correlation in Burland

Discussion

Examples

Compression data gathered for a wide stress range are fitted using these models in Fig. 1 for both remolded soils and for natural soils with distinct yield stress; fitting parameters are summarized in Table 1. Data points and fitted models are plotted in both log-linear and linear-linear plots; the last column in Fig. 1 shows computed trends for the tangent constrained modulusMtan. The following can be observed:

• Remolded soils: All four-parameter models addressed here can adequately fit experimental data gathered for remolded soils [Fig. 1(a)]. The tangent stiffness computed with these models shows monotonic stiffening.

• Natural soils: Hyperbolic, arctangent, power, and exponential (albeit to a lesser extent) models written in terms of σ 0β approx- wimate the brittle response of natural structured soils better than logðσ 0Þ models [Fig. 1(b)]. The tangent stiffness com- puted with hyperbolic and arctangent models shows early softening until the yield stress, followed by compaction-driven stiffening. Furthermore, the S-shaped models presented earlier are well

suited to fit the response of overconsolidated soils as well. These observations are confirmed with multiple cases compiled from the literature.

Correlations—Low-Stress Void Ratio and Compressibility

A database of consolidation tests was compiled from the literature for the purposes of this study [The complete data- base can be found in Chong (2014).] The classical Terzaghi model was fitted to data gathered with remolded, normally consolidated soils within the available effective stress range. Results show that the void ratio e1 kPa is closely related to the void ratio at the liquid limit eLL ¼ GsLL=100 (assuming 100% saturation)

Fig. 1. Wide stress range 1D compression data (ko, zero lateral strain boundary condition) fitted with different soil compressibility models: (a) re- molded clay—sodium montmorillonite at 0.001 N and pH ¼ 7 (data fromMesri and Olson 1971); (b) natural clay—Bothkennar soil from 6.5 m (data from Burland 1990)

© ASCE 06016003-3 J. Geotech. Geoenviron. Eng.

J. Geotech. Geoenviron. Eng., 06016003

D ow

nl oa

de d

fr om

a sc

el ib

ra ry

.o rg

b y

Po rt

la nd

S ta

te U

ni ve

rs ity

o n

05 /0

1/ 16

. C op

yr ig

ht A

SC E

. F or

p er

so na

l u se

o nl

y; a

ll ri

gh ts

r es

er ve

d.

e1 kPa ¼ 5

4 eLL ¼

1

80 Gs · LL

remolded soils ð28 cases; SD ¼ 0.36;R2 ¼ 0.92Þ ð13Þ

This equation compares well with the eLL versus e100 correlation in Burland (1990), as e1 kPa ¼ e100 kPa þ 2Cc. Furthermore, a direct link is obtained between the void ratio at 10 kPa and the void ratio at liquid limit

e10 kPa ¼ eLL remolded soils ð28 cases; SD ¼ 0.26;R2 ¼ 0.90Þ ð14Þ

The correlation between compressibility and liquid limit was recognized early on in the field (Skempton 1944; Terzaghi and Peck 1948; Burland 1990). The presented database shows a very similar trend

Cc ¼ 0.008 · ðLL − 5Þ remolded soils ð28 cases; SD ¼ 0.11;R2 ¼ 0.90Þ ð15Þ

Given the eLL − e1 kPa and Cc − LL correlations, this study ex- plored the relationship between model parameters e1 kPa and Cc [Fig. 2(a)]