-------
B
O
CJ
J3
H
to (x, a) u
b fit
u u a)
« w a
H
jj
H
u
UO
iH B M
H l-i iH iH
SfiS-0
0)
ta
B
s a
S3 X
U) 4J '4J
I IH -O
O >J
t-l « t
"id O W
3 IH a
O2 TJ AJ 41 C.
H O rf U
O W 1H 00
«3
c:s
«
U1 Bl
ft
§
o 8 S
r- H
. tu 01
in p^ c/i
s S
TJ w tr
EH iH «
O « 3> rH
>> 41 ^C O
,J OS M >
14 m
jt nj u
M 0)
o r-i u
4) CO M
u O 3
_- (0 -r4 W
O 3 JJ M
O\ f-t M 4) M
' 0) M C/3
> Pu Cu
*O
<
CJ -O
+ 4)
!Jo<-JQoocJ°°"^<:Jc
-------
CM
CM
^
fl
1
O
CM
cy»
/~s
5 CO
u ^
c:
o
CJ
N_X
C4
|
> r-
o <
i-l
VO
.«-4
in
*T
1*4
CO
<-4
£M
3
r-l
Q
CJ
f£ ^
U 3 B
U 04
01
B
B ft)
O i-l
H «J X
*! & a &
H O HI
"!-t a
a o o)
3 -H 0)
a- -o u a) tu
" .H o i-i w
O M vl CO Of
QJ H CJ tl)
oS £ -a o
^-i x
C4 3 M
r 13 U B
i-l 1-1 O (0 -H in
MO sum
CJ O 1-1 U 01 1
J.I ^3 (U V4 <
£1-3- < > Pi (
cj -o
4- 0)
CQ O
'^S-g
CO
O J3
r,Sa«co-n^<.c,coco--^
vO CO *tf *^ ^ *» ,*» '
r*« ex ^^ "^ ^ ^3 *^ *^ ^^ '*"* ^^ ^^ 55 ^J
o o o o o . .o-4 _ _
(sj r~i o in ^o o* r^ ^. r^ o m
^ Ti °. °. °. °. g 3 S 2 3 § S
338?°°° I'doddoo
*»v -_j -* r~i ^n t*« ««ff
l/tt O 00 ^3 f^1 U"l CO O^ "* * tJ w* ^ ^
SSS^cocScnSSSogoS
*-j« ^.y f*, co O **O *o o^ "*r *T «*i r** **^ ^
RSS^^SSSSSSSSS
^focnp^cn f?* fM ^^trtcoco-^ro o*
4 OI C"J C*4 CN C*l "' v*4 CN v*4
in«*»ri»o4OOciiii'^c^ou^
2 ^vf^^^cosr^oi^N-rsrco
5 ^.nSasssssssssss
» sr -a- sr » sr en * -a-
a co S3 o S 5 S SS o KSS co
n
-1
* co ^cr
o>CT>cr>o\CT>a>a>eno»'-<
B
4 S
CJ B
.C 4J a)
4J B 4-*
oo (d co
B -! tH
0) O. O.
Vi
4J 00 00
W5 5
U 01 01
tO M 1-1
CJ CO CO
£1-1 H
4J 4J
.ii 3 3
id o o
u M n
tu O, 3 o.
u -a u
J3 d) O
u i-l cd u
B CO 4J H U
Q) 3 U <4-l CJ
& ?, 11 >
O T3 > B
M cd a) o
-a o S u
O 4J 1-1
CO 4J O T3
-------
extent of hydrate formation. Also included in the initial mineral grain
density. j
3. Cured Specimen Void Ratio !
The void ratio for natural soils is often used for correlating with
permeability for a given type of soil. A cured specimen void ratio computed
from the apparent initial dry density of a loaded oedometer spjecimen and its
mineral grain density after curing is presented in Table V 1. j The angle of
internal friction at peak strength has also been correlated with void ratio
for some materials later. i
4. Secondary Compression Index i
The secondary compression coefficient Ca is determined frbm the semilog
slope of the secondary compression part of the consolidation curves of the
spring oedometer specimens. In general the secondary compression rate is
lower for cemented specimens. Many specimens showed increasing slopes at
longer times on the semilog plot. Some secondary compression curves are
shown in Figure V 1. Final Ca values are tabulated in Table V: 1.
3. Permeability Coefficient ;
The permeability of the spring oedometer specimens has generally been
determined at a hydraulic differential of 20 psi across the nominal one inch
thick specimen. Too low a pressure sometimes allowed no flow at all,
perhaps due to hydrophobicity of some specimens. Some specimens at inter-
mediate pressure would balk sometimes and flow again at other times. This
also is believed due to hydrophobicity. Obviously an emplaced; liner can not
rely on hydrophobicity which can reject only, say, 5 psi water head, or
whose hydrophobicity is perhaps fleeting. The permeability coefficients are
presented in Table V 2, column 5. j
!
6. Shear Modulus During Torsion :
Initial stiffness is measured by the initial stress/strain slope during
torsion. Cemented, highly compacted, or over consolidated specimens show
steeper slopes than soft, normally consolidated, or low brittleness index
specimens. The initial slope of the torsioned spring oedometer specimens is
given in Table V 2 may be an auxiliary measure of specimen softness. The
shear modulus 6 (column 23) has been calculated from this data!.
7. Peak Shear Strength ;
Beyond the initial steep part of the torsion stress/strain curve the
curve droops and often a maximum stress develops from which the peak shear
strength used here is calculated by the empirical formula having the same
form as the derived formula for T_ discussed later: ;
m 3 (moment) » moment '
T? 2u r3 2. ir(1.25)3; '
3 j
inch Ibs moment :
4.091 ;
- (52.17) (LVDT volts), piin lb/in2 '
The volts EMF output of the LVDT system of the torsion proving! ring is read
from the stress/strain curves recorded by an X-Y recorder, The X axis of
49
-------
10
20
30
N
50
60
70
3 80
ID
-------
the pen is synchronized to the extent of rotation of the worm 'driving the
' °f the
The peak shear strengths are tabulated in Table V 2. ;
8. Residual Shear Strength i
, ., A!ter a fertain amount of shear occurs in the torsion specimen
following peak shear strength development the stress subsides knd generally
asymptotes to a lower value than the neak eji-reca Th-sc, ? ~-ii j 2 ^
>-A.«j.ii uii«s pcctK. stress, inis is called for th
-------
TRIAXIAL TORSION SHEAR TEST WORKSHEET
Date of torsion test
Loading No.
Sheath No.
Shale mix and water, %
Compaction,, std. or mod. proctor?
Was Specimen permeated?
Spring loading hydraulic pressure, a
Consolidation pressure from spring 0.2579 a=A
Containing can pressure = 0.7 A = b
Instron pen above balance baseline after
pulling sheath observed, c
Initial mismatch in soil skeleton pressure after
Sept. 25. 1985 ;
36 [
21 [
Tosco 80/20 24 [
G, T Mod. Proc
Yes j
pulling sheath
- 0.11393a
Increase instron pen reading before torsioning
by 4.909 D = X by raising lower platen on
slow speed with manual knob X
This results in pen position above balance
baseline of X + C to hold during
torsioning by manual platen position
knob adjustment
X+C
1150
296.6
207.6
-85
-148.3
728.2
643.2
'PSI
PSI
;LB
iPSI
LB
52
-------
LOADING 42 0.5VOLT/INCH
24 30 36 42 48
TWJST, DEGREES
54
Tigure V 2. Some Torsion Stress-Strain Curves, Loadings 32; 33',
42, 64, 92 " i
53
-------
simple shear or zone shear action may occur. Even low amounts' of cement-
ation seem to prevent simple shear but promote shear planes arid development
of a counter system of jointing during strain in these materials. Table V
2, (column 20) presents the brittleness indices determined.
Figure V 2 shows some torsion, stress-strain curves for some oedometer
specimens in the triaxial torsion apparatus.. Several types of curve are
presented. Curve A is for a soft uncemented silty specimen, B for a mildly
cemented, specimen, C for a more strongly cemented specimen, and D for the
special mix of mellowed burned TOSCO with some burned TOSCO. iThe
brittleness index of curve A is zero, of curve B is low, of oirve C is high,
and of curve D is negative.. The negative brittleness index occurs with low
permeability clayish specimens which drain more slowly than the usual silty
specimen. This is believed to cause some pore pressure development during v
deformation at- the.. ra.gj,d.twlsfing rates of around--2 degrees pe^mnute"'
standardized.here when contraction of th'e specimen soil skeleton during
shear occurs. , , .
10. Photographs of Sheared Specimens .... ;
Before "loading" or compacting the material into the spring oedometer
sheath four yellow acrylic paint stripes were brushed longitudinally on the
inside of the sheath at a position including the future locatibn of the
specimen in the sheath. These stripes easily transferred from the Teflon-
coated inner sheath surface to the specimen when it was extruded from the
sheath by sheath raising before torsioning. The greased gum rubber membrane
did not attract the paint stripe either after the rubber was laid on the
specimen during the sheath raising operation nor during torsioning itself.
As a result, if the yellow stripes were not. disturbed during compacting by
the proctor hammer, the yellow stripes served as a marker of movement of the
specimen's outer surface during torsion shear. Shear planes are often
clearly distinguished that would be indistinct without the marker lines.
Simple shear or zone shear of soft specimens of low brittleness index is
indicated by development of diagonal marker stripes without discontinuities.
By this means it became apparent that only a little cementation or brittle- "
ness index served to inhibit simple shear and produced a few shallow angle
. shear planes in spiral configuration oriented such that the lowered normal
stress on the planes due to rotation of principal stresses was! contributing
to the development of the spiral fracture.
Figures V 3, 4, 5, 6, and 7 are photographs of marker stripes of the
typical torsioned specimens corresponding to the stress-strain curves of
Figure V 2.. When the piston was removed by pulling as nearly! vertically as
possible some tension developed in the specimen which sometimeis caused
cracking along the weakened slip planes and their easier observation.
Joints also were opened up which may not have been very obvious otherwise.
These tend to cut across the slip planes at 45°.
1.1. EGA Determined Hydrate Water ;
Hydrate species water evolution and carbon dioxide evolution curves of
some of the spring oedometer specimens after torsion testing ate presented
in Figures V 8-32. By integrating the area under peaks along the curves.the
hydrate water in a particular species formed during mixing with water and
curing may sometimes be approximated. Figures V 33-36 show the raw
materials autoclave mellowed Lurgi M 14, autoclave mellowed buirned TOSCO M
15,. unwetted Lurgi, and unwetted burned TOSCO, respectively* :
54
-------
Thus the peaks of water evolution contained in an EGA curve are an aid
in identification of hydrate species such as ettringite, gypsum, brucite,
etc. and their assay. This work has been done in conjunction with X-ray
diffraction as often EGA peaks of two species will overlap. To further aid
in identification of species evolved C0_ has also been recorded with the
evolved water on the same chart. Table V 1 summarizes the hydrate water
determined by EGA up to 500 C. One entry is for total evolved' water,
another is for water to 255 C, and another for 255 to 500 C.
B. Brazil Test Results ;
i
Table V 3 is a summary of Brazil tensile test results with sample
designation numbers paralleling the loading number of Tables B-V I and B-V
2. Suffix B indicates Brazil test, small letters distinguish Duplicate
specimens. . |
C. Pneumatic"Pedometer Compressibility Coefficients
!
