ORIGINAL PAPER Shear Modulus and Damping Ratio of Organic Soils P. Kallioglou Th. Tika G. Koninis St. Papadopoulos K. Pitilakis Received: 18 May 2007 / Accepted: 11 May 2008 Ó Springer Science+Business Media B.V. 2008 Abstract The paper presents results from a labora- tory investigation into the dynamic properties of natural intact and model organic soils by means of resonant-column tests. The natural intact organic soils were sands, cohesive soils and peats with varying content of calcium carbonate. The model organic soils were formed in laboratory by mixing kaolinite and paper pulp. The influence of various soil parameters, such as strain level, confining stress, void ratio, plasticity index, organic content and secondary consolidation time on shear modulus, G, and damping ratio, DT, is presented and discussed. The test results on natural organic soils show that only high organic contents (OC C 25%) have significant influence on G and DT at both small and high shear strains. For the model organic soils, however, a significant influence of even lower values of organic content (5% B OC B 20%) on G at small strains and DT at both small and high strains is observed. Keywords Shear modulus Damping ratio Dynamic properties Resonant-column Organic soils Peat 1 Introduction It is well known in Geotechnical Engineering that among the key parameters controlling the response of soil to dynamic loading is shear modulus and damping ratio. It is also widely accepted that the response of soil to both static and dynamic loading depends mainly on the level of strain induced to it. Soil behaviour ranges from linear elastic to inelastic, depending on the level of strain, and may be divided in three zones (Dobry 1991; Jardine 1992; Hight and Higgins 1994). At small shear strains, c, the stress– strain relation is linear elastic, soil shear stiffness has its maximum value, G = G max , and damping ratio its minimum value, DT = DT min . The upper strain limit of this range is called linear elastic threshold shear strain, c t e , and depends on soil type. When the strain exceeds the linear elastic threshold shear strain, but remains below another upper strain limit, the stress– strain relation becomes non-linear elastic, and is accompanied by stiffness degradation and damping increase. The latter upper strain limit is called volumetric threshold shear strain, c t v , and depends also on soil type and stress state. When the strain exceeds the volumetric threshold shear strain, c t v , the stress–strain behaviour becomes inelastic with sig- nificant stiffness degradation, damping increase and plastic strains development. Consequently, passage from elastic to inelastic behaviour corresponds to c t v , defined as the strain at which excess pore water pressures or plastic strains start to build up. P. Kallioglou (&) Th. Tika G. Koninis St. Papadopoulos K. Pitilakis Department of Civil Engineering, Laboratory of Soil Mechanics, Foundations’ & Geotechnical Earthquake Engineering, Aristotle University of Thessaloniki, 54124 Thessaloniki, Greece e-mail: [email protected]123 Geotech Geol Eng DOI 10.1007/s10706-008-9224-1
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
ORIGINAL PAPER
Shear Modulus and Damping Ratio of Organic Soils
P. Kallioglou Æ Th. Tika Æ G. Koninis ÆSt. Papadopoulos Æ K. Pitilakis
Received: 18 May 2007 / Accepted: 11 May 2008
� Springer Science+Business Media B.V. 2008
Abstract The paper presents results from a labora-
tory investigation into the dynamic properties of
natural intact and model organic soils by means of
resonant-column tests. The natural intact organic
soils were sands, cohesive soils and peats with
varying content of calcium carbonate. The model
organic soils were formed in laboratory by mixing
kaolinite and paper pulp. The influence of various soil
parameters, such as strain level, confining stress, void
ratio, plasticity index, organic content and secondary
consolidation time on shear modulus, G, and damping
ratio, DT, is presented and discussed. The test results
on natural organic soils show that only high organic
contents (OC C 25%) have significant influence on G
and DT at both small and high shear strains. For the
model organic soils, however, a significant influence
P2s 35.4–35.9 548 R 2.441 58 1.953 14 86 62 – Pt Brown-black, peat with shells
and intense fibrous structure
(fibers with horizontal
orientation)
a Specimens had nominal dimensions: Do = 7.1 cm, Ho = 14.2 cm, apart from that indicated with index s, which had nominal
dimensions: Do = 3.6 cm, Ho = 7.1 cmb Estimated in-situ vertical effective stress. Where data were not available, the unit weights, c, above and below ground water table
a Specimens had nominal dimensions: Do = 7.1 cm, Ho = 14.2 cmb Initial parameters at resonant–column testsc Organic matter contentd Mixture prepared with solution of distilled water and NaCl (0.18 N)
Geotech Geol Eng
123
organic content and calcareous content by the
following expression:
G ¼ A cð Þ � f r0o� �
� g eð Þ � h OCð Þ � k CCð Þ ð1Þ
where: A(c), f(r0o), g(e), h(OC) and k(CC) are
functions of c, r0o, e, OC and CC respectively.
