Safety and preservation of cultural heritage stone masonry buildings after fire events – POSTFIRE

Material-scale fire tests: Lutetian limestones (Saint-Vaast, Saint-Leu, Sébastopol, Saint Maximin Ferme/Franche, Liais de Ouachée, Croix Huyart) are compared with other stones (Tuffeau, Savonnière, Tervoux, Euville, Lens, Massangis). Classified by compressive strength (NF DTU20.1), their petrographic and thermal profiles are mapped by SEM, XRD, polarized microscopy, MIP, and TGA/DSC. Small-scale cylinders characterize mechanical properties (compressive/tensile strength, E modulus, v) in initial state and after 200, 400, 600, and 800°C. Porosity, thermal conductivity, water absorption, and thermal expansion coefficients are also measured.

On wallet scale: Blocks of low (Saint-Leu), medium (Tervoux), and high (Massangis) strength are used to build wallets with lime-based mortars (pure or cement-added), matching stone strength. Wallets (56×36×10 cm) are exposed unidirectionally to fire at slow and fast heating rates. Deformation and cracking are monitored with thermocouples and strain gauges on the unexposed side. Residual compressive strength is measured after heating. Damage progression and crack patterns, including thermally induced cracks and color changes, are mapped through the thickness.

Fire tests on full-scale walls: The stone-mortar types, firstly tested at wallet scale, are used to build full-scale walls (3×3×0.2 m). Each stone type is tested once without mechanical load, once under 50% of allowable load (Eurocode 6). Walls are exposed to EN 834-1 fire curve for 120 min, then cooled (24 hr). Extensive instrumentation (thermocouples, displacement sensors, DIC, thermal cameras, endoscopes) monitors heating and cooling. After cooling, walls are loaded to failure to assess residual capacity.

Numerical simulations: A multi-scale approach uses CODE_ASTER, integrating elastoplastic behavior at mesoscale and structural analysis at macroscale to model thermal effects. The methodology includes parametric studies, boundary condition evaluation, and thermal transient deformation incorporation, improving prediction accuracy. Simulation outputs (reaction kinetics, strength and deformability) are calibrated and validated by experimental data. Masonry-scale models account for thermal gradients at stone-mortar interfaces across different geometries and typologies, with wall stability assessed under varying conditions.

Notre-Dame de Paris case study: A meta-analysis reviews detailed reports, material databases, and bibliographic resources from the Notre-Dame Cathedral fire (2019). It correlates field observations and extracted cores with material- and structure-level lab tests, validating POSTFIRE’s universal methodology.

Recommendations for post-fire assessment: A report compiles research results and consortium discussions, providing tabulated data on stone thermal responses, damage profiles, and assessment protocols. It proposes a universal methodology and criteria to support future research and industrial practice for cultural heritage protection.

At material scale, calcite-rich limestones expand initially but contract at high temperatures due to decarbonation, causing up to 44% mass loss beyond 800°C. In quartz-bearing limestones, thermal expansion mismatch and the α-β phase transition at 573°C induce internal stresses. Thermal sensitivity is influenced by petrophysical factors: coarse grains amplify thermal expansion and stress, while strong cementation with low fine porosity intensifies boundary stresses. Beyond 600°C, thermal cracking connects previously inaccessible pores, causing pore network expansion and a higher capillary absorption coefficient. All limestones weaken at high temperatures, with tensile strength dropping from 200°C, compressive strength from 400°C, and stiffness loss with increased deformability after 600°C. Large grain sizes amplify deformation and microstructural damage.

At block scale, high heating rates create strong thermal gradients, causing differential expansion and macro-cracks. The heated region expands but is constrained by the cooler part, generating compressive stress on the hot face and tensile stress on the cold face. This is confirmed by LRMH’s report on the Notre-Dame de Paris fire, highlighting subsurface cracking at the compression-tension transition zones, indicated by colour changes and stiffness loss. Unidirectional heating of masonry wallets to 650°C showed premature mortar detachment from stone blocks, attributed to a 20% loss of stiffness and tensile strength in mortars at 200°C. Hard stones with low porosity may initially resist, but poor physico-chemical bonding leads to detachment.

At masonry scale, all walls achieved an REI 120 fire resistance rating. After one hour of exposure, thermal gradients showed temperatures below 100°C beyond 5 cm depth and 20°C on the non-exposed face. This caused thermal curvature in both horizontal and vertical directions, with maximum deflections at the center, leading to a circular bulge toward the fire. The resulting deformation, up to 20% of the wall thickness, induced load eccentricities and affected structural stability. This contrasts with the APL-POSTFIRE, showing that stone masonry walls cannot be calculated using Eurocode 6 tabulated values or methods for required fire resistance. Moreover, French CCH regulations lack specific criteria for estimating fire resistance of stone masonry, considering it non-combustible.

Numerical simulations showed that geometry and block size influence plastic deformation distribution and thermal stress dissipation. Block slenderness affects plastic deformations and thermal stress management. Boundary conditions play a key role in predicting wall behaviour. Parametric analyses highlighted that tensile strength and Young’s modulus are critical for thermomechanical behaviour. Finally, integrating transient thermal deformation improved simulation accuracy, especially for loaded walls.

