Professional Documents
Culture Documents
OF
Downloaded from ascelibrary.org by INDIAN INST OF TECHNOLOGY - CHENNAI on 08/06/13. Copyright ASCE. For personal use only; all rights reserved.
INTRODUCTION
In the past decade, dozens of papers have been written on
factors that influence the hydraulic conductivity of compacted
clays used for liners and covers in waste containment systems
[e.g., Boynton and Daniel (1985), Kleppe and Olson (1985),
Benson and Daniel (1990), Daniel and Benson (1990), Benson
and Othman (1993), Daniel and Wu (1993), Shelley and Daniel (1993), and Othman et al. (1994)], laboratory and field
hydraulic conductivity testing methods [e.g., Daniel (1989,
1994), Sai and Anderson (1990), Benson et al. (1994a), and
Trautwein and Boutwell (1994)], the correlation of laboratory
and field hydraulic conductivities [e.g., Day and Daniel
(1985), Elsbury et al. (1990), Reades et al. (1990), Benson and
Boutwell (1992), Benson et al. (1994a)], and methods of construction and construction quality assurance [e.g., Benson et
al. (1994b), Daniel and Koerner (1995), and Benson et al.
(1997)]. Despite this wealth of information, the hydraulic performance of compacted clay liners in the field is largely undocumented. In fact, only a few studies have characterized the
field performance of in-service compacted clay liners constructed using standards typical of industry [e.g., Gordon et al.
(1990), Reades et al. (1990), and Benson and Boutwell
(1992)].
There are two principal reasons for the lack of performance
data. First, compacted clays are frequently used in combination with geomembrane liners, and in such applications it is
difficult to separate the performance of the compacted clay
component from that of the geomembrane-clay composite.
Second, few landfills have lysimeters or other instrumentation
installed directly beneath the lining system that enable documentation of field performance. Although lysimeters have been
used to document field performance of several clay liners in
landfills (Gordon et al. 1990; Reades et al. 1990; Benson and
Boutwell 1992), not many clay liners (9% in this study) include lysimeters. Future opportunities to document field performance of clay liners using lysimetry will be limited because
clay liners are now infrequently used in lining systems without
a geomembrane.
1
The purpose of this study was to collect as much information as possible on the field performance of compacted clay
liners. In view of the dearth of information on actual performance of in-service clay liners, the next best source of data
was also used: large-scale hydraulic conductivity tests on fullscale field test pads. The test pads were constructed with materials, methods of construction, and quality assurance procedures that are typical of industry practice for landfill liners.
Test pads constructed for research purposes were not included.
This paper has two objectives: (1) Presentation and documentation of the database in sufficient detail to permit independent analysis by others; and (2) discussion of the key factors affecting KF that are particularly important to practicing
engineers trying to achieve KF 107 cm/s.
DATABASE
Data Sources
The database was assembled primarily from unpublished
data obtained from the writers files (85% of the database) and
supplemented by information in the literature (15% of the database). Most of the data were obtained from engineering reports documenting results from test pads. The data collection
process is thought to have captured the results of 50 75% of
all test pads in North America where large-scale field hydraulic
conductivity tests have been conducted.
Information was collected and screened from more than 120
sites. The requirements for inclusion in the database were (1)
construction methods in general accord with industry practices
for full-scale liners; (2) construction quality assurance in general accord with industry practices; (3) construction with the
objective of achieving KF 1 107 cm/s; (4) reasonably
complete documentation of construction test data; and (5)
availability of results from large-scale field hydraulic conductivity tests. Clay liners constructed from soil-bentonite
mixtures were not included in the database.
The database obtained using these requirements consists of
85 compacted clay liners, of which eight are actual in-service
liners for landfills and 77 are test pads. The database is summarized in Tables 1 and 2. Many of the numbers in Tables 1
and 2 are averages of multiple measurements (geometric mean
for KF and KL, arithmetic mean for all other data). Specific
sites are not identified due to potential sensitivities. The sites
are distributed geographically throughout the United States
and Canada. As a result, the soils included in the database
comprise a broad spectrum of material types. For example, the
average liquid limit (LL) varies from 21 to 101, the fines content varies from 48 to 99%, and the 2-m clay content varies
from 16 to 57%. Nearly all of the soils classify as either CL
Downloaded from ascelibrary.org by INDIAN INST OF TECHNOLOGY - CHENNAI on 08/06/13. Copyright ASCE. For personal use only; all rights reserved.
