Evaluation of the Compressive Strength and Rapid Chloride Permeability of High Replacement Ternary Mixtures

Transcription

1 Rupnow Evaluation of the Compressive Strength and Rapid Chloride Permeability of High Replacement Ternary Mixtures Submission Date: -- By: T. D. Rupnow Tyson D. Rupnow (Corresponding Author) Louisiana Transportation Research Center Gourrier Avenue Baton Rouge, Louisiana Ph. -- Fax -- Word Count Text Figures Tables Total

2 Rupnow ABSTRACT Portland cement concrete (PCC) is the world s most versatile and most used construction material. To meet the global demand for sustainability, engineers have looked to alternative supplementary cementitious materials (SCMs) such as fly ash, silica fume, ground granulated blast furnace slag, to increase pavement durability while lowering the initial and life-cycle cost. The objective of the study was to characterize the fresh and hardened concrete characteristics of high volume ternary cementitious combinations. Twenty-seven ternary mixtures with portland cement replacement rates ranging from to percent were produced and the fresh and hardened characteristics were determined. Fresh concrete properties of air content, slump, unit weight, and set time were measured. Hardened concrete properties measured included compressive strength at,,,, and days of age and rapid chloride permeability at days of age. Compressive strength results showed equal to or greater compressive strengths especially at later ages of and days. The compressive strengths of all mixtures with SCM replacements up to percent meet Louisiana Department of Transportation and Development (LADOTD) specifications of psi at -days. The ratios of the seven to day compressive strengths showed that they are more resistant to early age cracking due to the lower modulus at early ages allowing for more creep. The rapid chloride permeability results show that the ternary mixtures will easily meet the new permeability specifications for all structural class concrete requiring less than Coulombs at days. Results of this study indicated that a replacement level of up to percent SCMs are reasonable for LADOTD construction projects.

3 Rupnow INTRODUCTION Portland cement concrete (PCC) is the world s most versatile and most used construction material. Modern concrete consists of six main ingredients: coarse aggregate, sand, portland cement, SCMs, chemical admixtures, and water. Global demand for PCC sustainability has risen as of late. To meet that need, engineers have looked to alternative binders such as fly ash, silica fume, ground granulated blast furnace slag (GGBFS), and other SCMs to increase pavement durability while lowering the initial and life-cycle cost. Ternary mixtures are uniquely suited to address the sustainability and cost aspect of PCC. There is general agreement that the use of SCMs has the following effects of concrete: Improved workability and finish ability. Strength Gain - Despite early strength reduction, beyond days concrete incorporating SCMs tend to show increased strengths over Portland cement concrete. Effect of temperature rise in mass concrete The use of SCMs has been proven to reduce early rate of heat generation. Permeability is reduced in mature concrete and resistance to sulfate and chloride attack is improved. Freeze thaw resistance, modulus of elasticity and resistance to de-icing salts are all about the same as in ordinary portland cement concrete. Resistance to corrosion of reinforcing steel. The use of SCMs in concrete helps to reduce permeability and thus to the reduction of chloride penetration. Increased time of setting and unpredictable change in time between initial and final set. This is of particular concern for saw cutting operations. Detailed literature regarding portland cement and other SCMs can be found in works published by the Portland Cement Association (PCA) and the National Concrete Pavement Technology Center (-). A synthesis study detailing the use of ternary cementitious mixtures was conducted by the Canadian Cement Association and the results showed that the use of ternary mixtures was sporadic, and was generally confined to particular Departments of Transportation (DOTs) (). Since the initial work completed by Tikalsky et al. (), a second phase has been completed and the results showed that replacements of portland cement up to percent do not severely affect the PCC properties (). That study is currently finishing up the Phase III field trials in several states and is slated to be completed with a final report in late. The objectives of this research were to characterize the fresh and hardened concrete properties of potential ternary cementitious combinations. The replacement rates for class C and class F fly ash were set at,, and percent. The replacement rates for grade and grade slags were set at, and percent. The control mixtures were produced using current replacement rates set forth in LADOTD specifications. The total replacement rate of cement was varied from to percent. Concrete samples were produced in laboratory conditions with a w/cm ratio of.. Fresh concrete properties measured included: slump, unit weight, air content, and set time. Hardened concrete properties measured included: compressive strength, flexural strength, rapid chloride permeability, freeze-thaw durability, shrinkage, coefficient of thermal expansion, and modulus of elasticity. For purposes of this paper, the fresh concrete properties are discussed with compressive strength and rapid chloride permeability results.

