Fiber reinforced concrete


Fiber reinforced concrete (FRC) is concrete with short steel or polymer fibers added during mixing. The fibers bridge the cracks that form under load, so the cracked concrete keeps carrying tension. The group tests the material in beams, round panels and full-scale members, back-calculates its tension response from those tests, and develops the procedures used to design with it.

A concrete beam on two roller supports in a test frame, loaded at midspan, with gauges mounted on the beam
A fiber reinforced concrete beam in the three-point bending test, with gauges mounted on the beam and under midspan.
0.12 mm/min
rate of crack mouth opening in the closed-loop three-point bending test
817
28-day beams in one ASTM C1609 four-point bending data set
2.2 times
overestimate of the residual tensile strength by an elastic calculation, against the back-calculated value
2 million
load cycles on a full-size track slab mock-up

Fibers in cracked concrete

Fiber-reinforced concrete (FRC) is extensively used as primary or secondary reinforcement in infrastructure projects. Fibers have several advantages in comparison to traditional reinforcement, which include increased ductility and durability while reducing the construction time and labor costs. FRC also improves the post-cracking behavior considerably, and increases energy absorption, defined as toughness, several fold.

The fibers are introduced during the concrete mixing process, so they are dispersed through the whole member. When the concrete around them, called the matrix, cracks, the fibers that cross the crack bridge it and continue to transfer load. The stress they carry is the residual tensile strength, which represents the load carrying capacity due to fiber reinforcing mechanisms beyond the first crack strength. The residual strength is a function of the number and random orientation of fibers crossing the cracked section.

Side of a concrete beam with a crack running up from a notch at the bottom, fibers visible across the open crack
A crack that has grown upward from the notch of a tested beam. Fibers span the opening between the two faces.

Fiber types

Steel fibers and polymer fibers such as polypropylene are both used, and the type and volume fraction of fibers affect the level of energy absorption. The full-scale track slab tests for the Phoenix light rail compared a twisted steel fiber 2.5 cm long at 15 kg/m3 (0.2% by volume) with a hooked steel fiber 6 cm long at 39 kg/m3 (0.5%). Recent studies highlight the promising performance of polymeric fibers in full-scale tests, approaching those of steel fibers.

Two single steel fibers laid beside rulers, a short twisted one and a longer one with hooked ends
The two steel fibers of the track slab tests against a ruler. The twisted fiber on the left is 2.5 cm long and the hooked fiber on the right is 6 cm long.

Flexural beam tests

To assess post-cracking residual tensile strength, flexural tests on prismatic specimens such as EN 14651, ASTM C1609, and JCI-SF4 are used. ACI 544.4R recommends conducting the ASTM C1609 beam test on a closed-loop system, in which the machine is controlled by a deformation measured on the specimen. In the three-point bending tests run by the group on notched beams, that deformation was the crack mouth opening displacement (CMOD), which was increased at 0.12 mm/min.

Critical parameters such as crack length and width in the localization zone, where the deformation concentrates at one crack, can be measured using the Digital Image Correlation (DIC) technique. It gives the strain over the face of the beam from a sequence of photographs. In an unnotched beam the strain field shows cracks starting at more than one section before one of them opens. A notch fixes a single crack at midspan and reduces the first peak and the variability in the test data. The scatter in the results of beam specimens is attributed to the properties of the matrix, bond interface, fiber length, orientation, and distribution throughout the concrete.

Plot of load against crack mouth opening for six notched beams tested at ages of 8, 16 and 36 hours
Load against crack mouth opening displacement in three-point bending for a mix with 10 to 12 lb/yd3 of macro fibers, tested at 8, 16 and 36 hours. After the first crack the load falls and then holds at the level carried by the fibers.
Two beams with a colour strain map on the side face, one with several strain bands and one with a single band above a notch
Strain measured by digital image correlation before the main crack opens. The unnotched beam shows several cracks starting along the span, and the notched beam shows one crack at the notch.
Two scatter plots of load against deflection for many beams, each with a mean response curve
Four-point bending results according to ASTM C1609 for 243 beams tested at stripping age and 817 beams tested at 28 days, with the mean response of each set.

Round panel tests

In the ASTM C1550 test a round panel rests on three supports and is loaded at its center. The group used it with notched beams to evaluate fiber reinforced shotcrete, a concrete that is sprayed in place, for the initial support of a 2000 m deep shaft at Resolution Mine, Arizona.

Round panels 1500 mm in span and 150 mm thick on a continuous support were tested with 80 and 100 kg/m3 of steel fibers. The material properties back-calculated from the 100 kg/m3 panels were used in a finite element simulation of a full-scale elevated slab, which predicted a load capacity 15.5% higher than the test result.

A round concrete panel resting on three pivoted supports in a test frame, seen from below, with a displacement gauge under its center
A round panel on three supports in the test frame, seen from below, with the gauge that measures the deflection at the center.

Design values from test results

The structural design of FRC sections is based on post-cracking residual strength provided by randomly oriented fibers inside the cementitious matrix. A bending test records load and deflection and not the tensile stress in the material, so the tension response is back-calculated. A stress-strain model with a few parameters is assumed, the load-deflection curve it predicts is obtained from closed-form moment-curvature solutions, and the parameters are adjusted until the prediction fits the experiment.