The regression analyses of these data resulted in
Eq. 2.
G¼ 830
1þ17:35�c0:597�r0ð0:681�c0:045Þ
o �e�0:99
�OC�0:104�CC0:657 R2¼0:922� �
ð2Þ
where: c, OC and CC are expressed as percentages
(%) and r0o and G are expressed in kPa.
The normalized G values predicted from the above
equation are compared with the experimental data at
a shear strain close to the minimum shear strain used
in the tests on both large and small size specimens,
c = 6.6 9 10-5, in Fig. 4.
The normalized shear modulus, G/Gmax, is plotted
versus shear strain in Fig. 5a for all sands at an
effective isotropic stress approximately equal to or
higher than r0oinsitu. For sands S2s, S3s and S5s, the
measured G values were normalized by the Gmax
value, determined from equation (2) at c = 10-6. For
medium-organic sands, the degradation G/Gmax
curves agree with the range of corresponding curves
presented in literature for reconstituted inorganic
sands containing fines (Kallioglou 2003). High-
organic sands exhibit higher linearity in G/Gmax
curves than medium-organic. The magnitude of r0oaffects the position of the G/Gmax curves both for
medium-organic and high-organic sands, Fig. 6a. In
Fig. 2 Variation of (a) organic content, OC, with liquid limit,
LL, and (b) compression index, Cc, with void ratio, eo, for
model organic soils
Fig. 3 Variation of (a) small-strain shear modulus, Gmax, and
(b) small-strain damping ratio, DTmin, with isotropic effective
stress, r0o, for natural calcareous organic sands
Geotech Geol Eng
123
particular, at a given shear strain level, c, an increase
of r0o results in an increasing linearity of G/Gmax
curves. The influence of r0o on G/Gmax-c curves is
limited at stress levels lower than r0oinsitu or of the
order of 90 to 140 kPa and diminishes at stresses
exceeding the latter range.
The variation of small-strain damping ratio,
DTmin, with isotropic effective stress, r0o, at 24 h
confinement time is plotted in Fig. 3b. DTmin remains
either constant, or decreases with increasing r0o. The
range of DTmin values is from 1.4% to 1.8% for the
medium-organic and from 2.5% to 4.1% for the high-
organic sands at c B 10-5 for the range of stresses
examined (r0o = 30–400 kPa). These ranges com-
pare with the corresponding range of DTmin between
0.7% and 1.9% for inorganic sands containing fines
Fig. 4 Comparison of normalized shear modulus, G/[g(e) 9
h(OC) 9 k(CC)], obtained from Eq. 2 with the corresponding
results from the tests on natural calcareous organic sands
Fig. 5 Variation of (a) normalized shear modulus, G/Gmax,
and (b) damping ratio, DT, with shear strain, c, for natural
calcareous organic sands at effective isotropic confining stress,
r0o, approximately equal to or higher than insitu stress, r0oinsitu
Fig. 6 Effect of effective isotropic confining stress, r0o, on
variation of (a) normalized shear modulus, G/Gmax, and (b)
damping ratio, DT, with shear strain, c, for natural calcareous
organic sands S4 and S6
Geotech Geol Eng
123
(Tika et al. 2004). A small increase in DTmin of
cemented over non-cemented sands has been reported
in literature (Chiang and Chae 1972; Saxena et al.
