VERTICAL DIFFUSION OF SELECTED VOLATILE ORGANICCONTAMINANTS THROUGH UNSATURATED SOIL FROM A

WATER TABLE AQUIFER; FIELD AND LABORATORY STUDIES

Item Type text; Thesis-Reproduction (electronic)

Authors Thomson, Kirk Alan

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this materialis made possible by the University Libraries, University of Arizona.Further transmission, reproduction or presentation (such aspublic display or performance) of protected items is prohibitedexcept with permission of the author.

Download date 21/05/2018 15:59:46

Link to Item http://hdl.handle.net/10150/275331

INFORMATION TO USERS

This reproduction was made from a copy of a document sent to us for microfilming. While the most advanced technology has been used to photograph and reproduce this document, the quality of the reproduction is heavily dependent upon the quality of the material submitted.

The following explanation of techniques is provided to help clarify markings or notations which may appear on this reproduction.

1.The sign or "target" for pages apparently lacking from the document photographed is "Missing Page(s)". If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure complete continuity.

2. When an image on the film is obliterated with a round black mark, it is an indication of either blurred copy because of movement during exposure, duplicate copy, or copyrighted materials that should not have been filmed. For blurred pages, a good image of the page can be found in the adjacent frame. If copyrighted materials were deleted, a target note wili appear listing the pages in the adjacent frame.

3. When a map, drawing or chart, etc., is part of the material being photographed, a definite method of "sectioning" the material has been followed. It is customary to begin filming at the upper left hand corner of a large sheet and to continue from left to right in equal sections with small overlaps. If necessary, sectioning is continued again—beginning below the first row and continuing on until complete.

4. For illustrations that cannot be satisfactorily reproduced by xerographic means, photographic prints can be purchased at additional cost and inserted into your xerographic copy. These prints are available upon request from the Dissertations Customer Services Department.

5. Some pages in any document may have indistinct print. In all cases the best available copy has been filmed.

University Microfilms

International 300 N. Zeeb Road Ann Arbor, Ml 48106

1325627

Thomson, Kirk Alan

VERTICAL DIFFUSION OF SELECTED VOLATILE ORGANIC CONTAMINANTS THROUGH UNSATURATED SOIL FROM A WATER TABLE AQUIFER; FIELD AND LABORATORY STUDIES

The University oi Arizona M.S. 1985

University Microfilms

International 300 N. Zeeb Road, Ann Arbor, Ml 48106

Copyright 1985

by

Thomson, Kirk Alan

All Rights Reserved

VERTICAL DIFFUSION OF SELECTED VOLATILE

ORGANIC CONTAMINANTS THROUGH UNSATURATED SOIL

FROM A WATER TABLE AQUIFER; FIELD AND LABORATORY STUDIES

by

Kirk Alan Thomson

A Thesis Submitted to the Faculty of the

DEPARTMENT OF HYDROLOGY AND WATER RESOURCES

In Partial Fulfillment of the Requirements For the Degree of

MASTER OF SCIENCE WITH A MAJOR IN HYDROLOGY

In the Graduate College

THE UNIVERSITY OF ARIZONA

19 8 5

Copyright 1985 Kirk Alan Thomson

STATEMENT BY AUTHOR

This thesis has been submitted in partial fulfillment of re­quirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this thesis are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the copyright holder.

SIGNED: &X A. U

APPROVAL BY THESIS DIRECTOR

This thesis has been approved on the date shown below:

D. D. Evans Professor of Hydrology

Date

ACKNOWLEDGMENTS

I would like to thank Dr. Glenn Thompson, who served as thesis

director for the first two years of this research, for his advice and

inspiration during both field and laboratory studies. His expertise in

the fields of gaseous tracer research and subsurface contamination

problems has helped tremendously with the analytical aspects of this

work. It was a pleasure to work with him.

I thank Dr. Daniel Evans, who served as thesis director for the

third and final year of this research, as well as Dr. Gray Wilson and

Dr. Eugene S. Simpson for their consultation, support, and constructive

criticism throughout the studies.

The advice of Dr. David Kreamer, Dr. Klaus Stetzenbach,

Dr. Arthur Warrick, and Dr. Soroosh Sorooshian is gratefully acknowl­

edged. I would also like to thank a number of my colleagues, including \

Mark Malcomson, Timothy Sutko, Ronald Green, and Frans Lowman for their

patience and assistance in various phases of this research.

i i i

TABLE OF CONTENTS

Page

LIST OF TABLES vi

LIST OF ILLUSTRATIONS. vii

ABSTRACT viii

1. INTRODUCTION 1

Problem: Early Detection of Contamination by Volatile Organic Compounds (VOCs) 1

Soil Gas Sampling as a Preliminary Monitoring Technique for Volatile Organic Compounds 2

2. CHEMICAL PARTITIONING OR DISTRIBUTION OF VOCs AT THE GROUNDWATER-VADOSE ZONE INTERFACE 6

Equilibrium Chemistry 6 A Two-Phase Chemical Distribution Experiment 8

3. THEORY OF GASEOUS DIFFUSION THROUGH POROUS MEDIA 12

General Diffusion Coefficient 15 The Relationship Between the General Diffusion

Coefficient (D) and the Effective Diffusion Coefficient (D1) 16

4. SOIL GAS STUDY OF SELECTED VOLATILE ORGANIC CONTAMINANTS ABOVE A TRICHLOROETHENE CONTAMINATED AQUIFER, TUCSON, ARIZONA 21

Site Description 21

Location and Regional Hydrology . 21 Geologic Setting 22

Analytical System 24

Gas Chromatograph Desiccant 24 Column Development 25 Electron Capture Detector (ECD) 26 Calibration 26

iv

V

TABLE OF CONTENTS—Continued

Page

Field Sampling Techniques 27 Results and Discussion of Field Study 30

5. LABORATORY GASEOUS DIFFUSION COLUMN EXPERIMENT 40

Objective Method . 40 Experimental Apparatus and Procedure 41 Results and Discussion of Laboratory Studies 43 Modeling the Laboratory Experiment 48

6. CONCLUSIONS AND RECOMMENDATIONS 57

APPENDIX A: PRODUCTION STATISTICS AND CHEMICAL PROPERTIES OF SELECTED VOLATILE ORGANIC COMPOUNDS 59

APPENDIX B: BREAKTHROUGH CURVES IN COLUMN EXPERIMENTS FOR SELECTED VOLATILE ORGANIC COMPOUNDS 64

REFERENCES CITED 71

LIST OF TABLES

Table Page

1 Volatile organic compounds identified at the Carranza field study site 3

2 Dc values calculated assuming equilibrium conditions and based on measurements made in a two-phase laboratory experiment (25°C) 11

3 General diffusion coefficients (0) for seven selected VOCs 17

4 Effective diffusion coefficients (D') for seven selected VOCs according to various empirical equations 20

5 Vertical distribution of volatile organic compounds at the Carranza field study site 35

6 Distribution coefficients (Dc) measured in field and two-phase laboratory experiment 37

7 Distribution coefficients (Dc) for laboratory gaseous diffusion column experiment as a function of elapsed time 47

8 Effective diffusion coefficients (D*) calculated by one-dimensional model based on laboratory data shown with predicted D' and D* 54

9 Gaseous diffusion retardation factors for selected VOCs based on approximated D' values and measured D* values 56

vi

LIST OF ILLUSTRATIONS

Figure Page

1 Two-phase chemical distribution experiment 9

2 Geologic log of monitor well installed at Carranza study site 23

3 Apparatus for soil gas measurement in the field 28

4 Recognition of accurate soil gas contaminant concentration measurement 31

5 Gas chromatogram of 2 cc injection of air above ground at Carranza field study site 32

6 Gas chromatogram of 2 cc injection of soil gas from 25 ft. horizon at Carranza field study site 33

7 Gas chromatogram of 5 ul injection of groundwater from Carranza well 34

8 Experimental apparatus used to measure chemical partitioning and gaseous diffusion of VOCs 42

9 Aqueous concentration vs. time for TCE at Port 2 44

10 Breakthrough curves (concentration vs. time) for TCE at Ports 4, 6, and 8 45

11 Log difference determination of rate constant, k, for TCE breakthrough curves 52

12 Location of the Carranza Field Study Site (Pocket)

v i i

ABSTRACT

Distribution coefficients describing the ratio of gaseous phase

concentration versus aqueous phase concentration were measured for

selected volatile organic compounds (VOCs) at the saturated-unsaturated

zone interface in field and laboratory experiments. Effective diffusion

coefficients for selected VOCs were measured in laboratory column exper­

iments. Distribution coefficients measured ranged from 0.04 for methy­

lene chloride to all of the compound in the gas phase for 1,1,1 tri-

chloroethene. Laboratory gaseous diffusion column experiments yielded

effective diffusion coefficients ranging from 0.054 cm2/sec for tri-

chlorofluoromethane to 0.028 cm2/sec for methylene chloride. Results

indicate that trichlorofluoromethane, methylene chloride, chloroform,

1,1,1 trichloroethane, carbon tetrachloride, trichloroethene, and per-

chloroethene will partition from the aqueous phase into the gaseous

component of the unsaturated zone. Because vapor phase transport from

contaminant source areas may be faster than migration in the aqueous

phase, it is advantageous to monitor trace gases of the vadose zone.

