Preliminary Assessment of Potential Underground Stability (Wedge and Spalling) at Forsmark, Simpevarp and Laxemar Sites

R-05-71
Preliminary assessment of
potential underground stability
(wedge and spalling) at Forsmark,
Simpevarp and Laxemar sites
Derek Martin, University of Alberta
December 2005
Svensk Kärnbränslehantering AB
Swedish Nuclear Fuel
and Waste Management Co
Box 5864
SE-102 40 Stockholm Sweden
Tel 08-459 84 00
+46 8 459 84 00
Fax 08-661 57 19
+46 8 661 57 19
ISSN 1402-3091
SKB Rapport R-05-71
Preliminary assessment of
potential underground stability
(wedge and spalling) at Forsmark,
Simpevarp and Laxemar sites
Derek Martin, University of Alberta
December 2005
This report concerns a study which was conducted for SKB. The conclusions
and viewpoints presented in the report are those of the author and do not
necessarily coincide with those of the client.
A pdf version of this document can be downloaded from www.skb.se

Summary
In SKB’s Underground Design Premises /SKB 2004/ the objective in the early design
phase is to estimate if there is sufficient space for the repository at a site. One of the
conditions that could limit the space available is stability of the underground openings, i.e.,
deposition tunnels and deposition boreholes. The purpose of this report is to provide a
preliminary assessment of the potential for wedge instability and spalling that may be
encountered at the Forsmark, Simpevarp and Laxemar sites based on information from the
site investigations program up to July 30, 2004.
The rock mass spalling strength was defined using the in-situ results from SKB’s Äspö
Pillar Stability Experiment and AECL’s Mine-by Experiment. These experiments suggest
that the rock mass spalling strength for crystalline rocks can be estimated as 0.57 of the
mean laboratory uniaxial compressive strength. A probability-based methodology utilizing
this in-situ rock mass spalling strength has been developed for assessing the risk for
spalling in a repository at the Forsmark, Simpevarp and Laxemar sites. The in-situ stresses
and the uniaxial compressive strength data from these sites were used as the bases for the
analyses.
Preliminary findings from all sites suggest that, generally, the risk for spalling increases as
the depth of the repository increases, simply because the stress magnitudes increase with
depth. The depth at which the risk for spalling is significant, depends on the individual
sites which are discussed below.
The greatest uncertainty in the spalling analyses for Forsmark is related to the uncertainty
in the horizontal stress magnitudes and associated stress gradients with depth. The
confidence in these analyses can only be increased by increasing the confidence in the
stress and geology model for the site. From the analyses completed it appears that spalling
in the deposition tunnels can be controlled by orienting the tunnels approximately parallel
to the maximum horizontal stress. The analyses indicates however that the risk of spalling
for the deposition holes increases below a depth 550 m.
The spalling analyses for Simepvarp and Laxemar suggest that spalling will not be an issue
in the deposition tunnels or deposition holes down to a depth of 650 m for Stress Domain
II. In Stress Domain I, the analyses indicates that the risk for spalling in the deposition
holes increases significantly below a depth of 450 m. It should be noted the experience
with the underground excavations at a depth of 450 m at the nearby Äspö Hard Rock
Laboratory which is in a similar rock mass to that found at Simpevarp and Laxemar and
similar stress domain as Domain I, did not show any evidence of spalling. The mean
laboratory uniaxial compressive strength for the Simepvarp and Laxemar sites, which is
used to estimate the rock mass spalling strength, is considerably less than that used for the
Äspö diorite and hence may be under estimated. This needs further investigations.
The results from the Äspö Pillar Stability Experiment that utilized 1.8-m-diameter
boreholes suggest that should spalling be severe, as in this experiment, the increased
borehole volume due to spalling could approach 0.18 m3 . This volume increase is
approximately 18% greater then the 0.150 m3 proposed in /SKB 2004/ as an acceptable
volume increase per deposition hole. In the Äspö Pillar Stability Experiment the spalling
was caused by excavation-induced stresses, i.e., after drilling, and thermally-induced
3
stresses. The volume of spalling after drilling was approximately 30% of the total spalling
volume. Hence it is highly unlikely that the volume increase caused by drilling-induced
spalling will exceed the proposed allowable volume increase for any site.
Wedge analysis requires detailed knowledge of the fracture system at the repository level.
It was concluded that the DFN model may not contain sufficient information at this time
for detailed wedge stability analyses and hence may not be appropriate at this stage of site
investigation. Nonetheless a preliminary wedge analysis for Forsmark was carried out to
demonstrate the methodology that could be employed using the Discrete Fracture Network
(DFN) model. These analyses showed that the risk for wedge instability was very minor
and that these wedges could be handled by normal Scandinavian tunnel support practice.
These results are in keeping with construction experience from SKB underground facilities
at Forsmark (SFR), Clab and Äspö HRL suggests that wedge stability is not likely to
significantly influence the layout of repository or the preliminary design of the repository.
4
Sammanfattning
I SKB:s rapport Underground dEsign Premises /SKB, 2004/ är ett av målen med den
inledande projekteringen att bedöma om platsen är tillräckligt stor för ett slutförvar. En av
de faktorer som kan påverka tillgänglig plats är stabiliteten för undermarksutrymmena,
främst deponeringstunnlar och -hål. Syftet med denna rapport är att göra en preliminär
bedömning av risken för instabilitet på grund av kilar och spänningsinducerade spjälkbrott
vid Forsmark, Simpevarp och Laxemar, baserad på resultat som fanns tillgängliga upp till
och med juli 2004.
Bergmassans hållfasthet mot spänningsinducerat spjälkbrott har definiererats utifrån de
fullskaliga försöken av SKB - Äspö Pillar stability Experiement (APSE) och AECLs
Mine-by Experiment. Resultaten från dessa båda försök visar att bergmassans hållfasthet
mot spänningsinducerat spjälkbrott kan bestämmas till 0,57 av medelvärdet från enaxliella
tryckförsök. En sannolikhetsbaserad ansats har gjorts för att värdera risken för
spänningsinducerat spjälkbrott i ett slutförvar beläget i Forsmark, Simpevarp eller
Laxemar.
Preliminära resultat från analysen av samtliga platser visar generellt att risken för
spänningsinducerat spjälkbrott ökar med djupet, eftersom spänningsmagnituder ökar med
djupet. Det djup där risken för spänningsinducerat spjälkbrott kan vara signifikant varierar
mellan platserna, se nedan. I samtliga studerade fall kan risken för spänningsinducerat
spjälkbrott i deponeringstunnlarna minska radikalt, eller elimineras helt genom att
orientera dessa parallellt med största horisontalspänningen.
Den största osäkerheten i resultaten av spänningsinducerat spjälkbrott i Forsmark beror på
osäkerheten i magnitud för de horisontella spänningarna, samt dess gradient mot djupet.
Konfidensen i dessa analyser kan förbättras främst genom att öka tilltron till platsens
spännings- och geologimodeller. Enligt utförda analyser kan risken för spänningsinducerat
spjälkbrott kontrolleras om tunnlarna orienteras ungefär parallellt med den största
horisontalspänningen. Det finns dock en ökande risk för spjälkbrott i deponeringshål under
ca 550 m djup.
Analyser av spänningsinducerat spjälkbrott för Simpevarp och Laxemar indikerar att
spänningsinducerat spjälkbrott inte är ett problem inom spänningsdomän II åtminstone ner
till 650 m djup. Inom spänningsdomän I pekar analyserna på risk för spänningsinducerat
spjälkbrott i deponeringshål under ett djup av ca 450 m. Dock bör noteras, att
erfarenheterna från undermarksarbeten i det närliggande Äspö Hard Rock Laboratory på
samma djup i liknande berg, och med liknande spänningar som i domän I inte visar några
tecken på spänningsinducerade spjälkbrott. Medelvärdet för den enaxliga
tryckhållfastheten ifrån Simpevarp och Laxemar, som använts i denna rapport för analys av
risken för spänningsinducerat spjälkbrott, är signifikant lägre än motsvarande resultat från
Äspö dioriten, och kan därmed vara underskattad. Detta behöver utredas närmare.
Resultaten från APSE, som genomfördes i fullskaliga deponeringshål med diameter 1,8 m,
indikerar att en omfattande spjälkbrott kan orsaka ett utfall av storleksordninge 0,18 m3 för
ett deponeringshål. Detta är ca 18 % större än de 0,15 m3 som föreslogs i /SKB, 2004/
som den acceptabla volym överberg för ett deponeringshål. I APSE orsakades
spjälkbrottet av både själva borrningen och av termo-inducerade spänningar. Utfallande
5
volym i samband med borrningen var ca 30 % av den totalt avspjälkade volymen. I
jämförelse med de studerade platserna är det inte troligt att spänningsinducerat spjälkbrott
i samband med borrning av deponeringshål på någon av de studerade platserna skulle
orsaka utfall av överberg över de ansatta måtten.
Analys av risken för kilutfall kräver kunskaper om sprickfördelningen på förvarsdjupet.
Det bedöms att befintliga DFN-modeller troligen inte ännu innehåller tillräckligt detaljerad
information för kilbrottsanalyser, och kan därför vara olämpliga för detta syfte i det relativt
tidiga projekteringsskedet som denna studie gjorts för. En preliminär kilbrottsanalys
genomfördesändå för Forsmark med stöd av platsbeskrivningens DFN-modell. Analysen
visade på mycket liten risk för instabila kilar, och att dessa risker borde kunna hanteras
med sedvanlig skandinavisk praxis för bergförstärkning. Erfarenheter från SKB:s
undermarksanläggningar i Forsmark (SFR), Clab and Äspö HRL indikerar dessutom att
kilstabilitet troligen inte är en fråga som har signifikant påverkan på layout eller preliminär
design för slutförvaret.
6
Contents
1 Introduction 9
2 Risk terminology 11
3 Risk based methodology 13

3.1 Wedge stability methodology . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.1.1 Frequency of Wedges . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.1.2 Severity of Wedges . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.1.3 Wedge data uncertainty and probability . . . . . . . . . . . . . . . 13
3.2 Spalling assessment methodology . . . . . . . . . . . . . . . . . . . . . . 16
3.2.1 Spalling criterion . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2.2 Severity of Spalling . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2.3 EXCEL code for evaluating the spalling hazard . . . . . . . . . . . 22
3.2.4 Calibration of methodology . . . . . . . . . . . . . . . . . . . . . 25
3.3 Spalling for noncircular openings . . . . . . . . . . . . . . . . . . . . . . . 28
3.4 Spalling and thermally-induced stresses . . . . . . . . . . . . . . . . . . . 28
4 Forsmark analysis 31
4.1 Input data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4.1.1 Joint orientation and strength . . . . . . . . . . . . . . . . . . . . . 31
4.1.2 In-situ stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4.1.3 Uniaxial Compressive Strength . . . . . . . . . . . . . . . . . . . 33
4.2 Wedge stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.3 Spalling stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
4.3.1 Vertical Deposition Holes . . . . . . . . . . . . . . . . . . . . . . 38
4.3.2 Deposition Tunnels . . . . . . . . . . . . . . . . . . . . . . . . . . 38
4.3.3 Common central area caverns . . . . . . . . . . . . . . . . . . . . 40
5 Simpevarp analysis 41
5.1 Input data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
5.1.1 In-situ Stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
5.1.2 Uniaxial Compressive Strength . . . . . . . . . . . . . . . . . . . 43
5.2 Wedge stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
5.3 Spalling instability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
5.3.1 Vertical deposition holes . . . . . . . . . . . . . . . . . . . . . . . 45
5.3.2 Deposition tunnels . . . . . . . . . . . . . . . . . . . . . . . . . . 47
6 Laxemar analysis 51
6.1 Input data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
6.2 Wedge stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
6.3 Spalling instability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
6.3.1 Vertical deposition holes . . . . . . . . . . . . . . . . . . . . . . . 52
6.3.2 Deposition tunnels . . . . . . . . . . . . . . . . . . . . . . . . . . 52
7 Conclusions 55
7
References 56
Appendices 61
A Rock mass spalling strength 63

A.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
A.2 Rock mass spalling strength from in-situ experiments . . . . . . . . . . . . 65
A.2.1 AECL’s Mine-by Experiment . . . . . . . . . . . . . . . . . . . . 65
A.2.2 Äspö Pillar Stability Experiment . . . . . . . . . . . . . . . . . . . 66
B Forsmark Uniaxial Compressive Strength 69
C Simpevarp Uniaxial Compressive Strength 70
D Reliability of traditional in-situ stress measurements in high stress environment 71

D.1 Experience from 420 Level of AECL’s URL . . . . . . . . . . . . . . . . . 71
D.2 Forsmark stress data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
8
1 Introduction
In SKB’s Underground Design Premises /SKB 2004/ the objective in the early design
phase is to estimate if there is sufficient space for the repository at a site. One of the
conditions that could limit the space available is stability of the underground openings, i.e.,
deposition tunnels and deposition boreholes. Stability issues likely to be encountered at a
repository are discussed in detail in Martin et al. /2001/. The two major stability issues
considered in this report that could impact the design of the repository are (see Figure 1–1):
1. Structurally controlled gravity driven instability, referred to wedges or
falls-of-ground
2. Stress-driven instability referred to as spalling or slabbing.
Falling
Sliding
Gravity-induced Stress-induced spalling.

