applications
Sources
This chapter is made up of verbatim extracts from the following sources for which copyright permissions have been obtained as listed in the Acknowledgements.
• Feng, M. and Fredlund, D.G. (2003). Calibration of thermal conductivity sensors with consideration of hysteresis. Can. Geotech. J, 40, 1048–1055.
• Fredlund, D.G. and Rahardjo, H. (1993). Soil mechanics for unsaturated soils. John Wiley & Sons, Inc., New York, p. 517.
• Harrison, B.A. and Blight, G.E. (2000). A comparison of in situ soil suction measurements. Proc. Conf. on Unsaturated Soils for Asia, Singapore, Rahardjo, Toll and Leong eds, Balkema, Rotterdam, pp. 281–284.
• Ng, C.W.W. and Chen, R. (2005). Advanced suction control techniques for testing unsaturated soils (in Chinese). Keynote lecture, 2nd Nat.
Conf. on Unsat. Soils, Hangzhou, China, Zhejiang University Press, 144–167.
• Ng, C.W.W. and Chen, R. (2006). Advanced suction control techniques for testing unsaturated soils (in Chinese). Chinese J. of Geotech. Eng., Vol. 28, No. 2, 123–128.
• Ng, C.W.W. and Pang, Y.W. (2000a). Experimental investigation of soil–water characteristics of a volcanic soil. Can. Geotech. J., 37(6), 1252–1264.
• Ng, C.W.W., Cui, Y.J., Chen, R. and Delage, P. (2007). The axis-translation and osmotic techniques in shear testing of unsaturated soils:
a comparison. Soils and Foundations. Vol. 47, No. 4, 675–684.
• Ridley, A.M. and Wray, W.K. (1995). Suction measurement: a review of current theory and practices. Proc. Conf. on Unsaturated Soils, Paris, Alonso and Delage eds, Balkema, Rotterdam, pp. 1293–1322.
• Ridley, A.M. and Burland, J.B. (1993). A new instrument for the mea-surement of soil moisture suction. Géotechnique, 43(2), 321–324.
Theory of soil suction (Fredlund and Rahardjo, 1993)
Soil consists of solids and voids. Inside the voids of unsaturated soil, there are generally two fluids, i.e. air and water. As discussed and given by Equation (1.1) in Chapter 1, total suction can be quantitatively described by Kelvin’s equation (Sposito, 1981) as follows:
= − RT
w0vln
¯uv
¯uv0
(1.1)
At 20C, Equation (1.1) can be rewritten to give a fixed relationship between total suction in kilopascals and relative vapour pressure as follows:
= −135 022 ln
¯uv
¯uv0
(2.1)
The capillary model
Equilibrium at the soil–water interface (Ridley and Wray, 1995)
The commonly depicted picture of the soil–water system is that of granular particles separated by water. At an air–water interface, a meniscus will form between adjacent soil particles in a similar manner to water in a capillary tube (Ridley and Wray, 1995). It is therefore understandable that this model became known as the capillary model (Buckingham, 1907).
For equilibrium at the air–water interface of the capillary tube shown in Figure 2.1, the downward force exerted by the air must be equal to the upward force exerted by the water. The curved shape of the interface is the result of the upward force which exists at the boundary due to the wetting
α α
Capillary tube
u = uw (water) u = ua (air)
τ τ
r
Figure 2.1 Capillary suction (after Ridley and Wray, 1995).
Measurement and control of suction: methods and applications 33 of the surface. It is this upward force that holds the column of water up above the flat water surface outside the tube. Therefore
uar2= uwr2+ 2rTssin (2.2)
where Tsis the surface tension at the boundary, and is the angle of contact between the water and the boundary. For a perfectly spherical meniscus, the latter will be 90 and the equilibrium will reduce to
ua− uw=2Ts
r (2.3)
In the majority of cases, the air pressure will be atmospheric (or zero gauge), and the pressure which exists on a water molecule in the meniscus is a direct result of the surface tension and the radius of the capillary.
