Documente Academic
Documente Profesional
Documente Cultură
112]
The clay-interface shear resistance is an important parameter for the design of offshore pipelines, which
slide on the seabed as a result of thermally induced expansion, contraction and lateral buckling. This
paper presents a methodology for characterising the clay-interface resistance and quantifying the
effect of drainage and consolidation during or in-between shearing episodes. Models for describing
the clay-interface resistance during planar shearing are presented and compared to test data for a range
of drainage conditions from drained to undrained and including the case of episodic consolidation.
The test data are from two series of interface shear box tests carried out on marine clays. The effects
of normal stress level (in the low stress range), overconsolidation and interface roughness are also
examined.
18
Downloaded by [ Universidade de Sao Paulo] on [22/02/18]. Copyright © ICE Publishing, all rights reserved.
INTERFACE SHEAR BOX TESTS TO ASSESS AXIAL PIPE–SOIL RESISTANCE 19
effects. The influence of shearing rate has also been stress failure envelope to account for reducing friction angle
associated with a change in mode of shearing in soils with increasing stress level; (b) a drained–undrained tran-
exhibiting a sliding mode of shear at slow rate, characterised sition, associated with pore pressure generation and con-
by a brittle response and low residual shear resistance result- solidation; (c) a transition in interface strength associated
ing from reorientation of platy particles along the shear with interface roughness; (d) a wedging effect due to the
surface (Tika et al., 1996; Lemos & Vaughan, 2000; Fearon curved pipe surface. The first three elements of the frame-
et al., 2004). As first recognised by Lupini et al. (1981), soils work can be quantified for a particular soil through a series
with more rotund particles exhibit a turbulent mode of shear, of carefully specified interface shear box tests. This will be
characterised by high residual shear resistance and features of discussed in detail in the later section entitled ‘Testing
classical critical state soil mechanics such as effect of stress methodology’. The models used to describe the drained,
history. Owing to the relatively small displacements applied undrained and partially drained soil-interface resistance are
during shear box testing (tens of millimetres), it is possible described below.
that the low residual strengths that can occur in clays that
have platy particles may not be exhibited. The work pres-
ented in this paper is based on a critical state soil mechanics
model of soil behaviour, without the additional aspect of Drained and undrained soil-interface resistance
low residual strengths due to the alignment of platy particles Drained resistance. The drained interface resistance is a
during shearing. This approach is adequate to capture the function of the interface roughness and is described by an
behaviour apparent in a wide range of marine soils from effective stress failure criterion in which the friction varies
offshore regions throughout the world when tested at low with the effective normal stress, σn′ , in the low stress range. A
stresses (Najjar et al., 2007; Hill et al., 2012; White et al., power law relationship may be used to represent the variation
2012). of shear strength with σn′
In this paper, the authors present a quantitative analysis of
the consolidation process taking place during and in-between τ d ¼ aðσ′n Þb μmax σ′n ð1Þ
shearing events, which has not been reported before. The where σn′ is the effective normal stress on the shearing
analysis is based on the critical state framework and is interface, expressed in kPa, and a, b and μmax are constants
applied to data obtained from a series of interface shear varying with soil type and interface roughness. During
box tests for two marine clays tested at low normal stress. The drained shearing, no excess pore pressure is developed so
analysis also incorporates the effects of roughness and the initial normal stress, σn0 ′ , is also the normal stress at
normal stress level that are listed above. failure. The parameter μmax specifies an upper limit on
the stress ratio τd/σn′ , to avoid unrealistically high friction
coefficients as σn′ approaches zero. The drained stress
FRAMEWORK FOR ASSESSING AXIAL PIPE–SOIL path during interface shearing is shown schematically
RESISTANCE in Fig. 2(a) (path AB). The corresponding path in the
A framework for assessing axial pipe–soil resistance semi-logarithmic space of void ratio, e, plotted against ln
was proposed by White et al. (2012) and Hill et al. (2012) σn′ terminates on the critical state line (CSL), which has
that incorporates the effects of drainage conditions and slope λ (Fig. 2(b)). Drained shearing up to the critical
uses concepts of critical state soil mechanics to describe state induces a change in volume, which can be expressed
the changing shear strength of a pipe–soil interface through in terms of the void ratio, Δemax, or the volumetric strain,
episodes of shearing and consolidation. Four elements are Δεv,max, with Δemax = Δεv,max(1 + e0), where e0 is the initial
integrated in the framework (Fig. 1): (a) a curved effective void ratio.
