A shotcrete adhesion test system for

mining applications
B. Seymour, L. Martin, C. Clark, M. Stepan and R. Jacksha
Mining engineer, research mechanical engineer, mechanical engineer, engineering technician
and elctronics technician, National Institute for Occupational Safety and Health, Spokane, WA

R. Pakalnis, M. Roworth and C. Caceres

Mining engineers, University of British Columbia, Vancouver, BC

Abstract

The National Institute for Occupational Safety and Health (NIOSH) is conducting research to develop
safe practices for the use of shotcrete as ground support in underground mines, particularly mines
operating in weak host rock. As part of this research, a rugged, portable direct tensile test system was
developed for measuring shotcrete adhesion strength in underground mines. During the development
of this test system, more than 185 direct tensile tests were conducted with a common, commercially
avail- able macro-synthetic fiber-reinforced shotcrete that was applied to concrete test panels using a
dry mix process and machinery. The average bond strength of the shotcrete to the concrete substrate
typically increased as a function of the shotcretes curing age, ranging from 0.44 MPa (64 psi) after
one day of curing to 1.58 MPa (229 psi) after 90 days of curing. Adhesion strength increased
markedly between one and three days of curing, reflecting a similar trend of increasing shotcrete
tensile strength with cur- ing time. This robust direct tensile test system can improve mine safety by
providing a reliable means of measuring shotcrete adhesion strength and also supplying important
information about the quality of the applied shotcrete and the competency of the underlying rock.

Introduction

Historically, a significant percentage of the injuries and

fatalities that occur in underground mines are caused by falls
of ground (Fig. 1). To protect mine personnel from ground
fall hazards, particularly in underground mines where the
host rock is weak (RMR<40), research is being conducted
by the National Institute for Occupational Safety and Health
(NIOSH) to develop safe practices for the use of shotcrete as
ground support.
Shotcrete is a specially blended, cement-based product
that is pneumatically sprayed at a high velocity on the
exposed surfaces of underground openings to provide
ground support. In underground hard rock mines in the
western U.S., shotcrete is generally used as an integral part
of a ground support system consisting of multiple
components. When ground conditions are poor and the host
rock is weak, as in many of the underground gold mines in
Nevada, extensive ground support is required. In these
situations, shotcrete is typically applied in conjunc- tion with
other ground support elements, such as bolts and mesh, but it
is also used with spiling or cemented rockfill for extremely
weak ground. A more complete explanation con- cerning
the use of shotcrete in mechanized cut-and-fill stopes in
Nevada is provided by Clark et al. (2010). In raveling and
highly fractured ground, shotcrete is mainly used to provide
surface support or skin control between the roof bolts, which
serve as the primary ground support elements. By
supporting the rock near the surface of the mine opening,
shotcrete helps

prevent degradation of the anchorage for the other ground

sup- port components and bridges the span between the rock
bolts, thereby supporting the loose material that typically
causes many of the small ground falls (Bernard, 2008).
Shotcrete holds by adhesion, strengthens the rock by
preventing relative movement at the shotcrete/rock interface
and acts as a super mesh by providing a stiff retaining
component with substantial bending or flexural capacity
(Kaiser and Tannant, 2001).
When shotcrete is used as an integral part of a mines
ground support system, it is important to know the strength
proper- ties of the in-place shotcrete. Besides conventional
strength parameters, such as the shotcretes flexural,
compressive or tensile strength, the adhesion or bond
strength of the shotcrete to the host rock must also be
known, in order to adequately determine the shotcretes
ability to support the immediate ground near the surface of
the mine opening. Consequently, the adhesion strength of
the shotcrete is a necessary parameter for ground support
design.
Researchers have found that shotcrete applied in underground mines primarily fails in adhesion and that this initial
debonding of the shotcrete from the underlying substrate is
followed by a subsequent failure in flexure, as the shotcrete
bends under the weight of additional loading from loose
mate- rial (Fig. 2) (Barrett and McCreath, 1995; Holmgren,
2001). In another study, two basic types of shotcrete failure
modes were identified through a mapping program at the
Kiirunavaara Mine in Kiruna, Sweden (Fig. 3).

of the in-place strength prop- erties of shotcrete, particularly

the bond strength of shotcrete to the host rock, will lead to
improvements in ground support

Figure 1 Underground metal mining injuries by accident

class, 2004-2008 (MSHA).