The stress obtained this way is lower than an elastic calculation gives, because the cracked zone carries a low and nearly constant stress where the elastic formula assumes a linear distribution. In a comparison by the group, a calculation that treats the cracked beam as linear elastic overestimated the residual tensile strength by 2.2 times. The procedure is described in ACI 544.8R, and the parameters are then used in design under ACI 544.4R.

Left, measured and modelled load against deflection for a beam. Right, stress through the depth of the beam from an elastic calculation and from the softening model
Left, the load-deflection curve of a beam and the fit of the model. Right, the stress through the depth of the cracked beam. The elastic calculation gives a linear distribution, and the softening model gives a low constant stress over the cracked zone.

Shrinkage cracking

Concrete shrinks as it loses water, and it cracks when the shrinkage is restrained. The group measured the effect of fibers with a ring test, in which concrete is cast around a steel ring and the strain in the steel is recorded as the concrete dries. The ring without fibers lost its strain at once when the concrete cracked, while the rings with alkali-resistant glass fibers cracked later and released the strain gradually. After 14 days the average crack width was 1.15 mm in the control samples and 0.341 mm with fibers.

Plot of strain in the steel ring against drying time for a control mix and three fiber mixes, beside photographs of the shrinkage cracks
Strain in the steel ring against drying time for the control mix and three mixes with alkali-resistant glass fibers, and the cracks in the control samples (left pair) and the fiber samples (right pair).

Applications

There are numerous advantages of using a fiber reinforced concrete system in applications such as elevated slabs, industrial floors, and structural thin sections. The key advantage is in the time reduction of the installation as the double layers of traditional rebars and stirrups can result in several weeks of construction time. For Salt River Project the group developed FRC specifications for cast-in-place, precast and shotcrete work, which covers canal linings and precast panels for water tanks.

On the Valley Metro light rail extension in Phoenix, steel fiber reinforced concrete replaced the two reinforcing bar mats of the track slab, and the slab thickness was reduced from 14.5 to 12 inches. Full-size mock-ups were tested under increasing load and under 2 million cycles of fatigue load. Fibers are also combined with reinforcing bars in hybrid reinforced concrete and used in precast tunnel lining segments, which have their own pages.

A long concrete slab section with a steel rail cast into its top, supported under a hydraulic actuator in a laboratory
A full-size track slab mock-up with the embedded rail under the actuator in the laboratory.
Workers at night placing concrete from a pump hose between formwork beside a rail
Fiber reinforced concrete being pumped into the track slab forms of the Phoenix light rail at night.

Closed-form solutions for flexural response of fiber-reinforced concrete beams

C. Soranakom, B. Mobasher (2007)

Journal of Engineering Mechanics, ASCE, 133 (8), 933-941

Journal Featured DOI Paper page

Correlation of tensile and flexural responses of strain softening and strain hardening cement composites

C. Soranakom, B. Mobasher (2008)

Cement and Concrete Composites, 30 (6), 465-477

Journal DOI

Backcalculation of residual tensile strength of regular and high performance fiber reinforced concrete from flexural tests

B. Mobasher, M. Bakhshi, C. Barsby (2014)

Construction and Building Materials, 70, 243-253

Journal DOI

Comparative evaluation of early age toughness parameters in fiber reinforced concrete

Mehdi Bakhshi, Christopher Barsby, Barzin Mobasher (2014)

Materials and Structures, 47, 853-872

Journal DOI

Sequential Cracking and Their Openings in Steel-Fiber-Reinforced Joint-Free Concrete Slabs

Xavier Destrée, Yiming Yao, Barzin Mobasher (2016)

Journal of Materials in Civil Engineering, 28

Journal DOI

Analytical procedures for design of strain softening and hardening cement composites

Yiming Yao, Xinmeng Wang, Karan Aswani, Barzin Mobasher (2017)

International Journal of Advances in Engineering Sciences and Applied Mathematics

Journal DOI

Closed-form solutions for flexural fatigue mechanical degradation of steel fiber reinforced concrete beams

Vitor Moreira de Alencar Monteiro, Daniel Carlos Taissum Cardoso, Flávio de Andrade Silva, Barzin Mobasher (2023)

Construction and Building Materials, 409, 134200

Journal DOI

Mechanical response of precast tunnel segments with steel and synthetic macro-fibers

Devansh Patel, Chidchanok Pleesudjai, Van Bui, Paul Pridemore, Steve Schaef, Barzin Mobasher (2023)

Cement and Concrete Composites, 144, 105303

Journal DOI

Back-calculation of mechanical properties of fiber-reinforced concrete in tunnel lining segments

Devansh Patel, Chidchanok Pleesudjai, Mehdi Bakhshi, Verya Nasri, Barzin Mobasher (2025)

Structural Concrete, 26 (5), 6019-6040

Journal DOI

Test footage

All videos

Fibers bridging a crack in tension

Close-up of a dogbone tension coupon, where fibers bridge the opening crack and distribute the damage.

Multiple cracking in four-point bending

The strain field resolves a series of parallel cracks across the constant-moment region as the load-deflection curve climbs.

Drop-weight impact on a fiber reinforced beam

High-speed camera footage of crack nucleation in a fiber reinforced concrete beam under impact.

Rail track slab in the test frame

A full-scale slab section with an embedded steel rail, loaded in the fatigue frame.