Steel-fibre reinforcement has been used for many years in sprayed concrete in tunnels for temporary and even final lining. Multiple research studies and tests on the behaviour of steel-fibre-reinforced concrete (SFRC) have been carried out in recent years in various countries. They have greatly contributed to a better characterisation of SFRC, and have thus allowed a better understanding of the behaviour of this material and to specify minimum performance requirements for each project. The state of the art is well-known and lot of international standards provide clear guidance and performance criteria to use SFRC safely. Macrosynthetic fibre is also proposed today for different applications. Specific technical strength and weaknesses of the different fibres are often less wellknown, and lead to confusion. This paper discusses the important characteristics of steel- and polymer-fibre-reinforced sprayed concrete when used for ground support, and provides the latest test results from different laboratories.

Material properties

Young’s Modulus
The reinforcing ability of a fibre depends on the anchorage of the fibre into the concrete, the tensile strength and its Young’s Modulus.

The Young’s Modulus of concrete is typically 30,000MPa; of steel fibre it’s typically 210,000MPa and of polyolefin fibre it’s typically 3000-10 000MPa.

For well-anchored fibres, and equal demands on the fibre, the elongation of polymer fibre, and the corresponding crack width in concrete, might be considerably higher compared to steel fibres. This might have an impact on the durability of the concrete, especially in combination with traditional reinforcement.

Tensile strength
The tensile strength of steel wire is typically 1,000-2,000MPa compared to 300-600MPa for macrosynthetic fibre.

Specific density
The specific density of steel fibres is typically 7850kg/m3 compared to 910kg/m3 for polymer fibres, and 1000kg/m3 for water. Polymer fibres are light, which is favourable for health and safety, but they are lighter than water: the polymer fibres actually float on water, with potential risks for fibres at the surface.

Fire resistance
Metallic fibres have a neutral to positive impact on the fire resistance of structures. Due to a decreased spalling effect, a structure in metal fibrous concrete behaves rather better in the presence of fire than a mesh-reinforced structure for segmental lining, according to tunnelling specialists. Steel keeps its mechanical performance up to a temperature of 35°C to 40°C.

Macrosynthetic fibres, though, start to lose their mechanical properties as soon as the temperature reaches 5°C and even disappear at 16°C. In a fire a structure with macrosynthetic fibres alone soon becomes unreinforced, with no load bearing capacity left at all, and may result in an unsafe situation from the first hours onwards.

Polypropylene microfibres typically melt at temperatures around 16°C. Therefore these fibres (monofilament, length 6mm, diameter nominally <20µm) are proven to be suitable to improve the fire resistance. The exact reason is now fully-understood, as it’s generally accepted the microfibers start to melt in extreme fire conditions, thereby leaving small channels through which the pressurised vapour can escape. Consequently less damage from less spalling of the concrete is to be expected.

Macrosynthetic fibres melt at an equivalent temperature, but are not fine enough to provide the concrete under fire with the necessary network of channels. Moreover since the fibres melt, they are not suitable in those structures where the reinforcing effect of the fibres is important.

Oxidisation resistance
Polymer fibres don’t rust, even if the fibres are sticking out at the surface. Regarding metallic fibres, experience and research conclude:

• Steel fibres need only a concrete cover of 1-2mm compared to 30-40mm for normal rebar and mesh
• Corrosion of the fibres at the surface may cause discoloration but does not affect the mechanical properties of the SFRC structures
• Fibres in crack openings smaller than 0.25 mm do not corrode (see BRITE EURAM project)
• When no stains are required, galvanized steel fibres can be applied

Mix ability
Some macrosynthetic fibres tend to fibrillate during mixing. This fibrillation process can go on in the truck mixer until all fibres are completely destroyed. Quality degradation during mixing does not occur for steel fibres.

Concrete fibre content
European Standard 14721 specifies two methods of measuring the fibre content of metallic fibre concrete.

Method A measures the fibre content of a hardened concrete specimen. Method B measures the fibre content of a fresh concrete specimen.

