Discrepancy in the Critical State Void Ratio of Poorly

Graded Sand due to Shear Strain Localization

Wadi H. Imseeh, S.M.ASCE 1; Khalid A. Alshibli, M.ASCE 2; and Riyadh I. Al-Raoush 3

Abstract: The critical state (CS) concept is a theoretical framework that models the constitutive behavior of soils, including sand and other
granular materials. It supports the notion of a unique postfailure state, where the soil ultimately experiences continuous shearing with no
change in the plastic volumetric strain. However, the published literature has frequently noted the nonconvergence of sand specimens with
Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

different initial densities to a unique CS in the compression plane due to many factors such as specimen fabric, particle morphology, breakage,
and grain size distribution. This paper examines the CS for poorly graded (uniform) glass beads and 3 different types of silica sands using 50
conventional triaxial compression (CTC) experiments, 12 oedometer tests, and in situ synchrotron microcomputed tomography (SMT) scans
for 10 CTC experiments. The results of the 50 CTC experiments revealed a diffused CS zone in the compression plane, which was further
examined using the in situ SMT scans. A thorough three-dimensional image analysis of the SMT scans accurately quantified the evolution of
the local void ratio (elocal ) versus axial compression within zones of intensive shearing toward the center of the specimen. The evolution of the
void ratio was also measured using the entire volume of the specimen (eglobal ). At the CS, the elocal =eglobal ratio was assessed to be ∼1.25 when
a single shear band developed within the scanned specimens and ∼1.1–1.15 for specimens that failed via external bulging that was internally
manifested by the development of multiple shear bands. This finding suggests that the CS zone in the compression plane can be attributed to
the common wrong consideration of eglobal evolution in lieu of elocal within the developing shear bands. Furthermore, the lack of shear band
development in uniaxial compression has made the results of the oedometer test reliable in quantifying the CS parameters in the compression
plane. DOI: 10.1061/(ASCE)GT.1943-5606.0002280. This work is made available under the terms of the Creative Commons Attribution
4.0 International license, https://creativecommons.org/licenses/by/4.0/.
Author keywords: Sand; Critical state; Shear bands; Void ratio; Microcomputed tomography; Conventional triaxial compression;
Oedometer tests.

Introduction ratio, respectively. The CSL has conventionally been established in

the stress plane (p 0 − q) and compression plane (p 0 − e) separately,
In engineering practice, the constitutive behavior of sand has since this procedure enhances the understanding of general triaxial
been traditionally modeled using global-scale measurements of stress and compression paths when superimposed in planes versus
stress and strain at the boundaries of laboratory-size specimens. three-dimensional (3D) space representation (Poorooshasb et al.
The development of the stress-dilatancy relationship (Roscoe et al. 1966; Wood 1990). Many one-dimensional (1D) compression ex-
1958) offered a major enhancement to the Mohr-Coulomb criterion periments on sand have confirmed a similar response to clayey soil
and inspired the development of the critical state (CS) theory in soil in which the CSL and normal compression lines (NCLs) are parallel
mechanics. The CS theory was originally proposed for clays by when plotted in the compression plane (Been and Jefferies 1985;
Schofield and Wroth (1968), who postulated a material failure sur- Coop 1990; McDowell and Bolton 1998), with a certain allowance
face known as yield locus, and a unique postfailure state known as for the influence of particle breakage (Coop and Lee 1993). How-
the CS. The unique postfailure CS is attributed to conditions of ul- ever, the convergence of sand specimens to a unique CSL in the
timate shear distortion without changes in the volume or effective compression plane has been controversial in the current literature.
stress and theoretically represented by the critical state line (CSL) To mention a few studies, Martins et al. (2001) examined the com-
tracing the top of yield locus in p 0 − q − e space, where p 0 ; q, and e pression behavior of residual clayey sand specimens using oedom-
are the mean effective principal stress, deviatoric stress, and void eter tests, and the location of NCLs in the compression plane was
found to be a function of the specimen’s initial e. Ferreira and Bica
1
Ph.D. Candidate, Dept. of Civil and Environmental Engineering, Univ.
(2006) assessed the CS for sand-kaolin mixtures using triaxial
of Tennessee, 325 John Tickle Bldg., Knoxville, TN 37996. Email: compression experiments and reported a family of parallel CSLs
[email protected] in the compression plane depending on the specimen’s initial density
2 state.
Professor, Dept. of Civil and Environmental Engineering, Univ. of
Tennessee, 325 John Tickle Bldg., Knoxville, TN 37996. ORCID: https:// The nonconvergence of compression paths (NCLs and CSLs)
orcid.org/0000-0001-5351-1670. Email: [email protected] has been acknowledged when describing the constitutive behavior
3
Professor, Dept. of Civil and Architectural Engineering, Qatar Univ., of sand, and the published literature has introduced the term transi-
P.O. Box 2713, Doha 2713, Qatar (corresponding author). Email: riyadh@ tional to describe this mode of behavior (Nocilla et al. 2006).
qu.edu.qa
Several studies have demonstrated that the transitional behavior of
Note. This manuscript was submitted on July 28, 2019; approved on
January 24, 2020; published online on June 10, 2020. Discussion period sand is caused by particle breakage (Altuhafi and Coop 2011;
open until November 10, 2020; separate discussions must be submitted for Shipton and Coop 2012; Xiao et al. 2015), inherent effects of a spec-
individual papers. This paper is part of the Journal of Geotechnical and imen’s initial density state (Coop 2015; Shipton and Coop 2012;
Geoenvironmental Engineering, © ASCE, ISSN 1090-0241. Xiao et al. 2016; Xu and Coop 2017), the complex morphology

© ASCE 04020066-1 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

 of natural sand particles (Alshibli and Cil 2018; Kandasami and resin impregnation and two-dimensional (2D) section microscopy.
Murthy 2017; Santamarina and Cho 2004; Yang and Luo 2015), Numerous discrete element method (DEM) studies have also dem-
influences of grain size distribution (Altuhafi et al. 2010; Altuhafi onstrated the localization of e at higher values within the shear
and Coop 2011), variation in the mineralogy of sand-fine mixtures band that develops within specimens composed of 2D discs (Gu
(Ponzoni et al. 2017; Shipton and Coop 2015), and the percentage of et al. 2014; Iwashita and Oda 1998, 2000; Jiang et al. 2011) and
fine content in sand specimens (Kwa and Airey 2016; Zuo and 3D clumped spheres (Lu and Frost 2010) loaded in biaxial
Baudet 2015). The transitional behavior of sand has also been attrib- compression.
uted to the strong forms of microscale fabric that are difficult to In situ (i.e., in-position scanning during an experiment) X-ray
break even at high applied strains. For instance, Todisco et al. (2018) computed tomography (CT) has recently offered a powerful non-
used mercury intrusion porosimetry to investigate the evolution of destructive technique that imaged through the 3D internal structure
pore structure within sand specimens subjected to conventional tri- (e.g., particles and voids) of sand specimens, and confirmed the
axial and uniaxial compression tests. Todisco et al. (2018) advocated localization of e at high values within the developed shear band
for the complex evolution of pore void distribution, causing the tran- in sand specimens loaded under biaxial compression (Alshibli
sitional behavior of sand due to initial differences in the void struc- and Hasan 2008; Desrues and Viggiani 2004). However, sand spec-
ture that could not be erased during conventional testing. Although imens tested in biaxial compression exhibit much less volumetric
Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

there are still active quests for integrating fabric evolution as an es- strain at the CS in comparison to specimens loaded in triaxial com-
sential state parameter in CS theory (Dafalias and Manzari 2004; pression (Alshibli et al. 2003). Tagliaferri et al. (2011) collected
Fu and Dafalias 2011; Gao et al. 2014; Imseeh et al. 2017; Petalas in situ X-ray scans for conventional triaxial compression (CTC)
et al. 2019; Theocharis et al. 2016; Wang et al. 2017), the develop- experiments on two sand specimens: biocemented and nonce-
ment of these constitutive models remains incomplete because lab- mented. Images of the noncemented specimen reaffirmed the
oratory sand specimens tested in the conventional biaxial (Finno and localization of e at high values within the shear band that gradually
Rechenmacher 2003) and triaxial (Alshibli and Cil 2018) appara- developed with axial compression. In recent advances, sets of in
tuses fail to deliver a unique CSL in the compression plane. situ X-ray CT scans with excellent image quality were acquired
Most of the studies reported in the literature relied on global for sand specimens subjected to CTC experiments (Alshibli et al.
change in specimen volume to measure the evolution of e toward 2016; Andò et al. 2017). Compared to biaxial compression, the
the CS, with limited research investigating the accuracy of conven- scans for CTC specimens exposed rather complex internal shearing
tional triaxial or biaxial testing procedures (Garga and Zhang 1997; patterns named micro shear bands (MSBs) that developed during
Jefferies and Been 2000). Mooney et al. (1998) conducted a series the hardening stage of the experiments (Amirrahmat et al. 2018).
of biaxial compression tests in search of a unique CSL when locally Furthermore, in situ X-ray CT scans for conventional extension and
assessed for the zone affected by a high shear distortion, which is compression triaxial experiments on sand specimens in Salvatore
also known as the shear band. Their tests were conducted using a et al. (2017) revealed a unique trace of the CSL in the compression
special biaxial apparatus (Harris et al. 1995) that provided a local plane when e measured from the scans was limited to the zones
estimate of the volumetric strain within the shear band via several affected by the largest distortion (shear bands).
displacement sensors attached to the specimen surface at multiple The main objective of this paper is to accurately establish
vertical levels. The apparatus side walls and loading platens were the CSL and yield locus for four different types of poorly grad-
also lubricated to minimize friction, and the bottom loading platen ed granular materials based on conventional laboratory experi-
was connected to a free-sliding base that allowed for the onset and ments. The paper sheds light on a discrepancy in measuring the
growth of the shear band by minimizing the influence of boundary evolution of e toward the CS using the conventional triaxial
constraints. Furthermore, one of the apparatus side walls was made apparatus due to the development of internal shear bands within
of clear Plexiglas, which permitted monitoring of the shear strain the specimens.
localization from the specimen side using digital photographs.
Mooney et al. (1998) reported a significantly higher local e when
measured near the shear band region compared to e measured using CTC Experiments on Laboratory-Size Specimens
the global volume change. Furthermore, they advocated for the
notion of a CSL, but the CSL in the compression plane was found Experiments
to depend on the initial e and subsequent consolidation history of
the specimens (Finno and Rechenmacher 2003). Localization of e This paper uses the results of 50 CTC experiments that were re-
at higher values was also detected by optical microscopy near the ported in Alshibli and Cil (2018) on specimens composed of
center of sand specimens loaded in drained triaxial (Frost and Jang spherical glass beads (labeled GB) and three types of silica sands
2000) and biaxial (Evans and Frost 2010) compression using epoxy known as F-35 Ottawa sand (labeled F35), #1 dry glass sand

