Application of the dilatometer test for estimating undrained shear strength of Busan New Port clay

Ocean Engineering 115 (2016) 39–47

Contents lists available at ScienceDirect

Ocean Engineering journal homepage: www.elsevier.com/locate/oceaneng

Application of the dilatometer test for estimating undrained shear strength of Busan New Port clay Hyunwook Choo a, Woojin Lee a, Sung-Jin Hong b, Changho Lee c,n a

School of Civil, Environmental and Architectural Engineering, Korea University, Seoul 136-701, Korea Technology Research and Development Institute, Daelim Industrial Co., Itd., Seoul 110-150, Korea c Department of Marine and Civil Engineering, Chonnam National University, Yeosu 550-749, Korea b

art ic l e i nf o

a b s t r a c t

Article history: Received 3 March 2015 Accepted 17 November 2015

The dilatometer test (DMT) is an economical and fast tool that evaluates the stratigraphy and properties of soil. In this study, a series of DMTs, field vane tests (FVTs), and K0 consolidated-undrained triaxial compression (CK0UC) tests were performed to develop an empirical correlation for the undrained shear strength (su) of clayey soils from the Busan New Port site. The stress-normalized mobilized su(μsu =σ 0v ) for both laboratory tests and FVTs is determined to be around 0.22, which is consistent with previous studies. The exponent m for overconsolidation ratio (OCR), which is in the relation of su =σ 0v ¼ S  OCRm, is determined to be 0.83 by using the SHANSEP technique. Two different methods of estimating su are reviewed to make use of either the horizontal stress index (KD) or the bearing factor (Nc). Because the DMT results indicate that the Nc is linearly proportional to the material index (ID), representing the characteristics of soil, the empirical su estimating formula with ID is newly suggested in this study. According to the values for the mean absolute percentage error (MAPE), the empirical correlation with ID shows a slightly better accuracy than that using KD. However, a comparison between the values for the su that are measured and estimated by using two different methods shows good agreement in general. & 2015 Elsevier Ltd. All rights reserved.

Keywords: Dilatometer test (DMT) Field vane test (FVT) K0 consolidated-undrained triaxial compression test Mobilized undrained shear strength

1. Introduction The undrained shear strength (su) of soil is a key parameter that influences the design of the earth structures in soft soils. The su can be measured directly by using the field vane test (FVT), triaxial test, and simple shear test. In addition, it can be indirectly predicted through field penetration tests, such as the piezo cone penetration test (CPTu) and flat dilatometer test (DMT). These field penetration tests, which are fast and economical, can adequately reflect in-situ conditions. However, the measured results must be interpreted by using suitable empirical correlations to evaluate the su because the data obtained from these tests reflects complex interactions between the test equipments, various soil properties, and the in-situ conditions (Chung et al., 2010; Jamiolkowski et al., 1988; Lunne et al., 1997; Marchetti et al., 2001; Mayerhof, 1956). In addition, because the empirical correlations vary with regional soil characteristics, proper care must be taken to perform an effective evaluation of su. Since Marchetti (1980) first introduced the DMT as a soil testing technique in 1980, the DMT has been extensively used in many n

Corresponding author. Tel.: þ 82 61 659 7322; fax: 82 61 659 7329. E-mail address: [email protected] (C. Lee).

http://dx.doi.org/10.1016/j.oceaneng.2015.11.017 0029-8018/& 2015 Elsevier Ltd. All rights reserved.

geotechnical investigations (Marchetti, 2006; Robertson, 2009). The DMT measures the values for the horizontal pressures p0 (lifeoff pressure) and p1 (1.1 mm deflection pressure) at specific horizontal displacements of the soil. Three DMT indices, including the material index (ID), the horizontal stress index (KD), and the dilatometer modulus (ED), are evaluated by using the measured p0 and p1 values. A number of empirical correlations based on these DMT indices have been suggested to evaluate various geotechnical properties of soils, including the unit weight (γt), coefficient of lateral earth stress at rest (K0), overconsolidation ratio (OCR), undrained shear strength (su), constrained modulus (M), small strain shear modulus (G), and soil classification (Baldi et al., 1989; Hryciw, 1990; Iwasaki et al., 1991; Jamiolkowski et al., 1985; Kamei and Iwasaki, 1995; Lacasse and Lunne, 1989; Marchetti, 1980; Powell and Uglow, 1988; Roque et al., 1988). Since the DMT generates less disturbance in the soil due to the shape of blade and works on a small strain level than through other penetration tests such as the SPT (standard penetration test) and CPT, the stress history and deformation of clayey soil can be better evaluated through DMT (Baligh and Scott, 1975; Cruz et al., 2006; Iwasaki et al., 1991; Kamei and Iwasaki, 1995; Lacasse and Lunne, 1989; Lutenegger 1988; Marchetti, 1980; Marchetti et al., 2001; Monaco et al., 2007; Powell and Uglow, 1988). Because the stress history (or overconsolidation) is closely related to the undrained shear

40

H. Choo et al. / Ocean Engineering 115 (2016) 39–47

strength (Ladd and Foott, 1974), the su obtained through a DMT is reliable (Byeon et al., 2006). The purpose of this study is to evaluate the mobilized su (μsu) of Busan New Port clay using the DMT results. Therefore, a series of in-situ tests, including the DMT and FVT, were performed at the Busan New Port site. In addition, undisturbed samples were retrieved with depths to measure the basic index properties and to conduct CK0UC (K0 consolidated-undrained triaxial compression) tests. The DMT indices, including the material index (ID), dilatometer modulus (ED), and horizontal stress index (KD), were evaluated by taking the confining stress into account. The μsu was calculated using the measured su through FVTs and CK0UC tests with the recommended correction factors due to the discrepancy between the measured su from in-situ or laboratory tests and the mobilized su in the actual field. Previous μsu estimating formulas using the DMT results have been reviewed, and the new empirical formula for estimating μsu of Busan New Port clay is suggested in this study.

