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

net/publication/282393652

Study on the effect of principal stress orientation on tunnel stability

Article  in  Sichuan Daxue Xuebao (Gongcheng Kexue Ban)/Journal of Sichuan University (Engineering Science Edition) · June 2012

CITATIONS READS
9 215

3 authors, including:

Zheming Zhu
Sichuan University
308 PUBLICATIONS   1,566 CITATIONS   

SEE PROFILE

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

rock dynamics View project

All content following this page was uploaded by Zheming Zhu on 15 August 2016.

The user has requested enhancement of the downloaded file.

 Tunnelling and Underground Space Technology 49 (2015) 279–286

Contents lists available at ScienceDirect

Tunnelling and Underground Space Technology

journal homepage: www.elsevier.com/locate/tust

The effect of principal stress orientation on tunnel stability

Zheming Zhu ⇑, Yuanxin Li, Jun Xie, Bang Liu
School of Architecture and Environment, Sichuan University, Chengdu 610065, China

a r t i c l e i n f o a b s t r a c t

Article history: In order to investigate the effect of principal stress orientation on the stability of regular tunnels and
Received 1 November 2014 cracked tunnels, experiments by using square specimens with a centralized small tunnel were conducted,
Received in revised form 30 March 2015 and the corresponding numerical study as well as photoelastic study were implemented. Two kinds of
Accepted 4 May 2015
materials, cement mortar and sandstone, were used to make tunnel models, and three types of tunnel
models were studied, i.e. (1) regular tunnel models loaded by different orientation’s principal stresses,
(2) tunnel models with various orientation’s radial cracks in the spandrel under compression, and (3) tun-
Keywords:
nel models with a fixed radial crack loaded by various orientation’s principal stresses. In the numerical
Principal stress orientation
Tunnel model
study, the stress intensity factors of the radial cracks were calculated, and the results agree well with
Numerical simulation the test results. For regular tunnels, when the angle h between the major principal stress and the tunnel
Stress intensity factor symmetrical axis is 45°, the corresponding tunnel is the most unfavorable; for tunnels with a radial crack
in the spandrel, when the angle b between the crack and the tunnel wall is 135°, the corresponding tunnel
is the most unfavorable; for tunnels with a b = 130° radial crack, when h = 0° or h = 70°, the compressive
strengths of the tunnel models are comparatively low, whereas when h = 90°, it is the highest.
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction
The stress states of underground rock mass may vary with dif-
ferent locations because except for the weight of the overlying rock
mass, many factors are involved, such as the regional tectonic
movement, the geological structure, the nearby volcanic eruptions
or earthquakes (Brady and Brown, 1985; Roman et al., 2004;
Bohnhoff et al., 2006; Zhu, 2013). This indicates that the principal
stress orientation around a structure may not be in vertical plane
or in horizontal plane, but has a certain angle with those planes.
The stress states of rock mass could play an important role in
underground structure stability, e.g. a tectonic movement or earth-
quake would produce stress state changing which could cause
mass destruction, thus it is necessary to investigate the influence
of stress states on structure’s stability. The best approach to the
study is to first generate an extensive experimental database on
tunnel model failure properties, and then, based on these experi-
mental database, to investigate the failure mechanism through
numerical models so as to obtain a better understanding of the
dominant parameters that control tunnel stability.
In this paper, the effect of principal stress orientation on the sta-
bility of tunnels and cracked tunnels will be focused. The issue of
tunnel stability has been studied experimentally and numerically
by many researchers, and accordingly many significant research
⇑ Corresponding author.
E-mail address: zhemingzhu@hotmail.com (Z. Zhu).
results have been published, e.g. Song et al. (2001), Charpentier
et al. (2003), Sterpi and Cividini (2004), Zhou and Qiu (2006),
Chu et al. (2007), Wang et al. (2008), Lee and Schubert (2008),
Seki et al. (2008), He et al. (2010), Li et al. (2010) and Li et al.
(2012), etc. In the process of rock excavation, fragmentation by
using explosive is a widely applied method in mining and quarry-
ing due to its properties of easy operation, less cost, and high effi-
ciency. Under the action of explosive detonation, blast-induced
radial cracks will occur in the surrounding rock, which could lar-
gely affect tunnel stability (Zhu et al., 2007, 2008; Zhu, 2009a).
About the effects of cracks or weak interlayers, many research
results have been published. Suorineni et al. (1999) and Bruneau
et al. (2003) pointed out that weak interlayers have a great influ-
ence on tunnel stability due to their low strength and stiffness.
Many failures of underground openings were reported to be closely
related to the occurrence of weak interlayers nearby. Jeon et al.
(2004) investigated the effect of interlayer, weak plane, and grout-
ing on the stability of a tunnel, through scaled model tests and
numerical analysis. Huang et al. (2013) indicated that the weak
interlayer affected the stability of tunnel by increasing the failure
zones and causing asymmetrical stress distribution. The location,
dip and thickness of the interlayer as well as the distance to tun-
nels were proved to be important factors influencing tunnel
stability.
In numerical study, continuum-based and discontinuum-based
numerical analyses are two major methods. The continuum-based
methods include finite element method (FEM), finite difference

