Paper Ground Improvement
This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA
Overview
Download & View Paper Ground Improvement as PDF for free.
More details
- Words: 5,249
- Pages: 10
Ground Improvement (2007) 11, No. 1, 1–10
FS
1
Assessment of ground improvement work using radioisotope cone penetrometers A. K. SHRIVASTAVA M2C UMR 6143 CNRS, Caen, France
Keywords : ?
Notation
atomic weight mass number of ith element speed of light in vacuum (299 792 458 m/s) void ratio sleeve friction rest mass of electron Avogadro number (6.022 3 1023 ) corrected cone penetration resistance excess pore pressure natural water content atomic number atomic number of ith element energy of incident photon energy of deflected photon angle between incident and final direction of the photon number of electrons per unit volume weight fraction of ith element soil density dry density moist density total density reaction cross-section
PR
A Ai c e fs m N qT u wn Z Zi h h9 Ł re ri rs rd rm rt a
French summary
OO
The effectiveness of radioisotope cone penetrometers (RI cones) under various soil conditions is now well established. This paper briefly describes the design of RI cones, and the advantages associated with their use. Two case histories are presented in which these cone penetrometers were utilised to investigate their effectiveness in ground improvement works. Both case histories involve land reclamation work. In one, reclamation was effected by hydraulically placing the fly ash obtained from a nearby coal-fired power plant, and in the other, the reclamation of land from sea was effected by pluviating sand by a conveyor belt system. It has been shown that RI cones can detect any small changes in the density profile, which can result in large savings in a project involving millions of cubic metres of sand.
Introduction
Geotechnical engineers are always looking for ways to improve upon existing methodologies and procedures to
(GI 4217) Paper received ?? ?????????? ????; last revised ?? ?????????? ????; accepted ?? ?????????? ????
enhance the efficiency of projects. The author has been involved in developing radioisotope cone penetrometers (RI cones) to measure basic soil properties, namely density (rd ) and natural water content (wn ), along with regular cone penetration tests with pore pressure measurement, also known as piezocone test (CPTU) data. The RI cones system is composed of three different types of cone penetrometer:
(a) a neutron moisture cone penetrometer (NM cone) (b) a nuclear density cone penetrometer (ND cone) (c) a dummy cone, equipped with a detector to measure the natural background radiation. The virgin ground is first penetrated using the NM cone to measure the natural water content (wn ) and soil strength parameters such as corrected penetration resistance (qT ), sleeve friction ( fs ) and excess pore pressure (u) generation in real time. The same hole is then used for the ND cone to measure the density (r)m and for the dummy cone to measure the density and the natural background radiation. Their effectiveness under various soil conditions has been established by Shibata et al. (1994), Shrivastava (1994) and Mimura et al. (1995). Geotechnical engineers intuitively associate strength or density (rd ) with ground improvement work; in recent years other parameters, such as soil modulus or stiffness, are coming to be considered more realistic parameters to indicate the in situ soil conditions; however, the author believes that density is one parameter that has been in use for a very long time and will be difficult to dispense with. In land reclamation work, where the material requirement
www.groundimprovement.com 1751-7621 (Online) 1365-781X (Print) # 2007 Thomas Telford Ltd Article number = P4217
A. Shrivastava
Using neutron and gamma rays to measure natural water content and density of soil Physical principle of neutron moderation by hydrogen
3
He þ n ! T þ p þ 0:764 MeV;
3
a ¼ 5400 b
(1)
PR
where He is helium, T is tritium, n is the neutron, p is the proton, MeV is Mega electron volts, and a is the reaction cross-section, measured in barns (b) (Knoll, 1979). (See Appendix.) For details of the the various aspects of neutron detection, readers are referred to any basic text book in nuclear physics. The fast neutron source is californium-252 (252 Cf), which is a spontaneous fission source of neutrons with a half-life of 2.65 years. The detector used is a 3 He-filled proportional 252
Cf fast neutron source
Porous ceramic filter
Pre-amplifier
Fig. 1. Schematic diagram of NM cone (all dimensions in mm)
48·6
35·6
3 He-filled proportional tube
1184
2
Physical principle of ª-ray interaction with matter It is well known that a photon can interact with atoms through scattering, losing some or none of its energy (Compton or elastic scattering respectively) or it can disappear in a single interaction (by the photoelectric effect or by pair production). Each process that contributes to the attenuation is a function of Z, the atomic number of the absorbing material. Therefore, considering the chemical composition of the soil, almost all interactions lead to the Compton process and depend on the electron density. Compton scattering occurs when a photon interacts with an electron and its incident energy is considerably greater than the binding energy of the electron. Applying the laws of conservation of momentum and energy, the following relation can be obtained.
OO
When fast neutrons are emitted from a source, they go through the process of slowing down, thermalisation and diffusion. This process of slowing down from the initial energy to the thermal diffusion energy is governed mainly by elastic collisions of the fast neutrons with the hydrogen nuclei present in the water, which are considered to be free. The presence of elements other than hydrogen in the soil is of minor importance in the slowing-down process. Nonetheless, the neutrons do undergo collisions with the nuclei of these other elements in the vicinity of the fast neutron source, and their migration is impeded. Therefore the mean distance between the source and the point at which the neutrons reach epithermal energy depends not only on the hydrogen content, but also on the composition of the watercontaining matrix. However, apart from the moisture content, for most soils it is mainly the dry bulk density that determines this distance, and thus the chemical composition of the soil is less important. As the neutrons do not react appreciably with electrons, they are always detected through the effects caused by their collisions with the nucleus. Over the years various reactions have been proposed for the detection of neutrons. The reaction used to detect neutrons in the construction of the NM cone is
tube. Fig. 1 shows a schematic diagram of the NM cone penetrometer. Neutrons are discharged from the cone penetrometer under a constant voltage of 950 V. They move into the soil matrix randomly, and, because of the collision process described, above they lose their energy. The 3 Hefilled proportional tube captures the slowed-down, or thermal, neutrons. The effective zone of measurement is about 30 cm in diameter. The steel casing of the penetrometer does not impede the movement of the neutrons. The NM cone is capable of measuring the following soil properties simultaneously in real time: the cone penetration resistance; the excess pore pressure (through the ceramic filter); the sleeve friction; and the natural water content of the soil, wn . The type of reaction described in equation (1) takes place in the 3 He-filled proportional tube). This is designed so that it detects only the slow neutrons. For details of the various aspects of cone penetration testing see Lunne et al. (1997).
FS
is in millions of cubic metres of soil, a precise measurement of density is required to calculate the volume of material that is required and, as the work progresses, the material being deposited, and also to assess whether the design criteria have been met or not. What follows here is a brief description of RI cones, after which two case histories are presented.