2. Undrained shear strength (su) based on DMT The su varies according to the shearing mode, anisotropy, strain rate, boundary conditions, in-situ testing device, confining stress level, and other factors (Kamei and Iwasaki, 1995; Ladd, 1991; Mitchell and Soga, 2005). Therefore, the su is not unique for a given soil, but depends on the type of testing (Mayne, 2008). The su is 0 typically normalized with the vertical effective stress (σ v ) to obtain the following form: 0

ðsu =σ v ÞNC ¼ S

where m ¼fitting parameter, typically 0.8 (Ladd and Foott, 1974; Ladd et al., 1977). OCR of clays in Eq. (2) can be estimated with KD (Marchetti, 1980): ð3Þ

Additionally, Marchetti (1980) proposed S¼ 0.22 and m ¼0.8 in Eq. (2) by comparing the results of field vane tests. Therefore, the substitution of S and m values of Marchetti (1980), and Eq. (3) into Eq. (2) yields the following su estimating formula based on the DMT results (Kamei and Iwasaki, 1995; Lacasse and Lunne, 1989; Marchetti, 1980; Powell and Uglow, 1989): 0

su =σ v ¼ 0:22 Uð0:5K D Þ1:25

ð4Þ

However, it is important to note that the S in Eq. (2) can vary depending on the testing type or shearing mode as shown in Table 1 (Kamei and Iwasaki, 1995). The bearing capacity theory can be also used with the DMT results for an alternate method to estimate su, and this method is similar to the su estimating formulas based on the cone tip resistance (Ebrahimian et al., 2012; Mayne 2008). An inverted form of the equation for the bearing capacity is (Roque et al., 1988): su ¼

p1  σ h Nc

Type of testing

S in Eq. (1)

Triaxial compression Direct simple shear Triaxial extension Plane strain compression Plane strain extension Field vane test Self-boring pressure meter

0.33 0.20 0.16 0.33 0.19 0.21 0.42

the soil types. It is interesting to note that Eq. (4) uses p0 value to estimate su, while Eq. (5) employs p1 value. Because of the difficulty in a direct estimation of σh in Eq. (5), the coefficient of lateral 0 0 0 earth stress at rest (K 0 ¼ σ h =σ v , where σ h ¼ horizontal effective stress) has been used to evaluate σh (Ku and Mayne, 2013). Following the work of Marchetti (1980), K0 of clays can be expressed as a function of KD (Eq. (6)). Therefore, σh in Eq. (5) can be estimated using Eq. (7). K 0 ¼ ðK D =1:5Þ0:47  0:6 h

ð6Þ i

σ h ¼ σ 0v U ðK D =1:5Þ0:47  0:6 þu0

ð7Þ

where u0 ¼hydrostatic pore water pressure.

ð1Þ

where subscript NC¼normally consolidated state; and S ¼undrained strength ratio at NC, which is given in Table 1. Since the su/σ0 v depends on the overconsolidation ratio (OCR), the stress history and normalized soil engineering properties (SHANSEP) technique has been developed for use in practice (Ladd and Foott, 1974; Mitchell and Soga, 2005):
0 0  su =σ v ¼ su =σ v NC U OCRm ¼ S U OCRm ð2Þ

OCR ¼ ð0:5 UK D Þ1:56

Table 1 Undrained strength ratios for Boston Blue Clay (after Mayne 2008).

ð5Þ

where σh ¼total horizontal stress; and Nc ¼ bearing factor. Based on the comparison with the triaxial compression test results, Roque et al. (1988) suggested that the Nc ranges from 5 to 9 according to

3. Site description Busan New Port is located in the southeast coastal area of the Korean peninsula as shown in Fig. 1. The clayey soils in this region are typically referred to as Busan clay. Previous studies (Chung et al., 2002, 2007, 2005) have reported that the Busan clay layer is extended to a maximum depth of 70 m. The clay layer tested at the Busan New Port is deposited from EL-2 m up to EL-50 m and can be divided into two layers at EL-30 m (i.e., a soft upper clay layer and a stiff lower clay layer). The New Port was constructed by carrying out ground improvement with prefabricated vertical drains (PVDs) and the gravel surcharge load of 13 m in height. This surcharge load corresponds to a vertical effective stress increment of  260 kPa. Fig. 2 shows the profiles of typical properties of the Busan New Port clay. The natural water content (wn) ranges from 35% to 75%, and it is almost equal to or slightly lower than the liquid limit (wL) for the entire depth. The plastic limit (wp) is within a relatively narrow range, from 20% to 35%. These values increase as the depth increases down to EL-25 m and decrease with a depth below EL25 m (Chung et al., 2002; Kim, 2008; Tanaka et al., 2001). The total unit weight (γt) remains relatively constant up to EL-30 m, and then it increases with depth, which may reflect two different layers. The percent passing of # 200 sieve (0.075 mm in sieve size) ranges from 83% to 99%. The activity (¼ plasticity index/clay fraction) of New Port clay ranges from 0.6 to 1.1. Except some depths with sharp increases in cone tip resistance (qt) (or sharp decreases in measured pore water pressure behind the cone tip (u2)) reflecting the existence of layers with high silt or shell content, both qt and u2 increase with an increase in depth (Chung et al., 2007). It is notable that relatively low qt values (o5 MPa) and high u2 values ( 4u0) infer that the tested soil layers are clayey soils (Mayne, 2007).

