215
Proceedings of the 18
th
International Conference on Soil Mechanics and Geotechnical Engineering, Paris 2013
1
Misconceptions about experimental substantiation of creep hypothesis A
Les idées fausses justifiant l'hypothèse A de fluage au laboratoire
Degago S.A.
Norwegian Public Roads Administration, Trondheim, Norway
Grimstad G.
Oslo and Akershus University College of Applied Sciences, Oslo, Norway
Jostad H.P.
Norwegian Geotechnical Institute, Oslo and Norwegian University of Science and Technology, Trondheim, Norway
Nordal S.
Norwegian University of Science and Technology, Trondheim, Norway
ABSTRACT: Ample laboratory experiments as well as field observations show existence of rate effects or creep during the primary
consolidation phase of clayey soils. However, the role of creep during the primary consolidation phase has been a subject of active
debate among researchers. As a result, two totally different hypotheses referred to as creep hypothesis A and B have been defined as a
basis for the discussion. Despite being opposite extreme to each other, both creep hypotheses seem to be experimentally supported
leading to confusion as to which of them is correct. This paper aims to consistently clarify the apparent misconceptions involved in
the experimental substantiation of hypothesis A as well as discuss some of its unphysical implications. This means to provide simple
and convincing arguments as to why creep hypothesis A is not experimentally substantiated. It is shown that cohesive soils behave in
conformity with hypothesis B. It is also illustrated that a constitutive model based on hypothesis B can give excellent prediction of
long-term field measurements of settlements and excess pore pressure responses.
RÉSUMÉ : De nombreuse observations sur le terrain et au laboratoire ont montré que le fluage existe au cours de la phase de
consolidation primaire des argiles. Le rôle du fluage pendant la phase de consol
idation primaire fait toutefois l’
objet d'une vive
controverse dans la littérature. Au cours des ans deux hypothèses complètement différentes ont été définies pour discuter du fluage
dans les argiles ; l’hypothè
se A et B. Les deux hypothèses semblent être soutenues par des résultats en laboratoire et cela mène à
savoir laquelle des deux est correcte. Le but de cette étude est de clarifier certaines idées fausses concernant l’hypothèse A basé sur
des résultats expérimentaux. Ceci vise à renforce
r les arguments qui montre que l’hypothèse A n’est pas valide. L’étude montre
également que les sols cohérents se comportent pratiquement de la façon d’écrite par l’hypothèse B. Finalement, un modèle de
comportement basé su
r l’hypothèse B produit des résultats qui concordent très bien avec des mesures de tassements en chantier ainsi
qu’avec la réponse des pressions interstitielles due à la charge.
KEYWORDS: Creep, primary consolidation, settlement, clays
1 INTRODUCTION
Settlement of saturated soils under increased loading consists of
two successive phases, commonly referred to as the primary and
secondary consolidation phases. The primary consolidation
phase is dominated by pore pressure dissipation and effective
stress increase; whereas, the secondary consolidation phase is
dominated by creep at almost constant effective stress.
The existence of creep during primary consolidation is
evident, but there exist opposing opinions on the role of creep in
the primary consolidation phase. In 1977, Ladd et al. formally
proposed two creep hypotheses referred to as creep hypotheses
A and B. Creep hypothesis A implies that the end of primary
consolidation (EOP) strain and EOP preconsolidation stress (
p
′
c
)
are unique, independent of the consolidation duration; while,
hypothesis B implies that the EOP strain increases or EOP
p
′
c
decreases with increasing duration of the primary consolidation.
The two creep hypotheses have significance implications
when it comes to practical applications such as prediction of
field settlements where settlement behaviours of laboratory tests
(short primary consolidation duration) have to be extrapolated
to describe in-situ performance of clay deposits with very long
primary consolidation duration. Mesri (2003) and Leroueil
(2006) summarised several experimental and numerical
substantiations that have been independently presented to
advocate the two hypotheses. However, these voluminous
substantiations had little effect in deciding which of the two
hypotheses was correct as experimental observations were
presented that seem to advocate two opposing soil behaviours.
2 SUBSTANTIATIONS OF THE CREEP HYPOTHESES
The substantiations and arguments for hypothesis B are more
general as compared to the extreme hypothesis A which says
that the EOP strain and
p
′
c
are identical irrespective of the time
it takes to reach an EOP state. Hence, this motivates to take a
closer look at the experimental substantiation of hypothesis A.
Four groups of substantiations have been put forward by the
advocates of hypothesis A (Mesri, 2003). These arguments can
briefly be stated as; (1) comparing EOP strain of specimens
with different heights (here 127 and 508 mm thick specimens),
(2) studying sub-specimen compressibilities in interconnected
tests (3) predicting field settlements and excess pore pressures
using a numerical model developed based on hypothesis A
concept (referred to as ILLICON methodology) and (4)
comparing field and laboratory preconsolidation stresses.
Degago et al. (2009, 2010, 2011a and 2011b) and Degago
(2011) have thoroughly investigated the experimental and
numerical substantiations of hypothesis A and provided
explanations using a consistent framework as to why hypothesis
A seemed to be wrongly substantiated. Actually, the very same
data have been used to substantiate hypothesis B. However,
recent work by Mesri and Feng (2009) (published in 2011)
indicates a series of misconceptions by the advocates of
hypothesis A regarding substantiation of hypothesis A.
Therefore this paper attempts to briefly provide clarifications
for most of these misconceptions. In addition, it presents some
of the recent works performed with regard to the experimental
and numerical substantiations of the two creep hypotheses.