Table V 4 summarizes results of pneumatic arm oedometer .study of
standard proctor and modified proctor specimens of freshly mixed material.
55
-------
CO
CO
CU
ni
IM *H to
IH 4J T3 4J
H M cd iH
u o cu o o
CO 4J > (J >
j H co cd fe
CU 4->
r-1 60
H £3 s~*
CO tt» H
(3 H CO
CU 4-1 P4
an
4-1
£
4-1 Cd
CU 4-»
4J -H
§5
CJ >-*
&
a> i c
i-l 60 O
CU-H -H
g co w
ci) CU cd
CO Q fi
CO r-4 i-l r-H CM CM
coin"'mcM^vovo
OOCitir~i~»mm'-ivovO'n
^ co oo r^» ^f 3'cocoinincocococ->i
cMCMCMCMcMinmin-Ji'f^tN~in mo o in in
CM CM CM CM CM cococo.
i CM CM CM CM
tofflsestnwsggessesee eseeeeee^scotng^ss^s
in irt CM c*j CM irt trt cvi CM CM CM CM CM r** r** CM CM r*** i^1* i*** oo oo oo co oo oo co r^* I*** co co co oo oo
CMCMCMCMCMCMCMCMCMCMCMCN!CMCMCMCOCO eMCMCMCMcMCMcMCMCMcMcMCMCMcacsicMCM
1-1
r-l
O '
CJ
CO
O
E-l
81
CO
cd,acd,ncjcd,Qcd,ncd,ncd,jacd,ncd,Q cdrfJOcd^ao cd^3 cdja cd,ocdrO
ss^sss^^ssssssss ggggsssssg-§§gffss-s
t-4 f-H ^r _
r^*t r*^ in in in co co co co
ON ON
56
-------
^o
W
|
^
CO
i_j
5
H
CO
CO rH
CU Cd
C 0
i-H HH ^H ci
C3 iH (-1 cfl; r-
H W O CU O O
f-l CQ 4J > rJi >
rl
0
rH CU 0)
Cd rH £3 iH
JJ iH iH 3
CU CO 0) l-
U ti rt T
cd a) -u 4J cd
t-J EH CO cd F-i
»
o EH >-*
(J *
o -u
I ** O
B O <4H
Std <4H
p, W
' w cd
rl C -H
CU >
CO EH
CU 1 a
rH 60 0
a -H -H
S W 4J
cd CU cd
CO O C
^3" P^ CM ^D ^O
CM m in vo vo
CO O O O O
rH CT\ O rH
CO rH CO ^*
O^ CO CO CO CO ««^
crv cr^ vo vo vo in
co cr\ ^d* r>» co CT\
m -* CM co CM m
CO CO -H rH CM CM
VO VO CM CO CO CO
CM CM
s e e m s e
CO CO G\ *& '*& *&
CO CO CM CM CM CM
O
CM
""I.
CO
oo
o
CJ
CO
o
H
cd 43 cd 43
pq pa pq pq pq pa
o\ ON vo o in *n
ON cr> vo co r^ i^
vo CM in -H
vO vO vo vO
O O O O
o in vo co
rH rH CM CO
*«3" *^"' rH CO
m m o\ in
I-H -l ^J" IO to tO ^f ^J" ^t* t^ VO vO
iCMCMCMCMCMCM CMCM
i
g s s g M co EeegracogggegEggg 33
i ' ;
*^* ^^ in in co co CM CM CM CM o co P^ P^» i^* CM CM t"^ P^ CM CM in in
COCOCMCMCMCM CMCMCMCMCOCOCMCMCMCMCMCMCMCOCO m^
'
C-'
O i 3
CO . pQ
1 i m
O vH ; CM
r-» 60
0- ---- «_-__-__ _._ _____ g_
U i-4 i EH
M rH
O ! C1J
r I ^J
: "^
cd 43 cd 43 cd cd
pqpqpqpqpqpq pqpqpqpqpqpqpqpqpqoOrHrHcMCM pqpq
in f^> in r*» CO
-------
?
4-1
<3
O
^^
CO
ta
1^1
«
H
CO
CO
0)
js-
Cd
iJ H
f O
N
rH
S
0
E3
1
,3
i
e
8
0)
4J
.*
£3
CD
1
td
CO
cu
rH
ft
cd
CO
i-!
td
O
H V
4J -d U
!-l cd iH
cu o pc
cu
(3 M
i-l 3
cd t-
4J 4J CC
to cd fa
J3
0) 4J
r-l 60
H C y-N
CO CU M
tt) 4-1 (Li
»
CH *»'
(3 *
0 4J
4J O
O MH
ft W
4->
00 CO S CM So SIo ^ CO
-H
i^. i^. r^ r- o o 0 o r~ o
CMCMCMCMCMCMCMCM CMCO
ssseeese eco
inoomoomin mm
rl
W 60
H H
co 3
O hJ
H 0
-- m
o -^.
" m -H
r-\ h-5
r^ r*^ i*^ co CD ^^ CD in ^n
CO CO 00 CO ***^ *"^ *^ »^ CO CO
OA
^j.
CN
t l
CO
0
'
f*»
r-l
T-4
O
CO
E
O
CO
_
o
in
co
o
CM
O
ON
CO
00
i-H
CO
CO
CO
a
m
CM
_
m
co
CO ON *3"
^t* CO *^"
o o o
CO VO CO
o co cr>
co o\ ON
vo m in
CM CO CO
CO CO CO
6 w co
mom
r-l CO CM
1-1
60
3
0
m
Tt .
M»> ' B.
V
I-l
1 1
1-1
Q)
S
o
m
td 43 O
CO CO CO
ON OS ON
ON t-» vo ON r-
*%J" CO CM CO CO
O CM
O O O rH rH
ON in m ON m
co ^ r^ t*N CM
CO ON OS ON ^^
rH co co in m
OO -
t)
: 1
4-1
e
. -H
4-1
CO
rH
td
; o
H
! Jj
Vi S-<
4-1 4-1
,O O i-H
:o o td
ft ft -H
14-4
U CU IW
td -H O
TJ IW
(3 -H ft
td T3 O
'4J O rH
co S co
II II II
co S *
* *
58
-------
U
aj
O.
CU
1-
E CJ
O U
ti- -5
c
in » i
4->
r- C
=1 O
(/! -r*
CU I/I
a: v
ai
CU
_
O.
E
t.
i Q. C
ra S o
JS O "-
Cn O P-. C*J
in co <*i "-1
O O r-4 JH
o O o O
en co « a
r^ to ^T in CO
O -4 i 1 CM ^f
.0000
cn en to
in r-s. cn
0 0 0
0 0 0
rH3°s°:
rH O O o ° i
o « a» lOineMcocn cnOrncot-* co to co CM CM 2| r co r-. « to *r CM cn
-.os sScSSo SSJSSB ssssg °S!§ss sssssss-
^ ^ ^4 ^H" r-l 00000 ^4 ,-J f-H -* ,-J '"l ,-J .-I ,_! o* O* O* O o' °
v> u in in r-* o ^o ,_« r< «* *3> ,-* cn r-, CM cn co S^oiocnn r^coS0^
^-inm^10 iovor-*coo ir>ir)^or»»CT» Ssvn^oco*-* CT»O>OO
cofotofo**' «3-^-«3*^.fm cofy^cocnco 'cococo^r mco^*^-
* * *** _J_!«I^T_T -*_r_r_j__r t r * ,
,_! y«4 f«4 ^^ ^^ <>_4 i^>4 (-H (4 (I i' |i| **^ ^^ rl P^ r^ r~fl «~* r< r~^ T~<
O
»^*p««.p-* cn co ***<** cn **^ in ir> in if>
_-? inininvn"^ t^to^ovovo 2?°^ 5?°^
o. o (u g co cn co cn cn ^^ «»^ <^ «^ «cj* cn PQ
tO ' 2B 01
»
j^gjtr.(j lotoioujto cn o> r> i/«j tj CO' CO CO CO CO r^ ^H f^ k«-4 r-4 *"* rI ^* »^ r-H JC.J *»! »-H r-M »«H OJ to
ocMincstM >Ttor~.o-cM eDcocnr--^- rnococnrH Scovooc;)
OOO CSOOrHCM C3 rH CM «3" f C3 C3 rHCOr-, .rHro I
N
^r »
CU-r- rHCT»«3-rHCO COr^OCnrH lOlOCOCM^J- CnoCMCn^d- ^^."d'COrH^
op-*incMin r*o^3-*rO oincnoio Ooocor^ rncnto^rCO
cncococo^* cocoi toin cr»cof^-tom cor^io«3"rH cncococo^
1/1 l/l C/1 rH rH rH rH »4 rH rH rH rH f4 rH rH rH rH rH rH rH rH rH rH rH rH rH rH ,
4->
(
ra-3 cococococo CMCMCMCMCM ^s-^-T«a-^r cococococo minininif)
. £Z- 00000 C30000 rHrHrHrHrH tOlovotOtO CMCMCMCMfM
oc cncncncj>cn cncncncncn cncncncncn cacpcpcoco cncncncnen
i.
CU C
U 4-> 4->
CU U U
_.«_ coocsinin cocsointn coocsmm cocsomin coooinm
atr-t. cscomtoi*^ ocointor-. ocomioi Orointor^ cocointo
CUCUOE oocacao cacaooo ocscscso cscaocao o o o o
^3 C3 C^
(M CM CVJ CJ C^
0
eo
in in in in in
CM CM CM CM CM
CO CO CO CO CO
CM CM °J ^J CM
CO CO CO CO CO
E
o
E
0»
59
-------
o
1
1
vu
3
r
ul
vo
OJ
t- X
O. OJ
E T3 O
O C O
0 I-H
tO O
sic
TJ.2
'S£ o>
=-0: ,
^T» <-»
>> c-^.
S- OJ E
O CO O1
c
OJ
S -0 4J
f OJ JZ
U "O CO
OJ IO -r-
Q. O OJ =
in | -^ j;
V7 J .A Ol
*io +>
!_ =-r- 0
OJ * h-
N
4->
"S 4J >
N
4-*
r- OJ 1/J
,= #'=
OJ C
vu o u
E OJ OJ
O t S-
-0 'r- I.
vu vu o E
0 00 H
r- S- -O
4J OJ VU
r- 4J -O
C 10 T3
3 «:
t
u
OJ 10
r- 0. c
a E 0
j: o
00 tj -W
co in co ,_
m co co p^
1 CO CM CO ^,
o o o o
CO CO CO rH °^
"*"- S S CM
OOOOo
o cn rH cn ^
cn co co oo S
CO C3 O O o
CM O IO 2t CM
r-co co cn _
CO CO CO CO rf
CM XM CM CM CM
CO CO CO CO co
CM CM CM CM rj
CM CM CM CM jy
CM CM CvJ CM pj
o o ^~ r-^ tn
VO 0 rH VO r-»
OCM CO ^J- r^.
cn co rH r** co
«d- VO CO rH m
rH rH *" rH rH
CO CO fO CO CO
CO CO CO CO CO
CO CO OO CO CO
co o o in m
o co in 10 f»
0 0 0 0 0
0 0 0 0 o
CO CO .CO CO co
01
L.
cn CM CM CM
^- o cn cn
rH CM CM <3"
CO O o O
CO m rH CO CO
^l" *r 10 co "
i* CO CO o rH
O o O O CO
CM r-» rH co CO
ScoScow
n VQ f^4 QQ C3
CM oj n ro to
**> CO fO CO CO
tr> tr> in ifj to
CM c\j CM e\J CSJ
CM CM CM CM OJ
*
CM CM CM CM CM
O CM "3 r^. -
CM CO *O CO Ol
0 oi-*CM^f
CM O CO O «">
l« ^-l in 0% CO
o1* a* co r^ vo
»-* ^-1 i-< i-t r-H
o> cn cn cr> cn
-i i-i -* t-i «-*
CO o O to tO
o m 10 ^o r-^
o o cs o o
CM CM CM CM r-» r*»
rH rH.rH rH rH
CM CM CM CM CM
CM CM CM CM CM"
CO CO vo en in
rH VO *3" IO «3-
r^* rH CM co o
CM cn ^r r»» co
cn co co r*. vo
rH rH rH rH rH
O O O O o
cn cn cn en cn
rH rH rH rH rH
CO O co in m
o co in vo r-.