Cable leading to data collection system
h9 ¼
h 1 þ h=mc2 ð1 cos ŁÞ
(2)
where h is Planck’s constant (6.626 3 1034 J s or 4.135 3 1015 eV s); is the frequency; h9 is the energy of the deflected photon; h is the energy of the incident photon; c is the speed of light in vacuum (299 792 458 m/s), Ł is the angle between the incident and the final direction; and m is the rest mass of the electron (511 keV). For high-energy ª-rays, each electron participates in the scattering process. Consequently, the number of scattered photons is proportional to re , the number of electrons per unit volume. The mass density r is related to re through the equation Zi re ¼ Nrr i (3) Ai where N is Avogadro’s number, r i is the weight fraction of the ith element, Zi is the atomic number of the ith element, and A i is the mass number of the ith element. The value of Zi /A i is roughly equal to 0.5 for all the commonly occurring elements in the soil except for hydrogen (about 1.0). However, hydrogen does not exist in a free state in the soil, and is usually present in the form of water, whose Zi /A i ratio is 0.5521 (Shrivastava, 2005). The source of the ª-photons employed in the ND cone is caesium-137 (137 Cs), which has a half-life of 36.5 years. Fig. 2 shows a schematic diagram of the NaI(Tl) (thallium-doped sodium iodide) scintillator mounted on the photomultiplier tube used in the ND cone to detect incoming photons. A
Assessment of ground improvement work using radioisotope cone penetrometers NaI(T1)-mounted photomultiplier Pre-amplifier tube
35·6
FS
Cable leading to data collection system
Advantages of cone penetration testing
Fig. 2. Schematic diagram of ND cone (all dimensions in mm)
lead (Pb) shield is placed between the ª-source and the NaI(Tl) scintillator and photomultiplier to prevent direct measurement of ª-photons. The small blocks shown between the various components are mechanical connections to maintain the components in their respective positions. As described above in the NM cone section, the ª-photons are discharged randomly under a constant voltage of 950 V. These photons interact randomly with the soil matrix around the cone penetrometer, and the incoming ª-photons are detected by the NaI(Tl)-mounted photomultiplier tube. The incoming ª-photons hit the NaI(Tl) crystal, causing scintillation. This can be described as the flashes of light made by ionising radiation upon striking a crystal detector.
OO
1·7
PR
20·0
1·5
6·8
6·0
6·0
1·5
6·0
5·0
6·0
1·3 1·3
(a) It provides continuous subsurface data to aid rapid site characterisation. (b) It disturbs the subsurface conditions the least. (c) It requires no drilling fluid or mud. (d ) It is cheaper than drilling and sampling operations. (e) Real-time data are obtained.
5·2
5·2
Technology based on cone penetration testing (CPT) offers many advantages over sampling, and many researchers have already listed these; even so, authorities in many countries are still reluctant to use CPT-based technology. Some of the advantages related to the projects described in the following case histories are sunnarised below.
6·8
1184
These flashes of light must be converted into a corresponding electrical signal, as these signals are otherwise too faint to be detected. The photomultiplier tube is used to convert the light signals into a usuable current pulse without adding a large amount of random noise to the signal. For further details of both the NM and ND cones see Shrivastava (2005).
1·7
Lead shield
48·6
137 Cs gamma source
Zone A
6·0
Legend Sand compaction piles (SCP)
6·0
1·5
Sampling before SCP Sampling after SCP NM and ND cone testing before SCP NM and ND cone testing after SCP
1·5
Zone B
Fig. 3. Field investigation plan at fly-ash-enabled reclamation site (all dimensions in m)
3
A. Shrivastava
(a) No sample is obtained. (b) CPT-based instruments cannot be pushed into highly consolidated deposits, or deposits made up of boulders, cobbles or other large particles. (c) Data for natural water content (wn ) have to be corrected for the presence of chlorine ions (Cl ) in marine environments or suspected marine deposits.
5
0
15
20
25
2
Before SCP
4
6
After SCP
8
10
12
Case histories
14
OO
The first case study presented in this section is of a site where fly ash was used for reclamation work. As a major structure was planned for the site, ground improvement work was planned and carried out using sand compaction piles (SCPs) at various pitches to determine the optimum pitch for SCPs for further work. The second case study is of a site where reclamation work was affected by depositing sand from the nearby area. Density profiles at these sites were measured using RI cones as well as traditional sampling.
16
5
PR
qt: MPa 15
25
35
Before SCP
2
4
6
8
Depth: m
This area had been reclaimed by hydraulically placing fly ash, a by-product obtained from a nearby coal-based power plant. The fly ash in this area is underlain by a clay deposit. Approximately 5 m of fly ash was deposited, overlain by sandy silt, sand and gravelly sand, with the thickness of each layer varying between 1.5 and 4.5 m. This reclaimed area was chosen to receive another major structure. As the dynamic properties of fly ash are poorly understood, an extensive soil investigation was planned and carried out. Within this area, two sites (Zones A and B) were chosen as pilot study sites. The effectiveness of sand compaction piles (SCP) in clayey ground is well established, although not all the mechanisms involved are very well understood. In brief, to improve the clayey ground by the SCP method, a steel casing is driven into the ground to a predetermined depth, and then sand is pushed out from the casing into the ground and compacted by vibration to form compacted sand piles. The factors that affect the characteristics of the improved ground include the density (or strength) of the sand piles and their replacement area ratio, the external load conditions, and the strength of clay between the sand piles. To observe the effectiveness of SCP in improving the strength properties of fly ash ground, sand piles were installed in the soft ground at various spacings in both zones, to find the optimum spacing. Sand of very good grading (with less than 5% fines content) was used for easy ejection and densification. Figure 3 shows the set-up of the field investigation. In Zone A, the SCPs (shown as square blocks) were installed at various spacings (1.3, 1.5 and 1.7 m); in Zone B, they were all installed at a distance of 1.5 m from each other. The figure also shows the location of the RI cone testing
(a)
0
Reclamation work with fly ash
4
10
0
FS
There are some drawbacks with this technology, as with any other technology:
qt: MPa
Depth: m
( f ) Other sensors can be used with CPT, as in the cases discussed here. (g) For any large project, sampling is still essential. However, by using RI cones a quick initial assessment of a site can be made, and judicious selection of the sampling site can be made, based on the initial results.
