SFUND RECORDS CTR

^'^77'%, 88015087

; J ^ ^ ^ UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

75 Hawthorne Street SFUND RECORDS CTR

San Francisco, Ca. 94105-3901 0222-00525

December 12, 1991

AR0787 James G. Derouin, Esq. Meyer, Hendricks, Victor, Osborn & Maledon, P.A. P.O. Box 33449 Phoenix, AZ 85004

Re: Comments on the Technical Screening Memorandum Hassayampa Landfill, Maricopa County, AZ

Dear Mr. Derouin:

Please find enclosed the U.S. Environmental Protection Agency's (EPA's) comments on the "Technical Screening Memorandum, Hassayampa Landfill, Maricopa County Arizona," which was sub­mitted by Conestoga-Rovers and Associates on October 28, 1991. Also enclosed are comments on the Technical Screening Memorandum (TSM), submitted by the Arizona Department of Environmental Quality (DEQ) and Arizona Department of Water Resources (DWR). Additionally, I have enclosed letters from DEQ and DWR which identify potential Applicable or Relevant and Appropriate Re­quirements (ARARs) for the Hassayampa Landfill site.

I look forward to discussing EPA, DEQ and DWR comments with you during our meeting scheduled for December 19, 1991. I expect that these comments will be incorporated into the draft and final Feasibility Study (FS) reports, as is required by the Administra­tive Consent Order (ACO). Please be aware that EPA's comments on the TSM do not address Appendix A.A - Results of Hydraulic Cap­ture Zone Simulations and Recommendations for Extraction-Injection Wells. EPA considers this to be a separate design document, and additional comments on this portion of the document will be provided in the future.

In addition to providing comments on the TSM, DEQ has provided comments which pertain to the overall site approach. During our upcoming technical meeting, we will provide you with guidance as to how these comments should be addressed. The FS report should incorporate proposed ARARs identified by DEQ and DWR. Although EPA has provided some comments on ARARs at this time, EPA's comments do not represent a final ARARs analysis. This analysis will be performed during EPA's review of the FS report.

Printed on Recycled Paper

The ACO specifies several sections which should be included in the draft and final FS reports. In addition, the FS report should closely follow the format identified in the EPA document entitled "Guidance for Conducting Remedial Investigations and Feasibility Studies Under CERCLA" (EPA/540/G-89/004). Among other things, this guidance suggests that the FS report contain separate sections on Remedial Action Objectives, General Response Actions, and Identification of Volumes of Media Requiring Remediation.

Finally, I have enclosed an agenda which lists several issues which I would like to discuss during our December 19 meet­ing. If you have additional isssues which you would like to dis­cuss, please add them to the agenda and send me a copy of the revised agenda via telefax. If you have any questions, please call me at (415) 744-2395.

Thomas J. Dunkelman Project Coordinator

Enclosures

cc: W. Victor, Montgomery (w. enlcosure) S. Quigley, (w. enclosure) J. Maye, DEQ (w. enclosure) G. Gibson, DWR (w. enclosure) H. Karr, RC-3 (w. enclosure) R. Oglivie, RC-3 (w. enclosure)

AGENDA FOR HASSAYAMPA LANDFILL SITE MEETING December 19, 1991

Location: Meyer, Hendricks, Victor, Osborn & Maledon, P.A. 2929 North Central Ave. Phoenix, AZ

Time: 10:00 AM

Issues:

1. Results from Vadose Zone Monitor Borings

Hassayampa Steering Committee (HSC) questions on DEQ com­ments

2. Significance of soil gas detections north of Pit 1

Implication that liquid waste may have migrated north along basalt unit

3. HSC questions pertaining to EPA, DEQ, and DWR comments on the Technical Screening Memorandum

4. Discharge options for treated groundwater

5. Implications of hydraulic capture zone simulations

6. Schedule for Feasibility Study Report

7. Potential Groundwater Treatability Study

8. Remedial Design/Remedial Action Negotiation Issues

9. Questions pertaining to EPA annual billing

* EPA representatives will be available on 12/20 if necessary

U.S. EPA Comments on the Technical Screening Memorandum (dated October 28, 1991)

Hassayampa Landfill, Maricopa County, Arizona

1. p. 4., para. 2. It should be noted that while some lateral contaminant migration from the disposal pits has been ob­served, vertical contaminant migration appears to be much more significant.

2. p. 4, para. 3. There should be a brief discussion of the threshold limit value (TLV), including the significance of the TLV, the intended use of the TLV, etc. It should be pointed out that the TLV is not intended to be used for risk assessment purposes.

3. p. 9, para. 3. Add a sentence to the beginning of this paragraph indicating that this section presents a brief sum­mary of the Risk Assessment completed by PRC for EPA; however, the reader should be referred to the Risk Assess­ment report for a more complete discussion.

4. p. 10, para. 1. Delete "significantly" from the first and fourth lines of this paragraph.

5. p. 10, para. 1. Delete the last sentence of this paragraph: "This modification has only recently been proposed..."

6. p. 10, para. 3. The summary of the future residential use and future on-site commercial/industrial scenarios is incor­rect. Although specific values for carcinogenic and non­carcinogenic risk are calculated for both scenarios (several of which exceed acceptable levels); the Risk Assessment con­cludes that unacceptable levels of risk (carcinogenic and non-carcinogenic) would result under both scenarios, since the risk due to exposure to waste present below the soil cover was assumed to be significant, but was not quantified.

7. Table 1.1. For the reasons described in the preceding com­ment, this table is incorrect. This table should be re­placed directly by Table 5-1 of the Risk Assessment.

8. p. 11, para. 2. Delete the sentence, "These results do not clearly provide a motivation ..." This determination will be made in the Record of Decision (ROD), and the discussion is inappropriate here.

9. p. 11, para. 3. The content of the draft and final FS report should also be consisted with "The Guidance for Con­ducting Remedial Investigations and Feasibility Studies Un­der CERCLA" (EPA/540/G-89/004).

10. p. 12, para. 1. It is logical to treat soil and groundwater remediation technologies separately. However, in the FS report, groundwater information should be presented in the body of the report rather than as an appendix.

11. p. 13. This discussion of Remedial Objectives should be moved to and combined with Section 3.2 and 3.3. This Remedial Actions Objectives Section should be separate from and should follow the ARARs section. Much of the informa­tion presented on pp. 13-16 is simply a rehash of guidance and is not necessary.

12. p. 17- 21. The waste profile and soil profile sections are misleading. The use of "representative concentrations" does not present a complete representation of the nature of the waste and soil contamination. Tables 2.3, 2.4, 2.5, 2.6, 2.8, and 2.9 should also include maximum concentration values. This section should be rewritten to include a dis­cussion of the maximum waste and soil concentrations. The difference between representative and maximum concentration should be clarified.

13. p. 20, para. 3. Actually four distinct areas of soil gas contamination were identified. See comment pertaining to p. 21, para. 2.

14. p. 20, para. 4. The attempted correlation between soil gas and soil concentrations must either be supported further or the reference to this correlation should be removed.

15. p. 21, para. 2. The elevated detection of volatile organic compounds (VOCs) in soil gas north of the hazardous waste area is significant and cannot be written off as an anomalous event. This issue must be discussed further.

16. p. 21, para. 4. This summary should include a discussion of the fact that the highest concentration of contaminants in soil was most often found at the deepest depth.

17. p. 22, para. 3. No wells are found on Figure 1.2 as is stated here. Wells are shown of Figure A.1.1 (Appendix A).

18. p. 24, Section 3.0. The ARARs section should be separate from the Specific Remedial Objectives section. A detailed legal analysis of the ARARs section has not been completed by EPA at this time. This will be done during EPA review of the FS report. However, copies of letters in which DEQ and DWR identify their respective potential ARARs are included herein. The ARARs analysis from the North Indian Bend Wash (NIBW) Superfund Site ROD is also enclosed. It is expected that the ARARs selected for the Hassayampa site will be fairly similar to those selected for the NIBW site.

Furthermore, several preliminary comments on the ARARs sec­tion are provided below.

- The Federal ARARs identified in Table 3.1 appear to be fairly complete. Other Federal ARARs which may need to be included are identified below:

Water Quality Criteria NPDES National Pretreatment Standards National Primary and Secondary

Air Quality Standards Air Emission Standards for

Process Vents Air Emissions Standards

for Equipment Leaks Executive Order

Floodplain Management

40 CFR Part 131 40 CFR Part 122, 125 40 CFR Part 403 40 CFR Part 50

40 CFR Part 265 Subpart AA Subpart BB

Exec. Order 11988

- Table 3.1 identifies whether a requirement is applicable, but does not indicate if the requirement is relevant and ap­propriate. This should be corrected.

- A table indicating Federal and State criteria, advisories, and guidances which are "To Be Considered" (TBC) should also be included. Potential TBCs may include the following:

National Secondary Drinking Water Standards

National MCLGs

Water Quality Standards OSWER Directive 9355.0-28

Emissions from Air Strippers Health Advisories

ADHS Health Based Guidance Levels

40 CFR Part 143

Pub. L. 99-339, 100 Stat. 642 (1986)

40 CFR 131 EPA Directive

EPA and National Academy of Sciences

- The FS report should also include a table which indicates which ARARs are pertinent to which alternatives. This will make the ARARs analysis of each alternative easier in the FS and ROD.

19. p. 2 6, para. 2. Delete the sentence, "Table 3.1 provides a list of potential ARARs ..." as it is redundant with next sentence.

20. Table 3.2, A.2.2. This table needs to be expanded to in­clude all chemical specific ARARs for all contaminants detected in groundwater. As an example, see Table A-1 (enclosed) from the North Indian Bend Wash ROD. The maximum contaminant concentrations should also be included.

For your convenience a copy of the following document has been sent to Montgomery and Associates: Region 9, EPA Drinking Water Standards and Health Advisory Table.

21. p. 27. Sections 3.2 and 3.3 should be combined with Section 1.5 and presented separately from the ARARs section. Fur­thermore, a distinction needs to be made between remedial objectives and cleanup standards. Remedial objectives are identified in the FS report in order to estimate the volumes of media requiring remediation. Specific cleanup standards for the site will be selected in the ROD. The remedial ob­jectives presented in this document and the FS are not necessarily the same as the cleanup standards that will be selected in the ROD.

- The FS report must follow the format of the RI/FS Guidance which includes separate sections on Remedial Action Objec­tives, General Response Actions, and Identification of Volumes of Media Requiring Remediation. Volumes of waste, soil, soil gas, and groundwater requiring remediation should all be identified in the FS report.

- For groundwater, cleanup standards (and remedial objec­tives) must be developed for all contaminants detected in groundwater. While MCLs (when they exist) are often selected as cleanup standards, stricter standards may be selected when multiple contaminants are present. For chemi­cals for which no MCLs exist, alternate cleanup standards must be selected.

22. p. 28, para. 4. It should be clarified that TCLP levels were not designed to be used as soil cleanup standards.

23. p. 27-32, Tables 3.4, 3.5, 3.6, 3.7, and 3.8. The discus­sion, presentation, and conclusions of Section 3.3 -Remedial Objectives for waste and soil are presented in a biased manner. This is a result of exclusive use of repre­sentative concentrations. The analysis in this section ^ should be based on both representative and maximum con­centrations. All tables within this section should present both representative and maximum concentrations.

While representative concentrations may be useful in evaluating the effectiveness of a particular treatment tech­nology, maximum concentrations are more appropriate for determining which areas of waste and soil require remedia­tion. By not presenting an analysis that involves the use of maximum concentrations, EPA cannot evaluate which pits may or may not require remediation. Consequently, EPA can­not evaluate whether remedial alternatives have been properly developed.

EPA is not necessarily arguing that the waste in any or all pits must be excavated. A decision to leave waste in place would require careful justification in the ROD. The argu­ments presented in this section are not acceptable, and do not provide EPA with the proper information to evaluate rem-diation of waste within the pits.

24. p. 30, para. 4. This discussion of performance based objec­tives must be expanded. Exactly what are these objectives? More detailed remedial objectives must be developed for soil and soil gas. The FS must be able to show that performance based objectives for soil and soil gas will be protective of groundwater. Analytical models such as VLEACH (reference available upon request) have been used at other Superfund sites in Arizona to evalatue the continued threat to groundwater posed by soil gas contamination. The use of such models should be discussed in the FS report.

25. p. 31, para. 1. Previously in the TSM, it was stated that significant concentrations of VOCs were detected in three areas of the site (p. 21, para. 1). Why has the southwest portion of the special pits area been omitted here? Also, why has soil gas contamination north of the site been omitted here? Unless it can be shown (through the use of an analytical model or other method) that a certain area and concentration of soil gas contamination does not pose a threat to groundwater, then all documented areas of soil gas contamination must be remediated.

26. p. 31, para. 2. as follows:

The remedial criteria need to be redefined

Waste (in all pits)

Soil Contamination

(beneath all pits)

Soil Contamination

to be remediated to levels below HBGLS or to be covered as to miti­gate airborne exposure and prevent waste contact

to be remediated to levels which do

not result in significant future impact to groundwater

to be remediated to levels which do

(central and SW portion not result in significant future of Special Pits) impact to groundwater and to be

covered as to mitigate airborne ex­posure and prevent waste contact

Soil Gas Contamination (all site areas)

to be remediated to levels which do not result in a significant future impact to groundwater

27. p. 31, last para. The argument that single detection of a contaminant above a regulatory tolerance does not justify remediation of waste within a certain pits is very weak. Firstly, very few waste samples were taken per pit, and as a result there are few samples which exceed regulatory tolerances. If more samples had been taken per pit, it is probable that there would have been more hits in excess of these level. Secondly, representative rather than maximum concentrations were used in this analysis. As described above, this type of analysis is not sufficient.

28. p. 31. The list of technologies evaluated in this section appears to be fairly complete with the exception that steam injection and steam sparging have been excluded. These technologies may be appropriate for treating contaminated soils at depth. These technologies should be evaluated in the FS report.

29. p. 33, Table 4.1. This table needs to be clarified using bold facing or indentation. For example:

A.7 Soil Treatment Incineration Fixation/Solidification Soil Washing

30. p. 35, para. 2. This section should include a discussion regarding the long-term effectiveness of deed restrictions and the difficulty in maintaining such restrictions over a long period of time.

31. p.36, para. l. This section should include further discus­sion regarding the fact that the effectiveness of a fence is difficult to maintain over a long period of time.

32. p. 40, para. 1. Although it is appropriate to screen dif­ferent capping technologies, it is not appropriate for an FS document to "select" a particular technology or alternative. The ROD "selects" appropriate technologies and alternatives. It may be necessary to carry both RCRA and Arizona capping requirements through the FS. Furthermore, given the the number of pits located within the hazardous waste area and the uncertainty over the location and contents of the spe­cial pits, EPA feels that capping options which do not cover the entire hazardous waste area can be screened out at this stage.

33. p. 41, para. 4. This paragraph implies that removal of the waste in Pit 1 is not feasible since the volume of waste is greater than 15 feet and depth and 1,000 CY. in volume. Removal of wastes in Pit 1 is certainly technically feasible. This paragraph should be rewritten or deleted.

34. p. 42, para. 2. There does not appear to be any discussion of a requirement for incineration prior to disposal in Sec­tion 3, as is referenced here. This section should be clarified, to indicate that incineration would likely be re­quired prior to disposal at an offsite landfill.

35. p. 42, para. 3. The last sentence of this paragraph is con­fusing. Is the report saying that contaminated soil cannot be excavated due to the depth of contamination? This paragraph should be clarified.

36. p. 43, para. 5. Again, it is not made clear in Section 3.0 as reference here, that land disposal is not applicable to wastes from Pit 1. EPA is not suggesting land disposal as a potential technology, however this issue needs further clarification in this report.

37. p. 54. The description of soil vapor extraction (SVE) should include a discussion of the radius of influence of the vents and the number of vents which are likely to be re­quired. The discussion of SVE should be expanded so that EPA can evaluate whether this technology is capable of remediating all areas where soil gas contamination has been identified as a concern.

38. Table 4.6. As mentioned in previous comments, the areas of waste potentially requiring excavation and treatment cannot be properly evaluated given the type of analyses (exclusive use of representative concentrations) presented in this document.

39. Table 5.1. EPA generally concurs with the list of remedial technologies for waste and soil which survived the screening process, with the possible exception of the exclusion of steam injection/sparging. However, EPA does not agree with the manner in which technologies were assembled into alter­natives.

Capping should be a component of all alternatives except for No Action. Furthermore, two capping technologies should be should be presented for each alternative (a.- Arizona Cap, b. - RCRA Cap)

EPA suggests the waste and soil alternatives be assembled as follows: 1. 2. 3. 4. 5. 6.

No Action Deed Res. Deed Res. Deed Res. Deed Res. Deed Res.

Cap Cap Cap Cap

Vapor Vapor Vapor

Extrac. Extrac. Extrac.

Excavation Excavation

Soil Wash Incin.

40. p. 63, para. 2, Table 5.2. This section is awkward in that it appears to recommend a groundwater cleanup alternative. The FS should not propose or select a groundwater alterna­tive. Screening of groundwater technologies does occur in the FS, and the surviving technologies are assembled into alternatives. See subsequent comments below which pertain to the screening of technologies. Furthermore, Table 5.2 should be reformatted to be consistent with comments provided below.

41. p. A-1. The organization of the TSM is somewhat awkward in that several sections presented in the body of the document are repeated in the appendix. Although EPA agrees with the separation of soil and groundwater analyses, the FS report must combine the two analyses in a neater manner - ground water information should not be presented as an appendix. In the FS report, the initial analyses should be combined for all media (ie. background information, summary of risk, contaminant profiles, remedial action objectives, general response actions, and identification of volumes of media re­quiring remediation). Later in the FS analyses, separate presentations should be made for soil and groundwater (ie. identification and screening of technologies, description of alternatives, detailed analysis of alternatives, comparative analysis of alternatives).

41. p. A-2, last para. Why have SVOCs been omitted from analysis throughout the Appendix (reference to SVOCs occurs in tables but not in text)? This should be corrected in the FS report.

42. p. A-4, Table A.2.2. See previous comments pertaining to groundwater remedial objectives. Remedial action objectives and cleanup standards must be developed for all contaminants detected in groundwater

Corrections to Table A.2.2

dichlorodiflouromethane AZ MCL 1.0 chlorobenzene - EPA MCL 100 (effective 7/92) methylene chloride - EPA MCL 5 (proposed) toluene - EPA MCL 1,000 (effective 7/92) tribromomethane - EPA MCL sum of trihalomethanes = 100 xylene - EPA MCL 10 (effective 7/92)

43. p. A-4, para. 2. The point of compliance with groundwater cleanup standards will be selected in the ROD, and as a result the discussion is not appropriate here or in the FS report. EPA disagrees with the interpretation of the point of compliance presented here.

44. p. A-4, para. 4. Add a citation for EPA's discharge limit of 15 pounds of VOCs per day for an air stripper (OSWER Directive 9355.0-28).

8

45. p. A-5, para. 2. Identify the chemicals of concern and specify how they were selected. These should be the same chemicals of concern identified in the risk assessment.

46. Table A.2.3. This table should contain information on all contaminants detected. Those contaminants which have been identified as chemicals of concern should be highlighted. Maximum concentrations should also be presented in this table.

47. p. A-16, para. 4 The logic of the screening process for groundwater treatment technologies is not clear. This sec­tion needs to be clarified further. UV oxidation attains permanent destruction of contaminants, while air stripping and carbon adsorption do not. This distinction is important when it comes to performing the 9-criteria analysis in the FS. UV oxidation should be retained as a treatment technol­ogy. Carbon adsorption may be eliminated from further con­sideration if it is being argued that air stripping attains a greater level of treatment at lower cost. However, this argument is not clear in the TSM and needs to be clarified.

The FS report should not "select" treatment technologies; however, certain treatment technologies may be eliminated during the screening process.

48. p. A-17, para. 4. Given the strong opposition by DWR to discharge of treated groundwater to the Palo Verde Nuclear Generating Station (PVNGS), this disposal technology should be eliminated from further consideration unless the Has­sayampa Steering Committee can persuade EPA, DEQ, and DWR otherwise during our technical meeting scheduled for Decem­ber 19.

49. p. A-19, para. 3. Groundwater monitoring should also be conducted for Unit B wells.

50. p. A-19, para. 1. Based on the results of the technology screening process, technologies should be assemble into al­ternatives. This has not been done properly here. EPA sug­gest that the groundwater alternatives be assembled as fol­lows:

No Action (Monitoring) Monitoring Extract. Treatment (Air Strip.) Reinject Monitoring Extract. Treatment (UV Oxid.) Reinject

If the Hassayampa Steering Committee argues that discharge to the PVNGS should be retained as a disposal option, then the following two alternatives could be added:

Monitoring Extract. Treatment (Air Strip.) Discharge Monitoring Extract. Treatment (UV Oxid.) Discharge

51. p. A-19, para. 2. A discussion of the preliminary design of an extraction system is appropriate in the FS report. However, EPA does not necessarily agree with the recommended extraction and injection systems. EPA will provide comments on Appendix A.A in the future.

10

Appendix A ARARs AND OTHER CRITERIA FOR NIBW

This appendix identifies ARARs and other criteria to be considered (TBCs) for the selected remedial actions for NIBW.

CHEMICAL-SPECIFIC ARARs AND TBCs

Table A-1 presents chemical-specific ARARs and other criteria for water arranged by chemical compound. The major regulations which contribute to the list of potential chemical-specific ARARs are the Clean Water Act (CWA), the Safe Drinking Water Act (SDWA), and Arizona Water Quality Standards for Navigable Waters. The chem­ical-specific TBCs for the NIBW site include (1) Arizona Department of Environmental Quality (ADEQ) Human Health-Based Guidance Levels for Contaminants in Drinking Water and Soil (HBGLs), (2) Federal Health Advisories, and (3) proposed /JDEQ Water Quality Standards.

The SDWA Maximum Contaminant Level (MCL) standards are based on human con­sumption of water for drinking, cooking, bathing, etc. Economic considerations and technical feasibility of treatment processes are included in the justification for these levels. MCLs are applicable to the quality of drinking water at the tap pursuant to the Safe Drinking Water Act and are ARAR for treated ground water when the end use is drinking water.

Pursuant to 40 C.F.R. Section 300.430(e)(2)(i)(B), MCLs and non-zero Maximum Contaminant Level Goals are relevant and appropriate as in-situ aquifer standards for ground water that is or may be used for drinking water.

ADEQ Aquifer Water Quality Standards [A.R.S. Section 49-223 and implementing regulations] generally are identical to SDWA MCLs at this time, and therefore are not referenced in Table A-1. One notable exception is the 50 ng/1 chromium Aquifer Water Quality Standard, which is more stringent than the current MCL and therefore is an ARAR and the selected water treatment standard for chromium for NIBW.

The CWA Water Quality Criteria are designed to protect aquatic life (both marine and freshwater). These standards are expressed on the bases of acute and chronic toxicity levels. Both the Federal Water Quality Criteria and the State Water Quality Standards for Navigable Waters [A.R.S. Section 49-221 and implementing regulations] are ARAR for surface-water discharges.

RDD/R310/020.51 A - 1

•?-»*S"f^it-'-;'<'*!'' •-"^»V.v;«Hj;,'-.'<-^--- -r

> I

T . b k A-I

C i M m k a l - S p c c l l k ARARs . n d Other C:ritrrla for N I B W

(coDCcnlraUons l a i igf l )

ShfTl 1 or 3

1 CompoDDd

I . l . l - ' fhch loroe l tune

l . l -U i chk i ruc iha iK

1,1-Dichlorocihcne

1.1,2' l 'r i<rhloni-2.2,ITrinucini<:l l ianc

l ,2- l ) ich loroclh«ne

1,3-nichlonjbenzenc

1.2- Oichl i twpropane

Methy l Hlhyl Ketone

•1 4- D [>T

Acetone

l lenzcnc

His(2-ethylheiyl)pl i thalatc

RroinodiLhioromclhane

H fomofo rm

Carbon Tetrachloride

C:hlorubenzcne

Chl(>rc»(i)nn

Oihri>m<M.hfunimcthane

D i n hulylphthalalc

D i n t ictyl phthalate

Methylene chloride

A j>p l l c .bk or R e k v . n l • • > • App rop r i ab

SDWA M C L

200

7

5

5

5

IOI)

10(1

s

lOII

100

SDWA

M C U ;

200

7

IOO

AWQC

T o i k l t y

19.000

21.000

44.000

10"* C u K e r Risk

0 0 3 1

0.94

X I 0 1 2

0 6 7

0.42

6

Other C t l k r i a lo be C o t u h k r r d

SDWA

Propotieil

M C L

200

5

S

4

4

S

SI>WA

M C L t ;

fr

0

bO

tl

U.S. KPA Heal th AdvLsorle:.

1-d.y

l O k f

140000

1,000

740

8,9.10

75.000

211

4.IHKI

i.mio

IO.day

10 kc

.15.000

1,000

740

8,910

90

7.500

211

IftO

l.KOO

l.4inf!er T e r m

I O k ) i

.15,1)00

I.IKKI

740

8.910

2.500

N/A

71

9,000

70 Iw

125.1)00

.1.5(10

2.r*o

31.250

8.(<(NI

N/A

25(1

1.(100

L l le l lme

m k g

I.IXXI

350

N/A

.1.125

KA(I

N/A

N/A

1,150

M I K Q l l W i U ror

W . I r r

200

7

0 38

620

05 ( .

170

0 10

700

I I

\ 0.19

0 19

0.27

IOO

6

0 19

4 7

Selecled NIBW C k a n u p S u n d a r d r.r Treated

Waler alHl In Sl lu

C ruund Water

200

7

5

620

' 170

5

4

10(1

10(1

5

10(1

6

4

'

R1>U/K310A>21 511

i4A

lis

if-:-

K.

mi

> I

T a b k A- I

<:heink«l -Speel fk ARAR.< and O lhe r C r l k r l a ror N I B W

(cooccntral lo i ls In | i ] t / l l

Sheel 2 o l .1

_ Siyrcne

rc i rach lnroc ihcnc

Toluene

I r a n i 1,2-d ich lomelhcnc

T n c h loroethene

' [ ' r ichlornnuunimethane

Viny l Chki r ide

A luminum

An i imony

Arsenic

Bar ium

l lc ry l l iun i

Boron

Cadmium

(. 'hromium

Copper

l e a d

Mercury

Nickel

Selenium

Silver

AppUcabk or K e k v a n I and A p p n j p r l a U

SDWA M C I .

100

5

1,000

100

5

2

20

50

2.000

5

IOO

1,000

50

2

50

50

SDWA M C L U

100

1,000

100

0

3

2.000

5

IOO

2

50

AWyc

T o a k l l y

15,000

146

10

50

50

10

154

10

50

IO- *« : .nce r

Risk

0 8 8

2 8

2

0 0025

0 0039

O lher Cr i le r ia lo be Considered

SDWA

Proposed

M t T .

5

5/10

1

1,10(1

100

SDWA

Proposed

M I ' U :

140

0

2.000

70

50

0

I.UIO

20

3

KKI

45

U.S KPA Heal lh Advisories

| .day 10 k«

27.000

N/A

18.000

2.720

50

41

1.40(1

10-day

l « k «

20.0(«l

14.000

6.000

1,000

15

50

X

1,400

1 .(KKI

I x i n ^ r l e r m

l O k j l

20.0(KI

1.940

N/A

1.0(K)

50

,

5

240

20 pg/d;iy

70 kn

70.IHKI

6.K(K1

N/A

1.50(1

50

,8

840

20 t^^di iy

L i r n l m e

70 k f

7,1X10

N/A

lO.KMI

150

511

1.800

IK

1711

20 ^K/d.iy

5.5

A D K V I IB ( ; l . s ror

Waler

5

0 6 7

2.000

IOO

3.2

2.100

01)2

71

1

50

5.0IN)

0.IMI7

5

KKI

l..«K)

20

2

10(1

45

.50

S e l r r k d M K t V ( k a n u p Slandard fi>r Irealed

W a k r and In Sl lu 1 t^ruund Waler 1

KK)

"• I.IKKI

KKI

5

2.KKI

2

20 II

5

50 11

2.(KKI II

1

5

5(1

I.IKKI

5()

2

KK)

Sd

50

Kni)/l<310«21 512

>

T a b k A- I C b e m k a l - S p e e l l k ARARs U H I O lher C r l k r l a ror N IBW

(concentra l ions In p|^l) Sheel 3 uf 1

1 Stront ium

1 Vanadiura

/Jnc

A p p l k a b k or Rekvan I and A p p r o p r l a k

SDWA M C L

5,000

SDWA

M<:ixi

Awy<:

T o i k l l y

• ^

5,000

10"* Cancer

Risk

Other t. 'r l leria Iv be Considered

SDWA

Proposed

M<T.

SDWA

Propoaed

M<T;( :

II.S. KPA l l r a l l h Advisories

I J . ,

10 k (

lO-day

10 k«

Nmf ler Term

III i ... 70 kg

L i fe t ime

70 kg

WVH'.li. for Waler

7

5.000

SelecUd M B W C k a n u p

Standard lor I r ra ted

Waler and In Situ ( i f ound W a t , ,

5.(KK) 1

N o l e i ; A D t O = A r i i ona Departmenl of ILnvimnmcnlal Quality.

AWCH-' = Ambien i Water Quality Cr i tena; adjusted lo r consumption o l dnnking water only; fish ingestion component removed (U.S. l iPA , 1986).

A W Q C (10"* ) = The Ambient Water Qual i ty Cr i tena tesult ing in a IO"* acess l i fet ime cancer risk.

M C L = Maximum Contaminant 1 evel.

M C L G = Maximum Conlaminanl l e v e l Goal . |

S D W A = Safe Dr ink ing Water Ac l , 40 C I K 141, November 15, 1985.

U.S. B P A Heal th Advisories;

l-day/IO kg = Conccnirat ion of compound in drinking water that could pose a risk if consumed hy a 10 kg child for 1 day.

lO-day/IO kg = Concentrat ion o l compound in dr inking water that could putc a risk if umsumed by a lO-kg chi ld for 10 days.

lx>ngcr Tcrm/10 kg = ConcenlralitMl o l compound in dr inking waler ihal could fiose a risk if consumed hy a 10-kg child lor more than 10 days.

1 I JMiger ' rc rm/70 kg = Concentrai ion o f corapound in dr ink ing water Ihat could pose a nsk if consumed by a 70 kg adull lor more than 10 days.

1 I.(letime/70 kg = Concentration of contp^Hind in dr inking water Ihal cxHild pose a nsk if consumed by a 70-kg adul l lor a l i let ime. ||

KDI)/K3I0A)2I.51 3

s_ji -• •

Federal Health Advisories are criteria developed by either EPA's Office of Drinking Water Health Advisory Program or the National Academy of Sciences (NAS). The Federal Health Advisories are based on NAS-Suggested Non-Adverse Response Levels (SNARLS) at which no known or anticipated adverse human health effects would occur, given an adequate margin of safety. ADEQ HBGLs have been selected as water treatment standards for 1,3-dichlorobenzene, methyl ethyl ketone, and trichloro­fluoromethane. ADEQ HBGLs are also to be considered for direct exposure threats

I from potential soil ingestion.

LOCATION-SPECIFIC ARARs AND TBCs

Table A-2 identifies the location-specific AR2\Rs and other criteria for NIBW. Location-specific ARARs differ from chemical-specific or action-specific ARARs in that they are not as closely related to the characteristics of the wastes at the site, or to the specific remedial alternative under consideration. Location-specific ARARs are con­cerned with the area in which the site is located. Actions may be required to preserve or protect aspects of the environment or cultural resources of the area that may be threatened by the existence of the site, or by the remedial actions to be undertaken at the site.

ACTION-SPECIFIC AlURs AND TBCs

Table A-3 identifies action-specific ARARs and other for NIBW. The actions included in Table A-3 are components of remedial actions selected in this ROD and the remedial action selected in the 1988 ROD (the Scottsdale Operable Unit remedy).

Further identification and discussion of OSHA requirements, air emissions requirements, and additional State ARARs and other criteria are provided following Table A-3.

RDD/R310/020-51 A - 5

> 1

T a b k A-2

• .ocal lon-Speeirk ARARs and O l h e r C r l k r l a ror N I B W

Sheel 1 or 2

LocaUoa

II 1. W i th in lOO-ycar f loodplain

1 2. W i i h i n f loodpla in

1 3. W i i h i n area where acl ion may

1 cause i r r r pa raHc harm, kj«&, o r

1 destmct ion u l ftignincam ar t i fac t i

4. Cr i t ica l habi ia l upon which

endangered species or ihrca lencd

species depends

k tnu lnmtn l

r 'acij i ly musi be designed, m n -

s ln ic led, operaled, and mainlained

to avoid washout.

Ac i i f in IO avoid adverse cfrccis,

minimize poleni ial harm, reaiore

and preserve natural and beneficial

values.

Ac t ion to recover and preserve

artifacts.

Ac t ion to conserve endangered

species o i threatened species,

including consultat ion w i th the

Department of the Inter ior.

Prcm4uLslu<N)

K C R A hH/ardous wrtste;

trealmt-'nt. storage, o r

disposal.

Acii<in thai wi l l occur in a

f loftdplain, i.e.. lowlands, and

rclaiivety flat areas adjoinmg

in land and »tastal wntcrs and

o ther f loiHi-prone areas.

A l te ra t ion of terrain that

threatens stgnil icani scteniif ic,

prch is tonc, histor ic, or

archaeological data.

De te rmmai ion of endangered

species or threatened species.

<:tUlUtn

40 ( ; n < 2 M IH(h)

( m H 8 2 6 4 >

I j tccu l ive Order 11988,

Mroicci ion of llo<»d-

plains (40 C:i R 6.

Appendu A)

Nat ional Archaeological

and Histor ical

Preservation Act (16

u s e Section 4ftV); 36

( I K Part 65

i:ndangcred Species Act

o f 1973(16 DSC I V M

et set)) , 50 C i R Part

200, SO ( I K Part 402

AKAK

A R A R

A K A K

A R A R

A K A K

<!otnmrnLs 1'

Portions o f the N I H W site art- Unn\cd wi ih in JI KM) year ll

(l(Hidplain. A R ( ' R A faci l i ty li>calL-d in a lOO-ycar lltMidplain must 1

IH : designed, cons l ru t l cd , opcnt tcd , and mainl^ i i ie i l to prevent 1

w»sh^(ui of any hawrdous waste dy » tOO-yeitt ( I^HKI

/•'edfral agencies nre d i rcx fcd (o ensure rhat p(;inning [irogranis

and budget requests reflect considerat ion of flenid plain

management, inc luding the resioral ion and preservation ol such 11

(iind as natural undevelotHrd l l tKKlplains. l l newly constiucted II

facilities are to be located in a MiMKlplain, accepted no<Hfproolrng It

and o lher fltHid a>ntro | measures !>hall l e undertaken lo achieve 11

fltHid protect ion. Whenever pract ical, s lruclures shall bc elevHied II

above the base flotKl level rather ihan filling; Jand. As pun of any l j

1 ederal plan or act ion, the ptt lei i l ia) tor restoring; and preserving ll

fliMHlplatns Mt thctr natural t vnc l i c ia t valuer can be realized must

he considered.

(.'rossrng of the l l t W w i ih piping or Icxal ion o f wells in the II

lOO-year lltHHJplain wi l l be designed to lesult in no inipaci i') f lood

surtace prol i les. AJiy |Xi lential pipe or wel l breakage due to Hood

ing wi l l bkely nol tnlrviducc new a in tamina l i on tiecause o( the

regional nature of ihe l l A U contaminat ion.

The N I I I W is essentiaUy completely develo[»cd

Art i facts have been found in areas near N I H W .

No endangered sptxies are kn<pwn to exist on the N I I I W site.

R1>1)/R31D/022 51 1

r^ t ^ :M~~H~8i JB S JS JB iHT --jg^'-gi—j^—g|-

>

l i i b k A-2

1 g a l l o n - S p r e l f i c AKAHiv and (Mh«r Cr i t e r i a fur N I B W

II Shm 2 III 2

II I^Acailon

5. Wet land

6. A rea affecting • i rcam or nver

7. 1 lazardous waste site

R r q u l r r m r n l

Ac l ion to minimize Ihc destruc­

t ion, loss, or degradation of

wetlands. Ac t ion to prohibi t

discharge o f dredged or f i l l

material i n l o wetland wi lhout

pcnn i i .

Ac l ion to pro lec l fish or wi ld l i fe .

Acl ions (o l imi t worker exposure

to hazardous wasles or hazardous

substances, including training and

moni tonng.

P n r t q u t s l U { s }

Wet land as defined by

l-Jiecutive Order I I W O

Section 7.

Diversion, channeling, or

o lher activi iy that modif ies a

stream or nver and affects

fish o r wi ldl i fe.

Construct ion, operations and

maintenance or o lher

activi l ies wi th potent ial

worker exposure.