0 0 0 O 0
CM CM CM CM CM
CM CM CM CM CM
E
CO
CO
rH
co 2 r cn
rH * t-- rH
1 vo P; cn co
0 °. 0 rH
*3* vo in *a- r^.
in co OJ cn cn
rH rH l^ CM CO
CM 0 ° 0 CO
*
CO o ^ CO CO
0 0 ° CO CO
§^s°°
co Vo «*J 5 »
rH rH rH rH rH
» in "^ in in
10 vo" }° vo'io
rH rH rH rH t-H
CM CM |JJ CM CM
CM CM ^ CM* CM
CO «S- O CO I-.
vo co cn vo vo
rH
^f in CM CO rH
co in o co in
r«* 10 m co o
rH rH rH rH rH
CM CM CM CM CM
CM CM CM CM CM
CO CO CO CO CO
rH rH * rH rH
co co o m tn
o co tn vo r-
co o o o o
in vn m m in
CM CM CM CM CM
CO
o
rH
4-*
60
-------
E -o u
o
5» <
>>
M
** O
_ in u
>» c~,
Jr OJ g:
<=> Q I j>
C
111
H 13 *-»
"- (1) JC
o-o cn
<» a ^~
°- o 01 .
. _i 3: OT
i-"* 5*
~~a> ai .JE "£J o
S £££;£
« £ eg a 01
o
cj
o >
-
«= o c:
" 00 =1
S- .
to c:
*» +J +:>
j- s- -a
*J ai ai
r; *» -a
"
a. E:
1.2
CM io f* ^r
r^ cr» CM in
i-l CM * \O
O O O O
*& CM en cr» r^
^ CM en CM
- o in o
S co in o
o o o«-<
r~- CM r-. r-i
co o ti en
r-> CM co »a-
o o oo
VO O CO r-4 «-4
OOOOO
co f"* en co p-» ID CM en
P- CO CO cvj CO CM .-4 CM
SSgg '§oSS
in r-i oo «« en
r^ t-4 co vo tn
U? O O O t-H
co cz> o co o
co co 'cn p*« p«*
£i!S2cnr~ ioJo«3.p^r-jiocMinioco 10 ,_, co o in co CM co CM -<«3-cnu3u3 «^cocOrHco r-cooinm
o rt CM co * in >o us rcn CM CM CM co ^r in S «3 rt! co «? § S in S
in in in in in co co co co co ^j. «^- *a- ^j. ^. co co co co co co co co co co
""* "* f-H "* * "* *~* ' *H ""* »^ »* l< T-< *t «""* r-l *' I) II iI «I t"H 14 T-4
CMCMCMCMCM r^f^r^r^^* CMCMCMCMCM Ooooo ooooo
encncncnen co CO co co co 1010101010" m^iniXin i-«r--- co eo co co en m 4 (vi «f co co m co CM
SSS:S0 SS^°S cMOT^JoS SS«3?:2 SScoScn
00001^.^010 o^oooor-.*" p* <& *& m *& ooco^iotn' oocor^rxio
~* *"^ *""* *"* ^^ *"* J ^* ^* * ^^ «^ ""' »^< *~* *""* ^~t ^^ i^^ »4 41^ t| ,~4 <£J in in in o eo o o o o o o o o oo co co co oo encncncnen
cococoooco foSI?0^ co co ininmmin r*.p^i r^p^ ^-*a-«3-*r*3-
co co co co co cn cn cn cn en p^ p^ p««.p^ p^ co co co co co co co co 03 co
,-4 rt -H f-4 1-4 rt^Mrtrt ^4^41-4^,^, ^-4 ^ rt S S S S S S S
coooinin coOoinm cooOinin eooOmin cooomin
ocotnipr-. o co in 10 p^ o co in u> t o ro m 10 r-. o co in in K
cooooo ooooo ooooo ooeooo opooo
co co co co co p^ P"* r*^ r^» p^ co co co co co ^*" ^± ^" ^j~ 'a' o o co o co
CMCMCMCMCM CMCMCMCMCM CMCMCMCMCM CMcMCMcMCM CMCMCMCMCM
eo
S
S
o
O
CO
O
O
61
-------
01
V. X
Q-
OJ to
£ «.£
- * 2
i
o
CJ (Q
"a E .2
JO CM to to IO CO
~ in cn i^ cn to
- co l CM in r^ ?-*
O C3 O O O t-<
t cn r^ co 10 -^j. cncocncn cocntom"
CMto'-'m «a-cor-.cO'3- '-'cn'-'cM'
OOO^JCM r o 1-1 CM co '"'OCMco'
OOO^DO oooo o CM o cj o
O CO O O
oj in to co
CM CM m CO
1 g S o o
t vo i in r** co '
i co cn ' ^~ vo cn <
! o o ti»-» i ii o o <
' O O O O i CM O O <
coo^fCMr^, ,-1 CM in i-i vo £r r** i£ c^ **" «-< P-* co «3- in ' o ** ^H co o vo co -*f co cn CM *3* io CM cn «* vo «t cn cn co
cOcocnOcM cocnocM^r t^CMfOmcn CMcMcncoin '*-* r^. i CMCMCMCMCM
--jvovovo cococoroco cDoOoO r-*r*-.p^p^r-«*iCMCMCMCMCM
""7 co co co oo ^* ^ ^- ^ ^" ^s~ ^" "si" ^- ^i~ co co co co cn ^r '^ ^* **" ^~
r
. .-, » j co n vovovovovo cncncncncn CMcMeMCMcM.inininmin
> in tn *n tn co co co co co r2 »* t2 »-* ^ CMCMCMCMCM ' tovovo'^vo
»!«» f*^. r** r>*» p*» r*^ r^» r*» P1^ co co co co co vo 10 vo vo vo vo vo vo vo vo
CMCMCMCMCM CMCMCMCMCM CMCMCMCMCM CMcMCMCMCM'cMCMCMCMCM
|
!
CD oO ^H &* P** CM ^* CM co co co 10 CM p*» ^^ co io co co in r CM CM ^ cn r-»
"^" ".voCDco co^-vovoo CM ^- CD co ^° ^ co cn m co in ^~ P** m r**
I^COio f-4CM<3T-<.CM «-«CMVncO'* OO»-ICOin I Of-HCM^'f**
rt "--*" '*,*"*;
.tn cncor**401^
CMcMcvJ^JcM ra co co co co Scncncjicn Ooooo'ininmvotn
cococgcoco cococococo cococococo cncncn.cn en ; cncncncncn
coootom coooinin cooOtom coooinin coootnm
Ocoirtto^- ocointor oco"nto!~; 2coinior-.iOcotntQf^
OOG^OO ooooo oocsoo oesooo ' ooooo
CM Csj C\
CM CM
CM CM
CO CO CO ro CO
CO CO CO CO CO
O C3 O O O
o
CO
10 in in tn m
CM CM CM CM CM
in
CM
in
-------
Figure V 3. Torsioned Specimen of Oedometer Loading 33, A non Brittle
Material '
Figure V 4. Torsioned Specimen of Loading 64, A Mildly Cemented
63
Specimen.
-------
Figure V 5. Torsioned Specimen of Loading 42, A More Strongly
Cemented Specimen
Eigure V 6. Torsioned Specimen of Loading 92, A Rather Impermeable
Little Cemented Specimen
64
-------
Figure V 7. Torsioned Specimen of Loading 32, A Low
Brittleness Material Showing Shear Plane.
65
-------
EGA Detector Response (mV)
(Dotted Line)
CM
B:
CD
a*
<4-
00
c
-H
E
X.
CJ
CO
-------
CO
CO
CT)
CO
LO
in
C3)
T-l
\
C3
CO ffl
-p
0
cnct:
c
O (SJ
0 ID
_J
.. Q
CO CT)
ro c
».-< o
JC ££
EGA Detector Response (mV)
(Dotted Line)
(yyuj) esuodsey uoq.oe-%eg
Figure -V 9. EGA of Loading 31
67
-------
EGA-Detecton Response (mV)
(Dotted Line)
in
CD
(S3
CD
CD
in
& is
Si to
«
O)
T-t
CM sT
E
CJ
CM ..
CO
-------
ca
CO
ca
c\3
CD
in
ID
-:
(T)'
CM!
s c
Si -^
CO ..
CO (D
c
H
O El
O ID
_J
>. (D
(S3 CD
-------
EGA Detector Response (mV) '
(Dotted Lihe)
CM
*.
ca
O)
c\
CD
-------
00
CM
O)
S3
CD
00
E
U
O) ..
CD
-------
EGA Detector Responee (mV) '
(Dotted Line)
ta
CO
CM
C\!
El "
« -a?
00 ^
2S
in
in
£
CJ
»
S3
=& ffl
4^
OT O
c o:
O -H
0
-J ..
-------
EGA Detector Response (mV)
(Dotted Line)
(eut-|
esuddsey uoq.oeq.eg
Figure V 15. EGA of Loading 42
73
-------
LO
CM
cJ
in.
in
4-
CM
\
S3
E
rn
-------
EGA Detector Responae CmV) ;
(Dotted Line)
(SI
00
in
CM
in
in
in
Si
CM 3B
»i .
r-i £
V. \
»I * .
CO i
OT O
c o:
O »-H
0
_i ..
-------
EGA Detector Response (mV)
(Dotted Line)
(S3
(3
in
CD
in
cs
» ' i
in
r" -
CM 3c
0) C
ca -rt
E
CJ
ca
"*
o
OTQi
c
H
ID si
o s
o in
i
a
,
-------
EGA Detector Response (mV) i
(Dotted Line)
IS
C3,
C^
-+
CO
"*
s
*I
CO
CO ^
CO.
in -^
CM 3s
-
C3) C
ca -rt
£
C3) \
-r-. O
\ »
ca si
»-< «*
CD
"4-
-------
EGA Detector Response (mV) |
(Dotted Line)
ca
O)
CO
co
GO
to
c-
I-l
£
on
(D
OD 4i
C O
HI Q^
TJl
0
O «
_J
.. (D
.CO' CD
CM C
,-
-------
EGA Detector
Responae (mV) |
(Dotted Line)
ro
G3
CO
00
GO
CD
o
(D
a
in CT
do -a
O) C
(SJ -H
(T) ..