H. Choo et al. / Ocean Engineering 115 (2016) 39–47

41

Fig. 1. Location of Busan New Port in South Korea.

Fig. 2. Profiles of typical index properties. Note that wn ¼ natural water content; wL ¼ liquid limit; wP ¼ plastic limit; γt ¼ total unit weight; qt ¼ cone tip resistance; u2 ¼pore water pressure behind the cone tip.

4. Experimental methodologies

4.2. Laboratory tests

4.1. Sampling technique

The consolidation tests were performed to evaluate the overconsolidation ratio (OCR) of the undisturbed samples according to ASTM-D2435 Method A. A standard oedometer cell with 63.5 mm in diameter and 20 mm in height was used, and the specimens were tested in K0 (zero lateral strain) conditions at vertical stresses (σ 0v ) ranging from 4.9 kPa to 1255.7 kPa (load increment ratio ¼ 1). Settlements were measured with time at every loading step, and each loading was kept for 24 h. The preconsolidation stress (σ 0p ) of each testing specimen was interpreted from the relation between void ratio, which was calculated from the settlement data, and log σ 0v using the Casagrande graphical technique (Casagrande, 1936).

The undisturbed samples were obtained using a hydraulically operated piston sampler (ASTM-D6519, 2008). The stainless steel sampling tube, which has 73.1 mm inner diameter and 1.5 mm wall thickness, was used in this study. Two sampling boreholes were advanced using a hydraulic drilling rig. Each sampling was performed approximately 3 m apart from the location of field penetration tests such as FVTs and DMTs to prevent sample disturbance. After sampling, the specimens were sealed with a plastic cap and wax to maintain in-situ water contents until the laboratory experiments were conducted.

42

H. Choo et al. / Ocean Engineering 115 (2016) 39–47

OCR of each specimen was calculated using the simple definition of OCR¼ σ 0p / in-situ effective stress. CK0UC (K0 consolidated-undrained triaxial compression) tests were conducted to evaluate the su of the undisturbed samples. The CK0UC test is well known to be an appropriate method to evaluate the su because it reenacts the in-situ stress conditions (Bjerrum, 1973). A trimmed specimen was saturated with a 100–150 kPa backpressure, and then K0-consolidated with the confined stresses corresponding to the in-situ effective stresses. After consolidation, the specimen was sheared at a strain rate of 0.05%/min. In addition, to evaluate the undrained shear strength for the various stress conditions, the SHANSEP method was applied (Ladd and Foott, 1974) in this study since su/σ0 v depends on the OCR (Jamiolkowski et al., 1985; Ladd et al., 1977; Marchetti, 1980). 4.3. In-situ tests A series of DMTs (dilatometer tests) were conducted at the Busan New Port site. A standard dilatometer, which has 15 mm in thickness, 95 mm in width, 230 mm in length, and 24° in apex angle at the edge, was penetrated into the tested site with the penetration rate of 2 cm/s. The DMTs were carried out every 0.2 m intervals in two different holes, and the ID, KD, and ED were calculated by using the measured p0 and p1 values at each depth. In addition, fifty FVTs (field vane tests) were conducted to measure the su at the tested site from EL-7 m to EL-36 m. A Geonor type four-blade vane, which has 2 mm in blade thickness, 55 mm in diameter, and 110 mm in height, was used in this study. The FVT is the most widely used testing method to estimate the su of in-situ soft clays (Chandler, 1988). The FVT begun  5 min from the end of the vane penetration and rotated at a rate of 0.1°/s (i.e., 6°/min) (ASTM-D2573, 2008).

increase in the applied stress, and the values range from 700 kPa to 3500 kPa. 5.2. OCR values Fig. 4 presents the OCR values estimated through empirical correlations using the KD suggested by Marchetti (1980) and Powell and Uglow (1988) and those obtained from the conventional oedometer tests (load increment ratio¼1 and duration of each load¼1 day). It has been reported that the Busan clay is normally consolidated (Chung et al., 2012, 2007), and the measured OCR values by oedometer tests (i.e., OCR ¼ 0.7–1.5 in Fig. 4) also indicate that the tested Busan New Port clay is normally consolidated. Note that some measured OCR values which are smaller than 1.0 may reflect the effect of sample disturbance (Chung et al., 2007). The OCR values estimated through Marchetti (1980)'s method (i.e., OCR ¼(0.5KD)1.56) reasonably correspond to the OCR values that were measured. In contrast, the OCR values estimated using the empirical correlation suggested by Powell and Uglow (1988) (i.e., OCR¼0.24(KD)1.32) are an underestimate relative to the measured values (Fig. 4). 5.3. Measured undrained shear strength by CKOUC test and FVT Fig. 5 shows the su profiles obtained from the CK0UC tests (suðCK 0 UCÞ ) and FVTs (su(FVT)). The su(CK0UC) increases as the depth increases (Fig. 5(a)), while the suðCK 0 UCÞ =σ 'v remains almost identical at around 0.31–0.37, irrespective of the depth, as shown in Eq. OCR 0

