Ground Modification for Liquefaction Mitigation

www.HaywardBaker.com

Ground Modification for Liquefaction

Mitigation

January 11, 2013Kansas City, MO

Tanner Blackburn, Ph.D., P.E.Assistant Chief Engineer

Presentation Summary Determining liquefaction susceptibility

NCEER guidelines

Mitigation methods Densification Reinforcement Drainage

Geotechnical Seismic Hazards

Liquefaction Bearing capacity Excessive settlement Lateral spreading

Slope Stability Cyclic shear strength Kinematic loading of slopes/earth

Liquefaction Function of:

Earthquake magnitude Distance from site Groundwater conditions (current or ‘high

water’?) Depth to ‘liquefiable’ strata (svo , rd)

Common Input Parameters: Peak Ground Acceleration (PGA) Magnitude (M)

Liquefaction National Center for Earthquake Engineering

Research (NCEER) Summary Report (1997 Meeting, published in JGGE, 2001).

Seed and Idriss (1971):

Normalized by vertical effective stress:

dvocyclic rga s max65.0

dvo

voeqeq r

ga

CSR '65.0ss

Liquefaction Resistance to liquefaction

Referred to as Cyclic Resistance Ratio (CRR) or CSRfield

Function of: Geologic history (deposit type, age, OCR) Soil structure (relative density, clay content) Groundwater conditions

Factor of Safety = CRR/CSR

Liquefaction Evaluation of CRR (NCEER, 1997):

SPT blow count (N) Corrected blow count Need fines content Corrected clean sand blow count – N1(60)CS

CPT tip resistance (qc) and sleeve friction (fs)

Shear wave velocity (Vs) Corrections for magnitude (M)

Scaling factor (MSF) – apply to F.S.

Liquefaction – SPT Analysis

Liquefaction – CPT AnalysisTo address FC:

(qc1N)cs instead of qc1N

(qc1N)cs = Kc*qc1N

Kc = f(qc, fs, svo, s’vo)

This eliminates need for sampling to determine FC.

Liquefaction – Shear Wave

Liquefaction - MSF

Example Loose Sand

(N1)60 at 15’ depth = 10 Fines Content < 5% (SW/SP) Water table during earthquake @ 5’ depth

Soil Parameters: svo’=1176 psf svo= 1800 psf rd = 0.97 PGA=0.15g M=5.8

Example (cont’d)

CSR = (0.65)(0.15)(1800/1176)(0.97)

CSR = 0.15 Using NCEER figure for (N1)60=

10: CRR=0.11 MSF ≈2 FS = MSF*(CRR/CSR) =

2*(0.11/0.15) = 1.47 Note the influence of MSF!

dvo

voeqeq r

ga

CSR '65.0ss

Liquefaction - FS0 0 .5 1

C S R an d C R R

60

50

40

30

20

10

0

Dep

th [f

t]

C S RC R R

0 40 80 12 0 1 60 200q t [ts f]

60

50

40

30

20

10

0D

epth

[ft]

C P T -9

0 0 .4 0 .8 1 .2 1 .6 2R f [% ]

60

50

40

30

20

10

0

Dep

th [f

t]

0 0 .5 1 1 .5 2F a c to r o f S afe ty

60

50

40

30

20

10

0

Dep

th [f

t]

Liquefaction – Cohesive Materials

Strength loss – not technically liquefaction ‘Seismic softening’

‘Chinese’ Criteria (Seed et al. 1983) Function of wc, LL, clay content Not well accepted anymore...

Bray and Sancio (2006) No defined criteria, but good overview.

Boulanger and Idriss (2006, 2007) Chris Baxter at URI - Silts

Liquefaction – Lateral Spreading

Lateral spreading can occur in gradual slopes (<2°)

Must design for static and dynamic driving forces with residual undrained shear strengths Even for cohesionless materials

Liquefaction-induced Settlement

Tokimatsu and Seed, 1987

Ishihara and Yoshimine, 1992

Zhang et al., 2002

Liquefaction Mitigation Increase strength ( CRR)

Ground improvement (densification or grouting)

Decrease driving stress ( CSR) Shear reinforcement with ‘stiffer’ elements within

soil mass

Decrease excess pore pressure quickly Reduce drainage path distance with tightly

spaced drains

Mitigation - Densification Increase cyclic shear strength (CRR) by

increasing relative density of cohesionless materials

Advantages: Field Verifiable!

Conduct field testing before and after treatment Employed for over 50 years, through several large magnitude

earthquakes. Several peer-reviewed documents describing the methods,

efficiency, and mechanics of densification. Approved by CA Office of Statewide Health Planning and

Development (OSHPD) for hospital and school construction.