C l U l l u n

I j e c u t i v e Order IIV9U,

t ' ro tec i ion of Wetlands

(40 C I R 6, Appendix

A ) , Clean Water Act

Section 404; 40 C l R

Parts Z30, 2:^1

1 ish and Wi ldh le

(.'(Ktrdination Act (16

n S C ( » 6 l et seq.), 40

C I R 6 .U)2

28 CHR 1910.120

AHAI t

A R A R

A R A R

A R A R

CommrnLs 1

II wetlands are located wi ih in the area of pnt)>osed l edc ra l |

activi l ies, the agency must conducl a Wetlands Assessment. 11

there is no practical al lernal ive to l i tcaling in or ; i l tecl ing Ihc

wet land, thc Agcn t ^ shall acl to minimi/.e potcni ial harm to ihe

wetland Ihe ( l e a n Waler Ac l prohibit.s discharge of dredged or

( i l l matenal i n l o wetlands w i lhou l a |x:rmit.

Assessments wil) be perft trmed al potential arcas o t activi iy (e g ,

moni tor ing well instal lat ion) to identify wetlands and p«>lcnti;il

means o f m inmu i i ng impacls.

Ibe l i s h and Wi ld l i le (. 'oordination Act requires coi isul ini ion with

Ihe Depur tmcn i o t 1 ish and Wi ld l i fe pr ior lo any HCtion thai

would al ler a body of wnler o l the I h i i led Stales. This icqui rcmei i l

could f>e Hpplicahle to any action that wou ld resuli in mod i l i i a l ion

ol Ihe Aqua Tna or ( i i l a Rivers.

N I I I W aciums wi l l l ikely improve the qual i ty of IDW ponds |

Spillage lo the ( i i l a River is mfrequenl and would likely nol r i l led 1

the t;iila River, l i s h in N l l t W ptmds are mu (here l y natural

causes. Ih t7 arc s l tKkcd.

RDI5/R3I0A)22 5I 2

> I

00

-^

AcUon

A i r Str ipping

II Conta iner Slorvgc

(Ons i le )

Kequ l r f ments

R(. 'RA siandards fttr contro l o l emissnms of volati le

organics.

Con l ro l o f air emissicms o f volati le organics and

gaseous contaminants.

Con ta inen of hazardous waste must bc:

• Main la ined in good condi t ion

• C^>mpatible w i th hazardous waste to bc s iored

• Closed dur ing slorage (excepi to add or remove

wasle)

Inspect container slorage arcas weekly for

deter iorat ion.

Place ctmiainers on a sloped, crack free base, and

protect Ui.ym contact w i th accumulated l iqu id.

Provide containment system wi ih a capacity of

10 percent of Ihe volume of containers of free

liquids.

Kcmove spi l led o r leaked waste in a t imely manner

l o prevent overf low of ihe containment system.

Keep containers o f ignitable or reactive wasle al least

50 feel f rom Ihe facility's properiy line.

Keep incompai ib le malerials separate. Separate

incompatible malerials siored near each other by a

dike v»r other barrier.

At closure, remove al l hazardous waste and residues

fntn i thc conta inmcnl system, and decontaminate or

remove al l containers, liners

Table A-3

Act lon-Spei: inc AKAKs and O l h r r C r i U r l a for N I H W

Prerr«jul<(ll«!t

R C R A hazardous waste.

Fmission of V O C s o r gaseous air contami­

nants.

R < . : R A ha/^rdous waste ( l is icd or charac-

tensi ic) held for a temporary period bef(»re

treatment, disposal, or slorage elsewhere,

(40 CKR 264 10) in a container ( i e , any

por iable device in wh ich a material is

stored, t ransported, dispo»>ed (tf. o r

handled).

C lU i l l un

4 0 C I R Subparts A A & 111)

Mancopa ( .ounly Rules 210,

*2o, ^^o

40 C I K 264-171 ( K I 8 18-

264 170, et seq )

4 0 C I R 264 172

4 0 C I R 264 17 t

40 ( I K 264 174

40 C I K 264 175

4 0 C I K 264.176

4 0 C I R 264 177

4 0 C I R 264 178

AKAK

A R A K

T H C

A R A R

A R A R

A K A K

A K A U

A R A K

A K A R

A R A R

A R A K

Sheel 1 o l 2 |

f 'onimenls

ITic ()rnp<ist;d standard ra^uirc^ r c j u d i o n ttl V O t

cnn.ss s I rom "priHjutt atLumulHroi vt-s.st'li.'" and

leak di ' Icct ion and ic|>air ptngidnis. I*n.dut[

accuinulalor vL-sst'ls nictudtr dir strippers.

Ihcsc requirements nre applital i le ur relevant and 1

appropr iate lor dny coniamm.ned soil or ground water 1

or i rea lmenl syslem waste ll i. i l might In- containerized |

find stored cmsite pr ior lo trc' i tnicni or f inal dtsfX)&al. 1

( i r o u n d waler or soil roni^nning j l isted w;i<ile must 1

hc managed as i l i l were a hazardous waste sti long as

il contains Ihe l isicd waste. 1

1 Rni)/K31IW2.1.'il-l

<f'^g&^<'-c,;^mk';i,-^.-J^|^-,n:;;,,^^^-..,„^„^^^,^,l9^.,,^^HS

> I

[, 1 ' • 1 I I . Li_i; • : = —

T a b k A-J

Acl lon-Specinc ARARs. alHl O l h e r < r i U r U for NIHW

Sheel 2 of 2

Ac t i on

Direct Discharge of

I ' reatmcnt

System b fnuen t

Tncatmcnl

Ort>und Wate r We l l

1 Ins ia l la l ion,

l>cveli>pment. Testing.

1 and Sampl ing

( i r o u n d Water

Mon i i fH ing

Rt4|u l rc fnrn ls

Appl icable P'cderal water quality cr i ter ia lor the

pn>icci ion o f aqualic l i fe must be compl ied wi th

when environmental f ac ton are being ctmsidercd

Ar izona State Water Quali ty Siandards fo r Navigable

Waters

Sundards for miscellaneous units ( long- term re­

trievable storage, ihcnruit t reaimenl other ihan incm-

craiors, open buming, open detonat ion, chemical ,

physical, and biological Ireatment units using other

than tanks, surface impoundmenis, o r land i reatment

units) require new miscellaneous units l o satisfy

env inmmenia l performance siandards by pro tec l ion

o f ground water, surface water, and air qual i ty, and

by l imi t ing suriace and subsurface migrat ion.

Trea imenl of wastes subject lo ban on land disposal

must s i ia in levels achievable by best demonstrated

available ireatment techmilogies ( H D W ) for each

hazardous cvmstituent in each listed waste.

B D A r siandards arc based on one of four tech­

nologies or C(Hnbinations: for wastewaters (1) steam

str ipping; (2) biological treatment; or (3 ) carbon

adsorpt ion (alone or m combinat ion w i th (1) or (2 ) ;

and for al l other wastes (4) incineration. Any

technology may be used, however, i f i l w i l l achieve

Ihc concentrat ion levels spccil ied.

Regulations (or land based corrective acl ions at

R C R A facilities

Any nonwaste material (e.g., ground water or soi l)

that conlains a listed hazardous wasle musi be

managed as if i i were a hazardous waste.

( imund -wa ic r moni lor ing at new or existing K C R A

disposal units.

P n n n u L s i i t s

.Surface discharge of irealed ef f lueni .

Discharge to navigable waters.

I 'rcatmeni of hazardous wastes in units niK

regulaled elsewhere under R C R A (e.g., air

strippers).

Treaimenl of l , D K waste

l.and-bascd remedial acl ion. Nonwasic material containing listed

hazardous wasle

(.'leation of a new disp(«;i l unit , remedial

actions at an existing R C R A unit or

dis|)OKal of K C R A hazardous wasle

< i i a t ion

.SO I K .•Mt784 (July 29. 1985)

A R S 49 221

40 C l K 2b4 (Subpart X )

4 0 C I K 268 (Subpart D )

40 C I K Subpart S (Revised)

R C K A "continued in"

p i i i ic ip le

40 C l R. Subpiir l 1

AKAK

A K A R

A K A R

A K A K

AJ<AR

mc

A R A R

A R A R

Cuni i i i cn ls 1

Sec Ihe init ial screening lahlc lor chemical s)>ecifK j

A R A R s

Ihc iiulMHUlivc por l ions ol these requirenienls wi l l be

Hpplieahle or relevani and . ippropri; i le lo ifie 1

consTruclion, ofteration, mainlenanic, and closure ol 1

any miuel laneoui i treaimenl unit (a i iealnient nnil I

thai is nol el.sewhere legulaled) constructed on ihc

N l l i W site lor I rcalnienl and/or disposal o l ha/iuduus

site wa.sles.

Ihc substantive porl ions of ihese rc<piiicment.s are

applicable l o the dis(x>sal of any N I H W sile wastes 1

Ihal can be defined as i c s n u t c d h.iz^irdous w.^Mcs

Ihe substantive porl ions ol Ihese requircMients arc

relevant and appropriale lo the t iea in ic i i l prior lo and ||

dis)>os;d of any N I I I W siie wasles ihal coniain

comtKments o l reslr icled wasles in c«mccntiations thai

make the site wastes sunuien l ly similar l o ihe

regulated wasles Ihe requi iemenis specily levels of

treaimenl that must be attained prior to land dis|x>sal.

: : '

KI)l)/R.1IO/l)2,1')l 2

THE OCCUPATIONAL SAFETY AND HEALTH ACT (29 CFR 1910.120)

The Occupational Safety and Health Act (OSHA) requirements for worker protection, training, and monitoring are applicable to remedial actions at the NIBW site, and will also be applicable to the operation and maintenance of any treatment facilities, con­tainment structures, or disposal facilities remaining onsite after the remedial action is completed.

OSHA regulates exposure of workers to a variety of chemicals in the workplace, and specifies training programs, health and environmental monitoring, and emergency pro­cedures to be implemented at facilities dealing with hazardous waste and hazardous substances.

AIR EMISSIONS REQUIREMENTS

The Clean Air Act (CAA) has been implemented through a series of regulations (40 CFR 50-99) that define the air quality management programs used to achieve the CAA goals. The State of Arizona is responsible for preparation of a State Implemen­tation Plan (SIP), which describes how the air quality programs will be implemented to achieve compliance with primary standards. Upon meeting the primary standards, an area is classified as "in attainment." The SIP must also identify how the programs will maintain attainment status for each of the primary pollutants. NIBW remedial actions must comply with the substantive requirements of the CAA and its related programs, including the EPA-approved Arizona SIP.

RCRA standards for control of VOC air emissions from units such as air strippers are found at 40 CFR Subparts AA and BB. These standards require reductions, but do not include specific numeric standards.

Recent guidance on control of air emissions from air strippers used at Superfund sites for ground-water treatment is to be considered for air stripper emissions at NIBW. Controls are most needed on sources with an actual emissions rate of 3 Ib/hr or 15 lb/day or a potential rate of 10 tons per year of total VOCs because VOCs are ozone precursors (EPA OSWER Directive 9355.0-2.8, June 1989). The basis of the need for control indicates this guidance to be considered for SVE emissions at NIBW as well.

Maricopa County Rules 210, 320, and 330 are criteria to be considered for air emis­sions at NIBW. Maricopa County's January 1991 guidelines for implementing Rule 210 require VOC air emission controls for remediation sites where total uncontrolled VOC air emissions would exceed 3 pounds per day. The air emission controls must have an overall efficiency of at least 90 percent. These criteria are selected as the air emission standards for NIBW based on a consideration of the potential aggregate impacts of the numerous air stripping and soil vapor extraction systems that likely will be in operation at the site.

RDD/R310/020.51 A-10

ADDITIONAL STATE ARARs AND TBCs

Portions of the Arizona statutory code for cleanup of hazardous substances related to contaminated ground water ("Arizona superfund", Ariz. Rev. Statute Section 49-282, et seq.) and implementing regulations (TKriz. Ad. Code R18-7-109, et seq.) are applicable or relevant and appropriate to the NIBW site. The implementing regulations incorpo­rate by reference state law provisions that (1) establish that all definable aquifers are drinking water aquifers unless they qualify for an aquifer exemption and (2) establish water quality standards for these aquifers. Finally, the Arizona Superfund statute and regulations require that, to the extent practicable, NIBW remedial actions provide for the control, management, or cleanup of hazardous substances so as to allow the maximum beneficial use of the waters of the state.

Section 45-454.01 of the Arizona Groundwater Management Act (GMA) [A.R.S. Sections 45-454.01] is applicable or relevant and appropriate to the NIBW site. The remedial action selected in the 1988 ROD (Scottsdale Operable Unit remedy) requires an offsite use of the treated ground water. All offsite uses are subject to state law outside the context of the Superfund action. However, for activities conducted onsite, the substantive portions of the provisions referenced within Section 45-454.01 of the GMA shall be applicable or relevant and appropriate.

The Arizona Department of Water Resources well spacing guidelines are TBC.

RDD/R310/020.51 A - 1 1

ARIZONA DEPARTMENT OF ENVIRONMENTAL QUALITY

HFE SYMINGTON, GOVERNOR EDWARD Z. FOX, DIRECTOR

September 30, 1991 E-4110.6.2 RPU91,500

Tom Dunkleman U. S. Environmental Protection Agency 7 5 Hawthorne Street, H-7-2 San Francisco, California 94105

Re: Applicable, or Relevant and Appropriate Requirements (ARARs) Hassayampa Landfill

Dear Mr. Dunkleman:

The following State ARARs need to be included in the Record of Decision and any final remedy for the Hassayampa Landfill.

The ARARs for a proposed remedial action for the Hassayampa Landfill are hereby transmitted to the EPA. Any other ARARs developed through correspondence with other state agencies shall be presented under separate cover.

1. As stipulated under Arizona Revised Statutes (A.R.S. §49-224) , all aquifers in the state are classified as drinking water aquifers.

2. As stipulated in A.R.S. §49-282, Subsection C, remedial actions shall:

a. Assure the protection of public health and welfare and the environment.

b. To the extent practicable, provide for the control and management or cleanup of the hazardous substances so as to allow the maximum beneficial use of the waters of the state.

c. Be cost effective over the period of potential exposure to such hazardous substance.

The Depariment of Environmental Qualily is An Equal Opportunity Affirmative Action Employer.

Post Office Box 600 Phoenix, Arizona 85001-0600 Recycled Paper

Mr. Tom Dunkleman September 13 ,c 1991 Page 2

3. After remedial action, concentrations of the contaminants remaining in the aquifer(s) shall be in compliance with the maximum contaminant levels (MCLs) or, in the absence of an MCL, the Arizona Health-Based Guidance Levels for the following constituents:

1,1,1-Trichloroethane (1,1,1-TCA) 1,1,2-Trichloroethane (1,1,2-TCA) 1,1-Dichloroethane (1,1-DCA) 1,1-Dichloroethene (1,1-DCE) 1,2-DichlorGbenzene (1,2-DCB) 1,2-Dichloroethane (1,2-DCA) 1,2-Dichloroethene (1,2-DCE) Acetone Benzene bis(2-ethylhexyl)phthalate Chlorobenzene Ethylbenzene Lead Methyl ethyl ketone (MEK) Mercury Naphthalene Phenol Tetrachloroethene (PCE) Toluene Trichloroethene (TCE) Trichlorofluoromethane (Freon 11) Trichlorotrifluoroethane (Freon 113) Xylene

4. Source control of the volatile organic compounds (VOCs) present in subsurface soils at the site will also be required. Soil concentrations of the VOCs identified in #3 will be reduced to a level to which they are no longer considered to be a threat to groundwater. The techniques by which these levels will be determined will be left to the discretion of the Hassayampa Steering Committee, but will be reviewed by the regulatory agencies involved. In addition, if ADEQ Groundwater Protection Guidance Levels (GWPGLs) for soils are developed and adopted prior to remedial activities, the GWPGLs should also be considered for use as soil cleanup levels at the site.

5. During remediation of soil and groundwater, air emissions shall be treated to meet Maricopa County Air Quality Standards (Rules 200, 210,220 and 320) as dictated by the Federal Clean Air Act.

Mr.. Tom Dunkleman September 13, 1991 Page 3

6. Owners and operators must ensure that the concentration of methane gas does not exceed the lower explosive limit for methane at the facility property boundary.

7. The permit requirements under the provisions of A.R.S. §49-241, et seq., dealing with Aquifer Protection Permits shall be adhered to.

8. The entire contaminant plume will be characterized and treated as appropriate.

9. The need to withdraw groundwater for remedial activities requires Arizona Department of Water Resources (ADWR) approval under ARS §45-512. A Poor Quality Groundwater Withdrawal Permit (ARS §45-516) may be required by ADWR. Permit requirements must be met.

This list of ARARs is not necessarily a complete list and may be amplified and/or extended as remedial action alternatives are considered. Attached hereto is a list that also must be considered before the final definition of ARARs.

Sincerely,

-Jacqueline Maye Project Manager

cc: D. Marsin S. Eberhardt M. Leach J. Humphrey

ARARS

1. Maximize beneficial uses of the waters of the state in connection with remediation;

2. Health-based guidance levels for contaminants with no established MCLs;

3. Health-based guidance levels for cleanup of soil contaminants;

4. Within lOO-year floodplain must conform in general to 40 CFR 6 and to 40 CFR 264.18, specifically;

5. Preservation artifacts;

6. Preservation of critical habitats;

7. Preservation Wilderness Areas.

ARIZONA DEPARTMENT OF WATER RESOURCES 15 South 1 Sth Avenue, Phoenix, Arizona 85007

Telephone (602) 542-1553 Fax (602 256-0506

November 22, 1991

Thomas J. Dunkelman Remedial Project Manager U.S. EPA, Region IX 75 Hawthorne Street San Francisco, California

FIFE SYMINGTON Governor

ELIZABETH ANN RIEKE Director

94105

Re: Technical Screening Memorandum, Hassayampa Landfill, Dated October 28, 1991

Dear Mr. Dunkelman:

The Arizona Department of Water Resources (ADWR) has reviewed the above referenced document prepared by Conestoga-Rovers and Associates which includes a groundwater modeling study report by Errol Montgomery and Associates.

ADWR is frankly dismayed to find that the proposal to discharge treated groundwater to the Palo Verde Nuclear Generating Station effluent pipeline is still being considered as an acceptable option. ADWR has repeatedly stated that the discharge of groundwater to the effluent pipeline does not constitute beneficial use of the water. ADWR stated this position in the comments generated on the Draft Remedial Investigation Report (ADWR letter dated November 7, 1990) and at the December 11, 1990 Hassayampa Landfill Technical Committee meeting. ADWR strongly recommends that this groundwater disposal option contained in the Technical Screening Memorandum be deleted from further consideration.

Please find attached our specific comments on this document. If you have any questions, please contact me at (602) 542-1552.

Sincerely,

JUr-/l4^ Grant J. Gibson Water Resource Specialist Water Management Support Division

GG/eb

cc: Jackie Maye, ADEQ

ADWR Review Comments on the Technical Screening Memorandum, Hassayampa Landfill; Dated October 28, 1991

Table 3.1, Potentially Applicable or Relevant and Appropriate Requirements (ARARs), Hassayampa Feasibility Study:

Please note that Arizona Revised Statute Title 45 is applicable to the withdrawal and use of groundwater, minimum well construction standards, well driller licensing and well spacing requirements.

Table 4.1, Remedial Technologies To Be Considered, Hassayampa Feasibility Study:

The treated water discharge to surface water option has been eliminated from future consideration. Please amend this table so that it is consistent with page A-14, A.3.4.1 Surface Water Discharge.

Page A-4, A.2.0, Groundwater (ARARs):

The discussion of the groundwater ARARs is incomplete in this section. No mention is given to A.R.S. Title 45. which regulates the withdrawal and use of groundwater within the state.

Table A.3.1, Potentially Feasible Groundwater Treatment Technologies, Hassayampa Feasibility Study, Page 2 of 2, Item Number 8:

ADWR will not grant a permit to withdrawal poor quality groundwater unless the water is treated to at least drinking water standards. The proposal to discharge groundwater to the Palo Verde Nuclear Generating Station pipeline is not consistent with the beneficial use requirements of the Phoenix Active Management Area.

ADWR recommends that reinjection of the treated water be the favored groundwater end use for this site.

ARIZONA DEPARTMENT OF WATER RESOURCES 15 South 15th Avenue, Phoenix, Arizona 85007

Telephone (602) 542-1553 Fax (602 256-0506

November 14, 1991 FIFE SYMINGTON

Governor

ELIZABETH ANN RIEKE Director

Thomas J. Dunkelman (H-7-2) Remedial Project Manager U.S. Environmental Protection Agency, Region IX 75 Hawthorne Street San Francisco, California 94105

Dear Mr. Dunkelman:

Pursuant to your request dated September 1, 1991, please find enclosed a list of proposed ARARs submitted by the Arizona Department of Water Resources (ADWR) for the Hassayampa Landfill Superfund site. The following ARARs are part of Arizona's Groundwater Management Act (A.R.S. §§ 45-401, et seq.). In order to receive federal funding for the Central Arizona Project canal, the State of Arizona was required by the Federal Government to enact the Groundwater Management Act. The goal of this legislation was to reduce State reliance on federally funded water projects through conservation of the State's groundwater resource. Thus, both federal and state policy encourage the conservation of the State's groundwater resource as well as the beneficial end-use for all groundwater withdrawn in Arizona. ADWR hopes that as EPA analyzes ADWR's proposed ARARs, it recognizes the federal and state policies requiring conservation and beneficial use of Arizona's groundwater resource.

These ARARs Wash Record process, the Kathleen Joh (Allyn Stern The only sta A.R.S. § 45-prescribed c statute.

have been incorporated into the North Indian Bend of Decision. As part of the NIBW OU ARAR selection Department submitted a summary of Title 45 laws to nson. You might want to reference this summary should also have a copy) to assist your analysis, tute which the Department proposes as an ARAR, 454.01, incorporates much of Title 45 under ircumstances. You will find attached a copy of this

PROPOSED ARARs FOR THE HASSAYAMPA LANDFILL REMEDIAL ACTION: TITLE 45 OF THE ARIZONA REVISED STATUTES

45-454.01 Exemption of superfund remedial action activities, use requirements; definition

A. New well construction and withdrawal, treatment and reinjection into the aquifer of groundwater that occur as a part of and on the site of a remedial action undertaken pursuant to CERCLA are exempt from this chapter, except that:

1. A well that is exempt under this section is subject to Sections 45-594, 45-595 and 45-596, but no authorization to drill need be obtained before drilling.

2. If the groundwater that is withdrawn is not reinjected into the aquifer, the groundwater shall be put to reasonable and beneficial use.

3. A person who uses groundwater withdrawn in an active management area pursuant to this section shall pay the groundwater withdrawal fee for the groundwater the person withdrew or received and shall use the groundwater only pursuant to Articles 5 - 1 2 of this chapter. A city, town, private water company or irrigation district that serves groundwater pursuant to Article 6 of this Chapter is deemed to have used the groundwater for purposes of this paragraph.

B. For purposes of this section "CERCLA" means the Comprehensive Environmental Response, Compensation, and Liability Act of 1980, as amended (P.L. 96-510; 94 Stat. 2767; 42 United States Code Sections 9601 through 9657) commonly known as "Superfund".

Thomas J. Dunkelman Page Two November 14, 1991

If you have any questions concerning this letter or the enclosure, please call Howard Kopp, ADWR Deputy Counsel, at (602) 542-1507.

Sincerely,

Bruce S. Davis Chief Water Management Support Division

cc: Robert Ogilvie, EPA, Region IX RC-5 Jackie Maye, ADEQ Howard Kopp, ADWR

ARIZONA DEPARTMENT OF ENVIRONMENTAL QUALITY

HFE SYMINGTON, GOVERNOR EDWARD Z. FOX, DIRECTOR

December 6, 1991 RPU91,620 E-4110.4.6

Tom Dunkleman EPA Region IX H-7-2 75 Hawthorne Street San Francisco, CA 94105

RE: Hassayampa Landfill - Document Review and Comments

Dear Mr. Dunkleman:

I am writing regarding the review and comments by ADEQ on the following documents. These documents were submitted by the consultants for the Hassayampa Steering Committee (Montgomery & Assoc, and Conestoga-Rovers & Assoc):

1) Results From Vadose Zone Monitor Borings. Hazardous Waste Area, Hassayampa Landfill dated July 30, 1991

2) ADEQ Comments on the CRA Response (dated July 23, 1991) to the Final Draft Risk Assessment

3) Soil Gas Survey Report dated October 25, 1991

4) The Technical Screening Memorandum dated October 28, 1991

The attached memorandum addresses all of the above documents and represents ADEQ's comments at this time. In ADEQ's opinion, the elevated soil gas level in the area north of Pit 1 is significant and warrants further site investigation. The design of a remedial alternative may not sufficiently capture the contamination zone given the southeasterly flow direction which now seems evident. Based upon the above, it appears that prior assessment and investigative efforts have not sufficiently defined the full extent of the groundwater contamination at the site. ADEQ looks forward to working with the EPA in identifying areas needing further investigation.

The Department of Environmenial Quality is An Equal Opportunity Affirmative Action Employer.

Post Office Box 600 Phoenix, Arizona 85001-0600 Recycled Paper

Mr. Tom Dunkleman December 6, 1991 Page 2

If there are any questions please don't hesitate to call me at (602) 257-2167.

Sincerely,

Jacqueline Maye Project Manager

(attachments)

cc: M. Leach A. Brown D. Marsin L. Pollock

Arizona Department of Environmental Qualily

MEMORANDUM

DATE: Decembers, 1991

TO: Jacqueline Maye, Project Manager Remedial Projects Unit

THRU: Michele Kennard, Manager Superfund Hydrology Unit

FROM: Michael Leach, Hydrologist Superfund Hydrology Unit

RE: Hassayampa Landfill: Comments on Recent Data Submittals. Reports, and Field Activities Regarding the Site.

Upon hydrogeologic review, the following comments have been generated regarding recent reports, data submittals and memos received by both EPA and ADEQ from the HSC. These comments are given below for the indicated report or subject:

A. "Results From Vadose Zone Monitor Borings. Hazardous Waste Area. Hassayampa LandfiU" by EMA (8-30-91)

Al. Page 2. paragraph 2. The relevance of the content of this paragraph as it applies towards HVOC soil vapors is not understood. Statements in this paragraph attempt to construct a scenario in which downward liquid waste movement through the vadose zone to the underlying aquifer would not appear to be a viable contaminant migration route. In tum, this paragraph suggests that HVOC soU vapor movement may be an "important mechanism" for contaminant movement beneath the site. Although HVOC soil vapor transport may be a significant factor for the presence of contaminants in the vadose zone in no way can the presence of HVOC soil vapors be solely or even significantly responsible for the HVOC groundwater contamination detected at the site (as is implied in this portion of this report).

SpecificaUy, the report states that based upon the visual inspection of the soU samples obtamed from borings completed beneath Pits 1 and 3, soUs beneath the pits were of low moisture content. In tum, it is then concluded that present Uquid movement through the vadose zone must be "none or meager". In my opinion the key words here are present Uquid movement. I agree that recent subsurface investigations do indicate that very little or no Uquid movement is presently occurring through the soils beneath Pit 1. However, the recent investigations give no indication of what has occurred in the past. It is my opinion that the Uquid wastes discharged into the unlined pits

percolated rather quickly through the fine-grained zone overlying the basalt layer and then migrated to the north over the basalt, along with the dip of the basaU. Since both EMA and CRA have stated that the basalt is significantly fractured, it does not appear that the basalt would hinder this downward migration of Uquid wastes. This contaminant transport scenario is supported not only by the fact that the highest soU concentrations of HVOCs beneath Pit 1 were found immediately above the basalt layer, but also by the soU gas survey lab results obtained from the soU gas probes extending to the north of Pit 1. The data obtained from these probes will be discussed later in this memo.

A2. Once again in regards to the potential for downward Uquid waste migration beneath Pit I, an argument recently posed by the HSC is that the vertical hydrauUc conductivity values of the soUs in the lower fine-grained zone are too low to have aUowed the Uquid wastes to pass through this zone and be subsequently detected in HS-1. Similarly, I am certain that the statement made later in this same paragraph of the repprt (Page 2, paragraph 2) regarding the consideration of "other RI results," in conjunction with the statement regarding the low observed soU moisture conditions is referring to the vertical hydrauUc conductivity tests conducted on select vadose zone soils at the site. However, upon review ofthe vertical hydrauUc conductivity test results obtained from the onsite soil samples and after some additional research into the subject, it is my opinion that the laboratory values generated for the vertical hydrauUc conductivity of subsurface soils at the site are not vaUd and caimot be used for the purpose that the HSC and their consultants have tried to use them (to suggest a impermeable fine-grained soU barrier beneath the site). The foUowing is information supporting this opinion:

A2.a The vertical hydrauUc conductivity (k) data obtained from the laboratory analysis of the onsite soil samples is only vaUd for that discrete volume of soU which was sampled. Although the laboratory k values obtained may or may not be representative of the primary permeabiUty of the soU, this type of permeabiUty testing cannot determine the permeabiUty of the soUs over a large area, and certainly cannot be used to conclude that the fine-grained sediments directly above the basaU constitute an effective barrier to the downward migration of Uquids from Pit I.

Vertical hydrauUc conductivity values could be determined much better using an infiltration test method in a properly constmcted boring. In similar situations, infUtration test values of k may be a few orders of magnitude higher than the laboratory - determined values. Major differences are beUeved to be a result of higher vertical Uquid flow rates in secondary permeabiUty features in the fine-grained deposits such as root tubes and fracturing, and the greater volume of soil material being measured by field methods versus laboratory methods. Larger volumes of material measured by field methods would obviously be more representative of actual site conditions. A field study ofthis sort has been included as Attachment A.

A2.b The very technique by which soU samples are coUected in the field would tend to compact the soils and lower the k values calculated in the laboratory. In the field during sampling, soU is forced into brass tubes within a larger sampling device by repeatedly dropping a 140-pound weight onto the top of the sampUng device. This

sampUng process can only help to compact the soil samples and lower the laboratory-calculated k values.

A2.C There is some evidence that the presence of Uquid-phase organic chemicals (including HVOCs) can degrade clay soUs such as those comprising the fine-grained basal unit. Experiments indicate that the solvents can displace water in the clay stmcture leading to shrinkage, cracking and an increase in permeabiUty. This type of scenario is certainly possible given the extremely large volumes of Uquid-phase organic chemicals which were disposed into Pit I. Further discussion of this process is given in Attachment B.

A2.d The very laboratory technique used to analyze soUs for vertical hydrauUc conductivity tend to greatly underestimate actual soU k values. Attachment C is a reference which includes a good summary of potential errors encountered in laboratory permeabiUty tests.

Table 5 on page 59 of Attachment C is a summary of errors which can be made as a result of laboratory conductivity tests and the degree to which the lab error can result in a change in the soU hydrauUc conductivity. For the Hassayampa soU samples, it is doubtful that any of the Usted sources of error which would result in the lab k value being too high could actuaUy be occurring. In addition, other sources of lab error which would tend to underestimate actual k values (lab value of ^measured too low) are either insignificant or simply do not seem to apply to the Hassayampa situation.

Iri the case of the Hassayampa soil samples, review of this table indicates the use of distiUed water as permeant was probably the source of lab error. A telephone conversation with the laboratory (SHB) which conducted the permeabiUty tests on the Hassayampa soil samples confirmed that distUled water was indeed used as the permeant. Page 28 of Attachment C states that "leaching a sample with distiUed water may cause expansion of the diffuse cloud of absorbed cations around clay particles and reduce hydrauUc conductivity". Table 5 on page 59 of the same attachrhent indicates that the use of distiUed water as a permeant for such tests can underestimate the hydrauUc conductivity of a soU sample anywhere from at least one to over two orders of magnitude. Interestingly enough, as mentioned in part Al.a ofthis memo, Table 5 of Attachment C also Usts the use of laboratory methods rather than field methods for estimating hydrauUc conductivity of soUs as a source of error in determiiring actual k values. This table indicates that the use of laboratory tests rather than field methods can underestimate the hydrauUc conductivity by over four orders of magnitude.

A2.e Not only are the laboratory k values calculated for the Hassayampa soUs much lower than would be expected for the grain-size distribution of these soUs, the soU k values obtained for the onsite soUs do not make any sense when compared to one another given their respective grain-size distributions.

Table 1 is a summary of saturated hydrauUc conductivities calculated for many loam soUs (sand, silt, and clay mixtures) of several different types which are simUar in nature to the loam soils analyzed for the Hassayampa site. Upon comparison with the k values

obtained for the soils obtained at the Hassayampa site (Table 2), it is obvious that the "typical" k values for the types of soUs encountered at the Hassayampa LandfiU are generaUy a couple orders of magnitude higher than were actuaUy measured by the laboratory for the Hassayampa soil samples. This indicates that the laboratory k values for the onsite soUs are indeed lower than what would be expected under natural conditions.

In addition, upon review of Table 2, given the grain-size distributions of the Hassayampa soU samples, it appears that the onsite soU k values obtained don't make much sense when compared to one another. For example, soU samples obtained from borings SB-3 and SB-5 have the lowest k values calculated at the site yet these soU samples contain some of the highest percentages of sand and sUt and lowest percentages of clay found in any of the onsite soil samples. Conversely, the soil sample obtained from soU boring SB-7 had one ofthe highest lab k values calculated for the fine-grained unit yet the actual sample contained no sand and 70 percent clay. It is not typical by any means that a finer-grained soU would have a higher ^ valtie. Therefore it is my opinion that the k values obtained for the onsite soUs are not vaUd and cannot be used for any purpose, especiaUy to state or imply that the fine-grained unit beneath the landfiU is relatively impermeable to any downward percolation of Uquids.

A3. Pages 2-5. As you may remember, on 6-5-91 a tele-conference meeting was held by members of the technical committee for Hassayampa LandfiU to discuss the voluntary expansion of the soU gas survey program at the site. SpecificaUy, the HSC had aUeady authorized EMA to instaU four vadose zone borings in the Pit l/HS-1 area, and had completed these weUs in late May without the authorization of either ADEQ or EPA. In addition, during the conference caU the HSC and their consultants were proposing the InstaUation and sampling of additional soU gas probes emanating in three different directions away from Pit I. They indicated this additional work was required to gather data on HVOC soU gas concentrations near Pit I in order to evaluate the necessity and/or scope of any remedial activities for the vadose zone cleanup. Although I do agree that the soU gas work is necessary prior to any remediation of the vadose zone, I do not agree with the necessity of the vapor boring locations emanating out from Pit I being directed towards HS-1.

The report states that the HSC considered it to be "pmdent and cost effective" to constmct vadose zone monitor borings to obtain soU gas samples below the land surface between Pit 1 and the nearest area where volatUe orgaruc compounds had been detected in groundwater samples at the site. Upon review of Figure 1 of the Vadose Zone Boring report, it appears that monitor weU MW-6UA, not HS-1, is the closest monitor weU to Pit 1 to have had detected significant amounts of groundwater contamination. Considering that it took additional work to once again find and resurvey the original location of abandoned monitor weU HS-l, it seems strange that the HSC would go out of their way to target the HS-l area.

A4. Page 5. The last sentence on this page states that the results of this soU gas survey conducted at the landfill by Tracer Research wiU be submitted in a report by CRA. The soU gas samples were obtained by Tracer during the period of July 9th to the 18th of

this year. Up untU October 30, 1991 neither ADEQ nor EPA had received the results of this soil gas survey. Phone conversations between both ADEQ and EPA with BiU Victor (EMA) have indicated that the data was tied up in lab QA/QC. However, by no means should submittal of this data have been delayed so long.

A5. Page 7. first paragraph. This section of the report describes the constmction of the vapor borings instaUed at the site. The report states that due to constmction problems encountered during the InstaUation of VB-2f (the only vapor boring completed in the fine-grained layer), "results for soU gas samples obtained from this boring may not be completely representative for the fine-grained part of this unit". In addition, page 15 of the report states that "based on the results from the constmction of vapor boring VB-2f, these laboratory chemical results can not be used to make reUable conclusions regarding soU gas in the fine-grained part of the upper aUuvial deposits unit".

Given the stated objective of the vapor boring investigation (siting and design of ftiture remedial activities for the vadose zone), the HSC should authorize the instaUation of additional vapor borings completed in the fine-grained zone near Pit I to replace VB-2f, Although a vast majority of the HVOC soil contamination beneath Pit 1 is found in the lower fine-grained unit, four of the five vapor borings were completed in the upper coarse-grained unit instead of the fine-grained unit. Therefore, it seems that representative soU gas samples from the fine-grained unit would be essential for such an investigation and at least three or four additional soU vapor borings should be instaUed in the fine-grained soU layer around Pit I.