1

2

3

0

5. Experimental result and analysis

10

The profiles of DMT results are shown in Fig. 3. The DMT results measured in the two test holes are observed to be very similar. The ID values increase slightly as the depth increases, and these ranges are from 0.1 to 0.3, except at EL-12.75 m and EL-15.75 m, indicating that most of the soils in the tested site can be classified as clay, which is consistent with the results for the index properties as shown in Fig. 2. Additionally, the measured ID value suggests that a thin silt layer exists at a depth of EL-16 m. The values for KD range from 1.5 to 2.5, reflecting the fact that the clay layers at the test site are normally consolidated (Marchetti, 1980; Marchetti et al., 2001). The ED linearly increases with depth due to the I 0.0

0.6

Depth [m]

5.1. DMT results 20

30

Oedometer Marchetti (1980) Powell&Uglow (1988)

40

Fig. 4. Measured and estimated OCR values. The solid and hollow markers indicate each of the results obtained from hole nos. 1 and 2, respectively. E [kPa]

K

1.2

1.8

0 Hole 1 Hole 2

0

1 Hole 1 Hole 2

2

3

0

3000

6000

9000 Hole 1 Hole 2

Depth [m]

10

20

30

Fig. 3. Profiles of DMT results. The hollow and solid markers indicate each of the results obtained from hole nos. 1 and 2, respectively.

H. Choo et al. / Ocean Engineering 115 (2016) 39–47

su/σ 'v

su [kPa] 0

50

43

100

0.0 0

150

0

0.2

0.4

0.6 CK0UC

CK0UC

FVT

FVT

10

Depth [m]

Depth [m]

10

20

20

30

30

40

40

Fig. 5. Measured undrained shear strength from CK0UC tests and FVTs: (a) uncorrected su; (b) uncorrected su/σ0 v.

geostructures in the case of in-situ su measurements such as the FVT and based on the average su of three different shearing modes (i.e., triaxial compression, triaxial extension, and direct simple shear) with the correction factor for time to failure in the case of laboratory su measurements (Mesri, 1989; Mesri and Huvaj, 2007; Terzaghi et al., 1996). The average su of three different shearing modes can be expressed as follows under the assumption of that the undrained slip surface contains identical portions of triaxial compression, triaxial extension, and direct simple shear modes (Mesri, 1989; Terzaghi et al., 1996):

SHANSEP

1.2

s /σ '

0.9

0.6

su/σ'v=0.32(OCR)0.83 0.3

0.0

0

1

2

3

4

5

OCR Fig. 6. Dependence of su/σ0 v on OCR (result of the CK0UC SHANSEP tests).

(1) (Fig. 5(b)). The values measured for suðCK 0 UCÞ =σ 'v are consistent with the undrained strength ratio for the triaxial compression in Table 1 and that suggested by Jamiolkowski et al. (1985) (i.e., suðCK 0 UCÞ =σ 'v E0.32). Similar to the results of the CK0UC, the su(FVT) increases with an increase in depth (Fig. 5(a)). In contrast, the su (FVT) normalized with the vertical effective stress (σ 'v ) remains almost constant over the entire depth, ranging from 0.21–0.30 (Fig. 5(b)). The CK0UC SHANSEP tests were performed to develop a relationship between the su/σ0 v and OCR and to determine S and m in Eq. (2), as shown in Fig. 6. The values of S and m for the triaxial compression shearing mode of Busan New Port clay are determined to be 0.32 and 0.83, respectively, which are consistent with the results presented in previous studies (Jamiolkowski et al., 1985; Kamei and Iwasaki, 1995). Note that this S is very similar to the values for suðCK 0 UCÞ /σ0 v in Fig. 5(b), which reflects the fact that the clay layers that were tested had been normally consolidated. In addition, the measured m value is very comparable with that of previous studies (Ladd and Foott, 1974; Ladd et al., 1977). 5.4. Mobilized undrained shear strength by laboratory test and FVT Because of the discrepancy between the measured su by in-situ or laboratory tests and the mobilized su in the actual field, the undrained strength analysis (USA) is typically performed using the corrected su in order to consider the effects of strength anisotropy, time to failure (or strain rate), progressive yielding, and soil disturbance (Chung et al., 2012; Ladd, 1991; Terzaghi et al., 1996). Generally, the corrected su is determined based on the backcalculated factor of safety (FS) from the undrained failure of

suðavgÞ ¼

 1
suðCK 0 UCÞ þ suðDSSÞ þ suðCK 0 UEÞ 3

ð8Þ

where su(DSS) ¼undrained shear strength determined from direct simple shear test; and suðCK 0 UEÞ ¼ undrained shear strength determined from triaxial extension test. Previous studies showed that the average of suðCK 0 UCÞ and suðCK 0 UEÞ is almost the same as su(DSS) (i.e., suðCK 0 UCÞ þsuðCK 0 UEÞ E2  su(DSS)) (Chung et al., 2012; Lunne et al., 2003; Wang et al., 2014). Additionally, to evaluate the mobilized su based on laboratory tests, a factor of time to failure (μt) should be considered due to the discrepancy in the shearing time to failure between laboratory tests and actual field. Finally, due to the difference in stress–strain behaviors according to the shearing modes, a factor of progressive failure (λ) also needs to be considered (Chung et al., 2012; Ladd, 1991). Therefore, Eq. (8) can be:

