Tunnelling in Brittle Rock

Mass Conditions
Spalling Instabilities

Giovanni Barla
Department of Structural and Geotechnical Engineering

Lecture Outline
Introduction
Laboratory Results
Field Evidence
Modelling Brittle Failure
Case Study
Conclusions

Introduction

Definitions

BRITTLE FAILURE
BRITTLE FAILURE takes place around
underground openings in massive to
moderately jointed rock masses
subjected to high in situ stresses. It
manifests itself in the form of spalling
resulting in a revised stable geometry
with the onset of V-shaped notches. The
extent and depth of failure is a function
of the in situ stress magnitudes relative
to the rock mass strength

BRITTLE FAILURE
The onset of wall yield due to boundary
compression around an underground
opening is one of the primary design
issues in hard rock tunnelling at depth by
tunnel boring machines or conventional
excavation methods. In this lecture we
will discuss spalling instabilities,
although in cases the interest need be
centred on dynamically induced tunnel
failures such as rockbursts

BRITTLE FAILURE

Mine by Experiment at the URL, Canada (1990-1995)

Roof Tensile Failure

Sidewall Spalling

Laboratory Results

The starting point in the

understanding of this type of
behaviour is to analyse
laboratory test results on
cylindrical samples
It is worth to notice that a
significant contribution to this
understanding has been derived
from recent discrete element
simulations of Lac du Bonnet
granite (Diederichs et al.2004)

Laboratory Testing
Testing
Equipment

Spring
Model

Linear
Rock
Rock Specimen
Specimen

Non
linear

Unstable crack
propagation

11

peak
peak

A
A IV

III
B

cd
cd
Stable crack
propagation

ci
ci
II
cc
cc

33

Peak

Linear elastic
deformation
Crack closure

11

Stages of stress-strain and acoustic response

in uniaxial testing (Bieniawski 1967, Eberhardt et al. 1998)

Major thresholds within a

stress-strain test on rock
samples coupled with acoustic
emissions monitoring
Crack closure, cc
cc
Crack initiation, ci
ci
Crack damage, cd
cd
Peak strength, peak
peak

Laboratory Testing

No feedback
resistance

Diederichs et al. 2004

Dilating
crack

Feedback
confinement

Induced hoop
compression

Excavation Boundary

Small hole in plate

Biaxial
Biaxial tests
tests run
run for
for Politecnico
Politecnico di
di Torino
Torino
at
at Dresden
Dresden University
University

Biaxial
Biaxial tests
tests run
run for
for Politecnico
Politecnico di
di Torino
Torino
at
at Dresden
Dresden University
University

Field Evidence

Brittle Failure
Field Observations

ProcessLocalised
zone
Process
Process
stabilised
by confining
Buckling
damage
zonetip
zone
stress at notch

Stage IV - Stabilization: Development of the notch stops

Stagethe
III notch
- Spalling
: Development
the process
zone leads
when
geometry
provides of
sufficient
confinement
to
to the II
of zone
thin
slabs.
These
thin
slabs
form
Stage
stabilize
the
Dilation:
process
Shearing
at
the :and
notch
crushing
tip.
This
usually
occurs
means
inby:
a
Stage
Iformation
-- Damage
Initiation
Critically
oriented
flaws
shearing,
splitting
Themaximum
thickness
oftangential
the
slabs
there
is
a slight
tear-drop
like curvature
to dilation
the notch
shape.
very
narrow
process
zone.
Extensive
at
the
are
exploited
inand
thebuckling.
zone
of
varies.size
The
thickest
slabs
formthe
asnotch
the zone
notch
its
Alternatively,
if the
slabs
flanks reaches
are of
heldthe
in
grain
scale
occurs
infrom
this
process
stress.
The
process
initiates
at
the
boundary
maximum
size. Near
the notch
the slabs are
curved
place
by artificial
support,
notchtip
development
will
also stop
tunnel

Martin et al. 1997

The notch development is a 3D process which

is directly linked to the tunnel advance
Stages IV

III

II

V
Tunnel Advance
Initiation

Tunnel Longitudinal Section

Martin et al. 1997

Brittle Failure
Field Observations
1.0
df/a

max
df
= 1.25
- 0.51 0.1

a
c

0.8
0.6

Df Depth of overbreak

0.4
0.2

2a

0.2

+
+
+ +

0.4

0.6

max/c
0.8

1.0

Kaiser et al. 1995

This collection of available tunnel

overbreak data shows that
(1) Stress induced fracturing initiates at
approximately 0.3-0.5 cc and the critical
deviatoric stress for yield is essentially
independent of confining stress
(2) No overbreak occurs at a maximum
boundary stress equivalent to 40% of the
compressive strength. This is a lower bound
strength as unfailed tunnels are not plotted

Shear Failure

Slope of Spalling Limit

depends on Heterogeneity,
Surface Effects, Damage,
Stress Rotation

1/c

Distributed Damage
and Acoustic Emission
Axial Splitting

Spalling
Failure

Damage
Threshold (m=0)