Now consider the equilibrium that exists at the air–water interface between the liquid water molecules and the water vapour molecules. To escape from the surface of the liquid, a water molecule must have an energy equal to or greater than the latent heat of evaporation for water. If the space above the air–water interface is a closed system, equilibrium will be reached when it becomes saturated with water vapour molecules. If the water is pure and its surface is flat, then the partial pressure of the vapour at equilibrium is equal to the saturated vapour pressure of the liquid at the temperature of the system. However, soil–water is held in a meniscus and so the additional force caused by the surface tension of the curved surface reduces the vapour pressure reached in the enclosed system at equilibrium.
The stress holding a water molecule in the meniscus (i.e. the soil suction) is then directly related to the relative humidity (RH) in the space surrounding the soil by Equation (1.1).
Equilibrium of the water column and capillary height (Fredlund and Rahardjo, 1993)
For the vertical force equilibrium of the capillary water in the tube shown in Figure 2.1, the vertical resultant of the surface tension (i.e. 2rTscos ) is responsible for holding the weight of the water column, which has a height of hc (i.e. r2hcwg):
2rTscos = r2hcwg (2.4)
where
hc= capillary height
g= gravitational acceleration.
The equation can be rearranged to give the maximum height of water in the capillary tube, hc:
hc=2Tscos
wgr (2.5)
The contact angle between the contractile skin for pure water and tube is zero (i.e. = 0). When the angle is zero, the capillary height of pure water in a clean tube is
hc= 2Ts
wgr (2.6)
The radius of the tube is analogous to the pore radius in soils. Equation (2.6) shows that the smaller the pore radius in the soil, the higher will be the capillary height, as illustrated in Figure 2.2.
Assuming that the contact angle is zero, the capillary height can be plotted against the pore radius as shown in Figure 2.3 (Fredlund and
Radius of meniscus 0.0002 cm
Radius of meniscus 0.002 cm Magnified
soil particles
735 cm
73.5 cm
7.4 cm
1.5 cm
Radius of meniscus 0.02 cm
Radius of meniscus 0.1 cm
Figure 2.2 Capillary tubes showing the air–water interfaces at different radii of cur-vature of menisci (from Janssen and Dempsey, 1980; after Fredlund and Rahardjo, 1993).
Measurement and control of suction: methods and applications 35
Clay Silt Sand
Surface tension, Ts = 72.75 mN/m at t = 20¼C
Pore radius, r (mm)
Matric suction, (ua – uw) (kPa) Capillary height, hc (m)
102
10
0
10–1
10
0
10–1
10–2
10–3 10–2 10–1 0
Figure 2.3 Relationship between pore radius, matric suction and capillary height (after Fredlund and Rahardjo, 1993).
Rahardjo, 1993). The above explanation has demonstrated the ability of the surface tension to support a column of water, hc, in a capillary tube.
The surface tension associated with the contractile skin results in a reaction force on the wall of the capillary tube as shown in Figure 2.4 (Fredlund and
Compressive stress on the wall
Water Air Ts
Ts
Glass tube
Figure 2.4 Forces acting on a capillary tube (after Fredlund and Rahardjo, 1993).
Rahardjo, 1993). The vertical component of this reaction force produces compressive stresses on the wall of the tube. In other words, the weight of the water column is transferred to the tube through the contractile skin. In the case of a soil having a capillary zone, the contractile skin results in an increased compression of the soil structure. As a result, the presence of matric suction in an unsaturated soil increases the shear strength of the soil, pro-vided there is sufficient contractile skin present in the soil (i.e. the soil is not too dry).
Suction control and measurement methods (Ng and Chen, 2005, 2006; Ng et al., 2007a)
Axis-translation technique
Overview
In laboratory studies on unsaturated soils, an important issue is how to control or measure suction in an unsaturated soil specimen. Generally, total suction can be controlled by using the humidity control technique (Esteban and Saez, 1988). Matric suction can be controlled by using the axis-translation technique (Hilf, 1956) and the osmotic technique (Zur, 1966). Osmotic suction can be controlled by using different solutions as pore fluids or changing solute concentrations of pore fluid in the soil.