structually controlled Usually found in brittle, blocky
block movement and massive rock masses
(a) Wedge failure (b) Spalling failure
Figure 1–1: Illustration of the two significant modes of underground instability anticipated at a
repository.
The purpose of this report is to provide a preliminary assessment of the potential for wedge
instability and spalling that may be encountered at the Forsmark and Oskarshamn sites1 at
the depths of a repository (400, 500, 600 and 700 m) based on information from the site
investigations program up to July 30, 2004. The issues to be addressed in this report are:
1. The orientation of deposition tunnels that would minimise the volume of unstable
wedges in deposition tunnels.
2. The orientation of deposition tunnels with a view towards minimising the risk of
spalling.
3. The loss of deposition holes due to the risk of spalling.
1 Oskarshamn covers both the Simpevarp peninsula and Laxemar (the main inland) sites
9
At this stage in the site investigations and preliminary design it is not possible to answer
the questions above explicitly. Therefore, a probabilistic approach will be employed that
utilizes the range of data that are currently available.
10
2 Risk terminology
Geotechnical engineering is fundamentally about managing risk. Morgenstern /1995/

summarized risk assessment concepts using the framework for risk management adopted
by the Canadian Standards Association (Figure 2–1). Figure 2–1 is a generalised
framework for risk management that would need to be modified for individual projects.
But in all cases risk assessment requires identification of the hazard and quantifying the
risk associated with each hazard. Morgenstern noted that while quantitative risk analysis
(QRA) is one aspect of the framework, qualitative risk assessment is also a valuable
component of risk management. With risk defined as the combination of the probability of
occurrence of an undesired event and the possible extent of the event’s consequence, risk
can, in principle, be calculated. The full potential of QRA is best met with the
establishment of acceptable risk criteria. Establishing acceptable risk criteria in
geotechnical engineering is a challenge that is evolving. Relating consequences to
cost/benefit analysis provides a simpler basis for evaluating acceptable risk. The link
between risk and benefit must be balanced and within the context of geotechnical
engineering the risks are usually reduced to an acceptable standard by the best practical
means. This report is written within this context.
Swedish regulations state that probability based methods, such as the partial coefficient
method, should be used in geotechnical design. However, Stille et al. /2003/ suggest that
this approach is too restrictive and probability based methods should be used throughout
the whole design and construction process. In their report they use the following
definitions:
Hazard: a source of a potential damage event.
Damage event: an event that may cause damage to humans, properties or the environment.
Risk: the combined effect of the probability of a damage event and the consequences
caused by the damage event.
In the following section a risk based methodology is proposed that follows the general
principles suggested by Stille et al. /2003/ for assessing risk in rock engineering. Firstly
Risk Management
Risk Assessment Risk control
Risk Risk Decision-

Monitoring
analysis evaluation making
Hazard Risk Risk Risk

identification estimation acceptance estimation
Figure 2–1: A general framework for risk management proposed by the Canadian Standards Asso-
ciation Canadian Standard Association /1991/.
11
the hazards identified in Figure 1–1 are evaluated as potential damage events and the
severity of the hazard assessed by its magnitude, e.g., size of the wedge or depth of
spalling failure. The risk can only be assessed after these two steps are completed and is
beyond the scope of this report.
12
3 Risk based methodology
A general methodology that can be used to assess the wedge and spalling hazard is
illustrated in Figure 3–1. The methodology in this report is restricted to that noted above
the ’risk control options’ in Figure 3–1. For each hazard a criterion is required that can be
used to determine the frequency of occurrence. Once the frequency of the hazard has been
determined the next step is to assess the severity of the hazard. The approach for each
hazard is discussed in the following sections.
3.1 Wedge stability methodology

In this section the software program UnWedge was used for all the wedge related
calculations. UnWedge2 is a Windows based program that incorporates limit equilibrium
with in-situ stress to evaluate the stability of wedges around underground openings. Other
software programs may be equally suitable for this task.
3.1.1 Frequency of Wedges
When continuous fractures from three joint sets are intersected by an underground opening
a tetrahedral wedge can form. These three joints and the underground surface provide the
kinematic freedom for a wedge to slide or fall under the influence of gravity (see
Figure 3–2a). In Figure 3–2a the wedge is located in the centre of a 5-m diameter tunnel
and will simply fall under the weight of gravity and hence have the factor of safety of zero.
For this example, in-situ stress was ignored in the calculation; however, if the effect of
in-situ stress is included in the calculation the factor of safety for the roof wedge in
Figure 3–2a increases from zero to approximately 38. This improvement in the factor of
safety arises because the in-situ stresses increase the normal stress (confinement) acting on
the failure planes Martin et al. /2001/. Hence, in-situ stress can play a significant role in
the stability of underground wedges, and should be considered in the assessing the
frequency of occurrence, and the severity of the hazard, i.e., the size of the wedge.
3.1.2 Severity of Wedges
Once a wedge forms the next step was to evaluate the severity of the hazard, which is a
function of the size of the opening. In this case the severity was quantified by estimating
the size and weight of the wedge which are automatically calculated by UnWedge.
3.1.3 Wedge data uncertainty and probability
At this stage in the design the main objective is to assess if wedges are likely to be of
concern for detailed design or construction. For these analyses the shear strength of all
joints is assumed to have a constant friction angle with zero cohesion. This approach will
show the relationship of joint orientation and tunnel orientation on the formation of
wedges.
2 Available from WWW.RocScience.com
13
RISK ASSESSMENT FOR UNDERGROUND STABILITY
HAZARD IDENTIFICATION
* Fall-of-ground (Wedge)
* Spalling (Stress induced)
FREQUENCY ANALYSIS SEVERITY OF HAZARD

* Probability distribution * Size of Wedge to Support
- In-situ stress - Elevation
- DFN model - Tunnel orientation
- Strength
* Frequency of occurrence * Depth of Spalling
- % of tunnel length - Elevation
- vs Depth & tunnel orientation - Tunnel orientation
RISK CONTROL OPTIONS
* Depth of Repository * Tunnel orientation * Tunnel support

* Tunnel Size * Tunnel shape * Exposure time
* Excavation procedures * Organization of the tunnelling work
RISK EVALUATION
* Compare to levels of acceptable risk for underground construction
- Capacity
- Servicablity
Need for
Underground Risk
Yes No additional
Repository Acceptable
information
Design & analysis
Figure 3–1: Illustration of the steps involved in assessing the risk for wedge instability and spalling.
This report is focused on developing the methodology for assessing the severity of the hazard and
does not address “risk evaluation”
The major assumption is that the three joints required to form the wedge are present and
that the joint surfaces are sufficiently large to form the wedge. The probability that this
will occur resides with the development of the Discrete Fracture Network (DFN) model
used to establish the joint orientations. Hence, the orientation of the joints is held constant
over the depths investigated, unless the DFN has a depth dependency.
As indicated in Martin et al. /2001/ tunnel intersections can lead to wedge stability issues.
This special case has not been addressed in this report.
14
Table 3–1: The following input parameters were used for the wedge analy-
ses in this report. The input values for the fractures were obtained from the
Discrete Fracture Network Model (DFN).
Input parameter Values used in analysis

1. Number of joint sets, orientations and Based on DFN model. Inde-
trace length pendent of depth
2. Joint strength φj = constant, Cohesion=0
3. In-situ stress Mean magnitude and orienta-
tion of the principal stresses at
each depth of interest. Values
vary with depth.
4. Deposition Tunnel: 5-m-diameter Trend varies from 0 to 360 de-
grees in 30 degree intervals.
Plunge=0
5. Deposition Borehole: 1.8-m-diameter
Vertical Plunge = Vertical (90 de-
grees)
Horizontal Trend varies from 0 to 360 de-
grees in 30 degree intervals
(a) Deposition tunnel (b) Deposition borehole
Figure 3–2: Example of a wedge located in the roof of a deposition tunnel (left figure) and the wedges
formed in a vertical shaft (right figure) using the RocScience Software UnWedge. In the right figure
all the wedges are stable while in the left figure the wedge will fall from the roof if the effect of in-situ
stress is ignored.
15
3.2 Spalling assessment methodology
When the stresses on the boundary of an underground excavation reach the rock mass
strength failure occurs. In good quality hard rock, such as that found in the Scandinavian
Shield, the failure process is described as “spalling” and the associated rock mass strength
as the “spalling strength”. The word spalling is purposely used to indicate that the failure
process involves extensional splitting/cracking and is a different failure process from that
of shearing which is commonly observed in weak rocks. Appendix A.2 provides an
overview and the background to the rock mass spalling strength that is used in this report.
Spalling is generally defined as the formation of stress-induced slabs on the boundary of an
underground excavation (see Figure 3–3a). It initiates in the region of maximum tangential
stresses and results in a v-shaped notch that is local to the boundary of the opening (see
Figure 3–3b) . The slabs can vary in thickness from a few millimeters to a few centimeters,
for circular underground openings that range in diameter from 1 to 5 m. A detailed review
of the spalling process was provided by Martin et al. /2001/.
In order to determine if spalling will occur the rock mass spalling strength must be
established. Since 1990 two major experiments have been carried out to develop a
methodology for assessing the rock mass spalling strength for crystalline rocks: (1)
AECL’s Mine-by Experiment Martin and Read /1996/, and (2) Äspö Pillar Stability
Experiment (APSE) Andersson /2005/. The rock mass spalling strength used in this report,
i.e., the tangential stress required to initiate spalling on the boundary of an opening in
crystalline rock, is based on the findings from these experiments (see Appendix A.2 for a
brief summary of those experiments). Based on the findings from those experiments the
rock mass spalling strength (σsm ), given in Table 3–2, was used for the analyses in this
report.
In Table 3–2 the rock mass spalling strength (σsm ) is normalized to the mean laboratory
uniaxial compressive strength3 . This ratio provides a method for evaluating the rock mass
spalling strength for other sites when only the laboratory uniaxial compressive strength is
known. Note that the mean laboratory uniaxial compressive strength is used and not the
range in measured laboratory uniaxial compressive strength.
Because of the variability of the rock mass at some sites, the rock mass spalling strength in
Table 3–2 may need to be checked periodically. The onset of extension cracking (dilation)
in laboratory uniaxial samples can be used to indicate the boundary for the initiation of
damage in brittle rock (see Appendix A for a more complete discussion). Crack initiation
3 Themean uniaxial strength, rather then the full range of uniaxial strength, was found to provide better
agreement with field observations.
Table 3–2: The rock mass spalling strength used in this report.
σsm (MPa) σsm /UCSm

Coarse & medium grained crystalline rocks 120 ± 5 0.57 ± 0.02
UCSm= Mean laboratory uniaxial compressive strength
σsm = Rock mass spalling strength
± = Minimum, maximum
16
(a) Spalling slabs
(b) V-shaped notch
Figure 3–3: Examples of spalling observed around a 1.8-m-diameter borehole in the ÄPSÖ Pillar
Stability Experiment Andersson /2005/. Photos provided by C. Andersson.
17
in laboratory uniaxial tests represents a lower bound limit for the spalling rock mass
strength and may occur between 0.4 and 0.6 of the mean uniaxial compressive strength.
3.2.1 Spalling criterion
The potential for spalling is assessed using a traditional factor of safety approach that
compares Demand to Capacity. If the magnitude of the maximum tangential elastic
stresses (σθ θ ) on the boundary of an underground opening (Demand) reaches the rock
mass spalling strength (Capacity) the factor of safety can be expressed as:
σsm
Factor of Safety = (1)
σθ θ
The maximum tangential elastic stress on the boundary of circular opening in a
continuous, homogeneous, isotropic and linearly elastic rock can be expressed using the
well known Kirsch equations for plane strain:
σθ θ = 3σmax − σmin (2)
where σmax and σmin are the maximum and minimum far-field principal stresses,
respectively, in the plane of analysis.
In the criterion above, determining the maximum tangential stress is essential for
evaluating the potential for spalling. It should be noted that the maximum tangential stress
is a function of the tunnel profile shape and the orientation of the tunnel relative to the
in-situ stress tensor. Numerical analyses are required to determine the maximum tangential
stress on the boundary of non-circular tunnels, e.g., horse-shoe shaped tunnels.
3.2.2 Severity of Spalling
Depth of spalling
Once the stresses on the boundary of the excavation reach the rock mass spalling strength
(Factor of safety for spalling ≤ 1) and spalling initiates, the severity of the hazard must be
assessed, i.e., how deep will the spalling extend. The depth of spalling can be estimated
using the empirical correlations described in Martin et al. /2001/ and given in Figure 3–4.
These data were compiled from published case histories in a wide range of rock mass
conditions and in-situ stresses. The results from the Äspö Pillar Stability Experiment
(APSE) have been added to Figure 3–4 and Figure 3–4 shows that the empirical
correlations predicted the depth of spalling for the Äspö Pillar Stability Experiment
reasonably well.
In Figure 3–4 the depth of spalling is normalized to the tunnel radius and is measured from
the centre of the tunnel. The expression for the depth of spalling given in Figure 3–4 for
approximately circular openings can be rewritten as:

σθθ
Sd = a 0.5 − 0.52 for σθ θ > σsm (3)
σsm
where the Sd is measured from the boundary of the tunnel (see Figure 3–5).
Andersson /2005/ observed that the depth of spalling in the Äpsö Pillar Stability
experiment was influenced locally by existing fractures. However, as shown by Figure 3–4
it is not anticipated that existence of fractures will significantly affect the average depth of
spalling calculated using Equation 3.
18
-0.1
2 ûòò APSE +0.1
Df
1.8
Df =a
1.6
1.4
1.2 Df ûòò
= 0:48 æ 0:1 + 0:5 û
a sm
1.25 1.5 1.75 2 2.25 2.5 2.75 3

Maximum boundary stress ûòò (
Rock mass spalling strength ( ûsm
Figure 3–4: Empirical relationship used to establish the severity of the hazard, i.e., the depth of
spalling. Data from Martin et al. /2001/. Also shown are the new data from the Äpsö Pillar Stability
experiment (APSE) Andersson /2005/.
Volume of spalling
The volume of spalling in a deposition can be estimated using the data from the Äspö
Pillar Stability Experiment. In this experiment the slabs of rock were carefully removed to
expose the shape of the notch.. It is assumed that these slabs would also be removed at a
repository during the excavation process. Measurements of the extent of spalling after the
drilling and heating of the 1.8-m-diameter borehole showed that the spalling starts at some
depth below the tunnel floor and does not extend to the bottom of the borehole Andersson
/2005/ (see Figure 3–5). Similar findings were observed at AECL’s URL suggesting that
an Excavation Disturbed Zone will exist around all underground openings at the repository
depth that reduces the tangential stress concentrations near the boundary of the tunnel.
Based on the experience from AECL’s Mine by Experiment, this effect is more pronounced
in drill-and-blast tunnels than machine excavated tunnels. In addition, the spalling does
not extend to the bottom of the borehole because of the reduced tangential stresses at the
bottom of the borehole caused by the three dimensional stress effect. Using this data it
would appear that a reasonable approximation for the vertical extent of spalling in
deposition holes can be estimated by taking 75% of the total length of the deposition hole.
Figure 3–6 illustrates the spalling area observed on one side of a deposition hole and
sequence of calculations required to determine the area of spalling. The area of spalling as
illustrated in Figure 3–6 can be determined using the equations associated with the sector
and segment of a circle and the area of triangle.
Using the definitions in Figure 3–6, Area 1 can be determined by:
A1 = 1/2l(Sd + z)
= 1/2(R sin θ/2)(Sd + (R − R cos θ/2))
where Sd is the depth of spalling predicted by Equation 3 and θ is given in radians. Note
that the angle θ is given by 2 arcsin(l/2R).
19
Sd
A 1.8-m-diameter B
Plan View
~0.8 m
-1.41 m 0 +1.41 m Tunnel floor
A B
Extent of spalling
after drilling
1.0 m
2.0 m
Extent of spalling
after drilling and
heating
Note:
Spalling volume
3.0 m
based
on laser survey
was 99 litres.
4.0 m
5.0 m
6.0 m
Bottom of hole
6.7 m
Unrolled Perimeter Map
Figure 3–5: Extent of spalling observed in the Äspö Pillar Stability Experiment. Data provided by
C. Anderson (SKB).
20
Sd Area 1 (A1)
l z
-
Area 2 (A2)
Rf R Area 3 (As)
(Area of Spalling)
ò Area 1 - Area 2 = Area 3
Plan view of deposition hole
Figure 3–6: Illustration of the spalling area and labelling used in deriving Equation 4.
And, Area 2 is given by:
A2 = 1/2R 2 (θ − sin θ)
The area of spalling (As ) is then given by:
As = A1 − A2
= 1/2(R sin θ/2)(Sd + (R − R cos θ/2)) − 1/2R 2 (θ − sin θ)
Simplifying

