Long-term effects of geotechnical processes

Hokkaido University15 February 2005

Kenichi SogaReader in Geomechanics

Lunch Box (Bento)

• Appearance

• Price

• Expiration date

• Calorie

Geotechnical Construction

• Design

• Cost

• Sustainability/Maintenance

• Embodied Energy

Current

Source zone cleanup

Containment

Remediation

Plume restoration

Brownfield redevelopment

Maintenance & Reinforcement& Demolish

Design options• Construction – short period

– Design codes

• Maintenance – periodical, some uncertainty– Life-cycle cost analysis, system performance

• Long-term solution – rather slow. Large uncertainty. – Embodied energy evaluation, Life cycle environmental impact

analysis

Environmental solution/monitoring/maintenance

maintenance

Design options

construction Time

Future generation

Long-term consequences

Present stage

Resource management Protection Conservation

• Understand long term effects (geotechnical processes, embodied energy calculation)

• “Systematic” monitoring to deal with uncertainty in evaluating long-term effects

• Interpretation of data (direction and rate of change)

• Engineering ideas and solutions

Questions• Does the construction process affect the long-term

performance of geotechnical infrastructure?

• How much energy do we use to make a geotechnicalstructure? How much energy is used for its operation?

• What are the essential soil parameters and key indicators and how can we measure them?

• What long-term monitoring strategy should we adopt and at what scale?

• What are the emerging technologies that can be used for long-term monitoring?

Today’s Bento Menu

• Long-term effects of tunnel construction• Long-term effects of compensation

grouting• Long-term effects of piling• Innovation in monitoring

Embodied Energy of Channel Tunnel Rail Link Contract 220

Case Study: CTRL

• CTRL contract 220 drive.

Tunnel Map Stratford to London St. Pancreas

Stratford Box

Data Collection

Construction

TBM and GantriesSegment Loading

Twin Bore Tunnel

Construction

Segment Factory

Spoil Disposal at Stratford

Kawasaki TBM

Cross-section of TBM TBM in Harima Works, Japan

Breakdown of Tunnel Embodied Energy of CTRL 220 Construction

ConstructionEnergy

30%(266 GJ)

Material ProcessingEnergy

3%(30 TJ)

TransportationEnergy

2%(20 TJ)

Embodied Energyof Raw Materials

65%(584 TJ)

Equivalent CO2 Emission is 97.8 tonnes.

0.000059% of total UK greenhouse emissions in 2002

2.1% of all emissions associated with the UK construction industry in 1999.

Steel22%

Cement57%

Tail seal Grease5%

Aggregates3%

Sodium Silicate8%

Bentonite0%

Others5%

Aggregates*171%

Cement17%

Slag Cement1%

Bentonite1%

Others2%

Steel2%

Sodium Silicate2%

Pulverized fuel Ash4%

Total Mass

Embodied Energy

Comparison of the embodied energy of selected structures.

0.352,589,400[14]Large Office Building, 52 storeys, Australia, (130,000 m2)

1369,875[14]3 Storey Office,Australia (6500 m2)

1560,837[14]Large Shop, Australia (5918 m2)

3612,494[17]Single Storey Office, UK(584 m2)

3822,350[15]Standard Home, Toronto, Canada

7491,240[14]Residential attached two storey Unit, Australia (82 m2)

927942[16]Standard New US Home, Michigan, USA (227 m2)

1450625[13]Average House, Australia(125 m2)

2172414[12]Typical UK Masonry house (100m2)

1899,410[This project]CTRL, Stratford to London St. Pancreas Twin bore Tunnel

Embodied Energy(GJ)

Source ref.Building..

..EBuildingE

ETunnelE

1420

0

43 43

63

37

7

0

12

19

8

35

8.6 6.512 13

8.7

0

10

20

30

40

50

60

70

CTRL,Stratfordto London

St.PancreasTw in bore

Tunnel

Residentialtw o

storeyUnit,

Australia(82 m2)

SingleStoreyOffice,

UK, (584m2)

LargeShop,

Australia,(5918 m2)

3 StoreyOff ice,

Australia(6500 m2)

LargeOffice

Building,52

storeys,Australia,(130,000

m2)

Con

trib

utio

n to

Tot

al E

mbo

died

Eer

gy (%

%Contributionof Steel tototal E.E.

%Contributionof Cement tototal E.E.