Figure 2 Flexural failure of shotcrete resulting from

insufficient adhesion strength (after Kuchta,
2002).

Laboratory tests in Sweden have indicated that the

primary mode of failure of a good-quality shotcrete lining on
hard rock is governed by adhesion (Holmgren, 2001). As
discussed by Thomas (2009), the crown of an underground
opening pres- ents the worst condition for shotcrete stability,
because the shotcrete is loaded by its own weight from the
moment it is sprayed. Further research has identified a more
complete list- ing of possible shotcrete failure modes,
particularly for cases where shotcrete is used in conjunction
with roof bolts (Fig. 4).
Good bond strength depends on a number of factors, including proper surface preparation (Kuchta, 2002; Malmgren
et al., 2005), adequate compaction between the shotcrete and
substrate (Brennan, 2005) and also compatibility of the
shotcrete with the host rock. Studies have indicated that the
type of rock mineralogy can affect the bond strength (Hahn
and Holmgren,
1979). For example, the adhesion of shotcrete to weak
geologic formations, such as shale and mudstone, is
frequently poor (Spearing, 2001). Experience has also
shown that shotcrete bond strength can be poor in rock that
is structurally weak in tension or, in other words, rock that is
highly foliated, closely bedded or spalling (Kaiser and
Tannant, 2001; Kuchta, 2002).
As mentioned by Spearing (2001), key elements of a
shot- crete quality control program should include design
compliance for bond, strength and thickness of the sprayed
shotcrete. As a result, the ability to determine the bond
strength or adhesion of the shotcrete is a key component of
mine design and ground control methodology (Norwegian
ConcreteAssociation, 2007). A more thorough understanding

Figure 3 Two basic types of shotcrete failure

modes. A: Fallout of only shotcrete, indicating poor
adhesion; B: fallout of shotcrete and rock ,indicating
zones of weak rock (after Malmgren & Svensson, 1999).

Figure 4 Schematic of shotcrete loading and failure

modes (after Morton et al., 2008; Barrett and McCreath,
1995).

practices, thereby preventing falls of ground and reducing

mine roof fall accidents.

Background

Adhesion strength of sprayed shotcrete is generally determined by a simple pull test known as the tensile bond
strength test (Fig. 5). During this test, a direct tensile load is
applied to a core drilled through the shotcrete into the
underlying sub-

2004; American Concrete Institute, 2007); an adhesion test

method was developed in Sweden (SS 13 72 43), whereby a

Figure 5 Simplified schematic of a tensile bond

strength
test (after ACI 506.4R, 2004).

Figure 6 Average unconfined compression

splitting
tensile strengths for the
concrete substrate
(n=36).

and

strate. As this load is gradually increased, the test core

typically breaks or fails in tension. This tensile failure can
occur in the shotcrete, at the bond surface (interface), in the
substrate or at some combination of these locations. The
adhesion or bond strength of the shotcrete to the substrate
material is determined by measuring the maximum tensile
load at failure. Depending on where the core breaks, the
tensile strength at failure can represent either the actual
adhesion strength of the shotcrete or an assumed lower limit
of the adhesion strength. For example, if the core fails
predominantly at the bond surface, then the bond strength or
adhesion of the shotcrete to the substrate can be determined
directly. However, if the core fails primarily in the shotcrete
or substrate rather than at the bond surface (inter- face), then
the adhesion strength of the shotcrete is inferred to at least
exceed the measured tensile strength of the failed core.
Various test methods have been developed using this
basic test configuration. For example, a standard test was
established for the U.S. concrete industry, in which the
direct tensile load is applied through a steel disk that is glued
to the top of the shotcrete core (American Society for
Testing and Materials,

