Introduction
Fibre Reinforced Shotcrete (FRS) is extensively used throughout the underground operations of the Cannington Mine, Queensland, as an integral part of the ground support system.
A study to assess alternate FRS mixes was jointly undertaken between BHP Billiton, the mine owners, and Jetcrete (Australia) Pty Ltd, the principal con-tractors for the supply and application of FRS at the mine. The objective was to determine which of the mixes would deliver the same or better levels of performance at a lower cost.
At the commencement of the program, performance of FRS at the mine was measured in terms of uniaxial compressive strength (UCS). Bernard (1999) had argued that this property is inappropriate as a measure of performance, proposing instead a measure of the material’s flexural toughness or its resistance to crack propagation. Jetcrete (Australia) Pty Ltd, as the spraying contractor for the mine, also expressed an interest in a toughness measure of a mix as a replacement to reliance solely on UCS as the performance measure. The contact at the mine required the supply of a mix with a minimum UCS of 40 MPa. At the time there was little knowledge of what level of toughness was required at the mine.
With regard to the FRS mix, the test program considered two factors – fibre type and addition or dosage rate of fibres. A cost model was developed to assess any differences in cost.
The study also considered the substitution of silica fume with kaolite which could be supplied at a lower unit cost. The silica fume is used as a cement substitute in the FRS mix principally to increase strength and reduce porosity.
TEST PROGRAM
Test variables
The variables in the test program included: fibre type, dosage rate: mass of fibre (kg) per unit volume (m3) of FRS, replacement of silica fume by kaolite. The performance of each variable was compared against the “Standard” mix at Cannington which used EE256 slit-sheet steel fibres.
The types and dosage rate of the various fibres were selected by the mine owners and contractors and are summarised in Table 1. Three levels of kaolite were tested against the Standard mix – 40, 60 & 80 kg/m3. The Mining Department at the mine authorised the testing of each fibrecrete mix.
Table 1
Range of fibre types examined
|
Fibre type |
Manufacturer |
Fibre Length |
Dosage Rate |
|
EE256 |
BHP |
25 mm |
60 kg/m3 |
|
RC65/35 |
Bekaert |
35 mm |
30 & 35 kg/m3 |
|
HPP |
Synthetic Indust. |
50 mm |
6 & 7 kg/m3 |
|
BarchipHT48 |
Hagihara Indust. |
48 mm |
9kg/m3 |
Test method
Fibre mix
To assess the variation in performance, the ASTM C-1550 round panel test as developed by Bernard (1999) was used. This method was selected as it has the capability of assessing performance beyond the peak load. In this post-failure region, the FRS still offers resistance to loading thus requiring the input of additional force and energy to cause displacement or flexure of the FRS membrane beyond the point of peak load. This energy is absorbed by the fibres until they eventually debond from the concrete matrix; the greater the energy, the greater the toughness of the material.
Three samples of each FRS mix were sprayed into moulds at the mine site as part of a regular underground application so as to minimise any change in the environmental variables. Each sample had a nominal diameter and thickness of 800 mm and 75 mm, respectively. The samples were cured for a minimum of 28 days prior to testing.
The round panel test was carried out in accordance with ASTM C-1550 using facilities in the School of Civil Engineering and Environment at the University of Western Sydney. An MTS 250 kN servo-hydraulic actuator was used to apply a load at the centre of the sample at a controlled rate so as to achieve a constant rate of displacement of 4.0 mm/s. The sample was symmetrically supported at three points some 375 mm radius from the centre and restrained to prevent any lateral movement.
The peak load was measured for each sample and the energy calculated to cause displacements of 5, 10, 20 & 40 mm. As the peak load precedes crack formation, it represents the strength of the concrete matrix. With further loading beyond the peak load, the incremental energy absorbed in the FRS is measured by integrating the load/displacement curve from the peak load to the various levels of displacement. Further details on the method of testing are reported by Bernard (2000).
Kaolite substitution
The effect of various proportions of kaolite in the Standard mix was assessed in terms of its impact on overall strength of the mix. Kaolite, like silica fume, is a fine grain material that can be used to reduce porosity in the FRS mix. Hence the impact on strength in terms of UCS was measured rather than toughness.
The Standard UCS test (AS 1012.9) was performed by Bowler Geotechnical at Mount Isa. Samples were prepared at the mine site in accordance with AS 1012.8 in moulds with a core diameter and height of 100 and 200 mm respectively. Samples were moist-cured in lime-saturated water.
