A CRITICAL REVIEW ON THE DYNAMIC BEHAVIOUR OF OCR OF …

Sustainable Transportation Infrastructures Smart Driving Research Center (SDRC)

ISBN 978-967-2110-70-5

20

2017

CHAPTER 3

A CRITICAL REVIEW ON THE DYNAMIC BEHAVIOUR OF OCR OF

SOFT CLAY UNDER CYCLIC LOADING

3.1 INTRODUCTION

The soils are often vulnerable to undrained cyclic loading produced either by human activity

or natural geologic activity. These cyclic loadings have a wide range of amplitude,

frequency, and duration which its behaviour varies depends on specific factors. Therefore,

cyclic behaviour of soft clay is a critical interest in geotechnical engineering practice and has

been studied progressively over the years, [1]. At present, development of city has become

a major critical impact in Batu Pahat because of the growing economy, industry and

community. The problem that has been found after construction and during the working

period of a building is the very high settlement rate, which shortens the design life of

structure. In addition, Soft Clay are well known for their low strength and high

compressibility. Usually, due to sedimentary process on different environment, both physical

and engineering properties of the clays (namely void ratio, water content, grain size

distribution, compressibility, permeability and strength) show a significant variation.

Furthermore, they exhibit high compressibility (including an important secondary

consolidation), reduced strength, low permeability and compactness, and consequently low

quality for construction [2]. This paper reviews the research done on the effects of damping

ratio, D and shear modulus, G under dynamic cyclic loading with different frequencies, the

effects of effective vertical consolidation stress and the effect of over – consolidation ratio,

OCR.

3.2 FACTORS AFFECTING CLAY BEHAVIOUR UNDER CYCLIC LOADING

The soil deposits in many geotechnical engineering projects undergo dynamic cyclic

loadings during their design lifetime. These loadings may be due to environmental factors,

such as seismic activity and ocean storms, or human activities, such as passing traffic and

vibrating machinery installed on a structure or site. Importantly, the soil response generated

by these dynamic cyclic loadings is typically more complex than that considered when

conducting static analyses, requiring engineers to investigate the dynamic behaviour of soils


ISBN 978-967-2110-70-5

21

2017

in the laboratory, as well as in the field. A typical soft clay soil properties can be found in the

grounds of Universiti Tun Hussein Onn, UTHM which have low shear strength and bearing

capacity, and suffer large settlements when subjected to loading. The typical physical

properties of Batu Pahat Soft Clay at Research Centre of Soft Soil, RECESS UTHM have

been experimentally investigated by many researchers as shown in Table 3.1 are compared

with soft clay from other places. A study carried by Chan et. al. [2], found that clay soil at

RECESS, UTHM contained 10.8 % clay, 79.5 % silt and 10.7 % sand.

The dynamic behaviour of cyclic strength of the soft clay during cyclic loading depend on

various factors. It depends on parameter such as listed below:

i. Cyclic stress/strain amplitude

ii. Number of loading cycles

iii. Frequency/loading rate

iv. Effective vertical stress

v. Consolidation path

vi. Over consolidation ratio (OCR)

vii. Sensitivity

viii. Initial static/average shear etc.

Table 3.1: Some of soft soil geotechnical properties

Clay Batu Pahat Soft Clay, (BPSC)

Soft Bangkok Clay

Itsukaichi Japan Marine Clay

Cloverdale Clay

Researchers Chan and Ibrahim [2]

Thammatiwat.A, et.al, [3]

- -

Bulk Density (Mg/m3)

1.36 - - -

Specific Gravity 2.66 2.57 – 2.73 2.532 2.79

Plastic Limit (%) 31 30 – 42 51.4 24

Liquid Limit (%) 77 75 – 99 124.26 51

Plasticity Index (%) 46 45 - 57 72.86 27

Moisture Content (%) - 78 - 99 - -

Table 3.2 shows some type of clay tested by researchers on past studies in terms of

geotechnical properties, sample condition, mode of loading, loading period and number of

cycles.


ISBN 978-967-2110-70-5

22

2017

Table 3.2: Clay tested in undrained cyclic triaxial by previous researchers

3.3 SOURCE OF DYNAMIC CYCLIC LOADING

The dynamic cyclic loadings is the occurrence of repeated loading under specific

circumstances which built up of cycle, frequency, amplitude and time which consists of

stress – strain relationship under specific cyclic loading factors. They exist may be due to

the environmental factors, such as earthquake, wave, wind activity and etc. or human

activities, such as passing traffic, vibrating machinery installed on a structure or site,

building foundation and etc. Importantly, the soil response generated by these dynamic

cyclic loadings is typically more complex than that considered when conducting static

analyses, requiring engineers to investigate the dynamic behaviour of soils in the laboratory,

as well as in the field. The dynamic loading from wind, waves, equipment vibrations, or

earthquakes will lead to an accumulation of shear strains, and the shear modulus will

degrade once the threshold shear strain has been exceeded [4]. Large cyclic strains and

accompanied strength loss is a matter of concern for engineers especially in earthquake

response analysis. As a result, cyclic strength is usually defined in literature as the cyclic

shear required inducing a significant shear strain under triaxial or simple shear conditions.

