Definition of fines and liquefaction resistance of Maoluo River soil

Tzou-Shin Uenga, Chia-Wen Suna and Chieh-Wen Chena

a

Department of Civil Engineering, National Taiwan University, Taipei, Taiwan 10617

Tel/Fax: (02) 2362-1734
E-mails: [email protected]

Abstract
Two definitions of fines, soil particles finer than the No. 200 sieve (0.075 mm)
and No. 400 sieve (0.038 mm), are adopted to evaluate the effect of fines content on
the liquefaction resistance of the silty soil in the Maoluo River area of Nantou City.
The specimens of different fines contents according to these two definitions of fines
were prepared by moist tamping to a dry density of 1400 kg/m3. Comparisons were
made between the liquefaction resistances of soils containing fines according to
different definitions. The results of cyclic triaxial tests show that the effect of fines on
the liquefaction resistance of Maoluo River soil is more pronounced based on the
fines content defined as the portion finer than the No. 400 sieve than that with fines
defined as soil passing the No. 200 sieve.

Key words: Fines content, Fines definition, Liquefaction, Cyclic triaxial tests,
Earthquake

Corresponding author. Tel.: +886-2-2362-1734

E-mail address: [email protected] (T.S. Ueng)
1

1. Introduction
In the evaluation of liquefaction potential of a sandy soil, the effect of fines is a
important factor, e.g., [1]. It is usually considered that soil with higher fines content
(FC) will have a higher liquefaction resistance. However, in the recent earthquakes,
e.g., 1999 Kocaeli earthquake and Chi-Chi earthquake, liquefaction occurred in the
soils with very high FC [2, 3]. Therefore, the liquefaction behavior of fines and the
effect of fines on the liquefaction resistance of a soil were extensively studied recently,
e.g., [4, 5, 6].
The fines of a soil is presently defined as the portion of soil particles finer than
the No. 200 sieve (0.075 mm) [7], which was determined arbitrarily regardless of soil
behavior. It is not clear whether this definition of fines is suitable in the evaluation of
liquefaction potential or possibly, other definitions of fines can better reflect the effect
of fines on the liquefaction resistance of the soil. In this study, the portion of a soil
finer than the No. 400 sieve (0.038 mm) is considered as an alternative definition for
FC of a soil and the effect of FC on liquefaction resistance was evaluated. The silty
soil liquefied during the 1999 Chi-Chi earthquake in the Maoluo River area in Nantou
City was selected for the study. Dynamic triaxial tests were performed on the
specimens with various FC's based on the new definition of fines. The results of the

liquefaction resistances of these specimens were compared with those of the

specimens in which the fines content are defined as the portion finer than the No. 200
sieve.

2. Soil Samples
The soil used in the study was taken about 1.5 m below ground surface from the
bank of Maoluo River where severe liquefaction occurred during the 1999 Chi-Chi
earthquake. The SPT N-values of the liquefiable soils in this area are around 2 to 5.
The grain size distribution curve of this soil is shown in Fig. 1. The properties of this
soil are given in Table 1. The maximum density was tested according to ASTM
D4253 method 1B (wet method) that gives a higher maximum density than the dry
method. It can be seen that 48% and 37% of the soil are finer than the No. 200 sieve
and No. 400 sieve, respectively. The soil is a silty sand (SM) according to the Unified
Soil Classification System (USCS), but, with such a high FC, it is like a sandy silt
(ML). The soil in the field is in a loose state with a dry density of 1400 kg/m3 (Dr
31%).
Two series of dynamic triaxial tests were performed on the Maoluo River soil,
one with the fines defined as the soil finer than the No. 200 sieve and another with
fines defined as the soil finer than the No. 400 sieve. For the study of the properties of

