Khakimova, Ozyildirim, Harris 1

Transcription

1 Khakimova, Ozyildirim, Harris 0 INVESTIGATION OF HIGH PERFORMANCE FIBER REINFORCED CONCRETE PROPERTIES: VERY EARLY STRENGTH, TOUGHNESS, PERMEABILITY, AND FIBER DISTRIBUTION Evelina Khakimova, M.S., E.I.T., Corresponding Author Virginia Transportation Research Council 0 Edgemont Road, Charlottesville, VA 0 Telephone: () -; Fax: () -; emkrh@virginia.edu H. Celik Ozyildirim, Ph.D., P.E. Virginia Transportation Research Council 0 Edgemont Road, Charlottesville, VA 0 Telephone: () -; Fax: () -; Celik@vdot.virginia.gov Devin K. Harris, Ph.D. University of Virginia, Department of Civil and Environmental Engineering McCormick Road, Charlottesville, VA 0 Telephone: () ; Fax: () ; dharrris@virginia.edu Word Count: words text + tables/figures x 0 words (each) = words Submission Date: November, 0

2 Khakimova, Ozyildirim, Harris 0 ABSTRACT Concrete cracking, high permeability, and leaking joints allow for the intrusion of harmful solutions, resulting in concrete deterioration and corrosion of reinforcement. The development of durable concrete with limited cracking is a potential solution for extending the service life of concrete structures. Optimal design of very early strength (VES) durable materials will facilitate rapid and effective repairs, reducing traffic interruptions and maintenance work. The purpose of this study was to develop low-cracking durable materials that can achieve a very early compressive strength of,000 psi within hours. Various proportions of silica fume and fly ash, and steel fibers and polypropylene fibers were used to evaluate concrete durability and post-cracking performance. In addition, toughness, residual strength, permeability of cracked concrete, and fiber distribution were examined. VES durable concretes can be achieved with proper attention to mixture components (amounts of portland cement and accelerating admixtures), proportions (water cementitious material ratio), and fresh concrete and curing temperatures. Permeability values indicated that minor increases in crack width, above 0. millimeter, greatly increase infiltration of solutions. Adding fibers can facilitate crack width control. An investigation of fiber distribution showed preferential alignment and some clumping of fibers in the specimens and highlighted the need for sufficient mixing and proper sequencing of concrete ingredients into the mixer to ensure a uniform random fiber distribution. Results indicated that VES and durable fiber-reinforced concrete materials can be developed to improve the condition of existing and new structures and facilitate rapid effective repairs and construction. Keywords: very early strength, durability, fiber-reinforced concrete, steel, polypropylene, toughness, residual strength, permeability, fiber distribution

