IR 05000429/2005001

Trip Rept of 850429-0501 Visit to EPAs 11th Annual Research Symposioum on Land Disposal of Hazardous Waste in Cincinnati,Oh

ML20199B717
Person / Time
Site:	05000429
Issue date:	06/04/1985
From:	Dale Goode NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS)
To:	Fliegel M NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS)
Shared Package
ML20198R768	List: IR 05000429/2005001 ML20198R949
References
FOIA-99-21 NUDOCS 9901140063
Download: ML20199B717 (37)

Text

_ _

o .., - -

.

-

.

-

DISTRIBUTION:

lJ

!. JUN 0 41985 a 204 s/f Pa rW WM 406.3 s/f JStan er'

204/406.3/DG/85/06/03 NMSS r/f PJustus-1, WMGT r/f KJackson L

DGoode & r/f JGreeves MKnapp E0'Donnel l

JBunting WMGi Staff MEMORANDUM FOR: Myron Fliegel, Section Leader, WMGT MJBell REBrowning l

FROM:' Dan Goode, WMGT LHigginbotham SUBJECT:

TRIP REPORT - EPA'S ELEVENTH ANNUAL RESEARCH. SYMPOSIUM ON l LAND DISPOSAL OF HAZARDOUS WASTE, 29 APRIL - 1 MAY, 1985 l

!

I attended EPA's annual hazardous waste disposal symposium in Cincinnati, Ohi Papers were presented on about half of the projects funded by the EPA Hazardous Waste Research Lab. After the symposium, I visited the EPA offices and discussed EPA and NRC research with Walter Grubbe, Project Officer. Attached

,

are several documents, provided by Grubbe, which summarize EPA's research l

'

program out of Cincinnati, and which identify project officers (with telephone numbers) and principal investigators. I have distributed these documents to Ed O'Donnell, RE EPA's research program' includes many pratical experiments which c'an be readily applied to our regulation of LLW and uranium tailings. My summary notes and a program from the symposium are attached. I have the published proceedings and the attendees addresses.. Two areas of particular interest were demonstration of clay cover performance and comparison of laboratory measurements of ~

hydraulic conductivity (K) to field measurement EPA's research indicates that clay soil covers can be constructed to adequately limit infiltration into disposal units. Thc overriding factor in determining effectiveness of the cover is quality control during construction. Lifts should be thin (3 inches) and the soil should be compacted at or wet of optimum moisture content. It is probably necessary to break up large clods by han The surface of each lift should N acarified to ensu're continuity of the soil barrier. EPA is also beginning res. arch on the control of evapotranspiration through plant management and establishment of stable plant communities. These considerations are especially important in evaluating long tem performance and stabilit EPA's investigators indicate that laboratory measurenents, even on l'

'

" undisturbed" samples, do not indicate the effective hydraulic conductivity of natural soils or constructed covers. Field values detemined by water balance measurements on experimental and operating trenches are nominally 1000 times larger than. values determined in the. lab. The implications are that laboratory i

values of hydraulic conductivity should not be accepted for licensing decisions s unless they are discounted (multiplied) by several orders of magnitude. Simple field tests which accurately measure the large scale effective hydraulic

- conductivity of clay soils are highly reconsnende **

C -' :  : : : : : :

--:-- 9901140063 981217 -

---

:----------- *------------:-----------

4E :' PDR FOIA r.. : ........E9t!!MR,!? ,2,1,,,,, ;PpR,,,,,,,,,, ; ,,,,,,,,,,,, :;,,,,,,,,,,,, : ,,,,,,,,,,,, : ,,,,,,,,,,,

$:85/06/03  : : : : : :

990l/4DM3

. . -

._ . .. , __

.

' ~ ~

-204/406.3/0G/85/06/03

^

_2

,

I plan to' continue coordination with EPA researchers in Cincinnati. I have also coordinated ;this-information with Ed O'Donnell (RES) who is distributing it to RES staff. .Please contact me if you have any questions or would like to-take follow.up actio I Dan Goode, WMGT

!

.

.

.

f

.

%

l l

l Enclosures available in DC $

-

i

.

: : : : : : ;

..:. .....:......__....:__..____....:__...__....:..___.....__:............:___........

. . . . : : :

..

___ .e....:.........__.:............:............:.___._______........_____:...........

@;:85/06/q$L{ : : : : : :

N l-

._. _ _ __ _ .___ -_ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _

• .-

._ .

a

.. . .

EPA SYMPOSIUM ATTACH DANG-i-1-SUMMARY NOTES FROM EPA'S ELEVENTH ANNUAL RESEARCH SYMPOSIUM ON LAND DISPOSAL OF HAZARDOUS WASTE MAIN 7ENANCE FREE VEGETATIVE SYSTEMS FOR LAN0 FILL COVERS - John Rodgers (Los Alamos) presented experimental and theoretical work on the long term removal of moisture from trench covers by stable plant connunities. A stable comunity is one which would perpetuate itself and not recuire mainterance. Plants have metabolic cycles which control the amount of root extraction and ,

j evapotranspiratio Even in a very arid climate, like Los Alamos, deep l percolation will occasionally occur because a large precipitation event will coincide with a lull in plant metabolism. The plants cannot respond quickly i enough to prevent moisture from passing through the root zone into deeper horizons. The HELP model was applied to this experiment and its results, but HELP did not accurately predict moisture content in the cover. Rodgers believes that is due to some of the simplifying assumptions of HELP. This work is relevant to current NRC researc '

HYDRAULIC CON 00CTIVITY OF TWO PROTOTYPE CLAY LINERS - David E. Daniel (Ul Texas Austin) has, for several years, had the position that clay soils can function as appropriate covers (instead of flexible membrane liners due to '

their natural low conductivity. He reported results of field experiments on large (20' x 60') artificial trenches. The clay was installed in 2 three inch lifts and compacted with a hand operated vibrating sheep's foot roller. Clods were broken by hand. Two covers were used; the first with a lab conductivity of IE-8 cm/sec (Iow plasticity); the second with a lab conductivity of IE-9 cm/sec (high plasticity). Three inch Shelby tube sampics (undisturbed) yielded lab conductivities of IE-8 and 3E-9 cm/sec. Water was ponded on the cover and collected from the trcnches. Using Darcy's law, the hydraulic conductivities of the two cover systems were IE-S and 4E-6 cm/sec, respectively. These values are about 1000 times larger than the laboratory values. Infiltrometer tests (several sizes) yielded hydraulic conductivity values close to the actual field scale value. Daniel argued that a low conductivity clay cover could be constructed, and that only field tests are accurate in predicting field scale conductivity.

EFFECTIVE POROSITY 0F GE0 LOGIC MATERIALS - James P. Gibb (Ill State Water Survey) reported on lab experiments to define effective porosity. Borrowing concepts from chromatography, Gibb argued that a more appropriate term would be

" dynamic" porosity. This would reflect that fact that effective porosity is not a geologic media property, but is a characteristic of the media and the solute or tracer of interest. It is not constant for a given rock. For non-reactive tracers, spheric excletion causes large molecules to move faster than small molecules. This is due ta more thorough mixing of the small '

molecules with the water in the pores. Larger molecules are excluded from

- - . . - - - ~ - - - - . - . - - - - - . - - - - - - -

'

.

^

EPA SYMPOSIUM ATTACH DANG i-2-mixing and diffusing into small pores. If all other effects are ecual, l tritium, which is bound in a very small water molecule, will move slower than '

l other non-reactive tracers. This effect has been observed in the lab with ( columns of glass beads and Ottawa san FIELD VERIFICATION OF LAN0 FILL COVER SYSTEM CONSTRUCTION TO PROVIDE HYOROLOGIC ISOLATION - Richard Warner (V. Kentucky) has constructed three artificial l trenches with clay covers which have been monitored under natural precipitation l . conditions. One of the covers appear to be leaking along the edge of the l trench due to sealing problems. In the other two cases, no significant moisture moverant has been observed into the bottom of the trenches. The l covers consist of a thick layer of sand, compacted clay (capillary barrierl a j sand drainage layer, and topsoil. The next phase of the experiment will l involve artifically high precipitation. This work, since it is being perfomed in Kentucky, may have direct applicability to closure of the Maxey Flats sit MECHANISMS OF CONTAMINANT MIGRATION THROUGH A CLAY RARRIER - CASE $TUDY, WILSONVILLE, ILLINOIS - Robert Griffin (Il State Geological Survey, Champaign)

related results of investigations at the failed landfill site. The trenches were in the water table and the surficial geologic materials were glacial tills

! with high clay content. Laboratory measurements of hydraulic conductivity (K)

ranged from 1E-8 to 1E-10 cm/sec (very lo'w). Chemicals migrated at rates much higher than predicted. Field investigations (including 45 degree angled boreholes) indicated oxidation deep in the till around joints. Single well tests in the angled boreholes indicated hydraulic conductivity from 1E-7 up to 2.4E-5 cm/sec. Griffin reconmended that if laboratory values of conductivity are used for regulatory purposes, they should be discoented (multiplied) by a factor of 100 to 1,000. Contrary to previous thought, in-place tills appear to have natural secondary penneability which is a significant flow path. Griffin also reported that xylene measurements in monitoring wells increased for the l first 4-6 hours after pumping dry, and then slowly dropped to zer EVALUATION OF CHEMICAL GROUT INJECTION TECHNIOUES FOR HAZARDOUS WASTE CONTAINPENT - Philip Malone (Corps of Engineers, Vicksburg) conducted field experiments to determine if grout sealing could provide a barrier to moisture and contaminant migration. Both a small scale and a large scale field test indicated that the grout sealing was not successful in limiting infiltration.

l For the large scale test, grout was injected on 5 foot centers and at a depth of 8 feet in a sandy soil with a little clay. The pH was relatively hich, about The plot was excavated and large gaps were found between grout pod In addition, shells were formed around some of the pods with clean, completely ungrouted sand immediately around the well point (a flow path). There was no significant difference between the overall conductivity of the grouted plots 3 l

l-

'

f

,

!

. - - . _ _ -

_ - _ - - _ . - . .. . -. - _ - - - . . - . ....__ _ - - -.-

-'

'

-

..,,

,- .

!

'

EPA SYMPOSIUM ATTACH DANG-3- \

and uncrouted control olots. In addition to sodium silicate, several other types of grout were also tested on the small scal CONTROL OF FUGITIVE DUST EMISSIONS AT HAZARDOUS WASTE CLEANUP SITES - Keith Rosbury (PEI Assoc.) reported on field experiments of various techniques including 11 chemicals, vegetation, and windscreens. Vegetation destroys the crust of the chemical / soil mixture and also results in highly contaminated soil around the plant stem. Windscreens were not effective on small particles (less than 10 micrometers) which are eroded at a constant rate at wind speeds above about 7 mi/hr. The chemicals were 100 percent effecti,ve for 17 to 31 day LEACHATE COLLECTION SYSTEMS - SUPMARIZING THE STATE OF THE ART - Jeffrey Bass (A.D. Little) discussed internal drainage systems and their failure, i Failure causes included sedimentation, settling, equipment loading, pipe l deterioration, and biological clogging. Most problems appear to occur early in ;

the facility life; hence access during the post closure period for maintenance could significantly improve performance. EPA requires visual inspection of leachate collection systems weekly and after stonn '

SIMULATING LAN0 FILL COVER SUBSIDENCE - Harry Sterling (Daniel Engineering, Greenville) simulated cover subsidence with a centrifuge. Clay covers compacted at optimum moisture content were able to bridge 4 ft diameter gaps in the support. Significant cracking occurred with a 8 ft diameter gap. The strength of the clay cover is very dependent on moisture content.

t

!

i i

{

,

&'OR@O1-W NOJ 4 Aoril 1985 ,

.

.

LAND DISPOSAL OF HAZARDOUS WASTE Proceedings of the Eleventh Annual Research Symposium at Cincinnati, Ohio, April 29-May 1,1985 Sponsored by the U.S. EPA, Office of Research & Development Hazardous Waste Engineering Research Laboratory

.