3.4 OVER-CONSOLIDATION RATIO (OCR) OF SOIL

The over consolidation ratio (OCR) is a geotechnical parameter, related to historical

changes in the state of stress in the subsoil. The concept of over consolidation ratio was


ISBN 978-967-2110-70-5

23

2017

proposed by Cassagrande. The idea of recording the over consolidation ratio resulted from

the observation of changes in strength parameters of non-lithified deposits, depending on

the state of stress found in the subsoil. An example illustrating this dependence is an

increase in shear strength of a deposit homogenous in terms of its physic-chemical

properties with an increase in depth. Under such conditions determined strength parameters

of a given deposit, higher than expected, were thus considered to be connected with a

different state of stress in the subsoil than that generated as a result of normal consolidation

of the deposit. It needs to be emphasized that consolidation is considered to be normal if it

results from a successively increasing load, affecting the deposited sediment at a specific

depth. The over - consolidation ratio OCR was defined by Casagrande as a ratio of

maximum effective value of the vertical component of geostatic stress, found at any time in

a given subsoil point, to the present effective value of the vertical component of geostatic

stress.

3.4.1 EFFECTS OF OVER CONSOLIDATION RATIO, OCR ON SETTLEMENT

According to related studies, the over - consolidation state is an important effect for soil

liquefaction potential. If a soil mass has experienced stresses higher than its current state, it

is an over - consolidated soil (OCR > 1). The over - consolidated soils have fewer

settlements due to external loadings as compared with normally consolidated soils. Seed

and Idriss, showed that the liquefaction resistance increases as the over consolidation ratio,

OCR increases [5]. Sarsby [5] stated that the relationship between shear strength, pore

pressure and soil modulus depend on the stress history which is reflected by the over -

consolidation ratio (OCR). The over - consolidated soils, OC usually have lower porosities

than their normally consolidated, NC counterpart. This leads to their having a stiffer

behavior in respect of deformation under applied load, more dilatant behavior under shear

because of denser packing of the particles. The over - consolidation ratio of soil can affect

the settlement analysis for structures constructed on top of saturated cohesive soil.

Considerable settlement due to continued consolidation by the soil’s own weight and

applied structural load are expected if the cohesive soil is under consolidated. On the other

hand, if the cohesive soil is over - consolidated, there would be no significant settlement

when a load is applied to the soil [6].


ISBN 978-967-2110-70-5

24

2017

3.4.2 EFFECTS OF OCR ON STRESS – STRAIN RELATIONSHIPS

The over - consolidation ratio (OCR) seems to have similar effects on cyclic stress strain

behaviour of clays as it does in monotonic behaviour. For example, it is reported that NC

clays produced positive pore water pressure and shifted the stress path gradually towards

origin during cyclic shearing. On the other hand, OC clays developed negative pore water

pressure at the initial stages. Subsequently they generated positive pore pressure with

increasing loading cycles until failure. Also, magnitude of initial negative pore pressure

increases with OCR as in monotonic loading. Different observations have been reported in

literature regarding the effect of OCR on cyclic strength. The OC clays showed equal or

stronger behaviour than that of NC clays under equal cyclic stress amplitudes. However, it

is also found that number of cycles to failure is little bit lower in OC clays than that of NC

clays [1].

3.4.3 EFFECTS OF OCR ON STRENGTH

The influence of OCR and fines content on the variation of the initial target modulus is

shown in Figure 3.1. As OCR increases, the initial tangent modulus of clean sand

decreases due to the lower current confining stress and consequently, the soil specimens

become weaker. However, the OCR seems to have negligible effect on the initial tangent

modulus of soil specimens with clay fines content [6].

Figure 3.2 shows that the reduction in strength due to the increase of clay fines with

followed by specimens with 10%, 20%, 30% and 40% clay fines. According to Simons and

Menzies (2000), the strength depends on the tendency of soil in dilation, and it decreases

as the normal effective stress decreases [6].