Soil gas sampling is a fast, inexpensive, preliminary monitoring tech­

nique for several volatile organic contaminants.

vi i i

CHAPTER 1

INTRODUCTION

Problem: Early Detection of Contamination by Volatile Organic Compounds (VOCsT

Aquifer contamination by commercially-produced volatile organic

compounds is becoming increasingly common in industrial regions through­

out the United States. Valuable water resources are threatened by

contamination due to toxic or potentially toxic organic compounds.

Analysis and evaluation of contaminated ground water-vadose zones is

generally conducted by use of conventional monitoring wells and lysime-

ter techniques. Arranging a network of monitoring wells in a contami­

nated region is a very expensive and time-consuming practice. Suction

lysimeters collected very small, localized liquid samples from the

vadose zone and rarely yield representative samples. An effective,

inexpensive technique is needed to quickly identify the general extent

of contamination by VOCs and supply important information needed to

ensure the most effective placement of groundwater monitoring wells.

Much attention has been focused on halogeriated compounds of low

molecular weight. In particular, chlorinated hydrocarbons have gained

much interest due to their widespread use in industry and occurrence in

groundwater systems. Chlorinated hydrocarbons exhibit properties that

distinguish them as solvents. These compounds are nonflammable, have

relatively low water solubility, and are excellent nonpolar solvents for 1

2

fats, waxes, oil, and grease. Several fluorocarbons are used as

refrigerants and propellants.

The VOCs identified at the Carranza field study site described

later in this work are listed in Table 1.

The use of man-made organic compounds began more than 30 years

ago. Figure A.l (see Appendix A) shows the annual production of

selected synthetic organic compounds up to 1973. Tables A.l and A.2

(see Appendix A) list important physical and chemical properties of

selected chlorinated volatile organic compounds and their industrial use

patterns.

Soil Gas Sampling as a Preliminary Monitoring Technique for Volatile Organic Compounds

On-site soil gas analysis using gas chromatography has great

potential as a preliminary monitoring technique for contamination by

VOCs. Many halogenated aliphatic organic compounds are volatile at

standard groundwater temperatures and are only slightly soluble in

water. These compounds undergo significant transport in the gaseous

phase. Migration of volatile pollutants can proceed in all directions

from landfills and hazardous waste sites. Movement can be downward from

ground surface spills, dumps, and ponds, or upward and away from leaking

underground tanks and migrating contaminated groundwater.

Groundwater contamination by VOCs can often be identified by

analyzing trace gases in soils just below ground surface. This tech­

nique is possible because many VOCs will volatilize from contaminated

groundwater tables and move by molecular diffusion away from source

3

Table 1. Volatile organic compounds identified at the Carranza field study s i te.

Abbreviation Molecular Formula

F-ll Trichlorofluoromethane CC1 F 3

CH CI Methylene Chloride CH2C12

CHC1 Chloroform CHCI3

TCA Trichloroethane C H CI 2 3 3

CC1 Carbon Tetrachloride CCl^

TCE Trichloroethene C2^3

PCE Per ("tetra") chloroethene C2C14

4

areas towards regions of lower concentrations in the overlying soil

profile. Given sufficient time, a gaseous phase concentration gradient

from the water table to the ground surface will be established. In

unconfined aquifer systems, contaminant transport in the unsaturated

zone is often extensive. Barometric pressure changes and water level

fluctuations may act to draw gases out of the aqueous phase of the

groundwater into the gaseous phase of the vadose zone. Although the

pollutant vapor concentration may be several orders of magnitude lower

in the soil gas near the ground surface than in the groundwater, it can

often still be measured. For example, trichloroethene can be measured

in the soil gas in concentrations down to 5 x 10"^ ug/1 of gas. In

these situations, monitoring the gaseous component of the vadose zone

can be extremely beneficial towards understanding the movement of objec­

tionable material. Unfortunately, few studies have dealt with the vapor

phase transport of volatile organic contaminants through the vadose

zone.

Brief introductions to the principles of chemical partitioning

and the diffusion of gases in the vadose zone are presented in chapters

2 and 3 of this work. Approximations are used to estimate effective

diffusion coefficients (D1) for selected VOCs in a non-sorbing porous

medium.

In Chapter 4, a soil gas study of VOCs above a portion of a

trichloroethene (TCE) contaminated aquifer in the southwest part of

Tucson, Arizona, is described. Distribution coefficients (Dc),

describing the ratio of gaseous phase contaminant concentration (ug/1

5

gas) to the aqueous phase contaminant concentration (ug/1 water) are

calculated for selected VOCs based on field measurements.

In Chapter 5, a gaseous diffusion column experiment is

described. The column apparatus was used to measure distribution coef­

ficients (Dc) for the selected VOCs. Application of Fourier series and

boundary values as used to solve heat flow problems were then used to

calculate effective distribution coefficients (D*) for selected VOCs

based on the laboratory diffusion column experiment. In Chapter 6,

comparisons are drawn between values of Dc> D', and D* measured and

approximated in field and laboratory experiments.

CHAPTER 2

CHEMICAL PARTITIONING OR DISTRIBUTION OF VOCs AT THE GROUNDWATER-VADOSE ZONE INTERFACE

When studying the transport of VOCs in the vadose zone, we must

carefully examine the physical and chemical properties of the entire

contaminated groundwater-vadose zone system. The problem must be ad­

dressed as a two-part or two-phase system: (1) the groundwater as an

aqueous phase source of contamination, and (2) the overlying gaseous

phase of the vadose zone into which transport takes place.

Equilibrium Chemistry

The transfer or flux of contaminant from aqueous phase to

gaseous phase plays an important role in the resulting gaseous diffusion

that takes place in the vadose zone. The flux of VOCs into the gas

phase from an aqueous source is governed by principles of equilibrium

chemistry. Henry's Law (Freeze and Cherry, 1979), as shown in Equation

(2.1), can be used to calculate the liquid-gas partitioning coefficient

(Kw). Values of Kw describe the ratio of the concentration of the VOC

existing in aqueous solution to its concentration in the overlying gas

phase at equilibrium.

yvoc) = [VOC] p KV0C

(2.1)

7

where [VOC] = aqueous concentration of VOC (moles/1);

pV0C = Partial pressure of VOC (atm).

Values of Kw can be estimated using the solubility of a given

compound as its aqueous concentration and using its saturated vapor

pressure as its partial pressure. For example, in the case of TCE:

P-fCE = vapor pressure (atm).

Vapor pressure (Pygc) in atmospheres must be converted to a

value in terms of molarity (moles/liter). Based on ideal gas laws, the

volume (V), in liters, occupied by n moles of gas at a given temperature

(T), in K, and pressure (P), in atmospheres, is given by:

where R - ideal gas constant = 0.082054 1 atm/°K molo.

Therefore, at 25°C, T = 298.15°K, and P = 0.0623 atm.:

v = (1 mole) (0.082054 1 atm/°K mole) (298.15°K) = 3g2 gg m 0.0623 atm

8.4 x 10 moles/liter of water

where [TCE] = solubility (mg/1) = 1,100 mg/1

= 8.4 x 10"3 moles/liter of water;

( 2 . 2 )

8

or

.1 mole _3 = 2.5 x 10 moles/liter of gas 392.69 liters

Therefore, Kw(TCE) = _3 8.4 x 10 moles/liter of water

2.5 x 10~3 moles/liter of gas

and KW(TCE) = 3,4

For purposes of easy comparison to field and laboratory measure­

ments, an air-water distribution coefficient is defined as the ratio of

contaminant concentration in the gas phase as ug/liter of gas to the

contaminant concentration in the aqueous phase as ug/liter of water.

For TCE, Dc = Kw-1 = 0.30.

Distribution coefficients for each VOC were measured in a two-

phase laboratory experiment. A mixed solution of seven VOCs, each at a

concentration of 100 parts per billion (ppb) was placed in a sealed

container with equal volume of overlying headspace as shown in Figure

3.

The system was kept at 25°C and was given adequate time to

equilibrate. Known volumes of liquid and gaseous headspace were then

withdrawn from the sealed vial and analyzed by gas chromatography.

Temperature and partial pressure play important roles in the chemical

A Two-Phase Chemical Distribution Experiment

9

SEALED VIAL WITH SEPTUM

HEADSPACE

STANDARD SOLUTION OF VOC S

Figure 1. Two-phase chemical distribution experiment.

partitioning of VOCs from aqueous to gaseous phase. At higher

temperatures, VOCs tend to enter the gaseous phase more readily. Values

of Dc calculated assuming equilibrium conditions and calculated based on

measurements made in a two-phase laboratory experiment are listed in

Table 2.

It should be pointed out here that the partitioning (or distri­

bution) of VOCs is greatly affected by aqueous concentration. If, for

example, a VOC exists in aqueous concentration less than its solubility,

it will in most cases have a Dc value less than that calculated based on

equilibrium conditions. For this reason, field measurements of Dc in

situations where aqueous phase concentrations are below saturation will

likely yield coefficients smaller than those predicted by saturated

equilibrium conditions and closer to those measured in the two-phase

laboratory experiment (in which measurements were made using aqueous

concentrations below solubility values).

11

Table 2. Dc values calculated assuming equilibrium conditions and based on measurements made in a two-phase laboratory experiment (25 C).

VOC D- Assuming D_ Measured in Equilibrium Conditions Two-Pfiase Lab Experiment

F-ll NA 0.26

CH2C12 0.10 0.08

CHC13 0.12 0.12

TCA 1.44 All in gas phase

CC14 0.95 0.50

TCE 0.30 0.25

PCE 0.83 0.40

CHAPTER 3

THEORY OF GASEOUS DIFFUSION THROUGH POROUS MEDIA

Gases are transported in the vadose zone by barometric pressure

variations, soil temperature fluctuations, wind, water infiltration,

evaporation processes, and gaseous diffusion. Ordinary gaseous diffu­

sion has been identified as the mechanism of greatest importance for gas

transport in the unsaturated zone (Keen, 1931; Evans, 1964).

