The Use of Point Load Test For Dubai Weak Calcareo
The Use of Point Load Test For Dubai Weak Calcareo
The Use of Point Load Test For Dubai Weak Calcareo
net/publication/279312793
The use of point load test for Dubai weak calcareous sandstones
CITATIONS READS
8 462
1 author:
Amr F. Elhakim
Cairo University
14 PUBLICATIONS 51 CITATIONS
SEE PROFILE
All content following this page was uploaded by Amr F. Elhakim on 07 July 2015.
The use of point load test for Dubai weak calcareous sandstones
PII: S1674-7755(15)00070-0
DOI: 10.1016/j.jrmge.2015.06.003
Reference: JRMGE 173
Please cite this article as: Elhakim AF, The use of point load test for Dubai weak calcareous sandstones,
Journal of Rock Mechanics and Geotechnical Engineering (2015), doi: 10.1016/j.jrmge.2015.06.003.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
The use of point load test for Dubai weak calcareous sandstones
Amr Farouk Elhakim
PT
Abstract: Intact rock is typically described according to its uniaxial compressive strength (UCS). The UCS is needed in the design of geotechnical engineering problems
including stability of rock slopes and design of shallow and deep foundations resting on and/or in rocks. Accordingly, a correct measurement/evaluation of the UCS is essential to
a safe and economic design. Typically, the UCS is measured using the unconfined compression tests performed on cylindrical intact specimens with a minimum length to width
ratio of 2. In several cases, especially for weak and very weak rocks, it is not possible to extract intact specimens with the needed minimum dimensions. Thus, alternative tests
RI
(e.g. point load test, Schmidt hammer) are used to measure rock strength. The UCS is computed based on the results of these tests through empirical correlations. The literature
includes a plethora of these correlations that vary widely in estimating rock strength. Thus, it is paramount to validate these correlations to check their suitability for estimating
rock strength for a specific location and geology. A review of the available correlations used to estimate the UCS from the point load test results is performed and summarized
herein. Results of UCS, point load strength index and Young’s modulus are gathered for calcareous sandstone specimens extracted from the Dubai area. A correlation for
C
estimating the UCS from the point load strength index is proposed. Furthermore, the Young’s modulus is correlated to the UCS.
Keywords: point load test; uniaxial compressive strength (UCS); rock characteristic elastic modulus; Dubai calcareous sandstone; calcarenite
US
not always be possible to obtain specimens having a 50 mm in diameter
1. Introduction (Brook, 1980). Therefore, it is common to perform the test on specimens
of different sizes and determine the point load strength index Is. The size
AN
The uniaxial compressive strength (UCS) of rock is the most is corrected to obtain the value of Is which would have been measured by
commonly used parameter in the characterization of intact rock. It is a diametrical test with D=50 mm and is given the symbol Is50 (ASTM,
needed for different applications including the design of foundations 2008b). Several correction schemes for size have been developed over
M
resting on and/or in rocks and the stability of rock slopes. The UCS is the years since the beginning of the point load test (e.g. Broch and
typically determined from the axial loading of unconfined intact rock Franklin, 1972; Brook, 1985; Thuro and Plinninger, 2001; ASTM,
specimens. The uniaxial compression test should be performed on 2008b). The specimen size correction proposed by ASTM (2008b) is
cylindrical specimens with a length to width ratio of 2–2.5. The standard implemented in this paper.
D
specimen height should range between 100 mm and 300 mm. The
specimen ends perpendicular to the cylinder axis should be flat, smooth 3. Geology of Dubai in the United Arab Emirates
TE
especially useful when core specimens cannot be extracted from fractured sands, marine sands and silts. The Solid Geology is composed of
or weathered rock masses. Furthermore, the point load test is simpler and alternating beds of calcarenite/ calcareous sandstone with some carbonate
more economical compared to the UCS test. sand bands, gypsiferous sandstone with cemented sand layers, and
calcisiltite and siltstone, from the top down (Beau et al., 2008). A
C
2. The point load test geological map of the general location under investigation is presented in
Fig. 1, which shows the expected stratigraphy of the coastal areas of
AC
The point load test gradually applies a concentrated compressive force Dubai (Alsharhan and Kendall, 2003). It consists of Quaternary marine,
using conical platens on the rock specimen until it fails by splitting. The aeolian, sabkha and fluvial deposits that lie on the top of aeolianite and
loading system consists of a loading frame, a hydraulic jack, and a marine calcarenite (Ghayathi formation) which overlay the Barzman
pressure gauge. The test can be conducted in the field or in the laboratory. formation that is comprised of fluvial sediments characterized by poorly
The failure load is recorded and used to calculate the point load strength sorted conglomerates (Macklin et al., 2012). The rocks considered in this
index Is, according to the following equation (ASTM, 2008b): study belong to the Ghayathi formation described as marine calcarenite.