1 θ
As = R 2(R + Sd ) sin − Rθ (4)
2 2
The lateral extent of spalling (l in Figure 3–6) based on the data from Äspö Pillar Stability
Experiment was approximately 0.8 m. This would give θ = 53 degree. Using R = 0.9m,
θ = 0.921, and Sd = 0.07m, the area of spalling for the APSE borehole given by
Equation 4 is approximately 0.016m2 . Combining the area of spalling, with the vertical
extent after drilling (1.5 m) and heating (75% of 6.7 m) the volume of spalling (Sv ) for the
APSE borehole would be approximately 0.023m3 after drilling and 0.08m3 after drilling
and heating. The volume of spalling for the APSE borehole after drilling and heating
measured using laser scanning was 0.089m3 . Hence if the stress magnitudes created for
the Äspö Pillar Stability Experiment existed at the repository level the total volume of
spalling after drilling would be 0.046m3 and 0.160m3 after drilling and heating assuming
the hole remained open. This total spalling volume of 0.160m3 is approximately equal to
the 0.150m3 specified in SKB /2004/ as being an acceptable volume increase per
deposition hole. Hence it is unlikely that volume increase due to spalling of the deposition
holes will exceed the allowable volume increase. Nonetheless the approach described
above can be used estimate the volume loss due to spalling once the probability of spalling
is encountered.
The angle θ of 53 degree in Figure 3–6 for the APSE 1.8-m-diameter borehole is similar to
the angles reported at AECL’s Underground Research Laboratory. For example, the
spalling that formed during the diamond drilling of a 1.25-m diameter, 5-m-deep, borehole
gave θ ≈ 45 degree, while θ of approximately 65 degree was measured for the 3.5-m
diameter Mine-by tunnel. Brudy et al. /1995/ showed that for the deep KTB borehole θ
remained relatively constant at about 30 to 55 degree from a depth of 3000 m to a depth of
21
5500 m. Barton et al. /1988/ reported similar findings from the Fenton geothermal site
where θ averaged 76 degree between a borehole depth of 2829 and 3627 m. Hence,
θ = 53 degree for the APSE 1.8-m-diameter borehole is in keeping with existing
experience and may be used to calculate the spalling volume for deposition hopes.
It should be noted that by using a constant θ = 53 degree for the volume calculation,
implies that the area of spalling (As ) in Equation 4 will be negative when the depth of
spalling is less than 0.03 m. The depth of spalling cannot not be negative. It is not known
if these equations can predict such small depths of spalling and hence these equations
should only be used for depths of spalling greater than 0.03 m.
3.2.3 EXCEL code for evaluating the spalling hazard
Figure 3–7 illustrates the calculation logic used to establish the spalling factor of safety,
the probability of spalling and the average depth of failure. It should be noted that the
mean uniaxial compressive strength is used to establish the rock mass spalling strength
(σsm ), and the factor k is used to establish the range of rock mass spalling strength.
An EXCEL spreadsheet was developed for Figure 3–7 using Equations 1 and 3, i.e,
spalling factor of safety and the depth of spalling, respectively (Figure 3–8). The
probability for spalling and the potential range in the depth of spalling was determined
using the software @Risk4 which consists of a series of macros for EXCEL that conducts
Monte Carlo simulations. In Figure 3–8 the function RiskTriang(min,mean,max) is used to
represent a triangular distribution for the input parameter. An example of the input
parameters, the calculation sequence used in EXCEL, and the output in graphical form are
provided in Figure 3–8. The input parameters are site specific and for this example the data
were for the deposition holes at a depth of 350 m.
The Factor of Safety in Figure 3–8 was calculated using the @Risk Monte Carlo
simulations and hence the output is expressed as a probability distribution, shown
graphically in Figure 3–8. For these simulations there is no correlation between stress and
strength, i.e., the highest stress can be associated with the lowest strength. In reality, this
may not be the case as the highest in-situ stresses are often found in the most competent
rock mass and the lowest stresses in highly fractured rock masses. This level of
information is currently not available, and for these analysis this issue is not addressed.
The output in Figure 3–8 for the spalling factor of safety provides the mean factor of safety
as well as the probability of the factor of safety being less than 1. If there is 10%
probability that the spalling factor of safety is less than 1, this implies that for a 100 m long
tunnel, there is the potential that 10 m of the tunnel length will experience spalling. If there
is 0% probability of spalling there is no depth of spalling. However, once the probability of
spalling is greater than zero, a depth of spalling is determined. For the depth of spalling,
only the mean depth of spalling was used for assessing the severity of the hazard.
4 Available from www.palisade.com
22
Step 1: Stress and strength distribution
mean
ûsm = k UCSmean = x UCSmean
k
min max
mean mean
û òò = 3û 1 à û 3 = 3 à
û1 û3
min max min max
Step 2: Spalling Factor of safety

1.
Mean=1.11
0.8
ûsm 0.6
FOS = =
û òò 0.4
0.2
0.7
0.7 0.9
0.9
1 1.1
1.1 1.3
1.3 1.5
1.5
14.2% 85.8%
1.5
Probability of Spalling
Step 3: Depth of spalling

à û á
Sd = a 0:5 ûsmòò
à 0 : 52
a = radius of the opening in metres
16
Mean=0.043 m
14
12
10
Sd =
8
6
4
2
0
0 0.045 0.09 0.135 0.18
90% 5%
.0036 .0987
Depth of Spalling (m)
Figure 3–7: Illustration of the calculation flow chart used to establish spalling factor of safety, the
probability of spalling and the average depth of failure.
23
A B C D E F
1 INPUT PARAMETERS
2 Radius = 0.9 m
3 Depth = 350 metre
4 Min Mean Max
5 UCS = Mean 225
6 Shmax = RiskTriang(D6, E6, F6) E6-E6*0.1 35+0.02*C3 E6+E6*0.1
7 Shmin = RiskTriang(D7, E7, F7) E7-E7*0.1 17.5+0.01*C3 E7+E7*0.1
8 σsm/UCS = RiskTriang(D8, E8, F8) 0.55 0.57 0.59
9
10 OUTPUT
11 Ratio= σsm/UCS
12 Factor of Safety (FOS) = (Ratio*Mean UCS)/(3*Shmax-Shmin)
13
14 Distribution for FOS/C7
1.6
15 1.000
16 Mean=1.198548
17 0.800
18
Probability
19
0.600
20
21
22 0.400
23
24 0.200
25
26 0.000
27 0.8 1 1.2 1.4 1.6
28 2.62% 97.38%
29 1 1.6
Factor of Safety
30
31 Depth of Spalling = RiskOutput() +IF(FOS<1,(rad(0.5*(3*Shmax-Shmin)/(Ratio*Mean UCS)-0.52),0)
32
33
34 Distribution for DOF/C31
35 25
36 Mean=0.0288482
37 20
38
Frequency
39 15
40
41 10
42
43 5
44
45 0
0 0.03 0.06 0.09 0.12
46
47 Depth of Failure (m)
48
Figure 3–8: Example of the spreadsheet code including the @Risk functions used to assess the Factor
of Safety for spalling and the associated depth of failure. This fictitious example is for deposition
holes at a depth of 350 m.
24
11 MPa
59 MPa
11 MPa Extensive 11 MPa
59 MPa Minor Spalling Spalling
44 MPa 59 MPa 45 MPa 45

SE NW 48
m
oo
Half barrel
fR
Half barrel
o
is
Ax
204
1m
252 27
1m
225o
Room 413 Room 405 1-m-diameter borehole

Mine-by Test tunnel
in floor of Room 405
VENT
RAISE
SHAFT
N
E
Room 409 W
S
5
40
RM
Mine-by Test Tunnel σ1 σ3
σ2
Room 413
σ1 σ2 σ3
Magnitudes (MPa) 60±3 45±4 11±4
Trend/Plunge (o) 145/11 054/08 290/77
Figure 3–9: Location of AECL’s Mine-by test tunnel, Room 413, Room 405 and 1-m-diameter
borehole drilled from the floor of Room 405.
3.2.4 Calibration of methodology
The 420 Level of AECL’s URL provided an opportunity to calibrate the methodology used
in this report. The extent of spalling, indicated by the size of the v-shaped notch, varied for
tunnels in different directions. Figure 3–9 shows the location of various tunnels on the 420
Level of AECL’s URL, the observed spalling and the orientation of the in-situ stress
magnitudes. The extent of spalling can be readily seen in Figure 3–9 by comparing the
excavation profiles for Room 413 and Room 405. Note that for Room 413, which is
oriented parallel to the maximum horizontal stress, only very minor spalling occurred as
the half-barrels associated with the blast holes were still visible for nearly all of the
perimeter profile. While Room 405 which was excavated perpendicular to the maximum
principal stress, showed extensive spalling, i.e., large v-shaped notch. In addition to the
AECL data, an additional calibration analysis was carried out for the 5-m-diameter TBM
tunnel used in SKB’s Prototype Repository (see Figure A.5 for location). No spalling was
observed during the excavation of this tunnel.
25
Note that Martin and Read /1996/ established the confidence for the in-situ stress tensor for
the URL 420 Level through large scale back-analyses and provided the likely range in
stress magnitudes. These ranges were used in conjunction with the approach given in
Figure 3–7 to determine the probability of spalling for the AECL’s Mine-by test tunnel,
Room 413 Level and AECL’s 1-m-diameter borehole drilled from the floor of Room 405
(Figure 3–10a). The same approach was also used for the Äspö Prototype Repository
Tunnel using the stress tensor determined from convergence back analysis by Andersson
and Martin /2003/. The computed mean factors of safety for spalling are in agreement
with the observations. For example, in AECL’s Room 413 the methodology indicated a
mean Factor of Safety of 0.99, which is in agreement with the very minor spalling
observed. However the tunnels oriented perpendicular to the maximum stress, which
always exhibited spalling, had a mean factor of safety of 0.72. In addition there was no
evidence of spalling in the Äspö Prototype Repository tunnel which is also consistent with
the computed mean factor of safety of 1.43. The range in the factors of safety computed in
Figure 3–10a have been summarized in Figure 3–10b.
A circular access tunnel for the Äspö Pillar Stability Experiment while not shown on
Figure 3–10b would have had a Factor of Safety of approximately 1.5, similar to that
computed for the Äspö Prototype Repository tunnel. By modifying the tunnel geometry to
increases the tangential stresses on the tunnel rook and floor the factor of safety was
reduced to 1.3. Though the factor of safety was significantly decreased no spalling was
observed in the 50-m-long tunnel. However, by drilling the 1.8-m-diameter borehole in the
floor of the tunnel the tangential stresses on the the boundary of the borehole locally
exceeded the spalling strength and the factor of safety was reduced to approximately 1. By
drilling the second 1.8-m-diameter borehole the tangential stresses were increased again
and spalling was observed over 25% of the borehole depth (see Figure 3–3b). The
computed factor of safety based on elastic stresses would have decreased to approximately
0.8 at the end of drilling the second hole. While the Äspö Pillar Stability Experiment was
not used in the development of Figure 3–10b the spalling observations and the computed
elastic stresses reported by Andersson /2005/ are consistent with the trend in Figure 3–10b.
Inspection of Figure 3–10b reveals that if the mean Factor of Safety is greater than
approximately 1.15, the probability of spalling is essentially zero. In these analyses the
in-situ stress tensor is known with confidence. During the early stages of a site
investigation program the in-situ stress tensor may not be known with the same degree of
confidence as the examples used to develop Figure 3–10b. Hence it is suggested that when
the stress tensor is not well constrained a spalling Factor of Safety of 1.25 be used for
preliminary design purposes. Based on Figure 3–10b a mean Factor of Safety of 1.25
should give a probability of spalling of approximately zero.
26
Probability of Spalling - Aspo Prototype Repository Probability of Spalling - Room 413
1.000 1.000
Mean=1.43 Mean=0.99
0.800 0.800
0.600 0.600
0.400 0.400
0.200 0.200
0.000 0.000
1.25 1.35 1.45 1.55 1.65 0.75 0.875 1 1.125 1.25
5% 5% 5% 5%
1.319 1.550 .853 1.090
Spalling FOS Spalling FOS
Probability of Spalling - URL 420 Level

Probability of Spalling - Mine-by
1-m-diameter Borehole
1.000 1.000
Mean=0.89 Mean=0.72
0.800 0.800
0.600 0.600
0.400 0.400
0.200 0.200
0.000 0.000
0.8 0.845 0.89 0.935 0.98 0.55 0.6375 0.725 0.8125 0.9
5% 5% 5% 5%
.641 90% .810
.848 .935
Spalling FOS Spalling FOS
(a) Probability of spalling
1.6
Mean Factor of Safety
Minimum Mean
1.4 Aspo Prototype
No
Spalling
1.2
Maximum
1 Very Minor Spalling

Room 413
Extensive 1-m-diameter borehole
0.8 Spalling
Mine-by
0.6
0.6 0.8 1 1.2 1.4 1.6