direct tensile load is applied to a test core using a friction

grip or core sleeve (Nordstrm and Grnds, 2005) and
Canadian experiments have been reported in which shotcrete
is applied over drilled pucks to determine adhesion strengths
(Archibald et al., 1992). In the U.S., commonly specified
direct tensile strength values for shotcrete applied to
properly prepared con- crete substrates range from 0.69 to
1.00 MPa (100 to 145 psi) (Brennan, 2005). In Sweden, the
bond strength is commonly required to be a minimum of 0.5
MPa (73 psi) between shot- crete and rock, 1.0 MPa (145
psi) between different shotcrete layers and 1.5 MPa (218 psi)
for shotcrete applied to repair concrete (Nordstrm and
Grnds, 2005). Values that have been typically reported
for the adhesion strength of shotcrete applied to host rock in
underground mines range from about
0.2 to 1.5 MPa (29 to 218 psi) (Barrett and McCreath, 1995;
Malmgren and Svensson, 1999; Kuchta, 2003; Malmgren et
al., 2005; Saiang et al., 2005; Morton et al., 2008).
Although a number of adhesion test methods have been
developed, no universal procedure has been adopted or used
extensively by the mining industry. Usually, the adhesion
test equipment is unavailable or too expensive, complicated
or fragile for extended use underground. Some of the test
methods are not practical for typical mining conditions,
because they require special surface preparation, gluing or
curing time to allow the shotcrete to gain sufficient strength
before a test can be conducted. As a result, the adhesion
strength of shotcrete is seldom measured in underground
mines, even though this strength parameter plays a major
role in the stability of the shotcrete, particularly during the
initial phase of curing.
As explained by Clark et al. (2010), it is important to
measure the early-age strength characteristics of shotcrete in
order to conservatively determine when the applied shotcrete
is capable of providing ground support and, thus, when it is
safe to re- enter a sprayed area. Adhesion strength is
especially important during this initial phase of curing,
because the shotcrete must be securely bonded to the host
rock and held in place until the applied material can gain
sufficient internal strength to resist further loading. Before
an integrated, comprehensive design approach can be
developed for the use of shotcrete as ground support, more
complete information is needed regarding the shotcretes inplace strength properties, including the bond strength of the
shotcrete to the rock.

Research and development

To develop a practical method of measuring shotcrete

adhesion strength in underground mines, NIOSH researchers
conducted several series of direct tensile tests with shotcrete
applied to concrete test panels. The focus of this
investigation was to develop a suitable test method
incorporating lightweight, portable, robust equipment, along
with simple test procedures, so that mine personnel can
measure the bond strength of the shotcrete that is used to
support their underground workings. Concrete was used for
these tests, to provide a substrate material of known quality
and consistent strength properties, thus eliminating the
influence of many confounding factors that would normally
be present underground, such as varying rock type, geologic
structure, discontinuities, blasting-induced fracturing, loose
material, mud, dust and oil (Fig. 6).
A commercial shotcrete mix commonly used in western
hard rock mines was applied to the panels using a dry mix
process and an Aliva-252.1 shotcrete machine equipped with
a hopper and pre-dampener. For consistency, the same

brand of macro-synthetic fiber-reinforced shotcrete

(Superstick Shot- crete SCAPF) was used for all of the tests,
along with similar

Figure 7 Pull anchors investigated during

development of a direct tensile test system
determining shotcrete adhesion strength.

the
for

preparation methods and spraying procedures. Water was

sprayed on the test panels to clean and moisten the concrete
prior to applying shotcrete.
As explained by Seymour et al. (2010), five separate
series of direct tension tests were conducted using the
adhesion test fixtures or pull anchors shown in Fig. 7. The
first four series of tests were conducted with pull anchors
that were embedded in the shotcrete as it was sprayed on the
concrete test panels. These embedded test fixtures consisted
of an expanded metal anchor and two different-sized metal
washer anchors, which were held in place either manually or
with a mounting grid while the shotcrete was applied.
To address problems encountered while using these
embed- ded test fixtures, an alternative pull anchor was
developed, in which a threaded metal stud was glued in the
shotcrete after it was applied to the concrete substrate. To
center the pull anchor in the test core and align the
longitudinal axes of the pull anchor, test core and pulling
fixture, a drilling method was adapted from the Swedish
friction grip test, whereby three parallel concentric holes are
drilled from a single drill setup
(Fig. 8). The base of the pulling fixture was also redesigned
to seat in the kerf of the outer drill hole.
Utilizing these modifications, a fifth series of adhesion
tests was conducted. The improved drilling method and
revised pull- ing fixture automatically centered and aligned
the pull anchor with the test core and pulling fixture, thereby
limiting eccentric loading. Because the pull anchor is
installed after the shotcrete is applied, the shotcrete
application process is not disrupted, and mine personnel are
not exposed to unsupported ground while establishing the
position of the anchor. Furthermore, because the adhesion
test site is not predetermined, the location for conducting a
test can be specifically selected by the mine staff, such as in
a problem area where poor shotcrete adhesion strength is
suspected. The final version of the shotcrete adhe- sion test
system that was developed for use in underground mines is
shown in Fig. 9.
A total of almost 200 adhesion tests were conducted with
only five test cores failing prematurely, either during core
drilling or while setting up the pulling apparatus. All of
these failures occurred during the first three series of tests,
before centering of the anchor fixture and eccentricity of the
pulling axis were controlled more closely. As a general
rule, the di- rect tensile test measurements became more