Wedging factor
ζ = f(w/D)
Friction coefficent:
Pipe: T/W = ζ(τ/σ'n0)
Interface: τ /σ'n0 = tan δ
Embedment, w/D
Pipe wedging:
ζ = f(w/D) –ve
Virgin strength
T/W increases with Excess pore (dilatant) ‘Viscous’ rate
embedment (w/D) pressure effect (f(v))
and roughness (rn) +ve
Friction:
τ /σ'n = µ = tan δ Cr i t
ic
Rou stat al
gh e
δ=φ
l
na
(Cycles of sliding,
tio
consolidation and
si
Sm Virgin strength
an
h (contractile)
icti l
R ed a c e fr leve Dra
ined
with ucing i terf stress
g in
dec nt
reas erface d u cin asing
ing fr
roug iction Re incre Time
h
hne wit (drainage,
Interface ss consolidation)
Pipe weight, or
roughness, rn Und
rain
normal stress
(Or sliding velocity, ed
v, for given distance)
Fig. 1. Mechanisms controlling axial pipe–soil resistance (modified from Hill et al., 2012)
Downloaded by [ Universidade de Sao Paulo] on [22/02/18]. Copyright © ICE Publishing, all rights reserved.
20 BOUKPETI AND WHITE
τ
For the case of an overconsolidated material, the
undrained resistance increases as a function of overconsoli-
dation ratio (OCR)
µmax
τ u ¼ ðτ u ÞNC OCRbΛ ð6Þ
τ = a(σ'n )b
where NC stands for normally consolidated, Λ = (λ κ)/λ
and k is the slope of the one-dimensional unloading–
reloading response in the e–ln σn′ space. Note that for b = 1
B
(i.e. a stress-independent friction angle), equation (6) is the
classical critical state relationship for soil strength (Schofield
∆umax & Wroth, 1968; Wroth, 1984; Wood, 1990), or the Shansep
C equation given by Ladd et al. (1977).
Downloaded by [ Universidade de Sao Paulo] on [22/02/18]. Copyright © ICE Publishing, all rights reserved.
INTERFACE SHEAR BOX TESTS TO ASSESS AXIAL PIPE–SOIL RESISTANCE 21
0·7 Table 1. Summary of soil parameters
Planar shearing model
Normalised shear stress, τ /σ'n0
Downloaded by [ Universidade de Sao Paulo] on [22/02/18]. Copyright © ICE Publishing, all rights reserved.
22 BOUKPETI AND WHITE
respectively. The specimen was then transferred to the upper resistance in this paper) mobilised at large displacements.
half of the shear box, on top of the interface plate fitted into A displacement rate of 0·001 mm/s was used for the slow
the lower half of the box. The specimen was consolidated to case, which is likely to be drained, whereas displacement rates
a specified normal stress. In the case of an overconsolidated of 0·01 and 0·03 mm/s were applied for the two intermediate
specimen, the initial consolidation was followed by unload- cases. After the second shearing cycle was completed, a final
ing to the required normal stress level, and the specimen was horizontal displacement of 10 mm was imposed.
allowed to swell prior to the shearing stage.
Test no. Inter- Consolidation Normal stress, Velocity: mm/s No. of cycles Description
face ′ : kPa
stress, σnc ′ : kPa
σn0
1 R 4 4 0·001 2 Slow base case (SBC)
2 S 4 4 2 SBC with reduced roughness
3 R 8 8 2 SBC at higher stress
4 2 2 2 SBC at lower stress
5 4 4 0·1 20 Fast base case (FBC)
6 S 4 4 20 FBC with reduced roughness
7 R 8 8 20 FBC at higher stress
8 2 2 20 FBC at lower stress
9 4 4 0·03 2 Intermediate base case
10 4 4 0·01 2 Intermediate base case
11 8 4 0·1 20 FBC at OCR = 2
12 12 4 0·1 20 FBC at OCR = 3
Note: R, rough with roughness Ra . 80 μm; S, smooth with roughness Ra = 1·94 μm.
Downloaded by [ Universidade de Sao Paulo] on [22/02/18]. Copyright © ICE Publishing, all rights reserved.
INTERFACE SHEAR BOX TESTS TO ASSESS AXIAL PIPE–SOIL RESISTANCE 23
Time: min 10
0·1 1 10 100 1000 10 000
τd
0
t50
5
0·5
–5
–10
–10 –5 0 5 10
1·0 Horizontal displacement: mm
(a)
Fig. 5. Consolidation response of soil A (test 5)
10
Downloaded by [ Universidade de Sao Paulo] on [22/02/18]. Copyright © ICE Publishing, all rights reserved.
