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

net/publication/283347943

Corrosion assessment of ground support systems

Conference Paper · September 2004

CITATIONS READS

17 2,250

4 authors, including:

Rhett Hassell Ernesto Villaescusa

Northern Star Resources WA School of Mines - Curtin University Australia
10 PUBLICATIONS   110 CITATIONS    144 PUBLICATIONS   1,136 CITATIONS   

SEE PROFILE SEE PROFILE

Brian Kinsella
Curtin University
68 PUBLICATIONS   1,264 CITATIONS   

SEE PROFILE

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

Application of electrochemical noise for carbon steel corrosion monitoring View project

Mining at Great Depth View project

All content following this page was uploaded by Rhett Hassell on 01 November 2015.

The user has requested enhancement of the downloaded file.

Corrosion assessment of ground support systems
R. Hassell, E. Villaescusa and A.G. Thompson
Curtin University of Technology, WA School of Mines, Kalgoorlie, Western Australia
B Kinsella
Curtin University of Technology, WA Corrosion Research Centre, Perth, Western Australia

ABSTRACT: A Corrosion Assessment System has been developed to systematically measure and record
various parameters associated with the ground water and atmosphere in mines. The measurements are
complemented by qualitative and quantitative assessments of the condition of reinforcement and support
systems relative to their time of installation. These measurements and assessments are recorded. Data have
been collected at eight mine sites throughout Australia. These sites exhibit a wide range of ground water
qualities. A correspondingly wide range of observed degrees of corrosion for similar ground control systems
has been established.