ISBN 978-967-2110-70-5

25

2017

Figure 3.1: Initial tangent modulus vs OCR [6]

Figure 3.2: Strength vs OCR [6]

3.5 EFFECTIVE VERTICLE CONSOLIDATION OF DYNAMIC TRIAXIAL TEST

Table 3.3 shows the variance of effective vertical consolidation stress applied of the soft

clay under dynamic cyclic triaxial testing from past research within range of 40 to 800 kpa

depends on soil characteristic and geotechnical properties. Effective vertical stress (σ’vc) is

another factor that influences the cyclic behaviour of soils. For example cyclic stress ratio or

liquefaction resistance of sand has a correction factor for effective overburden stress (e.g.

Seed, 1983). However, reports show that the effect of (σ’vc) is not as important in the cyclic

behaviour of clays, as they have already been considered in their monotonic undrained

strength (su), as cyclic strength of clays is usually expressed in terms of su. It is observed


ISBN 978-967-2110-70-5

26

2017

that pore pressure generation during cyclic loading increased with effective confining stress,

and number of load cycles necessary for failure reached a minimum between effective

confining stresses of 50 and 100 kPa for Champlain Sea clay in Eastern Canada [1]


Researchers Type of Soil Effective vertical consolidation stress, σv’

(kpa)

J.Ni et.al, [7] Soft Clay 40

- Bangkok Soft Clay

50, 100

Theenathayarl, [1] Leda Clay 50, 100, 200, 400, 800

H. Soltani – Jigheh, [8] Mix Clay 100, 200, 350

- Kolkata Clay 50, 100, 150, 200

Table 3.3 shows the typical of test frequency that can be applied under dynamic

cyclic triaxial testing. Similar to the strain rate effects on the undrained monotonic strength

of clays, loading frequency or rate also influences cyclic strength of, even though this effect

is not as significant in granular soils. A number of investigations on the effect of loading

frequency on cyclic strength of clays show cyclic strength of clays increases significantly

with loading frequency. It has been reported that 30% cyclic strength increase when loading

frequencies have been raised through two log cycles (i.e. increased by about 100 times).

The cyclic strength increases with increasing frequency, the effect of frequency on strength

diminishes when number of load cycles are higher. It noted that shear stresses in excess of

about 40% of their undrained strength have been resisted for more than a cycle during

undrained cyclic simple shear tests due to effects of very high loading frequency. They also

found that very high loading rates in cyclic loading partially compensate the strength

degradation with number of loading cycles.

The loading frequency alters the strength envelope in structured or OC clays, but

not influences the pore water pressure generation much. As a result failure occurs below

the peak strength envelope in OC clays under slow loading rate. On the other hand in

structured or NC clays, changes in loading frequency affects pore pressure generation. As a

result, stress path is altered, but stress path meets a unique strength envelope at failure [1].

It is recognized that actual earthquake loading contains a range of frequency content (0.5 –

5 Hz range is probably the strongest modes in this region), but the effect of cyclic loading


ISBN 978-967-2110-70-5

27

2017

frequency on liquefaction characteristics of clays is rather. Further, the lower frequency

used would result in a conservative estimate of the cyclic resistance of the material.

Generally, even the most energetic earthquakes do not result in more than about 30

equivalent load cycles of shaking.

The shear modulus and damping of soil are expressed as a function of cyclic strain

amplitude and depend on various parameters such as effective confining stress, Plasticity

index (PI), OCR/Stress history, number of load cycles, loading rate etc. Various researchers

investigated modulus reduction and damping in various soils and the influence of the above

mentioned factors. It was originally provided modulus and damping curves as a function of

cyclic strain amplitude for coarse and fine grained soils such as sands and gravels. Some

analyses the available modulus data for sand and concluded that these curves mainly

depend on effective confining pressure and relative density or void ratio.

Furthermore they found modulus reduction and damping curves for sands falls

within a narrow band and no other factors significantly influence these curves. As a result,

an average modulus reduction curve and limiting curves based on relative density that can

be used in practice for cohesion less sands. Modulus and damping are a functions of shear

strain and can be obtained from stress or strain controlled cyclic loading [1].

The specimens were isotropically consolidated under confining stress of 0.5 kgf/cm2

and 1.0 kgf/cm2 were subjected to cyclic loading with a frequency of 0.1 Hz. Table 3.4

shows the relation between the double amplitude axial strain on the logarithmic scale and

the shear modulus during cyclic loading. It is seen in the table that shear modulus

corresponding to each effective confining stress increases as the effective confining stress

becomes higher. This behavior is in good agreement with the findings table 3.4 and suggest

that the dynamic shear modulus increases with increasing confining pressure. Their

relationship can be plotted as a straight line on a log-log scale at all levels of treatment and

strain amplitude. The relationship between the double amplitude axial strain on the

logarithmic scale and the damping ratio are presented in figure 3.3. It can also be seen that

the damping ratio decreases with increasing double amplitude axial strain. However, the

results presented in figure 3.3 show that at given loading frequency, the damping ratio

increases with increasing axial strain, and again when given axial strain, the damping ratio

increases with increasing loading frequency. When loading frequency increases, the axial


ISBN 978-967-2110-70-5

28

2017

strain decreases, damping energies decrease together, or damping ratios increase. The

results confirm the general findings by others for undisturbed cohesive soil.