24 BOUKPETI AND WHITE
for the undrained tests are listed in Table 3 (tests 5, 7 and 8) Using equation (2) for the undrained interface resistance,
and are in the range 0·35–0·45 for soil A and 0·33–0·36 the pore pressure ratio, Δumax/σn0 ′ , has been estimated for
for soil B. These values are smaller than the corres- each undrained test and is equal to 0·6 on average for both
ponding drained values by a factor of 0·47 (on average, soils (see Table 1). This leads to a ratio (σn0 ′ Δumax)/
see Table 1). ′ 0·4, which is slightly less than the ratio of undrained
σn0
to drained strength due to the non-linearity of the effective
8 stress failure envelope. The corresponding value of the state
parameter, S, determined using equation (3), is listed in Table 1
and is approximately S 0·85. Similarly as for the drained
resistance, the undrained interface resistance measured with
Residual stress, τ d: kPa
6
τ = 1·04 (σ'n)0·84 the smooth interface is smaller than the one obtained with the
τ = 0·87 (σ'n)0·90
rough interface (see tests 5 and 6 in Table 3).
4 For both soils, an increase in undrained interface resis-
tance with OCR was observed and can be modelled using
equation (6) (see Fig. 9). From curve fitting of the data
Soil A
2 plotted in Fig. 9 it can be inferred that for soil A, Λ = 0·68,
Curve fit for soil A
whereas for soil B, Λ = 0·57 (curve fitting is based on three
Soil B
data points for each soil and yielded R 2 = 0·99 for both soils).
Curve fit for soil B
0 These values fall within the range of values reported for
0 2 4 6 8 10 12 a wide range of clays (0·5 , Λ , 0·8; Schofield & Wroth,
Normal effective stress, σ'n: kPa 1968).
(a)
6
the undrained and drained values. The residual stress ratio
measured during ISB tests carried out at various velocities
4 spanning the range of drained to undrained response is
plotted in Fig. 10 with respect to the shearing velocities. The
data correspond to tests 2, 5, 9 and 10 performed with an
2 initial normal effective stress of 4 kPa and can be approxi-
Gulf of Mexico clay
Tilt table tests from
mated by an equation with a similar format as equation (8)
Najjar et al. (2007)
τ τ τ τ nv
0 ¼ 05ðv50 =vÞ
0 2 4 6 8 10 12 σ′n0 σ′n0 drained σ′n0 drained σ′n0 undrained
Normal effective stress, σ'n: kPa ð12Þ
(b)
where v50 is the velocity for which the stress ratio is midway
Fig. 7. Failure envelope: (a) drained interface resistance for soils A between the undrained and drained values and nv is a
and B; (b) drained shear strength of Gulf of Mexico clay (from Najjar constant. For soil A and soil B the values of v50 are 0·01 and
et al., 2007) 0·006 mm/s, respectively, and nv = 1·5.
1 2 3 4 5 6 7 8 9 10
Note: R, rough with roughness Ra . 80 μm; S, smooth with roughness Ra = 1·94 μm.
*At the end of the first cycle pair.
†At the end of the last cycle after episodic consolidation.
‡Consolidated under 8 kPa and unloaded to 4 kPa (OCR = 2).
§Consolidated under 12 kPa and unloaded to 4 kPa (OCR = 3).
Downloaded by [ Universidade de Sao Paulo] on [22/02/18]. Copyright © ICE Publishing, all rights reserved.
INTERFACE SHEAR BOX TESTS TO ASSESS AXIAL PIPE–SOIL RESISTANCE 25
10 1·0
Soil A
0·8
0·6
0
0·4
–5 Equation (12)
0·2
–10 0
–10 –5 0 5 10 0·0001 0·001 0·01 0·1 1
Horizontal displacement: mm Velocity: mm/s
(a) (a)
10 1·0
Soil B
8 0·8
6
0·6
4
0·4
τu
2
0·2 Equation (12)
0
0
0 10 20 30 40 50
0·0001 0·001 0·01 0·1 1
Cumulative horizontal displacement: mm
Velocity: mm/s
(b)
(b)
The interface resistance data are plotted again in Fig. 11 where Δumax is the pore pressure generated during undrained
with respect to normalised time T = cvt/(hs)2, where t is the shearing to the critical state, T* is the dimensionless time
time corresponding to 40 mm shearing displacement (two defined as T* = cvt/R 2 and βi are the roots of J0(β) = 0, with J0
cycles), when the residual stress is measured. The cv values for the Bessel function of order 0 (the six first roots are given by
each clay, estimated from the consolidation response, are listed Carslaw & Jaeger, 1959). Fig. 12 shows the interface resis-
in Table 1. The shear band thickness, hs, is chosen such that tance calculated using equations (13) and (2) as a function of
the data corresponding to mid-range between the drained dimensionless time T*, together with the test data. The
and undrained values matches the prediction from the planar dimensionless time T* was calculated using the specimen
shearing model of Randolph et al. (2012). This leads to values radius, R = 36 mm, and the values of cv listed in Table 1.
