See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/245410636

Discussion: The general and congruent effects of structure in

natural soils and weak rocks

Article  in  Géotechnique · January 1991

DOI: 10.1680/geot.1991.41.2.281

CITATIONS READS
255 375

2 authors, including:

Serge Leroueil
Laval University
239 PUBLICATIONS   11,148 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

COSTA-Canada View project

Structuration of sensitive clays View project

All content following this page was uploaded by Serge Leroueil on 15 March 2019.

The user has requested enhancement of the downloaded file.

 Leroueil, S. & Vaughan, P. R. (1991). Gtotechnique 40, No. 2, 281-284

DISCUSSION

The general and congruent effects of structure in natural soils and

weak rocks

S. LEROUEIL and P. R. VAUGHAN (1990) G.&echnique 40, No. 3, 467488

L. Picarelli, University of Naples indication of the size and consistency of the frag-
The Authors stress the role played by the struc- ments can be obtained by means of grain size
ture in the behaviour of natural soils and weak analyses carried out using different treatment
rocks. In particular, they state that ‘many of the procedures. Fig. 25 shows a case in which the fol-
naturally occurring materials treated in engineer- lowing procedures have been adopted
ing as soils have components of strength and
stiffness which cannot be accounted for by poro- dry sieving after rough hand separation of the
sity and stress-history alone’. single fragments (curve A)
In my experience this is certainly true for a air drying and grain size analysis by sedimen-
large number of geological formations out- tation without shaking (curve Be)
cropping all over the Italian peninsula, such as grain size analysis as in (b) after 3 or 10 cycles
pyroclastic soils and rocks, poorly cemented of wetting and drying (45”C, curves B, and
sandstones and conglomerates, mylonitized car- Bio)
bonatic rocks, marly clays and shales. Due to grain size analysis according to ASTM pro-
bond weakness, many of these materials may be cedures (curve C,)
defined as non-durable. In particular, in fine- as (6) after a prolonged working of the
grained hard soils and weak rocks the decay of material with a spatula to disaggregate all the
mechanical properties caused by softening or fragments completely (curve C,).
slaking may be remarkable. Accordingly, Botts The fragment size may vary noticeably with the
(1986) emphasized that for a correct classification energy of the treatment. Curve A gives an idea of
of these materials for engineering purposes, the the macroscopic appearance of the clay shale,
time effects in terms of degradability should be characterized by lenses of l-10 mm. Curves B,,
taken into account. B, and B,, show that simple dispersion in water
Actually, the clay shale formations along the follows cycles of wetting and drying giving rise to
Apennine chain in Italy are generally more more pronounced subdivision of the aggregates.
prone to landsliding than other apparently Curve C, gives the real grain size distribution; it
weaker materials. Among these formations, the is significantly different from that obtained using
so-called scaly clay shales are common and the ASTM procedure.
typical of the geological setting of Italy. Due to
intense tectonic activity imposing very high stress
states, these materials exhibit complex structure.
Frequently they are strongly fragmented, consist-
ing of l-10 mm hard shear lenses or flakes, with
an external polished surface. The individual frag-
ments are composed of other smaller fragments,
themselves representing aggregations of silty and
clayey particles. Furthermore, the soil mass is
crossed by an intricate system of very close prin-
cipal shears caused by local folds and shear dis-
placements.
Similar structures have been observed in single
shear zones of other, geologically very different,
deposits outcropping in different areas of the
I
world (e.g. see Cruden, Thomson & Tsui, 1989). 100
The presence of small aggregates of particles d, mm

resistant to the stresses imposed by the standard Fig. 25. Influence of pre-treatment on grain size distribu-
procedures for grain size determination has been tion of Brindisi di Montagna clay shale (after Coteccbia,
demonstrated by Rippa & Picarelli (1977). An del Prete, Federico, Fenelli, Pellegino & Picarelli, 1984)

281

Downloaded by [] on [15/03/19]. Copyright © ICE Publishing, all rights reserved.

 282 DISCUSSION

Undisturbed specimens o-7 \

9 Displacement normal to shear lenses
0 Displacement parallel to shear lenses 1 \

&moulded specimens

i”

L
0.6 y\

3 a\\
0.4 ‘.” \
‘1 ’
. ‘0’.
l \, ‘- ---0
0
0.2 l ul .----- 8 -._-’
.

ol0 100 200 300 400 500

a’. kPa

Fig. 26. Drained residual strength of Laviano clay shale

The influence of the aggregations on the

properties of the clay shale is shown in Fig. 26,
where the residual strength of undisturbed
remoulded specimens of Laviano clay shale is
shown. The residual strength of the remoulded
and 0.2---..-
10 100
Pressure
1000
(7” kPa
10000