3 Khakimova, Ozyildirim, Harris INTRODUCTION New construction and rehabilitation of existing structures are in need of high-performance durable materials that facilitate the extension of the service life of structures with minimal maintenance. The National Bridge Inventory estimated that in 0 about % of U.S. structures were structurally deficient. In comparison,.% of Virginia Department of Transportation (VDOT) structures were considered structurally deficient at the end of Fiscal Year 0 (). The percentage of structurally deficient structures is being reduced; however, there is still a large number of structures requiring immediate rehabilitation. Therefore, the use of high-performance concretes, preferably with high early strength and high durability, that facilitate extension of service life of transportation structures is essential. The Federal Highway Administration started the implementation of high performance concrete (HPC) in. In the transportation industry, early HPC developments were commonly described as concretes with low permeability and high strength (). Further, the use of VES concretes has become common within state departments of transportation. The development of VES depends on the cement type, amount of paste, water-cementitious material ratio (w/cm), fresh concrete and curing temperatures, and the addition of supplementary cementitious materials (SCMs) and chemical admixtures (). There are no minimum strength criteria for VES concretes; however, a compressive strength of,000 psi was reported to be sufficient for regular traffic loads (). Punurai et al. () conducted a study on VES concretes using high proportions of Type I cement, an accelerating admixture, and fresh concrete temperatures above F to achieve the required early strength. The VES concretes reached a compressive strength of more than,0 psi and a flexural strength of more than 0 psi in about. hours. VDOT has successfully used latex-modified concrete (LMC) with rapid set cement (LMC-VE) in concrete overlays to obtain VES and low-permeability mixtures. LMC-VE can achieve a minimum compressive strength of,00 psi in hours, which allows for rapid repairs and early lane opening to traffic (). Some of the main drawbacks of VES concretes are increases in thermal and autogenous shrinkage and elastic modulus. In addition, there is a high risk of cracking, potentially leading to a decrease in concrete durability. The long-term strength is also negatively affected by the high temperatures at early ages (, ). Associated increases in fresh concrete temperature also cause a decrease in slump and an increase in water demand (). The addition of SCMs generally enhances concrete ultimate strength; reduces concrete permeability; and improves resistance to corrosion, alkali-silica reactivity, and sulfate attack. Further, the addition of fibers minimizes cracks and reduces crack widths, leading to a lower permeability compared to concretes without fibers (, ). Cracks less than 0.-millimeter-wide do not have a great effect on permeability, whereas a rapid permeability increase is seen for concrete with cracks wider than 0. millimeter (). Further, the American Concrete Institute considers 0. millimeter as a reasonable crack width for water-retaining structures (). The lower the permeability of concrete, the better it resists cycles of freezing and thawing, penetration of chlorides, corrosion of reinforcement, sulfate attacks, and alkali-silica and other harmful reactions (). In addition, large volumes of steel (S) or polypropylene (PP) fibers (0.% to.0% by volume) are expected to improve the post-cracking performance of concrete, increasing its durability, tensile and flexural strengths, residual strength, and toughness (). Further, it has been shown that S fibers do not exhibit any corrosion with crack widths less than 0. millimeter (). Light corrosion occurs with wider cracks and the corrosion product does not penetrate more than millimeter into concrete. Mangat et al. () demonstrated that corrosion does not occur with a crack width less than 0. millimeter for low carbon steel FRC.

4 Khakimova, Ozyildirim, Harris The distribution of fibers in concrete mixtures can considerably influence the performance of concrete. Fiber clumping causes the formation of sections without fibers, which act as defects and detrimentally affect the mechanical performance of FRCs (-). At the same time, because of random fiber distribution, only a certain fraction of the fibers is optimally oriented and aligned to resist tensile and flexural stresses efficiently (0). It has been determined that the longitudinal fiber alignment is beneficial to the post-cracking concrete flexural and tensile capacities (-). Further, statistical spatial point pattern analysis can be used to examine the fiber distribution. Akkaya et al. () used statistical K- and F-functions to examine the fiber clumping tendency and describe areas without fibers, respectively. The first several cracks occurred in the areas with fewer fibers. These areas formed because of fiber clumping in the other areas of the specimen. The clumping of fibers is indicated by higher statistical K-function and lower F-function values. PROBLEM STATEMENT Concrete cracking and high permeability allow for the intrusion of harmful solutions, resulting in concrete deterioration and corrosion of reinforcement. The use of low-permeability durable concretes with controlled cracking for repairs and new construction would improve the service life of the structures. Further, the use of VES materials would facilitate rapid and effective repairs and reduce traffic interruptions. The purpose of this study was to develop durable VES concrete mixtures with resistance to cracking. The target minimum compressive strength was,000 psi within hours. The mixtures included SCMs for durability and different types and volumes of fibers for crack resistance. Compressive and flexural strengths and drying shrinkage were determined. Other characteristics such as toughness, residual strength, permeability of cracked concrete, and fiber distribution were also examined. MATERIALS AND METHODS Overview The study was divided into three main stages. The first stage focused on the development of VES fiber-reinforced concrete (VES-FRC) mixtures that can reach a compressive strength of,000 psi within hours. Concrete batches without fibers were prepared and tested for compressive strength to attain the very early strengths. When satisfactory strengths were obtained, batches with fibers were prepared to determine characteristics other than compressive strength, such as flexural strength, toughness, residual strength, drying shrinkage, permeability of cracked specimens, and fiber distribution. The second stage focused on testing of the water permeability of cracked FRCs. Controlled crack widths of 0. to 0. millimeters (0.00 to 0.00 inch) were formed in FRC cylinders using the splitting tensile test in accordance with ASTM C; the cylinders were then tested using the falling head permeability equipment. The third stage investigated the fiber density and distribution in the concrete mixtures, and spatial fiber dispersion was examined at the cross sections of the crack location. Laboratory testing was completed at the Virginia Transportation Research Council (VTRC). Table shows the ASTM and other test methods and specimen sizes used during this study. Two types of fibers were considered for this study: PP and S. The -inch-long PP fibers were crimped in shape to facilitate a mechanical bond to concrete. The S fibers were straight with hooked ends for improved anchorage, with the total fiber length of. inches.