of hs = 5 mm for soil A and hs = 6 mm for soil B. The The agreement between model and data is reasonably
prediction from the planar shearing model plotted in Fig. 11 good, showing that the radial flow solution captures adeq-
was obtained using the drained friction coefficient, the value uately the transition from undrained to drained regime.
Downloaded by [ Universidade de Sao Paulo] on [22/02/18]. Copyright © ICE Publishing, all rights reserved.
26 BOUKPETI AND WHITE
0·8 1·0
Residual stress ratio, τres/σ'n0 Soil A Soil A
0·8
0·6
0·4
0·4
0 0
0·01 0·1 1 10 100 1000
0·0001 0·001 0·01 0·1 1 10
Normalised time, T = cvt/(hs)2
Normalised time, T* = cvt/R2
(a)
(a)
0·8 1·0
Soil B Soil B
Residual stress ratio, τres/σ'n0
0·8
0·6
0·4
0·4
0·2 Planar shearing model
Equation (8) with n = 0·35 0·2
Equation (8) with n = 1·5 Solution for radial flow
0
0
0·01 0·1 1 10 100 1000
0·0001 0·001 0·01 0·1 1 10
Normalised time, T = cvt/(hs)2
Normalised time, T* = cvt/R2
(b)
(b)
Fig. 11. Variation in interface resistance with dimensionless time
T = cvt/(hs)2: (a) soil A; (b) soil B Fig. 12. Variation in interface resistance with dimensionless time
T* = cvt/R 2: (a) soil A; (b) soil B
Downloaded by [ Universidade de Sao Paulo] on [22/02/18]. Copyright © ICE Publishing, all rights reserved.
INTERFACE SHEAR BOX TESTS TO ASSESS AXIAL PIPE–SOIL RESISTANCE 27
6 2·4
Cycles 1–2 Soil A Equation (9)
σ'n0 = 8 kPa
2
1·8
0 Equation (9)
1·6
σ'n0 = 4 kPa
–2 1·4
Normal stress: kPa
–4 1·2 4
8
1·0
–6 0 2 4 6 8 10 12
–6 –4 –2 0 2 4 6
Consolidation cycle number, N
Horizontal displacement: mm
(a)
(a)
3·0
Soil B
6
Fast cycles 19–20 Equation (9)
Equation (9)
2 σ'n0 = 8 kPa
2·0
0
–2 1·5
Normal stress: kPa
4
–4 8
1·0
0 2 4 6 8 10 12
–6
Consolidation cycle number, N
–6 –4 –2 0 2 4 6
(b)
Horizontal displacement: mm
(b)
Fig. 14. Normalised resistance plotted against consolidation cycle
number for fast episodic tests: (a) soil A; (b) soil B
Fig. 13. Consolidation hardening for soil A: (a) fast test (test 5);
(b) comparison of final cycles of fast test and slow test (test 1)
25 Normal stress: kPa
using equation (10). The prediction from equation (10) is 4
8
compared with experimental measurements for two tests on 20
soil A in Fig. 15. The values of the parameters S, λ and κ are
Volumetric strain, ∆εv: %
Downloaded by [ Universidade de Sao Paulo] on [22/02/18]. Copyright © ICE Publishing, all rights reserved.
28 BOUKPETI AND WHITE
Soil A of vertical settlement. From Fig. 16 it can be seen that
5·5 End of consolidation the critical state reached at the end of the cyclic tests
End of slow test (settlement) with episodic consolidation agrees reasonably well with
5·0 End of slow test (water content)
λ End of fast episodic test (settlement) that obtained from the drained tests. Again, more scatter is
4·5
End of fast episodic test (water content) observed at low normal stress.