After SCP
10
12
14
16
18 (b)
Fig. 4. qT profiles before and after sand compaction piling: (a) zone A; (b) zone B
relative to the SCPs. RI cone testing was performed both before and after installation of the SCPs. Undisturbed samplings were also carried out in both zones, both before and after installation of the SCPs, for laboratory testing, and their location relative to the SCPs is also shown in Fig. 3. The case study presented here is only for the 1.5 m
Assessment of ground improvement work using radioisotope cone penetrometers Legend 0 Gravelly sand
FS
Sand
Sandy silt
5
Depth: m
Fly ash
Mixed sand, clay and fly ash
10
Clay
OO
Intrusion of sand in flyash
15
Before
After
Fig. 5. Columnar sections obtained using SPT logs in zone A before and after SCP
PR
spacing, as this is the only formation common to both zones. Figure 4 shows a comparison of the profiles of corrected cone resistance, qT , for both zones, both before and after installation of the SCPs. In Zone A the RI cone testing was terminated at a depth of 12–13 m after installation of the SCPs, owing to mechanical trouble in the pushing system. A considerable increase in penetration resistance was observed after installation of the SCPs, and therefore it was decided to drill a borehole until a soft layer was reached, to facilitate the penetration testing. A general increase in the qT profile is seen for both the zones. However, in Zone A, a reduction in qT is observed just below 9 m. To investigate this trend further, a columnar section of this zone was prepared, based on a standard penetration testing (SPT) log, after installation of the SCPs: see Fig. 5. The reduction in strength can be attributed to the intrusion of freshly deposited sand from the SCP in the fly ash at this depth. No effort was made to study the time-dependent strength phenomenon, however; the formation of silicic acid gel films on the particle surfaces and the precipitation of silica or other material from solution or suspension as cement at particle contacts, as suggested by Mitchell (1984), should result in a subsequent increase in strength. Figure 6 shows the natural water content wn and density rt as functions of depth before installation of the SCPs. Also plotted in these diagrams are the data obtained through sampling for both Zone A and Zone B. Undisturbed sampling was carried out at all locations both before and after installation of the SCPs, so that the results obtained
from laboratory tests could be compared with those obtained with the RI cones. In spite of some paucity in the sampling data, very good agreement is seen. Figure 7 shows the void ratio e as a function of depth before and after installation of the SCPs. As noted above, the freshly deposited sand from the SCPs shows the lower strength,which is reflected in the void ratio plot of Zone A at the corresponding depth (just below 9 m) of the sand intrusion intohe fly ash deposit. Apart from this, a decrease in the value of e is observed in both zones. Figure 8 shows the density profiles of both zones. Also plotted are the density data obtained from the laboratory work. In both cases (i.e. density measurement using both RI cones and sampling), a general increase in the density after installation of SCP is seen in both zones. However, the scatter in the density profile using the sampling data is rather large, and does not correspond with the increase in strength as given by qT for a similar depth (Fig. 4).
Reclamation work with sand A very large reclamation project has been undertaken to alleviate the problem of port congestion around Tokyo. These port facilities will reach their capacity in near future, and any enlargement will lead to further congestion of already congested port facilities. It is planned to reclain a total of about 500 ha. At the reclamation site the average water depth is about 12 m, and the sea floor at the site is generally made up of alternating layers of sand and sand plus gravel, although the occasional presence of a loam layer 5
A. Shrivastava wn: % 60
wn: % 20 0
40
60
80
100
120
0
0
20
40
80
100
120
2
2
Laboratory
NM cone
FS
4
4
6
Depth: m
Depth: m
6
8 Laboratory
10
8
NM cone
10
12
12
14
14
16
18
16
ρt: t/m3 1·2
1·6
ρt: t/m3
2·0
2·4
2
1·2
1·6
2·0
2·4
0
OO
0
2
ND cone
ND cone
Depth: m
6
8 10
4
6
Depth: m
4
8
Laboratory
10
Laboratory
12
14
16
12
14
16
(a)
(b)
PR
Fig. 6. Natural water content (wn ) and density (rt ) profiles before installation of sand compaction piles: (a) zone A; (b) zone B
has also been detected. Reclamation from the sea was effected by depositing sand obtained from two nearby sites, designated A and B. A conveyor belt system was built to transport the sand from the source sites, and the sand was deposited in an arc-sweeping fashion so that it was evenly distributed. The sands from the two sites (A and B) were mixed during transport for deposition through the conveyor belt system. The density (rt ) profiles of the deposited sand were measured using RI cones and a modified Bishop sampler once the required depth of deposited sand had been attained. The Bishop sampler (diameter 60 mm) was modified by Hanzawa & Matsuda (1977) using a stationary piston and reducing the diameter to 50 mm. One of the main advantages of this modified Bishop sampler is that a sample of very low density can be obtained. However, the sample must be small enough to be retained by capillary force. Fig. 9 shows a schematic diagramme of the various measurement points. The geotechnical properties of source sites A and B, based on borehole data, are summarised in Tables 1 and 2 respectively. 6
The RI cones data for the source sites A and B are given in Fig. 10. RI-CPTU was carried out until the refusal was met at 12 m at site A and just below 7 m at site B, which is reflected in the profiles of qT against depth (Fig. 10). The corresponding changes in excess pore pressure u are also given in Fig. 10. The sudden changes in excess pore pressure are well reflected in the changes of qT , and similar changes can be observed for the profiles of density and water content. Sand was evenly spread from the centreline. Measurement was carried out on both sides of the centreline, referred to in Fig. 9 as line 1 and line 2. As not all the data are allowed to be published, only data obtained at measurement point MP1-4 on line 1 and measurement point MP2-7 on line 2 are presented.
Discussion For both source sites, values of NSPT are also plotted in Fig. 10, and are shown (as squares) referred to the lower axis. For site A, there is a good match between the qT values
Assessment of ground improvement work using radioisotope cone penetrometers ρt: t/m3
Void ratio, e 0·5 0
1·0
1·5
2·0
2·5
1·0 0
1·2
1·4
1·6
1·8
2·0
2·2
2·4
2
2
FS
Before SCP
4
4
Before SCP
Depth: m
6
Depth: m
6
8
After SCP
10
8
12
10 After SCP
Laboratory Before SCP After SCP
14
12
16
(a)
16
OO
14
1·2 0
(a)
Void ratio, e 0·5 0
1·5
2
2·5
3·5
ρt: t/m3 1·4
1·6
Depth: m
Depth: m
After SCP
8
2·2
After SCP
6
4
6
2·0
Before SCP
4
2
1·8
8
10
12
14
10
16
Before SCP
PR
12
14
(b)
Fig. 8. Density (rt ) profiles before and after installation of sand compaction piles; aboratory data also shown: (a) zone A; (b) zone B
16
18
(b)
Fig. 7. Void ratio (e) profiles before and after installation of sand compaction piles: (a) zone A; (b) zone B
measured with the RI cones and the NSPT . However, for site B, there is considerable scatter between the values of qT and NSPT . However, in spite of this scatter, there is an almost linear relationship for the corresponding values of qT and NSPT : the value of qT varies between four and six times the
value of NSPT . Such a relationship has also been proposed by other researchers, such as Robertson et al. (1984). Comparative data are available for one point along each line (MP1-4 and MP2-7) on the reclamation line, and are plotted in Fig. 11. Post-reclamation data are plotted from the datum line, which is the surface of the reclaimed land above seawater level. At MP1-4, both the pre- and post-reclamation data are plotted. Pre-reclamation RI cone testing was done from a depth of 5.5 m until the refusal was met at 9 m. The modified Bishop sampler was used at this site, but unfortunately most of the obtained samples were disturbed in order to obtain meaningful in situ density for comparison with the density data obtainerd from the RI cones. At this point (MP1-4), although the changes in density profile are very 7
A. Shrivastava
1-6
1-7
1-5
27·5 m
10 m
1-2 1-1
5m 5m
Current Previous reclamation line reclamation line
Conclusions
Centre line
15 m
15 m
15 m
2-6
2-5
2-4-2
2-7
1-3
8m
25 m
RI cones Settlement gauge Surface-type RI measurement
50·0 m
70·0 m
27·5 m
1-4
1-4*
Line 2
16 m 9 m 5 m
20 m
2-4-1 2-4 2-3 2-2 15 2-4-3 m 1 2-4-4 5 m
FS
the density profile obtained using the Bishop’s sampler, which is not reflected in the RI cones measurement. As the RI cones measurement is made in situ, it is believed that it is more representative of the actual site conditions than the data obtained using the modified Bishop’s sampler.