IT)
O
COCE
. -rt
-D SI
0 (S3
o in
>< (D
CM CD
CM C
»-« O
(y\iu) esuodsey uoq.oeq.eg
Figure -V 21. EGA of Loading 53
79
-------
EGA Detector Response (rnV)
(Dotted Liine)
O)
(SJ
-*
in
O)
in
in
c\i
cri 3=
x! c
si -H
T-I E
\
O
(O
in
O O
o in
c\i en
.m c
i O
Figure : V 22. EGA of Loading 56
80
-------
:GA Detector Response (mV)
(Dotted Line)
in
O)
in
CM ^
"-1 (D
CO
. £
GJ
t^ >-*
in
«
(D
U> -P
C O
D
O
-------
EGA Detector.Responae (mV) ;
(Dotted Line)
S3
"
-------
EGA Detector Reaponse (mV) |
(Dotted Line)
in
GO
OJ
3?
00
C\I
CD
O)
m in
co
ca
CD ..
U") ffl
4*
O
mo:
c
trt
13 S
O
-------
EGA |Detecton Response (mV) .
(Dotted Ljine)
en
ca
(SI
in
O)
in
o«
<4_
" GO
«-H
*- ,4
^ in
r-1
H
03 3=
ca
o
CM ..
UD 26. EGA of Loading 59
84
-------
EGA
CM
si
mm"-
in
in
O)
c\i
ca
ca
CM
CO
CM
*
Or)
(O
c
H
£
*s.
CJ
(3
CD ft
(O
%
-------
EGA Detector Response CmV) ;
(Dotted Line)
(S3
CM
CD
CO
'<*.
CD
o
ca
-fr 3e
VI
C
H
£
O
CD
Gl
(O ^
03
*& Q
-P
CO O
c o:
TO
o *-H
o
_i ..
(D
.. CT)
CM C
00 0
C. Q^
i 1 1 1 1
ceun
; eeuodsey ~joq.oeq.eg
Figure , V 28. EGA of Loading 86
86
-------
(S3
C^
"*
C\I
3?
in
in
CO
03
CO
CJ
IT)
T-H
T-H O
»-H *
IS
QD »-«
OD
c a:
rN
13
0 i-<
O
_J ..
(D
CD
a) c
a> o
i. a:
'- . I - ,
EGA Detector Response (mV) ;
(Dotted Line)
i i 1 1
(eui-j
esuodsey ~ioq.oeq.eg
Figure V 29. EGA of Loading 88
87
-------
CM
in
ca
oj
83-
O)
o
(3
00
IS
O)
c
1-1
E
\
O
(D
0
cnor
c
r*
"D C3
O S
O ID
C
O
o:
EGA Detector Response CmV)
' (Dotted Line)
(Am)
esuodsey wjoq.oeq.en
Figure V 30. EGA of Loading 90
88
-------
S3
O)
CM
p.
CM
mm
89
-------
EGA Detector Reaponee (mV) ;
(Dotted Line)'
C3
,Tfr
CO
c\3
cs
*
CD
en
GO
CM
X. \
*-i CJ
* I »
s
CM <-«
cni
%! (D
4^
cr) o
c o:
0
-J> ..
(D
.« CD
(SI C
GO 0
c, o:
(^iu) esuodsey Joq.oeq.eg
Figure V 32. EGA of Loading 92
90
-------
EGA Detector Response GnV)
(3
CD
O)
csi
CD
CM
ca
-
ca
(S3
CD
CO
E
U
UD
CD .
=?fc 0)
-P
OT 0
C Qi
0 CO
0
-J ..
ffl
i. en
o? c
»-^ o
VO
M
en
cn
esuodsey uoq.oeq.eg
91
-------
EGA Detector Response (mV)
co
cy>
60
s
M-l
O
CO
flj
I
(/\tu) ©suodsey uoq.oeq.eg
92
-------
EGA Detector Response (mV)
1 1 T
o
o
I
(fl
p -
O
Ed
m
CO
>
0)
-
a
H
esuodeey uoq.oeq.eg
93
-------
EGA Detector Response
-------
(Si
in
*«
CM
en
CM
^ 0
-* 8
,-H &
CS1
07
C
H
E
O
(3
*«
-HI .,
I
(D
L
(D
Q_
E
(D
H-
-------
EGA Detector Response CmV)
(Dotted Line)
CM
CO
CM
S3
CM
CT)
IS
E--
O)
CM
CM,
O
O)
CM
O
"3-
CM
CO 3:
»i
C-x .rt
-* E
Xs. %
C3 CJ
»i »
S
O
4^
U"J 0
** DC
I
O
2: (3
<: in
va
..
-------
EGA Detector Response CmV) i
(Dotted Line)
*r
CO
in
SB.
L O
X,
L ..
O ffl
+>
(D O
OTQi
O
L
. c
oo o
L eg
(eui-j
Figure v 39
) esuodaey uoq.oeq.eg yj
EGA of Average Um'wetted Lurgl
97
-------
CD
CD
ft-
CM
in
O)
c\I
E
CJ
; X
> L ..
-a
-------
oa
CO
GO
CM
O)
in
c
rH
E
\
U
t-i O
o:
o
o
CD »-»
o
H- .-.
-------
VI. DISCUSSION OF RESULTS ;
A. Permeability
Permeability coefficients were more reproducible and self consistent
than expected. Perhaps the high pores ( 280 psi) consolidation pressure
contributed to minimum channeling along the wall of the oedometer sheath as
well as to minimizing flow between peds of soil fabric which may produce
variable permeabilities at lower vertical pressures.. Also bimodal
permeation channel development may be minimized by the compaction at optimum
water content often used in producing these specimens. !
During mixing of material wetter than optimum, granulation occurred in
which pellets of material formed which seemed to be wetter on itheir surface
than inside. This seeming tendency toward synergesis probably disappears as
mineral hydrates form during curing but during compaction and 'initial
consolidation the wet surface of the granules may allow smeariing of their
surfaces with development of parallel alignment of any clay-like mineral
platelets present during compaction by proctor hammer. This may account for
some of the observed lower permeability at wetter than optimum moisture
content compared to .a dryer moisture content.
Reduced permeability for compacted specimens of most spent shales at
wetter than optimum moisture content was observed. Figure VI 1 shows this
effect for both standard and modified proctored specimens made from mixtures
of burned TOSCO and unburned TOSCO II spent shale. However, the effect is
weak or nil for 100% unburned TOSCO II material, at four weeks curing time.
Of course some other factor may be influential with these cementing mate-
rials such as increased hydrate formation at higher water contents.
Figures VI 2, 3,. 4, 5, 6, and 7 show a general but not universal mild-
downward trend of permeability at increasing curing times, especially beyond
30 days with the exception of the more cementaceous 70% TOSCO II with 30%
burned TOSCO blend and the Lurgi spent shale which showed increased per-
meabilities beyond 30 days even though the latter had shown a decrease up to
30 days. The lower permeability at wetter than optimum water 'content is
evident again with the Lurgi material in Figure VI 6 when the :22% water
added is compared with the 27% water added curves within the modified
proctor constraint (square data points).
It is difficult to make sense of the sketchy permeability vs curing
time curves of Figures VI 2, 3, 4, 5 and 6. This is perhaps caused by the
irregular course of mineral hydrate formation and disappearance as curing
time increases. Reduction of the mineral grain density (determined by the
Beckman air pycnometer) of pulverized 48°C oven dried material from the
torsion test is assumed to be an indication of the extent of hydrate water
incorporated in cementaceous and/or bulk producing species in ;the specimen.
Figures VI 8, 9, 10, 11, and 12 plot the remarkable course of .the extent of
hydrate water present in the specimens after torsion testing as indicated by
mineral grain density. If a lower mineral grain density indicates more
mineral hydrate water then there is an appreciable maximum in mineral hydrate
water at around 20-30 days as curing progresses with time for iall unmellowed
TOSCO spent shale materials studied including unburned TOSCO II spent shale.
100
-------
10
-4
10
-5
o
Ul
V)
\
s
o
»-6
c.u
2 io-7
cc.
ai
Q.
10
-8
10
-9
% NOTATIONS ARE WATER
INITIALLY ADDED TO MIX
OPEN SYMBOLS - OPTIMUM MOISTURE
SOLID SYM3OLS - V/ET OF OPTIMUM
34%
100 90 . 80
PERCENT TOSCO II WITH BURNED TOSCO
c Pf^ability of Mixtures of Burned TOSCO and u,
prng Severs. ^oxmately Four Weeks Curing
Unburned
101
-------
in -5 ,,,
»' ' 'A3225% ' ' ' lJ T '
D 25 % . TOS 100
? O °A 71 I 1
A «S rt ? 7 °/ ^* ""> *"""--^
A^J^1 I4\^
^v=ir 22% ^~-~-~^^
^ ^ "^"""""*-1 ^. 2 2 °'«
22% , "" ^^^ 4g '
i io-s - ^^0 27% a* _ i
AD;
1 1 1 i i i i i ^ i
0
»
- Figure
IO-S
io-6
0
Figure
Sio-«
V)
^
O
>-
H
3
m 7
< IO"7
'UJ
S
lit
IX
io-8
' (
Figure \
10 20 ' 30 40 50 60 70 ' v 290 300 :
CURING TIME, DAYS ,
VI 2,. Permeability of 100% TOSCO II Spent Shale (TOSCO 100)
i '
\ ' . _
i 7o ' i i i i. r^^-^ 1 1 ;
^ ' TOS 90 i
27%
23% !
23% 39 '
'5723% "---^ -i
53 ^^-^- 23% !
28% ^~---^<»B !
D 27% 23%'
98 97
10 20 30 4° 50 60 70V/240 250/260 270
CURING TIME. DAYS -- S.EEilo"' VsixlO'';
VI 3. Permeability of . 90% TOSCO II - 10% Burned TOSCQiSpent
Shale (TOSCO 90) i K
i
i
i | 1.1 1 [ 1 i
89 8° - TO'S 80 !
^ A
2T% 24% i
N. 24%
X60 :
ssV 2"% ^ r~^_ 36
29%\ . " D
\ 24% .;
\
2g% TRIANGLES STANDARD PROCTOR
SQUARES MODIFIED PROCTOR i
SOLID SYMBOLS WET OF OPTIMUM !
1 1 t | , , ,
3 '0 20 30 40 50 . 60 70 80 ;
CURING TIME, DAYS
102 :
-------
10
-7
10'
10'
~T
58
TOS 70
25%
28%
3 4 %
597625% TRIANGLES STAND. PROC. i
* D SQUARES MODIFIED PROC.
I I I II !
10
70
80
20 30 40 50 SO
CURING TIME, DAYS
Figure VI 5. Permeability of 70% TOSCO II - 30% Burned TOSCO Spent
Shale (TOSCO 70) I
10'
o
UJ
-------
1 \J
ID'S
o
UJ
U)
\IO'S
0
>:
3
5
IT
UJ
Q.
io-8
io-9
I 1 T 1 1 f.{ , , , .