Land Pollution Control Division Containment Branch and Alternative Technologies Division Thermal Oestruction Branch ,

Coordinated by:

.

JACA Cor Fort Washington, Pennsylvania 19034 Contract No. 68-03-3131 Project Of ficer Naomi P. Barkley Cincinnai, Ohio 45268 HAZARDOUS WASTE ENGINEERING RESEARCH LABORATORY OFFICE OF RESEARCH AND DEVELOPMENT $

U.S. ENVIR0fMENTAL PROTECTION AGENCY CINCINNATI, OHIO 45268

1 l

l

_

'

,

,

. . ,

,

l'

'

' ~

..

i

~~. ,

l

! *

HYORAULIC CONDUCTIVITY OF TWO PROTOTYPE CLAY LINERS f

Steven R. Day Geo Con, In ,

Pittsburgh, PA 15235

David E. Daniel l University of Texas  !

Austin, Texas 78712 I'

ABSTRACT Two prototype compacted clay liners The various laboratory and field were constructed at a site near Austin, permeability tests showed that: (1) f f

Texas, using clays of high and low plasti- essentially all of the laboratory tests, city. The liners were 15 cm (6 inches) even on undisturbed samples, produced a {

thick and covered an area of approximately j

~

measured hydraulic conductivity that was 6 meters by 6 meters. The clays were one thousand times less than the actual I

,

compacted to 100% of standard Proctor den- hydraulic conductivity, and (2) field sity at a water content slightly wet of pemeability tests yielded an average hy-optimum using a sheepsfoot roller. Small draulic conductivity that was close to the

-

dikes were constructed around the p?ri- actual hydraulic conductivit meter of the test area, and water was ponded on the liners. An underdrain The findings of.this study raise system was installed to collect liquid important questions about whether labora-that flowed through the liner and to tory penneability tests on compacted clay i

measure the quantity of seepage. The are relevant to field problems and re-underdrain consisted of an impenneable inforce previous suggestions that geomembrane, freely draining filter fabric, compacted clay liners may contain numerous washed gravel, and perforated collection hydraulic defects such as fissures, pipes. The collected liquid flowed by slickensides, zones of poor bonding

-

gravity to a pit, where the quantity of between clods of clay, and zones of flow was measured and was used to calcu- relatively poor compaction. The desir-late hydraulic conductivity, ability of field permeability tests is evident from the results reporte The actual hydraulic Permeability tests on laboratory samples thetwolinerswas4x10'gonductivityof cm/sec for the clay of high plasticity and 9 x 10~0 that were tested at different effective stresses suggest that clay liners sub-cm/sec for the clay of low plasticity. A jected to significant overburden pressure variety of laboratory penneability tests may retain their hydraulic integrity much were perfonned. The tests included better than clay liners that have little permeability measurements on laboratory- or no overburden pressure, compacted specimens of clay, on samples of the actual clay liners obtained with a thin-walled sampling tube, and on hand-carved blocks of clay obtained from the actual liners. Field permeability tests were also performed using single- and ','

double-ring infiltrometer s

!

.

i t

265 l

\ ]

.

. _ . .- . . . . . _ -__ . - _ - . . .. .- - - -

'

April 1985 -

L

-

.)

!

LAND DISPOSAL OF HAZARDOUS WASTE l

' Proceedings of the Eleventh Annual Research Symposium at Cincinnati, Ohio, April 29-May 1,1985 Sponsored by the U.S. EPA, Office of Research & Development Hazarcous Waste Engineering Research Laboratory

.

Land Pollution Control Division Containment Branch and Alternative Technologies Division Thermal Destruction Branch ,

Coordinated by:

(

JACA Cor Fort Washington, Pennsylvania 19034 Contract No. 68-03-3131 l

Project Of ficer Naomi P. Barkley l' Cincinnai, Ohio 45268

HAZARDOUS WASTE ENGINEERING RESEARCH LABORATORY $

OFFICE OF RESEARCH AND DEVELOPMENT U.S. ENVIR0te4 ENTAL PROTECTION AGENCY CINCINNATI, OHIO 45268

.

. . _ - .-. __ _ __ _ _ . . _ . _ _ ____ _ '

o .

.-

,

-

~

a

,

- .

_

1 ,,. .

j

\

MECHANISMS OF CONTAMINANT MIGRATION THROUGH A CLAY RARRIER-.

- CASE STUDY, WILSONVILLE ILLIN0IS R. A. Griffin, B. L. Herzog, T. M. Jonnson, W. J. Morse, R. E. Hugnes, S. F. J. Chou, and L. R. Follmer Illinois State Geological Survey Champaign, Illinois 61820 ABSTRACT Routine monitoring of a hazardous waste disposal facility at Wilsonville, Illinois

' revealed that organic contaminants were migrating 100-1000 times faster than pre- 1

'

l-dicted. A detailed hydrogeologic investigation of the site revealed that the local groundwater flow and gradient .ere dominated by a 45-foot nign coal mine refuse pile that causes raatal flow from the pile through the trenches. Measurements of hydraulic head indicated tnat the vertical flow was much less significant than the lateral com-j

,

ponent of groundwater flow. Results of hydraulic conductivity tests and geologic '

investigations lead to the conclusion that the higner-than-predicted migration rates

' could be accounted for by the differences between laboratory and field measurements of

' hydraulic conductivities. These differences appear to be primarily due to the inability

' of small labort. tory necimens, whether " undisturbed" or recompacted, to simulate the flow through relatively large joints or partings present in natural material The distribution of contamination in the groundwater was clearly related to the flow patterns. Time series sampling of monitoring wells in slowly recharging formations showed thgt concentrations of volatile organics varied widely through the recovery perio conductivities of soils and clay could INTRODUCTION be greatly increased when exposed to Laboratory studies have shown that various organic solvents. Bridley, organic. chemicals can increase the h Wiewiora, and Wiewiera (3) and Griffin draulic conductivity of compacted clay et al. (10), found that clay minerals soils; several field studies have docu- that expand in water can undergo rela-tive collapse when exposed to organic mented instances where field-measured solvents. Murray and Quirk (15) have hydraulic conductivity of compacted clay liners exceeded the laboratory-measured shown that dry, pressed pellets of non-values used to design the fac411t expanding clays will expand when

-

These and other studies have resulted in solvated by various solvents in direct concern about the potential failure of proportion to the dielectric constant of clay barriers to control release of the solvent. These studies support a mechanism for producing higher hydraulic leachate from landfills and surface conductivities in which clay materials impoundment exposed to water, which has a relatively Interactions between clay, percola- high dielectric constant, shrink wnen ting water, organic solvents, and exposed to organic solvents with a lower constant, contaminants can' result in rates of '

migration greater than expecte Failure of clay barriers may also Anderson, Brown, and Green (2), Aca Olivert , and Field (1), and Foreman and result from measurement errors during

. Daniel (7) have reported that hydraulic design or' construction. Daniel (5)

- - ._ _

\

-

roported four case studies of clay-lingd PURPOSE .

tmpoundments chere unexpectedly high Igakage rates were traced to unde The purpose, objectives, and .

,

estimation of hydraulic conductivity approacn for the project have been because laboratory measurements were described in detail in earlier pud-used; field-measured hydraulic co lications (9 and 13).

ductivities were found to De mucn

greate This paper describes nydrogeologic conditions at the site and now tney )

Failure of some clay barriers to affect the distrioution of contaminants l j prevent leacnate migration may result in the groundwater. To determine tne l

'

d f rom natural as well as man-induced con- mecnanism responsible for tne nigne d i ditions. For example, evaluators of than-expected migration rates of organic potential landfill sites in unconsoll- contaminants at the site, nydraulic

_

L dated glacial deposits often overlook conductivity was measured by several tne existing natural joint syste metnods in the field and in the labo Contaminated aqueous wastes and organic atory, and tne effect of joints in tne solvents can move along tnese route till on the measured values was examin-Mditionally, potentia) clay-solvent ed. The results of these measurements interactions tnat can enhance these are particularly significant Decause existing natural joints or planar relatively few data have been publiened structures have usually been ignored that compare laboratory- and fiele~

during landfill and surf ace impounwent determined hydraulic conductivity n ues siting and design, from actual disposal site The case study reported here began This paper also discusses results as an outgrowth of a 1982 court order to of time series sampling of volatile exhume and remove the hazardous wastes organics from monitoring wells finisned buried at a disposal facility at in fine-grained earth materials. This Wilsonville, Illinois. The 130-acre sampling work was conducted to determine landfill operation was a trench-and-fill the ef fect of sample collection proce-

, procedurt; that basically relied on dures on the amounts of volatile l

natural attenuation of contaminants by a organics detected in water samples, clay-containing till deposit native to the sites Routine monitoring of the RESULTS l

site revealed that organic contaminants were migrating 100 to 1000 times faster The work described in this paper is l

tnan predicted. The Illinois State currently in progress. The results pr Geological Survey, supported by the sented should be considered as tentative Environmental Protection Agency, the and subject to possible reinterpretatio Illinois Environmental Protection Agencys and the site owner, SCA Hydrogeoloy1c/ Geochemical Stuoies l

l Services, Inc., began a study to deter-mine the cause or causes of contaminant Hydrogeologic and geochemical in-migration at the facility. Two obvious vestigations of the site have involved l

questions were posed: (1) Why were the completion of 17 piezometer / monitor-

! these organic contmainants migrating ing well nests (nests A to K, profiles y i faster than predicted, and (2) wnat were and W in Figure 1), with two to nine the implications for land disposal of piezometers and monitoring wells per similar wastes at other sites? nest. The details of the design, con-j l

struction, and experimental plan were The background events, overall site published previously (9,10, and 13).

l characteristics, and project description I

have been published previously (9,13, The piezometers were used initially l

and 21). The geology of the site has for in-situ hydraulic conductivity tests been described by Follmer (6) and by at various depths. After the water '

l Griffin et al. (10). levels stabilized, these piezometers were used to establish the long-term

..

piezometric surface and, in turn, the

E

.

'

V

e_ *

.e . -

% ,~

-

. g N

Coal mine cleaning refuse (Gob Pile) .'

,

Nest s Nest G A [

/og

' '

'

2*.-' ~@

% Cross

,

'

section

= hrenc, area f Nest s

, \ j' $

95 i j N st Trench

~187 i

-

est'

.,< .

'

K ' N ares A a) Trench !' s i'

.// area A l  !

. . Nest - i L )S g3 1 'l est\ < i ,\ Nest

~ . %. C ~- 40

"W fL Nyy,% .

- Water table (m), April 1984 0 Figure Site map showing: water table elevation (m) in April,1984; _

j locations of trenches, wells and cross section hydraulic gradient and flow across the This cross section shows the sequence of site. The elevation of the water table geologic materials in this area; these measured in all wells and piezameters at materials have been described in detati the site, including SCA wells in April, by Follmer (6) and Griffin et al . (10) .

1984,*is shown in Figure 1. The effect The cross section also shows the post-of the coal refuse plie on the shallow tion of the water table and equipoten-groundwater flow pattern is evident. A tial lines as measured in April, 198 groundwater mound is present beneath the Measurements of hydraulic head in wells coal refuse pile, resulting in shallow at different depths in each cluster of groundwater flow in a radial pa".ern wells in Profile V and at Nest I indi-toward trench area B to the west and cated a downward vertical nydraulic trench area A to the south. Some dis- gradient and groundwater flow. However, posal trenches in each area were apper- these were much less significant than ently excavated below tne water table; the lateral component of groundwater however, the trenches may have remained flow dominant in each geologic unit relatively dry during the disposal oper- because of the lateral hydraulic ation due to the very low hydraulic co gradien ductivity of the surrounding glacial til Hydraulic conductivity Tests ,

Figure 2 is a cross section of the The determination of nydraulic site along the line shown on Figure conductivity at the site was an

-

___

'.

SW NE

, .

[ 640 -195. _,, . uin, -

~

Spoil V

V3/ _ 2 VJ

?

