Cold Recycling & Bitumen

Stabilised Materials BSMs

Research and
Implementation?
Kim Jenkins

57th Annual Illinois Bituminous

Paving Conference
12th December 2016

tellenbosch
tellenbosch

Outline

1. What is BSM?
2. Mix Design
3. Structural Design
4. Application
5. Where to now?
 BSM Binder Options
BITUMEN EMULSION FOAMED BITUMEN
Colloidal Mill Expansion chamber
Acid or
Caustic Soda
Hot
Surfactants bitumen

Water Bitumen
Wat Water Air
Mill

5 microns

Orientation on BSM
4 Cement
3
stabilised
Cement %
Rigidity

Bound Non-
2 continuously
Bound
1
BSM
0
Asphalt
Granular
Bound
Unbound

0 1 2 3 4 5 6
Bitumen %
Flexibility
 Influence of Active Filler
BSM foam Strength versus Flexibility BSM

eb
800 4000
Foamed bitumen, Strain
700 Cement, Strain* 3500

Unconfined Compressive
Foamed bitumen, UCS
600 3000
Cement, UCS*
Strain-at-break

Strength (kPa)
500 2500

400 Cement < 1% 2000

300 1500

200 1000

100 500

0 0
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3
Cement : Foamed Bitumen Ratio
CSIR
* Cement treated with 2 percent cement and no foamed bitumen. Values plotted
at an arbitrary ratio of 1.25 for 2 percent cement and 1.2 for 1 percent cement.

BSM Triaxial Tests Shear properties

(monotonic tests at 25°C)

Cohesion C Friction Angle f

500 50
Friction angle [degrees]

400 45
Cohesion [kPa]

40
300
35
200
30
100 25
E E F E E F E E F E E F E E F E E F
0 20
A-75C-0

B-75C-0

C-75C-0

A-75C-1

B-75C-1

C-75C-1

A-75M-0

B-75M-0

C-75M-0
A-75C-0
B-75C-0
C-75C-0
A-75C-1

B-75C-1
C-75C-1
A-75M-0
B-75M-0
C-75M-0

25% RAP 75% RAP 25% RAP 75% RAP

Jenkins, 1999 & Ebels, 2006
 Resilient Modulus of BSM
Stress dependency: Foamed BC = 2%
HMA > 2500 MPa
Resilient Modulus Mr (MPa)

1350
BSM
1150
950
s3 GCS
750
12kPa
550 24kPa
48kPa
350
72kPa
150
0.0 200.0 400.0 600.0 800.0 1000.0
Sum of Principal Stresses q (kPa)

Research: Visco-elasto-plastic &

flexural properties on BSM-foam Tref = 20C

HMA/WMA
Fatigue cracks
HW

BSM
BSM

Rutting

Equiv HOT T or Equiv COLD T or

Slow Traffic Fast Traffic
 BSM test methods
C,f Mix design to
Reality Performance
Index Design BSM layers

h1
h2
h3

Testing >200
S
Vl
Z project mix
e
ie Sr designs!
be idSo
rv B
e tWL
Compaction ae a
o eoi
t s
e
f o n
o e
l
Md e
r P
o Re
y lo
1990 2000 H 2010 adnb
ul
d
a t
a
m e
m Years
s
e
e
r

Mix Design Flowchart

Sampling
Optimum Determine
bitumen shear
Blend addition properties

Sample No TRIAXIAL
preparation
Yes
SUITABLE? Effect of active
filler
Specification
Preliminary
ITS C (kPa) f (0)
tests
>250 >40
 Standardised Mixing Method
FOAMED BITUMEN UNIT

PUGMILL
MIXER

Lab Compaction: Vibratory Hammer

Vibratory Power Frequency Mass Point

hammer rating (W) (Hz) (Kg) Energy (J)

Kango 637® 750 45.83 7.5 27

Bosch GSH 11E® 1500 15 - 31.5 10.1 16.8
Bosch GSH 11VC® 1700 15 - 30 11.4 23

For PI >8%, cannot achieve 100% Mod. AASHTO density

 Influence of Frame
FRAME
Refusal Density
TYPE
2400
80% OMC, RFR,
2350 10kg Surcharge Rigid
Density (kg/m3)

