ML20202C160

Forwards NRC Comments on Preliminary Design for Falls City, Texas Umtrap Site.Nrc Offers No Specific Comments on Groundwater from Review Because Documents Do Not Address Groundwater Protection Aspects of Proposed Remedial Action

ML20202C160
Person / Time
Issue date:	11/30/1988
From:	Lohaus P NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS)
To:	Arthur J ENERGY, DEPT. OF
References
REF-WM-65, TASK-TF, TASK-URFO NUDOCS 9712030273
Download: ML20202C160 (31)

Text

(t)n)- 6 [

x l

4Wi%

Y

/;

/S?-

D[0 F3 d

1. 8 0 *8[

i

!ggyh b

1. ' kg GGo6 V

'f,go, W. John Arthur, III, Project Manager '

s J

Uranium Mill Tailings Project Office Department of Energy N

Albuquerque Operations Office 3

m P. O. Box 5400 Albuquerque, NM 87115

Dear Mr. Arthur:

~

O-Enciosed are NRC staff comments oo the preiiminary oesign for the raiis City, Texas UMTRAP site. These comen's are in the technical areas of surface water hydrology and gootechnical engineering.

The surface water hydrology coment was provided on October 20, 1988, by Ted Johnson of NRC staff in a telephone conversation with Frank Bosiljevac (DOE) and MKE representatives.

The NRC staff has no specific comments on ground water from review of the preliminary Design because these documents do not address the groundwater protection aspects of the proposed remedial action. Also, the staff has no coments with regard to geology, other than those previously made based on review of the draft Remedial Action Plan (dRAP) and draft Environmental Assessment (DEA). Though we have received additional information since review of the dRAP and DEA, to date DOE has not provided written responses to the NRC staff comments, dated July 15, 1988, on these documents.

O" Finally, DOE staff has informed NRC staff that a significant change in the proposed design for the Falls City site is being considered; a change from rock to soil cover. This design change is being considered late in the design schedule and, if adopted, would require reevaluation by NRC staff.

9712030273 081130 PDR WASTE WPt-65 PDR 0FFICIAL DOCKEi Con f cf-ony

7 f >.

't

'I l

.2 1

If you have any-questions regarding this review, please contact Susan Bilhorn at FTS 492 0573.

.i Sincerely, I

l 02 4-u: ht. Sue w t p 9 y Paul H. Lohaus, Chief Operations Branch

-O Division of to -tevei Waste Maansemeat and Decommissioning, NMSS

Enclosures:

As stated cc:

F. Bosijivac, DOE /A1 D. Mann, DOE /AL E. Bailey, Texas Department of Health O

Distribution:

Central file HMSS rf LLOB rf.

SBilhorn, LLOB MFleigel, LLOB

' PLohaus, LLOB DGillen, LLOB JStarmer, LLTB MTokar. LLTB JSurmeier, LLTB DWidmayer, LLTB JGrinn, LLTB W eber, LLTB TedJohnson, LLTB JGreeves, LLWM MKnapp, LLWM TURF t d 2 PLEASE SEE PREVIOUS CONCURRENCE

.,a UFC-:LLOB

LLOB

. LLO NAME:SBilhorn/jj:MFleigel :PLohaul' :

' 88 :l} /f0/88 :

DATE: '/_/88

-:- /

/

-OFFICIAL RECORD COPY-

9 e_

NRC STAFF COMMENTS ON THE PRELIMINARY: DESIGN-FOR THE-UMTRAP-SITE AT

-FALLS' CITY, TEXAS-

, Surface Water Hydrology;

1. : Design: of Eros _loh Protection for Northeast Apron.: Calculation'f:20-440-02-00

- NRCLstaff! review of-the proposed design of the northeast' apron indicates that-

.the-apron may not be appropriately designed to resist-potential erosional

. forces that could occur during the design life of:the structure.

It appears' thet several inappropriate assumptions may have been made and that insufficient

, designs have been provided to prevent erosion.

First, we notelfrom our review-of the calculatior.s that the design of the rock Ltoe:to resist erosive velocities in.the northeast-drainage ditch is based on W cthe occurrence of a-65-foot-wide ditch which may form as a result of erosion of V tthe triangular ditch initially = constructed. - An evaluation of Drawing-

1. FCT-PS-10-0412; indicates that the distance from the: rock apron to the

' ditch centerline is only about 20-25 feet; therefore, Lit appears that the ditch-could be considerably smaller than 65 feet wide after erosion occurs.

Regardless of the distance from the ditch centerline, it also ap> ears that a very different erosional configuration could develop.

One_ possi>ility is that the ditch centerline could migrate: closer to the toe.of-the apron as a result-i of erosion.over a long period of time.