- g,

~

feorin Loesh tj,r.,t:b.je ! A,p,fIU,2,,1,,9,,BS%

._w- a mamana satt to,,,

,

-190 rana isne -

,,

Vandalia ablation' l Trench Area B

. 620- till, stiff, clayey P I vandaha ablation t01,.igf.t r"hY

j Sg'7 I i o \

j610- \ l Basal Vandalia Till, altered, nointed g -185

$ g I

> N

,

600- c  ? Basal Vandiia Till,k unaltered i \ &J *

g-625- Equipotential lin ft ' @m 0 100f t

580 0 30m Figure Cross section from Profile V througn Trencn Area B to tne gob pile snowing equipotential lines of groundwater flo L l3 essential part of this project. There- Angle holes were drilleo at a 45'

fore, several methods were used to angle. The cased interval was sacx-measure i filled by pumping grout from the sottom

~

of tne casing to tne ground surface l Laboratory values of hydraulic con- rather than from the intervals of ductivity were deterinined from tests of expanding concrete and bentonite des-undisturbed and recompacted soil cores, cribed in earlier reports (9 and 13).

[I using a Harvard-type miniature permea- All other design and construction f meter. This metnod is described in details were identical to those

detail in Herzog &nd Morse (11). Addi- described for the vertical piezo-tional values were available from engin- meters. Construction details of the

.

eering. reports about the site (12 and monitoring wells and vertical 8). Field analyses included slug tests piezometers are shown in Figure analyzed by the method of Cooper, Brede- Figure 4 illustrates that the angle

,

hoeft, and Papadopulos (4) and by a holes have a much greater chance of y method proposed by Nguyen and Pinder intersecting vertical fractures because (16). Recovery tests were performed on the holes are longer and cover a greater monitoring wells that were pumped for lateral are the purpose of well development, using the technique described by Todd (22). Slug tests were originally designed to be analyzed by the method proposed by i Field tests were performed on both Cooper, Bredehoeft and Papadopulos (4),

vertical and angle holes. In addition and the type curves published by to the vertical piezometers installed at Papadopulos, Bredehoeft and Cooper 4, each nest shown in Figure 1, angle pie- (18). This method was chosen because it

- zometers were also installed at nests A, assumed boundary conditions that could B. E and K to attempt to measure the in- reasonably be met in the fiel '

fluence of vertical Joints on hydraulic j ,

, conductivity.

I

'41

30

>

.

- . . . _ _ - _ . - - _ _ - _ . . - - _ - - -

.

.., .~ * . ' . -

'

• Open hole pienometer

.

Morwtonne well

-

- during tn2 early part of the slug test

"' (usually witnin tne first nal f-nour) .

enMing coment Our slug tests were not originally c signed for analysis by this metnod, and many of tne data were less than 1oeat for analysis Dy tn15 method Decause few

• *g, early measurements nao ceen tamen. Ad pve cesing itional error may nave resulted from 4 2% in. lO using discrete measurements cotainea ey l a steel tape or electronic water level PVC casing -

meter instead of naving continuously recorded water levels, The Nguyen and Dinder method is de-

'*--- sendsiuny bntonia* 519ned to consider problems involving partially penetrating wells in aquifers for which the short-term effects of a water tante or leakage from a confining

- - . . layer can be ignored. It appears to be

_ expending , appropriate for analyzing tests in coment materials of moderate to low hydraulic

- - -- -

conductivity. The boundary conditions e- send M (except for full aquifer penetration)

,l lawwell screen and parameters to be measured are tne same as in the Cooper, Bredehoeft, ard

'l [

papadopulos metho Shelby tube hole e I l

~ ~'

- " ' " The third field metnod used was the recovery test (22). The recovery test, 4 Figure Construction details for mon- often used after an aquifer pumping l itoring wells and vertical test, applies to unsteady radial flow in piezometers, a confined aquifer, if the well fully penetrates the aquifer and well storage is negliviDie. For this study, tne r All Slug test data were reanalyzed covery test was used to analyze all i using the method of Nguyen and Pinder monitoring wells that were pumped for (16). The method is a direct calcula- developmen ; tion method that uses data obtained

.

- /*

/

! /

! /

1-- \- -----.1 - -- -/- 1-

! l's Wd answie /

'-;y U

vi

'f

/'

t-5 s

' _fr r -H 4 _Jl ,/

'

-

i w.e ,. * -- , n*

_, L :-!!wd u r_ e ,/3m1

-

, _]- __

-- ,

,

Figure 4 Parameters to be measured for slug test analysi .

l l

,

7 Yable 1 summar12es all the labora- On the basis of these tests, it ',

! tory and field hydraulic conductivity appears that the various labor 5 tory *

l data collected to date. Data for many methods used in this study can produce -

i (4 individual measurements have b;en re- comparable values of nydraulic condu .

'

ported previously (11). Field data are tivity for a given geologic materia presented as the geometric means because The various field tecnniques tested at nydraulic conductivity values in field Wilsonville also yielded similar soils are generally acknowledged to be values. However, in general, laboratory log-normally distributed (17 and 19), so

'

metnods produced much lower values than

'

b' arithmetic averaging would not be appro- field methods. Laboratory and fi el d m .

priate (14). For 5Me formations, ne results were similar only between un dis-h Il data nad a range of about three orders turbed laboratory samples and slug tests of magnitude. Because of the hetero- of the Sangamon Bt norizo h genity of glacial tills, this variabil-

l, '

1 ity is not unexpecte Laboratory results were generally less variable than field results. This h The number of values aveaaged may have occurred because fewer samples

]

! ranged from 1 for the Banner F3rmation- were run, but is proDably due to a g bedrock contact and some tests on the natural bias toward selecting conestve q Sangmon soil, to 13 for recovery tests samples for undisturbed analysis; crumb-V in the weathered, basal Vandalia Til ling or fractured samples were not

., Tne laboratory values are reported as chosen for testing, thus leading to more

"l ranges because they include data report- uniform results ano bias toward lower C" ed by consultants who did not report val ues . Recompaction of samples also I '

I individual values (8 and 12). yields more uniform results because the j original structure, including any

As reported in Table 1, the range natural joints present, are destroyed

'

of values determined in the laboratory during the sample preparation step '

for tests of each type of sample, di This suggests that laboratory measure-I turbed and undisturbed, is small . The ments of hydraulic conductivity should undisturbed samples yielded higher not be substituted for field measure-values of hydraulic conductivity than ments unless caution is used in their

. i the disturbed samples for soft sedi- interpretatio ments, but the reverse is trae for the

-

i hard till The biggest disadvantage of the

H Cooper, Bredehoef t, and Papadopulos

,'

Whereas the field data showed a method is the length of time necessary i wide range of values, the geometric to obtain results for materials that means calculated from the data from the have a very low hydraulic conductiv-g three methods were nearly equal. The ity. Some tests in this study ran for

-

most divergent value was found in the more than a year before water levels i Sangamon Soil St horizon for which there stabilized. Therefore, error may be

'

was only one slug test site and one re- introduced by seasonal changes in l

covery test site and these were widely regional water levels and by barometric

' '

ll separated (1/2 mile apart). The unalt- fluctuations. A second yoblem results ered basal Vandalia Till also shows from the assumption that wells fully disagreement between slug tests and penetrate the aquifer, recovery tests. For this unit, the j higher mean value from the recovery Recovery tests seem a reasonable tests can be explained by noting that

'

'

choice for monitoring wells, but it may the recovery tests were performed on be dif ficult to meet the assumption of a

monitoring wells and that three of the steady pumping rate, especially when the f nine wells used for recovery test forination has a very low hyoraulic con-I

!' analysis were finished in zones that ductivity. Another problem with the included a sand lens in their screened recovery method is its basis on the as-interval, thus yielding higher values seption that the aquifer is fully 3 i than from wells or piezometers finished penetrated by the well screen--an

.? in the matrix material, assumption that is not usually vali Constructing monitoring wells to meet i ,

-

, 32

.. . _ _ - _ _ . . - - - - -

_

' -

,

,,

.

-

Rnls latter condition usually reduces Pinder method because data that cannot tneir effectiveness as monitoring reasonably be analyzed using one slug wells. However, the test can De test metnod may De analyzed of the perfomed relatively quickly as part of other, a well development program to give additional information on tne geologic Contaminant Migration unit tnat the well monitor The nydraulic conductivity tests Ine Nguyen and Pinder method of fers Could expidin tne rate of contaminant several aavantages over the otner field migration. Howevar. It is equally metnods. Inis metnod was designed for important to determine tne pattern of partially penetrating wells--tne most contamination. Therefore, a separate common type of well construction. In set of monitoring wells was constructed addition, the test usually takes less to obtain water samples for chemical than a day to run; therefore, it is not analyses at each of the nests and pro-influenced by seasonal water level files (Figure 1). Cores from these changes, and barometric fluctuations are borings were analyzed enemically for minimized. The biggest disadvantage organic and inorganic constituents. The appears to be tnat small' errors in results, described in Grif fin et al .

water-level measurements will af fect the (10), indicated tnat the zones of nign-results of this method more than those est contaminant concentration were in from the other two field method the soft ablation till and tne wetn-However, these errors can De minimized ered, jointed basal till (figure'5).

by using a sensitive, continuous water- The results of in-situ field tests level recorde (Table 1) indicated that tre nyaraulic conductivities of the soft ablation zone Geometric mean values of hydraulic and weathered basal till are greater conductivity from the angle tests were th&n those of the underlying or over-greater than the corresponding values lying units; this probably accounts for from the vertical tests for both types the higher concentrations in the soil of slug test analyses. This was not and apparently higher migration rates in necessarily true for all piezameter this zon pairs. The mean dif ference ranged from less than. a f actor of two for the abla- The extent of contaminant migration tion zone to more than an order of in the groundwater flow system at pro-magnitude for lower casal till. This file V is illustrated in Figure 5. whicn difference was expected because angia shows the concentrations of trichloro-holes should intersect more of the H ethylene (vg/L) in grov*Ater in dominantly vertical fractures (Figui a September, 1984 The distribution of 4). Although these results are inter- contamination in this cross,section is esting ,they are not conclusive because clearly related to the my;Ured ground-the theory for analyzing angle piezo- water flow patterns in Figure 2. How-meters has not yet been fully developed ever, the highest levels of contamina-and the statistical significance of the tion were found in wells finished in differences between results for vertical sand lenses within the upper part of the and angle-holes has not been determine unweathered, basal till zone. Contam-inants from Trench Area B intenduced Although the Nguyen and Pinder into the soft ablation till and jointed method for the analysis of data from basal till must be able to move downward slug tests seems to be the best of tne into the unweathered Dasal till through three field methods tested, the recovery interconnected joints or sand lense test and the Cooper, Bredehoef t, and The extent of contaminant migration in Papadopulos methcd of slug test analysis this area is significantly affected by should not be aDandoned. The recovery the strong lateral component of ground-test is the only one of these tests that water flow away fr a the trench are can be performed on individual monitor- $

ing wells after p a ping. The Cooper, Bredehoeft, and Papadopulos methoc could be used to supplement the Nguyen and

. .

. . . . . . . , . . . . . - . . . . . . . _ . . . . . . . . . . . . .. . .

.

,

g

. 4 I e I -

1- -

T  ? T ?

'

B  ? ~ ~ ~

t5 2 2 2 = 2 'O 'O

~

'O -

., -[4 2 3 -

4 4

~ -

4 i

-

4 e

s

_

i E

& 6

. C

j . O O O TY oc TIoc I

% - -O - - - -- --

o

-

8. E -1 44 43 44 44  :

; - - ~ ~- -

E D'

x A N g :

. l

-$ . .

.*

$. O YO Y O TT Ii . - - - _o _o _o _o

=

I ,f E 3 -

-

- -

4 ,4,

~ 44

-~

t

w 4 ?o . e e g -

'O

- '.o 'a

-

x

"

I.-

-

-

i

-

e 3 2

. 2,

-

3. z .e

-

, , ,

3 .