http://dx.doi.org/10.1016/j.tust.2015.05.009
0886-7798/Ó 2015 Elsevier Ltd. All rights reserved.
280 Z. Zhu et al. / Tunnelling and Underground Space Technology 49 (2015) 279–286

Fig. 1. Three test models of tunnels.

Table 1
Material parameters of cement mortar and sandstone.

Material Young’s modulus E (GPa) Poisson’s ratio m Density q (kg/m3) Cohesion c (MPa) Internal UCS (MPa)
friction angle
w (°)
Cement mortar 2.7 0.2 2375 1.2 15 8.25
Sandstone 12.1 0.2 2280 2.01 38 38 11.7

method (FDM), boundary element method (BEM) and meshless mortar and sandstone will be used to make tunnel models. In the
method (MM), etc. Dhawan et al. (2002) analyzed the effects of numerical study, the finite element code ABAQUS will be applied
weak zones in rock mass and creation of multiple cavities in the in the analysis of tunnel stability. The numerical results are com-
inhomogeneous rock mass by using FEM. Lee and Kim (2003) pared with the test data, and some conclusions are presented.
assessed the effect of fault zones under different in-situ stresses
by using hybrid method of FEM and BEM. Jia and Tang (2008) used
2. Test models and materials
a FEM based code RFPA to study the influence of different dip angle
of layered joints and the lateral pressure coefficient on the stability
Principal stress orientations may have a large effect on tunnel
of tunnel in jointed rock mass. Huang and Xiao (2010) simulated
stability, and therefore, it is necessary to study tunnel failure prop-
the construction of a double arch tunnel through a weak interlayer
erty experimentally and numerically. In this paper, three typical
based on Mohr–Coulomb yield criterion by FLAC3D code. Zheng
tunnel models are studied, the first one is a regular tunnel loaded
et al. (2011) used strength reduction FEM to analyze tunnel dam-
by various orientation’s principal stresses as shown in Fig. 1(a), the
age and safety coefficient.
second one is a tunnel with various inclination b cracks as shown
Although many significant results have been published, less
in Fig. 1(b); and the third one is a tunnel with a fixed angle crack
attention has been paid to the effect of principal stress orientation
loaded by various orientation’s principal stresses as shown in
on the stability of tunnels and cracked tunnels. The objective of
Fig. 1(c).
this study is to experimentally and numerically investigate the
Considering the action of lateral pressures in the reality of tun-
relationship between principal stress orientations and tunnel sta-
nels, the tunnel models in Fig. 1 are confined by two steel plates
bilities. Two typical tunnels, i.e. regular tunnels and cracked tun-
during vertical loading, which has the function of limiting the
nels will be considered. In the experimental study, the cement

Fig. 2. A cement mortar tunnel model and its loading device. Fig. 3. The typical vertical stress–strain curves obtained from the model tests.
 Z. Zhu et al. / Tunnelling and Underground Space Technology 49 (2015) 279–286 281

Fig. 4. Failure patterns of cement mortar tunnel models with different inclination angles h.