Such erosion could result in.a more critical channel section'and higher velocities formed closer to the apron toe.- The design of the rock for the apron is based on-the velocity in the 165-foot ditch; however, if a smaller channel section is assumed after erosion

'of the ditch, it appears that the rock size required to resist erosion of the too could-need to.be considerably larger. The calculations should be revised,

- using the velocity that may be produced in a smaller ditch adjacent to the apron.1 Regardless of the design-selected, all assumptions regarding the ditch'

[ :o sire andilocation should be fully justified.

Second, it appears that a significant amount of energy dissipation could occur in the vicinity of the culvert. This could happen particularly during a

major floodiand/or if;the culvert becomes clogged, causing a backwater effect upstream of the roadway. Since +he slope of:the ditch is approximately-2%,

causing supercritical flow in i smaller channel section, it appears that significant energy-dissipation could occur-in a hydraulic. jump which would.be

- formed in thisiarea where the flow regime abruptly becomes subcritical.

The

--design,should:be.checkepand revised, as-necessary, to account for this

-phenomena.

Third, the' depth-of1 placement of. the rock apron should be based on the expected

. depth of scour, assuming the. smaller channel section discussed above.

If a Jsmaller channel-is assumed to form,;the maximum depth of scour should be-Jdetermined and the rock:placed at least to that depth.. The design of the toe should betchecked to account for scour and-should be revised, as necessary.

3 0

Y

---

s--a

Geotechnical Engineering 1.: Radon Diffusion Coefficients The Preliminary Design for. Review indicates that additional diffusion coefficient testing of radon barrier material will be performed. Since the radon barrier it the unly-layer of material to be placed over tailings material in subarea 5 of the stabilized pile, these tests a m critical in ensuring that the current 2 foot thickness of radon barrier material is sufficient. The results of these tests should be factored into future design documents as soon as they are available.

2.

Liquefaction p

The Preliminary Design for Review concludes that_ additional analysis must be d

performed on the slime tailings at the Falls City site to determine if liquefaction will be a aroblem in the stabilized pile, should it be determined that liquefaction will 3e-a problem, dewatering may be considered one remedy.

This additional liquefaction analysis and any analysis of the effects of dewatering on stability of the tailings pile must be included in future design documents.

This area is identified as an open item in Section E of the Preliminary Design for Review.

s.

Radon Barrier Material Characterization Results of triaxial testing performed on samples of radon barrier material from only one location (Location ID 005) at the La Mesa source area are presented in the Preliminary Design for Review in figures D.4.33, D.4.34, and D.4.59 through D.4.63.

Additional characterization of this material for properties other than radon diffusion coefficients should be performed and O

the results utilizcd in future design documents.

4.

Values in Sumary Tables Tables D.4.6 and D.4.7 both contain values that are confusing.

In Table D.4.6, Samples 3A and 3B from Location 54 are both designated as slime tailings on. laboratory sample results included with the Preliminary Design for Review.- Since this Table is a sumary of in-situ foundation material properties, this confusion should be corrected.

In Table 0.4.7, the values for Location 32, Sample 04 and Location 43, Sample 2B are not consistent with laboratory results-included with the Design documents.

This confusion should also be~ corrected.

2

f,*.

5.

Swell Potential-l

-Calculations included with the dRAP stated that, until the final moisture content was determined, no conclusion regarding the likelihood of swelling of the radon barrier material wrhi be made.

There is no further analysis of the potential for swelling'in the Preliminary Design for Review.

Future design

-documents-should include the conclusions regarding the swell potential of the-radon barrier material.

lO-O:

1 -

3

+

e j

A.$1.L$q' N..

Q)/y)-$S' s

<?

i ggnitD e

\\

g voc

\\

q V

% nta G

PREDICTION OF STRAINS IN EARTECH COVERS

,7 g,

s W LtB c

4

} s' 1

\\b berg Xeshian, Jr.,Hember ASCE Ned B. Larson,

, Nember ASCE cg DMD ABSTRACT cleanup and isolation of hydraulically placed uranium mill tailings and other contaminated materials at 24 uranium mill tallings sites are to be accomplished by the U.G.

Department of Energy (DOE)

Uranium Mill Tailings Remedial Action (UMTRA)

Project through a

series of 4

(

remedial action activities that will culminate in the licensinq of a

remediated site for surveillance and naintent. ace by the U.S.

Nuclear Regulatory Comnita ! on (NRC).

Generally, two methods are used to dispose of the tailings and contaminated materials:

(1) excavation, transportation to another site, compaction, and placement of a long-term earthen cover systems and (2) placement of off-pile contaminated materials over the existing tailings pile and covering the entire pile with an earthen cover system.

The second method is more difficult since it involves the characterization and prediction of how the undisturbed, hydraulically placed tailings will behave when a lected to the changes in stress condition as a result reshaping and covering.