Figure 8 Schematic showing a vertical cross-section of

the drill holes for a tensile bond strength test.

consistent as the

Figure 9 Field-expedient direct tensile test system.

testing progressed and the equipment and procedures were

improved. As indicated in Fig. 10, the shotcrete adhesion
strengths determined from these developmental tests ranged
from 0.15 to 2.22 MPa (22 to 322 psi).
One of the primary factors that must be controlled in any
shotcrete investigation is the quality of the applied product,
which is governed to a large extent by the application technique and the skill of the operator (Melbye, 2001; Thomas,
2009). Although measures were taken during these tests to
control the quality of the applied shotcrete, obvious defects
were observed in some of the panels and test cores.
However, all of the results from the completed tests have
been plotted in Fig. 10, including anomalously low strength
values that were

Figure 11 Schematic of direct tensile test system for

determining shotcrete adhesion strength.
Figure 10 Adhesion strength versus shotcrete curing
time for various anchor configurations (n=185).

obviously caused by poor shotcrete quality.

Table 1 provides a statistical summary of the average test
results organized in terms of test series and curing time.
Because direct tensile tests with a composite material are
influenced by a number of factors, there is an inherent
variability in adhesion test results, as indicated by the
coefficient of variation values listed in Table 1. This innate
variability demonstrates the im- portance of quality control
measures, not only for casting the concrete panels, preparing
the interface surface and applying the shotcrete, but also for
operating the adhesion test equip- ment and consistently
following well-defined test procedures.
As mentioned earlier, typical values reported for the
adhesion strength of shotcrete in underground mines range
from about
0.2 to 1.5 MPa (29 to 218 psi). Direct tensile tests
conducted using the epoxy stud pull anchor gave adhesion
strengths that were within a similar range, 0.19 to 1.99 MPa
(28 to 289 psi).

Therefore, the developed test method appears to provide a

credible means of measuring shotcrete adhesion strength.

Conducting a shotcrete adhesion test

This shotcrete adhesion test system consists of readily

available and relatively inexpensive components, primarily a
small stand-mounted core drill and a pulling unit equipped
with a precision pressure gage (Figs. 9 and 11). The
components are rugged and portable and they can be reliably
used to measure the adhesion strength of shotcrete applied to
the surface of an underground opening.
Once a desired test site has been selected, a handoperated rotary percussive drill is used to drill a 16-mm x
51-mm (0.625- in. x 2-in.) hole for anchoring the drill stand.
After installing a 13-mm- (0.5-in.-) diameter threaded stud
and expansion anchor in this hole, the drill stand is leveled
and secured in

Table 1 Average adhesion test results.

Adhesion strength, MPa

Failure surface predominant

Curing

Standard

Coefficient

location, %

Test

time,

Number

Mini-

Maxi-

deviation,

of variation,

series

days

of tests

mum

mum

Range

Avg

MPa

percent

Concrete

Interface

0.97

1.96

0.99

1.56

0.31

20.0

0.0

12.5

87.5

14

0.74

2.22

1.48

1.28

0.65

50.9

0.0

0.0

100.0

46

0.18

1.67

1.49

0.88

0.34

38.6

4.3

34.8

60.9

28

0.15

1.41

1.25

0.79

0.29

36.7

14.3

21.4

64.3

15

0.80

1.59

0.78

1.21

0.25

20.6

13.3

26.7

60.0

15

0.94

1.69

0.74

1.41

0.19

13.8

53.3

6.7

40.0

14

15

1.07

1.71

0.64

1.40

0.19

13.7

53.3

26.7

20.0

0.19

0.91

0.72

0.44

0.28

64.0

0.0

22.2

77.8

0.67

1.43

0.76

1.13

0.28

25.1

0.0

33.3

66.7

0.95

1.53

0.58

1.22

0.20

16.0

0.0

11.1

88.9

14

0.32

1.96

1.64

1.30

0.49

38.0

0.0

33.3

66.7

28

0.75

1.67

0.92

1.46

0.31

21.0

0.0

11.1

88.9

90

1.25

1.99

0.75

1.58

0.22

14.2

0.0

0.0

100.0

Shotcrete

Figure 12 Parallel and concentric drill holes with

extension rod connected to epoxied stud.