lateral deformation of the models, as shown in Fig. 1. This confine- made by using two kinds of materials, cement mortar and sand-
ment is equivalent to load the lateral pressure r2 on the models, stone. For cement mortar, the ratio of cement:sand:gypsum:water
and according to elasticity theory, the relationship between r2 is 1:2.6:0.52:0.3 by weight. The mechanical parameters of cement
and r1 is: mortar and sandstone are shown in Table 1.

r2 ¼ mr1
where m is Poisson ratio.
The models are square plates measuring 350 mm in length,
350 mm in width and 100 mm in thickness with a centralized
small tunnel, which is 60 mm in height and 50 mm in width, and
the tunnel’s arch consists of a semi-circle. The specimens were
ð1Þ
3. Effect of principal stress orientation on tunnel model
strength
In these tests, seven groups of tunnel models with different
inclination angle h (0°, 15°, 30°, 45°, 60°, 75° and 90°) between
the major principal stress and the tunnel symmetrical axis as

Fig. 5. Failure patterns of sandstone tunnel models with different inclination angles h.
282
Fig. 6. Test results of critical compressive stresses versus tunnel inclination angles
from cement mortar tunnel models.
Fig. 7. Test results of critical compressive stresses versus tunnel inclination angles
from sandstone tunnel models.
Fig. 8. Photoelastic experiment results of tunnels with different inclination angle h under the same compressive load
Z. Zhu et al. / Tunnelling and Underground Space Technology 49 (2015) 279–286
shown in Fig. 1(a) were made, and for each angle, there are 3 spec-
imens. The specimens were loaded vertically by a 200T
Electro-hydraulic servo control device, as shown in Fig. 2, and
the typical vertical stress–strain curves for the cases h = 0° and
h = 30° are shown in Fig. 3.
Figs. 4 and 5 show the failure patterns of a group tunnel models
with various inclination angles h from 0° to 90° for cement mortar
and sandstone, respectively. It can be seen that the failure patterns
of the specimens vary with the inclination angle h. When h = 0° and
h = 15°, the damages occur near the both side straight walls; when
h = 30°, h = 45° and h = 60°, the damages mainly occur in the span-
drel and the feet of the tunnels; when h = 75° and h = 90°, the dam-
ages mainly occur in the bottom and the arch head of the tunnels.
The test result of the curves of compressive strength versus tun-
nel inclination angle h for cement mortar models and sandstone
models are shown in Figs. 6 and 7, respectively. It can be seen that
the compressive strength varies with the tunnel inclination angle h,
and for both materials, the trendlines are almost the same. When h
is between 20° and 30°, the tunnel models have extreme values of
compressive strength for both materials, and when h is 45°, the
compressive strength is the lowest. The difference is about 30%
decrease when the inclination angle h increases from 25° to 45°.
This indicates that as h is 45°, the stability of the tunnel structure
is comparatively low.
In order to investigate the stress states around the tunnels in
the models under compression, photoelastic experiments were
carried out in this paper, and the results are shown in Fig. 8.
Photoelasticity is an optical technique for experimental stress anal-
ysis, and it can be used for getting the information of principal
stress difference for every points inside a domain based on the gen-
eral rule that the more the fringe number, the larger the stress con-
centration (Wang et al., 2012). It can be seen that the stresses are
concentrated in both sides of the tunnels; when h = 30° and 45°,
the fringe numbers near the right side of the tunnels are large as
compared to the others. This agrees with the test results that as
h = 45°, the tunnel model has the lowest strength.
4. Effect of crack inclination angles on tunnel stability
In the process of blasting excavation, blast-induced stress
waves could cause a large number of radial cracks around the
r1 .
 Z. Zhu et al. / Tunnelling and Underground Space Technology 49 (2015) 279–286
tunnels (Zhu et al., 2008; Zhu, 2009a), and these radial cracks could
weaken tunnel stability and cause large engineering disasters, such
as rock bursts. In order to find the effect of crack inclination angle b
on tunnel stability, we designed 5.0 cm radial crack with various
inclination angles b at tunnel spandrels, as shown in Fig. 1(b).
Using cement mortar and sandstone, we made 19 groups of tun-
nel models, respectively, and each group consists of 2 specimens
for cement mortar tunnel models and 3 specimens for sandstone
tunnel models. The angle b varies from 0° to180°, and the interval
is 10°. By using the same equipment, the critical compressive stres-
ses were measured, and the relationships between the compres-
sive strength and the crack inclination angle b were obtained, as
shown in Fig. 9 for cement mortar tunnel models and Fig. 10 for
sandstone tunnel models.
Figs. 9 and 10 show that the critical compressive stresses vary
with the inclination angle b. When b increases from 0° to about
135°, the critical stress decreases, and when b is about 135°, the
critical stress is the lowest, and when b increases from 135° to
Fig. 9. Testing results of the compressive strengths for cement mortar tunnel
models with different crack inclination angle b.
Fig. 10. Testing results of the compressive strengths for sandstone tunnel models
with different crack inclination angle b.
283
180°, the critical stress increases. Comparing cement mortar test
results (in Fig. 9) with sandstone test results (in Fig. 10), one can
find that the latter curve is much flatter.
In order to further study the effect of crack’s inclination angle
on tunnel stability, numerical study by using ABAQUS code was
implemented. Based on the tunnel model shown in Fig. 1(b),
numerical tunnel models were established. In this model, except
for the zone near crack tips where triangular element CPS6 is
applied, as shown in Fig. 11, quadrilateral element CPS8 is
employed. J-integration method is applied in the calculation of
crack tip stress intensity factor (SIF), and the total element number
is 10,800. The dimensionless stress intensity factor YI (mode I
crack) and YII (mode II crack) were calculated under different crack
inclination angle b, and the calculation results are shown in Fig. 12.
From Fig. 12, one can find that except for the case that b > 170°,
YI is negative. This is because cracks close under compression
which will not result in stress concentration at crack tips.
However, compressive load will cause compressive stress at crack
tips, which will affect the stress field. Therefore, the negative YI will
affect the crack propagation, but the effect extent of the negative YI
is not clear yet (Zhu, 1999, 2009b). The dimensionless stress inten-
sity factors YII vary with crack inclination angle b, and when b is
about 135°, the absolute value of YII is the largest, which basically
agrees with the test results as shown in Figs. 9 and 10.
Fig. 13 shows the photoelastic results of tunnels with different
crack inclination angle b under the same compressive load r1 . It
can be seen that as the crack inclination angle b is 110°, 130°,
and 150°, the fringe numbers near crack tips are comparatively
large.
Fig. 11. A crack and six node triangular elements.
Fig. 12. Calculation results of the dimensionless SIF YI and YII with different crack
inclination angle b.
284 Z. Zhu et al. / Tunnelling and Underground Space Technology 49 (2015) 279–286