In most geotechnical engineering applications, chemistry of pore fluids in the soil is not changed and soil–water content varies within a range in which concentrations of pore fluids are not altered significantly, and so osmotic suction appears not to be sensitive to changes in soil–water con-tent. Therefore, it is expected to control total suction and matric suc-tion in most geotechnical testing for unsaturated soils. The most com-monly used technique is axis-translation, followed by osmotic and humidity control.
In this section, the working principle, the development and applications of the three suction control techniques in the laboratory are introduced and reviewed. Experimental data using the axis-translation and osmotic techniques are compared and discussed. No matter which technique is used, suction equalization is a vital stage in testing unsaturated soils. To illustrate the influence of suction equalization on subsequent shearing behaviour, two direct shear tests were performed on a compacted expansive soil applying different durations of suction equalization under the same applied vertical stress and matric suction.
Measurement and control of suction: methods and applications 37
Working principle
Matric suction may be considered as an important variable in defining the state of stress in an unsaturated soil. Therefore, it is necessary to control or measure matric suction in laboratory studies on unsaturated soils. How-ever, difficulties associated with the measurement and control of negative pore water pressure present an important practical limitation. Water is normally thought to have little tensile strength and may start to cavitate when the magnitude of gauge pressure approaches −1 atm. Under some suitable conditioning (see Chapter 1), water can stand tensions of the order of 40–300 atm (Temperley and Chambers, 1946; Young, 1989). As cavita-tion occurs, water phase becomes discontinuous, making the measurements unreliable or impossible. Because it is required to control the matric suc-tion variable over a range far greater than 1 atm for many soil types and their applications, alternatives to measurement or control of negative water pressure are desirable.
Hilf (1956) introduced the axis-translation technique of elevating pore air pressure ua to increase pore water pressure to be positive, preventing cavitation in the water drainage system. Total stress is increased together with air pressure at the same amount so the net stress − ua remains unchanged. As shown in Figure 2.5, stresses on an unsaturated soil in the field are total stress 1, pore air pressure u1a(generally equal to atmospheric pressure) and pore water pressure u1w (generally negative gauge pressure).
When applying the axis-translation technique, total stress is increased from 1 to 2, pore air pressure is increased from u1a to u2a, and pore water pressure is increased from u1w to u2w (generally positive gauge pressure).
The net stress − ua and matric suction ua− uw remain unchanged.
This process is referred as ‘axis-translation’. Based on the axis-translation principle, the matric suction variable ua−uw can be controlled over a range far greater than the cavitation limit for water under negative pressure.
Unsaturated soil In the field
Soil specimen In the lab
σ1 u1a
u1w < 0
σ1 – u1a = σ2 – u2a
u1a – u1w = u2a – u2w
σ2 u2a
u2w > 0
Figure 2.5 Schematic diagram illustrating the axis-translation principle (Ng and Chen, 2005).
Axis-translation is accomplished by separating the air and water phases of the soil through porous material with a high air-entry value. When saturated, these materials allow water passage but prevent flow of free air when the applied matric suction does not exceed the air-entry value of the porous material, which can be as high as 1,500 kPa for sintered ceramics or 15 MPa for special cellulose membranes (Zur, 1966).
Applications
MEASUREMENTS OF SWCC AND SDSWCC
The axis-translation technique has been successfully applied by numerous researchers to study the soil–water characteristic properties of unsaturated soils (Fredlund and Rahardjo, 1993; Ng and Pang, 2000a,b), as well as the volume change and shear strength properties of unsaturated soils (Fredlund and Rahardjo, 1993; Gan et al., 1988; Ng and Chiu, 2001; Ng and Chiu, 2003a,b; Chiu, 2001; Zhan, 2003; Ng and Zhou, 2005).
The soil–water characteristic curve (SWCC) is the relationship between suction and water content or degree of saturation for an unsaturated soil.