1 INTRODUCTION 2 REVIEW OF EXISTING CORROSIVITY

CLASSIFICATION SYSTEMS
The corrosion of rock reinforcement and support
systems and the effect on their load bearing A wide variety of corrosivity classifications are
capacities has not been widely researched and is available in publications; however most are specific
generally not well understood. While there is much to certain environments and conditions. The
literature relating to the phenomenon of corrosion it following is a review of the main classifications that
is not always applicable to the hard rock may be useful in the assessment of ground control
underground mining environment. systems in underground mines.
The purpose of rock support and reinforcement is
to maintain excavations safe and open for their
intended lifespan. The types of support and 2.1 Soil corrosion classification
reinforcement required in a particular location are The probability of corrosion of metals in soil has
dependent upon several factors that include the been widely researched with numerous
strength of the rock mass, the geometry of the classifications. Parameters that are usually examined
excavation, the stresses present in the rock, the include soil type, resistivity, water content, pH,
blasting practices, and the weathering and corrosion buffer capacity, sulphides, neutral salts and
processes applicable at the site (Villaescusa 1999, sulphates, presence of groundwater, the horizontal
Windsor & Thompson 1993). Sundholm (1987) and vertical homogeneity (i.e. difference in soil
suggests that corrosion is one of the major factors structure), and the electrochemical potential
determining which reinforcement type can be used (http.njuct.edu.cn/MatWeb/soil/table603.htm).
as permanent support. The use of soil corrosivity classification for the
Corrosion of reinforcement is a poorly underground hard rock environment is not
understood mechanism and no systematic field study recommended because of the obvious different
has ever been undertaken within the underground environments in which the corrosion takes place.
mining industry in Australia. The WA School of
Mines is currently conducting a 3-year research
project into corrosion in underground mining. This 2.2 Atmospheric corrosion classification– ISO
study aims to determine the main corrosion The ISO 9223:1992 standard classifies the
mechanisms affecting rock reinforcement elements corrosivity of an atmosphere based on measurements
used underground and to assess the effectiveness of of time of wetness (TOW), and pollution categories,
existing corrosivity classifications. sulphur dioxide (SO2) and airborne chlorides. The
corrosivity of the atmosphere is divided into five
categories ranging from very low to very high with
corresponding corrosion rates for carbon steel and 2.4 Underground hard rock corrosion
zinc. classification– Li & Lindblad (1999)
The TOW is a key parameter as it determines the
Li & Lindblad (1999) have proposed two corrosivity
amount of time the electrolyte layer is present on the
classifications for the underground environment with
surface of the metal. For this standard it has been
relation to the corrosion of steel rock bolts. The first
defined as the time period during which the relative
classification is for wet rock conditions and the
humidity is in excess of 80% and the temperature is
corrosion-related parameters used are pH, dissolved
above 0ºC. This standard is widely used to classify
oxygen, resistivity, ambient temperature, rock mass
atmospheric corrosion potential in many different
quality and precipitation of calcium carbonate. The
environments; however, it often needs to be
second classification system is for dry rock
modified and calibrated to that specific environment.
conditions using the following parameters;
deposition rate of sulphur, nitrogen oxides and
2.3 Groundwater corrosion classification- DIN chloride, as well as the relative humidity and
ambient temperature.
The German DIN 50929 classification assesses the
Table 2 shows the Li & Lindblad (1999)
corrosive potential of water based on the flow (if the
classification for wet rock conditions using the same
location is submerged), the chloride (Cl-) and sulphate
Australian mines analysed in Table 1. The majority
(SO42-) content, acidity, calcium ion (Ca2+) content, and
of the mine groundwater is rated (Wwet) in the very
pH. Each parameter is given a positive or negative severe corrosion category, with a large weighting
numerical rating based on the effect they have on the being given to temperature. This classification has
corrosivity and then summed to obtain the probability of been developed for European groundwater and
corrosion. The more negative the number, the more appears to place too much emphasis on factors that
corrosive the water. The classification is separated into are not as relevant to Australian groundwaters.
four corrosivity levels with the most corrosive group
having a W1 less than –8. Table 2. Li & Linblad (1999) corrosion classification ratings
Table 1 shows the DIN values for various mine for eight Australian mine groundwaters.
groundwaters collected during this project within Mine Wwet Corrosion Corrosion Rate
Australia. W1 values are for steel, while WL values Site Description (mm/yr)
are for zinc. All the mine groundwaters have W1 1 8.6 Severe 0.15-0.30
values less than –8 and are placed in the highly 2 6.8 Severe 0.15-0.30
3 12.4 Very Severe >0.30
corrosive category. 4 11.9 Very Severe >0.30
5 10.6 Very Severe >0.30
Table 1. DIN corrosion classification values for eight 6 18.6 Very Severe >0.30
Australian mine groundwaters. 7 16.4 Very Severe >0.30
Mine 8 11.2 Very Severe >0.30
TDS pH Cl- SO42- Ca2+ DIN value
Site
mg/l mg/l mg/l mg/l W1 WL
1 230,000 5.80 180,000 24,000 310 -24 -14 2.5 Saturation indices
2 5,400 8.12 1,570 1,287 187 -11 -5 Water saturation indices relate the solubility of
3 13,000 8.80 7,700 170 430 -17 -7 dissolved ions to their tendency to precipitate. One of
4 44,000 7.40 23,000 6,000 910 -12 -7 the most notable is the Langelier Saturation Index
5 120,000 6.40 69,000 7,800 1,600 -17 -8 (LSI), which is an indicator of the degree of
6 6,100 7.70 1,500 800 300 -11 -5 saturation of water with respect to calcium carbonate.
7 28,000 5.90 16,000 1,500 2,700 -15 -23 The LSI is defined as the difference between the
8 160,000 7.00 93,000 3,100 4000 -21 -6 measured pH and the pHs at saturation in calcite or
calcium carbonate. If LSI is negative, then there is no
The indications are that the DIN classification potential to scale, as the water will dissolve the
system does not distinguish between widely calcium carbonate. If the LSI is positive, then scale
different mine waters, especially for the high total can form and calcium carbonate precipitation may
dissolved solids (TDS) groundwater generally found occur protecting the metal from corrosion attack. The
in Australian underground mines. majority of mine groundwaters showed the potential
for scale to form (Table 3). However, the general
consensus is that these types of indices are indicators
of the formation of scale and not of corrosivity and
they do not include other important corrosivity
parameters such as dissolved oxygen, temperature,
dissolved ions and water velocity.
