Durability Related Characterization of Cold Recycled Mixes ...

CEDR – Transnational Programme 2012: Recycling

Workshop, Praha, 24.-25.9.2015

Durability Related Characterization of Cold Recycled Mixes – Stiffness, Fatigue, Cracking,

Moisture Susceptibility

Jan Valentin – Faculty of Civil Engineering, CTU in Prague (CZ)Zuzana Čížková – Faculty of Civil Engineering, CTU in Prague (CZ)Jan Suda – Faculty of Civil Engineering, CTU in Prague (CZ)Konrad Mollenhauer, University Kassel (GE)Ciaran McNally, University College Dublin (IR)Fátima Batista, LNEC (Portugal)

This presentation has been supported by the research project COREPASOL carried out as part of the CEDR Transnational Road research Programme Call 2012. The funding for the research was provided by the national road administrations of Belgium (Flanders), Denmark, Finland, Germany, Ireland, Netherlands, Norway, Sweden and UK.

Moisture susceptibility

Moisture susceptibility – the selected approach

� FACT: Moisture susceptibility is considered to be one of the mostimportant characteristics of a road base or binder course if asphalt mixesare used in the pavement structure.

� Moisture susceptibility research of mixes with higher content of RAP orresearch focused fully on cold recycled mixes is significantly newer andresearch focused fully on cold recycled mixes is significantly newer andtherefore less developed, even in terms of available research findings andrecommendations.

� Standard tests for assessment of asphalt mix durability were considered.

Moisture susceptibility – influencing factors

� Chemical bond between aggregate and bitumen.� strength depends on chemical properties of used materials, particularly on

their surface tension;

� replacement of bitumen molecules by water molecules, which results inaggregate stripping, is caused by the fact that the dipolar water moleculesare more polar than the molecules of bitumen.are more polar than the molecules of bitumen.

� Voids content and the influence on moisture susceptibility (+ conectivityand structure of voids).

� Weather conditions during the paving, which can lead to poor compactionand hence the greater voids content or permeability.

� Aggregate porosity, because if too porous aggregate is used, a largeamount of bituminous binder gets right into the pores.

Moisture susceptibility – test protocols

� EN 12697-12 approach (with different compaction)

� AASHTO 283 T testo protocol (modified)

� TP208 protocol (Czech national procedure).

Selected test methods

� SR method (TG2)

� Tunnicliffe and Root test method (NCHRP 274)

� Moisture Induced Sensitivity Tester (MIST)

� Dynamic Asphalt Stripping Machine (DASM)

� etc.

Other existing test procedures

Moisture susceptibility – Czech study

Mix A Mix B Mix C Mix D Mix K Mix S Mix W

Reclaimed asphalt mix 91.0% 90.5% 94.0% 93.5% 95.5% 95.0% 94.0%

Water 2.5% 2.0% 2.5% 2.0% 2.0% 2.0% 2.5%

Bituminous emulsion 3.5% 0.0% 3.5% 0.0% 0.0% 0.0% 2.5%

Foamed bitumen 0.0% 4.5% 0.0% 4.5% 2.5% 2.0% 0.0%

Cement CEM II B32.5 3.0% 3.0% 0.0% 0.0% 0.0% 1.0% 1.0%

Cold recycled mix

Bulk density Maximum density Voids content

(g.cm-3) (g.cm-3) (%)

RAP 0/11 RAP 0/22 RAP 0/11 RAP 0/22 RAP 0/11 RAP 0/ 22

Mix A 2.154 2.220 2.449 2.377 12.1% 6.6%

Mix B 2.147 2.155 2.419 2.295 11.3% 6.1%

Mix C 2.098 2.203 2.444 2.412 14.2% 8.7%

Mix D 2.151 2.119 2.378 2.237 9.6% 5.3%


� Three moisture susceptibility test procedures.

� indirect tensile strength test at 15°C by using cylindrical specimens with aconstant load rate of 50 mm / min.

� Stiffness by indirect tensile stress test (IT-CY) in compliance with EN

What was assessed?

� Stiffness by indirect tensile stress test (IT-CY) in compliance with EN12697-26 at 15°C.