It is now generally accepted that unsaturated soil behaviour is governed by two independent stress state variables, i.e. net stress and matric suction (Fredlund and Morgenstern, 1977). Therefore, it is necessary to consider the influence of net stress on the SWCC. However, the SWCC of a soil is conventionally measured by means of a pressure plate extractor in which no external stress is applied, and volume change of the soil specimen is assumed to be zero. To investigate the influence of net stress on SWCC, Ng and Pang (2000a,b) developed a total stress controllable one-dimensional volumetric pressure plate extractor based on the axis-translation principle at the Hong Kong University of Science and Technology (HKUST) (see Figure 2.6). This apparatus can be applied to measure the SWCCs at various vertical stresses under K0 condition. An oedometer ring equipped with a high air-entry ceramic plate at its base is located inside an airtight chamber. Vertical stress is applied through a loading frame to a soil specimen inside the oedometer ring. To eliminate the error due to side friction of the loading piston, a load cell is attached near the end of the piston inside the airtight chamber for determining the actual vertical load applied to the soil specimen. Because the radial deformation is zero for the K0 condition, the total volume change of the specimen is measured from the vertical displacement of the soil spec-imen using a dial gauge. Using this apparatus, state-dependent soil–water characteristics curves (SDSWCCs) can be measured, and the assumption of zero volume change is no longer required. Similar to the conventional vol-umetric pressure plate extractor, pore air pressure ua is controlled through a coarse porous stone together with a coarse geotextile located at the top of the specimen. Pore water pressure uw is controlled at the atmospheric
Apply load
To data acquisition system
Supply air pressure through vapour saturator Coarse porous stone
2 bar high air entry disc
Dial gauge 'O' rings
Load cell Air trap
Burette
Level mark Specimen
Oedometer ring
(a) Schematic diagram
Ballast tube Level mark
Attachments Loading frame
Dial gauge
Chamber
Flushing system
(b) Photograph
Figure 2.6 A new total stress controllable one-dimensional volumetric pressure-plate extractor at HKUST (after Ng and Pang, 2000b; Ng and Chen, 2005).
pressure through the high air-entry ceramic plate mounted at the base of the specimen. In addition, some attachments are used for the purpose of study-ing the hysteresis of the SWCCs associated with the drystudy-ing and wettstudy-ing of the soil. These are a vapour saturator, air trap, ballast tube and burette. The vapour saturator is used to saturate the in-flow air to the airtight chamber to prevent the soil from drying by evapouration. The air trap is attached to collect air that may diffuse through the high air-entry disc. The ballast tube serves as a horizontal storage for water flowing in or out of the soil specimen. The burette is used to store or supply water and to measure the water volume change in the soil specimen.
Figure 2.7 shows the influence of stress state on the soil–water char-acteristics of natural completely decomposed volcanic (CDV) specimens (Ng and Pang, 2000a). The size of the hysteresis loops does not seem to be governed by the applied stress level. The specimens subjected to higher applied stresses possess a slight higher air-entry value and lower rates of desorption and adsorption as a result of smaller pore-size distribution.
These experimental results demonstrate that the stress state has a substantial influence on the soil–water characteristics of unsaturated soils. Ng and Pang (2000b) adopted the measured wetting SDSWCCs to perform numerical analyses on slope stability.
They found that during a prolonged low intensity rainfall, the FOS (fac-tor of safety) predicted by using the SDSWCC is substantially lower than predicted by using the conventional drying SWCC. Under highly intensive but short duration rainfalls, however, the two predicted FOS values are close. Therefore, in studying soil–water characteristics of unsaturated soils
Matric suction (kPa) CDV-N1(C) (0 kPa)
CDV-N2(M) (40 kPa) CDV-N3(M) (80 kPa)
Degree of saturation, S (%)
0.1 1 10 100 1,000
100
95
90
85
80
75
70
Figure 2.7 Influence of stress state on soil–water characteristics of natural CDV specimens (after Ng and Pang, 2000a; Ng and Chen, 2005).