1988). Since, however, all sands tested in this work
contained calcium carbonate, the higher DTmin values
observed for the high-organic sands indicate that the
organic matter starts to have an influence on DTmin at
an organic content exceeding 25% approximately.
The variation of damping ratio versus shear strain
is plotted in Fig. 5b. The DT versus c curves for the
medium-organic sands are within the range of
corresponding curves presented in literature for
inorganic sands containing fines (Kallioglou 2003),
whereas for the high-organic sands exhibit higher
linearity than the medium-organic sands. The mag-
nitude of r0o affects the position of the DT curves
both for medium-organic and the high-organic sands
and at a given shear strain level, c, with an increase of
r0o resulting in a decrease of DT, Fig. 6b.
5.2 Organic Cohesive Soils
5.2.1 Natural Intact Organic Soils
Figures 7a and 8a present the variation of small-strain
shear modulus, Gmax, with isotropic effective stress,
r0o, at 24 h confinement time for the tested medium-
organic and high-organic cohesive soils respectively. It
should be noted that the tests on the high-organic soils
of this group were conducted on small size specimens
and the c values (c = 53 9 10-6 – 230 9 10-6) are
up to two orders higher than the corresponding values
for the large size specimens of medium-organic soils
(c = 1.2 9 10-6 – 10 9 10-6). The insitu mean
effective stress, r0oinsitu, also indicated in the above
figures, was determined using the value of the coeffi-
cient of earth pressure at rest, ko, estimated either using
Jaky’s (1944) equation for normally consolidated clays
(1 – sin u0), or as function of PI and OCR for
overconsolidated clays (Brooker & Ireland, 1965).
The angle of shearing resistance, u0, was evaluated
from the empirical correlation of u0 and PI (Kenney
1959). A linear relation between Gmax and r0o in a log-
log plot is observed for each tested soil for both normal
consolidation and overconsolidation states.
Kallioglou et al. (2008) studied the small-strain
stiffness of natural intact inorganic cohesive soils
with calcium carbonate content less than 5% and
expressed Gmax by the following equation:
Gmax ¼ 6290� 80� PIð Þ � r00:50o � e�0:63 ð3Þ
where: Gmax and r0o are expressed in kPa and PI as
percentage values (%). Kallioglou et al. (2008) also
confirmed that the above equation holds for the
laboratory Gmax value of natural intact inorganic
cohesive soils containing calcium carbonate up to a
content of 25%.
To account for the effect of stress history, void
ratio and soil composition and plasticity on Gmax,
Gmax values at r0oinstitu for large size specimens only
were normalized by the void ratio and plasticity index
functions of the above equation and plotted versus
r0oinstitu at c B 10-5 in Fig. 9. As indicated in the
Fig. 7 Variation of (a) small-strain shear modulus, Gmax, and
(b) small-strain damping ratio, DTmin, with isotropic effective
stress, r0o, for medium-organic cohesive soils
Geotech Geol Eng
123
above figure, at a given r0oinstitu, the medium-organic
soils (OC = 6–15% and CC = 0–23%) showed
either comparable, or lower values of normalized
Gmax than the estimated from the above equation.