Mitigation - Densification Methods:

Dynamic compaction Vibro-compaction Vibro-replacement Blast densification Compaction grouting

0 0 .5 1 1 .5C S R an d C R R

50

40

30

20

10

0

P o s t T rea tm en tP re -T rea tm en t

0 10 2 0q t [M P a]

1 4

1 2

1 0

8

6

4

2

0

Dep

th [

m]

0 1 2F ac to r o f S a fe ty

1 4

1 2

1 0

8

6

4

2

0

1 4

1 2

1 0

8

6

4

2

00 2 0 4 0S e ism ic S e ttlem en t [m m ]

T rea tm en t D ep th

Loose sand zone

Hospital site Vibro-

replacementto 45 ft.

Liquefaction Mitigation-Densification

Liquefaction Mitigation-Densification Sandy site Compaction

grouting for liquefaction mitigation

Urban site, no vibrations

0 0 .5 1C S R a n d C R R

60

50

40

30

20

10

0

Dep

th [

ft]

C S RC R R P reC R R P os t

0 4 0 8 0 1 20 1 6 0 2 00q t [ ts f]

6 0

5 0

4 0

3 0

2 0

1 0

0

Dep

th [

ft]

C P T -9

0 0 .4 0 .8 1 .2 1 .6 2R f [% ]

60

50

40

30

20

10

0

Dep

th [

ft]

0 0 .5 1 1 .5 2F ac to r o f S afe ty

6 0

5 0

4 0

3 0

2 0

1 0

0

Dep

th [

ft]

Liquefaction Mitigation Increase strength ( CRR)

Ground improvement (densification or grouting)

Decrease driving stress ( CSR) Shear reinforcement with ‘stiffer’ elements within

soil mass

Decrease excess pore pressure quickly Reduce drainage path distance with tightly

spaced drains

Mitigation - Reinforcement Reduce cyclic shear stress

applied to liquefiable soil by installing ‘stiffer’ elements within soil matrix that attract stress.

Can be used in non-densifiable soils (silts, silty sands).

Large magnitude EQs Not verifiable

Post-installation CPT or SPT results will not differ from pre-installation.

Vertical load testing of elements is not applicable.

soil soilinc

GI for Large Earthquakes Large magnitude

earthquakes:· PGA ~0.3-1.0g· M >7

Typical CSR values ~ 0.3-0.6

High liquefaction potential for all soils N<30· Densification has

limited application

ReinforcementOriginal Design Methodology

Shear stress reduction factor (KG) (Baez and Martin, 1993):

GINC=Inclusion shear modulusGSoil=Soil shear modulusARR=Ainclusion/Atotal

Strain compatibility and force equilibrium

Assumes linear elastic soil and INC behavior

CSRapplied to soil = KG * CSRearthquake

11

1

Soil

INCG

GGARR

K

Mitigation - Reinforcement 10% Area

Replacement

GINC/GSOIL=5

KG=0.7

11

1

Soil

INCG

GGARR

K

0 0 .5C S R a n d C R R

60

50

40

30

20

10

0

Dep

th [f

t]C S R P reC S R P o s tC R R P re

0 4 0 8 0 1 2 0 16 0 20 0q t [ ts f]

60

50

40

30

20

10

0

Dep

th [f

t]

C P T -9

0 0 .4 0 .8 1 .2 1 .6 2R f [% ]

6 0

5 0

4 0

3 0

2 0

1 0

0

Dep

th [f

t]

0 0 .5 1 1 .5 2F ac to r o f S a fe ty

60

50

40

30

20

10

0

Dep

th [f

t]

Reinforcement Methods:

· Deep soil mixing· Stone Columns

(aggregate piers)– New research

indicates this reinforcement effect is limited

· Jet Grouting

Mitigation - Reinforcement Requires engineering judgment regarding

input parameters Is there a limit to the ‘inclusion’ stiffness? What is the deformation mechanism (bending or shear)? Is there a maximum spacing that should be used? If the soil liquefies around a stone column, what is the

strength of the stone column?

Few peer-reviewed publications or references regarding use and efficiency

Vendor/contractor ‘white-papers’ do not qualify as design standards or peer-reviewed methods

State-of-the-practice is developing

Liquefaction Mitigation-Reinforcement Example of required judgment:

Say we need KG=0.8, what ARR do we need?

Stone columns? Typical GSC/Gsoil ~ 5 (Baez/Martin,

Mitchell, FHWA) ARR = 6% (11’ grid spacing-36”

columns)

11

1

Soil

INCG

GGARR

K

Liquefaction Mitigation-Reinforcement Example of required judgment:

Say we need KG=0.8, what ARR do we need?

Piles? Typical GSteel/Gsoil ~ 2500 W14x120 – A=0.23 ft2 ARR = 0.01% 50’ Spacing!!

