BDV31 977-136 Final_Report.pdf

FINAL REPORT

Title: Testing Methods for the Next Generation of Concrete FDOT Contract Number: BDV31-977-136

Submitted to

The Florida Department of Transportation Research Center 605 Suwannee Street, MS 30 Tallahassee, FL 32399 c/o Tim Ruelke FDOT State Materials Office

Submitted by:

Dr. Christopher C. Ferraro (ferraro@ce.ufl.edu) (Principal Investigator) Dr. Kyle A. Riding (Co-Principal Investigator) Dr. Jerry M. Paris (Co-Principal Investigator)

Dr. Caitlin M. Tibbetts (Co-Principal Investigator) Engineering School of Sustainable Infrastructure and Environment

University of Florida Gainesville, Florida 32611

March 2021 Department of Civil Engineering Engineering School of Sustainable Infrastructure and Environment College of Engineering ii

DISCLAIMER

The opinions, findings, and conclusions expressed in this publication are those of the authors and not necessarily those of the State of Florida Department of Transportation or the U.S. Department of Transportation.

Prepared in cooperation with the State of Florida Department of Transportation and the U.S.

Department of Transportation.

iii

SI (MODERN METRIC) CONVERSION FACTORS (FROM FHWA)

Symbol When You Know Multiply By To Find Symbol

Length in inches 25.4 millimeters mm ft feet 0.305 meters m yd yards 0.914 meters m mi miles 1.61 kilometers km

Area in2 square inches 645.2 square millimeters mm2 ft2 square feet 0.093 square meters m2 yd2 square yard 0.836 square meters m2 mi2 square miles 2.59 square kilometers km2

Volume fl oz fluid ounces 29.57 milliliters mL gal gallons 3.785 liters L ft3 cubic feet 0.028 cubic meters m3 yd3 cubic yards 0.765 cubic meters m3

NOTE: volumes greater than 1000 L shall be shown in m3

Mass oz ounces 28.35 grams g lb pounds 0.454 kilograms kg

Temperature (exact degrees)

°F Fahrenheit 5 (F-32)/9 or (F-32)/1.8

Celsius °C

Illumination fc foot-candles 10.76 lux lx fl foot-Lamberts 3.426 candela/m2 cd/m2

Force and Pressure or Stress lbf pound-force 4.45 newtons N lbf/in2 pound-force per square inch 6.89 kilopascals kPa iv

TECHNICAL REPORT DOCUMENTATION

1. Report No. 2. Government Accession No. 3. Recipient's Catalog No.

4. Title and Subtitle

Testing Methods for the Next Generation of Concrete

5. Report Date

February 2021

6. Performing Organization Code

7. Author(s)

Christopher C. Ferraro, Kyle A. Riding, Jerry M. Paris, Caitlin M. Tibbetts

8. Performing Organization Report No.

9. Performing Organization Name and Address

Department of Civil and Coastal Engineering Engineering School of Sustainable Infrastructure & Environment University of Florida 365 Weil Hall P.O. Box 116580 Gainesville, FL 32611-6580

10. Work Unit No.

11. Contract or Grant No.

BDV31-977-136

12. Sponsoring Agency Name and Address

Florida Department of Transportation 605 Suwannee Street, MS 30 Tallahassee, FL 32399

13. Type of Report and Period Covered

Draft Final Report September 2020 – February 2021

14. Sponsoring Agency Code

15. Supplementary Notes – N/A

16. Abstract

The Florida Department of Transportation is working towards the evaluation of structural portland cement concrete to ensure the service life requirements for concrete structures are met based on material performance. As the industry moves toward performance specifications, there is a need to evaluate the performance of the composite concrete sample, rather than the individual components. Therefore, the objective of this project was to perform a thorough literature review to determine if there are existing test methods that can quantify the performance of concrete with respect to heat evolution, cracking, and durability at the mix design phase. Based upon the exhaustive literature search performed, there are a large number of tests available to evaluate composite concrete specimens for heat evolution, cracking potential, and durability. The majority of these test methods did not meet the criteria of FDOT without any modifications; the few methods that met all of the qualifications do not provide sufficient information to be recommended for FDOT quality assurance/quality control (QA/QC). Recommendations for a future study to investigate these methods further and determine acceptance criteria is as follows. To evaluate potential for cracking of concrete, a Florida Method based on AASHTO T 334 should be created. Additionally, a Florida Method based on the LCPC QAB semi-adiabatic calorimetry test should be developed to measure heat evolution of concrete. Lastly, characteristics regarding chloride and sulfate durability require fundamental research into the accelerated curing regimes, especially for concrete using SCMs prior to a Florida Method being developed to assess these characteristics.

17. Key Word

Concrete, heat generation, cracking, chloride, sulfate, durability, transport property

18. Distribution Statement

No restrictions.

19. Security Classif. (of this report)

Unclassified.

20. Security Classif. (of this page)

Unclassified.

21. No. of Pages

22. Price v

ACKNOWLEDGEMENTS

The Florida Department of Transportation (FDOT) is acknowledged for their funding and contributions to this study.

vi

EXECUTIVE SUMMARY

Background The Florida Department of Transportation is working towards the rapid evaluation and qualification of portland cement concrete to ensure the service life requirements for structures are met based on material performance. As the industry moves toward performance specifications, there is a need to evaluate the performance of the composite concrete sample, rather than the individual components. Therefore, the objective of this project was to perform a thorough literature review to determine if there are existing test methods that can quantify the performance of concrete with respect to heat evolution, cracking, and durability at the mix design phase. Test methods were evaluated based on the following criteria:

Testing will be conducted on conventional portland cement concrete – not mixture components, derivatives, or specialty mixes (e.g., paste, mortar, grout, UHPC, RCC).

