Hokkaido University
      
School of Engineering 
       Course of Civil Engineering / Public Policy and Engineering
      Graduate School of Engineering
         Division of Field Engineering for the Environment

Japanese / English

  This is a portfolio of some of the recent/ongoing projects in our group.


  Home

 
RESEARCH

  Facilities

 
Members

 
Why not work with us?

  Publications

 
Events

 
Lecture courses

 
Inquiry / access

  RESEARCH


Our group runs research projects on a variety of geotechnical problems by making use of our innovative laboratory. Recent topics include, for example, embankment and river dyke building on soft problematic soils, reclamation with new geomaterials, natural slope stability and submarine resource (methane hydrate) exploration. As modern geotechnical engineering practice ventures into ever wide-ranging problems, research requires more specilisation and integration at the same time. Our philosophy is to approach modern geotechnical problems by rationally understanding and explaining soils' true properties and characteristics through high-quality experiments and analytical demonstrations; hence our group's Japanese name "Jiban Busseigaku Kenkyusitsu (Soils Property Laboatory)". Some of our research activities are introduced here. The articles here are aimed at aspiring geotechnical students around the world as well as our profesional cohorts in the geotechnical community.



Natural Soils' Properties
Towards understanding natural soils at any state
There are a wide variety of soils with different properties deriving from different geology in the world, even if we limit our scope only to fine-grained soils (such as silts and clays). Their each peculiarity, combined with spatial variabilities in the ground, poses a challenging geotechnical setting, only safely approached by careful site investigation. However, the apparent variabilities do not necessaily mean complete randomness. Every charcteristic of a particular soil and its distribution arise from its own geological background. By paying attention to soils' geological origins, their inherent properties (such as mineralogy) and states, we seek the reasons for soil behaviour we observe in laboratory and field. The recent investigtions involve the topics such as presented below.



From soils to soft rocks: Understanding the ground


Small-strain behaviour of soils
Soils exhibits far higher non-linearity in their stress-strain relationships when compared to other building materials such as steel and concrete. It is known that soils behave elastically only when deformation is very small, typically at less than 0.001% of strain. In 'mapping' soil's mechanical characteristics, understanding the small-strain stiffness is an 'alpha', or the beginning of all. Among others, the following characteristics of small-strain behaviour are fundamental and important yet not fully established.

1. Anisotropy
Anisotropy means direction-dependency; in sedimentary soils' properties, this arises mainly because their main structure is formed during gravitational, one-dimensional sedimentation and compaction. This process renders soils 'cross-anisotropic', characterised by axi-symmetry of properties with regard to the vertical axis. This type of anisotropy in soils' stiffness is widely recognised in natural soils, but in many cases disregarded at every turn of design processes in practice, due to lack of confidence in measurement, lack of understanding of possible impact on  analytical results, lack of means to reflect in design and lack of reliable data accumulation. Our research ( (1)・(2) ) demonstrates that many of the natural soils share very similar anisotropy patterns in their stiffness, making it possible to roughly estimate the whole picture based on simple investigation. It also explains how the assumption of isotropic stiffness, as is routinely made in practice, sometime leads to apparently consistent observations (such as how E=3G, a relationship deriving from isotropic theory, holds).

2. Stress-dependency
Well-informed graduate students of soil mechcanics would know that most soils' small-strain stiffness is roughly propertional to the effective stress raised by an exponent of 0.5 or so. But if we pay attention to more details, it is found that such stress-dependency is clearly affected by soils' geological ages. Old clays, even if they are not necessarily cemented, exhibit clearly less pronounced dependency on stress levels. Such characteristics are found to be lost upon complete remoulding in laboratory. It is a recent fad to quantify ageing and cementation by looking at how the yield stress is larger in natural soils than in reconstituted soils, but our study suggests that the small-strain stiffness's dependency on stress is equally or better indicator of ageing ( (2) ).

3. Estimating stiffness from field investigation
We sometimes undertake an expedition to field in order to collect in-situ sounding data, including ground's shear stiffness measured by SCPT (Seismic Cone Penetration Test, measuring shear wave velocity in a down-hole manner) sounding, as well as conventionl CPT resistance. The database is examined to explore correlations between measured data in a variety of soil types so that small-strain shear stiffness may be estimated based on routine sounding tests ( (3) ). Such an approach is particularly useful in soils such as peats, for which there is no established way of assessing the potantial disturbance upon sampling, and which is difficult to sample anyway. We demonstrated how this approach of stiffness determination can be used successfully in analysis of far-field deformation associated with embankment construction ( (4) ).