the fines and preparation of the test specimens with different fines contents, the fines
were separated from the soil using the wet sieving method by washing the soil
particles through the No. 200 and No. 400 sieves. The portion of soil particles larger
than the No. 4 sieve (4.75 mm) was replaced with the coarse-grained material passing
the No. 4 sieve. The properties of the coarse-grained portion and fines according to
these two definitions are presented in Table 2. Table 3 also shows the mineral
compositions for three different ranges of grain sizes (D), i.e., (a) coarser than the
No.200 sieve (D > 0.075 mm), (b) between the No. 200 and No. 400 sieves (0.038
mm < D < 0.075 mm), and (c) finer than the No. 400 sieve (D < 0.038 mm). It can be
seen that the portion of soil passing the No. 400 sieve resembles a clayey soil of some
plasticity, while soil passing the No. 200 sieve is a low-plastic silt. There is mainly
quartz and very little clay minerals in the portion of Maoluo River soil between the
No. 200 and No. 400 sieves. It is essentially a very fine sand. That is, this portion of
fines will probably not significantly affect the liquefaction resistance of the soil.
The triaxial test specimens were prepared by the moist tamping method to the
field dry density of 1400 kg/m3. Based on the definitions of fines, the coarse-grained
materials were mixed with different amounts of fines according to the FC's of the
specimens. An 8% of moisture content was added to facilitate the sample preparation.
The specimens were consolidated to the estimated field effective stress of 78 kPa

before application of cyclic loading at a frequency of 1 Hz. Details of the sample

preparations and testing procedures are given in [8, 9]. The deviator stresses, pore
water pressure changes, and axial deformations were recorded during the cyclic
loading tests.

3. Test Results
For convenience of discussions hereafter, (FC)200 and (FC)400 represents the fines
contents of a specimen with fines defined as soil finer than the No. 200 sieve (D <
0.075 mm) and the No. 400 sieve (D < 0.038 mm), respectively.
3.1 Fines defined as D < 0.075 mm
For the specimens with fines defined as soil finer than the No. 200 sieve (D <
0.075 mm), six different (FC)200's were considered, i.e., 0%, 20%, 35%, 48%, 65%,
and 100%. The soil with (FC)200 = 48% is the original soil from the site. The relations
of cyclic stress ratio (CSR) versus number of cycles to 5% single amplitude axial
strain for different (FC)200's are presented in Fig. 2. Figure 3 shows the axial strain of
specimens with (FC)200 = 0%, 48%, and 100% under cyclic loadings. It can be seen
that the deformation of the 100 % fines increases gradually during the cyclic loading
test, and the strain reaches a substantial value before the effective stress reaches zero,
i.e., initial liquefaction. Whereas, the strains of specimens with (FC)200 = 0% and 48%

remain rather small and increase rapidly when they reached initial liquefaction. The
number of stress cycles to 5% axial strain (NL) appears more sensitive to the change
of CSR for sand with (FC)200 = 0%. NL increases less rapidly for soil with higher FC
when CSR decreases. Considering an earthquake of a magnitude of 7.5, an equivalent
number of stress cycles of 15 is usually taken to represent the duration of the
earthquake loading. The liquefaction resistances (CRR) under 15 cycles of loading for
Maoluo River soil with different (FC)200's are given in Fig. 4. The CRR of the soil
decreases when its (FC)200 increases from 0% and reaches the minimum value when
(FC)200 is around 30%. The liquefaction resistance increases when (FC)200 increases
beyond 30%.
3.2 Fines defined as D < 0.038 mm
Six different (FC)400's, i.e., 0%, 10%, 20%, 30%, 37% and 100%, were prepared
for the specimens with fines defined as soil finer than the No. 400 sieve (D < 0.038
mm). Here the coarse-grained material is the soil with D > 0.038 mm which is
different from that for fines defined as D < 0.075 mm above. The relations between
CSR and number of cycles to 5% single amplitude axial strain for different fines
contents are also shown in Fig. 5. As for the fines defined as D < 0.075 mm, Fig. 5
also shows that NL is more sensitive to the changes of CSR for specimens with a
lower (FC)400. The relation for soil with (FC)400 = 37%, which is the original in situ

soil, is the same as that with (FC)200 = 48% for fines defined as D < 0.075 mm. Figure
6 shows the axial strain of specimens with (FC)400 = 0%, 30%, and 100% under cyclic
loadings. The results show that for (FC)400 = 0 and 30%, the behavior of soil under
cyclic loading is quite similar to that of clean sand, i.e., there is very little deformation
developed before initial liquefaction, but large strains occurred rapidly at initial
liquefaction. Examining the stress-strain relationship, it was found that for (FC)400 =
100%, the soil behaved like a clayey soil with degradation of modulus under cyclic
loading; and the excess pore water pressure and strain developed gradually and no
initial liquefaction occurred even when the axial strain reached over 10%. The values
of CRR under 15 cycles of loading for Maoluo River soil with different (FC)400's are
also given in Fig. 4 that shows greater differences of CRR for specimens with
different (FC)400's. There is also a minimum liquefaction resistance for this soil with
FC near 20%.

