The inclusion of monofilament polypropylene (PP) fibres in concrete for fire resistance is not a new technology, with research and testing looking at concrete anti-spalling properties dating back to the mid 1980’s, for example Fibremesh 19851 and Schneider 1994. Eurocode 2: Design of Concrete Structures considers the use of polypropylene fibres for fire resistance, and recent projects such as Channel Tunnel Rail Link (High Speed 1) and A3 Hindhead have demonstrated that 1kg/m3 of polypropylene fibres is sufficient to meet the design requirements for resistance to hydrocarbon fires for tunnel linings.

All of the research and testing mentioned above to this present date, and to the authors knowledge has taken place on cast in-situ concrete elements such as pre-cast tunnel segments, cast in-situ secondary linings and slabs. In the past, cast in-situ secondary lined tunnels containing polypropylene fibres would have typically been used with a sprayed primary lining designed to take the short term load, and the secondary lining designed to take the long term load. However on Crossrail, the design philosophy is different, with the majority of station tunnels designed with a sprayed (SCL) steel fibre reinforced (SFRC) primary and secondary lining, which work compositely to take the long term load.

Steel fibre reinforced sprayed linings are also not a new technology, however sprayed concrete linings containing both steel and polypropylene fibres have historically rarely been used successfully and can cause issues surrounding pumping and placement of specific mix designs. On Crossrail the design philosophy removed this potential risk by introducing a finish 50mm regulating layer containing polypropylene fibres and calcareous aggregates and no steel fibres.

This report will look at the performance of the 50mm regulating layer against the requirements of the Crossrail specification, its resistance to spalling under fire when tested against Crossrail Specification KT244 in alignment with EFNARC5 fire testing guidelines. The paper will also look at the benefits of a composite lining containing a polypropylene regulating layer against a SCL steel fibre and polypropylene secondary lining.

Sprayed Concrete Lining Composite Design Philosophy
In recent years, the design philosophy for a sprayed concrete lined tunnel shell has changed. Originally a tunnel would have been typically designed with a temporary primary lining, sheet waterproof membrane and a cast in-situ secondary lining containing polypropylene fibres to take the permanent long term load and provide fire protection. However, due to improved technologies, materials and techniques more SCL tunnels are being designed with a SFRC primary and secondary lining and a spray applied waterproof membrane sandwiched between.

To ensure the sprayed concrete lining acts monolithically, there is a requirement for a bond between the spray waterproof membrane and the primary and secondary linings. There are several waterproof membranes on the market that can achieve this bond, as well as other testing parameters defined by the specifications.

To meet the Crossrail Specification (KT24) and EFNARC fire performance requirements, the design detailed the requirement for a 50mm regulating layer containing polypropylene fibres and calcareous aggregates (non steel fibre reinforced). This secondary lining mix containing both steel and polypropylene fibres, which was sprayed from a dry silo system and caused segregation, pumping and placement problems, resulting in a design change from SCL to cast in-situ secondary lining. There is very little information available detailing SCL secondary linings containing polypropylene and steel fibres from a wet mix batch source.

Mix Design and Constituent Materials
The specification required testing of concrete mixes, materials and processes similar to that being undertaken in the works, unless otherwise agreed. A brief description of the primary and secondary SCL mixes for each of the three Crossrail contracts can be found in Table 1 above.

The large scale fire test which will be discussed in the next section of this report required the testing of two steel fibre required the testing of two steel fibre reinforced and two non-fibre reinforced secondary lined panels. For each of the contracts, the non-fibre reinforced mix designs are that detailed in Table 1, without the inclusion of steel fibres.

The 50mm regulating layer for the finish of each of the panels differed between each contract in constituent materials; however each mix contained calcareous type aggregates and 1kg of monofilament polypropylene fibres per m3 of concrete for passive fire protection. The test method assessed the performance of the type of polypropylene fibres, fibre dosage, and aggregate type and bar reinforcement by looking at the thermal temperature at interface with the reinforcement.

The regulating layer also provided a finish free of steel fibres to the secondary lining.