Spalling Factor of Safety
(b) Spalling factor of safety
Figure 3–10: Comparison of the probability of spalling and the factor of safety for the Äspö Prototype
Repository Tunnel, AECL’s Mine-by test tunnel, Room 413 and 1-m-diameter borehole drilled from
the floor of Room 405.
27
250
Peak Parameters
φ p = 0o
cp = 60 MPa
200
σt = 10 MPa
Shear stress (MPa)

Residual Parameters
Spalling φr = 45o
150 Failure envelope cr = 5 MPa
φr=45o
100
50
σsm σci
σt
50 100 150 200 250
Normal stress (MPa)
Figure 3–11: The spalling failure criterion developed by Martin et al. /1999/ and Hajiabdolmajid
et al. /2002/ for implementation in numerical continuum software such as Phase2 or Flac.
3.3 Spalling for noncircular openings

The methodology used to assess the spalling hazard given in the previous sections was for
a single circular opening in a continuous, homogeneous, isotropic, linearly elastic rock.
For non-circular openings or when the interaction of multiple openings must be evaluated,
a numerical analysis is required. Appendix A describes the development of the spalling
criteria that should be used for numerical analysis. This criteria uses a Mohr-Coulomb
failure envelope with the peak friction angle (φp ) set to zero, to indicate that the strength
of the rock is entirely cohesive and the peak cohesion (cp ) of the rock set to 1/2 the
spalling rock mass strength (cp = 60 MPa). Once spalling occurs and the slabs form, the
strength of the broken rock slabs is mainly frictional and hence the cr is set to a nominal
5 MPa reflecting the small strain nature of the problem, and φr is set to 45◦ . The peak
tensile strength (σt ) goes to zero once spalling initiates.
Figure 3–11 shows the spalling-failure envelope that could be used to predict the depth and
extent of spalling using the finite element program Phase25 . Hajiabdolmajid et al. /2002/
and Hajiabdolmajid et al. /2003/ show how the failure criterion in Figure 3–11 can be
implemented in Flac6 . Once the depth of spalling has been assessed using this approach,
the length and capacity of support system can be determined assuming that the entire
weight of the spalled region must be retained by the support.
3.4 Spalling and thermally-induced stresses

At a nuclear waste repository, thermal stresses will develop in the rock mass after the waste
is placed. These thermal stresses will cause an increase in the tangential stresses on the
5 www.RocScience.com
6 www.itascacg.com
28
Tangential stress [MPa]
100
120
140
160
180
200
60
80
Unrolled Perimeter Map
0
EDZ effect
Extent of spalling
after drilling
1.0 m 1
Possible
Nonelasitc
reponse due
to
EDZ effects
2.0 m 2
Extent of spalling
after drilling and
heating Elastic tangential stress
at the end of
drilling
Depth (m)
3.0 m 3 (Determined using
the complete three
dimensional geometry
of the APSE Experiment)
4.0 m 4
5.0 m 5
6.0 m 6
Bottom of hole
6.7 m
Depth Geometry Thermal Spalling

(m) stress stress (MPa) strength
(MPa) (E=55 GPa) (MPa)
1.95 125 0* 125

3.0 111 13 – 15 124 – 126
3.55 106 24 130
(4.1 101 21 122)
*Prior to heating
Figure 3–12: The rock mass spalling strength determined for the Äpsö Pillar Stability Experiment
for geometry-induced stresses and for combined geometry and thermally-induced stresses.
boundary of the underground openings. Hence, underground openings that are stable at the
end of the excavation phase may become unstable during the heating phase, if the
tangential stresses reach the rock mass spalling strength. One of the objectives of the Äpsö
Pillar Stability Experiment was to evaluate the rock mass spalling strength under the
conditions of thermal loading. Figure 3–12 shows the results from this experiment7 . The
results indicate that the average rock mass spalling strength is 125 MPa regardless of
whether the tangential stresses are induced by geometry or by combined geometry and
thermally induced stresses. Because these results are preliminary, the rock mass spalling
strength given in Table 3–2 was used in this report for the spalling analyses reported in the
following sections.
7 Results
presented by C. Andersson (SKB) during the Äpsö Pillar Stability Experiment Meeting June 20
2005 at Äpsö
29
4 Forsmark analysis
4.1 Input data

4.1.1 Joint orientation and strength
The Discrete Fracture Model (DFN) in R-04-15 gave the following major joint sets (see
Table 5-24 in R-04-15):
Table 4–1: Joint sets obtained from Discrete Fracture Model
Joint Label Dip Direction (◦ ) Dip (◦ )

1 North-East 313 89
2 North-West 216 87
3 North-South 265 87
4 East-West 187 87
5 Sub-Horizontal 330 11
4.1.2 In-situ stress
In-situ stress data for this report was based on the results contained in Sjöberg /2004/ and
Carlsson and Christiansson /1986/. Those results are graphically shown in Figure 4–1.
The stress gradients illustrated in Figure 4–1 were taken from SKB’s Site Description
Model 1.2 and used to determine the stress magnitudes at the 350, 450, 550 and 650 m
depths in Table 4–2. At present, there is uncertainty in the horizontal stress gradients
below a 500 m depth. Table 4–2 presents the mean stress gradient as well as the
uncertainty in the gradient based on all available data from the Forsmark site and the
author’s experience. For example, it is highly unlikely the vertical stress gradient will have
a large variation with depth while there is considerable uncertainty in the minimum
horizontal stress gradient with depth, expressed as ±0.005z MPa/m and ±20%,
respectively in Table 4–2. This uncertainty is also shown graphically in Figure 4–1.
The maximum horizontal stress was determined from the recent Borre probe results given
in Sjöberg /2004/. The higher horizontal stress magnitudes reported by Carlsson and
Christiansson /1986/ for borehole DBT-1 in Figure 4–1 have not been included in
developing the maximum horizontal stress gradient, because of the limitation in the data
collection used at that time, i.e., only before and after measurements were taken. Hence, it
is possible that the maximum horizontal stress gradient may be higher than that shown in
Table 4–2. This will have to be checked in a future stress measurement program at
Forsmark.
31
Stress (MPa)
0 10 20 30 40 50 60 70
0
Hydraulic Fracturing
KFM01A-Psi
KFM01A-Psi-Shmin
100 Overcoring
KFM01B Smax
KFM01B Smin
DB T1 Smax
DB T1 Smin
200
300
400
Depth (MPa)
Repository Depth
500
600
700
σhmin = 19 + 0.025z σHmax = 35 + 0.02z
800
900
1000 σv = 0.0265z
Figure 4–1: Stress gradients compiled from overcoring Sjöberg /2004/and Hydraulic fracturing
results.
32
Table 4–2: Forsmark stress gradients used for spalling analysis.
σHmax σhmin σvert

Gradient (MPa/m) 35 + 0.02z 19 + 0.025z 0.0265z
Variability ±10% ±20% ±0.0005z
Depth, z (m) (MPa) (MPa) (MPa)
350 42 28 9.3
450 44 30 11.9
550 46 32 14.6
650 48 35 17.2
4.1.3 Uniaxial Compressive Strength
The uniaxial compressive strength were taken from laboratory test results from SKB SDM
1.2 SKB /2005a/. The data are summarized in Appendix B and shown graphically in
Figures 4–2 and 4–3. Also shown on Figure 4–3 is the P-wave velocity. These velocities
were measured on the core and can be used as an indicator of stress-induced damage i.e.,
microcracking, Martin and Stimpson /1994/. Notice that there is essentially no change in
p-wave velocity above 450 m depth, suggesting that the stress magnitudes are not sufficient
to induce microcracking.
33
350
Uniaxial compressive strength (MPa)
300
250
200
150
Tonalite
100
50
0
0 100 200 300 400 500 600 700 800
Depth (m)
(a) UCS versus depth
Normal(225.33, 22.11)
X <= 189.0 X <= 261.7
5.0% 95.0%
3
2.5
2
Values x 10^-2
1.5
0.5
0
160 180 200 220 240 260 280 300
Uniaxial Compressive Strength (MPa)

(b) UCS histogram
Figure 4–2: Uniaxial compressive strength (UCS) versus depth from ground surface (top figure) and
normal distribution fit to all the unconfined compressive strength data (Appendix B) for Forsmark
rock unit. Tonalite has been excluded in the normal distribution.
34
Strength (MPa)
0 50 100 150 200 250 300
0
CI Mean UCS Mean
119 MPa 225 MPa
100
200
300
400
Depth (m)
500
600
700
800 KFM01A - UCS

KFM01A - CI
KFM02A - UCS
KFM02A - CI
P-Wave KFM03A - UCS
900 KFM03A - CI
KFM04A - UCS
KFM04A - CI
KFM01A PWave
1000
3000 3500 4000 4500 5000 5500 6000 6500
P-Wave (m/s)
Figure 4–3: Uniaxial compressive strength (UCS) and the onset of dilation (Crack Initiation - CI)
versus depth. Also shown is the P-Wave velocity for KFM01A. The decrease in P-wave velocity below
a depth of 500 m may be indicative of the development of stress-induced microcracks in the core.
35
Wedge Information
8 2
4 Lower Right wedge [1]
5 Factor of Safety: 144.939
2 Wedge Weight: 0.008 MN
8 1
7
4 5 Upper Right wedge [2]
Factor of Safety: 14.153
7 1 Wedge Weight: 0.022 MN
Roof wedge [4]

Top Perspective Factor of Safety: 51.959
4 2 Wedge Weight: 0.000 MN
2 48
8
Floor wedge [5]
Factor of Safety: stable
Wedge Weight: 0.000 MN
Lower Left wedge [7]

Factor of Safety: 344.558
Wedge Weight: 0.022 MN
1
7
Front Side 5
5 71
Figure 4–4: Illustration of the perimeter wedges formed when the Deposition Tunnel is aligned to
North (0 degrees).
4.2 Wedge stability

Wedge analyses were carried out using the Discrete Fracture Network Model for
Forsmark. A typical example is provided in Figure 4–4 and shows all the perimeter
wedges that can potentially form when the horizontal 5-m-diameter tunnel is aligned to
North (0 degrees). Figure 4–5 shows the maximum wedge weight as the tunnel is rotated
360 degrees in 10 degree increments. The spikes in these plots indicate that the worst
orientation for the tunnel would be when it is approximately parallel to the strike of the
sub-horizontal joint set, i.e. for a tunnel at 60 or 240 degrees trend. For this analysis only
the tunnel trend was varied, all other parameters were kept constant. Also shown in
Figure 4–5 is the minimum factor of safety when stress is not included in the analysis and
support is neglected. For comparison purposes analysis were also carried out including the
effect of in-situ stress (Figure 4–5). It is important to note that in-situ stress can have a
significant stabilising effect on underground wedges and therefore should be included
when conducting wedge analyses.
The results in Figure 4–5 indicate that all the wedges were stable or can be supported with
nominal rock support, which is in agreement with tunnelling experience in the good
quality rock of the Scandinavian shield. Hence wedge stability should not significantly
influence design decisions related to repository depth or tunnel orientation.
36
Optimization for Tunnel Axis Plunge = 0°
0.22
0.20
Maximum Wedge Weight
0.18
0.16
0.14
Approximate
0.12 capacity
typical rock bolt
0.10
0.08
0.06
0.04
0.02
0
90 180 270 360
6
In-situ stress not

Minimum Factor of Safety
5 included in analysis
No support
4
0
200
0 90 180 270 360
6
In-situ stress included

5 in analysis
Minimum Factor of Safety
No support
4
0
0
90 180 270
360
Possible Tunnel Axis Trend (deg)
Figure 4–5: Maximum weight of potential wedges as a function of tunnel trend and the minimum
factor of safety. Note that the large wedges are stable and that all the small wedges would be handled
by typical tunnel support.
37
1.4 100
90
80
Probability of Saplling (%)

1.3
FOS 70
60
1.2 50
40
30
1.1
20
10
%PofS
1 0
300 400 500 600 700
Depth (m)
Figure 4–6: Forsmark deposition holes - Mean Factor of Safety for spalling and the probability for
spalling for repository levels at 350, 450, 550 and 650 m. Note that below a depth of 500 m there is
less confidence in the in-situ stress magnitudes used in the analysis.
4.3 Spalling stability

The spalling analysis was carried out using the spreadsheet functions shown in Figure 3–8.
The detailed results from the analyses for both the deposition boreholes and the deposition
tunnels are presented in terms of the mean factor of safety (see Equation 1), the probability
in percent that the factor of safety is less than 1, and the associated mean depth of failure.
4.3.1 Vertical Deposition Holes
The mean factor of safety and probability for spalling for the 1.8-m-diameter vertical
deposition boreholes at repository depths of 350, 450, 550 and 650 m are given in
Figure 4–6. These results are based on 10,000 simulations using the @Risk software
described in Figure 3–8. A probability of failure of 0%, implies that none of the 10,000
simulations had a factor of safety less than 1. At a depth of 650 m the probability of
spalling is less than 1%. Hence these results suggest that spalling will not be encountered
along the deposition holes, regardless of the repository depth.
It should be noted that these analysis assumes a uniform stress distribution along the
deposition hole. In reality the deposition holes are 8-m long and connected to a Deposition
tunnel. Hence the stress magnitudes along the deposition hole will not uniform. It is
recommended that three dimensional stress analysis using the tunnel geometry and
deposition hole spacing should be carried out in a later design phase for those cases where
the probability of spalling is significant.
4.3.2 Deposition Tunnels
The factor of safety, probability for spalling and associated depth of failure in metres for
the 5-m-diameter circular deposition tunnels at repository levels of 350, 450, 550 and
650 m is given in Figure 4–7. The tunnels are evaluated parallel (0deg), perpendicular
38
1.8
Deposition Tunnels
1.7
1.6
0deg
1.5
1.4
45deg
1.3
1.2
90deg
1.1
1
200 300 400 500 600 700
Depth (m)
45
90 deg
40
% Probability of Spalling
35
30
25
20
15
10
5 0 deg & 45 deg
0
200 300 400 500 600 700
Depth (m)
0.007
90 deg
Mean Depth of Spalling (m)
0.006
0.005
0.004
0.003
0.002
0.001
0 deg & 45 deg
0
200 300 400 500 600 700
Depth (m)
Figure 4–7: Forsmark 5-m-diameter circular Deposition Tunnels - Factor of safety, Probability for
spalling and associated depth of failure for repository levels at 350, 450, 550 and 650 m. The notation
of 0 deg,45 deg and 90 deg, refers to the orientation of the deposition tunnel relative to the orientation
of the maximum horizontal stress, i.e., 0 deg implies the tunnel is parallel to the maximum horizontal
stress and 90 deg implies that the tunnel is perpendicular to the maximum horizontal stress.
(90deg) and at 45 deg to the orientation of the maximum horizontal stress. Note that only
the tunnels aligned perpendicular to the maximum horizontal stress has a significant risk
for spalling. This risk increases with depth and the probability for spalling is
approximately 40% at the 650-m depth.
39
30 m 30 m 30 m 30 m 30 m 30 m
Repository Vehicle Electrical Ventilation Elevator Workshop Reloading