The time for completion does not exceed 30 days for testing.

The testing techniques do not utilize any hazardous materials.

Tests are relatively uncomplicated – a person with a high-school education can perform.

Cost of testing equipment should support a wide deployment – total cost not to exceed $30,000.00.

The project reviewed the available literature related to testing of concrete to assess potential for excessive shrinkage, potential to experience excessive temperature rise, and chloride and sulfate durability. The research team categorized identified test methods into categories as follows:

Category I – Test meets current needs of FDOT without any modifications.

Category II – Test can be modified to successfully meet FDOT needs.

Category III – Test shows some potential to meet FDOT needs, but does not meet one of the main criteria listed above.

Main Findings Based upon the exhaustive literature search performed, there are a large number of tests available to evaluate composite concrete specimens for heat evolution, cracking potential, and durability.

The majority of these test methods did not meet the criteria of FDOT without any modifications;

the few methods that met all of the qualifications do not provide sufficient information to be recommended for FDOT quality assurance/quality control (QA/QC).

Research Significance There are a large number of test methods that have been developed over the last 100 years for determining heat evolution, cracking, and durability; however, there is not a standard testing regimen that meets the criteria outlined above. Modifications of existing standards and development of new methods could provide expedient qualification of concrete mixtures, if these new methods prove effective.

vii

Conclusions Based on the literature reviewed, the following conclusions were drawn from this study:

The best available testing method for determining cracking is the Standard Method of Test for Estimating Cracking Tendency of Concrete as per AASHTO T 334. The method has a provision that allows for a standard size pipe to be used to construct the mold; if this provision is disallowed (or the rest of the mold is made bigger) to allow a 3-in.

minimum specimen thickness, this specification would meet all of the FDOT criteria.

The best available testing method for determining the heat production and resultant temperature rise of concrete is the LCPC QAB semi-adiabatic calorimetry test method.

The test provides the most robust data due to the large specimen size; however, modifications to allow for larger sizes would need to be developed prior to full adoption by FDOT.

The chloride and sulfate durability testing methods require further evaluation and development prior to being accepted with FDOT criteria; the largest barrier to acceptance is the 30-day requirement in combination with the ability to test concrete mixes with supplementary cementitious materials.

Some of the standard testing methods summarized in Table 39 may meet all of the FDOT criteria but do not give adequate testing information based on the available literature.

o Methods for semi-adiabatic heat rise meet all of the criteria, but the data obtained are insufficient for a quality control program.

o Methods regarding unrestrained shrinkage similarly meet FDOT requirements, but the amount of information is limited to one or two forms of shrinkage, and is not directly applicable to reinforced or restrained scenarios, which comprise the majority of FDOT applications.

Recommendations Based upon the findings from this study, the following recommendations are made:

Develop and create precision statement for a Florida Method based on AASHTO T 334 that does not allow for a standard size pipe to be used to construct the mold; using a standard size pipe restricts the specimen size and thus the maximum aggregate size.

Develop and create precision statement for a Florida Method based on the LCPC QAB semi-adiabatic calorimetry test method to allow for larger specimen sizes to ensure QA/QC at the design mix stage.

For chloride and sulfate durability, fundamental research into the evaluation of accelerated curing regimes is required, especially for concrete using SCMs.

o Further research should include investigation of the standard and accelerated curing methods prescribed in the Standard Practice for Developing Engineered Concrete Pavement Mixtures (AASHTO PP 84) for the acceptance of FDOT concrete. Currently, there is active research being performed with regards to accelerated curing methods as prescribed the Standard Test Method for Electrical Resistivity of a Concrete Cylinder Tested in a Uniaxial Resistance Test viii

(AASHTO TP 119) and the Standard Method of Test for Determining the Pore Volume in Hardened Concrete Using Vacuum Saturation (AASHTO TP 135).

Benefits to the State The literature review provides guidance for the rapid and robust qualification of concrete mixtures for FDOT-owned structures. This includes guidance with respect to the development of Florida Methods based on the test methods investigated for heat evolution, cracking potential, and durability. Modifications of existing standards and development of new methods provide a pathway towards the expedient qualification of concrete mixtures, especially those utilizing supplementary cementitious materials, if proven effective. Rapid and robust testing methods are critical for the efficient qualification of concrete mixtures, which ultimately leads to reduced repair and maintenance costs FDOT-owned structures.

ix

TABLE OF CONTENTS

DISCLAIMER ................................................................................................................................ ii

SI (MODERN METRIC) CONVERSION FACTORS (FROM FHWA) ..................................... iii

TECHNICAL REPORT DOCUMENTATION ............................................................................ iv

ACKNOWLEDGEMENTS ............................................................................................................ v

EXECUTIVE SUMMARY ........................................................................................................... vi

LIST OF FIGURES ...................................................................................................................... xii

LIST OF TABLES ....................................................................................................................... xiv

