See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/257445083

Case Histories of Four Extremely Intense

Rockbursts in Deep Tunnels

Article in Rock Mechanics and Rock Engineering · May 2012

DOI: 10.1007/s00603-011-0218-6

CITATIONS READS

29 85

5 authors, including:

Xia-Ting Feng Shili Qiu

Chinese Academy of Sciences Chinese Academy of Sciences
285 PUBLICATIONS 1,923 CITATIONS 20 PUBLICATIONS 144 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Shili Qiu on 05 March 2015.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
Rock Mech Rock Eng
DOI 10.1007/s00603-011-0218-6

ORIGINAL PAPER

Case Histories of Four Extremely Intense Rockbursts in Deep

Tunnels
Chuanqing Zhang • Xia-Ting Feng •
Hui Zhou • Shili Qiu • Wenping Wu

Received: 10 August 2011 / Accepted: 18 December 2011

 Springer-Verlag 2011

Abstract In the process of excavating seven parallel drainage and headrace tunnels were constructed. Rockburst
tunnels at the Jinping II Hydropower Station, several intensity in the auxiliary tunnels was classified into four
extremely intense rockbursts occurred, killing and injuring grades: light class (grade I), moderate class (grade II),
construction workers and damaging several sets of equip- intense class (grade III), and extremely intense class (grade
ment. Based on the characteristics and mechanisms of these IV) (Shan and Yan 2010).
rockbursts, four typical events were selected and their The grade of a rockburst is considered on evaluation of
temporal and spatial characteristics were here described in the effects of external influences, while the type of rockburst
detail. The geological conditions revealed after the rock- reflects the control mechanism of that event. In an analysis
bursts were surveyed carefully. The responses of support of rockburst events, it is equally important to determine the
elements were also analyzed. The details documented in type of rockburst. In general, the three basic mechanisms of
each case provide not only an important reference for rockbursts, i.e., strain burst, fault-slip burst, and their
understanding the development mechanisms of rockbursts combination, have been accepted (Board 1994; Tang 2000).
but also a basis for the selection and development of Ortlepp and Stacey (1994); Ortlepp (2001) thought that, due
rockburst prevention measures in deep hard rock tunnels. to the large-scale excavation, shear-rupture and fault-slip
types of rockbursts are more possible in deep mining tunnels
Keywords Deep tunnel
Hard rock
Rockburst
Support than in deep civil tunnel projects. According to Tang (Tang
response
TBM 2000) the majority of rockbursts during civil tunnel con-
structions were of the strain type. In fact, all three types of
rockbursts occur in deep civil tunnel projects in which,
1 Introduction nevertheless, a fault-slip burst mainly refers to a small-scale
slip on a structural plane. Using only site survey, it is often
The Jinping II Hydropower Station, which traverses difficult to distinguish a fault-slip burst from a rockburst
Jinping Mountain, has seven parallel tunnels, each 17-km with combination mechanism because both of them include
long with a maximum depth of 2,525 m. A number of the influence of structure plane.
extremely intense rockbursts occurred during the excava- Because a large number of intense and extremely intense
tions of the tunnels, killing several construction workers, rockbursts occurred in the Jinping II project, it is not
injuring many more, and damaging several sets of equip- practical to describe every event. Only four typical events
ment. The auxiliary tunnels were connected before the were selected for this study. Their characteristics and
mechanisms are representative of the other rockbursts that
occurred at this site during this period. The corresponding
C. Zhang (&)
X.-T. Feng
H. Zhou
S. Qiu
W. Wu spatial and temporal characteristics, support responses, and
State Key Laboratory of Geomechanics and Geotechnical geological conditions of these events are described in
Engineering, Institute of Rock and Soil Mechanics,
detail. The information revealed in these rockbursts can
Chinese Academy of Sciences, Wuhan 430071,
People’s Republic of China provide important references for the design and construc-
e-mail: [email protected] tion of other deep hard-rock tunnel projects.