Thi paper discusses the Falls City, Texas, site and the methods used to characterize the tailings as well as the prediction techniques used to predict strair.

and possible cover cracking as a

result of differential settlement in order to predict their performance for final design.

1 Deputy Manager.of Engineering, UMTRA project Jacobs Engineering Group Inc., Albuquerque, New Mexico 2

Senior Project Manager, Roy F. Weston, Inc., Albuquerque, New Mexico L

-v INTRODUCTION he Falls City, Texas, tailings pile is approximately 40 miles south of San Antonio, Texas.

The uranium mill-was constructed in the early 1960s, began operations in 1961r and continued operating until the shutdown in 1973 5due to economic problems.

Figure i shows the vicinity map for the site.

Figure 2 shows the location of the neven existing tailings piles that will be consolidated into one pile at the site.

Tailings piles 1, 2,

and 7 were placed on the ground surface while piles 3, 4, 5, and pond 6 were placed in old open pits from which the ore was removed.

Due to groundwater considerations,

• t was determined and that the tailings should be removed from pits 3, 4, 5,inal 6 and placed on &nd around piles 1, 2,

and 7.

The f shape of the pile is shown on Figura 3 and a cross section of the final pile is shown on Figure 4.

As can be seen on O

Figure 4, the remolded tailings would then be placed over approximately 40 feet of undisturbed tailings in rome areas, and placed directly on the Dundation materials and adjacent to the tailings in others.

This raises concern regarding= the differential settlement and ultimately strain in the cover caused by the differences in stiffness of the hydraulically placed tailings and the engineered compacted tailings.

Significant quantitiva of data were necessary to i

adequately calculate the settlements across the final reclaimed pile.

The piesocone was chosen as the prime characterisation tool and was supplemented with borehole and laboratory data in order to evaluate the tailings in their in-situ state while still maintaining reasonable exploration and laboratory testing costs.

A grid of 200-foot centers was imid out and a piesocone sounding was O

9 rrar=ed at each grid point.

Twelve boreholes were placed within three feet of the respective piesocone sounding locations so that corre2ations between the piesocone soundings. and laboratory tests could be made.

Not only would conventional drilling have been approximately eight to 10 times more expensive than the l

piesocone, but it would not have produced the continuous profile that-the piesocone provided.

CHARACTERIEATION i

The results- -of the piesocone correlations are

- discussed in detail by Karson and Mitchell (1986) and are I-briefly summarised he re. -

The parameters that were correlated were the material classification, udrained shear strength, over consolidation ratio, blow count, and.

friction. angle.-

parameters that - were determined from literature and not specifically correlated were modulus of elasticity, relative density, and liquefaction susceptibi-

y 4

n 4

MG4I

[

em 4.

e....

v tsen.

g,g a

goo.

%4404e1 l

eweim I

= a e==*. u,,,

m..u...

}'

caeag,enDt6iat e

Mom wa.

• ~..,, m, ei,,

een et e eNeie. #

e

)

'g e.,m.,

8o o 5o aoc o..

seman,sswes-ammempi

, w,,.I.

SCAtt IN unts P6A,h6 n..

m.

....ii.

FALLS CITY ~

TAILINGS &lTE s

to o

to 30 seatcsi.wass s

4 FIGURE 1 LOCATION OF THE FALLS CITY TAILlh o.,lTE NEAR FALLS CITY, TEXAS

J M

w 4l f

scau = ner

e c to co=tou===rtavat

/

TANTes l SUMACE C

)hll, I

5.

l.

/

k.-

,N y ['i_

,b PARCEL B PARCEL A

~.[

I a.

w Q,.. s

. [j

, x 7.-

m/ N b

(tw e

=:-

g

\\,,,

Q tusnm wu cs ntes y

v en.tutaat statau

~7

~

g

= :=t-FIGURE 2 PRESENT CONDITION OF THE FALLS CITY TAILINGS SITE NEAR FALLS CITY, TEXA5

O O

b l

f a

g 1

-[h I

.A I

i

}

j+

2~

eso -

t 47o E

,73 s

7 u

j

/

/

I N

~

rn d

{

4,,

k

(,

Y m.:~ m ATN

• O O'Tc" I

/

- soog 500 [

g 1"

g,

(

~~

I:

Lages

~

lII O % %

l

.a i

505 j

- - ~\\

N

' o.

~.

se j

\\

\\.

,q',

x t

f

\\

1 i

l y/

'W',,

/

s...

/

,1

~ 4,o gg,

/

N

, % met g

lgg

/

\\N

! Mg%

ese lI d'

i

~ 4eo

//

'IW

~

k,

.so - j l

~ ' " /

( *** %

[

47,

I l

A

/

\\

+

~*

[=oca arnou)g*.~

'e W

~~ ~1

(.,,-

\\

j scate i= e u, i.

