Periodic Reporting for period 1 - FiRe2C (Full-field experimental and numerical investigation of novel fire resistant fibre reinforced concrete for tunnel lining)
Reporting period: 2022-10-01 to 2024-09-30
Tunnel fires have attracted increasing attention in recent years, after a series of disastrous fires in road and rail tunnels. Some of the most catastrophic examples are the Mont Blanc tunnel fire in 1999 [39 fatalities - inoperable for 3 years], the Tauern bridge in the same year [12 fatalities - three-month closure] and the more recent one in China [31 fatalities, mostly children]. A key element in the spreading of a fire and its effect on the stability of an underground space is the design and performance of the concrete lining. Specifically, under fire, concrete linings spall (i.e. surface abrasions—explosive behaviour), resulting in the collapse of the tunnel structure, which causes a significant scourge in the economy, society, and the environment. Apart from the strength of the concrete (compressive and tensile), spalling is affected by the properties of the concrete such as permeability, age, mix homogeneity, size of aggregates, moisture content, pore pressure, pre-existing cracks, type of reinforcement (if any), size and shape of the exposed specimen.
In summary, the design of tunnel concrete linings is based on thermal calculations, which often ignore spalling. Based on such calculations new types of more durable, strong and hence denser concrete have been introduced on the market recently that are much more probable to spall due to their lower permeability. Reinforcing the concrete specifically with polypropylene fibres results in them melting during tunnel fires (fibres melt at approximately 160oC, before the concrete starts to spall) creating voids, through which water vapour can dissipate eliminating the expansion of pressure in the concrete and thus, spalling. However, the addition of fibres tends to decrease the workability and the environmental impact of the concrete and potentially its strength and durability too. The need to reduce CO2 emissions, or use sustainable materials to increase the life cycle of urban construction has led to the development of alternative types of concrete with properties such as self-compaction, self-healing, low-density, etc. Replacing limestone aggregates with lightweight ones, such as pumice, and therefore, introducing larger voids into the mortar mass, might be the solution to reducing spalling without using polycarbonate fibres, as well as making the concrete greener. The higher porosity values and subsequent lower density offer several advantages, including reduced construction costs, a high strength-to-weight ratio, a low coefficient of thermal conductivity and improved durability, yet cracking at early ages poses a significant concern for its structural use.
To enhance our understanding of the mechanisms involved in cracking and spalling it is necessary to employ full-field imaging techniques to study the size, distribution and orientation of the aggregates and pores within the concrete matrix and relate them to its thermo-mechanical response. This has not been done before and will contribute to the improvement of the design and performance of concrete linings. Specifically, in FiRe2C we have employed a holistic approach between state-of-the-art numerical (Discrete Element Method) and full-field experimental methods (X-ray computed tomography [XCT], utilising quantitative 3D experimental measurements at different length-scales to validate accurately the material response of the simulated material. Our main objectives were:
1) Manufacture a new type of high-strength, lightweight, fire-resistant concrete to minimise the need for additional costly fire protective measures (e.g. fireproof boards) and reduce the time and costs required for the restoration of a tunnel after a fire has occurred. The response of this material both under uniaxial compression and exposure to fire was compared against the response of normal-weight concrete and polycarbonate-fibre reinforced concrete.
2) Study experimentally the effect of size, distribution and orientation of the aggregates, pores and fibres on the strength and fire-resistance of concrete linings, by employing full-field imaging techniques (XCT) pre- and post-fire.
3) Create a novel DEM model to predict the mechanical response of concrete exposed to fire, utilising quantitative 3D experimental measurements to validate accurately the material response
In summary, the design of tunnel concrete linings is based on thermal calculations, which often ignore spalling. Based on such calculations new types of more durable, strong and hence denser concrete have been introduced on the market recently that are much more probable to spall due to their lower permeability. Reinforcing the concrete specifically with polypropylene fibres results in them melting during tunnel fires (fibres melt at approximately 160oC, before the concrete starts to spall) creating voids, through which water vapour can dissipate eliminating the expansion of pressure in the concrete and thus, spalling. However, the addition of fibres tends to decrease the workability and the environmental impact of the concrete and potentially its strength and durability too. The need to reduce CO2 emissions, or use sustainable materials to increase the life cycle of urban construction has led to the development of alternative types of concrete with properties such as self-compaction, self-healing, low-density, etc. Replacing limestone aggregates with lightweight ones, such as pumice, and therefore, introducing larger voids into the mortar mass, might be the solution to reducing spalling without using polycarbonate fibres, as well as making the concrete greener. The higher porosity values and subsequent lower density offer several advantages, including reduced construction costs, a high strength-to-weight ratio, a low coefficient of thermal conductivity and improved durability, yet cracking at early ages poses a significant concern for its structural use.