TABLE 1.
w
(%)
(9)
d
(kN/m3)
(10)
Po
(11)
10.3
19.8
44
26.6
13.8
14.4
19.4
28
98
SP
21.3
16.0
80
18.6
MP
17.3
17.3
95
13.5
19.5
MP
13.8
19.0
32
27
14.1
18.6
SP
17.2
17.7
88
75
45
14.5
18.8
MP
15.3
17.7
45
12.7
18.6
MP
19.6
17.0
90
21
29
16.6
18.7
MP
17.8
16.9
50
57
30
39
21.7
17.3
MP
25.4
16.0
75
12
55
28
33
23.0
16.6
MP
26.0
16.1
78
13
37
15
78
37
18.0
17.0
SP
20.7
16.7
100
14
40
20
70
25
16.2
16.7
SP
17.0
16.8
78
15
85
58
99
57
25.8
14.6
SP
30.8
14.1
98
16
41
22
77
38
15.8
17.0
SP
19.8
16.1
91
17
50
34
95
47
20.3
16.4
SP
23.3
15.7
100
18
30
18
52
16
13.0
18.7
SP
16.6
17.4
85
19
32
14
85
44
10.5
20.1
MP
13.6
19.0
81
20
21
49
51
23
26
94
90
43
36
18.5
11.8
17.2
18.5
SP
MP
17.6
19.5
16.9
16.9
8
80
22
23
63
39
42
18
96
73
30
20.5
20.0
16.3
16.5
SP
SP
22.0
16.4
89
24
67
46
94
53
81
41
88
36
18.9
16.9
71
26
33
19
85
37
15.5
17.6
17
27
31
18
74
26
13.5
18.0
28
35
19
89
41
16.2
17.7
57
29
27
10
76
28
13.9
18.8
84
30
31
32
40
19
24
58
23
MP
SP
MP
SP
MP
SP
RP
MP
SP
RP
MP
SP
RP
MP
SP
RP
MP
SP
15.8
53
18.4
16.3
19.8
18.0
19.3
17.7
17.1
19.5
17.7
16.8
19.4
17.5
17.0
20.5
19.1
18.6
18.6
19.3
23.6
25
16.0
21.5
11.5
16.1
12.2
17.5
18.5
12.1
16.7
18.9
11.5
16.6
18.5
9.0
13.0
14.4
14.0
12.4
16.2
13.1
18.6
19.1
65
75
32
45
27
99
42
11.0
19.9
MP
13.9
19.2
92
33
29
15
87
40
13.3
18.9
MP
13.4
18.7
80
34
44
16
96
17.3
17.1
SP
17.8
17.1
45
35
39
19
97
22.2
16.4
SP
20.7
16.8
78
36
36
17
74
30
13.2
18.3
SP
17.6
77
37
36
17
48
16
12.4
19.0
SP
15.5
15.5
14.1
18.2
45
Site
number
(1)
LL
(2)
PI
(3)
Fines
(%)
(4)
Clay
fraction
(5)
wo
(%)
(6)
24
10
65
37
2
3
58
25
29
10
85
85
50
22
10.2
9.0
26.8
12.3
20.1
21.3
14.6
19.0
SP
MP
SP
SP
50
34
95
47
17.9
16.8
43
26
87
32
14.3
32
19
88
35
33
13
77
35
22
55
31
10
43
11
Compaction
criterion
(12)
w > wo
PC > 90% dm MP
None
w > wo
PC > 95% dm SP
w > wo 2, <8
PC > 90% dm MP
w > wo
PC > 90% dm MP
w > wo
PC > 90% dm MP
w > wo 2
PC > 90% dm MP
w > wo 2 to 4
PC > 90% dm MP
w > wo
PC > 90% dm MP
w > wo
PC > 90% dm MP
w > wo
PC > 90% dm MP
w > wo
PC > 90% dm MP
w > wo 2, <5
PC > 90% dm MP
w > wo 2, <5
PC > 90% dm MP
w > wo
PC > 100% dm SP
w > wo 2, <5
PC > 90% dm MP
w > wo 2, <6
PC > 90% dm SP
Si > 78.5
PC > 90% dm MP
w > wo
PC > 90% dm MP
Si > 82.0
w > wo
PC > 95% dm SP
Si > 85.0
w > wo
PC > 90% dm MP
w > wo
PC > 90% dm MP
w > wo
PC > 90% dm MP
w > wo
PC > 90% dm MP
w > wo
PC > 90% dm MP
w > wo
PC > 90% dm MP
w > wo 2, 5
PC > 90% dm MP
PC > 90% dm MP
w > wo
PC > 95% dm SP
w > wo, <6
PC > 90% dm MP
w > wo
PC > 90% dm MP
w > wo
PC > 95% dm SP
w > wo
PC > 95% dm SP
w > wo, <6
PC > 95% dm SP
w > wo
PC > 95% dm SP
Downloaded from ascelibrary.org by INDIAN INST OF TECHNOLOGY - CHENNAI on 08/06/13. Copyright ASCE. For personal use only; all rights reserved.
TABLE 1.