A6. Page 14. last paragraph. The report states in this section that the HVOC soU gas concentrations detected in samples obtained from the vapor borings on July 30, 1991 are more representative of the soU vapor conditions in the Pit 1 area than the data from soU gas samples obtained on June 4, 1991. However, upon review of the summary of the HVOC laboratory results, consultation with Performance Analytical Inc. (PAI - the lab that conducted the analysis of the 7-30-91 soU gas samples) and other ADEQ professionals knowledgeable in HVOC soU gas science, it appears that the HVOC concentrations presented on Table 6 of the Vadose Zone Boring report are far from accurate. In creating Table 6 of the report, EMA appears to have taken the results from each soU gas analyses in parts per miUion (ppm) and simply multipUed the result by 1000 to obtain the value in micrograms per Uter (ug/l). This conversion from ppm to ug/l may work fine for dissolved chemicals in water, but it is not appUcable for gases. Upon discussions with other individuals, including the chemist at PAI who conducted the analysis on the Hassayampa soU gas samples, mUUgrams per meter cubed (mg/m^) is equivalent with ug/l. Therefore, the actual values that should be given on Table 6 ofthe report should be the values from the "mg/m^" column ofthe lab reports, not the "ppm" column (see example in Attachment D).

A7. Page 15. first paragraph. In this section of the report it is "noted" that the assemblage of HVOCs detected in soU gas samples is simUar to the assemblage of HVOCs detected in groundwater samples obtained from abandoned monitor weU HS-1. In my opinion, this statement attempts to imply a Unk between the groundwater contammation detected in HS-1 (which the HSC claims was improperly constmcted and therefore constituted

a potential contaminant conduit to groundwater) with the presence of HVOC soU vapors in the vadose zone. WhUe it may be tme that high levels of particular HVOCs were detected in both the vadose zone soU gas and in the underlying aquifer, there are several problems with attempting to link groundwater contamination at HS-l to the presence of HVOCs in soU gas near the HS-l site. These problems are discussed below.

A7.a WhUe it may be tme that the assemblage of HVOCs (in terms of individual HVOC concentrations) is simUar when comparing HVOCs detected in soU vapors near Pit I/HS-l and groundwater from samples obtained from monitor weU HS-l, the report faUs to also state that the assemblage ofthe HVOCs in the soU samples obtained beneath Pit I in SB-19 was also simUar (as would be expected) to the assemblage of HVOCs in groundwater samples from HS-1 and also to the soU vapor samples in the Pit l/HS-1 area. Table 3 is a summary ofthe most recent HVOC levels and assemblages obtained from VB-3 (soU gas), HS-l (groundwater), and SB-19 (soUs beneath Pit I at the deepest sampling interval). Table 4 is a summary of the total volume of each HVOC disposed into Pit I.

Upon inspection of Table 3, it does appear that the assemblage of HVOCs in soU gas (VB-3) and groundwater (HS-l) is simUar. However, when reviewing the assemblage of HVOCs in the soU samples from SB-19, it appears that the assemblage of the HVOCs is simUar in aU three types of envUonmental media. This is to be expected, since aU of the HVOCs found in the groundwater and soU vapors at the site were the result of Uquid waste disposal into the onsite pits. SpecificaUy, the groundwater was contaminated by way of Uquid waste disposal through the vadose zone to the underlying aquifer, and the soU vapors in the vadose zone were created from volatUization of the Uquid wastes as they initiaUy migrated through the vadose zone. HVOC soU vapors are StiU being generated in the vadose zone from the contaminants remaining in the vadose zone soUs beneath areas of the site after much of the Uquid waste has moved through them.

A7.b Other data which indicate that the soU gas/groundwater HVOC contaminant transport route is not vaUd for the HS-l scenario Ues in the relationship between the soU gas data itself and the groundwater quaUty data gathered at HS-1.

Upon release into the unsaturated zone, an organic contaminant wiU eventuaUy equiUbrate itself between three phases: soU, water, and pore air. The distribution of the chemical among these phases is dependant upon two properties ofthe chemical, the Organic Carbon Partition Coefficient (K^) and Henry's Law Constant (KH), and one property of the soU matrix, the percent content of organic carbon (Graf, 1991). Assuming that the HVOC soU vapor concentrations detected in the onsite vapor boring, VB-3 (VB-3 was used for this analysis since it was nearest to HS-l, about 6 feet away), were in equiUbrium with the soU and water in the vadose zone, the HVOC vapor concentrations detected in VB-3 should be representative of the gas concentrations which would come into contact with the groundwater within HS-1. Therefore, the KH value for each HVOC detected should govem the HVOC concentrations existing at the surface of the water table in HS-1.

Review of the Table 4 of the EMA Vadose Zone Boring report mdicates that a total of eleven HVOCs were detected in VB-3. Table 5 of this memo gives the Ust of HVOCs which were detected in VB-3 with the corresponding K , and KH values and observed soU gas concentrations, C„ at VB-3. By using Henry's Law Constant, the equUibrium water concentration of each contaminant in water, C^., can be calculated

C W = C J / K H

and compared to the HVOC concentrations in groundwater, C^„, which were actuaUy observed in HS-1. Calculations for aU 11 HVOCs are given m Table 5 including the anticipated mass contaminant balance for each of the detected HVOCs (f , f,, f ) utilizing values of organic carbon content and bulk density from on-site soUs. It should be remembered, however, that the equUibrium water concentration would be expected only at the surface of the water table. The concentration at any point below the surface would be governed by the rate of HVOC diffusion downward into the water column.

In reviewing Table 5 it appears that there are 4 different categories of HVOC contaminant detections:

HVOCs which were detected in significant concentrations in both soil gas (VB-3) and groundwater (HS-l):

This category of HVOCs includes:

1,1-DCE TCE

1,1,1-TCA PCE

FREON 11

For aU of these HVOCs, the calculated equUibrium water value is significantly greater than what was actuaUy observed in HS-1. One could either conclude that the equUibrium concentration is not uniformly distributed down through the water column in HS-1 or that the water concentrations are not due to equiUbrium with purported vapors in HS-1, but instead are due to some other mechanism altogether (i.e. contamination by migrating HVOC Uquids). Therefore, based on this analysis, the source ofthis category of contaminants detected in HS-l groundwater samples remains uncertain.

HVOC concentrations in soU gas which are too low to theoreticaUy produce the groundwater concentrations, observed in HS-1:

This category includes only Freon 113. Observed groundwater values for Freon 113 from HS-1 were significantly above the theoretical maximum that would be in equiUbrium with the soU gas values obtained from VB-3 (Table 5). In other words, the level of Freon 113 in soU gas detected at VB-3 was not sufficient to produce the groundwater concentrations observed in HS-1. Therefore, the theory of HVOC soU gas

contammatmg the groundwater within HS-l does not hold tme for this constituent.

HVOCs which were detected in significant amounts m soU gas samples at VB-3. yet Uttie or none ofthe contaminant was detected in HS-1:

From Table 5, this category of HVOCs includes:

1,1-DCA 1,2-Dichloropropane

Dichloromethane (DCM) 1,2-DCA

Given the VB-3 soU gas concentrations of the four HVOCs Usted above, high concentrations of these chemicals should have been detected in HS-1 if a soU vapor contaminant transport hypothesis has merit. SpecificaUy, equiUbrium water concentrations in HS-l for 1,1-DCA, 1,1-Dichloropropane, DCM, and 1,2-DCA should have been 1,033; 510; 6,571; and 342 ug/l, respectively. However, only 21 ug/l of 1,1-DCA was detected in HS-1 and concentrations of the other three HVOCs were not even detected. In my opinion, this is the most damaging evidence to a HVOC soU vapor contaminant transport theory. If HVOC concentrations in pore air are postulated to have caused the contamination in HS-l, why aren't aU of the HVOCs present in the vapor phase? Not only were the moderate to high water concentrations calculated for the HVOCs not detected in HS-1, three out of the four were not even present and one was detected at less than l/50th ofthe predicted concentration.

* HVOCs that were detected in soU gas at VB-3 but due to the chemical having a very high Henry's Constant. KH. this constituent would not be expected to be detected in groundwater samples from HS-1:

The only HVOC in this category would be vinyl chloride. Vinyl chloride was detected in soU gas samples from VB-3 but was not detected in HS-l, as would be expected.

A7.C Although somewhat aside from the point of a soU gas/groundwater contamination connection and more to the point of contaminant assemblages, a closer look at Table 3 reveals some additional points of interest. The HVOC of highest concentration in the three media were not the same - 1,1-DCE (DCE) was the highest in both soU gas and the groundwater yet the soUs beneath Pit 1 contained more 1,1,1-TCA (TCA) than any other HVOC. This is not surprising since DCE is a degradation product of TCA, and the formation of DCE would be expected to occur ftirther away from the discharge point of the HVOCs (Pit 1). This is also evident when looking oiUy at the soU gas data itself. Table 4 of the Vadose Zone Boring report shows that TCA soU gas levels are almost twice as high as DCE in the vapor boring closest to Pit 1 (VB-1) and yet in aU of the vapor borings fiirther away from Pit 1 the levels of DCE become greater than the TCA, High concentrations of TCA would be expected near Pit I since over 5,000 gaUons of this chemical was discharged into this pit.

Freon 113 was the HVOC detected at the second highest level for the three media types. This is not totaUy unexpected since over 16,0(X) gaUons of Freon 113 were discharged into Pit 1 (more than any other HVOC discharged into Pit 1).

About 5,000 gaUons of dichloromethane (DCM) were discharged into Pit 1, but DCM has only been consistentiy detected in the soU gas and soU boring samples, and was never confirmed to be present in any groundwater samples obtained from either HS-l or MW-6UA. Given the large volume of DCM discharged into Pit I, the relatively low KH (.07) and K^ (8.8) values for DCM, and the high NAPL mobUity of DCM (3.02), it would seem that there should be a great deal of DCM in the aquifer underlying the Pit 1/HS-l area. However, as is shown on Table 6, DCM degrades relatively quickly in both the vadose zone and in groundwater and, therefore, this characteristic of DCM may account for its non-detection in groundwater. The high biodegradation rate of DCM may also account for the relatively low soU concentration of DCM in SB-19 (Table 3) especiaUy given the large volume of Uquid-phase DCM placed into Pit 1 (Table 4).

PCE was placed into Pit 1 in relatively smaU volumes (Table 4) and thus relatively low concentrations of PCE were detected in aU three environmental media (Table 3).

About 1,300 gaUons of 1,2-dichloropropane (DCP) were discharged into Pit 1 and DCP was detected in SB-19 soU samples and VB-3 soU vapors, but was never detected in HS-1 groundwater. Subsequent groundwater sampUng at MW-6 UA has detected DCP in very smaU concentrations.

There is no record of TCE being discharged into Pit I in the RI/FS Uquid waste manifest summary, however, TCE has been detected in elevated concentrations in aU three environmental media in the Pit l/HS-1 area. The reason for this occurrence is unclear. It is my opiiuon that a load of TCE was mistakenly manifested to go into Pit 2 when it was actuaUy discharged into Pit 1 (where this type of waste was supposed to go). The waste manifest summary in the RI/FS for Pit 2 indicates that over 1,800 gaUons of TCE went into Pit 2. However, upon review of the laboratory results of soU samples obtained from soU boring AB-2 which was completed beneath Pit 2, there is no evidence of significant HVOC contamination beneath Pit 2. Given the large volume of TCE which was supposedly discharged, it seems possible that this discharge could have actuaUy occurred into Pit 1.

Only a smaU volume of Freon 11 (66 gaUons) was ever discharged into Pit 1. AUhough Freon 11 was detected in HS-1 groundwater and VB-3 soU gas, no Freon 11 was detected in soU samples beneath Pit 1 in either AB-3 or SB-19. However, given the high NAPL mobiUty of Freon 11 (3.55) and its relatively high Henry's Constant (4.62), it seems that much of this HVOC could have moved through the vadose zone very quickly leaving behind only smaU amounts within vadose zone which was in tum quickly volatilized.

B. ADEO Comments on the CRA Response (8-23-91) to the Final Draft Risk Assessment (RA)

I realize that it has been over two months since the ADEQ receipt of this letter, but my comments here are not being generated specificaUy in regards to the RA. Instead, the foUowing comments on the CRA responses to the final draft version of the RA are primarUy intended to comment on some of the new issues regarding the magnitude, fate, and migration of the HVOCs discharged into the pits at Hassayampa LandfiU.

Bl. Page 7. last paragraph. This section of the CRA responses is in regards to the "Geology and Hydrogeology" Section of the RA. The CRA comments state that "vadose zone contamination that may occur in the basaltic lava flow unit would be expected to occur chiefly in fractures and weathered zones". Although I agree with this statement, the foUowing sentence in the CRA comments is contrary to recent data gathered at the site. SpecificaUy, the sentence reads, "Although it is possible that the basaltic lava-flow unit may have influenced movement of Uquids in the vadose zone at the Site, it is likely that the unit is sufficiently fractured to not have been a substantial influence."

Data gathered from recent soU gas surveys (7/91) conducted at the site in the vicinity of Pit 1/ HS-l indicate that the basaU unit beneath Pit I may have indeed had a great influence on the movement of Uquid wastes beneath the site. On July 11, 1991, I visited the site whUe soU gas sampUng operations by Tracer were taking place. SoU gas sampling in the Pit 1 area had been taking place during the previous few days. As part of the HSCs voluntary expansion of the soU gas sampling activities required by EPA in the Special Pit Areas, additional soU gas probes were instaUed and sampled in three different transects emanating from Pit 1. Each of the three transects consisted of at least 3 soU gas probes spaced about fifty feet apart. One transect extended to the southeast towards HS-1, another was extended to the southwest of Pit 1, and the last transect extended to the north.

WhUe visiting the site on 7-11-91, I asked for and obtained from Tracer Research representatives, soU gas results from some ofthe probes instaUed in and around Pit 1. These resuUs are summarized in Table 7. Upon inspection of the data I noticed that the HVOC soU vapor concentrations in the northerly transect were increasing with increased distance away from Pit 1. At this point I asked the lab technician mnning the GC unit whether or not he had been mistaken in his designation of the soU gas probes for the northerly transect. He repUed that he had not been mistaken and added that the readings were quite pecuUar since HVOC soU vapor levels would be expected to decrease as you moved away from the HVOC disposal site (Pit I), as was observed in the southeast soU gas probe transect.

Although the data was preUmmary and Umited to only the three transects surrounding Pit I, it was and remains my opmion that the soU gas data mdicate significant amounts of HVOC liquids migrated to the north of Pit 1 with the dip of the basalt beneath the site. The significance of the soU gas data near Pit 1 wiU be discussed m greater detaU in my review of the SoU Gas Survey Report by CRA.

10

B2. Appendix A - Attachment 1. The techniques used by CRA to calculate the volumes of Uquid wastes discharged into each pit at the landfiU is incorrect. CRA calculated the volume of HVOCs discharged into the pits on a weight basis instead of a volume basis. There is no indication on the liquid waste manifests that Uquids were disposed at Hassayampa on a weight basis. Instead, indications are that wastes were disposed on a percentage, or volume basis. Since the primary HVOCs of concem have greater specific weights than water, calculation of the Uquid waste volumes on a weight basis results in lesser calculated volumes of HVOCs actuaUy discharged at the site. In addition, some of the HVOC Uquid waste discharges Usted for the site in the RI/FS were not documented in Table I of Appendix A of the CRA letter. These include:

• 44 gaUons of PCE from Syntex OphthamaUcs

• About 100 gaUons of PCE from Prestige Apparelmaster

• 5,400 gaUons of methylene chloride from HoneyweU

• 7,950 gaUons of various HVOCs from HoneyweU

• 66 gaUons of TCFM from APS

• 1,350 gaUons of propylene dichloride (same as 1,2-Dichloropropane: 1,2-DCP) from HoneyweU

C. "SoU Gas Survey Report" by CRA (10-25-91)

The soU gas survey conducted at the site by CRA was received by ADEQ on 10-30-91. This study includes the resuUs of soU gas sampling in both the Special Pit Areas and the area surrounding Pit 1.

Cl. Page 3. third paragraph:

Although orgaiuc solvents may have been approved for Pit 2, as is discussed in Comment A7.c of this memo, it does not appear that this disposal actuaUy took place mto Pit 2.

C2. Figures 3-1 through 3-7

From the appearance of these figures it seems that the grid spacing and number of soU gas probes completed and sampled in the Special Pit Areas was indeed adequate. The soU gas work conducted in the Special Pit Areas was able to delineate discrete areas or "hotspots" where a majority of the HVOC disposal most Ukely occurred. This area appears to be in the southem part ofthe 1980 Special Pit Areas and extendmg to the southwest of this area for about 200 feet. In addition, an area in the southwest portion ofthe Special Pit Areas appears to have greatly elevated levels of HVOCs. Therefore, discrete zones in the Special Pit Areas with considerable vadose zone soU contamination have been identified which wiU enable the efficient siting of both vadose zone and groundwater remedial devices. AU things considered, it appears that the soU gas survey in the Special Pit Areas was a success.

1 1

In addition, upon review ofthese figures it appears that the soU gas surveys in both the Special Pit Areas and near Pit 1 were able to distinctly iUustrate the presence of the three on-site HVOC source areas (two in the Special Pit Areas and Pit 1) and the presence of another off-site source area. The other source area appears north of Pit 1, which in my opinion indicates that a great deal of HVOC Uquid waste migration occurred over the basalt layer beneath Pit 1. This basalt layer dips to the north and appears to have controUed the migration of Uquids in the vadose zone which resulted in the movement of a great deal of Uquids to the north of Pit I.

Figures 3-1 through 3-7 of the report generaUy draw a smaU circular contour of high HVOC concentration around the northemmost soU gas probe (N-4) of the north transect emanating from Pit I in an attempt to make the HVOC concentrations in this area appear anomalous. This of course is not scientificaUy correct due to the Umited scope ofthe soU gas survey around Pit I. In my opinion, since the soU gas survey around Pit I was not intended to evaluate the potential for HVOC soU gas sources north ofthis pit, and given the disposal history of HVOCs into Pit I, the dip of the basalt to the north, and the presence of extremely high HVOC soU gas concentrations to the north of the site, it appears that the depiction of the soU gas concentrations in this area as being both localized and anomalous is not only premature and inaccurate, but is also an attempt to trivialize certain data which may have just discovered additional significant problems at the site.

C3. Page 13. paragraph 3. This paragraph states that although the HVOC soU gas concentrations in the Special Pit Areas are elevated, the HVOC levels are "much less sigmficant than the area surrounding the Pit 1". Although I do agree that the HVOC soU gas concentrations in the Special Pit Areas are lower than those associated with the Pit I area, I also beUeve that the use ofthe word "significant" in this paragraph ofthe report impUes that the presence of HVOCs in the Special Pit Areas is somehow not as important when compared to the contamination at Pit I.

The soU gas investigation conducted in the Special Pit Areas was very significant in that it not only confirmed that large amounts of HVOCs were discharged into these areas, but this study was also able to delineate discrete areas within the Special Pit Areas which could be considered source areas for the groundwater contamination which has been detected at monitor weU MW-IUA. This information wiU be extremely valuable for the siting of remedial faciUties for both the vadose zone and the groundwater in the Special Pit Areas. Therefore, 1 do not think it is appropriate to characterize the contamination in the Special Pit Areas as bemg less significant than any other contammant source at the site and I beUeve that the word "significant" should be removed from the text of this paragraph.

12

C4. Page 13. last paragraph. In reference to the greatly elevated levels of HVOCs detected in the soU gas probe about 200 feet north of Pit 1, the SoU Gas Report, as was illustrated in Figures 3-1 through 3-7, attempts to characterize the presence of "elevated" HVOCs detected in this area as being a singular event and discontinuous in nature. The report also concludes in this paragraph that "it is improbable that diffusive transport or Uquid transport of contaminants detected in the disposal pits are responsible for this single, discontinuous elevated detection".

The soU gas survey conducted in the northeast portion of the landfiU was intended to determine HVOC soU gas vapor concentrations immediately surrounding a known HVOC disposal and source area. Pit I. In order to characterize the HVOC soU concentrations around Pit 1, 11 soU gas probes and five vapor borings were instaUed and sampled. As a result, it appears we now have a good idea of the distribution of HVOC soU vapors in the coarse-grained deposits surrounding Pit I. Additional HVOC soU gas data is necessary in this area in order to support CRA's claim that the occurrence of greatly elevated HVOCs in soU gas probe N-4 is anomalous and unrelated to the site. Aside from the soU gas probe placed dUectly into Pit I, no other soU gas probe at the site detected greater amounts of total VOCs, not even those probes only 50 feet from Pit 1 or those probes in the most contaminated zones of the Special Pit Areas (both of which were in the vicinity of known HVOC source areas). CRA offers no explanation of how the HVOCs may have migrated to the N-4 area.

HVOC soU gas concentrations such as those detected in soU gas probe N-4 are obviously not a common occurrence and are no doubt associated with a great deal of HVOC contamination of subsurface soUs. In fact, using CRA's own evaluation of representative HVOC soU to soU gas concentration ratios for the Hassayampa site (Table 3-1 ofthe report) and given the soU gas concentrations at N-4 ofthe five HVOCs Usted in this table, the foUowing HVOC soU concentrations for these parameters should be present in the N-4 area:

Compound Soil Gas Conc. (ug/H SoU/Soil Gas Ratio

1,1-DCE

PCE

1,1,1-TCA

TCE

Freon.

77

1700

2200

540

7200

0.25

0.05

0.07

0.02

0.03

_j^: _-- —l,---p- --o.l

19

85

154

11

216

Needless to say, these estimated HVOC soU concentrations for the N-4 area are very significant and are comparable to the HVOC soU concentrations estimated m the SoU Gas Survey report for the Special Pit Areas soUs.

13

This paragraph in the SoU Gas Survey report also states that it is improbable that diffiisive transport or Uquid transport of the contaminants detected in the disposal pits was responsible for the detection of elevated HVOC levels in soU gas samples obtained from N-4. Given the HVOC soU gas data from the soU gas probes north of the Pit 1, it does appear that the greatly elevated HVOC detections in N-4 were not from diffusive transport of vapors from Pit I. However, it has long been the opiiuon of this writer and others at ADEQ that there was a good possibiUty that the basalt layer underlying the site could have had a significant effect on the movement of Uquid wastes in the vadose zone beneath the site. Since the basalt layer is known to dip to the north and given the extremely large volume of Uquid wastes disposed into Pit I (360,000 gaUons) in a relatively short period of time (18 months), it remains the opinion of myself and the ADEQ Superfund Hydrology Unit that these circumstances resulted in a great deal of Uquid wastes migrating to the north over this basalt layer. Since the basalt is known to be somewhat fractured, some portion of the Uquid wastes would leak through the basaU to groundwater as the remaimng Uquids migrated further to the north. It is also beUeved that the HVOC soU gas concentrations detected at N-4 supports this Uquid waste migration pathway.

In addition, the HVOC soU gas data summary presented in Table B-l ofthe SoU Gas Survey report tends to indicate a subsurface Uquid waste migration pathway to the north of Pit I. This Uquid waste migration pathway north to the N-4 area is supported by the foUowing:

Using one of EMAs techniques of data evaluation, the assemblage (by concentration) of the HVOCs are simUar when comparing the HVOCs detected in the soU gas probe instaUed into Pit 1 (SG-PTl-7) and N-4. SpecificaUy, the soU gas samples from both areas contained TCA and Freon 113 concentrations which comprised over 75 percent of aU of the HVOCs detected in these areas.

The only other soU gas probe sampled at the site where levels of PCE and Freon 11 were even close to those detected in N-4 was immediately within Pit 1. This not only indicates that the PCE and Freon 11 in both areas are derived from a common source, but it also shows that simUar magnitudes of vadose zone contamination ofthese HVOCs are present in each of these areas which in tum suggests the Uquid waste migration scenario. SimUarly, TCE and 1,1-DCA (DCA) are present in significant soU gas concentrations in the N-4 area indicating northward migration of Uquid waste.

In general, given a single source of HVOCs in subsurface soUs, HVOC soU gas concentrations should decrease with increased distance away from the source. Pit 1. This type of total HVOC concentration decrease was observed in the samples obtained from both the soU gas probes and vapor borings in the southeast transect towards the HS-1/MW-6UA area. However, in the northerly and southwesterly soU gas transects this phenomenon does not occur and HVOC soil gas concentrations generaUy increase with increased distance away from Pit 1. This occurrence is probably due to the presence of separate additional sources of HVOCs in soUs m these dUections from Pit I, the Special Pit Areas to the southwest and the soU contaminants to the north due to subsurface Uquid waste migration in this direction.

14

* Perhaps the most interesting piece of soU gas information indicating that the subsurface migration of Uquid wastes occurred to the north of the site is the relationship between TCA and DCE . TCA is an industrial solvent commonly used for degreasing purposes. Upon release to the environment, TCA is known to commonly degrade to both DCA and DCE. Although not common in many areas of the country, the prominent TCA degradation product observed in Arizona soUs and groundwater is DCE, not DCA. This phenomenon is evident in many ofthe HVOC contaminated sites around Arizona, including Hassayampa LandfiU. The best example at the site is MW-IUA, which monitors the results of Uquid waste disposal into the Special Pit Areas. Upon review of the groundwater quaUty results from this weU it is evident that the only primary contaminant detected in sigmficant concentrations or on a regular basis is TCA. In tum, the only other contaminants detected in significant concentrations are DCA and DCE, the degradation products ofthe TCA. However, DCE levels m the groundwater are generaUy an order of magnitude higher than the DCA levels, which supports the theory that DCE is the major degradation product of TCA in Arizona soUs and/or groundwater.

Figure 1 is an iUustration of the TCA/DCE concentration ratios obtained for soU gas probes around Pit 1. As would be expected, TCA/DCE ratios for the southeast and southwest transects decrease quickly with increased distance from Pit I. Although the TCA/DCE soU gas ratio decreases from soU gas probes N-l to N-3, at soU gas probe N-4 the ratio increases close to that observed at Pit 1. This not only indicates that the HVOCs in soU gases at N-l and N-4 are from the same source (Pit 1), it also indicates that the HVOC soU vapors in the N-4 area are being emitted from soUs that have been directly contaminated by Uquids disposed into Pit I that migrated to the north.

In closing, it is the opinion of ADEQ that there is a good possibiUty that a substantial portion of the Uquid wastes disposed into Pit 1 have migrated to the north of Pit I along the dip of the buried basalt flow. This potential migration is indicated by the elevated HVOC soU gas concentration detected in a soU gas probe to the north of the site. Therefore, for the purposes of eventuaUy remediating the site, it is the opinion of ADEQ that additional soU gas and groundwater investigations are necessary to the north of the site. An expanded soil gas investigation ofthis area would enable the delineation ofthe HVOC vadose zone contamination in this area and greatly assist in the subsequent siting of monitor weUs and the eventual groundwater remediation at the site.

C5. Page 14. paragraph I. This section ofthe report questions whether wastes actuaUy exist in the Special Pit Areas. Given the fact that HVOC wastes were manifested for these areas, and considering the HVOC groundwater contamination detected in MW-IUA (which monitors the Special Pit Areas) along with the recent soU gas survey results, it appears that wastes do exist in the vadose zone in the Special Pit Areas. Therefore, the phrase "(if wastes exist)" should be removed from the last sentence of this paragraph.

C6. Page 15. paragraph 1. As was previously mentioned in Comment A5 of this memo, the laboratory resuUs of the July 30, 1991 sampling of the vadose zone borings are incorrect. Therefore, the CRA explanations regardmg the discrepancy between these laboratory results and the results from the Tracer Research soU gas survey are

15

meaningless. In addition. Table 3.2 of the SoU Gas Survey report should either be revised or removed from the report since it includes this incorrect data.

C7. Page 15 and 16. The "Conclusions" section of the soU gas report states that there are 3 distinctive areas of soU gas contamination in the hazardous waste area. 1 beUeve the key phrase here is the reference to the soU gas contamination within the hazardous waste area itself. Obviously a great deal of soU gas contamination exists to the north of the site in the N-4 area. Reference to the existence of the HVOC soU gas contamination in the N-4 area north of Pit I should be included in the "Conclusions" section of the SoU Gas Survey report.

C8. Page 17. second paragraph. 1 once again object to the use of the word "significant" in this context. It appears that the authors are somehow trying to make the other sources of HVOC soU gas at the site appear unimportant when compared to the soU contamination encountered at Pit I. WhUe it is important to estimate the magnitude of soU contamination in each area with elevated soU gas concentrations, simply identifying additional discrete areas of soU contamination with the use of soU gas technology is also significant.

D. ADEO Comments on the "Technical Screening Memorandum." CRA (10-28-91)

The above referenced document has undergone hydrologic review. In general, the report is weU-written and informative, but as in recent documents submitted to ADEQ and EPA by the consultants for HSC, there are several areas of the report with which there are problems. Comments are presented below:

Dl. Page 3. last paragraph. As is discussed in Comment AI of this memo, it does not appear that the laboratory measurements of vertical hydrauUc conductivity of on-site soUs are vaUd.

D2. Page 4. paragraph 1. Recent evaluation of the Unit A water level elevations for the onsite monitor weUs indicates that the groundwater flow direction in the Pit 1 area is actuaUy to the southeast, not the south/southwest. This variation in flow dUection has serious impUcations on the design of any remedial groundwater extraction system proposed for the site and wiU be discussed later in this memo.

D3. Page 4. paragraph 2. The report states that as a resuU of soU boring sampUng and analysis it was revealed that lateral migration of contamination had occurred in the vadose zone. However, other than the soU borings instaUed directly into the waste disposal pits I am not aware of any soU samples from onsite borings which detected significant concentrations of HVOCs.

D4. Section 2 Tables. Tables 2.3, 2.4, 2.5, 2.6, 2.8, and 2.9 present different types of environmental media quaUty data in the form of an arithmetic mean of the associated data. For instance, Table 2.6 gives "representative" concentrations of HVOC in soUs beneath the pits at the Hassayampa site. Representative concentrations were calculated

16

UtUizing the HVOC soU quaUty data from aU of the samples obtained from a particular boring. WhUe this may not necessarily be an inappropriate way to present the data for a feasibiUty study, I beUeve that simply supplying the mean of the data in these tables does not give the reader a good "representation" of the maximum concentration of HVOCs beneath the pits and I think a column specifying maximum HVOC levels, or the range of levels, in each of these tables is appropriate.

D5. Page 20. paragraph 2. The report states that as a resuU of the recent soU gas survey at the site, three distinct and significant areas of soU gas contamination exist in the hazardous waste area. As I have explained earUer in Part C of this memo, I beUeve the area to the north of Pit 1 with the greatly elevated soil gas levels is also a significant area of subsurface HVOC contamination.

D6. Page 27. paragraph I. This paragraph Usts the HVOCs which have been detected in monitor weUs at the landfill in concentrations exceeding their respective MCLs. Upon review of the HVOC groundwater quaUty data for the onsite monitor weUs it appears that there are a few mistakes in this paragraph. Firstly, PCE has never been detected in MW-IUA. Second, both PCE and TCE have been detected in MW-6UA in excess of their respective MCLs.

D7. Page 31. paragraph I. This portion of the report states that the data coUected during the soU gas survey indicated that HVOCs were present in elevated concentrations in two areas of the landfiU, the immediate area of the Pit 1 and in the central portion of the Special Pit Areas. However, page 20 ofthis same report states that "three distinct and significant areas of soU gas contamination exist in the hazardous waste area". It appears that when it comes time to start discussing remediation at the site, the once "significant" area of elevated soU gas HVOCs in the southwest portion of the Special Pit areas suddenly became much less significant to the authors.

As it has been explained previously in this memo, it is the opinion of ADEQ that aU of the areas at and near the site which detected greatly elevated HVOC levels during the soU gas survey are significant. As such, in addition to the immediate Pit 1 area and the center portion of the Special Pit Areas, I beUeve that remedial activities should also be conducted in the southwestem portion of the Special Pit Areas which also detected elevated HVOC soU gas concentrations. In addition, as has been stated previously in this memo, it is the opmion of ADEQ that the area of elevated HVOC soU gas levels to the north of Pit 1 is indeed significant and warrants further site investigation.

D8. Table 4-2. Table 4-2 of the report indicates that for Freon 113 and dichloromethane (DCM), the authors had some difficulty in locating K^ values and water solubUities for these chemicals. The organic carbon partition coefficients for Freon 113 and DCM are 389 ml/g and 8.8 ml/g, respectively. Also, the solubUities ofthese HVOCs are 170 mg/l for Freon 113 and 13,000 mg/l for DCM. In addition. Table 4.4 of the report is missing vapor pressure and Henry's Law Constant values for Freon 113. The vapor pressure and Henry's Law Constant values for Freon 113 are 270 mm Hg and 0.333 atm mVmol, respectively. AU of these chemical properties are from a paper on the behavior of organic chemicals in the subsurface by Chuck Graf of ADEQ. This paper

17

is included as Attachment E.

D9. Page 62. paragraph 2. The word "infeasible" in this paragraph must be a typographical error and should be changed to "feasible".

DIO. Page A-4. paragraph I. The report states that the groundwater ARARs for the site should be the state and federal drinking water standards. However, some consideration should be given to using the ADEQ drinking water HBGLs which are generaUy more stringent than the state and federal drinking water standards. It should be noted that in this report the authors have impUed that the cleanup standards for the Pit 1 waste could be the ADEQ ingestion HBGLs for soUs.

Dll . Page A-7. paragraph 1. This section of the Appendix to the report describes a southerly groundwater flow direction for the site. SpecificaUy, the report states "contamination has not been detected in MW-2UA and MW-3UA indicating that the contamination detected in MW-6UA has not migrated south of the hazardous waste area". However, during my recent involvement with the legal case between the HSC and the State, it became apparent that the groundwater flow dUection of the Unit "A" Aquifer in the Pit 1/MW-6UA area is not to the south/southwest or even to the south, but instead the flow direction is somewhat to the southeast. This phenomenon was first brought to my attention by an expert witness for the state. Dr. Gary Walter. In analyzing the relationship between HVOC groundwater contamination in monitor weUs HS-l and MW-6UA, Dr. Walter generated groundwater elevation contour maps for the site for several different dates of groundwater elevation measurement rounds using a computer contouring package. Surprisingly enough, some of the computer generated contour maps indicated a south/southeasterly groundwater flow dUection in the Pit 1 area. Intrigued by the possibiUties and impUcations of such a flow dUection, I then hand-generated groundwater elevation contour maps for the five most recent rounds of water level measurements for monitor weUs in the hazardous waste area. These contour maps are presented as Figures 2 through 6. These figures depict groundwater elevation contours in 0.5 foot intervals and contain arrows indicating the groundwater flow directions in the Unit "A" Aquifer m various parts ofthe hazardous waste section ofthe landfiU. As can be seen m these figures, the groundwater flow dUection m the Pit 1/MW-6UA area appears to be to the south/southeast. In addition. Figures 2 through 6 agree faUly weU with the groundwater elevation contour map given in Figure 16 of the RI/FS. The exception to the simUarities ends, however, when contouring near the MW-6UA area. In my opUUon, there is no vaUd reason for EMA contouring the eastem end of the 843-foot contour in this manner (to make it appear that there as a southerly groundwater flow dUection in the MW-6UA area).

A southeasterly groundwater flow dUection m the vicinity of Pit 1 also explams the occurrence of HVOC groundwater contamination m the northeastem portion of the hazardous waste landfiU. HistoricaUy speaking, until very recently the only monitor weUs which have detected elevated concentrations of HVOCs m this area ofthe landfiU are HS-l and MW-6UA. These weUs have detected the highest levels of HVOCs ever measured at the site, yet given the southerly groundwater flow dUection depicted m the RI/FS there are no potential HVOC sources upgradient of these two weUs. In addition,

18

if the groundwater flow direction in the Pit 1 area was tmly to the south, the greatly elevated concentrations of HVOCs would be present in the MW-4UA, and yet only very low levels of HVOCs have just recently begun to show up in MW-4UA. The new discovery of a southeasterly groundwater flow dUection in the Pit 1 area readUy explains why HVOCs have been consistently detected in high concentrations in samples from HS-l and MW-6UA, both ofwhich were located to the southeast ofthe HVOC disposal point, Pit 1.

At this point it could be argued that both HS-1 and MW-6UA are situated too far in a southeasterly dUection (when compared to the proposed groundwater flow dUection in the Pit 1 area having a southerly component to the southeasterly flow) to be considered dUectly downgradient of Pit I. However, this would only be tme if the Uquid HVOCs percolated only in a downward dUection from PU 1 to the water table without any lateral movement. As has been repeated many times in this memo, it is the opinion of ADEQ that a great amount of the Uquid waste disposed into Pit I migrated to the north along with the dip of a buried basalt flow which is known to be somewhat fractured. This theory is somewhat supported by the recent soU gas survey north of Pit 1. As the Uquids migrated to the north of Pit 1, some ofthe Uquid waste would reach groundwater by way of percolation through the fractured basalt. The further north of Pit I the Uquid HVOCs entered the aquifer, the easier it would become to detect the HVOCs in both HS-1 and MW-6UA given the southeasterly groundwater flow dUection shown in Figures 2 through 6.