FIGURE 3 PROPOSED STABILIZED PILE CONFIGURATION FOR THE FALLS CITY TAILMGS SITE NEAR FALLS CITY, TEXA5 i

. -. _...= -

f 1

1 g

j

' il i I

il e

e i

f'i 1

a 6"

l 2

A

/

l

?

5 5

e a

l i

h#

l9 5

l 0

~

E !l 1

?

n I

o

[l I

i e

i e

O 5

8

/

l l

E

\\l i

i a

a s

8 1

we t

s t

)

4 CO

' w* E d s

g

\\

l b

h l

't f

i i

as h(

g -f' e

i i

e f

5! @5y G

.f.

y-(

i a

s 4

e gI W

I ; i<

jh en i

v, w I:.d i

~sa w

o I

O I i l?

• t i

5 5

s\\

i V

1 m

t

!g Is a

aal 1

8

g

(

l 5

e i

%l

\\

r a

8 l

8 i

e I

O a

a ii i

\\

$$5$

o i

i lity.

Each of the correlated parameters is discussed below.

t Material clammification i

The basis used for material classification was taken j

s from work done by Robertson (1985) and Douglas and Olsen (1981).

Although their work reduced the classification into several soil types, it was found that the tailings need only be divided into three different material typest sand, sand-slime, and slime.

The division for each of these types was based on the following percent passing the i

200 sieve divisions.

Sand 0 to 30 percent passing $200 sieve Sand-slime 30 to 70 percent passing (200 sieve slime 70 to 100 percent passing (200 sieve Figure 5 shows the results of the correlation between n

point stress and friction ratio when compared laboratory v

data for the three tailings types.

Blow counts The basis for the blow count correlations came from work done by Robertson and Campanella (1984).

The point stress to blow count ratio for the sands, sand-slines, and i

slines averaged 4.5, 3.0, and 1.5 for sai:h of the materials, respectively.

Undrained Shear strenath

]

The undrained shear strength was determined from the following equation:

O s =

-v Nx u

Su = undrained shear strength.

q,= point stress.

ey = total overburden stress.

Nk = cone factor.

factor average from 8 laboratory UU tests was The Nk 17 which correlates well with published data.

Friction Anale l

The friction angle was modified from Robertson and i

Campanella (1984).

The results of 10 unconsolidated undrained triaxial tests with pore pressure measurements i

were compared to the values-that were predicted using the i

....,a

,,..,,. -. - ~ +

... ~,. -. -. - -, -..

o 200 i

[A,[g SLIME-SAW iOO

~

~

R e

h 60 e

e e#y

~

s 5

a e-O 20 j.

,A o

z a

e A

4"'.,

04 10 I

be

~

8

~

a 4

6

=

A 3

1 1

i l

i f

0 1

2 3

4 5

6 7

PRICTION R ATIO (%I L t Ot tt0 8

SAND S LlWE e

SA ND SLOWE Wit TURE FIGURE 5 PIEZ0 CONE CLASSIFICATION CHART USED FOR TAILINGS

e.

method shown by Robertson and campanella and it was found that their method generally overpredicted the triaxial l

friction angle for the tailings.

To compensate for this overprediction, the friction angle predicted by the piezocone was multiplied by a factor of 0.9.

6 Over-consolidation Ratio The over-consolidation ratio (OCR) calculation was modified slightly from the method presented by Schmertman (1978).

This method uses the ratio of the undrained shear strength determined from the point stress (S )

and the u

and divides it total overburden stress at that point (oy)tio, ra which was by the normally consolidated S/o determined to be 0.33.

The OCR cou$d then be calculated u

frca the following equation:

OCR= ( S /'y ) / 0.33 )1'M u

O where Su = undrained shear strength e

= total overburden stress y

0.323 Normally consolidated S /'y ratio u

In general, this relationship slightly underpredicted the over-consolidation ratio; however, since there was not enough data to justify modifying it, the over-consolidation ratio was used as calculated since the rulationship would yield slightly conservative results.

ANALYSIS These parameters were calculated for all layers of O

the profile by using a specifically developed computer program (Larson and Mitchell, 1986) to analyze all data gathered in the field.

A typical profile of a piezocone sounding is shown in Figure 6.

The corputer was programed to classity every single data point gathered in the field

and, based on the resulting classification, a

corresponding color was placed in a bar code on the edge of the screen.

This method is simulated in Figure 6 except that the different material types were separated into different columns.

This simulation was necessary since the different colors for the materials could not be reproduced in the -- printing process.

By moving a

horizontal line up and down on the screen using a " mouse" peripheral, different layer horizons could be identified and the engineering parameters could be calculated and used in subsequent analyses.