Line 1
2-1
2-5-5
It has been shown that RI cones can be effectively employed to monitor ground improvement work under various soil conditions. With their help, good-quality data related to the quality of ground improvement can be obtained. As the data are obtained in situ, they are free of human error. Furthermore, the results can be observed in real time, which eliminates time-consuming laboratory processes. Although the economics of the two projects has not been discussed, from the above discussion it is clear that consierable savings can be realised.
Appendix
The concept of the reaction cross-section is used in nuclear physics to express the likelihood of interaction occurring between particles. Thus it can characterise the probability that a particular nuclear reaction will take place, and this cross-section is expressed in units of area. A barn (symbol b) is a unit of area. It is not an SI unit, but is accepted for use with SI units. 1 b ¼ 1028 m2 . The story goes that, in the early days of the field, a particular crosssection turned out to be much bigger than expected. An experimenter exclaimed, ‘Why, that’s as big as a barn!’, and thus a unit name was born.
OO
Fig. 9. Schematic diagram of various measuring points at sand-enabled reclamation site
small, with the deployment of RI cones such minor changes are also detectable (Fig. 11). These small changes can also be detected in the water content profiles (Fig. 11). At MP2-7 along line 2, no measurement was done using RI cones prior to the reclamation. Density profiles obtained from the RI cones and the modified Bishop’s sampler are plotted in Fig. 11. Between 5 and 7 m there is some scatter in
Table 1. Physical properties of sand at source site A Borehole no. ! #
P14
P18
P24
P33
4.15–4.45 2.699 11.0 6 79 10 5 1.556 1.192
13.15–13.4 2.724 8.4 2 92 4 2 1.632 1.248
18.15–18.47 2.682 28.2 0 46 38 16 0.994 0.696
24.15–24.45 2.729 6.0 0 83 11 6 1.247 1.041
33.15–33.24 2.691 10.6 33 52 12 3 1.478 1.070
PR
Depth: m rs : t/m3 wn : % Pebbles (2–75 mm): % Sand (75 m–2 mm): % Silt (5–75 m): % Clay (, 5 m): % rdmax : t/m3 rdmin : t/m3 )
P13
Table 2. Physical properties of sand at source site B Borehole no.
Depth: m rs : t/m3 wn : % Pebbles (2–75 mm): % Sand (75 m–2 mm): % Silt (5–75 m): % Clay (, 5 m): % rdmax : t/m3 rdmin : t/m3 )
8
!
#
P6
P10
P15
P19
P27
5.15–5.45 2.68 49.4 4 46 35 15 1.189 0.939
9.15–9.45 2.728 14.2 5 89 4 2 1.831 1.398
14.15–14.45 2.714 13.6 10 77 10 3 1.796 1.320
18.15–18.45 2.703 12.6 30 62 5 3 1.728 1.383
26.15–26.29 2.733 10.5 50 45 3 2 1.747 1.399
Assessment of ground improvement work using radioisotope cone penetrometers
0
qT: MPa 10 15
5
0
20
0
0
2
4
u: kPa 6 8
10 12
0
3 ρ: t/m 0·5 1·0 1·5 2·0 2·5
0
0
2
4
4
6
6
8
8
10
10
12
12
10
10
14
12
12
0
qT: MPa 50 100
0
0
4
150
0
1
1
2
2
3
3
4
4
5
5
6
50
u: kPa 100 150
6
0
5
ρd
8
t/m3 0·5 1·0 1·5 2·0 2·5 ρ:
200
0
0
10 15 20 25 30 NSPT
0
0
20
Wn: % 40 60
80
1
1
ρm
2
2
3
ρt
4
3
4
ρd
5
5
7
6
6
8
7
7
6
7 8
0
60
ρt
8
(a)
50
4
ρm
OO
Depth: m
5 10 15 20 25 30 35 NSPT
Wn: % 30 40
2
2
6
14
20
FS
Depth: m
2
10
(b)
Fig. 10. RI-CPTU at: (a) sand source site A; (b) sand source site B
0
0
5
qT: MPa 10
Depth: m
2 Postreclamation
4
0
0
u: kPa 5
0
2
4
4
6
6
Prereclamation
10
0
2·5
qT: MPa 5·0 7·5
0
0
u: kPa 5
Bishop sampler
ρm
ρd
0
10
Wn: % 30 40
50
Postreclamation
2 4 6 8
Prereclamation
Prereclamation
10
(a)
10
0
0
3 ρ: t/m 0·5 1·0 1·5 2·0 2·5
Bishop sampler
0
10
20
Wn: % 30 40
50
2
2
2
2
4
4
4
6
6
6
6
8
8
8
8
10
10
10 (b)
10
Depth: m
20
ρt
10
10
10
3 ρ: t/m 0·5 1·0 1·5 2·0 2·5
0
8
8
8
0
10
2
PR
6
15
ρd
4
ρm ρt
Fig. 11. RI-CPTU at: (a) measurement points MP1-4; (b) measurement point MP2-7
9
A. Shrivastava
Discussion contributions on this paper should reach the editor by ??????
PR
OO
Hanzawa H. and Matsuda E. (1977) Density of alluvial sand deposits obtained from sand sampling. Proceedings of the 9th International Conference on Soil Mechanics and Foundation Engineering, Tokyo, Soil Sampling Speciality Session 2, pp. 7–14. Knoll G. F. (1979) Radiation Detection and Measurements. John Wiley & Sons, New York. Lunne T., Robertson P. K. and Powell J. J. M. (1997) Cone Penetration Testing in Geotechnical Practice Blackie Academic and Professional. Mimura M., Shrivastava A. K., Shibata T. and Nobuyama M. (1995) Performance of RI-cone penetrometers in sand deposits. Proceedings of the International Symposium on Cone Penetration Testing (CPT’95), Linko¨ping, Vol. 2, pp. 55–60. Mitchell J. K. (1984) Practical problems from surprising soil behaviour: 12th Terzaghi Lecture. Journal of Geotechnical Engineering, ASCE, 112, No. 3, 259–289. Robertson P. K., Campanella R. G. and Wightman A. (1984)
SPT–CPT correlations. Journal of Geotechnical Engineering, ASCE, 109, No. 11, 1449–1459. Shibata T., Mimura M. and Shrivastava A. K. (1994) Use of RIcone penetrometers in foundation engineering. Proceedings of the 13th International Conference on Soil Mechanics and Foundation Engineering, New Delhi, Vol. 1, pp. 147–150. Shrivastava A. K. (1994) Development of Neutron Moisture and Nuclear Density Cone Penetrometers and Their Application to Foundation Engineering. DEng thesis, Department of Civil Engineering, Kyoto University. Shrivastava A. K. (2005) Development and calibration of radioisotope cone penetrometers. Nuclear Instruments and Methods in Physics Research A, 539, 421–426.