RESULTS
Fibre Reinforced Shotcrete mix
A load/displacement graph was recorded for each test. The variation in absorbed energy against displacement was then calculated. A graph of the variation in load and energy against displacement for one of the three tests based on the Standard mix is shown in Figure 1.
Figure 1. Load and absorbed energy versus displacement graph for the Standard mix, EE256 @ 60 kg/m3
Graphs for the other fibre mixes are shown in Figures 2 to 4. The graphs indicate some difference in behaviour between the different mixes particularly in the post peak load region.
All three alternate fibre types showed a sudden reduction in load with further displacement once the peak load was achieved. This residual strength behaviour was not as readily evident in the Standard mix. The level of residual strength relative to its peak load was also found to vary between the different mixes.
Figure 2. Load and absorbed
energy
versus displacement graph for RC65/35 @ 35 kg/m3
Figure 3. Load and absorbed
energy
versus displacement graph for HPP @ 7 kg/m3
Figure 4. Load and absorbed
energy
versus displacement graph for Barchip @ 9.1 kg/m3
Values for peak load and absorbed energy levels are contained in Table 2. The recorded values were adjusted for variations in diameter and thickness according to the method of Bernard & Pircher (2001). The values presented represent the average of three samples tested for each mix but exclude those samples considered invalid on the basis that less than three primary radial crack were developed.
Table 2
Summary of average test results
for the various fibre mixes
|
Fibre type |
Peak Load |
Absorbed Energy (J) |
||||||
|
|
|
5mm |
10mm |
20mm |
40mm |
max |
||
|
EE256 |
|
|
|
|
|
|
||
|
60 kg/m3 |
34.8 |
104 |
183 |
247 |
281 |
291 |
||
|
RC65/35 |
|
|
|
|
|
|
||
|
30 kg/m3 |
27.1 |
81 |
124 |
179 |
241 |
292 |
||
|
35 kg/m3 |
31.9 |
113 |
185 |
261 |
340 |
413 |
||
|
HPP |
|
|
|
|
|
|
||
|
6 kg/m3 |
25.7 |
44 |
82 |
132 |
174 |
188 |
||
|
7 kg/m3 |
27.8 |
63 |
117 |
163 |
187 |
198 |
||
|
Barchip |
|
|
|
|
|
|
||
|
9 kg/m3 |
21.3 |
59 |
133 |
222 |
257 |
257 |
||
In terms of peak load which is a measure of the strength of the concrete matrix, EE256 achieved the highest average peak load of 34.8 kN. It is interesting to note that the proportion of cement, aggregate and sand in all samples tested remained virtually unaltered and that in general there was little significant difference in the maximum peak load for the other mixes. The components in the Standard mix are shown in Table 3.
Table 3
Composition of the Standard mix
|
Component |
Proportion |
|
Cement |
420 kg |
|
Aggregate (7mm) |
390 kg |
|
Coarse sand |
730 kg |
|
Fine sand |
450 kg |
|
Water |
175 litre |
|
BHP EE256 fibres |
60 kg |
|
Silica fume |
20 kg |
|
Accelerator |
12 litre |
|
Sika 908 |
15 litre |
There was, however, some variation in the level of absorbed energy for the different mixes as illustrated in Figure 5.
Figure 5. Absorbed energy at varying displacements.
In terms of assessing toughness, two factors can be considered, these being:
· the energy absorbed by the FRS necessary to cause the FRS to deform; and
· the overall total energy that could potentially be absorbed, that is the maximum energy absorption capability of the FRS calculated as the level where the energy curve becomes asymptotic to displacement,
Both dosage rates for the HPP fibre were found to have a reduced capability to absorb energy when under load compared to the other fibre types. The higher of the two dosage rates of RC65/35 of 35 kg/m3 showed a higher overall energy absorption capacity over a range of displacements up to 20 mm as illustrated in Figure 5. Hence the higher dosage RC65/55 fibre should be considered where ground conditions are more susceptible to large displacements in excess of 20 mm otherwise there is little difference in performance compared to the Standard mix using EE256.
Kaolite
The results of the UCS test on samples with different cure times and proportions of kaolite in the Standard mix compared to silica fume are summarised in Table 4. In the table, n represents the number of samples tested.