Table 3. LSI ratings for eight Australian mine groundwaters. ground support is subjected. Because environmental
Mine No LSI Potential to Scale conditions in underground mines are never
1 -2.22 No potential homogenous and are constantly changing, only
2 0.21 Precipitation may occur
3 0.43 Precipitation may occur approximations can be made for classifying the
4 0.62 Precipitation may occur environment and such approximations must be
5 -0.77 No potential constantly reviewed.
6 -0.14 Borderline scale potential Essentially, corrosion in an underground
7 -1.11 No potential environment can be classified into two controlling
8 -0.01 Borderline scale potential
environments. Atmospheric corrosion, which can be
defined as the corrosion of materials exposed to air
2.6 Assessment of current corrosivity classifications and its pollutants. The second type of environment is
the combination of groundwater and the atmosphere.
The use of soil corrosivity classifications in the hard
Corrosion due to the presence of groundwater is
rock underground environment is not feasible due to
significantly more aggressive than atmospheric
the different environments in which the corrosion
corrosion and only occurs when the reinforcement
takes place.
and support are in direct contact with groundwater.
The ISO 9223 and the dry corrosion classification
The corrosion itself may be due to a combination of
of Li & Lindblad (1999) have similar input
groundwater and the atmosphere variables; however
parameters that appear to work reasonably well for
when groundwater is present it is usually the
determining the atmospheric corrosion environment
controlling factor to such an extent that atmospheric
in underground mines. A difficulty arises however
variables are rarely considered.
when determining atmospheric variables that are
constantly changing as mining progresses.
3.1.1 Atmospheric variables
The most destructive and complicated corrosion
The majority of reinforcement and support in
environment to predict is that related to the influence
underground mines is affected by corrosion from
of groundwater. Both the DIN and Li & Lindblad
atmospheric variables. The primary variable is the
(1999) classifications examine the corrosive
time of wetness (TOW), which is a function of
potential of water and generally rate Australian mine
temperature, relative humidity and the surface shape
groundwater as highly corrosive. These
of the metal. The TOW refers to the period of time
classifications have been developed in Europe where
during which the atmospheric conditions are
different groundwater conditions from Australia
favourable for the formation of a surface layer of
exist. Australian groundwaters, especially in the
moisture (electrolyte) on a metal, which is integral to
southern Yilgarn Craton, have high temperatures and
the corrosion process. The level of atmospheric
extremely high salinity. The mildest groundwater
contaminants often controls the rate of atmospheric
collected in Australia rates in the worst category for
corrosion. Pollutants such as sulphur dioxide (SO2)
the DIN classification and the second worst for the
mix with the electrolyte producing sulphuric acid.
Li & Lindblad (1999) classification. It is clear that
Other primary contaminants include atmospheric
these classifications were not developed for
chlorides, nitrogen compounds and dust particles. The
groundwaters with high TDS and temperatures, as
process of blasting and the use of diesel equipment
found in many Australian underground mines.
are the major sources of these contaminants, which
are further concentrated due to the restricted
ventilation system, especially at return airways.
3 CORROSION ASSESSMENT SYSTEM
3.1.2 Groundwater variables
A major objective of a Corrosion Assessment
Groundwater flowing from the rock mass is
System is to collect data in a systematic way that
present in most underground mines, however it
will enable observed corrosion of reinforcement and
interacts with only a small percentage of
support systems to be related to the various
reinforcement and support systems. The elements
conditions existing within the underground
are generally not submerged, but are located in a
environment. It should also provide an assessment of
water/atmosphere interface.
the current condition of the reinforcement and
Groundwater variables that influence the
support systems in various areas of the mine.
corrosivity of waters include:
- pH – the control the pH value exerts on the
3.1 Corrosion potential of a rock mass corrosivity of a system depends to a large degree
on the solubility of the corrosion product
All corrosion of support and reinforcement in
(usually the oxide) formed on the metal surface.
underground mines is due to contact with aqueous
Under normal conditions, the corrosion rate of
solutions. The actual rate and form of corrosion is
steel is independent of pH values between 4.5
controlled by the different environments to which
and 10.
- Dissolved Oxygen – dissolved oxygen is accelerate corrosion, however, such occurrences are
probably one of the most important factors that considered to be localised.
influence metal corrosion in a mine water
system. It significance lies in the fact that it is the
most common cathodic reactant present in 3.2 Condition of reinforcement and support
natural waters. The conditions of reinforcement and support
O2 + 2H2O + 4e- → 4OH- systems are dependent upon their age, the corrosive
The concentration of oxygen in seawater at potential of the environment and the types of
ambient temperatures is approximately 8 mg/l. reinforcement and support used. It is often difficult
Dissolved salts and temperature often control the to assess the condition of reinforcement, especially
oxygen concentration. for fully grouted and resin encapsulated elements.
- Water Velocity – as velocity causes a mass flow During this research project, observing the internal
of oxygen to the surface of the metal, and condition of ungrouted friction rock stabilisers with
corrosion is dependent on oxygen concentration, the use of a borehole camera has had some success,
the corrosion rate will increase with an increase but only the internal section of the bolt can be
in water velocity. Also high water velocity can observed.
cause erosion of corrosion products decreasing Assessing the condition of the surface support
their protective nature. and extrapolating to include the condition of the
- Chlorine and Sulphates – aggressive anions in reinforcement is not recommended. The excavation
the water such as chlorides and sulphates surface and the internal rock mass are two separate
increase the corrosion rate by lowering the environments with different rates and forms of
electrical resistivity of corrosion cells and corrosion. An example is given in Figure 1.
playing a significant role in the penetration and
breakdown of any protective film that might
have formed on the metal surface.
- Scaling – precipitation of salts such as
carbonates may produce a film or scale that
slows or prevents corrosion. Calcareous deposits
are sometimes formed at the cathodic sites.
However, the high total dissolved salts
encountered with mine groundwater greatly
impedes the ability of corrosion reducing
carbonate scaling by forming different chemical
deposits that are porous and less effective as a
protective coating.