(Continued )
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
38
21
60
26
10.3
20.4
SP
11.5
20.4
39
21
60
26a
10.3
20.4
SP
11.6
17.9
10
31.6
13.4
SP
35.5
100
w > wo
PC > 90% dm MP
w > wo 0.7, <wo 1.7
PC > 90% dm SP
w > wo, 5
PC > 92% dm SP
PC > 95% dm SP
w > wo 1
PC > 95% dm SP
w > wo 1
PC > 90% dm SP
w > wo, <2
PC > 95% dm SP
w > wo 1, <3
PC > 95% dm SP
w > wo 1, <5
PC > 90% dm SP
w > wo, <3
PC > 95% dm SP
PC > 95% dm SP
PC > 91% dm MP
w > wo 1, <5
PC > 95% dm SP
w > wo 1, <5
PC > 95% dm SP
w > wo 1, <5
PC > 95% dm SP
w > wo 1, <5
PC > 95% dm SP
w > wo 2, <4
PC > 90% dm MP
w > wo 2, <4
PC > 90% dm MP
w > wo, <6
PC > 95% dm SP
w > wo, <SL
PC > 95% dm SP
w > wo, <SL
PC > 95% dm SP
w > wo 3, <6
PC > 95% dm SP
w > wo 3, <5
PC > 95% dm SP
w > wo 0.4
PC > 98% dm SP
w > wo 1.5
PC > 94% dm SP
w > wo
PC > 95% dm SP
w > wo
PC > 95% dm SP
w > wo 4
40
101
71
98
49
41
42
43
47
69
62
30
45
42
66
79
86
49
19.5
23.4
22.4
16.3
15.1
15.4
SP
SP
SP
21.9
25.0
23.4
12.8
12.8
16.0
15.1
15.4
44
62
42
86
22.4
15.4
SP
24.2
15.0
47
45
44
28
70
19.5
16.4
SP
19.8
16.3
71
46
35
16
98
22
23.3
15.4
SP
27.3
15.4
100
47
39
24
70
14.6
17.7
SP
16.5
17.7
100
48
41
23
86
75
22
86
18.9
16.7
86
50
43
24
86
18.6
16.9
84
51
40
22
86
17.8
17.0
73
52
37
18
73
38
RP
SP
MP
RP
SP
MP
RP
SP
MP
RP
SP
MP
SP
17.0
42
16.2
16.7
18.7
16.2
16.7
18.7
16.2
16.7
18.7
16.2
16.7
18.7
16.5
17.8
49
20.0
18.0
13.3
20.0
18.0
13.3
20.0
18.0
13.3
20.0
18.0
13.3
19.9
21.2
16.1
67
53
54
55
56
57
58
59
60
61
62
63
64
65
66
54
66
66
69
69
69
68
68
51
51
47
47
50
31
35
35
38
38
38
35
35
20
20
30
31
29
93
93
98
98
98
95
95
73
73
66
66
75
40
19.9
27.4
27.4
26.8
26.8
26.8
26.6
26.6
20.2
20.2
19.5
13.5
19.0
16.4
14.5
14.5
14.6
14.6
14.6
14.6
14.6
15.9
15.9
16.3
19.2
16.1
SP
SP
SP
SP
SP
SP
SP
SP
SP
SP
SP
MP
SP
21.6
27.0
30.6
29.6
30.7
29.4
26.8
29.8
24.6
22.7
21.6
17.2
21.7
15.5
15.0
14.2
14.0
14.3
14.4
15.1
14.4
15.4
15.4
16.0
17.4
17.2
71
88
0
95
67
49
27
62
19.3
16.1
SP
21.4
17.2
98
68
35
17
67
22
100
50
16
11.5
19.4
75
70
42
26
88
45
SP
MP
SP
MP
MP
18.0
22
17.7
19.0
19.9
21.4
18.7
17.6
69
14.8
11.5
10.0
8.5
14.9
20.6
16.1
60
71
29
19
83
34
12.2
19.6
MP
14.3
18.0
64
72
36
20
85
35
18.0
16.5
SP
23.7
15.5
100
73
76
53
21.0
15.5
25.2
14.8
97
74
56
40
64
18.0
16.9
19.6
16.1
47
75
21.0
15.6
SP
25.4
15.2
100
76
37
17
92
19.2
16.6
SP
21.8
15.9
86
77
32
13
19
9.9
19.7
SP
11.0
19.2
51
78
32
16
25
11.5
19.6
SP
12.4
18.8
63
79
62
41
82
35
84
23.1
14.9
81
47
22
SP
MP
SP
MP
SP
16.2
52
14.9
16.5
15.9
18.0
15.3
28.2
80
25.0
17.8
19.6
14.4
25.0
28.0
14.2
92
81
63
TABLE 1.
(Continued )
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
82
83
84
85
39
16
84
81
54
48
18.2
17.6
SP
17.8
15.7
17.1
18.4
(12)
PC
PC
PC
PC
>
>
>
>
96%
96%
90%
90%
dm
dm
dm
dm
SP
SP
MP
MP
Downloaded from ascelibrary.org by INDIAN INST OF TECHNOLOGY - CHENNAI on 08/06/13. Copyright ASCE. For personal use only; all rights reserved.
Note: LL = liquid limit; PI = plasticity index; w = compaction water content; wo = optimum water content; PC = percent compaction; dm = maximum
dry unit weight; MP = modified Proctor (ASTM D 1557); SP = standard Proctor (ASTM D 698); and RP = reduced Proctor (Daniel and Benson 1990).
a
4-m clay fraction.
TABLE 2.
