Prediction of unconfined compressive strength by ultrasonic testing
Samir M. Ezziyani,
Swedish Geotechnical institute (SGI), Sweden, samir.ezziyani@swedgeo.se Martin Holmén
Swedish Geotechnical institute (SGI), Sweden, martin.holmen@swedgeo.se
ABSTRACT
At the Swedish Geotechnical Institute, soil mechanics laboratory, a number of specimens of stabilised sand with Portland cement have been tested with Ultrasonic testing method prior to unconfined stress tests in order to determine the seismic properties of the stabilised sand.
The objective of this study was to evaluate a non-destructive method, ultrasonic testing, for assessing the compressive wave velocity in stabilised sand in the laboratory. Ultrasonic measurements were done with the sample still in the test tube which allowed continuous
measurements of the curing process in stabilised sand. In parallel to the ultrasonic testing method, the resonant column free-free method was used for comparison. The tests showed that uncemented sand could not transmit an ultrasonic signal. The energy loss (damping) is large in contact points of the particles which prevents the signal from propagating through the specimen. The
compressive wave velocities measured on the stabilised specimens, were not the same for the ultrasound and the resonant column tests. The reason for the difference in compressive wave velocity in the two methods needs further investigations in future studies. The results obtained from this study suggested that ultrasonic wave measurements can be used in combination with other forms of laboratory testing.
Key words: Stabilised soil, bender element, resonant column free-free, ultrasound, UCS
1 INTRODUCTION
Measurement of the strength of stabilised soil is usually done by unconfined compression test (UCT) in laboratory. The objective of this paper is to present a non-destructive testing method, ultrasonic testing method, for prediction of unconfined compressive
strength. The feasibility of using ultrasonic testing to evaluate stabilised sand is
investigated.
The effectiveness of seismic testing methods for assessment of shear modulus and strenght in stabilised soil has been previously reported (Åhnberg and Holmén. 2011). The results showed that the seismic testing methods are well suited for the assessment of
compression as well as shear wave velocity in stabilised soils.
Ultrasonic velocity measurements have previously proven valuable in measuring the development of stiffness of cemented
mixtures, and have been used successfully to monitor the developement of stiffness in natural soil (Alramahi et al. 2008) and concrete mortars (D’Angelo et al. 1995;
Akkaya et al.2003).
Ultrasonic velocity measurements have also been applied to measure ultrasonic wave velocity in cemented paste backfill (Abdullah et al. 2011). The development of p- and s- waves was measured for specimens containing Portland cement, for a week period after mixing.
2 TESTING METHODS 2.1 Resonant column testing
The specimens, having a diameter to length ratio of two, are placed on a soft foam platform. A wave pulse is introduced either as a compression wave or as a shear wave by tapping on one of the end surfaces or on the side of the specimen, respectively.
The basic principle of the resonant column procedure is to excite one end of an
unconfined cylindrical soil specimen by means of transversal or longitudinal excitation (Ryden et al. 2006).
Figure 1 Free-free resonant column set up for frequency measurement. P-wave and S-wave excitation and recording.
Once the transmitted sound waves are recorded with a microphone at the opposite end of the specimen, as shown in Figure 1, the wave propagation velocities are
calculated using theory of wave propagation.
In the special case where the cylindrical specimens’ length is twice the diameter, the wave velocity can be calculated with the following relationship (Nazarian et al.2002).
V p= 2L. f p (1) where
Vp is the compression wave velocity (m/s), L is the specimen length (m), and
f
p is the resonant frequency (Hz).2.2 Ultrasonic velocity measurement An ultrasonic pulse is produced by an
electro-acoustical transducer, which is held in
contact with one surface of the sample under test. The generated pulse from the transducer is transmitted into the sample. Stress waves develop and propagate through the sample.
These waves are detected and converted into signals by a second transducer, and the travel time through the sample length is used to calculate wave velocity, v as:
v = l/ t (2)
where
v is the compression wave velocity (m/s), l is the path length (m),
t is the time taken by the pulse to pass through that length (s).
Figure 2: test set-up. The transducers placed directly opposite to each other on opposite sides of the specimen.
The equipment consists of: P-wave transducers, a pulse-receiver and a digital storage oscilloscope (figure 2).
The instrument’s controls were adjusted into the following settings: Pulse repetition frequency control set to 1 kHz, pulser voltage set to 400, transducer frequency 1 MHz.
3 MATERIALS AND METHODS
3.1 Production of stabilised soil specimens The test material used in this study is
Baskarp sand 25™ with a quartz ratio of
73%. The test specimens were 5 cm in diameter and 10 cm long. The dry sand was mixed with water. Previous studies show that the optimum water content for the stabilised soils varied between 16% and 19% (N.