4. Comparison of liquefaction resistance between different fines definitions

The differences of liquefaction resistance between soils with fines defined as D <
0.075 mm and D < 0.038 mm can be evaluated by comparing the relations of CRR
versus FC for these two fines definitions given in Figs. 4. It shows that the effect of
fines on the liquefaction resistance of Maoluo River soil is more pronounced using the

definition of fines of D < 0.038 mm, but the trends of the effect of FC on liquefaction
are quite similar, and there exist minimum liquefaction resistances for both definitions
of fines when FC is around 20%-30%. The liquefaction resistances of the specimens
with FC = 100% are approximately 1.28 and 1.45 times those minimum values for
fines defined as D < 0.075 mm and D < 0.038 mm, respectively. The differences in
grain size and plasticity of fines filling between the coarse particles might cause the
difference of effect on the liquefaction resistances.
For the specimens with fines defined as D < 0.038 mm, we can recalculate
(FC)200 of each specimen considering the portion of soil with D > 0.075 mm as the
coarse-grained material. Thus, for each (FC)400 of the soil specimen with fines defined
as D < 0.038 mm, there is a correspondent equivalent (FC)200 if we define fines as D <
0.075 mm, only the compositions of fines are different for specimens of different
equivalent (FC)200's. That is, the higher the equivalent (FC)200, the less portion of
grains between 0.038 mm and 0.075 mm. Table 4 shows the proportions of soil in the
three grain size ranges for each (FC)400 and its correspondent equivalent (FC)200. The
CRR of these specimens can then be plotted against the equivalent (FC)200's as shown
in Fig. 7. The relations for CRR versus FC for different fines definitions given in Fig.
4 are also shown on Fig. 7 for comparison. Figure 7 indicates that with different fines
compositions, CRR are substantially different even though (FC)200 are the same.

For (FC)400 = 0% or equivalent (FC)200 = 17.5%, the fines with grain size only in
the range of 0.038 < D < 0.075 mm is essentially a very fine sand as shown in Table 3.
There is a very slight difference in liquefaction resistance between this soil and the
clean sand with only particles larger than the No. 200 sieve. It depicts that fines
between the No. 200 sieve and the No. 400 sieve may not affect the liquefaction
resistance of sand significantly.
On the other hand, for the specimens with fines defined as D <0.075 mm we can
also obtain the equivalent (FC)400's by considering the coarse material as D > 0.038
mm. In this case, the compositions of coarse material are different for different
equivalent (FC)400's. Table 5 shows the proportions of soil in the three grain size
ranges and the corresponding equivalent (FC)400 for each (FC)200 . The relation of
CRR versus equivalent (FC)400 is plotted on Fig. 8 together with the relations for CRR
versus (FC)200 and (FC)400 from Fig. 4. It is interesting to find that the relation of CRR
versus equivalent (FC)400 is quite close to the relation for CRR versus (FC)400 even
though the compositions of the coarse materials are different in these two fines
definitions. That is, soil with the same proportion of fines with D < 0.038 mm,
regardless of the composition of the coarse material of D > 0.038 mm, will have about
the same liquefaction resistance. This infers that (FC)400 might be a better parameter
for considering the effect of fines on the liquefaction resistance of a soil.

It should be noted that the comparisons of the effect of FC are based on the same
dry density of 1400 kg/m3 for all specimens. Similar results of comparison might hold
for a denser soil but further studies are needed.