Figure 13 Adhesion test specimen showing a tensile

failure surface located predominantly in the concrete
substrate.

position. Three holes are then drilled from this single drill
setup, ensuring that all of the holes are parallel and
concentric (Figs. 8 and 12).
First, an 11.1-mm- (0.4375-in.-) diameter hole is drilled
dry into the shotcrete, using a rotary percussive bit, to a
depth of about 60 mm (2.375 in.), assuming a shotcrete
thickness of
75 mm (3 in.). Next, the hole is cleaned, filled with a quick
setting two-part epoxy adhesive and a 9.5-mm- (0.375-in.-)
diameter pull anchor is inserted. After the epoxy has
initially set or gelled (approx. 15 min), a 102-mm- (4-in.-)
diameter diamond core bit is used to wet-drill a second hole
through the shotcrete, to a depth of about 25-50 mm (1-2 in.)
into the underlying substrate. Finally, a 127-mm- (5-in.-)
diameter diamond core bit is used to wet drill a shallow kerf
for seating the base of the pulling fixture, typically to a
depth of about
3-6 mm (0.125-0.25 in.) depending on the irregularity of the
shotcrete surface (Fig. 12).
After the epoxy has fully set (30-60 min), a threaded
exten- sion rod is connected to the pull anchor with a
coupling nut and the pulling fixture is carefully placed over
the core sample, with the base of its reaction ring positioned
in the kerf of the outer drill hole. The hydraulic hose from
the hand pump is then connected to the loading ram and the
ram is cycled a few times to remove any extraneous air from
the system. Next, a collet and a slip-on, quick-threading
locknut are connected to the threaded extension rod to serve
as a mechanical stop for the pulling fixtures ram. To
conduct a test, the pressure gage is zeroed and an increasing
tensile load is applied to the core sample through a slow and
steady movement of the pump handle until the core breaks.
Test duration varies depending on the tensile strength of the
test core (typically 30 sec to 2 min).
The ultimate tensile force applied to the test core is determined by converting the maximum hydraulic pressure value,
saved on the pressure gages digital display, to the maximum
tensile force acting normal to the cores failure surface. To
simplify analysis of the test results, the tensile force is
assumed
to act in a direction parallel to the longitudinal axis of the

Prior to conducting these tests, the hydraulic pump and

load- ing ram were calibrated in a laboratory test machine
equipped with a certified load cell. The hydraulic
components were tested at several load values over the range
of the rated capacity of the loading ram, thus providing a
direct comparison between the hydraulic pressure reading on
the pumps digital pressure gauge and the corresponding
load or force reading measured by the test machines load
cell. Using a simple linear regres- sion equation obtained
from this calibration procedure, the actual force exerted by
the hydraulic ram can be accurately determined from the
measured hydraulic pressure. Equation
1 shows the simple linear relationship between measured
hydraulic pressure and applied tensile load for the hydraulic
components used in these tests.
Ft = (2.7181) p 10.058
where
and

(1)

Ft = tensile force (lbf)

p = hydraulic pressure (psi)

The maximum tensile stress at failure is then calculated

using Eq. (2).
T = FT / (d2 / 4)
where
and

(2)

T = ultimate tensile stress (psi)

FT = ultimate tensile force (lbf)
d = diameter of test core (in.)

To determine whether the maximum tensile stress

calculated above is a direct measure of the shotcretes
adhesion strength or an assumed lower limit of its bond
strength, the test core should be examined along with the
bottom of the drill hole to identify the location of the tensile
failure surface. The failure location is typically recorded as
a percentage of the shotcrete, interface and substrate that are
exposed on the tensile failure
surface (Fig. 13). The overall depth of the 102-mm- (4-in.-)

test core and the area of the failure surface is assumed to be

equivalent to the cross-sectional area of the test core.

diameter drill hole should also be noted, along with the

length of the test core and the thickness of the shotcrete layer
so that

Figure 14 Adhesion strength versus curing time for

direct
tensile tests using the epoxy stud fixture (n=54).