FS
References
10
FS
1
Assessment of ground improvement work using radioisotope cone penetrometers A. K. SHRIVASTAVA M2C UMR 6143 CNRS, Caen, France
Keywords : ?
Notation
atomic weight mass number of ith element speed of light in vacuum (299 792 458 m/s) void ratio sleeve friction rest mass of electron Avogadro number (6.022 3 1023 ) corrected cone penetration resistance excess pore pressure natural water content atomic number atomic number of ith element energy of incident photon energy of deflected photon angle between incident and final direction of the photon number of electrons per unit volume weight fraction of ith element soil density dry density moist density total density reaction cross-section
PR
A Ai c e fs m N qT u wn Z Zi h h9 Ł re ri rs rd rm rt a
French summary
OO
The effectiveness of radioisotope cone penetrometers (RI cones) under various soil conditions is now well established. This paper briefly describes the design of RI cones, and the advantages associated with their use. Two case histories are presented in which these cone penetrometers were utilised to investigate their effectiveness in ground improvement works. Both case histories involve land reclamation work. In one, reclamation was effected by hydraulically placing the fly ash obtained from a nearby coal-fired power plant, and in the other, the reclamation of land from sea was effected by pluviating sand by a conveyor belt system. It has been shown that RI cones can detect any small changes in the density profile, which can result in large savings in a project involving millions of cubic metres of sand.
Introduction
Geotechnical engineers are always looking for ways to improve upon existing methodologies and procedures to
(GI 4217) Paper received ?? ?????????? ????; last revised ?? ?????????? ????; accepted ?? ?????????? ????
enhance the efficiency of projects. The author has been involved in developing radioisotope cone penetrometers (RI cones) to measure basic soil properties, namely density (rd ) and natural water content (wn ), along with regular cone penetration tests with pore pressure measurement, also known as piezocone test (CPTU) data. The RI cones system is composed of three different types of cone penetrometer:
(a) a neutron moisture cone penetrometer (NM cone) (b) a nuclear density cone penetrometer (ND cone) (c) a dummy cone, equipped with a detector to measure the natural background radiation. The virgin ground is first penetrated using the NM cone to measure the natural water content (wn ) and soil strength parameters such as corrected penetration resistance (qT ), sleeve friction ( fs ) and excess pore pressure (u) generation in real time. The same hole is then used for the ND cone to measure the density (r)m and for the dummy cone to measure the density and the natural background radiation. Their effectiveness under various soil conditions has been established by Shibata et al. (1994), Shrivastava (1994) and Mimura et al. (1995). Geotechnical engineers intuitively associate strength or density (rd ) with ground improvement work; in recent years other parameters, such as soil modulus or stiffness, are coming to be considered more realistic parameters to indicate the in situ soil conditions; however, the author believes that density is one parameter that has been in use for a very long time and will be difficult to dispense with. In land reclamation work, where the material requirement
www.groundimprovement.com 1751-7621 (Online) 1365-781X (Print) # 2007 Thomas Telford Ltd Article number = P4217
A. Shrivastava
Using neutron and gamma rays to measure natural water content and density of soil Physical principle of neutron moderation by hydrogen
3
He þ n ! T þ p þ 0:764 MeV;
3
a ¼ 5400 b
(1)
PR
where He is helium, T is tritium, n is the neutron, p is the proton, MeV is Mega electron volts, and a is the reaction cross-section, measured in barns (b) (Knoll, 1979). (See Appendix.) For details of the the various aspects of neutron detection, readers are referred to any basic text book in nuclear physics. The fast neutron source is californium-252 (252 Cf), which is a spontaneous fission source of neutrons with a half-life of 2.65 years. The detector used is a 3 He-filled proportional 252
Cf fast neutron source
Porous ceramic filter
Pre-amplifier
Fig. 1. Schematic diagram of NM cone (all dimensions in mm)
48·6
35·6
3 He-filled proportional tube
1184
2
Physical principle of ª-ray interaction with matter It is well known that a photon can interact with atoms through scattering, losing some or none of its energy (Compton or elastic scattering respectively) or it can disappear in a single interaction (by the photoelectric effect or by pair production). Each process that contributes to the attenuation is a function of Z, the atomic number of the absorbing material. Therefore, considering the chemical composition of the soil, almost all interactions lead to the Compton process and depend on the electron density. Compton scattering occurs when a photon interacts with an electron and its incident energy is considerably greater than the binding energy of the electron. Applying the laws of conservation of momentum and energy, the following relation can be obtained.
OO
When fast neutrons are emitted from a source, they go through the process of slowing down, thermalisation and diffusion. This process of slowing down from the initial energy to the thermal diffusion energy is governed mainly by elastic collisions of the fast neutrons with the hydrogen nuclei present in the water, which are considered to be free. The presence of elements other than hydrogen in the soil is of minor importance in the slowing-down process. Nonetheless, the neutrons do undergo collisions with the nuclei of these other elements in the vicinity of the fast neutron source, and their migration is impeded. Therefore the mean distance between the source and the point at which the neutrons reach epithermal energy depends not only on the hydrogen content, but also on the composition of the watercontaining matrix. However, apart from the moisture content, for most soils it is mainly the dry bulk density that determines this distance, and thus the chemical composition of the soil is less important. As the neutrons do not react appreciably with electrons, they are always detected through the effects caused by their collisions with the nucleus. Over the years various reactions have been proposed for the detection of neutrons. The reaction used to detect neutrons in the construction of the NM cone is
tube. Fig. 1 shows a schematic diagram of the NM cone penetrometer. Neutrons are discharged from the cone penetrometer under a constant voltage of 950 V. They move into the soil matrix randomly, and, because of the collision process described, above they lose their energy. The 3 Hefilled proportional tube captures the slowed-down, or thermal, neutrons. The effective zone of measurement is about 30 cm in diameter. The steel casing of the penetrometer does not impede the movement of the neutrons. The NM cone is capable of measuring the following soil properties simultaneously in real time: the cone penetration resistance; the excess pore pressure (through the ceramic filter); the sleeve friction; and the natural water content of the soil, wn . The type of reaction described in equation (1) takes place in the 3 He-filled proportional tube). This is designed so that it detects only the slow neutrons. For details of the various aspects of cone penetration testing see Lunne et al. (1997).
FS
is in millions of cubic metres of soil, a precise measurement of density is required to calculate the volume of material that is required and, as the work progresses, the material being deposited, and also to assess whether the design criteria have been met or not. What follows here is a brief description of RI cones, after which two case histories are presented.