5 Khakimova, Ozyildirim, Harris 0 0 TABLE Fresh and Hardened Concrete Test Methods Property Test Method Specimen Size No. of Test Specimens Fresh Concrete Properties Air content ASTM C- - - Slump ASTM C- - - Unit weight ASTM C- - - Temperature ASTM C- - - Hardened Concrete Properties Compressive strength ASTM C- by in cylinder or per test age Elastic modulus ASTM C- by in cylinder at days Flexural strength ASTM C0- by by in beam or per test age Shrinkage (length change) ASTM C-0 by by in beam per FRC mixture Splitting tensile strength (crack formation) ASTM C- by in cylinder Total of Water permeability (falling head) Virginia Test Method 0 by in cylinder Total of Maturity test, Method ASTM C- - - Maturity test, Method ASTM C = not applicable. Stage I: Development of VES-FRCs The VES-FRC mixtures that can achieve a compressive strength of,000 psi within hours were developed. Initially, concrete batches without fibers were tested to achieve the very early strengths; then mixtures with fibers were made. The development of VES concretes mainly focused on the use of the following: increased cement content reduced w/cm increased fresh concrete temperature accelerating admixtures insulating the specimens during curing. The VES concretes with silica fume (SF) were developed based on the ongoing research at VTRC and by consultation with the industry. The mixture design of VES with fly ash (FA) was adopted from existing VDOT patching mixtures typically used for pavement repairs, which require a compressive strength of,000 psi within hours. The Class F FA and SF met the ASTM C and ASTM C0 standards, respectively. The VES with SF concrete batches without fibers were made at three temperatures:,, and 0 F. The temperatures were recorded using multiple thermocouples with a data logger. Two VES with FA concrete batches were made with fresh concrete temperatures at and F. To obtain elevated fresh concrete temperatures, the hot mix water was used; the aggregates and cementitious materials were kept at a room laboratory temperature of about F. After successful mixing of the VES concretes without fibers, two batches with PP or S fibers for each mixture with VES-FRC with SF and VES-FRC with FA were made. Based on the fiber manufacturers recommended dosages, lb/yd of PP and lb/yd of S fibers were considered. The fiber dosages were as recommended by the manufacturer and limited by proper mixing and workability (upper limit) and desired properties (lower limit). The cost will depend on the amount of used fibers for properties and project needs. The lb/yd of PP fibers appeared to be the maximum amount that can be added without any major mixing problems, such as fiber balling. The S fibers could be added at higher amounts; however, cost should be considered. Therefore, a small increase to 0 lb/yd was tried to ensure improvements in the post-cracking behavior. Table shows the VES-FRC mixture designs. Two types of FRC were used having a