%Contributionof DirectEnergy tototal E.E.

n

THE LONDON UNDERGROUND TUBE NETWORK

• Area served: 3240km2

• 45km N-S 72km E-W• line length: 392km• 35% (140km) in deep tunnels

• 80% (110km) cast iron• Deepest tunnel 67.4m bgl• Average tunnel depth 24.5m bgl• 2.5million passengers/ day

1889

1908

1926

Existing Tunnels• New construction interactions

– pile driving nearby– neighbouring tunnel construction

• Long-term survival: what is the unexpired life?– chemical environment– earth and water pressures create ground loading – affected by construction, consolidation, creep, ageing– loads on lining must change as groundwater changes

• Design of new works– what ground actions to assume in what design life?– what influence from new construction activities?

LUL Kennington

LUL Kennington

R rotary cored boreholeP cable percussion

borehole

tunnel

3m 3m 1.5m 5m 10m

R2

4m

4m

4m

R1

LC1 LC6LC5LC4LC3LC2

P1P6P5P4P3P2

PM1 PM3PM2 EP1

North

LC self-boring load cell pressuremeter testsEP self-boring expansion pressuremeter testsPM self-boring permeameter tests

EPSRC / LUL Project• London northern line at Kennington

– 75 year old tunnel– ground investigation, in situ measurements, cores

• Piezometric conditions– pressure profiles, in situ self-boring permeameter– degree of drainage into tunnel

• Comparison of ground near and far from tunnel– lateral pressure; self-boring load cells and

pressuremeter tests– very high quality cores; stiffness, and strength

• Predicted response to rising ground-water– FE based on in situ conditions and trends

LUL Kennington stratigraphy

0 5 10 15 20

-56

-50

-44

-38

-32

-26

-20

-14

-8

-2R1

Made GroundRiver Terrace Deposits

London Clay Formation

Silty fine SANDLambeth Group

MUDSTONE/LIMESTONE

R2

Silty SAND

Fissured CLAYFlint GRAVEL

Silty fine SAND with Gravel

Upper Chalk

Thanet Sand

(m)

(m)

0

5

10

15

20

25

30

35

40

45

50

55

0 50 100 150 200 250 300 350 400Pore water pressure u: kPa

Dep

th d

: m b

gl

P1 P2 P3 P4 P5 P6 R1 R2

RIVER TERRACE DEPOSITS

LONDON CLAY

LAMBETH GROUP

THANET SAND

CHALK

L: Linear regression of borehole piezometer data in London Clay

H1: Theoretical hydrostatic profileWT @ 4.5m

H2: Borehole piezometer data in underlying granular strata showing hydrostatic distribution with WT @ approximately 16m

Tunnel position

Pore waterpressures

(R1)P1 P2 P5P4P3 P6

(R2)0

20

25

30

15

10

50

45

40

35

5

-3.0 0 3.0 4.5 9.5 19.5

Distance from tunnel centre line: m

Dep

th d

: m b

gl

T1

T3T4T2

T5

18

19

20

21

22

23

24

100 110 120 130 140 150

Pore water pressure u: kPa

Dep

th d

: m b

gl

Tunnel piezometer data

Linear regression of borehole piezometer data (line L Figure 5)

LUL Kennington: load cell data

0

100

400

300

200

cell 1

cell 4

cell 5

cell 6

cell 3

cell 2

Data expressed in kPa

Permeability

0

5

10

15

20

25

30

1.E-12 1.E-11 1.E-10 1.E-09 1.E-08Coefficient of horizontal permeability kh: m/s

Dep

th d

: m b

gl

PM1: 1m from tunnel liningPM2: 2.5m from tunnel liningPM3: 7.5m from tunnel lining

linear regression of PM1 data(closest to the tunnel lining)

Top of London Clay

Base of London Clay

Tunnel position

Field Permeability Measurement

Picture-1: Self-boring pressuremeter

L

D

Max. depthdrilled

Bottom of cuttingshoe after pullingback

Water filled cavity

Lateralflow

Vertical flow

Lower end ofmembrane

Springline

Crown

Crown

Springline

Ward and Thomas (1965)

Long term effect of tunnel construction processes

Pore pressure

Horizontal displacementmeasurement points 4m and 16m

Instrument line

140m 120m

50m

40m

K0 = 1.5Water table 5 m below ground surfaceHydrostatic

St. James’s ParkDepth = 31mDiameter = 4.85m

Anisotropy

0

200

400

600

800

1000

1200

1400

1600

1800

0.001 0.01 0.1 1 10

Axial strain, ε a (%)

Norm

alis

ed s

ecan

t und

rain

ed Y

oung

's m

odul

us, E

u,se

c/p

o'

B-1

C-1

E-1 (Horizontally-cut)

Hight et al (1993)