3.1.3 Rock mass quality variables

The rock mass quality indirectly influences the Galvanised Friction Bolt 0.7m
corrosive potential of a mine environment. The rock
mass structures are primarily important as they
provide a conduit for ground and fill water flow.
Major geological structures such as faults and shears
allow the flow of groundwater into the mine from
surrounding aquifers. The extent of the area affected
by the water is increased by the presence of
interconnected, dilated joint sets, which allow the
ground water to travel and dissipate at significant
distances from the source. Opening of the joints can
occur during the initial excavation or subsequently at
a later date due to blasting or stress changes.
The mineralogy associated with the different rock
masses is not expected to influence the corrosion
Figure 1. Galvanised friction bolt showing no corrosion of
potential of an environment. The minerals are plate and severe internal corrosion located about 0.7 m from
generally inert and do not enhance the corrosion the collar.
processes in any major way. An exception to this
rule is for sulphide minerals. These reactive minerals Currently as part of this research project an
oxidise, creating very localised acidic conditions as overcoring drill rig is being developed at the
well as possibly creating an electrochemical Western Australian School of Mines (WASM) to
corrosion cell with the rock reinforcement. This may overcore reinforcement elements in-situ. This will
provide detailed information on how, where and be significantly higher than the rate of uniform
what type of corrosion is occurring on reinforcement corrosion.
elements, as well as the effectiveness of Pitting is the most common form of corrosion of
encapsulation by resin and cement grouts. steel in concrete and work by Ranasooriya et al.
(1995) identified pitting corrosion as the most
common of all types of corrosion on collected
3.3 Mechanisms of corrosion
samples of friction rock stabilisers.
The nature of the corrosion process will depend on
the interaction between a material and its 3.3.3 Stress Corrosion Cracking (SCC)
environment. When specific conditions in the Stress corrosion cracking (SCC) is used to
material and the environment occur characteristic describe service failures in materials that occur by
forms of corrosion result. Recent data collection at slow environmentally induced crack propagation
several Australian underground mines have (Jones & Ricker 1990). The cracking is induced
identified three main forms of corrosion, which are from the combined influences of tensile stress and a
supported in part by the work reported by corrosive environment and crack propagation
Ranasooriya et al. (1995) and Gamboa & Atrens proceeds slowly until it reaches its critical length
(2003). The main types of corrosion observed on and rapid fracturing occurs suddenly.
reinforcement and support are uniform or general The required tensile stresses may be in the form
corrosion, localised corrosion in the form of pitting, of directly applied stresses or in the form of residual
and stress corrosion cracking. These are not the sole stresses. Usually, most of the surface remains
types of corrosion observed in the underground unattacked, with fine cracks penetrating into the
environment, but are considered the most common material and thus is very difficult to detect and
and detrimental. damage is not easily predicted. (www.corrosion-
doctors.org/Forms/scc.htm). Crosky et al. (2002)
3.3.1 Uniform or General Corrosion Attack have found that steel with low fracture toughness is
Uniform or general corrosion occurs when the particularly susceptible to this type problem.
anodic and cathodic areas on the metal surface Recent experimental results from Gamboa &
change position continuously. As a result, the Atrens (2003) on SCC of rock bolts indicates that SCC
corrosion that occurs at the anodic areas is spread begins when sheared by moving rock strata and that
evenly over the surface of the metal. Other common the critical crack length can be of the order of only a
forms of corrosion occur when this exchange of few millimetres for rock bolts. However, laboratory
anodes and cathodes is prevented in some way, so testing indicates that SCC occurs in low pH waters
that the corrosion becomes localised. Commonly unlike most Australian mine groundwaters, which
uniform corrosion occurs when most metals are have an almost neutral pH. However, such pH may be
exposed to the atmosphere or submerged in water. reached locally due to the presence of bacteria or
The rate of corrosion often declines with time due to sulphides or by previous leaching mining methods
the build up of corrosion products on the surface of contaminating the groundwater.
the metal (CISA 1994).
3.4 Data collection sheet
3.3.2 Localised Pitting Corrosion
Pitting corrosion is the highly selective attack of To simplify the collection of data and to ensure that
passive metals at defects in the passive oxide layer. all possible variables affecting corrosion were
The corrosion attack is in the form of pits, usually identified, a Corrosion Assessment System was
covered by the corrosion products or remnants of the developed that systematically identified the
original protective layer. Pitting is considered to be following information for collection at mine sites:
more dangerous than uniform corrosion because it is
difficult to detect, predict and design against - Site Specification
(Roberge 2000). o Excavation type and age
Pitting usually occurs in solutions containing - Rock Mass
chlorides, and becomes autocatalytic (i.e. it o Intact Rock
stimulates itself once initiated). This occurs when a o Geological Discontinuities
charge imbalance occurs as the anodic reaction - Environmental Factors
becomes localised within the pit, and as cations are o Groundwater
liberated into the pit solution. As a result, anions o Atmosphere
such as chloride ions diffuse from the bulk solution o Stresses
to the pit to equalise the charge. The presence of - Ground Control Scheme
chlorides further stimulates corrosion by forming o Reinforcement Systems
soluble corrosion products (CISA 1994) as well as o Support Systems
acidification. The depth of penetration of pitting can o Other Component Systems
 The rock mass and environmental factors will Table 4. Temperature and relative humidity range of eight
characterise the potential of a particular underground Australian underground mines.
location to cause corrosion. Observations of actual Mine Temperature Relative Humidity
corrosion on installed reinforcement and support Site (ºC) (%)
systems can be used to identify the corrosion 1 18-24 50-70
2 15-23 65-90
resistance of different systems. The corrosion data 3 17-30 50-79
collection sheet is shown in its entirety in the 5 21-26 75-90
Appendix with rock mass condition diagrams taken 6 25-32 80-95
from Ortlepp (1992) and Heslop (1997). 7 30-45 80-95
8 20-29 60 - 90