Cold recycled mixes with max. 1 % of cement

Moisture susceptibility procedures Days of curing

3 days air storing Accelerated curing + 3 days in air 7 days

EN 12697-12 Accelerated curing + water saturation under pressure (6.7±0.3) kPa and 3 days in water in 40 °C 7 days EN 12697-12 and 3 days in water in 40 °C 7 days

AASHTO T283 Accelerated curing + 1 freezing cycle according to AASHTO 6 days TP 208 7 days in air + 7 days in water 14 days


Cold recycled mixes with max. 1 % of cement


� any curing of test specimens in water results in poorer values of measuredcharacteristics.

� ITS and stiffness values of the specimens cured according to AASHTOwere on average about 10-30 % lower than values of specimens cured

Facts and findings

were on average about 10-30 % lower than values of specimens curedaccording to the method specified in EN.

� The lowest values were caused by the longest immersion of testedspecimens in water according to the methodology defined in TP 208.

� better indirect tensile strength and stiffness were observed for mixescontaining finer RAP (0/11 mm).


Cold recycled mixes with 3 % of cement

Moisture susceptibility procedures

14 days air storing 14 days air curing

EN 12697-12 11 days air curing + water saturation under pressure (6.7±0.3) kPa and 3 days in water at 40 °C 12 days air curing + 1 freezing cycle according to modified procedure of AASHTO

AASHTO T283 12 days air curing + 1 freezing cycle according to modified procedure of AASHTO T283

TP 208 7 days in air + 7 days in water


Cold recycled mixes with 3 % of cement

� Water in the mix during the saturation, is partly used by cement which cangradually achieve higher strength. This compensates the negative effect,which causes the water in a cold recycled mix .


� Mix designs of K, S and W contain 2.0-2.5 % foamed bitumen (S and K) or2.5 % emulsion (W) and 0 % or 1 % of cement and therefore the samespecimen curing procedure as for mixes C and D was applied.

� Considering the fact that the ITS and stiffness modulus values depend

Cold recycled mixes with lower content of bituminou s binder

� Considering the fact that the ITS and stiffness modulus values dependsignificantly more on the hydraulic binder content, it is possible to placethis group of cold recycled mixes as an option between previously testedmixes without cement (C and D) and mixes with 3 % of cement (A and B).This is also proved by the determined values, which basically fell into theinterval limited on both ends by values gained for mixes A-D.

Moisture susceptibility – Irish study� Quick Viscoelastic Mixes (QVE) selected with 3% of cement. Option

without any cement was included as well included.

� Gyratory compaction used (600 kPa and 30 gyrations per minute).

� Stiffness and dry strength tests were performed on mixtures containing 1.5% cement.

� ITS tested at 25°C (standard for IR) and 15°C.� ITS tested at 25°C (standard for IR) and 15°C.

Mix Mix type

Mix Constituents (%)

Surface Material

Granular Material

Cement Water/

Moisture Binder Emulsion

A Emulsion Mix without cement 33.3 66.7 0 4 2.2 3.5

B Emulsion Mix with cement 32.8 65.7 1.5 4 2.2 3.5

C Emulsion Mix with cement 32.3 64.7 3 4 2.2 3.5

Note: 1. All percentages contents are by mass, 2. The emulsion used in the mix is the Irish Tar and Bitumen Suppliers Ltd. product: ‘Fuarflex’, of the binder specification

C60B4. 3. The amount of added water in cold recycled mix with bituminous emulsion was adjusted per mount of water in the

emulsion and moisture content of the surface and granular material.

4. The amount of added water in cold recycled mixes with foamed bitumen was adjusted per mount of moisture content of the surface and granular material.

Moisture susceptibility – Irish studyMix Mix type Curing Method Test temperature (°C) ITS (MPa) ITSR (%)

A Emulsion Mix without cement

3 days at 50°C

25 0.28(wet)

80 0.35(dry)

15 0.58(wet)

79 0.74(dry)

C Emulsion Mix with 3% cement

14 days at 20°C

25 0.47(wet)

85 0.65(dry)

15 0.83(wet)

88 0.94(dry)

� mixtures with and without cement show good ITSR values (79% to 88%);

� increase of ITS during curing period (7 – 14 days);

� increase of ITS with decrease of the test temperature (25°C to 15°C) –on ITSR the effect is minimal (same findings proven by German study aswell);

Moisture susceptibility – Portuguese study

� Specimens compacted by applying a compressive load of about 7.5 MPa(static compaction).

� ITS tested at 15°C.