Fig. 13. Photoelastic results of tunnels with different crack inclination angle b under the same compressive load r1 .

5. Effect of principal stress orientation on cracked tunnel
stability
From the above experiment study, one can find that when b is
about 130°, the strength of the tunnel model is comparatively
low, which means that the corresponding tunnel is unfavorable.
Therefore, we selected this angle 130° as the crack inclination
angle b0, as shown in Fig. 1(c). The test results of critical compres-
sive stresses for cement mortar tunnels with different inclination h
are presented in Fig. 14. One can find that the compressive strength
varies with the tunnel inclination angle h. When h is 15° and 70°,
the compressive strength is comparatively low, whereas when h
is 90°, it is the highest.
By using ABAQUS code, the dimensionless stress intensity factor
YI and YII were calculated, and for different crack inclination angle
h, the values of YI and YII are plotted in Fig. 15. It shows that YI and
YII vary largely with the tunnel inclination angle h, and YI are
always negative. When h = 0° and h = 75°, both the absolute values
of YII are the low extreme value, which means that the correspond-
ing cracked tunnel should have a low strength. Comparing with the
test results shown in Fig. 14, one can find that the calculation
results basically agree with the test results.
Fig. 16 shows the photoelastic test results of cracked tunnels
with different inclination angle h under the same compressive load
r1 . It can be seen that the fringe number near crack tips varies with
angle h, and as h is 0°, 15°, 60° and 75°, the fringe numbers near
crack tips are comparatively large, which generally agrees with
the test results shown in Fig. 14.