4 CASE STUDIES 4.1.1 Condition of Reinforcement and Support

Reinforcement and support under attack from
Using the corrosion assessment system described atmospheric corrosion were generally in fairly good
here a number of Australian underground mines condition. This is attributed to the low temperatures
have been inspected to determine the corrosive and relative humidity found throughout most mines,
potential of the underground environment and the the relatively short life expectancy of the ground
condition of reinforcement and support in those control in respect to the corrosion rate and the
environments. extensive use of galvanised and fully encapsulated
Based on this work two kinds of corrosive reinforcement for long term control.
environments have been identified. Those affected It is expected that few problems will occur from
by only the mine atmosphere and those affected by atmospheric corrosion at most mine sites. It has been
both the mine atmosphere and mine groundwater seen on numerous occasions that black friction bolts
flowing from the rock mass. form only uniform surface corrosion after two years
of installation in a mild atmosphere (Fig. 2).
However, some mines, for various reasons already
4.1 Mine Atmosphere discussed, have moderately corrosive atmospheres.
As indicated previously, atmospheric corrosion can When combined with a long mine life expectancy,
be defined as the corrosion of materials exposed to corrosion from atmospheric variables can cause
air and its pollutants. It was found that mine significant problems as illustrated in Figure 3. The
atmospheres do vary considerably with respect to ungrouted point anchor bolt is in excess of 15 years
moisture content, temperature, time of wetness and of age and has had severe localised corrosion along
contaminants throughout the mine. Over time as the element.
mining progresses and the ventilation circuit
changes, these parameters change.
Table 4 shows the range over which temperature and
Black Friction Bolt, 0.6m
relative humidity is present at various mine sites.
Generally for the Western Australian mines the
higher temperatures and relative humidity occur in
return airways, ore drives and where the air velocity
is low. Declines and larger crosscuts commonly
have the lowest temperature and humidity, which
makes them less favourable for atmospheric
corrosion attack. However two mines investigated
(Mine sites 6 & 7) have high temperatures and
humidity due to higher rock and water temperatures
and the common use of secondary ventilation. These
atmospheres are moderately corrosive and for such
mines the risk of atmospheric corrosion is higher.
Fortunately, the rate of atmospheric corrosion is
generally uniform and relatively slow with corrosion Figure 2. Black friction bolt after two years of installation in
rates for mild steel generally less the 0.05 mm/yr. 70-80% humidity atmosphere.
However, ground support within long-term
excavations can become severely corroded over
extended periods of time. Atmospheric corrosion is
usually not a problem for short-term excavations.
 At depth, groundwater becomes less common as
there is little weathering and high in-situ ground
stresses usually keep the structures tight, so that
ground water flow is severely restricted. Subsequent
mining activity redistributes the stress field and
structures open up allowing groundwater to flow
into previously dry areas. However, experience has
shown the rate and occurrence of groundwater flow
generally decreases with depth for all fractured rock
aquifers.
The presence of water is not always obviously
visible on the surface of excavations but may still be
present deeper in the rock mass, where corrosion of
Figure 3. Necking of ungrouted point anchor bolt due to
reinforcement may occur. These areas are classified
atmospheric corrosion.
as damp and are often revealed by white salt crystal
build up at the surface of excavations and a damp
appearance within the rock surface.
4.2 Mine Groundwater Mine groundwater samples were collected from
Corrosion attack due to mine groundwater affects a eight Australian underground mines and were tested
minority of ground support within a mine. However, the in-situ for total dissolved solids (TDS), pH,
attack can be highly aggressive and caused by a number dissolved oxygen (DO2) and temperature. The
of corrosion mechanisms with damage difficult to samples were also subsequently analysed for Na+,
predict. Mine groundwaters are generally encountered K+, Ca2+, Mg2+, Cl-, CO32+, HCO3-, SO42- and NO3+
in the presence of major structures and near draw points ions. A summary of the main constituents associated
of stopes that have been backfilled. The majority of with corrosion is shown in Table 5.
groundwater encountered by underground mining The testing and analyses of the groundwaters
activities in the Yilgarn occurs from hard or fractured showed a TDS range of 3,700 ppm to 230,000 ppm
rock aquifers in igneous and metamorphic rocks and with the major ions present being Na+, Ca2+, Cl- and
more than one aquifer may be controlling the flow of SO42-. In terms of corrosiveness the Cl- and SO42-
groundwater. The rock itself is generally impermeable, ions are the most important and generally the
but fractures, joints and weathering allow a degree of concentrations were very high. The dissolved
porosity and permeability. oxygen ranged from 0.73 – 4.1 mg/l and is directly
Groundwater flow from geological structures is related to the TDS and temperature of the water. The
rarely constant and regular over an extended time pH ranged from 5.80 – 8.12 indicating that the
period. The reasons may be environmental and groundwaters are relatively neutral and at this level
related to depletion of an aquifer, as well as pH no longer plays a direct role in corrosion
insufficient recharge or man-made changes such as although it can still affect the solubility and
nearby stoping diverting the flow of water. Excess equilibrium of other ions. The Ca2+ and HCO3+ ions
water from cemented backfill also flows into a rock are only minor constituents of the groundwater.
mass and while not technically groundwater, its Compared with the levels of chlorides and sulphates,
influence on surrounding ground support is similar. it is not believed they provide any corrosion
Areas affected by backfill water are usually protection.
restricted to drawpoint levels and often for limited
amounts of time.