Moisture susceptibility – Portuguese studyCold recycled mix ID CM-E4C0 CM-E4C1 CM-E4C1

Content of each mix

component

Reclaimed Asphalt 91.8 % 91.8 % 91.8 %

Filler 2.8 % 1.8 % 0.8 %

Cement - 1.0 % 2.0 %

Emulsion 3.8 % 3.8 % 3.8 %

Added water 1.6 % 1.6 % 1.6 %

Moisture susceptibility – German study

� Specimens compacted by double-plunger static compaction. For 3minutes a force of 50 kN was held constant (static compaction).

� Accelerated curing applied.

� ITS tested at 5°C and 15°C.

Content of bitumen Content of Content of Sample name

Content of bitumen emulsion (residual

bitumen) C 60 B1 – BEM

Content of cement CEM I

Content of reclaimed

asphalt

Content of limestone filler

I 2,0 (1.2) % 0.0 95 % 5 %

II 3.5 (2.1) % 0.0 95 % 5 %

III 3.5 (2.1) % 1.5 % 95 % 5 %

Moisture susceptibility – German study

Mix Mix type Curing Method

Test temperature (°C)

ITS (MPa) ITSR (%)

I 2 % Emulsion Mix without cement 3 days at 50°C

15 0.14(wet)

36 0.39(dry)

5 0.22(wet)

32 5 32 0.68(dry)

II 3.5 % Emulsion Mix without

cement 3 days at 50°C

15 0.31(wet)

41 0.76(dry)

5 0.59(wet)

44 1.34(dry)

III 3.5 % Emulsion Mix with 1.5 %

cement 14 days at

20°C

15 0.63(wet)

76 0.83(dry)

5 0.95(wet)

81 1.17(dry)

Moisture susceptibility – conclusions

Study Sample Mix

granulate

Bitumen emulsion content

Cement content

Compaction Curing ITS dry ITSwet ITSR [%]

CZ

A 100 3.5 3.0

Static

11d20 0,75 0,78 104

C 100 3.5 0,0 3d50 0,65 0,3 46

W 100 2.5 1,0 3d50 0,75 0,6 80

IR A 33 3.5 0

Gyratory 28d40 0,74 0,58 78

C 33 3.5 3.0 28d40 0,93 0,83 89

GER

I 95 2,0 0

Static

3d50 0,39 0,14 36

II 95 3,5 0 3d50 0,76 0,31 41

III 95 3,5 1,5 11d20 0,83 0,63 76

PT

CME4C0

100 3.8 0

Static

3d50 0,89 0,77 87

CME4C1

100 3.8 1.0 3d50 0,75 0,51 68

Moisture susceptibility – conclusions

� Procedure according to EN 12697-12 feasible:� Dry conditioning.

� Water conditioning at 40°C for 3 days after saturation

� Pay attention to the used reclaimed asphalt since it was repeatedly proventhat there is a crucial problem with its homogeneity.that there is a crucial problem with its homogeneity.

� Cold recycled mixes with combined binder will have less problem withmoisture susceptibility.

� Even for sam grading results influenced by the RA bitumen and itscontent.

� NECESSARY: continuous research of the voids content influence.

� Defining a fixed treshhold value for ITSR still problematic.

Stiffness

Stiffness modulus� Assessed by the repeated indirect

tensile stress test (IT-CY) according to EN 12697-26

� Characterizes short term rheological behaviour taking into account deformations lasting only for tens or deformations lasting only for tens or hundreds of milliseconds

� Non-destructive test with good reproducibility

� Better resistance to permanent deformations can be expected by the mixes with higher stiffness

� Cylindrical specimens (diameter 150 mm, height 60 mm)

� Complemented with values of Indirect Tensile Strength (ITS)

Stiffness modulus – influence of binder contentMixes with foamed bitumen

Foamed bitumen:

2.0% 2.5% 3.5% 4.5%

Cem

ent:

0% R K V D

1% S L P3 P1

Cem

ent:

1% S L P3 P1

3% T M P4 B

5% U N P5 P2

� Optimal foam content seems to bebetween 2.0 % – 2.5 %

� Using additional cement has biggerpositive impact on stiffness modulusand ITS than higher content ofbituminous binder.

Mix A Mix C Mix E Mix G Mix WRAP 91.0% 94.0% 93.0% 92.5% 94.0%Water 2.5% 2.5% 2.5% 2.5% 2.5%Bituminousemulsion

3.5% 3.5% 3.5% 3.5% 2.5%

Mixes with bituminous emulsion

Stiffness modulus – influence of binder content

emulsionCement 3.0% 0.0% 1.0% 1.5% 1.0%

Mix 2.5E

Mix 3E

Mix 3.5E

Mix 4.5E

RAP 94.5% 94.5% 94.0% 93.5%Water 3.0% 2.5% 2.5% 2.0%

Bituminousemulsion

2.5% 3.0% 3.5% 4.5%

Cement 0.0% 0.0% 0.0% 0.0%

� Optimal range of bituminous emulsion content is 2.5 – 3.0 %.