50o '''':-
°8482 *"tf ,
25%LURGI WITHnr-,25% ^
MELLOWED LURGTS'*^^ a5 ^°k/~ :
[__' j ^-* ] cV"k i
LJ ^ /=. . i
40%
40% 87 Mp0 TOSCO
1 93 m iipri BV3^^
0 ^rosco u «» M£UL° I
5L %^ !
\.%t,'
^\ ^3
\86 /> '
4, IO3 I
4^. D IO4 i
^x 40% n :
<0, 45%
<^
^0
\
SQUARES 'MODIFIED PROCTOR ^
^o !
^0
' 1 1 1 L 1.1 1° 1 i
° '° 20 30 40 50 ' ' 250 260 270 ;28<
CURING TIME, DAYS
Figure vi 7 Permeability of Mellowed Lurgi
-------
I
The Lurgi spent shale, however, showed not only less reduction in
mineral grain density after curing compared to the grain density of the
uncured dry raw material but no clear cut minimum and some scatter of the
data points. Apparently a different cementation mechanism is!involved with
the Lurgi material than with the TOSCO materials or else the Lurgi pilot
plant pre-hydrated the material to considerable extent during:post pyrolysis
operations. ;
The interpretation of the minimum in the mineral grain density plots of
the TOSCO spent shales (Figures VI 8, 9, 10, and 11) as due to mineral
hydrate maxima is substantiated by Figure VI 5, introduced later, which
shows maxima in a low temperature evolved gas analysis water peak (at
approximately 150 C) at curing times corresponding to the minima in
mineral grain density. This peak appears to be due to a tobermorite-like
species, CSH I (CaO S^-nH^O) and ettringite.
Figure VI 13, shows the permeabilities found for autoclave mellowed.
Lurgi (autoclave run M 14) mixtures with Lurgi, an autoclavedjburned TOSCO
(M15) mixture with unburned TOSCO II and an autoclaved burned | TOSCO (M15)
mixture with burned TOSCO spent shale plotted vs void ratio e ;for certain of
the spring oedometer specimens. These were all modified proctor specimens
and the water added and curing times are noted by the data points plotted.
The low permeability of the 75% autoclave mellowed burned TOSCO spent shale
mixed with 25% of burned TOSCO spent shale is to be particularly noted in
view of its low cementation to the time curing was stopped. Before con-
cluding that the mixtures containing burned TOSCO spent shale ;giving the
lowest permeability are most desirable in a liner the brittleness of the.
liner must be considered and also its ability to self heal after fracture or
during tension movement. ' .
The permeabilities were determined just before torsion testing, the
void ratios were based on mineral grain densities determined on 48 C oven
dried material after torsion testing and bulk dry density of the specimens
just after loading in the spring oedometers. The bulk dry density was
calculated based on water added to the wet mixture loaded into the
oedometer. I
Figure VI 13 shows that in general the greater the fraction of burned
TOSCO spent shale that is blended into the unburned TOSCO II spent shale the
less the permeability at a given void ratio. Lower void ratids give lower
permeabilities also, however, as is the well known trend for ordinary soils.
Fairly clearly the hydrate forming cementation reactions of burned TOSCO
containing spent shale reduce permeability beyond that to be expected by
simple reduction of void ratio determined using mineral grain 'density. This
may be due to deposition of fine precipitate or gel within the spent shale
particles interstitial spaces or it may be due to deposition at spent shale
particle contacts that grows to invade the interstitial spaces. Growth of
cementaceous hydrates within a given porous spent shale particle should not
Influence permeability of the specimen much. |
B. Peak Angle of Internal Friction P Related to Self Hdaling and
Its Trade off with Permeability '
Figure VI 14 shows a dilemma in trying to compound a liner material
made from any mixture made from burned TOSCO and unburned TOSCJO II spent oil
105
-------
2.64
CURING TIME, DAYS
Figure. VI 8. Mineral Grain density vs Time for TOSCO 100
Spent Shale
2.76
2.60
10
20
30 40 50
CURING TIME, DAYS
60
70
290
Figure VI .9.
Mineral Grain Density vs Time for TOSCO 90
Spent Shale
: 30C
106
-------
curing time, days
Figure VI 10 Mineral Grain Density vs Time
for TOSCO 80 Spent Shale ;
30 40 50 GO
curing time, days
Figure VI II. Mineral Grain Density vs Time for
TOSCO 70 Spent Shale.
107
-------
2.64
LURG!
o
o
\
V)
2
UNMOISTENED
tr
e>
z
tu
o
z
<
IT
O
2.62
= 2.(50 -
cc
iu
2.58
10
20
30 40 50
CURING TIME, DAYS
SO
Figure VI 12. Mineral Grain Density vs Time for Lurgi
108
230
Spent Shale
-------
o
r-
+J
03
o
>
T3
O)
(O
I
O
o
c/1
O)
oo
O)
Q.
OO
O
>>
01
s_
CD
a.
ro
0 3 C/W 3 ' A J. IT I 0 V 3 H U 3 d
109
-------
o
CD
t£
in
m
o
10
CO
ui
LU
cc
IT) (T,
^J" LU
c
O)
Q. d)
CO i
cn«=C
3 C
I O
<*- 43
O O
IT)
I
O
CO
I
o
Ol
O
in
ro
03s
wo
t- ta
_Q a)
«-» 03
r
H-< Q}
> S-
o
OJ
i- CU
3 i
CD rO
r- .C
LL. C/)
O)
Q..
in
o
o
oo
.o
o
4->.
O
<4-
O
^:
-^
03
O)
Q.-
-Q
(O
CU
0)
O.
T3
OJ
4->
fO
'o3
S-
o
<_>
03
WO 'AinieV3WM3d
«* (/)
t1 QJ
it 3
> (->
X
O)
s_ CD o>
3 r-f r
D1 «3 CT>
»- -C C
U- to =C
110
-------
shale. Both low permeabilities and low peak friction angles are desirable
but they seem to be nearly mutually exclusive. A low friction angle allows
easier or more extensive rapid self healing. A high peak friction angle
is shown by the more cemented specimens.
The peak angle of internal friction plotted in Figure VI
14 is more
easily reduced from the torsion data than a normalized peak shear strength.
Also 0 may be directly used in one model of the rapid self healing
process.
For a silty material (as many of the specimens here studied are) the
torsion shear strength measurements give shear strengths proportional to the
normal pressure on the failure plane (which is practically the same as the
vertical pressure on the specimen). This is because the specimens are small
enough so relief of pore pressure of a silty material is rapid and also
because cementation in these specimens is nil. For such specimens in such a
test that is not too quickly performed, the peak shear strength and residual
shear strength data obtained at known vertical pressure can be reduced to
jZL and 0_ respectively.
C K ' . !
Many of the peak friction angles of the more cemented specimens
measured are not as high as they should be because the looser, less
compacted, specimen material near the top pore stone tends to I fail before
the more representative center of the specimen. Thus the dilemma in trying
to attain low cementation as well, as low permeability is probably more
serious than even indicated by Figure VI 14. Also probably few if any of
the peak friction angles determined by the present torsion apparatus are as
high as they should be for the more cemented specimens due toid a more
complicated and hazardous case must be considered.
Ill
-------
Before the liner shears, and therefore before the soil skeleton in the
shear zone might be compressed (if its void ratio is above the "critical -
void ratio" for the material), the peak shear strength may be used to esti-
mate the maximum angle of inclination of a liner below a spent' shale pile
before the pile will slide through shear failure in the liner parallel to
the liner. But if the liner does begin to shear at some place due to
the peak strength being locally exceeded, and if the liner material does
contract during shear (the "critical void ratio" of the shearing liner being
lower than the emplaced liner void ratio), and if pore water pressure can
build up in the plane or zone of failure, trouble is probable. : Pore pressure
will reduce the effective vertical pressure '0" * and reduce the effective
peak shear strength TL,' of the section of liner beginning to shear. The
reduced strength of tne initially shearing section and the strain occurring'
shifts the stress toward the remaining intact section of liner and its
strength may also be exceeded. Thus a "progressive failure" occurs which
results in a slide.
The most conservative design strengthwise (and possibly the most
expensive) is (1) to design using the weak strength developed after shearing
occurs with no drainage where the effective normal stress cr is reduced and T
is also reduced (since T = a cos 0) or (2) to design with the low residual
shear strength T _ which is developed as the thrust of the failure shear
plane increases *and any cementation or particle to particle interlocking is
broken and any clayish mineral species platelets are aligned parallel to the
plane thus producing a more slippery shear plane), whichever i;s lowest. It
was beyond the scope of this present work to test the true residual shear
strength of the many materials and conditions of emplacement here surveyed
toy the ring shear apparatus ;
The change in void ratio from emplaced void ratio to the critical void
ratio at failure has not been measured in the torsion apparatus used in this
study. Probably it is best measured (for plastic specimens) in a
conventional "triaxial" soil testing apparatus in which the volume of the
specimen can be measured throughout a test. Attempts to make this
measurement on cemented specimens will probably be futile, however. It
should b'e done on good candidate (non cementing) liner materials. The
change in void ratio at critical conditions (where shearing, particularly
simple shearing, has caused the specimen to reach a steady state) helps
determine the pore pressure during failure and its degree of influence on
the shear strength during failure. ,
D. Brittleriess Index Related to Cementation and Permeability
In calculating the brittleness index two shea"1" strength values T and
c, mustnot only be normalized to the same vertical pressure, but good
a, and
-------
0.3 r
74
D
0.2
x
UJ,
Q
z.
V)
in
UJ
2:
UJ
o:
en
0.1
0
-0.1
MIXES WITH MELLOWED MATERIALS,
MODIFIED PROCTOR
A TOSCOS 8 LURGl, STD. PROCTOR
D " " " MOD. PROCTOR
-O.2
_L
35 40 45 50
PEAK INTERNAL FRICTION ANGLE,^.DEGREES
Figure VI 16. Brittleness Index Correlated with Peak Fricti
113
55
on Angle.
-------
Figure VI 17 is a plot of permeability vs brittleness index of
specimens made from TOSCO 70, TOSCO 80, TOSCO 90, TOSCO 100 mixtures con-
taining mellowed materials, and Lurgi spent shale raw materials. Difficulty
in selecting materials giving both low permeability and low brittleness
index from materials giving curves on the right of the plot is evident.
Some materials of the curve on the left of the plot are, on this basis,
perhaps acceptable for a liner, however. These materials give a negative
brittleness index calculated from the quick torsion shear strengths
0p and 0R. ;
i
The cause of a negative brittleness index seems to lie in low perme-
ability along with neglegible cementation. Low permeability of a specimen
in the torsion test is believed to prevent rapid drainage at the beginning
of torsion with development of appreciable positive pore pressure as speci-
men distortion and contraction occurs, which reduces the effective vertical
stress a . As twist precedes, the excess pore pressure drains, and
a increases causing T to rise. The result is an increasing torque vs
tY.me plot on the x-y recorder of the torsion test machine.