Void ratio, e
λ = 0·12 ratio is slightly larger than the ratio predicted by the original
1·7 Cam-Clay model (Roscoe & Schofield, 1963; Schofield
NCL
1·6
& Wroth, 1968), ru = e Λ, which gives the values ru = 2·0 for
S = 0·84
soil A and ru = 1·8 for soil B (using the values of Λ listed in
1·5 Table 1). The ratio predicted by the modified Cam-Clay
CSL
model (Roscoe & Burland, 1968) is ru = 2Λ, and yields the
1·4 values ru = 1·5 for soil A and ru = 1·6 for soil B, which again
1 10
are smaller than the measured values.
Effective vertical stress, σ'v: kPa
This suggests that the effective stress paths followed by
(b) soils A and B during the undrained ISB test differ from that
predicted by the Cam-Clay models. In fact, the undrained
Fig. 16. Normal compression line (NCL) and critical state line
response of soil A and B exhibits a peak in shear stress
(CSL): (a) soil A; (b) soil B
followed by softening (see Fig. 8), which cannot be modelled
using the standard Cam-Clay models and a normally con-
consolidation, adopting a soil specific gravity of Gs = 2·7. solidated initial state. During softening, additional excess
Fig. 16 indicates that the data can be reasonably well fitted by pore pressure is generated, causing the stress path to track to
a straight line, the NCL, with the best-fit parameters yielding a lower effective stress level and therefore a lower shear stress
a slope λ = 0·64 for soil A and λ = 0·12 for soil B. The latter after failure is reached. This is consistent with the effective
value of λ falls within the range of typical values observed for stress paths identified in undrained interface shear tests that
a wide range of clays (Schofield & Wroth, 1968). However, include internal pore pressure measurements, reported by
soil A, which has a high liquidity index (LI 3·9), exhibits a Tsubakihara & Kishida (1993).
relatively high compressibility (high λ). This is consistent with The value of the parameter S determined from the NCL
data reported by Hong et al. (2010) for reconstituted clays, and CSL in the e–ln σn′ space can be used to estimate the
which show that the slope λ increases as the water content resistance ratio τu/τd using equation (4). The predicted values
increases above the liquid limit (data show slope as high as are τu/τd = 0·50 for soil A and τu/τd = 0·47 for soil B and
λ 0·6 in the low stress range). The consolidation data compare reasonably well with the measured values reported
plotted in Fig. 16 exhibit more scatter in the low stress range, in Table 1.
where measurement error becomes significant in interpret-
ation of the measured vertical settlement.
The CSL was estimated from the drained residual states CONCLUDING REMARKS
reached during the slow ISB test (tests 2, 3 and 4). The void A series of ISB tests conducted on two marine clays have
ratio at the residual state was calculated in two ways: (a) from demonstrated the influence on the measured soil-interface
the volumetric strain, obtained by dividing the vertical shear resistance of normal stress level, overconsolidation
settlement measured after 40 mm shearing (two cycles) by ratio, interface roughness and drainage – both during and
the specimen height at the start of shearing; (b) from the in-between shearing episodes. The high interface resistance
water content at the end of the test (with Gs = 2·7). Fig. 16 (friction angle . 35°) measured at the shearing displace-
shows that the CSL has a similar slope, λ, as the NCL, ments achieved in the ISB suggests a turbulent mode of shear
consistent with conventional critical state theory and that and the conclusions given below apply to that case. The
the two methods of estimating the void ratio at the critical drained failure envelope in the normal stress range between
state yield reasonably close results for soil A (there is some 2 and 8 kPa was found to be slightly non-linear, leading to
scatter in the results for soil B). Also plotted in Fig. 16 is the ′ )
increases in resistance (as measured by the stress ratio τ/σn0
final state of the specimens subjected to cyclic shearing with by 15 to 25% as the normal stress decreases from 8 to 2 kPa.
episodes of consolidation. In this case, the total vertical The effect of OCR in the low OCR range (OCR 3) was
settlement measured during the consolidation periods is used well captured by a power law model similar to that proposed
to compute the volumetric strain and the final void ratio, in the critical state framework. Using a smooth interface
which is also determined from the final water content at the (Ra 2 μm) results in a lower soil-interface resistance com-
end of the test. No data point is shown for tests 5 and 8 on pared to the rough interface (Ra . 80 μm), with the decrease
soil B, as error in measurement led to negligible values in resistance varying with the soil type.
Downloaded by [ Universidade de Sao Paulo] on [22/02/18]. Copyright © ICE Publishing, all rights reserved.