Measurement and control of suction: methods and applications 41 and their applications, the effect of two independent stress state variables, i.e. net stress and suction, should be considered simultaneously.
STRESS PATH TESTING IN THE TRIAXIAL APPARATUS
The triaxial test and the direct shear test (shear box) are two commonly used shear strength tests. Figure 2.8 illustrates a triaxial system applying the axis-translation technique for testing unsaturated soils at HKUST (Zhan, 2003).
It is composed of a triaxial cell, four GDS pressure controllers, six transduc-ers, a digital transducer interface (DTI) and a computer, a new total volume change measuring system and a diffused air volume indicator (DAVI).
As shown in Figure 2.8, the triaxial cell is a Bishop and Wesley stress-path cell for testing up to 100 mm diameter specimens. Of the four GDS pres-sure controllers, two are automatic pneumatic controllers for controlling cell pressure and pore air pressure, and the other two are digital hydraulic
Pneumatic controllers
Digital transducer interface
GDS hydraulic pressure/volume controllers
Diffused air volume indicator Total volume change measuring system Bishop & Wesley
triaxial cell
Figure 2.8 A computer-controlled triaxial system for unsaturated soils based on the axis-translation principle at HKUST (after Ng and Chen, 2005, 2006).
pressure/volume controllers for controlling back pressure and axial stress.
With these four GDS controllers, cell pressure, axial stress, pore air pres-sure and pore water prespres-sure can be controlled independently. The six transducers consist of an internal load cell (measurement of axial force), a linear variable difference transformer (LVDT) (measurement of axial dis-placement), a differential pressure transducer (DPT) (a component of the total volume change measuring system, Ng et al., 2002a) and three pres-sure transducers (monitoring cell prespres-sure, pore water prespres-sure and pore air pressure). All the six transducers are connected to the DTI for data acquisi-tion. The DTI as well as the two pneumatic controllers are connected to the computer through a multiplexer. The two digital hydraulic pressure/volume controllers are connected to the computer by an IEEE interface card for computer control. All of these form a closed-loop controlling and feedback system, which is capable of performing strain-controlled and stress-path tests in triaxial stress space.
As shown in Figure 2.9a, a high air-entry disk was sealed onto the pedestal bottom of the triaxial cell using an epoxy. A spiral groove was produced at the bottom of the pedestal to serve as a water compartment as well as a channel for flushing air bubbles that may be trapped or accumulated as a result of air diffusion (Chiu, 2001; Zhan, 2003). The DAVI is used to measure the amount of diffused air (Fredlund, 1975; Zhan, 2003). Matric suction is applied to the test specimen through one water pressure controller and one air pressure controller using the axis-translation principle. Pore water pressure is applied or measured at the base of the specimen through the ceramic disc and the compartment. Pore air pressure is applied at the top of the specimen through a sintered copper filter.
Figure 2.9 shows the setup of the total volume measuring system (Ng et al., 2002a). The basic principle of the double-cell total volume measur-ing system is that the overall volume change in an unsaturated–saturated specimen is measured by recording the differential pressures between the water inside an open-ended, bottle-shaped inner cell and the water inside a reference tube using a high-accuracy DPT. The inner cell is sealed onto the pedestal of the outer cell in the triaxial apparatus. The high-accuracy DPT is connected to the inner cell and to a reference tube, in order to record changes in differential pressures between the water pressure change inside the inner cell due to a volume change in the specimen and the constant water pressure in the reference tube. Detailed calibrations have been carried out to account for apparent volume changes due to changes in cell pres-sure, fluctuation in the ambient temperatures, creep in the inner cell wall and relative movement between the loading ram and the inner cell (Ng et al., 2002a). The estimated accuracy of the volume change measuring system is in the order of 31 4 mm3 if the system is properly calibrated.
