ABSTRACT
Soft rock is a term that usually refers to a rock material with a
uniaxial compressive strength (UCS) less than 20 MPa. This low strength
range might be influenced by physical characteristics, such as size,
saturation, weathering and mineral content. A number of uniaxial
compression tests have been conducted onto soft rock samples. The
results showed that the strength reduced significantly in saturation.
The reduction was also caused by weathering, the strength of distinctly
weathered rocks were lower than that of partially weathered rocks. In
conjunction with the uniaxial compression test, point load strength
index tests, IS(50), have also been conducted in order to obtain a
correlation between the UCS and the point load strength index IS(50).
The results showed that the IS(50) could well be correlated with the
UCS. A conversion factor of 14 is proposed for soft rock materials.
Keywords: soft rock, uniaxial compressive strength, physical
characteristic, point load strength index, and conversion factor.
INTRODUCTION
The term soft rock is often referred to rock materials with a
uniaxial compressive strength (UCS) lower than that of hard rocks and
higher than that of soils. A rock material can be classified as a soft
rock if it has a uniaxial compressive strength below 20 MPa, determined
directly by uniaxial compression tests [1, 2, 3, 4].
Physical characteristics, such as sample preparation, size,
saturation, and mineral content, influence the uniaxial compressive
strength of soft rock materials. These factors can considerably reduce
the strength of soft rock materials.
The influence of moisture content on the behaviour of rocks has
been investigated for more than thirty years [5, 6, 7, 8, 9]. It is
frequently associated with mechanisms such as capillary suction and
crack propagation. Most researchers agree that the strength of any given
soft rock will decrease as the moisture content increases. However, the
correlation between strength and moisture content is not always linear.
Hawkins and McConnell [8] found that, for most types of sandstone,
a sudden loss of strength occurs between a moisture content of 0% and
1%, and only a slight strength reduction above a moisture content of 1%.
Similar results were also obtained by Schmitt et al. [10] for
Fontainebleau and Vosges sandstones, which show an exponential
correlation; but a relatively linear correlation was obtained for
Tournemire shale.
These findings [10] reveal that the sensitivity of rock strength
due to changes in moisture content seems to vary from rock to rock. This
sensitivity depends on the clay content of the rock being investigated
[8]. A similar result was also found by Bell and Culshaw [9]. However,
Dobereiner and DeFreitas [11] and Dyke and Dobereiner [12] pointed out
that weaker sandstones are more sensitive to changes in moisture content
than harder rocks. Dyke and Dobereiner [12] concluded that the texture
of the rock, that is the proportion of grain contact, is responsible for
reductions in the strength of sandstone. Further, Dyke and Dobereiner
[12] found that an increase in moisture content tends to decrease the
range of elastic behaviour of sandstone.
From the discussion above, it is interesting to note that the
critical condition for rocks containing fewer clay minerals is when the
moisture content increases up to 1%, where a sudden strength loss
occurs. The reason may be that suction, acting as a confining pressure,
suddenly disappears when the moisture content reaches 1%. However, for
rocks which are rich in clay minerals, suction disappears gradually up
to the degree of saturation of 100% [13].
Sample preparation for soft rock materials is more difficult
compared to that of hard rock materials. Agustawijaya [13] indicated
that the methods of sample preparation and laboratory testing for soft
rocks are similar to those of soils. Such difficulties have been
experienced by Agustawijaya [14] in preparing mudstone samples for
direct laboratory shear testing; and by Bro [15] in preparing samples
for triaxial testing. Soft rocks are sometimes so friable, that the
rocks can easily break apart. Instead of machining the samples, Akai
[16] suggested the use of hand sculpturing and fine sandpaper to flatten
the ends of the friable core specimen. Thus, it is crucial to protect
soft core samples from damage as much as possible prior to testing,
since excellent testing results will depend to some extent on the
quality of the samples.
The International Society for Rock Mechanics (ISRM) [1] suggested a
length/diameter (L/D) ratio of 2.5 for samples in compression tests.