For .screening candidate liner materials the ability of the torsion test
to indicate low permeability, non-cemented, simple shearing materials seems
useful. Whether the added ability of the torsion tester to transfer con-
solidating curing specimens from oedometer to torsion tester without much
sample disturbance is essential is perhaps debatable. It seems, however,
that every opportunity should be given the specimen to demonstrate any small
extent of cementation it has achieved during aging. Disturbance would tend
to break the cementation before testing thus reducing the peak of the
torsion stress strain curve. i
i
E. Relation of Peak Friction Angle 0 and Brittleness Index
BI with Initial Torsional Stiffness and Shear Modulus G
Figure VI 18 is a correlation of peak friction angle 0p with simple
shear modulus G obtained from the initial torsional stiffnessi derived from
volts/degree twist on the stress strain plot. There seems to!be a higher
curve for brittle material and a lower curve for soft material.
i
In comparing these G's with others,the increase of G with confining
stress (some 280 psi a usually existed here) must be remembered. Some of
the points plotted in ₯igure VI 18 no doubt correspond to specimens with
soft tops which too easily sheared and cause displacement of the points
downward. Soft tops are caused by poor emplacement of the top pore stone
and vanes into the proctor compacted specimen during loading of the spring
oedometer. ' !
, i
Apparent higher values of these slow shear moduli for more cemented
specimens, suggests that perhaps rapid shear moduli from resonant dynamic
tests might be used to nondestructively periodically assay a given curing
specimen for cementation. The very small samplea needed for an EGA assay
suggest this might be periodically done on the specimen also Without much
altering its integrity for the dynamic test. Perhaps the specimen should be
kept under high vertical or high confining pressure to simulate burial even
during dynamic G testing. Perhaps it can be shown that proctor compaction
is adequate to simulate this and such pressures are unnecessary. With
proctor compaction the persistant problem of weak specimen tops would .occur.
114
-------
-o.i
0 O.I
BRITTLENESS INDEX
0.2
Figure VI 17. Permeability Compared with Brittleness Index
115
-------
3OO
tO
(L
Vs-
X
10
2
111
H ZOO
CO
EC
<
LJ
:c
o
lot
a.
100
A
TRIANGLES STD. PROCTOR
SQUARES MOD. PROCTOR
A D TOSCO 100
A H TOSCO 90
A B TOSCO 00
A m TOSCO TO:
T ^ LURGI
O WJTH MEL.FRACTIONS
1
I
10,000
20.0OO 30.OOO
SHEAR MODULUS G.PSI
4O.OOO
5O.OOO
Figure VI.18. Peak Friction Angle Compared with Initial Shear
Modulus G '
116
-------
Probably static compaction at the high vertical pressures of interest would
simulate a real liner adequately.
Figure VI 19 plotting BI vs shear modulus G also suggests a distinction
between brittle and soft material. Here also some points are no doubt too
low. "
F. Relation of Peak Friction Angle with Twist at Peak Strength
Figure VI 20 shows the peak friction angles of the TOSCO 100, 90, 80
and 70 series and the Lurgi material as a function of twist produced during
torsion testing at peak strength. There seems to be a trend for greater
twist (or strain) before failure for the less compacted standard proctored
material than for the more compacted modified proctored material, especially
for material giving lower peak friction angles. Greater strain before
failure is advantageous in a liner. This does not necessarily correlate
positively with tensile strength, however, which has little relation to
strain before failure. Tensile strength is more related to shear strength
where a rule of thumb says that it is about. 1/20 of the shear \ strength of a
cohesive clay soil.
Also plotted in Figure VI 20 are points for loadings 86 and 92 for a
75% mellowed burned TOSCO - 25% burned TOSCO mix and the point for loading
87 which is of a 50% mellowed burned TOSCO - 50% unburned TOSCO II mix. A
large extent of strain before peak strength at low peak friction angles is
possible with these mixtures involving mellowed material. In !fact this sort
of material with low (in these cases negative) brittleness index tends to
exhibit simple shear or zone shear and larger twists than those plotted may
be more appropriate for this plot since for these materials peak strength
does not necessarily imply failure. :
G. Relation of Peak Friction Angle and Squashiness With '
Cured Void Ratio !
i
j
Figures VI 21, 22, 23 and 24 show a regularity in peak angle of
internal friction plotted vs void ratio for cured TOSCO spentishales, TOSCO
100, 90, 80, and 70, as the fraction of cementaceous burned spent shale
increases. Greater void ratios (determined after curing) are'associated.
with lower peak internal friction angles for the TOSCO spent shales. This
suggests that a liner made from spent shale should be as little compacted as
possible so a lower friction angle and resulting greater squashiness is had
which allows better self healing of the rapid type, other factors such as
not too much shrinkage at a shear zone being acceptable. The|squashiness is
inversely proportional to the shear strength which for a silty draining
material is cr tan 0p.
H. Hydrate Species Determined by EGA i
In the-early stages of curing of the series of mixtures.,TOSCO 100, 90,
80 and 70, rise, and fall of an EGA water peak at 115 to 135 C was found.
Since both tofaernite and ettringite may manifest themselves iiii this peak
some ambiguity exists without x-ray diffraction or other means of distin-
quishing between them. Figure VI 25 is a plot of this peak curing time for
crumbs of spring oedometer specimens after torsion testing. Timing of this
peak is similar to that in development of strength of 0 in Figure VI 26.
117
-------
0.20
0.15 -
'x 0.10
U]
o
to
Ui
ta
z
ffl 0.05
-0.05
nA TOSCO 100
gj ^TOSCO 9O
3 A TOSCO so
a A TOSCO 70
OV LURGI
SQUARES « MODIFIED PROC.
TRIANGLES-STANDARD PROC.
_L
10,000 20.0OO 30,000
TANGENT SHEAR MODULUS 6, PSI
Figure B-VI 19. Brittleness Index Compared with Initial Shea;r Modulus G
118
-------
60
lu 55
tU °°
CC
(.S
LU
50
LU
-J
! G
o 40
LL
tu 35
cu
30
WITH MELLOWED
TIP
SYMBOLS SAME AS FOR FIG, B-!VI 18
P DENOTES PERMEATED SPECIMEN
JL
JL
1
10 20 3O
TWIST AT PEAK STRENGTH, DEGREES
40
Figure B-VI 20.
Extent of Twist at Peak Strength Correlated wfith Peak
Friction Angle ; ,
119
-------
S50
111
-------
91 TOSCO 100
D
33Q
I
10 20 30 40 50
CURING TIME, DAYS
60
70
80
Figure VI 25.
EGA Hydrate Water Peak of Tobermorite and Ettringite
vs Time for TOSCO 100, 90, and 70 Specimens
88
10 20 30 40 50
CURING TIME, DAYS
60
70
2?»0 i 300
Figure VI 26.
Peak Friction Angle vs Time for TOSCO 100, 90, and 70,
Specimens, :
121
-------
Another EGA water peak at around 225°C becomes prominent'. with TOSCO
70 material. This is believed to be where the "carbonate ettringite"
(ettringite with sulfate replaced by carbonate) manefests itself as an
additionalQpart of its EGA curve. The main part, however, still is at the
115 to 135 C peak as for ettringite itself. There is a sudden appearence
of the 225 C peak at a sharp threshold with a burned spent shale content:
below 70%. This suggests that the pH of the water remaining in the mix
and/or the carbonate ion concentration level left after some of the
alkalinity and carbonate has attacked the unburned TOSCO II spent shale
component of the mixture may determine whether ordinary sulfate ettringite
or carbonate ettringite is formed. The identification of this peak is
substantiated by its reduction in side experiment curings where some gypsum
as a source of sulfate is added, its accentuation when Na-CO-|is added,
and its elimination when Bad- or Ba(NO-)0 as a sulfate scavenger is
added. 2 3 2 i
o Figure VI 27 is a plot of the initial shear modulus G vs:the 115 to
135 C EGA peak height for several of the spent shale mixes aflper curing.
Although there is some rise in stiffness at the highest quantities of
tobermorite and ettringite found, the effect is not very strong and moreover
for low peak heights which are in the low cementation region of most
interest the effect is nil. Thus study of cementation of this sort by
dynamic G testing does not seem straight forward. It must bejrecalled that
ettringite is not very cementaceous compared to tobermorite arid we have not
yet analyzed the 115-135°C EGA peak for these.
i
. . I
Figure VI 28 shows strong inverse correlation of twist at peak strength
vs the 115 to 135 C EGA peak height. Several high data points are
probably the result of slippage between specimen and upper pore stone. Any
cementation produced by these hydrates seems to operate against extensive
deformation, the action of interest in self healing, rather than against
small deformation involved in initial stiffness.
The peak strength friction angle is quite dependent on the extent of
formation of cementaceous hydrates under conditions of the tofsion test.
Peak strength also correlates fairly well with the void ratio:within types
of spent shale mixtures such as TOSCO 100, 90, 80, 70 or Lurgi. A better
correlation with strength than either of the above should be obtained by-
plotting the family of void ratio vs peak shear strength curves with the.
amount of tobermorite water present as a parameter (ignoring any ettringite
and hydromagnesite cementing action). The tobermorite might be determined
assuming all ettringite is the kind with also a peak at 225°C las well as
at 120 to 135 C (carbonate ettringite or hydroxy ettringite but not
sulfate ettringite). The early peak at 120 to 135°C or so (depending on
its height) results from both tobermorite and ettringite largely super-
imposed. The tobermorite peak area alone might be obtained by subtracting
out a calculated carbonate ettringite area for the 115 to 135 C part of
its water evolution calculated from a ratio of the 225°C part :of its water
evolution. .
I. Secondary Compression Index Cq related to
Cementation arid Mellowing :
Figure VI 29 summarizes Co, the secondary compression index of the
spring oedometer specimens just before the specimens removal from its LVDT
122
-------
O3
- 30
3 20
o
o
s
5 15
Ui
X
.0)
H 10
UI
o
z
** e
1_ 5
O
1 I I
..
86
Q9I 92 °
" D 33 0
rv 56
47 g "0 - -0
- Ol32 90 ^D"
a 35 ^49
- .
D
O
A
V
O
i 11
i i
58 59
40 A/V- -
/vv_i
^-V^ A 68
43 /^ /\b°
^j x'
\y >^ /\ fl fl
/\/ ' ^
X^
"O""^ ~
si 53
31
_
TOSCO 100 "~
TOSCO 90
TOSCO 70
LURGI ~
MEL.TOS 75%,TA 25%
1 |
234
II5-I35"C EGA PEAK HEIGHT
Figure '" VI 27. Initial Torsional Stiffness vs 115 to 135°C EGA Peak
Height . ;
. EXCESSIVE
SPECIMEN TOP
TO TOP
SUP?
SYMBOLS AS FOR FIG. B-.VI 27
1234
115-135° C EGA PEAK HEIGHT
Figure VI 28. Twist at Peak Strength vs 115 to 135°EGA Peak Height
123
-------
height measurement system prior to permeability determination : (if made) and
torsion testing. In general some specimens showed a uniform height vs log
time plot while others showed a fairly sharp downbend in this Iplot at around
of 1 to 5 days as though some kind of friction of cementation ^were broken
loose then in response to a particular environmental change. ;The nature of
these possible clock reactions has not been considered here but should be at
some time for whatever type of material might be considered a \further
possible candidate liner.
i
Figure VI 29 shows that at a burned TOSCO spent shale content of about
10% in TOSCO II spent shale shale a minimum Ca occurred for specimens
showing a given peak friction angle. Figure VI 29 is a cross 'plot of the
data shown in Figure VI 30, 31, 32, 33, 34 and 35 for the last Ca from the
LVDT measurements and the peak friction angle later determined by the
torsion tester. The scatter of the data in these latter figures is believed
to often be due to premature slippage of the specimen at its weaker top
rather than failing in a more representative portion in the middle of the
specimen. The "best" curves "have been drawn through reasonable higher 0p
values of presumed better failed specimens. !
