BSM Stabilisation (Mix) Design

Kim Jenkins

SARF Course on BSMs

KZN DoT13 and 14 August 2015

tellenbosch

Life is full of surprises…

Try to keep them to a minimum

Primary Design Objectives

• Load spreading

– Resilient Modulus (Mr)

• Rut resistance

– Shear Strength

• Flexibility

– Displacement at Ultimate Strength

• Durability

– Moisture resistance

Early South African pavements

CEMENT STABILISED BASE

40mm HMA SURFACING

Selected subgrade layer

REFLECTIVE CRACKING

Key pavement considerations

The South African “inverted” pavement

CEMENT STABILISED SUBBASE

40mm HMA SURFACING

CRUSHED STONE BASE

Selected subgrade layer

CRACK PREVENTION LAYER

The Concept

HIGHER COMPACTION HIGHER Mr WATER!!!!

The South African “upside-down” pavement

CEMENT STABILISED SUBBASE

40mm HMA SURFACING

Selected subgrade layer

The Upgrade

BITUMEN STABILISED BASE

Distress Mode?

UnboundPermanent deformation

(shear/ruts)

BoundFatigue

(cracking)

Durability??

MOISTURE

TEMPERATURE

TRAFFIC

AGGREGATESSUPPORT

BINDERS

Factors influencing mix selection

Traffic vs Classes of BSM

• BSM1

• BSM2

• BSM3

• > 6 MESA

• High shear strength

• Crushed stone or RAP

• 1 to 6 MESA

• Mod shear strength

• Graded nat gravel, RAP

• < 1 MESA

• Soil gravel or sand

AGGREGATES/COMPACTION

37.5

26.5

19.0

13.2

9.5

6.7

4.7

5

2.0

0

0.8

5

0.4

25

0.2

5

0.1

5

0.0

75

Sieve Size (mm)

0

10

20

30

40

50

60

70

80

90

100P

erc

en

t P

assin

g

Less suitable

Marginal

MarginalIdeal

37.5

26.5

19.0

13.2

9.5

6.7

4.7

5

2.0

0

0.8

5

0.4

25

0.2

5

0.1

5

0.0

75

Sieve Size (mm)

0

10

20

30

40

50

60

70

80

90

100P

erc

en

t P

assin

g

Less suitable

Marginal

MarginalIdeal

BSM Grading requirements (1)

Emulsion slightly coarser than foam

BSM Aggregate requirements (2)

Emulsion Foam

• PI < 7% PI < 10%

• P0.075> 2% P0.075> 4%

• Reclaimed asphalt RA + (15 to 25% Crusher dust)

• Triaxial tests at representative temperature (especially hot regions)

<30 RAP: Recovered pen >30

<5% RAP: Recovered BC(%) >5%

<2% Emulsion Residual BC >2%

No Rejuvenating Agent Yes

TG2/Wirtgen Mix Design Marshall/Superpave

Bitumen StabilisedNon-continuously

boundContinuously

bound

Asphaltic

BSM: RAP Influence

Vibratory Compaction Hammer

Rear View of Frame

Kango Hammer

Mould

Zero Line

Sleeve

Base Plate

Steel Rod

Side of Mould

Vibratory Hammer

Wooden base

To prepare specimens

Kelfkens, U Stell

Influence of Tamping Foot

Sleeve

Vibratory Hammer

Compaction time (vibratory)

Phase Level 1 Level 2 Level 3

Test ITS ITS UCS Triaxial

Foot φ 100mm 150mm 150mm 150mm

Height 65mm 95mm 125mm 300mm

Layers 1 2 2 5

Surchg 5 kg 10 kg 10 kg 10 kg

Foam 10 sec 25 sec 25 sec 25 sec

Emuls 10 sec 15 sec 15 sec 15 sec

Co

mp

Tim

e

BINDERS

Quality Assessment: Foamed bitumen

Emulsion Plant

Bitumen Emulsion

Emulsifiers, acid

Hot water

Bitumen

Heat

Oil in water emulsion (O/W)