6 Khakimova, Ozyildirim, Harris 0 0 different cementitious content, SCM, and w/cm to achieve the desired strength, permeability, and shrinkage. TABLE VES-FRCs Mixture Designs Component (lb/yd ) VES-FRC w/ SF VES-FRC w/ FA Cement Type I/II 0 0 Silica fume (%)* 0 - Class F fly ash (%) - Water Fine aggregate Coarse aggregate 0 0 Total cementitious material 00 w/cm PP (% by volume) (.0) - (.0) - S (% by volume) - 0 (0.) - 0 (0.) Admixture (oz/cwt) Set accelerating Hardening accelerating = not applicable; VES = very early strength; FRC = fiber-reinforced concrete; SF = silica fume; FA = fly ash; PP = polypropylene; S = steel. *SF is % of total cementitious content. For each batch, the compressive strength was tested starting from hours after casting and about every hour until a compressive strength of,000 psi was reached. Then the rest of the cylinders were tested at,, and days. In addition, samples for flexural strength, drying shrinkage, and falling head permeability tests were made. The first-peak and peak flexural strengths, residual strengths, toughness, and equivalent flexural strength ratio were determined in accordance with ASTM C0 (). All specimens were covered with plastic and insulating material and kept inside Styrofoam containers for the first hours or until a compressive strength of,000 psi was reached. Then the samples were demolded and moved to a moisture room for days. The moisture room relative humidity was maintained above % with an air temperature of ± F. During curing, the temperatures of the specimens were monitored. Stage II: Permeability of Cracked FRC Specimens For each FRC batch with S or PP fibers, eight to ten by inch cylinders were tested. ASTM C splitting tensile testing procedure was followed to form the cracks widths of 0. to 0. millimeter (0.00 to 0.00 inch). The MTS laser extensometer was used to capture the horizontal displacement between two reflective pieces of tape mounted at the center on one side of the specimen. In addition, an optical magnifier, with a 0-millimeter scale (in 0.-millimeter increments, numbered every.0 millimeter), was used to measure the crack widths with and without the applied load. Figure shows the splitting tensile test setup and the magnifier. The correlation between the results of the two measurement methods revealed a difference of about % on average, which was deemed to be acceptable because of the irregularity of the crack widths. The crack widths were measured at the top, middle, and bottom sections on both sides of the specimen. The crack width formed in the middle section of the specimen was the most observed width and was used as a final crack width of that specimen. The relaxation of the crack width after unloading was also determined. All VES-FRC specimens followed a similar trend of crack width recovery with an average of %.

7 Khakimova, Ozyildirim, Harris 0 (a) (b) (c) FIGURE (a) Splitting tensile test setup with MTS laser; (b) test specimen; (c) magnifier with scale. ASTM C0 was followed to saturate the cracked specimens for the permeability testing. Virginia Test Method 0, a falling head permeability test, was used to measure the coefficient of water permeability (CWP), i.e., the laminar water flow rate through the cracked saturated sample. The test was performed three times for each sample, and the average CWP, k, was determined in accordance with Equation (): k = a l A t ln (h h ) () where k = CWP, a = standpipe area, A = average specimen area, l = average sample thickness, t = average flow time, h = initial hydraulic head, and h = final hydraulic head. Stage III: Fiber Distribution Analysis To investigate fiber distribution, the flexural beams with PP and S fibers were sliced along the transverse and horizontal directions into -inch-thick sections. Figure shows the schematic of the ASTM C0 test method with a third-point flexural loading and the transverse (T) and horizontal (H) cutting orientations for the two slicing methods: HTH and THTHT. At least two specimens were analyzed for each method for each mixture type. H T H T H T H T (a) (b) FIGURE Schematic of the flexural test setup, and two slicing methods: (a) HTH, and (b) THTHT. H = horizontal cut; T = transverse cut. Digital images of the cross sections were taken and analyzed in MATLAB through image processing to determine fiber coordinates. Outputs of the MATLAB analysis code for a steel fiber specimen are displayed in Figure, with steps illustrating (a) an inverse grayscale of an original image, (b) a binary image after thresholding, (c) an image with removed pixels out of set