INTERFACE SHEAR BOX TESTS TO ASSESS AXIAL PIPE–SOIL RESISTANCE 29
The effect of partial drainage and consolidation was t50 time for 50% consolidation
investigated for two types of scenario; one where partial con- Uc degree of consolidation
solidation occurs during continuous shearing, and the other v shearing velocity
v50 shearing velocity corresponding to strength midway
one where episodes of undrained shearing are interspersed between undrained and drained values
with periods of consolidation. For the case of consolidation W weight on pipe
during continuous shearing, it was found that the solution w pipe embedment
for consolidation during planar shearing at the boundary βi roots of J0(β) = 0, where J0 is the Bessel function of order 0
of an infinite soil body was not directly applicable to ISB δ interface friction angle
tests, due to drainage allowed to take place between the upper Δemax maximum change in void ratio
and lower half of the shear box. The solution for radial Δuav average excess pore pressure
dissipation of excess pore pressure at the base of the specimen Δumax maximum excess pore pressure
leads to a reasonable prediction of the transition from Δεv volumetric strain
Δεv,max maximum volumetric strain
undrained to drained regime as a function of dimensio-
ζ wedging factor
nless time, with the characteristic length being the radius of κ slope of elastic compression line
the sample. This solution provides a means of interpreting Λ exponent for effect of OCR on undrained shear strength
ISB tests to determine the coefficient of consolidation, cv, λ slope of NCL and CSL
of the soil tested. For the case of episodic shearing and μmax maximum value of residual friction ratio
consolidation, a simple critical state framework is shown σn′ effective normal stress
to capture well the changing interface strength with con- ′
σn0 initial effective normal stress at start of shearing
solidation cycles, allowing ISB tests to be interpreted in ′
σnu effective normal stress at critical state during undrained
a consistent framework – deriving critical state-type strength shearing
parameters – for application in design. Comparison of τ shear stress
τd drained soil-interface resistance
predicted and measured volumetric strain is less successful τres steady (residual) soil-interface resistance
due to soil outside the initial shear layer contributing to the τu undrained soil-interface resistance
shearing and consolidation process. τui undrained soil-interface resistance during first shearing
There remains some uncertainty on the volume of cycle
soil within the ISB that is involved in the shearing process, τuN undrained soil-interface resistance after N cycles of
and further research investigating the spatial variation in shearing and consolidation
excess pore pressure, changes in moisture content and ϕ soil friction angle
microstructure during shearing of soil along an interface is
recommended.
REFERENCES
Carneiro, D., White, D. J., Danziger, F. A. B. & Ellwanger, G. B.
ACKNOWLEDGEMENTS
(2015). Excess pore pressure redistribution beneath pipelines:
The research presented here forms part of the activities FEA investigation and effects on pipe–soil interaction.
of the Centre for Offshore Foundation Systems (COFS), Proceedings of the 3rd international symposium on frontiers in
currently supported as a node of the Australian Research offshore geotechnics, Oslo, Norway, pp. 369–374.
Council Centre of Excellence for Geotechnical Science and Carslaw, H. S. & Jaeger, J. C. (1959). Conduction of heat in solids.
Engineering (grant CE110001009) and through the Fugro Oxford, UK: Oxford University Press.
Chair in Geotechnics, the Lloyd’s Register Foundation Chair Cheuk, C. Y. & White, D. J. (2011). Modelling the dynamic em-
and Centre of Excellence in Offshore Foundations and the bedment of seabed pipelines. Géotechnique 61, No. 1, 39–57,
Shell EMI Chair in Offshore Engineering, which is held by http://dx.doi.org/10.1680/geot.8.P.148.
the second author. The authors would like to thank Andy Cocjin, M. L., Gourvenec, S. M., White, D. J. & Randolph, M. F.
(2014). Tolerably mobile subsea foundations – observations of
Hill, from BP, for supporting this work. performance. Géotechnique 64, No. 11, 895–909, http://dx.doi.
org/10.1680/geot.14.P.098.
Corfdir, A., Lerat, P. & Vardoulakis, I. (2004). A cylinder shear
NOTATION apparatus. Geotech. Testing J. 27, No. 5, paper GTJ11551.
a coefficient for drained shear strength Dietz, M. (2000). Developing an holistic understanding of interface
b exponent for drained shear strength friction using sand within the direct shear apparatus. PhD thesis,
cv consolidation coefficient The University of Bristol, Bristol, UK.