Ordinary gaseous diffusion is defined as the movement of a gas

from a region of high concentration to a region of low concentration

resulting exclusively from the concentration gradient. One way of

analyzing the movement of pollutant vapor through the vadose zone in­

volves the application of equations originally derived to describe the

flow of heat in a solid.

Diffusion is caused by random molecular motions. Transfer of

heat by conduction is also due to random molecular motions, and there is

an obvious analogy between the two processes. Fick (Crank, 1956) put

diffusion on a quantitative basis by adopting the mathematical equation

of heat conduction derived some years earlier by Fourier (1822). The

mathematical theory of diffusion is therefore based on the hypothesis

that the rate of transfer of diffusing substance through unit area of a

section is proportional to the concentration gradient measured normal to

the section.

Fick's first law of diffusion is:

12

13

F = -D ill (3.1) a x

where F = rate of transfer per unit area of section;

C = concentration of diffusing substance;

x = space coordinate measured normal to the section; and

D = general diffusion coefficient.

For purposes of simplification, the transport medium will be

considered homogeneous and isotropic. The transport medium may be open

air or a gas-filled porous geologic formation. The mathematics of

diffusion in anisotropic media can be found in Crank (1956) but will not

be used here.

The fundamental differential equation for gaseous diffusion in

an isotropic medium is given by:

9C + + ifX + l£z _ Q a t a x - a y a z

(3.2)

The derivation of this equation can also be found in Crank

(1956). If the diffusion coefficient is assumed constant for a given

medium, temperature and pressure, Equation (3.2) becomes:

= D ^ + D 3 ^ + D 3 £ (3 3) 3 t x a x 2 * a y 2 a z 2 1 1

14

Considering only one dimension, Fick's second law of diffusion is:

— = D 3t 3X

(3.4)

In this chapter, general diffusion coefficients (D), describing

the diffusion of a gas into another gas, are approximated for seven

selected VOCs and are then used to approximate effective diffusion

coefficients (D'). Values of D' describe the diffusion of a gas through

a porous medium based solely on the porosity of the porous medium and

neglecting sorption onto solid or liquid phases which may be present.

In Chapter 5 of this work, a second effective diffusion coefficient, D*,

is measured for selected VOCs diffusing through a laboratory column

apparatus.

Retardation factors (R) are then calculated for the VOCs based

on approximated values of D' and laboratory measured values of D*. A

retardation factor, as defined here, describes the ratio between the

diffusion rate of a given gas through a non-reactive (non-sorbing)

porous medium and the diffusion rate through a potentially reactive

porous medium. Essentially, R = D' (approximated neglecting sorptive

effects of the porous medium) /D* (measured, considering sorptive

effects of the porous medium). The greater the R value, the greater the

sorptive effects of the porous medium. Direct measurement of diffusion

15

rates eliminates the need for tedious measurement of parameters used to

describe the sorptive properties of the porous medium.

The general diffusion coefficient, D, is a measure of molecular

diffusion of gas in free air. To this author's knowledge, there are no

experimental data available which directly yield values of D for the

organic compounds of interest in this paper. Slatterly and Bird (1958)

developed an expressionless (see Equation 3.5), however, to compute the

general diffusion constant, D, for the diffusion of a gas A into gas B

for nonpolar gases. Although this equation has been derived from the

equation of state and kinetic considerations, it also contains empiri­

cally-derived coefficients. The organic compounds of interest in this

paper will be considered nonpolar for purposes of this report. This is

a valid assumption in the light that these compounds are classified as

only slightly polar. Therefore, the general diffusion coefficient (D),

in cm^/sec, can be represented by:

General Diffusion Coefficient

D = W3 <w5/12(V^1/2 b

(3.5) P

where Pq = critical pressure for designated gas A or B (atm);

Tq = critical temperature for designated gas (°K);

M = molecular weight of designated gas (gm/moles);

P = prevailing pressure (atm);

16

T = prevailing temperature (°K);

a = 2.745 x 10"^ (a constant); and

b = 1.823 (a constant).

If the vadose zone atmosphere is considered to be predominantly

nitrogen, the general diffusion coefficients for selected VOCs can be

calculated. Values of D for gaseous contaminants identified at the

field study site are given in Table 3. It should be noted that values

of D presented here neglect chemical interactions that may occur when

more than one VOC or other substance is present.

The Relationship Between the General Diffusion Coefficient (D) and the Effective Diffusion Coefficient (D1)

Fick's second law can be generalized to describe gaseous diffu­

sion in a porous medium (dry or partially saturated) by replacing the

general diffusion coefficient, D, with an effective apparent diffusion

coefficient, D':

at 3x2 (3'5>

The effective diffusion coefficient (D1) describes the diffusion rate of

a given gas through a given non-reactive porous medium. There are many

differences between the diffusion of one gas into another gas and the

diffusion of that same gas into a gas-filled porous medium. Several

researchers have developed approximations which relate the effective

diffusion coefficient (D1) to the general diffusion coefficient and

properties of the porous medium.

17

Table 3. General diffusion coefficients (D) for seven selected VOCs.*

VOC General Diffusion Coefficient (D)

F—11 0.1025

CH.Cl„ 0.1350 2 2

CHC1 0.1260 3

TCA 0.0925

CC1 0.1180 4

TCE 0.0950

PCE 0.0875

* All values in cm2/sec.

Penman (1940) concluded that the decrease in diffusivity in

porous media as compared to air v;;" not only due to decreased cross-

sectional area but was also due to increased path length that a gas

molecule must follow. This increase in path length is referred to as

the tortuosity factor. Penman found that the ratio between effective

diffusion coefficient and general diffusion coefficient could be

approximated by:

wheree[)= drained or gas-filled porosity. ~

Blake and Page (1948) modified Penman's 0.66 coefficient to a

range of values between 0.62 and approximately 0.8, based on direct

measurements of gaseous diffusion on several soils (including clays)

using carbon disulfide as the diffusing gas. They found that in Brooks

ton clay, D' approached zero when 0g neared 10%.

Marshall (1959) proposed the relationship: <

which agrees with Penman's coefficient at well-drained or air-filled

porosity values, but is smaller than Penman's ratio at low values.

Marshall attempted to account for the size distribution of soil pores.

Millington (1959) arrived at the relationship:

D'/D = O. 6 6 0 D (3.7)

D ' / D = 0 D 3 / 2 (3.8)

D'/D = 0D 4/3 (3.9)

Several other properties of a porous medium have significant

effects on the diffusion of gases. These properties include: (1) total

organic content, (2) sorptive properties of the soil for VOCs, (3)

moisture content, and (4) overall permeability of the medium, to name a

few.

Effective diffusion coefficients (D') calculated using the equa­

tions presented above are given in Table 4. Parameters which define the

effective diffusion coefficient can be measured in field and laboratory

experiments. In Chapter 5, average values of D' calculated based on the

Penman approximations are compared to laboratory-measured values of D*

and are used to estimate values of retardation (R) for selected VOCs.

This was done by assuming, any difference between approximated effective

diffusion coefficients (D1) for a non-sorbing porous media (based on

physical properties of the compounds of interest) and effective diffu­

sion coefficients (D*) for a potentially sorbing porous media (calcu­

lated based on data obtained in the laboratory column diffusion experi­

ment) was the result of sorption or retardation caused by the presence

of the porous medium.

20

Table 4. Effective diffusion coefficients (D1) for seven selected VOCs according to various empirical equations*

Effective Diffusion Coefficient (D1)

(porosity = 0.45)

VOC F-ll CH CL CHC1 TCA CCV TCE PCE 2 2 3 4

Penman 0.0338 0.0446 0.0416 0.0305 0.0389 0.0314 0.0289

Taylor 0.0342 0.0451 0.0421 0.0309 0.0394 0.0317 0.0292

Marshall 0.0362 0.0477 0.0445 0.0327 0.0417 0.0336 0.0309

Millington 0.0407 0.0536 0.0500 0.0367 0.0468 0.0377 0.0347

* Values in c^/sec.

CHAPTER 4

SOIL GAS STUDY OF SELECTED VOLATILE ORGANIC CONTAMINANTS ABOVE A TRICHLOROETHENE CONTAMINATED AQUIFER,

TUCSON, ARIZONA

An experiment to investigate the concentration of volatile

halocarbons in the soil gas above a portion of the TCE contaminated

Tucson aquifer was initiated on February 2, 1983. The purpose of the

experiment was to learn what factors affect the soil gas concentration

of a contaminant emanating from the water table and to evaluate methods

of sampling the soil gas and groundwater. Soil gas sampling is poten­

tially the best preliminary investigative technique for volatile organic

compounds in groundwater because of the low cost and speed of the mea­

surement in comparison to drilling to the water table for each data

point.

Site Description

Location and Regional Hydrology

The site is located at the Carranza residence at 7019 S. 6th

Avenue in Tucson (U.S.G.S. location: D151313dbc) (refer to plate 1 in

pocket). The property is directly down gradient (northwest) of a known

source of TCE contamination. Over 500 ppb of TCE were detected in the

Carranza domestic well at the time of the study, indicating that the

Carranza property is over the contaminated groundwater plume. Because

of the proximity of the site to the contamination source, it is logical

that the TCE has moved under the study area with the groundwater flow

21

22

and has diffused upward from the water table through the soil in the gas

phase.