•Similar pattern but different magnitude between 2D and 3D•Different pattern between isotropic and anisotropic soil because

of difference in stress path directions

0.00 10.00 20.00 30.00-340.00-320.00-300.00-280.00-260.00-240.00-220.00-200.00-180.00-160.00-140.00-120.00-100.00-80.00-60.00-40.00-20.000.00

-30.00 -20.00 -10.00 0.00-50.00

-40.00

-30.00

-20.00

-10.003D Isotropic 3D Anisotropic

Excess Pore Pressure Contour (kPa)

0

200

400

600

800

1000

1200

0 200 400 600 800

Mean effective stress (kPa)

( σa-σ

r) (k

Pa)

Vertically cut sample

Horizontally cut sample

0

200

400

600

800

1000

1200

0 200 400 600 800

Mean effective stress (kPa)

( σa-σ

r) (k

Pa)

Anisotropic model

Isotropic model

Undrained Effective stress paths

Excess pore pressure generations will be different

Long-term

0 days7.5 days26.4 days48 days590 days19 years

•Permeable tunnel lining leads to further settlement with time

•There will be slightly outward displacement around the tunnel

-8

-6

-4

-2

0

2

4

6

8

1 10 100 1000 10000 100000

Time after short-term (days)

Cha

nge

in d

iam

eter

(mm

) Horizontal diameter

Vertical diameter-8

-6

-4

-2

0

2

4

6

8

1 10 100 1000 10000 100000

Time after short-term (days)

Horizontal diameter

Vertical diameter

Anisotropic Isotropic

Performance of the tunnel lining

Performance of the tunnel lining

-700

-600

-500

-400

-300

-200

-100

01 10 100 1000 10000 100000

Time after short-term (days)

Hoo

p fo

rce

(kN

)

Springline

Crown

Anisotropic-600

-500

-400

-300

-200

-100

01 10 100 1000 10000 100000

Time after short-term (days)

Hoo

p fo

rce

(kN

)

Springline

Crown

Isotropic

Crown

Springline

Springline

Crown

Ward and Thomas (1965)

145 7373145 198

145

145145145

145121

7373

73145 73

130145

14573

1457373

73141

7329

14514173

7373

68

7373

73

145

145

73

73

14573

145

145

145145

73

73

14573

14573

73

145145 145 145 145 145145 145145 73 97 73145 26145 127 73 BRIDGE STREET

JLE RUNNING TUNNEL

Sub

way

1

Speaker's Green

LLS.

c:\gcg\papers\harding\EPIS23.SRF

Grout Injections : Episode 2316 June 1997 to 20 June 1997

Scale 1:500

0m 10m 20m

Shaft 4/6

N

-10

0

10

20

30

40 Tilt

of Clock

Tower (

mm/55m

)

Nov-94 Nov-95 Nov-96 Nov-97 Nov-98 Nov-99 Nov-2000

Grouting Episodes

9 13 16 22 2531 35 39

WB EB WB EB

Tunnel Progress:Pilots Enlargements

Start ofGrouting

Box ExcavationProgress [m]:

ConstructionControlRange

Optical Plumb

May 27 June 13 June 17

45.5 m 17.5 mCase A Case B Case C

12.8 m

Shield tunnel20.0 m

3.5 m 7.0

m

Monitoring PointsInjection Points

Date of the tunnel face location

123

123

123

FillClayey siltFine sand

Alluvial clay 2

Sand

3 m

0 m

-10 m

-20 m

-30 m

Settlement Monitoring PointsSPT-N0 10 20

1

2

3 Alluvial clay 1

Alluvial clay 3

Alluvial clay 4

-20

-40

0

5 6 7 9Month

Dis

plac

emen

t mm

Tail passage

Grout injection (1)

-20

-40

0

5 6 7 9Month

Dis

plac

emen

t mm

Tail passage

(1)

Case A Case C

-20

-40

0

5 6 7 9Month

Dis

plac

emen

t mm

Tail passage

Grout injection(1)

Case B

Case B Point

Case C Point

17.0m

40.0m57.0

m

35.0m

-30

-20

-10

0

10

20

0 50 100 150

Time (days)

Set

tlem

ent (

mm

) Field dataFE data

-40

-30

-20

-10

0

10

20

0 50 100 150

Time (days)

Set

tlem

ent (

mm

) Field dataFE data

Case B

Case C

Drainage Hole

AppliedLoad

Piston withDrainage Holes

Clay Specimen

Porous Vyon Plate

Porous Vyon Plate

Copper Injection Needle

Drainage Plate

Cylindrical Casing(Brass or Steel)

Figure 4.3 Section of Modified Consolidometeri 4 3 S i f M difi d C lid

Note: Not to Scale

Different diameter specimens – simulate simultaneous injection at different spacings