However, it is not easy to obtain a sufficient number of samples with an
L/D ratio bigger than 2, as the drilling program into soft rock is
usually difficult, and it might only obtain 10-20% of the drilling
length [13]. Thus, the ratio could be less than 2.5.
Due to some degree of technical difficulties, the uniaxial
compression test for soft rock is often replaced by a point load
strength index test. This test is simpler in procedures than that of the
uniaxial compression test, and the test does not need necessarily
cylindrical samples. A conversion factor is then applied to the point
load strength index, Is(50), for estimating the uniaxial compressive
strength (UCS).
Broch and Franklin [17] suggested a conversion factor of 24 as a
correlation between UCS and Is(50). However, this conversion factor is
more effective for hard rock materials than soft rock materials. The
strength of soft rock materials is much lower than that of hard rock
materials, so the conversion factor should certainly be different. This
paper aims to evaluate the uniaxial compressive strength, the influence
of some physical characteristics on the strength of soft rock materials,
and the correlation between the point load strength index and the
uniaxial compressive strength.
METHOD
Methods for uniaxial compression test follows the suggested method
given by the ISRM [1]. The tests in this study were conducted under
drained conditions. Test specimens were argillaceous rocks (sandstone
and siltstone) obtained from rock drilling at five different locations,
namely, Desert View Motel (DV), McCormack's dugout (MC), Old Timers
Shop/Museum (OTM), Gunther Wagner's dugout (GW) and Les Hoad's
dugout (LH) [13]. These specimens were cut with an L/D ratio of
1.6:1-2.5:1. For the uniaxial compression test these samples were
grouped into two main groups based on the degree of weathering and the
degree of saturation for each type of rock.
Laboratory point load strength test follows the methods suggested
by the ISRM [18]. In this study the strength index test used a speed
control Instron machine with a maximum load cell of 5 kN. Specimens were
core samples with a diameter of 37.5 mm, and a thickness of 20 mm.
Similar rocks with that for the uniaxial compression tests were tested
in the point load tests.
The degree of weathering was described according to the method
given by GSEGWP [19]. The two main qualitative classifications were
partially weathered (PW) rock and distinctly weathered (DW) rock. Some
samples were classified into de-structured (DST) rock, where some parts
of the rock have changed to soil.
Samples were further divided into two types of saturation, dry (0%
saturation) and fully saturated (100% saturation). Dry specimens were
obtained by putting the sample in an oven at a temperature of
105[degrees]C for 12 hrs. Fully saturated specimens were obtained by
immersing samples in distilled water. They were then vacuumed over 24
hrs [13].
RESULT AND DISCUSSION
Results of uniaxial compression tests on 39 specimens are listed in
Table 1. As can be seen in this table, partially weathered rocks give
higher uniaxial compressive strength (UCS) values compared to distinctly
weathered rocks. The highest UCS value is about 11 MPa for the (PW) LH
rock, whereas the lowest UCS value is about 1 MPa for the (DW) GW rock.
From Table 1, it can be seen that the average UCS values are far
below 20 MPa, and only the UCS of LH siltstone is about 10 MPa. It seems
that weathering and saturation play an important role in strength
reduction. The discussion of strength reduction follows.
Influence of L/D Ratio
The ISRM [1] recommended an L/D ratio of at least 2.5:1 for
compression tests. However, a sufficient number of specimens with this
ratio are often difficult to obtain [13]. With L/D ratios of 1.6:1 to
2.5:1, no significant indication has been found that the UCS is
influenced by the L/D ratio.
Figure 1 shows a scatter of UCS data of some samples related to
their L/D ratios, and there is no general trend to indicate an influence
of the ratio on the UCS. For example taking the DV (DST, sat.) samples,
the UCS values for these specimens are relatively similar for each
different ratio. The scattered data may simply be due to the physical
characteristics of the samples. Matthews and Clayton [20] found a
similar result for chalk tested with ratios of 2:1 and 2.5:1.
