound concen trations commonly were lower than concentra tions obtained by using diffusion samplers This difference may be due to loss of volatiles by degassing under negative pressures in the sam pling lines induced while using the peristaltic pump mixing in the well screen or possible short circuiting of water from an adjacent depth Diffusion samplers placed in buckets of free phase jet fuel JP 5 and Stoddard solvent from observation wells did not show evidence of struc tural integrity loss during the 2 months of equilibration and volatile organic compounds detected in the free phase fuel also were detected in the water from the diffusion samplers INTRODUCTION Low density polyethylene diffusion samplers filled with deionized water or air have been shown to be an inexpensive alternative sampling method for volatile organic compounds VOCs in contaminated wells or in ground water discharge zones beneath surface water bodies Vroblesky and others 1996 Vroblesky and Robertson 1996 Vroblesky and Hyde 1997 Vroblesky and others 1999 Gefell and others 1999 The use of diffusion samplers in wells has gen erated substantial interest due to their capability to sample ground water without the need for prior
127. uare kilometer Flow foot per day ft d 0 3048 meter per day foot squared per day ft7 d 0 09294 meter squared per day gallon per minute gal min 0 06308 liter per second gallon per day gal d 0 003785 cubic meter per day inch per year in yr 25 4 millimeter per year Volume gallon gal 3 785 liter Temperature is given in degrees Celsius C which can converted to degrees Fahrenheit F by the following equation F 9 5 C 32 Sea level refers to the National Geodetic Vertical Datum of 1929 NGVD of 1929 a geodetic datum derived from a general adjustment of the first order level nets of the United States and Canada formerly called Sea Level Datum of 1929 Chemical concentration in water is expressed in metric units as milligrams per liter mg L or micrograms per liter ug L EDB AFCEE cDCE f d ITRC LDPE L um uL mg mL mL min MTBE NAVFAC NAPL PDB PCE TCE USEPA USGS VOA VOC Additional Abbreviations 1 2 Dibromomethane Air Force Center for Environmental Excellence cis 1 2 Dibromoethene cubic feet per day cubic feet per milligram degrees Celsius gram Interstate Technology Regulatory Cooperation low density polyethylene liter microgram micrometer microliter milligram milliliter milliliter per minute Methyl tert butyl ether Naval Facilities Engineering Command non aqueous phase liquid passive diffusion bag Tetrachloroethene Trichloroethene U S
128. uenb se onea N sem o dures q 8 penunuo2 o00z Menuer eoep puejs yoy Budwes pue pejoejes JO Results and Discussion B 11 N 005 N 00 c 000 000 8 N 005 N 005 006 5 009 9 N 00 c N 00 c N 00 c N 00 c 9 0 SS Md N 005 N 00 000 c 000 6 N 00 N 005 006 5 00 9 N OOST N 005 N 00 c N 00 6 8C N 008 N 00ST 000 000 6 N 005 N 00 00475 00 L N OOST N 00 c N 00 c N 00 c TLT SS Md N 005 N 005 00 L 00 L f OCI f 041 001 00t c N 00 N 00 00S N 005 ctt N OOS 005 008 S 00 N 005 N 005 008 1 006 1 N 005 N 005 00S N 005 STE N 005 N OST 00 000 N 005 f 65 009 1 006 7 N 005 N OST N 00S 46 TOE N OST N 001 00 c 08I N OST f 001 00c c 00 T 99 f 9E f LL 8c SI Md 6 cL c6 E 8 fl 058 051 g TS cS 99 c8 SI fc ns 007 fv tc VST N 005 N 00 000 11 0056 9 N 005 N 005 N 005 N 005 f 08 005 005 9 1 005 N 005 000 11 000 L 100 N 00 00 N 00 0c6 f OSE 00 N 005 6c C 089 MIN N 005 N 005 001 6
129. uipment to be decontaminated between wells and there is a minimum amount of field equipment required The samplers also have the potential to delineate contaminant stratification in the formation across the open or screened intervals of monitoring wells where vertical hydraulic gradients are not present In addition the samplers integrate concen trations over time which may range between about 48 to 166 hours depending on the compound of interest Because the pore size of LDPE is only about 10 angstroms or less sediment does not pass through the membrane into the bag Thus PDB samplers are not subject to interferences from turbidity In addition none of the data collected suggest that VOCs leach from the LDPE material or that there is a detrimental effect on the VOC sample from the PDB material Water filled polyethylene PDB samplers are not appropriate for all compounds The samplers are not suitable for inorganic ions and have a limited applica bility for non VOCs and for some VOCs For example although methyl fert butyl ether and acetone and most semivolatile compounds are transmitted through the polyethylene bag laboratory tests have shown that the resulting concentrations were lower than in ambient water variety of factors influence the ability of compoundis to diffuse through the polyethylene These factors include the molecular size and shape and the hydrophobic nature of the compound Unpublished lab oratory test data of