Oil phase

Water phase

Bitumen Emulsion Types

BITUMEN EMULSION

Fatty Acids+ NaOH

-ve

Amines+ HCl+ve

No charge

Cationic Anionic Non-IonicType

Emulsifier

pH10 -12ALKALI

2 – 6ACID

7NEUTRAL

RSA BSMsGlobal BSMs

Factors influencing mix design

TEMPERATURE

MOISTURE

Curing Protocol TG22009

CURING METHOD

Preliminary Mix Design

All BSMs72 hours at

40 °C, unsealed

Level 2 & 3 Mix Design

BSM-foam20 hours at 30 °C unsealed

48 hours at 40 °C sealed

BSM-emulsion26 hours at 30 °C unsealed

48 hours at 40 °C sealed

Level 1 Mix Design

CURING METHOD

Preliminary Mix Design

All BSMs72 hours at

40 °C, unsealed

Level 2 & 3 Mix Design

BSM-foam20 hours at 30 °C unsealed

48 hours at 40 °C sealed

BSM-emulsion26 hours at 30 °C unsealed

48 hours at 40 °C sealed

BSM-foam20 hours at 30 °C unsealed

48 hours at 40 °C sealed

BSM-emulsion26 hours at 30 °C unsealed

48 hours at 40 °C sealed

Level 1 Mix Design

“dry”

“equilibrium”“equilibrium”

Curing Protocol (BSM lab)ConditionsSimulated

“dry”

Constant mass @ 40oC

“wet”

Soak 24hrs @ 25oC

60% of OMC @ 40oC, seal

“equilibrium”

CURING: FIELD VALIDATION

Mr (MPa)

EMC (%)

Time

?BASE

PROPERTY

PSPA

Mr (field) versus cure

N7 PSPA Mr Analys is over 7 Months

0

500

1000

1500

2000

2500

3000

3500

4000

0 50 100 150 200 250

Tim e (days)

Mr

(MP

a)

B1-B3 B4-B6 Poly. (B4-B6)

PSPA

Moisture

Mr

MIX DESIGN

Mix Design(TG2, 2009)

PRELIMINARY STEPS

•Aggregate selection and blending

•Aggregate classification

•Pre-treatment

LEVEL 1 Mix Design

•Using Selected Aggregate Blend

•Consider Climate & Early Traffic

Treatment Type Foamed BitumenBitumen Emulsion

Expansion Ratio

& Half-life

Acceptable?

Aggregate-EmulsionCompatibility

Acceptable?Compaction & CureVibratory or Marshall

φ=100mm specimens•ITS dry and soak

NO

•Select bitumen type and content

•Select filler type and content

•Results acceptable?

LEVEL 2 MIX DESIGN

Vibratory Compaction & Cure at Optimums

φ=150mm h=127mm• ITS equil and ITS soak

LEVEL 3 MIX DESIGN

NO

Vibratory Compaction & Cure at Optimums

φ=150mm h=300mm•Triaxial (monotonic)

• MIST (wet) triaxial

FINALISE MIX SELECTIONDETERMINE DEMAC AND CERTAINTY

YES

YES

NO

Done

NO

•Optimise bitumen content

•Check ITS results

•Results acceptable?

Design Traffic < 3 MESA

DoneDesign Traffic < 6 MESA

YES

YES

YES

NO

YES

NO

PRELIMINARY STEPS

•Aggregate selection and blending

•Aggregate classification

•Pre-treatment

LEVEL 1 Mix Design

•Using Selected Aggregate Blend

•Consider Climate & Early Traffic

Treatment Type Foamed BitumenBitumen Emulsion

Expansion Ratio

& Half-life

Acceptable?

Aggregate-EmulsionCompatibility

Acceptable?Compaction & CureVibratory or Marshall

φ=100mm specimens•ITS dry and soak

NO

•Select bitumen type and content

•Select filler type and content

•Results acceptable?

LEVEL 2 MIX DESIGN

Vibratory Compaction & Cure at Optimums

φ=150mm h=127mm• ITS equil and ITS soak

LEVEL 3 MIX DESIGN

NO

Vibratory Compaction & Cure at Optimums

φ=150mm h=300mm•Triaxial (monotonic)

• MIST (wet) triaxial

FINALISE MIX SELECTIONDETERMINE DEMAC AND CERTAINTY

YES

YES

NO

Done

NO

•Optimise bitumen content

•Check ITS results

•Results acceptable?