123
 C. Zhang et al.

2 Project Overview TBM machine. In fact, this rockburst followed the con-
tinuous development of several extremely intense rock-
The Jinping II Hydropower Station has seven parallel bursts, as shown in Fig. 2.
tunnels, including two auxiliary tunnels, a drainage tunnel, In October 8, 2009, a rockburst occurred at about 1 o’clock
and four headrace tunnels. The axes of these tunnels are north of Stake SK9 ? 302 during the TBM tunneling pro-
orientated at N58W. Here, the entrances at the two ends of cess, accompanied by a sound similar to blasting. The failure
each tunnel are called east end and west end, respectively zone had a maximum depth of 1.7 m. Another rockburst
(Fig. 1). When facing the west end, the left side is called occurred at about 1–2 o’clock north of Stake SK9 ? 314,
the south side and the right side the north side. Like the where the failure zone with a depth of 1.2–1.7 m extended to
drainage tunnel (with a 7.2-m diameter), the headrace the tunnel face. A rockburst also occurred at about 7–10
tunnels #1 and #3 were excavated with TBMs. They have a o’clock south of Stake SK9 ? 311–9 ? 322, where the
circular section with a 12.4-m diameter. The drilling and cauldron-shaped failure zone had a maximum depth of about
blasting (D&B) method was adopted for the upper and 1 m, as shown in Fig. 3a.
lower bench excavations in the headrace tunnels #2 and #4. On October 9, an extremely intense rockburst occurred
Both tunnels have a 13-m diameter with the bench heights at Stake SK9 ? 301–9 ? 322, where the tunnel face was
of 8.5 m and 4.5 m, respectively. located at SK9 ? 296. The 14-beam steel sets buckled
This project was described in more detail by Wu et al. severely due to the heavy impact of the rockbursrt, as
(2010); Shan and Yan (2010), Zhang et al. (2011a), (b). shown in Fig. 4a. The major failure zone with a 2–3 m
Interested readers can refer to these literatures. depth was located at 10–2 o’clock on the crown of tunnel
A number of extremely intense rockbursts occurred section SK9 ? 301–9 ? 314. A collapse with a 1-m failure
during the excavation of these seven tunnels. Four typical depth was induced by the seismic waves from the major
ones among them with distinctive characteristics were rockburst zone. It was located at about 10–7 o’clock of
selected for case studies: the ‘‘11.28’’ event in the drainage Stake SK9 ? 311–9 ? 322, as shown in Fig. 3b. This
tunnel, the ‘‘7.14’’ event in the headrace tunnel #4, the rockburst was accompanied by loud blast-like sounds. The
‘‘4.16’’ event in the headrace tunnel #3, and the ‘‘2.4’’ tunnel was instantly filled with gray dust.
event in the headrace tunnel #2 (Fig. 1). The clock cali- Before the rockburst, shotcrete and H-section steel sets
bration method was used to describe the different positions were introduced into this tunnel section. The surrounding
of the cross section of the tunnels excavated by TBMs. rock mass had been reinforced by cement-grouted rock-
bolts with a spacing of 1 m and rod length of 3 m. How-
ever, the bolts were so short that they could not cross the
3 The ‘‘11.28’’ Rockburst in the Drainage Tunnel rockburst failure zone and they could not have played a
role in resistance to impact. Even so, as viewed from the
On November 28, 2009, an extremely intense rockburst integrity of the support structures after a rockburst, steel
(often referred to as the ‘‘11.28’’ rockburst or the ‘‘11.28’’ sets indeed played an important role. Although bending
event) occurred at Stake SK9 ? 283–9 ? 322 in the deformation and damage occurred to some steel sets, the
drainage tunnel at a depth of 2,330 m. It caused seven integrity was still maintained. They resisted the impact of
deaths and one injury as well as the total destruction of the fractured rock mass without flying out or collapsing, which

Fig. 1 The locations of the West End East End

extremely intense rockbursts
z
x
y

The “4.16” rockburst The “7.14” rockburst in the The “2.4” rockburst in
in the headrace tunnel headrace tunnel #4 and the the headrace tunnel #2
#4 at a 1,980 m depth. “11.28” rockburst in the at a 1,900 m depth
drainage tunnel at a 2,300 m
depth.

Marble Greenschist
Slate and sandstone Fault

123
Case Histories of Four Extremely Intense Rockbursts

Fig. 2 Development of the SK9+283 Rockburst on November 28 SK9+322

‘‘11.28’’ rockburst in the
drainage tunnel (F1–F4 are the
locations of the advancing
Rockburst on November 15
tunnel faces)
SK9+287 SK9+292
Rockburst on November 6 and 7 Rockburst on October 8 and 9

SK9+296 SK9+301 SK9+322 E

W
SK9+291
Drainage
tunnel
F4 F3 F2 F1 Advancing direction 35m

#B
35m
#A

Fig. 3 Failure zone

morphologies in the drainage Tunnel profile Profile on October 9
Structural plane
tunnel: a rockburst on October 8 after rockburst
b rockburst on October 9
c rockburst on November 6 and
7 and d rockburst on November
15