1. INTRODUCTION

1.1. BACKGROUND

1.2. RESEARCH OBJECTIVES

1.3. RESEARCH APPROACH

2. HEAT EVOLUTION TESTING

2.1. BACKGROUND

2.1.1. Adiabatic Temperature Measurement

2.1.2. Semi-Adiabatic Temperature Measurement

2.1.3. Isothermal Temperature Measurement

2.2. ADIABATIC TEST METHODS

2.2.1. USBR Method

2.2.2. RILEM Method

2.2.3. CRD-C Method

2.2.4. EN 12390 – Part 15

2.2.5. Existing Available Equipment

2.3. SEMI-ADIABATIC TEST METHODS

2.3.1. RILEM Method

2.3.2. LCPC QAB

2.3.3. NT Build 388

2.3.4. EN 12390 – Part 14

2.3.5. ASTM C1753

2.3.6. Existing Available Equipment

2.4. ISOTHERMAL TEST METHODS

2.4.1. ASTM C1679

2.4.2. Existing Available Equipment

x

2.5. SUMMARY

3. CRACKING TENDENCY TESTING

3.1. BACKGROUND

3.2. RESTRAINED SHRINKAGE

3.2.1. ASTM C1581

3.2.2. AASHTO T 334

3.2.3. AASHTO T 363

3.2.4. Cracking Frame

3.2.5. NT Build 433

3.2.6. ASTM C1579

3.3. UNRESTRAINED SHRINKAGE

3.3.1. ASTM C157/AASHTO T 160

3.3.2. EN12390 – Part 16

3.3.3. AASHTO T 336

3.3.4. ASTM C827

3.3.5. Existing Available Equipment

3.4. OTHER METHODS

3.4.1. German Angle Test

3.4.2. SPS Plate Test

3.4.3. ASTM C596

3.4.4. ASTM C1090

3.5. SUMMARY

4. DURABILITY TESTING

4.1. BACKGROUND

4.1.1. Chloride Durability

4.1.2. Sulfate Durability

4.2. CHLORIDE DURABILITY

4.2.1. ASTM C1202/AASHTO T 277

4.2.2. NT Build 492

4.2.3. AASHTO T 358

4.2.4. AASHTO TP 119

4.2.5. Chloride Binding

4.2.6. Existing Available Equipment

xi

4.3. SULFATE DURABILITY

4.3.1. USBR 4908

4.3.2. MNS Test

4.3.3. Soaking and Drying Test

4.3.4. Rapid Sulfate Electrochemical Test

4.3.5. Partial Immersion

4.3.6. Complete Immersion

4.4. TRANSPORT PROPERTY

4.4.1. AASHTO PP 84

4.4.2. EN 12390 – Part 8/CRD-C 48-92/ACCI water permeability

4.4.3. ASTM C1585

4.4.4. ASTM C642

4.5. SUMMARY

5. MAIN FINDINGS, CONCLUSIONS, AND RECOMMENDATIONS

5.1. MAIN FINDINGS

5.2. CONCLUSIONS

5.3. RECOMMENDATIONS

REFERENCES

xii

LIST OF FIGURES

Figure 1. General diagram of adiabatic calorimeter design adapted from RILEM TC 119

Figure 2. Plan view of insulated concrete specimen with location of RTDs per CRD-C 36-73. . 12

Figure 3. Sure Cure system with the computer, hydration chamber, and controller box by Products Engineering, Inc. [6]

Figure 4. Example of semi-adiabatic measuring device from RILEM TC 119-TCE

Figure 5. Example of semi-adiabatic measuring device used for LCPC QAB

Figure 6. Differences in maximum heat generation between an effectively uninsulated and well-insulated concrete specimen [19]

Figure 7. Relationship between autogenous and chemical shrinkage, adapted from [29]

Figure 8. General diagram of a concrete ring cracking test

Figure 9. Ring test setup for ASTM C1581

Figure 10. Initial layout of dual-ring test from AASHTO T 363

Figure 11. General testing setup for concrete cracking frame

Figure 12. Typical plot of cracking frame stress development with time and temperature

Figure 13. Typical plot of cracking frame stress development with time and temperature

Figure 14. Concrete panel mold specifications. Adapted from ASTM C1579

Figure 15. Example of a specimen cabinet that must be inside an environmental chamber to test two concrete panels in accordance with ASTM C1579

Figure 16. Schematic of a CTE setup as per AASHTO T 336

Figure 17. Test setup for ASTM C827

Figure 18. German angle test

Figure 19. SPS plate test setup

Figure 20. Diagram of ASTM C1202/ AASHTO T 277 specimen cell setup with rubber gaskets.

Figure 21. Picture of PROOVEit voltage application and data readout apparatus used for ASTM C1202/AASHTO T 277 and NT Build 492 xiii

Figure 22. Diagram of NT Build 492 test setup [86]

Figure 23. Diagram of four-point Wenner probe for surface resistivity test setup

Figure 25. Diagram of bulk resistivity test setup

Figure 24. Picture of Resipod surface resistivity apparatus by Proceq for AASHTO T 358

Figure 26. Diagram of typical casting mold used for USBR 4908 specimens

Figure 27. Comparison of w/cm to residual compressive strength, data from [108]. Control mixes have no SCM

Figure 28. Diagram of unsealed specimen for water penetration depth test per EN 12390-8 [118].