Table 1 shows a typical set of results for a pietocone sounding.

(V g

N,../

PIE 3XOf fE 500f tDIrfG TEST i

6 LOCATI0tt ID: 004 l

SITE It's TtW t 1

FDITtT STFE M 'tG r.t$7e19 rOLT pn f3tst ee G Ctd ?

FFI<.7 ItT1 FHTIO ' %

• R FACTTV' t*7t1O o / ? ? ? ? P T F ? 'Po, 1, F

?, f, 70 ! ? ? ? ? F 7 ? ?'Po ! ? ? ? F F T ? P'P n

p l

I

'> W h

j s

c.

~

Y 4

~%

-}

b.

l' r

t 1

>~%

I s

E i

1s -

s i

g

- C'

~;w 5

=

- b,

_s e

- 20 _- q<

Q 5

m@

I E

- (-

g

• =-

=

a:

O

~

SMO

~

see-stm SLM 30 -

FIGURE 6 TYPICAL PIEZOCONE PP,0 FILE AND COMPUTER GRAPHICS USED TO AID IN THE IDENTIFICATION OF S0ll LAYERS FROM PIEZ0 CONE DATA

3 O

O Table 1.

Typical engineering parameters for each layer in Figure 6 site 10 rcr.o1. tecetten Is 004 Average sof t seight (pcfl 110.0 Grougeseter Septh (ft) 50.0 Avg.

Av9 avg.

Sadretned Thtct-pot et sleeve pere meg.

81oe sel.

Frte.

Med. of sheer From To ness stress-stress presswee friction Meterial cemets dew ity angle _ elast. strength Layer (ft) (ft) (ft)

(rG/tMZ) (tG/CM2)

(rG/CM2) ratio Type per f t.

(1)

(DEG) (RG/042)

GC8 TAILING 5 1

0.0 3.0 3.0 50.4 0.6 0.1 1.3 Sand.stfee 16 63.

42.

177 4054 2

3.1 3.6 0.6 32.9 0.6 0.0 1.9 Send.sitee 10 51.

37 115.

3950.

3~

3.7 8.0 4.3 56.7 0.5 0.2 0.9

• Saad 11 62.

42.

198.

4 8.0 8.8 0.8 30.0 0.6 0.1 1.9 Saad.sitee 10 a0.

35.

105.

3601.

5 8.8 14.2 5.4 58.0 0.5 0.1 0.9 saad 11 St.

37 203.

6 14.3 15.2 0.9 32.0 0.4 0.1 1.3 Sand.si tue 10 28.

33.

112.

3884 7

15.2 24.6 9.3 6.4 0.2 1.0 2.7 Silee 4

23.

773.

1.1 a

1 Thickness = 39.9 9.7 57.4 0.5 0.1 0.9 Sand 11 56.

39.

201.

5 Thttkness = 21.8 5.3 42.2 0.6 0.0 1.5 Sand.sitee 13 52.

39.

148.

5070.

1 Thickness = 38.3 9.3 6.4 0.2 1.0 2.7 51 tee 4

23.

773.

1.1 l

l.

FOUuBATitpt 8

24.6 25.0 0.4 58.4 0.0 1.0 0.0 Seed

• 11 33.

33.

204 t

1 Thickness = 100.0 0.4 58.4 0.0 1.0 0.0 Saad 11 33.

33.

204 i

?

4

~. _ _., _ _,,. _ _

.m t

SETTLEMENT ANALYSIS once a profile for a grid point was determined from the piesocone, a settlement analysis could be performed on that profile.

Both primary and secondary settlements must be determined.

Primary settlement begins during the i construction of the embankment.

As long as primary settlement is completed near the end of embankment construction, it has a minimal impact on_the final pile.

since hydraulically placed tailings piles are generally highly layered, a computer program capable of modeling

(

multiple layers and different drainage patterns _was used.

The program CONSOL (Wong and Duncan,1985) was capable of modeling the tailings layers and the rate at which the pile was to be built.

As the-pile is built, the tailings will settle and additional material can be placed.

If the tailings experience all primary settlement by the end of the recontouring and placement phase of construction, the O-tailings piles can be final graded and the cover system placed.

Because of this, the only impact that primary settlement has on the pile design is the tims required for primary settlements

however, the amount of primary settlement is not of significant importance.

The profiles of the piesocone soundings were checked; those profiles with_the thickest amount of slines and the largest loadings were analyzed for primary settlement.

At Falls City, the soil profiles with the thickest slime layers were also the ones directly under the areas to receive the most fill for the final shaping.

These profiles were analyzedt the time required for primary consolidation is only three months after completion of construction.