The results for the shear modulus at high strains
are shown in Figs. 10a and 11a & b for the medium-
organic and the high-organic cohesive soils respec-
tively at an effective isotropic stress approximately
equal to or higher than r0oinsitu. For the high-organic
soil C7s the measured G values were normalized by a
Gmax value at c = 7 9 10-5, because of the plateau
observed in G-c curve at strains c\ 2 9 10-4,
Fig. 11a. In the above figures, the G/Gmax-c curves
are compared with those presented in literature for
Fig. 8 Variation of (a) small-strain shear modulus, Gmax, and
(b) small-strain damping ratio, DTmin, with isotropic effective
stress, r0o, for high-organic cohesive soils
Fig. 9 Variation of normalized small-strain shear modulus,
Gmax/[f(e) 9 g(PI)], with insitu mean effective stress, r0oinsitu,
for natural organic cohesive soils
Fig. 10 Variation of (a) normalized shear modulus, G/Gmax,
and (b) damping ratio, DT, with shear strain, c, for medium-
organic cohesive soils at effective isotropic confining stress,
r0o, approximately equal to or higher than insitu stress, r0oinsitu
Geotech Geol Eng
123
inorganic cohesive soils (Kallioglou et al. 2008;
Vucetic and Dobry, 1991; Sun et al. 1988). The
degradation curves follow the trend of increasing
linearity with increasing plasticity; in particular for
the medium-organic soils, the G/Gmax curves either
agree with, or exhibit a higher linearity than the
corresponding curves, proposed in literature for soils
of the same plasticity, whereas for the high-organic
soils the G/Gmax curves exhibit a higher linearity for
c\ 10-3. The magnitude of r0o was observed to have
a negligible effect on the position of the G/Gmax
curves for the stress range studied (r0o = 110–
400 kPa).
The variation of small-strain damping ratio,
DTmin, with isotropic effective stress, r0o, at 24 h
confinement time is plotted in Figs. 7b and 8b for the
medium-organic and the high-organic cohesive soils
respectively. DTmin either remains constant or
decreases with increasing r0o. The range of DTmin
values is between 1.8% and 4.2% for the medium-
organic cohesive soils at c B 10-5 and at r0oinsitu.
This range is similar with the corresponding of
DTmin = 1.4–4.7% for inorganic cohesive soils (Kal-
lioglou et al. 2008). No clear effect of the organic
content on DTmin is observed possibly due to
different composition of natural soils tested.
The DT versus c curves for the medium-organic and
high-organic cohesive soils, shown in Figs. 10b and
11c respectively, indicate that there is no clear effect of
soil plasticity on them. Moreover, these curves differ
significantly from the corresponding curves presented
in literature for inorganic cohesive soils on the basis of
soil plasticity (Vucetic and Dobry 1991).
5.2.2 Model Reconstituted Organic Soils
Figure 12a presents the variation of small-strain shear
modulus, Gmax, with isotropic effective stress, r0o, at
24 h confinement time for the tested model organic
cohesive soils as well as kaolinite for comparison. A
linear relation between Gmax and r0o, in a log–log plot,
is observed for each tested soil for both normal
consolidation and overconsolidation states. At a given
r0o, Gmax decreases with increasing organic content
above 5%. The addition of salt (NaCl) in soil P-10-B
results in an increase of Gmax to values equal or above
the corresponding for soil P-10-A and even kaolinite.
The normalized shear modulus, G/Gmax, is plotted
versus shear strain in Fig. 13a at r0o = 90–110 kPa,
apart from soil P-20 for which data at r0o = 40 kPa
only were available. The degradation curves of the
model organic cohesive soils are practically coincident
with that of kaolinite, irrespectively of organic content
(OC = 5–20%) and stress level. This behaviour is in
agreement with the results on natural medium-organic
cohesive soils showing either negligible or small effect
of medium organic content (OC = 6–15%) on degra-
dation curves. It can thus be concluded that for
medium-organic soils the plasticity of the mineral soil,
rather than of the organic soil, controls the position and
Fig. 11 Variation of (a) shear modulus, G, (b) normalized
shear modulus, G/Gmax, and (c) damping ratio, DT, with shear
strain, c, for high-organic cohesive soils at effective isotropic
confining stress, r0o, approximately equal to or higher than
insitu stress, r0oinsitu
Geotech Geol Eng
123
shape of G/Gmax curves. The magnitude of r0o has
either a slight (P-10-B), or a negligible effect (P-5 & P-
10-A) on the position of the G/Gmax curves for the
stress range studied (r0o = 90–200 kPa), Fig. 14a.