Site
number
(1)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Compactor
weight
(Mg)
(2)
32.4
30.0
19.8
36.0
32.4
32.4
12.6
32.4
39.0
19.8
25.0
32.4
32.4
32.4
18.9
18.9
18.9
18.9
18.9
32.4
27.0
32.4
19.8
19.8
32.4
17.1
19.8
19.8
7.2
7.2
19.8
10.9
19.8
19.8
19.8
19.8
19.8
19.8
19.8
19.8
19.8
Passes
per lift
(3)
Lift
thickness
(cm)
(4)
Number
of lifts
(5)
6
6
4
6
5
4
4
6
4
5
8
6
8 12
8 12
46
46
46
6
4
6
8
4
12
12
40
16
8
6
8
6 10
58
3
3
45
12
6 (lifts 1 3)
21 (lift 4)
6
15
15
15
15
15
15
15
15
15
15
15
15
15
17
20
17
15
15
15
15
15
15
15
15
15
15
15
17
15
30
15
15
15
15
6
6
15
13
15
15
15
15
15
15
20 25
8.5
15
15
15
15
15
16
13
13
14
6
5
8
4
10
6
8
6
10
10
10
10
8
6
7
5
4
5
10
6
6
10
5
5
6
6
6
9
6
2
8
8
3
6
6
5
6
6
6
4
4
5
5
4
5
10
4
4
4
4
4
4
4
4
4
4
Thin-wall
tube
(6)
3.2
3.6
8.0
5.0
8.7
2.4
8.4
9.0
1.0
8.0
2.0
3.0
1.3
4.8
4.4
3.7
3.0
1.5
1.9
3.0
3.1
2.4
1.5
9.0
2.3
2.9
3.0
1.9
2.2
3.0
1.6
3.0
1.3
1.5
3.0
9.1
4.9
2.6
3.5
5.5
2.4
2.4
5.8
1.5
1.1
5.1
7.4
4.1
1.7
8.1
2.8
2.8
3.4
108
109
109
109
109
108
108
109
108
109
109
109
108
108
109
108
109
108
108
109
107
108
108
109
109
109
108
108
108
108
108
108
108
108
108
109
108
108
109
109
109
109
109
108
108
108
108
108
108
108
108
108
108
SDRI
(7)
2.8 107
1.5 107
1.1 108
9.0 109
2.7 107
5.8 108
1.2 107
1.3 108
2.0 108
3.3 109
3.0 108
9.8 109
8.0 107
2.5 107
2.0 108
1.5 108
8.0 109
2.0 107
1.8 107
9.0 108
1.7 108
1.1 107
6.0 108
3.9 108
3.9 108
4.0 107
3.7 108
3.0 108
1.3 108
<3.6 108
3.0 108
2.2 108
1.0 107
8.0 108
7.0 108
2.0 107
3.7 108
2.0 108
5.0 108
4.0 108
5.0 108
2.6 107
3.0 107
1.1 107
2.2 108
7.0 108
1.3 107
2.4 108
2.4 108
5.6 108
Lysimeter
(8)
9.0
7.0
3.0
3.0
2.0
6.0
4.4
1.4
109
109
108
109
109
109
108
108
TSB
(9)
4.3
1.4
1.6
5.0
9.2
4.7
1.6
2.1
3.2
7.5
1.1
1.2
108
108
108
109
109
108
108
108
107
108
107
108
300-mm
block
(10)
2.6 107
4.0 108
1.4 108
2.2 107
1.1 108
6.0 109
1.8 107
1.5 107
1.7 107
1.7 108
3.5 107
4.1 109
4.8
7.7
3.1
5.3
108
108
106
107
Downloaded from ascelibrary.org by INDIAN INST OF TECHNOLOGY - CHENNAI on 08/06/13. Copyright ASCE. For personal use only; all rights reserved.
TABLE 2.
(1)
(2)
(3)
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
19.8
19.8
19.8
19.8
19.8
19.8
19.8
19.8
19.8
19.8
19.8
19.8
32.4
19.8
19.8
19.8
19.8
19.8
19.8
19.8
10.2
10.2
19.8
19.8
16.2
32.4
32.4
19.8
19.8
19.8
19.8
12
12
12
12
12
12
40
80
2
2
6
6
7 10
7 10
8
22
22
22
22
6
10
4
8
8
4
6
6
74
75
76
77
78
79
80
81
82
83
84
85
1
6
1
6
1
3
1
3
1
3
(4)
(5)
15
17
17
19
15
23
15
15
10
10
15
15
15
15
15
at 20
at 10
at 20
at 10
at 20
at 15
15
10
10
15
15
15
15
15
at 30
at 20
at 30
at 20
4
4
4
4
4
4
4
4
10
11
4
4
6
6
4
7
7
7
7
4
4
4
9
9
8
8
4
4
4
4
4
4
4
(Continued
(6)
2.5
2.7
3.4
4.3
1.6
1.7
5.5
3.7
3.0
7.8
2.1
2.0
2.0
1.4
4.7
3.3
1.8
4.2
1.5
1.7
108
108
108
108
107
107
109
108
108
109
108
108
108
108
108
109
109
108
108
108
(7)
5.0
9.4
1.2
3.7
3.1
3.9
2.3
1.8
1.2
8.3
2.3
1.3
4.0
8.3
2.0
8.0
8.0
9.0
9.0
5.0
5.0
3.0
3.0
3.0
4.5
4.0
1.5
3.0
4.5
1.3
1.3
2.8
2.8
108
108
107
108
107
107
107
107
108
108
108
108
108
108
108
108
108
108
108
108
108
108
108
108
108
108
107
108
108
107
107
108
108
(8)
(9)
108
108
108
108
1.1
8.5
2.6
5.6
(10)
4.1 109
FIG. 1.