Table 4
Result of UCS test with Kaolite in the Standard mix
|
Sample |
UCS (MPa) |
n |
|
Standard mix - silica fume @ 20 kg/m3 |
||
|
7 days |
47.0 |
1 |
|
28 days |
58.8 |
2 |
|
Kaolite @ 40kg/m3 |
|
|
|
7 days |
44.0 |
1 |
|
28 days |
48.3 |
2 |
|
Kaolite @ 60kg/m3 |
|
|
|
7 days |
52.5 |
1 |
|
28 days |
63.0 |
2 |
|
Kaolite @ 80kg/m3 |
|
|
|
7 days |
60.5 |
1 |
|
28 days |
69.0 |
2 |
The contract specification required a minimum UCS of 40 MPa. The test results indicated all samples tested exceeded the specification. A minimum proportion of kaolite of 60 kg/m3 in the mix was required in order to achieve comparable strength levels to that achieved when 20 kg/m3 silica fume is used in the Standard mix. Even so, the minimum strength specified by the mine operators for the FRS was 40 MPa and all the samples tested were found to exceed this value.
Cost model
A cost model was developed to compare the different fibre types and dosage rates and the use of kaolite in the FRS mix. The purpose of the model was to determine the likely changes in cost of using alternate FRS mixes in ground support. An underlying assumption in the model was that there would be no change in the thickness of FRS and rate of application per unit area. Hence cost for the purposes of comparison was calculated on a dry unit volume basis.
The model examined three levels of estimated costs:
· the component cost - the cost of purchasing and transporting materials to site expressed in terms of dollars per unit volume;
· the agitator cost - the cost of mixing on site including maintenance, quality assurance, margin and stock loss; and
· the pump cost - the total cost of supply, mix and applying the mix in the underground mine environment.
For the purposes of comparison, the costs of each mix have been normalised against the cost of the Standard mix. A summary of the modelling is shown in Table 5.
Table 5
Result of cost comparison against the Standard mix
|
Mix |
Material |
Agitator |
Pump/Total |
|
Standard mix - EE256 |
|
|
|
|
60 kg/m3 |
1.000 |
1.000 |
1.000 |
|
RC65/35 |
|
|
|
|
30 kg/m3 |
0.950 |
0.954 |
0.957 |
|
35 kg/m3 |
0.988 |
0.989 |
0.990 |
|
HPP |
|
|
|
|
6 kg/m3 |
0.880 |
0.890 |
0.897 |
|
7 kg/m3 |
0.907 |
0.914 |
0.920 |
|
Barchip |
|
|
|
|
9 kg/m3 |
0.954 |
0.958 |
0.961 |
|
Standard mix |
|
|
|
|
silica fume |
1.000 |
1.000 |
1.000 |
|
Kaolite |
|
|
|
|
40 kg/m3 |
0.993 |
0.994 |
0.994 |
|
60 kg/m3 |
1.011 |
1.010 |
1.009 |
|
80 kg/m3 |
1.028 |
1.026 |
1.025 |
The modelling indicated that all the alternate fibre types and dosage rates appeared to be less expensive than the standard mix with savings in the total cost (i.e. for the supply and application) ranging between 1% and 10%.
In terms of substituting kaolite it appeared that given the proportions tested, the cost appeared to be of the same order of cost of the Standard mixture using silica fume.
It should be noted that the cost model used data as supplied by Jetcrete and BHP in 2000. Since this time there may have been relative changes in the cost of the different materials used in the various FRS mixes.
ANALYSIS
The cost modelling indicated apparent cost differences arising from the use of the three alternate fibre types at the dosage rates examined. Differences in total cost varied from 1% for the RC65/35 at a dose rate of 35 kg/m3 up to 10.3% for the HPP fibre at 6 kg/m3.
However, the more significant cost savings would be achieved in conjunction with a reduction in the performance measures. Table 6 endeavours to correlate any differences in cost per unit volume of FRS against changes in the performance measures with the different fibres. Table 7 correlates difference in cost against performance when kaolite was substituted for silica fume.
Table 6
Comparison of cost and
performance of fibres
on the basis of a unit volume of FRS
|
Mix |
Cost |
Peak load |
Absorbed energy (J) |
|
|
|
|
|
@ 20 mm |
@ max |
|
Standard mix - EE256 |
|
|
|
|
|
60 kg/m3 |
100% |
100% |
100% |
100% |
|
RC65/35 |
|
|
|
|
|
30 kg/m3 |
-4.3% |
-22% |
-28% |
0% |
|
35 kg/m3 |
-1.0% |
-8% |
+6% |
+37% |
|
HPP |
|
|
|
|
|
6 kg/m3 |
-10.3% |
-26% |
-47% |
-35% |
|
7 kg/m3 |
-8.0% |
-20% |
-34% |
-32% |
|
Barchip |
|
|
|
|
|
9 kg/m3 |
-3.9% |
-39% |
-10% |
-12% |
Table 7
Comparison of cost and performance of Kaolite
|
|
cost |
UCS |
|
Standard mix |
|
|
|
silica fume |
100% |
100% |
|
Kaolite |
|
|
|
40 kg/m3 |
-0.6% |
-18% |
|
60 kg/m3 |
+0.9% |
+7% |
|
80 kg/m3 |
+2.4% |
+17% |
Table 6 indicates that while there might be apparent savings in total cost of FRS with the use of alternate fibres this was in all but one case at the expense of a reduction in performance.