[FIGURE 1 OMITTED]
Some researchers [7, 9, 11, 21, 22] have conducted compression
tests on soft rocks with an L/D ratio of 2.0. Chiu et al. [7] conducted
drained triaxial tests on Melbourne mudstone with L/D ratios from 0.5 to
3.0, and found that the trend of peak deviator stresses and secant
Young's modulus tend to be constant when the ratio is at least 2.0.
Chiu et al. [7] pointed out that increased strength with a shorter
specimen is due to lateral restraint at the ends of the specimen. It is
understood that this lateral restraint is caused by the platens, which
may cause non-uniform stress distributions under compression [20]. The
volume of the specimen may also influence the results of the test.
However, it is the material properties that seem to cause the major
effects on the mechanical behaviour of soft rocks. Matthews and Clayton
[20] found that samples with ratios of 2.0 and 2.5 both displayed
similar behaviour under uniaxial compression stresses. They noted that
the uniaxial compressive strength of chalk was more likely to be
influenced by the dry density and porosity, rather than by the ratio of
the sample dimensions.
Influence of Saturation
The influence of water on rock strength has been known for years
[13], it is also known that strength reduction due to saturation varies
from one rock to another. The strength reduction for weathered
argillaceous rocks can be seen in Table 2.
From Table 2, the UCS reduction could be up to 75%, although, for
de-structured rocks, DV (DST), the reduction might only be 8%. Under dry
and fully saturated conditions, these DV (DST) rocks had very low UCS
values, below 2 MPa. Thus, in general, the mean value of strength
reduction due to water for all rock samples would be about 38%, which is
close to the strength reduction of sandstone (36%) found by Bell and
Culshaw [9].
Influence of Weathering
Data in Table 3 show that strength reductions due to weathering
could be quite high, to about 82%. Dry and fully saturated samples show
similar characteristics.
Dry density and porosity as weathering indicators [13, 20] were
correlated with uniaxial compressive strength, and both parameters show
a good correlation with uniaxial compressive strength (Figure 2).
From Figure 2, it can be seen that when the dry density increases,
the UCS increases. Similarly, as the porosity increases, the UCS
decreases, although the correlation between UCS and porosity has a lower
[R.sup.2], which is about 0.66. These correlations could mean that when
rock weathers, its dry density and porosity change, so does it's
UCS.
[FIGURE 2 OMITTED]
Influence of Mineral Content
As the degree of weathering increases, the clay content of the
weathered rock increases. For example, the (PW) DV rock may contain a
greater proportion of quartz than the (DW) GW rock, which contains
dominant kaolinite. It is important to note that gypsum is commonly
found in argillaceous rocks. It appear in every level of weathering, but
it is most commonly found in distinctly weathered rocks. Robertson and
Scott [23] noted that the presence of sulphate minerals, such as gypsum,
is due to oxidation of pyrite, which is commonly found in unweathered
sandstone. As sandstone weathers, gypsum may become more apparent.
However, weathered sandstone is pre-dominated by kaolinitic silty or
sandy minerals, which may become a good indicator for weathering in
argillaceous rocks [23].
Hawkins and McConnel [8] pointed out that the sensitivity of
sandstone to change in moisture content is controlled mainly by the
mineralogy and to a lesser extent by texture and microstructure.
However, Dobereiner and DeFreitas [11] and Dyke and Dobereiner [12]
found that the weaker the sandstone, the more sensitive its strength to
moisture content variation.
The important phenomenon observed during the current investigations
on argillaceous rocks are that weathering, is a complex feature that may
involve not only one process, but also other processes which can occur
simultaneously. Referring to Agustawijaya [13], chemical weathering and
physical weathering may occur one after the other. As products of
weathering, it may be difficult to quantify clay minerals that may
dominate over dry density and porosity in controlling the mechanical
behaviour of weathered rocks. Thus, both mineralogy and texture should
have, to some extent, an equal contribution to the mechanical
characteristics of rocks, particularly the uniaxial compressive
strength, depending on the degree of weathering.