2300

2250 80% OMC, LFR,

10kg Surcharge Loose
2200

2150
80% OMC, RFR, Rigid
20kg Surcharge
2100

2050
G2 G4 80% OMC,LFR,
Material Type 20kg Surcharge Loose

Comparison of refusal density for G2 and G4 material

(Stell Univ)
 Inter-Layer Roughening (ILR) Device

ITS 2 layers

6 layers
Triaxial X 50mm

Inventor: Wynand van Niekerk

S1A

CT Scans 85

BSM-emulsion
75

65
Scan slice nummer (boorkern lengte in mm).

55

Poor
45
attention to
35
interlayer
preps
25

15 voids
Mortar
stone
5

0 20 40 60 80 100
-5
Volume in %
Why is curing important?
Mr (field) versus cure
PSPA
N7 PSPA Mr Analysis over 7 Months

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

4000
% OMC
3500 100
3000
MC Mr
80
Mr (MPa)

2500
2000
1500 60
1000
500 Moisture 40
0
0 50 100 150 200 250
Tim e (days)

tellenbosch

New Triaxial
Apply Load (stress s1)

Test at
25ºC

Confining Pressure s3
(inflate tube)
 Validation
Research Triaxial Test RTT versus
Simple Triaxial Test STT
Applied Stress [kPa]

1200
1400

Applied Stress [kPa]

1000 RTT 1200
800
STT 1000 RTT
600 800 STT
400 600
200 400
0 200
0
0.0 1.0 2.0 3.0 4.0
0.0 1.0 2.0 3.0 4.0 5.0
Strain [%] Strain [%]

BSM Crushed Hornfels with 3.3% Emulsion

s3 =50 kPa
tellenbosch and 1% Cement s3 = 200 kPa and 0% Cem

APPLY LOAD (3mm/min)

σ1

σ3

APPLY CONFINING PRESSURE (AIR)

 Determine shear properties
σ
(C and φ)
1

BSM
t
Shear σ3
stress UNBOUND
f Friction σ1
angle

σ3
Cohesion
0 50 100 200 s
Normal
stress
(kPa)

Durability of BSM
t
Shear f Friction
stress angle

Effect of
Moisture
Cohesion Loss = 25% max
Retained Cohesion CR = CR*100/CBSM
CBSM
Cohesion s
Normal
stress
 Structural Design 90mm Asphalt
Considerations
250mm CIPR:
2.5% Foam 1% Cem

BSM Design for Max Rut Depth

Lab Triaxial
(same principle as Granular Design)
Analysis

Permanent deformation Design Life for 10mm rut

(rutting) design for granular
material
 Design Function for BSM
Relative Density Plastic Strain (a/b)

a
b

𝑁 = 𝑓 (𝑅𝐷, 𝑅𝑒𝑡𝐶, 𝑃𝑆, 𝑆𝑅)

Retained Cohesion Stress Ratio

Mr change with trafficking (triaxial)

2000.0
BSM1-foam (2%) SR=49%
1800.0
SR=57%
1600.0 SR=65%
SR=67%
Resilient Moduli (MPa) .

1400.0
1200.0
1000.0
800.0
600.0
400.0
200.0
0.0
100 1,000 10,000 100,000 1,000,000
Load repetitions

Jenkins et al, IJPE Jenkins, TU Delft, 1999

 In service behaviour of Mr
Influence of support & traffic
2000
Conceptual
Effective Stiffness (Mpa)

Traffic
1500 Poor Support
High sd/sd,f
1% cem CTSB
1000
1% cem G5SB
N7 1% cem G7SB
500

0 Cem % Support
0 1 2 3 4 5 6 7 8 9 10 BSM1 type
Years or Traffic

Effective Long Term Mr for BSM base

Mr from FWD back-calcs

2%cem

2%cem
1%cem
1%cem

SAPDM - R35 : Theyse, 2015 & Lynch, 2014

 Effective Long Term Mr Stiffness (MPa)
for BSM base
Supporting Layer

BSM Class Cemented Granular Subbase

Subbase
BSM 900 – 1750 700 – 1200
(RAP + GCS)
BSM 800 – 1200 600 – 900
(GCS Grade Crushed
Stone)

ELT Mr = f (aggregate type and quality, RAP %,

bitumen %, support, traffic, climate)

Case Study – Ayrton Senna Highway

Brazil’s most heavily-trafficked
highway

8-lanes / divided

30
 Key Data

AADT > 200,000vpd (15% heavy)

(> 15,000 heavies / day in each direction)

Milling & Replacing 100mm HMA lasts < 3 months

Lane closure only between 21:00 – 05:00

8 HOURS

Results from Pavement Investigation

HMA ± 100mm

CEMENTED CRUSHED STONE ± 250mm

6% CEMENT

GRADED CRUSHED STONE ± 200mm

SELECTED COARSE GRAVEL (CBR >25) ± 200mm

EMBANKMENT (RIVER LEEVEE) (CBR > 15)