The presence of salt (NaCI) in soil P-10-B results in
higher degradation of G/Gmax curve.
The variation of small-strain damping ratio, DTmin,
with isotropic effective stress, r0o, at 24 h confine-
ment time is plotted in Fig. 12b. A slight to moderate
decrease of DTmin with increasing r0o is observed.
DTmin increases with increasing organic content and
this may be attributed to the increasing flexibility of
the soil mass due to the presence of organic fibers.
However, the addition of salt (NaCl) in soil P-10-B
results in a decrease of DTmin. These observations
indicate a significant effect of both the organic matter
content and pore water chemistry on DTmin. The range
of DTmin values at c B 10-5 is between 2.1% and
4.8% and is similar with the corresponding of natural
intact organic cohesive soils tested, Fig. 7b.
The variation of damping ratio versus shear strain
for model organic cohesive soils is plotted in Fig. 13b.
The damping ratio curves shift to higher values of DT
with increasing organic content. Similarly with the
natural organic cohesive soils, these curves differ
from the corresponding curves, presented in literature
on the basis of soil plasticity (Vucetic and Dobry
1991). The magnitude of r0o has slight or negligible
effect on the position of the DT versus c curves for the
stress range studied (r0o = 90–200 kPa), Fig. 14b.
Fig. 12 Variation of (a) small-strain shear modulus, Gmax, and
(b) small-strain damping ratio, DTmin, with isotropic effective
stress, r0o, for model organic cohesive soils and kaolinite
Fig. 13 Variation of (a) normalized shear modulus, G/Gmax,
and (b) damping ratio, DT, with shear strain, c, at effective
isotropic stress r0o = 40–110 kPa for model organic cohesive
soils and kaolinite
Geotech Geol Eng
123
5.3 Peats
Figure 15a presents the variation of small-strain shear
modulus, Gmax, with isotropic effective stress, r0o, at
24 h confinement time for the tested peats. The insitu
mean effective stress, r0oinsitu, also indicated in the
above figure, was determined using the value of the
coefficient of earth pressure at rest, ko, estimated using
Jaky’s (1944) equation for normally consolidated soils
(1 - sin u0). A linear relation between Gmax and r0o, in
a log–log plot, is observed for both soils. The Gmax
values of the tested peats (Gmax = 15.9–27.8 MPa at
r0o = 39–374 kPa and wo = 101% for P1 and
Gmax = 60.7–73.1 MPa at r0o = 111–396 kPa and
wo = 58% for P2s) are higher than the corresponding
values reported in literature for peats (7–11.3 MPa for
r0o = 66–200 kPa and w = 152–240% by Boulanger
et al. 1998, and 0.15–11 MPa for r0o = 1.5–12.5 kPa
by Kramer 1993). Obviously, the high Gmax values of
peats P1 and P2s may be attributed to their lower water
contents, as well as to the fact that for most of the stress
range examined, apart from the highest ones, the peats
were overconsolidated.
The results for the shear modulus at high strains
for both peats are shown in Fig. 16a & b at an
effective isotropic stress approximately equal to
r0oinsitu. The measured G values were normalized by
the G value at c = 2 9 10-5, Fig. 16a. Both peats
exhibit strong linearity, similar to those reported for
very plastic clays (PI = 200%) and other peats in
literature, Fig. 16b. The effect of r0o on the position
of the G/Gmax curves was not studied due to the upper
limit of confining pressure (ro \ 700 kPa) in the
resonant-column apparatus.