Cable leading to data collection system
h9 ¼
h 1 þ h=mc2 ð1 cos ŁÞ
(2)
where h is Planck’s constant (6.626 3 1034 J s or 4.135 3 1015 eV s); is the frequency; h9 is the energy of the deflected photon; h is the energy of the incident photon; c is the speed of light in vacuum (299 792 458 m/s), Ł is the angle between the incident and the final direction; and m is the rest mass of the electron (511 keV). For high-energy ª-rays, each electron participates in the scattering process. Consequently, the number of scattered photons is proportional to re , the number of electrons per unit volume. The mass density r is related to re through the equation Zi re ¼ Nrr i (3) Ai where N is Avogadro’s number, r i is the weight fraction of the ith element, Zi is the atomic number of the ith element, and A i is the mass number of the ith element. The value of Zi /A i is roughly equal to 0.5 for all the commonly occurring elements in the soil except for hydrogen (about 1.0). However, hydrogen does not exist in a free state in the soil, and is usually present in the form of water, whose Zi /A i ratio is 0.5521 (Shrivastava, 2005). The source of the ª-photons employed in the ND cone is caesium-137 (137 Cs), which has a half-life of 36.5 years. Fig. 2 shows a schematic diagram of the NaI(Tl) (thallium-doped sodium iodide) scintillator mounted on the photomultiplier tube used in the ND cone to detect incoming photons. A
Assessment of ground improvement work using radioisotope cone penetrometers NaI(T1)-mounted photomultiplier Pre-amplifier tube
35·6
FS
Cable leading to data collection system
Advantages of cone penetration testing
Fig. 2. Schematic diagram of ND cone (all dimensions in mm)
lead (Pb) shield is placed between the ª-source and the NaI(Tl) scintillator and photomultiplier to prevent direct measurement of ª-photons. The small blocks shown between the various components are mechanical connections to maintain the components in their respective positions. As described above in the NM cone section, the ª-photons are discharged randomly under a constant voltage of 950 V. These photons interact randomly with the soil matrix around the cone penetrometer, and the incoming ª-photons are detected by the NaI(Tl)-mounted photomultiplier tube. The incoming ª-photons hit the NaI(Tl) crystal, causing scintillation. This can be described as the flashes of light made by ionising radiation upon striking a crystal detector.
OO
1·7
PR
20·0
1·5
6·8
6·0
6·0
1·5
6·0
5·0
6·0
1·3 1·3
(a) It provides continuous subsurface data to aid rapid site characterisation. (b) It disturbs the subsurface conditions the least. (c) It requires no drilling fluid or mud. (d ) It is cheaper than drilling and sampling operations. (e) Real-time data are obtained.
5·2
5·2
Technology based on cone penetration testing (CPT) offers many advantages over sampling, and many researchers have already listed these; even so, authorities in many countries are still reluctant to use CPT-based technology. Some of the advantages related to the projects described in the following case histories are sunnarised below.
6·8
1184
These flashes of light must be converted into a corresponding electrical signal, as these signals are otherwise too faint to be detected. The photomultiplier tube is used to convert the light signals into a usuable current pulse without adding a large amount of random noise to the signal. For further details of both the NM and ND cones see Shrivastava (2005).
1·7
Lead shield
48·6
137 Cs gamma source
Zone A
6·0
Legend Sand compaction piles (SCP)
6·0
1·5
Sampling before SCP Sampling after SCP NM and ND cone testing before SCP NM and ND cone testing after SCP
1·5
Zone B
Fig. 3. Field investigation plan at fly-ash-enabled reclamation site (all dimensions in m)
3
A. Shrivastava
(a) No sample is obtained. (b) CPT-based instruments cannot be pushed into highly consolidated deposits, or deposits made up of boulders, cobbles or other large particles. (c) Data for natural water content (wn ) have to be corrected for the presence of chlorine ions (Cl ) in marine environments or suspected marine deposits.
5
0
15
20
25
2
Before SCP
4
6
After SCP
8
10
12
Case histories
14
OO
The first case study presented in this section is of a site where fly ash was used for reclamation work. As a major structure was planned for the site, ground improvement work was planned and carried out using sand compaction piles (SCPs) at various pitches to determine the optimum pitch for SCPs for further work. The second case study is of a site where reclamation work was affected by depositing sand from the nearby area. Density profiles at these sites were measured using RI cones as well as traditional sampling.
16
5
PR
qt: MPa 15
25
35
Before SCP
2
4
6
8
Depth: m
This area had been reclaimed by hydraulically placing fly ash, a by-product obtained from a nearby coal-based power plant. The fly ash in this area is underlain by a clay deposit. Approximately 5 m of fly ash was deposited, overlain by sandy silt, sand and gravelly sand, with the thickness of each layer varying between 1.5 and 4.5 m. This reclaimed area was chosen to receive another major structure. As the dynamic properties of fly ash are poorly understood, an extensive soil investigation was planned and carried out. Within this area, two sites (Zones A and B) were chosen as pilot study sites. The effectiveness of sand compaction piles (SCP) in clayey ground is well established, although not all the mechanisms involved are very well understood. In brief, to improve the clayey ground by the SCP method, a steel casing is driven into the ground to a predetermined depth, and then sand is pushed out from the casing into the ground and compacted by vibration to form compacted sand piles. The factors that affect the characteristics of the improved ground include the density (or strength) of the sand piles and their replacement area ratio, the external load conditions, and the strength of clay between the sand piles. To observe the effectiveness of SCP in improving the strength properties of fly ash ground, sand piles were installed in the soft ground at various spacings in both zones, to find the optimum spacing. Sand of very good grading (with less than 5% fines content) was used for easy ejection and densification. Figure 3 shows the set-up of the field investigation. In Zone A, the SCPs (shown as square blocks) were installed at various spacings (1.3, 1.5 and 1.7 m); in Zone B, they were all installed at a distance of 1.5 m from each other. The figure also shows the location of the RI cone testing
(a)
0
Reclamation work with fly ash
4
10
0
FS
There are some drawbacks with this technology, as with any other technology:
qt: MPa
Depth: m
( f ) Other sensors can be used with CPT, as in the cases discussed here. (g) For any large project, sampling is still essential. However, by using RI cones a quick initial assessment of a site can be made, and judicious selection of the sampling site can be made, based on the initial results.