Additional testing (different curing):

Stiffness modulus – influence of binder content

� Optimal foam content seems to be between 2.0 % – 2.5 %

� Optimal range of bituminous emulsion content is 2.5 – 3.0 %.

� Increase in bituminous binder content above optimum increa ses thematerials flexibility which results in a decrease of stiffn ess.

� Using additional cement has bigger positive impact on stiffness modulus� Using additional cement has bigger positive impact on stiffness modulusand ITS than higher content of bituminous binder.

� With increasing cement content stiffness and ITS are increasing as well,but higher cement content results in increased brittleness

� According to literature it is recommended to add max. 6 % of cement,because too rapid growth in initial strength could lead to formation ofhydration cracks or microcracks.

Stiffness modulus – Influence of curing time

Mix A Mix G Mix E Mix C Mix W Mix B Mix F Mix DRAP 0/22 91.0% 92.5% 93.0% 94.0% 94.0% 90.5% 92.5% 93.5%Water 2.5% 2.5% 2.5% 2.5% 2.5% 2.0% 2.0% 2.0%Bituminous emulsion 3.5% 3.5% 3.5% 3.5% 2.5% 0.0% 0.0% 0.0%Foamed bitumen 0.0% 0.0% 0.0% 0.0% 0.0% 4.5% 4.5% 4.5%Cement 3.0% 1.5% 1.0% 0.0% 1.0% 3.0% 1.0% 0.0%

� Significant difference in stiffness after 7 days in mixes with cement and without cement. This difference then gradually decreases

� Addition of cement has markedly bigger positive influence on stiffness than on ITS

Stiffness modulus – Effect of fines and temperature

� With increasing testing temperature stiffness decreases

� Adding fine aggregate seems to be quite advantageous

� Adding of waste filler causes

REC1 REC2 REC3 REC1a REC2a REC3aWater 5.00 5.00 5.00 5.00 5.50 5.50Cement 3.00 2.00 1.50 3.00 2.00 2.00Bituminous emulsion 2.50 2.50 2.50 2.50 2.50 2.50RAP 0/11 80.55 72.40 72.80 80.55 72.00 63.00Aggregate 0/2 0.00 0.00 0.00 8.95 18.00 18.00Waste filler 8.95 18.10 18.20 0.00 0.00 9.00

REC4 REC5 REC6 REC7 REC8 REC9

� Adding of waste filler causes even more significant increase and can partly substitute using of cement

Water 5.10 5.10 5.10 5.00 5.00 5.00Cement 3.00 3.00 3.00 3.00 3.00 3.00Bituminous emulsion 2.50 2.50 2.50 2.50 2.50 2.50RAP 0/11 89.40 71.50 62.60 0.00 0.00 0.00RAP 0/22 0.00 0.00 0.00 89.50 71.60 62.65Aggregate 0/4 0.00 17.90 26.80 0.00 17.90 26.85

Stiffness modulus – Effect of fines and temperatureEffect of waste filler content

� With increasing testing temperature stiffness decreases

� Adding fine aggregate seems to be quite advantageous

� Adding of waste filler causes

REC1 REC1a REC2 REC2a REC3 REC3a

Water 5.00 5.00 5.00 5.50 5.00 5.50

Cement 3.00 3.00 2.00 2.00 1.50 2.00

Bituminous emulsion 2.50 2.50 2.50 2.50 2.50 2.50

RAP 0/11 80.55 80.55 72.40 72.00 72.80 63.00

Fine aggregate 0/2 0.00 8.95 0.00 18.00 0.00 18.00� Adding of waste filler causes

even more significant increase and can partly substitute using of cement

Fine aggregate 0/2 0.00 8.95 0.00 18.00 0.00 18.00

Waste filler 8.95 0.00 18.10 0.00 18.20 9.00

Stiffness modulus – Mixes with recycled concrete, recycled gravel/sand and pulverized concrete

Stiffness modulus – Mixes with recycled concrete, recycled gravel/sand and pulverized concrete

� Unlike mixes with RAP – very different trends in stiffness compared to ITS trends

� Content of RAP is necessary for high ITS, but highest stiffness was achieved by mixes without RAP

� Influence of aggregate skeleton is more important than amount of pulverized concrete added, pulverized concrete has very good effect on stiffness but it caused similar or lower values of ITS

Complex modulus

Complex modulus – what can be the reason for this test?� Cold recycled mixes are affected by time-dependant behavior.