Fig. 14 Effect of effective isotropic confining stress, r0o, on
variation of (a) normalized shear modulus, G/Gmax, and (b)
damping ratio, DT, with shear strain, c, for model organic
cohesive soils Fig. 15 Variation of (a) small-strain shear modulus, Gmax, and
(b) small-strain damping ratio, DTmin, with isotropic effective
stress, r0o, for peats
Geotech Geol Eng
123
The variation of small-strain damping ratio,
DTmin, with isotropic effective stress, r0o, at 24 h
confinement time is plotted in Fig. 15b. As shown,
DTmin decreases with increasing r0o. The range of
DTmin values is between 2.3% and 2.4% at c B 10-5
and at r0oinsitu and agrees with the values for peats
given in literature.
The variation of damping ratio versus shear strain
for peats is plotted in Fig. 16c. As shown, the peats
exhibit high linearity in DT curves. These curves
agree with the lower bound curve presented in
literature for peats and high plastic clays with
PI = 200% (Vucetic and Dobry 1991).
5.4 Effect of Secondary Consolidation Time
on Shear Modulus and Damping Ratio
For all the organic soils tested in this work, primary
consolidation was completed within 24 h. The effect of
time on Gmax and DTmin was studied by conducting
tests of long duration under constant confining stress
and recording the variation of Gmax and DTmin with
secondary consolidation time of soils. A linear increase
of Gmax with logarithm of secondary consolidation
time was observed. DTmin either remained constant or
decreased with logarithm of secondary consolidation
time, Fig. 17. The influence of secondary consolida-
tion time on Gmax can be expressed in terms of the
following parameter NG (Marcuson and Wahls 1972):
NG ¼Gmax tð Þ � Gmax t ¼ 48 hð Þlog t
48 h
� �� Gmax t ¼ 48 hð Þ
ð4Þ
where Gmax tð Þ: small-strain shear modulus at a given
time t; Gmax t ¼ 48 hð Þ: small-strain shear modulus at
48 h consolidation time.
Fig. 16 Variation of (a) shear modulus, G, (b) normalized
shear modulus, G/Gmax, and (c) damping ratio, DT, with shear
strain, c, for peats at effective isotropic confining stress, r0o,
approximately equal to insitu stress, r0oinsitu
Fig. 17 Variation of (a) small-strain shear modulus, Gmax, and
(b) small-strain damping ratio, DTmin, with consolidation time
at effective isotropic confining stress, r0o, equal to and higher
than insitu stress, r0oinsitu
Geotech Geol Eng
123
Similarly, the following parameter ND can be used
in order to quantify the effect of secondary consol-
idation time on DTmin:
ND ¼DTmin t ¼ 48 hð Þ � DTmin tð Þlog t
48 h
� �� DTmin t ¼ 48 hð Þ
ð5Þ
The variation of parameter NG with plasticity
index of both natural and model organic soils at the
insitu mean effective stress, r0oinsitu, is shown in
Fig. 18. The results of medium-organic soils are
within the range of inorganic clays given in literature,
whereas high-organic soils exhibit higher values of
NG. The effect of time on DTmin for the above soils
was either negligible (ND = 0 for S1 & C3) or small
(ND = 0.15 for C2 & P1, ND = 0.21–0.25 for S6, C1
& C7s) irrespectively of organic content.
The variation of NG with of secondary compres-
sion index, Ca, obtained from RC tests for model and
natural organic cohesive soils, as well as peats is
presented in Fig. 19. An increase of NG with
increasing Ca is observed.
The influence of secondary consolidation time on
G/Gmax-c and DT-c curves was studied for the natural
high-organic cohesive soil C7s, Fig. 20. It is shown
that with increasing consolidation time the G/Gmax
curve moves upwards, whereas the DT curve down-
wards. This influence is stronger for DT curves and it
diminishes for G/Gmax curves with increasing con-
solidation time.
5.5 Linear Elastic and Volumetric Threshold
Shear Strains
The variation of linear elastic and volumetric thresh-