2 j e{ s5 * - i = t m a

~

2 2 2 2 I 4 4 4 4 o $. ~ e 4 4,

• '

. -

" *

-

o a ., a ?o 3 - o

-

- -

. C -- =.

I w .

-

m l m  ;

~

a ~

.

. =( y *. a

,

e 9a

.

5 -

o

- .

a

- -

. A 2 2 4 4 3 a 4

-

. .

3 3

.. . -

.

.

..

.- .

-

. .

% 2 3 3r 3

- -

b

-

b

-

b

.F -

b

.F -

b

-.

b

- b

-

b

-

b t 2 %.  %. 2 %. 3 %,

o 8 ,,,

. .

O OB B

~

o < o - - 2.- 8-

.

,

8 .

8-

-

-

-

• F

--

--

-*

8 : I

-.- , . ,

$*I

.

2*

- .-

. 8.,

-

.

.!

.,

W "

"X 32 3 8 -

- - --

"S. ,g -

j~

-Jj 2j

-

s

• 2

.- --

IE eg

-

a l

-

l$

- E{~a ~

E,[ 8 a~{i 3". "1.f se -41 E

.

e I e -

,

• i

_ . . . _ _ _ _ _ _ _ _ . _ __

'

,' I

'

'.. '

.' SW NE

-

ft m /

.

640 -195

-

Mine

'

l 3 630- -

l feoria Loess ggelableJAgrel 12,1984)

• - " ' ' #eo 190

" " ""' S d '

Vandalia ablation' 7 Trench Area 8 M'

J " ens

' Loess sar; 620- till, stoff, claf /

e N

== (? W [ . [/ / Van'daha ablation till. sof t. mushy

~

.$  ?:- l / - *

~ 610- / al Wndalia,T, ill, altered, jointed-185 /

A -A h hu.EL fohvandaua S*

590 -180 ss ~

s

- d riil, N

l Trichloroethylene O 100ft (sept., ggg4; "*

580- 0 30m Figure Cross section from Profile V through Trenen Area B to the 900 pile showing distribution and Trienloroethylene (ppd) in groundwate Volatile Organic Samoling in Fin Gralned Mater 1al '8 0.xyLgNE, WELL V30 n-A protocol nas not oeen establisned for collecting water samples from 20 -

monitoring wells fintsned in fine-

"-

grained materials with slow recharge rates for analysis of the water's .: ,- -

volatile organic content. Schuller,

, , ,

Gibb, and Grif fin (20) have recommended

-

evacuation of 4 to 6 well volumes to n 2 o xvLane, weLL v20

. purge the well of stagnant water prior  % to collection of smples. However, this 3,

, is not possible for wells with very slow j *"

i

. recharge rates; most wells at the W11sonv111e site could be pumped down to  :-

the top of the well screen after remov- r~

( '

'

_

ing less than 2 well volumes. Thus, the ' ' '

question arises as to what impact the well recovery time has on the enemical is o.xyLapes, wsLL v'io composition of water samples collected ..

for analysi ,,

_ _ _

To address this question, sets of

~~ ~

a-time-series water smples were collected  ;

at several wells. Representative re- ...

suits obtained for o-xylene from three '

. 4 de se wells are shown in Figure 6. The re- wem mearvta*0***3 ,

sults clearly indicate that changes in

, o xylene concentration cccur as a Figure 6. Volatile Organic Concentra-

function of well recovery tim tioc,vs. Time

i a;

_ _ _ _ _ - _ _ _ _ _ _

- - - - _ _ - - _ _ _ - _ - _

i_ *

.

-

Sampleswerecollectgdusinga conductivity of materials underlying tne

'I double eneck valve Teflon batier with a site, evaluation of distribution of con-bottom draining device (Timco Mfg., taminants, and testing of experimental -

Prairie Du Sac, Wisconsin)*. Samples protocols for sampling of volatile weredrainedinto40-mLvolatileogganic organic compounds. Tne hydraulic con-analysis (V0A) vials with a Teflon - ductivity portion of the study is r faced s.eptum and a screw cap. The vials essentially completed. Results of fI were placed on ice immediately after laboratory tests of hydraulic conduc-

,

collection and transported to the lab- tivity were found to be consistently oratory witnin 48 nours of collection, lower than those f rom in-situ field tests for most materials at tnis site, f Samples collected before purging This is thougnt to be due to the f act

.g i

the well of stagnant water (time 0) that laboratory tests do not generally t contained o-xylene concentrations that indicate the effects of macrostructures l were below detection limits at two of such as joints and sand lenses. In-situ the tnree wells for wnich data are shown tests using angle holes were used to

'

lL I in Figure 6. This is probably due to assess the significance of vertical

. volatilization losses of the compound joints. The in-situ hydraulic conduc-l resulting from the water standing in the tivity tests and the geologic investiga-

'

l ;' well. Af ter the initial samples were tion both revealed that the upper collected, the well was pumped down to portion of the glacial till underlying the top of the 2-foot well screen witn a the trenches was highly jointed. The bladder-pump; the pump was then removed differences between predicted and actual l f rom the well . Samples were tnen col- migration rates, therefore, can ce e l lected by oailer at intervals of plained by the fact that only laooratory ,

q .

approximately 1, 2, 4, 6, 24, 31, 48, determined values of hydraulic conduc- '

i 124, 290, and 460 hours0.00532 days <br />0.128 hours <br />7.60582e-4 weeks <br />1.7503e-4 months <br />. Maximum tivity were used for the prediction <6 concentrations mere ootained af ter i approximately 2-6 nours of recharg Groundwater sampling revealed that g Concentrations fell to low values the hignest degree of contamination was i

between 24 and 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> of recharge, within the unaltered, casal vandalia i

Concentrations of o-xylene in all Till. The depth of contaminant migra-samples collected after 124, 290, and tion was greater than expected based on

-

460 hour0.00532 days <br />0.128 hours <br />7.60582e-4 weeks <br />1.7503e-4 months <br />,s were below detection limits, measurements of hydraulic gradient and, ( We interpret the data to indicate as such, the pathway is uncertain. The pathway may involve migration enrougn that fresh formation water containing o- joints and along shear planes, basically xylene enters the well during the controlled by the fracture network, initial hours of recharge after the well Flow may also have occurred through sand

, is pumped down. The concentration of o- lenses and through the fine-grained xylene in the water column is then

,

matri I appare'ntly depleted over time due to volatilization losses. The sampling Time-series sampling of monitoring protocol can affect the results obtained wells prior to, during and after punping

), from slowly recharging wells relative to revealed that o-xylene concentrations

volatile organic concentrations, reached a maximwe af ter 2 to 8 nours of recharge to the well. Additional stud-SumAAT AND CONCLUSIONS ies of the behavior of other compounds, l

currently underway, are necessary before This paper has presented results of a protocol can be proposed for sampling

'

three phases of the investigation of the of volatile organics in fine-grained, Wilsonville hazardous waste disposal low-permeability materials, site: determination of the hydraulic jl ACKNOWLEDGMENTS

.;

- -

• Mention of trade names or conenercial The authors wish to acknowled9e $

products does not constitute endorsement partial support of this project by SCA

. ,

or recoassendation for us Chemical Services, Inc., Wilsonville,

,

Illinois; the Illinois Environmental l e

'

, , . -

t

_ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _

.,. - -.~ - _ _ - . - . . . - . . . . . . - - . ~ ._.-.-. . - - .. ___ . - . -.

-

-

e -

s.,. ,'"

'

~ '

.

.

-

- Protection Agency; and the U.S. Envir- 7 Forman, D. E and D. E. Daniel, onmental Protection Agency, Cincinnati, 1984 Effect of nyaraulic graate,nt OH, under cooperative Agreement No, and metnod of testing on tne ny-R810442-01; Dr. Micnael Roulier is the draulic conductivity of compacted U.S. EPA project of fice clay to water, methanol, and ne p- !

tan U.S. Environmental pret<c-

.The authors also gratefully acknow- tion Agency, Cincinnati, OH, EPA-ledge the contributions of C. J. Ston /9-84-0U ,

D. N. Cote, W. J. Su, P. B. DuMontelle,  ;

,

and K. Cartwrignt to tnis manuscrip .. Geoengineering, Inc. 1982. Pro- 1 posed remedial action plan, Eartn-l~

l REFERENCES line Site, Wilsonville, Illinois,

! - ,- .

SCA Chemical Services, Inc.,

!

' Acar, Y. B.,1. Olivieri, and S. Boston, MA, 22pp with appendice ! ,

Field. 1984. Organic fluid ef- i

'

f acts on the structural stability Grif fin, R. A., K. Cartwright, ;

of compacted kaolinite. In Pro- B. DuMontelle, L. R. Follmer, C. i

'

ceedings of the Tentn AnnTal Re- Stohr, T. M. Johnson, M. M. Killey, l

,

search Symposium, Ft. Mitchell, KY, R. E. Hughes, B. L. Herzog, and $

April 3-5,1984; U.S. Environmental ' J. Morse. 1983. Investigation of

Protection Agency, Cincinnati, OH, clay soil behavior and migration of

EPA-600/9-84-00 industrial chemicals at Wilson- i ville, Illinois. U.S. Environmen- l l Anderson, 0., K. W. Brown, and tal Protection Agency, Cincinnati, Greene.' 1982. Effect of organic- OH, EPA-600/9-83-018, pp70-7 fluids on the perweability of clay soil liners. In Land Disposal of 10. . Gri f fin, A., R. E. Hughes , L. Hazardous Waste " EPA-600/9-82-002, Follmer, C. J. Stohr, W. J. Morse, U.S. Environmental Protection T. M. Johnson, J. K. Bartz, J. Agency, Cincinnati, OH, pp179-19 Steele, K. Cartwright, M. Killey, and P. B. DuMontell . Brindley, G. W., K. Wiewiora, and 1984 Migration of industrial A. Wiewiora. 1969. Intracrystal- chemicals and soil-waste inter-line spiling of montmorillonite in actions at Wilsonville, Illinois.

, some water-organic mixtures (clay- U.S. Environmental Protection

! organic studies. XVII): Ame Agency, Cincinnati, OH. EPA-600/9-

! ,Mineraloeist, V 54, pp163M .

l Cooper, H. H., J. D. Bredehoeft, 11. Herzog, 3. t. and W. J. Mors and I. S. Papadopulos. 1967. "Re- 1984. A comparison of laboratory

,. sponpa of a finite-diaseter well to and field determined values of

- an instantaneous charge of water," hydraulic conductivity at a waste l' Water Resources Research, 3:1, disposal site. Seventh Annual

-

i .

'

, p pz63-26 Madison Weste Conference Proceed-Ain s. University of Wisconsin- Daniel. D. E. 1984 Predicting Extension, Madison, WI, pp30-5 hydraulic conductivity of clay ,

'

liners. ASCE Journal of Geotech- 12. John Mathes and Associate nical Epeineerine,110:Z, pp255- 1976. Report of subsurface inves-30 tigations: proposed site of industrial residue management Follmer, L. Soil - an facility, Earth 11ne Corporation, uncertain medium for waste dispos- Wilsonville, Illinois. 38p al . Seventh Annual Madison Waste Conference Proceedings, Univers1Iy l

of Wisconsin-Extension, Madison, 3 WI, pp296 311.

l .

{

j 37 t

.

,

.

-.m 4 .-

.

_

_ _ ___ _ _- - - - - --

• f_ 13 Johnson, T. M., R. A. Gri f fin, 21 Stohr, C. . Applications

. .-

K. Cart::risht, L. R. Follmer, 8. Herzog and W. J. Morse. 1983 of close-range photogrammetry for Hydrogeologic investigations of geologic invostigations during the ,

failure mechanisms and migration of exhumation of a nazardous-waste organic chemicais at Wilsonville, disposal site. Technical Papers of Illinois. National Water Well the 49th Annual Meeting of the Association Third National Sympos- American Society of Photogrammetr ium and Exposition on Aquifer March 13-18, 1983, at Washirgton, 0.C .