� If cold recycled mixes are subjected to a very small loading, then it comesto a combination of elastic, delayed elastic and viscous behavior.

� In the range of low temperatures and high loading frequencies thesemixes behave like elastic material in a solid phase with almost fullymixes behave like elastic material in a solid phase with almost fullyreversible response => impact on deformation behavior of these mixes.

� Can we evaluate this behavior and ideally design master curves.

� Is 4PB test applicable?

Complex modulus – possible limits of testing

� The testing beam has to be locked in by the clamps. Such clampingconstitutes locally stress and deformation which is introduced to the beammaterial (extra stress and deformation can represent fatigue damage orlocal non-linear effects).

� It is not possible to design and fabric 4PB-PR apparatus, which wouldexcept any friction and at the same time allow free displacement + duringexcept any friction and at the same time allow free displacement + duringits life-time it would not deteriorate.

� Shear forces act on the beam in the area between the outer and innerclamps. These forces are causing testing beam deformation, whereassuch deformations are not taken into account in the prescribed testmethodology. The test beam in the 4PB-PR apparatus is subjected tothese shear forces in the area between the clamps.

� Clamps limit the possibility of displacement in cross-section. This leadsaccording to our experience to ineligible deformations of the testing beamin the areas close to the clamps and this is not in accordance with thedeflection theory.

Complex modulus – results of testingTesting temp. Frequency BCSM-BE BCSM-FB BSM-BE BSM-FB

(ºC) (Hz) E*

(MPa) δ (º)

E* (MPa)

δ (º) E*

(MPa) δ (º)

E* (MPa)

δ (º)

10

50 4 620 0.03 5 669 0.10 4 495 5.80 5 222 0.15 30 4 210 3.03 5 203 1.22 3 960 5.78 4 723 5.43 20 3 729 4.08 4 582 3.87 3 503 8.95 4 156 7.35 10 3 482 7.13 4 152 6.76 3 164 12.24 3 651 11.97 8 3 403 8.12 4 066 7.63 3 053 13.05 3 565 12.96 5 3 243 9.38 3 851 9.29 2 855 14.88 3 261 14.90 2 3 314 11.32 3 551 11.23 2 515 16.98 2 794 17.20 1 2 844 11.56 3 247 12.28 2 298 18.21 2 438 19.10

0,5 2 614 11.87 2 842 12.82 1 981 18.81 2 037 20.91

20

50 3 870 0.00 2 852 12.69 2 829 6.44 3 892 5.20 30 3 441 4.20 2 411 14.71 2 605 11.95 3 416 9.93 20 2 955 4.85 2 039 26.11 1 983 14.86 2 827 13.29 10 2 631 8.92 1 946 14.57 1 693 19.41 2 432 17.10 8 2 550 10.10 1 955 15.97 1 637 20.28 2 343 18.06 5 2 413 11.42 2 421 13.33 1 500 22.37 2 157 20.18 2 2 178 13.47 2 174 13.58 1 239 24.15 1 780 22.87 1 2 001 13.95 2 181 16.28 1 062 25.18 1 565 23.78

0,5 1 704 12.89 2 078 16.60 809 23.67 1 241 24.40

� However, comparison of complex dynamic modulus values to st iffness (IT-

CY) is very difficult or closely impossible.

Complex modulus – possible limits of testing

Complex modulus – what can be concluded?

� In general gained results very well demonstrate the thermal dependenceof the material and the applicability of linear viscoelastic theory. In term oftemperature effects even the thermal susceptibility can be demonstratedby the master curves.

� The transition from viscoelastic to elastic behavior is evident from the lastmeasured values in the decreasing imaginary part of the master curves.measured values in the decreasing imaginary part of the master curves.

� If we compare the mixes in terms of used binders, the influence of cementis clearly evident respectively it enhances elastic properties at hightemperatures and low frequencies. By the application of cement, thethermal susceptibility is decreased.

� The difference between foamed bitumen and bituminous emulsion is clearespecially in the elastic part of resulting master curves.