Since experience from test fills on hydraulically placed uranium tailings (Chen, et al.,

1988) has shown that the time for primary consolidation is O

. a r tiv ov rvredict a 6:e 5 to 2o ti u in,1 dor tory determined Cv

values,

.t was concluded that primary consolidation at the site would not be a problem.

If the analysis had shown that the time for primary consolidation was significant, then the construction sequence would have had to been broken up so the the final cover would not be placed-until primary consolidation was-completed and test fills would have been recommended to verify the predictions.

Secondary settlement tends to be more critical to the final performance of the remediated tailings pile-and can have an adverse offact on the-pile.

However, unlike primary settlement, it is more difficult to control.

If secondary settlement is excessive, there are. few things that can be done

+a mitigate its effects.

Because of this, special attention needs to be given to secondary settlement.

The amount of secondary settlement was calculated using_a combination of piezocone data and

2 r

f 2

laboratory data.

The-following equation was used to calculate the amount of secondary settlement.

j S=C-

• N

• log (t /t )

2 y

where i -

8.=

secondary settlement.

C=

Strain secondary compression coefficient.-

H=

thickness of' layer to be analyzed.

c t=

Time of the beginning of secondary i

settlement.

t=

Time at the end of secondary settlement.

2 The secondary compression coefficient was determined from laboratory data; the height-of-each layer was determined from the piezocone profiles.

Since it takes approximately three months for primary consolidation to be completed, t1 was set to three months; t was 1000-years 2

. C (the desired design life of the embankment).

Using the above equation, the secondary settlement was calculated

- for every soil profile made inside the final embankment footprint.

i secondary settlement in itself is not necessarily all that serious; however, differential secondary settlement is very critical.-

Unlike primary settlement, it is not possible to simply analyze the profile that is the most 1

critical because it is the difference in settlement between two profiles that is critical.

It is for this reason that the secondary settlement was analyzed at every point on the grid and then compared to the settlements for the surrounding profiles.

The results of the settlements were analyzed so that the strains could be calculated and compared to those needed to cause cracking of the cover.

STRAIN CALCULATION The method used to calculate the final strain in the cover was - generally based on approaches to analyze the effects from subsidence due to mining or pumping of underground fluids (e.g.,

water or oil)

(Brauner, 1976; Ime and Shen, 1969).

Figure 7 shows the relationship between vertical settlement, horizontal displacement, and horizontal strains.

As a

trough is formed due to settlement of the underlying soils, subsidence is also accompanied by horizontal displacements.

These horizontal displacements are_ larger than those normally expected from the-curvature caused by the subsidence.

The magnitude of the horizontal displacements are proportional. to the f

slope of the subsidence profile-or the first derivative of the ' trough - created by - the subsidence profile.

Likewise with the-horizontal

strain, the point of maximum 4

4 as-** *'-%

..., W m L

,m

-~m NH*

• O O

h t

i I

l t

I I

XTENTI EXT EMS 80N t Al HORIZONTAL COMPR ESSION g

STR AIN I

g g

I I

I

{

I I

I I

I l

8 I

I I

I I

f (St HORIZONTAL l

MOVEMENT I

I I

I a

i E

I a

w I

g I

IC3 VERTICAL SUBS!DENCE -

I I

g I

I I

I I

I I

I I

I I

I LOADING I

l 1

I

!!Ill!Illl 1113131 i

UPPER STIFF LAYER h

i DEEP COMPRESSIBLE LAYER *

.. ~..

. ~..(FrDm Lee & Shen) ph/4W/4WM/4W4V/AWXNWh@g FIGURE 7 IDEALIZED SURFACE MOVEMENTS AT A SUBSIDING AREA t

i

F l

horisontal strain is the point of steepest slope on the horisontal displacement curve or the second derivative of the subsidence profile.

The shapes of the curves and relative positions of the movement curves agree with both observed data and theories.

In order to evaluate the strains, the shape of the 4 settlement trough must be calculated by determining the settlement in the tailings caused by the added load of the shaped fill.

Once the shape of the settled area is determined, then the horizontal displacements in the cover can be calculated by taking the first derivative of the i

curve.

After that, the second derivative of the curve is calculated; it is this curve that represents the horizontal strain caused by settlement.

If a

cross i

section is drawn through a series of holes suspected of being the most critical and the deformed shape of the surface is plotted,- then the derivative can be obtained 1

using graphical methods.

In some

cases, it is very O

difficult to determine which is the most critical surface.

Therefore, many sections have to be analyzed.

Even then, the most critical section may not have been identified.

Since the effects of differential settlement are critical, a method of analysis was devised such that the entire cover could be checked, thereby eliminating the need to i

guess which section was the most critical.

With the use of computers and advanced analytical software, the methods used to evaluate tensile strains for 2

cover cracking can-be improved over graphical methods used in the past.