Downloaded from ascelibrary.org by INDIAN INST OF TECHNOLOGY - CHENNAI on 08/06/13. Copyright ASCE. For personal use only; all rights reserved.
FIG. 3. Field Hydraulic Conductivity versus Hydraulic Conductivity Measured in Laboratory on Small Undisturbed Specimens. Percentage of Water Content-Dry Unit Weight Points Unknown for Sites Marked Po = ?
Downloaded from ascelibrary.org by INDIAN INST OF TECHNOLOGY - CHENNAI on 08/06/13. Copyright ASCE. For personal use only; all rights reserved.
for one of the highest KF in the database (Site 20, Fig. 4 and
Table 1). The sites with higher KF/KL were generally compacted drier relative to the line of optimums (i.e., data shown
as Po < 80% in Figs. 3 and 4) a condition known to result
in macroscopic features and scale-dependent KF (Benson and
Daniel 1990; Benson and Boutwell 1992; Benson et al.
1994a).
Laboratory tests on 75-mm-diameter specimens obtained
from compacted clay liners are a routine part of many construction quality assurance programs, and the results of these
tests often form the principal basis for pass/fail decisions.
Twenty-two clay liners in the database failed to achieve a KF
107 cm/s. Of the 18 failing clay liners for which KL data
are available, the laboratory tests incorrectly indicate that 15
of them pass the 1 107 cm/s requirement. That is, KL provided a false positive (pass) in 15 of 18 (83%) of the failing
test pads. The database clearly shows that laboratory hydraulic
conductivity tests have a strong tendency to yield unconservative (passing) values for failing clay liners (KF > 107 cm/s).
Consequently, they have limited value for assessing construction quality, especially for detecting failing liners. The best
that can be said for small-scale laboratory tests is that a failing
KL is occasionally measured and that in all three such instances
in the database KF was also >107 cm/s (i.e., there were no
false negatives in the KL values). Thus, laboratory tests are
occasionally successful in identifying unsatisfactory KF in
those rare cases where KL > 107 cm/s.
STATISTICAL ANALYSIS
Various studies including those cited in this paper indicate
that the factors influencing KL are soil composition, water content relative to the line of optimums, compactive effort, and
effective stress. For KF, the thickness or number of lifts must
also be considered (Boutwell and Rauser 1990; Benson and
Daniel 1994). Statistical analyses were therefore conducted to
identify if these variables had a significant effect on KF.
The first analysis consisted of t-tests to determine which, if
any, of the aforementioned variables were statistically different
in the 53 clay liners where KF was 107 cm/s relative to
those for the 22 clay liners where KF was >107 cm/s. Neither
soil composition [in terms of LL, plasticity index (PI), and
percent fines] nor dry unit weight was significantly different,
even at a significance level of 0.30. Relative wetness (in terms
of average saturation or percent of compaction data falling wet
of the line of optimums) and thickness (or number of lifts)
were significant, even at the 0.005 significance level.
A more comprehensive analysis was subsequently conducted using stepwise regression. The analysis was similar to
that used by Benson et al. (1994b) using principles described
in Draper and Smith (1981). In short, stepwise regression identifies independent variables (e.g., compaction parameters, index properties, etc.) having a statistically significant effect on
a dependent variable (in this case log KF) by examining the
linear correlation between the independent and dependent variables using the partial-F statistic. At the common significance
level of 0.05, statistically significant variables have partial-F
4. Borderline significant variables have partial-F < 4 and
>2, whereas statistically insignificant variables have partial-F
< 2. The regression proceeds in steps, identifying the most
significant independent variable in the first step and the less
significant variables in subsequent steps after removing the
correlation attributed to the most statistically significant independent variable. The base-10 logarithm of KF was used in the
regression analysis, because it is nearly normally distributed
(Fig. 1). More detail on stepwise regression can be found in
Benson et al. (1994b) and Draper and Smith (1981).
The first regression analysis was conducted with the entire
database, using the following independent variables: soil com-
FIG. 5.
TABLE 3.
Entire Databasea
Independent
variable
(1)
Percent wet line of
optimums
Liner thickness
Liquid limit
Plasticity index
Percent fines
Lift thickness
Percent wet of optimumc
Percent compaction
Initial saturation
a
Partial-F
statistic
(2)
Significant?
(3)
Partial-F
statistic
(4)
Significant?