5. Concluding remarks
Two definitions of fines for Maoluo River soil, i.e., D < 0.075 mm and D < 0.038
mm, are considered to evaluate the effect of FC on the liquefaction resistance
subjected to cyclic loading. The soil with the grain size between 0.038 mm and 0.075
mm is essentially a very fine sand and its presence will probably not significantly
affect the liquefaction resistance of Maoluo River soil. The results of dynamic triaxial
tests show that the effect of fines on the liquefaction resistance of the soil is more
prominent using (FC)400 than (FC)200, probably due partly to the difference in
plasticity of the fines. Even though (FC)400 is probably a better indicator for the effect
of FC on liquefaction resistance of a soil, there is no sure overall benefit yet to change
the present definition of fines, i.e., D < 0.075 mm. However, in liquefaction potential
evaluation, corrections of CRR solely based on FC without considering the properties
of fines cannot truly reflect the effect of fines on the liquefaction resistance of a soil.
Further studies are needed for a better way of considering the effect of fines on
liquefaction resistance.

10

Acknowledgement
This study was supported by the National Science Council, Taiwan, Grant Nos.
NSC 90-2211-E-002-093 and NSC 91-2211-E-002-049. The authors also thank
Department of Geosciences, National Taiwan University for their helps in mineral
analyses of the soil samples.

References
[1] Youd TL, Idriss IM, Andrus RD, Arango I, Castro G, Christian JT, Dobry R, Finn
WD Liam, Harder Jr. LF, Hynes ME, Ishihara K, Koester JP, Liao SSC, Marcuson III
WF, Martin GR, Mitchell JK, Moriwaki Y, Power MS, Robertson PK, Seed RB,
Stokoe II KH. Liquefaction resistance of soils: summary report from the 1996
NCEER and 1998 NCEER/NSF workshops on evaluation of liquefaction resistance of
soils. J Geotech Geoenviro Engng, ASCE 2001;127(10):817-833.
[2] Sancio RB, Bray JD, Stewart JP, Youd TL, Durgunoglu HT, Onalp A, Seed RB,
Christensen C, Baturay MB, Karadayilar T. Correlation between ground failure and
soil conditions in Adpazari, Turkey. Soil Dynamics and Earthquake Engineering 2002;
22(9-12):1093-1102.
[3] Ueng TS, Lin ML, Chen MH. Some geotechnical aspects of 1999 Chi-Chi, Taiwan

11

earthquak. In: Proceedings of the 4th International Conference on Recent Advances in

Geotechnical Earthquake Engineering and Soil Dynamics. San Diego, CA, 2001,
SPL-10.1.
[4] Prakash S, Guo T, Kumar S. Liquefaction of silts and silt-clay mixtures. In:
Dakolas P, Yegian M, Holtz RD, editors. Geotechnical Earthquake Engineering and
Soil Dynamics III. ASCE Geotech Spec Plubl 1998;75, Vol 1:337-348.
[5] Thevanayagam S, Fiorillo M, Liang J. Effect of non-plastic fines on undrained
cyclic strength of silty sands. In: Pak RYS, Yamamura J, editors. Soil Dynamics and
Liquefaction 2000. ASCE Geotech Spec Publ 2000;107:77-91.
[6] Yamamura JA, Kelly M. Monotonic and cyclic liquefaction of very loose sands
with high silt content. J Geotech Geoenviron Engng, ASCE 2001; 127(4):314-323.
[7] ASTM D 653-02. Standard terminology relating to soil, rock, and contained fluids.
Annual Book of ASTM Standards 2003; 04.08:43-77.
[8] Chen CW. Effect of characteristics of fines on soil liquefaction resistance. MS
Thesis, National Taiwan University, Taiwan; 2002.
[9] Sun CW. Effect of fines definition on liquefaction resistance of sand. MS Thesis,
National Taiwan University, Taiwan; 2003.

12

Figure Captions
Fig. 1. Grain size distribution of Maoluo River soil
Fig. 2 . Cyclic stress ratio (CSR) versus No. of cycles to 5% axial strain for Maoluo
River soil with fines defined as D < #200 sieve
Fig. 3. Axial strains of specimens with various (FC)200's
Fig. 4. Cyclic resistance ratio (CRR) causing 5% axial strain under15 cycles of
loading
Fig. 5. Cyclic stress ratio (CSR) versus No. of cycles to 5% axial strain for Maoluo
River soil with fines defined as D < #400 sieve
Fig. 6. Axial strains of specimens with various (FC)400's
Fig. 7. Cyclic resistance ratio (CRR) versus equivalent (FC)200
Fig. 8. Cyclic resistance ratio (CRR) versus equivalent (FC)400