The computer program used to analyze the tensile strains was CPS (Contour Plotting System) from Radian corporation and was run on a Micro VAX.

All northing and easting coordinates and settlements were entered into the computer and were contoured and l

O plotted.

(Figure 8 shows the results of the secondary settlement contours.)

Next, the horizontal displacement was calculated by taking the first derivative of the settlement data and contour plotted as shown in Figure 9.

Figure 10 shows the second derivative, which represents i

the strains in the cover due to differential settlement.

The positive values represent tensile-strains; the negattve. values represent compressive strains.

The contour plot of the strains can then :be closely examined e

and values.of maximum strain can be located.

This method allows the entire cover _ to be of factively. analyzed and i

evaluated rather than simply taking a -few sample sections and - then extrapolating their interpretation to the whole pile..

1 P

~,,

-c

.m

,~m,

,,,,y

..,ww,,

.,.,,,y ym. % m,,.m m

...,,m,,%.,,.,m,.,w

,,,,m,,,,

,,..m.,,_,._.._,._m

,,.#'^~

%W

'N

\\

\\

U k

&Me/

o

% #*ESN)

</ A

((

a

\\

n N

q Q.

o

~

P.e

\\

A&

4

+

l-il a %%

e ~

p

(.

4ybl*fj:*

'\\

%*p

-,~

EVAWATING STRAINS once all strains have been calculated, they must be evaluated and analyzed to determine if they will have harmful effects on the final pile configuration.

Table 2

.shows the results from several researchers (Noore and Hor, 1984; Ajas and Parry, 1975; carrigan 1972; Leondards and 8

Narain, 1963) who have investigated the amount of tensile strain that compacted soil can experience before cracking.

These results were also compared to many years of observations made on the effects of tensile strains caused by subsidence from underground mining.

Table 2 also shows the allowable tensile strains for concrete and masonry walls.

Although these - structures are considered to be much more rigid that compacted soils, a comparison of the allowable strains shows that they are not significantly different.

O

.x - paring the maximo strain ca1cu1sted on ngure 10 (0.003 percent), it was concluded that cracking of the final cover was not a

problem for the life of the embankment since this strain is much less than any listed on Table 2.

As an added assurance, the cover will be constructed above optimum moisture

content, therefore increasing the cover's ability to withstand tensile strains.

If cracking from secondary settlement had been a problem, then few steps could be taken-to eliminate its effects on the embankment.

The only reasonable thing that can be done to reduce secondary e asolidation ist (1) to reetce the height of the in-situ tailings, and (2) to replace the material with compacted material that has a lower secondary coefficient of consolidation.

O CONCWSIONS Although hydraulically placed tailings are often highly variable in both material type and behavior, they can be effectively and economically characterized using the piezocone as a logging tool.

By using both published and site-specific data, the degree of confidence in the results can be increased.

Not only can the material classification be obtained from the piezocone for engineering

purposes, but other parameters ~ such as friction angle, blow counts, undrained shear strength, and over-consolidation ratios can be calculated for settlement and stability.

The piezocone also enabled significant amounts of data to be collected quickly and economically.

By using the methods discussed in this paper, the.

long-term performance, settlements, and strains of the cover can be calculated and evaluated.

Any potential

~

^

O O

Table 2.

Tensile strains at cracking Compaction water content relative to optinum Tensile strain water content 1 Material type at cracking %

Reference

-3 to +4 Limestone caly 0.13 to 0.31 Leonards & Narain (1963)

. OMC = 25.91)

(bending tests)

(

t

-3 to 0 Kaolin 0.21 to >0.55 Carrigan (1972)

(OIC = 27.5%)

(model tests)

-3 to +7 Gault clay 0.5 to 1.7 Ajaz and Parry (1975)

(OMC = 24.6%)

(bending tests)

-3 to +3 Balderhead cla.

0.14 to 1.6 (OMC = 13%)

-6 to +2 Gault clay 0.08 to 0.34 Ajaz and Parry (1975)

(tension tests)

-1 to +5 Balderhead clay 0.08 to 0.48 N/A Plaster O.1 Voight and Pariseau (1970)

N/A Reinforced 0.25 to 0.40 Voight and Pariseau (1970) con rete i-

I l

l

)

problems can be identified during the design and steps can be taken to mitigate the problems before construction.

ACKNOWLEDGEMENTS This study was completed as part of the DOE UMTRA 6 Project, headquartered in Albuquerque, New Mexico, and was supported under DOE Contract No.

DE-AC04-82AL14086 to Jacobs Engineering Group Inc.,

the Technical Assistance Contractor (TAC).

The TAC consists of Jacobs Engineering j

Group Inc.; Roy F. Weston, Inc.; and sergent Hauskins and Beckwith Geotechnical Engineers, Inc.