I s = Pult / De2 (1)
where Pult is the failure load (N), and De is the equivalent core diameter 4. Mechanical and chemical characteristics of Dubai calcareous
(mm). sandstone
It has been found that the point load strength index depends on the
specimen size (Thuro and Plinninger, 2001). According to ASTM Boreholes were drilled at a number of sites at Dubai in the United Arab
(2008b), the standard specimen diameter D is 50 mm. However, it may Emirates. Rock samples were extracted using a double tube core barrel
headed with diamond bit, producing a nominal core diameter of 76 mm.
The boreholes extended into the rock at a depth of approximately 8 m
Corresponding author. Tel: +20 1111921320; E-mail: [email protected]
ACCEPTED MANUSCRIPT
(approximately −13 to −21 DMD (Dubai Municipality Datum)). define the stress-strain response up to failure to determine the
Chemical testing of 21 specimens shows that the calcium carbonate deformation characteristics of intact rock specimens in addition to its
(CaCO3) contents vary between 50.62% and 93.57% with an average of UCS (ASTM, 2002b). The test is conducted by applying an axial load to
71.32%. the specimen and recording the corresponding deformation. Young’s
According to ARGEMA classification of carbonate formations shown modulus (either secant or tangent) can be determined by plotting the
in Table 1 (Le Tirant and Nauroy, 1994), cemented formations with stress-strain curves. In this study, the elastic modulus of the intact rock
carbonate contents higher than 30% are defined as soft to hard carbonate specimens is computed as the tangent value at 50% of the unconfined
rock. For geomaterials with calcium carbonate contents lower than 30%, compressive strength and is known as the characteristic elastic modulus
the material performs as a silicate. Accordingly, it is warranted to (ER50). The measured values of the characteristic elastic modulus (ER50)
describe such materials as “carbonate rock” (Le Tirant and Nauroy, 1994). vary between 34.4 MPa and 1911.5 MPa with an average of 372 MPa,
The unconfined compressive strength is generally considered as the which are within the typical values reported for sedimentary rocks
PT
reference value for the compressive strength which is typically measured according to AASHTO (2012).
using uninstrumented uniaxial compression tests (ASTM, 2002a). The For the tested specimens, the Poisson’s ratio ranged between 0.22 and
recorded values of the unconfined compressive strengths varied between 0.365 with an average of 0.297. These values lie within the typical range
RI
0.13 MPa and 15.75 MPa. Thus, the tested rock specimens were of 0.1 0.46, with a mean of 0.29 proposed by Carmichael (1982) and
classified as weak to very weak ones according to Mayne et al. (2001). Mayne et al. (2001).
Alternatively, instrumented uniaxial compression tests may be used to
C
US
AN
M
D
TE
C EP
AC
Fig. 1. Geological map of the United Arab Emirates showing the area under study (Alsharhan and Kendall, 2003).
Table 1. ARGEMA classification of carbonate formations (Le Tirant and Nauroy, 1994).
Designation of materials Grain size Carbonate content Cementation
Fine carbonate sediments (muds and fine silts) From 1 µm to 2 µm Greater than 10% Uncemented or weakly cemented
Uncemented or weakly cemented
From a few ten microns to
Carbonate sands Greater than 30%
a few millimeters or irregularly cemented
Cemented carbonate formations All grain sizes Greater than 30% Soft to hard rock
ACCEPTED MANUSCRIPT
5. Existing correlations between the point load strength index and with zero intercept, qu=kIs50 (e.g. Broch and Franklin, 1972; Bieniawski,
UCS of rock 1975; Beake and Suttcliffe, 1980; Puech et al., 1988). The value of the
constant k varies between 3.1 and 25. This means that using the
A plethora of correlations for estimating the UCS using the point load inappropriate equation may result in overestimating the UCS by
strength index Is50 have been developed over time. A selected summary of approximately 800%. Other equations used to express the correlation
the available UCS–point load strength index correlations is presented in between Is50 and the UCS also include linear with non-zero intercept (e.g.