After SCP
10
12
14
16
18 (b)
Fig. 4. qT profiles before and after sand compaction piling: (a) zone A; (b) zone B
relative to the SCPs. RI cone testing was performed both before and after installation of the SCPs. Undisturbed samplings were also carried out in both zones, both before and after installation of the SCPs, for laboratory testing, and their location relative to the SCPs is also shown in Fig. 3. The case study presented here is only for the 1.5 m
Assessment of ground improvement work using radioisotope cone penetrometers Legend 0 Gravelly sand
FS
Sand
Sandy silt
5
Depth: m
Fly ash
Mixed sand, clay and fly ash
10
Clay
OO
Intrusion of sand in flyash
15
Before
After
Fig. 5. Columnar sections obtained using SPT logs in zone A before and after SCP
PR
spacing, as this is the only formation common to both zones. Figure 4 shows a comparison of the profiles of corrected cone resistance, qT , for both zones, both before and after installation of the SCPs. In Zone A the RI cone testing was terminated at a depth of 12–13 m after installation of the SCPs, owing to mechanical trouble in the pushing system. A considerable increase in penetration resistance was observed after installation of the SCPs, and therefore it was decided to drill a borehole until a soft layer was reached, to facilitate the penetration testing. A general increase in the qT profile is seen for both the zones. However, in Zone A, a reduction in qT is observed just below 9 m. To investigate this trend further, a columnar section of this zone was prepared, based on a standard penetration testing (SPT) log, after installation of the SCPs: see Fig. 5. The reduction in strength can be attributed to the intrusion of freshly deposited sand from the SCP in the fly ash at this depth. No effort was made to study the time-dependent strength phenomenon, however; the formation of silicic acid gel films on the particle surfaces and the precipitation of silica or other material from solution or suspension as cement at particle contacts, as suggested by Mitchell (1984), should result in a subsequent increase in strength. Figure 6 shows the natural water content wn and density rt as functions of depth before installation of the SCPs. Also plotted in these diagrams are the data obtained through sampling for both Zone A and Zone B. Undisturbed sampling was carried out at all locations both before and after installation of the SCPs, so that the results obtained
from laboratory tests could be compared with those obtained with the RI cones. In spite of some paucity in the sampling data, very good agreement is seen. Figure 7 shows the void ratio e as a function of depth before and after installation of the SCPs. As noted above, the freshly deposited sand from the SCPs shows the lower strength,which is reflected in the void ratio plot of Zone A at the corresponding depth (just below 9 m) of the sand intrusion intohe fly ash deposit. Apart from this, a decrease in the value of e is observed in both zones. Figure 8 shows the density profiles of both zones. Also plotted are the density data obtained from the laboratory work. In both cases (i.e. density measurement using both RI cones and sampling), a general increase in the density after installation of SCP is seen in both zones. However, the scatter in the density profile using the sampling data is rather large, and does not correspond with the increase in strength as given by qT for a similar depth (Fig. 4).
Reclamation work with sand A very large reclamation project has been undertaken to alleviate the problem of port congestion around Tokyo. These port facilities will reach their capacity in near future, and any enlargement will lead to further congestion of already congested port facilities. It is planned to reclain a total of about 500 ha. At the reclamation site the average water depth is about 12 m, and the sea floor at the site is generally made up of alternating layers of sand and sand plus gravel, although the occasional presence of a loam layer 5
A. Shrivastava wn: % 60
wn: % 20 0
40
60
80
100
120
0
0
20
40
80
100
120
2
2
Laboratory
NM cone
FS
4
4
6
Depth: m
Depth: m
6
8 Laboratory
10
8
NM cone
10
12
12
14
14
16
18
16
ρt: t/m3 1·2
1·6
ρt: t/m3
2·0
2·4
2
1·2
1·6
2·0
2·4
0
OO
0
2
ND cone
ND cone
Depth: m
6
8 10
4
6
Depth: m
4
8
Laboratory
10
Laboratory
12
14
16
12
14
16
(a)
(b)
PR
Fig. 6. Natural water content (wn ) and density (rt ) profiles before installation of sand compaction piles: (a) zone A; (b) zone B
has also been detected. Reclamation from the sea was effected by depositing sand obtained from two nearby sites, designated A and B. A conveyor belt system was built to transport the sand from the source sites, and the sand was deposited in an arc-sweeping fashion so that it was evenly distributed. The sands from the two sites (A and B) were mixed during transport for deposition through the conveyor belt system. The density (rt ) profiles of the deposited sand were measured using RI cones and a modified Bishop sampler once the required depth of deposited sand had been attained. The Bishop sampler (diameter 60 mm) was modified by Hanzawa & Matsuda (1977) using a stationary piston and reducing the diameter to 50 mm. One of the main advantages of this modified Bishop sampler is that a sample of very low density can be obtained. However, the sample must be small enough to be retained by capillary force. Fig. 9 shows a schematic diagramme of the various measurement points. The geotechnical properties of source sites A and B, based on borehole data, are summarised in Tables 1 and 2 respectively. 6
The RI cones data for the source sites A and B are given in Fig. 10. RI-CPTU was carried out until the refusal was met at 12 m at site A and just below 7 m at site B, which is reflected in the profiles of qT against depth (Fig. 10). The corresponding changes in excess pore pressure u are also given in Fig. 10. The sudden changes in excess pore pressure are well reflected in the changes of qT , and similar changes can be observed for the profiles of density and water content. Sand was evenly spread from the centreline. Measurement was carried out on both sides of the centreline, referred to in Fig. 9 as line 1 and line 2. As not all the data are allowed to be published, only data obtained at measurement point MP1-4 on line 1 and measurement point MP2-7 on line 2 are presented.
Discussion For both source sites, values of NSPT are also plotted in Fig. 10, and are shown (as squares) referred to the lower axis. For site A, there is a good match between the qT values
Assessment of ground improvement work using radioisotope cone penetrometers ρt: t/m3
Void ratio, e 0·5 0
1·0
1·5
2·0
2·5
1·0 0
1·2
1·4
1·6
1·8
2·0
2·2
2·4
2
2
FS
Before SCP
4
4
Before SCP
Depth: m
6
Depth: m
6
8
After SCP
10
8
12
10 After SCP
Laboratory Before SCP After SCP
14
12
16
(a)
16
OO
14
1·2 0
(a)
Void ratio, e 0·5 0
1·5
2
2·5
3·5
ρt: t/m3 1·4
1·6
Depth: m
Depth: m
After SCP
8
2·2
After SCP
6
4
6
2·0
Before SCP
4
2
1·8
8
10
12
14
10
16
Before SCP
PR
12
14
(b)
Fig. 8. Density (rt ) profiles before and after installation of sand compaction piles; aboratory data also shown: (a) zone A; (b) zone B
16
18
(b)
Fig. 7. Void ratio (e) profiles before and after installation of sand compaction piles: (a) zone A; (b) zone B
measured with the RI cones and the NSPT . However, for site B, there is considerable scatter between the values of qT and NSPT . However, in spite of this scatter, there is an almost linear relationship for the corresponding values of qT and NSPT : the value of qT varies between four and six times the
value of NSPT . Such a relationship has also been proposed by other researchers, such as Robertson et al. (1984). Comparative data are available for one point along each line (MP1-4 and MP2-7) on the reclamation line, and are plotted in Fig. 11. Post-reclamation data are plotted from the datum line, which is the surface of the reclaimed land above seawater level. At MP1-4, both the pre- and post-reclamation data are plotted. Pre-reclamation RI cone testing was done from a depth of 5.5 m until the refusal was met at 9 m. The modified Bishop sampler was used at this site, but unfortunately most of the obtained samples were disturbed in order to obtain meaningful in situ density for comparison with the density data obtainerd from the RI cones. At this point (MP1-4), although the changes in density profile are very 7
A. Shrivastava
1-6
1-7
1-5
27·5 m
10 m
1-2 1-1
5m 5m
Current Previous reclamation line reclamation line
Conclusions
Centre line
15 m
15 m
15 m
2-6
2-5
2-4-2
2-7
1-3
8m
25 m
RI cones Settlement gauge Surface-type RI measurement
50·0 m
70·0 m
27·5 m
1-4
1-4*
Line 2
16 m 9 m 5 m
20 m
2-4-1 2-4 2-3 2-2 15 2-4-3 m 1 2-4-4 5 m
FS
the density profile obtained using the Bishop’s sampler, which is not reflected in the RI cones measurement. As the RI cones measurement is made in situ, it is believed that it is more representative of the actual site conditions than the data obtained using the modified Bishop’s sampler.