Fig. 14. Test results of compressive strength of cement mortar tunnel models with Fig. 15. Calculation result of the dimensionless SIF with different tunnel inclination
different crack inclination angle h. angle h.
 Z. Zhu et al. / Tunnelling and Underground Space Technology 49 (2015) 279–286
Fig. 16. Photoelastic results of cracked tunnels with different inclination angle h under the same compressive load
6. Conclusion
The underground stress states vary with locations due to many
factors involved, and therefore, the principal stress orientations
may not be in vertical or horizontal planes, but they may have a
certain angle with those planes. The principal stress orientations
play an important role in tunnel stability, and it is necessary to
investigate because clear understanding of the effect will be of
benefit to the design and construction of tunnels.
In order to study the effect of principal stress orientation on
tunnel stability, three types of tunnel models were considered,
and two kinds of materials, cement mortar and sandstone, were
used to make tunnel models. Numerical study by using ABAQUS
code, and photoelastic tests were implemented, and based on the
experiment and numerical simulation results, the following con-
clusions can be obtained:
(1) For regular tunnels under compressions, as the angle h
between the major principal stress and the tunnel axis is
45°, the tunnel stability is comparatively low.
(2) For tunnels with a radial crack in tunnel spandrel under
compressions, as the crack inclination angle b is 135°, the
crack stress intensity factor KII is the highest (absolute
value), and therefore, the corresponding tunnel stability is
the lowest.
(3) For tunnels with a b = 130° radial crack under compressions,
as the angle h between the major principal stress and the
tunnel axis is 15° and 70°, the corresponding tunnel stability
is comparatively low, and as h is 90°, the corresponding tun-
nel stability is high.
Acknowledgements
This work was financially supported by the Open Fund of State
Key Laboratory of Oil and Gas Reservoir Geology and Exploitation
(PLN1202), by the project of Science and Technology of Sichuan
285
r1 .
province (2014JY0002), and by the National Natural Science
Foundation of China (51074109).
References
Bohnhoff, M., Grosser, H., Dresen, G., 2006. Strain partitioning and stress rotation at
the North Anatolian fault zone from aftershock focal mechanisms of the 1999
Izmit Mw = 7.4 earthquake. Geophys. J. Int. 166, 373–385.
Brady, B.H.G., Brown, E.T., 1985. Rock Mechanics for Underground Mining. George
Allen and Unwin Lid.
Bruneau, G., Tyler, D.B., Hadjigeorgiou, J., Potvin, Y., 2003. Influence of faulting on a
mine shaft—a case study: Part I—background and instrumentation. Int. J. Rock
Mech. Min. Sci. 40 (1), 95–111.
Charpentier, D., Tessier, D., Cathelineau, M., 2003. Shale microstructure evolution
due to tunnel excavation after 100 years and impact of tectonic paleo-
fracturing, Case of Tournemire, France. Eng. Geol. 70 (1/2), 55–69.
Chu, B.L., Hsu, S.C., Chang, Y.L., Lin, Y.S., 2007. Mechanical behavior of a twin-tunnel
in multi-layered formations. Tunn. Undergr. Space Technol. 22 (3), 351–362.
Dhawan, K.R., Singh, D.N., Gupta, I.D., 2002. 2D and 3D finite element analysis of
underground openings in an inhomogeneous rock mass. Int. J. Rock Mech. Min.
Sci. 39 (2), 217–227.
He, M.C., Gong, W.L., Zhai, H.M., Zhang, H.P., 2010. Physical modeling of deep
ground excavation in geologically horizontal strata based on infrared
thermography. Tunn. Undergr. Space Technol. 25 (4), 366–376.
Huang, F., Zhu, H.H., Xu, Q.W., 2013. The effect of weak interlayer on the failure
pattern of rock mass around tunnel-scaled model tests and numerical analysis.
