Large-scale Implementation of Semi-proprietary UHPC with virgin and recycled fibers for full precast bridge columns

Project Information 

Link to Latest Report: Coming Soon!


The use of ultra-high performance concrete (UHPC) is currently expanding worldwide from bridge deck joints and overlays or small architectural applications to full components and larger applications. In the US, the FHWA is developing mega bridge girders up to 300-ft spans (El[1]Helou et al. 2022) while also developing the first AASHTO design guidelines for UHPC bridge applications. Another large initiative funded by PCI developed non-proprietary UHPC mixes for precast/prestressed girders (Tadros et al. 2022). Such efforts will help expand the UHPC market and encourage researchers to look at other applications such as full UHPC columns and their applications for precast construction especially in seismic areas. To contribute towards filling the general knowledge gap on columns, our team at UNR has completed several projects over past 5 years (e.g. Aboukifa and Moustafa 2022; Aboukifa et al. 2020, and Subedi et al. 2019) that focused mostly on fundamental axial behavior of UHPC columns and accelerated bridge construction (ABC) connections. However, more work is still needed to scale-up columns construction and investigate structural and seismic behavior of full precast UHPC bridge columns and different ABC column-to-footing connections. Seismic columns are of particular interest since UHPC has been previously considered for retrofit applications and previous studies (e.g. Aboukifa et al. 2020) demonstrated UHPC seismic columns as viable candidates for future resilient and state-classified important bridges.

Meanwhile, with the increased interest in larger and full structural UHPC applications, economy and sustainability become of paramount significance. As such, we just started working with a major UHPC vendor (Cor-Tuf) to develop semi-proprietary UHPC mixtures with local sand and cement along with recycled steel fibers from landfill tires, which is a major initiative that we will leverage in this project. If demonstrated and implemented successfully, recycled steel fibers could be the next big leap in the UHPC world. This is because not only a pound of recycled fibers is about 25 cents versus ~2.5 dollars for virgin or manufactured fibers, but also expanding tires recycling and clean landfill operations is of a great environmental benefit. It is noted that the steel industry is among the largest contributors of CO2, making up 8% of the total carbon emissions according to the DOE (2023). Federal agencies like the DOE have created roadmaps to achieve decarbonization by improving material composition and increasing material lifespan (DOE 2023). Thus, one way to contribute to the easement of the carbon footprint associated with UHPC steel fibers is the use of recycled steel wires/fibers from landfill tires. It is noted that while old and worn tires from cars and trucks contribute to increasing waste in landfills, tire recycling can help with clean landfill operations and prevent incinerations that can release toxic pollutants and carbon into the atmosphere. In used tires, 12-21% of the overall tire composition is steel fibers (US Tires 2023). Technology such as pyrolysis and shredding allows used tires to be decomposed, separating the steel fibers (Williams 2013). The steel fibers generated from this process can be recycled and efficiently used as an alternative to manufactured fibers in UHPC to significantly reduce material costs and decrease carbon emissions towards a more sustainable and net zero carbon UHPC.

In late 2022, we started performing several UHPC trials with different types of recycled fibers from landfill tires and then tested a pair of large-scale axial columns to compare the behavior of UHPC with high-end manufactured fibers versus the recycled ones. Figure 1 shows a sample of the different fibers along with pictures and preliminary results from previous UHPC buildings columns tests at UC Berkeley, which we completed using couple ACI-funded projects. The material tests showed that only tensile behavior of UHPC with recycled fibers might not be at exact same level as virgin manufactured fibers. Nevertheless, at the structural level, since axial columns are less dependent on tensile behavior, the UHPC column with recycled fibers almost outperformed the column with manufactured fibers which is very promising and in turn, motivates this project.


The specific objectives of this project are to:

  1. Collaborate with Con-Fab Precast in CA to implement UHPC at scale and fabricate six to eight full precast UHPC columns. Half of the columns will be axial columns to be tested under pure axial loading at UC Berkeley, and the other half will be seismic columns to be built using both recycled and virgin fibers and assembled with conventional RC footings using ABC connections;
  2. Expand an ongoing exploratory ABC-UTC project from cycle 5 and study two different ABC seismic connections, grouted ducts and socket, for precast UHPC columns;
  3. Conduct both pure axial and seismic testing for the various specimens to investigate and compare, for the first time, the structural response of precast UHPC columns with virgin and recycled fibers for future bridge columns.


TASK #1 – Material characterization and testing

A robust semi-proprietary UHPC mix from COR-TUF with local CA sand and cement will be used along with manufactured steel fibers (acquired from HiPer Fiber) and recycled steel wires (acquired from Liberty Tire Recycling). The first task is to fully characterize the material and mechanical properties of UHPC with different volumetric ratios of both types of fibers. In anticipation of the full columns construction, the material samples will be obtained from trial batches at Con-Fab Precast. We have established a working relationship with Con-Fab to introduce UHPC to their plant, and this collaboration will be leveraged in this project. The material tests include instrumented compression tests to obtain full stress-strain behavior in addition to flexural and direct tension tests, and will be used later to interpret results from all of the different structural tests; both the axial loading tests as well as the seismic loading tests.