(5)
11.3
4.7
0.63
0.64
3.10
1.11
Yes
Yes
No
No
Borderline
No
5.43
0.31
3.13
0.88
0.37
Yes
No
Borderline
No
No
No
No
No
0.47
0.32
0.15
No
No
No
0.74
0.65
0.46
Line of Optimums
The first regression analysis indicates that the only two independent variables are statistically significant (Table 3): Po
and liner thickness. The relationship between KF and Po is
illustrated in Fig. 6(a); KF decreases with increasing Po , and
KF is almost always <107 cm/s when Po > 80%. The data
indicate that lower KF should be expected when a greater portion of a test pad or liner is compacted wet of the line of
optimums. Discussion of the importance of thickness is in a
subsequent section.
Numerous laboratory studies over the past 40 years have
shown that the hydraulic conductivity of compacted clay varies orders of magnitude from dry to wet of the line of optimums (Lambe 1954; Mitchell et al. 1965; Garcia-Bengochea
et al. 1979; Acar and Oliveri 1990; Benson and Daniel 1990),
whereas composition (e.g., index properties) usually has a
much smaller effect on hydraulic conductivity (Daniel 1987;
Benson et al. 1994b). Because compaction water content is so
influential, it masks the importance of the other variables [e.g.,
Benson and Trast (1995)].
A statistical inference regarding the relationship between KF
and Po can be made from box plots of KF obtained from test
pads having similar percentage Po [Fig. 6(b)]. The central horizontal line of a box corresponds to the median, whereas the
upper and lower ends of the box correspond to the 75th and
25th percentiles. The outermost horizontal lines correspond to
5th and 95th percentiles; that is, the outermost lines delineate
the range of nearly all of the data. The circles represent outliers. The box plots provide a means to describe the distribution of KF for particular ranges of Po. The box plots show that
the distribution of KF shifts downward nearly an order of magnitude when Po increases beyond 60%. Moreover, the box
plots and the data in Tables 1 and 2 indicate that <33% of the
liners have KF 107 cm/s when Po < 60%, whereas when
Po > 80% the percentage of liners with KF 107 cm/s is
>90%.
The data shown in Fig. 6 suggest that specification writers
should consider requiring a minimum Po. Based on this analysis, the writers suggest a minimum Po of 70 80%, which is
slightly lower than the threshold recommended by Benson and
Boutwell (1992) (Po 85%). This criterion is relatively easy
to establish because the line of optimums is practically unique
for most clayey soils (Blotz et al. 1998). For example, Fig. 7
shows that the optimum water content-maximum dry unit
weight points for the sites in the database can be represented
with a simple equation nearly identical to the relationship reported by Blotz et al. (1998).
Ensuring the compaction water content is wet of the line of
optimums is not a panacea. For example, the four test pads at
Sites 48 51 were constructed with the same soil and methods
and at essentially the same water content and Po. Nevertheless,
their KF varied by more than an order of magnitude. Differences in hydration time were the apparent cause for the variation in KF (Benson et al. 1997). Thus, seemingly minor construction variables can have a profound influence on KF, even
if the gross measurements of water content and dry unit weight
indicate that compaction is wet of the line of optimums.
Percent Compaction/Wet-of-Optimum Specifications
Downloaded from ascelibrary.org by INDIAN INST OF TECHNOLOGY - CHENNAI on 08/06/13. Copyright ASCE. For personal use only; all rights reserved.
FIG. 8. Compaction Curves, Compaction Data, and Acceptable Compaction Zone for Site 27
Downloaded from ascelibrary.org by INDIAN INST OF TECHNOLOGY - CHENNAI on 08/06/13. Copyright ASCE. For personal use only; all rights reserved.
CONSTRUCTION FACTORS
The regression analyses conducted on the entire database
and that portion of the database with Po > 80% showed that
lift thickness was not statistically significant. Compactor mass
and number of passes were not included in the regression analysis due to lack of data (Table 2). For brevity, graphs of these
variables have not been included, but neither variable significantly influenced KF. Compaction conditions (e.g., Po and
wo) appear to be much more important than the details of
how the soil was compacted. For at least 95% of the sites,
four or more compactor passes were used. In addition, heavy
compactors (180 MN) were used for 90% of the sites (Table
2). Field hydraulic conductivity may have been significantly
related to these variables if a fewer number of passes or lighter
compactors had been used at a greater number of sites. The
data in Table 2 do show, however, that there is no apparent
advantage of using very heavy compactors (e.g., 315 MN)
relative to the moderately heavy compactors (weight 195
MN) that are commonly used.
Examples exist, however, where the number of compactor
passes was the only variable that differed substantially between two similar test pads having different KF. For example,
Sites 43 and 44 are two test pads that were constructed from
the same soil at essentially the same water content and compacted with the same compactor. The only variable was the
number of compactor passes: 16 for Site 43 and eight for Site
44. KF for Sites 43 and 44 are 7 108 and 2 107 cm/s,
respectively, illustrating that increasing the compactive effort
can reduce KF. The soil compacted with 16 passes had Po =
63%, whereas Po was 47% when eight passes were used. The
additional compactive effort used for Site 43 resulted in more
w-d points falling above the line of optimums.
Liner thickness was found to be statistically significant,
which is consistent with the conclusion Benson and Daniel
(1994) made in their study regarding an appropriate minimum
thickness for clay liners. Thicker liners typically have a greater
number of lifts, which results in a lower probability that a
continuous permeable pathway penetrates the entire thickness
of the liner (Boutwell and Rauser 1990; Benson and Daniel
1994).