Restoration and Ground Water Moni-toring My 25-27, 1983, at 22. Todd , D. K. 1980. Grounewater Columous , C Hydrology, (2nd edition). Jonn

Keisling, T. C., J. M. Davidson, Wiley and Sons 535p L. Weeks and R. D. Morrison. 1977 3 Precision with which selected soil parameters can be estimated. Soil Science 124:4 pp241-24 I 15. Murray, R. S. and J. P. Outc . The physical swelling of clays in solvents; Soil Sci. So Am. J., V. 46, pp865-86 . Nguyen 1984 V. and G. F. Pinde Ofrect' calculation of aquifer parameters in slug test analysis I.,s. Rosensneim and ,

D. Bennet S s.), Groundwater Hydraul ics . American Geopnysical l Union Water Resources Monograph 9 l pp222-23 Nielsen, D. R., J. W. Biggar and j T. Erh. 1973. Spatial variability f of field-measured soil-water  ;

properties. Hilgardia, 42:7  !

pp215-25 . Papadopulos S. S., J. i Bredehoeft and H. H. Coope i 19i3. On the Analysis of ' Slug 1 Test' Data, Water Resources Research, 9:4 pplos7 10 3 ~

19. Rogowski, A. . Watershed physics: soil variability criteria. Water Resources Research 8:4. pp1015-102 . Schullw, R. M., J. P. Gibt, and A. Griffin. 1981. Recommended sampling procedures for monitoring wells. Ground Water Monitoring  ;

Review 1:42-4 i

)

[' __ _ _ _ _ _ _ _ . - - - - - _ _ - - - - - - - - - - - - - - - - -

___

-

- - - -

,

,

, , )

l . -

' .

.- .

,i PaEDicnNc HYDRAUUC CONDUCTIVITY OF CLAY,

,

Disasselen b'y Gabstel Y. Auvinet,' M. ASCE ' ', '

1 I '

I -

I e rhaciemmansen l i ne author should be copmended for his thorough evaluation q different factors which explain the differences between the labor (

.

Dinannatone stay be submitted on any paper or sedmical ngee published in any jour-

-

Permeability test results and the actual performance of thin clay q nel or on any per ,- . .m - f at any opedetty conference or oeher meetmg the pr '

De writer agrees that field tests give more realistic values of hyda cuedings of w have been ' ' " :f by ASCE. re - = of a paper /techn& cal nose le open to anyone who l'ese Fonductivity than laborato'Y tests. De P'rmeability field test for d ne comunenw or sauseolone regardmg the cones of she paper /sechnacalnees. --  ; ese accepsed for a period of 4 mor:the Proposed described byby thethe author writer a,n Ref. is2.as a matter of fact quite similar to t%j somewing the dose of ; " ei a paper /sechnical noee, and shey should be sene so the Man ger, journals. ASCE. 345 East 47th Semt. New York. NY 10H7-2398. h . To conclude that thin clay liners should be avoided seems h .

diensessen pened may be essended by a wateien reipseet froen a discuseer ynuch too drastic to the wnter. This is besa diustrated by the em h setginal and esinee copese of the disonesian eheuld be sishmilated on 220 mm b performance of the Rio Escondido 300-ha coohng pond in Northern Y

""" _

• ""

,,8 resericeed two paierns[

'8 "' ico that the author includes in his case histories revision. Measurcy

, four pi the rate of seepage through the 60 cm clay liner were performed l

-_-;,,; pages, inchiding Egiares ased ambisel; ebe odisers we delete senteer entr peoise to the subject under discussion. M a Ale ===saat le over two long it will ing several weeks et the end of 1981 after the pond had been in d

Stion for two years. A correction for evaporation was introduced &

be returned for shortening All discueelone ese soviewed by the J and the di- 8 v6eion's or counare ; "' "- comundense. In samme cases, dear ==sa=ie wiu be r I pn measurements in a 10 rig concrete pool and a series of evaporimes viarned to disosemere for wwr96ng, or they may be =-- ;,-f se submit a paper or %e average rate of acepage was 2.4 mm/ day. Taking into accoud

3,,,,as,g",'"ge, s,e average 5 m depth of the pond and neglecting the seepage througD es ie as those der pepers. A h le subject { parthen dikes, the hydraulic conductivity of the liner could be bacQ o ,egocnon g it eenemine sneten readey sound elsewhose, advocmass opedal ensereens, le canelessly prepared, controverte estabbehed fact le purely speodetive introduca culated to be less than 3 x 10 cm/s, which is totally adequate fy

,

,,.. 5. or le foreign to the purposes of the Sodsty. AB dienassione should be Pmec pro This was a better result than those obtained in the te wneten in the shard person. and the discusser shoesid use eine term the writes when ,

3 ably due to the smalle,r relative irnportance of the seepage thrq

$he dike ,

refemns to himoeit. h author of the crismal P*Per/sechnical nose le referred to se i

'

pia have a speofic format. The title of the original paper /sechnical note it should be emphasized that this result was only achieved thrh ne at the top of the hree Strict in8Pection and control quality testing of the hner and a veg ka the month. ye rite .en. seer.e , ionis s), and number of she '

- e,auth,ge be e.id.ca usewith

,2 to a footnote ow diea27e'se.per/esdinical superscript that pa

.e-- he.e w nos culiar compaction technique which allows permeabdity to be nu,nism as discussed in Refs. 2 and 1 ,  !

es an emaanple) together w6eh his or her ASCE -- M; grade (if a w)-

h discuseer's aisle, connpeny "". and businese addsees.elicedd

-

r on --Repeneasca the aree page o* ehe manuscrips. elong we the papa number eme papa /

technical nose, the date and nasne of the jonarnelin whide le , , _ ased the original 11. Auvines, G. Y., r.nd ihnart, G., An Arhficia! Coolm Pond for the b condado Coat Fired Power Impoundesmis. Mennea nt." Prmce.fmgs ASCE Sy cuum am Surfag 9 thee the discuseer's identification footnote should foaow consecutively freen MN, Vol. II.1980. pp.10tl9-l&#t ahe esiginsi paper /sechnacel nose. If the paper / technical note esader d&=a-Aa i con-

1 emined doesnose nusnbers I and 2. the Area d'an==aa=i would begin wtah footnote 3, and

' me descueenons would continue in eespaenc ,, ; ,

, ,

Figuses s'upphed by the discuseer eliould be designated by lettere. etarting with [ This also a separately to tables and references. In eeferring to a ngure, table, or Discueelon by Wesley G. Ifolez,' F. ASCE esterence e appeared an the on$ mal paper /nedinacal riose use the easie number no sugges e posennat discussere request a copy of the Anthers' Guide to #Ac Peshincaissas of ASCE for more detailed information on preparation and outmuneson og . The writer would like to compliment the author for bringing tol manuscnpes, attention of the professton certain fundamental prmeiples which sud f, times have been forgotten. The author also points out certain necess p testing practices for obtaining reliable permeabihty dat , '

For the design of impermeable soil barners, there is a mis.o ice pi i ' February,1984, by David E. Daniel (Paper 18571).

'Research Prof., instituto de Ingeneria Ciudad Univerusaria, fLtemi.o ('

Menic 'Consultang Engr. (Fosmerly Chi-, Soils Engr , thi1R)

1456 1 1 1457 d

i

_ _ _ _ _ _ _ _

~

"  :

!

that the soils must be heavy clay soils Such soils are oft:n subjeSt to Gener:Ily, sod or ci:y Oners should be 2-5 f t thick (04-152 m). O

! undesirable shrinking and cracEng. IEss plastic soils, even well graded struction should be closely controlled by empenenced engmeer cisy:y sands cnd clayey grsvels, c:n ' permeabilities just es low, ,

l if properly moistened and compact . This fact was discussed as early j as 1932 by R. R. Proctor in his classic Engineering News Record article < --Recanences ,

l (12). Sometimes heavy compaction will be required to produce the Hec-  ; 12. Proctor, R. R., "Tundarnental Pnnaples of Sod Compaction.** LN R. V

ensary impermeability.1.arge particles block volds and thus reduce per- J Au meability (13). * *

"" 13. Hokz. 31-Sept. 28,19333. W. C., and towitz, C. W.. Comp

'

It has long been known that incra ased moistute at placement reduG Sous," ASTM STP No. 232, 195 . Hamihon, l W.; ** Percolation. Consohdation and Shear Studies.,, Ug f permeability of sod compacted to a certain density. Some of the earlies)

research work on this principle was carried,out by the US Bureau iof i 1 C., 'k' onstru of Compacted Soit Lanms5 for C'"*I'-

Reclamation in 1940. nose studies showed the'reintionships (9r a ntini- j d s Tlurd int. Conference on Sod Medum(s aad foundation Engmecri

,; ber of soils placed at different snoasture and compactive effort conditions <

u,IN3 (14). Clayey gravel soils, properly moistened and compacted have been t 16. Ilottz. W. G., " Volume Change in limpansive Clay Soils and Control b t very desirable for lining canals, ponds and equalismg reservoirs where k Treatment." Srmnd int. Conf. on Eqwnsue Clay Sods, Procredings. Vol I, soone light wave action has been antici ted (15). *e-

, it should be remembered that ge, in reality, is densa,licatio .

,Ay g % I ion and Related Properties of Resorted Od 2 joun 'I of Geotechn E gineering. ASCE, Vol.109. No.10. Ost.,196D herefore, if we compact clays to high densita'es, shrinkage shouki be q

,

reduced, as should cracking. If the facili is such that the lining'is riot '

loaded or protected against moisture launediately, perhap

,

heavily cornpacted less plastic soils'should be conaldseed -

-

, ptscussion by John A. htundell,' A. ht. AS E Another method often suitable for teducing E-p *= and shnnkage of highly plastic CH clays is the addition of lisne to the soll'(16).-Sous .

treated in this manner will act as less plastic soils and heavier compec- '

ne. author should bel commended for bringmg to hght the d tion may be required to achieve a destedd imp - ' *t '

-*-

task of predicting the in- Aace hydraube conductivity of clay Imer in all cases whether the liner be clay; clayefsoil or treated clay / the ever, the writer is afrak that the true significance of the case h'

j

'

laboratory design tests to determine permeability must be made with's the author presents may be overlooked, and that the author's .

permanent liquid having the same chemical properties as those antici- {

! tation may be used by some as evidence that clay Imers cannot poted for the facility. His includes not ertty acid and alkahne wasta lig- l signed and constructed to achieve a field permeability of less t>

t ulds but also canals, ponds and small reservoirs which are required to  ; 10-* cm/s, or that the use of laboratory permeability tests cannot -

hold waters which may have significant amounts of dissolved solids. For i instance, calcium ions can have an effect on sodium montmorillonite to predict the field hydraube charactenstics of a compacted cla'

l he writer would like to address the impact that the nature of th clays and bentonites which tend to change the clays and increase per- j defects has on these question meabilit De author discusses Officulties encountered in determining the' lab- gg, oratory and field permeabilities of liner ~matedals. It is obvious that if liner cracking is allowed to occur, the laboratory permealdlities will not It is clear from the data presented by the author that the cause l

.

represent the ultimate field condition. However, if the laboratory testing high leakage rates experienced by the referenced ponds is the p i is properly performed and a clay liner is protected, or a less plastic linin 5 of certain " defects" or zones of greater hydraulic conductivity wit l material is used, good correlations can be obtatoed. For example, labo- clay liner that allows the passage of significant quanuties of j ratory tests were conducted in 8-in. diam ste'el ring permeameters on pounded water. he nature or type of defects primanly respons t retorted oil shale proposed for a disposal pile lining, utilizmg high com- )

the high leakage rates has not been quanti *atively assc> sed by -

I pection. De matenal was classified as a silty gravel, GM. A test pond, thor, although he implies that these defects are most probably

about 100 ft x 70 ft (30.48 m x 21.34 m and 7-1/2 ft (2.29 m) deep was tion cracks, fissures, slickensides, or similar hydrauhc discontinui

, lined with 31/2 ft (1.07 m) of highly c)ompacted retorted shalel on addition, he also mentions that sods contsning rocks. roots, o the

[ bottom and 2:1 side slopes. He pond was filled with w and losses '

t ' deleterious substances, or large clod size clay chunks may im

measured through a bottom drain placed over a Hypalon lining. Four hydraulic conductivity of the line in (15.24 cm) diam cores were ontled'from the lining and were tested To assess the effect that a given magnitude of defects would h<

for permeability in the laboratory. De average rates were: Init al labo-  !

the hydraulic conductivity of a (lay liner, a simphised sccpage ratory tests = 4.5 x 10; laboratory tests on cores - 5.9 x 10; pond j tests = 3.3 x 10-' cm/s (1 x 10" = 1 ft/yr approx.) (17). .Envuonmental Research Assoc . Dept of Civ. L"6'g , Umv ..t Noise

Notre Dame. IN 4655 ,

.