8 Khakimova, Ozyildirim, Harris limits, and (d) the original image with overlaid calculated fiber coordinates. Optimal orientation was taken into consideration; hence only fibers oriented at the angle or less were included in the fiber count. 0 (a) (b) (c) (d) FIGURE Image analysis process: (a) inverse; (b) binary; (c) pixel removal; (d) final. Fiber density was calculated, and spatial point pattern trends through statistical K- and F- functions were determined (). Typically, there are three types of spatial point patterns: clustered, regular (ordered), and random. The K-function represents the tendency of fiber clumping and measures the distance between fibers. The F-function can be used to measure the empty spaces between the fibers. The calculated K- and F-function values were compared to the values obtained under the complete spatial randomness (CSR) condition: K-CSR and F-CSR, respectively. For the K-function, the clustering is observed for values greater than K-CSR, whereas the F-function values that drift below the F-CSR curve indicate a more clustered pattern, with more fiber-free areas. RESULTS AND DISCUSSION Stage I: VES-FRCs Development VES Concretes Without Fibers The laboratory batches with VES concretes with SF were made at,, and 0 F fresh concrete temperatures and reached a compressive strength of,000 psi over,, and hours, respectively (Figure a). ASTM C and ASTM C maturity test methods, using the Nurse- Saul function, indicated the time the mixture with the fresh concrete temperature of F reached the required strength (Figure b). Further, the successful results of the VES with FA concrete patching mixtures led to the implementation of the mixtures in this project, with the compressive strength requirement of,000 psi within hours.

9 Khakimova, Ozyildirim, Harris Temperature [ F] VES w/ 0 F VES w/ F VES w/ F 0 0 Age [hours] Initial Set Time,000 psi reached (a) (b) 0 FIGURE (a) Trial VES w/ SF temperature developments; (b) strength-maturity relationship for the F batch. VES = very early strength; SF = silica fume. Further, the time of initial set was about,, and hours for the,, and 0 F mixtures, respectively. The temperature rise after the set was greater for the mixtures with higher initial fresh concrete temperatures, which indicates faster strength development. The trial VES with FA mixtures reached a compressive strength of,000 psi within hours. Therefore, based on maturity predictions, it was expected that these mixtures would gain the desired,000 psi compressive strength within hours. VES Concretes with Steel and Polypropylene Fibers The PP and S fibers were added to the optimized VES with FA and VES with SF concrete mixtures. The VES-FRCs fresh concrete properties are presented in Table. The air content values were within the VDOT specifications (). The slump values were low, ranging from to inches. The mixtures with FA had lower values than the ones with SF attributed to a lower dosage of high-range water-reducing admixture, the type of fiber, and a lower w/cm. However, the mixtures remained sufficiently workable for easy placement in the molds. The molds were then capped to limit evaporation. The shrinkage specimens during drying were kept in a special room with the relative humidity set to 0% ± % as required by ASTM C.

10 Khakimova, Ozyildirim, Harris TABLE VES-FRCs Fresh Concrete Properties Properties VES-FRC w/ SF VES-FRC w/ FA w/ PP w/ S w/ PP w/ S Air content (%)... Unit Weight (lb/ft ) Slump (in) Mix temperature ( F) 0 Air temperature ( F) VES = very early strength; FRC = fiber-reinforced concrete; SF = silica fume; FA = fly ash; PP = polypropylene; S = steel. All VES-FRC specimens reached a compressive strength of over,000 psi within. hours (Table ). Typical VES-FRCs flexural load-deflection plots are presented in Figure. The increase in first-peak and residual flexural strengths with age is shown. TABLE VES-FRCs Hardened Properties Test Age VES-FRC w/ SF VES-FRC w/ FA w/ PP w/ S w/ PP w/ S. hours,0 0,0,0. hours,0,0,0,0 hours - - -,0 Compressive strength (psi). hours,0,0 - - hours,00,0,0,0 days,0,0,0,0 days,0,0,0,0.-. hours 0 0 / 0 / First-peak / peak flexural strength hours 0 /,00 0 /, (psi) days,000,0 /,00 days,0,0 /,0,0, /,0 Elastic modulus ( psi) days.... Drying shrinkage (%) days* Months * Length change values at days after casting, including the first days of moist curing. - = data not available; VES = very early strength; FRC = fiber reinforced concrete; SF = silica fume; FA = fly ash; PP = polypropylene; S = steel. The flexural strength values met the strength requirements of 0 to 00 psi reported by Van Dam et al. () for the to -hour concrete repair materials. The VES-FRCs with S fibers had better post-cracking performance than the VES-FRCs with PP fibers, with deflection hardening behavior and higher residual strength and toughness values. Toughness values, T D 0, were almost double for the S fiber samples (0 inch-pounds) compared to the samples with PP fibers (0 inch-pounds). The flexural capacity, represented by the equivalent flexural strength D ratio (), R T,0, of the VES-FRCs with S fibers was % on average after the first-peak, whereas the capacity of VES-FRCs with PP fibers was about % on average.