SOIL COMPOSITION
Engineers usually specify minimum values for LL and/or PI
(LL typically is 25 30, PI typically is 12 15), fines (typically 50% passing the No. 200 sieve), and sometimes percentage clay (e.g., 20 25% 2-m clay) to ensure that suitable soils are used for construction of clay liners (Daniel 1990;
Benson et al. 1994b). These specifications are made based on
the belief that these variables influence KF (Benson et al.
1994b; Benson and Trast 1995). In their analysis of KL from
laboratory-compacted specimens, Benson and Trast (1995)
found that KL correlates reasonably well with these index properties when compaction is wet of the line of optimums. Boutwell and Hedges (1989) and Benson et al. (1994b) found similar results for KL of field-compacted specimens.
The second regression analysis (sites where Po 80%) was
conducted to assess whether relationships exist between KF
and index properties when compaction is wet of the line of
optimums. Results of the analysis (Table 3) show, however,
that liner thickness (or number of lifts) is the only statistically
significant variable even when Po 80%, and only PI is borderline significant. LL was found to be borderline significant
when a similar regression was conducted without PI as an
independent variable. Similar information is in the LL and PI,
because these variables are correlated.
Graphs of KF versus LL and PI are shown in Fig. 12. Slight
trends of decreasing KF with increasing LL and PI exist that
Downloaded from ascelibrary.org by INDIAN INST OF TECHNOLOGY - CHENNAI on 08/06/13. Copyright ASCE. For personal use only; all rights reserved.
Downloaded from ascelibrary.org by INDIAN INST OF TECHNOLOGY - CHENNAI on 08/06/13. Copyright ASCE. For personal use only; all rights reserved.
The third lesson follows directly from lesson two and the
data shown in Fig. 6(b).
Lesson 2: Avoid the conventional percent compaction specification (minimum d and wet of a single optimum water content) and craft a specification that ensures compaction on or
above the line of optimums.
CONCLUSIONS
FIG. 14. Schematics of: (a) Percent Compaction Specification; (b) Recommended Compaction Specification
Downloaded from ascelibrary.org by INDIAN INST OF TECHNOLOGY - CHENNAI on 08/06/13. Copyright ASCE. For personal use only; all rights reserved.
APPENDIX.
REFERENCES
Acar, Y., and Oliveri, I. (1990). Pore fluid effects on the fabric and
hydraulic conductivity of laboratory-compacted clay. Transp. Res.
Rec. 1219, Transportation Research Board, Washington, D.C., 144
159.
Benson, C., and Boutwell, G. (1992). Compaction control and scaledependent hydraulic conductivity of clay liners. Proc., 15th Annu.
Madison Waste Conf., University of Wisconsin, Madison, Wis., 62
83.
Benson, C., and Daniel, D. (1990). Influence of clods on hydraulic conductivity of compacted clay. J. Geotech. Engrg., ASCE, 116(8),
1231 1248.
Benson, C., and Daniel, D. (1994). Minimum thickness of compacted
soil liners: II. Analysis and case histories. J. Geotech. Engrg., ASCE,
120(1), 153 172.
Benson, C., Gunter, J., Boutwell, G., Trautwein, S., and Berzanskis, P.
(1997). Comparison of four methods to assess hydraulic conductivity. J. Geotech. and Geoenvir. Engrg., ASCE, 123(10), 929 937.
Benson, C., Hardianto, F., and Motan, E. (1994a). Representative specimen size for hydraulic conductivity assessment of compacted soil liners. Hydraulic conductivity and waste contaminant transport in soils,
ASTM STP 1142, D. Daniel and S. Trautwein, eds., ASTM, West Conshohocken, Pa., 3 29.
Benson, C., and Othman, M. (1993). Hydraulic conductivity of compacted clay frozen and thawed in place. J. Geotech. Engrg., ASCE,
119(3), 276 294.
Benson, C., and Trast, J. (1995). Hydraulic conductivity of thirteen compacted clays. Clays and Clay Minerals, 43(6), 669 681.
Benson, C., Zhai, H., and Wang, X. (1994b). Estimating the hydraulic
conductivity of compacted clay liners. J. Geotech. Engrg., ASCE,
120(2), 366 387.
Blotz, L., Benson, C., and Boutwell, G. (1998). Estimating optimum
water content and maximum dry unit weight for compacted clays. J.
Geotech. and Geoenvir. Engrg., ASCE, 124(9), 907 912.
Boutwell, G., and Hedges, S. (1989). Evaluation of water-retention liners by multivariate statistics. Proc., 12th ICSMFE, Balkema, Rotterdam, The Netherlands, 815 818.
Boutwell, G., and Rauser, C. (1990). Clay liner construction. Proc.,
Geotech. Engrg. in Todays Envir., ASCE, Reston, Va., 1 7.
Boynton, S., and Daniel, D. (1985). Hydraulic conductivity tests on
compacted clay. J. Geotech. Engrg., 111(4), 465-478.