TASK #2 – Specimens construction and assembly

As mentioned above, six to eight columns will be constructed at Con-Fab using either a 6 cubic[1]yard pan mixer or ready mix trucks (see Figure 2) using different types of steel fibers to further UHPC mixing practicality and scalability. For the axial columns, only precast columns will be considered without footings, and such columns will be shipped to Berkeley for testing as explained more in Task 3. For the seismic columns, both precast UHPC columns and conventional RC footings will be delivered to UNR where we will mix UHPC for the connections and assemble the final test specimens. The proposed test matrix and preliminary dimensions are illustrated below in Figure 3 and discussed later in Task 4. It is noted that selected columns rebars will be instrumented using strain gages before construction. It is also noted that for the seismic columns, two ABC seismic column-to-footing connections, that are selected based on recent AASHTO design guidelines (Saiidi et al. 2020), will be used to connect the full precast UHPC columns to conventional RC footings as illustrated in Figure 3. The selected connections are: grouted ducts (UHPC-filled ducts) and socket connections with UHPC interface.

TASK #3 – Axial columns testing at UC Berkeley

Through two recently completed ACI-funded projects, the PI successfully collaborated with all major UHPC vendors in the US, i.e. Ductal from LafargeHolcim, carbon nanofibers enhanced UHPC from CeEntek, Steelike UHPC, and Cor-Tuf UHPC, to test about 25 columns for buildings under axial loading. In addition to the varying UHPC mixture type, transverse reinforcement detailing as well as columns slenderness ratio and stability limits were varied. Figure 4 shows the 4000-kip testing facility at UC Berkeley which was used for those 25 columns tests and is proposed to use again in this project for testing 3-4 axial bridge columns. The precast UHPC columns will be shipped directly from Con-Fab to Richmond, CA where the UC Berkeley structures lab is located. Both spirals versus tie/hoop transverse reinforcement will be considered to better understand the behavior of confined UHPC bridge columns. All tests will be extensively instrumented using axial string potentiometers and rebars strain gages to properly capture the axial behavior and stiffness of the tested columns, which will be needed to inform the design of future bridge UHPC columns.

TASK #4 – Large-scale seismic testing at UNR

Once the construction and assembly of the seismic columns is completed in Task 2, extensive experimental testing will be pursued. The 3-4 seismic columns specimens will be tested at UNR Earthquake Engineering Laboratory either under combined axial and quasi-static cyclic loading (similar to previous Caltrans test illustrated in Figure 5) or at one of our shake tables under dynamic uniaxial ground motions. The decision to conduct either quasi-static or dynamic tests will depend on final budget, time limitations, and lab schedule. Various loading protocols will be considered as well as part of finalizing the most feasible seismic testing method. Regardless of the test type (qusi-static or dynamic), our team has conducted tens of seismic large-scale tests in the past 8 years and in turn, this task is expected to be conducted smoothly with no foreseen challenges. Comprehensive instrumentation will be used to capture the comparative structural and seismic behavior of UHPC columns with different steel fibers types and ABC connections.

TASK #5 – Experimental Data Processing and Interpretation

The objective of this task is to carefully reduce and process the different types of data expected to be generated from the material characterization tests along with both axial and seismic structural tests. The data processing and interpretation will aim at understanding the structural and seismic response of the UHPC columns as well as ABC connections. For example, data from the seismic tests will be used to estimate lateral flexural stiffness of UHPC columns and how it relates to the modulus of elasticity obtained form material tests, and such tests will also provide new knowledge on stiffness degradation at different drift demands. Strain gage data will be used to track longitudinal and transverse reinforcement yielding and rupture (if any). Global force[1]displacement relationships and hysteretic behavior will be obtained and studied as well to determine force and moment capacities. Overall, a good understanding of the confinement behavior of axial columns as well as the plastic hinge damage and ABC seismic connection performance will be established at the end of this task so that it can be summarized and reflected in a potential ABC guide.

TASK #6 – Results dissemination and Final report

A final report will be prepared and submitted first to the advisory panel for review and comments then a revised version will be widely disseminated through the ABC-UTC. The report will be complemented with ABC-UTC guide (see section 2.2.2) for the design guidance for precast UHPC columns and ABC seismic connections. At least two journal papers are expected to be produced from this project and will be submitted for potential publication in reputable peer[1]reviewed journals

Research Team:
Principal Investigator: Dr. Mohamed Moustafa