11 Khakimova, Ozyildirim, Harris (a) 0 (b) FIGURE VES-FRCs with PP (left) and S (right) fibers for mixtures with (a) SF and (b) FA. VES = very early strength; FRC = fiber-reinforced concrete; PP = polypropylene; S = steel; SF = silica fume; FA= fly ash. High paste and cement contents for VES-FRCs result in higher shrinkage values. At months, the systems had length change values up to 0.0% on average, which satisfied the suggested limit of 0.0%; however, the shrinkage values after days of drying were greater than the recommended value of 0.0% (). VDOT uses a maximum value of 0.0% for low cracking bridge decks (). In addition, the paste content, amount of cement paste expressed as percent volume of the whole mixture (0), was determined. The values of 0. and 0. for VES- FRCs with SF and with FA, respectively, exceeded the recommended 0. paste content limit (). Stage II: Permeability of Cracked FRC Specimens Figure shows the CWP results for the cracked VES-FRCs. The CWP values follow approximately the same trend for all concretes, and the observed variation could be due to the irregularity of crack size and pattern. As expected, a large increase in the CWP for increased crack widths was observed. The increase of CWP by 0 times on average (except for the VES- FRC with FA and PP) was observed for crack widths increasing from 0. to 0. millimeter.

12 Coefficient of Water Permeability, k [cm/s] Khakimova, Ozyildirim, Harris The CWP values for the solid samples without cracks were on the order of - cm/s, corresponding to the typical CWP values of normal strength concrete (). The CWP values for crack widths greater than 0. millimeter were above - cm/s and for widths greater than 0. millimeter were above - cm/s. In this case, the corrosion of primary reinforcement attributable to leakage through cracks wider than 0. millimeter is highly probable. Crack Width [µm] E E-0 E-0 E-0 E-0 E-0 0 E-0 VES-FRC w/ SF - S VES-FRC w/ FA - S VES-FRC w/ SF - PP VES-FRC w/ FA - PP FIGURE CWP versus crack width. Conversion factors: 0 µm = 0. mm = 0.00 in. VES = very early strength; FRC = fiber-reinforced concrete; SF = silica fume; FA = fly ash; PP = polypropylene; S = steel. Stage III: Fiber Distribution Analysis The image analysis of the cross sections with steel fibers indicated a tendency of long rigid fibers to align parallel along the length of the beam. This alignment is beneficial to the flexural and tensile capacities of the specimens because of a greater number of fibers at the beam crack face effectively resisting the applied normal stresses after cracking. The flexible PP fibers were more equally distributed in all directions. Figure illustrates the average fiber density for transverse and horizontal cross sections for S and PP fibers. (a) FIGURE Fiber density per cross-section side for HTH and THTHT: (a) S and (b) PP. H = horizontal cut; T = transverse cut; VES = very early strength; FRC = fiber-reinforced concrete; SF = silica fume; FA= fly ash; PP = polypropylene; S = steel. (b)