At this time it is unknown how far to the north of Pit 1 the slug of liquid wastes above the basalt has actuaUy migrated. However, it does appear that the data from the recent soU gas survey around Pit 1 may have detected soU vapor emanating from the "taU end" ofthe slug of vadose zone contaminants. As such, there could be a great deal of Uquid waste from Pit 1 stiU moving over the basalt to the north of soU gas probe N-4 which is continuing to contaminate the Unit "A" Aquifer. Given such a scenario in conjunction wUh the southeasterly groundwater flow dUection in the northeast portion of the landfUl, it appears the vast majority of the HVOC groundwater contamination produced as a result of Uquid waste disposal into Pit 1 is probably in the aquifer to the north and east of the site. In such a case, only MW-6UA and MW-6UB are properly situated to detect this HVOC groundwater contamination, and the extent of groundwater contamination is therefore unknown. Thus, none of the groundwater extraction design options in Appendix A.A of the report are appropriate since aU ofthe proposed options were designed upon the assumptions of a southerly groundwater flow dUection and that aU of the HVOC groundwater contamination associated with Pit 1 was presently within the boundaries of the hazardous waste area. As stated before, addhional soU gas surveys and groundwater investigations are necessary north and east of the landfUl before any groundwater remedial devices can be effectively sited.

D12. Page A-18. last paragraph. In addition to the groundwater sampUng schedule for all of the Unit A monitor weUs, select Unk B monitor weUs should also be included on the sampUng Ust. These weUs should mclude MW-IUB, MW-4UB, MW-6UB, and possibly some of the Unit B monitor weUs that may be mstaUed as a result of addhional investigations at the landfiU.

a:hassrev.ml#mll 1 9

Table 1

1 Typical K Values for Loann Soils |

Soil Type

[(Sand)

(Sandy Loann)

(Loamy Sand)

(Loam)

(Silt Loam)

(Clay Loam)

Soil Name

Fine Sand |

AASHO - A 4

Rillito Grandly

Cecil

Rock Valley

Gueloh

Adelanto

Pachappa

Grenville

Caribou . .

Grenville

Columbia

Glendale

Panoche :

Particle Dist %

65.1 Sand/34.9 Silt & Clay

62 /20 /18

75/13/12

•:...7MZ/5 : „

42/19 /39

50/24/26

77/18/5

all sizes

•^;:;:;:::::s;::::::--I:^y.l!.-

y . . . . . . . . . . . ^ . . . . : - : ^ : . . . . — ^ • • ' • ' : ' • • : ' ' • '

. . .

::iiiiii:iAAXi::T}--^..:;.

—

39/30/31

,, . 24 /30 /46

K, (cm/s)

2 . 1 8 x 1 0 - ^

9 x 1 0 - ^

4.5 X 10"^

.....2.75..x..1..Q:i..

1 . 2 x 1 0 ^

3 x 10-5

9.94 X 10-5

8.76 X 10"^

5 x 1 0 - 5

8.7 X 10 5

.3...7..X..1.0.1

4 x 10- '

4.9 X 10 5

1.51 x l O - ^

5.5 X 10-«

1.13 X 10-^

5 X 10-"

1.79 X 10""

1.5 X 10-^

2 . 9 x 1 0 - *

2 . 4 x 1 0 - ' '

2.6 X 10"^

7,3...K...ml

1.17 X 10-"

\ ' 2.1 X 10-"

1 . 5 x 1 0 - ^

4 :4x10-= '

7.3 X 10-'

3 x 1 0 "

4 . 4 x 1 0 - "

2 . 4 x 1 0 ^

Comments

..(Q-.a..c.rnJ

(33-89 cm)

(1931)

(37.1 cm)

•(62.5)

(plot 18)

(91.4 cm)

{152 cm)

(.182 cm)

Above data is taken from "Unsaturated Flow Properties Catalog" Compiled by the Desert 1 Research Institute in Reno, Nevada.

Table 2

Laboratory-Calculated Vertical Hydraulic Conductivity Values for On-Site Soils

Soil Boring Sample

SB-14 (24')

SB-5 (49')

SB-3 (49')

SB-9 (49')

SB-14 (40')

SB-8 (49')

SB-11 (49')

88-10(49')

SB-12 (49')

SB-7 (60')

Particle Size Dist %

70/30 (30% Silt & Clay)

10/70/20

15/65/20

0/70/30

20/80 (80% Silt & clay)

777f7AciiMimx^77/

10/55/35

-• ;. .0/60/40

0/65/35

.0/30/70

Lab K (cm/s)

2.01 X 10-5

1.81 X 10-5

2.58 X 1--5

< 1 X 10-5 :

< 1 X 10-5

2.4 X 10-'

3.4 X 10-^

1.8 X 10-'

2.75 X 10-'

1.86 X 10-'

2.3 X 1 0 '

2 . 4 3 x 1 0 - 5

;.;2.49xJ!..a5

2 ; 0 3 x 10-5

4.8 X 10-5

3.8 X 10-5

2.8 X 10-5

1.9 X 105

5.4 X 1 0 '

2:4 X 1 0 '

2.1 X 1 0 '

6.4 X 1 0 '

3.5 X 1 0 '

2.4 X 1 0 '

4.5 x: 10-5.

4.5 X 10-5

3.7 >( 10-5

Lith Desc.

Sandy Loam

Silt Loam..

Silt Loam

Silt Loam

Silt/Clay Loam

Silt Loam

Silt Loam

Silt Loam ;.

Silt Loam

Clay Loam

Table 3

Comparison of HVOC Levels and Assemblages in Different Media in the Pit 1 /HS-1 Area

Soil HVOCs" (05/11/91)

SB-19

1.

2.

3.

4 .

5.

6.

7.

8.

9-

10.

1 1 .

1,1,1-TCA

Freon-113

DCM

1,1-DCE

PCE

TCE

Toluene

Xylene

MEK

1,2-DCP

1,1-DCA

18,300

10,500

840

660

480

440

400

350

292

160

47

Soil Vapor HVOCs" (06/04/91)

VB-3

1,1-DCE 3,400

Freon-113 980

DCM 540

1,1,1-TCA 500

Freon-11 218

1,1-DCA 186

TCE 182

PCE 85

1,2-DCP 49

Vinyl Chloride ; 25

1,2-DCA 1 22

Groundwater HVOCs" (05/26/88)

HS-1

1,1-DCE 910

Freon-113 600

1,1,1-TCA 170

TCE 52

PCE ; 23

1,1-DCA \ 21

Freon-11 20

' All soil concentrations of HVOCs are in mg/kg.

^ All soil vapor and groundwater concentrations of HVOCs are in ug/l .

Table 4

Total Volumes of Liquid HVOCs Disposed Into Pit 1

HVOC

1.

2.

3.

4.

5.

6.

7.

Freon-113

DCM

1,1,1-TCA

1,2-Dichloropropane

PCE

Freon-11

Misc. HVOCs

Volume Disposed (gallons)

-16 ,000

5,400

5,333

1,350

250

66

7,950

Table 5

Phase Distribution of HVOCs at VB-3 Calculated f rom Soil Gas Concentrations

HVOC

1,1-DCE

TCE

Freon-113

1,1,1 -TCA

PCE

Freon-11

1,1-DCA

1,2-DCP

DCM

1,2-DCA

Vinyl Chloride

K..

65.0

126.0

359 .0

152.0

364 .0

159.0

30 .0

51.0

8.8

14.0

2.5

KH

0.882

0.420

21 .500

0 .770

0.340

4 .620

0 .180

0.096

0.070

0.038

99 .000

C„ (ug/l)

3 ,400

182

520

390

85

218

186

49

460

13

25

C„a (ug/l)

3 ,854 .00

430 .00

242 .00

506 .00

250 .00

47 .00

1,033.00

510 .00

6 ,571 .00

342 .00

0.25

C„o (ug/l)

910

52

600

170

23

20

21

ND

ND

ND

ND

f. {%)

26.21

8.53

63 .80

12.50

2.70

67 .37

12.00

4 .72

10.10

4.55

99 .60

f . (%)

9.90

6.79

0 .90

5.40

2 .60

1.94

22 .00

15.72

48 .08

3 9 . 9 4 .

0 .34

f, (%)

63.89

84.70

35 .30

82.10

94.70

30.68

66.00

79.56

41.83

55.51

0.06

f„^ = fraction of organic carbon in soil (assumed to be 0 . 0 0 7 1 , equivalent to 0.71 %)

p = soil particle density (assumed to be 2.33 g/cm')

Koc = Organic Carbon Partition Coefficient

KH = Henry's Law Constant

C, = observed concentration of HVOC in VB-3 soil gas

C „ . = calculated concentration of HVOC at water table surface in HS-1

C„<, = observed groundwater concentration of HVOC in HS-1

f, = calculated mass fraction of HVOC in gaseous phase

f „ = calculated mass fraction of HVOC dissolved in pore water

f, = calculated mass fraction of HVOC sorbed by soil

Table 6

HVOC Soil Half-Lives

HVOC

DCM

PCE

TCFM

TCE

TCA

1,2-DCP

1,1-DCA

1,1-DCE

Freon-113

High Estimate of Half-Life'

4 weeks

1 year

1 year

1 year

39 weeks

3.5 years

22 weeks

6 months

1 year

Low Estimate of Half-Life'

7 days

6 months

6 months

6 months

20 weeks

167 days

32 days

4 weeks

6 months

• Scientific judgement based upon aqueous aerobic biodegradation half-life.

From: Handbook of Environmental Degradation Rates (Howard, 1991).

Table 7

HVOC Soil Gas Concentrations (ug/l) Hassayampa Landfill (07/11/91)

Soil Gas Probe #

SE-3

SE-2

SE-1

PT-1

N-1

N-2

N-3

N-4

B-1

TCA

250

770

2600

54,000

2,700

2,000

430

2,200

300

F-11

33

30

26

210

8

15

31

160

5

1,1-DCE

32

35

42

< 710

7

19

31

< 70

16

1,1-DCA

< 7

< 7

41

< 6,900

74

2

< 7

< 640

< 7

F-113

3,300

3,400

3,600

57,000

1,300

2,300

3,500

7,200

355

TCE

25

6

80

< 570

75

35

21

540

84

PCE

14

20

77

3,600

83

32

14

1,700

48

a:hassrev.ml#ml1

^ i c , ^ . . \ . - X c f ^ l - ^ C i . ^ c . V , . ^ ^ , , S . A t c . ' V ^ e . S . ^ . e ^ ^ V . KJ^cK. P ' . - V l ,

2%..i

p / r A

A S B - 4 - ^ — — ^

I A S B - 7

'? . -^1

J<28

0 .37

SPEC/AL P IT AREAS

SPECIAL PITS AREA

/980

GMW-7UA

CD n ( ) H S - I

P / r 2

MW-4UB ® O M W - 4 U A

A S B - I O A S B - 9

_ C^^^PJJ ^^

14

- 2

P/T 3o

SECTION 3, TOWNSHIP I SOUTH, RANGE 5 WEST

E X P L A N A T I O N

SAy.i . . . - .^ ' , r , r v ^ . . . : l . . ,,A-.,- ? \ . . . , . . . . , .C-.,.^•^.•..•.r . - L U : V ' / V < f. v A , , A . - ^ . - ^ i O )

SECTION 3 , TOWNSHIP I SOUTH, RANGE 5 WEST

E X P L A N A T I O N

SECTION 3, TOWNSHIP I SOUTH, RANGE 5 WEST

E X P L A N A T I O N

SECTION 3 , TOWNSHIP I SOUTH, RANGE 5 WEST

E X P L A N A T I O N

^ . c y . . .1. .x... . tr \ , . . . , ,v .^ r^-.,.v+ Wv^.v'A'^ C\ i . y ( ' 4 - \ ^ - a i ^

SECTION 3, TOWNSHIP I SOUTH, RANGE 5 WEST

E X P L A N A T I O N

\ ^ ^ . . ( D - G-<-r,v -sA.'jjy N'.r ^ \ ? v c, TI u . „ , C v^VV,. . . c •:> - V.V'^ ."r" K'~ .X, 7 ' ^^'77,1 m

Attachment A

Investigation of Failure Mechanisms and Migration of Organic Chemicals at Wilsonville, Illinois

by B.L. Herzog, R.A. Griffin, C.J. Stohr, L R . Follmer, W.J. Morse, and W.J. Su

Abstract Ground water contamination was discovered in 1981 in a monitoring well al the Earthiine disposal faciiity near

Wilsonville, Illinois. Organic chemicals had migrated at a rate 100 to IOOO times greater than predicted when the site received its permit to operate in 1978. Postulated failure mechanisms included migration through previously unmapped permeable zones, subsidence of an underground mine, organic-chemical and clay-mineral interactions, acid-mine drainage and clay interactions, trench-cover settlement, and erosion.

In this investigation, the Illinois State Geological Survey found the primary reason for the rapid migration; the presence of previously undetermined fractures and joints in glacial till. The inaccurate predictions of hydraulic conductivity were based on laboratory-determined values that did not adequately measure the effects of fractures and joints on the transit time calculations. Field-measured hydraulic conductivity values were generally 10 to 1000 tinies greater thari their laboratory-measured counterparts, thus largely accounting for the discrepancy between predicted and actual migration rates in the transit time calculations. The problem was compounded, however, by the burial of liquid wastes and by trench covers that allowed excess surface runoff to enter the trenches. Organic-chemical and clay-mineral interactions may also have exacerbated the problem in areas where liquid organic wastes were buried.

Introduction Earthiine Corp., a subsidiary of SCA Services Inc.

began operating a 130-acre (53 hectare) landfill in south­west Illinois near Wilsonville on November 15, 1976. It was a trench-and-fill operation that relied on a clayey till deposit for natural attenuation of leachate. The material, Vandalia Till, is native to the site. Supplementing the till in at least one of the 26 trenches was a liner of compacted native clay soil.

The Illinois Environmental Protection Agency (IEPA) had granted the company a permit to dispose of industrial and hazardous wastes at the site. Several months after the landfill began operation, the citizens of nearby Wilsonville objected to the disposal of hazardous wastes so close to their town. To stop the disposal of wastes and have wastes removed from the site, the citizens group and the Illinois Attorney General filed suit. A long court battle ensued during which Earthiine continued to bury wastes.

In 1981, the Illinois Supreme Court affirmed a 1978 trial court ruling that the hazardous wastes be exhumed from the 26 trenches (each approximately 10 to 20 feet [3 to 6m] deep, 50 to IOO feet [15 to 30m] wide, and 175 to 400 feet [53 to 122m] long) and removed from the site. SCA Services announced in March 1982, after discovering that contaminants had migrated from the disposal trenches, that they were dropping further appeals and would comply with the court order. Preparation for

82 Spring 1989 GWMR

exhumation began in the summer of 1982. The actual process of exhumation and removal, which began on September 7, 1982, took more than four years.

In January 1982, the IEPA confirmed that organic pollutants had migrated up to 50 feet (15m) in a three-year period. These migration rates were 100 to 1000 times faster than predicted by engineers who were consulted when the site was planned. The levels of volatile organic priority pollutants detected were reported by Johnson et al. (1983). This discovery, which was a separate issue from the court proceedings and exhumation order made before contamination was discovered, raised two obvious ques­tions: (1) why were these organic compounds migrating faster than predicted, and (2) what were the implications for land disposal of similar wastes at other sites?

A research project was designed by the Illinois State Geological Survey (ISGS) for the U.S. Environmental Protection Agency to provide answers to these and many other questions regarding the efficacy of land disposal of hazardoiis wastes—particularly organic liquids. The scope of work included sludies of several aspects of the site; • Geological characterization—detailed descriptions of

geologic materials and geomorphology. • Hydrogeological characterization—determination of

ground water flow directions and gradients, comparison of field and laboratory measurements of hydraulic conductivity, and evaluation of the significance of

SjaMMBEmi^i^^ias^gj^EE

iVacuiic lluw. • Ciiound water quality—clcicrnimaiioii ol ihc disiriini-

tion of organic chemicals across tiic site, and cslab-' lislimcni ofa sampling nicthotiology i'or vt)latilc organic

compounds in (inc-graincd scdimcius. • Gob pile effects—investigation of theeffccts of acidity

and high inorganic salt content of leachate from an adjacent coal refuse pile on the trench materials.

• Trench cover sludies—observation of the condition of trench covers and the degree of differential settlement, determination of erodibility of cover malerials, and delermination of the effects of mine subsidence.

• Condition of drums and wastes—photographic doc­umentation of the effects of leachate on drums. Some individual items in the scope of work have been

presented in previous reports. This paper summarizes all the work performed to determine the causes of sile failure, and discusses the relative impact of each postulated failure mechanism.

Site Characterization Geology

An extensive geologic investigation was conducted lo place the sile in the regional geologic framework and to collect sufficienl baseline data for extrapolating the results from this site investigation to other sites. Four principal approaches were laken; (1) examination of all existing data and relaled information; (2) investigation of outcrops and exposures on and near the site (within about 5 miles); (3) study of the trenches, including the walls of backhoe pits located away from the trenches on and around the site; and (4) study of drillhole samples collected on-site. Samples were collected from each well nest (Figure I).

Site geology is shown in cross section in Figure 2, along the line shown in Figure 1. The first unit encountered at the land surface is the Peoria Loess, a windblown sill that has been weathered into a modern soil wilh the' texture of clayey sill. Il is brown lo gray and ranges from 1 to 4 feet (.3 to 1.2m) in thickness across the sile. Beneath the Peoria Loess is the Roxana Silt, throughout which the Farmdale Soil formed. This silt, which ranges from 1 to 3 feel (.3 to .9m) thick, has a higher sand content than the Peoria Loess.

Underlying the silly deposits is the Vandalia Till of Illinoian Age; it can be divided inlo four zones; (1) a brown, stiff, upper weathered zone; (2) a sofl weathered zone; (3) a brown, brittle, lower weathered zone; and (4) a gray unweathered zone. A distinct buried soil, the Sangamon Soil, often referred to as a paleosol, forms the main body of Zone I. Because of leaching, developmenl of the Sangamon Soil in Zone 1 has infiuenced the character of Zones 2 and 3. The upper three zones of the till range from 8 lo 12 feet (2.4 lo 3.7m) in thickness al the site. Occasional bedding planes and inclusions with con­trasting textures occur. Although discontinuous sand lenses are common throughout the Vandalia Till, they are more common at the base of Zone 4 and constitute a large part of Zone 2. Joints and fractures are numerous in the Sangamon Soil and decrease in frequency toward the base of Zone 3. The main part of Zone 4 has very few joints or fractures; it is uniform, stiff, and somewhat

v,aiei taDie elevation

LifW ol cross secl-On

\ AppfO'iniale tioundary ot Ounal areas

riiimeiiit fXaies

i T.edov.^ o* luimng oo'oi

Iefept>ooe pole

Panel points

Figure 1. Site map showing elevation of potentiometric surface (feet) in April 1984; locations of trenches, wells and cross section. Thc first letter in each well or piezometer number designates the nest or profile. For wells within a profile, the second number desig­nates the nest within the profile. The last item denotes well depth: S for shaUow, M for medium, and D for deep. For piezometers, the second item is either P for piezometers or A for angled hole. The last ilem in the identification is a number signifying the order in which the piezometers were installed.

Figure 2. Cross section from Profile V through Trench Area B to the gob pile showing distribution of trichloroethylene (//g/L) in ground water.

plastic. The thickness of Zone 4 typically varies 5 to 20 feet (1.5 to 6m).

Beneath the Vandalia Till are older tills ofihe Banner Formation. The thickness of this formation at the site is

Spring 1989 GWMR 83

w l A B l . E 2

S u n u n a r y of Labora to ry Test Kesuits ( from Herzog and M o r s e 1986)

Material

Hydraulic Conductivity

from Falling-Head Tests (cm s')*

Hydraulic Conductivity from Triaxial Tests (cm s') ' '

Hydraulic Conductivity from ISGS Miniature Permeameter Tests

Undisturbed Sample (cm s"')

Recompacted Sample (cm s ' )

Peoria Loess

Roxana Silt

Vandalia Till Ablation Zone, Sangamon B, Horizon (Zone I)

Vandalia Till Lower Ablation Zone (Zone 2)

Weathered Basal Vandalia Till (Zone 3)

Unaltered Basal Vandalia Till (Zone 4)

1.1 X 10^ 7.5 X l0-« to 2.2 X 10-

1.6 X 10'

1.2 X 10- to 7.4 X 10-8

1.3 X 10-«to 2.7 X 10-« 4.1 X [0 -9

5.8 X 10- to 1.4 X 10^

3.3 X 10- to 5.1 X 10''

3.3 X 10-8 to 3.6 X 10-8

7.0 X 10-9 to 8.6 X 10-9

6.0 X lO-'O to 1.2 X 10-9

2.6 X 10-8 to 3.5 X 10-8

1.4 X 10-8 to 4.7 X 10-8

8.9 X 10-9 to 1.1 X 10-8

^ Tests conducted by John Maihes Associates (1976). ^ Tests conducted by Geoengineering Inc. (1982).

for waler qualily samples. During well developmenl, hydraulic conductivity at the monitoring wells was mea­sured using the recovery test as described by Todd (1980). Recovery test results from the monitoring wells were generally higher than slug lest results (Table 1), probably due to the location of monitoring well screens in more permeable zones.

Monitoring wells and piezometers installed by the ISGS and monitoring wells installed by the site owner were used to measure the potentiometric surface in each zone. The shallow potentiometric surface is shown in Figure 1. Potentiometric surfaces of deeper zones showed the same general pattern of flow directions as did the shallow potentiometric surface, but with lower water levels, indicating that the zones are connected and the vertical gradient is downward. Figure 1 also shows the influence ofthe gob pile on ground water flow at the site; ground waier flows radially outward from the gob pile.

Geochemistry Sampling Procedures

Because no protocol existed for collecting volatile organic compound samples from wells finished in fine­grained sediments, it was necessary to establish procedures for this project. Ground water was sampled in a time series from wells VID, V2D, and V3D, and analyzed for several volatile organic compounds. These samples were retrieved using dedicated point-source bailers equipped with bottom-emptying devices. Samples were collected in VOA vials and analyzed by gas chromatograph in the laboratory.

These tests, described fully in Griffin et al. (1985), showed dramatic changes in concentrations with time

after the wells were purged. Most volatile organic com­pounds were found at their lowest concentration before purging and their greatest concentration after two to 24 hours of recharge. Because ofthe result ofthese tests, water samples were collected two to 24 hours afler each well was purged. Contaminant Distribution

Soil samples from the deepest well in each nest and liquid samples from each monitonng well were analyzed for organic compounds. The soil was sampled during drilling of boreholes for placement of the monitoring wells. As reported by Griffin et al. (1984), the highest concentrations of organic chemicals were found over most of the site in the upper three zones of the Vandalia Till. Along profile V, the zone of highest contamination was in Zone 3 of the till. These zones had higher hydraulic conductivities than the underlying zones (Table 1). The overlying Roxana Silt and Peoria Loess are generally unsaturated. Figure 4 shows a typical pattern, using tri­chloroethylene as an example.

Monitoring well results indicated that organic com­pounds were also found in Zone 4, the unweathered Vandalia Till. The highest levels of contamination were found on the southwest corner of the site at Nests Wl and B in Zone 4. In these wells, endrin and dieldrin were present at concentrations greater than 1 percent and phase separation occurred. High levels ofcontamination in the unalterered Vandalia suggested that contaminants were moving downward through permeable materials or sand lenses, interconnected joints, or clay materials. The clays may have been altered by chemical interactions with a free liquid organic phase, thereby increasing the hydraulic conductivity in that area of the sile.

Spring 1989 GWMR 85

M

I

Figure 3. Pinhole tests of non-erosive materials on left, and highly (.Tosive materials on right. The 1mm pinhole for each sample was (he same diameter as a paper clip. The hole in soil sample at extreme right was formed in two minuies al a 2-inch (SOmm) clcva-lion head (from Stohr et al. 1985).

line Subsidence Subsidence ofan underground coal mine, located

approximaiely 300 feel (90m) below the site, was consid­ered for its possible effects on trench stability. Soil frac­turing associated with trench instability might increase the rates of chemical migration. Stereoscopic examination of prelandfill aerial photographs indicated only one depression adjacent to the landfill site. Tree canopies obscured the ground in some areas; however, no pattern of tilting trees was observed. A ground reconnaissance survey prior to exhumation activities found no indication of mine subsidence at the landfill site.

Monthly precision (third order) vertical surveys of deep settlement probes set al or below the bottom ofthe burial trenches were performed to measure near-surface movements that would indicate possible instability from collapse of underground mines. Statistical analysis by Pearson correlation coefficients showed that mosl ofthe deep probes correlated highly with each other. The movement of one anomalous probe was believed to have been caused by slope instability due to its proximity lo a steep slope. The conclusion was that no recenl mine subsidence had occurred at the site, and therefore, that mine subsidence had no effect on the rate of chemical migration.

Condition of Drums and Wastes The condition of the drums and other wastes was

documented photographically as they were removed from the trenches. The data were used lo help interpret the effecls of leachate on the drums and earth materials, the relative strength of the leachate, and the life expectancy of the drums. The condition of drums exhumed was highly variable, ranging from well-preserved and intact to highly degraded. During the excavation of Trench 24, the paint was found intact on some drums buried three years earlier and silling in an unidentified orange-brown liquid. The condition of these drums and some intact sacks of herbicide suggests that the longevity of waste conlainers in a landfill depends on their original condition, handling.

ami coniciits as well as (he composition of fluids sur-roiiiKJiiiU the coiiiainci".

Sicico jiliolography was used to record drum orien-laiion using a 2-l'oot (().6in) aluminuni cube assembled on-site I or oriental ion ofstcreo photography (Stohr 1983). I'lioiogia|5hic observations ol drums and other wasle containers were made periodically during the landfill exhumation. Results from a study of drum orientation in liench 24 indicated that inclination varied fromO° lo 32° from the vertical as measured from a photograph exter­nally oriented by means of the aluminum cube. Void space belween the drums w;is calculated lo range from 17 lo 38 percent of ihc trench volume.

Conclusions The primary reason that the Wilsonville industrial

waste-disposal site failed lo perform as predicted was that the earth materials were more hydraulically conductive than early laboratoiy tests had indicated. The original predictions, which were based on laboratory-determined values of hydraulic conductivity, were made in accordance wilh the accepied practices ofthe time. These laboratory values were reproduced in this study. However, the field-determined values of horizontal hydraulic conductivity were 10 to 1000 times greater than laboratory-determined values. In addition, the original investigation did not recognize the importance of vertical joints and uncon­nected sand lenses. These joints caused the vertical hydraulic conductivity lo be up to 10 limes greater than the horizontal value. Joints and sand lenses also presented preferential pathways for both horizontal and vertical movement of chemicals al the site.

Rales of chemical migration may also have been enhanced by differential settlement. Highly erodable earth materials allowed freely draining depressions to develop in the trench covers permitting surface water to enter the trenches, interact with the waste, and increase local ground water gradients out of the irenches.

Reactions of organic chemicals wilh clay minerals may also have been a contributing factor. The highest levels ofcontamination, found at Nest B and Profile W, were immediately downgradient of an area where large quantities of free liquid wastes (no drums) were buried. Interactions between these immiscible chemicals and the clay minei-als may have opened joints, allowing an increase in downward flow. This mechanism does not appear to be significanl elsewhere on the site where wastes were buried in drums.

The site was not affected by acid mine drainage or subsidence associaied with the abandoned coal mine; however, the coal refuse pile created a ground water mound that affected the local flow direction and gradients of the shallow ground waler flow system.

References Cooper, H.H., J.D. Brcdehoefl, and I.S. Papadopulos.

1967. Response of a finite-diameter well to an instan­taneous charge of water. Waler Resources Research, V. 3, pp. 263-269.

Follmer, L.R. 1984. Soil—An uncertain medium for waste disposal. In Proceedings ofthe Seventh Annual Madi-

Spring 1989 GWMR 87

' i if i

MWiMnam

Roherl A. Grijfin i.s principal cheiiiist and licad of the .'Environmental Gcolog}' and Geochemistry Branch ofthe iriirwis State Geological Survey (615 E. Peabody Dr., Champaign, IL 61820). He has a B.S. (P)66) and an M.S. (1968) in sod chemistry from the University of California, and a Ph.D. (1973) in soil physical chemistry from Utah State University. His primary research interests are in contaminant transport processes in earth materials.

Christopher J. Stohr obtained a B.S. in geology from St. Joseph's College, Rensselaer, Indiana, and an M.S. in engineering geology from Purdue UniversUy, West Lafayette, Indiana. As an engineering geologist with the Missouri Geological Survey he 'mvestigated many landfdls and waste disposal facilities, directed a statewide inventory of 3600 dams, and conducted a regional investigation of the state for hazardous waste isolation. He is an associate engineering geologist and remote sensing coordinator at the Illinois State Geological Survey (615 E. Peabody Dr., Champaign, IL 61820) where he has conducted research in soil properties, waste management, and remote sensing since 1980. Stohr also lectures at the University of Illinois Geography Department on photointerpretation.

Leon R. Follmer is coordinator of Quaternary Studies, vice chair of the Quaternary Geology Committee, and chair of the Radiocarbon Dating Conmiittee, Illinois State Geological Survey (615 E. Peabody Dr., Champaign, IL 61820). He obtamed a B.S. from Western Illmois University, andanM.S . andPh.D. in pedology from the

University of Illinois, l lc holds adjunct appointments in the Department of Geology ami Geography, and colla-horator status with the College of .Agriculture al the University of Illinois, and participates in graduate student research programs and serves as a field research advisor in pedology, soil-geomorphology, and glacial geology. Current research activity is in glacial mapping, pedostra-tigraphy, andfield identification of paleosols.

Walter J. Morse is an associate staff geologist in the Groundwater Resources Section of the Illmois State Geo­logical Survey (615 E. Peabody Dr., Champaign, IL 61820). He has lived most of his life in Illinois, and has a B.S. from the Geology Department of Illinois State Uni­versity. Current research projects include explorationfor and mapping of ground water resources, and investigation of ground water depletion by overpumping of aquifers. Surface and downhole geophysics are important compo­nents of these projects. Other research interests include ground water contaminate migration and hydrogeology of karst terrain.

Wen-June Su, an assistant engineering geologist at the Illinois State Geological Survey (615 E. Peabody Dr., Champaign, IL 61820), was born in Taiwan in 1951. He obtained a B.S. and an M.S. in geology in Taiwan. He worked as an engineering geologist in a consulting com­pany in Taiwan from 1978 to 1980. He also taught an engineering geology course at the University of Chinese Culture, Taiwan, in 1979.

WMWttWIiH'

Leak Detection and Monitoring Methods for Underground Storage Tanks

- EPA Sponsored by

United States Environmental Protection

Agency

JACA Corp. Environmental Consultants

& Engineers

iVlay 8 -9 , 1989 Boston

May 17-18, 1989 Dallas

May 24-25, 1989 Orlando

Cost: $295 (includes two-day seminar, handouts, two lunches, four coffee breaks)

L_

June 12-13, 1989 San Diego

For further information please contact: Ursula S. Thomas (215) 643-5466 JACA Corp. 550 Pinetown Road Fort Washington, PA 19034

Circle card no. 72 Spring 1989 GWMR 89

At tachmen t B

CHAPTER 6

SUBSURFACE CHEMICAL PROCESSES: FIELD EXAMPLES

Richard L. Johnson, Carl D. Palmer, and William Fish

Introduction Understanding chemical processes in the subsurface is essential for accurate site characterization and for the design and implementation of efficient remedi­ation systems. This chapter will focus on examples of how chemical processes can control t r anspor t and fate in the subsurface.

Petroleum Hydrocarbons In the Unsaturated Zone Gasoline in the subsurface has been examined from both experimental and theoretical points of view. For example, theoret ical ana lyses of spills by Baehr (1987), Corpacioglu and Baehr (1987), and Baehr and Corpacioglu (1987) demonstrate that volatilization is a dominan t mechan i sm of t r a n s p o r t for m a n y hydrocarbon spills (Figure 58). Gasoline is a complex mixture of hundreds of compounds, some of which are more a m e n a b l e to v o l a t i l i z a t i o n t h a n to dissolution, and vice versa. Also, other components within gasoline are not particularly prone to either and will tend to persist in the subsurface.

As discussed in C h a p t e r 5, the i m p o r t a n c e of volatilization is determined by the composition ofthe free-product, the vapor p r e s s u r e of the specific compound, and the r a t e a t which the chemica l difTuses in the vapor phase. For example, aromatic hydrocarbons in gasoline are moderately volatile, but also d issolve readi ly into the pore w a t e r . Therefore, they diiTuse more slowly and may persist in the unsaturated zone longer than the less-soluble alkanes and alkenes (Baehr, 1987).

Biodegradation also plays an important role in the fate of gasoline. Many components of gasoline are

< cc o

50

40

30

20

10

2000

1000 2000

05 ALKENES

C5-C6 ALKANES

06 NAPHTHENES

0 1000 2000

^ DAYS Figure 58. (a) Transport pathways (or subsurface gaso­

l ine; (b) masses o l a romat i cs in the unsaturated zone after a hypothetical spill; and (c) masses of hydrocarbons in the unsaturated zone after a fiypothetical spill (after Baehr, 1987).

readily degraded by subsurface microorganisms. In the saturated zone, biodegradation frequently makes the aquifer anaerobic, resulting in much slower rates of degradation. This trend towards anaerobic condi-

57

- i

^

§ 1410 < UJ W IU 1400 > o OQ <

1390 H-lU UJ U.

1380

-

'

OXYGEN (ATM) ^ — r " — " ' '

1 "V \

° a

V

f

" \

-"-

\

\

\

7 \

50 FEET

TOTAL VOLATILE HYDROCARBONS (g/m')

CARBON DIOXIDE (ATM) 1420

Figure 61. Oxygen and carbon dioxide concentrations in the unsaturated zone above an oil spill at Bemidji, IVlinnesota (after Hult, et al., 1985).

analyses. However, the sum of all priority pollutants in landfill leachates often represent only 1 to 5 percent of the total dissolved organic material present in the ground water (e.g., Reinhard, et al., 1984; Johnson, et al., 1989). The bulk of the unidentified organic contaminants are made up of a wide variety of primarily hydrophilic compounds (Figure 63). An examination of major chemicals produced in the United States (Webber, 1986) indicates that 7 of the top 35 are priority pollutants and only 5 are purgeable priority pollutants (PPP). Furthermore, many of the top 35 chemicals are significantly more hydrophilic than the PPP (Figure 64). This means that if sorption plays a role in retarding chemical movement, there are many common compounds that will move more rapidly than the PPP. Many of those non-PPP compounds cannot be detected by standard EPA methods. Nevertheless, they may have significant impacts on ground-water quality due to their mobility, solu­bility, degradability, toxicity, etc. A better under­standing of the complexity of landfill leachate is necessary before an accurate assessment of the transport and fate of the wide range of compounds found there can be made.

UJ > UJ - I < UJ w Ul > o CO < j -Ul UJ UL

1420

1410

1400

1390

1380

Figure 62

METHANE (PPM)

Hydrocarbon and methane concentrations in the unsaturated zone above an oil spill in Bemidji, Minnesota (after Hult, et al., 1985).

Alcohols

Analines Acetates Amines Thiols Furans NKriles Phenols Aldehydes Ketones Acids

Figure 63. Polar and ionizable compound classes commonly present in leachates.

Organic Transport on Microparticles A growing number of field data suggest that very hydrophobic compounds can be transported in the subsurface at rates which are significantly greater

59

m

minor importance, except where b reaks in the barrier exist.

There is some experimental evidence that solvents p r e s e n t in l a n d f i l l s c a n d e g r a d e o t h e r w i s e impermeable clay l iners (Anderson, et al., 1985; Brown and Thomas , 1984; Brown and Anderson, 1983; Green, et al. , 1983). The main question is whether or not organic solvents can cause the clay to shrink and crack. Experiments indicate that if pure solvents penetrate into the clay, they may be able to displace water within the clay structure, leading to shr inkage, cracking, and an increase in pe rme­ability. In these exper iments , the solvents were forced into the clay at high pressures equivalent to many tens of meters of fluid head. This behavior is consistent with the cri t ical pressure required for entry o f the solvents into the clay (estimated from the grain diameter ofthe clay) (Chapter 3; Villaume, et al., 1983), but is probably not realistic in most landfill situations.

In the absence of sufTicient pressure, ent ry of the solvents occurs by simple Fickian diffusion in the aqueous phase. However, the ex ten t of e n t r y is limited by the solubilities of the solvent in the water and vice versa. For many solvents, solubilities are thousands of milligrams per liter or less, in which case, the impact of the so lvents on the clay is expected to be minimal.

The presence of mechan ica l f a i l u r e s or o t h e r imperfections in the clay may provide focal points for degradation ofthe clay liners. Miscible liquids, such as alcohols, acids, bases, and ketones, also have the potential to affect clay structure. Because they are not limited by solubility, large concentrations can diffuse into the clay. Miscible compounds generally are polar molecules which can interfere with the electrostatic forces within the clay (Fernandez and Quigley, 1985). This interference can resul t in a breakdown of the clay s t ruc tu re and subsequent failure ofthe liner.

