Performance of high-capacity tensiometer in constant water content oedometer test
Matric suction is one of the stress-state variables which governs the behaviour of unsaturated soils. In order to evaluate the effect of matric suction on the behaviour of unsaturated soil under constant water content condition, rapid measurement of change in matric suction during the test is required. High-capacity tensiometer can be used to rapidly measure the change in matric suction of unsaturated soil.
In this paper, an oedometer apparatus was modified to incorporate a high-capacity tensiometer to investigate the compression behaviour of unsaturated soil under constant water content condition. The design of the oedometer apparatus was described. Two soils which are kaolin FP grade (commercial kaolin which has a fine grain size) and residual soil from Bukit Timah Granite were used to evaluate the performance of the high-capacity tensiometer.
The results showed that kaolin paste is helpful in improving the contact condition between the soil specimen and high-capacity tensiometer. The equilibrium time of the high-capacity tensiometer was about 10 min. Evaporation effect reported by other researchers was not observed with the high-capacity tensiometer.
The change in matric suction due to loading and unloading was accurately captured by the high-capacity tensiometer.
KeywordsUnsaturated soil High-capacity tensiometer Pore-water pressure measurement Constant water content Oedometer test
The properties of unsaturated soils such as shear strength, volume change and permeability are different from those of saturated soils due to the presence of matric suction s which is the pore-water pressure (uw) referenced to the pore-air pressure (ua). Thus, matric suction measurement is important in investigating the behaviour of unsaturated soil. However matric suction measurement faces challenges such as limited range of matric suction which can be measured, and stability of matric suction reading during measurement.
In order to investigate the behaviour of unsaturated soils, constant matric suction condition is commonly implemented either using the axis-translation technique [8, 11, 14, 21, 22] or by controlling the relative humidity of the soil [5, 11]. The reasons for using constant matric suction instead of constant water content condition are due to the long equilibrium time teq needed to measure matric suction and the difficulties in interpreting the test results for constant water content condition . However, the use of axis-translation technique has its critique. The axis-translation technique hinders cavitation in the unsaturated soil specimen which is unnatural as the air pressure in the field condition is not artificially elevated  and can only be justified if the water is rigid . The effect of cavitation on the unsaturated soil specimen is unclear but some experimental tests using osmotic suction gave different results compared to tests using the axis-translation technique [17, 25, 26].
High-capacity tensiometer (HCT) [6, 7, 19, 20, 24] measures negative pore-water pressure of soil beyond the pressure commonly associated with cavitation and at the same time allows rapid measurements of the pore-water pressure. The principle of HCT is to use a very small water reservoir between the transducer sensing diaphragm and the high air-entry (HAE) disk so that it is difficult for cavitation to take place. By pressurizing the water in the water reservoir at a high pressure, all the remaining air is dissolved. Thus, when the specimen is placed in contact with the high-air entry disk, water in the HCT starts to equilibrate with the water in the soil specimen and the water inside the water reservoir can sustains a high negative pore-water pressure. However, the use of HCT has not been standardized as there are several issues that remain to be solved such as: How small the water reservoir of the HCT should be? What is the best method to saturate the HCT? And, how to ensure good contact between the soil specimen and the HCT?
With regards to the size of the water reservoir of the HCT, at first it is believed that the water reservoir should be extremely small (i.e. 200 μm gap or about 3 mm3 water volume) to inhibit the formation of air bubbles . However, further investigation by Mendes and Buzzi  indicated that it is possible to increase the size of the water reservoir up to approximately 1100 mm3.
Saturating the HCT can be challenging especially when the air-entry value of the ceramic disk is very high (i.e. around 1500 kPa). The most common method used to saturate the HCT is by applying a high water pressure and sometimes followed with applying a vacuum (e.g., . The water pressure used should be high enough such that the water can penetrate the high–air entry disk into the water reservoir and dissolved the air in the water reservoir. Acikel and Mancuso  proposed to circulate CO2 through the HCT. Unfortunately, Acikel and Mancuso  only showed measurement of matric suction of less than 700 kPa, which is less than the maximum matric suction of 1500 kPa that can be measured by using high water pressure to saturate the HCT [7, 15, 19, 23]. Thus, it is not certain whether the proposed procedures of using CO2 can be used to achieve higher matric suction measurement.
It helps to maintain hydraulic conductivity between the water in the specimen and the water in the HCT,
It reduces the evaporation of water in the HCT prior to placing the specimen in contact with the HCT, and
It improves the contact condition between the HCT and the soil specimen especially when the surface of the soil specimen is rough.