Figure 29. Diagram of water permeability test assembly for CRD-C 48-92 [120]

Figure 30. Diagram of University of Florida water permeability cell with concrete specimen cast in epoxy ring (similar to the ACCI water permeability test)

Figure 31. Diagram of partial water exposure for water absorption test ASTM C1585 [127] xiv

LIST OF TABLES

Table 1. Summary of required equipment and approximate costs using USBR 4911-92

Table 2. Summary of required equipment and approximate costs using RILEM TC 119-TCE. . 10

Table 3. Differences between CRD-C 38-73 and USBR 4911

Table 4. Summary of required equipment and approximate costs using CRD-C 38-73

Table 5. Summary of required equipment and approximate costs using RILEM TC 119-TCE. . 18

Table 6. Summary of required equipment and approximate costs using LCPC QAB

Table 7. Summary of required equipment and approximate costs using NT Build 388

Table 8. Summary of criteria for heat evolution test methods

Table 9. Potential for cracking classification adapted from ASTM C1581

Table 10. Summary of required equipment and approximate costs using ASTM C1581

Table 11. Summary of technical differences between ASTM C1581 and AASHTO T 334

Table 12. Summary of required equipment and approximate costs using AASHTO T 334

Table 13. Summary of required equipment and approximate costs using AASHTO T 363 assuming one specimen is tested at a time rather than three

Table 14. Summary of required equipment and approximate costs using a basic cracking frame.

Table 15. Summary of required equipment and approximate costs using NT Build 443 with three specimens

Table 16. Summary of required equipment and approximate costs using ASTM C1579

Table 17. Summary of required equipment and approximate costs using ASTM C157 or AASHTO T 160 with six specimens

Table 18. Summary of required equipment and approximate costs using AASHTO T 336

Table 19. Summary of required equipment and approximate costs using ASTM C827

Table 20. Summary of criteria for cracking test methods

Table 21. Chloride ion penetrability classifications per ASTM C1202 / AASHTO T 277 [80], [81] xv

Table 22. Summary of required equipment and approximate costs using ASTM C1202 /

AASHTO T 2771

Table 23. Summary of required equipment and approximate costs using NT Build 492

Table 24. AASHTO T 358 chloride ion penetrability classifications based on surface resistivity values for 4 in. x 8 in. and 6 in. x 12 in. cylinders [87]

Table 25. Summary of required equipment and approximate costs using AASHTO T 358

Table 26. AASHTO TP 119 chloride ion penetrability based on uniaxial (bulk) resistivity values [95]

Table 27. Summary of required equipment and approximate costs using AASHTO TP 119

Table 28. Summary of required equipment and approximate costs using USBR 4908

Table 29. Summary of required equipment and approximate costs using the MNS test

Table 30. Summary of required equipment and approximate costs using the soaking and drying test (four – 2-in. cubes soaked, four 2-in. cubes not soaked, three – 6-in. cubes compression). .. 75

Table 31. Summary of required equipment and approximate costs using the rapid sulfate electrochemical test

Table 32. Summary of required equipment and approximate costs using the partial immersion sulfate exposure of concrete specimens proposed by Hartell et al. [78]

Table 33. Summary of required equipment and approximate costs using the complete immersion sulfate exposure of concrete specimens proposed by Park et al. [110]

Table 34. Summary of required equipment and approximate costs using the formation factor per

AASHTO PP 841

Table 35. Summary of required equipment and approximate costs using water permeability test methods

Table 36. Summary of required equipment and approximate costs using ASTM C1585

Table 37. Summary of required equipment and approximate costs using ASTM C642

Table 38. Summary of criteria for durability test methods

Table 39. Summary table of test methods and FDOT category classification

Table 40. Final summary table of experimental methods

1. INTRODUCTION

1.1. BACKGROUND

The Florida Department of Transportation (FDOT) is working towards the evaluation of cracking of portland cement concrete for use in FDOT-owned structures. Accordingly, the assessment of concrete at the design and qualification stages is needed to ensure the service life requirements for concrete structures are met based on material performance. There are a myriad of different test methods used to evaluate the components of concrete, which are intended for specifications that are mostly prescriptive in nature. However, as the industry moves toward performance specifications, there is a need to evaluate the performance of the composite concrete sample, rather than the individual components.

Early age shrinkage of concrete is a consequence of hydration wherein the initial volume of concrete is greater than the final volume of concrete as the concrete transitions from a plastic to an elastic material. This is generally known as “autogenous shrinkage” in hardened concrete where exterior water is unable to migrate towards the inner pore structure, and which is understood to occur without being subjected to external forces and under constant temperature [1]. Additionally, when evaporation of water from fresh concrete happens faster than the water can proceed to the surface through “bleeding”, “plastic shrinkage” will result and begin to produce inner tensile forces on the concrete before set. In either scenario, cracking may occur, resulting in immediately reduced durability through increased susceptibility to sulfate, chloride, and water ingress.

Concrete structures are adversely affected by chlorides and sulfates which may be present during mixing, borne from environmental exposure, or a combination of both conditions. With respect to sulfate degradation, internal sulfate attack (concrete degradation associated with sulfates present during mixing) is typically associated with high curing temperatures and dissolution of ettringite, forming monosulfate, which then latter reverts to ettringite forming needle-like crystals inside pores, inducing tensile forces. External sulfate exposure can result in more than one degradation mechanism. Damage to the cement paste through crystallization pressure (thenardite and mirabilite phase change) is also known as “physical salt attack” [2]. Whereas chemical degradation in which sulfate-bearing soils or water chemically dissolve cement hydrates (predominantly calcium aluminate phases) and form less soluble phases can produce expansive forces on the hardened paste [3]. Additionally, thaumasite can form as a conversion of ettringite (a normal byproduct of hydration) when exposed to carbonate from the atmosphere, from limestone aggregates within the concrete, or from carbonate soils that the concrete is in contact with.