Even though chemical concentrations of thousands of mil l igrams per liter w i th in clay l iners a r e not expected to affect the hydraul ic propert ies of the clay, such concentrations could have a major impact on ground-water qual i ty because of the diffusive movement of chemicals through the liner. In the absence of advection, mass transport is controlled by Fick's Second Law:

XL 77 R ii7 dt

(2)

A n u m b e r of field s t u d i e s d e m o n s t r a t e d t he importance of simple Fickian diffusion (Johnson, et ah, 1989; Goodall and Quigley, 1977; Crooks and Quigley, 1984; Desaulniers, et al., 1981). Johnson, et al. collected clay cores from beneath a five-year-old waste disposal cell and determined concentration profiles for chloride and several organic contam­inants. The chloride diffused nearly one meter while the organics moved much shorter distances. The principal reason for the slower diffusion of the organics was their sorption onto the clay (fo,.—0.01). For non-sorbed contaminants (including organics when the f(,c is low), breakthrough of a one-meter liner could occur in less than 10 years. Within a few decades, steady-state concentrations could develop across the clay, possibly resulting in significant mass transfer into the underlying aquifer.

Under s teady-state conditions, mass fiux follows Fick's First Law:

ac D 0^ — (3)

As the calculation in Figure 66 suggests, the mass flux through a liner via diffusion can be on the order of 1 gram per square meter per year. For a large liner, a mass transfer of thousands of g r ams of individual contaminants per year to the aquifer

1000 ppm

l m

FLUX = -nD

0 ppm

dC

P-" dX

= - 0 . 3 7 x 1.5x10'® cm^/sec X I g / L -m

D 3 C

dx

ac dt

(1)

As seen in Chapter 5, for sorbing compounds, Fick's Law can be modif ied to h a n d l e e q u i l i b r i u m partitioning:

- 12 2

= 5.55x10 g/cm /sec

= 1.75 g/m /year Figure 66. Example calculations of steady-state diffu­

sive mass flux through a one-meter thick clay liner.

61

TOTAL VOLATILE HYDROCARBONS

LU O < u. cc

lil CQ h-LU LU LJ.

RELATIVE CONCENTRATION Figure 68. Vertical hydrocarbon profiles in the unsaturated zone above an oil spill near Bemidjii Minnesota site (adapted

(rom Hult, et al., 1985) .

pr imari ly in metal plat ing and l e a the r t a n n i n g app l i ca t ions for over 100 y e a r s . As a r e s u l t , numerous waste lagoons, dumps, and landfills a re contaminated with chromate wastes (Black and Heil, 1982; Cook and DiNitto, 1982; Owen, 1982; Massa­chuset ts Depar tment of E n v i r o n m e n t a l Qua l i ty Engineering, 1981; Keely and Boateng, 1987). At many sites, the chromium coexists with a variety of other inorganic and organic wastes, and under these conditions, a wide range of chemical interactions are possible.

Chromium transport in aqueous systems strongly depends on sorption, chelation, and redox reactions. The redox reactions are only poorly understood, yet, they are of key importance because the redox state of c h r o m i u m d i c t a t e s i ts s o r p t i v e a n d c h e l a t i o n behavior. Despite the strong tendency for chromium to partition to many mineral surfaces, there are reports of extensive chromium migration (Keely and Boateng, 1987).

Surface s o r p t i o n is an i m p o r t a n t e l e m e n t of chromium behavior. Electrostatic forces near the

solid-liquid interface form a transition range near neutral pH in which both positively and negatively charged surface sites coexist (Swallow, et al., 1980). Several published studies discuss adsorp t ion of chromium species at oxide interfaces (Bartiett and Kimble, 1976; Bartiett and James, 1979; Huang and Wu, 1977). If the mineral composition ofthe aquifer is known, appropriate sorption parameters for Cr(III) or Cr(VI) can be roughly estimated. Unfortunately, the redox state ofthe chromium also must be known and this is difficult to predict because natural redox chemistry of Cr(III)/Cr(VI) is not well understood.

Although Cr{III) and Cr(VI) are each quite stable and tend to be kinetically inhibited from undergoing redox t ransformations, there are systems which catalyze oxidation-reduction reactions of chromium. In strong acid solutions, Cr(VI) will oxidize organic compounds and be reduced to Cr(III) (Bartiett and Kimble, 1976). This reaction eliminates toxic Cr(VI) and genera tes the relatively insoluble t r i v a l e n t species. The reverse reaction can be driven by Mn02 in a process in which the Mn02 appears to act as both an oxidizing agent and a catalytic surface (Bartiett

63

gJiiKiMffi^lJ!

Baehr, A. L. and M. Y. Corpacioglu , 1987. "A Compositional Multiphase Model for Ground-water C o n t a m i n a t i o n by P e t r o l e u m P r o d u c t s . 2: Numerical Solution." Water Resources Research, Vol.23, No. 1, pp. 201-213.

Baehr, A. L., 1987. "Selective Transport of Hydro­carbons in the Unsaturated Zone Due to Aqueous and Vapor Pbase Partitioning." Water Resources Research, Vol. 23, No. 10, pp. 1926-1938.

Bar t ie t t , R. and B. J a m e s , 1979. "Behavior of C h r o m i u m in Soils; III. O x i d a t i o n . " J . Enu. QuaUty, Vol. 8, pp. 31-35.

Bartiett , R. and J. M. Kimble, 1976. "Behavior of Chromium in Soils: II. Hexavalent Forms." J . Env. Qua/i<y, Vol. 5, pp. 383-388.

Black, J. A. and J . H. Heil, 1982. "Municipal Solid Waste Leachate and Scavenger Waste: Problems a n d P r o s p e c t s in B r o o k h a v e n T o w n . " In : Proceedings, N.E. Conference on the Impact of Waste Storage and Disposal on Ground-wa te r Resources, R. P. Novitsky and G. Levine, Editors. U.S. Geological Survey/Cornell University.

Brown, K. W. and D. C. Anderson, 1983. "EfTects of Organic Solvents on the Permeab i l i ty of Clay Soils." EPA/600/2-83-016, Office of Research and Development , U.S. E n v i r o n m e n t a l Pro tec t ion Agency.

Brown, K. W. and J. C. Thomas, 1984. "Conductivity of Three Commercially Available Clays to Petrol­eum Products and Organic Solvents ." H a z a r d . Waste, Vol. 1, No. 4, pp. 545-553.

Cook, D. K. and R. G. DiNitto, 1982. "Evaluation of Ground Water Quality in East and North Woburn, MA." In: Proceedings, N .E . Conference on the Impact of Wa^te Storage and Disposal on Ground-Water Resources, R. P. Novitsky and G. Levine, Editors . U.S. Geological Survey/Corne l l Uni­versity.

Corpacioglu, M. Y. and A. L. Baehr , 1987. "A Compositional Multiphase Model for Ground-water Contamination by Petroleum Products. 1; Theo­retical Considerations." Water Resources Research, Vol. 23, No. l ,pp . 191-200.

Crooks, V. E. and R. M. Quigley, 1984. "Sal ine Leachate Migration through Clay: A Comparative Laboratory and Field Investigation." Can. Geotech. J., Vol.^21, pp. 349-362.

Desaulniers, D. E., J. A. Cherry, and P. Fritz, 1981. "Origin, Age, and Movement of Pore Water in Argillaceous Quaternary Deposits at Four Sites in Southwestern Ontario." J . Hydrology, Vol. 50, pp. 231-257.

Fernandez, F. and R. M. Quigley, 1985. "Hydraulic Conductivity of Na tu ra l Clays Permeated with

Simple Liquid Hydrocarbons." Can. Geotech J Vol.22, No. 2, pp. 205-214.

Fish, W., 1987. "Subsurface Transport of Gasoline-derived Lead at a Filling Station Contamination Site in Yakima, Washington." In: Proceedings, NWWA FOCUS: Nor thwes te rn Ground Water Issues Conference, National Water Well Associ­ation, Portland, OR, May 1987.

Godsy E. M. and D. F. Goerlitz, 1984. "Anaerobic Microbial Transformations of Phenolic and Other Selected Compounds in Con tamina ted Ground Water at a Creosote Works, Pensacola, FL." In: Mouemenl and Fate of Creosote Wasle in Ground Water, Pensacola, FL, U.S. Geological Survey Open File Report 84-466.

Goodall, D. C. and R. M. Quigley, 1977. "Pollutant Migration from Two Sanitary Landfill Sites near Sarnia, Ontario." Can. Geotech. J., Vol. 14, pp. 223-236.

Green, W. J., et al., 1983. "Interaction of Clay Soils with Water and Organic Solvents: Implications for the Disposal of Hazardous Wastes." Environ. Sci. Technol., Vol. 17, No. 5, pp. 278-282.

Gschwend, P. M. and S.-c. Wu., 1985. "On the Constancy of Sediment-water Par t i t ion Coeffi­cients of Hydrophobic Organic Pollutants ." Enu. Sci. Tech., Vol. 19, pp. 90-96.

Huang, C. P. and M. H. Wu, 1977. "The Removal of Chromium (VI) from Dilute Aqueous Solution by Activated Carbon." Water Res., Vol. 11, pp. 673*-678.

Hult, M. F. and R. R. Grabbe, 1985. "Permanen t Gases and Hydrocarbon Vapors in the Unsaturated Zone." In: Proceedings, U.S. Geological Survey Second Toxic-Waste Technical Meeting, Cape Cod, MA, October, 1985.

James, B. R. and R. J. Bartiett, 1983. "Behavior of Chromium in Soils: IV. In te rac t ions Between Oxidation-reduction and Organic Complexation." J . Env. Quality, Vol. 12, pp. 173-176.

Johnson, R. L., C. D. Palmer, and J. F. Keely, 1987. "Mass Transfer of Organics Between Soil, Water, and Vapor Phases: Implications for Monitoring, B i o d e g r a d a t i o n , a n d R e m e d i a t i o n . " In : Proceedings, Petroleum Hydrocarbons and Organic Chemicals in Ground Water, National Water Well Association and the American Petroleum Institute, Houston, TX, November, 1987.

Johnson, R. L., J. A. Cherry, and J. F. Pankow, 1989. "Diffusive Contaminant T ranspo r t in N a t u r a l Clay: A Field Example and Implications for Clay-l ined W a s t e Disposal S i t e s . " Accep ted for publication in Environ. Sci. Technol.

65

mwrni ^xMosm

Pt^fc^C>vv>,j\r CL

R. E. Olsofi' a n d D . E. Daniel '

---^.._^M oi:-. .

c;OMi>;f: i^n(^K.; ,[ . , ,

Measurement of the Hydraulic Con(ductivity of Fine-GrainecJ Soils

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 19

where q is the flow rate [lengtli-'/time ( L V T ) ) ; / i.s the hydraulic gradient (dimensionless); A is the total cross-sectional arca of fiow (L-); arui k is the constant of propoitioiiality (L/T), which is termed the hydraulic conductivity in most disciplines but is often termed the permeability by civil engineers. In Eq 1, A: is dependent on the properties of both the fluid and the porous medium. An alternative form of Daicy's law is

q = - K ^ ^ i A (?.)

lUiriiUCNCE: Olson, R. E. and Daniel, D. E., "Measurement of (he Hydraulic Conduc-llvilv of Fliie-Graincd Soils," Permeahilily and Groinulwater Coniaininanl Tniiisporl, •l.s7A/ ,STI' 7-16, v. F. Ziminie and C. O. Riggs, Eds., American Society for Testing and M.ilciials. 1981, pp. 18-64.

ADS TRACT: The purpose of this paper is to review the state of the art in the measiircnie/it of livdiaiilic conductivity of fine-grained soils. Both field and laboraiory tests for saturated and pnrtiaily saturated soils are considered.

For saturated soils, field tests are to be prefened because they permeate a larger volume of .soil, lluis taking into account the effects of niacrostructure better than laboratory tests. Field tests are generally best performed by using a cylindrical piezometer tip, installed by methods that minimize disturbance, and measuring flow under a constant head. Labora­tory tests offer the advantage of economy. Laboratory specimens should be as large as prac-liial nnd should be oriented to produce flow in the direction of ma.ximum hydraulic con-ducli\ity. The permeant should be a fluid similar to that encountered in the field. Without proper experimental technique, the conductivity measured in the laboratory may differ from the field value by several orders of magnitude.

Field tests for unsaturated soils are not well developed and can only be recommended for cases where water will be ponded on the surface of a site. The mosl versatile laboratory techniques are Ihe instantaneous profile method using tensioinelric or psychromelric probes, and the pressure plate outflow method. The besl method to use depends on the soil suction expected in the field.

KKY WORDS: permeability, hydraulic conductivity, soils, fine-grained soils, clays, sam­ple preparation, laboratory tests, field tests, suction, unsaturated soils, groundwater

Analyses of water flow in saturated soils are usually based on Darcy's [7]^ law which, in turn, is based on the experimental observation of a linear rela­tionship between the rate of flow and the driving forces. The law has been written in many forms depending mainly on the discipline of the user and the date of usage. The form most commonly encountered is

q = —kiA (1)

Professor and assistant professor of Civil Engineering, respectively, Department of Civil En­gineering. University of Texas, Austin, Tex. 78712.

"Ihc italic numbers in brackets refer to the lisl of references appended to this paper.

where 7 , [units of mass/length^ titne^ (M/L^T^)] and n (ML/'f) are tiie unit weight and viscosity of the fluid, respectively, and K is the constant of propoi-tionality (L^). In most disciplines K is termed thc coefficient of pcrincabililv but is sometimes termed intrinsic permeability [2\. A third form of Darcy's law is [.7]

q -k dp IL d.x

A (3)

where q is the flow rate (cubic centimetres per second), /t is viscosity (ceiiti-poises), p is pressure (atmospheres), .v is flow distatice (centimetres.), ,4 i.s total cross-sectional area (square centimetres), and A- is the permeability in units of "darcies." For a permeant of pure water at 20^C the conversions ate

1 cm/s = 1.02 X 10--^cm2 = 1.04 X 10^ darcy

Because of its simplicity and the fact that we are concerned with the flow nf a single fluid, water, we prefer to use Eq 1. To conform with general usage wc will also term k the hydraulic conductivity, or, where thc meaning is clear, simply conductivity. However, when considering general matters not con­cerned with a specific equation or parameter we will use the term permeability, as, for example, "permeability tests."

Most engineers and geologists are familiar with techniques for measuring k in coarse-grained soils—for example, constant or falling head tests in the laboratory and pump tests from wells in the field—but are less faniilitir with techniques for measuring k in fine-grained soils. The reason for this, in the case of geologists, is probably that priinary interest has traditionally been in development of groundwater supplies, which cannot be extracted economi­cally from fine-grained soils due to their low hydraulic conductivities. In engi­neering, past practice has frequently been to assume that fine-grained soils are effectively "impervious" and to foigo attempts to measure k. The need for measurements of hydraulic conductivity in fine-grained soils seems to be iti-cicasini? ;is ii result of recenl developments. One such developmenl is in-

20 PEnWEABlLlTY AND GROUNDWATER CONTAMINANT TRANSPORT

creased concern over the long-term environmental effects associated with buiyiiig toxic wasles in the ground. When consideration is given to water and poiliilanl niovenientsover periods of up to hundreds of thousands of years, the finc-giained .soils can no longer be treated as impervious and their conduc-livilics imist be measured. The attractiveness of sites in arid regions for disposal of radioactive and other toxic wastes has also created interest in measuring the conductivity of partially saturated soils.

Another area of increased inteiest involves consolidation problenis; evidence now indicates that the accuracy of field predictions may be improved by using cither laboratory [5J or i?i situ [6] measurements of conductivity, as opposed to evaluating the coefficient of consolidation directly by fitting theoretical curves fo iaboraloiy time-settlement data [7]. Measurements of hydraulic conduc-livily in fine-grained soils have always been, and will continue to be, of concern ill design of earth dams, slun^ trenches used as groundwater cutoffs, and cotnpactccl clay used to line water-storage reservoirs or waste-disposal pits, as well as ill hydrological investigations of groundwater recharge.

In the ijagcs that follow, we will examine the state of the art for measuring Ihc hydraiiiic conductivity of both saturated and partially saturated fine­grained soils, inchiding both field and laboratory methods. It is convenient to discuss the relative merits of laboratory versus field tests first. Then laboratoiy (esling foi- both saturated and unsaturated soils will be considered, followed by a similar discussion of field testing. Finally, laboratory and field values will be compared. Testing techniques and testing errors will be discussed. No attempt will be made to cover in detail all methods that have been used, but references wiil be made to numerous papers to aid in a more expanded study.

Liiljoraloiy Versus Field Measurements

Soils lend to be nonhomogeneous. Fine-grained soils may be stratified on a laige scale but may also be nonhomogeneous on a small scale as a result of sand pat tings, fissures and joints, and root holes. It is desirable, therefore, to lesl a volume of soil that is large enough to contain a .statistically significant (listiibtilion of these features. Such volumes are almost always too large to be inchided in laboratory tests. Further, the resultsof laboratoty tests may be in-flnciiccd by effects of sampling disturbance, by various laboratory errors to be discussed subsequently, and by a tendency to select the most uniform, intact samples for testing. Tbe soil is also likely to be anisotropic, and laboratoiy saniiiles are not likely to be oriented so that flow occurs in the direction of highesi conductivity. Clearly, field tests will be required in many cases.

I here are several cases in which laboratory tests seem appropriate. In the case of compacted soils, permeability tests on soil prepared at various densities and water contents are mo.st readily performed in the laboratoiy. Comparative lests to determine the effects of different permeants on conductivity are also lypically perlormed in the laboratory.

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 21

Two practical reasons may also be advanced for use of laboratory tests. The first is economics. Large numbers of laboratoiy tests can be performed in a well-equipped laboiatoiy with minimal expense, but field tests can be pro­hibitively expensive. A second rea.son for using laboratoty tests is ignorance of field testing methods. Essentially all soil mechanics texts mention acceptable laboratoiy testing methods, but none currently provides a similar treatment for field measurements of the hydraulic conductivity of fine-grained soils.

Laboratory Tests for Saturated Soils

Penneahility Cells

The standard laboratoiy tests for saturated fine-grained soils generally have the soil in the form of a disk with radial boundaries of metal, or occasionally plastic, and with the flow vertically upwards. Consolidation cells work well for undisturbed samples, whereas compaction molds are used directly for com­pacted soils.

A simple design for a consolidation-cell permeameter is shown in Fig. 1. The ring has a sharpened upper edge to facilitate trimming. With such a))paratus and hand trimming, values of k less than 1 X I0~'- cm/s have been measured. Mechanical devices [S.9.I0\ may be used to pie\ent tilting of the ring during trimming and thus minimize the |:ossibility of creating voids be­tween the soil and the ring.

Clamping Nuts (3 at 120°)

Porous Stones

Loading Cap

Water

Clamping V Flange

Drainage Port

Ring O-Ring

F I G . 1—Cnn.si>/i(i(i[n>n cr / / /JcrnKUimrlcr .

22 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

Cells for samples from about 4 to 10 cm (1.5 to 4 in.) in diameter, and heights up to about 10 cm (4 in.), can usually be mounted in standard con­solidation loading frames if the effect of void ratio or effective stress on con­ductivity is to be determined. The base should be provided with two drainage lines to facilitate flushing with deaired water.

For compacted soils, a standard 10-cm (4-in.) diameter mold can be used with a special base containing a porous stone. If desired, a top plate may be sealed against the top of the mold and a vacuum may be applied to the top during the saturation stage. For soils containing coarse particles a larger mold may be used [11,12]. With such apparatus it is common practice to prevent soil swelling by clamping a plate against the top of the sample, but calibrated siblings may also be used [12].

Permeability tests are also conventionally performed in standard triaxial cells \ l .^ . This apparatus has the advantages that back pressure can be used to pif)mote saturation and the applied total stresses can be controlled. Similar advantages accrue from use of back-pressure consolidation cells [14,15.16].

Slandard Test Methods

Constant Head—In the constant head test, the hydraulic gradient / is main­tained constant at a value li7L, where /; is the head loss associated with flow through the soil sample and L is the length of the sample, and the total volume of flow, Q. is measured during a time period, /. From Eq 1

k = QL7hAt (4)

For fine-grained soils the constant head is typically applied using a Mariotte bottle, one example of which is shown in Fig. 2 [12]. Such equipment is designed to apply only small heads (a few feet of water) so it is most useful with rather pervious soils or in cases where prolonged testing times can be tolerated. The main advantages of constant head tests are tlie simplicity of interpretation of data and the fact that use of a constant head minimizes confusion due to changing volume of air bubbles when the soil is not saturated.

Falling Head—A more coinmon test for fine-grained soils is the falling head lest in which the time, /, for the head loss, //, to decrease from //] to //2 in a volumetric tube, typically a pipet or a buret with cross-sectional area, a, due to flow through a sample of area. A, and length, L, is measured. Equation 1 is used with q = adh/dl to obtain

a l h, In —

At hi (5)

Tlic advantage of using this procedure is that small flows are easily measured using the pipet or buret. The observation time may still be long, in which case corrections for water losses due to evaporation or leakage may be needed.

Scale

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 23

Inner Tube Open to Atmosphere

•Outer Tube

Water Leve l

: 2 \

Rubber _ _ Stopper"""""—--~^Jffl H

•Head Measured between Here and E levat ion of Ta i lwa te r

3 To Sample

FIG. 2—Mariotte bottle fo r maiiilainiiig coitslaiU head ancl nn-asnrinR j h w laics / 1 2/.

The testing time may be reduced by increasing the flow rate, for example, by superimposing an air pressure, ;/„, on top of the water in the pipet, thus in­creasing the heads by an amount ii /'Y-.',- An apparatus such as that shown in Fig. 3 works well for this purpose; it allows the pipet to be refilled without removing the air pressure. However, two problems still exist. First, there is a tendency to superimpose excessive heads. Second, the water in the pipet tends to become saturated with gas at elevated pressure, in accord with Henry's law [17.18], As the water flows through the soil sample, and thc water pressure drops, there may be a tendency for air bubbles to evolve in the sample. This problem is minimized by replacing the water in the storage resenoir with deaired water periodically and flushing that water through the base to remove water containing excessive gas. Alternatively, the volume change device

24 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

Vent

GiadL ja ted Pipette To Measure Flow

To Cell t r

Reservoir

SL

® 0 Open to Ref i l l Pipette

To d g t = J Ai r

Supply

Mercury Manometer lo Measure Air Pressure

Dra in

FIG. 3—Appara tus fo r superimposing air pressure on water.

teported by Mitchell, Hooper, and Campanella [19] may be used or pressure may be obtained from the hydrostatic pressure of a column of mercury and a "single buret volume change" device [20,21] may be used. In cases where the npplied pressure substantially exceeds that of the water head, the test becomes essentially a constant head test and Eq 4 is used.

A falling head test naturally lends itself to automatic data recording using a differential pressure transducer mounted in the base of the cell with the tailwater reservoir connected to the reference port.

Special Test Methods

One-Dimensional Consolidation Testing—The hydraulic conductivity can be calculated from the coefficients of consolidation, c, and compressibility, a, nnd void ratio, e

A: =r<7r„./(l -Fe) (6)

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 25

(see Ref 7, page 228). Terzaghi |22| found that the measured k (k,„) from con­ventional permeability tests and the value calculated using Eq 6 (/CKI) weie essentially equal for one soil. Casagrande and Fadum (Ref 2J, page 480) also found substantial equality, piovided there was a distinct break between primaiy and secondary consolidation, but presented no supporting evidence. Taylor [24] found significant differences between k,„ tmd ky^y.

We have compared k„., and ky^j for numerous samples of undisturbed, remolded, and resedimented clays. Typical data are shown in Fig. 4. For highly overconsolidated clays k„^/k^-y ranges from perhaps 2 to IOOO, wtiereas for normally consolidated clays the ratio varies from about 0.9 to 5. flic discrepancy between k„ and k^y presumably results, at least in part, from the fact that the classical theory of consolidation makes no adjustment for the structural viscosity of the .soil.

Radiol Flow Tests—The horizontal conductivity can be measured in (he laboratory using a cell (Fig. 5) with a central sand drain (radius = /„) and a

2.8 ~ i I I I l l l l 1—I l l l l l l "I I M i l l

0.4 10'

I I I l l l l l l I l l l l l I I l l l l l l

10 ° 10

k, c m / s e c

-7 10

FIG. 4 — Typicfj/ conipatisons nf lut'iisurcd cnnt/nclivifv (tifwn SVHI/ID/.^I and Cf)iu/ucil\-i!v CDnt-p u t c d from Tcrzaf^/ii's t / teon' fso/id syn\/><)ls).

26 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

C o n s o l i d a t I o n

P r e s s u re

P o r o u s P l a s t i c

Dra inage Con t ro l Valve

FIG. 5—Radial flow pernteanieter.

porous outer boundaiy (radius = T-Q)- For a sample height of £ and a constant head of // on the sand drain, the conductivity is

k = Q ro

l irLht r ,

wiieic Q is tbe volume of flow in a time period t. For a falling head test

; - " 1 " 1 , ' '0 k — -——- In — In — 2 TrLt h -I r„.

(7)

(8)

In principle, radial flow tests can also be performed in the triaxial apparatus using a central sand drain and a continuous outer filter paper drain, but the pcrmcabilify of the paper may not be high enough to provide free drainage ex­cept for fairly impervious clays [25.26]. Use of a triaxial cell offers the advan-

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 27

tages that no special aj^paratus is needed and that large enough samples may be used to include tiie effects of maciostrucliire, siicii as fissures.

Low Flow Tests—\n the case of relatively impei-vious clays, substantial time may be needed lo obtain measurable flows. A po.ssible solution \27,2<S.29\ is to perform the equivalent of a falling head permeability test by replacing the pipet with a compliant pressure transducer. The pressure on the upstream (transducer) side is elevated suddenly, thus deflecting the diaphragm of the transducer. The pressure is measured as water slowly leaks out of the transducer through the sample. The volume of flow is determined from cal­ibrations of volume change versus transducer readings.

Sources of Error in Laboratoiy Tests Using Saturated Soils

"Sources of error" is here laken to mean all errors that cau.se the hydraulic conductivity measured in the laboratoiy to differ from the conductivity in the field. The term "saturated" is taken to mean nominally saturated.

Nonrepresentative Samples

The overriding source of error in laboratoiy permeability tests involves use of samples that are not representative of actual field conditions. Provided reasonable care is taken in the performance of the laboratory tests, the chances of making gross errors are probably controlled by this factor alone. The problem of unrepresentative sampling is best minimized by thorough field investigation, by attention to details (sampling along faults, fissures, sand partings, and so on), by prudent selection of samples for testing, and by use of large samples.

Laboratoiy Testing Errors

Voids Fonned Dtiring Sample Preparation —For undislurbed samjiles, voids may be formed around the edges due to inadequate control of trimming, and fissures may open as a result of stress relief, thus leading to unrealistically high measured hydraulic conductivities. The first problem can be minimized by |)roper technique during trimming and the second by suhjecting the samjjlcs to stresses ajJiiroximating those in the field.

Smear Zones—U the sample contains such features as thin sand partings or root holes, the trimming operation may smear clay across the surface and tend to block entrance to these zones. Van Zelst [30] considered the effect of distur­bance during trinnning of the faces of specimens of clay for one-dimeiisioii;il consolidation testing and concluded that each of the flat faces of his clay was remolded to a depth of about 0.2 cm (0.1 in.). Chan and Kcnney [.V | trimmed samples of varved clay using a vibro-tool moved parallel to the stratification

28 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

and found that the thickness of disturbed soil on each face was 0.06 cm (0.025 in.).

Id minimize the effects of smear, (I) use a sharp knife for final trimming and cul the soil rather than trowel it; (2) include open root holes and other visi­ble /ones of higher conductivity in the specimen to be tested; and (3) use as large a s|5ccimen as possible.

Alti'idtiniis in Clay Chemistty—A belief apparently held by some is that the permeability test should be performed using distilled water because such water is incil. Actually, leaching a sample with distilled water may cause expansion of (he diffuse cloud of ab.sorbed cations around clay particles and reduce hydiaiilic conductivity. Further, in some soils the leaching may increase parti­cle mobility, either because of expansion of diffuse double layers or because of iemo\al of cements, and lead to particle migration. An example of these ef­fects is shown in Fig. 6 [32]. A solution is to use a permeant of the same chcmisdy as the original pore water but the time and expense involved in ex­tracting, analyzing, and duplicating the pore water makes this solution im-piaclical. Alternatively, samples of groundwater may be obtained in the field and used as a permeant. Another possible .solution is to perform laboratory (csis using veiy.small amounts of flow and using sealed permeameters in which Ihc flow can be reversed, thus cycling the same fluid to and fro [19,33].

Large changes in conductivity are likely to occur if a permeant is used with a chemislry that is widely different from that of the pore fluid. For example, Fiieni.iii \34\ leached samples of Hesperia sandy loam with various aqueous soliKions and found the following conductivities: 4 X 10""^ cm/s (originally), (1 X l ( ) - ' cm/s (800 ppm calcium chloride), 3 X 10~^ cm/s (tap water), 1 X 10" ' cm/s (4500 ppm sodiuin chloride), and 2 X 10~^ cm/s (distilled water). Numeious studies have shown changes in conductivity for samples originally prepared with different chemistries [35-38].

When it is not feasible to determine the chemistry of the natural pore fluid and duplicate it as a permeant, agronomists have often used 0.01 A' calcium siiHale as the permeant [39.40]. Some prefer to use tap water, which, though iK)l ideal, generally seems a much better choice than distilled water.

Air in the Sample—In testing compacted samples, engineers often assume (ha( soaking from the bottom, with the top open to the atmosphere, will lead to satiiiated samples. Smith and Browning [41] used this procedure to "salurale" 200 samples. They found that the average degree of saturation of I heil samples was 91 percent with the lowest value 78 percent. Because water cannot flow through an air bubble, the bubbles effectively reduce the void space thtit can be occupied by water and thus reduce hydraulic conductivity. Bjeiiiim and Huder [42] noted that air bubbles may tend to accumulate near the end of tbe sample where the water emerges, causing a clogging and er­roneous measurements. Christiansen [43] soaked a number of air-dried sani­iiles from the bottom up and then ran permeability tests. As flow continued

o CO

~ E o

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 29

, - 6

10

10 -8

10"

; _ _ --\ \ - \

\

-O o

-

r-

_ ---

"

1 1 1

P o r m n n n l : D i s l i l l e d W . i l e '

' - '— ' S a m o ' e No. 2

X Permean t :

\ ^ N a l u r a l Pore

\ I S a m p l e No 7

\ \

\ ° \

VV^^^^^ ' —Q.

1 1 1

W n l e r

1

-----

-z

--'

'

20 40 60 80 C u m u l a t i v e I n f l o w , cc

100

FIG. b —Influence of using distilled water Ifrom Wilkin.ion IMII.

the entrapped gas bubbles were slowly removed, and thc measured hydraulic conductivities went up by factors ranging from 2 to 40 times. A typical range for many soils may be closer to 2 to 5 tinies [44]. When Christianson .soaked his samples under vacuuni he found no time-dependent increase in conduclivity.

If water is forced through the soil using compressed air, the water entering the sample may contain a higher gas concentration than that corresponding to gas saturation at a lower pressure, and thus gas bubbles may form as the pressure in the flowing water decreases.

Growth of Microorganisms —Prolonged performance of permeability tests may result in a substantial reduction in hydraulic conductivity due to clogging of the flow channels by organic matter that grows in the soil during the test. Allison [45] reported on tests in which a variety of disinfectants were added to the permeating water to stop such organic growth. He found that phenol (IOOO ppm) and formaldehyde (2000 ppm) were the most effective agents in delaying growth of organic matter, but eventually even these soils "sealed up." Finally, he sterilized samples of three soils, including one that was rich in organics and two that contained little organic matter, and used elaborate procedures to keep the permeating fluid sterile. In the stenle samples there was an insignifi-

30 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

cant decrease in k with time. However, the long-term conductivities of sterile soils were 8 to 50 times the values for soil that was unsterile or was originally sterile but had been allowed to regain organisms (Fig. 7).

The imiilications of these tests for field problems depends on the application of interest. For some problems, such as ponding of water on the surface of a site, microorganisms are just as likely to plug up the soil in the field as they are a labottitoty specimen [45]. Hence, it might be besl not to liy lo prevent giowth of such organisms in laboratoiy tests. In other applications, growth of microorganisms in the field may be unlikely, in which case a disinfectant should bc added to the permeant and testing times should be minimized.

Meniscus Problems in Capillaiy Tiihes^ln an effort lo minimize olher er­rors some investigators have used low hydraulic gradients and allempled to measure the outflow by obsen'ing the movement of the air-water interface in a capillary tube. Olsen [33.46] has demonstrated that significant errors can oc­cur in the calculated pressure drop across the specimen because of essentially unavoidable contaniination of the capillary lubes, which leads lo an inde­terminate, but nonzero, contact angle belween the waler and the glass. To eliminate this problem Olsen used a constani flow rale and measured the pressure drop.

Olsen [33] also pointed oul thai waler flow rates should not be measured by obseiTing the rate of movement of an occluded air bubble in a capillary tube because water can bypass the bubble.

Use of E.xcessive Hydraulic Gradients—In an effort lo reduce testing time, large hydraulic gradients may be imposed on samples. If Darcy's law is valid, such gradients will not alter the measured conductivity. Schwartzendruber [47] sur\'eyed the then-existing literature and found many experiments in which k, defined as in Eq 1, increased as fhe gradient increased, with ratios of the maximum to minimum measured k lypically between I and 5 but with one value of 84. Other studies, such as those by Mitchell and Younger [48] and Gairon and Schwartzendruber [49], found decreasing values of k as the gra-

o <D W

E o

20 40 60

Time (days)

FIG. 7—Time de.pe.ndcnce of conductivity of ( l j sterile soil and water. I2j initially sterile soil ni-rmealed with un\u-rile wnter nnd I t l iin\lerile uid and wnler Ifrnni .Allison lAFill

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 31

dient was iticretised, apparently as a result of |)article migration, causing clog­ging. Il seems desirable to use gradients as close to those encountered in the

• field as is economically feasible. Temperature—On occasion, engineers correct the measured conductivity to

a standard temperature by adjusting for the effect of temperature on the viscosity and density of pure water'(Ref 7, page 113, nnd Ref 12, page 592). We have measured the effect of temperature on conductivity of three fine­grained .soils (Fig. 8) and found that simple viscosity and density adjusdiiciils are generally adequate for taking into account the effects of temperature. Note that conductivity is not particularly sensitive to small or moderate changes in temperature when water is used as the permeant; thc viscosity of wnter decreases approximately 3 percent per degree Celsius rise in temperature from 21 °C. However, the conductivity of fine-grained soils is probably influenced by complex interaction between the water, adsorbed and free ions, and the mineral surfaces. Consequently, it is a good idea lo perform permeability tests at approxiniately the relevant temperature when the results are to be applied lo the solution of a problem in the field.

Volume Change Due to Stress Change — If a change in pore pressure is im­posed on a sample under a constant tolal stress, the resulting change in.effec­tive stress must result in a change in volume of the sample. Thus, in a constani head test some of the initial measured inflow is making up for volume change rather than steady-stale seepage. In a falling head test the appnrent k would depend on the current applied head [50.51 \.

3 h

KEY

• Taylor Marl

• Kaol ini te

* Tokyo Si l t

Relat ionship Predicted by Viscosi ty Correct ion

Average Measured Curve

J_ 20 30 40 50

Tempera ture ( °C )

60

FIG. ^ — Effect of t empera ture on conductivity. Conilucliviiies al temperalure I ^kyJ are noi-

ualized with respect to the measured conductivities at 2.5°C'.

32 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

Al-Dhahir and Tan [52] have presented a solution for the consolidation (or swelling) of a sample subject to a constant total stress and an instantaneous change in pore pressure al one boundaiy. They suggest thai a plot be prepared of the flow rate at the boundary where the pore pressure is changed, q, versus t~^'^. The relationship should have a sloping portion, representing the lime period in which the soil is undergoing volume change, followed by a leveling off at small values of <~"' 2, where volume change ceases and steady-slate seepage occurs (Fig. 9).

Flow Direction—h is nearly always easier to perform laboratory permeabil­ity tests with the soil in the same orientation as in the field and with the flow vertical. However, sometimes the horizontal conductivity, ky,, is larger than the vertical value, k„, which usually leads to predominantly honzonlal flow in the field. Data published on the ratio k^ /k., are summarized in Table I \31,.53-60], For varved or stratified clays, the ratio may exceed 10, whereas for less stratified soils, the ratio is likely to be closer to 1. For soils containing rool holes, k, may e.xceed Aj,. Clearly, laboratory specimens should be onented to produce flow in the direction that will dominate in the field.

Laboratoiy Tests for Partially Saturated Soils

Methods available for laboratory measurement of the conductivity of par­tially snfuraled soils are similar to those used wilh saturated soils, and the problems are similar but more severe. Two of the problems require immediate discussion because they influence testing procedures strongly; these are (1) measurement of pore water pressure and (2) the effecl of the degree of satura­tion. We will examine these problems first, and then consider testing methods and errors as for saturated soils.