To enhance our understanding of the mechanisms involved in cracking and spalling it is necessary to employ full-field imaging techniques to study the size, distribution and orientation of the aggregates and pores within the concrete matrix and relate them to its thermo-mechanical response. This has not been done before and will contribute to the improvement of the design and performance of concrete linings. Specifically, in FiRe2C we have employed a holistic approach between state-of-the-art numerical (Discrete Element Method) and full-field experimental methods (X-ray computed tomography [XCT], utilising quantitative 3D experimental measurements at different length-scales to validate accurately the material response of the simulated material. Our main objectives were:
1) Manufacture a new type of high-strength, lightweight, fire-resistant concrete to minimise the need for additional costly fire protective measures (e.g. fireproof boards) and reduce the time and costs required for the restoration of a tunnel after a fire has occurred. The response of this material both under uniaxial compression and exposure to fire was compared against the response of normal-weight concrete and polycarbonate-fibre reinforced concrete.
2) Study experimentally the effect of size, distribution and orientation of the aggregates, pores and fibres on the strength and fire-resistance of concrete linings, by employing full-field imaging techniques (XCT) pre- and post-fire.
3) Create a novel DEM model to predict the mechanical response of concrete exposed to fire, utilising quantitative 3D experimental measurements to validate accurately the material response
Initially, a thorough lightweight concrete synthesis design campaign was carried out, where several different mixtures were created to reach appropriate levels of strength, confirm reproducibility and have comparable volumetric attributes (either amount of cement or grading of aggregates) with normal-weight concrete. Previous work of the supervisor defined the decisions regarding the limits of the concrete synthesis design. Once the right mixture was identified, cylindrical specimens were created and tested under uniaxial compression, and ultrasound pulse velocity before and after exposing them to 200, 400, and 600oC under controlled conditions. The specimens that were exposed to heat were sent to Laboratoire 3SR in France to be scanned by an X-ray computed micro-tomographer, from which 3D grey-scale images were acquired. The grey-scale images were then analysed using a combination of open-source software and bespoke Machine Learning algorithms, developed specifically for the scope of this project, revealing distinct patterns of post-fire failure and mechanisms responsible for it. The final part of the FiRe2C project involves the simulation of the mechanical response of lightweight concrete using the commercial software PFC3D to create a bonded model representing the real 3D structure of concrete. The only way this would be achievable was by utilising the results from the full-field experiments to create a realistic fabric (i.e. shape, orientation and position of aggregates, and shape and location of pores). The experimental findings were additionally used to calibrate the material parameters and validate the numerical results.
The 3D images enhanced our understanding of the distribution of fractures within the concrete matrix and the effect of pores and particle orientation on the evolution of failure. Fractures due to mechanical loading were mainly located in the upper half of the specimen and areas with fewer fibres (in the fibre-reinforced specimens). In the specimens created with a mixture of lightweight (pumice) and normal-weight (limestone) aggregates three different failure mechanisms are observed: a) breakage of pumice, b) breakage of cement along the boundaries of limestone aggregates, and c) breakage of a few of the larger limestone aggregates. Breakage, however, is mainly located around areas of high concentration in pumice where most large macroscopic pores are also located. Pumice has a lower strength in comparison with limestone aggregates, which explains why most fractures pass through pumice. The results also suggest that smaller particles and those with a higher aspect ratio break less, a phenomenon also observed in granular assemblies but not previously considered for concrete. There are also strong indications that require further research that the internal porosity of pumice is highly anisotropic. This is a significant finding, as the vibration necessary to create the specimens will orient the lightweight aggregates in such a way resulting in a highly anisotropic specimen. Whilst it is well known that anisotropy in porous rocks will influence the development of pre-peak damage in low confining stresses, it is not typically considered in concrete technology, yet the search for alternative materials due to the climate crisis will bring forward new considerations, such as anisotropy, in the way we design our concrete mixes.