Table 1. Properties of tested granular material

Material Soda lime glass beads F35 Ottawa sand #1 dry glass sand GS40 Columbia grout sand
Label GB F-35 DG GS#40
Specific gravity, Gs 2.55 2.65 2.65 2.65
Minimum void ratio, emin a 0.554 0.490 0.626 0.643
Maximum void ratio, emax 0.800 0.763 0.947 0.946
Roundness index, I R 0.965 0.959 0.937 0.924
Surface texture index, Rq 0.381 2.084 1.990 1.923
Sphericity index, I sph 1.096 1.872 1.704 1.674
Grain sizes Size fraction between US sieves #40 (0.420 mm) and #50 (0.297 mm)
Source: Data from Alshibli et al. (2014).
a
Data from Alshibli and Cil (2018).

© ASCE 04020066-2 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

 (labeled DG), and GS#40 Columbia grout sand (labeled GS40). and tracking). A detailed description of particle morphology for
Table 1 summarizes the properties of the tested materials. The glass each material can be found in Alshibli et al. (2014), including scan-
beads and three silica sands represent poorly graded (uniform) ning electron microscope (SEM) images.
granular materials with grain size between US sieve #40 (0.420) The CTC experiments were conducted on dry cylindrical
and #50 (0.297 mm) and have different particle morphologies rang- specimens measuring 70 mm in diameter × 140 mm in height
ing from rounded to angular classes. Poorly graded granular ma- under drained conditions. The specimens were tested at constant
terials were selected in this study because the uniform grain sizes confining pressures (σ3 ) of 15, 25, 50, 100, and 400 kPa as well
limit the experimental parameters and significantly increase the as different initial relative density (Dr ) states: loose (Dr < 30%),
quality of the X-ray CT image as well as the efficiency of image medium dense (30% ≤ Dr ≤70%), and dense (Dr > 70%) (Table 2).
processing procedures (e.g., algorithms for particle segmentation Alshibli and Cil (2018) presented a detailed description of the

Table 2. Summary of the 50 CTC experiments

Tested materiala Experiment label Confining stress, σ3 (kPa) Initial void ratio, eo Visual observation on specimen surface at failure
Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

F35 F35_L_15 15 0.735 Slight bulge, no apparent shear band

F35_L_50 50 0.731 Slight bulge, no apparent shear band
F35_L_100 100 0.726 Slight bulge, no apparent shear band
F35_L_400 400 0.738 Slight bulge, no apparent shear band
F35_MD_15 15 0.617 Moderate bulge, no apparent shear band
F35_MD_50 50 0.612 Moderate bulge, possible internal shear band
F35_MD_100 100 0.611 Shear band at 61°
F35_MD_400 400 0.615 Shear band at 45.5°, possible second shear band
F35_D_15 15 0.491 Shear band at 57°
F35_D_50 50 0.496 N/A
F35_D_100 100 0.498 Significant bulge, possible internal shear band
F35_D_400 400 0.491 Two shear bands at 49° and 58°
DG DG_L_15 15 0.935 Slight bulge, no apparent shear bands
DG_L_50 50 0.916 Slight Bulge, no apparent shear bands
DG_L_100 100 0.935 Slight bulge, no apparent shear bands
DG_L_400 400 0.925 Slight bulge, no apparent shear bands
DG_MD_15 15 0.777 Significant bulge, possible internal shear band
DG_MD_50 50 0.786 Moderate bulge, possible internal shear band
DG_MD_100 100 0.774 Two shear bands at 63.5° and 66°
DG_MD_400 400 0.775 Moderate bulge, possible internal shear band
DG_D_15 15 0.635 N/A
DG_D_25 25 0.645 Two shear bands at 77° and 63°
DG_MD_50 50 0.639 Significant bulge, possible internal shear band
DG_MD_100 100 0.645 N/A
DG_MD_400 400 0.631 Shear band at 56°
GS40 GS40_L_15 15 0.921 Slight bulge,no apparent shear bands
GS40_L_50 50 0.916 Slight bulge,no apparent shear bands
GS40_L_100 100 0.933 Slight bulge,no apparent shear bands
GS40_L_400 400 0.915 Slight bulge,no apparent shear bands
GS40_MD_15 15 0.764 Moderate bulge,possible internal shear band
GS40_MD_50 50 0.762 Moderate bulge,possible internal shear band
GS40_MD_100 100 0.758 Moderate bulge,possible internal shear band
GS40_MD_400 400 0.762 Moderate bulge,possible internal shear band
GS40_D_15 15 0.638 N/A
GS40_D_25 25 0.640 N/A
GS40_MD_50 50 0.648 N/A
GS40_MD_100 100 0.635 N/A
GS40_MD_400 400 0.634 Shear band at 55.5°
GB GB_L_15 15 0.788 Slight bulge,no apparent shear bands
GB_L_50 50 0.795 Slight bulge,no apparent shear bands
GB_L_100 100 0.792 Slight bulge,no apparent shear bands
GB_L_400 400 0.780 Slight bulge,no apparent shear bands
GB_MD_15 15 0.707 Moderate bulge,possible internal shear band
GB_MD_50 50 0.706 Moderate bulge,possible internal shear band
GB_MD_100 100 0.721 N/A
GB_MD_400 400 0.701 Moderate bulge, possible internal shear band
GB_D_15 15 0.624 Significant bulge, possible internal shear band
GB_D_50 50 0.622 Significant bulge, possible internal shear band
GB_D_100 100 0.612 Significant bulge, possible internal shear band
GB_D_400 400 0.621 Apparent shear band at 62°
Source: Data from Alshibli and Cil (2018).
a
See Table 1 for a detailed description of tested material.

© ASCE 04020066-3 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

 specimen preparation and testing procedures. Briefly, dry sand was water pressure); V 0 = initial volume of specimen calculated using
deposited using air pluviation to prepare the cylindrical specimens caliper measurements of specimen’s initial diameter and height;
with different initial Dr . Water was used to apply a constant con- wS = weight of solids; γ w = unit weight of water; and Gs = specific
fining pressure (σ3 ) via a standard triaxial cell, then the specimens gravity of solids (Table 1). Referring to Figs. 1(d) and 2(d), one
were sheared under axial compression at a constant displacement notices that the curves describing the glass bead experiments ex-
rate of 0.5 mm=min. The CTC apparatus was connected to a data hibited an oscillatory behavior caused by slip-stick that occurs be-
acquisition system that recorded accurate measurements of σ3 , bulk tween particles as they roll/slide against each other. This sudden
volume change (δv), axial stress (σ1 ), and axial strain (ε1 ). δv was slip-stick behavior is attributed to the high uniformity in the round-
measured by continuously recording the change in the volume of ness and sphericity of the glass beads (manufactured material) as
water pumped in/purged out of the confining chamber. Figs. 1 and 2 well as their smooth surface texture. Alshibli and Roussel (2006)
display the evolution of q=p 0 and e versus ε1 for the experiments, presented a detailed experimental study of the slip-stick behavior in
respectively, where specimens composed of glass beads.
Deviator stress ðqÞ ¼ σ10 − σ30 ð1Þ
Critical State Assessment
σ10 þ 2σ30
Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

Effective mean stress ðp 0 Þ ¼ ð2Þ The results of the 50 CTC experiments were examined to establish
3

the CSL for the tested materials. The CS attainment is mathemati-
ws
ðV 0 − δvÞ − γ w Gs
cally described by the conditions at which (Wood 1990)
e¼ ws

ð3Þ
γ w Gs
∂p 0 ∂q ∂e 2ðε1 − ε3 Þ
where the effective stresses σ10 and σ30 = total stresses σ1 and σ3 ¼ ¼ ¼ 0; εq ¼ ð4Þ
since dry specimens were tested in drained conditions (no pore ∂εq ∂εq ∂εq 3

