Вы находитесь на странице: 1из 4

Indian Geotechnical Conference IGC2016

15-17 December 2016, IIT Madras, Chennai, India

EFFECT OF CONFINING PRESSURE ON POST PEAK BEHAVIOUR


OF IGNEOUS ROCK

RAKESH KUMAR
Head Geotechnical & Tunnels, AF Consult India Pvt. Ltd., Noida-201301 (UP)
rakesh_iitd@rediffmail.com

K.G. SHARMA
Professor, Department of Civil Engineering, Indian Institute of Technology, Delhi-110016.
kgsharma@civil.iitd.ernet.in

A.VARADARAJAN
Former Dogra Chair Professor, IIT Delhi, Professor, Civil Engg. Dr.M GR Educational and Research University, Chennai
avr2466@yahoo.co.in

ABSTRACT: M any hydroelectric, metro rail and nuclear repository projects are under construction in the country. The
structures for these projects, which include dams, tunnels, power house caverns, are mostly constructed on/in rocks. The
characterization of the behaviour of rock forms essential part of the analysis and design of these structures. M any of the
rocks around tunnels and underground caverns exhibit strain softening behaviour under loading. Strain softening is
defined as the progressive loss of strength when material is compressed beyond peak.
The present paper deals with the testing of an Igneous rock i.e. Basalt from hydroelectric project located in M adhya
Pradesh in India. The testing has been conducted using closed-loop-servo-controlled testing machine under strain
controlled loading. The specimens were prepared as per ISRM standards. The diameter of the specimen was 54 mm and
the length of the specimen was 108 mm. The index properties of the rock were determined as per ISRM standards.
Triaxial tests were conducted under various confining pressures at a constant axial strain rate of 9.259 X 10 -6 /s. The
strains were measured with the help of strain gauges and specially designed extensometers. The stress -strain-volume
change relationships of the tests are presented and discussed in this paper.
The behaviour of Basalt is also predicted using an elastoplastic constitutive model i.e. M ohr-Coulomb strain softening
model using FLAC. The material parameters for the model are determined from the experimental results. The stress-strain
volume change response of the Basalt is then predicted using the material parameters and is compared with the observed
results. The predictions using the constitutive modes are found satisfactory and comparable with the experimentally
observed results.
KEYWORDS: Strain softening, Closed-loop-servo-control, Strain controlled, Constitutive Modeling.

not possible to capture strain softening behaviour with


1 INTRODUCTION
conventional equipments which are useful to characterize
The understanding of the behaviour of rocks plays an behaviour upto peak only. The strain softening can be
important role in the analysis and design of structures captured only by using stiff testing machine or closed-
constructed in/on rocks. M any of the rocks around tunnels loop servo-controlled testing machine (Hudson et al.
and underground caverns exhibit strain softening 1972). The constitutive models used to characterize the
behaviour under loading. Strain softening is defined as the strain softening behaviour are elastic, elastoplastic and
progressive loss of strength when material is compressed elasto-viscoplastic (Shang & Yao 1999). The paper deals
beyond peak(Karstunen et al. 1997). with characterization of material behaviour of Basalt in
The use of strain softening behaviour is of paramount the laboratory under confining pressures of 0 M Pa, 10
importance in arriving at rational design of the structures M Pa, 20 M Pa and 30 M Pa, using closed-loop servo-
constructed on the rocks. The rock behaviour can be controlled testing machine.
obtained only after testing the rock specimens in the
laboratory under various loading conditions. However it is

1
Effect of Confining Pressure on Post Peak Behaviour of Igneous Rock

2 INDEX PROPERTIES OF ROCK

The rock used for testing is Basalt. The rock is from a


hydroelectric project located in M adhya Pradesh in India.
The rock is fine grained and greenish black in colour. The
X-ray diffraction and microscopic studies have been
carried out to determine the mineral composition of the
rock. The mineral composition is quartz (52%), iron oxide
(1%), amphiboles (26%), pyroxenes (16%), olivine and
serpentine (5%).The specific gravity of the rock is 2.89.
The index properties of the rock as per ISRM standards
are determined in the laboratory and are presented in
Table 1.
Table1 Index Properties of Basalt