� Comparison of complex dynamic modulus values to stiffness (IT-CY) isvery difficult or closely impossible.

Complex modulus – what can be concluded?

� First problem related to cold recycled asphalt mixes and 4PB-PR testingappeared already during demoulding of the test slabs. Because thematerial is a bit more brittle, it is necessary to secure sufficient separationbetween the steel bottom plate of the mould and the compacted asphaltmaterial.

� Another problem occurred when beam specimens were cut from the slabs� Another problem occurred when beam specimens were cut from the slabsto get the necessary test specimens. During the cutting the material hastendency to brake off edges of the beams This finding is not unique.

Fatigue behaviour

Fatigue behaviour – the general practice

� Important characteristic for asphalt mixes, but so far not tested for coldrecycled mixes.

� Related to long term material behavior and cracking failure of thepavement.

� Crack initiation occurs when the specimen is subjected to repetitiveloading, which elicits stresses lesser than the strength of materialloading, which elicits stresses lesser than the strength of material

� Standard fatigue testing specified by EN 12 697 standard for differentmodes (controlled stress or controlled strain):

� 2PB test method,

� 4PB test method,

� Indirect tensile fatigue test on cylindrical specimens.

� Resistance to fatigue is observed as a change in the stiffness modulus ofasphalt mixtures to a certain critical value.

Fatigue behaviour – approach and selected test method� Based on experience with complex modulus testing 4PB (or even 2PB)

test method was seen as hardly applicable.

� Additionally such testing would be limited to narrow range of laboratories.

� ITFT test selected as the most suitable test.

� Among the concept of 50% loss in modulus the properties are studied with� Among the concept of 50% loss in modulus the properties are studied withthe help of linear viscoelasticity and Weibull analysis.

� Several research works showed that large amount of permanentdeformation is accumulated, thus the test method is not considered aspurely fatigue analysis.

� Impossibility of performing the test in the controlled strain mode.

Nevertheless at this moment the only test method which allow edreliable assessment of fatigue behavior.

Fatigue behaviour – principles of testing

� Cylindrical specimen with 150 mm diameter and 60 mm thickness issubjected to vertical loading pulse where the rise time is 50 ms.

� Resulting strain is measured in both vertical and horizontal direction.

� For the fatigue life the maximal strain is calculated according to ČSN EN12697-24+A1 2007.

� The load condition during the ITFT test includes 0.5 s rest periods, therepetition pulse period is 0.6 s.

� Test temperature 10°C.

� The Poisson’s ratio, or rather called lateral contraction ratio, was selectedequal to 0.23.

� reference number of cycles equal to one million determined as anequivalent strain from the Wöhler curve.

Fatigue behaviour – principles of testingData record and data use

� Data recorded in National Instruments TDMS data format.

� Converted by LabVIEW to space delimited text format .

� Then data imported into Microsoft Excel spreadsheet with preprogramedVisual Basic (VBA) modules that were used for data analysis.Visual Basic (VBA) modules that were used for data analysis.

� Data were fitted with the use of nonlinear solver aiming subsequently boththe maximal correlation coefficient and minimal residuals.

� The strain and stress amplitudes, its position in time and permanentdeformation were consequently derived for each cycle and used in thefurther analysis

Fatigue behaviour – principles of testingTest criteria – loss modulus

� Results of 4PB tests and axial compressive and tensile tests can beplotted on a logarithmic scale as the dependence of the number of cyclesto stress or strain.

� Those data, plotted in a logarithmic scale, can be approximated by astraight line known as the Wöhler curve or SN curve. The number ofstraight line known as the Wöhler curve or SN curve. The number ofcycles where the stiffness is half of the initial stiffness is chosen as thevalue representing resistance of material to fatigue.

Fatigue behaviour – resultsBSM-BE

Epsilon6 value

Fatigue behaviour – resultsBSM-FB

Epsilon6 value

Fatigue behaviour – conclusions

� The BSMs exhibits very abrupt decrease of stiffness during the fatiguetesting. The loss of the stiffness cause very low fatigue life when assessedby 50% modulus loss concept.

� The other approaches determine the fatigue life between 50% lossmodulus concept and the fracture.

� The Hopman and Pronk method seems to be most suitable for studied� The Hopman and Pronk method seems to be most suitable for studiedBSMs. The Hopman and Pronk method presents the best ability todetermine interface between crack initiation and propagation (goodcorrelation coefficients).

� All of the test and analytical methods show very high deviation.