Table 5. Groundwater characteristics of eight Australian underground mine groundwaters.

Mine TDS DO2 Temp pH Ca2+ HCO3- Cl- SO42-

No ppm mg/l ºC mg/l mg/l mg/l mg/l
1 230,000 0.73 26 5.80 310 50 180,000 24,000
2 5,400 3.54 26 8.12 187 73 1,570 1,278
3 13,000 3.92 27 7.67 430 10 7,700 170
4 44,000 1.13 26 7.40 910 250 23,000 6,000
5 120,000 3.23 22 6.40 1600 75 69,000 7,800
6 3,700 4.1 30 7.70 300 48 1,500 800
7 28,000 4.13 33 5.90 2,700 35 16,000 1,500
8 160,000 3.2 28 7.00 4000 35 93,000 3,100
4.2.1 Condition of Reinforcement and Support Friction bolts are highly susceptible to corrosion
In the case of fully encapsulated reinforcement, attack because of their thin steel thickness and open
dilation of geological discontinuities is likely to design of the bolt. Figure 6 shows the internal
create cracks within the cement grout or resin condition of two galvanised friction bolts of
column leading to corrosion (Fig. 4). This may leave different ages. The upper bolt is shown after two
the elements exposed at the areas where the months of installation and the bottom after four
groundwater is more likely to flow. The timing of months of installation in a wet shear zone with water
such occurrences is difficult to predict, but stress at mine 6. A considerable increase in the corrosion
changes and blast damage may promote dilation of rate from essentially non-corroded to strong
rock mass discontinuities. corrosion was observed in the interval of two
months. As a general rule, corrosion on galvanised
friction bolts is not experienced before 2-3 months
regardless of the environment. However, once the
protective zinc coating has been compromised,
corrosion of the steel may follow very quickly. The
corrosion rate actually decreases over time due to
the protection provided by the corrosion products.
The corrosion products are a mixture of iron rust and
deposited salts. These products provide some
protection from uniform corrosion but at the same
time promote pitting corrosion
Figure 4. Cracking of a resin grout column exposing the black
rebar.

Galvanised Friction Bolt, 1.0m

In the case of cement grouted elements, the
cement itself is a corrosion inhibitor due to its high
alkalinity and is also self-healing, as the grout can
reform in the crack protecting the element. Results
from laboratory testing suggest that a minimum
crack width of greater than 1-2 mm is needed before
self-healing cannot occur.
Research into corrosion along the axis of fully
encapsulated bolts has not been completed, as trials
of bolt overcoring (Fig. 5) have just started at the
WA School of Mines. Nevertheless, numerous
friction bolts have been examined internally to
gauge their level of corrosion in relation to their
environment.

Galvanised Friction Bolt, 1.0m

Figure 6. The internal condition of a galvanised friction bolt

Figure 5. Overcored resin encapsulated rock bolt using the after 2 months (top) and 4 months (below) in an environment
WASM bolt overcoring rig. strongly affected by groundwater (same mine site).
 The susceptibility of weld mesh to corrosion tested on four common reinforcement elements.
appears to be moderate. Strongly corroded mesh These were black thread bar, black friction bolt,
samples have been laboratory tested to determine black strand cable and galvanised strand cable. The
their failure loads, which range from 0-40% less corrosion rates for each synthetic water and each
than failure loads determined from non-corroded reinforcement element are shown in Figure 7. The
samples. results indicate that there is a steady decrease in the
corrosion rate of the black reinforcement as the
water quality become worse. This is attributed to a
4.3 Summary lowering of the dissolved oxygen content due to the
Two corrosive environments have been established higher TDS and an increase in scaling that prevents
for the underground environment; those affected by oxygen diffusion to the metal surface. The
atmospheric variables and those affected by galvanised strand gave much lower corrosion rates
groundwater variables. Often a clear distinction and, if anything, the corrosion rate increased with
between the two environments cannot be salt concentration. The precision of these
established. Mine atmospheres range from mildly to measurements was quite poor due to the non-
moderately corrosive and generally do not cause a uniform surface of the galvanised samples.
problem for short-term support. Groundwater
corrosion is considerably more aggressive and is a Oxygen Corrosion of Steel Supports
problem for both short and long term ground control, Ga lva nis e d S te e l S tra nd B la c k Stra nd Ca ble
especially for thin walled reinforcement such as 8.000
B la c k Thre a d Ba r B la c k Fric tio n Bo lt

friction bolts. While cement grout and resin 7.000

encapsulation offer some protection from the 6.000
5.000
corrosive environment, dilation of joints causes 4.000
cracking of the protective barrier so the element is at 3.000
2.000
risk. It has been found that galvanising of ground 1.000
control elements helps protect the underlying steel 0.000
-1.000
from corrosive attack by natural groundwaters. Dilute Wa te r No rma l
Wa te r
Mo de ra te
Wa te r
High Wa te r Extre m e
Wa te r