Figure 15 Comparison of adhesion strength and estimated shotcrete tensile strength.

drilling depth(s) for the pull anchor and/or test core can be
adjusted, if necessary, for further tests.

some measure of its compressive strength, it is more

appropriate to compare adhesion with tensile strength,
particularly when a direct tensile test is used to determine
adhesion strength. In Fig. 15, the results of the epoxy-stud
adhesion test series are plotted with shotcrete tensile
strengths estimated from unconfined compression and
splitting tensile tests with cored

Discussion of test results

Results of direct tensile tests using the epoxy-stud pull

anchor are shown in Fig. 14 for macro-synthetic fiberreinforced
shotcrete applied to concrete test panels. The adhesion tests
were conducted after the shotcrete had cured for 1, 3, 7, 14,
28 and 90 days. Average adhesion strength values increased
with shotcrete curing age and ranged from 0.44 MPa (64 psi)
after one day of curing to 1.58 MPa (229 psi) after 90 days
of curing. The range of these results are comparable to
previ- ously published values for the adhesion of shotcrete to
concrete test panels (Malmgren and Svensson, 1999; Kuchta,
2002) and are within the normal range of bond strengths
specified for shotcrete applied to concrete substrates
(Brennan, 2005; Nordstrm and Grnds, 2005).
Measuring adhesion strength in terms of shotcrete curing
time helps identify when the bond strength of the shotcrete is
sufficient to support more than its own weight. In contrast
with other methods that require a relatively stiff shotcrete in
order to conduct adhesion tests (Malmgren et al., 2005),
adhesion strengths were successfully measured using the
epoxy stud pull anchor after only one day of shotcrete curing
time. These tests indicated that the shotcretes average
adhesion strength increased markedly between one and three
days of curing, from
0.44 to 1.13 MPa (64 to 164 psi); or, in other words, from 28
to
72% of the average 90-day adhesion strength (Fig. 14 and
Table
1). As previously noted, it is important to measure shotcrete
adhesion strength during the initial phase of curing, in order
to identify safe re-entry times for underground workings that
have been recently sprayed with shotcrete.
Because it is difficult to sample and test a weakly consolidated material such as freshly sprayed shotcrete, the strength
properties of early-age shotcrete are usually related to some
direct or indirect measure of its compressive strength.
Adhesion strength gain with shotcrete curing time has been
identified in other studies, and this trend has been related to

samples of similar shotcrete. The concrete test panels had

an increase in the shotcretes compressive strength over time
(Malmgren et al., 2005; Bernard, 2008).
Rather than comparing the bond strength of the shotcrete
to

generally cured for over 28 days before they were sprayed

with shotcrete, so the average tensile strength of the concrete
ranged from at least 4.8 to 6.2 MPa (700 to 900 psi) when
the adhesion tests were conducted (Fig. 6). Although
adhesion strength closely follows the strength gain trend
estimated for the tensile strength of the shotcrete, the results
of the adhesion tests are much lower in magnitude (Fig. 15).
These lower strength values are at least partially caused
by inherent differences between the two types of tests and
the complexity of testing composite samples. During a
direct tensile test, failure occurs at the weakest element in
the test specimen. Unlike tests such as the flexural test, the
entire volume of a direct tensile test specimen is subjected to
the maximum stress; therefore, the probability of a weak
element occurring in the test specimen and influencing the
test results is relatively high (Neville, 2009). For a direct
tensile test with a composite material, this issue is
compounded because the weakest element can occur in any
of the individual materials or at their interfaces. Therefore,
in a shotcrete adhesion test, a tensile failure can occur in the
shotcrete, at the bond interface (contact surface), in the
substrate, or in a combination of these locations. For
concrete, splitting tensile tests provide more uniform results
and give strengths that are 5 to 12% higher than those
obtained from direct tensile tests (Neville, 2009). Further
testing is needed to clearly define the relationship be- tween
the adhesion strength at the bond surface and the tensile
strength of the shotcrete.
If a high-quality shotcrete is applied using correct
procedures and the bond surface is prepared properly (i.e.,
clean and free from loose materials, mud, dust or oil, with
sufficient rough- ness and moisture to permit a good bond),
the location of the tensile failure should depend on the
relative difference between the tensile strength of the
shotcrete and that of the substrate. Results reported for
shotcrete adhesion tests in underground

of direct tensile tests were conducted with this equipment to

measure the adhesion strength of macro-synthetic fiberreinforced shotcrete applied to concrete test panels using a
dry mix process and machinery.