Geologic Setting

The Carranza study site is located in the southwest portion of

the Tucson basin. Geologic materials in this region consist primarily

of quarternary basin fill alluvial deposits. The domestic well on the

property reaches a depth of 106 feet with a depth to groundwater of 95

feet. No geologic log was available for this well. The monitor well

installed at the time of the field study (February 2, 1983) reaches a

depth of 110 feet, with a depth to groundwater of 95.5 feet. The

monitor well is located approximately 100 feet from the domestic well on

the property. Generally, the geologic materials encountered in each

well can be considered very similar, although there are most probably

localized variations in stratigraphic thicknesses. The geologic log

recorded for the monitor well is shown in Figure 2. The well penetrates

gravel, sand, clay, and caliche units.

Moisture content measurements could not be made at the time of

this study but it should be noted here that the presence of water in the

formation has significant effects on the sorptive properties of the

porous media for organic contaminants. Chiou (1979) suggests that, when

water is present in porous media, it competes with organic compounds for

sorption sites on geologic materials. In the absence of water, sorption

of organic compounds on geologic materials may be more extensive.

23

DEPTH (FEET)

WELL DETAILS

LITHOLOGIC LOG

SAMPLE DESCRIPTION

•10

.20

30 6 Seh 80

PVC sating

40

SO

•60

70

80

•0

•100

110

6 Seh 80 PVC SerMn

0.50 «lo1»

w

jm

mm

mm

UN

10

20

SO

40

SO

eo

7a

80

•0

100

110

MEDIUM BROWN SI-TV CLAY CALICHE

UGHT BROWN SILTY CLAY

LIGHT TAN SILTY CLAY AND GRAVEL

DARK SLTY CLAY AND FINE GRAVEL

LIGHT CLAY, SILT, FINE SAND AND GRAVEL

I/4M-I"GRAVEL AND 8ILT

REDDISH BROWN SILT AND CLAY (INCREASING MOISTURE WITH OEPTH)

LIGHT BROWN FINE SANDY SILT

REDDISH BROWN SILTY CLAY

BROWN AND RED COARSE SAND AND GRAVEL

Figure 2. Geologic log of monitor well installed at Carranza study site.

24

Analytical System

Analyses of gas and liquid samples containing VOCs were made

using a Varian 3700 series gas chromatograph equipped with electron

capture detectors (ECD). A dessicant system was designed to remove

water vapor from the sample gas, thus allowing the direct aqueous injec­

tion of relatively large sample volumes.

The improvement is significant because it means that the method

will be practical for a large percentage of the volatile priority pollu­

tants. If TCE can be measured at the 0.1 ppb level, compounds with more

chlorine or bromine substitution will be measurable at lower levels

(parts per trillion), while compounds with fewer halogen substitutions

such as dichloroethane and methylchloride will be measurable in the low

(1-5) parts per billion range. Furthermore, the speed and simplicity of

the measurement will enable inexperienced operators to easily learn and

very effectively use the technology to collect the much needed field

data.

Gas Chromatograph Dessicant

The dessicant systlem installed is a new des^n which employs

Nafion tubing (DuPont de Nemors, Wilmington, Delaware) surrounded by a

nitrogen gas purge jacket. The Nafion tubing is composed of an ionic

polymer which has a high affinity to allow polar compounds such as water

to diffuse through the walls, but does not pass nonpolar compounds such

as most organic halocarbon solvents. The Nafion tubing is essential to

the analysis because it removes water from the sample that otherwise

would interfere with the chromatography and impair the operation of the

25

•EC detector. Without the Nafion tubing to remove water, one must

perform an extraction to remove the organic solute from the water sample

before the measurement can be made. There are a variety of standard

procedures to do this, but all are time-consuming. Also, the extraction

techniques are often plagued by contamination problems due to increased

handling, use of solvents, and contact of the sample with easily contam­

inated surfaces such as the elastomer materials in valves and glass-to-

metal seals.

Column Development

Different chromatographic column material, column lengths, and

column temperature were tested for optimum analysis of sample gas.

Criteria for optimum analysis are: (1) that the important peaks to be

observed are sharp, distinct, and separate, and (2) that the analysis

time is reasonable for the entire sample to be analyzed.

After some testing, the analytical column chosen for use in

these studies was a 2.6 meter long, 4 mm I.D. column filled with 10% SP

2100 on Supelcoport 100-120 mesh (Supelco, Inc.). Optimal column tem­

perature was 70°C. A flow rate of 35 ml/min was used for N2 carrier gas

through the system. A pre-column/backflush system was designed to allow

the sample gas stream to vent out of the system after the compounds of

interest were integrated, thus eliminating unnecessary delays and prob­

lems with late eluting peaks.

Electron Capture Detector (ECD)

The ECD is an extremely sensitive, highly specific gas

chromatographic detector. Its specificity is for gas molecules (often

the halogenated gases) which are electron acceptors and readily form

negative ions when exposed to a concentration of free electrons. The

detector cell emits energetic beta particles from an 8 mci ^Ni radio­

active foil which ionize N£ carrier gas molecules. This produces free

electron concentration which is monitored by the electronic circuitry of

the detector. Sample and carrier gases, which are effluent from the

chromatographic column, enter from the bottom of the detector tower.

pass through the collector and foil cylinders (which are coaxial and

displaced), and exit through the top of the detector tower. When elec­

tron absorbing sample gas molecules enter the detector, they absorb

electrons, and this results in a decrease in free electrons in the

ionization volume. The electronic circuitry of the ECD senses this

reduction and produces an output signal proportional to the concentra­

tion of electron absorbing molecules.

Sensitivity of the detectors for the compounds of interest in

this study allowed distinguishable, separate peaks at concentrations

lower than a part per billion. At this sensitivity, minute contaminants

begin to interfere with accurate measurement, and a further increase in

detectability is generally not desirable.

Calibration

Initial identification of VOCs encountered at the Carranza field

study site was accomplished by analysis on a gas chromatograph/mass

spectrometer. After identification of the compounds present, subsequent

analyses were made on the analytical system previously described. One

hundred parts per million (ppm) certified standards of each compound

identified at the study site were diluted down to lower concentrations

when necessary and run throughout testing for calibration of aqueous and

gaseous samples.

Field Sampling Techniques

Soil gas was withdrawn through a drive-point screen (probe)

driven or buried in the ground at the desired depth (Figure 3). Gas was

collected by pumping the soil gas out of the ground and through a sample

container by means of a vacuum pump. A gas sample was periodically

collected in a syringe from the sample bottle in the evacuation line and

analyzed in the field. The field analysis is critical to the method in

order to determine when a representative sample has been obtained and to

direct the investigation as it progresses.

A hollow stem auger was used to drill the access hole. Soil gas

samples were collected at various depths through an air probe (well

point) lowered down the center of the auger. Generally, the work pro­

ceeded as follows: The auger hole was advanced to the desired depth,

the air piezometer which consisted of a standard 30" drive-point screen

on 1-1/4" steel pipe was lowered to the bottom of the hole and either

driven with a 150 lb. hammer or backfilled to bury the screen in the

bottom of the hole.

After the probe was in place, the soil gas was pumped at 5 to 20

1/min for a period of 30 to 50 minutes with analyses being made as

28

VALVE

HIGH SPEED VACUUM

PUMP GENERATOR SOIL GAS

SAMPLE BOTTLE

GROUND SURFACE

HOLLOW FLIGHT AUGER

SOIL PROFILE

VADOSE ZONE

AUGER BIT

GAS INTAKE

DRIVE POINT SCREEN (SOIL GAS PROBE)

^7 95 ft

CONTAMINATED GROUNDWATER

Figure 3. Apparatus for soil gas measurement in the field.

29

frequently as possible during this period. The series of measurements

were needed to determine if uncontaminated air was being drawn into the

sample from above ground. If surface air is being drawn down the bore­

hole, the contaminant concentration will show a decrease after about

five minutes of pumping when the surface air reaches the probe screen.

If there is no open connection to the surface, the concentrations will

remain constant for at least 50 minutes of pumping. Two examples which

illustrate the behavior described are given in Figure 4.

In Figure 4, sample A (soil gas collected at a depth of 25 ft.

below ground surface) shows air leakage down the borehole. Sample B,

soil gas collected from a depth of 50 ft. in the same location using the

technique described above, represents a sample collected with no air

leakage; thus, the contaminant level remained nearly constant for the

entire sampling period. This ability to know if air is being drawn from

above is extremely important to the problem of collecting meaningful

data in vadose gas sampling programs because undetected air leakage can

easily cause 100% error in a sample measurement. After gas samples have

been collected, the borehole is augered down to the next desired depth

and the sampling procedure is repeated.

All of the TCE measurements were made in the field using conven­

tional laboratory equipment mounted in a vehicle and operated from a

generator. A Varian 3700 series gas chromatograph and Hewlett-Packard

integrator were the principal equipment items. The gas chromatograph

was modified with a Nafion tube dryer to remove water, thus allowing

direct injection of either soil gas or water. The practical detection

30

limit for TCE by this method is 0.1 ug/1 in water 1 x 10"^ ug/1 in soil

gas. The analysis time is the same for either water or soil gas typi­

cally taking about 10 minutes if no more than five to 10 compounds are

present in the sample. Figures 5, 6, and 7 show representative chromat-

ograms of air, soil gas, and groundwater, respectively.