Effect of OCR on Compensation EfficiencyGrouting Efficiency vs TV(R=25mm, epoxy injection)

-20%

0%

20%

40%

60%

80%

100%

120%

1 10 100 1000 10000 100000

Time (sec)

Gro

ut E

ffici

ency OCR=1

OCR=1.5OCR=2OCR=5OCR=10

Epoxy injection, R = 25 mm

OCR=1

OCR=1.5

OCR=2

OCR=5

OCR=10

Time (sec)

Gro

ut e

ffici

ency

100%

80%

120%

60%

40%

20%

0%

-20%

Very low efficiency in NC clays confirmed by field trials

Effect of Grout Spacing

100%

80%

60%

40%

20%

0%

-20%

-40%

-60%1

Fina

l Gro

ut E

ffici

ency

Closer spacing

OCR = 2

OCR = 1.5

OCR = 1

OCR = 5

2 3 4 5 6 7 8n

Single injection in infinite space

Similar FE results

Pile set-up - Field Observation

1

1.5

2

2.5

3

3.5

4

4.5

0.1 1 10 100 1000 100001

1.5

2

2.5

3

3.5

4

4.5

0.1 1 10 100 1000 10000Time (days)Time (days)

Pile capacity increase Qt/QtoShaft capacity increase Qs/Qso

• Driven piles in sand exhibit shaft capacity increases of typically 60% per log cycle time, though highly variable (20-170% per log cycle).

• Greater set up appears to occur in denser sands, with saturated and dry sands and not above 3m.

• References: Chow et al (1998), Jardine & Standing (1999)

Observation of soil movement during pile driving

Model pile is jacked into calibration chamber. Digital images captured through reinforced window.This example: silica sand, base resistance = 14 - 17 MPa.

Deformation of soil element around a pile

Images converted to displacement measurements using Particle Image Velocimetry (PIV) and close-range photogrammetry.

Measured strain paths

-12 -10 -8 -6 -4 -2 0 2 4-30

-20

-10

0

10

20

30

40

50

Soil element position relative to pile tip, h/B

εΙ

εΙII

εhεv

εΙεΙII ε

γ/2

Plane of projection

εv

εh

Compression

Extension

-8

-6

-4

-2

0

2

4

6

8

10

-12 -10 -8 -6 -4 -2 0 2 4Soil element position relative to pile tip, h/B

εΙεΙII ε

γ/2 Plane of projectionεv

εh

εΙ

εΙII

εh

εv

Compression

Extension

h

B

offset

Near the pile (offset = 1.5B) Far from the pile (offset = 3.5B)

Tests on silica sand and carbonate sand showed similar behaviour, but carbonate sand produced large strains and lower stresses.

Triaxial tests• The stress path was determined based on the pile driving mechanism. Creep was performed at various stages of the stress path.

•Note that an approximation was made to convert the 3D stress conditions to triaxial condition.

•A variety of granular materials (two silica sands, glass balls, glass shards) were tested to examine the effect of shape and particle strength.

0 200 400 600 800-200

0

200

400

600

800

1000Triaxial Stress Path

Mean Effective Stress, p (kPa)

Extension creep stage(48 hours to 3 weeks)

CSL

Soil element with local strain devices

Triaxial test data

Time (s)

100 101 102 103 104 105

101 102 103 104 105

0.00

-0.05

0.05

-0.10

-0.15

-0.20

-0.25

-0.30

0.00

0.05

0.10

0.15

0.20

0.25

Dev

iato

ric S

train

(%)

Vol

umet

ric S

train

(%)

Glass Ball

Leighton Buzzard Uniform Silica Sand

Glass Ball

Dilation

Montpellier Natural Sand

Leighton Buzzard Uniform Silica Sand

Montpellier Natural Sand

Stress State at Creep : p' = 600 kPa and q = 800 kPaAll Samples Were Prepared With Relative Density of Approximately 70%.

Pile-set-up due to dilatant creep1. Triaxial creep test (stress constant)

εv

time

contraction

dilation

σr’

time

relaxation stage

set-up by mean pressure increase

σr’

2. Soil around a pile (kinematic restraint)σr’

Fibre optic strain sensing in piles

• Installed at BRE test site, Chattenden

• Monitor minipiles in group during load tests

• Attached to rebar• Optical fibre

pretensioned• Comparison with and

complementary to vibrating wire strain gauges

Depth 7m,Minipiles143mm dia,Central pile300mm dia,1.2m cube pile cap

Strain sensor technology comparison

DistributedDiscreteDiscreteMeasurement

Distributed measurementHigh strain accuracyEstablished techniqueFeatures

Analyser £50kFibre ~£.0.1-10/m

Analyser £20kGratings ~£50-300 each

Analyser ~£1-10kSensor £80-250

Cost

~10,000µε~10,000µε3,000µεMaximum strain

Brillouin gain spectrum frequency shift

Bragg reflection frequency shift

Change of resonance in wireDetected physical quantity

4-25 minutesReal timeReal timeMeasurement time

1 measurement every 20cm

Range 10km (max 20,000)