Semi-infinite
 Rehabilitation Options?? (8-hour working window)
HMA
20mm
50mm
BSM-b 130mm
100mm
350mm HMA ?
BSM-a 200mm

Step 1. Mill off asphalt layers

100mm
Impact crusher (20mm setting)
Grading Correction using Single Stage Crushing

Normal CONTINUOUS
AFTER
RAP CRUSHING

Wirtgen KMA 220 plant mixer

2.0% / 2.1% Foamed Bitumen
1.0% OPC

Mixed material placed in stockpile

Step 2. Mill and remove CTB layer

250mm
Step 5. Import / pave / compact 130mm BSM layer

130mm BSM (RAP / crushed stone blend)

Step 6. Import / pave / compact 20mm HMA

20mm HMA (gap-graded / fine mix)

Temporary surfacing
18th November 2011
31ST January 2012

Currently (3.75 years later)

> 100 lane-km rehabilitated using this method

PROBLEM SOLVED !

Way Forward: Research

Monotonic Load Cycle (triaxial)

GCS BSM
(Bredenhann & Jenkins, 2016)
 Way Forward: Research
Dynamic Conditioning (triaxial)

GCS BSM
(Bredenhann & Jenkins, 2016)

Way Forward: Research(2)

Dynamic Triaxial – Permanent Deformation
0.035
Deformation rate

0.030
(ep/N x 10-6)

0.025

0.020

0.015
GCS
0.010
BSM
0.005

0.000
0.5 0.6 0.7 0.8 0.9 1

Stress ratio

(Bredenhann & Jenkins, 2016)

 Flexural Strain-at-break
 All beams compacted in a mould

 Testing temperature: 25°C

 LVDT on top of the beam to accurately measure displacement in the middle of

the beam.

LVDT

(Campher, 2014)

Flexural Strain-at-break & DE

Material parameter
Specimen
Average Average Strain- Average Average
specification Maximum Stress at-break Dissipated stiffness
(kPa) (µԐ) energy (Pa) (MPa)
0.9% Emulsion; 1%
Cement 174.4 376.5 39.1 524.2

2.4% Emulsion; 1%
Cement 254.9 537.2 89.8 473.1

2.4% Emulsion; 2%
Cement 320.4 391.1 Increase in 78.8 821.6
bitumen emulsion (specimens
containing 1% cement)
2.4% Foamed; 1% Increase in cement (stabilised with bitumen
Average stiffness (MPa)
emulsion)
Flexibility related

Cement 211.6 480.8 Increase in 68.7

cement (stabilised with foamed
447.4
bitumen
parameters

emulsion)
2.4% Foamed; 2%
Average Dissipated energy (Pa)
Cement 383.8 508.7 151.3 761.9

Average Strain-at-break (µԐ)

-40.00 -20.00 0.00 20.00 40.00 60.00 80.00 100.00 120.00 140.00
% Change in parameter value (Campher, 2014)
Flexibility (triaxial)

(Llewellyn, 2016)

Flexibility (triaxial)

(Llewellyn, 2016)
Factors Influencing BSM Flexibility
Analysis of Variance
Summary of P values for variables in ANOVA analysis

0.85
0.8
0.75
0.7
0.65
0.6
0.55
0.5 Strain at Break
P- 0.45
value 0.4
0.35 Dissipated
0.3 Energy
0.25
0.2
0.15
0.1
0.05
Significant
0
Variables
<0.05
Fenton, 2013

Conclusions
• Mix design system in place
– Aim for flexibility not high strength
– Update of equipment (vib hammer & triax)
• Pavement design
– New ME design function
– Link of mix- and pavement-design (C & f)
• Application (field performance data)
• Way forward: flexibility focus

tellenbosch
 tellenbosch

Pavement Balance

Granular Base BSM Base

Mr (MPa) Mr (MPa)
Asp 3000 3000 Asp
800
CTB
G1 2800
500 Base >1000 BSM1

350 ---- >1500

350 CTSB
G5 Subbase G5

200 200

SG 150 Subgrade 150 SG

tellenbosch
 Research on BSM Flexibility
How can we benefit from?

(Llewellyn, 2016)

Strain-at-break vs Fatigue
25%RA & 0%Cem

1E+07
emulsion A
1E+06
4PB Fatigue emulsion B
Number of repetitions

1E+05 foamed bitumen C

1E+04

1E+03

1E+02
A
B
1E+01
C
Strain at break
1E+00
100 1,000 10,000
Strain [x 10 -6 ]Stellenbosch University