130. ursn Aq papajo o dures y rung uorojop popou pue uonejnuenb ejdures st jou WN sem N sem ojdwes q Jod 8 0002 Menuer pulls UUON pue ui ej nejoA papajas JO suoneJueouo Gg Diffusion Sampler Testing at Naval Air Station North Island B 10 San Diego County California November 1999 to January 2000 0 N OST 0011 oorr N 05 N OST N OS OST N 05 OST N 05 N OST 0 65 89 05 N 001 088 004 N 05 001 N os 001 f OI N 001 N os N 001 CLS 89 N 06 N 05 089 1 10 N OS N OS 05 05 0 N 05 05 0 95 89 0005 n 000 5 000 c9 000 0 I 0005 000 N 000 S N 00075 f 000 S 008 000 n 000 lt 889 MIN 000 0005 000 v 000 c9 000 A o00 c N 000 N 000 f oor v OOL L 0005 n 0005 OLE 889 MIN N 000S N OST 000 82 000 67 N 0005 f 6 N 000 5 f 8 002 008 5 N 0005 f S8 S889 MIN ns N 05 061 061 I N OS 18 091 5 N OS 5 N 05 0 c V89 MN ns N sc Occ OLS c L
131. well purging The Naval Air Station NAS North Island in San Diego County California has been used since 1917 as an air station harbor and training base The base is approximately 1 mile west across San Diego Bay from the San Diego metropolitan area fig 1 Activities at the base have resulted in ground water contamination by a variety of compounds including chlorinated aliphatic hydrocarbons and petroleum hydrocarbons In addition free phase JP 5 jet fuel and Stoddard solvent mineral spirits are present locally in the ground water Stoddard solvent is a refined petroleum product typically used as a thinning agent for paints coatings waxes printing inks and adhesives a solvent in photocopy toners and in dry cleaning and as a degreaser for engine parts The purpose of this report is to present the find ings of an investigation to determine whether the use of polyethylene deionized water filled diffusion Introduction B 1 0 75 MILES IL 0 75KILOMETERS P D gt 75 5 PP igs mL EE AR SAN ens m 4 000 FEET 8 1220 METERS APPROXIMATE SCALE Figure 1 Location of observation wells Naval Air Station North Island California samplers is a viable meth
132. wing water to flow through the tube into the VOA vials In each case flow rates can be controlled by tilting or manipulating the sampler Alternatively the PDB sampler can be cut open at one end using scissors or other cutting devices which have been decontaminated between use for different wells Water can then be transferred to 40 ml vials by gently pouring in a manner that mini mizes water agitation Acceptable duplication has been obtained using each method Preserve the samples according to the analytical method The sampling vials should be stored at approximately 4 C in accordance with standard sampling protocol Laboratory testing suggests that there is no substantial change in the VOC concentrations in PDB samplers over the first several minutes after recovery however the water should be transferred from the water filled samplers to the sample bottles immediately upon recovery 5 A cost effective alternative when using multi ple PDB samplers in a single well is to field screen water from each sampler using gas chromatography These results can be used to decide which of the multi ple PDB samplers should be sent to an EPA approved laboratory for standard analysis Typically at least the sample containing the highest concentration should be analyzed by a laboratory 10 User s Guide for Polyethylene Based Passive Diffusion Bag Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 1 Deployment Recover
133. wn however two explanations can be postulated One explanation is that the contaminant concentra tions in well MW 68C2 may have been shifted down ward as a result of a vertical hydraulic gradient within the well Water level measurements are not shown for well MW 68C2 because they would reflect only com positing across the screened interval however evi dence for such a hydraulic gradient can be seen in the water level data for wells MW 68B and MW 68C The water level in well MW 68B is 0 34 ft higher than the water level in well MW 68C indicating a net down ward hydraulic gradient between the two depths table 1 Water levels remeasured in March 2000 confirmed the hydraulic gradient Because well MW 68C2 is only about 5 ft from wells MW 68B and MW 68C and because the screened interval of well MW 68C2 hydraulically connects the depths sampled by wells MW 68B and MW 68C the probability is high that there also is a downward hydraulic gradient within well MW 68C2 An alternative explanation is that lithologic heterogeneities in the screened zone place the contamination at slightly different depths in differ ent wells Evidence for such heterogeneity is the clay layer at a depth of 37 5 to 40 ft in well MW 68C driller s log Richard Wong OHM Remediation writ ten commun 2000 Despite the uncertainty regarding B 24 Diffusion Sampler Testing at Naval Air Station North Island concentration differences between wells the diffusion sa
