A Novel Field Device for the Measurement of Soil Collapsibility

  • Mehdi Mokhberi
  • Seyed Ayuob Rafieean
Conference paper
Part of the Sustainable Civil Infrastructures book series (SUCI)


Soil collapsibility or soil swelling is typically measured in the laboratory by using a single or double consolidation test. Such a test requires time and is costly: its stages include preparing the samples, transferring them to the laboratory, and conducting related tests. In most cases, geotechnical engineers have to carry out experiments during the stage of site investigation to evaluate the soil in terms of collapsibility. However, soil collapsibility might be easily estimated by a simple desert test. The device designed, developed, and used by the authors of the present study has the ability to determine the soil’s collapsibility and offers its primary percentage. The experiments that were conducted on the downstream alluvium of Seevand Dam, with high collapsibility, to assess the performance of the device indicated that the device is able to appropriately evaluate the collapsibility of soil. Furthermore, comparing the results of these experiments with those of experiments using the consolidation method confirm the accuracy of the findings present study, with a high percentage. Given the probability of disturbing soils in laboratories, it might be contended that this device obtains results that are more valid than laboratory results.


Collapsibility Field test Unsaturated soil Double consolidation Seevand Dam 

1 Introduction

Collapsible soils are widespread in Iran and the world. These types of soils are usually known for their low natural humidity percentage and also low unit weight. Structures placed on collapsible soils, in the case of soil saturation, might settle unexpectedly. In the planning of most engineering structures, it is essential to pay special attention to soil collapsibility as it might destroy the foundation of a structure, destroy a dam, or produce road subsidence.

Furthermore, the existence of other technical buildings in these areas, and also given the development of cities and the necessities of the development of big cities, building residential areas, water and waste water pipes, etc. on these types of soil is of great importance (Derbyshire 2001). In the process of the development of soil mechanics science, which formally started in 1936, it was in the late 1960’s and during the seventh and eighth international soil mechanics conferences that the investigation of the scientific principles of unsaturated soil was taken into consideration. In 1959, one of the first criteria of collapsibility based on dry density was offered by Clevenger (1958). After that, in 1962, Gibbs and Bara (1962) suggested using the unit weight of dry density and soil smoothness limit as criteria for separating collapsible and non-collapsible oils. The trend of offering criteria for distinguishing collapsible and non-collapsible soils continued during the 1970’s, 1980’s, and 1990’s, and different benchmarks were introduced by Denisov (1963), Feda (1966) and Fookes and Best (1969). In a Russian conference on soil mechanics and foundation engineering, the instability of a collapsible soil’s skeleton was investigated and discussed. Jennings and Knight (1957) suggested the double consolidation test to assess the behavior of soil under conditions of saturation and loading with different levels of stress. Since the late 1980’s, considerable attempts have been made to offer mathematical and computer models. In these models, the attention was on modeling the behavior of collapsible soils and the distribution process of these soils’ porosity (Handy 1973; Houston et al. 1988). Since then, many studies have been carried out to estimate collapsibility, numeric modeling, failure model offering, etc. Similarly, different studies have also been conducted in Iran about the behavior of soils. For example, Habibagahi and Mokhberi (1998) investigated the effect of humidity on collapsibility estimation by modifying the relation offered by Duncan and Chang (1970) and also involving important parameters. They could offer a new model for soil failure. Tarantino et al. (2005) attempted to model the behavior of collapsible soils through laboratory and computational methods, especially the finite element method and using remolded specimens of loose soil and computer programs continue to be used. They have offered related research studies related to conducted experiments in their book.

The common methods of determining soil collapsibility require taking disturbed and non-disturbed soil samples. Such methods are not only time consuming and costly, but also change the behavior of the soil. Additionally, recreation of collapsible soils in the laboratory needs experience and special accuracy. The present research study is, in fact, an attempt to provide a small portable device to determine the collapsibility of soil. The soil collapsibility field identification device offered in the present study is used to evaluate the soil’s collapsibility in non-laboratory and field contexts. The device is able to evaluate soils in terms of their potentials.

2 Materials and Method

2.1 Development of Device

The offered device consists of two sections, one for collapsibility and the other for loading. Figure 1 presents the whole of the device. In brief, the main parts of the device are as follows:
Fig. 1.