Figure 7. Corrosion rates of various reinforcement elements in

5 LABORATORY TESTING a range of synthetic mine groundwaters

5.1 Wheel tests 5.1.2 Mine groundwaters

A second set of tests was conducted using mine
A number of laboratory-based experiments were groundwater collected directly from underground
undertaken in conjunction with the WA Corrosion rock masses. The corrosion rate was tested on mild
Research Group at Curtin University using the steel and the corrosion rates ranged from 2 – 7
accelerated corrosion wheel test methods. In this
mm/yr (Table 6.). This rate is quite extreme and
test, a metal coupon is placed within a glass bottle
containing the water sample. The bottle is capped, overstates the actual rate of corrosion. The predicted
placed on a wheel inside an oven and rotated for the rate of corrosion by the oxygen wheel test is
required period (generally 6 hours). The test assumed to be the initial rate of corrosion. As the
provides stirred conditions at a controlled process continues the corrosion products actually
temperature. The test is excellent for screening the slow the rate of corrosion. This statement is
performance of a large number of metals, in supported by field evidence, which did not detect the
different waters, under the same condition. It is predicted rate of corrosion reinforcement on any
important to appreciate that the test vessel is a closed environment underground.
system and the chemistry inside the vessel will When comparing the results with the groundwater
change according to the corrosion process. Two analysis it is concluded that one groundwater
series of testing has been completed using synthetic parameter does not control the rate of corrosion.
and actual mine groundwaters.
Table 6. Corrosion rates of mild steel in actual mine
5.1.1 Synthetic groundwaters groundwaters.
Five synthetic mine groundwater types were Mine Site 1 2 3 4 5
Corrosion Rate 1.35 2.57 7.34 3.65 4.24
prepared based on data from Australian underground (mm/yr)
mine groundwaters. They ranged from dilute (Cl-
500 ppm, SO42- 50ppm) to extreme (Cl- 125,00 ppm,
SO42- 12,500ppm) with a pH of 7 at 25ºC and were
6 OUTCOMES FROM CORROSION
CLASSIFICATIONS AND LABORATORY

Corrosion rate (mm/yr)

8
TESTING 7
6
5
Comparing the existing corrosion classification 4
rankings to the actual corrosion rate measured from 3
2
the oxygen wheel test the relevance of each 1
classification for Australian mine groundwaters is 0
appreciated. The DIN classification (Fig. 8) shows -3.00 -2.00 -1.00 0.00 1.00
no trend. With this method the corrosion rate is
LSI rating
expected to increase as the rating decreases.
The Li & Linblad (1999) classification shows an
increase in ranking for the increasing corrosiveness Figure 10. Relationship between the Langelier Saturation Index
and wheel test corrosion rates.
of the water (Fig. 9) as predicted by the
classification.
No correlation is seen when comparing the
7 CONCLUDING REMARKS
corrosion rate to the Langelier Saturation Index
rating (Fig. 10). With this method those waters
A corrosivity assessment system has been developed
having a positive rating should have a lower
and tested on a number of underground
corrosion rate if protective calcium carbonate
environments. The method characterises the rock
scaling is forming.
mass, environmental and ground control schemes
It is accepted that the sample volume is small,
contributing or affected by corrosion. The
although research is progressing and early
parameters collected have been used to evaluate
indications are that the classification that best
existing corrosivity classifications systems, which
predicts the corrosiveness of the waters is that
are generally not appropriate to the Australian
proposed by Li & Linblad (1999).
mining conditions. The corrosivity classification by
Li & Linblad (1999) appears to be the most
applicable.
Corrosion Rate (mm/yr)

8
7
6
5 ACKNOWLEDGEMENTS
4
3
2 The authors wish to acknowledge the support of the
1 Minerals and Energy Institute of WA (MERIWA),
0 WMC Resources Ltd, PlacerDome Asia Pacific,
-30 -20 -10 0 Xstrata Ltd, Barrick Gold Australia, GoldFields
DIN ranking W1 Australia, MBT (Australia) and Strata Control
Systems.
Figure 8. Relationship between DIN ranking W1 and wheel test
corrosion rates.
REFERENCES

CISA, 1994. Corrosion Control in Southern Africa, Corrosion

Corrosion Rate (mm/yr)