Drainage Room Power Room Room Storage Station
Room Room Room
Figure 4–8: Geometry and spacing of the tunnels used in the Central Area.
Perpendicular to SHmax - Depth 650 m

20
0
17
Tensile cracking 48
Localized spalling
-20
35
Figure 4–9: Spalling around the common central area caverns at a repository depth of 650 m and
the caverns oriented perpendicular to the maximum horizontal stress.
4.3.3 Common central area caverns
A central area of the repository will be utilize a series of caverns with varying cross
sections. These caverns and there relative spacing is shown in Figure 4–8.
Because the stresses around these tunnels cannot be determined using the Kirsch
equations, spalling analyses were carried out using Phase2, the spalling criterion given in
Table 3–2 and the failure envelope given in Figure 3–11. The results from the Phase2
analysis are given in Figure 4–9, and indicates that minor localized spalling may occur at
the corners of the excavations where the stress magnitudes are elevated by the corner
geometry. However, this is not considered significant and is seldom a problem in practice.
As, with the deposition tunnels, this potential for spalling can be eliminated by orienting
the tunnels parallel to the maximum horizontal stress.
When excavations, such as the repository drainage vault, have high flat walls the potential
for tensile failure is elevated as shown in Figure 4–9. In practice, tensile failure is seldom
observed as a single fracture surface. It often occurs as a diffuse zone of cracking, seldom
requiring support. However, when joints occur in these flat sidewalls the potential for
wedge failure is elevated significantly as the clamping stresses in sidewalls are
significantly decreased. Replacement of these flat sidewalls with slightly curved sidewalls
would improve stability. Such issues should be addressed during the detailed design phase.
Due to the importance of the central area during operations, it is anticipated that normal
rock reinforcement (shotcrete and bolts) will be employed. It is anticipated that local
instabilities can be handled by such reinforcement.
40
5 Simpevarp analysis
5.1 Input data

5.1.1 In-situ Stress
Two in-situ stress domains have been determined for the Simpevarp area8 (Figure 5–1).
These domains have been established using the overcore data in Figure 5–2.
Stress Domain I
Stress Domain II
Figure 5–1: Location of stress domains for Simpevarp area, (taken from Simpevarp SDM 1.2 SKB
/2005b/).
Table 5–1 gives the stress gradients used for the analyses in this report based on the data in
Figure 5–2. At present there is considerable uncertainty in the gradient for the maximum
horizontal stress in Stress Domain I. The uncertainty in the horizontal stress components
was taken into account by assuming the mean stress given by the gradient ±25%. Hence a
range of stress magnitudes was used in the analyses.
Table 5–1: In-situ stress gradients used for Simpevarp spalling analysis.
Stress Domain I Stress Domain II

σHmax σhmin σvert σHmax σhmin σvert
Gradient (MPa/m) 3 + 0.06z 0.019z 0.0265z 0.0314z 0.011z 0.0265z
±25% ±25% ±0.002z ±15% ±15% ±0.002z
Depth,z (m) (MPa) (MPa) (MPa) (MPa) (MPa) (MPa)
350 24 6.3 9.1 11 3.9 9.1
450 30 8.1 11.7 14.1 5.0 11.7
550 36 9.9 14.3 17.3 6.1 14.3
650 42 11.7 16.9 20.4 7.2 16.9
8 Based on SKB - Simpevarp Site Description Model
41
Overcoring Data
Maximum Principal Stress [MPa]
0 5 10 15 20 25 30 35 40 45 50 55 60
0
KAS05
KA3579G
100 KK0045G01
KF0093A01
KA3376B01
KLX04
200 Model Domain I
Data from CLAB
KSH02
KAV04
300 KOV01
Model Domain II
Depth [m]
400 ASPO/APSE
500
600
700
800
σHmax = 0.0314z σHmax = 3 + 0.06z
Stress Domain II Stress Domain I
Overcoring Data Overcoring Data

Interm ediate Principal Stress [MPa] Minim um Principal Stress [MPa]
-5 0 5 10 15 20 25 -5 0 5 10 15 20 25
0 0
KAS05
KAS05
KA3579G KA3579G
KK0045G01 KK0045G01
100 100
KF0093A01 KF0093A01
KA3376B01 KA3376B01
KLX04 KLX04
200 200
Model Domain I Data from CLAB
Data from CLAB KSH02
KSH02 KAV04
300
Depth [m]
300
Depth [m]
KAV04 KOV01
KOV01 Model Domain I
Model Domain II Model Domain II
400 400
ASPO/APSE
500
500 ASPO/APSE
600
600
700
700
σvert = 0.0265z
800 Stress Domain I & II σhmin = 0.019z
800
σhmin = 0.011z Stress Domain I
Stress Domain II
Figure 5–2: Stress domains for Simpevarp area based on overcore test results, (data taken from
Simpevarp SDM 1.2 SKB /2005b/).
42
300
Uniaxial compressive strength (MPa 250
APSO/APSE
200
150
100
50
0
0 200 400 600 800
Depth (m)
Normal(182.995, 45.149)
1.2
1.0
0.8
Values x 10^-2
0.6
0.4
0.2
0.0 120 140 160 180 200 220 240 260 280
Uniaxial compressive strength (MPa)
Figure 5–3: Simpevarp: Uniaxial compressive strength (UCS) versus depth and a normal distribution
fit to all the data. The large scatter in the UCS at depth is thought to be related to the large number
of healed fractures in the samples tested.
5.1.2 Uniaxial Compressive Strength
The uniaxial compressive strengths were taken from laboratory test results (see
Appendix C) and summarized in POM 1.2. The data are shown graphically in Figure 5–3
and Figure 5–4. The data shows considerable scatter which is likely caused by the number
of inclined healed fractures that were present in the core. Note that in Figure 5–4 the
P-wave velocity decreases significantly below a depth of about 500 m. This decrease likely
reflects the increase in microcracks with depth. In this case the microcracks are probably
associated with the fracturing observed in the core, rather than increasing in in-situ stress
magnitudes, as is normally assumed.
The rock mass spalling strength used for these analyses was taken as 0.57 ± 0.02 × UCS.
Using a mean UCS = 183 MPa, the rock mass spalling strength ranges from 100 to
108 MPa.
43
Strength (MPa)
0 50 100 150 200 250 300
0
100 KSH01A - UCS

KSH01A - CI
KSH02A - UCS
KSH02A - CI
CI Mean
200 KSH01A PWave
300
400
APSE APSE
σsm UCS
Depth (m)
(120 MPa) (211 MPa)

500
CI Mean UCS
84 MPa Mean
600 183 MPa
700
800
900
1000
3000 3500 4000 4500 5000 5500 6000 6500
P-Wave (m/s)
Figure 5–4: Uniaxial compressive strength (UCS) and onset of dilation (Crack Initiation) versus
depth. Also shown is the P-Wave velocity for Borehole KSH01A, the mean uniaxial compressive
strength for Äpsö Diorite and the rock mass spalling strength obtained from the pillar stability
experiment.
44
5.2 Wedge stability
The results from Forsmark, indicated that the wedges encountered at the repository depth,
based on the discrete fracture network model, could be adequately supported by the typical
support system (shotcrete and bolts) used for tunnel construction in the Scandinavian
Shield. The Äspö Hard Rock Laboratory has excavations to a depth of 450 m. The
underground openings at Äspö HRL were supported with spot-bolts and shotcrete in
localized areas, i.e., systematic application of support was not required. This construction
experience in close proximity to the Simpevarp site suggests that wedge stability should
also not be a significant issue at the repository depth at Simpevarp. This practical
experience is considered more relevant at this stage of design then the results from wedge
analyses based on the DFN model for Simpevarp. In addition there is some uncertainty
with regard to the reliability of DFN model at the depth of the repository. This will be
resolved as new information from the Site Characterisation program becomes available.
5.3 Spalling instability

tunnels are given in following sections.
5.3.1 Vertical deposition holes
The results from the spalling analyses are summarized in Figures 5–5 and 5–6. These
results indicate:
1. Stress Domain II: Spalling in the deposition holes will not be encountered regardless
of repository depth.
2. Stress Domain I: The risk for spalling in the deposition holes increases below a
repository depth of 450 m. At 550 m depth the probability of spalling is
approximately 20% while at a repository depth of 650 m the probability for spalling
increases to approximately 70%.
It should be noted that if the rock mass is highly fractured the in-situ stress
magnitudes maybe lower than magnitudes given for Stress Domain I. Hence the risk
for spalling may be significantly reduced at depths below 500 m if the stress
magnitudes are lower (e.g., see the results for Stress Domain II)
Also shown on Figure 5–5 is the Factor of Safety for the deposition holes in the
Prototype Repository located at a depth of 500 m in the Äspö Hard Rock Laboratory.
The increase in the Factor of safety for the Prototype Repository occurs because of
the increase in the UCS (211 MPa) and an increase in the minimum horizontal stress.
It should be noted that no spalling was observed in the eight 1.8-m-diameter
8-m-deep deposition holes drilled for the Prototype Repository. These holes were
spaced at 6-m centre-to-centre which is the typical spacing for the SKB KBS3
concept.
45
2 1
Stress Domain I
FOS 0.9
Mean Factor of Safety Aspo
FOS Prototype %PofS 0.8
Repository
Probability of Saplling
0.7
1.5
0.6
0.5
0.4
1
0.3
0.2
0.1
0.5 0
300 400 500 600 700
Depth (m)
80 0.045
Stress Domain I
%PofS
70 0.040
Depth of Spalling(metres)
0.035
60
0.030
50
0.025
40
0.020
30
0.015
DofS
20
0.010
10 0.005
0 0.000
300 400 500 600 700
Depth (m)
Figure 5–5: Simpevarp: Spalling instability for vertical deposition holes for Stress Domain I.
4 1
Stress Domain II
0.9
3.5
0.8
Probability of Saplling
0.7
3
Factor of Safety
0.6
2.5 0.5
0.4
2
0.3
0.2
1.5
0.1
1 0
300 400 500 600 700
Depth (m)
Figure 5–6: Simpevarp: Spalling instability for vertical deposition holes for Stress Domain II.
46
5.3.2 Deposition tunnels
The spalling analyses for circular Simpevarp deposition tunnels, for both Stress Domains I
and II, are summarized in Figure 5–7. There is no probability of spalling in the deposition
tunnels in stress Domain II at any depth or orientation. However, in Stress Domain I, the
tunnels at 550 m and 650 m depth and oriented perpendicular to the maximum horizontal
stress will likely encounter minor spalling. In Figure 5–7, the deposition tunnel in Stress
Domain II orientated at 45 degrees to the maximum horizontal stress has the highest factor
of safety against spalling. While in Stress Domain I the tunnels oriented parallel to the
maximum horizontal stress have the highest factor of safety. Also shown in
Figure 5–7-Stress Domain I, for reference purposes, is the Factor of safety for the spalling
for the TBM tunnel at Äspö at a depth of 450 m. This tunnel is oriented approximately 30◦
from the maximum horizontal stress. No spalling was observed in the TBM tunnel, see
Figure 5–8.
The analyses indicate that the probability for spalling in deposition tunnels for Stress
Domain I at a depth of 650 m is greater than 50% when those tunnels are oriented
perpendicular to the maximum horizontal stress. At a depth of 550 m the probability of
spalling for these tunnels is reduce to approximately 10% (Figure 5–9). The relationship
between the probability of spalling and the depth of spalling is also shown on Figure 5–9.
At 550 m depth the average depth of spalling is 15 mm. This increases to 40 mm for a
repository at a depth of 650 m.
47
5
Stress Domain I
FOS
4 Aspo
Prototype FOS-Parallel
Repository
Factor of Safety
3
FOS-45
2 FOS-Perpendicular
1
Deposition tunnels
0
300 400 500 600 700
Depth (m)
6
Stress Domain II
5
FOS-Parallel
FOS-45
Factor of Safety
3
FOS-Perpendicular
1
Deposition tunnels
0
300 400 500 600 700
Depth (m)
Figure 5–7: Simpevarp: Spalling instability for deposition tunnels for Stress Domains I and II. Also
shown, for reference purposes, is the the TBM tunnel at Äspö at a depth of 450 m. This tunnel is
oriented approximately 30◦ from the maximum horizontal stress. No spalling was observed in the
TBM tunnel, see Figure 5–8.
48
Figure 5–8: Example of the 5-m-diameter TBM tunnel profile at Äspö at a depth of 450 m. No
spalling was observed in the TBM tunnel
2 60
Domain I
Perpendicular to SHmax
%PofS
50
Probability of Saplling (%)
FOS
Factor of Safety
1.5 40
30
1 20
10
0.5 0
300 400 500 600 700
Depth (m)
60 0.045
0.040
Depth of Spalling (metres)
50
0.035
40 0.030
0.025
30
DofS 0.020
20 0.015
%PofS 0.010
10
0.005
0 0.000
300 400 500 600 700
Depth (m)
Figure 5–9: Simpevarp: Spalling instability for deposition tunnels oriented perpendicular to the
maximum horizontal stress in Stress Domain I.
49
6 Laxemar analysis
6.1 Input data

Two in-situ stress domains have been determined for the Laxemar area9 . A summary of the
stress gradients given in the SDM 1.2 is given in Table 6–1.
The mean laboratory uniaxial compressive strength for the Granite to Quartz monzodiorite
(Ävrö granite) is 195 MPa.
Table 6–1: In-situ stress gradients used for Laxemar spalling analysis.
Stress Domain I Stress Domain II