Profile on
October 9
Profile on October 8
(a) (b)

Tunnel profile The first rockburst

after rockburst

The second rock burst

(c) (d)

may have improved the safety of both equipment and rockburst occurred near the cutter head and shield with very
personnel. This event only caused damage to the south roof loud sounds, resulting in damages to the reducer and brake
bolter platform, part of the oil circuit, and the electrical of a bolter at the south of the TBM, as shown in Fig. 4b.
apparatus circuit. Meanwhile, TBM tunneling was suspended. The failure
When the tunnel face advanced 4 m at Stake SK9 ? 292 zone profile after this rockburst is shown in Fig. 3c within
on November 6, a rockburst occurred at about 6–10 o’clock the range of Stake SK9 ? 296–9 ? 291 (extending 1 m in
at Section SK9 ? 292, leading to the migration of the front of the tunnel face).
TBM centerline with a 13.4-mm horizontal displacement On November 13, TBM tunneling resumed, but when
and an 8.9-mm height difference. On November 7, another the face advanced by 4 m on November 15, rockbursts

123
 C. Zhang et al.

Fig. 4 Photos after the

rockbursts in the drainage
tunnel: a the steel sets at the
arch crown were damaged by
the rockburst on October 9,
b the south bolter was damaged
by the rockburst on November 7
and c the failure zone on the
north sidewall caused by the
rockburst on November 15

Steel sets at the arch

crown was bent or
damaged Arch crown subsiding

(a)

Steel sets
Shield
Blasting pit

Shield

Ejected
fragments

(b) Roof bolter in L1 area (c)

occurred at 7–3 o’clock of Stake SK9 ? 288 and 12–4 30 m behind the cutter head was buried in stones, as shown
o’clock of SK9 ? 292, respectively, accompanied by huge in Fig. 5. The shockwave of this rockburst broke off the
sounds. The deepest failure zone was located at the 2:30–3 door of the duty room on the second floor of the backup
o’clock position at a depth of approximately 3 m, and the system. The total volume collapsed was more than 400
failure zone at the south waistline was 2-m deep and cubic meters. The seismic events before and during the
extended 1 m in front of the tunnel face, as shown in rockburst were recorded using microseismic monitoring
Figs. 3d, 4c. After a number of rockbursts, an accumulated equipment, which indicated a Richter magnitude of 2.0.
horizontal displacement of 19.8 mm for the TBM center- The tunnel section at Stake SK9 ? 283–9 ? 322 con-
line was observed with an 18.4-mm height difference. This sisted of thick, gray and off-white medium-fine-grained
rockburst caused the smashing of the north rig tubing and crystalline massive marbles in the Baishan Formation
resuspension of TBM tunneling. (T2b), including fresh and hard calcite, pinstripe biotite,
On November 28, when the drainage tunnel was exca- and other minerals. There were some flat rigid structural
vated for 0.9 m at Stake SK9 ? 283, an extremely intense planes in the rock mass with scratches on the surface. A
rockburst occurred to the spandrel south of Stake rigid fault sub-parallel to the tunnel axis with a NWW-
SK9 ? 285, inducing another extremely intense rockburst orientation was exposed after the ‘‘11.28’’ rockburst. Its dip
in the back range of 28 m. Due to the instant impact of rock angle was about 50. It was straight and smooth with no
shock, all support systems were destroyed, the main beam fillings. All the rock mass beneath the fault collapsed,
of the TBM equipment was broken, and a space of about creating a ‘‘V’’-shaped failure zone with a depth of 7 m, as

123
Case Histories of Four Extremely Intense Rockbursts

Fig. 5 The ‘‘11.28’’ rockburst in the drainage tunnel

shown in Fig. 6. The series of rockburst events in this 4 The ‘‘7.14’’ Rockburst in the Headrace Tunnel #4
tunnel section might be related to the sudden slip of this
fault induced by excavation activities. This is very similar 4.1 Description of the Event
to the fault-slip rockburst that occurred in the headrace
tunnel at the Tianshengqiao II Hydropower Station (Lee To speed up the construction schedule, construction con-
et al. 1996). tractors excavated the branch tunnels from the drainage
In addition, this tunnel section was located in the core of tunnel to the headrace tunnels to increase the number of
a fold syncline with more intense tectonics than the other tunnel faces. The drainage-headrace branch tunnel #2 is
tunnel sections, causing the concentration of local in situ indicated in Fig. 7. Because it was adjacent to the location
stress (Goodman 1989). This was an important external of the ‘‘11.28’’ rockburst, a number of moderate rockbursts
cause of the rockburst. occurred during the excavation of this branch tunnel.