Stiffness anisotropy at small strains for a marine clay (Ch/Cv and Shh/Svh correspond to Eh'/Ev' and Ghh/Gvh at isotropic stresses, respectively)


Stress-dependency of stiffness and influence of geological age (m and n are the exponents applied to stress to compute Young's and shear moduli, respectively)


Correlation between CPT resistance and the small-strain stiffness (the hollow symbols represent peats)



Large deformation behaviour of soils
Under large deformation, what matters is strength; Textbooks of soil mechanics describe soils' strength by defining 'strength constants', such as effective stress parameters, c', φ', or undrained shear strength, su. Our reseach involves seeking these impotant engineering parameters for a variety of soils, including those which have been barely known to geotechnical community  (such as deep-water soils and a new family of artificially made geomaterials) at a variety of states. In doing so, it is impotant to recognise that these simple strength constants are not really 'constant' and assigning values to them requires profound understanding of soil mechanics and wide knowledge. So when we research soil strength, we also need to understand the engineering context of the projects to which application of the results are planned.

1. Strength of 'new' soils
By 'new', we mean 'new to us the geotechnical engineers' here. One of them is deep-water soils. Through work with JOGMEC and Shimiz Corporation, we investigated submarine sediments retrieved from 1000m below the sea level at Nankai Trough, off the pacific coast in Japan. Nankai trough is now a household name for its potential to cause a mega-earthquake and tsunamis that can wreck major cities in Japan. But it is also known to embrace a vast quantity of methane hydrate not far from the sea bottom. Physical and mechanical tests on the sediments between the sea bottom and the methane hydrate layer exhibited the soil's peculiar characteristics; despite a large share of very fine particles, it exhibits low plasticity, due to abandunt non-plastic remnants of organisms (coccolith) ( (5) ).

2. Strength of soils with a variety of structures, under a variety of conditions.
If you want to obtain strength parameter values for a specific soil, you can perform, say, triaxial compression tests, which are routine tests in geotechnical practice and for which very informing standard and manual books exist. However, manuals are useful only so far as 'standard' conditions for standard soils are concerned. For example, how much do we know about the soils' failure behaviour under very intense stress concentration, or conversely, at very low stress levels as may be observed in river dyke surfaces undergoing flood water infiltration, and, how can we accurately measure them? We also need to remember that the shear strength is also anisotropic normally. Although investigation of strength anisotropy by, for example, HCA (Hollow Cylinder Apparatus) has been undertaken for 50 years at least, high-quality data on natural soils are very limited. The concept of "soil structure", which is a sort of buzzword now in geotechnical research, can only be vague at best without addressing the above issues. We make persistent efforts to understand strength of soils with a variety of structures and geological origins (marine/lacustrine/dilluvial/articially-compacted) to contribute to more confident design in geotechnical practice ( (6) ).



SEM Photograph of sediments from Nankai Trough, retrieved from 1,000m below the sea level: Characterised by abundant coccolith


Shear test with rotated principal stress axes


Soil strength at various states: From at low confining stresses (as beneath a river dyke surface undergoing water infiltration during flooding) to fierce cyclic loading conditions under breakwaters


Soil behaviour at extremely large deformation:
Rheological properties of high-WC soils
Soils at very high water content are no longer solid in everyone's eyes. Solid mechanics is not the most appropriate famework to interpret and describe the behaviour of such a material. Rheology (a study of viscous flows, to put it simply) provides a better set of ideas and terms, and this is what we chose to adopt. Rheology of high-water-content soils is useful in addressing, for example, run-off behaviour of mudflows and pressurised transport of dredged soils (flow in pipelines and stability in stock yards need to be considered). Along with low-stress yield and viscosity, thixiotropy, or recovery of strength after a period of rest, is also a factor to be taken into account. To investigate these properties, we adopt a set of tools such as a low-capacity, high-resolution laboratory vane shear device, industrial viscometer, fall cones, bender elements, etc. We have published observations on how thixotropy affects the strength development in high-water-content clays, and how to quantify it in a non-destructive way ( (7)(8) ).


Mudflow in Jakarta, Indonesia. The inset shows application of laboratory vane shear testing to evaluate clays' rheological characteristics



Prediction of settlement in soft soils:
Effects of temperature and time
Strain rates are knwon to affect soils' strength even when soils are sufficiently solid. The strain rate effects, combined with ageing effect (similar to thixotropy described above; change in soils' stiffness and strength during a long resting period), are generlly called time effects. The time effects are often considered to be a secondary factor in soil deformation and consolidation analysis, but they in fact play an essential role when we think over infrastructure's life spans (50-100 years). Consider how Kansai International Airport is subsiding after 20 years of constuction, and how pipelines burried in peat ground in Sapporo city undergo continued straining. Without considering the real long-term behaviour of soils, these problem can only be half solved. Our group aims to provide rational interpretation of time effects through specialised tests such as ultra-slow CRS (Constant-Rate-of-Strain) oedometer tests and long-term trixial tests ( (9)(10) ). The time effects seem to be closely related and coupled with temperature effects; see how heating accelerates the ageing. So we also study how the temprature affects the long-term deformation behaviour of soils through temperature controlled mechanical tests ( (11) ). We also sometimes conduct model tests to see the reality of how problems of two-dimensional geometry develop.