13

Table 1
Properties of Maoluo River Soil
Specific gravity

2.68

In situ d

1400 kg/m3

D50

0.078 mm

Cu

43

Maximum d

1717 kg/m3

Minimum d

1290 kg/m3

LL

23

PI

NP

FC (< 0.075 mm)

48 %

USCS

SM

14

Table 2
Properties of soils of different grain size ranges
Grain size

> #200 sieve

(> 0.074 mm)

< #200 sieve

(< 0.074 mm)

> #400 sieve

(> 0.038 mm)

Gs

2.67

2.69

2.68

2.70

LL

36

29

38

PI

15

USCS

SP

ML

SM

CL

15

< #400 sieve

(< 0.038 mm )

Table 3
Mineral compositions in percent of soils of different grain size ranges
Grain size

Rock
fragments

Quartz

Feldspar

Mica

Illite

> #200

24

68

87

41

28

25

< #200
> #400
#400

16

Chlorite Kaolinite

Table 4
Grain size compositions in percent of specimens with various (FC)400
(FC)400

10

20

30

37

100

Equivalent (FC)200

17.5

25.7

34

42.2

48

100

D > 0.075 mm

82.5

74.3

66

57.8

52

0.075 mm > D > 0.038 mm

17.5

15.7

14

12.2

11

10

20

30

37

100

D < 0.038 mm

17

Table 5
Grain size compositions (%) of specimens with various (FC)200
(FC)200

20

35

48

65

100

Equivalent (FC)400

15.4

27

37

50.1

77.1

100

80

65

52

35

0.075 mm > D > 0.038 mm

4.6

11

14.9

22.9

D < 0.038 mm

15.4

27

37

50.1

77.1

D > 0.075 mm

18

Percent passing (%)

100

80

60

40

20

10

0.1

Grain size (mm)

Fig. 1. Grain size distribution of Maoluo River soil

19

0.01

0.001

0.4

'c = 78 kPa

FC = 0%
FC = 20%

0.35

FC = 48%
FC = 65%

CSR

0.3

FC = 100%

0.25

0.2

0.15

0.1
1

10

100

No. of cycles to 5% strain

Fig. 2 . Cyclic stress ratio (CSR) versus No. of cycles to 5% axial strain for Maoluo
River soil with fines defined as D < #200 sieve

20

20
FC = 0%, CSR = 0.147

15

FC = 48%, CSR = 0.130

FC = 100%, CSR = 0.192

Axial strain (%)

10
5
0
-5
-10
-15
0

10

15

20

Time (s)

Fig. 3. Axial strains of specimens with various (FC)200's

21

25

30

35

0.25
Fines definition
D < #200 sieve
D < #400 sieve

CRR

0.2

0.15

0.1
0

20

40

60

80

100

FC (%)

Fig. 4. Cyclic resistance ratio (CRR) causing 5% axial strain under15 cycles of
loading

22

0.30
FC = 0%

Dry density = 1400 kg/m

'c = 78 kPa

FC = 10%
FC = 20%

0.25

FC = 30%

CSR

FC = 100%

0.20

0.15

0.10
1

10

100

No. of cycles to 5% axial strain

Fig. 5. Cyclic stress ratio (CSR) versus No. of cycles to 5% axial strain for Maoluo
River soil with fines defined as D < #400 sieve

23

20
FC = 0%, CSR = 0.161

Axial strain (%)

15

FC = 30%, CSR = 0.157

10

FC = 100%, CSR = 0.193

5
0
-5
-10
-15
-20
0

10

15
Time (s)

Fig. 6. Axial strains of specimens with various (FC)400's

24

20

25

30

0.25
Fines definition
D < #200 sieve
D < #400 sieve
Equivalent D < #200

CRR

0.2

0.15

0.1
0

20

40

60

FC (%)

Fig. 7. Cyclic resistance ratio (CRR) versus equivalent (FC)200

25

80

100

0.25
Fines definition
D < #200 sieve
D < #400 sieve
Equivalent D < #400

CRR

0.2

0.15

0.1
0

20

40

60

80

FC (%)

Fig. 8. Cyclic resistance ratio (CRR) versus equivalent (FC)400

26

100