There is no method available for polymer fibres at present in order to meet the quality control requirement for many projects.

Waterproofing membrane
Double-shell tunnel construction includes a waterproof membrane in between the sprayed concrete support and the final cast in situ concrete layer combined with a geotextile as a buffer to level out the irregularities of the sprayed concrete.

There was a concern about the danger of steel fibres protruding from the sprayed concrete surface to punch through the waterproof membrane.

In 1993 CETU (the French tunnel administration) had already financed puncture tests (under hydraulic pressure) on the geomembrane (600g/m2) placed on fibre-reinforced sprayed concrete support. These revealed the importance of the lower protective geotextile, without, however, revealing puncturing risks when it was placed between the PVC geomembrane and the fibre-reinforced sprayed (at that time Dramix fibres type RC65/35BN were tested).

As other tests were performed and with practical experience on many sites, test results confirm clearly that there are no problems with the membrane/protection sheets in combination with steel-fibrereinforced sprayed concrete.

However for some projects 30mm of non-fibre concrete are applied. In fact the key issue remains the irregularities of the sprayed concrete. It should be noted that several experiments have been carried out on the use of macropolymer fibres in sprayed concrete. The presence of these polymer fibres may increase the roughness of the support irregularities formed by the fibre/concrete mix.

Recommendation on this subject remains in discussion.

Properties of fibre concrete
Fibre concrete is well-known for its ductility. The effect of fibres is a combination of reinforcement and networking. Steel fibres in particular change the behaviour of the concrete: steel fibres transform a brittle concrete into a ductile material, which is able to withstand fairly large deformations without losing its bearing capacity. Ductility means load redistribution and a higher bearing capacity of the structure with the mechanical properties of the basic concrete material unchanged.

Usual performance criteria
The test plate usually used (600 by 600 by 100mm panels) (see European Standard EN 14.488-5) is designed to determine the energy absorbed from the load/deflection curve. Slabs intended for the punch-flexure test shall be made in receptacles of 600 by 600 by 100mm. In this case care will be taken to obtain an even surface and a thickness of 100 mm.

Spraying shall be carried out rigorously under the same conditions as recommended for the works: constituents, machine, lance holder and spraying methods in particular.

This approach tries to simulate the real lining behaviour. It gives a good idea of the load bearing capacity and the energy absorption of a sprayed concrete lining.

Instead of determining a materialcharacteristic, which requires a proper design model in order to calculate the allowable solicitation of the structure, the EN plate test approach allows skipping that step and immediately checks the energy absorption and the load bearing capacity of the lining.

It has to be stated very clearly that the statically indeterminate slab test is a structural test to check the behaviour of a construction. It is not a test to determine material properties to be used as design values. Based on this plate test, three SFRSC classes (E500, E700, and E1000) are defined for a C30/37 mix:
• 500 Joules for sound ground/rock conditions
• 700 Joules for medium ground/rock conditions
• 1,000 Joules for difficult ground/rock conditions

These values are proposed for the C30/37 class, usually specified for a temporary support. Compressive strengths with a too low and too high strength class may have undesired side effects.

In the case of higher compressive strength, the performance criteria proposed by the EN standard should be increased in order to keep the same level of ductility required for safety.

The plate test is also appropriate for a comparison of different fibre types and dosages. It allows for a comparison between mesh reinforcement and fibre reinforcement concrete, provided that the failure mode is the same according to EN 14 487-1 “Sprayed concrete, definition, specification and conformity.” That is why the performance criteria based on this test and the currently proposed test should only be used to compare steel mesh and steel fibres (material with same the same Young’s Modulus – E).

The relative importance of load carrying capacity at small crack widths—and hence small deflections and rotations—is, in recent times, assuming much greater importance to the designers of civil engineering tunnels.

Due to the very low Young’s Modulus – E of macrosynthetic fibres and the mode of failure observed with this type of fibre, the plate test is not sufficient to compare steel fibres and macrosynthetic fibres. In the case of polymer fibre another criteria should be added in order to have complete information, such as residual strength.