Subsidence of structure in peat ground:
Image analysis of model test


Different appearance of strain rate effects
on compression behaviour in two clays




Properties of Artifical Geomaterials
Small- to medium deformation behaviour
Soils treated with binders, such as cement and lime, normally exhibit high stiffness and hence smaller deformation in ground when compared with surrounding untreated soils. It is therefore all the more important to understand their pre-yield behaviour correctly to enable more informed design. Our recent research involves a study on pre-yield compressibility of SGM (Super GeoMaterial, or light-weight cement-treated soil containing microbubbles), which was used in Haneda Airport's latest reclamation project. It has been known to be very difficult to reliably measure compressibility of stiff, low-permeability soils at small strains in conventional tests. Our test results from novel oedometer setting ( (12) : See also an later article) demonstrated that the actually measured small compression of the SGM layer can be fully explained by elastic compression of the material.

Cement-treated soils, stiff as they may be, are still soils, not concrete or stone. We recognise this fact when we need to deal with hard-to-stabilise soils such as peat and organics soils, in which humin prevents cement hydration and development of solid bonding. The strength of such soils, even treated with specially made binders, is measured in hundreds of kPa rather than in MPa. For these soils, an updated framework of mechanics is required in which cementation and inter-particle confinement have competing influence on stiffness and strength. We are trying to propose a new model that describes cemented soils' stiffness and its non-linearity in a rational and simple way ( (13)(14) ).




SGM (light-weight cement-treated soil): Even a small
differential settlement is not tolerated in airport runways.
Best skills and novel methods are necessary to measure
the stiffness of this new family of soils.


Model prediction of cement-treated soils' stiffness under various stress levels: Only unconfined compression strength is used as model parameter. The deviation of two points (■) suggests more work is necessary!



Large-deformation behaviour: Origin of strength
Japan boasts the highest standard of in-situ ground stabilisation (i.e. ground improvement by admixure) techniques. However, we have not been able to provide an explanation of why there is so large a discrepancy between the strength of labortory-made speciemens and core specimens taken from the treated ground ( (15) ). One of the assumed reasons for this is potentially variable nature of original soils and mechanical mixing processes. At least for clays, the in-situ stress is assumed not to affect the stabilised soils' strength, as the cement hydration proceeds faster than consolidation. In some situations, however, the in-situ stress and consolidation during  hydration cannot be ignored altogether; this is so, for example, in stabilisation of organic soils in which hydration does not lead to very strong structure. We explore the interations between strength development and stress development in hydrating cement-soil mixture by mechanical experiments ( (16) ). The study naturally leads to more profound question of what the cement-treated soils' strength originates from, and what mechanical framework can describe the process.



Triaxial compression tests on cement-treated soils with different stress history during curing: In other test seriess, the stress was applied at different timing during curing



Ground Improvement Techniques
Ground improvement by vertical drain installation:
Predicting the process based on soils' properties

Vertical drain installation combined with application of surcharges or vacuum is a preferred method of improving very soft soils such as normally consolidated clays and peats. The consolidation model with horizontal pore water migration requires the coefficient of permeability in horizontal direction. Back analysis of field executions suggests that the anisotropy in peameability does not follow a pattern common to all the soils. In particular, the permeability in peats and its anisotropy are little understood.  Our research involves further analysis of field trials and systematic laboratory tests with special arrangements ( (17) ).



Installation of plastic drains: It is difficult to predict
precisely the time taken in consolidation in peat ground




Developing New Laborratory Techniques
Measuring soil deformation: From small to large strains
As explained above, soils' stiffness depends on applied effective stress. So investigating the stiffness and its general rule of evolution requires appartus that can control the confining pressure, such as triaxial apparatus. So far many studies, including ours, have found that most of the sedimentary soils' small-strain stiffness can be modelled successfully by cross-anisotropic elasticity theory. A method for determining five independent parameters of the cross-anisotropic model by triaxial apparatus has been proposed by Kuwano & Jardine (1998) and Lings et al. (2000). However, performing a drained loading in clays is a difficult feat. The data of clays' full stiffness anisotropy and its evolution are severely limited, and many of the existing data are of questionable quality. So our group optimised the method and proposed alternative processes of parameter derivation ( (1) ).