1458 3459

.

. -

_____ __ _ _ __ __ ______ ______ _- -_ - ----- -- - - - - --- - -- - - - -

. _ , . _ - _ _ ._ - - - m -.-..-.- - -

g

---

"-----.- $

.

j based on Darcy's Law can be used. This modet..shown in Fig. 8, es- id , , -, -

'g. , (l

,

-,

sumes that the pore pressurz et the base of the liner is zero. From the .

1M it can be seen that >

.

,

,

q .k,IA,............................ ~~...~. ..~..$ ..~ ' Y sd - n= ,l q . kfA ,.......... ....................... n . .. ~ ~ ~ . ~ ..; (1) .

<

q - t,lA,....................... ~ ........ ~ ~ n - . - - ~ .(3)

,

.

~

g . Q + G . . . . . . . . . . . . . . . . . . . . . . . . . . ~ . - ~ ~ ~ - . - ~ ~ s (4)

'

A,.A,eA,................................~..~..-.-.4) .

,. o*

'

where k, = hydras he ce:~lucthrity of intact compacted clay tirier; k, = l'

hydraulic conductivity of defects;1, - equivalent hydraulic conductivity  ; 8 ,s f ,o

of the pond; A,, A . A, = total area of intact compacted clay liner, de- ,/ <

(j

,,

fects, and pond, remectively; G,Qn.Q = seepageihrough intact clay

-

" ,

liner. defects, and pond. respectively; and I =. hydraulic gradient,4/ +.,

In addition, the following es, '== may be used'to relaty the ama ,

,

-

of the defects to the total area of the Pond, and the abihty of the +

,

defects to the permeability of the Intact compacted y: .

A, . xA, . . . . . . . . . . . . . . . . . . . . . . . . . . . . .n : . . . . . . . ~ ~ ~ ~ ~ . ~ . (6)

.

to

'

-

k, - g ,................................ ~ ~ ~ ~ . - ~ ~ ~ - ~ - (7) '

..,

Setting Eqs. 3 and 4 equal to each other, =adundng, and edting ^-

yields the fonowing equation for the T. '4ydraulic conductivity -

? 4- "noo

. i . of the pond: I * '"- 10 80 , 18 80 ' to'

t, - (1 - x + sy)k, . .~.--. -: - ~~ I-""""")N) "*"*

d Thc " of both the quantity of defects over a given area and.the  ;

FIO. e.-Eftect of Defect Quenury and Hydraulic Conductivity on Equiv P relativ hw iic conductivity of the defect to that of the intact ca <

dreune er' ~ r, of a uned Pond

) pacted c' . . the equivalent hydraulic conductivity of a lined pond,ans a

' shown in h3 9 I x 10-8 cm/s. the pond will exhibit an equivalent hydrauhc condu The following example illustrates the use of the figure: , ity of 1.0 x 10-' cm/s,. even though 99% of the area is lined wi Given: k, = 1 x 10~* cm/s; A, = 1 x 10-, cm/s; x = 0.01 (i.e.,1Wpl

.

. ,' compacted clay having a 1 x 10-'cm/s hydraube conductivity! Althe several other rnodels and assumptions may be applicable to specific g 3 the pond area has defects); k,/A, = 10'. .

ects (18,19), the concepts presented generally illustrate the effect ilE From Fig. 9, t,/k, = 1.001 or 1. - 1.0 x 10.s cm/s. Therefore, if only relatively small area of defects can have on the overall pond hydre g,gy;,y, 1% of the area of the pond has defects with hydrauhc conductivities of l y ,

Deeslution hkWh & ses m i% M hm cracking is the major contributor to the high seepage losses from a l ponds discussed. To compare the effect of dessication cracking on

'

+

equivalent pond hydraulic conductivity with that of the general de-Ia * g ) - - - - -

o

.

i model previously presented. another 5?epage model for dessicat cracking is shown in Fig.10. From this model the following equati

,

/ /

[

_

[  ! are apparent:

j j' omcr

,

J

'

lLa / ,I a - n. ' ... . .

-

cto t-

$ '

,

,

no. s.-s peo. Modei *< oven cter t.iaer esects Q, = k.1 A, . ..... ..

,

M60 46:

t  !

- _ - _ _ _ - - - - _ - - - - - - - - - - - - - - - - - - --

~] F

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - - - . --- -

- - - - - - - - - - - - - '

. .

.  ;

l

__ __

The following conclusions may be made sespect to 11,

. -

phenomenon: 8

_

4. F- --dae,l-- 1. 'A very smill res tzge of defects of high conductivat

,

/ ./

{ ------rr ..... d , through the com e thickness of the clay hnct can cau g

e

!

1 I have an equivalent hydr'aulic conductivity several orders a#

I gg m _[ more than the intact clay liner conductmt p 1 2. A relatively large occurrence (10-40% of the pond ar(

$5,yT80N d I

~

cation cracking whicitdoes not entend completely through t[

HQ.10.-Seepage asodel W Casy I.iner e r" will cause a pond to have an equivalent hydrauhc conductig

- wu

-

,

one to three times greater than the intact clay liner conducts fore, based on the dessication crack model, it appears thac Using Eqs. 3-6 defined previously, the following expression for the n8 main c nNot t ee @ seepage bues exps equivalent hydraulic conductivity can be generated: Pondhued, udess the aacks exteged canpletely de ( ,, gg _ ,)

,821,4"'""""""""-" .(11)

' 3. Two of the case studies presented by the author stron that dessication cracking was not the cause of the high rates For the two ponds in central Texas, hydrauhe conductivitie Fig.11 shows the effects of the greater than would have been expected occurred even alto ntity of dessication cracks and their

.thschness and recompacting the liner, and then pumping water into 4 an the figure,on the com ted equiv nt hydraulicconductivity. As shown if the ssication cracks extend 6..-.. one-half to five-j almost immediately after construction, so that no cracking d

<

samths the way through the liner thicknees, the ratio of the equivalent '

In addition, the eastern brine pond in southern Texas contic

.

Pond hydraulic conductivity to that of the httact coenpacted clay would excessively, even though measures were taken to reconsert

'

range between 2-6 if 100% of the area contained dessication ancks. Poe and keep it from undergoing any dessicatio ,

r announts of cracking (10-40%), this ratio would be in the range Cosestsoucnoes Quauty tr ,

The conclusions presepted form the basis for the writer's

, ,

,

the ** defects" in the clayliners which cause hig are not dessication cracks, fissures, or stickensides, but rate a, s y

• * T-

. 3 high permeability caused by inadequate control of the cons (

io cess.1hese zones are erost likely one of the following:

1. Areas of low compactive effort and low moisture cont

• -

.

2. Areas where drastically different matenals (e.g., sands,

-

I topsoil, etc.) have been allowed to pass as " clay."

3. Areas where the joining of the areas of the hner hat

s -

" defective" seam, such is where adjacent hits are brought te different areas (e.g., where the pond bottom and sides conc

1. '

J4 _

,,

Which one (or ones) of the precedmg areas is the pnmary cs i high conductivities is npt known at this time. Itowever. ttV

. point is that the high conductivities expenenced by the pond i

been avoided by closely controlhng all aspects of constr

, _

..oio

-

thickness, compactive effort, compaction water content, soi

_ ---- ..o na ' acterization, and lift placement control). As pointed out prev 1% of the pond liner area need exhibit higher conductivities -

,

o i j .  ; ' equivalent pond conductivity to be several orders of magmt than would be expect t/t

,, , This strongly suggests the need for degree of construction uality control to achieve the desired ductivity that has historically not been provided on sua h pr nG.11.-Effect draulic Conduestvityof Dessication of a Uned PondCrack Quantity and Thickness on Equivalent H The question posed by the author's presentatmn then is i laboratory permeability ~ tests can predict field performanse. '

i 1462 I 1463

'

+

'

• .

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - _

.

., *\ NUREG-1200 U.S. Nuclear Regulatory Commission

% . 7.'. . Office of Nuclear Material Safety and Safeguards LOW-LEVEL WASTE DISPOSAL LICENSING PROGRAM STANDARD REVIEW PLAN 3.2 - APPENDIX A GUIDANCE ON SOIL COVER SYSTEMS PLACED OVER LOW-LEVEL'RADI0 ACTIVE WASTE INTRODUCTION

!

Several studies (e.g., NUREG/CR-4701) have analyzed tne individual components of a designed low-level radioactive waste disposal facility and have concluded that the cover component is one of the most important engineered barriers. All scientific and engineering disciplines involved with the safe dis agree that careful consideration needs to be given to the design,posal of waste construction and maintenance of a waste cover syste Agreement with the U.S. Army Corps of Engineers (COE) in March quested the COE to provide recommendations to the NRC for the selection, place-ment, compaction, testing, and acceptance of soils proposed to be placed in cover systems over low-level radioactive wastes. In addition, because several areas of con Woversy and technical differences were kr en to exist when assessing the pcceptability of soil cover materials, provisions were made in the A with the COE to provide for peer review of the COE recommendations. greement of the COE recommendations and the Seer review efforts are summarizedThe results in the three volumes of NUREG/CR-5432 whic1 were published in February 199 NRC staff reviewing the integrity of dis >osal unit covers in a license application should be thoroughly familiar w'th t)e guidance presented in these three volume The review guidance provided in the discussions and recommendations of Appendix A is not intended to be a complete summary of NUREG/CR-5432, but rather the high-lighting of selected recommendations that the staff believes is desirable because of the uniqueness and importance of certain review issues that necessitate a conservative approach, to ensure the safe long-tern performance of the soil cover systems over-low-level radioactive wastes. The omission of any COE recommendation from the discussions in this Appendix A should not be taken as a position that the recommendation is less important, but rather that the recommendation covers a technical ites that has general acceptance on what sound ,

engineering practio would typically require. In this appendix guidance is provided were a specific review item is considered to ha especIally important-to satisfying the regulatory technical requirements and the long-term stability objectiveof10CFR61.4 include: The waste cover functions required by the regulations Minimizing infiltration through the cover from precipitation and surface runoff or runo .2-11 ,

Rev. 3-September 1991

, , _, _

.. .

.

.

.. .. . . . .

._ _ . _ _ - - -- Minimizing the contact of water eith tastes, through removal of water as *

runoff before it infiltrates, through drainage layers after it infiltrates (percolation), and through the use of low permeability barriers around the waste . Minimizing surfaca erosio . Minimizing differential settlement and subsidence of the cover, and more importantly, damage to the cover as a result of differential settlement and subsidence of the wastes or highly compressible foundation soil . Limiting the radioactivity dose rate at the ground surface of the cover to acceptable level . Providing resistance to damage to the cover as a result of burrowing animals or root penetration (biointrusion). Providing resistance to damage to the cover as a result of freezing and thawin . Providing long-term stability over the covered wastes without the need for active maintenanc It is recognized that geosynthetics (e.g., geomembranes, geotextiles) may effectively be used to supplement and improve the performance of soil cover ,

The use of these manufactured materials require special considerations that are beyond the scope of the soil cover discussions in this Appendix.