It should be commented that none of the Ox values seen to date seem too
high for use of a liner for a period of 10,000 years. There is some pos-
sibility, however, that the same secondary compression vs log ;time curves
may turn downward even further were specimens studied for longer periods.
The secondary compression index Gx, that can be calcualted from
Townsend and Peterson (1979) data for their unscalped TOSCO II spent shale
for modified and standard proctor material at 10 minutes consolidation.
and interpolated to a a of 280 psi are about two or three times the
values we have found (Figure VI 30) . These higher Got values would be
explained if much of the larger chunks of spent shale in their unscalped
sample soften when water soaked so the bridging between them is weakened at
their point to point contacts. The 0 values reported by these authors for -
standard and modified proctor are roughly comparable with 0 measured with
our torsion apparatus although a number of variables are important to the
exact values which are found.
J. Indirect Tensile Strength (Brazilian) Tests :
l
In Figure VI 35 Brazil indirect tensile strengths of specimens
compacted and cured with no confining pressure are compared with the twist
at peak strength from the torsion test of counterpart specimens cured in the
spring oedometers at around 280 psi vertical pressure and tested near that
pressure. i
There is a trend in Figure VI 35 for more cementaceous TOSCO 80
material to give greater tensile strength than less cementaceous TOSCO 90
material and for it to give greater tensile strength than the | TOSCO 100
material. Lurgi spent shale gives relatively high tensile strengths while
mixtures including mellowed material as noted give low to medium tensile
strengths. The negative slope of the trend-lines drawn for the different
materials suggests that the kind of tensile strength measured'by the
Brazilian tests is a brittle cementaceous type strength associated with
rapid attainment of peak cementaceous shear strength as strain progresses.
124
-------
'
s
^J
< 0.001
z
ik
0
1 1 !
C^ AT 45" efD
O
D
O
0 ,
-------
I
o
i
O
\ . * t I i 5 i
CO
UJ
K
stss
<9
Z
UJ
a:
0
CM
UJ
Q.
07
i
O t-
~ 05
U
O
w
a:
o
H
C 0) 4->
>- I D>
O i-
C/> -^ +J
l/l CO
(1) i.
i O -^
i I rtJ
UJ O)
s= (o a.
a) re
i- 3 +->
N r (/I
«3 T-T-
i- IV3d J.V 1SIAM 1S3J. NOISaOl
CT)
QJ
S_
C/)
(O -t->
QJ en
Q- c
o>
+-> s_
to -t-J
cn QJ
5^
c
O)
C N
O BJ
cn en
to
CO
OJ
i-
01
iZ
126
-------
This slope is opposite to that expected for a correlation of the ishear test
twist and tensile strength for a cohesive plastic material. I
For the silty and sometimes cemented materials of most of the specimens
studied here high tensile strength primarily indicates cementation. A more
clayish material might show both high twist before peak shear strength and
relatively high tensile strength. Tensile strength, as weak as it is, in a
material at the overburden depths involved for the liner being considered
does not seem to be an important consideration in the face of the high
vertical stress involved at depth which greatly increases shear strength.
Figure VI 36 is a plot of the tensile strain at failure with the Brazil
tests vs the twist at peak strength for the torsion tests. It is; concluded
that: only a little cementation reduces the tensile strain at failure
considerably. The largest torsion twists (strains) of cemented specimens
and less cemented specimens at peak torsion shear strength seen are about
the same but the Brazil test tensile strains at tensile failure are much
greater for less cementaceous material.
Figure VI 37 shows some classical shaped tensile strength vs! water
content curves for specimens made from mixtures involving mellowed
materials. Even for the 75% mellowed TOSCO -25% burned TOSCO specimens
which are undergoing some cementation, a peak tensile strength vs: water
content is observed. The points corresponding to specimen water additions
used in the main series of oedometer - torsion tests are indicated by the
oedometer loading number of this main series. According to data ifor these
little aged specimens, optimum tensile strength was not usually attained.
But as observed above, the tendency of the liner to shear at higher
vertical pressure which will far over-shadow the weak tensile strengths.
i
K. CompressibilityCoefficients ;
Pneumatic arm oedometer tests were made on standard proctored or
modified proctored specimens one inch high and 2% inch diameter. > Table II 1
lists the types of mixes studied by arm oedometer. Table V 4. is a summary
of some results and some calculations. In the first column t sighifie *"
TOSCO, m modified proctor, s standard proctor, etc.
Primary consolidation was finished after 80 minutes for the JDne inch
thick double pore, stone specimens used. For this time the total compression
of the specimen from the data collecting computer print out for ekch loading
increment was read and entered in a computer spread sheet program', Table VI
1 column D, as steady state Schaevitz units. After converting thjase to mm
compression and correcting for arm oedometer apparatus deflection; at the
particular load (column C) the net compression was calculated (column F) for
that; loading increment. Load increments producing vertical pressures on the
specimens of 19.4, 38.8, 77.5, 155.0, and 310 psi were used. From the
initial water added, mineral grain density of the initial dry mixture before
wetting, specimen loaded-weight, cell volume of 80.44 cm and specimen top
surface area of 31.67 cm the dry densities and void ratio for eajch
loading were calculated and for each change in load the delta void ratio and
compression index C were calculated (columns K and L). :
From the data calculated in Table V-4 a variety of correlations may
be made. Void ratio e plotted against consolidation stress (load) gives
127
-------
curves typical of silt. Figure VI 38 shows the compression index for the
last loading increment (155 to 310 psi) for each specimen, plotted against
the initial water added to the specimen in mixing it. There is much
similarity in the position of the curves for the various types of spent
shale. I
128
-------
o
*
cc
UT
o
to
H-:
z
i
OJ
-C S_
E LO
tO C1J if)
CD E I
S_ !-
Q. O +->
E
-------
VII. CONCLUSIONS AND RECOMMENDATIONS !
1. A softer less brittle material after placement seems desirable for
most of the specimens tested. Apparently the angularity and harshness of
silt and sand sized particles in most mixes is the cause of a relatively
high angle of internal friction for the peak shear strengths and'residual
shear strengths found after yielding. Such high strengths do not: seem
necessary nor desirable and should be traded off for lower angles of
friction through less compaction, some heap mellowing time whichiwill allow
more particle surface roughening and floculation and looser compaction while
retaining some measure of swelling and shrinkage stability, and/or addition
or generation of some quantity of clayish material. The latter can
apparently be made from certain spent oil shales by autoclave mellowing but
it -may be more economical to add clay from other sources. Trial'of further-
autoclave mellowed or other mellowed materials seems desirable, i
2. Trial of addition of clay or other similar fines to reduce
permeability in the case of the TOSCO II spent shale or the average Lurgi
spent shale is desirable. This could be an added component or could be
generated from autoclaving an especially active burned spent shale such as
the burned TOSCO material. '
3. Ring shear tests should be made to getT at large displacements
for several examples of candidate liner material.
4. The cementing characteristics of burned spent shale seem to be a
detriment to self healing as any shear movement needed for closure of a
vertical tension crack by "caving in" of liner material would produce shear
plane separated fragments. The planes may be possible water channels'.
Moreover-the depth into the liner away from the tension crack for a source
of fill material will be less for high shear strength liner material.
Extrusion of a non brittle plastic liner material, into a tension crack, on
the other hand, should not produce such distinct planes. Autoclave
mellowing inhibits cementation of materials. :
i
5. The high friction angles observed in most of the liner specimens
tested are useful in that there will be less tendency for a pilejof spent
oil shale founded on a liner to slip down a valley. Silty sandy;materials
producing high friction angles are generally rather permeable and uncementsid
specimens of the materials here tested are no exception. To reduce perme-
ability clay sized material can be mixed in or a non strength producing fine
precipitate or colloid within the interstitial spaces of the silt grains
might conceivably be internally generated. In this way a synthetic boulder
clay having both low permeability yet a reasonably high friction;angle might
be produced. . i
6. Quick clay inadvertently made in any of the above ways should be
avoided. Even a material that has only a little above critical void volume
should be suspect until proven unlikely to soften or liquify when subjected
to slip or earthquake produced shear. Even though a liner material in
unsaturated condition may not soften or liquify the same material when
saturated may soften under shear strains. This must be predictable and
eventually probably must be studied for any candidate liner material.
130
-------
7. The "final" secondary compression rates or Cafor none of the
materials, when measurements ceased just before torsion testing, was ex-
cessive. Rates below 2% settlement at 10,000 years by extrapolation were
about as large as observed. Much lower rates were more typical. However
since the settlement vs log time curves plotted were often concave downward
specimens should be followed for longer periods and chemistry of the
apparent softening with age ascertained.
i
8. The effect of additives on spent shale should be further studied in
a more methodical experimental design. Gradations of mixes between silty
spent shale and clayish material should be tested for the following:
1. Permeability '
2. Shear strength by a quick method relative to the permeability
so any softening due to reduction of soil skeleton -volume
during shear is observable. ;
3. Shear strength of both saturated and unsaturated material
should be studied. ;
Well mixed or pugged materials should be used or fines should be .generated
internally. . '
\
9. Possible methods for in situ fines generation include th'e
following: ,
a. Mix two slow precipitating liquid intereactants in ;with the
spent shale. . i
b. Mix in one liquid reactant which reacts with the spent shale
itself. ' .
c. Mix in a solid reactant which reacts with the spent shale.
The spent shale base for the mixture should not be burned spent shale
to avoid the cementation already demonstrated but one such as TOSCO II
material. Perhaps mildly autoclave mellowed burned spent shale would not
cement even though much ettringite may be initially formed.
i
10. Periodic non destructive resonant measurement of shear !modulus G
may be valuable for following development of any cementation in a given
specimen as curing or aging precedes. EGA and to a lessor degree x-ray
diffraction may also be done on the same specimen without affecting G since
such small samples are required.
11. Experiments with physical models of liner materials should be made
in which a tension crack is induced in brittle containment strata and the
ability of liner material to suppress water flow as the crack opens is
measured. It is desirable to be able to perform meaningful experiments of
this sort without the need for continuous high vertical pressures during
aging of the liner. Strong forces at the time of, say, flexing or
stretching a model to generate a tension crack would, of course Tpe necessary
however. Proof is desirable that proctor or other compacted liner material
in such a model approximates a real liner aged with considerable I overburden
pressure. !
131
-------
VIII. QUALITY ASSURANCE AND QUALITY CONTROL , :
A. Objective ;
I -
The overall objective of the quality control and quality asisurance
program was to assure that the data obtained were of known quality and integ-
rity that would permit valid scientific conclusions to be made regarding the
potential for using retorted oil shale as liners below retorted shale disposal
sites. This study was exploratory in nature and intended to proyide a general
assessment of the potential for using new materials (mellowed and burned retort-
ed oil shales blended with standard retorted shales) in a unique^ new manner
(liners below several hundred feet of retorted shale) for long term stability
(centuries). No standard methods exist to perform such as assessment. Therefore,
some of the methods used, as well as much of the equipment and test procedures,
were developed under, and are original to, this effort. Therefore.an important
QA objective was to assure internally consistent and reproducible results
permitting valid conclusions to be made. ;
B. Activities \
Since the experiments described in this report are of an exploratory and
unique nature, the general QA/QC procedures involved internal correlation, cross-
checking and duplicate testing as required to provide internally self consistent
results. This was accomplished to the degree required to assure that the trends
were internally consistent thus indicating reproducible results.: Table VIII-I
provides a summary of the most significant QA objectives, methods, and results.