Correlation Between UCS and Point Load Strength Index
The point load strength index, Is(50), has been used to predict the
uniaxial compressive strength (UCS), with a conversion factor of about
24 [17, 24, 25, 26]. However, this conversion factor may only adequately
predict the UCS of hard rocks. For soft rock, the conversion factor
could be much less than 24.
Forster [27] found that the conversion factor for sandstone falls
in the range between 7.4 and 17.6. More recently, Bowden et al. [28]
found that the conversion factor for chalk is from 5 to 24.
In the current research, the UCS and the Is(50) of the argillaceous
rock samples were correlated, as can be seen in Figure 3. The Is(50) has
a good correlation with the UCS, with a correlation coeficient of about
0.9. This is a good correlation that has also been indicated by Bell and
Culshaw [9].
From Figure 3, the conversion factor for weathered argillaceous
rocks is about 13.4. The conversion factor agrees well with the
theoretical conversion factor given by Chau and Wong [29], which is
about 14.9. Comparing between the current conversion factor and
published data [27, 28, 29] it seems that the value of 14 is the median
of all values falling in the range from 4 to 24. This value is 58% of
the conversion value for hard rocks, (24), given by Broch and Franklin
[17].
[FIGURE 3 OMITTED]
CONCLUSION
The uniaxial compressive strength (UCS) values of soft rocks have
been found to be far below 20 MPa. The UCS value of some samples
approached 1 MPa. This value is certainly very low for rock materials.
Weathering and saturation seem to play a significant role in reducing
the strength of soft rock materials. The strength reduction could be up
to about 80%.
The UCS values were then correlated with the point load strength
index values, which show a good correlation. The conversion factor for
soft rocks is found to be about 14, which is about 58% of the conversion
factor for hard rocks.
REFERENCES
[1.] International Society for Rock Mechanics (ISRM), Rock
Characterization, Testing and Monitoring, ISRM Suggested Methods, Brown,
E. T. (Editor). Pergamon Press, Oxford, 1981.
[2.] Bieniawski, Z. T., Engineering Rock Mass Classifications, John
Wiley & Sons, New York, 1989.
[3.] Franklin, J. A. and Dusseault, M. B., Rock Engineering,
McGraw-Hill Publishing Company, New York, 1989.
[4.] Bell, F. G., Engineering in Rock Masses,
Butterworth-Heinemann, Oxford, 1994.
[5.] Colback, P. S. B and Wiid, B. L., The influence of moisture
content on the compressive strength of rock, Proceedings of the 3rd
Canadian Rock Mechanics Symposium, 1965, pp. 65-83.
[6.] Chiu, H. K., Johnston, I. W. and Donald, I. B., Appropriate
techniques for triaxial testing of saturated soft rock, International
Journal of Rock Mechanics and Mining Science & Geomechanics
Abstract, 20, 3, 1983, pp. 107-120.
[7.] Chiu, H. K. and Johnston, I. W., The uniaxial properties of
Melbourne mudstone, 5th Congress of the International Society for Rock
Mechanics, Melbourne, 1983, pp. A209-A214.
[8.] Hawkins, A. B. and McConnell, B. J., Sensitivity of sandstone
strength and deformability to changes in moisture content, Quarterly
Journal of Engineering Geology, 25, 1992, pp. 115-130.
[9.] Bell, F. G. and Culshaw, M. G., Petrographic and engineering
properties of sandstones from the Sneinton Formation, Nottinghamshire,
England, Quarterly Journal of Engineering Geology, 31, 1998, pp. 5-19.
[10.] Schmitt, L., Forsans, T. and Santarelli, F. J., Shale testing
and capillary phenomena, International Journal of Rock Mechanics and
Mining Science & Geomechanics Abstract, 31, 5, 1994, pp. 411-427.
[11.] Dobereiner, L. and DeFreitas, M. H., Geotechnical properties
of weak sandstones, Geotechnique, 36, 1, 1986, pp. 79-94.
[12.] Dyke, C. G. and Dobereiner, L., Evaluating the strength and
deformability of sandstone, Quarterly Journal of Engineering Geology,
24, 1991, pp. 123-134.