� BSM-FB fatigue resistance has very high variability. This is contributed tothe heterogeneity of RAP coating when the foam binder productiontechnology is used.

� The variability of test results may be implemented in the pavement designmethodologies with the use of partial variance coefficient.

Low temperature cracking

Cracking – is this an issue for cold recycled mixes?

� cracking is for a long time and well known complex issue;

� important for colder regions but also where large temperature gradientscan be expected during the year;

� bitumen related aspects => thermal susceptibility and ageing.

Cracking – test review

� SCB is an acronym for Semi-Circular Bending test, which is used as adestructive test for assessment of resistance to crack origin andpropagation. The test follows the standard EN 12697-44: Crackpropagation by semi-circular bending test.

� The standard focuses on hot mix asphalt but the test might be applicable

SCB test

� The standard focuses on hot mix asphalt but the test might be applicableto cold recycled mixes as well.


� The test specimen is put directly on cylinders, which are not attached in itsaxes, but the horizontal shift is enabled. The loading is not provided withconstant vertical strain, as defined in EN 12697-44, but it is provided withconstant growth of the crack mouth. The strain velocity is defined by 0.5mm/min (controlled by crack mouth opening displacement - CMOD)

SCB test (U.S.)

mm/min (controlled by crack mouth opening displacement - CMOD)measuring equipment. The test equipment is provided also by additionalsensors.


� The principle of the test is to determine deformation characteristics of abituminous mixture at low temperatures (usually in the range of 0°C to -20°C) by flexural three-point strength test. For the test beam specimensare used and they are loaded by three point bending.

� The basis for the FS test performance is a standard press apparatus. TP

Flexural strength test (3 PB test)

� The basis for the FS test performance is a standard press apparatus. TP151 requires its maximum force to be at least 100 kN and ability to keepconstant strain velocity of between 1.25 mm/min and 50 mm/min


� The best description of this test is its name itself - uniaxial tension stresstest. The test procedure is described in EN 12697-46. It can be found inanother technical literature or reports under a different name, e.g. directtension test.

UTST test

� The whole name of TSRST is thermal stress restrained specimens tensiletest. It is sometimes branded cooling test. In Europe the test is defined bythe harmonized standard EN 12697-46. It is possible to find the test also inAASHTO TP 10-93.

TSRST test


� Disc-shaped compact tension test (DCT) was developed in the USAbetween 2008 and 2012 as a practical method for determination of lowtemperature behavior of bituminous mixtures. By several authorities in theUSA this test was accepted as suitable for testing of the given parameter.

DCT test

Cracking – tested materials/mix options

� BCSM-BE: Bitumen-Cement Stabilized Material – Bitumen Emulsion

� BCSM-FB: Bitumen-Cement Stabilized Material – Foamed Bitumen

� BSM-BE: Bitumen Stabilized Material – Bitumen Emulsion

� BSM-FB: Bitumen Stabilized Material – Foamed Bitumen.

Standard cold recycled mixes

� BSM-FB: Bitumen Stabilized Material – Foamed Bitumen.

� Mix design of mixes with reduced binder contents

� Cold recycled mixes containing recycled concrete

� Containing recycled concrete and cement or pulverized concrete

� Job site mix.

Standard cold recycled mixes

Cracking – tested materials/mix options

Mix designation Mix E Mix K Mix S

RAP 0/22 93.0 % 95.5 % 95.0 %

Water 2.5 % 2.0 % 2.0 %

Bituminous emulsion C60B7 3.5 % - -

Foamed bitumen - 2.5 % 2.0 %

Cement CEM II B32.5R 1.0 % - 1.0 %

Mix designation Mix DA Mix DE Mix DB Mix DO

Recycled concrete 0/22 44.75 % 45.75 % 68.625 % 43.75 %

Sand 44.75 % 45.75 % 22.875 % 43.75 %

Water 4.0 % 4.0 % 4.0 % 4.0 %

Bituminous emulsion 3.5 % 3.5 % 3.5 % 3.5 %

Cement 3.0 % 1.0 % 1.0 % -

Pulverized concrete - - - 5.0 %

Cracking – specimen preparation

� 150 mm cylindrical specimens compacted by static pressure

� Curing according to recommendations of CoRePaSol project

� Additional curing for 28 days

SCB test

Cracking – specimen preparation

� Manufacturing of beam specimen follows EN 12697-33+A1.