Property Value

Dry density (kN/m3) 28.100 Fig. 1. S tress-Strain Behaviour for Basalt


Saturated density (kN/m3) 28.100
Void Ratio (e) 0.0041
Porosity (n) % 0.4100
Tensile strength (M Pa) 13.700
UCS (M Pa) 69.900

3 EXPERIMENTAL PROGRAMME

The strain controlled compression tests under confining


pressures of 0 M Pa. 10 M Pa, 20 M Pa and 30 M Pa were
conducted on Basalt using closed-loop servo-controlled
testing machine. The machine works on the principle of
closed-loop control which is described elsewhere (Hudson
et al. 1972). The machine used for testing has a loading
capacity of 1000 kN and has loading rate capability in the Fig. 2. Volume Change Response for Basalt
range of 0.001mm/s to 10.000 mm/s. The stiffness of the
machine is 1700 kN/mm. High pressure triaxial cell which As seen from Figures 1 & 2, it is observed that there is
has the capability of applying 0-140 M Pa confining gradual loss of strength from peak to residual strength
pressure and has sufficient space to accommodate with increase in strains. The effect of confining pressure
extensometers inside, is used for the testing. The is evidently noticed on the stress-strain-volume change
confining pressure was applied through hydraulic oil with behaviour of rock. The effect of softening and dilatancy
the help of high pressure confining unit. The specimens decreases with increase in confining pressure.
were prepared with 54 mm diameter and 108 mm height
as per ISRM . The strains were measured with the help of
HM M strain gauges and specially designed extensometers. 4 CONSTITUTIVE MODELING
The axial and lateral extensometers having a gauge length
of 82.0 mm and 25.4 mm respectively, were placed on the The material behaviour can be characterized by various
specimen and held in position with the help of springs. constitutive models like local models, continuum damage
The tests were conducted under a constant axial strain rate mechanics models (CDM ), smeared crack models,
of 9.259 X 10-6 /s. The stress-strain-volume change Cosserat continuum models, visco-plastic models and
response of the rock are presented in Figures 1 to 2. models based on the disturbed state concept (Desai 1995
& 2001, Varadarajan et al. 2001). In the present paper
M ohr-Coulomb strain softening model have been used to
characterize the material behaviour of Basalt. The
complete details of the M ohr-Coulomb model are given in
FLAC Version 4.0 manual (2002) and Kumar (2007). The
salient features of the model are discussed herein.

2
Indian Geotechnical Conference IGC2016
15-17 December 2016, IIT Madras, Chennai, India

4.1 Mohr-Coulomb StrainSoftening Model elsewhere (FLAC Version 4.0 manual 2002 and Kumar
2007). The material parameters for Basalt based on M ohr-
In this model, the non-associated shear and associated Coulomb model are presented in Figures 3 and 4. The
tension flow rules are used. The difference, however, lies other details are given in Kumar (2007).
in the possibility that the cohesion, friction, dilation and
tensile strength may harden or soften after the onset of
plastic yield. In the conventional M ohr-Coulomb model,
the properties are assumed to remain constant. Here, the
user can define the cohesion, friction and dilation as
piecewise-linear functions of a softening parameter
measuring the plastic shear strain. The yield and potential
functions, plastic flow rules and stress corrections are
identical to those of the M ohr-Coulomb model.