2 2

DG_L_15
F35_L_15 1.5
1.5 DG_L_50
F35_L_50
DG_L_100
F35_L_100
DG_L_400
F35_L_400
DG_MD_15
F35_MD_15
DG_MD_50
q/p'

F35_MD_50
q/p'

1 1 DG_MD_100
F35_MD_100
DG_MD_400
F35_MD_400
DG_D_15
F35_D_15
DG_D_25
F35_D_50
DG_D_50
F35_D_100
0.5 0.5 DG_D_100
F35_D_400
DG_D_400

0 0
0 5 10 15 20 0 5 10 15 20

(a) Axial Strain, 1

(%) (b) Axial Strain, 1
(%)

2 2

GS40_L_15
GB_L_15
1.5 GS40_L_50 1.5
GB_L_50
GS40_L_100
GB_L_100
GS40_L_400
GB_L_400
GS40_MD_15
GB_MD_15
GS40_MD_50
GB_MD_50
q/p'
q/p'

1 GS40_MD_100 1
GB_MD_100
GS40_MD_400
GB_MD_400
GS40_D_15
GB_D_15
GS40_D_25
GB_D_50
GS40_D_50
GB_D_100
0.5 GS40_D_100 0.5
GB_D_400
GS40_D_400

0 0
0 5 10 15 20 0 5 10 15 20

(c) Axial Strain, 1

(%) (d) Axial Strain, 1
(%)

Fig. 1. (Color) Evolution of q=p 0 versus ε1 for the 50 CTC experiments listed in Table 2, which were conducted on specimens of (a) F35; (b) DG;
(c) GS40; and (d) GB.

© ASCE 04020066-4 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

Fig. 2. (Color) Evolution of e versus ε1 for the 50 CTC experiments listed in Table 2, which were conducted on specimens of (a) F35; (b) DG;
(c) GS40; and (d) GB.

where εq and ε3 = deviatoric and radial strains for CTC conditions, representation of the CSL in the compression plane for granular
respectively. ∂εq in Eq. (4) degenerates to ∂ε1 since ε1 ¼ −2ε3 for materials. Furthermore, the normalization of p 0 with respect to pa
the specific condition of ultimate shear distortion at a constant vol- in Eq. (7) makes the intercept parameter eλ independent of the unit
ume, which is postulated by CS theory. Therefore, the tested spec- chosen for the measurement of stress, unlike the case for the Cam-
imens are assumed to reach the CS when curves in Figs. 1 and 2 Clay model [Eq. (6)].
approach a relatively constant slope (∼ε1 > 15%). For each set of
curves/material in Figs. 1 and 2, constant p 0 , q, and e at the CS
were determined and plotted in the stress and compression planes Discussion
as depicted in Figs. 3 and 4, respectively. In the stress plane (Fig. 3),
the CSL is modeled using the CS parameter M as For each tested material, the CTC results showed a distinctive CSL
in the stress plane as depicted in Fig. 3. Still, a trend can be noticed
q ¼ Mp 0 ð5Þ
in Fig. 3 in which dense specimens tend to have a slightly higher
In the compression plane (Fig. 4), several analytical models slope (M). This trend was also reported by Alshibli and Cil (2018)
were proposed in the literature to fit a linear yield locus including and correlated well to the applied σ3 as well as the specimen’s
the CSL (eλ ¼ Γ for the CSL): initial Dr . Overall, the linear regression models in Fig. 3 show
significant statistical correlations: R2 ∼1, p-value <0.05 for the
e ¼ eλ − λ ln p 0 → Cam-Clay model ð6Þ F-statistics (the regression model is significant), and narrow 95%
confidence limits for the estimate of M. However, the CSL
 0 α
p appeared as a diffused stress-dependent zone in the compression
e ¼ eλ − λ → Linear model ð7Þ plane (Fig. 4 and Table 3) rather than a linear representation (low
pa
R2 values, p-value>0.05 for the F-statistics, a wide 95% confi-
where pa = atmospheric pressure; and α ¼ 0.6 is a model fit dence interval for the estimates Γ and λ. The extensive published
parameter. This paper adopts the linear model Li and Wang (1998) literature has reported a similar pattern of response in the com-
since it has been widely recommended to enhance the linear pression plane in which sand specimens with different initial

© ASCE 04020066-5 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

Fig. 3. (Color) CSL modeled in stress plane ðq − p 0 Þ using the 50 CTC experiments listed in Table 2, which were conducted on specimens of (a) F35;
(b) DG; (c) GS40; and (d) GB.

e do not approach a unique CSL when tested under general triaxial based on the results of biaxial compression experiments. In sum-
compression paths. For instance, Marschi et al. (1972) conducted mary, they characterized the CS zone by multiple CSLs depending
drained CTC experiments on Pyramid Dam rockfill granular on the initial e of the specimen.
material and reported that the CTC paths in the compression plane Visual observation of specimen failure at the CS manifested
of dense versus loose specimens did not approach the same CSL. different modes for the 50 CTC experiments depending on the
Wood (1990) examined results of constant p 0 triaxial compression applied σ3 and the specimens’ initial Dr (Table 2). In summary,
experiments on Chattahoochee River sand (Vesic and Clough loose specimens exhibited slight bulging with no externally ob-
1968), and the results showed different e for initially dense versus served shear bands, whereas medium dense and dense specimens
loose specimens at the CS. That is, Wood (1990) reported that failed via apparent single or multiple shear bands. Characterizing
dense specimens needed ∼17% dilation by volume to attain the the failure mode of specimens by visual observations on their
same e as the loose specimens at the CS, which is unlikely to surfaces can be misleading because they are just an external mani-
occur due to testing difficulties (e.g., rigid loading endplates, rub- festation of more complex internal shearing patterns that cause
ber membrane) hinder the ultimate dilation of the specimens. a nonuniform distribution of e within the sheared specimens.
Although the testing difficulties were fairly alleviated by the spe- Therefore, the diffused CS zone in the compression plane can be
cial biaxial compression apparatus used by Mooney et al. (1998), attributed to the false reliance on the global volume change (δv)
they still failed to deliver a unique CSL in the compression plane. [Eq. (3)] to measure e at failure. The following section evaluates
In fact, findings similar to those presented in Fig. 4 were reported the latter hypothesis that would explain the formation of the CS
by Mooney et al. (1998) and Finno and Rechenmacher (2003) zone in the compression plane using accurate measurements of

© ASCE 04020066-6 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

Fig. 4. (Color) CSL modeled in compression plane ðe − p 0 Þ using the 50 CTC experiments listed in Table 2, which were conducted on specimens of
(a) F35; (b) DG; (c) GS40; and (d) GB.

e based on in situ synchrotron microcomputed tomography (SMT) SMT scanner stage of beamline station 13BMD, Advanced Photon
scans that were collected for a series of CTC experiments on the Source (APS), Argonne National Laboratory (ANL), Downers
tested materials. Grove Township, Illinois. The specimens were confined at a con-
stant σ3 inside an air-presssurized chamber, then axial compression
was applied by a stepper motor at a displacement-controlled load-
In Situ SMT Scans for CTC Experiments ing rate of 0.2 mm=min. The axial compression was paused at cer-
tain loading steps, then the special apparatus was rotated over 180°
at 0.2° rotational increments to acquire 900 radiographs using a
In Situ Scanning
monochromatic X-ray beam at an enegry of 33 keV. The radio-
Alshibli et al. (2016) collected sets of in situ SMT scans for 10 graphs were reconstructed into 3D images with an excellent spatial
drained CTC experiments on dry specimens composed of the four resolution ranging between 8.16 and 11.18 μm=pixel.
granular materials studied in this paper (Table 1). The CTC experi- The set of scans collected by Alshibli et al. (2016) has excellent
ments were conducted using a specially fabricated triaxial compres- resolution (Table 4) and sharp grayscale contrast distinguishing the
sion apparatus, which was described in Hasan and Alshibli (2012) solid and air phases inside the specimens. The ability to image
and had capabilities to test 10-mm ðdiameterÞ × 20-mm ðheightÞ through the internal structure of sand specimens at this high quality
specimens under drained CTC conditions similar to a conventional have yielded deep experimental insights into the evolution of 3D
laboratory apparatus. The special apparatus was mounted on the particle-scale characteristics in the course of CTC, such as the