Results and Discussion of Field Study

Seven compounds were identified in the soil gas and in ground­

water at the study site. These were:

trichlorofluoromethane (F-ll)

methylene chloride (CHoJJ.^)

chloroform (CHC13)

1,1,1, trichloroethane (TCA)

carbon tetrachloride (CC14)

trichloroethene (TCE)

perchloroethene (PCE)

The approximate depth and concentration observed for these com­

pounds in the soil gas and in the groundwater are given in Table 5.

Although a moist clay layer was encountered at depths from 65 to approx­

imately 90 feet, volatile organic contaminants have emanated upward

through the overlying soil profile. Significant contaminant concentra­

tions were detected in gas samples collected just 5 feet below ground

surface.

In the case of CHCI3, TCE, and PCE, the concentration increased

with depth down to the water table. For F-ll, TCE, and CH2Cl2, the

CONCENTRATION

X10'3

(ug/l gas I

10 "ioT

SAMPLE B

L

SAMPLE A

30 40

TIME (min)

Figure 4. Recognition of accurate soil gas contaminant concentration measurement.

Figure 5. Gas chromatogram of 2 cc injection of air above ground at Carranza field study site.

Figure 6. Gas chromatogram of 2 cc injection of soil gas from 25 ft. horizon at Carranza field study site.

Figure 7. Gas chromatogram of 5 ul injection of groundwater from Carranza wel1.

35

Table 5. Vertical distribution of volatile orqanic compounds at the Carranza field study site.

AIR MOVE GROUND1 F-ll . 0.004

CH2CI2

O.OOS

CCLJH

m

TCA

0.01

CC14

0.01

TCE PCE

0.001

SOIL MATERIAL

SOIL GAS

n~ 10 ft 0.007 1 0.007 0.02 0.008 0.006 0.01

SILT, SAND FIRAVEL

25 ft 0.006 0.2 0.009 0.01 0.009 0.02 0.04

+ 50 ft 0.005 0.1 0.03 0.001 0.09 0.03 1

CLAY 1

_1_ 3 AND 90 ft 0.004 0.08 0.3 0.001 2 9 5

SILT CLAY

UATER TABLEB 100 ft 0.003 2 1 m 0.1 142 0.05

CARRANZA HELL

100 -106 ft 0.009 6 1 0.1 0.2 558 0.2

a Concentrations expressed In ug/L gas • 20X (one standard deviation).

b Concentrations expressed In ug/L water *20!.

36

reverse trend was observed, the soil gas concentration was greatest near

the surface. The contaminant concentration from two samples of ground­

water is provided in Table 4. The first sample named "water table" is

water that was bailed from the first water to flow into the auger hole

drilled the same day. The Carranza well is a domestic well (about 300

feet away) which intercepts approximately the upper six feet of the

water table. Both samples are included for comparison. The "Carranza"

sample is probably a better representation of the local water, but the

"water table" sample is probably a better sample for comparing relative

concentrations of contaminants across the water table, i.e., the air-

water distribution coefficient underground.

The data are most easily interpretable for TCE because the

groundwater concentration is high enough to produce a strong gradient

from the water table to the ground surface. There is no TCE in the

atmosphere (free air) and the source is clearly from the groundwater.

Distribution coefficients observed for seven volatile organic

compounds across the water table surface in the field and in laboratory

experiments are listed in Table 6. A lower Dc value would be expected

in the field because of the slower aqueous phase transport of solute

through the aquifer material to the water-table surface where the gas

phase concentration is established. This was true for CHgC^ and TCE.

The other VOCs studied showed greater Dc values in the field as compared

to Dc values measured in the laboratory. This may have been caused by

the lateral diffusion of gaseous contaminant from nearby sources or by

gaseous phase transport complexities induced by subsurface geologic

37

Table 6. Distribution coefficients (Dc) measured in field and two-phase laboratory experiments.

VOC Dc Field Dc Laboratory

F-ll 1.33 0.26

CH2C12 0.04 0.08

CHC13 0.30 0.12

TCA All in gas phase All in gas phase

CC14 20.00 0.50

TCE 0.06 0.25

PCE 100.00 0.40

38

materials. It is possible that the dense clay layer encountered from 60

feet to 90 feet at the field study site acted as a semi-impermeable trap

for VOCs. This would account for the high gas phase contaminant concen­

trations measured for F-ll, CHC^, and PCE in the field. Equilibrium

may never be achieved in the field, assuming that diffusion and escape

through the unsaturated sediment is too rapid to allow the soil gas

concentrations to reach equilibrium above the water-table surface.

The other compounds which showed increasing concentration with

depth in the unsaturated zone (chloroform, carbon tetrachloride, and

PCE) also appear to have a subsurface source. However, in these cases,

the groundwater concentration at the site appears too low to be the

principal source for most of the gas observed in the soil. Lateral

diffusion from a nearby higher contamination source is a more plausible

explanation. Clearly, a horizontal gradient would have to be measured

to determine if lateral diffusion was a principal factor in producing

the gas concentrations observed. An influx of contaminated runoff into

the subsurface from a nearby wash might also be a plausible explanation

for the lower level contaminants observed at this site.

The F-ll, TCA, and the methylene chloride showed decreasing

concentrations with depth, indicating an atmosheric source, yet the

subsurface concentrations were higher than the concentrations in the

atmosphere. This seemingly paradoxical situation occurs quite commonly

for halocarbons originating in the atmosphere. Their concentration in

groundwater near recharge areas is often found to be several times

higher than would be expected for water in equilibrium with the

39

atmosphere from which they are derived. This phenomenon has been

demonstrated by Russell and Thompson (1983) to occur naturally as a

result of sorption-desorption mechanisms occurring in the three phase

soil-water-air system. Even though the natural processes can be respon­

sible for anomalously high halocarbon concentrations in groundwater,

this mechanism should be invoked with caution in areas where subsurface

dumping of contaminants has occurred.

CHAPTER 5

LABORATORY GASEOUS DIFFUSION COLUMN EXPERIMENT

Objective Method

A gaseous diffusion column experiment was designed to measure

the distribution and the one-dimensional gaseous diffusion of volatile

organic compounds (VOCs) emanating up through dry porous media from a

contaminated aqueous source. This was done to further evaluate the

extent of gaseous diffusion of VOCs that may be encountered in true

field situations.

The chemical distribution (partitioning) of VOCs across the

water table-vadose zone interface is an important process to consider

when studying the transport and fate of VOCs. Distribution coefficients

(Dc) were determined in laboratory column studies for seven selected

VOCs as a function of time since the introduction of contaminants to the

system. As discussed in Chapter 2, the diffusion of gaseous contami­

nants emanating into the overlying vadose zone from a groundwater source

is directly related to the distribution (partitioning or flux) of con­

taminant across the water table. The concentration of contaminant

entering the gas phase just above the water table establishes the lower

boundary condition and creates a concentration difference across the

vadose zone establishes the lower boundary condition and creates a

concentration difference across the vadose zone. This gradient is the

"driving force" of simple gaseous diffusion. 40

Thermal diffusivity of heat (K) in a solid is analogous to the

diffusion coefficient (D*) of gases in a porous media. Gaseous diffu­

sion of VOCs in a packed column can be mathematically modeled using

equations derived for the one-dimensional flow of heat in a cylindrical

solid bounded by two parallel planes of changing temperatures.

With a model of this kind, effective diffusion coefficients (D*)

were calculated based on contaminant breakthrough curves (concentration

versus time) generated from data collected from sampling ports along the

length of the diffusion column. Retardation factors, describing the

ratio between the general diffusion coefficient (as estimated by the

Slatterly and Bird (1958) approximation) and the effective diffusion

coefficient measured in the laboratory experiments were calculated.

Experimental Apparatus and Procedure

A schematic diagram of the experimental apparatus used to mea­

sure the chemical partitioning and gaseous diffusion of VOCs is shown in

Figure 8. The column consists of a 2 meter long, 10,2 cm internal

diameter glass tube with sampling ports every 20 cm along the length.

The column was packed with 2-5 mm size gravel to a height of 50 cm up

the column. The remaining column length was packed with 90 mesh washed

sand to a height of 190 cm.

At time t = 0, an aqueous solution of seven VOCs (about 100 ppb)

was introduced through the inlet at the base of the column. For the

purpose of this study, we assume instantaneous introduction of the

contaminated aqueous source to the diffusion column. The solution

continuously entered the column at a relatively constant flow rate of 5

200

190 L'llOcm

Sampling Port Port 8 '

160 X=80cm

90 Mesh Sand

Feeder Solution (Organic Compound

Mixture) Port 6 120

Port 4 80 X"0 cm

Height of Capillary Rise

§1 •.Port 2*

60 •Inlet Valve

50 Outlet Valve

- 40

Gravel Reservior Port 1° 20

Inlet-Outlet Valve

Figure 8. Experimental apparatus used to measure chemical partitioning and gaseous diffusion of VOCs

43

ml/min. Once the gravel-packed reservoir became saturated, the solution

then drained through the outlet valve located 50 cm up the column. This

was done so as to maintain a constant water level and a relatively

continuous contaminant source.

A capillary fringe developed in the column reaching just below

port 4, allowing gaseous samples to be taken at a point just above the

aqueous source. As time elapsed, aqueous or gaseous samples were with­

drawn from ports along the length of the column and analyzed for VOCs

immediately by gas chromatography. Chromatograph area counts can be

considered units of contaminant concentration assuming a linear inte­

grator response. This has been proven valid within the range of contam­

inant concentrations dealt with in these studies. Rapid analysis of

samples is essential to a study of this kind due to the volatile nature

of the compounds and their relatively fast travel time through a column

of this size.

