Vertical compression of soft clay within PVD-improved zone under vacuum loading: Theoretical and practical study

https://doi.org/10.1016/j.geotexmem.2019.12.003Get rights and content

Abstract

Reliably estimating soil vertical compression under vacuum loading is still challenging for geotechnical engineers. In this paper, the soil vertical compression within the prefabricated vertical drain (PVD)-improved zone under vacuum pressure was investigated based on the theoretical analysis, reported laboratory tests and the case histories. A theoretical equation was developed to evaluate the effect of lateral displacement on the soil vertical compression. The observed small overall lateral-to-vertical strain ratio at the perimeter of the PVD-improved zone soil showed that the vertical compression of PVD-improved zone was close to one-dimensional (1-D) state, but 1-D compression method yielded smaller predictions than the measurements from the case histories. The back-analyzed soil modulus of the natural soft clay in Shanghai Disneyland Resort project showed considerable degradation as compared to the corresponding constrained modulus from the laboratory test. Such degradation could be due to the soil disturbance caused by PVD installation. The 1-D compression method was modified by introducing two factors to consider the effect of soil lateral displacement and to correct the constrained modulus due to the soil disturbance caused by PVD installation, respectively. The recommendations were provided after applying the proposed method to a case history.

Introduction

When constructing infrastructures, such as roadways and buildings, on soft clays, geotechnical engineers may face design challenges of low bearing capacity, excessive settlement and slope instability. Various techniques (e.g., lightweight materials, stone columns, deep mixed (DM) columns and surcharge/vacuum preloading) have been used to improve such unfavourable geotechnical condition (McCabe et al., 2009; Shen et al., 2012; Neramitkornburi et al., 2015; Wu et al., 2015; Chai et al., 2018; Cai et al., 2018; Wang et al., 2019). However, for the projects with large treated area, such as airports, highways and storage tanks, vacuum preloading combined with prefabricated vertical drains (PVDs) has been the most commonly-used method around the world (Chu et al., 2000; Indraratna et al., 2004; Rujikiatkamjorn et al., 2007; Chai et al., 2010; Griffin and Kelly, 2014; Lei et al., 2017; Deng et al., 2018; Ni et al., 2019). In last decade, growing innovations in vacuum preloading enhanced its improvement effect and extended its applications, including membraneless technique, thermo-PVDs, vacuum preloading combined with electro-osmosis, and vacuum preloading with air injection and/or lime treatment (Artidteang et al., 2011; Long et al., 2015; Sun et al., 2017, 2018; Wang et al., 2017; Wu et al., 2017; Cai et al., 2018; Zhang and Hu, 2019).

In spite of such developments, reliably estimating soil vertical compression under vacuum loading is always challenging for geotechnical engineers. The classical one-dimensional (1-D) compression method is widely used in practice to estimate ground settlement as its simplicity and ease to determine its parameters (Terzaghi et al., 1996; Chai et al., 2005; Tang and Shang, 2000; Indraratna et al., 2011). According to this method, the total vertical compression is the sum of the vertical compression of each sublayer:SP1D=i=1nΔσziEciliin which, SP1D is the total vertical compression of ground, l is the thickness of a soil sublayer, Ec is the constrained modulus of soil obtained from laboratory oedometer test, and σz is the average effective vertical increment in soil sublayer, and i represents the soil sublayer number. However, as the ground might have inward lateral deformation under vacuum loading, the laterally-constrained condition cannot be maintained like 1-D compression, so that Eq. (1) cannot be directly used in design. Chai et al. (2005) indicated that the soil near the ground surface was isotropically compressed while at deep position it was close to 1-D consolidation. Based on laboratory tests, some researchers indicated that soil vertical deformation induced by vacuum pressure was less than or equal to that induced by surcharge pressure with equal amount (Mohamedelhassan and Shang, 2002; Chai et al., 2005; Robinson et al., 2012). It is noted that as compared to the soil sample in laboratory test, treated ground in practice has big area (usually over 2500 m2) with a large number of PVDs and different boundary condition. Whether the laboratory test can represent the real condition in field is still an arguable issue, but it is crucial to consider the real condition in calculating the soil vertical compression.

A large number of PVDs are installed into the natural soft clays in the use of vacuum preloading method. PVD installation process by pushing a mandrel with a PVD into the soils has been proved to have appreciable disturbance to the natural soft clays and change the soil compressibility significantly (Terzaghi et al., 1996; Ghandeharioon et al., 2011; Perera et al., 2014). Therefore, such disturbance effect as a practical issue needs to be considered in the calculation method. The soil sample prepared for laboratory test is commonly sampled before PVD installation, as a result the influence of PVD installation on soil properties is hardly reflected in the laboratory test. Soil disturbance caused by PVD installation is complex and relative to many factors, including soil properties, mandrel shape, PVD types and installation methods. These factors are difficult to be completely reproduced in laboratory. However, back analysis based on field monitoring data is regarded as an effective mean as the field conditions can be included (Voottipruex et al., 2014), through which geotechnical engineers can enhance the understanding of soil deformation under vacuum preloading.

This paper aims at investigating the characteristics of soil vertical compression of PVD-improved zone under vacuum preloading based on theoretical analysis, reported laboratory tests, and case histories. Two important issues are explored, one is the different boundary condition between the real condition in field and the 1-D compression condition, and the other is the soil disturbance by PVD-installation. Accordingly, a modified 1-D compression method is proposed for calculating the soil vertical compression of PVD-treated soil under vacuum loading, and its feasibility and applicability is verified through comparing with field data.