The failure envelope for shear is defined by the M ohr-


Coulomb yield function
Fig. 3. Variation of Cohesion (c) with e ps for Basalt
f s
1 3 N 2c N (1)

and tension yield function of the form

f t
t 3 (2)

where is the friction angle; c the cohesion; t the tensile

strength and
1 sin
N (3)
1 sin

Plastic shear strain is measured by the shear softening


parameter eps, whose incremental form is defined as
Fig. 4. Variation of Friction Angle () with e ps for
1

e ps 1

e1ps e mps
2 1

e mps
2 1
2 2
e 3ps e mps Basalt
2 2 2
(4) 6 PREDICTIONS
where, The stress-strain-volume change response of the Basalt
e ps
m 1
3 (e 1
ps
e ) ps
3 (5)
was carried out on the simulated test sample using M ohr-
Coulomb strain softening model using FLAC for various
confining pressures i.e. 0, 10, 20 & 30 M Pa. The results of
and e jps , j = 1, 3 are the major and minor principal the analyses are presented in terms of deviatoric stress,
plastic shear strain increments, respectively. axial, lateral and volumetric strains.

The other details of the model are given in FLAC Version 7 RESULTS AND DISCUSSION
4.0 manual (2002).
The predicted results of Basalt for 3 = 0, 10, 20 & 30
MPain terms of stress-strain volume change response
5 MATERIAL PARAMETERS using FLAC along with experimental results are presented
in Figures 5 to 6. The predictions are found to be
The material parameters of Basalt were determined using generally satisfactory and comparable with experimental
one set of experimental results viz., confining pressures of results.
0 M Pa, 10 M Pa and 20 M Pa and predictions are made for
these confining pressures (Group A) and remaining test
result i.e., confining pressure of 30 M Pa (Group B). The
details of the determination of the parameters is described

3
Effect of Confining Pressure on Post Peak Behaviour of Igneous Rock

Microstructure, H.B. M uhlhaus (Editor), John Wiley,


U.K.
Desai C.S. (2001) M echanics of M aterials and Interfaces:
The Disturbed State Concept, CRC Press, Boca RatonFL,
USA.
FLAC (version 4.0). 2002 Theory and Background
M anual.
Hudson J.A., Crouch S.L. and Fairhurst C. (1972)
Review-Soft, Stiff and Servo-Controlled Testing
M achines: A Review with Reference to Rock Failure,
Engg. Geology, 6: 155-189.
Karstunen M ., Pande G.N. and Desures J. (1997) Strain
Localisation and Rotation of Principal Stress Axis in
Fig. 5. Predicted S tress-S train Behaviour for Basalt
Biaxial Test, Proc. 9th Int. Conf. Comput. Meth. Adv.
Geomech., J. X. Yuan (Ed.), Wuhan, China.
Kumar Rakesh. 2007 Testing and Constitutive M odelling
of the Strain-softening Behaviour of Some Rocks, Ph.D.
Thesis, submitted to Indian Institute of Technology, Delhi.
Shang D.G. and Yao W.X. (1999) A Non-linear
Cumulative M odel for Uniaxial Fatigue, Int. J. Fatigue,
21: 187-194.
Varadarajan A., Sharma K.G., Desai C.S. and Hashemi M
(2001) Constitutive M odeling of a Schsitose Rock in the
Himalaya, Int. J. of Geomechanics, 1(1): 83-107.

Fig. 6. Predicted Volume Change Response for Basalt

8 CONCLUSIONS

The strain controlled compression tests have been


conducted on Basalt under unconfined state and confining
pressures of 10 M Pa, 20 M Pa and 30 M Pa in the
laboratory using closed-loop servo-controlled testing
machine. It is observed that there is gradual loss of
strength from peak to residual strength with increase in
strains. The effect of confining pressure is evidently
noticed on the stress-strain-volume change behaviour of
rock. The effect of softening and dilatancy decreases with
increase in confining pressure.
The M ohr-Coulomb strain softening model in FLAC have
been used to predict the rock behvaiour. The predicted
behaviour and observed behaviour for Basalt were found
to be in satisfactory agreement for both Group A and
Group B predictions.

References

Desai C.S. (1995) Constitutive M odelling using the


Disturbed State as M icrostructure Self-Adjustment
Concept, Chap. 8 in Continuum Models for Materials with