Results and Discussion of Laboratory Studies

Data obtained for trichloroethene (TCE) at ports 2 (Figure 9),

4, 6, and 8 (Figure 10) will be used as the data base for sample calcu­

lations of Dc, D*, and R. Figures 1 through 6 of Appendix B represent

contaminant breakthrough curves on which calculations were based for 6

other VOCs. Unexpectedly high concentrations measured at sampling port

4 early in the experiment may have been caused by a piston effect

generated by the development of the capillary fringe. The migration of

liquid into the fine-grained sand possibly created a pulse of gaseous

TCE, PORT 2

AQUEOUS SOURCE

o r-H

X 1.0

o

(/) +-> c 3 o o <0 CD S-<c

0.5

i 1000 2000 TIME (MINUTES)

Figure 9. Aqueous concentration vs. time for TCE at port 2.

3000 4000 5000

-C4

PORT 4-1

PORT 6- ©

PORT 8-A

IOOO 4000 2000 TIME 3000 (MINUTES) ;

Figure 10. Breakthrough curves (concentration vs. jtime) for TCE at ports 4, 6, and 8.

5000

-c* cn

VOCs past sample port 4. Farther up the column, this effect seemed to

dampen out.

Values plotted on contaminant breakthrough curves are based on

one sample analysis. For this reason, no error analysis could be con­

ducted to determine the reliability of the data. Together with such

problems as contamination of samples by outside sources (chromatographic

syringes, lab air, etc.) and error introduced by slight variations in

gaseous sample size, the data are left open to significant interpreta­

tion.

For purposes of modeling the experiment using equations

d^cribTng"monitonically increasing functions, a solid line was drawn

through data points obtained from port 4, neglecting an anomalous data

point measured early in the experiment (see Figure 10). The dotted line

shown on Figure 10 was not used in calculations. All values used in

calculations described in this chapter were taken from solid lines hand-

drawn for best fit through the data points.

Values of gaseous phase concentration measured at port 4 divided

by the aqueous phase concentration (measured at port 2) in the source at

any time, t, were used to calculate a Dc value for each compound as a

function of elapsed time into the experiment. These values are given in

Table 7.

These values give some indication of the relative flux of these

compounds across the aqueous-gas interface. F-ll, C^C^, and CCl^ seem

to partition into the gas phase to a significant degree. The other

compounds show moderate preference for the gaseous phase.

47

Table 7. Distribution coefficients (Dc) for laboratory gaseous diffusion column experiment as a function of elapsed time

VOC

Elapsed Time F-ll CH.C1_ CHC1_ TCA CC1. TCE PCE (minutes) 3 4

50 0.02 0.04 0.01 0.02 0.03 0.01 0.01

250 0.06 0.32 0.01 0.03 0.06 0.02 0.02

450 0.10 0.48 0.03 0.07 0.13 0.06 0.06

1400 0.12 0.20 0.03 0.07 0.11 0.04 0.03

3500 0.17 0.27 0.03 0.06 0.12 0.03 0.03

48

Modeling the Laboratory Experiment

Fick's (1855) second law of molecular diffusion can be rewritten

to describe the retardation imposed upon the diffusing gas by the pre­

sence of the porous medium as follows:

aC _ n a2C v aC

2 (1 + K) || = D 2-C where: (1 + K) = R

ax and R = a retardation factor

n 3 C - n i f c 8t V

a£ _ D a£c at"R ax2

Therefore, O'/R = D*, where D* is an effective diffusion coeffi­

cient describing the diffusion of a gi'ST through a porous medium con­

sidering the sorptive properties of the porous medium.

Equations governing the conduction of heat in a solid bounded by

two parallel planes (as described by Carslaw and Jaeger [1959, p. 104])

can be applied to the case of one-dimensional gaseous diffusion between

changing boundary values. The equation used to model the laboratory

column experiment with the appropriate changing boundary conditions is

as follows:

49

C4(t) = e^t), when x = 0

Cg(t) = ©^(t)» when x = L

C = f(x), when t = 0

where e^t) = contaminant concentration in gas port at port 4 at time, t;

©2(t) = contaminant concentration in gas at port 8 at time, t;

L = distance between sampling ports 4 and 8; and

x = distance between sampling ports 4 and 6.

The solution from Carslaw and Jaeger (1959) in terms of the effective

diffusion coefficient (D*) is as follows:

C(t)

Letting f(x') = 0 and rearranging, the solution becomes:

i

O

50

where f(*n) = e0*"2"^^2 - (-ln«|,2(An)

X N = dummy variable of integration (time);

L = distance between two ports (4 and 8);

x = distance between ports 4 and 6 (halfway between ports

4 and 8);

D* = effective diffusion coefficient considering the sorptive

capacity of the porous medium;

t = elapsed time (time since introduction of VOCs);

<l>lUn) = +1 C"t) = gaseous VOC concentration at port 4 at time, t;

<j>2(X) = <}>2(t) = gaseous VOC concentration at port 8 at time, t.

The distance (L) from port 4 to port 8 is a distance of 80 cm.

Port 6 lies halfway between these ports and is a distance (x) of 40 cm

from port 4. The concentration at any time t at port 6 is a function of

the concentrations at ports 4 and 8 over time. Given x,L, a time

increment (tj to t2) and the average concentration at ports 4 (C^(t))

and port 8 (C®(t)) over this time increment, it is possible to choose D'

values that generate concentrations which fit the contaminant break­

through curve for port 6 (Figure 10).

Gaseous phase concentrations of VOCs measured at sampling ports

4 and 8 as a function of time were used as the lower and upper bound­

aries, respectively, in the model. These boundary values, represented

by contaminant breakthrough curves at sampling ports 4 (C4(t)) and 8 Q

(C°(tt)), can be described by the following expressions (as per verbal

communication with Sorooshian, 1984).

The lower boundary value (source) is represented by:

51

(t) = c4(t) = c|j (l-eV)

where = maximum concentration achieved at sampling port 4; o = rate constant for contaminant breakthrough curve measured

port 4; and

t = elapsed time.

The upper boundary is represented by:

<P2(t) = C8(t) = C® ( l -e k8( t 'V)

where C® = maximum concentration achieved at sampling port 8;

kg = rate constant for contaminant breakthrough curve measured

at port 8; and

t0 = time of first contaminant breakthrough at port 8.

Rate constants, k, were calculated for each curve using a log

difference as shown in Figure 11. Details of this method can be found

in Ramalho (1977). In most cases, more than one k value was found for

each curve. At early times (0-500 minutes), high values of k

( > 0.0020) were calculated for breakthrough curves generated from data

from port 4. At later times ( > 500 minutes), values were calculated as

low as 0.0004. Values of k for curves generated by measurements of

samples from sampling port 8 were generally less than 0.0010 at all

52

100

SLOPEa(L06 44-LOG l0)/-200, k 0.0032

AA.C. AT SLOPE-(LOG 10-LOG 3)/- 379, fc--0.0008

SLOPE-(LOG S-L063)/-500,ki"-0.0004

k --0.0008

k4(0VflJ— 0.0013

k (avg.)"-0.0007 SLOPE-(LOG 4-L0G 2}/-500, k--0.0006

TIME (MINUTES)

1000 1500 250 500

Figure 11. Log difference determination of rate constant, k, for TCE breakthrough curves.

53

times. Average values used in the model to calculate D* for TCE were as

follows:

k4 = 0.0015

k8 = 0.0007

t0 = 150 minutes

tQ = 1000 minutes

CQ(t) = 1.8 x 106 area counts

CQ(t) = 7.0 x 105 area counts

Values of D' approximated by the Penman (1940) relationship were

used as a "first guess" in the model calculations of D*. Values were

then adjusted until the calculated concentration fit that of data

obtained for port 6. An effective diffusion coefficient (D*) of 0.038

cm2/sec yielded a C6 (t) value which matched that of the contaminant

breakthrough curve at port 6 for t = 1000 minutes.

The following values of D* were -ilculated for seven VOCs.

Here, results are shown with predicted D' values based on approximation

equations derived by Penman (1940) and with corresponding general diffu­

sion coefficients (D) as calculated by Slatterly and Bird approximations

(1958) as described in Chapter 3.

F-ll showed a significantly higher D* value than the D' value

predicted. Methylene chloride showed a significantly lower D* value

than predicted. Values of D* calculated for CHC13, CC14, TCE, and PCE

were slightly higher than predicted. TCA showed a slightly lower D*

value than predicted. Generally, values of D* (considering sorption)

are very close to the predicted values of D' (assuming a non-sorbing

54

Table 8. Effective diffusion coefficients (D*) dimensional model based on laboratory predicted D' and D.*

VOC D* (by model) D1 (predicted)

F-ll 0.054 0.0338

CH CI 0.028 0.0446 2 2

CHC13 0.045 0.0416

TCA 0.035 0.0305

CC1, 0.039 0.0389 4

TCE 0.038 0.0314

PCE 0.032 0.0289

calculated by one-data shown with

D (approximated)

0.1025

0.1350

0.1260

0.0925

0.1180

0.0950

0.0875

55

porous medium). This suggests that sorption on the column material was

essentially negligible.

Using the equation, R = retardation factors (R) were calcu­

lated. Values of R based on D* values (considering sorption) measured

in this study are presented in Table 9.