� Using segment compactor and after curing slabs were cut to required shapeof beam specimens according to EN 12697-26

� Curing regime I: Cold recycled mixes stabilized by cement and bituminous

FS test

� Curing regime I: Cold recycled mixes stabilized by cement and bituminousemulsion or foamed bitumen which were stored two days at 90-100 %relative humidity and temperature of (20±2) °C. Further the test slabswere stored in dust free area in the laboratory at 40-70 % relative humidityfor additional 26 days at the same temperature.

� Curing regime II: Cold recycled mixes stabilized by bituminous emulsion orfoamed bitumen for 1 day at 90-100 % relative humidity and temperature of(20±2) °C. Further the test slabs were stored at (50±2) °C in a climaticchamber for additional 4 days. The test specimens were then stored at 40-70 % relative curing and the temperature of (20±2) °C for 14 days afterthis accelerated curing.

Cracking – results

� Several types of crack propagation occurred.

� OPTION I: crack was almost ideally straight.

� OPTION II: crack propagated in required area,

� OPTION III: Crack half in and half out the allowed area. This kind of

SCB test

� OPTION III: Crack half in and half out the allowed area. This kind ofbehaviour is caused by big aggregate particles, which lie right above thenotch, so the crack has to run around. The aggregate coherence is not thatbig and the crack cannot run through the bigger particles. Measured valueswith this behaviour were included to evaluation.

� OPTION IV: crack is searching areas in the mix with the lowest stressresistance.


SCB test


SCB test

� Fracturaltoughness andcomparison of 4days or 14 dayscuringcuring


SCB test

� Fractural toughness after 28 days curing


SCB test

� Time dependence of fracture toughness values


FS test


FS test

Cracking – conclusions� Crack propagation (SCB) test according to EN 12697-44 is suitable for cold

recycled mixes. Similarly flexural strength test which is not standardizedwithin Europe and was done according to Czech technical specifications TP151 is applicable to cold recycled mixes.

� It is recommended to motivate and support within Europe further researchin the field of more advanced tests, e.g. TSRST test according to EN12697-46, to be assessed for cold recycled mixes.

� There might be some limitations with respect to use one of the tests inrelation to applied mix designs. In general none of the tests will be suitablefor cold recycled mixtures with low bituminous binder content (≤ 2.5 %) andhydraulic binder content (≤ 1.0 %. In such cases cracking might be even ofvery low importance (similarly to granular base layers).

� It is recommended to carry out selected tests always on specimens whichhave at least 28 days curing period. In this case it is possible to avoid anydeterioration of the specimens especially during the test preparation andspecimen manipulation.

Cracking – conclusions� In case of flexural strength test it is recommended to proceed the test with

1.25 mm/min loading velocity on specimens with cross sectional dimensionsof 50x50 mm at -5°C. The minimum flexural strength value should be 1.0MPa or 1.1 MPa.

� The crack propagation test according to EN 12697-44 should be done onspecimens of 150 mm diameter which are cut in two semi-circularspecimens. A loading velocity of 5 mm/min has to be applied and the testshould be carried out at least for the temperature of 0°C. Proposedminimum fractural toughness value at this temperature is 12 N/mm3/2 forcold recycled mixes with >3 % bituminous emulsion or foamed bitumen and>1 % cement. If the cold recycled mix contains 2.5-3.0 % of bituminousemulsion or foamed asphalt minimum fractural toughness can be reducedto 9-10 N/mm3/2.

� Proposed values have to be understood as preliminary and it is highlyrecommended to continuously collect data for cold recycled mixes in thefield of crack propagation and behavior in low temperature range for aperiod of at least 5 years.

And the possibilities for And the possibilities for future...

Further functional tests – the possible future � Despite of some disadvantages, the monotonic triaxial test that determines

shear parameters of granular mixtures could be one of the suitable tests toclassify this type of material, particularly if the mixture is characterized bylower content of bituminous and/or hydraulic binder.

� The idea of creating a modified triaxial device for BSM testing is based onthe work of JENKINS and his team for more than 15 years.

� In the second step maybe even cyclic triaxial test that determines theresilient modulus Mr (in case of short-term tests) or permanent deformationεp (in case of long-term tests.

Mohr-Coulomb Failure Criterion Principle

Further functional tests – the possible future

Further functional tests – the possible future

Mix D Mix C

Department of Road Constructions

Faculty of Civil Engineering in Prague

[email protected]

[email protected]

www.corepasol.eu

Durability Related Characterization of Cold Recycled Mixes ...

Documents