The following discussion provides an overview of some additional, QA activities.
1. Compaction and Dry Density Measurements
Due to the somewhat "non standard" diameter of the test speciments, a
miniature proctor hammer was constructed for sample compaction. | Compacting
efforts were conducted at 656 kV/m3 for the standard proctor and 2929 kV/m3 for
the modified proctor as described in section IV-A. Since specimens used were
of non standard size they were checked, against specimens of standard size to
assure that they were comparable.
!
The apparent dry density was obtained by weighing all the damp mixed
material added in lifts for proctoring. Care was taken to avoid evaporation of
the moisture. The volume was calculated from the internal diameter of the
specimen sheath and the measured specimen height. Linear measurements were
made to the nearest 0.025 mm and weights were determined to the'nearest 0.1
gram. i
132
-------
2. Mineralogical Species Identification ;
Mineral species were determined by X-ray analysis using spiked standards
of known quantities of known mineral species to provide diffraction peaks
for calibration. Cross correlation was established through differential
thermal analysis and evolved gas analysis (DTA-EGA) as appropriate. Finally,
as an" internal check, a material balance was performed. |
3. Soil Mechanics \
Soil mechanics tests of consolidation, zone shear strength,;cohesive
intercept, and brittleness index were made on molded compacted specimens of
fine-grained spent shales in specially designed and built zone shear cells.
During the consolidation period the movement of the anvils were measured using
linear variable differential transformers. The transformers werfe calibrated
against displacements produced by a micrometer to the nearest 0.025 mm. Con-
solidation time was measured in days. After various periods of ageing time
samples were zone sheared to obtain the zone shear strength, cohesive intercept,
and brittleness index. The proving ring of the shear fixture was calibrated
with the load cell of the Instron testing machine. The load cell of the Instron
testing machine was calibrated with dead weights. ;
C. Accomplishment of.the Quality Assurance and Quality Control |
Objectives -i
The quality assurance and quality control objectives were achieved as
evidenced by the internal consistency of results. Further, the data is consis-
tent with the anticipated behavior of the materials based on their physical and
chemical properties. Based on the QA/QC procedures used herein, this data is
valid for making initial assessments regarding the utility, design, and potential
for using retorted oil shale as a liner below retorted shale piles. However the
results are not intended for use in regulatory decisions, litigation, or design
of specific retorted shale liners.
133
-------
UJ
o
AND PERFORMAI
»»
ASSURANCE OBJECT
^*.
1
cr
"
i i
UJ
_i
ca
i
RESULTS
o
CJ
o-
. ^
LU
>
*
i-
(_
u.
^"
e£
C
<
c
1 1
1
1
d
1
»*
o
+ 1
e
j= e>J
4-1 1
C *
41 _
4-1 1
in >
in t i
c
O 41
:i
t_ i
s«
u
in c
4_> 41
, t-
3 41
VI 14-
41 41
ce. 02
o
41
*J
U
41
"a!
in
C
o
in
4)
in
>>
iO
c
10
.
4) in
4-1 C
SJ
* o
CL 41
3 O.
o in
.
c
0
£
T3 in
41 O
in Q.
3 e
O
IO
each raw mater
<\ and consistent
41
^ "
in .
41 1
tf. ;»
in
c
41
JB
o
41
Q.
in
41
4-1
10
U
"a. en
3 C
O f»
41 in
t- 41
19
<_
IO
<5 H
=» lO
27
a"
in
4-> 41
i t_
3 3
in 01
41 «-
a. u.
!>
i selected specimens
'Stems and compare
>
CU 7^ VI
<- VI 4->
IO
CL.C 3
41 4-> m
C. O 41
Q. JJ t_
O
u
o
t, 41
Q.M
41 V)
N
in
3
"S"-
10 .C
13 4-1
«'!
So,
4->
C iO
o m
C 41 -0
t. 0
41 u J=:
J= O 4->
4J O 41
Assure that
compactions
(standard)
en
ind conclusions are similar
values vary slightly due to
;er1a1s and methods. See
references and specific
5 Townsend & Peterson
7.
£?iH a u en
3 v- 10 "-> 3 1O
in o i- i » vi CL
41 41 *-
u CL4J =» in e
in c: CO
, 41 C 0
10 .C U 0 "-
c. en 41 t- 4J Ol
41 3 <4 *J IO r-.
C O *- 0 4J O
41 -C i- 41
O 4-J *O I/) O *-*
in .
c in
41 41
U
(_
IO C.
i 41
11
in
o5>
4J
41 "S
I- 4->
10 1.
Q. O
E 0.
O 41
0 L.
4J
IO
in
c
o
i- 4)
in j=
1j .c in
C 4-J «
o .- n
in 4l u
4J 41
r- t. »-
in 10 4->
41 C
t. 41 41
4** *^' O
10 10 in
5 o -o
in 41
41 10 4->
t Q) Q.
3 U 41
in o
in c u
<~ -0
*
'
strength changes in
transfer were avoided.
n Section IV-D
Dimensional and :
specimens during
See discussion i
L.
o in
^y 41
B is's t-"
41 > C 4J 41
14- 4J O 3 in >*-
O 10 3 m
3 c 14 * c:
0^ 1 0 «
O iO O - «-
£ > c m 4J
.j 41 ^i O » *
4) *- 1 U
£ O 4J 4-1 > O
4J lO 10 II *-
a. «-
uj^-^r 4i-a
(/) ^> 4J C C« 41
z r-* cn o 3 m
«i 0% c -f- en 3
co ' u 10 u. in
4-1 -O 3
5 o c- o a! fc-
0 » w > 2
41 U« 41 C O CL
in J= O ja Q.
i3 =o in o « 10
i.
c o
i 5 i»i
4J _*
c c
21 o s
in 10 u
J3 41
C «- 0.
4i 3 in
O -r- 41
41 *O C.
CL 3
vi 4) o en
c "a. 41 ~-
i- E J= 4J
to 4>J in
in in 4)
41 en*4>
en >, c
< "O IO *-
O 41 O X
in f 10
54^§C
41 o cn o
U C =f-
3 .r- in
in 41 > <-
in t- o O
< 10 E 4J
in
134
!_,
41 C 41 W> Or-
f O f 4-> VI C IO
4H»>r*4JC:bv4l < C.
VI4I4IVI4IE4I
§l.U_VI4J4IJ=>
OO414lf4.>3O
t. 4-> I. E 4-> O
14 in a. o "O -C 41
4-> -a e: e in in
VIIO^-^41010 O
4J -r- 3 1 O VI i VI
,_ X =» -ain-ooa)
31041 E^mO »--
in-r-^4)C-CIVIJ=4-14J
41 U 10 3 >,* 3 t.
uwmn ^-^-gjaaj
4j r: .c >> 3 4-», 41 «
T3 41 r CH IO J3
10 § |2 u «. t o^~
.tlj 4JI4-4I 4IC4-)
i o m «<- ( f _o c
^ cn in i 4-1 o in **§
2 H a) * in "o.S o t- t.
10 CL4-I t_ 41 3 10 <- iO 01
H- m CQ t- o * O > 10
1 C- O 1
4J E 41 in T3 *-
4I>--
4J 4J -O VI 41 M- C
41 41 t. >>4J E 4>1 C.
4J 41 IO O £ IO
in , L. 4-> T3 O
aj j^ in 4J O 4J 41
14 41 41 O O O 41
« "''«,: r= * S .1
^ e CL4-> CO (0 S U
J>
ex
C 0* >i
* 1 0) O **
fQ S > U U
^- O O» «
t_ O Q. C
OJ 0) XJ O O
(0 C rtS *
41 41 C 4-> «
io 4J t- 4) *
"42 a,^"0 b
73 in u u 4J
41 10 41 O iO
4J W ^ I* E
5 -C C 4J O «
4-> O ^41
4) cn*^> .c 4^ -C
u c in cn 41 4->
3 4) in 3 E .
vt t- a> e> <-
in 4-> i. J£ "O *o
«< in CL4-I 41 14-
10*
snts the overall experimen-
ting intermediate values
shale mixed" with unburned
igeing) times, moisture
iunt of mellowed shale used
[he extensive use of graphs
(perlmental results demon-
>ults are consistent and
41 O/41 3 '-
«_ CiC 01 ^ «J ^
°""" : 3 »- e c 10 10
1 Ol' 3 4J 4J
_ n_ o ^a m
vi'o 4-> 3 in
41 1 » C 41 41
4)T3ia* 41 4) JC 4J
S *- ""3 ^ ^" *-
10 lo'o-c: o c 04^
I 4^14- in u 4-> in
3 : J
.1 S! o 1 S
in 3 4-1 !O vi
4) "Oi i-
o Oi in in in
I.I4J = C
i i*f- 41 IJ
r- O< O 4->
IO 4-1-0.4-1 41
U iO t
c: 41 CL 10
4-* IO 4J
in , : 10 4i -o
^- CL-O J= 41
1041 <-
4J (O 4-> 4^ 4J
tn 4^ 10 10 in
c .e ja
10 41 -0 4-> O
41 4) -F- c
41 CL4J t. 41
5! X = 4) t-
rs 41 »- > 4-»
!
;? !
c ,
t* <
c
41 41
in 6
-i
3 Q
. t.
41
J= 13
' 5 41
41 in
3 VI
in c
in O
< 9
i
i
41
U
=1
0
CL
L.
-------
IX REFERENCES
Culbertson, W.J., Jr., C. Habenicht, R. Nye, F. Bonomo, E. Barrow afcd C. Ruff.
Fifth Progress Report: Development of Liner Materials from Spent Oil Shale.
Some Spent Oil Shale Properties and Development of a "Torsion Triaxjial" Testing
Procedure. Under Cooperative Agreement CR 809233 by Denver Research Institute,
Chemical and Materials Science Divisio-, Univ. of Denver, Nov. 22, 1983,
unpublished. , .
Insle-3 O.G. , "Soil Chemistry Relevant to the Engineering Behavior jof Soils"
in I.k! Lee, Editor, Soil Mechanics Selected Topics., American Elsevier
Publishing Co., Inc., New York 1968. ;
Krishnavya, A.V.G., Eisenstein, Z., and Morgenstern, N.R. "Bef^ of
Compacted Soil in Tension." Journal of the Geotechnical Division, ,Am. -Soc.
: same
Journal GT9. p. 1020-22, Sept. 1976 ;
pp. 763-86, (1974)
SM6, Nov. 1975-'
Townsend, F.C. and R.W. Peterson, Geotechnical Properties of Oil Shale
Retorted by the Paraho and TOSCO Processes - , Tech. Kept. GL 79 "»
S^s^^s^fe.
Research Center, Spokane Wash. Under Contract No.
135
-------