4 Rupnow MATERIALS AND TEST METHODS Cementitious Materials X-ray fluorescence (XRF) was used to determine the chemical constituents of the cementitious materials used in the laboratory produced concrete. Table shows the results of the XRF analysis. Note that all materials are representative of those used on construction projects in Louisiana and conform to applicable ASTM, AASHTO, and LADOTD standards and specifications. TABLE XRF Results for the Cementitious Materials Used in the Laboratory Test Factorial Type I/II Portland Class C Fly Class F Fly Grade Grade Oxide Cement Ash Ash Slag Slag SiO..... Al O..... Fe O..... CaO..... MgO..... Na O..... K O..... TiO..... SO..... LOI..... Test Matrix Table shows the laboratory test factorial used to develop a greater understanding of the behavior of ternary mixtures, especially those with portland cement replacement rates greater than percent. The mixture id notes the name of each mixture and the numbers in each of the columns indicate what percentage of that material is used. For example, Mixture ID TI- GS-C contains percent type I/II portland cement, percent grade slag, and percent class C fly ash. Each mixture was produced and cured at ºF. The reference mixtures contained pounds of total cementitious material per cubic yard, a # limestone coarse aggregate, natural sand fine aggregate. The coarse to fine aggregate ratio was set at : for the control mixtures. For ternary mixtures the coarse aggregate content was kept constant and the sand fraction was adjusted to keep a constant mortar volume for comparison of length change results. The water/cementitious material ration (w/cm) was kept constant at. and air entraining agents and water reducers were used to obtain mixtures conforming to LADOTD standards.

5 Rupnow TABLE Laboratory Test Factorial Type I/II Portland Mixture ID Cement Class C Fly Ash Class F Fly Ash Grade Slag Grade Slag TI* TI-C* TI-F* TI-GS* TI-GS* TI-GS-C TI-GS-C TI-GS-C TI-GS-C TI-GS-C TI-GS-C TI-GS-C TI-GS-C TI-GS-C TI-GS-C TI-GS-C TI-GS-C TI-GS-F TI-GS-F TI-GS-F TI-GS-F TI-GS-F TI-GS-F TI-GS-F TI-GS-F TI-GS-F TI-GS-F TI-GS-F TI-GS-F TI-C-F TI-C-F TI-C-F *Denotes a control mixture Test Methods For this study, all samples were produced in the laboratory and then cured in a percent relative humidity room until the age of testing. Slump, air content and unit weight were measured on the fresh concrete according to ASTM C (), ASTM C (), and ASTM C (), respectively. Set time was measured according to ASTM C (). Compressive strength specimens ( x in. ( x mm) cylinders) were tested according to ASTM C () at,,,, and days of age. ASTM C () was used to determine the rapid chloride

6 Rupnow permeability at -days of age. Samples were tested in triplicate for both the compressive strength and rapid chloride permeability. RESULTS AND DISCUSSION Fresh Concrete Property Results The fresh concrete properties of slump, air content, unit weight, and set time are shown in Table. All mixtures met the slump and air content requirements set forth by LADOTD standards and specifications for portland cement concrete. The control mixtures exhibited typical laboratory set times between six and nine hours. The ternary mixture results show that as the portland cement replacement percentage is increased, the set times generally increase. The ternary mixtures exhibit set times ranging from equal to the control mixtures all the way up to more than twice the control mixtures. The extended set times steadily increase up to between and percent replacement of portland cement. At portland cement replacement percentages greater than percent, the time to initial and final set were dramatically increased. The field results for these mixtures are expected to be slightly different due to the increased temperatures in summertime construction conditions. The author believes that for Louisiana environmental conditions, a portland cement replacement up to percent will not be detrimental to performance, and will actually aid contractors in the hot summer conditions. The three mixtures with final set times greater than hours indicate potential mixture incompatibility. The individual set time of field produced concrete at these replacement levels (greater than percent) will vary according to the chemistry of the individual cementitious materials and resulting mixture design. The author is quick to note that concrete produced with high replacement rates do have other potential drawbacks, especially when doing paving or flatwork. Research, and experience shows that a higher rate of evaporation may occur during concrete placement and finishing leading to an increased tendency for plastic shrinkage cracking. Extreme care must be taken to avoid plastic shrinkage cracking by good placement and curing practices. A double coat of curing compound may be required in certain circumstances. For concretes produced with long times to initial and final set, use of an accelerator may be warranted to aid in mitigation of cracking potential.