σHmax σhmin σvert σHmax σhmin σvert
Gradient (MPa/m) 5 + 0.058z 3 + 0.014z 0.027z 0.032z 0.01z 0.027z
±25% ±25% ±0.002z ±25% ±25% ±0.002z
Depth,z (m) (MPa) (MPa) (MPa) (MPa) (MPa) (MPa)
350 25.3 7.9 9.5 11.2 3.5 9.5
450 31.1 9.3 12.2 14.4 4.5 12.2
550 36.9 10.7 14.9 17.6 5.5 14.9
650 42.7 12.1 17.6 20.8 6.5 17.6
6.2 Wedge stability

The results from Forsmark, indicated that the wedges encountered at the repository depth,
based on the discrete fracture network model, could be adequately supported by the typical
support system (shotcrete and bolts) used for tunnel construction in the Scandinavian
Shield. The Äspö Hard Rock Laboratory has excavations to a depth of 450 m. The
underground openings at Äspö HRL were supported with spot-bolts and shotcrete in
localized areas, i.e., systematic application of support was not required. This construction
experience in close proximity to the Laxemar site suggests that wedge stability should not
be a significant issue at the repository depth at Laxemar. This practical experience is
considered more relevant at this stage of design then the results from wedge analyses based
on the DFN model for Laxemar. In addition there is some uncertainty with regard to the
reliability of DFN model at the depth of the repository. This will be resolved as new
information from the Site Characterisation program becomes available.
6.3 Spalling instability

tunnels are given in following sections.
9 Based on SKB - Laxemar Site Description Model (SDM) 1.2
51
5
4
Stress Domain II
3
2
Stress Domain I
400 500 600 700

Depth (m)
Figure 6–1: Laxemar: Spalling instability for deposition holes in Stress Domains I and II.
2 Stress Domain I 1.
1.8
0.8
1.6 FOS
0.6
1.4
1.2 0.4
1
0.2
0.8 POS
0.6 0.
350 400 450 500 550 600 650
Depth (m)
Figure 6–2: Laxemar: Probability of spalling for deposition holes in Stress Domain I.
6.3.1 Vertical deposition holes
The results from the spalling analyses for the vertical deposition holes are summarized in
Figures 6–1 and 6–2. These results indicate:
1. Stress Domain II: Spalling in the deposition holes will not be encountered regardless
of repository depth.
2. Stress Domain I: The risk for spalling in the deposition holes increases below a
repository depth of 450 m. At 550 m depth the probability of spalling is
approximately 20% while at a repository depth of 650 m the probability for spalling
increases to approximately 60%.
6.3.2 Deposition tunnels
The spalling analyses for Laxemar circular deposition tunnels are summarized in
Figure 6–4 for both Stress Domains I and II. There is no probability of spalling in the
deposition tunnels in stress Domain II at any depth or orientation. However, in Stress
Domain I, the deposition tunnels oriented perpendicular to the maximum horizontal stress
52
1 Stress Domain I
0.2
Mean depth of spalling (m)

0.8
0.15
0.6
POS 0.1
0.4
DOS 0.05
0.2
0 0.
350 400 450 500 550 600 650
Depth (m)
Figure 6–3: Laxemar: Probability of spalling for deposition holes in Stress Domain I and the
corresponding depth of spalling.
6
Stress Domain I
5
3
0deg
2
45deg
1
90deg
400 500 600 700

Depth (m)
6
Stress Domain II
5
4
45deg
3
90deg
0deg
2
400 500 600 700
Figure 6–4: Laxemar: Spalling instability for deposition tunnels in Stress Domains I and II.
will likely encounter minor spalling below 500 m depth (Figure 6–5). The anticipated
depth of spalling is given in Figure 6–6.
53
2 Stress Domain I 1.
1.8
Probability of spalling
0.8
1.6 FOS
0.6
1.4
1.2 0.4
1
0.2
0.8 POS
0.6 0.
350 400 450 500 550 600 650
Depth (m)
Figure 6–5: Laxemar: Probability of spalling for deposition tunnels oriented perpendicular to the
maximum horizontal stress in Stress Domain I.
1 Stress Domain I 0.2

Mean depth of spalling (m)
0.8
0.15
0.6
0.1
0.4 DOS
0.05
0.2
0 POS 0.
350 400 450 500 550 600 650
Depth (m)
Figure 6–6: Laxemar: Probability of spalling for deposition tunnels oriented perpendicular to the
maximum horizontal stress in Stress Domain I and the corresponding mean depth of spalling.
54
7 Conclusions
Underground excavations in a fractured sparsely rock mass may develop stability issues
related to gravity induced falls-of-ground (wedges). The discrete fracture network for
Forsmark has identified several joint sets. These joint sets were used in the underground
wedge stability analysis program UnWedge. The results from these stability analysis
indicated that any potentially unstable wedges would be adequately handled by standard
rock bolt support systems. Hence, based on the current DFN model, there is no reason to
base the depth of a repository or the orientation of the repository tunnels or caverns on
wedge stability analyses. This notion is also supported by the construction experience
from the Äspö Hard Rock Laboratory to depths of 450 m. This conclusion may need to be
re-evaluated as the DFN model is refined and new information is added during the course
of site investigation.
The results from SKB’s in-situ Äspö Pillar Stability Experiment in a fractured crystalline
rock mass supports the findings from AECL’s Mine-by Experiment in a massive granitic
rock mass and indicates that the rock mass spalling strength for crystalline rocks can be
estimated as 0.57 ± 0.02 of the mean laboratory uniaxial compressive strength. This value
is only slightly above the onset of crack initiation (dilation) measured in laboratory
samples. A probability-based methodology utilizing this in-situ rock mass spalling
strength has been developed for assessing the potential for spalling for Forsmark,
Simpevarp and Laxemar sites. This methodology was calibrated against the underground
experience at Äspö and AECL’s Underground Research Laboratory. The in-situ stresses
and the laboratory uniaxial compressive strength data from the Forsmark, Simpevarp and
Laxemar sites were used as the bases for the analyses.
It should be noted that the confidence in the results from these analyses is directly related
to the confidence in the input data. The greatest uncertainty in the input parameters is
related to the in-situ stress magnitudes and the variation of these magnitudes with depth.
This uncertainty will be reduced as the Site Description Models are improved.
Findings from the analyses to evaluate the potential for spalling indicates the following:
Forsmark site
1. At a repository depth of 650 m the probability for spalling in the vertical
deposition holes increase to 10%.
2. The risk for spalling in the deposition tunnels oriented perpendicular to the
maximum horizontal stress increases significantly below a repository depth of
450 m. This risk is eliminated at all repository depths if the deposition tunnels
are oriented parallel to the maximum horizontal stress.
3. A central area of the repository will utilize a series of caverns/rooms with
varying cross sections. These caverns should be oriented parallel to the
maximum horizontal stress to minimize the risk for spalling
Simpevarp Site
1. Two stress domains have been identified for the Simpevarp site. In stress
55
domain II, there is no potential for spalling in either the vertical deposition
boreholes or the deposition tunnels, regardless of tunnel orientation relative to
the maximum horizontal stress.
2. Stress Domain I: The risk for spalling in the vertical deposition holes increases
below a repository depth of 450 m. At 550 m depth the probability of spalling
is approximately 20% while at a repository depth of 650 m the probability for
spalling increases to approximately 70%.
3. Stress Domain I: The analyses indicate that the probability for spalling in
deposition tunnels at a depth of 650 m is greater than 50% when those tunnels
are oriented perpendicular to the maximum horizontal stress. At a depth of
550 m the probability of spalling for these tunnels is reduce to approximately
10%. The risk for spalling is eliminated if the deposition tunnels are oriented
either parallel or at up to 45 degrees to the maximum horizontal stress,
regardless of the repository depth.
Laxemar Site
1. Two stress domains have been identified for the Laxemar site. In stress domain
II, there is no potential for spalling in either the vertical deposition boreholes or
the deposition tunnels, regardless of tunnel orientation relative to the maximum
horizontal stress.
2. Stress Domain I: The risk for spalling in the vertical deposition holes increases
below a repository depth of 500 m. At 550 m depth the probability of spalling
is approximately 20% while at a repository depth of 650 m the probability for
spalling increases to approximately 60%.
3. Stress Domain I: The analyses indicate that the probability for spalling in
deposition tunnels at a depth of 650 m is greater than 50% when those tunnels
are oriented perpendicular to the maximum horizontal stress. At a depth of
550 m the probability of spalling for these tunnels is reduce to approximately 5
%. The risk for spalling is eliminated if the deposition tunnels are oriented
either parallel or at up to 45 degrees to the maximum horizontal stress,
regardless of the repository depth.
General Oskarshamn Region
The in-situ stress magnitudes measured at the Äspö Hard Rock Laboratory (HRL)
appear to be similar in magnitudes to the in-situ stress magnitudes measured in
Stress Domain I at Simpevarp and Laxemar. The rock mass quality and strength also
appear to be similar at all sites in the Oskarshamn region. Tunnels and caverns of
various profiles were excavated as part of the construction of the Äspö HRL. These
openings have remained stable since being excavated in the mid-1990s. Spalling has
not been observed in any of the Äspö HRL excavations except the Pillar Stability
Experiment, where it was purposely created. There are no reasons to suspect that
underground excavations in the Simpevarp and Laxemar area would perform any
differently from the performance of the excavations at the Äspö HRL down to a
depth of at least 450 m. An evaluation of the Äspö Prototype Repository (depth 450
m) deposition holes gave a factor of safety against spalling of approximately 1.5.
56
References
Andersson C, 2005. Äspö pillar stability experiment. In Bergmeckanikdag, Stockholm

(Ed. Franzén T), pp. 69–78, SveBeFo, Stockholm.
Andersson, C, Martin, CD, 2003. Stress variability and the design of the Äspö Pillar
Stability Experiment. In Proc. 3rd Int. Symp. on Rock Stress, Kumamoto, Eds. Sugawara,
K, Obara, Y, Sato, A,pp. 321–326, A.A. Balkema, Rotterdam.
Andersson C, Martin CD, Christiansson R, 2004. SKB s Äspö Pillar Stability

Experiment, Sweden. In CD-ROM Proc. 6th North American Rock Mechanics
Symposium - NARMS04, Houston (Eds. Yale DP, Willson SM, Abou-Sayed AS), pp.
1–8, NARMS 04-503.
Barton CA, Zoback MD, L BK, 1988. In situ stress orientation and magnitude at the
Fenton geothermal site, New Mexico, determined from wellbore breakouts. Geophysical
Research Letters, 15(5):467–470.
Brace WF, Paulding B, Scholz C, 1966. Dilatancy in the fracture of crystalline rocks.
Journal Geophysical Research, 71:3939–3953.
Brudy M, Zoback MD, Fuchs K, Rummel F, Baumgärtner, 1995. Estimation of the

comlpete stress tensor to 8 km depth in the KTB scientific drill holes: Implications for
crustal strength. Journal Geophysical Research, 102(B8):18453 – 18475.
Canadian Standard Association, 1991. Risk analysis requirements and guidelines. Tech.
Rep. CAN/CSA-Q634-91, Quality Management-A National Standard of Canada,
Rexdale (Toronto), Canada, 42p.
Carlsson A, Christiansson R, 1986. Rock stress and geological structures in the

Forsmark area. In Proc. Int. Symp. on Rock Stressand Rock Stress Measurements,
Stockholm (Ed. Stephansson O), Centek Publishers, Lulea.
Hajiabdolmajid V, Kaiser PK, Martin CD, 2002. Modelling brittle rock failure.
International Journal Rock Mechanics And Mining Science, 39(6):731–742.
Hajiabdolmajid V, Kaiser PK, Martin CD, 2003. Mobilised strength components in

brittle failure of rock. Géotechnique, 53(3):327–336.
Hallbauer DK, Wagner H, Cook NGW, 1973. Some observations concerning the
microscopic and mechanical behaviour of quartzite specimens in stiff, triaxial
compression tests. International Journal Rock Mechanics Mining Science &
Geomechanics Abstracts, 10:713–726.
Lindfors U, Perman F, Sjöberg, 2005. Forsmark site investigation: Evaluation of the

overcoring results from borehole KFM01B. Project P-05-66, Swedish Nuclear Fuel and
Waste Management Company, Stockholm.
57
Martin CD, 1997. Seventeenth Canadian Geotechnical Colloquium: The effect of
cohesion loss and stress path on brittle rock strength. Canadian Geotechnical Journal,
34(5):698–725.
Martin CD, Chandler NA, 1994. The progressive fracture of Lac du Bonnet granite.
International Journal Rock Mechanics Mining Science & Geomechanics Abstracts,
31(6):643–659.
Martin CD, Christainsson R, Söderhäll J, 2001. Rock stability considerations for siting
and constrcuting a KBS3 repository: Based on experiences from Äspö HRL, AECL’s
URL, tunnelling and mining. Technical Report TR-01-38, Swedish Nuclear Fuel and
Waste Management Company, Stockholm.
Martin CD, Christiansson R, 1991. Overcoring in highly stressed granite – The influence
of microcracking. International Journal Rock Mechanics Mining Science &
Geomechanics Abstracts, 28(1):53–70.
Martin CD, Kaiser PK, McCreath DR, 1999. Hoek-Brown parameters for predicting the
depth of brittle failure around tunnels. Canadian Geotechnical Journal, 36(1):136–151.
Martin CD, Read RS, 1996. AECL’s Mine-by Experiment: A test tunnel in brittle rock.
In Proc. 2nd North American Rock Mechanics Symposium, Montreal (Eds. Aubertin M,
Hassani F, Mitri H), vol. 1, pp. 13–24, A.A. Balkema, Rotterdam.
Martin CD, Read RS, Martino JB, 1997. Observations of brittle failure around a circular
test tunnel. International Journal Rock Mechanics And Mining Science,
34(7):1065–1073.
Martin CD, Stimpson B, 1994. The effect of sample disturbance on laboratory properties
of Lac du Bonnet granite. Canadian Geotechnical Journal, 31(5):692–702.
Morgenstern N, 1995. The 3rd Casagrande Lecture: Managing risk in geotechnical

engineering. In Proc. 10th Pan-American Conference on Soil Mechanics and Foundation
Engineering, Guadalajara, Mexico, vol. 4, p. 102 126, Mexican Society of Soil
Mechanics, Mexico City.
Peng SS, Johnson AM, 1972. Crack growth and faulting in cylindrical specimens of
Chelmsford granite. International Journal Rock Mechanics Mining Science &
Geomechanics Abstracts, 9:37–86.
Potyondy DO, Cundall PA, 2004. A bonded-particle model for rock. International
Journal Rock Mechanics And Mining Science, 41(8):1329–1364.
Read RS, 1994. Interpreting excavation-induced displacements around a tunnel in highly

stressed granite. Ph.D. thesis, Department of Civil & Geological Engineering,
University of Manitoba, Winnipeg, Manitoba, Canada.
Read RS, Martino JB, Dzik EJ, Chandler NA, 1997. Excavation stability study –
Analysis and interpretation of results. OPG 06819-REP-01200-0028-R00, Ontarion
Hydro, Nuclear Waste Management Division, Toronto.
58
Scholz CH, 1968. Microfracturing and the inelastic deformation of rock in compression.
Journal Geophysical Research, 73(4):1417–1432.
Sjöberg J, 2004. Forsmark site investigation: Overcoring rock stress measurements in

borehole KFM01B. Tech. Rep. SKB P-04-83, Swedish Nuclear Fuel and Waste
Management Company, Stockholm, Sweden.
SKB, 2004. Deep repository. Underground design premises. Edition d1/1. R-report
R-04-60, Svensk kärnbränslehantering AB, Stockholm.
SKB, 2005a. Forsmark Site Description Model v1.2. R-report R-05-18, Svensk
kärnbränslehantering AB, Stockholm.
SKB, 2005b. Simpevarp Site Description Model v1.2. R-report R-05-08, Svensk
kärnbränslehantering AB, Stockholm.
Stacey TR, 1981. A simple extension strain criterion for fracture of brittle rock.
International Journal Rock Mechanics Mining Science & Geomechanics Abstracts,
18:469–474.
Stille H, Andersson J, Olsson L, 2003. Information based design in rock engineering.