The versatility of the HCT is in its small size which allows it to be installed into any device using minimum modification. Thus, HCT can be installed in oedometer and triaxial apparatuses to measure the matric suction. Under constant water content condition, the pore-air pressure of the soil specimen ua is at atmospheric condition. Thus, it is possible for the soil specimen to dry due to evaporation during the test . Such a situation is not desirable as it becomes difficult to interpret the relationship between the measured matric suction and the water content of the specimen. Latex or aluminium membrane covering the exposed surfaces of the soil specimen is commonly used to reduce the evaporation of the water in the soil specimen during the test [9, 16, 20, 23].
In this paper, an oedometer apparatus was modified to incorporate a HCT with efforts to minimize the problems of the HCT described above. Kaolin FP grade (commercial kaolin which has a fine grain size) and residual soil from Bukit Timah Granite were used to evaluate the performance of the HCT.
Evaporation test was performed by repeatedly drying and wetting the HCT and then letting the HCT dry until it cavitates. The drying was conducted by exposing the HCT to the atmosphere while the wetting was conducted by placing a drop of water on the top of the HAE disk of the HCT. The result of the evaporation test is shown in Fig. 2c. The evaporation test indicates that the HCT is responsive and can measure negative pore-water pressure of up to 1300 kPa before it cavitates.
A piece of dry filter paper was placed at the top of the specimen to separate the soil specimen from the dry porous stone. The purpose of using dry filter paper is to let the pore-air be in a drained condition when the soil specimen is loaded in a constant water content condition. On the other hand, once the soil specimen reaches the fully saturated condition, it allows the excess pore-water pressure to drain during compression.
Soil specimens and testing procedures
Basic soil properties of the kaolin and the residual soil
Basic soil properties
In order to minimise sample disturbance, the soil specimen was first cookie cut with a 6.3 cm internal diameter ring that has a sharpened edge. The soil specimen was then air dried until it reached the desired water content. Once the soil specimen reached the desired water content, the soil specimen was then cookie cut with a 5 cm internal diameter oedometer ring. For the kaolin specimen, the specimen was directly placed on the HCT. However, the residual soil specimen has a rough surface and kaolin paste was used to improve the contact condition between the residual soil specimen and the HCT.
Once the soil specimen was placed on the HCT, the pore-water pressure of the soil specimen was monitored until it reached equilibrium. The soil specimen was then loaded incrementally under constant water content condition in order to observe the change in pore-water pressure. The load increment ratio used was unity. The duration for each loading step was determined based on the pore-water pressure reading and was deemed completed when the HCT reached equilibrium. Long loading duration may cause evaporation effects on the soil specimen. If the HCT reads positive excess pore-water pressure, the loading duration follows the conventional saturated oedometer test (approximately 1 day/loading step following  /D2435M (2011) method A).
The kaolin specimens were named using “KwXsY” where K stands for kaolin, w stands for the gravimetric water content, X is the water content of the specimen, s stands for matric suction, and Y is the matric suction of the specimen. The naming of the residual soil specimens from Bukit Timah Granite were named using similar convention that is “BTwXsY” where BT stands for Bukit Timah Granite.
Equilibrium time of kaolin and residual soil specimens
Lourenço et al.  observed three variations of the HCT response during the initial pore-water pressure measurement before it reaches equilibrium which are equilibrium types A, B and C. Equilibrium type A is the normal equilibrium which is desired where the negative pore-water pressure reaches the negative pore-water pressure of the soil specimen slowly. Equilibrium type B is when a higher initial negative pore-water pressure was registered before rebounding to the equilibrium negative pore-water pressure. Lourenço et al.  suggested that this phenomenon is due to the water inside the soil specimen not in equilibrium condition prior the suction measurement. Equilibrium type C is when the negative pore-water pressure never reaches equilibrium. Lourenço et al.  suggested that equilibrium type C is due to the continuous loss of water from the specimen. Such an effect can be attributed to evaporation.
Equilibrium Type B appears to be due to poor contact condition between the HCT and the soil specimen. When the contact condition between the HCT and the soil specimen is poor, some air is trapped in the contact and causes the air pressure to rise beyond atmospheric air pressure and cause the HCT to give a higher suction than the suction of the soil specimen. However, with time the air pressure reduces and the water in the soil specimen equilibrates with the water inside the HCT water reservoir, and the HCT will read the matric suction of the specimen. The concern of this type of equilibrium is that there is a possibility of the HCT cavitating prior to the pore-water pressure rebound to the equilibrium pore-water pressure reading.