11

1

Soil

INCG

GGARR

K

Current research by Boulanger,Elgamal, et al.

34

Spatial distribution Rrd

35

Reinforcement – Panels and Grids

Figure : Basic Treatment Patterns (Bruce 2003)

Linear Elastic Soil Profile DSM Half Unit Cell

Linear Elastic FE DSM ModelBoulanger, Elgamal, et al.

Shear reduction - panels

Ratio of shear stress reduction coefficients; (a) Gr = 13.5, (b) Gr = 50

Conclusion – Soilcrete Grid per Boulanger, Elgamal et. al

DSM grids affect both:· seismic site response (e.g., amax)

· seismic shear stress distributions (e.g. spatially averaged Rrd)

DSM grids on seismic site response can be significant and may require site-specific FEM analyses

The reduction in seismic shear stresses by reinforcement can be significantly over-estimated by current design methods that assume shear strain compatibility.

A modified equation is proposed for estimating seismic shear stress reduction effects. The modified equations account for non-compatible shear strains and flexure in some wall panels.

The top 2m-3m of DSM wall could potentially be the critical wall section in term of tension development.

Thanks to Masaki Kitazume, Tokyo Institute of TechnologyProvided images to HBI.

Thanks to Masaki Kitazume, Tokyo Institute of TechnologyProvided images to HBI.

Thanks to Masaki Kitazume, Tokyo Institute of TechnologyProvided images to HBI.

Brunswick Nuclear PlantSouthport, NC

Batch Plant

Intake Canal

N

Spoil Deposit

Ventura Cancer Center, CA

Liquefaction Mitigation Increase strength ( CRR)

Ground improvement (densification or grouting)

Decrease driving stress ( CSR) Shear reinforcement with ‘stiffer’ elements within

soil mass

Decrease excess pore pressure quickly Reduce drainage path distance with tightly

spaced drains

Mitigation - Drainage Limit excess pore pressure increase and duration of

increased pore pressure during cyclic shearing by providing short drainage paths in cohesionless materials.

Not verifiable with in situ testing Limited peer-reviewed publications or design standards.

Methods: EQ Drains – perforated pipe installed on tight grid Stone columns – additional feature, but not relied on for

design Permeability of stone column material Contamination with outside material.

EQ Drain Theory Reduce the excess pore pressure

accumulation during earthquake

0 5 1 0 1 5 2 0 2 5S h e ar s tre ss c y c le s

0

0 .2

0 .4

0 .6

0 .8

1

Pore

pre

ssur

e ra

tio

0 5 1 0 1 5 2 0 2 5S h e ar s tre ss c y c les

0

0 .2

0 .4

0 .6

0 .8

1

Pore

pre

ssur

e ra

tio

EQ Drain Details Typically 75-150 mm diameter Slotted PVC pipe with filter fabric Typical spacing 1-2 m triangular Installed with large steel probe with wings (densification also

intended)

EQ Drain Installation

EQ Drain Design Concept Based on radial dissipation theory (just like vertical consolidation, but

radial geometry)

tu

tum

ru

rk

rg

vw

h 1

tu

tu

ru

rrc gh

1wv

hh m

kc

2sin

2tan

testundrainedin on liquefacti causing cycles stress uniform ofNumber 7.0~

arcsin2

2

'

21

'

u

u

dl

eqog

l

lo

g

d

eq

gg

r

r

tNN

Nu

N

NNu

tN

tN

tN

Nu

tu

s

s

DeAlba et al., 1975

Assume periodic wave form

• Change in PP per cycle depends on PP of previous cycle

• NL based on CSR of soil, SPT, Fines

• Neq, td are functions of earthquake, but there are correlations to magnitude

Derivations Factor of safety is inverse of Ru Settlement

'

'

'

'

vouNewvlayerlayer

vou

v

vNewvlayerlayer

RmT

Ru

u

mT

s

s

s

s

EQ Drain Design Graphical solutions to diff equation (JGS):

· Address drain size, well resistance· Provides Ru, but no settlement calculations

FEQDrain – Finite Element software program· Provides Ru and settlement calculations

Both methods need the following:· Soil permeability, kh

· Soil compressibility, mv,· Earthquake duration, td

· Number of earthquake cycles, Neq

· Drain spacing (trial values)

EQ Drain with Stone Column Installation

Stone Column Installation with EQ Drains

Liquefaction Mitigation Increase strength ( CRR)

Ground improvement (densification or grouting)

Decrease driving stress ( CSR) Shear reinforcement with ‘stiffer’ elements within

soil mass

Decrease excess pore pressure quickly Reduce drainage path distance with tightly

spaced drains

Questions