13 Khakimova, Ozyildirim, Harris The calculated K-function and F-function average values were compared to the theoretical functions values under the CSR condition, K-CSR and F-CSR, respectively. Here, values of the K-functions are presented in the form of ( K(s) )/. For the S fiber specimens, the VES-FRC with π SF samples displayed more fiber clumping compared to the samples with FA. The VES-FRC with FA displayed values closer to the K-CSR curve, indicating more of a random distribution (Figure a). The specimens with VES-FRCs with PP fibers displayed similar spatial fiber dispersion for both SF and FA mixtures, indicating some degree of fiber clumping (Figure b). Further, during mixing, concretes with PP fibers exhibited more fiber clumping compared to the mixtures with S fibers. From the F-function results it is apparent that all specimens displayed fiber clumping and some number of empty areas without fibers, with the slopes less steep than the F-CSR curves slopes. (a) (b) FIGURE Spatial Fiber Distribution Analysis for K- (left) and F-function (right): (a) S, (b) PP. K-CSR = K-function under CSR condition; F-CSR = F-function under CSR condition; VES = very early strength; FRC = fiber-reinforced concrete; SF = silica fume; FA = fly ash; PP = polypropylene; S = steel.

14 Khakimova, Ozyildirim, Harris CONCLUSIONS VES-FRCs with different fiber types and dosages and SF or FA can provide durable concretes, facilitate rapid and effective repairs, and reduce traffic interruptions and maintenance work. A compressive strength of,000 psi within hours can be achieved with the use of increased cement contents, a low w/cm, increased fresh concrete temperatures above 0 F, accelerating admixtures, and insulated curing. High residual strengths and resistance to cracking can be achieved with the addition of fibers. VES-FRCs with S fibers had superior post-cracking performance compared to VES-FRCs with PP fibers. VES-FRCs had high paste contents and high shrinkage values, making them prone to cracking. However, the addition of fibers is expected to facilitate crack control through residual flexural strength and toughness. Water permeability increases with increase in crack width. Cracks wider than 0. millimeter allow for the intrusion of harmful solutions. The long rigid steel fibers had a tendency to align parallel along the length of the beam, which is beneficial to the flexural and tensile capacities of the specimens; flexible PP fibers were more equally distributed in all directions. All FRC systems, especially the mixtures with PP fibers, had a tendency for fiber clumping to various degrees. ACKNOWLEDGMENTS The authors thank VTRC research and technical staff for their assistance and guidance in this study, specifically William Ordel, Michael Burton, Kenneth Herrick, Andrew Mills, Troy Deeds, Stephen Lane, James Copeland, Xuemeng John Xia, and Keith Peres. The authors also thank the industry for their input and assistance with this study. The authors thank the University of Virginia, specifically Muhammad Sherif, and personnel of VDOT s Materials Division and Structure and Bridge Division, VDOT s Staunton District, and the Federal Highway Administration. REFERENCES. Virginia Department of Transportation. State of the Structures and Bridges Report. Richmond, 0.. Ozyildirim, C. High-Performance Concrete for Transportation Structures. Concrete International, January,, pp... Kosmatka, S. H., and M. L Wilson. Design and Control of Concrete Mixtures. Portland Cement Association, Skokie, Ill., 0.. Parker, F., and W. L. Shoemaker. PCC Pavement Patching Materials and Procedures. Journal of Materials in Civil Engineering, Vol., No.,, pp... Punurai, S., W. Punurai, and C.-T. T Hsu. A Very Early Strength Concrete for Highway Construction. Journal of Testing and Evaluation, Vol., No., 00, pp... Sprinkel, M. M. Very Early Strength Latex-Modified Concrete Bridge Overlays. Transportation Research News, Vol., December, 00, pp... Klieger, P. Effect of Mixing and Curing Temperature on Concrete Strength. Journal of the American Concrete Institute, Vol., No.,, pp... Nawy, E. G. Fundamentals of High-Performance Concrete. John Wiley & Sons, Inc., New York, 00.