Fig. 6 The major fault exposed

in the ‘‘11.28’’ rockburst in the Rigid fault
drainage tunnel

Drainage tunnel

Buried TBM
Facing west

123
 C. Zhang et al.

Fig. 7 Schematic diagram of

W Drainage-headrace
the location of the ‘‘7.14’’ E
rockburst in the headrace tunnel Advancing direction branch tunnel #2
#4 K9+728 K9+801 K9+810
Headrace tunnel #4

Drainage-headrace
branch tunnel #3 TBM rescue 45m
Drainage tunnel branch tunnel

SK9+283
The tunnel section with the “11.28” rockburst

Rock burst

North sidewall to
Non-excavated section
Spandrel

crown
Excavated section
Multi-beam jumbo drill smashed

Ejected fragments
Tunnel face

Floor
Extremely fierce rockburst
from south sidewall to
Intense rockburst on July 6
spandrel on July 14

South sidewall to
crown
Extremely intense rockburst from Spandrel
south sidewall to arch foot on July 14
Shotcrete on south sidewall
impacted to drop, slabs of
rock mass slipped down
K9+728 K9+742 K9+766 K9+801 K9+810

The headrace tunnel #4 was excavated from the drain- • the fourth sound occurred at 15:08 on July 14, 2010.
age-headrace branch tunnel #2 westward. When it
This rockburst destroyed a multi-beam jumbo drill.
advanced to K9 ? 728 on July 6, 2010, an intense rock-
After the cessation of the rockburst, a failure zone with a
burst occurred on the south sidewall at K9 ? 742–9 ? 766
with a failure depth of up to 2 m, causing a shutdown
(Fig. 8).
On July 13, another intense rockburst occurred at the
section around Stake K9 ? 780–9 ? 790 and at the Rockburst zone

intersection between the headrace tunnel #4 and the TBM

rescue branch tunnel, as shown in Fig. 9. The entrance to
this branch tunnel was excavated only 2 m using the D & B
method.
On July 14, in succession, an extremely intense rock-
burst occurred from the south sidewall to the south arch
foot at Stake K9 ? 728–9 ? 766, when on-site workers
heard four very loud sounds and felt intense vibrations: Multi-beam jumbo drill
Ejected rock fragments
• the first sound occurred at 03:54 on July 14, 2010;
• the second sound occurred at 05:42 on July 14th, 2010; Fig. 8 Photo of the support construction to the intense rockburst in
• the third sound occurred at 14:52 on July 14, 2010; and the headrace tunnel #4 on July 6

123
Case Histories of Four Extremely Intense Rockbursts

Entrance of the TBM rescue branch tunnel

Expansion shell anchored rockbolt

Shotcrete and fragments

Fragments
retained by steel mesh

Fig. 9 Photos of the intense rockburst at the intersection between the headrace tunnel #4 and the TBM rescue branch tunnel

maximum depth of 6.0 m was found on the south sidewall horizontally ejected to the opposite north sidewall (by
at K9 ? 755–9 ? 740 (Fig. 10a). Due to the drastic impact a horizontal distance of 13.4 m). The rockbolts were bro-
of the rockburst, some rock fragments and rockbolts were ken and showed serious distortion (Fig. 10b). Ejected

The maximum failure depth is 6 m Seriously bent rod

Multi-beam jumbo
drill damaged

(a) (b) Ejected fragments

(c) (d)

Fig. 10 Photos after the ‘‘7.14’’ rockburst in the headrace tunnel #4: a failure zone, b ejected fragments and pieces of rod, c rockbolts at the
spandrel and d slabbing to the shotcrete and shallow rock mass at the south arch foot in the tunnel section K9 ? 755–9 ? 770