Although the above method provides an accurate set of stiffness data, the elasticity theory is useful to soils only so far as very small strains are concerned. So we introduced a strain measurement system adopting digital photography and high-resolution image analysis. By appropriate data processing, we managed to resolve the strains to order of 0.001% in oedometer with a transparent ring, thereby observing compatibility between the measured elastic one-dimensional compressibility and that calculated from the cross-anisotropic matrix obtained by the special triaxial tests described earlier. The image analysis is versatile, allowing to follow deformation to very large strains. We also adopt the system to long-term monitoring of peats' creep in triaxial cells, in which very large deformation prevents use of local sensors.

  Bulletin of Faculty of Engineering, "Engineer-ring"
 
Innovation Forum poster, 2012



Characterising cross-anisotropic elasticity in triaxial apparatus with local sensors and shear wave velocity measurements


Compliance error-free CRS oedometer aided by image analysis



REFERENCES (Please contact us if you need any of these)

(1) Nishimura, S. (2014) "Assessment of anisotropic elastic parameters of saturated clay measured in triaxial apparatus: Appraisal of techniques and derivation procedures," Soils and Foundations, to appear in Vol.54, No.3.
(2) Nishimura, S. (2014) "Characterising anisotropic structure of natural clays based on their elastic stiffness and geological backgrounds," in preparation.
(3) Hieda, N., Tanaka, H. and Kaneko, H. (2013) "Shear modulus at a small strain for peaty ground,"
Technical Report of JGS Hokkaido Branch, Vol.53, pp.109-112. (in Japanese)
(4) Yamazoe, N., Tanaka, H., Nishimura, S. and Hayashi, H. "A constitutive model formulation for predicting far-field deformation in peaty soft ground and its application to field embankment test," submitted to Geotechnical Journal, JGS. (in Japanese)
(5) Hattori, T., Tanaka, H. and Kaneko, H. (2013) "Geotechnical characteristics of deep water sediments,"
Technical Report of JGS Hokkaido Branch, Vol.53, pp.67-70. (in Japanese)
(6) Fukutomi, Y., Nishimura, S. and Tanaka, H. (2014) "Measuring undrained strength anisotropy of natural normally consolidated clays by hollow cylinder apparatus,"
Proceedings of the 49th Annual Conference of JGS, Kitakyushu (Submitted; in Japanese).
(7) Seng, S. and Tanaka, H. (2011) "Properties of cement-treated soils during the initial curing stages," Soils and Foundations, Vol.51, No.5, pp.775-784.
(8) Seng, S. and Tanaka, H. (2012) "Properties of vey soft clays: a study of thixotropic hardening and behavior under low consolidation pressure," Soils and Foundations, Vol.52, No.2, pp.335-345.
(9) Tsutsumi, A. and Tanaka, H. (2011) "Compressive behavior during the transition of strain rate changing," Soils and Foundations, Vol.51, No.5, pp.813-822.
(10) Tsutsumi, A., Tanaka, H. and Kawaguchi, T. (2010) "Development of an extremely slow constant rate of strain oedometer and its application to several clays," Proceedings of JSCE, Series C, Vol. 66, No.3, pp.660-670. (in Japanese)
(11) Tsutsumi, A. and Tanaka, H. (2012) "Combined effects of strain rate and temperature on consolidation behavior of clayey soils," Soils and Foundations, Vol.52, No.2, pp.207-215.
(12) Takashino, S., Shinohe, A. and Nishimura, S. (2013) "Evaluation of SGM's and PFMM's compressibility under low confining pressure," Proceedings of the 48th Annual Conference of JGS, Toyama, pp.723-724. (in Japanese)
(13) Sugiyama, Y. and Nishimura, S. (2014) "Stress dependency and non-linearity of shear modulus in cement-treated high organic content soils,"
 Technical Report of JGS Hokkaido Branch, Vol.54, pp.19-26. (in Japanese)
(14) Nishimura, S. and Sugiyama, Y. (2014) "A stiffness model for cemented soils and simplified estimation of its parameters based on unconfined compression strength,"
Proceedings of the 49th Annual Conference of JGS, Kitakyushu (Submitted; in Japanese).
(15) Kitazume, M. and Nishimura, S. (2012) “An application of wet grab sampling to quality assurance of wet type cement stabilized soil,” Proceedings of 4th International Conference on Grouting and Deep Mixing, New Orleans.
(16) Abe, K. and Nishimura, S.
(2014) "> Influence of effective stress evolution during curing in cement-treated soils' strength," Proceedings of the 49th Annual Conference of JGS, Kitakyushu (Submitted; in Japanese).
(17) Tsutsui, K., Tanaka, H. and Yamazoe, N. (2014) "Anisotropy of permeability by using triaxial apparatus," Technical Report of JGS Hokkaido Branch, Vol.54, pp.41-44. (in Japanese)









Copyright © 2011 Laboratory of Soil Mechanics, Hokkaido Univerisity