Figure 16 Predominant location of the tensile failure

surface for shotcrete adhesion tests with concrete panels
(n=185).

mines indicate that the majority of the tensile failures occurred in the host rock or at the contact surface with the host
rock (Kuchta, 2002; Clements et al., 2004; Malmgren et al.,
2005). In contrast, the majority of the tensile failures in
these tests occurred in the shotcrete, more than likely
because the shotcrete had not yet developed sufficient
strength with curing time to match the tensile strength of the
concrete (Figs. 6 and
15). As the test series progressed, the concrete, shotcrete
and test procedures were controlled more closely, improving
the consistency of the test results, as shown in part by the
values listed in Table 1. This was also reflected in the
predominant locations of the tensile failure surfaces, which
averaged 64% in the shotcrete, 23% at the interface and 13%
in the concrete for the entire test series and 81%, 19% and
0%, respectively, for the final epoxy stud test series (Fig. 16
and Table 1).
According to Swedish Standard SS 13 72 43, the result of
a direct tensile test is reported as adhesion strength if more
than
80% of the tensile failure surface is located at the interface
(bond surface). Otherwise, the test result represents a lower
limit for adhesion strength. In other words, the actual
adhesion strength at the bond surface is larger than a tensile
failure that occurs at some other location (shotcrete, host
rock or some combination of locations).
Using this
criterion, only 4% of these tests failed at the bond surface
(eight of the 185 total tests). This restriction produced
significantly lower adhesion strengths than the average
adhesion strength curves shown in Figs. 10 and 14-16. By
averaging all the test results, a more consistent and
representative value was reported for adhesion strength,
because anomalously low values for tensile failures at the
bond surface were not given an undue weight in the data
analysis.

Conclusions

NIOSH researchers have developed a rugged, portable

direct
tensile test system, consisting of readily available,
inexpensive components that can be used in conjunction
with simple test procedures to determine the adhesion
strength of shotcrete in underground mines. Several series

The average bond strength of the shotcrete to the concrete

substrate typically increased as a function of the shotcretes
curing age, ranging from 0.44 MPa (64 psi) after one day of
curing to 1.58 MPa (229 psi) after 90 days of curing. A substantial increase in adhesion strength was noted between one
and three days of shotcrete curing time (approximately a
44% increase in terms of the shotcretes average 90-day
adhesion strength). The increase in adhesion strength as a
function of curing time appeared to reflect a similar trend of
increasing tensile strength in the shotcrete with curing time.
This robust direct tensile test system provides a reliable
means of measuring the adhesion strength of freshly applied
shotcrete that has cured for as little as one day. Because the
epoxy stud pull anchor is installed after the shotcrete has
been applied, the test procedures do not interfere with the
shotcrete application process, and the adhesion test can be
conducted at a desired location rather than at a site which is
predetermined by the test method. As a result, this adhesion
test system can improve mine safety by providing a viable
means of evaluating the in-place strength and quality of the
applied shotcrete, as well as the tensile strength and
competency of the underlying rock.

Disclaimer

The findings and conclusions presented in this document

have not been formally disseminated by the National
Institute for Occupational Safety and Health and should not
be construed to represent any agency determination or
policy. Mention of any company name or product does not
constitute endorse- ment by NIOSH.

Acknowledgments

The authors thank James Schumacher and Paul Meyer with

Thiessen Team USA, Elko, NV; Vince Mendive with
Thiessen Team USA, Big Timber, MT; and Mark Mudlin
with BASF, Master Builders, Elko, NV for their technical
support regard- ing commercial shotcrete mixes and
application practices. We also thank Justin Deemer and
Rocky Wegland with Westech Automation in Spokane, WA,
for their assistance developing a base for the adhesion test
pulling fixture.