Schematic design and image of the offered device to assess the soil collapsibility

  1. 1.

    Stabilizer panel: This is a square rigid metal panel with sides of 30 cm and thickness of 1 cm, made from iron. There is a hole whose diameter is equal to the outer diameter of the cylinder in the limiting center of the cylinder within which the sampling cylinder is inserted.

  2. 2.

    Sampling cylinder: This is a hollow cylinder made from steel and is completely resistant to the materials of the soil. Its length is 13 cm.

  3. 3.

    Porous stone: This is placed into the sampling cylinder and above the soil sample.

  4. 4.

    Loading panel: This is made from steel and its diameter is equal to that of the porous stone. It is completely resistant to the soil materials and is an important point in the design of the device.

  5. 5.

    Moving loading bar: This is a solid metal bar of steel and is completely polished so that it can easily move as the soil settles.

  6. 6.

    Device Weights: The weight is imposed through the metal weights placed on the moving loading bar.

  7. 7.

    Sample Thickness Change Measuring Gauge: The gauge used in this device has 0.01 mm accuracy. There is a magnet in the foot of this gauge that might be connected to the cylinder limiting panel.


The information obtained from this device is used to assess the soil’s collapsibility and also to calculate the indices that are about the relation between effective the stress and the strain or vacuum symptom. Additionally, the information resulting from this device might be used to increase or estimate speed, determine unsteady settlement and total settlement of a structure or collapsible soil piles.

2.2 Method

First, the device was placed and leveled in the intended area. Then, the user gradually puts pressure steadily on the two handles inserted on the cylinder stabilizer panel so that the cylinder penetrates into the soil. After that, filter paper and the porous stone were placed on the cylinder. Next, the primary load is done for 5 kPa. Then, within a 5-minute time period after the primary loading, further loading steps are gradually done at intervals of one hour in natural humidity conditions. The loading process continues until the settlement stops. The loading steps should be 10, 25, 50, 100, 200 kPa. Before each step, any change of form is observed. The amount of stress on the sample before its saturation is determined based on conditions. One hour after reaching the intended stress, any changes of the soil form in natural humidity is measured. After that, the sample is drowned by pouring water into the cylinder. Water steadily enters into the porous stone through the three holes on the panel. In this way, the whole of the sample is steadily saturated downwards in line with natural patterns. After drowning the sample, the sample form changes are read and observed at time intervals of approximately 0.1, 0.25, 0.5, 1, 2, 4, 8, 15, 30, and 60 min. It should, however, be mentioned that in soils with high penetration, collapsibility might happen quickly and makes reading the time difficult. Adding water to the sample should be in a way that it is saturated downwards and air is not imprisoned in the soil. After adding water into the cylinder and drowning the sample, the necessary time for putting on a load would continue for one day or until the primary consolidation (based on D2435 standard) is finished. The degree of sample form change is read in the above-mentioned time intervals. Finally, the information obtained from the device is used to determine the collapsibility percentage of non-saturated soils after being drowned.

2.3 Evaluation of Test Results

2.3.1 Results of Test Device by Using the Device

In this section, the results obtained from the tests using the developed device and those of an experimental consolidation device are presented. These tests were done on three points of soil in the downstream area of Seevand Dam located 100 km to the north of Shiraz, which is known to be collapsible soil. The area soil is of CL type. Table 1 presents the soil consolidation properties in this area.
Table 1.

The consolidation properties of soil of the studied area


\( e_{0} \)

\( e_{f} \)

\( \omega_{0} \left( \% \right) \)

\( C_{s} \)

\( C_{c} \)

\( G_{s} \)






















To find the soil collapsibility using the device, the soil changes were recorded and measured after placing the device and putting the load on the soil. Figures 2a to c and Table 2 show the soil changes before saturation, during saturation, and after saturation.
Fig. 2.

Comparison of soil collapsibility from offered device (left) and single consolidation tests (right)

Table 2.

Comparison of soil collapsibility from offered device and single consolidation test


Collapse potential from oedometer test (%)

Collapse potential from proposed device (%)










2.3.2 Laboratory Consolidation Device Results

The single consolidation tests were carried out based on standard (ASTM D 5333-03) to determine the soil’s collapsibility. In this experiment, a sample of soil is placed in the oeodometer device and vertical pressure is increased until reaching the probable pressures of real earth. At this loading level, water is injected into the device, the sample is saturated, and resulting changes are recorded.