Guidance on specialized technical topics that also directly affect the safe performance of waste cover systems is separatel These topics would include erosion resistance (y provided SRPs 3.4.4, 5.1.1, in other and SRP sections 6.3.1);

minimizing voids in and around waste containers (SRP 4.3); and the settlement and subsidence aspects of cover design (SRP 6.3.3). DISCUSSION The following review guidance is provided to address the developmental phases that would logically be completed to design and construct a soil cover system. These phases are closely interrelated, and typically a decision or selection in one phase can have a direct effect on important considerations in the other phases. The development phases to be discussed include (1) cover design; (2) cover material selection; (3) laboratory and field testing; (4) field placement control and acceptance; and (5) penetrations through constructed covers.

2.1 Cover Design To satisfy the waste cover functions required by 10 CFR Part 61 regulations, the designer's initial preference would likely be to use a low permeability soil layer, such as an inorganic clay, that could be compacted to achieve a 3.2-12 Rev. 3-September 1991

,

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ._. _ _ . _ _ _ _ _ .. _ .._ _ _

SRP Appendix A

low saterated hydraulic conductivit listed would be satisfied with the selection of the low permeability c because of biointrusion and freeze / thaw considerations.except for It The need for resis+ance l to cover damage is crucial because of the long period of time over which tue covers would be expected to perfor The COE in NUREG/CR-5432 when evaluating the various functions of a cover advanta ultimately recomm, ended that a multi-layer cover be used to take system, erosionge of both the low permeability soil and the good drainage and resistant characteristics of a coarse grained soil. The intent of the multi-layer approach is to use the best material in layers that complement and improve the performance of the adjacent layers within the cover, as well as the entire cover itsel a multi-layer cover Le that differential settlements would be minimi requiring (1) a firm and stable foundation beneath the wastes to be placed, (2) stable waste forms, and (3) minimization of void spaces in and around

,

waste containers. Conversely, a multi-layer cover should not be installed at

!

a waste disposal facility if differential settlements were not to be minima As an example, placing a multi-layer cover over unstable Class A waste should

!

be avoided until such time that actual settlements and subsidence transpire would h!

This time frame would reasonably be expected to be very long-much longer than conventional estimates of soil settlements, because of the slow deterioration and decomposition processes that would be anticipated for the unstable wastes.

i A conceptual design for a multi-layer waste cover system is provided in l NUREG/CR-543 !

The staff considers the conceptual multid ayer cover that '

is presented to have essential components that take best advantage of available soil material properties to fulfill the required waste cover functions. For this reason the staff, in a license application review of

a cover design, would compar,e the actual design submitted in a SAR with the recommended conceptual design in NUREG/CR-5432 to assist in the staff's eval-uation and acceptance determination of the proposed cover design. Alternatives to the recommended designed conceptual design would be found acceptable if properly and documente '

Some of the important and deliberate considerations that the COE and peer review panel weighed in the development of the conceptual design may not be readil multi y apparent. For example, the low permeability soil layer component of the layer cover that provides the greatest contribution toward minimization of infiltration and avoidance of water contacting the waste needs to be located at a depth below the maximum frost penetration at the site of the proposed facilit If the low permeability soil were placed within the frost penetration zone, eventual cycles of freezing and thawing of the low permeability soil could be expected to cause large increases in the hydraulic conductivity of this layer, and thereby significantly impair the cover's intended performanc The thicknesses of the individual soil layers comprising the conceptual

. multi-layer waste cover system were carefully chosen. For example, although lift thicknesses for the various filter, bedding, and drainage materials could

he thinner, and this would still permit these layers to perform their intended

functions, the COE's and the peer review panel's experience with actual

construction practice in the placement of thin layers suggested that the recommended minimum values shown on the conceptual design be used. The field i

3.2-13 Rev. 3-September 1991

, - . _ . - .

_ _ _ _

SRP Appendix A ,

experience recognizes the limitations of the construction equipment and the .

imprecise operations in the placement of the soil layers in the field. This experience is the basis for recommending the minimum thicknesses, to compensate for the-field placement limitations, and to have reasonable assurance that minimum coverage of each material will be availabl An important recommendation in NUREG/CR-5432 is to use a minimum thickness of 3 feet for the low permeability _ soil layer, to minimize the effect of defects through the full cover resulting from construction practice, and to increase the potential length of-seepage paths (Elsbury, B.R. et al, 1990). The staff reviewer will find additional guidance and informative discussions in NUREG/CR-5432 related to cover design consideration These discussions cover such topics as recommendations on (1) the hydraulic conductivity contrast needed between the low permeability soil layer and drainage layers within the cover system and (2)-how the placement and compaction of soils on slopes require special consideratio .2 Cover Material Selection A very important step, after the decision to use a multi-layer cover, is to select acceptable materials to fulfill the intended functions uf the various component layers of the cover system. The obtainability or scarcity of cover materials that are locally available to a proposed site are crucial consider-ations that significantly influence the decisions on the cover design. It is important that the selection process, when evaluating costs and the locally available material characteristics, not result in the selection of materials that are marginal or questionable in fulfilling the required cover function '

As an example, a preferred or an acceptable low permeability clayey soil with ,

resistance to damage from cracking, heaving, or erosion may not be locally available; however, there may be a position taken in a license application to substitute a material that is locally available, but of which the long-ters ,

'

cover performance is questionabl This could be a difficult situation, the '

resolution of which warrants a regulatory position that requires the borrowing of an acceptable material from an offsite distant sourc ortant and

The crucialCOE and the peerrelated considerations reviewtopanel coverhave extensively material selection addressed in NUREG the imp /CR-5432.

'

In Volume 1, Tables 1-1 and 1-2, the range of fine grained soils and coarse-grained soils are ranked according to their ability to fulfill the required cover functions. A summary of the rankings indicating the overall best-to-worst choice for the two material types is also provided. In Volume 3 of NUREG/CR-5432, a good summary of the desirable characteristics for both the low permeability soils and the filter and drainage soils is provided and this summary has been included in this appendix (Tables A-1 and A-2). The characteristics that are summarized in the tables include soil classification, plasticity, gradation, hydraulic conductivity, allowable percentage of organic material, and shear strengt NUREG/CR-5432 also provides helpful guidance to staff reviewers on other .

material selection concerns including (1) the use of dispersive clay soils in cover layers, and (2) the addition of modifiers to marginal or undesirable soils to improve engineering properties important to long-term performanc .2-14 ,

Rev. 3-September 1991

-. ._ _ _ _ _ _ - - _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _

_ _ _ . _ _ _ _ _ _ _ . _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ - _ _

'

SRP Appendix A

'

2.3 Laboratory and Field Testing

,

l Although separately discussed in this appendix, laboratory and field testin

! considerations can not reasonably be separated from the important concerns,g

expressed previously regarding cover design and materials selection.

! testing is typically needed to evaluate the capability of soils proposed forLaboratory fulfilling the required waste cover functions. The major portion of field i tasting is typically needed to verify that desired engineering properties are l actually being obtained as a result of the construction operations, and that ,

controls are being carried out in the field.

!

The COE and the peer review panel in Volume 3 of NUREG/CR-5432 provide very t

~

useful tables (Tables 2-3 through 2-5) that list laboratory tests that should be considered in determining index properties, classification, and engineering propertie In addition, because of its importance in cover design and performance, expanded discussions are provided on the laboratory testing for hydraulic conductivity (Table 2-6). Tables 2-8 and 2-9 in Volume 3 of

.

'UREG/CR-5432 identify field tests for evaluating test fills and controlling construction of soil cover advantages and disadvantages of the various in situ hydraulic test condu Reviewers of a license application should find the tables that list

.

t the various laboratory and field tests to be very helpful; however, a word of caution in applying the complete test listing in a licensing review is offere When a designer is faced with the many decisions on what soil samples need to be tested and for what engineering purposes, there is a wide range of answers, and the answer is very much influenced by the experience and philosophical approach of the individual designers. There is no exact number of tests to be performe There are overall goals and accepted practice which would warrant at least a minimum level of testing. A regulatory reviewer, therefore would be expected to carefully review the scope and adequacy of a testing pro, gram completed by a license applicant in order for the staff reviewer to develop reasonable confidence that the testing program completed is acceptable. Important review questions to be considered in this evaluation should include: (1) Were sufficient and representative samples taken and tested (2) Were the proper tests to establish needed engineering properties run; a;nd (3) Was the testin that was completed properly conducted and do the results appear reasonable? g Favorable answers to these questions are netded if the staff reviewer is to be able to make an acceptable evaluation finding in a SER on the adequacy of a completed laboratory and field testing program.

t

!

!

l

'

3.2-15 Rev. 3-September 1991

._ , . -- , --

, - -.. - . - - - - - .

'

-SRP Appendix A -

Table A-1 Desirable Characteristics of. Low *

Permeability soils for Waste Covers .

Characteristic Ratings Comments Preferred Acceptable Undesirable ,

USCS Soil Classi- CL CH, SC, MH, ML, SM Local-availability '

fication . C L-ML impacts choice Plasticity Index~ 15 to 25 7 to 40 <7 PI <7 or LL <20 may (PI) >40 result in difficulty in '

meeting hydraulic con-ductivity requirement PI >40 or LL >70 may

,

result in workability problems, i.e., hard-when dry, sticky when i wet, and difficult to  ;

adjust moisture conten Liquid Limit (LL) 30 to 50 20 to 70 <20

>70 Coarse Fraction:

+1-in. size

.

None 53% by w >3% by w Larger percentages of coarse fraction may '

result in difficulty in meeting hydraulic con-

+1/4-in. size 55% by w to 10% >10% by w ductivity criteria and

'

by w may lead to damage of '

geomembranes if use Maximumparticlesize must be much less than lif. thicknes Fine Fractio to 65% 15 to 100% <15% Fine fraction <15% may  ;

.(% finer than result in difficulty in No. 200 sieve meeting criteria for size) hydraulic conductivit Fine fraction >85% may -

>

result in workability problem Hydraulic Con- Dependentonproject- >1x10-7 Higher values of '

' ductivity specific conditions en/sec hydraulic conductivity (under expecte could result in difficulty icng. term field in satisfying long-term

conditions) performance requirement ,

.

3.2-16 . Rev. 3-September 1991

.

_ ,_ ______ '- -- " '-~ ~~ ~ ~

'

SRP Appendix A

!

Table A-1 (Continued)

l Characteristic Ratings Preferred Comments Acceptable Undesirable Organic Material None <1% by w !

>1% by w Organic material i increases hydraulic conductivity, compres- ,

sibility and decreases long-term stability and shear strength Shear Strengths Dependentonprojectspecific conditions Minimum strength criteria must be based oa site-specific considerations for stable slopes ade-quate bearing capa, city, limiting settlements and crackin .

l 3.2-17 . Rev. 3-September 1991

.

. . - - . - - - .._ - - . . - - -. .- . - . . . . - - . -

Appendix A

,

Table A-2 Desirable Characteristics of Filter and Drainage Soils for t!aste Covers '

Characteristic Ratings Comments '

Preferred . Acceptable Undesirable For Drainage:

USCS Soil Cobbles, SP, SW Classification GM, GC, SM, Local climate, avail-GW, GP SC ability and location of layer within cover cross-section impact choice For Filters: Apply accepted criteria For example, cobbles for selection, based on provide excellent drain-characteristics of soils age, but are not satis-to be protected and draine factory as filter Hydraulic Hydraulic conductivity

>l cm/sec

- >1x10-2 <1x10-3 most important facto Conductivity cm/sec cm/sec Hydraulic conductivity

'

value of drain should be at least 10,000 times higher than hydraulic conductivity value of soil to be drained, and high enough to quickly drain estimated infil-trating water w/large safety factor. Thickness is an important consider-ation for selecting mini-num hydraulic conductivity of draira.