© ASCE 04020066-7 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

 Table 3. Summary of regression model results fitted in Fig. 4 0.05 to 5 mm at 0.1 mm increments of the REV side length
Regression equation e ¼ Γ − λðp 0 =pa Þα to examine the effect of the REV size on elocal measurement.
Regression Tested The change in elocal value versus the REV size is displayed in
plot materiala Γ λ
Fig. 5, which shows REV sizes >2 mm having a relatively con-
Fig. 4(a) F35 0.694 (0.648, 0.739) 0.0064 (0.0317, −0.0188) stant elocal for the processed images. Therefore, a REV size of
Fig. 4(b) DG 0.866 (0.809, 0.923) 0.0085 (0.0401, −0.0231) 2 mm was selected to generate local distribution fields of e within
Fig. 4(c) GS40 0.862 (0.808, 0.916) 0.0073 (0.037, −0.0228) the scanned specimens, as illustrated in Figs. 6 through 8 at a
Fig. 4(d) GB 0.786 (0.749, 0.822) 0.0171 (0.0388, −0.0047)
central axial (Y-Z) section across the 3D images acquired for
Note: Values in parentheses denote 95% confidence limits of parameter. the F35_D_400_SMT, F35_D_15_SMT, and F35_MD_15_SMT
a
See Table 1 for a detailed description of tested materials. experiments, respectively. Furthermore, elocal values correspond-
ing to 2 mm REV sizes were averaged for each image to generate
fabric of normal to contact vectors between particles (Imseeh et al. the evolution curves of elocal versus ε1 in Fig. 9, with error bars
2017) and localization of shear strain into MSBs (Amirrahmat et al. displaying the interquartile range (IQR) of elocal with respect to
2018). This paper exploits the rich 3D experimental data in this set different REV sizes, whereas dashed curves represent the evolu-
tion of eglobal .
Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

of scans and processes the scans to accurately examine the evolu-

tion of e toward the CS within the CTC specimens in light of po-
tential reasons for the CS zone in the compression plane (Fig. 4).
Table 4 summarizes the scans collected for the 10 experiments, Discussion
which were tested on medium dense and dense specimens at σ3 ¼
15 and 400 kPa. Figs. 6 through 8 show clear evidence that e within the scanned
specimens was nonuniformly distributed, particularly when the
specimens approached the CS (ε1 > 15%). Higher e values can be
Image Processing and Data Analysis seen in Figs. 6–8 at the center of the specimens, which is expected
The SMT scans provided excellent grayscale images with crisp due to the effects of the rigid loading endplates and the flexible
boundaries that separate the solid and air phases within the spec- latex membrane surrounding the specimens. Moreover, the evolu-
imens. The grayscale images were processed using AVIZO version tion of elocal versus ε1 in Fig. 9 is significantly different from the
9.7.0, a commercial computer software developed by Thermo evolution of eglobal , similar to findings reported on sand specimens
Fisher Scientific for high-performance 3D image visualization and loaded in biaxial compression (Finno and Rechenmacher 2003;
processing. An anisotropic filter module was executed on grayscale Mooney et al. 1998). In an attempt to assess the uniqueness of
images to reduce image noise, and then the solid and air phases the CSL in the compression plane using elocal , Fig. 10 shows box-
were segmented using an interactive thresholding module. Voxels plots of elocal for the scanned specimens at failure (last loading
belonging to the solid and air phases were assigned a value of 1 step), with the scattered points representing the variation in elocal
and 0, respectively. A detailed description of the image acquisition versus the REV size (2–5 mm). Regarding the boxplot components,
and processing modules can be found in Druckrey et al. (2016). the green diamond represents the mean, the red line marks the
Processed images were used to accurately measure the volume median, the blue box represents the IQR, and the black whisker
of solid and air phases within the scanned specimens at a microm- bounds the data points within 1.5 × IQR of the upper and lower
eter level of accuracy. For each processed image, eglobal was calcu- quartiles. Interestingly, equal elocal was assessed at failure for the
lated as the volume ratio between the air and solid phases within the F35_MD_15_SMT and F35_D_15_SMT specimens in Fig. 10(a)
entire specimen, and the evolution of eglobal versus ε1 is plotted in as well as the DG_MD_15_SMT and DG_D_15_SMT specimens
Fig. 2 for the scanned specimens to compare with the behvior of in Fig. 10(b), which are the specimens with different initial e tested
laboratory-size CTC experiments presented earlier in the section at the same σ3 . The equality in elocal for these two pairs of spec-
“CTC Experiments on Laboratory-Size Specimens.” imens was assessed via the Wilcoxon Rank Sum (WRS) test, which
A cubical representative elementary volume (REV) was parti- tested the null hypothesis (Ho ) of equal median for the two spec-
tioned at the center of each processed image, and e was calculated imens’ data of elocal . In statistical hypothesis testing, the failure to
for the REV to investigate the evolution of elocal versus ε1 within reject Ho is assessed by the test probability value (p-value) and
the scanned specimens. REV sizes were gradually increased from a predefined significance limit (αs ¼ 0.05). That is, a p-value

Table 4. Summary of SMT scans acquired during CTC experiments on specimens of tested granular materials
Tested Experiment Confining stress, Initial void Axial strain at which SMT Resolution
materiala label σ3 (kPa) ratio, eo scans were acquired, ε1 (%) (μm=pixel)
F35 F35_D_400_SMT 400 0.493 0, 1, 2, 3.4, 4.9, 6.9, 8.9, 11.8, 17.2 11.18
F35_D_15_SMT 15 0.525 0, 1, 2, 3.5, 5, 6.9, 8.9, 11.9, 17.4, 22.3 11.14
F35_MD_15_SMT 15 0.573 0, 1, 2, 3.6, 5.1, 7.1, 9.2, 12.2 15.8, 19.9, 22.9 8.16
DG DG_D_400_SMT 400 0.689 0.0, 1.0, 2.0, 3.6, 5.1, 7.1, 9.2, 12.2, 17.8 11.18
DG_D_15_SMT 15 0.688 0.0, 2.0, 3.5, 5.0, 7.0, 9.0, 11.9, 17.4 11.14
DG_MD_15_SMT 15 0.737 0.0, 1.0, 2.0, 3.6, 5.1, 7.2, 9.2, 12.3, 15.8, 19.9 8.16
GS40 GS40_D_400_SMT 400 0.626 0.0, 1.0, 1.9, 3.3, 4.7, 6.7, 8.6, 11.4, 14.7 8.16
GS40_D_15_SMT 15 0.675 0.0, 1.0, 2.0, 3.5, 5.0, 7.0, 9.0, 12.0, 17.5 11.14
GB GB_D_400_SMT 400 0.547 0.0, 1.0, 2.1, 3.6, 5.2, 7.3, 9.3, 12.4, 18.1 11.18
GB_D_15_SMT 15 0.525 0.0, 1.0, 2.0, 3.6, 5.1, 7.1, 9.1, 12.2 11.14
a
See Table 1 for a detailed description of tested materials.

© ASCE 04020066-8 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

Fig. 5. (Color) Change in elocal versus REV size for (a) F35_D_400_SMT; (b) F35_D_15_SMT; (c) F35_MD_15_SMT; (d) DG_D_400_SMT;
(e) DG_D_15_SMT; (f) DG_MD_15_SMT; (g) GS40_D_400_SMT; (h) GS40_D_15_SMT; (i) GB_D_400_SMT; and (j) GB_D_15_SMT experi-
ments. Refer to Table 4 for labels of experiments.

greater than αs indicates a failure to reject H0 , and one can con- specimens in Figs. 10(a and b) supports the notion of a unique
clude that the two specimens’ data of elocal have equal medians. The CSL in the compression plane, which confirms the findings
WRS test exhibited a p-value > 0.8 for the F35_MD_15_SMT and reported in Salvatore et al. (2017), and goes a step beyond the
F35_D_15_SMT boxplots in Fig. 10(a) and p-value > 0.1 for conclusions reported in Mooney et al. (1998) and Finno and
the DG_MD_15_SMT and DG_D_15_SMT boxplots in Fig. 10(b). Rechenmacher (2003) of multiple CSLs, depending on the initial
The equality between elocal for the medium dense and dense e of the specimens tested in biaxial compression.

© ASCE 04020066-9 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

Fig. 6. (Color) Distribution fields of e versus ε1 at central Y-Z cuts across the SMT scans collected during the F35_D_400_SMT experiment.
Refer to Table 4 for labels of experiments: (a) ε1 ¼ 0.0%; (b) ε1 ¼ 1.0%; (c) ε1 ¼ 2.0%; (d) ε1 ¼ 3.4%; (e) ε1 ¼ 4.9%; (f) ε1 ¼ 6.9%;
(g) ε1 ¼ 8.9%; (h) ε1 ¼ 11.8%; and (i) ε1 ¼ 17.2%.