Measurement of Pore Water Pressures

Pote water pressures in partially saturated soils are negative compared with the pore air. To avoid use of negative numbers we will use the lerm "suction" as the negative of the pore water pressure. Techniques for measurement of suction have been reviewed in the engineering literature by Croney and Cole­man 16/J and in this volume by Daniel, Hamilton, and Olson [i52]. References to agricultural literature will be included in Ihe following discussion.

Only a few of the methods discussed in the literature can be used when con­ductivities are to be measured. Typical problems wilh other methods include slow respon.se, inadequate sensitivity, and instability. The useful apparatuses include tensiometers, pressure plates, and p.sychrometers.

Tensioincters—A tensiometer consists of a porous sensing elemeni con­nected to a pressure measunng device by a tube. The sensing element is typicnily a ceramic probe. The pores in the probe must be fine enough to pre­vent nir from blowing through the stone and draining the measuring systeni.

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 33

1.0 1.5 2 0 2.5

t - ' / 2 ( , ^ , , - 1 / 2 )

FIG. 9 — Tinu! dependency nf rate of water flow fnr remolded clav tested in cunsolidatinn cell permeameter,

TABLE 1—Data on the ratio of horizontal to vertical conductivity of fine-grained soils.

Reference k,-'k,. Notes

Subbaraju ct al |.S.?|

Lumb and Holt |.S/| Bazctt and Brodie 155|

Tsicn 1561

Chan and Keniiey |,7/| Keiiney and Chan |.S7] Haley and Aldrich |.S(S'| Wu et al I.S9I

Casagrande and I'oiilos (6^1

1.05

1.2 1.5

1.2 10 1.7

1.5 to ,1.7 1.5

0.7 10 3.,1 ^ lo 15

•1 to '10

soft marine clav, u' = A , . /. . = 65 lo "0%. /'.. = 2'4 to 55"/,,

higlilv plastic marine clay •soft clav, t„ = 60 lo 80%. /» =

35 lo 50"/n organic sill wilh peal, w =- 191 lo

570".;. vancd clay, lahoralory tesls vancd clay, field tesls Boston blue clay, w — 40 to 't5"/n vari'Ctl clay, w ~ 20 lo .10"ii.

I , , - 25 lo.35"/n, /-'„ = 8 to 20",;i vaiTcd clay, w = -15 to 75"n, /,„ =

50 lo 8()'v;,

However, tensiometers are generally limited to suctions less tlinii aboul 0.9 aim 16.7| because larger suctions lead to niiclcation of air bubbles in the mctisiiting system ("cavitation"), and the immediate ex|iansion of these bubbles reduces the suction in (he me;isuring .system essentially to zero. A tyjiical tensiometer i. shown in Fig. 10.

34 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

^" i ' ; s ^

'7 ' ' 7^ ^ ' " />|j '.,i ^ ' V ^ o I

' 1 /w1 in. H 1 ^

l i e . \0~Pl io iograph of la) tensiometer in brass housing and (bl therniocouple psvchrometer.

Pressure Pliiies—V/hen used lo measure suctions, the pressure plate device | 6 / . 6 / | may bc the same as a tensiometer except that cavitation of the measur­ing system is prevented by superimposing an air pressure within the soil sample until the iJtessure in the measuring system is near zero. The suction is defined ;is (he iJoie air pressure minus the pore water pressure, and is unaffected by in-ciensing the pore air pressure. The melhod may also be called the "axis (ranslation" method [65]. The probe may be made of ceramic for suctions up to about 15 atm. For higher suctions a Visking membrane may be placed over a ceramic probe [64.66,67]. Allhough air may not blow Ihrough the fine probe, it may pass through in solution and reform as air bubbles in the mea­suring systeni.

Psycl i rometers-Suct lons up to aboul 80 atm can be measured by determin­ing the relative humidity, H , of the pore air using a psychrometer [62.68.69] and calculating the suction, /;, using

R T p ^ — - h , H

M (9)

w here R is the gas constant, T is temperature, M is the molecular weight of water, and H is expressed as a ratio, Psychromelers are inaccurate for suctions lower than I or 2 atm because of the difficulty of measuring relative humidities near IOO perceni. A photograph of a typical psychrometer is included in Fig, 10.

Fffcct of Degree of Saturation

Measurements show that ( I) the degree of saturation decreases as the suc­tion increa.ses (Fig. Wa), and (2) the conductivity decreases rapidly as the degree of saturation decreases (Fig. 11/;). The imposition of a hydraulic gra-

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 35

100 ir

100

D e g r e e of S a t u r a t i o n , %

FIG. 1 1 —.S'i/(f/ij/( and hydraulic conductivity versus degree of sal unit ion for compacted fin

clay.

dient on a sanqile leads to spatial varialions in suction, degiee of satutalion, and thus conduclivity. riierefore, either tests must be iierformed using small gradients, or water flow-gradients-suclions must bc measured simultaneouslv al a point, or some other method used to resolve this |)inblem.

Measurement of Conductivity

Steady-State A-/<?//;o(:/s—Steady-state methods arc similar to those used for saturated soil except the head is ncgtitive and is conirolled at both ends of (hc samjjlc |7(^|. For fine-grained soils, thc sample is ty|iically cylindrical, with tiie dinmeter of (he order of 25 lo 100 mm and (he lengdi 50 to 500 mm. with a

36 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

horizon I al axis and flow directioii. The change in elevation head is lypically negligible and Darcy's law is written

k dp q = - - A

Tw d.\ (10)

where /; is suction, .v is flow distance, and olher variables have been defined previously.

The pore air is typically vented to the atmosphere, and suctions at Hie two ends are maintained at values between about 0 and 0.9 aim using porous stones and manometers (Fig. 12 [71]). To avoid end effecls the suction is typicnily measured al two or more points along the length of the sainple. For suctions greater than about 0.9 aim, the pore air pressure may be raised lo dcvelo]5 suctions up to about 15 atm. The lest may be repealed at different suctions to yield a relationship between conductivity and suction.

II a iclationsliip between conductivity and water content is desired, then (1) tesls m;iy be performed on replicate samples with the water content measured destiucti\ely after each test; (2) the tests may be performed by using a single sample, with the water contents measured nondestructively using neutron backscaltcring techniques, or by weighing the enlire sample and apparatus and obiaining the dry weight at the conclusion of the lesl; or (3) the relation­ship lietween water content and suction may be measured on a separate sam­ple. Tlic iclationsliip belween water content and suction may have hysteresis, so measu rements are taken by either wetting or drying a sample through the range of suctions of interesl.

Several variations of these testing procedures have been used. In one [72] a pore water pressure gradient is ajiplied in one direction and an air pressure giadient of the same amount in the opposite direction, leading to flow with a constant suction, thus constani waler content. In another [73], water is in­troduced ;i( a cons(anl suction at one end and is removed at a constant rate of c\apoin(ion n( the exit end.

Instantaneous Profile Method—In this method [74-76] n long cylindrical sample of soil, typically with a horizontal axis, is provided with a number of suclion probes arranged along the length of the tube (Fig. 13). The soil is ini-(inlly in hydraulic equilibrium, and then the hydraulic conditions al one end ("near end") are altered. The alteration may be the imposition of a constant or lime-dependent suction, either above (outflow) or below (inflow) the suction in the soil. Of a constant or time-dependent impo.sed inflow. Non-steady-state sce|Kige develojis within the samiile.

fhe \-olumc of water, V„ , between any probe and the "far end" (away from Ihe end uheie hydraulic conditions were altered) is

r i

V„ i = OA d.y ( I I )

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 3 l

To C o n s t a n t H e a d W a t e r S u p p l y

To A i r Supp ly

T e n s i o m e t e

Out f l ow

FIG. \2 —Cell for steady-slate method of measurement of hydraulic conductivity of unsaiuraied soil (after K l u t e l 7 \ / I ,

- S u c t i o n P r o b e s

m I n f l o w ^ :

O u t f l o w I Soi l

N e a r E n d '

/ ^ V W \ \ \ \ \ \ \ \ \ \ \

1 ^

Far E n d '

FIG. ]^—A|lpurlllns for laboratoni measurements oJ hydraidic conduclivny nf irn.iiiluraird soils with the instantimcous jirnfile metiiod.

where .\-, is (hc .v coordinate of any jjiobc, " i " ; L is thc length of the s;im|3lc: 6 is the volumetric water content (rtitio of the volume of pore water to the total volume); nnd A is the tolal cross-.sectional area. I he change in V„| in relation to time, dV,,.-,/dl, is the flow rate. q-,.{Ec] I). The volumetric walei

38 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

coiiieiils (Ei | 11) arc oblained from thc measured suctions, p- . and curves of /' versus /), eidier by using n numericnl integration .scheme or by fitting an niialylic function to the ^-.v c u u c and integrating the function. The average hydraulic gradient at the probe, i, can be approximated as

(dp/d.x)-, = (pi+i - p i - | ) / 2 A x (12)

where /); , , and / ) i_ | are the measured suctions at adjacent probes, and A.v is liie spacing of the probes. Again, higher order finite difference equations can bc used or a fitted analytic function can be differentiated. The conductivity al the probe, A'i, is then calculated from

d V ^

dt

dji

d.x (13)

I he condtictivily may be calculated for each time step. A/, for each node past whicli there has been a measurable flow, thus leading to a large nuniber of obscrvnlioiis which can bc plotted ngninst waler content, suclion, or degree of saltiialion.

Ill using (bi.s method it is essential lo have an accurate d-p relationship. Ihe most satisfactoty procedure is to introduce, or remove, moisture at a

slow p.-ice so the wetting or drying front is spread oul. Just before the leading edge of (he front reaches the far end of the sample, the test is stopped and water contents and suctions are measured al each probe, thus yielding a O-p relationship for the sample tested. Measurement of the O-p relationship on a replicate sample .set up solely for this purpo.se may provide useful supplemen­tary data.

A sleady-rate-of-inflow test, starting with a nearly diy sample and conlinu­ing until the soil near tbe entrance end is nearly saturated and the welling front has reached the far end, requires aboul 2 weeks.

I'ressine-Platc Outflow T e s t ~ \ n this test, a sample of soil is placed on a saiuintcd, fine, porous plate in a pressure vessel (Fig. 14). An appropriate air picssiiie is api^lied, the water pressure in the plate is maintained at atmos­pheric incssure, and the sample is given time to come to equilibrium. Then the ail pressure in the vessel is suddenly increased (or decreased), thus gener-nlinu a uniform excess pore water piessure in the sample. The excess water drains out through the porous plate. The rate of outflow from nn element of area, dA , and height, dz , is

d(dV,,)

dt

3 ^

dz dz

Substiditc

7HdVJ_ ^ d(dVJ c[p_ dt dn ,ll

(14)

(15)

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 39

To Air S u p p l y

In terconnecte Grooves for F l u s h i n g

F lush L ine

] To O u t f l o w M e a s u r e m e n t

Sys te m

FIG. \A — Tyincal cell for pressure filale outflow tests (after Klule p \ / l .

Insert Eqs 9 and 15 inlo Eq 14, assume k is constant, and factor the resulting equation lo

dp k

dt " y J d O / d p ) dz^ " d-J TTlL = o - ^

where 0 is again the volumetric water content, and D is called thc diffusiv­ity. Equation 16 is identical to the differential equation governing onc-di-meiisional consolidation (Ref 7, page 228) and has the same solution. 1 he av­erage degree of drainage, U, is thus

oo j

U = (AV„,),/(A V , ) , = E - r - T e x p ( " M 2 7 - ) ).i = n.i M ' -

( 1 7 )

where AV„. is (he volume of water outflow with siilisciipts t nnd u indicating - ' - - I i : 1 . . u : I . . 1 . . „ . , , !

40 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

M = — ( 2 m + I)

and

T = Dt/L^

(18)

(19)

where T is the time factor, t is elap.sed time, and L is the height of the sam­ple. The diffusivity can be calculated using a log t or VT-fitling methods [77], or by methods recommended in the agricultural literature [77j.

Sources of Error in Laboratoiy Tests Using Partially Saturated Soils

The sources of en-or involved in laboratory testing of partially saturated soils are similar to those previously discussed for saturated soils, but consid­erably less information is available. Errors involved with use of nonrepresen­tative samples, smear zones, and incorrect flow directions need no further commenl. Growth of microorganisms again resulls in difficulties, and use of a O.l percent phenol solution [78] or solutions of mercuric chloride, thymol [40\. or formaldehyde is recommended to niinimize biological aclivity. .Several sources of error desen'c special mention, either because of availabil-ily of data or becau.se of their unique nalure in testing partially saturated soils.

Chemical Effects

If the suction in a sample is constant, but there exists a variation in electro-l}le concentration, the pore water will flow in the direction of increasing elec-liolyte concentration, and the electrolyte will diffuse in the opposite direction until equilibrium is finally established. Thus, permeation of a sample with a fluid of different electrolyse concentraiion will lead to diffusive flow of water in addition to flow induced by variation in suction. When the conductivity is high the flow induced by electrolyte gradients is probably too sniall to be of much inlerest, but as k drops, the importance of diffusive flow may increase. Letey et al \78] wrote Darcy's law in a form similar to

q = k dp

Tvv d.x

Jh dw d.x

(20)

uhere TT is the osmotic pressure, and k„ is the osmotic conductivity. Measure­menls of A, were obtained by maintaining a constant difference belween the electrolyte concentration across a sample of partially saturated soil. The ralio of A-„/A- increased from essentially zero for nearly saturated soil to about 0.16 at a suction of about 0.66 atm. No data exist for higher suctions.

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 41

Temperature Effects

What little evidence that exists of temperature effects [79] indicates that an increase in lemperature may reduce the thickness of water films at con-slant suclion, thus decreasing the conductivity, but also reduces the viscosity of the water, thus increasing conductivity. The nel result is that temperattire changes of the order of ten degrees Celsius cause changes in the p-k relation­ship that are smaller than experimental scatter.

Filter Impedance

In steady-state and instantaneous profile measurements, tensiometers or psychromelers are placed at various localions along the length of a sample to eliminate end effecls. In the pressure plate outflow lest, however, a nonfreely draining porous stone may retard flow and tead to significant errors. At­tempts lo account for filler impedance [SO] have produced a tedious method which, however, does not require knowledge of the filter impedance.

An allernalive approach is to derive a Fourier series solution for the one-dimensional non-steady-slate flow problem with an impervious upper bound­ary and a non-freeiy draining lower boundaiy. The solution is

U = 1 E C„sin^(r„)exp(-;-„27') r i = l

(21)

where

u = Q7Qu (22)

and Q, and Q^ are the volumes of outflow of water at times t and ultimately, respectively. Further, ;•„ represents successive roots of the equations

Rr„ ian r„ - 1 = 0 (23)

in which R is termed the impedance ralio and is defined as

R = {k7k,0iT7i-) (2 1)

where k^ and k are the conductivities of the porous stone and soil, respec­tively, and La and L are the thickness of the porous stone and soil, respec­tively. Also

C„ = 2(/?V„2 -b l)/(/-„2)(/? V + R + \) (25)

and

T = Dt/L^ (26)

where T is dimensionless time (conveniently termed (hc "(ime factor"), and D is the diffusivity, given by

42 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

D — k/my^^. (27)

where

m = dO/dp (28)

and 0 and /) are the volumetric water content and suction, respeclively. Equa­tion 21 is solved for the U - T relationship for suitable range in values of R (Fig. 15). For a particular test, Q, is plotted versus log t and the time, tc,^, corrcsi)onding to half of the ultimate flow {U = 50 perceni) is read from the Cline. From Eqs 26 to 28

T^oy^.mL^ (29)

'50

The solution proceeds as follows:

). Delermine the hydraulic conductivity, k^, and length, L^, of the filter by direct measurement before the test begins.

2. Determine the length of the soil sample, L , and the slope of 0-p curve, m, from measurements for this particular increment of pressure.

3. Assume a trial value of k. Use the value from the last pressure incre­ment if you are on the second or greater pressure step.

4. Calculate the impedance ratio, R (Eq 24).

1 10

T I M E F A C T O R

1 0 0

FIC. 15—Theoretical cun 'es of degree of outflow versus time fncior fnr a pressure pla te outflow test with impeded drainage.

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 43

5. Read 7'5o from Fig. 16. 6. Calculate k using Eq 29. 7. If the new k is the same as the previously assumed value, quit. If not,

use the new value and return to Step 4.

The accuracy of Ihis approiich hinges on knowing Ihe chnracterislics of (he filler, ensuring essentially perfect contact between Ihc soil and the filter, and measuring Q ticcuralcly.

One-Step Pressure Plate Outflow Tests

Based on an analysis by Gardner [81], Doering |<V2| suggested that the pressure plate outflow test could be modified by using only a single large step of air pressure. However, the basis for calculation of A includes so many er­roneous assumptions (such as constant (properties and no filter impedance) that this melhod .seems of little value and should nol be used.

Variable Properties in Incremental Outflow Method

To avoid some of the problenis associated with the one-step outflow method, outflow tests are usually performed with small steps in the applied pressure.

1 0 0

o to

o CO

Lu

OJ

E i -

0.01 0.1 1 10

I m p e d a n c e R a t i o ( R )

1 0 0

FIG. 16—Influence of boundaiy im/iedance on the time factor at which 50 perceni of t i f outflow has occuired.

44 PERMEABIUTY AND GROUNDWATER CONTAMINANT TRANSPORT

However, Ihe increments of pressure must be large enough to produce meastiiable outflow. As a result, substantial changes in water content, and at­tendant changes in k and D, usually occur in each increment. Hence, iiitcr-pie(nlioi) of the data is often difficult and reproducibility of results is less than ideal [S3.84\.

Evolution of Gas ifi Pressure Plate Outflow Test

Prcssine plate experiments performed at suctions greater than 1 atm require suiicriiosilion of an air pressure on the soil lo prevent cavitation of water in the measuring system. A major problem exists with evolution of air bubbles in oiitfloxv tests and thc errors that these bubbles tend to cause in fhe measure­ment of outflow quantities [71[. To minimize this problem, a trap ( 7 / | or [.lump and trap 12 . 1 may be used to remove the bubbles.

Validity of Darcy s Law

D;Ua on the effecl of the hydraulic gradient on flow rates are much more meager for pnrlially saturated soils than for saturated soils. Schwartzendruber |(V5| used data collected by Rawlins and Gardner [86] and concluded that, for the one soil tested, Darcy's law was valid for 35 < 0 < 55 percent but flow rales increa.sed more than proportionally to gradient for 15 < ^ < 35 percent. Similar non-Darcy behavior was reported by Schwartzendruber [47] but Weeks and Richards [75] found Darcian behavior (without presenting diagnostic diagrams), and OLson and Schwartzendruber [87] presented definitive data showing the validity of Darcy's law for narrow ranges in degree of sniiiralion (80 to 89 percent, 73 to 87 percent, 66 to 89 percent, 66 to 83 per­cent) for four soils of rather low plasticity. Hamilton, Daniel, and Olson [76] leport measurements of hydraulic conductivity on a clay compacted over a range in saturation of 25 to 95 percent; the data do not suggest any tendency for k to vaiy wilh hydraulic gradient.

The existing evidence thus suggests that Darcy's law is a useful approxima­tion fot (he q- i relationship in partially saturated soils but is probably nol valid in all cnses, thus leading to the conclusion that gradients used in measuring liydiaiilic conductivity should be as close to those encountered in the field as feasible.

Field Measurements of the Hydraulic Conductivity of Saturated Soils

The hydraulic conductivity is generally measured in the field by drilling a hole in the ground, measuring the rate of flow of water into or out of the hole, ;uul using an appropriate formula to calculate the conductivity. Tesls may be liciformed at constant head, generally by establishing a high head of water in " ' •'•"'" "•-"' ".••"nino nt n r.ntp sufficient lo maintain this head, or wMth a

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 45

variable head, that is, with the head set at a nonequilibrittm vnlue initially nnd then mensured ns a function of time widi no furdier pumping. Acltlidonnl variables to be included in thc equations for k must account for the presence or absence of casing, thc location of the bottom of the casing in relation to thc bottom of the boring, (he shape of a ]iiezonielcr tip if one is used, anisottopy in the soil, soil compressibility, presence of impervious surfaces near the tijx amount of air in the soil, secondaty effects, and, doubtless other effects as well. Analysis shows that rigorous solutions can be obtained in only a few cases with unrealistic soil properties or test geometries. Solutions for more realistic conditions have generally been one of the following three types: (I) replacing the aciual geometiy with a simpler one and obtaining an approximate analytical solution; (2) using a three-dimensional electrical model; and (3) us­ing a numerical method, such as finite differences and a digital computer. The inevitable result of having numerous special cases, complex geometry, and ap­proximate .solutions, is a rather complicated, and .sometimes con(ra(lictoiT, literature. We will review a few of the methods that seem to be in most general use.

Testing Methods

Auger Method—In principle the simplest field test is performed by drilling a hole, withoul the use of casing, and then performing either a constant head or variable head test, using eitiier inflow or outflow. The method is termetl (he auger method by agronomists [88]. In its usual form, (he method involves bor­ing a hole to beneath the water table, pumping the water level down several times to flush out the voids in the soil, and then pumping (he hole doun tigaiti and measuring the water level in the hole as a function of time. The equation for A: was derived by Kirkhnm and van Bavel ILS'91 and applications have been discussed by van Bavel and Kirkham [90] and Kirkham 19/ | . Tlie relevant equation is

k = TT^ r A l l

JbTd W (30)

where A- is the conductivity (L/T) , /• is the radius of the well (L). S is n shape factor (dimensionless), d is the depth of thc bottom of the hole below the water table (L), // is the height of water in the hole (L), and i is the time elapsed since the cessation of pumping (T). Values for the shape factor are shou'ii in Fig. 17 (Ref 92, page 141). The solution applies only for an incompressible soil, a hole drilled down to an impei-vious base, and no drawdown of the water table (keej.i h / d less than 0.2). fo simplify analysis, most users of the auger method nppeni lo assume the presence of an impervious base, l^oersma (Ref 99. pages 223-229) has presented shape factors for an impervious base below the bottom of the hole but only for a range in values of d / r from 6 to 14.

48 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

As a result of the wide use of cylindrical lips there have been a number of studies to evaluate the shape factors. Numerical values of these shape faclors are shown in Fig. 18 [10.50.103-106]. The factors presented by Smiles and Youngs [106] and Al-Dhahir and Morgensleni [104] seem to have Hie best dieoiedcal base and to represent the most reliable values at present.

A number of special conditions may need to be evaluated. Several of these arc discussed in the following section.

Special Conditions and Testing Errors

Anisoiropv — For cross-anisolropic soils, where the vertical and horizontal conductivities are k . and Ai,, respectively, the actual soil may be replaced by an equivalent isotropic soil of conductivity, k„,, where

k„ = TTM (33)

u-<D a 03 .c CO

28

26

24

22

20

18

16

14

12

10

8

6

4

2

0

1 1 I 1

Recommended X ' C u r v e s - > ^ / ^

7 77 73 jS*

: / /Yyy^'-7 7 ^ ^ 7 ^ \7/ '-'' ^ ' ' ^

' y y - ^ 1 Wilkinson [ l03 ] ( t ) J Al-Dhahir » Morgenslern [104]

J ^ 3 HvofSlev [50] ' * Lulhian « Kirkham [105]

5 Raymond & Azzouz [ to] 5 Smiles S Youngs [106]

l . l 1 1 10

L/ D

FIG. 18—Shape factors used by various investigators.

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 49

The transformation is performed by multiplying all horizontal dimensions by 7k77k\, [24.93.94]. For Cases 5 and 6 in Table I, Hvorslcv \.V)] shows that the'equations for isotropic soils can be u.scd for anisottopic soils, provided that the terms L7D are replaced by mL/D where

m = 77777,. (-^^J Values for m can be estimated by perlorming field permeability tests using probes with differing L/D ratios [6.10\.

Compressible S o d - U thc .soil is compressible, then changes in wnter pres­sure in the probe cause swelling or con.solidation in the surrounding soil, anci part of the waler entering or leaving the probe results from volume change in the soil rather than exclusively from steady-state seepage. The resulting prob­lems of inlerprelalion lead lo the conclusion that field lests should preferably be of the constant head type. Gibson [107] originally analyzed this problem for a spherical probe in a semi-infinite soil. He assumed that the change in pressure in the probe simply altered the pore pressure at the probc-soil inter­face (constant total stress). He also assumed validity of a consolidation equa­tion of the heat flow type. The resulting analysis yielded the solution

a = 4wa (1 + -7=^) ^ " (35)

where q is the flow rate (LVT), a is the radius of the spherical probe (L). A is conductivity (L/T), 7,,, is the unil weight of water [force/volume (F/L-^)), Au is the changed pore pressure in the probe (force/area (F/L^)], and T is the time factor (dimensionless) given by

T = cl7a' (- '- )

where c is the coefficient of consolidation, and t is time. A plot should bc made of q versus t - ' ' 7 and the intercept at r ' '^^ = 0 defines 7,,. The per­meability is then given by

k = qn-y.74:raAu C 7i

The variation of q with time, .such as in Fig. 19 [108]. indicates possible cr rors as.sociated wilh ignoring the compressibility of tbe soil. Subsequently, Gibson [109] modified his solution to account for thc fact that a change in water pressure in the probe may also change the total stres.ses in the soil. The shape and slope of the q versus f' ' curve is altered, but the intercept re­mains unaffected.

Head Los.ses in Probe and Surrounding Zone of Incompressible Material-Gibson [110] has afso analyzed the case of a spherical probe of finite conduc­tivity and a further thickne.ss of some other incompressible material of finite conductivity (sand, disturbed soil). In this case, fhe slope, shape, and in-

50 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

let cept of the q versus t~^^^ cur\'e are all affected. The solutions are probably mainly of use in designing a probe that does not retard flow, but a trial solution can be used to find k if necessary. It may be noted that forcing a probe into place is likely to result in the formation of a zone of reduced conductivity around the probe (/f.^2|.

U.w of E.i'cessive Heads—With a probe, use of excess pressures that are near the inilial minor principal effective stress in the soil is likely to cause hydraulic fracturing of the soil and a measurement of a value of conductivity thai is loo high [111,112]. The problem may be particularly acute in the case of measure­meni of conductivity in a slurry trench where part of the weight of the trench backfill may be supported through shear along the sides of the Trench, which leads to arching and lo lower vertical effective stresses in the trench backfill than ex]iccted. BjeiTum el al [7/2] report that the vertical effective stress in such cases may be so low that hydraulic fracturing occurs merely by filling a cased borehole with water. For tesls where the water flows into the probe from the soil, u.se of excessively low heads may lead to problenis wilh cavitation and to formation of a zone of less permeable soil near the probe due lo the in­creased effective stress [10].

Use of "Dirty" Water—Use of dirty vvater in lests where waler flows from the probe into the soil may lead to clogging of the pores of the soil [ 9 1 . 1 0 5 . 1 1 3 ] .

Head Loss in Entrance Tubes—In exceptional cases involving use of long cnleiing tubes of small diameter, significant head losses may occur in these tubes [99].

Sealing—In soils of conductivity less than about 1 X 10~^ cm/s there may bc serious problems in sealing the entrance tubes to the probe [99].

Field Measurement of Hydraulic Conductivity in Partially Saturated Soils

All of the methods discussed previously for field measurements in saturated soils could also be u.sed in partially saturated soils, provided consideration is restricted to inflow tests. However, data from simple inflow tests are difficult to interpret because the waler content, and hence hydraulic conductivity, is con­tinually changing during the lest. Even if steady-state seepage were eventually achieved, water content and conductivity would vary spatially. Thus, more elaborate testing methods are required with partially saturated soils. Two methods are discussed in the following sections.

Inslantaneous Profile Method

Various forms of the instantaneous profile melhod have been used in agronomy to measure the conductivity of shallow, partially saturated soils in Ihc field. Typically, the procedure is as follows. A plot of land, several metres or more across, is ringed with a low dike. Probes for measuring suction (usually

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 51

c E o o

a

2.5

2.0

15

1.0

0.5

0 0 2 0 4 0.6

FIG. 19—Fielil observations of flow rate versus the inverse of the square root of time ifrom .-1/ fJhahir, Kennard , and Morgensleni / \08/J.

tensiometers) arc inserted into thc ground near the center of thc plot nt sevcrnl depths. Probes for measuring water content may also be inserted, fhe plot of land is then flooded with several inches of waler. After the water has seeped into the ground, the plot is covered with a sheet of plastic to prevent evapora­tion. As water percolates downward, suctions are measured as a function of depth and time. The water content is eilher measured directly or estimaied from field measurements of suction and laboratory correlations between water content and suction [114]. The data are reduced using Eqs 10 lo 12.

Several variations in this procedure have been tried (Table 3) [115-123]. In­stead of flooding a plot of land with water, one can wait for a heavy rain. Evaporation may be allowed, but interpretation of the data is unceilain unless the rale of evaporation is known. Some have tried measuring the water content i/l situ with neutron probes and estimating the suction from water con­tent-suction curves measured in the laboratory [118,1I9[.

Advantages of the instantaneous profile method for field measurements in­clude modest equipment requirements and relatively straightfonvard inter­pretation of data. Problems include die following: (1) tensiometers are often installed improperly; (2) water frequently flows inlo access tubes housing ten­siometers; (3) if tensiometers are used, suctions are restricted lo less (hnn about 0.9 atm; (4) water flow at the probe locations may not bc purely one-dimensional; (5) the method is restricted to shallow depth; (6) the plot of land must be level; and (7) testing times may be long in relatively impen'ious soils [114].

Infiltration Through Impeding Layer

In this method, described by Gardner [124], Hillel and Gardner |/25], and Bounia el al [126[, a column of soit in the field is i.solated by pushing a thin-walled tube to a suitable depth. Baker (/271 reports that the best tube

52 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

I

<

•7 E

lil S Q .

Q . 5

2<

GO LD LD 3 0 . O ^ o o

o o -J _: — — —

o o o j= j : .1= i J CJ i J

c E E

£; O C 3 3 3 g -O XJ C C C C C

OJ

I )

H .9

a> QJ

h .9.

u c .2 .,

OJ

9J

C

.. .^

C a j

OJ

b o

c c >, >, c c

^£) ^J3 r--r-i r^ r^

X X X ^ -C! t~-

fN r^j r^

O

X X X O lO

« E-g -g • « 5

rt rt rt !y

o T; T; o

Qi

I F T

S rt •o „ c u

: §

E.I O z

=>D

2. rt

s s

Di

c rt

• c

•o rt

J! o Gi

"c5

u

OJ

5 CQ

C

>

r-1

"c5

OJ

c s

-o

Q

•"n

r\) .— '—' "rt a •-' " (U

OJ c

_ o

ir £

^ cu

"rt

QJ

a. o

Z

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 53

dianieter and length are about 24 and 30 cm, res|iectively. The column is capped with a relatively impervious porous stone, membrane, crust of soil, or other impeding material. Water is ponded on the impeding layer and a small, constant head is maintained long enough for steady-stale seepage (o develop. The rate of flow through the stone is measured with a Mariotte bottle or other suitable device. Developers of the method claim that (his procedure produces a hydraulic gradient of unity in the soil directly beneath thc porous stone: hence the conductivity of the soil is equal to the measured velocity of inflow. Suction in the soil directly beneatii the imiieding layer is a function of the con­ductivity of the impeding layer. Tyincally, tensiometers arc inserted into the ground to confirm that the gradient is one and (o measure the suction coi­responding to the obseived conductivity.

This method appears to offer little advantiigc over laboratoiy tests because a Ihin-walled lube must be pushed into the ground to ensure one-dimensional flow. Essentially identical results could probably be obtained by removing the tube filled with soil and testing the material in the laboratory. For fine-grnined soils, the impeding porous stone would lypically have to be so impervious that accurate measurements of flow rates might be imiiossiblc to obtain. Similarly, in many fine-grained soils, it may be impractical to wait for steady-stale seepage to develop.

Comparison of Hydraulic Conductivities Measured in (lie Laboratory' and Bacl{-Calcula(cd from Full-Scalc Field Observations for Saturated Soils

When the permeability lests are performed to allow a designer to estimate flow rates in the field, the ultimate check on the validity of the laboratoiy measurements is clearly a coniparisoii of predicted flows wilh values measurcti in the field for full-scale projects. For fine-grained soils we find no such com­parison. The reasons seem to include the following:

1. The total amount of water that moves is too small lo be of interest if (he soil is fine grained. Inlerest is coiiceiitrated in cases where waler flow lends to settlement, change in stability conditions, or transport of pollutants.

2. Field conditions are ofteii so complex that there is no means availnble for collecting flows. Even in cases where the obser\'ation could consist of measur­ing the arrival time of a pollutant, the hydrogeologic conditions arc often too complex to allow the field conductivities to be backed out of an analysis.

3. The costs involved in obtaining the laboratory and field measurements have precluded obtaining the data in many cases.

4. In some cases, such as flow around toxic waste disposal sites, the flows will occur over times ranging from decades to centuries, thus making it dif­ficult (o obtain useful obseiTations in reasonable periorls of time.

54 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

One of the best opportunities for estimating field conductivities is obtained by measuring time rates of sctllemenl of wide embankments above soft clays, although in this case the coefficients of compressibility must be oblained from other measurements. Sophisticated analytical techniques are required lo ac­count for the dependency of soil properties on effeclive stress. Several cases have been analyzed to compare laboraiory and field curves of void ratio and coefficients of consolidation versus vertical effective stress [5,128,129], but no values for the conduclivity were reported.

Comparison of Hydraulic Conductivities Measured ui the Laboratory anil /// .Situ

All attempt was made to tabulate data from various sites where field and laboratoiy conductivities had both been measured. In many cases it was necessary to simplify data by reporting average values when there was signifi­cant scatter or by reporting data at only one effective stress. It was often unclear how certain measurements were made, and inferences were drawn from general discussions in some cases. The data are presented in Table 4 \f), IU,UO, 102.103.108,1.30-137]. The range in the ratio of field A/laboratory k is from 0.3 lo 46 000, but nearly 90 perceni of the obseiTations lie in the range from 0.38 to 64. It appears that the major causes of the higher values of the field A arc (I) a tendency lo run laboraiory tests on more clayey samples [6]; (2) the presence of sand seams, fissures, and olher macroslructures in the field which are not represented properly in laboratory tests; (3) the use of laboratory A values back-calculated from consolidation theory rather than directly measured values; (4) measurement of vertical flow k in the laboratory and hori^onlal flow k in the field; (5) the use of distilled water in the laboratory; and (6) air entrapment in laboratory samples.

Larger scale field tests may be performed by isolating an area of soil using a metal wall, flooding the isolated area, and then measuring the inflow and the pore pressure distribution with depth. One of the mosl complete sludies of this kind was reported by Ritchie, Kissel, and Burnett [135]. They isolated two areas, one a 10-m square and the other a 2.5-ni square. Field pore pressure measurements showed thai the flow was straight down with a hydraulic gra-fiient of one. The upper soil was Houston black clay; free drainage occurred at a depth of 175 cm. They also performed laboratoiy permeability tesls. The value of A back-calculated from field observations was 3 X 10~^ cm/s at both sites, compared with laboratoiy measurements of k that varied with sample size but fell within the range of 8 X 10~^ to 3 X IO"** cm/s. To investigate flow through fissures, they also permeated samples with fluorescein, a material (hn( fluoresces in ultraviolet light, and found that the fluid was apparently flowing througii the following percentages of the total area:

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 55

Depth,

5 20 35 50

cm Flow Area, %

IOO 60. 10 2

In this fissured clay, the flow apparently concentrated in the fissures.

Summary and Conclusions

There are significant margins for error in both laboratoiy and field tests. For saturated soils, field tesls are to be preferred, provided they are performed and interpreted properly, because they permeate a larger volume of soil than laboratoiy tests, thus taking into account the effects of macrosfrticture, such as roots and fissures. Field tests are generally best performed by using a cylin­drical piezometer tip, installed by methods that niinimize disturbance, and us­ing the constani head technique, Cui-ves should be prepared of q-i~^'~ and flow continued until dq/dt = 0 and a reasonable estimate of the steady-state q is obtained. More research is needed to develop improved methods of field testing and correlation of predicted and aciual flows.

Laboraiory tests offer the advantage of economy, and for many current ap­plications this consideration is a dominating factor. Laboratory tests on nalural samples should use undisturbed samples of the largest practicable size, and samples should be oriented in the proper direction, typically so that the flow is in the directioii of maximum hydraulic conductivity. The permeant should be a fluid similar to that encountered in the field. Care should be taken to avoid accumulation of air bubbles in the sample. Hydraulic gradients should be kept as low as possible while still allowing lests to be performed within a reasonable time. If these precautions are not heeded, the laboratoiy k may differ from the field values by as much as several orders of magnitude (Table 5).

Field testing for measurement of conductivity in unsaturated soils is at such a rudimentary stage of developmenl that field measurements cannot currenlly be recommended except for agricultural purposes or cases where water will be ponded on the surface of a site. Laboratoty testing methods for unsaturated soils are belter developed than field methods, but many of the problems mentioned previously for saturated soils also apply to unsaturated soils. The best laboratoty techniques presently appear to be as follows: (I) for suctions between 0 and 0.9 atm, the instantaneous profile method with ten-siomelric probes; (2) for suctions between 2 and 80 atm, thc instantaneous profile method with psychrometric probes; and (3) for suctions between 1 and 15 atm, the pressure plate outflow method.