Table 2. More details about these relationships are given in Kim et al., 2004; Cha et al., 2007), power (e.g. Tsiambaos and
corresponding references. One of the earliest and most widely used Sabatakakis, 2004; Min and Moon, 2006; Salah et al., 2014), and
correlations was proposed by Broch and Franklin (1972) and Bieniawski exponential equations (e.g. Salah et al., 2014). The estimated UCSs using
(1975), based on tests performed on different rock types. Since then, these correlations result in values with large variability. Thus, it is highly
several other correlations were formulated either for different types of recommended to develop correlations to be used for specific rock types
PT
rocks or for local formations. Most of the developed equations are linear and local geologies.
Table 2. Correlations between the point load strength index Is and the UCS.
Rock type Correlation Reference Rock Type Correlation Reference
RI
Broch and Franklin (1972), qu = 16.656Is50 + 21.7 for qu =
Hard strong rock qu = 24Is50 Gneiss Cha et al. (2007)
Bieniawski (1975) 29 209.3 MPa
Calcarenites of the Arabian-
qu = 12.2Is50 Beake and Suttcliffe (1980) All rock types qu = 24Is50 ASTM (2008b)
Persian Gulf
Calcarenites of the Arabian- Limestone of Imgye area,
C
qu = 3.1Is50 Puech et al. (1988) qu = 26Is50 Kim et al. (2012)
Persian Gulf Korea
Egyptian Code for Soil
Mechanics, Design, and qu = 22.8Is50
All rock types qu = 25Is50 Quartzite
Construction for qu = 32.5–98.9 MPa
US
of Foundations (2001)
qu = 21.431Is50 qu = 21.9Is50
Mudstones Kahraman (2001) Sandstone
for qu = 2–115 MPa for qu=17.6–56.4 MPa
Singh et al. (2012)
Innsbruck quartz phyllite qu=14.4Is50
qu = 19.9Is50 Shale
(perpendicular to foliation) for qu= 9.9–18.8 MPa
Thuro and Plinninger (2001)
qu = 14.4Is50
AN
All rock types qu = 18.7Is50 Gabbro
for qu = 17.3–137 MPa
qu = 22Is50 + 49 qu = 22.3Is50
Tuffs Kim et al. (2004) Limestone
for qu = 80.4–208 MPa for qu = 86.9–129.8 MPa
Limestones, marlstones and 1.71 Meta-siltstone and meta-
q u = 7.3I s50 qu = (20–21)Is50 Li and Wong (2013)
sandstones Tsiambaos and Sabatakakis sandstone
Limestones, marlstones and (2004)
qu = 23Is50 All rock types qu = 9.459Is50
M
sandstones
1.71 Salah et al. (2014)
q u = 7.3I s50 for Is50 > 3.5 MPa 0.75
Sedimentary rocks Min and Moon (2006) All rock types q u = 9.459 I s50
and qu = 2.1–254 MPa
Dandot sandstone, Sakkessar
qu = 11.076Is50 Akram and Bakar (2007) Very weak rock qu = 5.833sqrt (0.57Is50)
sandstone, marl
D
Elkateb (2009)
Siltstone qu = 8.6Is50
Very weak sandstone qu = 5.679 sqrt (Is50)
Limestone qu = 6.5Is50
6. Correlation between the UCS and point load strength index for q u = 2.86 ls50 (3)
Dubai calcareous rock Eq. (3) has an important advantage over Eq. (2) which yields a non-
zero value for the UCS at Is50=0, implying that the rock has a quantifiable
The relationship between the point load strength index (Is50) and the strength although the measured strength is zero. The developed
C
UCS (qu) for Dubai calcareous sandstone is shown in Fig. 2. As expected, correlation is in close agreement with the equation proposed by Puech et
the UCS increases with the point load strength index (Is50). Linear al. (1988) for calcarenites of the Arabian-Persian Gulf. Considering the
AC
regression analysis is used to provide mathematical representation of the correlations reported for extremely soft rock (e.g. Puech et al., 1988;
correlation between the point load strength index and the UCS. The best Elkateb, 2009; Salah et al., 2014), a conversion factor (K) of 2.86 6.5 is
fit correlation is represented by Eq. (2). The coefficient of determination suitable for such formations.