D pipe diameter Eid, H. T., Amarasinghe, R. S., Rabie, K. H. & Wijewickreme, D.
e0 initial void ratio (2015). Residual shear strength of fine-grained soils and soil–
f function describing the variation of the wedging factor solid interfaces at low effective stresses. Can. Geotech. J. 52,
with depth No. 2, 198–210.
Gs specific gravity Fearon, R. E., Chandler, R. J. & Bommer, J. J. (2004). An
H height of soil specimen investigation of the mechanisms which control soil behaviour
hs shear band thickness at fast rates of displacement. In Advances in geotechnical
N number of consolidation cycles (i.e. undrained shearing engineering: the Skempton conference (eds R. J. Jardine,
followed by consolidation) D. M. Potts and K. G. Higgins), pp. 441–452. London, UK:
n exponent in interface resistance as a function of Thomas Telford.
dimensionless time Ganesan, S., Kuo, M. & Bolton, M. (2014). Influences on pipeline
nv exponent in interface resistance as a function of velocity interface friction measured in direct shear tests. Geotech. Testing
Ra interface roughness J. 37, No. 1, paper GTJ20130008.
rn pipe roughness Gibson, R. E. & Henkel, D. J. (1954). Influence of duration of tests
ru ′ /σnu
effective normal stress ratio σn0 ′ at constant rate of strain on measured ‘drained strength’.
S state parameter Géotechnique 4, No. 1, 6–15, http://dx.doi.org/10.1680/geot.
T non-dimensional time for planar shearing model; 1954.4.1.6.
in Fig. 1, T is the axial load on the pipe Hill, A. J. & White, D. J. (2015). Pipe-soil interaction: recent
T50 non-dimensional time for 50% consolidation and future improvements in practice. Proceedings of the
T* non-dimensional time for radial flow model 3rd international symposium on frontiers in offshore geotechnics,
t time Oslo, Norway, pp. 387–392.
Downloaded by [ Universidade de Sao Paulo] on [22/02/18]. Copyright © ICE Publishing, all rights reserved.
30 BOUKPETI AND WHITE
Hill, A. J., White, D. J., Bruton, D. A. S., Langford, T., Meyer, V., Roscoe, K. H. & Burland, J. B. (1968). On the generalised stress–
Jewell, R. & Ballard, J. C. (2012). A new framework for axial strain behaviour of ‘wet’ clay. In Engineering plasticity (eds
pipe-soil interaction illustrated by a range of marine clay J. Heyman and F. A. Leckie), pp. 535–609. Cambridge, UK:
datasets. Proceedings of international conference on offshore Cambridge University Press.
site investigation and geotechnics, SUT, London, UK, Roscoe, K. H. & Schofield, A. N. (1963). Mechanical behaviour of
pp. 367–377. an idealised wet clay. Proceedings of the 2nd European conference
Ho, T. Y. K., Jardine, R. J. & Anh-Minh, N. (2011). on soil mechanics and foundation engineering, Wiesbaden,
Large-displacement interface shear between steel and granular Germany, pp. 47–54.
media. Géotechnique 61, No. 3, 221–234, http://dx.doi.org/ Schofield, A. N. & Wroth, C. P. (1968). Critical state soil mechanics.
10.1680/geot.8.P.086. Glasgow, UK: Blackie Academic.
Hong, Z. S., Yin, J. & Cui, Y. J. (2010). Compression behaviour of Skempton, A. W. (1985). Residual strength of clays in landslides,
reconstituted soils at high initial water contents. Géotechnique folded strata and the laboratory. Géotechnique 35, No. 1, 3–18,
60, No. 9, 691–700, http://dx.doi.org/10.1680/geot.09.P.059. http://dx.doi.org/10.1680/geot.1985.35.1.3.
Jardine, R., Chow, F., Overy, R. & Standing, J. (2005). ICP design Stark, T. D. & Eid, H. T. (1994). Drained residual strength of
methods for driven piles in sands and clays. London, UK: cohesive soils. ASCE J. Geotech. Engng 120, No. 5, 856–871.
Thomas Telford. Taylor, D. W. (1948). Fundamentals of soil mechanics. New York,
Ladd, C. C., Foot, R., Ishihara, K., Schlosser, F. & Poulos, H. G. NY, USA: Wiley.
(1977). Stress-deformation and strength characteristics. Tika, T. E., Vaughan, P. R. & Lemos, L. J. L. (1996). Fast
Proceedings of the 9th international conference on soil mechanics shearing of pre-existing shear zones in soil. Géotechnique
and foundation engineering, Tokyo, Japan, vol. 2, pp. 421–494. 46, No. 2, 197–233, http://dx.doi.org/10.1680/geot.1996.46.