For a 100 mm diameter by 200 mm high test specimen, this corresponds to a volumetric strain of 0.002 per cent. In this system, apparent volume
To pneumatic controller
Load cell Loading ram Outer cell Inner cell
Top cap
Pedestal
Air pressure line Valves
Specimen Line to reference tube
Differential pressure transducer (a) Schematic diagram
'O' ring High air-entry value disc Coarse porous disc Water level in
outer chamber Reference water level Air
Water level in inner chamber
Reference tube
(b) Photograph Inner cell
Differential pressure transducer
Figure 2.9 A new double-cell total volume change measuring system for unsaturated soils at HKUST (after Ng et al., 2002a).
changes due to changes in cell pressure, fluctuations due to variations of ambient temperature and creeping are all smaller than other existing double-cell volume change measuring systems. More detailed comparisons are made by Ng et al. (2002a).
Based on the axis-translation principle, Ng and Chiu (2001) conducted triaxial stress path tests on a loosely compacted unsaturated CDV specimen.
Figure 2.10 shows the results of field stress path (wetting) tests at constant deviator stress (i.e. field stress path tests which simulate rainfall infiltration on a slope element). As shown in Figure 2.10a, when the suction decreases from its initial value (i.e. 150 kPa), there exists a threshold suction above which only small axial strain is mobilized. As the suction drops below this threshold value, the rate of increase in axial strain accelerates towards the end of the test. The threshold suction increases with applied net mean stress. Figure 2.10b shows the variation of volumetric strain with suction.
Similar to the variation of axial strain with suction, there exists also a threshold suction above which only small volumetric strain is mobilized.
As the suction drops below such threshold suction, contractive behaviour is observed for the two specimens (ua1 and ua2) compressed to net mean stress smaller than 100 kPa. In contrast, the other two specimens (ua3 and ua4) show dilative behaviour.
SHEARING TEST IN THE DIRECT SHEAR BOX
Compared to the triaxial test, the direct shear test is simpler to perform and requires shorter test durations due to the smaller drainage paths. Figure 2.11 shows a direct shear apparatus for unsaturated soils based on the axis-translation principle at HKUST. It is modified from a conventional direct shear box for testing saturated soils (Gan et al., 1988; Zhan, 2003). The cylindrical pressure chamber is built of stainless steel and was designed for pressure of up to 1,000 kPa. Three holes are drilled through to provide the necessary housing for a vertical loading ram and two horizontal pis-tons. Teflon ring seals are installed to ensure the airtightness around the loading ram and pistons. A high air-entry value ceramic disk is installed in the lower portion of the shear box. The water chamber beneath the ceramic disk is designed to serve as a water compartment as well as a channel for flushing air, similar to the pedestal of triaxial apparatus for testing unsaturated soils (refer to Figure 2.9). The desired matric suction is applied to a soil specimen by maintaining a constant air pressure in the air pressure chamber and a constant water pressure in the water cham-ber below the ceramic disk. The pore air pressure and pore water pressure in the soil are then allowed to come into equilibrium with these applied pressures. Matric suction in the soil is equal to the difference between the applied air and water pressures. The modified direct shear apparatus has five measuring devices. There are two LVDTs for horizontal and vertical
Measurement and control of suction: methods and applications 45 25
20
15
10
5
0
Suction, s (kPa)
Axial strain (%)
150 100 50 0
(a)
p = 25 kPa (ua1) p = 50 kPa (ua2) p = 100 kPa (ua3) p = 150 kPa (ua4)
6
4
2
0
–2
Volumetric strain (%)
Suction, s (kPa) (b)
150 100 50 0
p = 25 kPa (ua1) p = 50 kPa (ua2) p = 100 kPa (ua3) p = 150 kPa (ua4)
Figure 2.10 Stress path tests on unsaturated CDV triaxial test specimens showing the relationships between (a) axial strain and suction; (b) volumetric strain and suction (after Ng and Chiu, 2001).
displacements, a load cell for shear force, a pressure transducer for water pressure and a water volume indicator for water volume changes in the soil specimen. All these transducers are connected to a data logger for data acquisition.