Amirrahmat et al. (2018) used the relative particle translation the difference between elocal and eglobal gradually increased with
gradient (RPTG) concept proposed by Druckrey et al. (2018) to ε1 . At the CS (last loading step scans), elocal =eglobal was ∼1.25
provide a thorough assessment of the internal shearing patterns that for the F35_D_400_SMT [Fig. 11(a)] and GS40_D_400_SMT
developed within the scanned specimens reported in this paper. [Fig. 11(d)] specimens, which failed via a single shear band. On
Briefly, RPTG refers to the incremental displacement of each the other hand, elocal =eglobal was less throughout compression for
particle relative to its neighboring particles and normalized with specimens that exhibited external bulging (∼1.10–1.15 at the CS).
respect to the global axial compression. When a shear band devel- This discrepancy between eglobal and elocal can actually explain the
ops within a specimen, particles within the shear band rotate and formation of the CS zone in the compression plane in Fig. 4. Fur-
translate as if the bulk specimen is divided into multiple frictionally thermore, the slight bulging in the F35_MD_15_SMT specimen
sliding wedges, which produces higher RPTG values along the de- [Fig. 11(c)] produced the least severe discrepancy as the curve of
veloped shear band and constant volume flow as postulated by CS elocal =eglobal was the closest to unity in Fig. 13. The lesser deviation
theory. Figs. 11 and 12 presents side-by-side color maps for RPTG between elocal and eglobal for the F35_MD_15_SMT specimen can
and e distribution within the same central Y-Z vertical cut across also be noticed in Fig. 11(c) as the distribution field of e shows less
the images acquired at the last loading step for the 10 scanned spec- variation, and RPTG reveals a dispersed development of multiple
imens. RPTG clearly exposed the development of a single shear shear bands within the specimen. Therefore, the loose laboratory-
band in the F35_D_400_SMT [Fig. 11(a)], GS40_D_400_SMT size CTC specimens that manifested slight surface bulging at fail-
[Fig. 11(d)], and GB_D_400_SMT [Fig. 12(d)] specimens, while ure (Table 2) supposedly provided a more accurate (representative
external surface bulging in the other specimens was internally man- of the whole specimen) measurement of e at the CS than the dense
ifested by the development of multiple shear bands in opposite specimens.
directions. To quantify the influence of the failure mode (single
versus multiple shear bands) on the discrepancy in e at the CS,
Fig. 13 depicts the evolution of elocal =eglobal versus ε1 for the spec- Oedometer Experiments
imens presented in Fig. 11. The discrepancy curves in Fig. 13 re-
present the solid curves/error bars (elocal ) from Figs. 9(a and c)
Experiments
normalized by their respective dashed curves (eglobal ). Initially,
elocal =eglobal was ∼1 − 1.05 since isotropic normal compression The in situ SMT scans revealed a discrepancy in measuring the
under σ3 produces no shear (no shear strain localization), then evolution of e using the CTC apparatus due to shear strain

© ASCE 04020066-10 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

Fig. 7. (Color) Distribution fields of e versus ε1 at central Y-Z cuts across SMT scans collected during F35_D_15_SMT experiment. Refer to
Table 4 for labels of experiments: (a) ε1 ¼ 0.0%; (b) ε1 ¼ 1.0%; (c) ε1 ¼ 2.0%; (d) ε1 ¼ 3.5%; (e) ε1 ¼ 5%; (f) ε1 ¼ 6.9%; (g) ε1 ¼ 8.9%;
(h) ε1 ¼ 11.9%; (i) ε1 ¼ 17.4%; and (j) ε1 ¼ 22.3%.

localization within sand specimens at the CS. Alternatively, the medium dense, and loose), as summarized in Table 5. Specimens
CSL in the compression plane can be determined using an oedom- with different initial Dr were prepared by controlling the deposition
eter test. The restriction of lateral expansion in the oedometer test height (small deposition height for loose specimens) and slightly
hinders the development of shear bands, which results in a rela- tapping on the mold sides with a rubber mallet after depositing
tively uniform distribution of e within the specimens and allows each lift (no tapping for loose specimens, four taps for medium
accurate quantification of NCLs in the compression plane. Twelve dense specimens, eight taps for dense specimens). The specimens
oedometer tests were conducted on dry specimens of the same were also capped with a cylindrical steel spacer measuring 49 mm
granular materials studied in this paper (Table 1). The specimens in diameter × 7 mm in thickness to provide a level contact surface
were prepared by depositing the granular materials in four lifts us- when loading the specimen in uniaxial compression.
ing a funnel into a thick-wall cylindrical steel mold with an inner The oedometer experiments were performed using a universal
diameter (D) of 50 mm and a height (Ho ) of 25 mm. For each testing machine (UTS) with a dual column loading frame that can
material, three specimens were tested at different initial Dr . (dense, apply a vertical compressive load up to 600 kN. The UTS machine

© ASCE 04020066-11 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

Fig. 8. (Color) Distribution fields of e versus ε1 at central Y-Z cuts across SMT scans collected during F35_MD_15_SMT experiment. Refer to
Table 4 for labels of experiments: (a) ε1 ¼ 0.0%; (b) ε1 ¼ 1.0%; (c) ε1 ¼ 2.0%; (d) ε1 ¼ 3.6%; (e) ε1 ¼ 5.1%; (f) ε1 ¼ 7.1%; (g) ε1 ¼ 9.2%;
(h) ε1 ¼ 12.2%; (i) ε1 ¼ 15.8%; (j) ε1 ¼ 19.9%; and (k) ε1 ¼ 22.9%.

consists of a hydraulic table equipped with a built-in load cell and semitransparent lines. As proposed by the CS theory, the gray lines
can move up/down at a displacement-controlled rate to vertically in Fig. 14 represent the K 0 -NCL that extends along the yield locus
load/unload specimens against a fixed cylindrical top loading pis- of the tested materials as it grows with uniaxial compression.
ton that has a diameter of 49 mm. The specimens were loaded in Assuming K o compression conditions (zero lateral expansion of
confined uniaxial compression at a displacement loading rate the sand specimen), p 0 can be calculated as
of 0.05 mm=min, including an unloading cycle at the end of each
test (ε1 ¼ 25%). Fig. 14 shows the evolution of e versus σ1 for the σ10 þ 2ð1 − sin ϕcs Þσ10
p0 ¼ ð9Þ
12 oedometer tests in which e was calculated based on the height 3
of solids (Hs ) as
where ϕcs = CS angle of internal friction, which is expressed as
H 0 ð1 − ε1 Þ − H s ws 3M
e¼ ; Hs ¼ ð8Þ sin ϕcs ¼ ð10Þ
Hs ðπD2 =4Þðγ w Gs Þ 6þM
where M = slope of CSL in stress plane, determined earlier in Fig. 3
Critical State Assessment
using the results of the CTC experiments. Fig. 15 displays the re-
The compression curves depicted in Fig. 14 show an initial elastic lationship between the CSL and K 0 -NCL in the stress plane, which
behavior, followed by stiff yielding, which is denoted by the gray can be expressed by the yield locus equation. Table 6 summarizes

© ASCE 04020066-12 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

Fig. 9. (Color) Comparison between the evolution of eglobal and elocal versus ε1 for (a) F35; (b) DG; (c) GS40; and (d) GB experiments. Refer to
Table 4 for labels of experiments.

the models presented in Fig. 15 that have been proposed in the lit- the dimensionless stress ratio type parameter β. The size of the
erature to fit a yield locus in the stress plane for granular materials. yield locus is determined by the stress type parameter po0 , which
The MIT-S1 model (Pestana and Whittle 1999) was adopted: is analogous to p 0 along the isotropic-NCL in the modified cam
  0 n  clay (MCC) model. However, the MIT-S1 model assumes a unique
ðq − βp 0 Þ2 2 þ β 2 − 2βηÞ2 1 − p lateral earth pressure coefficient (K o ) for specimens loaded in
¼ ðm ð11Þ
p 02 p00 1D compression (Fig. 14), for which p 0 ¼ po0 and η ¼ β. Accord-
ingly, the data points along the yielding portion of the compression
where η ¼ q=p 0 = stress ratio (e.g., η ¼ M ¼ 6 sin ϕcs =3 − sin ϕcs curves in Fig. 14 (gray lines) were mapped from the K o -NCL
for the CSL and η ¼ ηKo ¼ 3 sin ϕcs =3 − 2 sin ϕcs . for the K o -NCL). (medium gray line in Fig. 15) to CSL (black line in Fig. 15)
n and m are fitting parameters that respectively control the slender- using Eq. (11) (red curve in Fig. 15), and the CSL in the com-
ness and aperture of the yield locus (e.g., n ¼ 0.7 and m ¼ 1.55 pression plane was established for each material, as shown in
in Fig. 15), while the orientation of yield locus is controlled by Fig. 16. The MIT-S1 model was selected since it alleviates concerns

© ASCE 04020066-13 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

Fig. 10. (Color) Boxplots of elocal at last loading step for (a) F35; (b) DG; (c) GS40; and (d) GB experiments. Refer to Table 4 for labels of
experiments.

associated with the limited ellipsoidal shapes assumed by the MCC this trend supports the transitional behavior of sand because of
and MIT-E3 models (Fig. 15), in addition to significantly enhanc- particle breakage, the effect of the specimens’ initial Dr , and
ing the postulation of yield locus for sand, particularly in the super- fabric, for example. However, the linear models in Fig. 16 do
critical region (η > M) (Pestana and Whittle 1994, 1999; Whittle not produce a diffused CS zone like the case in the CTC results
et al. 1994). (Fig. 4) due to the lack of shear band development in the oedom-
eter tests.
To compare the oedometer with CTC results, the CSL in the
Discussion compression plane was fitted with a linear model in Fig. 17 using
the CS data points of the CTC experiments that were conducted on
The linear regression models in Fig. 16 established the CSL in each initial Dr state separately (dense, medium dense, and loose).
the compression plane very well (R2 ∼ 1, p-value <0.05 for the The results of the linear fit among the CS data points of loose spec-
F-statistics, and narrow 95% confidence limits for the estimates imens in Fig. 17(a) closely agree with Γ and λ determined by the
Γ and λ). A trend can still be seen in Figs. 16(a–c) in which the oedometer tests (Table 7). This agreement supports the earlier
data points of the tested sands show a higher slope (λ) for the conclusion of loose specimens providing more accurate CS assess-
loose, medium dense, and dense experiments, respectively. Again, ment in the compression plane when tested in CTC conditions due

© ASCE 04020066-14 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

Fig. 12. (Color) Distribution fields of RPTG and elocal calculated

Fig. 11. (Color) Distribution fields of RPTG and elocal calculated using processed SMT images collected at last loading steps for
using processed SMT images collected at last loading steps for (a) DG_D_400_SMT; (b) DG_D_15_SMT; (c) DG_MD_15_SMT;
(a) F35_D_400_SMT; (b) F35_D_15_SMT; (c) F35_MD_15_SMT; (d) GB_D_400_SMT; and (e) GB_D_15_SMT. Refer to Table 4 for
(d) GS40_D_400_SMT; and (e) GS40_D_15_SMT. Refer to Table 4 labels of experiments.
for labels of experiments.

to the slight bulging failure mode, which produced less discrepancy

between eglobal and elocal throughout the compression of the
F35_MD_15_SMT specimen (Fig. 13).