μt U suðavgÞ ¼

λ U μt
2

suðCK 0 UCÞ þ suðCK 0 UEÞ



ð9Þ

where μtsu(avg) ¼mobilized undrained shear strength determined from laboratory test; μt was expressed as the function of plasticity index (PI) as shown in Fig. 7(a) (Terzaghi et al., 1996); and λ ¼progressive failure factor, where λ ¼1 when μtsu(avg) is determined based on the combined stress–strain relation of different shearing modes and λ E0.9 when μtsu(avg) is determined based on the arithmetic average of the peak shear strengths of different shearing modes (Chung et al., 2012; Ladd, 1991). In this study, the CK0UE (K0 consolidated-undrained triaxial extension) tests were not performed; therefore, the suðCK 0 UEÞ values were estimated according to the relation between suðCK 0 UEÞ and suðCK 0 UCÞ as (Ladd, 1991): suðCK 0 UEÞ ¼ 0:37 þ 0:0072 UPI suðCK 0 UCÞ

ð10Þ

Note that Chung et al. (2012) showed that the measured ratios of suðCK 0 UEÞ to suðCK 0 UCÞ for Busan clay can be effectively captured by Eq. (10). Additionally, λ in Eq. (9) was assumed to be 0.9 because

44

H. Choo et al. / Ocean Engineering 115 (2016) 39–47

the μtsu(avg) was determined based on the arithmetic average of suðCK 0 UCÞ and suðCK 0 UEÞ . Therefore, the corrected average su (or mobilized su) of tested Busan New Port clay was calculated using Eqs. (9) and (10) as shown in Fig. 8(a). Fig. 8(a) demonstrates that the μtsu(avg) increases from 15 kPa to 90 kPa with an increase in depth due to the increase in the applied effective stress. In contrast, the stress-normalized μtsu(avg) (i.e., μt suðavgÞ =σ 'v ) show almost constant values of around 0.22 (Fig. 8(b)). Although the su of clays can be directly estimated via the FVT, unexpected failures of earth structures have occurred as a result of

the use of an uncorrected undrained shear strength measured from the FVT (su(FVT)), indicating that uncorrected su(FVT) overestimates the FS of the structures. Note that similar to Eq. (9), the corrected su(FVT),which is the mobilized undrained shear strength based on the FVT, can be expressed as μsu(FVT), where μ is a correction factor. The first attempt to correct the measured su(FVT) was made by Bjerrum (1972). Bjerrum (1972) expressed μ as the function of PI. Later, Aas et al. (1986) expressed μ as the function of suðFV TÞ =σ 'v based on a number of undrained failure records as shown in Fig. 7(b). Because previous studies on Busan clay (Chung et al., 2010, 2007) showed that the μ suggested by Aas et al. (1986) efficiently represents the undrained shear strength of Busan Clay, the measured su(FVT) of tested Busan New Port clay was corrected using Fig. 7(b) in this study. Fig. 8(a) shows the corrected su(FVT) (i.e., μsu(FVT)) profile obtained from the FVT. The μsu(FVT) values increase from 13 kPa to 72 kPa as the depth increases; while, similarly to the results of the μtsu(avg), the stress-normalized μsu (FVT) (i.e., μsuðFVTÞ =σ 'v ) show almost constant values of around 0.22 (Fig. 8(b)), which is consistent with the undrained strength ratio for the field vane test in Table 1 and with the suggestions provided in previous studies (Azzouz et al., 1983; Mesri, 1975). It is interesting to note that previous studies (Mesri, 1989; Mesri and Huvaj, 2007) showed that the preconsolidation stressnormalized mobilized su values (μsu =σ 'p ) for inorganic soft clays determined from the FVT and from the laboratory tests with three different shearing modes show almost identical numbers as around 0.22. Consistent with these studies, the measured μtsu(avg)/ σ0 v and μsu(FVT)/σ0 v of tested Busan clay are around 0.22 (Fig. 6(b)), which may imply that the tested Busan clay is normally consolidated (σ0 p E σ0 v). Therefore, the use of μsu/σ0 v E0.22 is recommended for the undrained stability analysis of tested Busan New Port clay (Chung et al., 2006).

6. Correlation between the mobilized undrained shear strength and DMT results

Fig. 7. Undrained shear strength correction factors for: (a) su(avg); (b) su(FVT). Note that PI¼ plasticity index (liquid limit–plastic limit); NC¼ normally consolidated state; and OC ¼ overconsolidated state.

The μsu of tested soils can be estimated using Eq. (4) with an input of KD values; while, to estimate the μsu using Eq. (5), the Nc must first be determined. Vesic (1972) noted that the Nc tends to increase with an increase in the rigidity index (Ir ¼ G/su, where G is the stiffness of soils). Among the DMT indices the ID is known to best reflect the Ir of soils (Marchetti, 1997) because ID can be

Fig. 8. Mobilized su profiles for: (a) μsu; (b) μsu/σ0 v.