[13.] Agustawijaya, D. S., The Development of Design Criteria for
Underground Excavations in Coober Pedy Arid Soft Rocks, Ph.D.
Disertation, University of South Australia, 2001.
[14.] Agustawijaya, D. S., The Mechanical and Geometrical
Properties of Discontinuities in Leigh Creek Coalfield Rocks, M.Eng
Thesis, University of South Australia, 1996.
[15.] Bro, A., A weak rock triaxial cell, International Journal of
Rock Mechanics and Mining Science & Geomechanics Abstract, 33, 1,
1996, pp. 71-74.
[16.] Akai, K., Testing methods for indurated soils and soft
rocks--Interim report, Geotechnical Engineering of Hard Soils--Soft
Rocks, Anagnostopoulos et al. (Eds.), Balkema, 1997, pp. 1707-1736.
[17.] Broch, E. and Franklin, J. A., The point load strength test,
International Journal of Rock Mechanics and Mining Science &
Geomechanics Abstract, 9, 1972, pp. 669-697.
[18.] International Society for Rock Mechanics (ISRM), Suggested
methods for determining point load strength, International Journal of
Rock Mechanics and Mining Science & Geomechanics Abstract, 22, 1985,
pp. 53-60.
[19.] Geological Society Engineering Group Working Party (GSEGWP),
The description and classification of weathered rocks for engineering
purposes, Quarterly Journal of Engineering Geology, 28, 1995, pp.
207-242.
[20.] Matthews, M. C. and Clayton, C. R. I., Influence of intact
porosity on the engineering properties of a weak rock, Geotechnical
Engineering of Hard Soils--Soft Rocks, Anagnostopoulos et al. (Eds.).
Balkema, Rotterdam, 1993, pp. 693-701.
(21.) Yoshinaka, R., Osada, M. and Tran, T. V., Deformation
behaviour of soft rocks during consolidated-undrained cyclic triaxial
testing, International Journal of Rock Mechanics and Mining Science
& Geomechanics Abstract, 33, 6, 1996, pp. 557-572.
(22.) Hayano, K., Sato, T. and Tatsuoka, F., Deformation
characteristics of a sedimentary soft mudstone from triaxial compression
tests using rectangular prism specimens, Geotechnique, 47, 3, 1997, pp.
439-449.
(23.) Robertson, R. S. and Scott, D. C., Geology of The Coober Pedy
Precious Stones Field, Result of Investigations, 1981-86, Report of
Investigations 56, Department of Mines and Energy, Geological Survey of
South Australia, 1990.
[24.] Bieniawski, Z. T., Estimating the strength of rock materials,
Journal of the South African Institute of Mining and Metallurgy, March
1974, pp. 312-320.
[25.] Brook, N., The use of irregular specimens for rock strength
tests, International Journal of Rock Mechanics and Mining Science &
Geomechanics Abstract, 14, 1977, pp. 193-202.
[26.] Greminger, M., Experimental studies of the influence of rock
anisotropy on size and shape effects in point load testing,
International Journal of Rock Mechanics and Mining Science &
Geomechanics Abstract, 19, 1982, pp. 241-246.
[27.] Forster, I. R., The influence of core sample geometry on the
axial point load test, International Journal of Rock Mechanics and
Mining Science & Geomechanics Abstract, 20, 6, 1983, pp. 291-295.
[28.] Bowden, A. J., Lamont-Black, J. and Ullyott, S., Point load
testing of weak rocks with particular reference to chalk, Quarterly
Journal of Engineering Geology, 31, 1998, pp. 95-103.
[29.] Chau, K. T. and Wong, R. H. C., Uniaxial compressive strength
and point load strength of rocks, International Journal of Rock
Mechanics and Mining Science & Geomechanics Abstract, 33, 2, 1996,
pp. 183-188.