The shear modulus decreases with the

increasing of confining stress

Th

e shear modulus decreases with the

increasing of frequencies

The damping ratio increases with the

increasing of confining stress

The damping ratio increases with the

increasing of frequencies

The hysteresis loops can be considered in order to calculate the secant shear

modulus, G and damping ratio, D from the slope and the total area of the loop respectively.

It has been observed an increase in shear modulus values with loading frequency from tests

on Leda clay, but they are not significant. It has been investigated that the influence of

loading frequency on modulus and damping ratio at very small strains using cyclic torsional

shear tests. They concluded that shear modulus values of soil are not sensitive to frequency


ISBN 978-967-2110-70-5

29

2017

variations, however, damping increases with decreasing frequency. However, it was found

damping is independent of changes in loading frequency from tests conducted at very small

strain amplitudes. As OCR increases, the strength of both clean sand and sand with

different clay fines decreases. The addition of clay fines in sand in over – consolidated, OC

state changes the excess pore pressure response significantly. Normally consolidated soil,

NC specimens yield highest peak strength. The peak strength of specimens decreases with

an increase in OCRs. The soil specimens generate positive maximum excess pore pressure

as clay fines increases. Skempton pore pressure parameter, A, values for soil specimens

decrease as OCR increases, but they increase with increasing clay fines content in soil

specimens. The initial tangent modulus of soil specimens increases as clay fines content

decreases. Heavily over - consolidated sand specimens yield lower initial tangent modulus,

but the effects of OCR on initial tangent modulus of sand with clay content is negligible [9].

Figure 3.3: Damping ratio vs axial strain under different OCR

3.6 CONCLUSIONS

Commonly, on past research, shows that normally consolidated, NC specimen were less

deformable than that over consolidated, OC ones but it depends on specific dynamic cyclic

parameter. The cyclic loading can produce a substantial decrease in the shear modulus, G

of the soil. The amplitudes of τcy degrade with the accumulation of pore pressure in clay

specimens as the number of cycle’s increases. The degradation of τcy and pore pressure

accumulation become more significant as strain amplitude increases in both normally and

over consolidated specimens. However, negative pore pressure is developed in over


ISBN 978-967-2110-70-5

30

2017

consolidated specimens particularly when strain amplitude is small and OCR is high which

can be attributed to dilatant tendency of clay. The hysteresis loops gradually tilt towards the

horizontal axis and gradual softening of soil is noted [9].


ISBN 978-967-2110-70-5

31

2017

REFERENCES

[1] Theenathayarl, T. (2015).

Behaviour of Sensitive Leda Clay

under Simple Shear Loading.

University of Peradeniya Sri

Lanka. Master of Applied Science

in Civil Engineering.

[2] Chan, C. et.al (2011). Some

Mechanical Properties of Cement,

Malaysian 5(2), 76–83.

[3] Thammathiwat, A., et.al (2004).

Behavior of Strength and Pore

Pressure of Soft Bangkok Clay

under Cyclic Loading. Stress: The

International Journal on the

Biology of Stress, 9(4).

[4] Rees, S. (2014). Part Three:

Dynamic triaxial testing. White

Paper, 5. Retrieved from

http://www.gdsinstruments.com/wh

ite-paper-dynamic-triaxial-testing

[5] Moradi, G., et.al (2015). The

Influence of Overburden Pressure

on Liquefaction Potential, 1–15.

[6] Thian. S. Y, et.al (2011). Stress

History Effect on Mining Sand

with Fines Contents, 2(1), 1–10.

[7] Jiang, M. (2012). Stiffness

Degradation of Soft Marine Clay

under Uniaxial Cyclic Loading.

Vol 17.

[8] Soltani-Jigheh, et.al (2010). Cyclic

behavior of Mixed Clayey Soils.

International Journal of Civil

Engineering, 8(2), 99–106.

[9] Thian. S. Y, et.al (2016). Effect of

OCR on Cyclic Shear Strength

Degradation of Marine Clay.

(2016), 4(1), 280–285.

A CRITICAL REVIEW ON THE DYNAMIC BEHAVIOUR OF OCR OF …

Documents