Wilson et al. (1981) describe retardation factors of various

VOCs in the aqueous phase. For aqueous transport, a retardation factor

of less than 2 is considered very low. Apparently, retardation of VOCs

in their gaseous phase is less than retardation in their aqueous phase.

In all cases, retardation induced by sorptive processes was very low.

The flux or production of gaseous contaminant above the aqueous source

plays a major role in governing the resulting diffusion that takes place

in the unsaturated zone above. The greater the flux of a given compound

across the saturated-unsaturated zone interface, the larger the gaseous

phase concentration gradient across the vadose zone. This results in a

faster possible rate of gaseous diffusion through the vadose zone.

56

Table 9. Gaseous diffusion retardation factors for selected VOCs based on approximated D' values and measured D* values

VOC Retardation Factor (R)

F-ll 0.6

CH2C12 1.6

CHC1 0.9 0

TCA 0.9

CC14 1.0

TCE 0.8

PCE 0.9

CHAPTER 6

CONCLUSIONS AND RECOMMENDATIONS

Field measurements show that in every case (with the exception

of TCA) where halocarbons could be measured in the soil gas, they were

detectable in the groundwater. In the case of TCE, which showed high

concentration in the groundwater, the soil gas component appeared to be

derived from the contaminated groundwater immediately below the sampling

site. The groundwater appears to be the TCE source because the concen­

tration ratio measured between the soil gas and the groundwater corre­

sponded reasonably well to expectations which are based on laboratory

measurements of the gas/liquid distribution coefficient, Dc.

For chloroform, carbon tetrachloride, and PCE, a subsurface

source appears likely because the highest concentrations were measured

near the water table, but the groundwater immediately below the gas

sampling location appears to be too low to be the main contributor of

contaminants to the soil gas. Lateral movement in the gas phase from a

nearby source could have produced the profile observed. More sampling

locations along a horizontal transect would be needed to verify this

hypothesis.

Laboratory gaseous diffusion column experiments yielded effec­

tive diffusion coefficients ranging from 0.054 cm^/sec for F-ll to

0.028 cm2/sec for methylene chloride. Results agreed reasonably well

with predicted values based on approximations derived by Slatterly and

57

Bird (1958) and Penman (1940). Calculated retardation factors ranged

from 0.6 for F-ll to 1.6 for CI^Cl;? based on effective diffusion rates

measured in the laboratory. Values of D' and R are greatly dependent on

the flux rate of contaminant from aqueous source to gaseous phase at the

water table-vadose zone interface.

The ease of collecting soil gas samples coupled with sensitivity

of the measurement technique indicates that the gas sampling method will

be useful in contaminant investigations. The method may provide a rapid

survey technique for determining the approximate area! extent of a sub­

surface contamination problem. If the vertical and horizontal soil gas

profiles can be developed, considerable information about the source of

contamination may also be derived. The soil gas measurement at the very

least could provide a far more effective substitute for conventional

"soil sampling" as a technique for locating volatile contaminants in the

saturated zone.

In general, this field study suggests that shallow soil gas

sampling may aid in tracing contaminated groundwaters. However, re­

search is necessary to evaluate the effects of subsurface geology and

other varying parameters that may limit conclusions which can be drawn

from data obtained by this field method. Many volatile organic contami­

nants undergo significant transport in the gaseous phase. Soil gas

sampling can be used to determine the extent of this transport and may

be used in developing a contaminant "budget" to be incorporated into

models describing contaminant transport and fate.

APPENDIX A

PRODUCTION STATISTICS AND CHEMICAL PROPERTIES OF SELECTED VOLATILE ORGANIC COMPOUNDS

59

600

500

400

x|08

METRIC TONS

300

200

100

ANNUAL PRODUCTION OF SELECTED SYNTHETIC ORGANIC COMPOUNDS F-ll (CCLgF) TRICHLOROFLUOROMETHANE

CH£CL2 METHYLENE CHLORIDE

1954

CHCL. CHLOROFORM

TCA (CgHgCLg) 1,1,1 TRICHLOROETHANE

CARBON TETRACHLORIDE

TCE (C^HCLg) TRICHLOROETHENE

PCE (C2CL4) TETRACHLOROETHENE

CHCL

i . i I i i 1958 I9T0

1200

1000

800

KlO6 POUNDS

600

400

200

1974 1962 1966

PRODUCTION YEAR

Figure A-l. Annual .'production of selected chlorinated volatile organic compounds,

! TABLE A1

Properties and Use Patterns of Several Common Volatile Organic Compounds (from Yaws, 1975 and Lowenheim, 1975)

Compound

Property F—11 CH0C10 2 2 CHCI3 TCA CCl, TCE PCE

Molecular Weight 137.4 84.9 119.4 133.4 153.8 131.4 165.8

State at 25°C, 1 atm. gas lqd. lqd. lqd. lqd. lqd. lqd.

Melting Point, °C, 1 atm. -111 -97.7 -63.5 -30.4 -22.9 -73.0 -22.8

Boiling Point, °C, 1 atm. 23.8 39.8 61.7 74.1 76.5 87.0 121.0

Solubility in Water (mg/1) at 25°C, 1 atm.*

1600 13,200-20,000

8200 480-4400

785 1100 150-200

Critical Pressure (Pc) atm. 43.5 60.5 54.0 43.5 45.0 48.3 44,2

Critical Temperature (TC)°K 474.1 517.1 539.5 553.1 559.3 574.0 616.1

Vapor Pressure (Vp)25°C, atm. NA 0.5737 0.1980 0.1263 0.1184 0.0631 0.0184

NA - Not Available •Solubility ranges based on changes in solubility as a function of temperature (20°C-25°C)

62

TABLE A2

Use Patterns of Several Common Volatile Organic Compounds

Compound Use Pattern Percent

T ri chlorof1uoromethane Refri gerants Propel!ants

Methylene Chloride Paint remover 40 Exports 20 Aerosol sprays 17 Chemical specialties (mainly ' 10

solvent degreasing) Plastics processing 6 A11 other : 7

100

Chloroform Fluorocarbon refrigerants 52 and propellants

Fluorocarbon resins 41 Miscellaneous plus exports 7

TOO

1,1,1, Trichloroethane Solvent, including metal 70 degreasing and electrical and electronic equipment cleaning

Aerosols, solvent for adhesives 15 and polishes and other uses

Exports 15 W

Carbon Tetrachloride Fluorocarbon 11 28 Fluorocarbon 12 52 Miscellaneous 20

TOO

Trichloroethene Vapor degreasing 87 Extraction sol vent 3 Exports 8 Miscellaneous 2

TOO

63

TABLE A2 (continued)

Compound Use Patterns . - Percent

Per ("tetra") Chloroethene Dry cleaning solvent 75 Chemical intermediate, 8

particularly fluoro-carbons

Vapor degreasing 7 Miscellaneous 10

TOO"

APPENDIX B

BREAKTHROUGH CURVES IN COLUMN EXPERIMENTS FOR SELECTED VOLATILE ORGANIC COMPOUNDS

64

F-II

5.0

4.0 PORT

PORT 6

3.0

- PORT 8

2.0

1.0

2000 TIME 3000 (MINUTES)

4-000 5000 1000

Figure B-1. Breakthrough curve (concentration vs. time) for F-11 at ports 4, 6, and 8.

1.75

CH_C1

1.5 PORT 4

o

o 1.0

ol PORT e

o t(,T *K>

S A.C.

0.5

PORT 0

TIME (MINUTES)

Figure B-2. Breakthrough curves (concentration.-.vs. time) for. CH2C12 at ports 4, 6\ and 8.

0 1000

Fi " "

2000 4000 3000 5 000

2 Breakthrough (concentrati gure curves for. CH CI on vs at ports 4 6 and 8 2 2 9

CHCL

PORT 4

A.C. xlO®

PORT 6

OS

PORT 8

0 IOOO 2000 TIME 3000 4000 5000 (MINUTES)

Figure B-3. Breakthrough curves (concentration vs. time) for CHCl3 at ports 4, 6, and 8.

TCA

4.0

PORT-4

PORT-6 o XIO «c A.C. (/)

I 2.0 o

PORT-B

TIME (MINUTES)

Figure B-4. Breakthrough curves (concentration vs. time) for TCA at ports 4, 6, and 8.

4000 1000 3000 5000 2000

3.0

port 4

o r-H

X

port 6

+-> c 3 A.C.

port a

1 1000 2000 TIME 3000 (MINUTES)

Figure B-5. Breakthrough curves (concentration vs. time) for CC14 at ports 4, 6, and 8.

4000 500Q

curves

PCE

4.0

o

3.0 PORT 4

o

port 6

port 8

1000 4000 TIME (MINUTES)

Figure B-6. Breakthrough curves (concentration vs. time) for PCE at ports 4, 6, and 8.

2000 3000 5000

REFERENCES CITED

Ali, Syed Mohammed, 1980, Fate of trichloroethylene and other pollutants in soil and water: California Regional Water Quality Control Board— Central Valley Region.

Blake, G. R., and Page, J. B., 1948, Direct measurement of gaseous diffusion in soils: Soil Sci. Soc. Am. Proc., 13:37-42.

Braids, Olin C., 1982, Behavior of contaminants in the subsurface: Geraghty and Miller, Inc., Syosset, New York.

Carlslaw, H. S., and Jaeger, J. C., 1959, Conduction of Heat in Solids: Oxford University Press, Ely House, London, pp. 103-104.