(R2) for this equation is 0.65, which indicates a relatively good
relationship. It is also shown that most of the measurements are bounds
by one standard deviation around the regression line. The standard error
(calculated as the ratio of the standard deviation to the mean of the
collected data) was computed as 40.6%.
q u = 2.59 ls50 + 0.21 (2)
Alternatively, Eq. (3) represents a linear relationship between the UCS
and point load strength index with a zero intercept with the results
presented in Fig. 3. This correlation has almost the same coefficient of
determination of 0.64 compared to Eq. (2). The standard error associated
ACCEPTED MANUSCRIPT
The relationship between the rock characteristic elastic modulus ER50
and the UCS is investigated using the ratio of rock elastic modulus to the
UCS (modulus ratio), which was first introduced by Deere (1968). The
modulus ratio of the tested samples varied between 81 and 265 with an
average of 140, which lie within the typical values reported by Mayne et
al. (2001). The measured UCSs are plotted versus the characteristic
elastic modulus ER50 as presented in Fig. 4. The data are compared to the
trend lines reported by Deere and Miller (1966) for sedimentary rocks.
This classification describes the rock to be having high modulus ratio,
which is greater than 1:500. For modulus ratios smaller than 1:200, the
PT
rock is described to be having low modulus ratio. Accordingly, the tested
specimens, generally, have low modulus ratios. Lower values of the
modulus ratio indicate that the rocks can sustain higher deformation
Fig. 2. Correlation between the point load strength index and UCS for Dubai calcareous
before fracturing compared to rock specimens with higher modulus ratios
RI
sandstone (linear correlation with non-zero intercept).
(Tziallas et al., 2009).
C
US
AN
M
Fig. 3. Correlation between the point load strength index and UCS for Dubai calcareous
sandstone (linear correlation with zero intercept).
D
Bieniawski (1975), which is one of the commonly used formulae, Fig. 4. Characteristic elastic modulus versus the UCS of Dubai calcareous sandstone.
overestimates the UCS by approximately 800% when used for such rock A best fit linear equation is used to represent the relationship between
formations. Such a large error is unacceptable and emphasizes the need the rock elastic modulus and UCS, as presented in Fig. 5, yielding the
for developing local correlations to estimate rock strength based on the following relationship:
EP
to its heterogeneous nature as indicated by Idris et al. (2011). The degree be 124 times its UCS.
of natural variation in a specific rock property depends on many factors
such as the type of rock, the degree of weathering, the property
investigated, etc. Kulhawy and Prakoso (2001) stated that there are
substantial depositional variations in calcareous rocks leading to
considerable ranges in rock properties. Their study showed that the
coefficient of variation reaches 39% and 59% for the strengths of
calcareous rocks obtained using uniaxial compression and point load tests,
respectively. Thus, the error in the predicted strengths may be partly
explained by the natural variability in the rock properties.
PT
USA: ASTM International, 2002a
ASTM. D3148 Standard test method for elastic moduli of intact rock core specimens in
uniaxial compression. West Conshohocken, USA: ASTM International, 2002b.
RI
ASTM. D4543 Standard practices for preparing rock core as cylindrical test specimens
Fig. 5. Correlation between the characteristic elastic modulus and the UCS of Dubai and verifying conformance to dimensional and shape tolerances. West Conshohocken,
calcareous sandstone. USA: ASTM International, 2008a.
ASTM. D5731 Standard test method for determination of the point load strength index of
C
rock and application to rock strength classifications. West Conshohocken, USA: ASTM
8. Conclusions International, 2008b.
US
Beake RH, Suttcliffe G. Pipe pile drivability in the carbonate rocks of the southern
The point load test provides a fast and economic method for evaluating Arabian Gulf. In: Proceedings of the International Conference on Structural
the UCS of rock specimens. The point load strength index (Is50) is Foundations on Rock, Sydney, Australia. Rotterdam, Netherlands: A.A. Balkema, 1980.