SveBeFo Rapport 61, SveBeFo, Swedish Rock Engineering Research, Stockholm,
Sweden, 147p.
59
Appendices
61
A Rock mass spalling strength
A.1 Background
Failure of underground openings in hard rocks is a function of the in-situ stress magnitudes
and the characteristics of the rock mass, i.e., the intact rock strength and the fracture
network. At low in-situ stress magnitudes, the failure process is controlled by the
continuity and distribution of the natural fractures in the rock mass. However as in-situ
stress magnitudes increase, the failure process is dominated by new stress-induced
fractures growing parallel to the excavation boundary. This fracturing is generally referred
to as spalling failure or extension fracturing. Initially, at intermediate depths, these
stress-induced spalling regions are localized near the tunnel perimeter and tend to form
v-shaped notches (Figure A.1).
Martin et al. /1997/ described the brittle-failure process observed around AECL’s Mine-by
test tunnel. The test tunnel was extensively instrumented and Martin /1997/ showed that
the onset of spalling was related to the onset of cracking in the rock mass as recorded by a
microseismic monitoring array. Their results clearly showed that the onset of cracking
in-situ commenced well before the peak strength recorded in uniaxial compressive tests.
Several researchers, e.g. /Brace et al. 1966; Scholz 1968; Peng and Johnson 1972;
Hallbauer et al. 1973; Martin and Chandler 1994/, have shown that the onset of dilation in
brittle rocks is represented by the initiation of extension fracturing (crack initiation) in
uniaxial laboratory compression tests and that these extension fractures also initiate well
before the peak strength is reached as illustrated in Figure A.2. Laboratory tests by these
researchers have shown that the onset of dilation (Crack Initiation) typically occurs
between 0.4 and 0.6 of the peak strength for a wide variety of rock types and concrete.
Stacey /1981/ using elastic analyses introduced the concept of extensional strain to
estimate the extent of spalling around South African underground mines. Stacey /1981/
determined the extensional strain from laboratory compression tests using plots of lateral
Failed Region
1.5 m Wide 4 m Wide 2 m Wide
10 m Diameter 3.5 m Diameter 3.5 m Wide
Figure A.1: Extent and shape of stress-induced spalling failure observed around various shaped
underground openings, after Martin et al. /1999/.
63
Axial
Stress (MPa)
Peak
200
Long-term strength
Onset of Post-Peak
Crack
On set of dilation Initiation
100
caused by extension 0.4-0.6Peak
fracturing
-0.2 -0.16 -0.12 -0.08 -0.04 0 0.1 0.2 0.3 0.4

Lateral Strain % 0.2 Axial Strain %
Contraction
σ1 0.1
Total
Measured
ΔV/V
ΔV/V %
Axial 0
Strain Gauge Calculated
Crack Volumetric
Dilation
Crack Strain
Lateral
Strain Gauge Closure Crack
-0.1 Growth
0 0.1 0.2 0.3 0.4

Axial Strain %
Figure A.2: Stress-strain curves for Lac du Bonnet granite, data from Martin and Chandler /1994/.
Note that the onset of dilation occurs at approximately 50% of the peak uniaxial strength.
strain versus axial strain. Stacey /1981/ defined the the critical extension strain as the
lateral strain at the onset of nonlinearity in the lateral strain versus axial strain plot. It can
be seen in Figure A.2 that if one makes such a plot the critical extensional strain defined by
Stacey /1981/ is identical to the onset of dilation. Martin /1997/ used the same concept,
expressed in terms of crack initiation10 to evaluate the depth of spalling around the
Mine-by Test tunnel. Martin et al. /1999/ applied this logic to develop the depth of
spalling-failure given by:
Df σθ θ
= 0.48 + 0.5 (A.1)
a σsm
The depth of spalling given by Equation A.1 is an empirical criterion restricted to

approximately circular excavations. For non-circular excavations or when more detailed
information is required the depth of spalling can be determined using a numerical analysis
and a spalling failure criterion. Because the spalling process occurs via extension
fracturing, the cohesive strength component dominates the peak-failure envelope while the
frictional component can be ignored. However, once spalling has occurred the cohesive
strength component of the residual-failure envelope is significantly reduced while the
frictional component is fully mobilized. Figure A.3 shows the spalling-failure envelope
used to predict the depth of spalling for the Mine-by test tunnel compared to the measured
depth of spalling. Once the depth of spalling has been assessed using this approach, the
support can be determined assuming that the entire weight of the spalled region must be
retained by the support system. Alternatively, if no support will be used this method can be
10 At
the time of the development of Equation 3, Crack initiation was considered a lower bound estimate
and approximately equal to the rock mass spalling strength
64
250
Peak Parameters
Yielding Measured
φp = 0o
in Phase2 spalling limit
cp = 60 MPa
200
σt = 10 MPa
Shear stress (MPa)

Residual Parameters
φr = 45o
150 Failure envelope cr = 5 MPa
φr=45o
100
50
σsm σci
σt
50 100 150 200 250
Normal stress (MPa)
Figure A.3: Predicted brittle failure around the Mine-by test tunnel compared with the observed
failure. The predicted failure used Phase2 and a spalling failure criterion.
used to estimate the depth of the v-shaped notch that will develop.
Since the mid-1990’s with the development of the discrete element numerical codes (such
as Particle Flow Code) researchers can now numerically replicate the stress-strain curve in
Figure A.2 numerically Potyondy and Cundall /2004/. This ability has provided additional
insight into the behaviour of brittle rocks in compression and supports the spalling-failure
envelope discussed above. In particular, it has shown that the onset of dilation (crack
initiation) is controlled more by the geometry of the grains than the material properties.
This helps to explain why the onset of dilation in rocks where the grain size is small and
relatively uniform, typically occurs near the peak strength rather than at the 0.4 to 0.6
observed in medium to coarse grained crystalline rocks.
A.2 Rock mass spalling strength from in-situ experiments

Two in-situ have experiments have been carried out to investigate brittle failure, i.e.,
spalling: AECL’s Mine by Experiment and SKB’s ASPO Pillar Stability Experiment.
A.2.1 AECL’s Mine-by Experiment
The first phase of AECL’s Mine-by Experiment, the excavation of a 3.5-m-diameter tunnel,
was carried out at the 420 Level of AECL’s Underground Research Laboratory from 1990
to 1995 to investigate progressive failure around a circular opening in brittle, unfractured
Lac du Bonnet granite (Figure A.4). The tunnel was excavated without explosives, and
state-of-the-art instrumentation comprising both geomechanical and geophysical
instruments was used to monitor the failure process. The in-situ experiment was
supplemented by extensive laboratory tests. The major conclusion from the Mine-by
Experiment was that spalling initiated when the maximum tangential stress on the
boundary of the tunnel reached 120 MPa. The mean uniaxial compressive strength of Lac
du Bonnet granite was given as 212 MPa, therefore the onset of spalling could be
65
σ1
VENT
RAISE
SHAFT
N
E
W
S
σ1
5
40
RM
Mine-by Test Tunnel
Mine-by Test Tunnel
Figure A.4: Layout of the Mine-by Experiment and a photo of the excavated test tunnel.
expressed as 120/212 = 0.56UCSm.

This 120 MPa value was also confirmed by Read et al. /1997/ by excavating tunnels with
various shapes and various orientations relative to the in-situ stress state at the 450 Level.
Their work concluded there was little variation in the tangential stress required to initiate
spalling in the medium-grained granite suggesting that the local variation in the granite
rock mass strength and in-situ stress magnitudes was minor. However, Read et al. /1997/
noted that the tangential stress required to initiate spalling in the fine grained granodiorite
was approximately 150 MPa, despite the granoditoite only having a modest increase in the
uniaxial compressive strength. Only a few observations were made at the URL, but it was
clear that the grain size of the rock had a significant effect on the stress magnitude required
to initiate spalling.
A.2.2 Äspö Pillar Stability Experiment
The Äspö Pillar Stability Experiment (APSE) is in progress and only preliminary results
are reported here. The APSE was excavated at the 420-m depth of Äspö Hard Rock
Laboratory and examined spalling in a fractured rock mass and thermo-mechanical
coupling (Figure A.5). The excavation of two 1.8-m-diameter boreholes, separated by
1-m-thick pillar, was purposely over-stressed, using excavation- and thermally-induced
stresses, to induce spalling in the pillar Andersson et al. /2004/ (Figure A.6). The stresses
in the pillar were controlled and the onset of spalling monitored using acoustic emission
instrumentation and deformations. The major difference between the APSE and the
Mine-by Experiment is that APSE was located in fractured rock but the intact uniaxial
strength of Aspo Diorite is similar to that of Lac Du Bonnet Granite, i.e., 210 and
212 MPa, respectively.
Table A.1 compares the rock mass spalling strength determined from APSE and the
66
Prototype
Repository
σ1
Aspo Pillar
Stability Experiment
TBM Tunnel
Figure A.5: Location of the Äspö Pillar Stability Experiment and the Prototype Repository.
Pillar separting
two 1.8-m-diameter
deposition holes
v-shaped
notch
(Spalled region)
Figure A.6: Photograph of the Äspö Pillar Stability Experiment, showing the extent of the notch in
the 1-m-thick pillar.
Mine-by Experiment. This data was then used with the methodology described in
Figure 3–7 to evaluate the approach.
67
Table A.1: Summary of the on set of dilation (crack initiation) measured in
laboratory tests and the rock mass spalling strength from in-situ tests. The
mean peak uniaxial compressive strength (UCSm) is used as the reference
strength.
Laboratory Mean Values In-situ Spalling Strength