8
7 Institute of Southern Africa (Mintek: South Africa).
6 DIN 50929-3, 1985 Corrosion of metals; probability of
5 corrosion of metallic materials when subjected to corrosion
4 from the outside; buried and underwater pipelines and
3 structural components.
2 Gamboa, E. & Atrens, A. 2003. Environmental influence of the
1
0 stress corrosion cracking of rock bolts. Eng. Failure
Analysis, 10(5):521-558.
0.0 5.0 10.0 15.0 Heslop, G. 1997. Personal communication.
Li & Linblad rating Wwet ISO 9223, 1992 Corrosion of metals and alloys; corrosivity of
atmospheres; classification.
Jones, R.H. & Ricker, R.E. 1990. Stress Corrosion Cracking.
Figure 9. Relationship between Li & Linblad ranking and ASM Metals Handbook, vol 13, pp145-163, (ASM: USA).
wheel test corrosion rates. Li C. & Lindblad K. 1999. Corrosivity classification of the
underground environment. Proc. Int. Symp. Rock Support
and Reinforcement Practice in Mining, Kalgoorlie.
(Villaescusa, Windsor and Thompson Eds), pp 69-75,
Balkema, Rotterdam.
Method for Determination of Soil Corrosivity.
(httd.njuct.edu.cn/MatWeb/soil/table603.htm) 29/4/2004.
Ortlepp, W.D. 1992. The design of support for the containment
of rockburst damage in tunnels – An engineering approach,
in Proc. Int. Rock Support in Mining and Underground
Construction, Ontario, pp 593-609, (Balkema: Rotterdam).
Ranasooriya, J., Richardson G.W. & Yap L.C. 1995. Corrosion
Behaviour of Friction Rock Stabilisers used in
Underground Mines in Western Australia, in Proc
Underground Operators Conference, Kalgoorlie, pp 9-16
(The Australian Institute of Mining and Metallurgy).
Roberge, P R, 2000. Handbook of Corrosion Engineering,
(McGraw-Hill: New York).
Stress corrosion cracking. (www.corrosion-
doctors.org/Forms/scc.htm) 29/4/2004.
Sundholm, S. 1987. The quality control of rock bolts. Proc. Int
Congress on Rock Mechanics, ISRM Montreal, (Ed. Herget
& Vongpaisal), vol. 2, pp 1255-1264.
Villaescusa, E. 1999, Keynote lecture: The reinforcement
process in underground mining. Rock support and
reinforcement practice in mining. Kalgoorlie, (Ed.
Villaescusa, Windsor & Thompson), pp 245-257, Balkema:
Rotterdam.
Windsor, C. R. & Thompson, A. 1993, Rock reinforcement -
technology, testing, design and evaluation. Comprehensive
Rock Engineering. (Ed. Hudson, J.), pp 451-484,
Permgamon Press, Oxford.
 APPENDIX

SITE SPECIFICATION
Date of Date of Excavation Expected
Mine Site Location Inspected By Current Age
Inspection Excavation Type Life

ROCK MASS
NATURE OF THE ROCK MASS
Layered Blocky
Massive Thin/Medium/Slabby Cross Joint Spacing Small/Medium/Large
<0.2/ />0.5 m <1/ />2 m <0.5/ />1.5 m

INTACT ROCK
Density Compressive Strength Stiffness Ratio Failure Mode
Type Minerals L/M/H L/M/H L/M/H
3
Brittle/Ductile
<2.5/ />3 t/m <50/ />100 MPa <200/ />500

MAJOR DISCONTINUITIES
Type: Condition
Weathering Profile Roughness Aperture Filling Seepage
L/M/H S/U/P VS/S/R Width (m) None/Type None/Drip/Flow

JOINTS
Geometry Condition
Joint
FF Persistence Weathering Profile Roughness Aperture Filling
Number of Connectivity
Sets L/M/H L/M/H
L/M/H F/S/M/H S/U/P VS/S/R Tight/Open None/Type
0-4/4-7/>7 <5/5-10/>10

ENVIRONMENTAL CONDITIONS
GROUND WATER
Dry Damp Wet Dripping Flowing
Location
Amount
Source
Salt Precipitation
Comments_____________________________________________________________________________________________
______________________________________________________________________________________________________
Properties
Source
Temperature pH Electrical Conductivity Redox Potential Dissolved Oxygen TDS Salinity
°C µS/cm or mS/cm mV mg/l ppm Ppm

ATMOSPHERE
Ventilation Type Flow Rate Particulates Temperature Relative
Gases Dew Point
Fresh/Exhaust L/M/H L/M/H Dry Bulb Wet Bulb Humidity
°C °C % °C
 STRESSES
Global Local Seismic Potential
Stresses Low Moderate High Low Moderate High Low Moderate High

OVERALL ROCKMASS CONDITION

Rock Mass
Stress
Massive Layered Jointed

Stable Arch Formation Discrete Large Blocks

Low Stress

Flexural Toppling and Sliding Unravelling of Small Blocks

Moderate
Stress

Sliding, crushing and squeezing

Crushing and spalling Buckling perpendicular to layers Crushing and spalling

High Stress

Strain Burst

Violent Rock Burst Tensile Splitting shearing and sliding Rock Burst
 REINFORCEMENT
System Date of
Component Type Dimensions Material Coating Corrosion Description Photo
Number Installation
Element
Internal Fixture
R1
External Fixture
Plate
Element
Internal Fixture
R2
External Fixture
Plate
Element
Internal Fixture
R3
External Fixture
Plate
Element
Internal Fixture
R4
External Fixture
Plate

SUPPORT
System Date of
Component Type Dimensions Material Coating Corrosion Description Photo
Number Installation
S1 Plate
S2 Strap
S3 Mesh

General Comments________________________________________________________________________
________________________________________________________________________________________
________________________________________________________________________________________
________________________________________________________________________________________

View publication stats