Figure 2.12 shows direct shear results for a completely decomposed gran-ite (CDG) test specimen (Ng and Zhou, 2005). In Figure 2.12, stress ratio
10 mm Teflon seal Teflon seal Hardened steel plate
Rubber O-ring
Upper box
Lower box
To motor Chamber
cap
186 mm
To load cell
Epoxy
(a) Schematic diagram
273 mm
Soil
Shear box base 220 mm
Chamber body
Teflon seal
Rubber packing Coarse porous disc
Chamber base
Water channels High air
entry disc
Horizontal LVDT
(b) Photograph
Flushing system Vertical LVDT
Loading frame
Chamber
Figure 2.11 A direct shear apparatus for unsaturated soil test specimens: (a) schematic diagram (after Gan et al., 1988); (b) photograph at HKUST (after Zhan, 2003). Interpreting the results uses the axis-translation principle (Ng and Chen, 2005, 2006).
Measurement and control of suction: methods and applications 47
v− ua . Dila-to incremental horizontal displacement x . Negative sign (or negative dila-tancy) means dilative behaviour. As shown in Figure 2.12a, at zero suction and suctions of 10 and 50 kPa, the stress ratio–displacement curve displays strain hardening behaviour. As suction increases, strain softening behaviour is observed at suctions of 200 and 400 kPa. Generally, measured peak and ultimate stress ratios increase with suction, except the ultimate stress ratio measured at suction of 200 kPa. Figure 2.12b shows the effects of suction on dilatancy of the CDG in the direct shear tests. Under the saturated conditions, the soil specimen shows contractive behaviour (i.e. positive dila-tancy). On the other hand, under unsaturated conditions, all soil specimens exhibit contractive behaviour initially but then dilative behaviour as hor-izontal displacement continues to increase. The measured maximum neg-ative dilatancy is increased by an increase in suction. The measured peak stress ratio in each test does not correspond with its maximum negative dilatancy.
Advantages and limitations of the axis-translation method
In the axis-translation technique, both pore water pressure and pore air pressure are controlled and measured independently. This enables controlled
0 0.5 1.0 1.5 2.0 2.5
0 1 2 3 4 5 6 7 8 9 10 (a) Horizontal displacement, Δx (mm)
Stress ratio
s = 0 kPa s = 10 kPa s = 50 kPa s = 200 kPa s = 400 kPa
Figure 2.12 Evolution of (a) stress ratio and (b) dilatancy of CDG subjected to direct shear under different controlled suctions (after Ng and Zhou, 2005).
(b) Horizontal displacement, Δx (mm)
0 1 2 3 4 5 6 7 8 9 10 –0.4
–0.2 0.0 0.2 0.4
s = 0 kPa s = 10 kPa s = 50 kPa s = 200 kPa s = 400 kPa
Dilatancy, δy/δx
Figure 2.12 (Continued).
variation of suction. When a feedback system is used, suction can be con-trolled automatically. The majority of experimental results for unsaturated soils have been obtained by the application of axis-translation technique because of the easy measurement and control of suction.
One limitation of the axis-translation technique pertains to the maximum value of suction that can be applied. It is limited by the maximum value of cell pressure and the air-entry value of porous material. Hence, this technique generally is used for controlling suction in the order of several hundred kilopascals.
Another disadvantage of the axis-translation technique is that by elevat-ing the pore water pressure from a negative to a positive value, the possibil-ity of cavitation is prevented not only in measuring system but also within soil pores. This implies that any influence of cavitation on the pore water by altering the desaturation mechanism of a soil under in situ conditions is not accounted for in the laboratory tests using the axis-translation technique (Dineen and Burland, 1995). It is then fundamental to understand whether experimental results obtained by using the axis-translation technique can be extrapolated to unsaturated soils under atmospheric conditions in the field.
Some researchers (Zur, 1966; Williams and Shaykewich, 1969; Ng et al., 2007) compared experimental data by using the axis-translation technique and the osmotic technique where the pore air pressure is at the atmospheric