Given the range of fibres tested and dosage rates examined, the Bekaert RC65/35 fibre at a rate of 35 kg/m3 was shown to perform better while achieving a slightly reduced cost. While significant cost reductions were indicated with the other fibres there were significant reductions in performance in terms of strength and toughness..
In terms of the replacement of the silica fume with kaolite, it would appear that this would be feasible without compromising performance at a concentration of 60 kg/m3 in the FRS mix but at a marginal increase in cost of 0.9%.
POST TRIAL CHANGES
Further tests undertaken principally by the Contractor at Mt Isa and in Western Australia lead to a change in early 2002 to the use of a synthetic fibre. In late 2002, Eroc was awarded the contract for the supply and placement of FRS at the mine. While the measure of UCS has remained unchanged, toughness as measured by the Round Determinate Test has been incorporated into the quality test procedures at the mine. The current composition of FRS at the mine site is shown in Table 8.
Table 8
Composition of the FRS mix
|
Component |
Proportion per m3 |
|
|
|
2000 |
2004 |
|
Cement |
420 kg |
420 kg |
|
Aggregate (7mm) |
390 kg |
390 kg |
|
Coarse sand |
730 kg |
730 kg |
|
Fine sand |
450 kg |
450 kg |
|
Water |
175 litre |
175 litre |
|
BHP EE256 fibres |
60 kg |
- |
|
Kyoto Barchip |
- |
7 kg |
|
Accelerator |
12 litre |
- |
|
Silica fume |
20 kg |
- |
|
Sika 908 |
15 litre |
- |
|
Sigunit L55AF (accelerator) |
14 litre |
|
|
Sika Pump (silica replacement) |
2.1 litre |
|
|
Sika NN (superplasticiser) |
6.1 litre |
|
|
Sika 930 (stabliser) |
1.9 litre |
|
CONCLUSIONS
The study indicated significant variations in the performance of FRS when different types and dosage rates of fibres were used.
In terms of assessing performance both the toughness and strength of FRS were measured using the round panel test. A cost model was developed to ex-amine differences in the cost of supply and application on a dry unit volume basis of FRS mix.
Over the range of variables studied, only one combination of fibre and dosage rate, the RC65/35 at 35 kg/m3, was found to outperform the Standard mix used at the Cannington mine. An 8% improvement was observed in the peak load that could be sustained by the FRS as well as a 37% increase in the total amount of energy that could be absorbed before failure of the FRS membrane. It would appear that its use would result in a cost saving, albeit only a marginal saving of 1%.
In terms of the other two fibres investigated, there may be scope for an improvement in performance at higher dosage rates given the indicative cost saving. Further study of these fibres at higher dosage rates would be necessary.
The study found that substitution of the silica fume by kaolite in the FRS mix could achieve comparable levels of performance. Cost was found to be particularly sensitive to the proportion of kaolite added to the FRS mix. The results indicate that while kaolite at a rate of 60 kg/m3 improved the strength by 7%, there was a slight cost increase of 0.9%. There is scope for further work examining proportions between 40 and 60 kg/m3 where performance could be improved while remaining cost neutral.
ACKNOWLEDGEMENTS
The authors wish to acknowledge the co-operation of BHP Billiton and Jetcrete (Australia) Pty Ltd in facilitating this research project which formed the basis of an Honours research thesis begun during industrial training. Particular thanks are extended to Natasha Jakubowski and Scott Jeffery at the Cannington Mine and Tony Finn at Jetcrete (Australia) Pty Ltd.
REFERENCES
Bernard, E.S. 1999. Use of panels for FRS toughness assessment in Proc. Conf. Ground support with fibre reinforced shotcrete. Australian Shotcrete Society.
Bernard, E.S. 2000. Behaviour of round steel fibre reinforced concrete panels under point loads. Materials and Structures 33(227):181-188.
Bernard, E.S. & Pircher, M. 2001. The influence of thickness on performance of fibre-reinforced concrete in a round de-terminate panel test. Cement, Concrete and Aggregates 23(1): 27-33.