123
 C. Zhang et al.

fragments were of different sizes and shapes, including this process, if the axial force in a rod exceeds its tensile
slices, slabs, and blocks. The fragments ejected to the strength, it would break.
greatest distance were mainly smaller. The upper rock The impact of a rockburst is similar to the tensile force,
residues were mostly blocks with larger volumes. but it has different loading speeds. Under the force of an
The rock mass and shotcrete on the south arch foot in instant impact, the failure of the interface and the re-dis-
tunnel section K9 ? 755–9 ? 770 were damaged, forming tribution of the shear stress are not gradual. The instant
slabs under the severe vibrations, and slipped down from high shear stress leads to a rapid increase in the axial force,
the surrounding rock mass, causing void in the undamaged and eventually the rod breakage. Therefore, under the
shotcrete on the sidewalls with a depth of about 0.5 m, as direct impact, this type of rockbolt is easy to break, but
shown in Fig. 10d. cannot be pulled out. Figure 9 indicates that the rockbolts
The geological survey indicated that the rock mass was installed in this section had insufficient anchoring force.
fresh and intact, and no major fault was found. This tunnel The factors affecting the anchoring force include the
section is parallel to the drainage tunnel in which the following:
‘‘11.28’’ rockburst took place. As mentioned above, due to
• the shear strength between the cement grout and the
intense geological tectonics, the in situ stress levels in this
rod;
tunnel section are higher than in other sections at the same
• the shear strength between the cement grout and the
depth. This rockburst was classified as strain-type because
surrounding rock mass;
it occurred in an intact rock mass and was caused by high
• the anchorage length;
stresses.
• the mechanical properties of the surrounding rock
mass;
4.2 Analysis of the Support Response
• the material strength of the rod; and
• the material strength of the cement grout.
Before this rockburst occurred, the exposed surrounding
rock mass had been supported. Figs. 8, 9, 10 show that the Therefore, the state of the on-site anchoring system after
expansion shell-anchored rockbolts that had been installed destruction indicates that lack of anchoring force was
were the main reinforcement component in place. The rods mainly due to the low shear strength between the cement
were 6 m in length and spaced by 1.0 m. The faceplates grout and the rod or rock mass. That was related to the
were 150 mm (length) 9 150 mm (width) 9 8 mm strength of the cement grout and the construction quality.
(thickness). The steel fiber nano shotcrete was 8-cm thick. Moreover, because the rock masses broke into small frag-
The integrity of the support structures in different locations ments under high-stress compression, a systematic
after the rockburst indirectly reflected their response anchorage body could not form with the rockbolts. In this
characteristics under the impact of the rockburst. situation, the rockburst impact force could not be trans-
As shown in Fig. 9, about 11 expansion shell-anchored ferred to the rockbolts.
rockbolts enclosed in a white frame remained anchored in As shown in Fig. 10c, most of the expansion shell-
the rock mass, and they were connected to a collapsing anchored rockbolts exposed on the south spandrel were still
megalith through the faceplates. The rockbolts slipped, but completely retained on the tunnel walls. Most of the
were not pulled out. The megalith was supported by the faceplates were still well preserved, and small rock blocks
underlying residue and the construction bench. Although remained hanging on the faceplates. Although slips had
the rockbolts showed a tensile state, no cement grout occurred, the rockbolts still had a certain bearing capacity,
residuals were found on the exposed rods. This indicates mainly because of the anchoring effect of the undamaged
that the cement grout between the rods and rock had been mechanical anchors.
damaged and that the mechanical anchors at the ends had In the Jinping II support system, rockbolts are the main
slipped or been damaged under the impact of the rockburst, component capable of withstanding the impact force. To
diminishing the bearing capacity. maximize the anchoring ability of the bolts, their design
In fact, the expansion shell-anchored rockbolts belong to and construction must be considered carefully. Given that
a type of rigid rockbolt with a very high bearing capacity, the strength of cement grout and construction quality can
but the deformation and energy absorption capacities were meet the design requirements, there are two ways to
not outstanding. Under the tensile load, the increase in the transfer impact force to rockbolts: one is through the sur-
interfacial shear stress led to progressive shear yield on rounding rock mass and the other is through the surface
the interface from the outer anchor end inwards, softening support. The first means of transfer is difficult, as shown in
the interfacial shear strength and causing the maximum Fig. 10c. However, in this case, the impact force on surface
shear stress to migrate to the inner anchor end (Li and supports was not transferred to the bolts. Rather, it was
Stillborg 1999). As a result, the rods were pulled out. In independently supported. This indicates that the strength of

123
Case Histories of Four Extremely Intense Rockbursts

the surface supports and the connection between those

supports and the rockbolts were insufficient. An effective The NWW-oriented rigid
Fractured rock
transfer of force failed to form between the shotcrete, steel mass rusted by closed structural plane
mesh, faceplates, and rockbolts. water seepage