Response of the HCT in constant water content oedometer test
There is a difference between the duration required for the specimen to reach the end of primary compression (tend) and the duration required for the pore-water pressure to reach equilibrium (teq) for the saturated and unsaturated states. In Fig. 9, the difference between tend and teq is indicated as ∆t. It is important to ensure that the load is applied until both teq and tend have been reached. Based on the reported tests results, it appears that teq of up to 20 min is sufficient for each load increment for the unsaturated state. For transition and saturated states, it is recommended to use 24 h loading duration, same as conventional oedometer tests, to ensure that the positive excess pore-water pressure has fully dissipated.
Placing the high-capacity tensiometer at the bottom of the specimen avoids effects of evaporation and improves the contact condition between the HCT and the soil specimen.
Using kaolin paste as an interface between HCT and soil specimen reduces the effect of evaporation on the HCT prior to the placement of the specimen and also improves the contact condition of the specimen.
There is a difference between duration for the specimen to reach the end of primary compression (tend) and for the pore-water pressure to reach equilibrium. Based on the test results, it appears that 20 min is sufficient for each load increment when the soil is in the unsaturated state.
Once the soil specimen reaches the saturated state, use 24 h duration between load increments as used in the conventional oedometer test.
In this paper, the performance of a high-capacity tensiometer (HCT) in a modified oedometer apparatus developed to conduct constant water content test on unsaturated soils was evaluated. A modified base plate was constructed in order to incorporate the HCT into the oedometer apparatus to measure the pore-water pressure of the unsaturated soil specimen. The modified base plate was made such that the HCT is at the bottom of the specimen to reduce evaporation effect on the HCT and also to ensure a good contact condition between the HCT and the soil specimen.
Two consolidated specimens which are kaolin FP grade and residual soil from Bukit Timah Granite were used to evaluate the performance of the HCT. A kaolin paste was used as an interface between the residual soil specimen and HCT while no soil paste was used as an interface between the kaolin specimen and HCT. The kaolin paste allows more time to set up the specimen and improves the contact condition between the residual soil specimen and the HCT.
The equilibrium time which is needed by the HCT to read the initial matric suction of the soil specimen is about 10 min. Once the soil specimen is loaded, there is a difference between the durations to reach end of primary compression and for the pore-water pressure to reach equilibrium. Based on the test results, 20 min appear to be sufficient for each load increment for the soils tested when the soil is in the unsaturated state. However, once the pore-water pressure becomes positive (transition and saturated states), it is recommended to follow the test duration of a conventional saturated oedometer test (approximately 24 h) for each load increment.
MW carried out the tests, conducted the analyses and drafted the paper. ECL provided directions for the conduct of the tests, guided the analysis and the drafting of the paper. Both authors read and approved the final manuscript.
The first author acknowledges the research scholarship from the Nanyang Technological University, Singapore.
The authors declare that they have no competing interests.
- 1.Acikel AS, Mancuso C (2010) Use of CO2 gas to improve saturation of high capacity tensiometer (HCT). In: Buzzi O, Fityus D, Sheng D (eds) Unsaturated soils. Taylor & Francis group, LondonGoogle Scholar
- 3.ASTM-D2435/D2435 M (2011) Standard test methods for one-dimensional consolidation properties of soils using incremental loadingGoogle Scholar
- 4.Burland JB, Ridley A (1966) The importance of suction in soil mechanics. State of the art lecture. In: Twelfth Southeast Asian geotechnical conference, Kuala Lumpur, p 27–49Google Scholar
- 7.He L, Leong EC, Elgamal A (2006) A miniature tensiometer for measurement of high matric suction. American Society of Civil Engineers, Reston Va, p 1897–1907Google Scholar
- 10.Leong EC, Rahardjo H, Tang SK (2002) Characterisation and engineering properties of Singapore residual soils. In: Tan TS, Phoon KK, Hight DW, Leroueil S (eds) International workshop on characterisation and engineering properties of natural soils. Taylor & Francis, Singapore, p 1279–1304Google Scholar
- 12.Lourenço SDN, Gallipoli D, Toll DG, Evans FD (2009) On the measurement of water pressure in soils with high suction tensiometers. Geotech Test J 32(6). doi: 10.1520/GTJ102372
- 13.Marinho FAM, Take WA, Tarantino A (2005) State of the art presentation on tensiometer and axis-translation technique. In: International symposium on advanced experimental unsaturated soil mechanics. TrentoGoogle Scholar
- 18.Oliveira OM, Marinho FAM (2008) Suction equilibration time for a high capacity tensiometer. Geotech Test J 31(1):1–5Google Scholar
- 24.Tarantino A, Mongiovi L (2002) Design and construction of tensiometer for direct measurement of matric suction. In: Proc., proc. 3rd int. conf. on unsaturated soils. Recife, p 319–324Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.