Summary and Conclusions

This paper seeks to accurately measure the CSL and yield locus for
poorly graded spherical glass beads and three types of silica sands
with different particle morphologies. The CSL was quantified in the
stress plane based on the results of CTC experiments that were con-
ducted on specimens with different initial densities and multiple
levels of σ3 . However, the CTC results revealed a diffused CS zone
in the compression plane that was clearly dependent on the applied
σ3 and initial density of the specimen. Potential causes of the
CS zone in the compression plane were investigated by analyzing
high-resolution 3D images of in situ SMT scans collected for CTC
experiments on the same tested materials. The SMT scans provided
Fig. 13. (Color) Discrepancy in evolution of e toward CS for experi-
excellent 3D images that offered interesting insights into the dis-
ments presented in Fig. 11 and conducted on F35 and GS40 sand.
crepancy in measuring the evolution of e within sand specimens

© ASCE 04020066-15 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

 1
Table 5. Summary of oedometer tests
CSL [Slope=M= ]
0.75
Tested Initial void
materiala Experiment label ratio, eo 0.5 K o-NCL [ Slope= K = ]
0
F35 F35_L 0.757
0.25 Iso-NCL [ Slope= ]
F35_MD 0.630

o
/p'
F35_D 0.575

q
0 Modified Cam Clay [M=1.2]

DG DG_L 0.819 -0.25 MIT-E3 [M=1.2, = K , n=0.7, N=0.95]

0
DG_MD 0.926
MIT-S1 [M=1.2, = K , n=0.7, m=1.55]
DG_D 0.716 -0.5 0

GS40 GS40_L 0.930 -0.75

0 0.25 0.5 0.75 1
GS40_MD 0.778
p'
GS40_D 0.694 /p'
o
GB GB_L 0.771
GB_MD 0.721 Fig. 15. (Color) Illustration of common models used in literature to fit
Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

GB_D 0.690 a yield locus for sand in stress plane ðq − p 0 Þ. Refer to Table 6 for more
a
See Table 1 for a detailed description of tested materials. details on each model.

0.8 0.9

0.7 0.8

0.6 Void Ratio, e 0.7

Void Ratio, e

0.5 0.6

0.4 0.5

0.3 F35_D 0.4 DG_D

F35_MD DG_MD
F35_L DG_L
Yield Locus Yield Locus
0.2 0.3
10 -2 10 -1 10 0 10 1 10 2 10 3 10 -2 10 -1 10 0 10 1 10 2 10 3
(a) Axial Stress, 1
(MPa) (b) Axial Stress, 1
(MPa)

0.9 0.8

0.8 0.7

0.7 0.6
Void Ratio, e

Void Ratio, e

0.6 0.5

0.5 0.4

GS40_D GB_D
0.4 0.3
GS40_MD GB_MD
GS40_L GB_L
Yield Locus Yield Locus
0.3 0.2
10 -2 10 -1 10 0 10 1 10 2 10 3 10 -2 10 -1 10 0 10 1 10 2 10 3
(c) Axial Stress, 1
(MPa) (d) Axial Stress, 1
(MPa)

Fig. 14. (Color) Evolution of e versus σ1 for the 12 oedometer experiments listed in Table 5, which were conducted on specimens of (a) F35; (b) DG;
(c) GS40; and (d) GB.

© ASCE 04020066-16 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

 Table 6. Summary of common models used in literature to describe yield locus of sand in stress plane ðq − p 0 Þ
Model Yield locus equation Source
p0 M2
Modified cam clay (MCC) ¼ Roscoe and Burland (1968)
po0 M 2 þ η2
   
ðq − βp 0 Þ2 p 0 n
MIT-E3 ¼ N2 1 − a
Whittle and Kavvadas (1994)
p 0 p00 p00
  0 n 
ðq − βp 0 Þ2 2 þ β 2 − 2βηÞ2 1 − p
MIT-S1 ¼ ðm a
Pestana and Whittle (1999)
p 02 p00
Note: Equations are plotted in Fig. 15.
a
N, n, and m are fitting parameters.
Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

Fig. 16. (Color) CSL modeled in compression plane ðe − p 0 Þ using the 12 oedometer experiments listed in Table 5, which were conducted on
specimens of (a) F35; (b) DG; (c) GS40; and (d) GB.

© ASCE 04020066-17 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

 0.95
e = 0.937 - 0.0121 (p' / pa )
R-squared = 0.862
0.9 e = 0.934 - 0.0136 (p' / pa ) F35 Loose Exp.
R-squared = 0.871

Void Ratio, e
0.85 DG Loose Exp.
e = 0.815 - 0.0126 (p' / pa )
R-squared = 0.837

0.8 GS40 Loose Exp.

e = 0.742 - 0.0093 (p' / pa )
R-squared = 0.960
0.75 GB Loose Exp.

0.7
0 0.5 1 1.5 2 2.5 3 3.5
p'
(a) ( /p )
a
Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

0.9
e = 0.871 - 0.0098 (p' / pa )
R-squared = 0.940

0.85
e = 0.863 - 0.0079 (p' / pa )
F35 Medium Dense Exp.
R-squared = 0.835
Void Ratio, e

0.8 e = 0.777 - 0.0105 (p' / pa ) DG Medium Dense Exp.

R-squared = 0.841

GS40 Medium Dense Exp.

0.75
e = 0.687 - 0.0016 (p' / pa )
R-squared = 0.090 GB Medium Dense Exp.
0.7

0.65
0 0.5 1 1.5 2 2.5 3 3.5
p'
(b) ( /p )
a

0.85
e = 0.813 - 0.0094 (p' / pa )
R-squared = 0.987
0.8 F35 Dense Exp.
e = 0.810 - 0.0069 (p' / pa )
R-squared = 0.407
Void Ratio, e

0.75 e = 0.761 - 0.0246 (p' / pa ) DG Dense Exp.

R-squared = 0.986

0.7 GS40 Dense Exp.

e = 0.647 - 0.0054 (p' / pa )
R-squared = 0.673
0.65 GB Dense Exp.

0.6
0 0.5 1 1.5 2 2.5 3 3.5
p'
(c) ( /p )
a

Fig. 17. (Color) CSL modeled in compression plane ðe − p 0 Þ using the 50 CTC experiments summarized in Table 2, which were conducted on
(a) loose; (b) medium dense; and (c) dense specimens.

Table 7. Summary of regression model results fitted in Fig. 16 subjected to CTC conditions. Processed 3D images affirmed a non-
Regression equation e ¼ Γ − λðp =pa Þ 0 α uniform distribution of e within the scanned specimens as they ap-
Regression Tested proached the CS due to shear strain localization. elocal measured at
plot materiala Γ λ the center of the specimen exhibited a very different evolution with
Fig. 16(a) F35 0.736 (0.727, 0.745) 0.0097 (0.0010, 0.0095) ε1 in comparison to eglobal , particularly when a single shear band
Fig. 16(b) DG 0.931 (0.920, 0.943) 0.0165 (0.0170, 0.0160) developed within the specimen at the CS. As proposed by CS
Fig. 16(c) GS40 0.940 (0.9289, 0.9518) 0.0180 (0.0185, 0.0175) theory, scanned specimens with initial medium dense and dense
Fig. 16(d) GB 0.857 (0.841, 0.873) 0.0110 (0.0115, 0.0105) states attained equal elocal at the CS when tested at the same σ3 .
Note: Values in parentheses denote 95% confidence limits of parameter. Therefore, the CS zone in the compression plane was attributed to
a
See Table 1 for a detailed description of tested materials. the reliance on measurements of global change in volume provided

© ASCE 04020066-18 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

Fig. 18. (Color) Yield locus for (a) F35; (b) DG; (c) GS40; and (d) GB.

Table 8. Summary of CS parameters for tested granular materials

CS parameters
Tested
materiala Yield locus M Γ λ
F35 Fig. 18(a) 1.281 (1.310, 1.252) 0.736 (0.727, 0.745) 0.0097 (0.0010, 0.0095)
DG Fig. 18(b) 1.354 (1.392, 1.316) 0.931 (0.920, 0.943) 0.0165 (0.0170, 0.0160)
GS40 Fig. 18(c) 1.361 (1.393, 1.329) 0.940 (0.9289, 0.9518) 0.0180 (0.0185, 0.0175)
GB Fig. 18(d) 1.064 (1.111, 1.018) 0.857 (0.841, 0.873) 0.0110 (0.0115, 0.0105)
Note: Values in parentheses denote 95% confidence limits of parameter.
a
See Table 1 for a detailed description of tested materials.

by the conventional triaxial apparatus to calculate the evolution of e plane compared to the CTC results. Accordingly, the yield locus
toward the CS. Alternatively, the CSL was quantified in the com- and the CSL in p 0 − q − e space were established in Fig. 18 for
pression plane using the results of oedometer tests on specimens the four granular material by synthesizing the results of the
composed of the same granular materials. The oedometer speci- CTC and oedometer experiments. That is, the CS parameter M
mens inhibit the development of internal shear bands, which leads was measured using the CTC results, while Γ and λ were deter-
to accurate quantification of the CS parameters in the compression mined based on the oedometer tests, as summarized in Table 8.