H. Choo et al. / Ocean Engineering 115 (2016) 39–47

expressed as: P1  P0 1 ED G U ¼  αU ID ¼ su P 0  u0 34:7 K D U σ 'v

ð11Þ

where α ¼fitting parameter. Therefore, the interrelation between the Nc and ID is analyzed in this study as shown in Fig. 9. Note the Nc values were calculated using Eq. (5) with the measured mobilized su. The values determined for Nc with μtsu(avg) or μsu(FVT) range from 7 to 12.5, and these are slightly greater than those suggested by Roque et al. (1988) because they calculated the Nc from the uncorrected su determined from the triaxial compression tests. Although some data points are scattered, the Nc in Fig. 9 linearly increase as the ID increases, regardless of the test method, as: N c ¼ 9:0  I D þ7:4

for

both

μt suðavgÞ and μsuðFVTÞ

ð12Þ

Eq. (12) is substituted into Eq. (5) to obtain the empirical correlation to estimate the mobilized su for Busan New Port clay using ID values as:

μsu ¼ μt suðavgÞ ¼ μsuðFV TÞ ¼

p1  σ h 9:0I D þ7:4

ð13Þ

Fig. 10 presents the measured and estimated profiles for μtsu (avg) and μsu(FVT) using the KD (Eq. (4)) and ID (Eq. (13)). Note that Marchetti (1980) developed Eq. (4) using S ¼0.22 and m¼ 0.8 in Eq. (2), and the determined S and m values for the tested clay are 0.22 and 0.83 (Figs. 6 and 8(b)). Therefore, the overall trends for all of the estimated mobilized su are in good agreement with the 25

20

Lab. FVT Noksan

Nc

15

10

5

0 0.0

0.1

0.2 ID

0.3

0.4

Fig. 9. Interrelation between Nc and ID. For comparison, data from related investigation with tests on Noksan clay in Korea are also included. Note that the trendline (Eq. (12)) was developed using all data points shown in above figure.

45

measured values, reflecting that both Eqs. (4) and (13) provide a suitable prediction for the mobilized su of Busan New Port clay. A quantitative validation is conducted for Eqs. (4) and (13) by calculating the mean absolute percentage error (MAPE). The MAPE, a measure of accuracy of the estimating formula, is defined as:  n   1X su ðmeasuredÞ  su ðpredictedÞ ð14Þ MAPE ¼  ni¼1 su ðmeasuredÞ where n ¼number of data points. The MAPE values of Eqs. (4) and (13) are 10.1% and 8.6%, respectively, reflecting that all su estimating formulas that are suggested in this study can predict the mobilized su of soils within a difference of 10%. In addition, the MAPE values appear to indicate that the accuracy of the ID (or Nc)based su estimating formula (Eq. (13)) is slightly better than that of the KD-based su estimating formula (Eq. (4)).

7. Application of the suggested empirical correlation to the improved ground Because the mobilized su estimating formula using KD values (Eq. (4)) has been well recognized by many researchers (Kamei and Iwasaki, 1995; Lacasse and Lunne, 1989; Marchetti, 1980; Powell and Uglow, 1989), the newly suggested empirical formula based on ID (Eq. (13)), which was developed using the results of the unimproved Busan New Port clay, is verified through the comparison between the measured μtsu(avg) of the improved Busan New Port clay and the estimated μtsu(avg) using Eq. (13). After the completion of the ground improvement, which was confirmed by monitoring of settlement data, a series of DMT and CK0UC tests were performed. The ground improvement with the vertical drains and preloading enhances the engineering properties of soils (e.g., increase in strength and decrease in compressibility) due to the consolidation of soils. Therefore, it can be observed in Fig. 11 that the undrained shear strength increases by around 200% after the ground improvement. Additionally, after the ground improvement, ID values increase reflecting an increase in rigidity of soils and KD values increase due to the increase in horizontal stress of soils (Marchetti et al., 2001). The calculated Nc values for the improved Busan clay are plotted as a function of ID in Fig. 12. The suggested linear relationship between the NC and ID of Eq. (12) is also included in Fig. 12. Both ID and Nc values of the improved clay are greater than those of the unimproved clay. However, Fig. 12 clearly demonstrate that the interrelation between the Nc and ID of the improved Busan clay can be also adequately captured by Eq. (12),

Fig. 10. Profiles of the estimated and measured mobilized su: (a) hole no. 1; (b) hole no. 2. The MAPE values were calculated using Eq. (14), and the reported MAPE values in above figures are the average of hole nos. 1 and 2.

46

H. Choo et al. / Ocean Engineering 115 (2016) 39–47

Fig. 11. Comparisons between the measured and estimated μtsu(avg) values for the improved Busan New Port clay: (a) hole no. 1; (b) hole no. 2. Note that closed markers¼ data for the improved clay; hollow markers¼data for the unimproved clay (Fig. 9). 25

20

15

Nc

Nc=7.4+9.0×ID

10

5

0 0.0

0.2

0.4

0.6

ID

3. The bearing factor (Nc) in the relation between su and p1 is observed to linearly increase with an increase in the material index (ID) with a relation of Nc ¼ 9.0  ID þ7.4; therefore, the μsu estimating method based on the ID, which is μsu ¼ (p1–σh)/ (9.0  ID þ 7.4), is newly suggested in this study. 4. The overall trends for all of the estimated μsu using both KD and ID are in good agreement with the measured μsu values, regardless of the test methods. The MAPE values indicate that the empirical correlation using the ID provides slightly higher accuracy than the previously suggested empirical correlation using the KD. 5. The application of the newly suggested μsu estimating formula using the ID yields the reliable estimates of μsu for the improved Busan clay, which may reflect the reliability of ID-based μsu estimating formula.

Fig. 12. Nc values as a function of ID for the improved Busan New Port clay. Note that closed markers¼ data for the improved clay; hollow markers¼ data for the unimproved clay (Fig. 9). The trendline included in above figure ¼ Eq. (12).

Acknowledgments which may reveal the validity of the newly suggested mobilized su estimating formula using ID. Finally, the measured μtsu(avg) and the estimated μtsu(avg) with Eq. (13) are compared in Fig. 11. It can be clearly observed that Eq. (13) yields a good agreement between the measured and the estimated μtsu(avg) with the calculated MAPE value ¼ 5.5%.