7 Rupnow TABLE Slump, Air Content, Unit Weight, and Set Time Results For All Mixtures Air Unit Time to Time to Slump Content Weight Initial Set Final Set Mixture ID (in) (%) (pcf) (hrs:mins) (hrs:mins) TI... : : TI-C... : : TI-F... : : TI-GS... : : TI-GS... : : TI-GS-C... : : TI-GS-C... : : TI-GS-C... : : TI-GS-C... : : TI-GS-C... : : TI-GS-C... : : TI-GS-C... : : TI-GS-C... : : TI-GS-C... : : TI-GS-C... : : TI-GS-C... : : TI-GS-C... : : TI-GS-F... : : TI-GS-F... : : TI-GS-F... : : TI-GS-F... : : TI-GS-F... : : TI-GS-F... : : TI-GS-F... : : TI-GS-F... : : TI-GS-F... : : TI-GS-F... : : TI-GS-F... : : TI-GS-F... : : TI-C-F... : : TI-C-F... : : TI-C-F... : : Compressive Strength The compressive strength results show that a wide range of ternary combinations will meet LADOTD compressive strength requirements. Figure shows the comparison between average compressive strength and age for the control mixtures. Note that these mixtures all meet the psi specification for structural concrete within seven days of age. Figure shows the average compressive strength results for mixtures containing grade GGBFS and class C fly ash. Note that the only mixture not meeting the psi

8 Average Compressive Strength (psi) Rupnow compressive strength at -days was the mixture containing only percent portland cement. Although it did not meet at -days of age, the mixture passed compressive strength requirements at -days indicating that this mixture would be an ideal candidate for a mass concrete placement. The results also indicate that the mixtures will continue to gain strength due to pozzolanic action based on the shape of the strength gain curves. TI TI-C TI-F TI-GS TI-GS FIGURE Average compressive strength results for the control mixtures.

9 Average Compressive Strength (psi) Rupnow TI-GS-C TI-GS-C TI-GS-C TI-GS-C TI-GS-C TI-GS-C FIGURE Average compressive strength results for mixtures containing grade slag and class C fly ash. Figure shows the average compressive strength results for mixtures containing grade GGBFS and class F fly ash. The only mixture not meeting the psi compressive strength at -days was the mixture containing only percent portland cement. The remaining mixtures produced strengths exceeding psi at days of age. Figure and Figure show the average compressive strength results for mixtures containing grade GGBFS and class C and F fly ash, respectively. The results show that a replacement of portland cement up to percent produced compressive strengths in excess of psi at days of age. The percent portland cement replacement mixtures still made psi, but at much later ages of and days for the mixtures containing class C and class F fly ash, respectively. Figure shows the average compressive strength results for mixtures containing both class C and class F fly ash. The results show that an increase in percentage of portland cement replacement greatly effects the compressive strengths at early ages. The results indicate that a maximum replacement rate for these mixtures is between and percent. Although these results are somewhat low for the laboratory, field results at percent fly ash replacement have obtained over psi on a project located in Lake Charles, LA.

10 Average Compressive Strength (psi) Average Compressive Strength (psi) Rupnow TI-GS-F TI-GS-F TI-GS-F TI-GS-F TI-GS-F TI-GS-F FIGURE Average compressive strength results for mixtures containing grade slag and class F fly ash. TI-GS-C TI-GS-C TI-GS-C TI-GS-C TI-GS-C TI-GS-C FIGURE Average compressive strength results for mixtures containing grade slag and class C fly ash.

11 Average Compressive Strength (psi) Average Compressive Strength (psi) Rupnow TI-GS-F TI-GS-F TI-GS-F TI-GS-F TI-GS-F TI-GS-F FIGURE Average compressive strength results for mixtures containing grade slag and class F fly ash TI-C-F TI-C-F TI-C-F FIGURE Average compressive strength resutls for mixtures containing class C and class F fly ash.