5 The ‘‘4.16’’ Rockburst in the Headrace Tunnel #4

On April 16, 2011, the tunnel face of the headrace tunnel

#4 was excavated westward to Stake K6 ? 010 (Fig. 11).
After the tunnel face was blasted, the tunnels were venti-
lated and the smoke exhausted. After 27 min, workers
heard an intense rockburst inside the tunnel. However, they Facing west
found the tunnel face had not collapsed. A moderate
rockburst had occurred at the south sidewall at
Fig. 12 Middle rockburst on the south sidewall in the drainage
SK5 ? 560–5 ? 540 and SK5 ? 535–5 ? 530 (stake tunnel on April 16
number in the drainage tunnel = the stake number in the
headrace tunnel #4—470) behind the tunnel face of the
drainage tunnel (Fig. 12). The average failure depth was slightly dipping toward the north, as shown in Fig. 13.
0.5 m and new cracks were generated on freshly sprayed Except for the failures on both sides of the arch foot, no
shotcrete on the north sidewall at SK5 ? 560–5 ? 540. significant deformity or damage was found in the rockbolts
Rusty structure planes were found to have developed in the or shotcrete.
rock mass of the drainage tunnel, indicating that self-ini- This section of the headrace tunnel #4 was located
tiated rockbursts could not occur. Fortunately, because this 174 m west of the auxiliary-headrace branch tunnel #3,
rockburst occurred during the ventilation process, no whereas the rockburst in the neighboring drainage tunnel is
casualties to personnel or damage to equipment occurred. located 169 m west of this branch tunnel. It can be con-
In the process of clearing away the broken bits, the floor cluded then that the initiation source of the rockburst was
at K6 ? 025 of the headrace tunnel #4 was found to have in the headrace tunnel # 4 and at least 2 m beneath the
been uplifted with fresh rock outcrops. According to the south arch foot. During the rockburst, the huge release of
field survey, the fresh floor strata in the 30-m range behind energy and severe vibrations caused the moderate rock-
the face was uplifted by about 2 m with a 10-m width, burst in the drainage tunnel.

Fig. 11 Schematic diagram of

the location of the ‘‘4.16’’ W E
rockburst on the floor of the
headrace tunnel #4 Headrace tunnel #3

K6+010 174m 60m

Headrace tunnel #4

The “4.16” extremely intense 169m 45m

rockburst on the floor

Drainage tunnel

Middle rockburst in the drainage tunnel on April 16 Auxiliary-headrace branch

tunnel #3

Advancing direction

Rockburst Non-excavated section Excavated section

123
 C. Zhang et al.

Fig. 13 Floor uplift in the

‘‘4.16’’ rockburst in the
headrace tunnel #4

Original floor
0.5 m, original
pad dregs
1.5 m, the fresh
rock mass on
floor uplift by 1.5
m caused by the
rockburst

North arch
South arch foot foot

Because of the unique location of this rockburst, there rock mass with exposed fissures was hard and intact with
was no severe ejection impact on the rock mass, but the no obvious structural planes.
resulting uplifting of the floor was high and the range was When the rockburst occurred, the floor vibrated
very large. Meanwhile, seismic waves reached the drainage severely, causing a dump truck laden with broken to shift
tunnel 45 m away, causing a moderate rockburst in the and bounce three times. The truck was originally parked
drainage tunnel. This rockburst was graded as ‘‘extremely parallel to the tunnel’s axis, but the rockburst changed the
intense’’. angle of the truck relative to the tunnel’s axis by 30. The
truck was severely damaged, the windshield was shattered
and a number of connecting cylinders destroyed (Fig. 16).
6 The ‘‘2.4’’ Rockburst in the Headrace Tunnel #2 Three workers sitting in the truck were injured by the
violent shakes.
An extremely intense rockburst occurred in the headrace The expansion shell-anchored rockbolts and cement-
tunnel #2 on February 4, 2010, when the tunnel face was at grouted steel rebars had been installed about 5 m in front of
Stake K11 ? 006, as shown in Fig. 14. Before this rock- and behind Stake K11 ? 023. An 80-kN pre-load had been
burst, a number of rockbursts had occurred from the north imposed on the expansion shell anchored rockbolts, and the
sidewall to the spandrel of K11 ? 027–11 ? 046 m with a cement grouting was completed by contractors. Some
local failure depth of 2 m (Fig. 15). Swellex dowels had also been installed at this section as the
According to the on-site construction workers, broken temporary support.
bits were being cleared away after the blasting of the tunnel After the rockburst, a large number of bolts remained
face, ventilation, and exhausting of smoke. After 12 on the rock mass in this tunnel section, but they under-
truckloads of broken bits were removed (with three truck went significant slipping. Some faceplates were bent by
loads remaining), an extremely intense rockburst occurred the impact (Fig. 17). However, systematic connections of
at Stake K11 ? 023. The sound was louder than the noise the support components were lacking; there was extre-
of the explosions during excavation. The rockburst caused mely thin shotcrete, lagged mesh hanging, and small
the ejection of rock mass on the south sidewall and the arch faceplates. As a result, the rockburst impact force at the
foot, and a NWW-oriented structural plane was then surface could not be transferred to the rockbolts, and the
exposed (Fig. 17). fractured rock mass easily broke through the surface
This rockburst also created three fissures in the floor of supports and ejected out from the gap among the rock-
the upper bench. One fissure extended from the south arch bolts. Because of the highly concentrated stress, the
foot to the north arch foot with the strike being approxi- shallow surrounding rock mass was crushed and broken
mately perpendicular to the tunnel’s axis, and cross-tra- before the development of anchoring strength. Therefore,
versing the tunnel with a maximum fissure width of 25 cm the rockbolts had an insignificant effect on the rein-
and an available explored depth of 1.5 m (Fig. 16). The forcement of this part of the rock mass, and so the impact