specimens, measured by both the direct shear and Fig. 27. Swelling index of Fiumarella clay shale
Bromhead ring-shear apparatus, is significantly
lower than that of undisturbed specimens. These
clay aggregations. Such phenomena could involve
results do not agree with previous researches,
significant loss of strength.
which have demonstrated that the residual fric-
Picarelli (1986) showed that the effective shear
tion angle should be independent of the initial
strength parameters of Bisaccia clay- a highly
conditions of the samples, and is influenced only
plastic clay of the southern Apennines with
by the index properties (Skempton, 1964; Lupini,
wL = 128% and I, = 88%-may decay signifi-
Skinner & Vaughan, 1981). According to Chan-
cantly if it is allowed to swell for a long time
dler (1969) they could be explained only if the
under stresses lower than the swelling pressure
effective grain size of the undisturbed specimens is
(Fig. 29). This reduction in strength seems to be
considered to be coarser than that of the
due to the viscous deformations related to sec-
remoulded samples.
ondary swelling, but could also be due to
The results of many routine investigations of
destructuration phenomena during swelling.
the mechanical parameters of several scaly clay
However, other experiments carried out on the
shale formations give some indication that yield
same clays show that a noticeable shear strength
either in compression or in swelling could occur.
decrease may occur after sustained swelling under
Typically, in oedometer tests performed on undis-
very low stresses, and reconsolidation to higher
turbed samples, swelling is as noticeable as it is in
stresses (Cicolella & Picarelli, 1990) or compres-
the Culebra shales quoted by the Authors; fur-
sion under a load higher than ai and gradual
thermore, in some cases the swelling index may
stress removal.
be as high as the compressibility index and higher
All the data reported agree with the Authors’
than the swelling index of remoulded specimens
ideas about destructuration. In particular, the
(Fig. 27), i.e. it seems that in the compression (or
coupled effects of stress removal and time could
the swelling) phase a sort of destructuration
play a fundamental role. The engineering signifi-
occurs, increasing the swelling potential of the
cance of these phenomena is evident, mostly in
material. However, if compression-swelling
problems in slope stability and deep excavations.
cycles are carried out while the maximum pres-
sure is gradually increased, an increase in the
swelling index may be observed (Fig. 28); it prob- Authors’ reply
ably depends on a gradual failure of the dia- Professor Picarelli extends the arguments m
genetic bonds between the particles forming the the Paper in two important directions. The first

Downloaded by [] on [15/03/19]. Copyright © ICE Publishing, all rights reserved.

 DISCUSSION 283

0
,---
/ -\
/
--iI
\ &moulded /
/
Fiemoulded
/O

Undisturbed

01 I 1 1
1 10 100 1000
OCR

(b)

Fig. 28 (left and above). Results of oedometer tests on

Laviano clay shale

0.2 Vancoloured clay (6s): (7. = 06 MPa

m
4 0 Swellhlg. 2 days
l Swelhng- 10-100 days/O
i I

I
100
Pressure
1000
(1’ kPa
I I
10000 ~“/Az/; 1 ,
(a)
O-l o-2 0.3 o-4 0.5
Normal stress u’: MPa

Fig. 29. Influence of swelling time on drained peak

involves the destructuring of weak mudrocks of strength of Bisaccia clay shale
high plasticity through swelling, and particularly
through slaking with cyclic wetting and drying.
The development of additional macrostructure in strength would be measured as in the same
the form of fissures and fractures during these material which had been disaggregated to a clay
processes may dominate eventual material behav- before shear. However, this strength may be
iour. Quite rapid chemical changes can also irrelevant. If the properties of the destructured
occur, and all these processes take time. They soil are used as a reference framework to deter-
need to be better understood when one is dealing mine the contribution of structure to the behav-
with those many argillaceous materials which iour of the structured soil, the degree and manner
appear at first sight to be rock but which, with of destructuring need to be selected with care. It
exposure, change to clay. would seem logical to destructure to the aggre-
The second point concerns the aggregated gated but particulate state to determine the con-
structure of many of these materials. When such tribution of bonding between aggregations, and
materials are destructured there is a primary to disaggregate to a clay if the effects of aggre-
breakdown to the aggregated particles, followed gation are to be determined. This area requires
by further breakdown of the aggregates to a clay. more research.
Similar effects occur in many residual soils. Pro-
fessor Picarelli shows the effect of this even on
residual strength. It is probably true that, if shear REFERENCES
of the undisturbed material were continued long Botts, M. (1986). The e&~s of slaking on engineering
enough, then the aggregations would be slowly hehavior of clay shales. PhD thesis, University of
degraded, and, eventually, the same residual Colorado.

Downloaded by [] on [15/03/19]. Copyright © ICE Publishing, all rights reserved.

 284 DISCUSSION

Chandler, R. J. (1969). The effect of weathering on the Lupini, J. F., Skinner, A. E. & Vaughan, P. R. (1981).
shear strength properties of Keuper marl. Gtotech- The drained residual strength of cohesive soils. G.&o-
nique 19, No. 3, 321-334. technique 31, No. 2, 181-213.
Cicolella, A. & Picarelli, L. (1990). Decadimento mecca- Picarelli, L. (1986). Caratterizzazione geotecnica dei
nice di una tipica argilla a scaglie di alta plasticita. terreni strutturalmente complessi nei problemi di sta-
Riv. Ital. Geotecn. 24, No. 1. bilita dei pendii. Proc. 16th Ital. Conf: Soil Mech..
Cotecchia, V., de1 Prete, M., Federico, A., Fenelli, G. B., Bologna 3,&155-169.
Pellegrino, A. & Picarelli, L. (1984). Some observa- Rippa, F. & Picarelli, L. (1977). Some considerations on
tions on a typical mudslide in a highly tectonized index properties of Southern Italy shales. The gko-
fomation in Southern Apennines. Proc. 4th Int. technics of structurally complex formations 1, 401-
Symp. Landslides, Toronto 2, 394%. 406.
Cruden, D. M., Thomson, S. & Tsui, P. C. (1989). The Skempton, A. W. (1964). Long-term stability of clay
geotechnical characteristics of an ice-thrust mud- slopes. Gkotechnique 14, No. 2, 77-101.
stone, Wabamun lake area, Alberta. Can. Geotech. J.
26, No. 2, 227-234.

Downloaded by [] on [15/03/19]. Copyright © ICE Publishing, all rights reserved.

View publication stats