For problems of practical interest, it is clear thai permcafiilhy (ests (lab-

Reference

Skempton and Henkel [130]

Colder and Gass [13I\ Weber 16)

-

-

-

Wilkinson [103\

Al-Dhahir, Kennard and Morgenstem {108]

Raymond and Azzouz [10]

TABLE 4

Site

Bradwell

Netherlands Pismo

Lafayette

Atascadero

La Trianon

Napa River

Frodsham

Fiddler's Ferry

Lyndhurst

—Comparison

Soil

clay

sandy clay silty clay silty clay silty clay silty clay

sandy silty c sandy silty c silty clay silty clay silty clay bay mud

organic silty

silty clay

peat mad

of lah

ay ay

clay

>rator\' and

Laboratory Test"

M KT

KT

KT

M

field 1 vdraulic coi

Field Test*

piezometer rising head suction bellows P

P, C

constant head. flow into cell

ductiviiies.

k cm/s

Laboratory

4.5 1.2 3.2 6.9 2.0 1.1 5.5 4.0 3.9 8.5 2.8 1.6 1.2 3.3 2.9 2.1 3.4 1.4 1.9 1.9 4.2 1.2 1.8 8.0 2.7 1.1

5.0 3.3

X X X X X X X X X X X X X X X X X X X X X X X X X X

X X

10 -^ 1 0 - " 1 0 - ' 10 - ' ' 10-8 io-» 10-8 1 0 - ' 10-8 10-^ 1 0 - ' 1 0 - ' 1 0 - ' 1 0 - ' 1 0 - ' 1 0 - ' 1 0 - ' I 0 - ' 1 0 - ' 1 0 - ' I 0 - ' 1 0 - ' 1 0 - ' 10-8 10-8 10-8

10-5 10-*"

3.7 3.7 1.2 3.4 2.7 1.1 2.5 1.7 1.8 8.5 4.2 2.0 6.1 7.1 9.6 9.1 4.2 4.1 6.2 4.3 3.9 3.3 5.0 5.0

Field

X X X X X X X X X X X X X X X X X X X X X X X X

5.0 X 1.2

3.0 2.1

X

X X

1 0 - ' 1 0 - " 1 0 - ' 10-5

1 0 - ' 1 0 - " 1 0 - ' 10-5

10-^ 10-5

10 -* 10-^ IO-*" 1 0 - ' 1 0 - ' 1 0 - ' 1 0 - ' 1 0 - ' 1 0 - ' 1 0 - ' 1 0 - ' 1 0 - ' 1 0 - ' 1 0 - ' 1 0 - ' 1 0 - '

10- ' ' 1 0 - "

Field k

Laboratory k

0.8 3.1 0.4

4 9 0 0 14 0.7 4.6

43 46 000

1 IS

1 200 51

2.2 3.3 4.3 1.2 2.9 3.3 2.3 0.9 2.8 2.8 6.2

19 11

6.0 64

cn CT)

TI m 3

s m > t33 1 -

^ > CJ

o ID

o c z o s > -1 m 3 O

o 2 -1 > S 2

> 2 H

3J3 > 2 C/3 "D O 31

-\

Bishop and Al-Dhahir [132\

Casagrande and Poulos |601

Balderhead M6 Fiddler's Ferry Selsct Selset Diddington

New Jersey Tumpikc

James |/J.?| Murray ( / j ' / |

Ritchie. Kissel and Burnett Houston

\n5\

Jezquel and Mieu.ssens 1/^21 Bordeau.x

algae AEC(? clav

clay fill clay fill alluvium core clav foundation clay core foundation clay varved clay

" Malava Avonmouth

Houston

-s i l tv clay brown clav blue clav peat

-sil tv clav clav

clav

M M KT M KT KT KT M. H M. V

KT KT KT KT KT M

KT

P, C. 1

p . jetted P. driven W P. jetted w p . driven SD. jetted SD. driven

p-v P P P P 10 X 10 m

constani head. flow into .soil

5.0 5.7 4.2 3.7 3.0 1.0 1.0 3.4 1.0 7.7 3.8 1.2 4.0 2 2

1.0 3.0 1.7 2.0 3.4 5.7 3.5 8.1 4.0 3.0 H.O 5.0 8.0 1.7

5.0

X X X X X X X X X X X X X

>< ><

X X X X X X X X X .X

;< X X'

X

I 0 - ' 1 0 - ' 1 0 - ' 1 0 - ' 1 0 - ' 1 0 - ' 10-8 10-8 10-8 10-8 1 0 - " 10-'8 1 0 - "

10-0 0 - '

1 0 - ' 1 0 - ' IO"" 1 0 - ' 1 0 - ' 10^' ' 1 0 - " 1 0 - ' 1 0 - " 1 0 - ' 10-8 10-8 10-8 10-8 10-8

1.9 1.4 5.2 1.7 8.7

1.1 1.0 5.0 8.0 7 9 8,4 3.3 1.8

2 2

6 2 4 4

1.0 2.0 1.5 2.0 1.6 3.1 3.1 3.1 3.1 2.6 2.6 2.5 4.0 2.0 2.0

X X X X X X X X X X X X X

X X X X X X X X X X X X

X X X X X X X X X

10- ' ' 10-0 1 0 - ' 1 0 - ' 10-8

1 0 - ' 10-8 10-8 10-8 10-8 1 0 - " 1 0 - " I 0 - "

I Q - " 1 0 - ' 10-0 lO-o 10-5

1 0 - " 10-5

IQ-o 10-5 10-0 10-5 10-5 1 0 ' 5 10-5 10-5

10"-' 1 0 " ' 1 0 - ' I 0 - ' 1 0 - ' u ) - '

38 2.5 1.2 0.5 0.3 1.1 1.0 l.s 8.0 1.0 2.3 2.8 0.4

1.0 0.1 3.0 1.0

20 2

100 6.1 8.6

10 47 5.4 8.9

38 7.8 0.9 3.2 5.0 5.0

11.8 4.0

O 2

> 2 O O > 2 m i— O 2 X -< O

> C

O

O

o 2 a c o

tn - . J

58 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

I

5--

0 0 0 0 t ^ 0 0 0 ~ ' ' ^ 0 0 0 0

TO 30 X l 00 ( 0

X r-1

1 0

X (-•1

.~.

1 0

X L O

_

1 0

X vO

- r

1 0

X -c r - l

1 0

X • o

r g

1 0

X I D

n

1 0

X 0

. T

1 0

X

0 r v l

1 0

X 0

(-Nl

OC

0

X i r ,

I C

0

X -c

» 0

X

-.

» 0

X • a

r—

0

X 0

« 0

X 0

oc

0

3 C

0

X X 0 q

oc

0

X t ^

oc

0

X

q —• — ( - " • j r ^ r n c o i o o o — ' l O

> > > > QL CL CL 0^ I

^ 5 S

c O

O -o

O S

3/) C

•o i3 • = u 3 -O 3 -H .V

E ^

2 ^ 1.1

c -^ o y E 2 3 -S O u

• 5 , rt c 0 CJ

c^ Xl :=: "O

S'S I g

I I I H l l l - l I T I I I ^ Q ^ S X > >..a. < J. O u > ? t/1

u;

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 59

TABLE 5—Summary' of published data on potential errors in laboratoiy permeabihty tests on saturated soils.

Source of Error and References Measured k Too

Low or Too Migh? Published Data on Tyincal (Measured t)/(Cnncct k)

1. Voids formed in sample preparation 2. Smear zone formed during Irimniing 3. Use of distilled water as a permeant

1.72,.? | - 4. Air in sample \4-t] ' 5 . Growth of microorganisms 1- .51

6. Use of excessive hydraulic gradieni [-17,^S]

1, Use of wrong temperature (Eig. 8) 8. Ignoring volume change due to

stress change {Fig. 9) 9. Flowing water in a direclinn other

than Ihe one of highest permeability (Table I)

10. Performing laboratoiy rather than in situ tests (Table 4)

high low low-

low low

low or high

varies hiuh

> 1 < 1

5/1000 lo 1/10

1/10 lo 1/2 1/100 to 1/10

< 1 lo 5

1/2 to 1 1/2 1 to 20

low

usual ly low

1 111 40

< 1/10 CXK) to 3

oratory or field) must be performed with a grea( degree of care and a((en(ioii lo detail. However, just performing the tests properly does nol ensure suc­cessful results. Thorough field investigation to identify zones of ma.ximum and minimum conductivity, and careful selection of samples or layers for testing, are in some respects more im|iortaiil than experimental technitpie. Even with a comprehensive field investigation and suitable experimental technique, some degree of judgment must inevitably be exercised before the resulls are used for field predictions.

Acknowledgments

Allhough the authors have had a long-standing interest in nietisiitemcnt of hydraulic conductivity of soils, their current emphasis has lestiltcd from an effort lo minimize pollulion resulting from burial of toxic industrial and ra­dioactive wasles in fine-grained soils. The work involving radioactive uaste has been sponsored by the Los Alamos Scientific Laboratoty. The nuthois express their npprecintioii to Lamar Johnson, Merlin Wheeler, and Jtimes Steger, of LASI^, for tbeir support. T he project was supervised by Professor J. O. Ledbetter. Jed M. Hamilton was a graduate research assistant on this proj­ect and contributed greatly to the work on measuremciif of hydraulic conduc­tivity of partially saturated soil.

References

| / | Darcy. II.. Ilistoire Des Fouiulataines I'ublii/ue de Diion, Dalmont. Paris. I85(i. pp. .590-591.

60 PERMEABILITY AND GROUNDWATER CONTAMmANT TRANSPORT

|21 (le Wicst. R. J. M. . Flow Through I'brous Media. R. J. M. de Wiest, ed. . Academic Press, New 1'ork. 1969. pp. 1-52.

| i | W\koff, R. D. , Botset, H. G. , Muscat, M. and Reed, D. W., Bulletin of the Ametican .•\ssociation nf Petroleum Geologists, Vol. 18, 1934, pp. 161-190.

Ml ne.H, J.. Hydraulics of Groundwater , McGraw-Hill, New York, 1979, pp . 68-69. 1. 1 Pcllcticr. J. 11., Olson, R. E. , and Rixner, J. J., Geolecliuical Testing Journal, Vol. 2, No.

1. 1'179. pp. 34-43. Ull Weber. W. G. "In Situ Permeabilities for Determining Rates of Consolidation," Highway

Riscnc l i Record No. 243. N.'ilioiial Research Council, Washinglon, D . C . 1968, pp. • U l - l i l .

I ." l liiylor. D. W., Fundamenta l s of Soil Mechanics, Wiley, New York, 1948. |'V| Anclresen, A. and Simons, N. E., l^ioceedings. Conference on Shear Strength of Cohesive

Soils. American Society of Civil Engineers, Boulder, Colo., 1960, p . 705. T'l Landv.i. A.. "Ec)iiipmci)t for Cutting nnd Mounting LIndisturbed Specimens of Clay in

Testing Devices." Nonvegian Geotechnical Institute Publ. No. 56, Oslo, 1964, pp. 1-5. \lll\ Raymond. G. P. and Azzouz. M. M., Proceedings, Conference on In-Situ Investigations

of Soils and Rocks, British Geotechnical Society, London, 1969, pp . 195-203. I / / I Jones. C. W., Special Procedures for Testing Soil and Rock fo r Engineering Purposes,

.A.SIM S T P 47'i, American Society for Testing and Materials, Philadelphia, 1970, pp . l(.-l-171.

| / 2 | U.S. Ucpa i tmcnt of Interior. Earth Manual . Bureau of Reclamation, Denver, Colo., 1974.

| / . i | Bishop. A. W. and Henkel . D. J., The Triaxial Test. 2nd ed. , Edward Arnold, London. 1957.

\l-l\ I.owe. J.. Jonas, E., and Obrician. V., Journal of the Soil Mechanics and Foundations Division, Proceedings of the American Society of Civil Kngiiieers, Vol. 95, No. S M I , 1969. p p . 77-97.

l/.il Wissn. A. E. Z. . Christian. J. T . . Davis. E. H. . and Heiberg, S., Journal of the Soil Me­chanics and Foundat ions Division, Proceedings of the American Society of Civil Ettgi-neei-s. Vol. 97, No. SMIO, 1971, pp. 1393-1413.

\lii\ Rowe. P. VV. and Barden. L., Ceotcchiuque, Vol. 16, No. 2, 1966, pp. 162-170. | / 7 | Milf, I. W., "An Invesligation of f'ore-Walcr Pressure in Compacted Cohesive Soils,"

Technical Memorandum No. 654, U.S. Department of the Interior, Bureau of Reclama­lion, Denver, Colo., 1956.

|/A'| Gihbs. H. J.. Hilf, J. W. . Holtz, W. G. . and Walker, F. C , Proceedings, Conference on Shear Strength of Cohesive Soils, American Society of Civil Engineers, Boulder, Colo., 1960, p. 75.

| /V| Mitchell. J. K., Hooper, D. R., and Campanella, R, G. , Journal of the Soil Mechanics imd Foundat ions Division, Proceedings of the American Society of Civil Engineers, Vol. 1)1. No. SM4. 1965. pp. 41-65.

\2l)\ Bishop. A. W. and Donald. I. B. . Pi-oceedings, Fifth International Conference on Soil

Mechanics and Foiiiidations Engineering, Vol. 1, 1961, pp. 13-21. 12/1 Malyas. E. L,, Air and Water Permeability of Compacted Soils, ASTM S T P 417. Aineri-

ciii Society for Testing and Malerials, Philadelphia, 1967. p p . 160-175. 1221 Terzaghi. K.. Die ISerechnnng der Durchlassigkeitziffer des l o n s aus Dem Verlauf der

llvdrodynanusc-hen Sjiannungserscheinungen, Ditz, Akad, Wissen, Wien Math-Naturw. D l . , Part l lA, Vol. 32. 1923. pp. 125-138.

|2.il ("asagraiule, A. and Fadum. R. E.. Transactions, American Society of Civil Engineers, Vol. 109, 1944, pp . 383-490.

| 2 ' / | Taylor. D. W., "Research on Ihe Consolidation of Clays," M.I . T. Serial No. 82, Massa­chusetts liistilnte of Technology. Boston, Mass., 1942.

12^1 Rowe. I'. VV., Geotechnique , Vol, 16. No. 2, 1959. pp. 162-170. |2o | liishiip. A. W. and Gibson, R. E.. I'he Inflin-nce of the Provisions fo r lloiindai-y Drainage

nn .Strength and Consolulation Characteristics oJ Soil Mensined ut the Tria.vial .Apfuira-iiis. . '\S'I'M S l I ' .161. Americaii Societv (or Testing and Malerials, Philadelphia, 1963, pp. 435-451.

|.?7| lii.'iiichi, NV. C. and H.'iskell. 1. . 1"'.. Jr.. Proceedings of the Anwrican Society for I'esring and Materials, Vol. 63, 1963, pp. 1227-12.14.

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 61

British Geotcchnicil Society.

V(,l, 29. 1965, pp 1.15 |.|0

19. 1971. pp, 151-158. , J, W., Soil Science Sinii-tv of

\2S\ Nightingale, 11. 1. and Bianchi, VV. C . Soil Science, Vol. 110. No, 4. 1970, pp. 221-228, 1291 Rcmy. J. P., Geotechnique, Vol, 23. No. 3, 1973, pp. 454-458, \30\ Van Zelsl, T, VV., Proceedings, Second International Conference on Soil Mechanics and

Foundations Engineering, Rolleidam, Vol. 7. 1948. pp. 52-61. lJ /1 Chan , 11. T. and Kenney. T. C Canadian Geotechnical Journal. Vol, 10. No. 3. 1973.

pp . 453-472. l.?2| VViikinson. W. B. . Jn Situ Investigation in Soils and Pocks

Institiilion of Civil Engineers. London. 19()9. pp. 311-313. 1.7?I Olsen. 11. W. , Soil Scii-nce .Society of ,'\nierica I^roceedings \3-1\ Fireman. M.. ,Sod Science, Vol. 58, 1944. pp. ,137-.J55, | 2 5 | Harris, A. E , . S o d Science, Vol. 32. 1931. pp. 425-446, \36\ l^eevc, R. E.. Bower, C. A., Brooks, R. II, , and Gschwend, T". (i.. Soil Science Society n/

America Proceedings, Vol. 18. 1954, pp. 130-132. 1^7] Lambe. T. W. , Symposium on Permeability of Soils, .ASIM S T P 163, American Society

for Testing and Malerials, Philadelphia. 1954. pp. 56-67. \3S\ Mesri, G. and Olson, R. E.. Clays and Clay Minerals. Vol i.?91 Cassel, K. K., Warrick, A. W.. Niel.seii, D.' R., and Biggar

America Proceedings, Vol. 32, No. 6, 1968, pp. I I A - I V . \40\ Klule, A,. Sod Science, Vol. 113. 1972, pp. 264-276. \4 l \ Smith, R. M. and Browning, D. R.. Soil Science Society of .Anu-rica I 'rnciedmgs. Vol. 7,

1942, pp. 114-119. (•^2) BjeiTum, L., and Hyder. J. Proceedings. Fourth liilcrnatiorial Conference on Soil Me­

chanics and Foundations Engineering. L(nidon. Vol. 1, 1957, )ip. 6-8. |4J1 Christiansen, J. E. , Sod Science, Vol. 57, 1944, pp. 381-390. I'/'/l Johnson, A. 1., Symposium on Soil Permeability, A S l ' M S I P 163, American Society foi

Testing and Materials, Philadelphia, 1954, pp. 98-114. \45\ Allison, L. E., Sod Science, Vol, 63, 1947. pp. 439-450. [• 6] Olsen. H. W., Water Resources Reseaich, Vol. 2. No. 2, 1966, pp. 287-294. K 7 | Schwariy.endniher. D. , Soil Science Societv of .America Proceedings, Vol. 32. No. 1. 19()8,

pp . 11-18. y S ] Mitchell, J. K. and Younger, J. S., Permeability and Capillarity of Soils. ASTM .Sl P -117.

Americnn Society for Testing and Materi.ils, Philadelphia. I' '67. pp. 106-1,19. \49\ Gairon, S. and Schwartzendruber , D. , Soil Science Societv of .Anu-rica Proceedings. Vol,

39, No. 5. 1975, pp . 811-817. | 50 | Hvorslcv, M. J., "Time Lag and Soil Permeability in Ground Water Obsena l ions . " Bulle­

tin No. 36. VVater^vays Experiment Station. Vicksburg. Miss., 1951, 15/1 Al-Dhahir, Z. A., /;; Situ Investigations in Soils and Rocks, British Geolechnical SocieH.

Institution of Civil Engineers. London, 1969, p. 305. 1.521 Al-Khahir. Z. A. and Tan , S. B., Geotechniqtn-, Vol, 18, No, 4. 1968. pp, 499-505, \53\ Subbaraju, B. H., Natarajan, 1'. K., and Bhandari , R. K., Troceedings. 8lh Intcrnatioiuil

Conference of Soil Mechanics and Foundations Engineers. Moscow. Vol. 2,2. pp, 217-220.

1541 Lumb, P. and Holt, J. K., Geotechnique, Vol. 18, 1968, pp. 25-36, 1551 Bazctt, D. J. and Brodie, A. F. , Ontario Hydro Reseaich News. Vol. 13. No. 4. 1951. pp.

1-6. 156| Tsicn, S. I., Stabilization of Marsh Dcfiosit. Highway Research Board, Bullelin 115.

1955, pp. 15-43. 1571 Kenney, T. C. and Chan, H. T. , Canadian Geotechnical Journal. Vol. 10, No. 3. 1973.

p p . 473-488. (.Wj "Reporl No. I — Engineering Pmpcrties of T'oundalion Soils al Long Cieck-Foic River

Areas and Back Cove." Halev and Aldrich, Rcporl to Maine State Highwav Commission. 1969.

15'71 VVii, T. H. . Chang . N. Y.. and Ali. E. M., Journal of Geo t r chnua l Fnginecr.s Division.

American Society of Chemical i:ngineers. Vol. IOI. No. GT7. I97H. pp. H99-905, \60\ Casagrande. L. and I'oiilos, S. .1., Canadian Geotechnical Jnunud, Vol, 6. No. 3. I')(i0.

pp . 287-326. lrt/1 Croney. P . , and Coleman, J. D. in Pore Pressure and Suclion in Snd. Iiistitiiliiui nl Civil

Engineers, l intlerworllis, London. 1961. pp . 31-37.

62 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

162] D.iniel, D. E. , Hamilton, J. M. , and Olson, R. E., in this publication, pp , 84-100. 163] Penier. E. R. and Evans, D. D. , Soil Science Society of America Proceedings. Vol. 25,

No. 4. 1961, pp. 173-175. |6-/ | Richards, L. A., Soil Science, Vol. 51 , 1941, pp. 377-386. \63] Olson, R. E. and Langfelder, L. i , . Journal of SoU Mechanics and Foundations Division,

Proceedings of the American Society of Civil Engineers, Vol. 91 , No. SM4, 1965, pp. 127-150.

1661 Fukada, Tl., "Mechanism of Soil Moisture Exiraction from a Pressure Membrane Appa­ra tus . " Highway Research Board, Special Report 40, National Research Council, Wash­inglon, D . C , 1958. pp. 73-77.

|671 Coleman. J. D. and Marsh. A. D , . Journal of Sod Science. Vol. 12, 1961, pp. 343-362. 16rS'l Spanner, D. C . Journa l of E.xperimental Botany. Vol. 2, 1951, pp . 145-168. 1691 Wiebe, H. H. . Campbell', G. S., Gardner, W. H., Rawlins, S. L., Cary, J. VV., and

Brown, R. W. . "Measurement of Plant and Soil Water Sta lus ," Utah Agricultural Exper­iment Station, Bulletin 484, Utah State University, Logan, Utah, 1971.

17 ;) Nielsen. D. R. and Biggar, J. VV., Soil Science. Vol. 92, 1961, pp. 192-193. |7/1 Klule. A., Methods of Soil Analysis, C. A. Black Ed. , Monograph 9, American Societv of

Agionomy, Madison, Wis. , 1965, pp. 210-221. 172| Corey, A. T . , Soil Science Society of Ametica Proceedings. Vol. 21, 1957, pp. 7-10. 17.?| Nielsen, D. R., Ki rkham, D. , and Perrier, E. R., Soil Science Society of America Pro­

ceedings, Vol. 24, No. 3, I960, pp . 157-160. 17^1 Richards, S. J. and Weeks, L. V., Soil Science Societv of America Proceedings, Vol. 17,

1953, pp. 206-209. | 75 | Weeks. I. V. and Richards, S. J., Sod Science Socierv of America Proceedings, Vol. 31 ,

1967, pp . 721-725. |761 Hamilton, J. M., Daniel, D . E. , and Olson, R. E., in this publication, pp . 182-196. |77 | Gardner , VV. R., Soil Science Society of America Proceedings. Vol. 20, 1956, pp .

317-320. 17('il Letey, J., Kemper , W. D. , and Noonan, L., Soil Science Societv of America Proceedings,

Vol. 33. 1969, pp . 16-18. |791 Haridasan, M. and Jensen. R. D . , Sod Science Societv of America Proceedings, Vol. 36,

1972, pp . 703-708. |W1 Knnze, R. J. and Kirkham, D. , Soil Science Society of America Proceedings, Vol. 26,

1962, pp . 421-426.

]Kl] Ca rdne r , W. R.. Soil Science Society of America Proceedings, Vol. 26, No. 4, 1962, p , 4 0 4 . |.S21 Doering, E. J., Soil Science, Vol, 99, No. 1. 1965, pp . 322-326.

Irf,?! Jackson, R. D . , Van Bavel, C. H. M., and Reginato, R. J., Soil Science, Vol. 96, 1963 pp . 249-256.

|*/1 Davidson, J. M. , Biggar, J. W . . Nielson, D. R., Warrick. A. VV., and Cassel, D. R., Water in thc Unsa tura ted Zone, International Association of Scientific Hydrology, Vol, 1 1968, pp. 214-223.

1.V51 Schwartzendruber , D . , Sod Science Socieiy of America Proceedings, Vol. 27, 1963, pn 491-495.

1.T6I Rawlins, S, L. and Gardner , W. H., Soil Science Society of America Proceedings, Vol. 27, 1963. pp. 507-511.

1.V71 Olson, T. C, and Schwartzendruber , D., Sod Science Society of Ametica Pioceedings Vol. 32, No. 4, 1968, pp . 457-462. " '

\iS,S\ Boersma, L. in Methods of Soil Analysis, C. A. Black Ed., Monograph 9, American Socieiy of Agronomy, Madison, Wis. , i965, pp . 222-233.

[•W| Kirkham, D. and van B.avel, C, H. M., Sod Science Socieiy of America Proceedings Vol

13, 1948, pp . 75-82.

| W | Van Bavel, C. H. M. and Kirkham, D. , Soil Science Society of America Proceedings Vol

13, 1948. pp . 90-96.

l y / | Kirkham, D. , .Symposium on Permeahilily of Soils, A S T M STP 163, American Society for Testing and Materials, Philadelphia, 1954, pp. 80-97.

]'/2\ Spangler, M. G. , Sod Engineering, International Textbook Co., Scranton, 1951. ]'.I3] Saiiisioe, A. P. , Zeitschrift fiir Angewandlc Maiheniat ik und Mechanik, Vol 11 April

1931. pp, 124-135.

OLSON AND DANIEL ON HYDRAULIC CONDUCTIVITY 63

]'J'I] Dachler. R., Grundwasserstromung. Julius Springer, VVicn. 1936. 1951 Foichcimer. P., Hydtaidik, 3rd. ed. , B, G. Teubner, Leipzig. 1930. 196] Harza, L. F. . Ttansactions. American Society of Civil T-jigiiieers. Vol. 100. 1935. pp,

1352-1385. 197] Casagrande, A,, Journal of thc Poston Socieiy of Civil Engmeers. Vol. 36, April 1949. pp,

192-221. 1951 Bjerrum, L. and Johanncssen. I., Pore Pressme and Suction in Soils. Institution of Civil

Engineers, London. I960, pp. 108-111, 1991 Wilkes. P. F,. Geotechnique, Vol, 20. No. 3, 1970, pp. 3.10-.113.

]100] Pany . R. Tl. G. . Geotechnique. Vol. 21, No, 2, 1971, pp. 103-167, [IOl] Vaughn, P, R., Geotechnique, Vol. 14. No. 3. 1969, pp. 405-113. | / 0 2 | Jezcqucl. J. F. and Mieussens, C . In Situ Measurement of Soil Ptiipenies. American

Society of Civil Engineers. Vol, 1, 1975. pp. 208-224. ]I03\ Wilkinson. VV. B., Geotechnique, Vol. 18. No. 2. 1968. pp. 172-194, ]l()7] Al-Dhahir, Z. A. and Morgenslern, N. R.. Soil Science, Vol. 107, No. 1. 1969. pp. 17-21, | / 0 5 | Lulhian. J, N. and Kirkham. D. . Sod Science, Vol. 68, 1949, pp. 349-358. 1/061 Smiles. D. E. and Youngs, E, G. . Sod Science, Vol. 99, 1965, pp, 83-87, ]107] Gibson, R. E.. Geotechnique, Vol. 13, 1963, pp. 1-11. ]10S\ Al-Dhahir. Z. A., Kennard , M. F, . and Morgensleni . N, R.. In Situ Investigations in

Soils and Rocks, Institution of Civil Engineers, London, 1969, pp. 265-276, 1/091 Gibson, R. E., Geotechnique, Vol. 20, No. 2, 1970, pp . 193-197. ]II0] Gibson, R. E.. Geotechnique, Vol. 16, 1966, pp. 256-259, ] l l l ] Bjerrum, L. and Andersen, K. 11.. Proceedings, Fifth European Confeience on Soil Me­

chanics and Foundation Engineering, Madrid, reprinted in Norwegian Geotechnical In­stitute Publ. No. 91 , 1972, pp . 29-38.

1//2) Bjerrum, L., Nash, J. K. T. L., Kennard , R. M.. and Gibson, R. E., Geotechnique. Vol, 22, No. 2. 1972, pp . 319-332.

l//.?l Frevert, R. K. and Kirkham, D. , Proceedings, Highway Research Board. Vol. 28. 194& pp . 433-442.

] lM] Baker, F. G. , Veneman, P. L. M. , and Bouma, J. ceedings. Vol. 38, 1974, pp. 885-888.

1//51 Richards, L, A., Gardner , W. R., and Ogata , G. , ceedings. Vol. 20, 1956, pp. 310-314.

I//61 Ogata , G. and Richards, L. A., Soil Science Society of Anu-rica Proceedings. Vol. 21. 1957, p p . 355-356.

1//7) Nielsen, D. R., Davidson, J. M., Biggar, J. W. , and Miller, R. J., Hilgardin. Vol. 35. 1962. p p . 491-506.

]118] Rose, C, W., Stern, W. R., and D r u m m o n d , }, . Australian Journal of Soil Research. Vol.

3 , 1965, pp . 109. | / / 9 j F^ose. C VV. and Krishnan, A., Soil Science, [120] Van Bavel, C. H. M. , Stirk, G. B. . and Brust. K. J., Soil Science Society of America Pro­

ceedings. Vol. 32, 1968, pp. 310-317. 1/2/1 Davidson, J. M. , Stone, L. R., Nielson, D. R., and Larue. M. E., Water Re.umtces Re­

search, Vol. 5, 1969, pp . 1312-1321. 1/221 Hillel, D . , Krentos, V. D. . and Stylianou, Y., Sod Science. Vol, 114, 1972, pp, 395-40(i, 1/2J1 Nielsen, D. R., Biggar, J. W. . and Erb, K. T. . Hilgardia. Vol. 42. No. 7, 1973. pp,

215-259. ]124] Gardner . W. R., Soil Science Societv of America Pioceedings, Vol. 34, 1970, p|,i,

832-833. 1/25) Hillel, D. and Gardner , W. R., Soil Science. Vol. 109, 1970, pp. 149-153. | /26t Bounia, J., Hillel, D. I., Hole, F. D . , and Amerman, C. R., Soil.Science Socieiy of .Amer­

ica Proceedings. Vol. 35,1971, pp. 262-264. 1/271 B-.ikcr.V. G. . SoU Science Society of Anu-rica Proceedings. Vol, 41. 1977. 1029-1032, |/2<.S'| Olson, R, E., Daniel, D. E.. and Lin. 1 . K.. Proceedings, Specially Conference on Anal\-

sis and Design in Geolechnical Engineering, American Society of Civil Engineers, Austin, I'ex., 1974, pp. 85-110.

1/291 Olson, U. E. and Ladd, C. C , Journa l of Geotechnical Engineeritig Division. Proceedings of Ihc American Socieiy of Civd Engineers. Vol. 105, No. 1, 1979, pp. 11-30.

, Soil Science Socieiy of .Ann-rica Pro

Soil Science Society of .'\merica Prn

Vol. 103, 1967, pp. 369-373,

6 4 PERMEABILITY AND GROUNDWATER CONTAMINANT TRANSPORT

130\ Skempton. A. VV. and Henkel, D. J., Pore Pressure and Suction in Soils, Institution of Civil Engineers, London, I960, pp. 81-84.

131] Cjolder. H. A. and Gass, A. A., Field Testing of Soils, A S T M STP 322, American Society for Testing and Malerials, Philadelphia, 1962, pp. 29-45.

/,72| Gisliop. A. W. and Al-Dhahir, Z. A., In-Sitti Investigations in Soil and Rocks, Institution of Civil Engineers. London, 1969, pp. 251-264.

/j,?| James. P. M., Quarterly Journal of Engineering Geology, Vol. 3, No. I, 1970, pp. 41-53. 134] Murray, R. T.. "Embankments Constructed on Soft Foundations: Settlement Sludy at

.Avonmouth," Road Research Laboratoiy, Report LR4I9, 1971. /.i.^l Ritchie. J. J . . Kissel, D. E.. and Biirnelt, E., Soil Science .Society of Amcricn I'roceed-

ings. Vol. 36, 1972, pp. 874-879. / J 6 | Goodall, D, C and Quigley, R. M. , Canadian Geotechnical Journal, Vol. 14, No. 2,

1977, pp. 223-236. /,'71 Mieus.sens, C , and Diicasse, P,, Canadian Geotechnical Journal, Vol. 14, No. 1, 1977,

pp. 76-90.

(Vi;\c.cVsrr.^D

Performance Analytical Inc. Env i ronmcn t j l Tcsrin^: inJ (,j.>n,^ijlnnc

PERFORMANCE ANALYTICAL INC.

RESULTS OF ANALYSIS

V6-3 f 7

Client: Analytical Technologies, Inc.

Client Sample ID: 91212003-02 (07/30/91)

PAI Sample ID: 9102707

Test Coda: Analyst: Instrument IDi Verified by:

GC/MS Mod. EPA TO-14 Michael Tuday Finnigan 4500C/Tekmar 5010 Chris Casteel

Matrix: Date Received: Date Analyzed: Volume Analyzed:

Tedlar Bag 07/31/91 07/31/91 S, 08/01/91 0.001 Liter 0.00005 Liter

CAS /

74-87-3

75-01-4

75-00-3

74-83-9

67-64-1

75-59-4

75-35-4

75-09-2

75-15-0

76-13-1

156-60-5

156-59-2

75-34-3

108-05-4

78-93-3

67-66-3

107-06-2

71-55-6

71-43-2

56-23-5

78-87-5

7S-27-4

79-01-6

1 10061-01-5

COMPOUND

CHLOROMETHANE

VINYL CHLORIDE

CHLOROETHANE

BROMOMETHANE

ACETONE

TRICHLOROFLUOROMETHANE

1,1-DICHLOROETHENE

METHYLENE CHLORIDE

CARBON DISULFIDE

TRICHLOROTRIFLUOROETHANE

TRANS-1,2-DICHLOROETHENE

CIS-1,2-DICHLOROETHENE

1,1-DICHLOROETHANE

VINYL ACETATE

2-BUTANONE

CHLOROFORM

1,2-DICHLOROETHANE

1,1,1-TRICHLOROETHANE

BENZENE

CARBON TETRACHLORIDE

1,2-DICHLOROPROPANE

BROMODICHLOROMETHANE

TRICHLOROETHENE

ClS-1,3-DICHLOROPROPENE

RESULT

(MG/M^)

ND

ND

ND

ND

ND

190

4900

1100

ND

25,000

ND

ND

210

ND

ND

ND

ND

1900

7.3

ND

71

ND

360

ND

DETECTION LIMIT (MG/M^)

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

RESUL.T

(Pt'M)

ND

ND

ND

ND

ND

33

1200

330

ND

3300

ND

ND

54

ND

ND

ND

ND

360

2.3

ND

16

ND

68

ND

DETECTION LIMIT (PPM)

2.5

2.0

1.9

1.3

2.1

0.90

1.3

1.5

1.6

0.65

1.3

6.3

1.3

1.4

1.7

1.1

1.3

0.90

1.6

0.80

1-1

0.75

0.95

1.1

ND " Not Detected TR • Trace Level - Below Indicated Detecion Limit

2:954 Osborne Street. C^ni-a.i Pirk, C,A 915C4 • Th.^ne "<1:: 709-11"- ' • Fax >•-:•< TO^-I'-ilS

Performance Analytical Inc. Environment,li Te^tini; , 'nj t. .in-olrini;

PERFORMANCE ANALYTICAL INC.

RESULTS OF ANALYSIS (Continued)

VB-3

Client: Analytical Technologies, Inc.