Yesiller. 2001). In this study the mixtures were prepared at 16% water content. The stabilising agent was Portland cement. Four sets of four samples were prepared with the following amount of Portland cement (PC) as binder: 1%, 3%, 7% and 10%. The dry binder was mixed with the wet sand for five
minutes.
A total of 16 specimens were prepared; 4 samples for each stabilised soil mixture which allowed for multiple specimens to be tested for each testing condition and curing duration. Additional three 10% PC samples were prepared to follow the curing process during the first hours. The stabilised soil mixtures were compacted in plastic tubes, which were sealed and stored in room temperature (22oC).
3.2 Initial testing
Before the main test series, ultrasonic tests were conducted on the plastic tubes to measure the velocity through the plastic material. The reason was to separate the signal through the plastic material from the one through the specimen. A value of wave velocity for each plastic tube that was used in this testing was determined.
3.3 Specimen testing
After the predetermined curing time, one specimen of each mixture was removed from the plastic tube and subjected to ultrasonic testing, resonant column testing and finally to unconfined compression tests. These tests were conducted 24 hours after compaction, subsequent to 7 days, 14 days and 28 days of curing time. Ultrasonic measurements were also conducted on the specimens in their plastic tubes to follow the curing possess during 28 days and results were assembled for further evaluation. The specimens were weighed after compaction and before unconfined compression test to follow the development of the moisture content.
The data from the three different tests was assembled and evaluated continuously during the curing time.
4 TEST RESULTS
The test results showed that uncemented sand could not transmit an ultrasonic signal. The energy loss (damping) is large in the contact points of the particles which prevents the signal from propagating through the sample.
It was also shown that the cemented sand needed to develop a certain degree of
stiffness and compressive strength to transmit a sound wave that is stronger than the sound wave transmitted through the plastic tube.
In this study the minimum compressive strength where the wave velocity could be determined was between 400 and 500 kPa.
The ultrasonic wave velocity through the specimen was between 1500 and 1600 m/s.
Those values were achieved approximately 9 hours after the mixing of sand and cement, for the 10% PC specimens.
4.1 Results evaluation
An example of results from the tests is presented in figure 3. Three lines are drawn to limit the measured maximum and
minimum values for plastic tubes. A first dashed upper line limits the maximum values and a second dashed lower line limits the minimum values. A third solid red line is drawn to function as a limit when the curves deviate from the lower line as much as half the distance between the upper and the lower dashed lines.
There was no received signal during the first hours of curing, but after about 9 hours some strength began to develop in the 7% and 10%
PC specimens and the velocity started to increase. It was difficult to measure any change in velocity in 1% and 3% PC specimens.
A steady change in velocity started to occur 9 hours after sand/cement mixture compaction (figure 3). The curves started to deviate from the average and continued to move towards the solid red line which was taken as a start line to conduct unconfined compressive testing. The first obtained UCT-value was 438 kPa. 2 hours later, a second unconfined
compressive testing was conducted. The maximum compressive strength obtained was 486 kPa.
An example of received signals through 10%
Portland cement during the first 24 h after compaction is shown in figure 4
After about 9 hours the sound through the specimen was distinguished from the sound through the plastic tube.
Figure 3 Variation of wave travel time with time in logarithmic scale (10%PC)
Figure 4 Normalised compressive waveforms through a 10 % PC specimen during the first 9 hours after compaction.
4.1 Moisture content
The changes in moisture content over the curing period for 1%, 3%, 7% and 10 % PC specimens are shown in figure 5. The figure show a rapid decrease in moisture content, of about 30 % of the initial moisture content,
shortly after mixing. After about one week the change decreased to a stable value. That shows that access to water is important for the chemical reactions that occur during the curing process in the first hours.
4.2 P-wave velocities and unconfined compression testing
The variation of velocity with strength for the four different mixtures is presented in figure 6. The velocity increased with strength of the soil. This increase of was more prominent in
the 7% and 10% Portland cement specimens.
The results indicate a relationship between the compressive wave velocity and the unconfined compressive strength.
0 2 4 6 8 10 12 14 16 18
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Time (days)
Moisture content (%)
1% PC 3% PC 7% PC 10% PC
Figure 5 Change in moisture content in samples during the curing time (28 days).
0 500 1000 1500 2000 2500 3000
1000 1500 2000 2500 3000 3500 4000 4500
Velocity (m/s)
quc(kPa)
1% PC 3% PC 7% PC 10% PC
Figure 6 Variation of Strength with ultrasonic velocity measurement.
4.3 RCFF vs. Ultrasonic testing
The resonant column testing was conducted subsequent to ultrasonic testing to compare the two methods. As mentioned before, it was difficult to get any results from ultrasonic testing in the 1% and 3% PC specimen. The same difficulty was encountered when trying
to measure the resonant frequency. Also in this case, it is believed to be due to that the energy loss (damping) is large in the contact points of the particles which prevent the signal from passing through the specimen.