Large Scale Fire Test
The large scale fire testing for each contract was carried out in accordance with Crossrail Specification KT24. The specification detailed large scale fire test panels tested to the Eureka temperature/time curve for 170 minutes as seen in Figure 2. The test method is intended to assess the temperature profile in the furnace, which provides information on the thermal transmission and its effect on:

i. the interface with the structural concrete

ii. the residual strength of the concrete

iii. the steel reinforcement placed

iv. the sprayed waterproof membrane

Large scale samples were manufactured by each contractor on site. The samples represented each concrete mix that would be used in the works. Specification KT24 detailed a minimum of four samples containing:

i. Two non-steel fibre (plain) with traditional reinforcement (i.e. representative of local reinforcement at junctions and thickening areas).

ii. Two steel fibre reinforced with 12mm reinforcement on a nominal 200mm grid at 50mm cover from secondary lining surface.

Each test sample was 800mm x 1700mm with an area of 800mm x 1000mm exposed to the furnace during testing. The large scale test panel and material arrangements were manufactured on site as detailed in Figure 3, unless otherwise agreed by designer Mott MacDonald and Crossrail.

All panels were stored on site and struck as per specification KT24, then shrink wrapped before transportation to the Building Research Establishment (BRE) in Watford (north of London) for humidity storage (40oC and 60 per cent humidity) for testing at 28 days.

Each panel from each of the SCL contracts under went the same process and procedure for testing at BRE as defined in the specification.

The acceptance criteria for resistance to fire testing for each panel as defined by Crossrail specification KT24 was:

i. The 50mm surface regulating layer was considered sacrificial. The depth of spalling to the main secondary lining should not exceed 25mm.

ii. The temperature of the waterproof layer shall not exceed a temperature at which it may degrade as defined by the manufacturer’s instructions.

iii. The compressive strength of samples after fire testing should not be less than 70 per cent of the original design compressive strength at 28 days (design strength at 28 days – C28/34).

iv. The temperature of the steel reinforcement should not exceed 450oC.

The spalling depth of the regulating layer and secondary lining, temperature at interface with waterproof membrane, residual compressive strength and reinforcement temperature was recorded during and after testing.

Conclusions
This paper details fire resistance testing of sprayed concrete panels representing the design of three different contracts on Crossrail. Based on the results and information contained in this report, it can clearly be seen that a 50mm regulating layer containing monofilament polypropylene fibres and calcareous type aggregates can provide passive fire protection to an SCL composite lining subject to a hydro carbon fire. The regulating layer should be extensively trialled to ensure there are no issues with pumping and placing of the material.

This report also aligns with other research papers as mentioned previously, which identify that 1kg of polypropylene fibres per cubic metre of concrete can provide adequate fire resistance and anti-spalling properties in the event of a fire.

The design of SCL composite linings is still relatively new with the primary and secondary linings working monolithically together. This report helps those involved in SCL tunnelling understand that a regulating layer of approximately 50mm in depth with the inclusion of polypropylene fibres ensures the composite structure is fire resistant. This knowledge will help avoid SCL secondary linings being designed with an over-specified quantity of polypropylene fibres which can cause potential problems with workability, pumping and placement. Providing the design requirements are the same for the primary and secondary linings, the primary lining mix could be used for the full thickness of the secondary lining as well, reducing the risk associated with producing and trialling differing mixes for the primary and secondary lining.

The regulating layer provided enough thermal resistance to ensure the steel bar reinforcement and waterproof membranes were not affected by the temperature. The regulating layer also provided enough resistance to ensure the structural integrity of the primary and secondary linings was not affected by the thermal transmissions, with all the post-fire test compressive strengths exceeding the required 70 per cent of design required.

On contract 1, panel 1A failed to meet the allowable <25mm spalling to secondary lining, exhibiting approx. 30-40mm of spalling. After an extensive review of the manufacturing, transportation and testing processes, it was identified by the contractor that the failure was caused by a number of factors:

i. There was a significant time lapse between the spraying of the secondary lining and the spraying of the 50mm regulating layer.

ii. The joint between the secondary lining and regulating layer was not adequately cleaned, to that extent that would typically be seen in construction.

iii. The location and orientation of the panels made it difficult to clean and remove complete debris and dust.

On contract 2, all samples met the specified acceptance criteria, however on the steel fibre free (plain) panels after test completion, they were visually inspected and a de-bonding between the secondary lining and regulating layer was observed.

The regulating layer was very friable and easily removed by hand, turning to dust particles. The secondary lining showed no spalling and remained unaffected, thus meeting the specification.

Although the specification was met, it can be questioned that if a fire were to occur in the actual tunnel environment, would the regulating layer remain intact in the crown of the tunnel, or would a risk be introduced for the structure after a fire with potential falling of the regulating material in the crown area?

This analysis is outside the remit of this paper; however it may provide a level of discussion if de-bonding is seen from future testing on various projects