The presence of chlorides in a cementitious system is generally broken into two main categories:

water-soluble (or available) chloride, and acid-soluble (or total) chloride. It is a general agreement that only water-soluble chlorides present within a cementitious system during hydration are able to freely move throughout the pore fluid system. The remaining chlorides present are considered to be “bound” within inaccessible phases and pose less likelihood for damage. In Florida, chlorides are a common concern for structures near marine environments where water-borne or air-borne chlorides produce a damage mechanism by external chloride attack. The main concern with either chloride origination is damage to the concrete structure due to the corrosion of reinforcing steel as the corrosion product of steel is approximately seven times more voluminous than the original steel itself.

The hydration of portland cement is an exothermic chemical reaction, where excessive heat generation can result in a number of durability concerns. As discussed previously, internal sulfate attack (also referred to as “delayed ettringite formation”) is considerably more likely to result in cracking of concrete when the concrete experiences temperatures exceeding 158°F (70°C) and the magnitude of deleterious expansion is exacerbated by continued thermal heat rise [4]–[6]. Due to the nature of the exothermic reactions, increased heat rise tends to make the hydration of portland cement accelerate; combined with thermal expansion of the concrete mass and the rate of thermal emissivity, this can differential stress internally in hydrating concrete where the inner core of the placement is considerably hotter, more hydrated (and therefore less plastic) while the outer surfaces can radiate heat away (leading to less maturity and higher likelihood of cracking) [7]. These compounding factors are more pronounced in placements with large volumes with relatively small exposed surface areas (predominantly drilled shafts, large footings, and bridge superstructure elements) collectively known as “mass” or “thermally controlled” concrete. These particular elements necessitate the use of additional control systems (temperature monitoring, thermal modeling for mixture design, precooled aggregates, post-cooled circulating water in embedded pipes, etc.) [8], [9].

With all of these concerns, various methods can be employed to mitigate or eliminate the likelihood of damage. The methods of mitigation often vary, but some methods such as the addition of supplementary cementitious materials such as fly ash or silica fume can result in reduction of potential for damage from multiple degradation mechanisms. However, the implementation of these materials also further complicate evaluation of performance due to their longer time to maturity.

1.2. RESEARCH OBJECTIVES

The objective of this project is to perform a thorough literature review to determine the existence of test methods that can qualify the performance of a portland cement concrete mixture with respect to cracking, durability and heat generation at the mix design phase. Preference will be given to test methods based on the following criteria:

Testing will be conducted on conventional portland cement concrete containing coarse aggregate – not mixture components, derivatives, or specialty mixes (e.g., paste, mortar, grout, UHPC, RCC)

The time for completion of testing does not exceed 30 days from casting

The testing techniques do not utilize any hazardous materials

Tests are relatively uncomplicated – a person with a high-school education can perform

Cost of testing equipment should support a wide deployment – total cost not to exceed $30,000.00

The project reviewed the available literature related to testing of concrete to assess potential for excessive shrinkage, potential to experience excessive temperature rise, and chloride and sulfate durability. The research team will categorize any successfully identified test methods into categories as follows:

Category I – Test method meets current needs of FDOT without any modifications.

Category II – Test method can be modified to successfully meet FDOT needs.

Category III – Test shows some potential to meet FDOT needs, but does not meet one of the main criteria listed above.

1.3. RESEARCH APPROACH

The research approach included a literature review of existing test methods that can quantify the performance of concrete with respect to heat evolution, cracking, and durability at the mix design phase. Chapter 2 documents the test methods found in the literature for heat evolution for adiabatic, semi-adiabatic, and isothermal calorimetric testing. Chapter 3 discusses restrained and unrestrained shrinkage test methods to evaluate the potential for cracking. Chapter 4 chronicles durability test methods with respect to chloride and sulfate penetrability as well as general transport characteristics. The five criteria outlined in the Research Objectives are summarized for each test method and then categorized based on the FDOT needs. Chapter 5 presents the conclusions and recommendations based on the findings herein.

2. HEAT EVOLUTION TESTING

2.1. BACKGROUND

Portland cement concrete is composed of a variety of materials that can be grouped as aggregates, water, reinforcement (whether it be steel reinforcement or fiber reinforcement), and cementitious/binder material. The aggregate, water, and reinforcement portions of the composition do not take part in hydration chemical reactions and therefore the cementitious content is responsible for the entirety of the heat generation of a concrete mix (ignoring environmental and initial conditions).

The hydration reaction of portland cement and water is exothermic in nature and therefore will produce heat as the hydration proceeds [1], [2]. This chemical reaction also tends to accelerate with thermal input [3, p. 207], and as concrete can be a thermal insulator, this can result in concrete self-insulating causing higher temperatures and accelerated hydration. In an unchecked situation, this can result in temperatures exceeding 200°F (93°C), and thermal cracking of the concrete element.

The factors that affect the heat evolution of a concrete mixture include cementitious material type and content, material fineness, material chemistry, initial placement/reaction temperature, water-to-binder ratio, and the thermal properties of the concrete constituents, amongst other factors. The heat rise of concrete directly impacts the rate of strength gain, form removal, thermal gradients, and can influence the cracking and durability. Higher temperatures generally result in higher early compressive strengths (but lower ultimate strengths), which allow for faster removal of formwork. However, hydrating cement exposed high temperatures typically results in less dense phase assemblages to form causing higher permeability and lead to reduced service life when chloride and sulfate exposure are concerns. Unregulated thermal rise in concrete can lead to high thermal gradients in which interior portions of concrete placements are much higher temperatures than exterior surfaces [2]. This differential temperature causes a thermal expansion incompatibility wherein the interior portions expand beyond what the exterior strength allows for, resulting in tensile stresses to develop to a point where the concrete cracks. These cracks considerably reduce the service life of concrete by allowing accelerated ingress of deleterious substances.