12 Rupnow The compressive strength results are very encouraging and indicate up to percent of the portland cement can be replaced for a large portion of concrete mixtures used on LADOTD projects. The results show that LADOTD has been very conservative with only allowing up to to percent fly ash and up to percent slag respectively, in binary mixtures. The main drawbacks to lower early age strengths are the need to keep forms in place longer for structural concrete applications and an increased set time for sawing operations for paving applications. These problems are easily addressed with an adjustment in paving and sawing operations for paving concrete and the introduction of another set of forms, or leave-inplace forms for structural concrete applications. Additional curing may be required to prevent plastic shrinkage cracking. Set accelerating admixtures may be used to shorten the time to initial set for sawing operations, or early removal of forms. Table shows the average and day compressive strengths and the ratio between the two. The portland cement control mixture had a ratio of about one. This indicates that nearly all the strength is being gained within the first seven days after batching and placement. The binary fly ash mixtures had a ratio of. and. for the class C and class F fly ash combinations, respectively. These values are still very typical being under.. Note that the binary mixtures containing slag had ratios. and.. The to seven day compressive strength ratios for the ternary mixtures are significantly different. The lowest ratio for a ternary mixture is. and the highest is.. These values indicate that the concrete mix design is a slower strength gain. One advantage of a slower strength gain is the concrete has a lower modulus of elasticity value at early ages allowing for more creep. This increased ability to creep at early ages may lead to a reduction in cracking potential ().

13 Rupnow TABLE Average Seven and -Day Compressive Strengths and the to Seven Day Compressive Strength Ratio for All Mixtures Mixture ID Day Day : TI. TI-C. TI-F. TI-GS. TI-GS. TI-GS-C. TI-GS-C. TI-GS-C. TI-GS-C. TI-GS-C. TI-GS-C. TI-GS-F. TI-GS-F. TI-GS-F. TI-GS-F. TI-GS-F. TI-GS-F. TI-GS-C. TI-GS-C. TI-GS-C. TI-GS-C. TI-GS-C. TI-GS-C. TI-GS-F. TI-GS-F. TI-GS-F. TI-GS-F. TI-GS-F. TI-GS-F. TI-C-F. TI-C-F. TI-C-F. Rapid Chloride Permeability The RCP results are shown in Figure to Figure. The results for the control mixtures shown in Figure illustrate typical permeability results for a straight portland cement mixture compared to other binary mixtures. The replacement of portland cement leads to a reduction in permeability as expected.

14 Chloride Ion Penetrability (Coulomb) Rupnow Figure to Figure shows the influence of ternary mixtures on the permeability of the resulting concrete. Note that all ternary mixtures containing combinations of slag and fly ash fell below the very low permeability threshold of Coulombs and will be acceptable under the new LADOTD specifications for structural concrete which call for a permeability value of less than Coulombs. TI TI-C TI-F TI-GS TI-GS FIGURE Rapid chloride permeability results for the control mixtures. The results for ternary mixtures containing both class C and class F fly ash are shown in Figure. The results are as expected with an increase in portland cement replacement a decrease in permeability is observed. It is important to note that use of class F fly ash decreased the permeability of the concrete more so than the use of class C fly ash in combination with slag for mixtures tested in this study. This is due to the class F fly ash being more pozzolanic with a greater portion of the ash being in the glass phase.

15 Chloride Ion Penetrability (Coulomb) Chloride Ion Penetrability (Coulomb) Rupnow TI-GS-C TI-GS-C TI-GS-C TI-GS-C TI-GS-C TI-GS-C FIGURE Rapid chloride permeability results for mixtures containing grade slag and class C fly ash. TI-GS-F TI-GS-F TI-GS-F TI-GS-F TI-GS-F TI-GS-F FIGURE Rapid chloride permeability results for mixtures containing grade slag and class F fly ash.

16 Chloride Ion Penetrability (Coulomb) Chloride Ion Penetrability (Coulomb) Rupnow TI-GS-C TI-GS-C TI-GS-C TI-GS-C TI-GS-C TI-GS-C FIGURE Rapid chloride permeability results for mixtures containing grade slag and class C fly ash. TI-GS-F TI-GS-F TI-GS-F TI-GS-F TI-GS-F TI-GS-F FIGURE Rapid chloride permeability results for mixtures containing grade slag and class F fly ash.

17 Rupnow FIGURE Rapid chloride permeabiltiy results for mixtures containing class C and class F fly ash. Although the ternary mixtures exceed the proposed new LADOTD permeability specifications ( coulombs at days or k -cm at days), the author strongly cautions against the use of the combinations without first conducting trail batches. While ternary combinations will greatly assist in reduction of permeability of concrete, other factors influence the concrete permeability such as paste content, w/cm, and curing conditions. It is important to note that the permeability results shown in this report will continue to improve at later ages. It is common knowledge that the permeability can improve up to days after concrete placement in ideal conditions. If the samples tested for this study were to be retested at later ages such as one year of age, the permeability values would be significantly better. CONCLUSIONS The results of this study warrant the following conclusions. The fresh concrete results showed adequate workability, air content, and set times for all ternary mixtures with portland cement replacements less than or equal to percent. Compressive strength results showed equal to or greater compressive strengths especially at later ages of and days. The compressive strengths of all mixtures with SCM replacements up to percent met LADOTD specifications of psi. The ratios of the to seven day compressive strengths showed that they may be more resistant to early age cracking due to the lower modulus of elasticity at early ages allowing for more creep.