Beau P, Maghazy M, Chiriotti E. Importance of geological and geotechnical model
empirically correlated to the UCS. There is a plethora of correlations for
AN
upgrading for the tunnel/station interface during the TBM breakthrough on the Dubai
estimating the UCS from the point load strength index, as summarized in
metro project. In: Proceedings of the World Tunnel Congress 2008—Underground
Table 1. A review of published literature shows the available correlations
Facilities for Better Environment and Safety. 2008. pp. 1130–8.
including linear, exponential and power functions. Existing correlations
Bieniawski ZT. The point-load test in geotechnical practice. Engineering Geology 1975;
are typically developed using specific data of certain rock types (e.g.
M
9(1): 1–11.
Tsiambaos and Sabatakakis, 2004) or regions (e.g. Puech et al., 1988; Broch E, Franklin JA. The point-load strength test. International Journal of Rock
Elkateb, 2009). Alternatively, some of the published correlations are Mechanics and Mining Sciences and Geomechanics Abstracts 1972; 9(6): 669–76.
based on data collected from all rock types and/or different regions (e.g. Brook N. Size correction for point load testing. International Journal of Rock Mechanics
D
Thuro and Plinninger, 2001). The literature review shows that there is a and Mining Sciences and Geomechanics Abstracts 1980; 17(4): 231–5.
large degree of variability in estimating the UCS from the point load Brook N. The equivalent core diameter method of size and shape correction in point load
TE
strength index. Based on the data collected in Dubai, Eqs. (2) and (3) testing. International Journal of Rock Mechanics and Mining Sciences and
were derived for estimating the UCS of Dubai calcareous sandstone from Geomechanics Abstracts 1985; 22(2): 61–70.
Carmichael RS. CRC handbook of physical properties of rocks (Volume 2). CRC Press,
the point load strength index. As discussed above, it is shown that the
1982.
developed Is50-qu correlation is considered acceptable despite the
EP
Cha AR, Song JH, Choi HS, Chun BS. The study on the correlation between unconfined
relatively large scatter. It is in close agreement with the equation
compressive strength and point load strength within the geniss complex in Gyunggi
proposed by Puech et al. (1988) for calcarenites of the Arabian-Persian
Province. Journal of the Korean Geo-Environmental Society 2007; 8(1): 21–6 (in
Gulf.
Korean).
Furthermore, it is also shown that the tangent elastic modulus of the
C
Deere DU. Geological consideration. In: Stagg KG, Zienkiewicz OC editors. Rock
intact rock specimens at 50% of the unconfined compressive strength Mechanics in Engineering Practice. New York: John Wiley & Sons Ltd., 1968. pp. 1–
(ER50) is approximately equal to 125.7qu. Thus the results of the point
AC
20.
load test are used to estimate the strength and deformation parameters of Deere DU, Miller RP. Engineering classification and index properties of intact rock.
intact rock through the proposed empirical relationships. Finally, it is Technical Report No. AFWL-TR-65-116. Urana, USA: University of Illinois, 1966.
important to develop site-specific correlations for different rock types Diamantis K, Gartzos E, Migiros G. Study on uniaxial compressive strength, point load
because of the large variation in the published correlations. Using strength index, dynamic and physical properties of serpentinites from Central Greece:
Test results and empirical relations. Engineering Geology 2009; 108(3–4): 199–207.
“generic” correlations could lead to unrealistic evaluations of rock
Egyptian Code for Soil Mechanics, Design, and Construction of Foundations, Volume 2.
strength and deformation parameters.
Cairo, Egypt: Ministry of Housing, 2001. pp. 272–6.
Elkateb T. The point load test in the Persian Gulf carbonate rocks. In: Proceedings of the
Conflict of interest 17th International Conference on Soil Mechanics and Geotechnical Engineering,
Volume 1. 2009. pp. 372–5.
The authors wish to confirm that there are no known conflicts of Idris MA, Saiang D, Nordlund E. Numerical analyses of the effects of rock Mass property
variability on open stope stability. In: The 45th US Rock Mechanics /Geomechanics
interest associated with this publication and there has been no significant
Symposium, San Francisco, USA. American Rock Mechanics Association (AMRA),
financial support for this work that could have influenced its outcome.