CCarse Fraction Physical and chemical stability more important than actual percenta of coarse fraction. ges Fine Fraction <5% <8% >12% Permeability greatly (% finer than reduced by clay, silt, No. 200 sieve and even fine sand size size)

i l

3.2-18 Rev. 3-September 1991

. --

.__ _ _ - _ . _ _ _

. _ . _ . _ _ _... _ __ _ _ . _ .._ __. - _ _ y

.

SRP ,

Appendix A i $

j 2. 4 Field Placement Control and Acceptance

An essential review step directed at ensuring that the designad waste cover

system will fulfill the required functions is a check of the actual license application to verify that acceptable procedures have been established to i control field placement of the cover materials. This verification effort

! should enable the staff to develop reasonable assurance that the cover will be the SA constructed as designed, and fulfill the important commitments given in

!

' It is important that an applicant provide sufficient information in an SAR to j

,

show that sources for each of the proposed cover materials are available and

'

contain sufficient volumes. The information may consist of detailed borrow area excavation plans and sections that will support the volume estimates of i the needed materials. During construction, some testing of the borrow source

j should be periodically performed to verify that materials changes are not i

occurring that would adversely impact important material properties.

i Testing for field quality control during actual material placement needs

to be performed on the low permeability soils and the filter and drainage materials. The COE's and the peer review panel's recommendations for the types I

of tests and their frequency in construction are provided on Tables 3-5 and i 3-6 in Volume 3 of NUREG/CR-5432. The testing listed on these tables include

!

i (1)visualchecksonliftthicknessandbondingoflayers,(2)observationson j completeness (4)grainsize,(5)Atterberglimits,and(6 and number of compactor passes,)(3)

hydraulic conductivity. moisture-density re j

and 3-9 of NUREG/CR-5432, provide important and useful COE recommendations Tables 3k8

' covering placement and compaction specifications, for both the low permeability I and the drainage soils. General recommendations are also offered for handling i

deficiencies in meeting established specification as revealed in quality control test results and inspections. An important recommendation that both the COE i

!

and the peer review panel continuously stressed was the need to require a guality assurance program that would be implemented and executed by trained, experienced i professionals, including field inspectors and their supervisor Important i

considerations to be addressed in a quality assurance program include: (1) a clearidentificationofresponsibilitiesoftheinvolvedparties;(2)the  ;

- establishment of the actual parameters to be tested, and under which test methods,  !

and at what frequency the arameters are to be tested; (3) the defining of i allowabletolerancesfort$emeasuredparameters;and(4)explicitinstructions i for taking corrective actions whenever deficiencies are measured or observed.

-

Construction inspection personnel need to be properly trained and familiar with  !

' the important aspects inherent in design criteria field performance require-ments,andspecialdetailspertinenttocoverdesIgnandconstructio '

t i

Helpful guidance is also provided in Volume 3 of NUREG/CR-5432 on special

problems associated with conditioning and processing of the cover, material i

soils, during field placement. The guidance includes recommendations for adjusting the field moisture of soils placed for compaction that are either i- initially too wet or dry of optimum moisture content, the breaking up of clods in low permeability soils, and the removal of oversize or objectionable particles from the fill materials after the fill is placed in lift i i

t i

.

3.2-19 Rev. 3-September 1991

. . - . - , , . - - , .-- , - - - -

o- , y _ __ _-.

- ~ 7 F y.74 Appendix A ,

-2.5 -Penetrations Through Constructed Covere s

Every effort should be made, during the design and construction stages of soil cover development, to produce a cover system that will be effective in ful-filling the functions required by the regulations. Yet, in spite of these important and concerted efforts the best planned and constructed cover could berenderedineffectiveinfulfillingtherequiredfunctions,becauseof man-made penetrations that compromise the ccvers integrity. The penetrations may have been necessar maintaining the cover'y to accommodate consiocrations that compete with s intagrity, such as the necessary installation of environmental monitoring device Every possible effort should be made to avoid making penetrations through the soil cover that could potentially lead to the development of preferential flow paths along, the alignment of the penetrations. If the importance of competing regulatory provisions requires cover penetrations to be made, the penetrations need to be carefully located, constructed and sealed, to maintain the covers integrity, and to prevent formation of preferential flowpath In Appendix A to Volume 3 of NUREG/CR-5432, the COE provides recommended methods for sealing penetrations in waste cover systems. Staff reviewers of cover design are encouraged to become familiar with the recommendations that includecontrol quality discussions testing,on (2)

construction of seals for (1) holes needed to accommodate monitoring field performance, placement of geotechnical instruments for and(3)monitoringwell . RECOMENDATIONS Based on the information and discussions provided in NUPEG/CR-5432, the following procedures and positions are recommended for use by the staff when reviewing a license application related to the design, construction and maintenance of a waste cover syste Deviations from the recommendations may be acceptable, uphold conclus ions that the required cover functions will be metbut wou

.

3.1 Cover Design .

n To take adva'tage 'of the definite benefits provided by a multi-layer cover, it is recommended that a multi-layer cover system with components similar to the conceptual ~ design shown in Figures 1-la and 1-lb of NUREG/CR-5432, Volume 1, be used in the design of a low-level radioactive vaste disposal facility. This recommeedation presup>oses that differential settlements would be minimized by properly addressing tie pertinent design consideration Minimusthicknesses(beforecompaction)forthelowpermeabilittsoillayer and filter / drainage layers should be 3 feet and 1 foot, respectively. The low permeability soil layer component needs to be placed at a depth where its entire thickness would be below the depth of maximum frost, for a specific site locatio .2 Cover Material Selection Soils that are proposed in a Safety Analysis Report to fulfill the required waste cover functions should meet the preferred or acceptable material ratings 3.2-20 . Rev. 3-September 1991 ___ ___ -- -

SRP Appendix A

..

'

for each of the soil characteristics listed in Tables A-1 and A-2. Soils that are not listed as either preferred or acceptable low permeability or filter and drainage soils should not be used in a cover design, unless an applicant proposes modifications or improvements acceptable to the staff for the un-desirable material, with sufficient documentation to support a regulatory evaluation conclusion of the proposed that the required waste functions will be met because modificatio Selected soils that are improved or amended with bentonite, to reduce hydraulic conductivity, would be acceptable if (1) properly designed, (2) .sufficiently controlled during construction placement operations, and (3) adequately tested to demonstrate achievement of engineering properties required to meet waste cover functions. The process of blending bentonite with certain soils to reduce hydraulic conductivity, has been proven to provide satisfactory performanc Because questions remain about the long-ters durability and performance these other amendedof soils soils amended may be questionablewith other additives and unreliabitumen means fnr satisfying long-term cover functions. Dispersive clays that are or lime) (e.g., ble a highly erosive, but otherwise may be acceptable, should not be used in cover designs where they may be directly exposed to erosive force: .3. Laboratory and Field Testina j

The use of Tables 2-3 through 2-5 and Tables 2-8 and 2-9 that are presented '

in Volume 3 of NUREG/CR-5432 is recommended as guidance for a staff reviewer, when evaluating the scope and adequacy of laboratory and field testing programs, for the low permeability and drainage soils to be placed in a multi-layer cover design. Reasonable caution when deciding on the required I number and types of tests listed on these tables needs to be followed, keeping '

in mind the overall objective of developing reasonable assurance on the -

acceptability of the tasting program completed by a license applican l Both laboratory and field testing to establish hydraulic conductivity are acceptable practice, provided they are conducted on representative samples and recognize the limitations inherent in the selected testing equipment and procedures. It is recommended that flexible wall permeameters be used in the laboratory testing of undisturbed soil samples and that rigid wall permeameters be used for remolded, compacted soil specimen To best duplicate the condition and structure of field compacted soils and to have the capability of testing much larger areas and volumes of soil than can be tested in the laboratory, field tests for hydraulic conductivity of low- )

t permeability soils should be performed on test fills (not the final cover) l that are constructed using the same materials and methods as would be required  !

by the full-size cover specifications. The recommended field tests to  !

establish hydraulic conductivity are the pan lysimeter or the sealed, double-  !

ring infiltrometer. At least one field test for each soil type proposed for the low permeability layer should be ru This is considered a minimum number that is essential for verifying important design good quality control check on field performance. parameters and to provide a  !

!

3.2-21 Rev. 3-September 1991

~ - . - - . - - - . .

SRP Appendix A . ,

i 3.4 Field Placement Control and Acceptance d An SAR for a low-level waste facility should provi6 information on the quality control testing to be performed for the van'ous materials that are proposed-for placement in the waste cover system. ~Ihe staff reviewer should i use Tables 3-5, 3-6, 3-8, and 3-9 in NUREG/CR-5432 to assist in the comparison i of the information provided in the SAR and to help in the staff's evaluation i conclusions on the adequacy and acceptability of the applicant's quality control testing program. The applicant's quality control program should address the l

!

qualifications of construction personnel who will implement and execute the '

program'and the information in the SAR should demonstrate that the personnel will be trained and experienced and have familiarity with important design  !

criteria and with the field perfoi s nce requirement Sufficient information (e.g., detailed plans and cross-sections) on the plans for borrow excavations needs to be provided in an SAR to se , port volume estimates on the availability of materials that are proposed and shown to have important engineering propertie . 5 Penetrations through Constructed Covers Whenever possible,' man-made penetrations through a properly constructed soil cover system should be avoided. This goal implies the serious consideration of locating and installing required environmental monitoring devices away from  !

the immediate area of the waste emplacement, wherever possibl Shallow, partially penetrating holes in the waste cover that are required for l quality control testing (e.g., field density tests) should be located to avoid having these penetrations directly over one another, as the soil cover is raised during construction. The holes upon com)letion of testing should be

!

filled with the same type of soil remov,ed from tie hole and recompa,cted in thin ,

lifts, with suitable equipment, to produce the required density and moisture i equivalent to the surrounding fil !

When environmental monitoring instruments and cables must be installed to  !

penetrate the cover, the be carefully considered. guidance provided The planning in NUREG/CR-5432, of monitoring Volume installations should 3 should anticipate potential differential settlements and deformation movements of the installations, and provide provisions that will safely accommodate conservative estimates of the movements. Installations cables and casings should be placed at slight declines away from the emplaced waste, where possible, to prevent development'of preferential flow paths. Any penetrations of geomembranes that are necessary should be made during the initial construction operations, and not drilled at a later date in the completed cove Because of the importance of making ' positive seals av .d penetrations through the cover, the guidance in NUREG/CR-5432 covering sea:Ing operations and methods should be carefully considered in a license application. In reviewing i an SAR, the staff reviewer would need to verify that the license application l has given careful consideration to avoiding, penetrations in the cover, and where the penetrations are unavoidable, verify that the appli m t has carefully planned and designed for the penetrations with the least pots xial of disrupting the cover's performanc .2-22 . Rev. 3-September 1991

- . - - .

,

SRP Appendix A

' REFERENCES Elsbury, B.R, Daniel, D.E, Straders, G.A, and Anderson, D.C., " Lessons Learned No. 11, New York, N.Y., November 1990.from Compacted Clay Liner," J Idaho National Engineering Laboratory, NUREG/CR-4701, " Safety Assessment of Alternatives to Shallow Land Burial of Low-Level Radio 1 U.S. Nuclear Regulatory Commission, NUREG/CR-5432, Vol.1, "Recommendatinns to the NRC for Soil Cover Systems Over Uranium Mill Tailings and Low-Level Radioactive Wastes:

Covers,"

February, . Bennett, U.S. Army Engineer Waterways Experim

-- , NUREG/CR-5432, Vol. 2 Over Uranium Mill Tailings,and Low-level Radioactive Wastes:" Recommej Field Tests for Soil Covers," R.D. Bennett and R.C. Horz, U.S Army EngineerLab Waterways Experiment Station, February 199 !

Over Uranium Mill Tailings and Low-Level Radioactive Construction Wastes:-- ,

Methods and Guidance for Sealing Penetrations in Soil Covers," R.D. Bennett 1 and A.F. Kimbrell, U.S. Army Waterways Experiment Station, February,199 l

!

i i

I e

i i

!

3.2-23 Rev. 3-September 1991 i