Client Sample ID: 91212003-02 (07/30/91)

PAI Sample ID: 9102707

Test Code: Analyst: Instrument ID; Verified by:

GC/MS Mod. EPA TO-14 Michael Tuday Finnigan 4500C/Tekmar 5010 Chris Casteel

Matrix: Date Received: Date Analyzed: Volume Analyzed:

Tedlar Bag 07/31/91 07/31/91 & 08/01/91 0.001 Liter 0.00005 Liter

CAS #

108-10-1

10061-02-6

79-00-5

108-88-3

124-48-1

119-78-6

127-18-4

108-90-7

100-41-4

75-25-2

100-42-5

1330-20-7

79-34-5

541-73-1

106-46-7

95-50-1

COMPOUND

4-METHYL-2-PENTANONE

TRANS-1,3-DICHLOROPROPENE

1,1,2-TRICHLOROETHANE

TOLUENE

DIBROMOCHLOROMETHANE

2-HEXANONE

TETRACHLOROETHENE

CHLOROBENZENE

ETHYLBENZENE

BROMOFORM

STYRENE

TOTAL XYLENES

1,1,2,2-TETRACHLOROETHANE

1,3-DICHLOROBENZENE

1,4-DICHLOROBENZENE

1,2-DICHLOROBENZENE

RESULT

(MG/M3)

ND

ND

ND

ND

ND

ND

210

ND

ND

ND

ND

ND

ND

ND

ND

ND

DETECTION LIMIT^ (MG/M^)

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

5.0

S.O

5.0

RESULT

(PPM)

ND

ND

ND

ND

ND

ND

31

ND

ND

ND

ND

ND

ND

ND

ND

ND

DETECTION LIMIT (PPM)

1.2

1.1

0.90

1.4

0.60

1.2

0.75

1.1

1.2

0.49

1.2

1.2

0.75

0.84

0.84

0.84

ND » Not Detected TR • Trace Level - Below Indicated Detection Limit

ZCS'54 O s b o r n e S t r ee t . Oanoi;D Pnrk. C A 91304 • P U m c ^L'' 70^-1 1 5-> • F.,:v 5!.^ IC-^-l 'M'

P\-W c^cVv'ffv-^ H,

BEHAVIOR OF ORGANIC CONTAMINANTS

IN THE SUBSURFACE

Notes for Lecture by

Charles Graf, Manager Groundwater Hydrology section

Arizona Departinent of Environmental Quality

October 1991

CONTENTS

I. Introduction

II. Definitions

PAGE

III. Distribution of Contaminant Mass in the Soil, Water, and Air Phases

IV. Calculation of Concentrations 14

V. Retardation of an Organic Chemical in Groundwater 18

VI. Effect of Equilibrium Partitioning on Groundwater Cleanups^ 21

VII. Effect of NAPLs on Mass Distribution Equations

Appendix A. Properties of Selected Organic Contaminants of Soil and Groundwater

Appendix B. Mass-Distribution Diagrams for Selected Contaminants

Appendix C. Retardation Factors for Common Soil and Groundwater Contaminants

BEHAVIOR OF ORGANIC CONTAMINANTS IN THE SUBSURFACE

I. Introduction

V'Jhen an organic contaminant is released to the subsurface, it will

eventually equilibrate itself between two phases in the saturated

zone (soil and water) and among three phases in the unsaturated

zone (soil, water, and pore air). The precise distribution of 'the

chemical among these phases is primarily dependent on two

properties of the chemical, Organic Carbon Partition Coefficient

(K<,) and Henry's Law Constant (K^) , and one property of the soil

matrix, the content of organic carbon.

These notes outline how to:

1. Calculate the percentage of contaminant mass in each of

the three phases — soil, water, air { S e c t i o n I I I ) .

2 . Calculate concentrations of the contaminant in soil,

water, and pore gas knowing any one of the three

concentrations ( S e c t i o n IV).

3 . Determine the velocity of the contaminant in the

groundwater with respect to the velocity of the

groundwater itself (the velocity of the contarainant is

retarded with respect to the velocity of groundwater)

(Section V).

4. Estimate the effect equilibrium partitioning has on

prolonging cleanup time-frames.

II. Definitions

M, = Mass of contaminant in soil phase [mg]

M^ = Mass of contaminant in water phase [mg]

M3 = Mass of contaminant in air phase [mg]

MJ = Total mass of contaminant [mg]

fj = Fraction of contaminant mass sorbed to soil

f - Fraction of contaminant mass dissolved in water

f = Fraction of contaminant mass in soil air

f^ = Fraction of organic carbon in a soil

f ^ = Fraction of organic matter in a soil

C„ = Concentration of contaminant in water phase

[ug/cm^ (= mg/L) ]

CJ = Concentration of contaminant on air phase

[ug/cm^ (= mg/L) ]

CJ = Lab-reported concentration of a contaminant in a soil

sample analyzed by standard methods [ug/g (= mg/kg)]

C3 = True concentration of contaminant in soil phase

[ug/g (= mg/kg)]

K = Hydraulic conductivity (cm/sec)

i = Hydraulic gradient

V = Darcy velocity (cm/sec), or equivalently, specific

discharge (cm-'/cm -sec)

Vp = Actual groundwater velocity (cm/sec)

v = Contaminant velocity (cm/sec)

Kp = Soil partition coefficient (also often referred to as the

soil distribution coefficient, Kj) [cmVg]

III. Distribution of Contaminant Mass in the Soil, Water, and Air

Phases

For a one-centimeter volume of soil or aquifer:

K = c ,n e.

M^ = C^ n (1 - 0 7

M^ = C. i i - n ) p - C„ p s y b

M.J. = M^ + M^ + M^

(1)

(2)

(3)

(4)

Substituting (1), (2), and (3) into (4) yields;

MJ. - C^(l-22)p + C ^ n Q ^ + C^ n { l - Q 7 ) (5)

Consider now the equilibrium partitioning of a chemical between

air/water and soil/water. Assume that a 1 cm^ volume of air

overlies a 1 cm^ volume of water containing a dissolved organic

chemical.

A i r

W a t e r

Cl.

' '

• J '7 '

C ^Cug/cm33

c .^Cua^cmS^

K» = H ^ H

R<jT (8.2054 K 10-5) (T) (7)

I f T ^ 293''K (= 20' 'C o r 68°P)

KJJ = 4 1 . 5 9 / / (8)

The water/soil interaction is somewhat more complex. Consider the

diagram below indicating a volume of soil in contact with water:

V/ater

O"O0nic carbon adher ing t o

s o i i por t i c l e

So I I p a r t I d e

C ,^Cug/cm3;)

C c-Cug/g])

In a soil-water system, an organic chemical eventually reaches

equilibrium between the soil and water phases in an analogous

manner to air-water partitioning. The equilibrium equation

describing the relationship between soil and water phases can be

particles. A portion of the organic matter consists of active

organic carbon which is the actual sorbing medium. The greater the

organic carbon content of the soil, the greater the amount of

chemical sorbed by the soil. In fact, the sorption relationship is

linear - doubling the amount of soil organic carbon will double the

chemical concentration in the soil. The content of organic carbon

in a soil is usually represented as the weight fraction, f^. In

some literature, this property is referred to as fraction of

organic matter, f „. The fraction of organic carbon can be

calculated from the fraction of organic matter by the following

empirical (i.e., much scatter 1) equation:

f,, = 0.58f^ (11)

The second property affecting Kp is the inherent difference among

organic chemicals in their affinity for the organic carbon phase of

the soil. The more hydrophobic the chemical (dependent on the

chemical's degree of non-polarity), the greater the affinity for

soil organic carbon than pore water. This affinity for organic

carbon is represented by K^, the Organic Carbon Partition

Coefficient. The greater the value of K^, the greater the

chemical's sorption in soil.

The above two relationships allow us to calculate Kp without

actually performing batch or column tests of the soil. First, an

estimate or laboratory determination of a soil's f is needed.

Mass fraction of contaminant in water:

f = ^ . EkiL?£ (16)

f = ? (17) K^ f^ ( l - n ) p + n 6 + /C 72 (1- e )

Mass fraction of contaminant in air:

" M., K , , f,, C,(l-72) p ^ C ^ n Q , ^ K „ C , n (1-6,)

f - ^ u n i l - B 7 (,,) K^ f^ (1-n) p + 71 e^ + iC 72 (1-0^)

Note that in the saturated zone, where no pore air is assumed to

exist ((?3 = 1) , the rightmost term of the denominator reduces to

zero and the value for f, becomes zero.

Appendix A lists K^ and K„ (as well as H, S, and p) values for many

common soil and groundwater contaminants. Values are for 20"C

unless noted. Vapor pressure, Henry's Law Constant, and viscosity

are often highly temperature dependent, so ambient soil temperature

may be important to consider if significantly different from 20°C.

11

T h e r e f o r e ,

0 .1887 __o 532 0 .3671

^ ^ ^ 0 9 0 0 ^ Q _ 2 6 3 0 . 3 6 7 1

^ 0 . 3 6 7 1

Interestingly, even at this low organic carbon level, more than

half (55.2 %) of the TCE mass resides in the soil (leaving the

other half of the mass split between water and air).

As a few calculations with equations (15), (17), and (19) will

demonstrate, the K^ and Kj of a chemical, as well as the organic

carbon content of the soil, greatly affect the distribution of the

contaminant in the soil, water, and air phases. This is shown

graphically in Appendix B, which contains mass distribution

diagrams for several common contarainants. (Programming and

graphics output for Appendix B was by Wyn Ross... Thanks!) As an

example of the range of variation, the diagrams show that virtually

all DDT resides in the soil phase, while vinyl chloride will

largely.partition to the air phase if a reasonable volume of air-

filled porosity exists.

13

IV- Calculation of Concentrations

Given only the water concentration, C„;

C. = K^ f^ C, (21)

C, = K,C^ (22)

Recall that the correct units for C„, C,, and C are ug/cm^ (=mg/L) ,

u q / c r 7 (= mg/L), and ug/g (= mg/kg), respectively.

B. Given only the air concentration, C (for example, from a soil

gas analysis):

C, == CjKj , (23)

C = —5£ o c _ ^ (24)

C. If only the concentration from a soil sample is known:

For this case, calculations for the soil, water, and air phases are

not as direct. This is because the laboratory procedure for

analyzing a soil usually involves driving off or stripping the

chemical from all three phases during analysis. For example, one

typical laboratory procedure involves cutting a small disk of soil

14

S o l v i n g fo r C„ y i e l d s :

C ~- ^^^^ -^^P (29) " /C^ /^ (1 - II) p + 71 e, + iC 72 (1 - 8,)

Air and true soil concentrations readily follow;

C, = K„ C, (30)

Cs = ^ o . f o . C^ (31)

Example

Q. A soil boring is properly collected and preserved from the site

described in the previous example. The laboratory reports a TCE

concentration of 2.30 mg/kg from the saraple. What are the

corresponding water and air concentrations in the soil? What is

the true soil (sorbed) concentration?

A. The concentration of the chemical in water is calculated from

equation (29) :

C^ ( 1 ~ n ) p

K ^ f^ (1 - 72) p 4- n e, + iC 72 (1 - 0,)

16

V. Retardation of an Orqanic Chemical in Groundwater

Darcy's Law states that the specific discharge (Darcy velocity) of

groundwater in an aquifer is the product of the hydraulic

conductivity of the aquifer and the hydraulic gradient:

V = K i (32)

The actual fluid velocity of the groundwater is obtained by

dividing the Darcy velocity by the effective porosity of the

aquifer:

Vp = - ^ (33)

The retardation factor, R, for an organic chemical in groundwater

is calculated as follows:

1 + ^^ ^ P = i-f Pj^ "^o^ ^ o c R = 1 + J-5 £ = l-f -L^ 2£ 2£ (34) ^e -"e

The retardation factor defined in equation (34) is valid only if

sorption of the organic chemical is described by the relationship

C = Kp C^; that is, sorption must follow a linear isotherm. In

fact, the calculations outlined in this entire set of notes are

predicated on the assumption of sorption linearity (which

conveniently holds for a great many organic compounds).

18

R = I A- (1-5) (126) (Q.OQl) _ ^ ^3 0.3 0

^c 1 = 0.61

Therefore, TCE travels at 61% of the velocity of groundwater, or

looking at it another way, in the time it takes groundwater to

travel 100 feet, the dissolved TCE in the groundwater will travel

only 61 feet.

20

VI. Effect of Equilibrium Partitioning on Groundwater Cleanups

Because of equilibrium partitioning, an organic contaminant

dissolved in groundwater travels slower than the groundwater, and

a certain percentage of the contaminant is tied up in the soil

phase rather than dissolved in groundwater. Therefore, groundwater

cleanup plans that do not account for these -phenomena will

underestimate both contaminant mass and cleanup timefraraes.

Due to variations in lithology and groundwater flow regime, the

most precise method of estimating the effects of sorption on

cleanup timeframes is through the use of a digital groundwater flow

and solute transport model. If such a model is to be accurate, it

must be based upon an accurate representation of spatial and

temporal variation in hydraulic and chemical source parameters.

For the purpose of these notes, however, we desire to know the

effect of equilibrium partitioning on cleanup timeframes in a more

general way, so that these effects can be roughly estimated without

resorting to a detailed groundwater model of the site. In the

following discussion, we will perform a highly simplified analysis

using a one-dimensional computer model, then relate this result to

the retardation factor.

Let us proceed by first considering an aguifer containing a

contaminant plume of TCE with the following characteristics:

21

Volume of TCE in the plume

V. M, w 1861 kg (37 J

^^ PTCB ( I . i e g/c:m' ) (1000 cm^/L) (.001 kg/g)

= 1275 L

= 336.6 gallons (Conversion: 3.785 L = 1 gal)

Note: This indicates how little of a chemical, in this case about

six 55-gal drums, is needed to substantially contaminate a vast

volume of groundwater.

A. Scenario 1: Extraction Without Retardation of TCE

Assume four extraction wells are constructed, each of which pumps

1000 gal/min. Also assume (unrealistically) that the extraction

wells remove groundwater only from the plume area (i.e., the

pumping does not affect groundwater beyond the limits of the

plume).

Q = 4000 X 3.786 - ^ X ^^^IBH x 1 ^ x ^^^ ^ ^ mm gal hr day yr

= 7.9594 X 10^ L/yr

In order reduce the concentration of TCE in the aquifer from 400

ug/L to the MCL of 5 ug/L, 98.75% of the TCE mass in the plume

23

B. Scenario 2: Extraction with Consideration of Retardation

In this scenario, we consider the effect of retardation. To

simplify the problem, the contaminated groundwater is assumed to

flow from the bottom to the top of the aquifer, where water is then

removed. In other words, the contaminant plume is considered to be

a column with elliptical cross-section and a height of 100 feet.

Clean water is drawn into the bottom of the column while

contaminated water is removed from the top. The organic carbon

content of the aquifer is assumed to be 0.1 % (f,. = 0.001); the

particle density used in the Section V example (p = 2.14 g/cm-') is

maintained. By equation (17), the mass fraction of TCE in water

can be calculated as:

f _ (0.30) (1) (126) (0.001) (1-0.30) (2.14) +- (0.3) (1)

f„ = 0.614

Thus the dissolved mass of TCE in the plume, 1861 kg (determined

previously), represents only 61.4% of the total system mass. The

other 38.6% of TCE, or 1170 kg, is sorbed to aquifer raaterial

within the plume. A total of 3031 kg of TCE therefore exists in

the system. Because this additional mass must be removed from the

system, cleanup tiraes will obviously increase.

Using the coraputer program BIO-lD, a column 100 feet long (the

thickness of the aquifer) and a pumping rate of 4000 gal/min were

25

required to clean up a TCE plume if sorption is neglected, at least

16.3 years would be required with sorption considered (assuming the

same organic carbon content used in this example).

[Additional retardation may occur due to flow through lower

permeability layers (Bouwer, H. , "Simple Derivation of the

Retardation Equation and Application to Preferential Flow and

Macrodispersion,", Groundwater, Vol. 29, No. 1, Jan.-Feb. 1991),

sorption directly onto mineral surfaces (Ball, W.P. and P.V.

Roberts, "Long-Term Sorption of Halogenated Organic Chemicals by

Aquifer Material: 1. Equilibrium," Environmental Science &

Technologv, Vol. 25, No. 7, 1991, pp. 1223-1237), and diffusion

into and out of deadend pores and stagnant "pockets" within the

aguifer. None of these retardation effects are fundamentally

related to carbon-based sorption.]

27

VII. Effect of NAPLs on Mass Distribution Equations

If NAPL (non-aqueous phase liquid or "free product") is present in

the subsurface in either the vadose or saturated zones, the

equations presented in Sections III through VI do not apply in the

region of the NAPL.

Both LNAPL (light non-aqueous phase liquid or a "floater") and

DNAPL (dense non-aqueous phase fluid or a "sinker") may occur as

residual fluid in the unsaturated and saturated zones. The NAPL

becomes immobilized by capillary forces in the small interstices

present in the soil matrix.

LNAPLs include petroleum fuels (gasoline, diesel, oil) and

unchlorinated aliphatic and aromatic hydrocarbons such as benzene,

xylene, naphthalene, hexane, ketones, ethers, etc. If enough LNAPL

is disposed so that it reaches the water table, it floats as a

separate fluid phase on the water table.

Most halogenated fluids are DNAPLs - trichloroethene (TCE),

tetrachloroethene (PCE), trichoroethane (TCA), methylene chloride,

carbon tetrachloride, trichlorotrifluoroethane (Freon-113),

pentachlorophenol and many others. If DNAPLs reach the water

table, they tend to sink through the aquifer until totally

dissolved in the groundwater or until impeded by less permeable

zones (including bedrock). If DNAPL reaches a less permeable zone

in the aquifer, it will tend to spread laterally in a downslope

28

Appendix A. Properties of Selected Organic Soil and Groundwater Contaminants

C h o m i c a l

A c e t o n e ( D M K )

B o n z o n e

B o n z o ( Q ) p y r e n e

B r o m o d i c h l o r o m e t h a n e

B r o m o f o r m

C a r b o n t e t r a c h l o r i d e

C h l o r o b e n z e n e

C h l o r o f o r m

D D D

D D E

D D T

D i b r o m o c h l o r o m e t h a n e

D i b r o m o c h l o r o p r o p a n e ( D B C P )

1,2- D:b»-omoethone ( E D B )

1.2- D i c h l o r o b e n z e n e ( o - O C B )

1.3- D i c h l o r o b e n z e n e ( m - D C B )

1.4- D i c h l o r o b e n z e n e ( p - D C B )

D i c h l o r o d i f l u o r o m e t h a n e ( F r e o n - 1 2)

1.1- D i c h l o r o e t h a n e (1 , 1 - D C A )

1.2- D i c h l o r o e t h a n e (1 , 2 - D C A )

1.1- D i c h l o r o e t h e n e ( 1 . 1 - D C E )

c i 5 - l , 2 - D i c h l o r o e t h e n e (e -1 ,2 -DCE)

t r Q n 6 - l , 2 - D i c h l o r o e t h e n e ( t - 1 , 2 0 C E )

1.2- D i c h l o r o p r o p a n e

t r a n 5 - l , 3 - D i c h l o r o p r o p e n e

E t h a n o l (E thy l a l c o h o l )

E t h y l b e n z e n e

b is -2- E t h y l h e x y l p h t h a l a t e ( D E H P )

M e t h a n o l ( M e t h y l a l c o h o l )

M e t h y l - t e r t - b u t y l e t h e r ( M T B E )

• So lub i l i t y ••

in?a/i)„_.. M i s c i b l e

1 7 8 0

0 , 0 0 3 8

4 5 0 0

3 0 1 0

7 8 5

4 s a

8 2 0 0

0 . 0 2 0

0 . 0 4 0

0 , 0 0 7 7

4 0 0 0

1 2 3 0

3 3 7 0

1 4 8

1 3 3

7 4

2 8 0

550O

8 6 9 0

4 0 0

8 0 0

6 3 0 0

2 7 0 0

280O

MIoc lb le

1 5 2

0.041

M i s c i b l e

4 3 0 O 0

_:_-[2 ]

[21

[2 ]

[11

...l^L [1 ]

[1]

[11

[2]

_i21_ [2 ]

[1 ]

[3]

[2]

_.JHL [21

* [ 2 ]

- [ 2 ]

[11

[11

- [ 1 ]

-[21

(21

_J21. [4 ]

[21

[3 ]

("1 [ 1

, F lu id .

P o n s l t y . ;

( g / c m 3 ) :

0 ,79 [2 ]

0 .88 [2 ]

S o l l d [2J

1.97 [1 ]

2 .89 [1 ]

1.59 [1 ]

1.11 [1 ]

1.49 [1 ]

S o l i d [ 2 ]

S o l l d [2 ]

S o l i d [ 2 ]

2 .38 [1 ]

Z 0 5 [2 ]

2 .18 11]

,...l:51_..[i]_. 1.29 [2 ]

S o l l d [ 2 ]

Q Q S [2 ]

1.17 |1 ]

1.26 [1 ]

1.22 [1 ]

1.28 [ I J

1,26 [2 ]

1,16 [4 ]

1.18 (2 ]

0 .79 [4 ]

0 .87 [2 ]

0.99 [2J

0.79 [4 ]

0 .74 [6 ]

Bo l l i n

Rp in t

(Do_q

5 6

8 0

4 9 5

9 0

1 5 0

7 7

1 3 2

6 2

1 9 3

2 6 0

1 1 9

1 9 6

1 3 2

1 8 0

1 7 3

1 7 4

- 3 0

5 7

8 3

3 2

6 0

4 8

9 6

1 1 2

7 8

1 3 6

3 8 5

6 5

5 5

g

91^. [21

[2 ]

[21

[1]

J l l . (11

[11

[1 ]

[21

(21

[11

[5 ]

[11

.1?]., [21

[2 ]

[2 ]

[11

[1 ]

(11

(11

(21

[21

[2 ]

Cl (2)

[21

(41

[61

A b s o l u i o

V I s c o s i

-J?Pl.. 0 .33

0 .65

— 1.71

2 . 0 7

0 .97

0 . 8 0

0 .56

.-.. — —

1.72

0 .75

— —-

O.SO

0 .84

0 .36

0 .48

0 . 4 0

0 .87

0 . 8 0

1 .20

0 . 6 9

0 . 6 0

ty

[6 ]

(4]

(11

JAL (11

[ 1 ]

(11

[11

i' J...

[11

(11

[11

[11

[1 ]

[41

[4 ]

[6 ]

[4 ]

[6 ]

, k lnennat l c

V l a c q s i t y

.,7 :(cs),, „

0 . 4 2

0 .74

-— 0 .87

0 .72

0,61

0 .72

0 .38

—

—

0 .79

0 .57

— —

0 .42

0 .67

0,3O

0 .38

0 , 3 2

0 .75

0 ,68

1.52

0 .79

0 ,76

V a p o r , :

P r e f l s u r e :

( m m H g ) :

1 6 0

7 6

5 , 0 E - 0 7

5 0

5

9 0

1 2

151

1 . 0 2 E - 0 6

6 . 4 9 E - 0 6

1 .9E -07

7 6

0.8

11

1

Z 3

0 .6

4 2 5 0

1 8 0

61

5 9 0

2 0 0

2 6 5

4 2

2 5

4 0

7.1

2 E - 0 7

9 5

(21

[2]

[ 2 ]

(11

_. , iU, [11

[ 1 ]

[ 1 ]

* * [ 2 ]

_*,*.i?l. (21

[11

(51

[11

[ 2 ]

*[21

[21

[ 2 ]

[11

JIL [ 1 ]

* [ 1 1

(2 ]

[ 2 ]

_-.i?L (41

[21

(21

(41

H o n r y l s L a w

C o n s t a n t , H

( o t m - m 3 / m o l )

3 . 9 7 E - 0 5

0 . 0 0 5 4 8

< 2 . 4 E - 0 6

0 . 0 0 2 4

0 . 0 O 0 5 6

0 . 0 2 3

0.OO36

0 , 0 0 2 8

2 . 1 6 E - 0 5

2 . 3 4 E - 0 5

4 . 8 9 E - 0 5

0 .0OO99

0 ,OO0249

0 , 0 O 2 5 0

0 . 0 0 1 2

0 . 0 0 3 6

0 .0031

0 . 4 2 5

0 . 0 0 4 3

0.0OO91

0 .021

0 . 0 0 2 9

0 . 0 0 5 3

0 . 0 0 2 3

0 .001 3

0 . 0 0 6 6

1.1 E -05 *

4..5.5R.06

• [ 2 ]

- [ 2 ]

(2 ]

(11

[11_._ (11

[ 1 ]

(11

[ 2 ]

.[21...

[ 2 ]

[11

[ 2 ]

[ 2 ]

.EL. [ 2 ]

[ 2 ]

[ 2 ]

[ 1 ]

[ 1 ]

[11

(2 ]

[ 2 ]

i.?L..

[ 2 ]

[ 2 ]

[ 2 ]

D i m e n a i o n l o s s

H o n r y ' s L a w

C o n s t a n t . K h

0,001 62

0 .224

< O.OOOI

0,10

0 .023_

0,96

0.15

0.12

0 . 0 0 0 8 9 8

0 . 0 0 0 9 7 3

0 . 0 0 2 0 3

0,041

0 ,0104

0,1 04

0 ,050

0 ,15

0.13

1 7,7

0,18

0 .038

0 ,87

0 .12

0 ,22

0 .096

0 .054

0 .27

0 . 0 0 0 4 5

0 . 0 0 0 1 8 9

K o o

( cm3 /g )

0.37

91

8 9 0 0 0 0

61

116

439

3 3 0

* 4

4 3 7 0 0

1 0 0 0 0 0 0

2 4 5 0 0 0

64

129

44

166

1 7 0

158

363

30

1 4

65

59

27

48

95

1 0 0 0 0 0

21

21

2]

11

J,l 11

11

11

21

?i 21

11

2]

2]

?]. 21

2]

2]

11

11

1]

21

2]

11

2]

21

Page 1

Chemicals Sorte(j by Increasing Fluid Density

1.4-

m-

o-

P-

bia-2-

1.2-

1.1-

trar,s-l.3-

1.1-

1,2-

t raT^ l .2-

ds-1.2-

1.3-

1.2-

1. l . i ­

l . i .2-

1.1.2.2-

1.2-

Chemtca)

Dichlofodifluofomethane (Freon-12)

Vinyl chlonde

Ben2o(a) pyrene

ODD

DDE

DDT

Dtchkxobenzer>e (p-DCB)

Toxapherte

MethyHerl-butytethef (MTBE)

Acetone (DMK)

Ethanol (Ethyl alcohoO

Methanol (Methyt alcohol)

Methylethylketooe (MEK)

Xylene

Ethylbenzene

Toluene

Benzene

Xylene

Xylene

Ethylhexylphthalate (DEHP)

Phenol

Chlorobenzene

Dichkxopfxjpane

Naphthalene

Dichloroeth.ane (1,1-DCA)

Dichloropropene

Dichloroethene (1.1-DCE)

Dtchlofoettiane (1.2-DCA)

Dichloroethene (t-1.2-0CE)

Dichloroethene (c-1.2-DCE)

Dichlorobenzene (m-DCB)

Dichlorobenzene (o-DCB)

Methylene chloride

Tnchlcroethane fTCA)

T r^ lo roe thane (1.1.2-TCA)

Trichlofoethene (TCE)

Chloroform

Trichlorofiuoromethane (Freon-11)

Pentachlofobiphenyl (PCB-1254)

Trichkxotrifluoroethane (Freon-113)

Ca i t on tetrachloride

Tetrachloroethane

Tetrachkxoethene (PCE)

Bromodichloromethane

Dibfomochloropropane (DBCP)

Obfomoethane (EDB)

Di bromoch lorofnethane

Bromoform

Fluid

Density*

(g/cm3)

Gas

Gas

SotKi

SoJid

Sd id

Solid

Solid

Solid

0.74

0.79

0.79

0.79

0.81

0.66

0.87

0.87

0.86

0.88

0.88

0.99

1.06

1.11

1,16

1,16

1.17

1.18

1,22

1.26

1,26

1.28

1.29

1.31

1.33

1.35

1.44

1.46

1.49

1,49

1.51

1.56

1.59

1.60

1.63

1.97

^ 0 5

2.18

2.38

2 8 9

Chemicals Sorted by Increasing NAPL Mobility (1/Kinematic Viscosity)

Chemical

Kerosene (JP-1)

Gasoline (Auto)

Aviabcxi gasoline

Mercury

' Watef = 1

MobiUty*

(1/K. Vtsc)

1.1.2.2-

o-

1.2-

1.2-

P-

n>

tranG-1.3-

1,2-

1.1.1-

1.2-

1.1-

CG-1.2-

trans-1.2-

1.1-

Phend

Ethanol (Ethyl alcohol).

Tetrachtoroethane

Xylene

Bromodichloromethane

Ethylbenzene

Dibromoethane (EDB)

Methanol (Methyt alcohol)

Dichtofopropane

Benzene

Xylene

Xyteoe

Chlorobenzene

EJromoform

Toluene

DichkxoprDpene

Dichloroethane (1.2-DCA)

Trichloroethane (TCA)

Carbon tetrachloride

, Dichlorobenzene (o-DCB)

Tetrachloroethene (PCE)

Methylethylketooe (MEK)

Trichlorotrifluoroethane (Freon-113)

Dichloroethane (1,1-OCA)

Acetone (DMK)

Trichloroethene (TCE)

Chloroform

Dichloroethene (c-1.2-DCE)

Methy lef>e chloride

Dichloroethene (t-1,2-DCE)

Dichloroethene (1.1-DCE)

Trichlorofluoromethane (Freon-11)

Other fluids

0,09

0.66

0.91

1.07

1.15

1.26

1.27

1.32

1.33

1.35

1.38

1.39

1.39

1.40

1.47

1.48

1.50

1.61

1.64

1.75

1.31

1.93

2.29

2.38

2.39

2.55

2.66

2,67

3,02

3.15

3.39

3.55

0 27

0.81

1,29

9,02

"Water ;-- 1 00

Chem.icais Sorted by Increasing Vapor Pressure

Chemicals Sorted by Increasing Dimensionless Henry's Constant

b.s-2-

1.4-

1.2-

1.3-

1.1.2.2-

o-

m-

P-

1.2-

1.1.2-

trans-1,3-

1.2-

1.2-

1.1.1-

1.1-

cis-1.2-

trans-1.2-

1.1-

CTiemical

Methyl-tert-butylether (MTBE)

DDT

Ethylhexylphthalate (DEHP)

Benzo(a)pytene

DDD

DDE

Toxaphene

Pentachlorobiph-3nyl (PCB-1254)

Naphtiialene

Phenol

Dichlorobenzene (p-DCB)

Dibromochloropropane (DBCP)

Dichlorobenzene (o-DCB)

Dichlorobenzene (m-DCB)

Bromolorm

Tetrachloroethane

Xylene

Ethylbenzene

Xylene

Xylene

Dibromoethane (EDB)

Chlorobenzene

Tetrachloroethene (PCE)

Trichloroethane (i .1.2-TCA)

Toluene

Dichloropropene

Ethanol (Eth-/1 alcohol)

Dichloropropane

Bromod ichloromeSiane

Trichloroethene (TCE)

Dichloroethane (1,2-OCA)

f/ethylethylketone (MEK)

Benzene

Dibfomochloromettiano

Cartx)n tetrachloride

Methanol (Methyl alcohol)

Trichloroethane (TCA)

Oiloroform

Acetone (DMK)

Dichloroetharw (1.1-DCA)

Dichloroethene (c-1,2-DCE)

Dichloroethene (1-1.2-OCE)

Trichlorotrifluoroethane (Freon-113)

Methylene chloride

Dichloroethene (1.1-DCE)

TricNorofluorometfiane (Freon-11)

Vinyl chloride

Dichlorodtftuoromethane (Freon-12)

ND = No dnta

Vapor

Pressure

(mm Hg)

ND

1.9E-07

2E-07

5,0E-07

1.02E-06

6.49E-06

3.3E-05

6E-05

0.054

0.2

0,6

0.8

1

2.3

5

5

6,6

7.1

3,3

8 8

11

12

14

19

22

25

40

42

50

58

61

72

76

76

90

95

100

151

180

180

200

265

270

349

590

667

2580

4250

b<3-2-

1,1.2.2-

1,1,2-

1,2-

1.2-

trans-1,3-

1.2-

1,2-

cis-1.2-

1.4-

1.3-

1,1-

trans-1,2-

o-

rrv

P-

1.1.1-

1.1-

Ct>emical

Ethanol (Ethyl alcohoO

Methyl-tert-butylether (MTBE)

Phenol

EJenzo(a)pyrene

Methanol (Methyl alcohol)

Toxaphene

Methylethylketone (MEK)

Ethylhexylphthalate (DEHP)

DDD

DDE

Acetone (DMK)

DOT

Dibromochkxopropane (DBCP)

Tetrachkxoethane

Naphthalene

Bromoform

Trichk)roethane(l,l,2-TCA)

DK^Iofoethane (1,2-OCA)

Dibromochkxomethane

Dichlofobenzene (o-DCB)

Dichloropropene

Methylene chkxide

Dichloropropane

BrortK)dk;hlofomethane

Dibronwethane (EDB)

PentacKiorobiphenyl (PCB-1254)

Chkxoform

Dichloroethene (c-l.2-DCE)

Dichtorobenzene (p-DCB)

Chkxobenzene

Dichkxobenzene (nvDCB)

Dichkxoethane (1.1-OCA)

Dtc))k>roethen9 (1-1,2-DCE)

Xylene

[Jenzene

Ethylbenzene

Toluene

Xylene

Trichloroethene (TCE)

Xylene

Tnchkxoethane (TCA)

Tetrachkxoethene (PCE)

Dichkxoethene (1.1-OCE)

Carbon tetrachkxkle

Trichkirofluoromethano (Freon-11)

Trichlorotrifluoroettiane (Freorvt 13)

Ochkxodifluoromethar>e {Fr6on-12)

Vinyl chlofKle

NO = No d j U

Dimensionless

Henry's Law

Constant, Kh

ND

NO

1.IE-05

<0.0001

• O.OOOI 89

0.00025

0.00O437

0.00045

0.000898

0.000973

0.00162

0.00203

' 0.0104

0.016

0.019

0,023

0.031

0.038

0.041

0.050

0.054

0.071

0.096

0.10

0.104

0.11

0.12

0.12

0.13

0.15

0.15

0.18

0.22

0.22

0.224

0,27

0.28

0.29

0,30

0.30

0,54

0.545

0.87

0.96

4.6

13,9

17.7

116

APPENDIX B

CONTAMINANT MASS DISTRIBUTION

DIAGRAMS FOR SELECTED

ORGANIC COMPOUNDS

o I— o < rv

(71 in <

FOC = .0007 "1 i I I \ I r

A i r

Water

S o i l

1—i I I r 0 5

SATURATION

FOO = . 0 0 2

O

u < Cr:

in in <

~] r 0.5

SATURATION

O \— u < cr

LT)

<

" 1 — I — i — I I I \ r 0,5 1 0

SATURATION

SATURATION = .15 SATURATION =

O h-u < cr

cn in <

SATURATION = .75

O I — o < cr

in in <

FOC FOC FOC

BENZENE: K,,= 91.0 c m V g . KH= .224

o u < cc

on

<

3 o

ll

ll

l

.,.

FOC = ,0007 l i l l l l l l

S o i 1

1 1 1 1 I 1 1 1

1

-

FOC = .002

o I — o < cr

in in <

0,5

SATURATION

z o I — o < cr

in in <

05 —

FOC =

S o i 1

°'° l l l l j 0,0 0,5

.01 1 1 1

1 1 1

1

-

1,0

SATURATION SATURATION

SATURATION =

z o u < cr

in m <

SATURATION = .45 SATURATION = .75

O o < cr

in in <

1 — 1 — I — I I r

-^-^__ W a t e r

"1 r

S o i 1

1 — I — \ — I — I — i \ r 0,005 0,010

FOC FOC

DDT; Ko,= 245,000 cmVg> KH= .00203

o I— o < cr

in in <

n — I — \ I—I I r

W a t e r

So i

"1 \ I I I I I r 0,005 0,010

FOC

A i r.

FOC = .0007

o o <

in IT)

<

FOC = .002

SATURATION

n \ I I I \ \ I — r

0,0 0.5 1,0

SATURATION

S o i 1

~ |—I—\—\—r 0,5 1 0

SATURATION

O I—

o < a:

c/) LO <

0,5 —

-

SATURATION = I I I I I I

A i r ^

W a t e r ^ _ ^

y ^

/ So i 1

1 1 1 1 1 1

.15

1 1

1

- ^

-

-

o \— CJ

< cr in

<

FOC

Ai r.

O I—

O < cr

in in <

1 0 —

0.5 —

-

i SATURATION = ,75 — 1 — 1 — \ — i — 1 — 1 — 1 — \ —

W a t e r

^ ^ ^ S o i 1

1 1 1 1 [ 1 1 1

1

-

_ . -

-

0,005

FOC

ETHYLENE CHLORIDE: Koc== 8.8 c m V g . K H - .071

A i r

z o

< cr L^ 0,5 in in < 2

—

—

-

-

FOC --

Ai r

——

So i 1

l l l l

= .0007 I 1 ' i—-

W a t e r

1 1 1 1

^ -

-

=

-

1 0 0 0,5

SATURATION

SATURATION

O

o < cr

in in <

O

o < cr

in m <

A i r

A i r-

SATURATION

• SATURATION = ,45

O I — o < cr

in in <

FOC = ,01

z o \— o < cr U_ 05

in in < 2

0,0

-

1 '1 ,1 ,—. \ i i L_

W a t e r —

So i 1

0,0 0.5

- ^ -^

" -

--

1 1

A 1 r

SATURATION

SATURATION = ,75

FOC

PCE: Koc= 3 6 4 c m ^ / g , K H = 0 . 5 4 5

A i r

FOC = 0007

o I

FOC = ,002

o — U < or

CA)

o I—

o < cr b - 0,5 —

t/) ir, <

V a t e r -

S o i 1

SATURATION SATURATION

n i—j 1 \ r 0 5

SATURATION

Ai r

O

SATURATION = ,15 _ _ _ _ I — ^ I |_

I ' I 000

I I I I I I 0 005 /OOIO

FOC

SATURATION = ,45

O h -u < cr

in in <

"00

' - ^ o c - 2 5 cm'^/g, K H = .3