Underground Facility CI CI/UCS UCSm σsm σsm /UCSm
(MPa) (MPa) (MPa)
Äspö Pillar Stability Experiment
Aspo Diorite 94 0.44 211 119 ± 6 0.56 ± 0.03
Mine-by Experiment
Lac du Bonnet granite 97 0.45 212 120 0.56
Granodiorite 120 0.52 228 150 0.65
68
B Forsmark Uniaxial Compressive Strength
Table B.1: Summary of the test results from uniaxial uncon• ned compression
tests for Forsmark. The peak strength is given as the uniaxial compressive
strength (UCS).
Borehole ACTUAL Crack CI Long- UCS Young’s Poisson’
Initation UCS term modulus s ratio
Depth DEPTH Density (CI) Strength
3
Specimen ID (m) (m) (kg/m ) MPA (MPa) (MPa) (GPa)
KFM04A-113-1 162.02 140.41 2670 130 0.52 233 249.1 76.56 0.24
KFM04A-113-3 162.62 140.93 2690 130 0.54 208 241.1 80.68 0.25
KFM04A-113-5 163.51 141.70 2680 112 0.50 200 221.8 78.22 0.19
KFM04A-113-6 163.65 141.82 2700 116 0.50 190 230.1 73.44 0.23
KFM01A-113-1 225.92 224.96 2660 142 0.53 244 270.1 81.90 0.24
KFM01A-113-2 226.09 225.13 2660 187 0.69 243 269.7 78.15 0.21
KFM01A-113-3 227.23 226.27 2650 182 0.83 220 220.4 79.09 0.25
KFM01A-113-4 228.00 227.03 2660 100 0.45 210 224.1 74.05 0.22
KFM01A-113-5 229.60 228.63 2650 128 0.56 215 227.0 79.79 0.28
KFM02A-113-1 333.51 332.42 2660 120 0.54 195 221.3 78.56 0.24
KFM02A-113-2 333.66 332.57 2660 115 0.52 200 221.5 76.06 0.22
KFM02A-113-3 334.11 333.02 2660 - - 195 226.9 74.13 0.16
KFM02A-113-4 334.52 333.43 2660 115 0.53 192 218.1 73.53 0.26
KFM02A-113-5 334.78 333.69 2660 112 0.54 185 209.2 75.28 0.21
F01A-113-13 495.61 493.51 2662 130 0.54 239.2 74.20 0.29
KFM01A-113-8 496.01 493.91 2660 102 0.54 165 187.4 78.91 0.24
F01A-113-9 496.3 494.19 2660 120 0.51 234.3 75.00 0.29
KFM01A-113-10 496.45 494.34 2650 120 0.52 201 229.3 76.11 0.20
F01A-113-11 496.74 494.63 2642 125 0.51 247.3 75.90 0.29
KFM01A-113-12 496.74 494.63 2660 160 0.61 229 261.6 76.47 0.22
KFM01A-113-14 497.14 495.03 2660 101 0.42 229 238.9 74.23 0.18
F01A-113-15 497.42 495.31 2664 125 0.53 236.9 75.80 0.30
KFM01A-113-16 497.42 495.31 2660 126 0.53 204 235.6 75.31 0.23
F01A-113-17 497.71 495.60 2656 128 0.54 237.0 75.00 0.28
KFM01A-113-18 497.76 495.65 2660 125 0.43 220 288.6 78.00 0.18
F01A-113-19 498.04 495.93 2661 126 0.53 237.7 75.20 0.30
KFM04A-113-7 586.46 508.24 2650 110 0.51 176 217.8 73.02 0.24
KFM04A-113-8 586.60 508.37 2650 123 0.51 188 238.9 74.90 0.29
KFM04A-113-9 586.73 508.48 2650 133 0.54 200 244.9 75.00 0.26
KFM04A-113-10 586.95 508.67 2660 120 0.54 182 221.0 77.26 0.26
KFM04A-113-11 587.42 509.07 2660 113 0.51 172 221.3 75.11 0.28
KFM03A-113-15 523.82 522.37 2650 123 0.54 190 227.8 76.28 0.27
KFM03A-113-16 523.96 522.51 2650 120 0.57 180 209.1 71.36 0.25
KFM03A-113-17 524.09 522.64 2660 105 0.52 168 203.5 73.43 0.21
KFM03A-113-18 524.87 523.42 2650 124 0.57 190 217.4 77.74 0.25
KFM02A-113-7 527.72 526.00 2650 110 0.50 210 219.5 77.18 0.23
KFM02A-113-8 528.00 526.28 2650 92 0.48 175 193.3 78.60 0.20
KFM02A-113-9 528.91 527.18 2650 85 0.51 133 166.4 76.62 0.24
KFM02A-113-10 529.33 527.60 2660 134 0.55 215 241.7 77.42 0.21
KFM03A-113-22 670.54 668.69 2660 110 0.54 177 202.6 70.61 0.24
KFM03A-113-23 670.68 668.83 2660 120 0.48 200 251.1 72.81 0.22
KFM03A-113-24 670.82 668.97 2660 122 0.54 195 226.2 77.08 0.23
KFM03A-113-25 671.08 669.23 2650 128 0.56 190 227.2 80.10 0.25
KFM01A-113-20 690.32 687.39 2660 120 0.53 203 224.6 77.83 0.27
KFM01A-113-21 690.46 687.53 2650 130 0.52 219 248.0 76.51 0.23
KFM01A-113-22 690.60 687.67 2660 108 0.51 200 213.3 76.13 0.14
KFM01A-113-23 690.74 687.81 2650 120 0.48 223 249.1 77.86 0.24
KFM02A-113-13 705.55 703.25 2650 117 0.53 192 222.0 77.00 0.25
KFM02A-113-14 705.70 703.40 2660 107 0.53 186 202.0 73.27 0.20
KFM02A-113-15 705.85 703.55 2650 102 0.46 200 220.5 71.67 0.18
KFM04A-113-13 812.78 704.39 2660 114 0.55 170 207.8 72.93 0.28
KFM04A-113-14 812.92 704.51 2660 105 0.53 170 199.2 73.48 0.25
KFM04A-113-15 813.06 704.63 2660 100 0.49 175 205.4 72.09 0.25
KFM04A-113-16 813.20 704.75 2660 95 0.49 160 192.1 69.45 0.23
KFM04A-113-17 813.64 705.13 2660 95 0.52 174 183.1 71.29 0.21
KFM02A-113-16 708.25 705.94 2650 112 0.53 191 209.4 77.24 0.26
KFM02A-113-17 708.40 706.09 2650 113 0.53 187 213.6 75.28 0.22
69
C Simpevarp Uniaxial Compressive Strength
Table C.1: Summary of the test results from uncon• ned compression tests
for Simpevarp.
Borehole BH Depth Wet Density Dilation UCS Young 50 Poissons

(m) (kg/m**3) (MPa) (MPa) (GPa) Ratio 50
KSH01A 299.27 2770 79.0 158.3 70.7 0.31
KSH01A 301.66 2770 80.0 167.7 68.5 0.33
KSH01A 303.04 2780 65.0 170.4 69.0 0.28
KSH01A 311.21 2790 92.0 192.9 86.1 0.32
KSH01A 319.59 2810 95.0 186.0 80.7 0.33
KSH01A 401.48 2780 95.0 246.8 84.0 0.25
KSH01A 413.16 2780 48.0 109.3 81.5 0.24
KSH01A 486.05 2810 89.0 181.6 77.6 0.31
KSH01A 491.95 2800 62.0 142.1 80.5 0.26
KSH01A 494.81 2790 110.0 229.5 78.6 0.28
KSH01A 495.57 2790 90.0 171.5 100.5 0.27
KSH01A 699.67 2830 52.0 141.6 86.4 0.22
KSH01A 700.35 2830 65.0 133.1 77.0 0.27
KSH01A 702.38 2840 77.0 177.6 83.3 0.25
KSH01A 703.00 2830 87.0 160.2 82.6 0.24
KSH01A 703.15 2830 54.0 118.3 80.5 0.19
KSH02A 313.19 2790 194.0 244.4 85.4 0.26
KSH02A 313.33 2790 35.0 235.4 85.1 0.21
KSH02A 313.47 2790 95.0 263.8 90.9 0.27
KSH02A 323.23 2780 125.0 229.6 81.0 0.22
70
D Reliability of traditional in-situ stress
measurements in high stress environment
D.1 Experience from 420 Level of AECL’s URL

Once the sampling depth encounters in-situ stress magnitudes that are sufficient to start
inducing microcrack damage in the core samples, the properties of these damaged samples
no longer reflects the in-situ properties. Martin and Stimpson /1994/ clearly showed the
effect of stress-induced sample disturbance on the properties of Lac du Bonnet granite.
In-situ stress test results that rely on the interpretation of strains based on linear elastic
theory also become suspect when stress-induced damage is encountered Martin and
Christiansson /1991/.
Prior to the execution of AECL’s Mine-by Experiment a series of overcore measurements
were carried out using AECL’s Modified CSIR Triaxial Strain cell at the 420 Level of
AECL’s URL. The results, based on the classical linear elastic solution, are shown as
histograms in Figure D.1 and it is clear there was a large range in the stress magnitudes
measured. At the end of the Mine-by Experiment, the in-situ stress tensor was back
calculated using a variety of techniques based on large-scale displacement measurements
Read /1994/. Martin and Read /1996/ gave the likely range in this back-calculated stress
tensor magnitude as that shown in Figure D.1. It is clear from Figure D.1 that the
back-calculated stress magnitudes for Sigma 1 and Sigma 2 were greater than the stress
magnitudes given by the overcore test results.
This example is perhaps extreme as core discing was observed in nearly all the overcore
samples and it was obvious that the linear elastic theory was not appropriate for the
interpretation of the measure strains. However, despite the error the overcore results did
provide stress magnitudes that were adequate for the preliminary design of development of
the 420 Level, and the execution of the Mine-by Experiment. The important lesson, is that
the over core results were indicating that the stress magnitudes were elevated, and that the
measured stress magnitudes turned out to be lower bound stress magnitudes.
D.2 Forsmark stress data

In-situ stress data for this report were taken from Sjöberg /2004/. The magnitudes and
orientations were determined in borehole KFM01B at a depth between 238-242 m (Level
1) using the Borre probe triaxial strain cell Lindfors et al. /2005/. Three successful tests
were completed.
The average magnitude of σ1 for Level 1 was reported as approximately 40 MPa. Below
the Level 1 depth, overcoring encountered core-discing during the overcoring process. The
author’s experience suggests that if core discing is present the overcore data were
immediately suspect because the linear elastic theory used to interpret the measured strains
has been violated. It is also the author’s experience that the overcore results, when core
discing is present, show unrealistic scatter in magnitude and orientation. Hence, for the
purposes of this report only the Level 1 triaxial data were used to establish the statistical
variation at a given Level.
71
X <= 38. 53 X <= 58. 39
0.12
Overcore In Situ
0.1 Mean = 48.4
StDev = 6.0 60 æ 3
0.08
54 æ 13
0.06 Hydrulic Fracturing
Mean = 54
0.04 StDev =13
0.02 Sigma 1
0
0 5 10 15 20 25 30 35 40 45 50 55 60 65
X <= 22. 09 X <= 32. 38

0.25
Overcore In Situ
0.2 Mean=27.9,
Stdv=3.5 45 æ 4
0.15
0.1
Hydrulic Fracturing
Mean = 36
0.05
Sigma 2 StDev =16
0
0 5 10 15 20 25 30 35 40 45 50 55 60 65
X <= 11. 070 X <= 17. 397

0.35
In Situ Overcore
0.3 Mean = 14.2,
11 æ 4
0.25 Stdev = 1.1
0.2
Hydrulic Fracturing
0.15 Mean = 13.5
0.1 StDev =0
0.05 Sigma 3
Number of triaxial overcore tests=17
0
0 5 10 15 20 25 30 35 40 45 50 55 60 65
Stress magnitude (MPa)
Figure D.1: Comparison of the magnitudes from 17 triaxial overcore tests using AECL’s Modified
CSIR triaxial strain cell, with the back calculated in-situ stress magnitudes at the end of the Mine-by
Experiment.
72
The confidence interval for the Level 1 was established using the Excel macros provided
by SKB Lindfors et al. /2005/. The results are given in Figure D.2. It was assumed that the
range in the principal and horizontal stress magnitudes, determined from Level 1, was
similar at all depths.
73
KFM01B - Level 1 - Hole Depth between 238-242 m:
Meas. No. σ1 Dip Bearing σ2 Dip Bearing σ3 Dip Bearing
1 41.3 6.0 104.0 21.9 34.0 198.0 6.9 55.0 6.0
2 38.7 12.0 282.0 22.3 19.0 187.0 15.6 67.0 43.0
3 40.2 12.0 289.0 32.4 17.0 195.0 19.0 69.0 53.0
Average 39.5 4.6 284.0 25.4 25.4 191.8 14.6 64.1 23.5
Confidence Intervals for Level 1 Principal stresses

Principal stresses
σ1 Dip Bearing σ2 Dip Bearing σ3 Dip Bearing
[MPa] [°] [°] [MPa] [°] [°] [MPa] [°] [°]
Average: 39.5 4.6 284.0 25.4 25.4 191.8 14.6 64.1 23.5
95% lower 37.8 *) 14.6 0.5
95% upper 45.0 38.0 24.4
90% lower 38.0 *) 16.0 2.1
90% upper 44.1 36.7 23.1
80% lower 38.4 *) 17.6 3.9
80% upper 43.2 35.0 21.5
Measured values & deviation
Max : +1.8 +7.0 +6.0 +7.0 +8.4 +6.3 +4.4 +4.7 +28.4
Min : -0.8 1.5 -180.0 -3.5 -7.9 -3.0 -7.7 -8.6 -18.5
Max [%] : +5% +28% +30%
Min [%] : -2% -14% -53%
Confidence Intervals for Level 1 Horizontal stresses

Vertical and Horizontal components
σH Bearing σh Bearing σv
[MPa] [°] [MPa] [°] [MPa]
Average: 39.3 105.2 23.4 15.2 16.8

95% 37.1 *) 11.5 5.1
lower
95% 42.6 34.3 28.5
upper
90% 37.4 *) 12.9 6.6
lower
90% 42.2 32.9 27.1
upper
80% 37.8 *) 14.6 8.2
lower
80% 41.8 31.3 25.4
upper
Measured values &
deviation
Max : +1.7 +10.2 +7.7 +10.2 +4.3
Min : -1.5 -3.0 -6.2 -3.0 -4.8
Max +4% +33% +26%
[%] :
Min -4% -26% -29%
[%] :
Figure D.2: In-situ stress from KFM01B at Level 1 and the associated confidence intervals. Data
from Martin et al. /2001/.
74

Preliminary Assessment of Potential Underground Stability (Wedge and Spalling) at Forsmark, Simpevarp and Laxemar Sites

Uploaded by

Copyright:

Available Formats

Preliminary Assessment of Potential Underground Stability (Wedge and Spalling) at Forsmark, Simpevarp and Laxemar Sites

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Preliminary Assessment of Potential Underground Stability (Wedge and Spalling) at Forsmark, Simpevarp and Laxemar Sites

Uploaded by

Copyright:

Available Formats

R-05-71

Derek Martin, University of Alberta

Derek Martin, University of Alberta

A pdf version of this document can be downloaded from www.skb.se

3 Risk based methodology 13

A Rock mass spalling strength 63

B Forsmark Uniaxial Compressive Strength 69

C Simpevarp Uniaxial Compressive Strength 70

D Reliability of traditional in-situ stress measurements in high stress environment 71

Gravity-induced Stress-induced spalling.

Geotechnical engineering is fundamentally about managing risk. Morgenstern /1995/

Risk Assessment Risk control

Risk Risk Decision-

Hazard Risk Risk Risk

3.1 Wedge stability methodology

3.1.1 Frequency of Wedges

3.1.2 Severity of Wedges

3.1.3 Wedge data uncertainty and probability

FREQUENCY ANALYSIS SEVERITY OF HAZARD

RISK CONTROL OPTIONS

* Depth of Repository * Tunnel orientation * Tunnel support

Input parameter Values used in analysis

(a) Deposition tunnel (b) Deposition borehole

σsm (MPa) σsm /UCSm

(b) V-shaped notch

3.2.1 Spalling criterion

3.2.2 Severity of Spalling

1.25 1.5 1.75 2 2.25 2.5 2.75 3

Unrolled Perimeter Map

Plan view of deposition hole

And, Area 2 is given by:

The area of spalling (As ) is then given by:

3.2.3 EXCEL code for evaluating the spalling hazard

4 Available from www.palisade.com

Step 2: Spalling Factor of safety

Step 3: Depth of spalling

Depth of Spalling (m)

44 MPa 59 MPa 45 MPa 45

Room 413 Room 405 1-m-diameter borehole

Mine-by Test Tunnel σ1 σ3

3.2.4 Calibration of methodology

Probability of Spalling - URL 420 Level

1 Very Minor Spalling

0.6 0.8 1 1.2 1.4 1.6

Shear stress (MPa)

3.3 Spalling for noncircular openings

3.4 Spalling and thermally-induced stresses

Depth Geometry Thermal Spalling

1.95 125 0* 125

3.55 106 24 130

(4.1 101 21 122)

4.1 Input data

Table 4–1: Joint sets obtained from Discrete Fracture Model

Joint Label Dip Direction (◦ ) Dip (◦ )

4.1.2 In-situ stress

σhmin = 19 + 0.025z σHmax = 35 + 0.02z

σHmax σhmin σvert

4.1.3 Uniaxial Compressive Strength

Uniaxial Compressive Strength (MPa)