Lehane, B. M. & Jardine, R. J. (1992). Residual strength character- 2.197.
istics of Bothkennar clay. Géotechnique 42, No. 2, 363–367, Tsubakihara, Y. & Kishida, H. (1993). Frictional behaviour between
http://dx.doi.org/10.1680/geot.1992.42.2.363. normally consolidated clay and steel by two direct shear type
Lehane, B. M. & Liu, Q. B. (2013). Measurement of shearing apparatuses. Soils Found. 33, No. 2, 1–13.
characteristics of granular materials at low stress levels in a shear Wesley, L. D. (1977). Shear strength properties of halloysite and
box. Geotech. Geol. Engng 31, No. 1, 329–336. allophane clays in Java, Indonesia. Géotechnique 27, No. 2,
Lemos, L. J. L. & Vaughan, P. R. (2000). Clay-interface shear 125–136, http://dx.doi.org/10.1680/geot.1977.27.2.125.
resistance. Géotechnique 50, No. 1, 55–64, http://dx.doi.org/ White, D. J., Campbell, M. E., Boylan, N. P. & Bransby, M. F.
10.1680/geot.2000.50.1.55. (2012). A new framework for axial pipe–soil interaction
Lupini, J. F., Skinner, A. E. & Vaughan, P. R. (1981). The drained illustrated by shear box tests on carbonate soils. Proceedings of
residual strength of cohesive soils. Géotechnique 31, No. 2, international conferemce on offshore site investigation and
181–213, http://dx.doi.org/10.1680/geot.1981.31.2.181. geotechnics, SUT, London, UK, pp. 379–387.
Najjar, S. S., Gilbert, R. B., Liedtke, E., McCarron, B. & White, D. J., Leckie, S. H. F., Draper, S. & Zakarian, E.
Young, A. G. (2007). Residual shear strength for interfaces (2015). Temporal changes in pipeline–seabed condition
between pipelines and clays at low effective stresses. ASCE and their effect on operating behaviour. Proceedings of
J. Geotech. Geoenviron. Engng 133, No. 6, 695–706. 34th international conference on ocean, offshore and arctic
Pedersen, R. C., Olson, R. E. & Rauch, A. F. (2003). Shear and engineering, St John’s, Newfoundland, Canada, paper
interface strength of clay at very low effective stress. Geotech. OMAE2015-42216.
Testing J. 26, No. 1, 71–77. Wijewickreme, D., Amarasinghe, R. & Eid, H. (2014). Macro-scale
Potts, D. M., Dounias, G. T. & Vaughan, P. R. (1987). Finite element direct shear test device for assessing soil-solid interface
analysis of the direct shear box test. Géotechnique 37, No. 1, friction under low effective normal stresses. Geotech. Testing J.
11–23, http://dx.doi.org/10.1680/geot.1987.37.1.11. 37, No. 1, paper GTJ20120217.
Potyondy, J. G. (1961). Skin friction between various soils and Wood, D. M. (1990). Soil behaviour and critical state soil mechanics.
construction materials. Géotechnique 11, No. 4, 339–353, Cambridge, UK: Cambridge University Press.
http://dx.doi.org/10.1680/geot.1961.11.4.339. Wroth, C. P. (1984). The interpretation of in situ soil tests.
Ramsey, N., Jardine, R., Lehane, B. & Ridley, A. (1998). A review of Géotechnique 34, No. 4, 449–489, http://dx.doi.org/10.1680/
soil–steel interface testing with the ring shear apparatus. geot.1984.34.4.449.
Proceedings of international conference on offshore site investi- Yan, Y., White, D. J. & Randolph, M. F. (2014). Cyclic consolidation
gation and foundation behaviour, ’98:‘New frontiers’, London, and axial friction on seabed pipelines. Géotechnique Lett. 4,
UK, pp. 237–258. No. 3, 165–169.
Randolph, M. F., White, D. J. & Yan, Y. (2012). Modelling the axial Yoshimi, Y. & Kishida, T. (1981). A ring torsion apparatus for
soil resistance on deep-water pipelines. Géotechnique 62, No. 9, evaluating friction between soil and metal surfaces. Geotech.
837–846, http://dx.doi.org/10.1680/geot.12.OG.010. Testing J. 4, No. 4, 145–152.
Downloaded by [ Universidade de Sao Paulo] on [22/02/18]. Copyright © ICE Publishing, all rights reserved.