Tensile Failure

Damage Initiation
Threshold depends on
Mineralogy, Grain Size,
Bond Type

In situ
Strength

Diederichs et al. 2004

3/c

Brittle Failure Modelling

A number of possible
approaches
(1)
(1) Phenomenological
Phenomenological Models:
Models: use
use
appropriate
appropriate constitutive
constitutive equations
equations which
which
should
should describe
describe the
the brittle
brittle failure
failure processes
processes
based
based on
on back
back analysis
analysis of
of carefully
carefully
documented
documented case
case histories
histories

(2) Micromechanical Models: account for

microscopic aspects of rock fracture and/or
crack propagation mechanisms

We discuss (1) only

One common approach to estimate the yield

potential and the depth of disturbance for a
tunnel is the Hoek and Brown Criterion for
rock mass

11

= + cc mbb
+ s
cc

''
1
1

33

cc

''
3
3

''
3
3

Standard uniaxial
compressive
strength

s
mbb

11

Costants which
depend upon the
characteristics of
the rock mass
determined using
the GSI index

33

It is found that this approach

is of limited reliability to
predict brittle failure for rock
masses of good quality,
typically GSI > 70-75

mb=0 Model
At low confinement levels, the accumulation
of significant rock damage, equivalent to loss
of cohesion (i.e. mbb=0), ocurs when 11- 33 =
1/3 to 1/2 cc (i.e. when s= 0.11 to 0.25)
11/
/ cc
1.5
1.5

- = s c = 0.33-0.50 c

1.0
1.0

'
1

0.5
0.5
0.4
0.4

1/2

0.3
0.3

0.5
0.5 1.0
1.0

33/
/ cc

'
3

Kaiser et al. 2000

CWFS Model
The cohesive strength is gradually
destroyed by tensile cracking and crack
coalescence. The frictional strength can be
mobilized only when the cohesive strength
is significantly reduced

aa

III

IV

II
Onset of
microcracking

cr
ci

Strength components
are strain- dependent

Frictional
Strength

aa
Hajiabdolmajid et al. 2002

Case Study

Pont Ventoux - Susa Hydropower System

DIAMETER: 4.75 m
TOOLS: 35 cutting disks
MAXIMUM THRUST: 6950 kN
TOTAL POWER: 895 kW

Free Flow Tunnel

Pont Ventoux - Susa

Hydropower System

F2 - Clarea
Length

Pont-Ventoux
F2 Length

Free Flow Tunnel

1500
1400
1300
1200
1100
1000
900
800
700
600
500
400
300
200
100
0

100
90
80
70
60
50

RMR6

Overburden [m]

Val Clarea F2 LENGTH

40
30
20
10
0

500

1000

1500

2000

2500

3000

3500

4000

4500

5000

5500

0
6000

Chainage [m]
Ground Surface

Gravity

Spalling

RMR 6

Instability of rock wedges occurs with Rock Mass Quality

Index RMR < 55
Spalling instability is observed to take place when the
overburden is greater than 500-600 m and RMR > 65

SECTION 1 (530-545 m)

Fault

Spalling

3
Spalling
The overbreak zones give the directions of 1
and 3 . From Flat Jack measurements the Stress
Ratio is in the range 0 25 to 0 50

SECTION 2 (550-565 m)
Spalling

Spalling
The overbreak zones give the directions of 1
and 3 . From Flat Jack measurements the Stress
Ratio is in the range 0 25 to 0 50

Intact Rock
(Ambin Formation)
500

Peak

400

1 - MPa

Gneiss

300

Residual

200
100

-10

10

ci
ci=135MPa

15

20

33 - MPa

25

30

mii=8.1

35

 1 - MPa

Rock Mass
(Ambin Formation)
250
225
200
175
150
125
100
75
50
25

-10 0

Mohr-Coulomb
Criterion

Gneiss

Hoek-Brown
Criterion

GSI=70
Ed=36.7GPa
10

mbb=2.8

30

50

33 - MPa

70

s=0.04

90

Elasto-Plastic
Ideally Brittle

Elastic
Ideally Plastic

11/
/ cc
1.5
1.5
1.0
1.0
0.5
0.5

0.5
/ cc
0.5 1.0
1.0 33/

m=0 and CWFS

Models
sb=0.25, c=80MPa

Folding
Directions

Tunnel Axis

3D Discrete Fracture Network model created by the Fracman code

with plot of joint sets. This is typical of rock conditions on the right wall

J1
J2

J3a

TUNNEL
FACE

J3b

(b)

(a)
Detail of deterministic model

Detail of DFN model

YIELDED BLOCKS AND SHEAR DISPLACEMENTS

AROUND THE TUNNEL

LEFT WALL
Q =10

CH 2359 m

Q = 0.007

(a)

Q =10

Q =0,007

(b)

RIGHT WALL