134. y Data Interpretation and Quality Control and Assurance 6 If a comparison is being made between concentrations obtained using PDB samplers and concentrations obtained using a conventional sampling approach then the well should be sampled by the conventional approach soon after preferably on the same day recovery of the PDB sampler The water samples obtained using PDB samplers should be sent in the same shipment as the samples collected by the conventional approach for the respective wells Utilizing the same laboratory may reduce analytical variability 7 Any unused water from the PDB sampler and water used to decontaminate cutting devices should be disposed in accordance with local state and Federal regulations DETERMINING APPLICABILITY OF PASSIVE DIFFUSION BAG SAMPLERS AND INTERPRETATION OF DATA When attempting to determine whether the use of PDB samplers is appropriate at a particular well a common approach is to do a side by side comparison with a conventional sampling method during the same sampling event This approach is strongly suggested in wells having temporal concentration variability In a well having relatively low temporal concentration vari ability comparison of the PDB sampler results to historical concentrations may provide enough infor mation to determine whether the PDB samplers are appropriate for the well In general if both PDB and conventional sampling produce concentrations that agree within a rang
135. y 500 mL of water also is available if larger sample volumes are required The sampler is placed in flex guard polyethylene mesh tubing for abrasion protection attached to a weighted rope and low ered to a predetermined depth within the screened interval of a well The rope is weighted to ensure that the sam pling devices are positioned at the correct depth and that they do not float upward through the water column FIGURE 2 USGS DIFFUSION SAMPLER USGS sampler Zip tie securing sampler to rope x ee 2 a tubing Flex Guard Monitoring well b 4 polyethylene L mesh tubing USGS sampler prior to sample collection Assembled USGS sampler being lowered into monitoring well Materials and Methods A 3 For this evaluation multiple 0505 samplers were placed end to end in three test monitoring wells to develop vertical contamination profiles Upon recovery the samplers were cut open and water samples were transferred into 40 mL volatile organics analysis VOA vials The samples were preserved and submitted for analysis DMLSTM Sampler The DMLS sampler shown in figure 3 uses dialysis cells as passive collection devices The dialysis cells are composed of a polypropylene cylinder that holds 38 mL of deionized distilled water The cells have 0 2 micrometer cellulose acetate filters attached to each end of the cell that serve as the permeable mem branes The cells are mounted in cylindrica
136. y has been observed in an aquifer in Cape Cod Massachusetts on the scale of centimeters Wolf and others 1991 Smith and others 1991 Hess and others 1992 Multiple PDB samplers have been used to show a change in TCE concentration of 1 130 ug L over a 6 ft vertical screened interval in Minnesota Vroblesky and Petkewich 2000 Tests using PDB samplers in screened intervals containing VOC stratification showed that the PDB sampler data appeared to be point specific whereas the pumped sample integrated water over a larger interval Vroblesky and Peters 2000 Determining Applicability of Passive Diffusion Bag Samplers and Interpretation of Data 11 The decision on whether to use samplers in such situations depends on the data quality objectives for the particular site If the goal is to determine and monitor higher concentrations or to examine contami nant stratification within the screened interval then the PDB samplers may meet this objective If the goal is to determine the average concentrations for the entire screened interval then a pumped sample or an average from multiple diffusion samplers may be appropriate As an aid in the decision making process the following section examines the influences that hydrau lic and chemical heterogeneity of an aquifer can have on sample quality in long screened wells Because VOC concentrations from PDB samplers commonly are compared to VOC concentrations from other sampling methodo
137. ydraulic and chemical heterogeneity This section examines potential sources of concentration differences between traditional methodologies and the PDB methodology The purge and sample approach to ground water monitoring differs from the diffusion sampler approach primarily because the area of the screened or open interval that contributes water to the purged sample typically is greater than for the PDB sampler and the potential for mixing of stratified layers is higher When pumping three or more casing volumes of water prior to collecting a sample chemical concen trations in the discharging water typically change as the well is pumped Keely and Boateng 1987 Cohen and Rabold 1988 Martin Hayden and others 1991 Robbins and Martin Hayden 1991 Reilly and Gibs 1993 Barcelona and others 1994 Martin Hayden 2000 due to mixing during pumping and other factors such as the removal of stagnant water in the casing and changing patterns of inflow and outflow under ambient and pumping conditions Church and Granato 1996 The induction of lateral chemical heterogeneity during pumping also may produce variations in the sampled concentrations The amount of mixing during purging can be highly variable Barber and Davis 1987 Church and Granato 1996 Reilly and LeBlanc 1998 Martin Hayden 2000 and may result in concentra tions that are not locally representative Reilly and Gibs 1993 Substantial vertical hydraulic gradients even

User's Guide for Polyethylene-Based Passive Diffusion Bag

Contents

Download Pdf Manuals

Related Search

Related Contents