8. Conclusions A series of dilatometer tests (DMTs), field vane tests (FVTs), and comprehensive laboratory experiments were conducted in order to characterize the undrained shear strength (su) of Busan New Port clay using the results of DMTs. The results of this study demonstrate the following key observations: 1. The stress-normalized mobilized su(μsu/σ0 v) for both laboratory tests and FVTs are determined to be around 0.22, which is consistent with previous studies. The exponent m for the overconsolidation ratio (OCR), which is in the relation of su/ σ0 v ¼ S  OCRm, is determined to be 0.83 by using the SHANSEP technique. 2. Two different methods of estimating μsu are reviewed to make use of either the horizontal stress index (KD), which is μsu/ σ0 v ¼ 0.22  (0.5KD)1.25, or the bearing factor (Nc), which is μsu ¼(p1  σh)/Nc.

This research was supported by a Grant (15-RDRP-B076564-02) from Regional Development Research Program funded by the Ministry of Land, Infrastructure and Transport of Korean government.

References Aas, G., Lacasse, S., Lunne, T., Hoeg, K., 1986. Use of in situ tests for foundation design on clay. In: Proceedings of the Use of In Situ Tests in Geotechnical Engineering. ASCE, pp. 1–30. ASTM-D2573, 2008. Standard test method for field vane shear test in cohesive soil. Annual Book of ASTM Standards. ASTM-D6519, 2008. Standard practice for samling of soil using the hydraulically operated stationary piston sampler. Annual Book of ASTM Standards. Azzouz, A.S., Baligh, M.M., Ladd, C.C., 1983. Corrected field vane strength for embankment design. J. Geotech. Eng. 109 (5), 730–734. Baldi, G., Bellotti, R., Ghionna, V.N., Jamiolkowski, M., Lo Presti, D.F.C, 1989. Modulus of sands from CPTs and DMTs. In: Proceedings of the 12th International Conference on Soil Mechanics and Foundation Engineering, pp.165–170. Baligh, M.M., Scott, R.F., 1975. Quasi-static deep penetration in clays. J. Geotech. Eng. Div. 101 (11), 1119–1133. Bjerrum, L., 1972. Embankments on soft ground. In: Proceedings of the ASCE Specialty Conference on Performance of Earth and Earth-Supported Structures, pp. 1–54. Bjerrum, L., 1973. Problems of soil mechanics and construction on soft clays. In: Proceedings of the 8th International Conference on Soil Mechanics and Foundation Engineering, pp. 111–159.

H. Choo et al. / Ocean Engineering 115 (2016) 39–47 Byeon, W.Y., Lee, S.R., Kim, Y.S., 2006. Application of flat DMT and ANN to Korean soft clay deposits for reliable estimation of undrained shear strength. Int. J. Offshore Polar 16 (1), 73–80. Casagrande, A., 1936. The determination of the pre-consolidation load and its practical significance. In: Proceedings of the International Conference On Soil Mechanics and Foundation Engineering. Harvard University Cambridge, pp. 60– 64. Chandler, R.J., 1988. The in-situ measurement of the undrained shear strength of clays using the field vane. Vane Shear Strength Testing in Soils: Field and Laboratory Studies, pp. 13–44. Chung, S.G., Chung, J.G., Jang, W.Y., Lee, J.M., 2010. Correlations between CPT and FVT Results for Busan Clay. Mar. Georesour. Geotechnol. 28 (1), 49–63. Chung, S.G., Giao, P.H., Kim, G.J., Leroueil, S., 2002. Geotechnical properties of Pusan clays. Can. Geotech. J. 39 (5), 1050–1060. Chung, S.G., Hong, Y.P., Lee, J.M., Min, S.C., 2012. Evaluation of the undrained shear strength of Busan clay. KSCE J. Civ. Eng. 16 (5), 733–741. Chung, S.G., Kim, G.J., Kim, M.S., Ryu, C.K., 2007. Undrained shear strength from field vane test on Busan clay. Mar. Georesour. Geotechnol. 25 (3–4), 167–179. Chung, S.G., Kim, S.K., Kang, Y.J., Im, J.C., Prasad, K.N., 2006. Failure of a breakwater founded on a thick normally consolidated clay layer. Geotechnique 56 (6), 393–409. Chung, S.G., Ryu, C.K., Jo, K.Y., Huh, D.Y., 2005. Geological and geotechnical characteristics of marine clays at the Busan New Port. Mar. Georesour. Geotec. 23 (3), 235–251. Cruz, N., Devincenzi, M.J., Fonseca, A.V., 2006. DMT experience in Iberian transported soils. In: Proceedings of the 2nd International Conference on the Flat Dilatometer, pp. 198–204. Ebrahimian, B., Movahed, V., Pasha, A.Y., 2012. Evaluation of undrained shear strength of marine clay using cone penetration resistance at South Pars field in Iran. Ocean Eng. 54, 182–195. Hryciw, R.D., 1990. Small-strain-shear modulus of soil by dilatometer. J. Geotech. Eng. 116 (11), 1700–1716. Iwasaki, K., Tsuchiya, H., Sakai, Y., Yamamoto, Y., 1991. Applicability of the Marchetti dilatometer test to soft ground in Japan. In: Proceedings of the GEO-COAST 91, pp. 29–32. Jamiolkowski, M., Ghionna, V.N., Lancellota, R., Pasqualini, E, 1988. New correlations of penetration tests for design practice. In: Proceedings of the 1st International Symposium on Penetration Testing, pp. 263–296. Jamiolkowski, M., Ladd, C.C., Germaine, J.T., Lancellotta, R, 1985. New developments in field and laboratory testing of soils. In: Proceedings of the 11th International Conference on Soil Mechanics and Foundation Engineering, pp. 57–153. Kamei, T., Iwasaki, K., 1995. Evaluation of undrained shear strength of cohesive soils using a flat dilatometer. Soils and Found. 35 (2), 111–116. Kim, S.K., 2008. Characterization of deltaic deposits in the Nakdong River mouth, Busan. In: Proceedings of the 3rd International Conference on Site Characterization, pp. 75–88. Ku, T., Mayne, P.W., 2013. Evaluating the in situ lateral stress coefficient (K  0) of soils via paired shear wave velocity modes. J. Geotech. Geoenviron. Eng. 139 (5), 775–787. Lacasse, S., Lunne, T., 1989. Calibration of dilatometer correlations. Publ.-Nor. Geotek. Inst. 177, 1–10. Ladd, C.C., 1991. Stability evaluation during staged construction. J. Geotech. Eng.— Asce 117 (4), 540–615. Ladd, C.C., Foott, R., 1974. New design procedure for stability of soft clays. J. Geotech. Eng. Div. 100 (7), 763–786.