123
Case Histories of Four Extremely Intense Rockbursts

Fig. 14 The location of the

W E
‘‘2.4’’ rockburst in the headrace
tunnel #2
Headrace tunnel #1 Advancing direction

A number of rockbursts
K11+006 occurred at the north spandrel 60m
Headrace tunnel #2

K11+023 The “2.4” rockburst 60m

K11+060
Headrace tunnel #3

Rockburst

Non-excavated section

North sidewall to
Floor fractures with
local openness at 10 Excavated section
Spandrel

crown
cm and an available
explored depth of 1.5
m

A number of intense rockbursts occurred

from the north sidewall to the spandrel with a
Tunnel face

local failure depth of 2 m

Floor
Surface rock mass from the
south sidewall to the arch
foot was loosened and even
ejected. The drain on the
floor foot was uplifted.

South sidewall to
Spandrel crown

Rock mass from the south spandrel to

the arch foot was ejected with a
maximum depth of about 0.5 m.

K11+006 K11+023 K11+060

force could not be effectively transferred to the rockbolts sprayed with shotcrete cracked and peeled off the sur-
during the rockburst. This was the reason why the rock rounding rock mass (Fig. 18).
mass was rapidly detached from the bolts during the Overall, bulking and cracking were evident in the drains
rockburst. on the south side of the tunnel section (Fig. 18), where the
Influenced by the vibrations from an extremely intense groundwater had been more developed and flowed out of
rockburst at K11 ? 023, a rockburst occurred at the south the tunnel through a 1-m-wide drain. However, because of
arch foot in the tunnel section at K11 ? 023–11 ? 060 m. the rockburst, the drains no longer functioned properly and
Some Swellex dowels were pulled out of the south sidewall the groundwater flowed out through the bottom of the
(at a height of about 4 m). Some cement-grouted steel bulking and cracking rock.
rebars broke and faceplates dropped. The rockbolts On the south sidewall of the tunnel section behind Stake
anchored in the expansion shell slipped. The hanging steel K11 ? 060, serious slabbing occurred due to high stress
mesh support was impacted by collapsing rocks; some were before the rockburst and remote seismic wave during the
deformed and others fell off. A large area that had been rockburst, where significant bulking occurred as well.

123
 C. Zhang et al.

7 Conclusion

Based on the information revealed in the four representa-

tive extremely intense rockbursts, some characteristics of
these rockbursts have been summarized and a few of
conclusions have been drawn as follows:
• The major structural planes generally invisible in the
deep tunnels at the Jinping II Hydropower Station fill an
important role during rockbursts. The resulting destruc-
tion is very sudden and the seismic focus is very deep,
with huge amounts of energy accumulated and released.
The failure intensity is also somewhat higher than that of
rockburst caused by a shallow structural plane.
A number of intense rockbursts occurred from the
• The floor rockburst has special features exposed in the
north sidewall to the spandrel with a maximum
failure depth of 2m. headrace tunnels at the Jinping II Hydropower Station.
However, the failure phenomena are different for the
Fig. 15 The intense rockburst from the north sidewall to the ‘‘4.16’’ rockburst in the headrace tunnel #4 and for the
spandrel in section K11 ? 027–11 ? 046 m in the headrace
tunnel #2 ‘‘2.4’’ rockburst in the headrace tunnel #2. The former
featured an uplift of the floor. Horizontal cracking and
vibration of the floor were observed in the latter.
• Under high-stress conditions, when the installation of
However, these fractured rock masses were still held in rockbolts lags behind tunnel face excavation, the
place by the support system without being ejected or col- rockbolts will not able to effectively reinforce the
lapsing (Fig. 19a). shallow surrounding rock mass because they are often
Serious fracturing also occurred in the rock mass on the crushed and broken. Thus, the impact force cannot be
north sidewall and the spandrel, which were subjected to transferred to the rockbolts through this part of
high stress before the rockburst and to remote seismic surrounding rock mass during the rockburst. Where
waves during the rockburst. This caused the shotcrete to there are no systematic supports, neither the integrity of
crack into plates (Fig. 19b). The faceplates and rockbolts the surface supports nor their ability to transfer the
were non-functional because they were detached from the impact force to the rockbolts can be guaranteed. This is
sidewall and slipped, respectively. why the on-site rockbolts and faceplates must be