The compressive wave velocities measured on the 7% and 10% specimens, were not the
same for the ultrasound and the resonant column tests. This was not expected. One reason for the different compressive wave velocities in the two methods could be strength anisotropy between axial and radial directions in the specimen. This needs further investigations in future studies.
4.4 Ultrasonic velocity vs. time As the curing process continued the
specimens became stiffer. This is verified by the velocities during the curing period, which increased with time, see figure 7. The
increase was more pronounced during the first 7 days for the 7% and 10% PC samples.
The wave velocity measured in the 10% PC mixture increased rapidly the first 48 hours.
A more stable increase in wave velocity was
measured in the 7% PC mixing during the curing time. In the 1% and 3% PC mixings a very small increase in wave velocity and strength was measured, because of insufficient amount of Portland cement mixed with sand.
4.5 Unconfined compressive strength vs.
time
The strength increased with time and amount of cement. This increase occurred rapidly in the 7% and 10% PC mixings. A smaller increase in strength was measured in the 1%
and 3% PC mixture (figure 8).
1000 1500 2000 2500 3000 3500
0 5 10 15 20 25 30
Curing days
Velocity m/s
1% PC 3% PC 7% PC 10% PC
Figure 7 Development of p-wave velocity (mean value) measured with ultrasonic tests vs. time.
0 500 1000 1500 2000 2500 3000
0 5 10 15 20 25 30
Curing days
quc(kPa)
1% PC 3% PC 7% PC 10% PC
Figure 8 Measured unconfined compressive strength vs. curing days.
5 CONCLUSIONS
A number of specimens of stabilised sand have been investigated by ultrasonic testing.
The tests showed that uncemented sand could not transmit an ultrasonic signal. The energy loss (damping) is large in contact points of the particles which prevents the signal from propagating through the specimen. The cemented sand need to develop a certain degree of stiffness and compressive strength to transmit a sound wave that is faster than the sound wave transmitted through the plastic tube.
In this study (and in this laboratory configuration), the minimum compressive strength that could be resolved with ultrasonic testing was found to be between 400 and 500 kPa. The compressive wave velocity through the specimen was between 1500 and 1600 m/s.
The compressive wave velocities measured on the specimens, were not the same for the ultrasound and the resonant column tests. The reason for the difference in compressive wave velocity in the two methods needs further investigations in future studies.
Additional testing on other types of soils stabilised with different binders need to be performed in order to get a better
understanding of ultrasonic testing of stabilised soil.
ACKNOWLEDGEMENT
The authors thank Helen Åhnberg and David Bendz for their contribution to this paper.
REFERENCES
Abdullah M. Galaa, Ben D. Thompson, Murray W.
Grabinsky, and William F. Bawden. 2011.
Characterizing stiffness development in hydrating mine backfill using ultrasonic wave measurements.
Can Geotech , J 48 1174-1187 (2011).
doi:10.11139/T11-026.
Akkaya, Y., Voigt, T., Subramaniam, K.V., and Shah, S.P. 2003.
Nondestructive measurement of concrete strength gain by an ultrasonic wave reflection method. Materials and Structures, 36 (262): 507–514. doi:10.1617/13854.
Alramahi, B., Alshibli, K.A., Fratta, D., and Trautwein, S. 2008. A suction-control apparatus for the measurement of P and S-wave velocity in soils.
Geotechnical Testing Journal, 31(1): 12–23.
doi:10.1520/GTJ100646.
D’Angelo, R., Plona, T.J., Schwartz, L.M., and Coveney, P. 1995. Ultrasonic measurements on
hydrating cement slurries: onset of shear wave propagation. Advanced Cement Based Materials, 2(1):
8–14. doi:10.1016/1065-7355(94)00015-6.
Nazarian, S., Yuan, D., Tandon, V., and Arrellano, M.
(2002). Quality management of flexible pavement layers with seismic methods. Development of structural field testing of flexible pavement layers.
Research Report 1735-3F, The Center for Highway Materials Research, The University of Texas at El Paso.
Nazli Yesiller, James L. Hanson, and Mumtaz A.
Usmen. "Ultrasonic Assessment of Stabilized Soils"
Proceedings of the ASCE Geo-Institute Soft Ground Technology Conference.. May. 2001.
Rydén, N., Ekdahl, U. and Lindh, P. (2006). Quality control of cement stabilised soil using non-destructive seismic tests. DGZfP – Proc. BB102-CD, Lecture 34, Advanced testing of fresh cementitious materials, Stuttgart 2006.
Åhnberg.H & M. Holmén. (2011).Assessment of strength of stabilised soils by geophysical laboratory testing. Ground improvement. Volume 164 issue GI 3