Tunn. Undergr. Space Technol. 35, 207–218.
Huang, R.Q., Xiao, H.B., 2010. Deformation mechanism of a shallow double-arch
tunnel in a sloping rock mass. Bull. Eng. Geol. Environ. 69 (1), 89–97.
Jeon, S., Kim, J., Seo, Y., Huang, C., 2004. Effect of a fault and weak plane on the
stability of a tunnel in rock – a scaled model test and numerical analysis. Int. J.
Rock Mech. Min. Sci. 41 (3), 658–663.
Jia, P., Tang, C., 2008. Numerical study on failure mechanism of tunnel in jointed
rock mass. Tunn. Undergr. Space Technol. 23 (5), 500–507.
Lee, I.M., Kim, D.H., 2003. A simulation using a hybrid method for predicting fault
zones ahead of a tunnel face. Int. J. Numer. Anal. Meth. Geomech. 27 (2), 147–
158.
Lee, Y.Z., Schubert, W., 2008. Determination of the round length for tunnel
excavation in weak rock. Tunn. Undergr. Space Technol. 23 (3), 221–231.
Li, Z.H., Zhu, W.C., Feng, X.T., 2010. Effect of lateral pressure coefficients on damage
and failure process of horseshoe-shaped tunnel. Rock Soil Mech. 31 (S.2), 441–
448+468.
Li, Y.Q., Zhu, Z.M., Hu, R., 2012. Study on the effect of principal stress orientation on
tunnel stability. J. Sichuan Univ. 44 (S1), 93–98.
Roman, D.C., Moran, S.C., Power, J.A., Cashman, K.V., 2004. Temporal and spatial
variation of local stress fields before and after the 1992 eruptions of Crater Peak
vent, Mount Spurr volcano, Alaska. Bull. Seism. Soc. Am. 94, 2366–2379.
286 Z. Zhu et al. / Tunnelling and Underground Space Technology 49 (2015) 279–286
Seki, S., Kaise, S., Morisaki, Y., Azetaka, S., Jiang, Y., 2008. Model experiments for
examining heaving phenomenon in tunnels. Tunnell. Undergr. SpaceTechnol. 23
(2), 128–138.
Song, J.J., Lee, C.I., Seto, M., 2001. Stability analysis of rock blocks around a tunnel
using a statistical joint modeling technique. Tunn. Undergr. Space Technol. 16
(4), 341–351.
Sterpi, D., Cividini, A., 2004. A physical and numerical investigation on the stability
of shallow tunnels in strain softening media. Rock Mech. Rock Eng. 37 (4), 277–
298.
Suorineni, F.T., Tannant, D.D., Kaiser, P.K., 1999. Determination of fault-related
sloughage in open stopes. Int. J. Rock Mech. Min. Sci. 360, 891–906.
Wang, J.L., Chen, J.P., Su, S.R., 2008. Study of collapse mechanism of tunnels in
jointed rock mass by distinct element method. J. China Univ. Min. Technol. 37
(3), 316–319.
Wang, M., Zhu, Z.M., Liu, J., 2012. The photoelastic analysis of stress intensity factor
for cracks around a tunnel. Appl. Mech. Mater. 142, 197–200.
View publication stats
Zheng, Y.R., Wang, Y., Wang, C., 2011. Stability analysis and exploration of failure
law of jointed rock tunnel. Chin. J. Undergr. Space Eng. 7 (4), 649–656.
Zhou, Y.L., Qiu, X.H., 2006. Study on stability of stratified rock tunnel. Chin. J.
Undergr. Space Eng. 2 (2), 345–348.
Zhu, Z.M., 1999. New biaxial failure criterion for brittle materials in compression. J.
Eng. Mech., ASCE 125 (11), 1251–1258.
Zhu, Z.M., Mohanty, B., Xie, H., 2007. Numerical investigation of blasting-induced crack
initiation and propagation in rocks. Int. J. Rock Mech. Min. Sci. 44 (3), 412–424.
Zhu, Z.M., Xie, H., Mohanty, B., 2008. Numerical investigation of blasting-induced
damage in cylindrical rocks. Int. J. Rock Mech. Min. Sci. 45 (2), 111–121.
Zhu, Z.M., 2009a. Numerical prediction of crater blasting and bench blasting. Int. J.
Rock Mech. Min. Sci. 46 (6), 1088–1096.
Zhu, Z.M., 2009b. An alternative form of propagation criterion for two collinear
cracks under compression. Math. Mech. Solids 14 (8), 727–746.
Zhu, Z.M., 2013. Evaluation of the range of horizontal stresses in the earth’s upper
crust by using a collinear crack model. J. Appl. Geophys. 88 (1), 114–121.