Line 1
2-1
2-5-5
It has been shown that RI cones can be effectively employed to monitor ground improvement work under various soil conditions. With their help, good-quality data related to the quality of ground improvement can be obtained. As the data are obtained in situ, they are free of human error. Furthermore, the results can be observed in real time, which eliminates time-consuming laboratory processes. Although the economics of the two projects has not been discussed, from the above discussion it is clear that consierable savings can be realised.
Appendix
The concept of the reaction cross-section is used in nuclear physics to express the likelihood of interaction occurring between particles. Thus it can characterise the probability that a particular nuclear reaction will take place, and this cross-section is expressed in units of area. A barn (symbol b) is a unit of area. It is not an SI unit, but is accepted for use with SI units. 1 b ¼ 1028 m2 . The story goes that, in the early days of the field, a particular crosssection turned out to be much bigger than expected. An experimenter exclaimed, ‘Why, that’s as big as a barn!’, and thus a unit name was born.
OO
Fig. 9. Schematic diagram of various measuring points at sand-enabled reclamation site
small, with the deployment of RI cones such minor changes are also detectable (Fig. 11). These small changes can also be detected in the water content profiles (Fig. 11). At MP2-7 along line 2, no measurement was done using RI cones prior to the reclamation. Density profiles obtained from the RI cones and the modified Bishop’s sampler are plotted in Fig. 11. Between 5 and 7 m there is some scatter in
Table 1. Physical properties of sand at source site A Borehole no. ! #
P14
P18
P24
P33
4.15–4.45 2.699 11.0 6 79 10 5 1.556 1.192
13.15–13.4 2.724 8.4 2 92 4 2 1.632 1.248
18.15–18.47 2.682 28.2 0 46 38 16 0.994 0.696
24.15–24.45 2.729 6.0 0 83 11 6 1.247 1.041
33.15–33.24 2.691 10.6 33 52 12 3 1.478 1.070
PR
Depth: m rs : t/m3 wn : % Pebbles (2–75 mm): % Sand (75 m–2 mm): % Silt (5–75 m): % Clay (, 5 m): % rdmax : t/m3 rdmin : t/m3 )
P13
Table 2. Physical properties of sand at source site B Borehole no.
Depth: m rs : t/m3 wn : % Pebbles (2–75 mm): % Sand (75 m–2 mm): % Silt (5–75 m): % Clay (, 5 m): % rdmax : t/m3 rdmin : t/m3 )
8
!
#
P6
P10
P15
P19
P27
5.15–5.45 2.68 49.4 4 46 35 15 1.189 0.939
9.15–9.45 2.728 14.2 5 89 4 2 1.831 1.398
14.15–14.45 2.714 13.6 10 77 10 3 1.796 1.320
18.15–18.45 2.703 12.6 30 62 5 3 1.728 1.383
26.15–26.29 2.733 10.5 50 45 3 2 1.747 1.399
Assessment of ground improvement work using radioisotope cone penetrometers
0
qT: MPa 10 15
5
0
20
0
0
2
4
u: kPa 6 8
10 12
0
3 ρ: t/m 0·5 1·0 1·5 2·0 2·5
0
0
2
4
4
6
6
8
8
10
10
12
12
10
10
14
12
12
0
qT: MPa 50 100
0
0
4
150
0
1
1
2
2
3
3
4
4
5
5
6
50
u: kPa 100 150
6
0
5
ρd
8
t/m3 0·5 1·0 1·5 2·0 2·5 ρ:
200
0
0
10 15 20 25 30 NSPT
0
0
20
Wn: % 40 60
80
1
1
ρm
2
2
3
ρt
4
3
4
ρd
5
5
7
6
6
8
7
7
6
7 8
0
60
ρt
8
(a)
50
4
ρm
OO
Depth: m
5 10 15 20 25 30 35 NSPT
Wn: % 30 40
2
2
6
14
20
FS
Depth: m
2
10
(b)
Fig. 10. RI-CPTU at: (a) sand source site A; (b) sand source site B
0
0
5
qT: MPa 10
Depth: m
2 Postreclamation
4
0
0
u: kPa 5
0
2
4
4
6
6
Prereclamation
10
0
2·5
qT: MPa 5·0 7·5
0
0
u: kPa 5
Bishop sampler
ρm
ρd
0
10
Wn: % 30 40
50
Postreclamation
2 4 6 8
Prereclamation
Prereclamation
10
(a)
10
0
0
3 ρ: t/m 0·5 1·0 1·5 2·0 2·5
Bishop sampler
0
10
20
Wn: % 30 40
50
2
2
2
2
4
4
4
6
6
6
6
8
8
8
8
10
10
10 (b)
10
Depth: m
20
ρt
10
10
10
3 ρ: t/m 0·5 1·0 1·5 2·0 2·5
0
8
8
8
0
10
2
PR
6
15
ρd
4
ρm ρt
Fig. 11. RI-CPTU at: (a) measurement points MP1-4; (b) measurement point MP2-7
9
A. Shrivastava
Discussion contributions on this paper should reach the editor by ??????
PR
OO
Hanzawa H. and Matsuda E. (1977) Density of alluvial sand deposits obtained from sand sampling. Proceedings of the 9th International Conference on Soil Mechanics and Foundation Engineering, Tokyo, Soil Sampling Speciality Session 2, pp. 7–14. Knoll G. F. (1979) Radiation Detection and Measurements. John Wiley & Sons, New York. Lunne T., Robertson P. K. and Powell J. J. M. (1997) Cone Penetration Testing in Geotechnical Practice Blackie Academic and Professional. Mimura M., Shrivastava A. K., Shibata T. and Nobuyama M. (1995) Performance of RI-cone penetrometers in sand deposits. Proceedings of the International Symposium on Cone Penetration Testing (CPT’95), Linko¨ping, Vol. 2, pp. 55–60. Mitchell J. K. (1984) Practical problems from surprising soil behaviour: 12th Terzaghi Lecture. Journal of Geotechnical Engineering, ASCE, 112, No. 3, 259–289. Robertson P. K., Campanella R. G. and Wightman A. (1984)
SPT–CPT correlations. Journal of Geotechnical Engineering, ASCE, 109, No. 11, 1449–1459. Shibata T., Mimura M. and Shrivastava A. K. (1994) Use of RIcone penetrometers in foundation engineering. Proceedings of the 13th International Conference on Soil Mechanics and Foundation Engineering, New Delhi, Vol. 1, pp. 147–150. Shrivastava A. K. (1994) Development of Neutron Moisture and Nuclear Density Cone Penetrometers and Their Application to Foundation Engineering. DEng thesis, Department of Civil Engineering, Kyoto University. Shrivastava A. K. (2005) Development and calibration of radioisotope cone penetrometers. Nuclear Instruments and Methods in Physics Research A, 539, 421–426.
FS
References
10
Related Documents
Paper Ground Improvement
May 2020 9