D. S. Agustawijaya
Lecturer, Department of Civil Engineering, Faculty of Engineering,
Mataram University, Mataram, Indonesia Email: agustawijaya@telkom.net
Note: Discussion is expected before June, 1st 2007, and will be
published in the "Civil Engineering Dimension" volume 9,
number 2, September 2007.
Received 1 Mey 2006; revised 21 July 2006, 18 October 2006;
accepted 21 November 2006.
Table 1. Uniaxial Compression Test Results
Rock type Sample Weather- Satura- L/D UCS Ave
ing tion (MPa) (MPa)
DV DV9^ PW Dry 1.77 10.54 9.78
Sandstone DVR1 PW Dry 2.00 9.01
DV3-3DW^ DW Dry 1.72 3.15 3.72
DV3-DW0 DW Dry 2.33 3.07
DV8A^ DW Dry 1.64 4.93
DV2-DW1^ DST Dry 1.75 1.79 1.79
DV3-1S1 DW Sat. 2.17 2.45 2.84
DV3-2 DW Sat. 2.00 3.32
DV3-3 DW Sat. 1.92 2.76
DV2-S1 * DST Sat. 2.34 1.48 1.64
DV2-S2 * DST Sat. 2.22 1.85
DV3-1S2 DST Sat. 2.03 1.52
DV3M-S3 DST Sat. 2.12 1.63
DV2-S7 DST Sat. 1.95 1.74
GW GW-7M6 DW Dry 2.20 5.80 4.70
Siltstone GW1 DW Dry 1.67 5.08
GW3 DW Dry 1.62 4.53
GW2 DW Dry 2.04 3.38
GW-9M3 DW Sat. 2.29 1.46 1.16
GW-10M1 DW Sat. 2.03 1.00
GW-9M5 DW Sat. 1.97 1.01
LH LH3-9 PW Dry 2.13 9.92 10.10
Siltstone LH3-11 PW Dry 2.16 9.29
LH3-7 PW Dry 2.14 11.09
LH1-1 DW Dry 2.13 4.07 4.26
LH1-22 DW Dry 2.09 4.33
LH1-13 DW Dry 2.19 4.39
MC MC2-3-5 DW Dry 1.68 5.07 5.36
Siltstone MC22 DW Dry 2.27 5.65
MC4 PW Sat. 1.98 6.97 6.05
MC19 PW Sat. 2.09 7.20
mc16 DW Sat. 2.03 3.99 3.99
OTM OTM2-10-2 DW Dry 2.27 4.31 4.92
Siltstone OTMi-14 DW Dry 2.26 5.53
OTMo-15 PW/DW Sat. 2.50 5.50 5.25
OTMo-17 PW/DW Sat. 2.21 5.00
OTMo12 DW Sat. 1.89 1.58 2.05
OTM2-8A DW Sat. 1.66 2.13
OTM2-11 DW Sat. 1.87 2.43
Table 2. Influence of Saturation in Uniaxial Compressive Strength
UCS ave.
Rock type Weathering (MPa) Reduction Mean
Dry Sat (%) (%)
DV Sandstone DW 3.72 2.84 23.5 38.2
DV Sandstone DST 1.79 1.64 8.2
GW Siltstone DW 4.70 1.16 75.4
MC Siltstone DW 5.36 3.99 25.6
OTM Sandstone DW 4.92 2.05 58.4
Table 3. Influence of Weathering on Uniaxial Compressive Strength
Rock type Satura- UCS ave. (MPa)
tion PW PW/DW DW
DV Sandstone Dry 9.78 -- 3.72
DV Sandstone Dry 9.78 -- --
DV Sandstone Sat -- -- 2.84
LH Siltstone Dry 10.10 -- 4.26
MC Siltstone Sat 6.05 -- 3.99
OTM Siltstone Sat -- 5.25 2.05
UCS ave.
(MPa) Reduc- Mean
Rock type DST tion (%) (%)
DV Sandstone -- 62.0 56.5
DV Sandstone 1.79 81.7
DV Sandstone 1.64 42.2
LH Siltstone -- 57.8
MC Siltstone -- 34.1
OTM Siltstone -- 61.0