© ASCE 04020066-19 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

 Acknowledgments Held at St Catherine's College, 186–198. London: Thomas Telford
Publishing.
This material was partially funded by the US National Science Dafalias, Y. F., and M. T. Manzari. 2004. “Simple plasticity sand model
Foundation (NSF) under Grant CMMI-1266230. Any opinions, accounting for fabric change effects.” J. Eng. Mech. 130 (6): 622–634.
findings, conclusions, and recommendations expressed in this https://doi.org/10.1061/(ASCE)0733-9399(2004)130:6(622).
paper are those of the authors and do not necessarily reflect the Desrues, J., and G. Viggiani. 2004. “Strain localization in sand: An over-
views of the NSF. The SMT scans presented in this paper were view of the experimental results obtained in Grenoble using stereopho-
togrammetry.” Int. J. Numer. Anal. Methods Geomech. 28 (4): 279–321.
collected using the X-Ray Operations and Research Beamline
https://doi.org/10.1002/nag.338.
Station 13-BMD of the Advanced Photon Source (APS), a US Druckrey, A., K. Alshibli, and R. Al-Raoush. 2016. “3D characterization of
Department of Energy (DOE) Office of Science User Facility sand particle-to-particle contact and morphology.” Comput. Geotech.
operated by the Argonne National Laboratory (ANL) under 74 (Apr): 26–35. https://doi.org/10.1016/j.compgeo.2015.12.014.
Contract DE-AC02-06CH11357. We acknowledge the support Druckrey, A., K. Alshibli, and R. Al-Raoush. 2018. “Discrete particle trans-
of GeoSoilEnviroCARS (Sector 13), which is funded by the lation gradient concept to expose strain localisation in sheared granular
NSF Earth Sciences (EAR-1128799), and the DOE Geosciences materials using 3D experimental kinematic measurements.” Géotechni-
(DE-FG02-94ER14466). We thank Dr. Mark Rivers for his que 68 (2): 162–170. https://doi.org/10.1680/jgeot.16.P.148.
Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

guidance at APS. Evans, T. M., and J. D. Frost. 2010. “Multiscale investigation of shear
bands in sand: Physical and numerical experiments.” Int. J. Numer.
Anal. Methods Geomech. 34 (15): 1634–1650. https://doi.org/10.1002
/nag.877.
References Ferreira, P., and A. Bica. 2006. “Problems in identifying the effects of struc-
ture and critical state in a soil with a transitional behaviour.” Géotech-
Alshibli, K., S. Batiste, and S. Sture. 2003. “Strain localization in sand: nique 56 (7): 445–454. https://doi.org/10.1680/geot.2006.56.7.445.
Plane strain versus triaxial compression.” J. Geotech. Geoenviron. Eng.
Finno, R. J., and A. L. Rechenmacher. 2003. “Effects of consolidation his-
129 (6): 483–494. https://doi.org/10.1061/(ASCE)1090-0241(2003) tory on critical state of sand.” J. Geotech. Geoenviron. Eng. 129 (4):
129:6(483). 350–360. https://doi.org/10.1061/(ASCE)1090-0241(2003)129:4(350).
Alshibli, K., and M. Cil. 2018. “Influence of particle morphology on the Frost, J. D., and D.-J. Jang. 2000. “Evolution of sand microstructure during
friction and dilatancy of sand.” J. Geotech. Geoenviron. Eng. 144 (3): shear.” J. Geotech. Geoenviron. Eng. 126 (2): 116–130. https://doi.org
04017118. https://doi.org/10.1061/(ASCE)GT.1943-5606.0001841. /10.1061/(ASCE)1090-0241(2000)126:2(116).
Alshibli, K., A. Druckrey, R. Al-Raoush, T. Weiskittel, and N. Lavrik. Fu, P., and Y. F. Dafalias. 2011. “Fabric evolution within shear bands of
2014. “Quantifying morphology of sands using 3D imaging.” J. Mater. granular materials and its relation to critical state theory.” Int. J. Numer.
Civ. Eng. 27 (10): 04014275. https://doi.org/10.1061/(ASCE)MT.1943 Anal. Methods Geomech. 35 (18): 1918–1948. https://doi.org/10.1002
-5533.0001246. /nag.988.
Alshibli, K., and A. Hasan. 2008. “Spatial variation of void ratio and shear Gao, Z., J. Zhao, X. S. Li, and Y. F. Dafalias. 2014. “A critical state sand
band thickness in sand using X-ray computed tomography.” Géotech- plasticity model accounting for fabric evolution.” Int. J. Numer. Anal.
nique 58 (4): 249–257. https://doi.org/10.1680/geot.2008.58.4.249. Methods Geomech. 38 (4): 370–390. https://doi.org/10.1002/nag.2211.
Alshibli, K., M. Jarrar, A. Druckrey, and R. Al-Raoush. 2016. “Influence of Garga, V. K., and H. Zhang. 1997. “Volume changes in undrained triaxial
particle morphology on 3D kinematic behavior and strain localization of tests on sands.” Can. Geotech. J. 34 (5): 762–772. https://doi.org/10
sheared sand.” J. Geotech. Geoenviron. Eng. 143 (2): 04016097. https:// .1139/t97-038.
doi.org/10.1061/(ASCE)GT.1943-5606.0001601. Gu, X., M. Huang, and J. Qian. 2014. “Discrete element modeling of
Alshibli, K., and L. Roussel. 2006. “Experimental investigation of shear band in granular materials.” Theor. Appl. Fract. Mech. 72 (Aug):
slip-stick behaviour in granular materials.” Int. J. Numer. Anal. Methods 37–49. https://doi.org/10.1016/j.tafmec.2014.06.008.
Geomech. 30 (14): 1391–1407. https://doi.org/10.1002/nag.517. Harris, W. W., G. Viggiani, M. A. Mooney, and R. J. Finno. 1995. “Use
Altuhafi, F., B. A. Baudet, and P. Sammonds. 2010. “The mechanics of of stereophotogrammetry to analyze the development of shear bands
subglacial sediment: An example of new ‘transitional’ behaviour.” in sand.” Geotech. Test. J. 18 (4): 405–420. https://doi.org/10.1520
Can. Geotech. J. 47 (7): 775–790. https://doi.org/10.1139/T09-136. /GTJ11016J.
Altuhafi, F., and M. R. Coop. 2011. “Changes to particle characteristics Hasan, A., and K. Alshibli. 2012. “Three dimensional fabric evolution of
associated with the compression of sands.” Géotechnique 61 (6): sheared sand.” Granular Matter 14 (4): 469–482. https://doi.org/10
459–471. https://doi.org/10.1680/geot.9.P.114. .1007/s10035-012-0353-0.
Amirrahmat, S., A. Druckrey, K. Alshibli, and R. Al-Raoush. 2018. “Micro Imseeh, W. H., A. M. Druckrey, and K. A. Alshibli. 2017. “3D experimen-
shear bands: Precursor for strain localization in sheared granular mate- tal quantification of fabric and fabric evolution of sheared granular
rials.” J. Geotech. Geoenviron. Eng. 145 (2): 04018104. https://doi.org materials using synchrotron micro-computed tomography.” Granular
/10.1061/(ASCE)GT.1943-5606.0001989. Matter 20 (2): 24. https://doi.org/10.1007/s10035-018-0798-x.
Andò, E., E. Salvatore, J. Desrues, P. Charrier, J.-B. Toni, G. Modoni, Iwashita, K., and M. Oda. 1998. “Rolling resistance at contacts in simu-
and G. Viggiani. 2017. “Strain localisation in sand in cycles of triaxial lation of shear band development by DEM.” J. Eng. Mech. 124 (3):
compression and extension: Continuum and grain-scale analysis.” 285–292. https://doi.org/10.1061/(ASCE)0733-9399(1998)124:3(285).
In Proc., Int. Workshop on Bifurcation and Degradation in Geomate- Iwashita, K., and M. Oda. 2000. “Micro-deformation mechanism of shear
rials, 489–497. New York: Springer. banding process based on modified distinct element method.” Powder
Been, K., and M. G. Jefferies. 1985. “A state parameter for sands.” Technol. 109 (1–3): 192–205. https://doi.org/10.1016/S0032-5910(99)
Géotechnique 35 (2): 99–112. https://doi.org/10.1680/geot.1985.35.2.99. 00236-3.
Coop, M. 1990. “The mechanics of uncemented carbonate sands.” Jefferies, M., and K. Been. 2000. “Implications for critical state theory from
Géotechnique 40 (4): 607–626. https://doi.org/10.1680/geot.1990.40 isotropic compression of sand.” Géotechnique 50 (4): 419–429. https://
.4.607. doi.org/10.1680/geot.2000.50.4.419.
Coop, M. 2015. “Limitations of a Critical State framework applied to the Jiang, M., H. Yan, H. Zhu, and S. Utili. 2011. “Modeling shear behavior
behaviour of natural and ‘transitional’ soils.” In Proc., 6th Int. Symp. and strain localization in cemented sands by two-dimensional distinct
on Deformation Characteristics of Geomaterials, IS-Buenos Aires, element method analyses.” Comput. Geotech. 38 (1): 14–29. https://doi
115–155. Amsterdam, Netherlands: IOS Press. https://doi.org/10.3233 .org/10.1016/j.compgeo.2010.09.001.
/978-1-61499-601-9-115. Kandasami, R. K., and T. G. Murthy. 2017. “Manifestation of particle mor-
Coop, M., and I. Lee. 1993. “The behaviour of granular soils at elevated phology on the mechanical behaviour of granular ensembles.” Granular
stresses.” In Predictive Soil Mechanics: Proc., Wroth Memorial Symp. Matter 19 (2): 21. https://doi.org/10.1007/s10035-017-0703-z.