15 Khakimova, Ozyildirim, Harris Rappaport, J., C.-M. Aldea, S. P. Shah, B. Ankenman, and A. Karr. Permeability of Cracked Steel Fiber-Reinforced Concrete. National Institute of Statistical Sciences, Durham, NC, 00.. Ozyildirim, C., C. Moen, and S. Hladky. Investigation of Fiber-Reinforced Concrete for Use in Transportation Structures. Publication VTRC -R-. Virginia Transportation Research Council, Charlottesville,.. Aldea, C.-M., S. P. Shah, and A. Karr. Permeability of Cracked Concrete. Materials and Structures, Vol., No.,, pp. 0.. American Concrete Institute. Control of Cracking in Concrete Structures. Publication ACI R-0. Farmington Hills, Mich., 00.. Shah, S. P., and K. Wang. Microstructure, Microcracking, Permeability, and Mix Design Criteria of Concrete. Fifth International Conference on Structural Failure, Durability and Retrofitting,, pp. 0.. American Concrete Institute. Report on Fiber Reinforced Concrete. Publication ACI.R-. Farmington Hills, Mich., 00.. Granju, J.-L., and S. Ullah Balouch. Corrosion of Steel Fbre Reinforced Concrete from the Cracks. Cement and Concrete Research, Vol., No., 00, pp... Mangat, P. S., and K. Gurusamy. Permissible Crack Widths in Steel Fibre Reinforced Marine Concrete. Materials and Structures, Vol. 0, No.,, pp... Akkaya, Y., J. Picka, and S. P. Shah. Spatial Distribution of Aligned Short Fibers in Cement Composites. Journal of Materials in Civil Engineering, Vol., No., 000, pp... Akkaya, Y., A. Peled, and S. P. Shah. Parameters Related to Fiber Length and Processing in Cementitious Composites. Materials and Structures, Vol., No., 000, pp... Akkaya, Y., S. P. Shah, and B. Ankenman. Effect of Fiber Dispersion on Multiple Cracking of Cement Composites. Journal of Engineering Mechanics, Vol., No., 00, pp.. 0. Portland Cement Association. Fiber Reinforced Concrete. Skokie, Ill.,.. Kang, S. T., and J. K. Kim. The Relation Between Fiber Orientation and Tensile Behavior in an Ultra High Performance Fiber Reinforced Cementitious Composites (UHPFRCC). Cement and Concrete Research, Vol., No., 0, pp. 0.. Kang, S. T., and J. K. Kim. Investigation on the Flexural Behavior of UHPCC Considering the Effect of Fiber Orientation Distribution. Construction and Building Materials, Vol., No., 0, pp... Kang, S. T., B. Y. Lee, J.-K. Kim, and Y. Y. Kim. The Effect of Fibre Distribution Characteristics on the Flexural Strength of Steel Fibre-Reinforced Ultra High Strength Concrete. Construction and Building Materials, Vol., No., 0, pp. 0.. ASTM International. ASTM Standard C0/C0M-: Flexural Performance of Fiber- Reinforced Concrete (Using Beam with Third-Point Loading. West Conshohocken, Penn., 0.. Virginia Department of Transportation. Virginia Test Method 0: Method of Test for Measurement of Permeability of Bituminous Paving Mixtures Using a Flexible Wall Permeameter - (Asphalt Lab). Richmond, 00.. Diggle, P. J. Statistical Analysis of Spatial Point Patterns. Hodder Education Publishers, London, 00.. Virginia Department of Transportation. 0 Road and Bridge Specifications. Richmond, 0.

16 Khakimova, Ozyildirim, Harris. Van Dam, T. J., K. R. Peterson, L. L. Sutter, A. Panguluri, J. Sytsma, N. Buch, R. Kowli and P. Desaraju. Guidelines for Early-Opening to Traffic Portland Cement Concrete for Pavement Rehabilitation. Publication NCHRP Report 0. Transportation Research Board, Washington, D.C., 00.. Babaei, K., and A. M. Fouladgar. Solutions to Concrete Bridge Deck Cracking. Concrete International, Vol., No.,, pp American Concrete Institute. ACI Concrete Terminology. Publication ACI CT-. Farmington Hills, Mich., 0.. Darwin, D., J. Browning, and W. D. Lindquist. Control of Cracking in Bridge Decks: Observations from the Field. Cement, Concrete, and Aggregates, Vol., No., 00, pp. -.