Fig. 16 The ‘‘2.4’’ rockburst

on the sidewall of the south arch
foot in section K11 ? 023 in The failure zone caused by the
the headrace tunnel #2 rockburst from the north
spandrel before this rockburst

The failure location of

the rockburst on the
south sidewall

Ejected fragments Due to the intense floor vibrations, the truck

for removing slag shifted and fell down.

123
Case Histories of Four Extremely Intense Rockbursts

Fig. 17 The support

components in the failure zone
on the south sidewall in section
K11 ? 023 in the headrace
tunnel #2 NWW-oriented structural plane

Retained Swelling dowel

Retained expansion shell anchored rockbolts

Fig. 18 The local rockburst on

Steel mesh damaged Expansion shell anchored rockbolt
the south sidewall in section
K11 ? 023–11 ? 060 in the
headrace tunnel #2

Local rock
mass ejected

Drain bulging at the

south arch foot

123
 C. Zhang et al.

Fig. 19 Fractures and bulking of the shotcrete and rock masses: a on the south sidewall in section K11 ? 023–11 ? 060 and b from the north
sidewall to the spandrel

retained on the tunnel walls after any rockburst. More Goodman RE (1989) Introduction to Rock Mechanics, 2nd edn. edn.
attention should be paid to systematization of various Wiley Press, New York
Lee CF, Wang SJ, Yang ZF (1996) Geotechnical aspects of rock
support components in support design and construction. tunnelling in China. Tunn Undergr Sp Tech 11:445–454
• This analysis of four extremely intense rockburst events Li C, Stillborg B (1999) Analytical models for rock bolts. Int J Rock
provides an important reference for understanding the Mech Min Sci 36:1013–1029
initiation conditions and development mechanisms of Ortlepp WD (2001) The behavior of tunnels at great depth under large
static and dynamic pressures. Tunn Undergr Sp Tech 16:41–48
rockburst events. It also lays a foundation for the study Ortlepp WD, Stacey TR (1994) Rockburst mechanisms in tunnels and
of rockburst prevention measures in deep tunnels. shafts. Tunn Undergr Sp Tech 9:59–65
Shan Z, Yan P (2010) Management of rock bursts during excavation
of the deep tunnels in Jinping II Hydropower Station. B Eng
Acknowledgments The authors gratefully acknowledge the finan- Geol Environ 69:353–363
cial support from the National Science Foundation of China under Tang B (2000) Rockburst control using distress blasting. Ph.D.
Grant Nos. 2010CB732006, 51079144, 50979104, and the National McGill University, Montreal
Science Foundation of Hubei Province in China under Grant No. Wu S, Shen M, Wang J (2010) Jinping hydropower project: main
2010CDB10404. The work in this paper was also supported by technical issues on engineering geology and rock mechanics.
funding from the Ertan Hydropower Development Company, Ltd., B Eng Geol Environ 69:325–332
and the important pilot project (for youth talent) of knowledge Zhang C, Zhou H, Feng X (2011a) An index for estimating the
innovation project of the Chinese Academy of Sciences (CAS) (No. stability of brittle surrounding rock mass: FAI and its engineer-
KZCX2-EW-QN115). We are also grateful for the support and ing application. Rock Mech Rock Eng 44:401–414
assistance in the headrace tunnel design and construction from the Zhang C, Feng X, Zhou H, Qiu S, Wu W (2011b) A top pilot tunnel
engineers at the East China Investigation and Design Institute. Special preconditioning method for the prevention of extremely intense
thanks go to Prof. Giovanni Barla and the two anonymous reviewers rockbursts in deep tunnels excavated by TBMs. Rock Mech
for their constructive comments. Rock Eng (published online) doi:10.1007/s00603-011-0199-5

References

Board MP (1994) Numerical examination of mining-induced seis-

micity. PhD thesis, University of Minnesota

123