18 Rupnow The rapid chloride permeability results show that the ternary mixtures will easily meet the new permeability specifications for all structural class concrete requiring less than Coulombs at -days of age. The above results point to a reasonable portland cement replacement level with SCMs of about percent for LADOTD concrete projects. Care should be taken when interpreting these results and the results apply only to the materials used and tested through the course of this study. Producers and contractors wanting to implement these results are strongly encouraged to produce trail batches with their locally available materials to ensure the mixture s ability to meet and exceed the standards and specifications. RECOMMENDATIONS The author recommends full implementation of the results of this study and suggests a maximum portland cement replacement of percent. Ternary combinations containing class C and class F fly ash should be allowed, but be incorporated in equal amounts. Slag and fly ash combinations may be used with the exception being that the fly ash content cannot be greater than the slag content. Lastly, the cold weather limitation should be set such that risk of cracking and delayed set times are minimized. To this end, the author suggests a cold weather limitation of about degrees F; the temperature at which ternary concrete operations should cease. ACKNOWLEDGEMENTS The author would like to thank the Federal Highway Administration, Louisiana Department of Transportation and Development, and the Louisiana Transportation Research Center for sponsoring this research. The efforts of Randy Young, Matt Tircuit, Shane Laws, Scott Reech, Benjamin Shearer, Kelly Goudeau, Joel Taylor, Jacob Newgard, Anthony Saladino, Brandon Pitre, and Brennon Hughes in the concrete laboratory are greatly appreciated. The efforts of Mike Bailey are greatly appreciated for the chemical and physical testing of the cementitious materials used in the laboratory test matrix. The author would also like to thank Holcim, LaFarge, and Buzzi Unicem for their participation in the study by providing the grade slag, portland cement, and grade slag, respectively. The opinions, findings and conclusions presented here are those of the author and do not necessarily reflect those of LADOTD or the Federal Highway Administration. REFERENCES. Kosmatka, S.H., Kerkhoff, B., and Panarese, W.C. Design and Control of Concrete Mixtures. th Edition. Engineering Bulletin. Skokie, IL; Portland Cement Association,.. Tikalsky, P., Schaefer, V., Wang, K., Scheetz, B., Rupnow, T., St. Clair, A., Siddiqi, M., and Marquez, S. Development of Performance Properties of Ternary Mixtures: Phase I Final Report. Iowa State University, Ames, IA,.. MacLeod, N. A Synthesis of Data: On the Use of Supplementary Cementing Materials (SCMs) in Concrete Pavement Applications Exposed to Freeze / Thaw and Deicing Chemicals. Canadian Cement Association, Ottawa, ON,.

19 Rupnow. Tikalsky, P., Taylor, P., Hanson, S., and Ghosh, P. Development of Performance Properties of Ternary Mixtures: Laboratory Study on Concrete. Iowa State University, Ames, IA,.. ASTM C/M Standard Test Method for Slump of Hydraulic-Cement Concrete. Annual Book of ASTM Standards, Vol.., ASTM, Philadelphia, PA,.. ASTM C Standard Test Method for Air Content of Freshly Mixed Concrete by the Pressure Method. Annual Book of ASTM Standards, Vol.., ASTM, Philadelphia, PA,.. ASTM C Standard Test Method for Density (Unit Weight), Yield, and Air Content (Gravimetric) of Concrete. Annual Book of ASTM Standards, Vol.., ASTM, Philadelphia, PA,.. ASTM C/M Standard Test Method for Time of Setting of Concrete Mixtures by Penetration Resistance. Annual Book of ASTM Standards, Vol.., ASTM, Philadelphia, PA,.. ASTM C Standard Test Method for Compressive Strength of Cylindrical concrete Specimens. Annual Book of ASTM Standards, Vol.., ASTM, Philadelphia, PA,. ASTM C Standard Test Method for Electrical Indication of Concrete s Ability to resist Chloride Ion Penetration. Annual Book of ASTM Standards, Vol.., ASTM, Philadelphia, PA,.. Krauss, P. D., and Rogalla, E. A. Transverse Cracking in Newly Constructed Bridge Decks. NCHRP Report. National Research Council, Washington, D.C.,.