Residual strength
To determine the residual strength, the European EN 14651 is mainly used: Test method for metallic fibered concrete – Measuring the flexural tensile strength (limit of proportionality [LOP] residual).

This test procedure is mentioned in the final recommendation of Rilem TC162TDF “Test and design method for steel fibre reinforced concrete.”

This European Standard specifies a method of measuring the flexural tensile strength of metallic fibre concrete with a wide-beam moulded test specimen. The method provides for the determination of the LOP and of a set of residual flexural tensile strength values.

This testing method is intended for metallic fibres no longer than 60mm. The method can also be used for a combination of metallic fibres and, a combination of metallic fibres with other fibres.

The characterisation test enables the contractor who proposes a fibre-reinforced concrete (FRC) to check that this FRC satisfies the “mechanical” specification resulting from dimensioning.

In order to improve this approach, we could propose to follow the following requirements:
• The geometry and dimensions of the specimens, as well as the casting method adopted, should ensure distribution of the fibres in thematrix that is as close as possible to that encountered in the actual structure as spray concrete or flooring
• The dimension of the test specimen is acceptable for handling within a laboratory (no excessive weights or dimensions)
• The test is compatible, as far as the experimental means permit, with use in a large number of normally equipped laboratories (no unnecessary sophistication)
• The geometry should be the same as in the EN 14 488-5 plate test for energy absorption
• One geometry for isostatic and hyperstatic tests
• An easily managed test programme could use spraying on the job site with the same procedure as the plates test giving lower scatter than the beam test. These are results (Figure 2) for different CMOD (crack mouth opening displacement) according to EN 14 651 using: PP = Macro-polymer fibre dosage 6kg/m3; and SF20/30/40 = Steel fibre (Dramix RC65/35BN) at dosages of 20, 30 and 40kg/m3.

After the first cracking, the load bearing capacity of macrosynthetic fibre concrete dropped down about 60 per cent rapidly. This means that the 6kg/m3 fibres have lower influence on the residual strength than steel fibres.

A higher dosage of macrosynthetic fibres will have big influence on the concrete mix workability and pumpability.

Creep of fibre concrete
Sample plates have been tested in a displacement-controlled manner as described in EN 14488-57. At a deflection of 3mm the load was removed.

The plates were then ready to be subjected to the creep test and were reloaded with 60 per cent of the applied load at a deflection of 3mm. The deflection was measured and shown on the y-axis in 1/100mm increments as on the graph.

The consequence of creep is that the material will not provide significant reinforcement with the aim of stabilising the ground and minimising any future movement that may well be necessary.

Design rules
Since October 2003, Rilem TC 162-TDF1 design guidelines have been available for steel fibre concrete. No such guideline is available yet for macrosynthetic fibre concrete.

Quality control
As part of the production quality control wash-out tests are quite common in order to check the dosage of fibres in fresh concrete. This is possible when the fibres can be removed by a magnet, as is the case for steel fibres.

Conclusion
Steel fibre used for sprayed concrete has proven over the years to be a reliable construction material for tunneling applications. After 30 years of experience, the return of experience is very positive. Official international standards are now available.

Macrosynthetic fibres may be used in sprayed concrete support for some mining applications (perhaps in combination with mesh) or specific technical needs.

However, only steel fibres (no macropolymer fibres) can act as structural reinforcement of concrete for the following reasons:
• Polymer fibres melt at 165C; in a fire any reinforcing effect of the macrofibres fades away as the temperature rises.
• The Young’s Modulus is 3-10MPa, which is largely insufficient to reinforce concrete material with a modulus of 30MPa.
• Macro-polymer fibres creep (see elaboration above).

Clear test procedures and performance criteria should be specified for each project in order to meet technical requirements.


Figure 1-Three-point bending test for residual strength Figure 2 – Flexural strength results for crack-mouth opening displacement (CMOD) Figure 3 – Creep test with a square panel according to EN 14488-57 Figure 4 – Creep results on a square panel