WHRP Year in Review - Wisconsin Concrete Pavement ... • Wisconsin Highway Research Program • WisDOT (Jim Parry, Barry Paye, Chad Hayes, Myungook Kang, Heidi Noble) • WCPA (Kevin

• These research projects were funded through WisDOT and the USDOT. Neither WisDOT nor the USDOT assume any liability for the contents or the use thereof nor does this presentation reflect official views, policies, standard specifications or regulations of either department.

Disclaimer

1

Acknowledgements

• Wisconsin Highway Research Program• WisDOT (Jim Parry, Barry Paye, Chad Hayes,

Myungook Kang, Heidi Noble)• WCPA (Kevin McMullen)• Industry contributors and partners• UW Madison staff and students

2

Better Concrete Mixes for Rapid Repair in Wisconsin

Project 0092-15-08Le Pham, Research AssistantSteven Cramer, PhD, PEMark Snyder, PhD, PE

University of Wisconsin-MadisonDepartment of Civil and Environmental Engineering

July 2017

Research Objectives

• Provide a basic inventory -materials, and procedures

• Evaluate performance of recent rapid-repair pavements in Wisconsin

• Identify high-quality rapid-repair mixtures

• Analyze life-cycle costs of cast-in-place vs precast concrete pavement 4

Field Review - Project Locations

5

Project # Completion year Road1 2011 STH 292 2011 STH 293 2013 US 514 2010 US 1415 2011 US 456 2010 STH 237 2014 I 39/908 2013, 2014 I 39/90

9 multiple generations I 90/94

10 multiple generations STH 26

11 2010 US 12/1812 2013 STH 96

Field Review

6

• No serious durability issue except for one project (#7 - scaling)

• Many pavement patches were ground

• No correlation between field performance and project records could be established

Informal Survey

7

• Mix design:– No one used type III cement– Most used 9-bag type I cement

with calcium chloride accelerator to achieve 3000 psi in 8 hrs

– Some used heated water, HRWR, non-chloride accelerator

– Some advocated for the use of structural fibers

• 4 rapid repair concrete suppliers in Wisconsin• 15-20 minute phone discussion

Informal Survey

8

• Challenges:– Restricted ability to make adjustments

for air entrainment – Rapid hydration and concrete setting up

in the truck before discharge – Very tight timing between travelling to

site, discharging the load and achieving strength

– Profitability concerns – No known durability issue

Laboratory Test Matrix

9

Mix # Coarse Aggregate w/c Slump Accelerator

Curing scheme for C672

specimens

1 Crushed Limestone 0.32 Normal None Standard(a)

2 Crushed Limestone 0.32 Normal CaCl2 solution Standard

3 Crushed Limestone 0.32 Normal CaCl2 dry Standard

4 Crushed Limestone 0.32 Normal Non-chloride Standard

5 Igneous gravel 0.32 Normal None Standard

6 Igneous gravel 0.32 Normal CaCl2 solution Standard

7 Igneous gravel 0.32 Normal CaCl2 dry Standard

8 Igneous gravel 0.32 Normal Non-chloride Standard

9 Igneous gravel 0.32 Normal CaCl2 solution Short(b)

10 Igneous gravel 0.32 High CaCl2 solution Short

11 Igneous gravel 0.36 Normal CaCl2 solution Short

12 Igneous gravel 0.36 High CaCl2 solution Short

13 Igneous gravel 0.32 Normal CaCl2 solution Curing compound(c)

• Compressive strength • Drying shrinkage• Scaling resistance

(C672)

• standard procedures• 3 samples per set

Tests

10

Concrete Strength Results

11

• 8 hrs to 3000 psi? - All mixes using CaCl2 and w/c ratio of 0.32

• Dry flakes of CaCl2 (vs liquid form) equally effective for strength development

• 10 hrs to 3000 psi - Non-chloride accelerator• 14-20 hrs to 3000 psi – 9-bag mix without

accelerator

Effect of Accelerators on Strength

12

Effect of Accelerators on Strength

13

Drying Shrinkage Results

14

• Addition of CaCl2 significantly increased drying shrinkage, by 40-240%*

• Addition of non-chloride accelerator slightly increased drying shrinkage, by 8-18%*

*(depending on age of concrete)

Scaling Resistance at 60 cycles

15

Scaling Resistance – Effect of Curing

16

• Rapid slump loss (within 45 minutes), especially in mixes using calcium chloride.

• Strength development temperature sensitive• Curing samples in insulated box increased strength at 8

hrs by 30% compared with wet-room curing.

• Low-range water reducer is counter productive – significantly delayed setting and hydration.

Other Observations

17

Economic Analysis Results

18

• Among tested mixtures: benefits of using longer-lived materials far exceed those of using materials with short expected service lives

• CIP vs precast: CIP pavements would have life-cycle costs comparable to precast pavements that have service lives about 3 times longer (given the cost data)

Conclusions

• No serious widespread durability issues found in field

• No serious durability issue associated with the materials in the laboratory tests.

• Mixtures using type I portland cement can surpass WisDOT strength requirements and have satisfactory scaling resistance.

19

Conclusions

• Addition of CaCl2 significantly increased drying shrinkage, a potential for cracking

• Mixtures using CaCl2 in dry form surpassed WisDOT strength requirement and exhibited satisfactory scaling resistance.

• Mixtures using non-chloride accelerator had slower strength development and smaller drying shrinkage than those using CaCl2.