Chiu, T.-F., and Shackelford, C. (1994). Practical aspects of the capillary
barrier effect for landfills. Proc., 17th Annu. Madison Waste Conf.,
University of Wisconsin, Madison, Wis., 357 375.
Daniel, D. (1984). Predicting hydraulic conductivity of clay liners. J.
Geotech. Engrg., ASCE, 110(2), 285 300.
Daniel, D. (1987). Earthen liners for land disposal facilities. Proc.,
Geotech. Pract. for Waste Disposal 87, R. Woods, ed., ASCE, Reston,
Va., 21 39.
Daniel, D. (1989). In situ hydraulic conductivity tests for compacted
clays. J. Geotech. Engrg., ASCE, 115(9), 1205 1227.
Daniel, D. (1990). Summary review of construction quality control for
earthen liners. Proc., Waste Containment Sys., R. Bonaparte, ed.,
ASCE, Reston, Va., 175 189.
Daniel, D. (1994). State-of-the-art: Laboratory hydraulic conductivity
tests for saturated soils. Hydraulic conductivity and waste contaminant transport in soils, ASTM STP 1142, D. Daniel and S. Trautwein,
eds., ASTM, West Conshohocken, Pa., 30 78.
Daniel, D., and Benson, C. (1990). Water content-density criteria for
compacted soil liners. J. Geotech. Engrg., ASCE, 116(12), 1811
1830.
Daniel, D., and Koerner, R. (1995). Waste containment systems: Guidance
for construction, quality assurance, and quality control of liner and
cover systems. ASCE, Reston, Va., 354.
Daniel, D., and Wu, Y. (1993). Compacted clay liners and covers for
arid sites. J. Geotech. Engrg., ASCE, 119(2), 223 237.
Day, S., and Daniel, D. (1985). Hydraulic conductivity of two prototype
clay liners. J. Geotech. Engrg., ASCE, 111(8), 957 970.
Draper, N., and Smith, H. (1981). Applied regression analysis. Wiley,
New York.
Elsbury, B., Daniel, D., Sraders, G., and Anderson, D. (1990). Lessons
learned from compacted clay liner. J. Geotech. Engrg., ASCE,
116(11), 1641 1660.
Garcia-Bengochea, I., Lovell, C., and Altschaeffl, A. (1979). Pore distribution and permeability of silty clays. J. Geotech. Engrg. Div.,
ASCE, 105(7), 839 856.
Gordon, M., Huebner, P., and Mitchell, G. (1990). Regulation, construction and performance of clay lined landfills in Wisconsin. Proc.,
Waste Containment Sys., R. Bonaparte, ed., ASCE, Reston, Va., 14
27.
Johnson, G., Crumley, W., and Boutwell, G. (1990). Field verification
of clay liner hydraulic conductivity. Proc., Waste Containment Sys.,
R. Bonaparte, ed., ASCE, Reston, Va., 226 245.
Kleppe, J., and Olson, R. (1985). Desiccation cracking of soil barriers.
Hydraulic barriers in soil and rock, ASTM STP 874, ASTM, West
Conshohocken, Pa., 263 275.
Lambe, T. (1954). The permeability of compacted fine-grained soils.
ASTM STP 163, ASTM, West Conshohocken, Pa., 56 67.
Leroueil, S., Bouchard, R., and Bihan, J. (1992). Discussion of Water
content-density criteria for compacted soil liners, by D. Daniel and C.
Benson. J. Geotech. Engrg., ASCE, 118(2), 963 965.
Mitchell, J., Hooper, D., and Campanella, R. (1965). Permeability of
compacted clay. J. Soil Mech. and Found. Div., ASCE, 91(4), 41
65.
Othman, M., Benson, C., Chamberlain, E., and Zimmie, T. (1994). Laboratory testing to evaluate changes in hydraulic conductivity of compacted clays caused by freeze-thaw: State-of-the-art. Hydraulic conductivity and waste contaminant transport in soils, ASTM STP 1142,
D. Daniel and S. Trautwein, eds., ASTM, West Conshohocken, Pa.,
227 254.
Reades, D., Lahti, L., Quigley, R., and Bacopoulos, A. (1990). Detailed
case history of clay liner performance. Proc., Waste Containment Sys.,
R. Bonaparte, ed., ASCE, Reston, Va., 156 174.
Sai, J., and Anderson, D. (1990). Field hydraulic conductivity tests for
compacted soil liners. Geotech. Testing J., 13(3), 215 225.
Shelley, T., and Daniel, D. (1993). Effect of gravel on hydraulic con-
tests for compacted soil liners and caps. Hydraulic conductivity and
waste contaminant transport in soils, ASTM STP 1142, D. Daniel and
S. Trautwein, eds., ASTM, West Conshohocken, Pa., 184 223.
Wang, X., and Benson, C. (1995). Infiltration and field-saturated hydraulic conductivity of compacted clay. J. Geotech. Engrg., ASCE,
121(10), 713 722.
Downloaded from ascelibrary.org by INDIAN INST OF TECHNOLOGY - CHENNAI on 08/06/13. Copyright ASCE. For personal use only; all rights reserved.