The consolidation graphs of three different natural soils have been presented below. It should, however, be pointed out that in the consolidation tests, stresses of 25, 50, and 100 kPa were loaded in natural humidity status and then the samples were consolidated with stresses of 100, 200, 400, and 800 kPa. After that, the loading process was done to the weight of one kilogram on one cubic centimeter. Figures 2d to f and Table 2 show the related results.

2.3.3 Comparison of Soil Collapsibility Determined by the Two Methods

The soil collapsibility was determined by using (Lutenegger and Saber 1988) relation (1).
$$ {\text{I = }}\frac{{\Delta {\text{e}}}}{{1 + {\text{e}}}} \times 100 $$

As Fig. 2 and Table 2 show, the collapsibility of the reconstructed sample is approximately 6.3 in the laboratory and 7.4 in the field, which implies the correct performance of the device.

3 Conclusions

The comparison of the results obtained from the developed device and those from an experimental consolidation device shows that the developed device has appropriately assessed the collapsibility of the area soil. Additionally, due to its description of the soil collapsibility, it has a better and more complete capability than other comparable devices. Therefore, it could be used in similar projects of the district as a suitable criterion to document the soil collapsibility. In fact, the soil collapsibility field device is a device that has removed the technical problems of existing related devices and could simulate the real and natural conditions of soil and as a result, the obtained results are completely coincident with the natural patterns of the soil. This device, in essence, increases the accuracy and output of the work considerably.


  1. Clevenger, W.A.T.: Experiences with loess as foundation material. Trans. Am. Soc. Civ. Eng. 123(1), 151–169 (1958)Google Scholar
  2. Denisov, N.Y.: About the nature of high sensitivity of Quick clays. Osnov. Fudam. Mekh. Grunt. 5, 5–8 (1963)Google Scholar
  3. Derbyshire, E.: Geological hazards in loess terrain, with particular reference to the loess regions of China. Earth Sci. Rev. 54(1), 231–260 (2001)CrossRefGoogle Scholar
  4. Duncan, J.M., Chang, C.-Y.: Nonlinear analysis of stress and strain in soils. J. Soil Mech. Found. Div. (1970) Google Scholar
  5. Feda, J.: Structural stability of subsident loess soil from Praha-Device. Eng. Geol. 1(3), 201–219 (1966)CrossRefGoogle Scholar
  6. Fookes, P.G., Best, R.: Consolidation characteristics of some kate Pleistocene periglacial metastable soils of East Kent. Q. J. Eng. Geol. Hydrogeol. 2(2), 103–128 (1969)CrossRefGoogle Scholar
  7. Gibbs, H., Bara, J.: Predicting surface subsidence from basic soil tests. In: Field Testing of Soils. ASTM International (1962)Google Scholar
  8. Habibagahi, G., Mokhberi, M.: A hyperbolic model for volume change behavior of collapsible soils. Canad. Geotech. J. 35(2), 264–272 (1998)CrossRefGoogle Scholar
  9. Handy, R.L.: Collapsible loess in Iowa. Soil Sci. Soc. Am. J. 37(2), 281–284 (1973)CrossRefGoogle Scholar
  10. Houston, S.L., Houston, W.N., Spadola, D.J.: Prediction of field collapse of soils due to wetting. J. Geotech. Eng. 114(1), 40–58 (1988)CrossRefGoogle Scholar
  11. Jennings, J., Knight, K.: The additional settlement of foundations due to a collapse of structure of sandy subsoils on wetting. Proceedings (1957) Google Scholar
  12. Lutenegger, A.J., Saber, R.T.: Determination of collapse potential of soils (1988)Google Scholar
  13. Tarantino, A., Romero, E., Cui, Y.J.: Advanced experimental unsaturated soil mechanics. In: Proceedings of the International Symposium on Advanced Experimental Unsaturated Soil Mechanics, Trento, Italy, 27–29 June 2005. CRC Press (2005)Google Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  • Mehdi Mokhberi
    • 1
  • Seyed Ayuob Rafieean
    • 1
  1. 1.Islamic Azad UniversityTehranIran

Personalised recommendations