20

Conclusions

• Construction and policy constraints appear to be the biggest difficulties in obtaining high quality rapid-repair pavement.– Strict time constraints drive use of CaCl2 which in

turn potentially leads to other mix and pavement issues

– The process is not particularly robust where the unexpected likely causes compromises

21

Recommendations

• To alleviate difficulties in construction of rapid repair pavement:– Consider higher slump limits (6”) for rapid-repair concrete mixtures

(current limit is 4”).– Consider the use of CaCl2 in dry form.– Consider extending time to 3000 psi to allow non-chloride accelerators

• What’s going on with PAMS curing compound and CaCl2?

• Precast a viable alternative?22

Comparison of Fresh Air Content Test Methods and Analysis of Hardened Air Content in Wisconsin Pavements Project 0092-14-05

Le Pham, Research AssistantSteven Cramer, PhD, PE

University of Wisconsin-MadisonDepartment of Civil and Environmental Engineering

2018

Summary of Phase 1 – field study

• Cores from pavements– significantly higher air

contents than pressure meter or sample cylinders made from concretes before slip-form paver

– synthetic AEAs were used

24

Phase 2 – laboratory study

• The goal is to identify the mechanisms causing discrepancy

• Hypothesis: air bubbles entrained with a synthetic AEA are less resilient than those entrained with NVR

25

Phase 2 - tests

• Hardened concrete air void analysis (ASTM C457)

• Foam drainage test• Atomic force microscopy

and nanoindentation of air bubbles – Yes! we are testing

individual and pairs of air bubbles

26

• Mixture: typical for slip-form paver– Cement = 395 lb, fly ash = 170 lb, w/cm = 0.36– 4.5 – 7.5 % air content (pressure method)– 1 – 3 in. slump– ¾-in. crushed stone coarse aggregate

• ASTM C457 air void analysis– Unconsolidated concrete prism– Hand-rodded 6 x 12 in. cylinder

Concrete tests

27

Air content of voids < 1mm

28

AEA Batch # Rodded UnconsolidatedRelative air

loss

Average relative air

loss % % % %

SYN_1 1 4.9 5.9 17%18%SYN_1 2 4.9 6.0 18%

SYN_1 3 4.8 6.0 20%SYN_1 Prelim 1 5.8 7.5 23%SYN_1 Prelim 2 4.9 5.5 11%SYN_2 1 3.6 4.8 25%

22%SYN_2 2 4.2 4.9 14%SYN_2 3 4.4 6.0 27%NVR 1 3.5 3.9 10%

7%NVR 2 6.4 6.6 3%NVR 3 8.0 8.7 8%NVR Prelim 1 4.6 4.8 4%

Foam Drainage Test - Background

• Measures time of water drainage from a foam made by mixing water and an AEA

• Might have link with stability of air bubbles in concrete: Cross et al. (2000), Taylor et al. (2006), Taylor et al. (2015)

• Correlations with field or lab performance of air bubbles in concrete not available 29

FDT – Original protocolCross et al. (2000), Taylor et al. (2015)

• Mix 290 ml water with 10 ml AEA in a blender for 10s, pour the mixture into a 1-liter graduated cylinder, record time (t) & volume of liquid drained (Vd) to bottom of the foam within 60 min.

• Indicators:– V0: volume of drained liquid after t = ∞– (1/k): indicates rate of drainage– Higher (1/k) and lower V0 mean more stable foam30

Foam Drainage Test

31

Drained liquid after 3 min. (left) & 1 hr. (right)

FDT Original protocol

• Cannot tell which mixture was the best/worst based on V0 and (1/k).

• (1/k) was an incorrect indicator for NVR w/ cement.

32Vd = Vo – 1/(k × t)

Mix 1/k Vo R2

Synthetic w/o cement 1.71 0.85 0.86NVR w/o cement 0.76 0.96 0.98

Synthetic w/ cement 0.77 0.78 0.74NVR w/ cement 0.58 0.54 0.79

FDT – Original protocol

33

• Issues:Using 2 indicators often leads to inconclusive results. – Neither V0 nor (1/k) alone is a proper indicator of

drainage for the mixtures tested– Using a fixed amount of AEA is not representative

of the practice of concrete because AEAs have different foaming capacities.

FDT – Proposed changes

34

• Fix the volume of the foam instead of the admixture

• Use only one indicator: time at which 75% of the liquid drains (t75). Higher t75 means more stable foam.

FDT – Revised protocol

35

FDT – Revised protocol

36

FDT–Revised protocol

37

No cementAEA t75 (min.) RankNVR 6

SameSYN_1 6SYN_2 6

With cementAEA t75 (min.) RankNVR infinitely large 1

SYN_1 30 2SYN_2 8 3

• Without cement, the foams with synthetic and NVR AEAs had similar drainage rates

• With cement, the synthetic AEA foam drained much faster than the NVR, an indication that synthetic bubbles are less resilient

FDT– Revised protocol

38

Summary

39

• Correlation was found between the revised FDT and concrete air void tests regarding performances of different AEAs

• FDT and ASTM C457 results of laboratory concrete appeared to support the discrepancy in air content found in Phase 1

What’s next?

40

• Why were synthetic AEA entrained bubbles less resilient than NVR?– Fundamental tests of air bubbles (ongoing)– Theoretical analysis of bubble behaviors

(ongoing)• Deliverables:

– A simple test for practitioners to evaluate AEAs used in concrete

– A fundamental test for researchers/developers of new AEAs

THANK YOU!

41