Typically 40 sensors1 per copper cableNo. of measurements

1m ~2-20mm (length of grating)

50-250mmLimit of spatial resolution

30µε0.1-10µε0.5-1µεStrain resolution

Optical fibreFibre Bragg gratingVibrating wireSensor

BOTDR (eg. Ando AQ8603)

FBG (eg. Ando FB200)Vibrating wire

Brillouin optical time domain reflectometry (BOTDR)

Distributed strain sensor – BOTDRAverage strain over 1m every 20cmRange ~5-10kmResolution 30µε (0.003%)Low cost sensors - optical fibre5 - 25 minutes per measurementCan link or switch between fibres

Diagram courtesy of ntt.co.jp/news

Field installation

Monitoring strain during load tests

-0.1

-0.08

-0.06

-0.04

-0.02

0

0.02

0.04

0.85 0.87 0.89 0.91 0.93 0.95 0.97 0.990

0.25

0.5

0.75

1

1.25

1.5

Change

0t

90t

0

0.2

0.4

0.6

0 0.5 1

Distance / km

Distance / kmStra

in /

%

Comparison with vibrating wire strain gauges (VWSG)

What can it be used?• Research

– Load tests – construction of load transfer functions– Clearer understanding of pile group behavior

• Monitoring lifespan of the structure (part of smart structures approach)– Post-earthquake diagnostic of a structure– Behavior due to long-term changes, (consolidation, capacity

increase with time etc.) • Adaptive Design

– Tunnel excavation below an existing structure.– Changes in building designation. – Additional floors/adding removing walls.

• Reuse of old piles– Increased pile capacity due to kinematic hardening

Piles are subjected to additionaloads due to bridging effect

l

P

z

t

t(z)

z

t

z

t

z

t

z

t

z

t

P=0

First Loading Unloading

z

t

z

t

z

t

P

Reloading

Bankside123 Installation

• Long term bearing pile monitoring– φ1.6m, 50m deep piles,

Strain sensors: BOTDR, vibrating wire, FBG

Monitoring of existing ThamesLink tunnel

• Tunnelling obliquely under Victorian masonry tunnel– Existing tunnel loaded by canal basin retaining wall– Directly below Midlands Main Line (MML)

MML above

Existing ThamesLinktunnel

Canal basin

New tunnel

King’s CrossSt Pancras Reproduced from Ordnance Survey of Northern Ireland mapping with the

permission of the Director and Chief Executive, © Crown Copyright.

3.6m

6.5m

8.5m

Innovation in Monitoring of Aging Infrastructure

• Fiber optics• Micro Electro Mechanical Sensors (MEMS)• Wireless Network Systems

MEMS

Connection Tab for Bottom Electrode

Connection Tab for Microbridge(Upper Electrode)

50 62.5 125 62.5 50

15 15

Bottom Driving Electrode

Microbeam

2550

253 1

60560

1532

.55

5

22.5 5 22.5

12.5

15

Substrate

Schematic drawing of a mircobeam (all units in micrometer)

Note:1. Electrostatic excitation and capacitive detection tecnhique is adopted.2. Microbeam thickness is 1 micrometer.3. Gap between microbeam and bottom electorde is at least 1.5 micrometer.

Micro-machined Reflectors

Warnecke et al., 2001From David Moore (CUED)

Lucent Technologies Lucent Technologies

Fibre optics MEMS

David Moore (CUED)Lucent Technologies

Axsun.com/David Moore (CUED) Axsun.com

Wireless Network Systems

Modelling (Better understanding of long-term soil behaviour

Monitoring (optimum numberof measurements)

Important soil properties?• Permeability• Undrained-partially

drained-drained behavior

• Stiffness anisotropy

Boundary conditions• Interface permeability

Key Performance Indicators?

• Ground or structure movements

• Crack developments

Future of Geotechnical Engineers?

• Need to perform life cycle assessment and embodied energy evaluation of geotechnical structures/construction

• Requires risk assessment (social, environmental, economic)

• Long-term prediction and monitoring• How can we incorporate this into design? • Do we have enough technical knowledge on

predicting (uncertain) long-term future events?

Thank you