As ultra-high-performance concrete (UHPC) becomes more widely used, the need for good concrete tensile tests for fiber-reinforced concrete has become apparent. However, there is not yet one universally used test method. This research compares the results between standard direct tension, four-point flexure, and double-punch tests. Each test method was run on batches of UHPC made with two fiber types each at five different doses by percent volume, resulting in a testing matrix of 10 different mixes. The results of each test method were plotted to see how well different outcomes, such as peak strength and toughness, would reflect differences in fiber content. The results of the two indirect tension tests were then compared to the results of the direct tension test to see how well they correlated. Results showed that the average maximum stresses of each test method did not have good correlation but that measures of toughness or postcracking strength did have reliable correlation between the direct tension test and both of the simpler test methods.

The quantity and orientation of steel fibers in a structural UHPC element will impact its ability to carry and distribute tensile stresses. UHPC is typically placed as a highly flowable material. As a result, the fibers in UHPC tend to align along the direction of flow, and in undesirable conditions may segregate. Both of these properties can significantly influence fiber alignment and thereby impact the local tensile behavior of the material. Quality assurance and quality control (QA/QC) methods for characterizing the quantity and alignment of fibers in UHPC elements are needed for assessing the impact of the flow properties and placement methods on the structural performance of a UHPC element. Accordingly, two techniques have been independently developed to characterize fiber distribution and alignment in UHPC cores extracted from structural members: (1) three-dimensional imaging of concrete samples by x-ray microtomography and (2) twodimensional imaging of individual cut surfaces. This paper presents a collaborative research effort in which the fiber quantity and fiber orientation of ten samples of UHPC, extracted from four different full-scale UHPC members, were characterized using both the three-dimensional and twodimensional imaging techniques. Parallel analyses were performed to quantify the total fiber content and "axial alignment factors" for each sample using each method, and the results were compared. The results indicate that while three-dimensional imaging by x-ray microtomography of fiber alignment provides a comprehensive assessment of fiber alignment, two-dimensional analysis of cut planes can provide a reasonable estimate for use as a QA/QC method or as a tool to investigate or explain UHPC performance, provided that results are appropriately normalized.

UHPC has been found to have excellent freeze–thaw durability in cold regions. Previous UHPC testing performed has mostly focused on concrete with compressive strength above 21 ksi (145 MPa). In this study, testing was conducted to determine at what strength level concrete transitions to provide excellent freeze–thaw (F–T) performance. Non-proprietary concrete samples were made for freeze–thaw durability from four different concrete mixture designs: 12–15 ksi, 15–18 ksi, 18–21 ksi, and 21+ ksi (83–145+ MPa), and these were tested according to ASTM C666, using 1.5% steel fibers. The samples were made for three different curing regimens: limewater curing in a fog room, simulated precast curing, and steam curing. Low-temperature differential scanning calorimetry (DSC) and mercury intrusion porosimetry (MIP) tests were carried out to reveal the freeze–thaw mechanism of the concrete samples. All mixtures with compressive strength above 15 ksi (103 MPa) performed excellent in freeze–thaw testing with no damage seen. Steam curing was found to negatively affect the freeze–thaw performance at the lowest strength level tested.

This paper studies and compares several machine learning algorithms to characterize and cluster different types of concrete based on their ultrasonic response. We focus on identifying properly fabricated ultra-high-performance concrete (made with extremely fine aggregate and containing thin steel fibers). Ultra-high-performance concrete achieves much higher strength characteristics