Section snippets

Theoretical analysis on effect of lateral movement on soil vertical compression

Elastic analysis was conducted to investigate the effect of lateral movement on soil vertical compression. Though it has some limitations, elastic analysis is still widely used in practice as its simplicity and ease to determine its parameters (Terzaghi et al., 1996). The vertical compression of soil element under laterally-constrained condition (i.e., zero lateral deformation, and also called 1-D compression condition) can be expressed byΔεz1D=ΔσzEcEc=(1ν)E(1+ν)(12ν)in which, Δεz1D is the

Obtained ζ1 based on laboratory test and case histories

Based on Eq. (8a), (8b), ζ1 can be determined by two methods: (a) directly determine ζ1 using Eq. (8a) based on the vertical compressions of soil sample under vacuum pressure and 1-D compression from test; or (b) calculate ζ1 by substituting the obtained Δεh/Δεz from the test into Eq. (8b).

Observed and calculated soil vertical compression

The above analyses revealed that the soil displacement within the PVD-improved zone in the field under vacuum pressure was close to 1-D compression state. In this section a comparison is made between the measured soil compressions within the PVD-improved zone (SP) and the calculated results by the 1-D compression method (SP1D). SP was obtained by subtracting the measured ground surface settlement from the measured sublayer settlement at the PVD tips after consolidation was complete. SP1D was

Soil disturbance effect analysis in a real project

In this section, the compression curves of soil layer in the PVD-improved zone are established based on six treated blocks in the ground improvement project of Shanghai Disneyland Resort. The back-calculated modulus of soil based on the compression curve is obtained to evaluate the soil disturbance caused by PVD installation by comparing with the corresponding constrained modulus from the laboratory test.

Modification of the 1-D compression method

According to the above analyses, the effects of soil lateral strain and the soil disturbance caused by PVD installation should be properly considered in estimating the soil compression within the PVD-improved zone under vacuum preloading. The 1-D compression method (Eq. (1)) is modified as following to consider of the two influence factorsSPV=ζ1(i=1nΔσziζ2iEcili)=ζSP1Din which, SPV is the soil compression of the PVD-improved zone under vacuum preloading, ζ1 is the factor to consider the

Conclusions

In this study, soil vertical compression of PVD-improved zone under vacuum preloading were investigated with two important issues, the different boundary condition between the real condition in field and the 1-D compression condition, and the soil disturbance by PVD-installation. Based on the analysis results, the following conclusions can be drawn:

  • 1.

    A theoretical equation was deduced to convert soil compression under 1-D compression condition to that under vacuum pressure with a correction

Acknowledgements

The authors appreciate the financial supports provided by the Natural Science Foundation of China (NSFC) (Grant No. 40702051, No. 41972272 & No. 41972273) and by the Fundamental Research Funds for the Central Universities (Grant No. 22120180106) for this research.

References (40)

  • Z.F. Wang et al.

    Enhancing discharge of spoil to mitigate disturbance induced by horizontal jet grouting in clayey soil: theoretical model and application

    Comput. Geotech.

    (2019)
  • L. Zhang et al.

    Laboratory tests of electro-osmotic consolidation combined with vacuum preloading on kaolinite using electrokinetic geosynthetics

    Geotext. Geomembranes

    (2019)
  • Z. Zhang et al.

    Comparison analysis on performance of vertical drain-reinforced soft soil subjected to vacuum and surcharge preloading

    Geotext. Geomembranes

    (2018)
  • S. Artidteang et al.

    Enhancement of efficiency of prefabricated vertical drains using surcharge, vacuum and heat preloading

    Geosynth. Int.

    (2011)
  • ASTMD1687. “Standard Practice for Thin-Walled Tube Sampling of Fine-Grained Soils for Geotechnical Purposes”. American...
  • Y. Cai et al.

    New approach of vacuum preloading with booster prefabricated vertical drains (PVDs) to improve deep marine clay strata

    Can. Geotech. J.

    (2018)
  • J.C. Chai et al.

    Ground deformation induced by vacuum consolidation

    J. Geotech. Geoenviron. Eng.

    (2005)
  • L. Chen et al.

    Soil consolidation and vacuum pressure distribution under prefabricated vertical drains

    Geotech. Geol. Eng.

    (2019)
  • J. Chu et al.

    Soil improvement by the vacuum preloading method for an oil storage station

    Geotechnique

    (2000)
  • Y. Deng et al.

    Colloid effect on clogging mechanism of hydraulic reclamation mud improved by vacuum preloading

    Can. Geotech. J.

    (2018)
  • Cited by (4)

    • Effectiveness of a Geocomposite-PVD system in preventing subgrade instability and fluidisation under cyclic loading

      2022, Geotextiles and Geomembranes
      Citation Excerpt :

      A system of vertical drains combined with vacuum pressure and surcharge loading can increase the rate of consolidation by accelerating both vertical and horizontal drainage (Chai et al., 2020; Chu et al., 2004; Rujikiatkamjorn and Indraratna 2007). PVDs installed at shallow subgrade depths within 6–8 m can help to stabilise heavy-haul rail embankments to curtail the development of EPWP, especially in low-lying areas with soft estuarine deposits (Indraratna et al. 2009, 2010), while also reducing the fluidisation potential under various forms of dynamic loading (Brennan and Madabhushi 2002; Gao and Zhang 2020). This study examines the combination of a geocomposite (a filter membrane sandwiched between nonwoven geotextiles) with PVDs to alleviate the occurrence of soil fluidisation.

    View full text