Design Traffic < 3 MESA

DoneDesign Traffic < 6 MESA

YES

YES

YES

NO

YES

NO

PRELIMINARY STEPS

•Aggregate selection and blending

•Aggregate classification

•Pre-treatment

LEVEL 1 Mix Design

•Using Selected Aggregate Blend

•Consider Climate & Early Traffic

Treatment Type Foamed BitumenBitumen Emulsion

Expansion Ratio

& Half-life

Acceptable?

Aggregate-EmulsionCompatibility

Acceptable?Compaction & CureVibratory or Marshall

φ=100mm specimens•ITS dry and soak

NO

•Select bitumen type and content

•Select filler type and content

•Results acceptable?

LEVEL 2 MIX DESIGN

Vibratory Compaction & Cure at Optimums

φ=150mm h=127mm• ITS equil and ITS soak

LEVEL 3 MIX DESIGN

NO

Vibratory Compaction & Cure at Optimums

φ=150mm h=300mm•Triaxial (monotonic)

• MIST (wet) triaxial

FINALISE MIX SELECTIONDETERMINE DEMAC AND CERTAINTY

YES

YES

NO

Done

NO

•Optimise bitumen content

•Check ITS results

•Results acceptable?

Design Traffic < 3 MESA

DoneDesign Traffic < 6 MESA

YES

YES

YES

NO

YES

NO

Mix Design(Wirtgen, 2012)

Bitumen Stabilised Material

Non-continuously bound

Continuously bound

Asphalt

DIFFERENT BEHAVIOUR PATTERNS

Testing the Material

BSM = non-continuously bound

0

50

100

150

200

250

300

350

400

450

0 500 1000 1500 2000 2500 3000 3500

UCS (kPa)

150m

m I

TSw

et (

kPa)

MR439 Zambia Greece

UCS versus ITS

Developing the Right Tools

Mix Design Approach(TG2, 2009)

• Preliminary tests

• Level 1

• Level 2

• Level 3

• Grading, MDD, OMC, Atterberg Limits (PI)

• 100mmφ Marshall ITS

• Binder, Active filler

• 150mmφ ITS for OBC

• > 3 MESA refine OBC

• 150mmφ 300H Triaxial

• >6 MESA reliability

2013 BSM Stabilisation Design1

Natural Material Charactr

Grading

Moisture density (Mod)

Atterberg Limits (PI)

ITS – select active filler

Step 1

150mmφdry

wet

Typical Bitumen %

0% 1%Lime 1%Cem

3 reps

Typical Bitumen % for BSM-foam

Typical Bitumen % BSM-emulsion

2013 BSM Stabilisation Design1

Natural Material Charactr

Grading

Moisture density (Mod)

Atterberg Limits (PI)

ITS – select active filler

Step 1

150mmφdry

wet

ITS – select bitumen %

Step 2

150mmφdry

wetequil

Typical Bitumen %

0% 1%Lime 1%Cem

3 reps

Selected active filler

-0.4% -0.2% 0% +0.2%

3 reps Bit %

Ref: (Wirtgen,

2012)

Binder type & content – Level 1

ITS

BSM Binder Content

ITSdry

ITSwet

BSM2

Min ITSdry

Min ITSwet

Min BC

2013 BSM Stabilisation Design3

Step 3

150mmφ

300mmH

σ1

σ2 σ2=σ3

Triaxial: Shear properties

Final mix selection

Reliable performance related properties

Flexibility?

C

ϕτ

σ0 50 100 200 kPa

4 σ3 x 2 reps = 8 specimens

Equilibrium

C

ϕτ

σ100 kPa

1 σ3 x 2 reps = 2 specimens

Wet

Level 1 and 2 Classification

Test Dia φmm

BSM1 (kPa)

BSM2 (kPa)

BSM3 (kPa)

Comments

ITSdry 100 >225 175 to 225

125 to 175

Indicates OBC

ITSwet 100 >100 75 to 100

50 to 75

Indicates active filler type & amt

TSR 100 Not applicable Prob mat TSR < 50 % ITSdry > 400 kPa

ITSequil 150 >175 135 to 175

95 to 135

OBC refined

ITSsoaked 150 >100 75 to 100

50 to 75

Adjusted to ITSwet

Critical Material Properties• Tri-axial test to determine:

– Shear parameters ( C & ϕ )– Resilient modulus ( Mr )