Chiou, C. T., Peters, L. J., and Freed, V. H., 1979, A physical concept of soil-water equilibria for nonionic organic compounds, Science, Vol. 206, November issue.

Crank, John, 1975, The Mathematics of Diffusion, 2nd Edition: Clarendon Press, Oxford, England.

Currie, J. A., 1960a, Gaseous diffusion in porous media. Part I. A non-steady state method: Br. J. Appl. Phys., 11:314-317.

Currie, J. A., 1960b, Gaseous diffusion in porous media. Part II. Dry granular materials: Br. J. Appl. Phys., 11:318-324.

DeVries, D. A., 1950, Some remarks on gaseous diffusion in soils: Tran. 4th Int. Cong. Soil Sci., 4:41-43.

Evans, D. D., 1965, Gas movement, jm Black, C. A., ed., Methods of soil analysis: American Society of Agronomy, Madison, Wisconsin, pp. 319-330.

Fick, A., 1855, Ueber Diffusion. Annalen der Physik. XCIV. pp. 59-86.

Fourier, J. B.s 1822, Theorie analytique de la chaleur: English Trans­lation by A. Freeman, Dover Publ., New York, 1955.

Freeze, R. A., and Cherry, J. A., 1979, Ground Water: Prentice-Hall, Inc., Englewood Cliffs, New Jersey, p. 95.

Gallant, R. W., 1968, Physical Properties of Hydrocarbons: Gulf Pub­lishing Co., Houston, Texas.

71

72

REFERENCES CITED—Continued

Keen, B. A., 1931, The Physical Properties of the Soil. Longmans, Green and Co., London-New York, p. 380.

Lowenheim, Frederich A., and Moran, Marguerita K., 1975, Industrial Chemicals. Faith, Kayes, and Clark's 4th Edition: Wiley Inter-science Publications, New York.

Marshall, T. J., 1959, The diffusion of gas through porous media: J. Soil Sci., 10:79-82.

Mellan, Ibert, 1977, Industrial Solvents Handbook: Noyes Data Corpora­tion, Park Ridge, New Jersey.

Millington, R. J., 1959, Gas diffusion in porous media: Science, 130:100-102.

Penman, H. L., 1940, Gas and vapor movements in the soil. I. The dif­fusion of vapor through porous solids: J. Agric. Sci., 30:437-462.

Ramalho, R. S., 1977, Introduction to Wastewater Treatment Processes: Academic Press, New York, pp. 47-51.

Russell, A. D., and Thompson, G. M., 1983, Mechanisms leading to enrich­ment of atmospheric fluorocarbons CC13F and CCl2F2ground water: Water Resources Research, p. 57, February.

Slatterly, J. C., and Bird, R. B., 1958, Calculation of the diffusion coefficient of dilute gases and of the self-diffusion coefficient of dense gases: American Geophysical Union Transactions, 18th Annual Meeting, Vol. 4, No. 2, pp. 137-142.

Sorooshian, S., 1984, As per verbal communication, Department of Hydrology and Water Resources, University of Arizona, Tucson, Arizona.

Thompson, G. M., and Stiles, G. K., 1981, Final evaluation of six fluo­rocarbons for use as groundwater tracers: Report for Environmental Sciences Division, Oak Ridge National Laboratory, under terms of University of Arizona, Subcontract 7224.

Weeks, E. P., Earp, D. E., and Thompson, G. M., 1982, Use of Atmospheric Fluorocarbons F-11 and F-12 to Determine the Diffusion Parameters of Unsaturated Zone in the Southern High Plains of Texas: Submitted Jto_ the Water Resources Research Center, University of Arizona, Tucson Arizona.

73

REFERENCES CITED-Continued

Wilson, J. T., Enfield, C. G., Dunlap, W. J., Cosby, P. L., Foster, D. A., and Baskin, L. B., 1981, Transport and fate of selected organic pollutants in a sandy soil: Journal of Environmental Qual­ity, Vol. 10, No. 4.

Yaws, Carl L., 1975, Physical Properties: A Guide to the Physical, Thermodynamic, and Transport Property Data of Industrially Important Chemical Compounds. Department of Chemical Engineering, Lamar Uni­versity, Beaumont, Texas.

Cft \ §\

J . j>W- %***. t

««&& £ U>x£&»£->iSE3ia7 g3 *faF»f —v t5ts^n^ !?grsfl?K.,CTMEavKjgni«rco>caaryt«Via !ifi v ... s v •

3 v. #

ystua -e,u 3is& £ u im fc iatatss^^g&MS^<%,<wy«3flyT^ iT^T>r*gc»MB»g?j»tt»gpag>aax; _ S0xzu¥v xmjsGr:&*Wiite is&>aii&a caffaa*m*7z&a&aFza&fi>e&£issG*3a • ' bwtj

I OI fcl?

j t.o i $ l<ll

. - r |ON O'^ ' ANS-''''

0

v LiNtS)

0 CD

'i\: '' A L • "1 MO:"-;' i

X3A1} ]

' i . J 4 r

o o o o

/, <> cxi

•<o '•'/' • 5 ' . - Z .

'** '5-

I g / -

| A311VA

o* ro

i— XT r

CO

<

\ \

© ®

V Ai. k

r x : s < I N ix; vv N v o w r A I RPOR T

; h, A CI' T i v' L : ENTION SA.SIN

• r - f I f ' ' , r j f \

tl" i s*

i i r 8 p i 9 '

I, .,~4 I '

i v

4—

* ' C.

&r:. § • #=*

8 t *

V

RPO R f D^TFNTiON BASIN

J- a |

JIRI-20

•43

X

M \

\ R -8

:'iC cvuv-tiot'

- ;*t - v •

ftSC .»

-,-rrp; 2

cys'j:' '.*rr jrr:

• i)

' Vt'*suu?v vrii' ajwi-'

4 Rhl

Jtt&hS A

I .

15

S.

• • • yri MARTINEZ H-flL / •Hb 4

SAM R A N C 11 i.

Di. MARTINL, LA N D ; - Pi A ' ,/•'

A EXPLANATION

EXISTING WELL

• PUBLIC SUPPLY

Q nnuceTir

Vv /\

(A ND X

X RfS

t c/vad o sono

GRANCE

c

V A v 's £10 A r \ • > ' J i \

j i?

! K

DN RI3E RI4E

REN

PR

REMEDIAL INVESTIGATION-PHASE I PROPOSED MONITOR WELL SITES

©

A Hi

v*::;

L INVESTIGATION-PHASE I 3ED MONITOR WELL SITES

EXPLANATION /\

\ EXISTING WELL

• PUBLIC SUPPLY

© DOMESTIC

0 IRRIGATION/INDUSTRIAL

0 UNUSED/ABANDONED

MONITOR WELL ° TUCSON WATER 8 IBM

• REMEDIAL INVESTIGATION

^ EPA

IV.

LOCATION OF THE

CARRANZA FIELD

STUDY SITE Source.: A2. Dept. of yeaLlfi Services

Mat/ lOorfcplan. far JC *f Xcmeci/a/ JZstfe*-4*jA&*n of

'T'ucstn /hrpfirt /Ire*.

A

36 MAXIMUM TRICHLOR IN XIG/L FOR A TUC

<0 6= LESS THAN ND = NOT DETECTI

RI -4 : REMEDIAL \TI

_|_ : WELL SITE

N '• PRIORITY =*

R|~4 ;WELL NAME

• POTENTIAL SC

AME OF

A=BURR BROWN B=WEST CAP C=HANGERS-CONSOL D=TURKEY FARM Dl E= ARIZONA AIR NA1

A

ON RI3E RI4E

A \

SICHLOROETHENE CONCENTRATION R A TUCSON WATER MONITOR WELL

THAN 0.6^g / I ETECTED (DE T E C T I O N L I M I T O F < 0 . 3 « g / l )

EDIAL INVESTIGATION WELL SITE

TRICHLOROETHENE CONCENTRATION CONTOUR

5 5= 5MG/ I OF TC E

L SITE

R I T Y # F R O M 8 - 8 3 " P R O P O S E D " S I T E M A P

L NAME

1983 GROUND WATER ELEVATION CONTOUR

2450 TIAL SOURCE AREA OF CONTAMINATION 2 4 5 0 = 2 4 5 0

AME OF SOURCE CONTOUR INTERVAL-10 FEET

*OWN

-CONSOLIDATED AIRCRAFT CO., GRAND CENTRAL AIRCRAFT CO., DOUGLAS AIRCRAFT CO.. FARM DUMP

AIR NATIONAL GUARD

INTERNATIONAL. AIRPORT

REMEDIAL INVESTIGATION-PHASE I PROPOSED MONITOR WELL SITES

-vf-;; * /'•% % :v"" i: :y-'\V"-' /i'. * j :4<>. :• >• U r, S -"'U

ARIZONA DEPARIIVENT Or HEALTH SERVICES BUREAU OF WATER QUAUTY CONTROL |j

MAP 2-9 o FEET o 500 1000 2000 3000

"••••mil I I I I I FEET

I « 12.000

MARCH 1984

/ESTIGATION-PHASE I MONITOR WELL SITES

y&fv'i' HEALTH SERVICES I

H'ER Q.UA i; TV CONTROL i

IAP 2-9 >0 1000 2000 . 3000

FEET I < 12000

MARCH 1984 Kirk Alum. 7% 0 m tan

Aia icr mf Science.

MjJrtkjy i Madtr Xcs.

*f fjrtzan*.

J90S-