References

American Concrete Institute, ACI 506.4R-04, 2007, Guide for the Evaluation of
Shotcrete, ACI Committee 506.
American Society for Testing and Materials, ASTM C1583/C1583M - 04, 2004,
Standard Test Method for Tensile Strength of Concrete Surfaces and the
Bond Strength or Tensile Strength of Concrete Repair and Overlay Materials
by Direct Tension (Pull-off Method).
Archibald, J., Mercer, R., and Lausch, P., 1992, The evaluation of thin polyurethane surface coatings as an effective means of ground control, Rock
Support in Mining and Underground Construction: Case Studies, Kaiser &
McCreath (eds.), pp. 105-115.
Barrett, S.V.L., and McCreath, D.R., 1995, Shotcrete support design in blocky
ground: towards a deterministic approach, Tunnelling and Underground
Space Technology, Vol. 10, No. 1, pp. 79-89.
Bernard, E., 2008, Early-age load resistance of fibre reinforced shotcrete linings, Tunnelling and Underground Space Technology, Vol. 23, pp. 451-460.
Brennan, E., 2005, Testing shotcrete for bond, Shotcrete Magazine, Winter,
pp. 18-19.
Clark, C.C., Stepan, M.A., Seymour, J.B., and Martin, L.A., 2010, Report on
early strength performance of modern day weak rock mass shotcrete
mixes, SME Preprint 10-138, SME Annual Meeting, Phoenix, AZ.
Clements, M.J.K., Jenkins, P.A., and Malmgren, L., 2004, Hydro-scaling an
overview of a young technology, Shotcrete: More Engineering Developments, Bernard (ed.), A.A. Balkema, Leiden, pp. 89-96.
Hahn, T., and Holmgren, J., 1979, Adhesion of shotcrete to various types of
rock surfaces and its influence on the strengthening function of shotcrete
when applied on hard jointed rock, in Proceedings of the 10th International
Congress on Rock Mechanics, Montreux, Switzerland, pp. 431-439.
Holmgren, J., 2001, Shotcrete linings in hard rock, Underground Mining
Methods, Engineering Fundamentals and International Case Studies, Hustrulid & Bullock (eds.), Society for Mining, Metallurgy and Engineering Inc.,

Littleton, CO, pp. 569-577.

Kaiser, P.K., and Tannant, D.D., 2001, The role of shotcrete in hard-rock
mines, Underground Mining Methods, Engineering Fundamentals and
International Case Studies, Hustrulid & Bullock (eds.), Society for Mining,
Metallurgy and Engineering Inc., Littleton, CO, pp. 579-592.
Kuchta, M.E., 2002, Quantifying the increase in adhesion strength of shotcrete
applied to surfaces treated with high-pressure water, SME Preprint 02-35,
SME Annual Meeting, Phoenix, AZ.
Kuchta, M., 2003, Shotcrete adhesion tests at NIOSHs Lake Lynn Laboratory
(TR4-2), Project Report.
Malmgren, L., and Svensson, T., 1999, Investigation of important parameters
for unreinforced shotcrete as rock support in the Kiirunavaara Mine,
Sweden, Proceedings of the 37th U.S. Rock Mechanics Symposium,
Vail. Amadei, Kranz, Scott, & Smeallie (eds.), A.A. Balkema, Rotterdam, pp.
629-635.
Malmgren, L., Nordlund, E., and Rolund, S., 2005, Adhesion strength and
shrinkage of shotcrete, Tunnelling
and
Underground Space
Technology, Vol. 20, pp. 33-48.
Melbye, T., 2001, Sprayed concrete for rock support, MBT International
Underground Construction Group, 9th ed., Division of MBT, pp. 186-193.
Morton, E., Thompson, A., and Villaescusa, E., 2008, Static testing of shot-

crete and membranes for mining applications, Proceedings of the 6th

International Conference on Ground Support, Cape Town, Johannesburg:
SAIMM, pp. 195-212.
Neville, A.M., 2009, Properties of Concrete, 4th ed., Pearson Prentice Hall,
Harlow, England.
Nordstrm, E., and Grnds, J., 2005, Effective in-place testing of bond
strength with Swedish friction grip, Shotcrete Magazine, Summer, pp.
18-19.
Norwegian Concrete Association, 2007, Sprayed Concrete for Rock Support Technical Specification, Guidelines, and Test Methods, Publication Number 7.
Saiang, D., Malmgren, L., and Nordlund, E., 2005, Laboratory tests on shotcreterock joints in direct shear, tension and compression, Rock Mech Rock
Eng, Vol. 38, No. 4, pp. 275-297.
Seymour, B., Martin, L., Clark, C., Stepan, M., Jacksha, R., Pakalnis, R.,
Roworth, M., and Caceres, C., 2010, A practical method of measuring
shotcrete adhe- sion strength, SME Preprint 10-137, SME Annual
Meeting, Phoenix, AZ.
Spearing, S., 2001, Shotcrete as an underground support material,
Underground Mining Methods: Engineering Fundamentals and International
Case Studies, Hustrulid & Bullock (eds.), Society for Mining, Metallurgy
and Engineering Inc., Littleton, CO, pp. 563-568.
Thomas, A., 2009, Sprayed Concrete Lined Tunnels, Taylor & Francis, New
York, NY.