The authors wish to express special thanks to Ron Rager and Jack Caldwell for valuable technical input; to Dianna Muller for help in manipulating and plotting the data on the computers and to Pam Preisen for editing assistance.

O O

.-.,---.-m-,..~,_,r.-.,_,-.,_...r

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

,m.._

REFERENCES Ajas, A. and Parry, R.H.G.

(1975). Stress-Strain Behaviour of Tso compacted Clays in Tension and compression.

Geotechnique. Vol. 25, No. 3, pp. 495-512.

a

Brauner, G.,

1971, " Subsidence Due to Underground Mining,

1. Theory and Practice in Predicting Surface Defermation,"

U.S. Bureaue of Mines IC 8571, Washington, D.C.

i

Carrigan, L.F.

(1972).

Tensile crackina in clav Cores of Earth and Rockfill Dams.

Thesis (M.Eng.Sc.) University of Melbourne.

O

Chen, P.K.,

B.

Keshian, and F.B. Guros, 1988, " Predicting Settlement of Hydraulically Placed Uranium Tailings",

Hydraulic Fill Structures 888",

Geotechnical Engineering Div., ASCE, August, 1988, Ft. Collins, Co.

Douglas, B.J.

R.S.

Olsen 1981.

" Soil Classification Using Electric Cone Penetrometer,"

Symposium on Cone Penetration Testing and Performance, Geotechnical Engineering Div., ASCE, October 1981, St. Louis, MO.

Lae, K.L.

and C.K.

Shen, 1969.

" Horizontal Movements Related to subsidence," Journal of Soil Mecahnics and Foundations, ASCE, New York, New York.

O

Leonards, G.A.

and warain.

J.

(1963).

rlex1hility of Clays and Cracking of Earth Dams.

Jnl. Soil Mech. and Found. Div. ASCE. Vol. 89, No. SM2, pp. 47-98.

Moore, P% and A.Y.T.

Hor, 1984.

" Cracking Behaviour of compacteo Clay," Fourth Australia-New Zealand Conference on Geomechanisms.

t Robertson, Peter K.

1985.

"In-situ Testing and its l

Applications to Foundation Engineering,"

Invited Colloquium Paper at the 38th Geotechnical Conference, Edmonton, Alberta, Canada.

l.

l

^i

.... ~ T-

_ z ;_;=;-

i i

i Robertson, P.K.

& R.G.

Campanella, 1984.

" Guidelines for Use and Interpretation of the Electric Cone Penetration Test,"

prepared for Nogentogler and Co.,

Inc.,

Gaithersburg, MD, at the University of British columbia, i

,Voight, B.

and W. Pariseau, 1970, " State of the Art of l

tPredictive Subsidence Engineering",

Journal of Soil Mechanics and Foundation Engineering, ASCE, SM2, New York, New York.

J.M.

Duncan 1985.

"CONSOL: A Computer

Wong, K.S.

and Program for 1-D Consolidatlon Analysis of Layered Soil Masses,"

Department of Civil Engineering, Virginia Technical Institute, Blacksburg, Virginia.

O O

.m-..,

r

-74 m.--

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

.,m_,..____.ym=_wm.,yyg,,,...,,--,,

E 6d M/** ' E' UMTRA@OE/ AL gg0g20.

O n ; - w :)

r'

~

United States Department of Energy o

Remedial Action Plan and

' Site Conceptual Design for Stabilization of the Inactive Uranium Mill Tallings Site O

Near Falls City, Texas Addenda Di Through D5 to Appendix D Draft Appendix B of the Q

Cooperative Agreement No. DE-FC04-85 A1205LL ks O

i w

UMTRAPI Uranium MillTailings Remedial Action Project g

UMTRA-DOE / AL g50g20.

'l n : G ',

United States Department of Energy, -

/4, l

)

7

f_,

q Remedial Action Plan and Site Conceptual Design for Stabilization of the O

inactive Uranium Mill Tallings Site Near Falls City, Texas Addcoda D1 Through D5 to Appendix D Draft Q

Appondix B of the Cooperative Agreement No. DE-FC04-85 A120533 W

,' O Uranium Mill Tailings Remt.fial Action Project R WTRAP yj gia 1

UMTRA-OOE/ AL-Oggg20.

'~

a... o United States Depertment of Energy 9"

'%@)

Remedial Action Plan and

' Site Conceptual Design for Stabilization of the O

Inactive Uranium Mill Tailings Site Near Falls City, Texas Addenda D1 Through D5 to Appendix D Draft fQ Appendix B of the Cooperative Agreement No. DE-FC04-85 A120533 E

e s' O w

i *'

Uranium Mill Tailings Remedial Action Project lU WTRAP l

l f - c;l } l

~

l

_