Evaluating the heat rise of concrete must strive to standardize as many external variables as possible to gauge a concrete mix purely on the mixture composition rather than factors such as environmental wind/precipitation loading, insulation level, size/shape of placement, initial/transient temperature differentials, or any compounding factors that would influence the final placement temperature. Most measurement methods attain this by constraining the initial temperature conditions, the sample size and shape, level of insulation, and prevention of moisture loss in an effort to seal the system from external variables. With all constraints in place, the only variables remaining are the mixture components themselves; therefore, the resulting heat given off is characteristic of the mixture composition.

2.1.1. Adiabatic Temperature Measurement

The measurement of the total heat of the concrete mixture components, given an initial temperature, constant specimen shape and size, without loss or gain of temperature to the environment is known as the adiabatic temperature rise of concrete. The technical definition of adiabatic is “a condition in which heat neither enters or leaves the system” [4]. RILEM Technical Committee 119-TCE defines an adiabatic calorimeter as one that prevents the specimen from having a maximum temperature loss of 0.02 K/hr [5]. This is the maximum heat rise that can be attained from a concrete mixture given the initial conditions with no external factors. In practice, perfectly adiabatic conditions are essentially impossible to attain; however, very close approximations are performed by measuring the heat evolved by concrete and adjusting the environmental temperature to match.

Adiabatic temperature rise measurement has significant drawbacks and potential pitfalls due to the nature and precision of the measurements. Since the device must have a maximum temperature loss of 0.02 K/hr, active heating and cooling systems are required to maintain environmental conditions. Furthermore, as the chemical hydration reactions are affected by temperature, having systemic errors in environmental temperature will result in erroneous results [6]. This necessitates highly calibrated temperature sensors and tightly controlled heating and cooling systems that do not over heat or over cool the system, which would affect the reaction.

One method to ease the temperature sensitivity is using large specimens, sometimes exceeding 1 yd³ (as they are less susceptible to temperature swings from the environment); but this requires large curing spaces, heating and cooling systems, mixers, molds, and specimen handling equipment. All of these modifications translate into higher initial and “per test” costs. While larger specimens are not strictly necessary, it has been shown that smaller specimens generally result in lower evolved heat when using commercially available equipment [7].

Despite the drawbacks, a properly conducted adiabatic temperature rise measurement will result in quite accurate hydration curves. RILEM recommends the use of adiabatic calorimeters in situations where high degrees of accuracy are required, or when attempting to simulate conditions that approximate adiabatic conditions such as in mass concrete placements where the majority of concrete is self-insulated [5].

2.1.2. Semi-Adiabatic Temperature Measurement

A method of measurement that typically requires less experimentation set up, involves placing a specimen into a well-insulated container and measuring the heat evolved over time without adjusting the temperature of the environment. Consequently, some amount of heat energy is lost to the environment as a tradeoff for lower cost of experimentation. This method is known as semi-adiabatic temperature rise because heat loss/gain is minimized, however, active systems are usually not in place to assure that the heat remains constant. RILEM TC 119-TCE defines a semi-adiabatic calorimeter as a calorimeter that has a maximum heat loss that is less than 100 J/(hr-K) [5].

One of the major benefits to the semi-adiabatic temperature rise measurement is that there are several variants of “off-the-shelf” calorimeters that can be purchased to measure the heat evolved from concrete specimens. One of downsides is that the sample size is generally a 4 in. x 8 in.

cylinder, which is considered to be a small sample of concrete as true adiabatic temperature rise of concrete is dependent on size. As a result, using a small specimen size could result in large variability, especially with extended measurements as the total heat output is low. However, semi-adiabatic calorimeters can be fabricated to house specimens of any size by using large amounts of insulation. This comes with the same mixing and handling issues as the larger adiabatic testing, but will not require large air conditioning systems to match the temperature, making it less expensive.

Additionally, several semi-adiabatic calorimeters that are available off-the-shelf have the capability to test more than one specimen at a time. While this may seem advantageous, there is a concern for “thermal crosstalk” where a sensor will detect heat from a neighboring specimen erroneously. Furthermore, even with multiple testing channels, the calorimeter should not be opened while a test is in progress; therefore, the number of channels available for use is limited to the number of concrete mixes or specimens that will be tested for the same duration.

The largest detraction from semi-adiabatic calorimetry is that the data has to be converted from semi-adiabatic heat rise to an approximated adiabatic heat rise. Due to this approximation, relaxation in specimen insulation and heat measurement are often employed lending to considerably reduced costs, but the data obtained is not “true” adiabatic heat rise.

2.1.3. Isothermal Temperature Measurement

Isothermal temperature measurement involves the determination of the heat of evolution of the chemical reaction of a mass of cement paste, mortar, or concrete under constant temperature.

According to ASTM C1679 (primarily used for cementitious paste experiments), samples are placed into an isothermal calorimeter, which measures heat flow from a specimen maintained at a constant temperature by intimate thermal contact with a constant temperature heat sink [8]. The heat moving through the sensor is recorded as thermal power, which is plotted as a function of hydration, providing an indication of the rate of hydration over time at a given temperature [8].

Isothermal calorimetry is almost universally utilized for cementitious pastes and mortars;

however, there are some companies that offer calorimeters with cells for concrete.