47

Ladd, C.C., Foott, R., Ishihara, K., Schlosser, F., Poulos, H.G., 1977. Stress-deformation and strength characteristics. In: Proceedings of the 9th International Conference on Soil Mechanics and Foundation Engineering, pp. 421–494. Lunne, T., Long, M., Forsberg, C., 2003. Characterisation and Engineering Properties of Onsøy Clay. vol. 1, pp. 395–428. Lunne, T., Robertson, P.K., Powell, J.J.M., 1997. Cone Penetration Testing in Geotechnical Practice. CRC Press. Lutenegger, A., 1988. Current status of the Marchetti dilatometer test. In: Proceedings of the 1st International Symposium on Penetration Testing, pp. 137– 155. Marchetti, S., 1980b. In situ tests by flat dilatometer. J. Geotech. Eng. Div.—ASCE 106 (3), 299–321. Marchetti, S., 1997. The Flat Dilatometer: Design Applications. Citeseer. Marchetti, S, 2006. Origin of the flat dilatometer. In: Proceedings of the 2nd International Conference on the Flat Dilatometer, pp. 2–5. Marchetti, S., Monaco, P., Totani, G., Calabrese, M., 2001. The flat dilatometer test in soil investigations. In: Proceedings of the International Conference on In Situ Measurement of Soil Properties. A Report by the ISSMGE Committee TC16, pp. 1–41. Mayerhof, G.G., 1956. Penetration tests and bearing capacity of cohesionless soils. J. Soil. Mech. Found. Div. 82 (1), 1–19. Mayne, P., 2007. NCHRP Synthesis 368: Cone Penetration Testing. Transportation Research Board Washington, DC. Mayne, P.W., 2008. Piezocone profiling of clays for maritime site investigations. In: Proceedings of the 11th Baltic Sea Geotechnical Conference, Gdansk, Poland, pp. 15–18. Mesri, G., 1975. New design procedure for stability of soft clays. J. Geotech. Geoenviron. Eng. 101 (Discussion). Mesri, G., 1989. A Reevaluation of Su(Mob) ¼ 0.22-Sigma'p Using Laboratory Shear Tests. Can. Geotech. J. 26 (1), 162–164. Mesri, G., Huvaj, N., 2007. Shear strength mobilized in undrained failure of soft clay and silt deposits. In: Proceedings of Advances in Measurement and Modeling of Soil Behaviour (GSP 173), pp. 1–22. Mitchell, J.K., Soga, K., 2005. Fundamentals of Soil Behavior. John Wiley and Sons. Monaco, P., Totani, G., Calabrese, M., 2007. DMT-predicted vs observed settlements: a review of the available experience. Stud. Geotech. Mech. 29 (1–2), 103–120. Powell, J., Uglow, I., 1989. The interpretation of the Marchetti dilatometer test in UK clays. In: Proceedings of the Geotechnology Conference. Penetration testing in the UK. Birmingham, United Kingdom. Powell, J.J.M., Uglow, I.M., 1988. The interpretation of the Marchetti dilatometer test in UK clays. In: Proceedings of Penetration Testing in the UK, ICE, pp. 269–273. Robertson, P.K., 2009. CPT-DMT Correlations. J. Geotech. Geoenviron. Eng. 135 (11), 1762–1771. Roque, R., Janbu, N., and Senneset, K., 1988. Basic interpretation procedures of flat dilatometer tests. In: Proceedings of 1st International Symposium on Penetration Testing. Orlando, Florida, pp. 577–587. Tanaka, H., Mishima, O., Tanaka, M., Park, S.Z., Jeong, G.H., Locat, J., 2001. Characterization of Yangsan clay, Pusan, Korea. Soils Found. 41 (2), 89–104. Terzaghi, K., Peck, R.B., Mesri, G., 1996. Soil Mechanics in Engineering Practice. John Wiley and Sons. Vesic, A.S., 1972. Expansion of cavities in infinite soil mass. J. Soil. Mech. Found. Div. 98 (SM3), 763–786. Wang, Q.S., Du, X.L., Gong, Q.M., 2014. Undrained shear strength of K-0 consolidated soft clays under triaxial and plane strain conditions. Int. J. Appl. Mech. 6, 3.