– Permanent deformation behaviour

σ1

t

σ1

tC

ϕτ

σ

εi

t

εp

log t

Triaxial Testing

Effect of using BSMτ

σ

Shear stress

Normal stress

C = Cohesion

φ Friction angle

Higher

Cohesion

Effect of Binder

Typical target binder contents

>75% RAP + Cr Dust

Granular GM > 2

Granular GM < 2

Typical Binder Cont. (%)

2% 2.5% 3%

Shear properties (monotonic triaxial at 25°C)

Cohesion C

0

100

200

300

400

500

A-7

5C

-0

B-7

5C

-0

C-7

5C

-0

A-7

5C

-1

B-7

5C

-1

C-7

5C

-1

A-7

5M

-0

B-7

5M

-0

C-7

5M

-0

Co

hesio

n [

kP

a]

E E E E E E FFF20

25

30

35

40

45

50

A-7

5C

-0

B-7

5C

-0

C-7

5C

-0

A-7

5C

-1

B-7

5C

-1

C-7

5C

-1

A-7

5M

-0

B-7

5M

-0

C-7

5M

-0

Fri

cti

on

an

gle

[d

eg

rees]

Friction Angle φ

E E E E E E FFF

Ebels

25% RAP 25% RAP75% RAP 75% RAP

Simple triaxial trials - with latex membrane as confining tube

(Mulusa)

New “Simple triaxial”

Research Triaxial Test RTT versus Simple Triaxial Test STT

0

200

400

600

800

1000

1200

1400

0.0 1.0 2.0 3.0 4.0 5.0

Strain [%]

Ap

pli

ed

Str

ess [

kP

a]

RTT

STT

0

200

400

600

800

1000

1200

0.0 1.0 2.0 3.0 4.0

Strain [%]

Ap

pli

ed

Str

ess [

kP

a]

RTT

STT

BSM Crushed Hornfels with 3.3% Emulsion

σ3 =50 kPa and 1% Cement σ3 = 200 kPa and 0% Cem

BSM Classification ito Shear Properties

Cohesion (kPa)

> 250

100 – 250

50 – 100

Angle of Internal Friction (º)

> 40

30 to 40

< 30

Equivalent BSM Class

BSM 1

BSM 2

BSM 3

MOISTURE DAMAGE

Durability of BSMs: Improved Tests- Untreated Material Properties - Moisture sensitivity tests

Moisture Induction Sensitivity Test MIST

(Twagira)

50mmlayer

0.20

0.40

0.60

0.80

1.00

1.20

0 10 20 30 40 50

Foam +1%Lime

Foam +1%Cement

Pulsing time (sec)

De

gre

eo

fs

atu

rati

on

(%)

MIST saturation time

Effect of moistureτ

σ

Shear stress

Normal stress

C

Cohesion

φ Friction angle

Lower Cohesion

Effect of Moisture

Retained Cohesion (Twagira)

0

10

20

30

40

50

60

70

80

90

Q+ 0C -F

Q+ 1C -F

Q+ 1L -F

Q+ 1C-E

Q+ 1L -E

H+ 0C -E

H+ 1C -E

H+ 1L -E

BSM 1

BSM 2

BSM 3

Res

idua

l coh

esio

n [%

]

Tested BSMs

BSM Classification ito Moisture Resistance

Retained Cohesion (%)

> 75

60 – 75

50 – 60

< 50

Equivalent BSM Class

BSM 1

BSM 2

BSM 3

Unsuitable

SUPPORT

0

200

400

600

800

1000

1200

1400

1600

1800

0 0.1 0.2 0.3 0.4 0.5 0.6

Strain

Str

ess

(k

Pa

)Dissipated Energy

W = 0.5 σεA1A2

BSM+2%cem

BSM+1%cem

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 0.2 0.4 0.6 0.8 1 1.2

Ve

rtic

al

Dis

pla

ce

me

nt

at

ITS

ma

x (

mm

)

Active Filler Content (%)

ITS wet (All)

85% RA Emul

85% RA Foam

75% RA Foam

25% RA Emul

85% RA Emul

85% RA Emul

85% RA Foam

ITS displacement measures (from Mix Design)

Triaxial data from Mix Design

Can refine by separating data based on σ3

So we need quality from structural design

Conclusions

• Understanding of material behaviour of BSMs has increased significantly

• Active filler versus bitumen content is very important

• More advanced test methods (triaxial)

• Mix Design is linked to Structural Design method for BSMs

Thank you

Questions??