With isothermal testing, the specimen is meant to remain at a constant temperature which is in stark contrast to the previous two test methods that measure the increase in specimen temperature. To maintain constant temperature, the heat energy must theoretically be removed as soon as it is created, but also simultaneously measured. The majority of isothermal calorimeters rely on a single directional thermal sensor (below the specimen) to record temperature, and therefore, some heat is lost to the calorimeter through the sides or top of the specimen. This effect is amplified for larger samples such as concrete. Similar to the multi-channel semi-adiabatic chambers, isothermal chambers will usually have more than one channel so crosstalk is also a concern with these experiments. The most common off-the-shelf isothermal calorimeters have specimen capacities of approximately 7.5 in³ (0.125 L) whereas a 4 in. x 8 in. cylinder is approximately 100 in³ (1.64 L). This sample size restriction would limit the aggregate size required to get a representative sample.

2.2. ADIABATIC TEST METHODS

2.2.1. USBR Method

The United States Bureau of Reclamation (USBR) developed a method of measuring the adiabatic temperature rise of concrete. This method, simply called “Procedure For Temperature Rise of Concrete” is designated as procedure USBR 4911-92 [9]. In summary, the test involves a cylindrical specimen of concrete having dimensions of 21.5 in. dia. x 21.5 in. tall and approximate 700 lb that is placed into a well-insulated calorimeter wherein an electrical resistance thermometer (resistance temperature device, RTD) is embedded into the specimen to measure temperature. This calorimeter is placed into a room where the temperature of the air is maintained to be the same temperature as the concrete specimen; as the cementitious materials react and produce heat, the RTD measures the temperature and relays the information to the air temperature controller which adjusts the temperature of the room. When no specimens are present in the room, the temperature control box shall maintain the room temperature at 70°F (21.1°C).

This method requires the concrete mold to be 21.5 in. dia. by 21.5 in. tall, made of 20-gauge “black iron”, with a 7 in. hole in the top for filling the mold in two layers in a manner such that exact quantities are known.

Note: The term “black iron” does not designate a specific grade of metal; colloquially black iron is used in piping to indicate a pipe that is coated with a lacquer paint to prevent corrosion.

However, ASTM A123 – Standard Specification for Zinc (Hot-Dip Galvanized) Coatings on Iron and Steel Products defines “black” as “the condition of not galvanized or otherwise coated. For purposes of this specification the word ‘black’ does not refer to the color or condition of surface, or to a surface deposit or contamination” [10]. It should be noted that this specification may be recommending the use of “black plate” which is low-carbon steel, however, this material is noted in ASTM A625 as being “highly susceptible to rusting” and ASTM A650 notes that it is “relatively brittle” but both specifications note its use for can bodies [11], [12]. Additionally, there are a number of manufacturers in the US that supply black plate in thicknesses up to 20-gauge.

The lid must have two 12 in. long brass tubes soldered through the lid extending into the container which will house temperature measuring devices. Additionally, an air valve is soldered on the lid so the container can be pressurized to 2 - 3 inHg to ensure that the container is air-tight. Optionally, a manometer is also soldered to the lid to allow the study of pressure development during hydration.

Once sealed, oil is added to each thermometer well to act as a heat transfer fluid between the brass tubes in the specimen and the temperature measuring devices. A control thermometer (or RTD) is immediately placed in one of the wells and connected to the temperature controller of the room so that the room temperature will match the specimen temperature. In the second well, a second thermometer is placed and read to ensure that the air temperature of the room is the same temperature of the specimen; if not, adjustments to the temperature controller (temperature offsets) must be made.

Temperature readings are then taken at hourly intervals at the beginning of the test with increasingly greater intervals as the test progresses. Temperature monitoring should continue for a minimum of 28 days and the values of temperature rise and heat of hydration of cementitious material should be reported at 1, 3, 7, 14, and 28 days. Table 1 provides a summary of the approximate costs to run the USBR 4911-92 adiabatic test method for one sample.

Table 1. Summary of required equipment and approximate costs using USBR 4911-92.

Non-Consumables – Approximate Cost Consumables – Approximate Cost

Ultra-precise temperature sensor – $1,000 (similar to Honeywell Type R4 or R6 RTD; if these do not meet specifications, Correge Sensors can custom make probes to desired specifications)

Specimen Mold3 – $500 - $1,500

Temperature controlled room1 – $50,000 - $100,000

Computer with controller and software2 – $5,000 - $8,000

Specimen insulation – $200

Data acquisition modules and chassis – $1,500 - $2,500; depending on number of thermal inputs

Heated oil bath for calibration of temperature sensors – $800 (similar to Yamato BO400)

Total Approximate Cost: $60,000 - $110,000 ≤ 30 Days Concrete ≤ $30,000 Uncomplicated No Hazards

✓ ✓ X X ✓ 1The temperature controlled room is only specified to be “heavily insulated”. An example room comprises of two layers of masonry units with 5 - 7 in. of rockwool insulation between, and the floor is a 4 in. thick concrete slab upon 10 in. thick vermiculite concrete (as insulation) which is placed on another 4 in. reinforced concrete floor. The interior walls are 1 in. thick plaster that are sprayed with aluminum paint. The heating/cooling system is installed in the ceiling (thermally insulated from the interior) and delivered through perforated wall panels.

2 Computer requirements vary depending on thermal controller and software. Additionally, any processing of the signal may require additional software that FDOT may or may not have licenses for such as LabView to generate graphical user interfaces for the computer.