2011. pp. 1530–40.
ACCEPTED MANUSCRIPT
Kahraman S. Evaluation of simple methods for assessing the uniaxial compressive strength of Highway Administration, 2001.
rock. International Journal Rock Mechanical and Mining Sciences 2001; 38(7): 981–94. Min TK, Moon JK. A review of strength estimation method on Ulsan sedimentary rocks.
Kim HG, Koh YK, Oh KH. A study on the mechanical properties of the cretaceous tuffs in Journal of the Korean Geotechnical Society 2006; 22(8): 63–75 (in Korean).
Goheung area. The Journal of Engineering Geology 2004; 14(3): 273–85 (in Korean). Puech A, Beuce JP, Colliat JL. Advances in the design of piles driven into non-cemented
Kim GM, Kim DH, Kang JS, Kang SS, Baek H. A study of the correlation between to weakly cemented carbonate formations. In: Proceedings of the International
uniaxial compressive strength and point load strength index of limestone of Imgye area. Conference on Calcareous Sediments, Volume 1, Perth, Australia. 1988. pp. 305–12.
Tunnel and Underground Space 2012; 22(5): 330–8 (in Korean). Salah H, Omar M, Shanableh A. Estimating unconfined compressive strength of
Kulhawy FH, Prakoso WA. Foundations in carbonate rocks and karst. In: Brandon TL sedimentary rocks in United Arab Emirates from point load strength index.
editor. Foundations and Ground Improvement. Reston, USA: American Society of Civil Journal of Applied Mathematics and Physics 2014; 2: 296–303.
Engineers (ASCE), 2001. pp. 1-15. Singh TN, Kainthola A, Venkatesh A. Correlation between point load index and uniaxial
Le Tirant P, Nauroy JF. Design guides for offshore structures: Foundations in carbonate compressive strength for different rock types. Rock Mechanics and Rock Engineering
PT
soils. Technip Editions, 1994. 2012; 45(2): 259–64.
Li D, Wong LNY. Point load test on meta-sedimentary rocks and correlation to UCS and Thuro K, Plinninger RJ. Scale effects in rock properties. Part 2: Point load test and point
BTS. Rock Mechanics and Rock Engineering 2013; 46(4): 889–96. load strength index. In: Rock Mechanics–A Challenge for Society, Proceedings of the
Macklin S, Ellison R, Manning J, Farrant A, Lorenti L. Engineering geological ISRM Regional Symposium EuroRock 2001. CRC Press, 2001. pp. 175–80.
RI
characterisation of the Barzaman Formation, with reference to coastal Dubai, UAE. Tsiambaos G, Sabatakakis N. Considerations on strength of intact sedimentary rocks.
Bulletin of Engineering Geology and the Environment 2012; 71(1): 1–19. Engineering Geology 2004; 72(3–4): 261–73.
Mayne PW, Christopher BC, DeJong J. Manual on subsurface investigations– National Tziallas GP, Tsiambaos G, Sarglou H. Determination of rock strength and deformability of
C
Highway Institute Publication No. FHWA NHI-01-031. Washington, DC, USA: Federal intact rocks. Electronic Journal of Geotechnical Engineering, 2009; 14: 1–12.
Dr. Amr Farouk Elhakim is an assistant professor at the Faculty of Engineering, Cairo University. In July
US
1995, he earned his B.Sc. in Civil Engineering with honors degree. He worked as a teaching / research
assistant in the Soil Mechanics and Foundations Research Laboratory of Cairo University. In the same
year, he enrolled for a Masters program in Geotechnical Engineering in Cairo University, where he earned
his M.Sc. in July 1998. He then joined the Geosystems Engineering Group at Georgia Tech, where he
worked under the supervision of Prof. Paul W. Mayne. In May 2001, he earned another Masters in
AN
Geotechnical Engineering from Georgia Tech. He then finished his Ph.D in Summer 2005 from the same
university. He has several publications in the fields of in-situ testing, deep excavations, design of axially
and laterally loaded piles, displacements of shallow foundations and the use of small strain soil modulus in
foundation design. Furthermore, he has worked in geotechnical consulting in several projects.
M
D
TE
C EP
AC