© ASCE 04020066-20 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066

 Kwa, K., and D. Airey. 2016. “Critical state interpretation of effects of fines Schofield, A., and P. Wroth. 1968. Critical state soil mechanics. New York:
in silty sands.” Geotech. Lett. 6 (1): 100–105. https://doi.org/10.1680 McGraw-hill.
/jgele.15.00176. Shipton, B., and M. R. Coop. 2012. “On the compression behaviour of
Li, X. S., and Y. Wang. 1998. “Linear representation of steady state line for reconstituted soils.” Soils Found. 52 (4): 668–681. https://doi.org/10
sand.” J. Geotech. Environ. Eng. 124 (12): 1215–1217. https://doi.org .1016/j.sandf.2012.07.008.
/10.1061/(ASCE)1090-0241(1998)124:12(1215). Shipton, B., and M. R. Coop. 2015. “Transitional behaviour in sands with
Lu, Y., and D. Frost. 2010. “Three-dimensional DEM modeling of plastic and non-plastic fines.” Soils Found. 55 (1): 1–16. https://doi.org
triaxial compression of sands.” In Proc., Soil Behavior and Geo- /10.1016/j.sandf.2014.12.001.
Micromechanics, 220–226. Reston, VA: ASCE. https://doi.org/10 Tagliaferri, F., J. Waller, E. Andò, S. A. Hall, G. Viggiani, P. Bésuelle, and
.1061/41101(374)33. J. T. DeJong. 2011. “Observing strain localisation processes in bio-
Marschi, N. D., C. K. Chan, and H. B. Seed. 1972. “Evaluation of proper- cemented sand using x-ray imaging.” Granular Matter 13 (3): 247–250.
ties of rockfill materials.” J. Soil Mech. Found. Div. 98 (1): 95–114.
https://doi.org/10.1007/s10035-011-0257-4.
Martins, F. B., L. A. Bressani, M. R. Coop, and A. V. D. Bica. 2001.
Theocharis, A. I., E. Vairaktaris, Y. F. Dafalias, and A. G. Papadimitriou.
“Some aspects of the compressibility behaviour of a clayey sand.”
2016. “Proof of incompleteness of critical state theory in granular me-
Can. Geotech. J. 38 (6): 1177–1186. https://doi.org/10.1139/t01-048.
chanics and its remedy.” J. Eng. Mech. 143 (2): 04016117. https://doi
McDowell, G., and M. Bolton. 1998. “On the micromechanics of crushable
.org/10.1061/(ASCE)EM.1943-7889.0001166.
Downloaded from ascelibrary.org by 147.83.201.96 on 04/09/24. Copyright ASCE. For personal use only; all rights reserved.

aggregates.” Géotechnique 48 (5): 667–679. https://doi.org/10.1680

/geot.1998.48.5.667. Todisco, M., M. Coop, and J.-M. Pereira. 2018. “Fabric characterisation in
Mooney, M. A., R. J. Finno, and M. G. Viggiani. 1998. “A unique critical transitional soils.” Granular Matter 20 (2): 20. https://doi.org/10.1007
state for sand?” J. Geotech. Geoenviron. Eng. 124 (11): 1100–1108. /s10035-018-0786-1.
https://doi.org/10.1061/(ASCE)1090-0241(1998)124:11(1100). Vesic, A. S., and G. W. Clough. 1968. “Behavior of granular materials
Nocilla, A., M. Coop, and F. Colleselli. 2006. “The mechanics of an under high stresses.” J. Soil Mech. Found. Div. 94 (3): 661–688.
Italian silt: An example of ‘transitional’ behaviour.” Géotechnique Wang, R., P. Fu, J.-M. Zhang, and Y. F. Dafalias. 2017. “Evolution of vari-
56 (4): 261–271. https://doi.org/10.1680/geot.2006.56.4.261. ous fabric tensors for granular media toward the critical state.” J. Eng.
Pestana, J. M., and A. J. Whittle. 1994. “Model prediction of anisotropic Mech. 143 (10): 04017117. https://doi.org/10.1061/(ASCE)EM.1943
clay behavior due to consolidation stress history.” In Proc., 8th Int. -7889.0001342.
Conf. Computer Methods and Advances in Geomechanics. Rotterdam, Whittle, A. J., D. J. DeGroot, C. C. Ladd, and T.-H. Seah. 1994. “Model
Netherlands: A.A. Balkema. prediction of anisotropic behavior of Boston blue clay.” J. Geotech.
Pestana, J. M., and A. J. Whittle. 1999. “Formulation of a unified constit- Eng. 120 (1): 199–224. https://doi.org/10.1061/(ASCE)0733-9410
utive model for clays and sands.” Int. J. Numer. Anal. Methods Geo- (1994)120:1(199).
mech. 23 (12): 1215–1243. https://doi.org/10.1002/(SICI)1096-9853 Whittle, A. J., and M. J. Kavvadas. 1994. “Formulation of MIT-E3 con-
(199910)23:12<1215::AID-NAG29>3.0.CO;2-F. stitutive model for overconsolidated clays.” J. Geotech. Eng. 120 (1):
Petalas, A. L., Y. F. Dafalias, and A. G. Papadimitriou. 2019. “SANISAND- 173–198. https://doi.org/10.1061/(ASCE)0733-9410(1994)120:1(173).
FN: An evolving fabric-based sand model accounting for stress principal Wood, D. M. 1990. Soil behaviour and critical state soil mechanics.
axes rotation.” Int. J. Numer. Anal. Methods Geomech. 43 (1): 97–123. New York: Press Syndicate of the Univ. of Cambridge the Pitt Building.
https://doi.org/10.1002/nag.2855. Xiao, Y., M. Coop, H. Liu, H. Liu, and J. Jiang. 2016. “Transitional behav-
Ponzoni, E., A. Nocilla, and M. Coop. 2017. “The behaviour of a gap iors in well-graded coarse granular soils.” J. Geotech. Geoenviron. Eng.
graded sand with mixed mineralogy.” Soils Found. 57 (6): 1030–1044.
142 (12): 06016018. https://doi.org/10.1061/(ASCE)GT.1943-5606
https://doi.org/10.1016/j.sandf.2017.08.029.
.0001551.
Poorooshasb, H., I. Holubec, and A. Sherbourne. 1966. “Yielding and flow
Xiao, Y., H. Liu, X. Ding, Y. Chen, J. Jiang, and W. Zhang. 2015.
of sand in triaxial compression. I.” Can. Geotech. J. 3 (4): 179–190.
“Influence of particle breakage on critical state line of rockfill material.”
https://doi.org/10.1139/t66-023.
Roscoe, K. H., and J. Burland. 1968. On the generalized stress-strain Int. J. Geomech. 16 (1): 04015031. https://doi.org/10.1061/(ASCE)GM
behaviour of wet clay. Berlin: ScienceOpen. .1943-5622.0000538.
Roscoe, K. H., A. Schofield, and C. Wroth. 1958. “On the yielding of soils.” Xu, L., and M. Coop. 2017. “The mechanics of a saturated silty loess with a
Géotechnique 8 (1): 22–53. https://doi.org/10.1680/geot.1958.8.1.22. transitional mode.” Géotechnique 67 (7): 581–596. https://doi.org/10
Salvatore, E., G. Modoni, E. Andò, M. Albano, and G. Viggiani. 2017. .1680/jgeot.16.P.128.
“Determination of the critical state of granular materials with triaxial Yang, J., and X. Luo. 2015. “Exploring the relationship between critical
tests.” Soils Found. 57 (5): 733–744. https://doi.org/10.1016/j.sandf state and particle shape for granular materials.” J. Mech. Phys. Solids
.2017.08.005. 84 (Nov): 196–213. https://doi.org/10.1016/j.jmps.2015.08.001.
Santamarina, J., and G.-C. Cho. 2004.“Soil behaviour: The role of particle Zuo, L., and B. A. Baudet. 2015. “Determination of the transitional
shape.” In Proc., Advances in Geotechnical Engineering: The Skempton fines content of sand-non plastic fines mixtures.” Soils Found. 55 (1):
Conf., 604–617. London: Thomas Telford. 213–219. https://doi.org/10.1016/j.sandf.2014.12.017.

© ASCE 04020066-21 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 2020, 146(8): 04020066