3The specimen molds are not standard sizes and would need to be fabricated for each test. The material specified for the molds is not widely available (limited suppliers in North America) and would need to be either welded or soldered. Furthermore, fixtures need to be placed onto the lid to hold the temperature sensors which can be fabricated ahead of time. As the material specified is not corrosion resistant, it would be preferred to fabricate the specimens shortly before testing, however this tends to make cost rise due to short run fabrications.

Drawbacks of the Method:

The largest drawback of the USBR adiabatic test method is cost; this method requires a dedicated room with a custom heating and cooling system designed to match a thermocouple.

Following the room costs, the method requires relatively costly one-time use molds, which are custom made. As the molds are almost complete cylinders (with two lids welded or soldered in place) they cannot be nested or stored efficiently, thus bulk purchasing would require a large amount of conditioned storage space. The storage space would need to be conditioned because the material specified (“black iron”) is not corrosion-resistant (such as more widely available material including aluminum, stainless steel, or an aluminum coated steel sheet product).

Furthermore, the method requires the presence of a technician at the time of casting have the necessary skills to solder the lid in place in a water-tight manner. In all likelihood, this would require several technicians in the laboratory to be trained to have this skill to ensure that at least one person is present and available for the task.

2.2.2. RILEM Method

RILEM TC 119-TCE describes a method of adiabatic testing and includes general design considerations for performing an adiabatic test with prescriptions on calorimeter design, materials, controls, and concrete size, shape, and characteristics [5]. The concrete specimen is to be cast into a cylinder of approximately 245 in.³ (4,000 cm³); with a minimum dimension of three times the maximum aggregate size, with a recommended maximum aggregate size of 1 ¼

in. (32 mm).

The sample must have a PT100 resistance thermometer or thermistor inside of a metal tube in the middle of the specimen; to ensure good thermal conductivity with the concrete, the tube is filled with oil. The container is then sealed to be watertight. Once sealed, the container is placed on a thermal heater and surrounded with insulating material as shown in Figure 1. A flexible polyurethane foam jacket is placed around the specimen as insulation, and the entire assemblage is placed into a thermal jacket, which is plumbed to circulate water to create an isothermal surface at the interface with the flexible polyurethane foam. Finally, the water jacket is insulated with more foam and placed inside of a metal container for the remainder of the test. A computer controller monitors the temperature of the center of the concrete specimen and applies voltage to the heater as well as adjusts the water circulation temperature to warm the specimen (if necessary). Since the water jacket temperature should never exceed the temperature of the specimen, the heated water temperature is set slightly below that of the specimen; ultimately resulting in a small thermal loss from the system. Due to the requirement that the sample must not lose more than 0.02 K/hr, the offset temperature of the water jacket must be adjusted so that the adiabatic conditions are met, and the temperature sensor must have a sensitivity of ± 0.01 K.

Given these design considerations, Table 2 provides a summary of the approximate costs to run the RILEM adiabatic test method for one sample.

Figure 1. General diagram of adiabatic calorimeter design adapted from RILEM TC 119.

Table 2. Summary of required equipment and approximate costs using RILEM TC 119-TCE.

Non-Consumables – Approximate Cost Consumables – Approximate Cost

PT100 Sensor – $110/ea (similar to OMEGA Item# PR-21SL-3-100-A-0600- M12-1)

Thermowell – $35/ea (similar to OMEGA Item# 1/2-260S-U71/2-304SS-F);

dependent on temperature sensor size

Foam Insulation, varies with specimen size – $200

Specimen Mold1 – $15 ea + VAT + Int’l shipping

Thin heater2 – $75 (similar to OMEGA Item# SRFRA-10/10)

Circulating water bath – $5,500 - 6,000 (similar to PolyScience PS:AP15R-30- A11B or PS:AD15R-30-A11B)

Water Jacket – $3,000 (custom fabricated)

Computer with controller and software3 –

$1,500 - $2,500

Total Approximate Cost: $16,500 - $22,000

✓ ✓ ✓ X ✓ 1 One potential solution for a 4 L cylindrical container that is air tight is to purchase European paint tins. U.S.

supply of 4 L cylindrical molds that are air-tight is limited and would likely require custom fabrication.

2 Depending on the voltage of the heater (some are AC some are DC), a power supply and inverters or transformers may be required to drive the heater unit.

3 Computer requirements vary depending on thermal controller and software. Additionally, any processing of the signal may require additional software that FDOT may or may not have licenses for such as LabView to generate graphical user interfaces for the computer. This software would also have to control the thin electrical heater and is required for calibration.

This estimate does not include any associated labor, programming, or fabrication costs, or additional items such as modifying the concrete mold to hold the thermowell (for instance by drilling a hole in the lid and using bulkhead pipe fittings which the thermowell can be screwed into), or ancillary items like wiring or electrical connectors, uninterruptable power supplies, cabinets for the computer, tools for wiring, or other miscellaneous items that may be required for construction and execution.

Drawbacks of the method:

The accuracy of the method is largely dependent on sample size; Lee et al. researched commercially available adiabatic calorimeters in sizes of 6 L (1.6 gal), 30 L (7.95 gal), and 50 L (13.2 gal) and found that small specimen sizes (6 L) should not be used [7]. The results showed that the difference in ultimate adiabatic temperature rise between the 6 L and 30 L adiabatic calorimeters was approximately 10°C, 8°C, and 7°C for concrete mixes containing 500 kg/m³ (843 lb/yd³), 400 kg/m³ (674 lb/yd³), and 300 kg/m³ (506 lb/yd³) of…