POSTFIRE validated an integrated methodology that reports on stone masonry damage in fire, however, the damage phenomenology merges the impact of both firing and cooling, therefore, online monitoring of both fire testing and cooling phase will provide an insight view of the damage kinetics. Further testing on masonry walls should consider the ISO 834 fire curve as a baseline, simulating the sudden temperature rise during a fire, followed by rapid water cooling that mimics firefighter intervention during fire extinguishment.

The stone-mortar compatibility should be further explored by testing a larger gamma of historic mortars, and masonry building practices (e.g. ribbing/grooving, surface treatment, consolidation, etc.).

Fatigue phenomena should be explored beyond the linear fire-extinguishing scenario, adding weathering and durability schemes at the testing procedure.

An integrated strain gauge/LVDT and digital image correlation methodology should be coupled with ultrasonic and other NDT solutions in order to provide a full-field mapping of damage progress.

Both experimental and numerical methodologies should be further tuned by integrating real fire case scenarios.

Numerical predictive models could explore the importance of boundary conditions, the evolution of stone properties with temperature, and transient thermal deformation in predicting the thermomechanical behavior of walls. This study provides a solid foundation for improving numerical models and forecasting the behavior of limestone structures in fire situations.

Repair and consolidation of stone masonry at post-fire state should be tuned taking into account the ethics and value of built heritage.

Sciarretta, F., Eslami, J., Beaucour, A.L., Nomowé, A., State-of-the-art of construction stones for masonry exposed to high temperatures, Construction and Building Materials, 2021, 304, 124536.
DOI: 10.1016/j.conbuildmat.2021.124536
HAL: hal-04383399v1
Cited 28 times (Google Scholar)

Vigroux, M., Eslami, J., Beaucour, A.L., Bourgès, A., Noumowé, A. High temperature behaviour of various natural building stones, Construction and Building Materials, 2021, 272, 121629.
DOI: 10.1016/j.conbuildmat.2020.121629
HAL : hal-04383418v1
Cited 47 times (Google Scholar)

Vigroux, M., Sciarretta, F., Eslami, J., Beaucour, A.L., Bourgès, A., Noumowé, A., High temperature effects on the properties of limestones: post-fire diagnostics and material’s durability, Materials and structures, 2022, 55 (10), 253.
DOI: 10.1617/s11527-022-02086-5
HAL: hal-04383410v1
Cited 9 times (Google Scholar)

Daoudi, A., Eslami, J., Beaucour, AL., Vigroux, M., Noumowé, A., 2023. High-temperature behaviour of various limestone used in heritage buildings at material and block scales, AIP Proceedings 2848 (1)
DOI: 10.1063/5.0149082

Daoudi, A., Eslami, J., Beaucour, A.L., Vigroux, M., Noumowé, A., High-temperature behaviour of various limestones at the block scale. Construction and Building Materials, 2024, 416, 135109.
DOI : 10.1016/j.conbuildmat.2024.135109
HAL : hal-04753693v1
Cited 4 times (Google Scholar)

Obaei, A., Eslami, J., Beaucour, A. L., Hoxha, D., Noumowé, A., Pimienta, P., Pham, D. T. (2024). Étude du Comportement Thermomécanique des Murs en Maçonneries de Pierres Calcaires. Academic Journal of Civil Engineering, 42(1), 607-618.
DOI : 10.26168/ajce.42.1.53

Daoudi, A., Sciarretta, F., Eslami, J., Beaucour, AL., Noumowé, A., 2024. Experimental Characterisation of Physical, Thermal, Transport and Mechanical Properties of 13 French Limestones. Masonry Research and Innovation, submitted 22/01/2025

Daoudi, A., Ngoudjou, V., Eslami, J., Beaucour, A.L., Pani, C., Noumowé, A., High-temperature performance of limestone masonry: investigations at the material and assembly scales. Construction and Building Materials, submitted 31/03/2025

Daoudi, A., Eslami, J., Beaucour, A.L., Henin, J., Vigroux, M., Noumowé, A., Comprehensive study of the post-heating mechanical performances and pore network changes of a wide variety of limestones: influence of petrographic and thermoelastic properties, Engineering Geology, submitted 04/12/2024

Obaei, A., Pimienta, P., Eslami, J., Pham, D.T., Beaucour, A.L., Noumowé, A., Experimental Investigation of the Thermo-Mechanical Behavior of Full-Scale Limestone Masonry Walls Exposed to Fire, Engineering Structures, submitted 19/03/2025

The POSTFIRE project addresses the vulnerability and resilience of urban systems, with the main objective of facilitating the recovery of cultural heritage buildings after fires. The project will investigate the behaviour of historic stone masonry after high temperature exposure at the material and structural scale, accounting for the effect of water quenching. The project will investigate the relations between microscopic phenomena and property changes; it will create a property database for the selected materials in post-fire conditions, and material models of immediate applicability in analytical and computer-based methods for structural performance assessment and calculation. The reliability of the database and models will be tested at the structural scale. Finally, the project will propose post-fire assessment guidelines for heritage buildings of France, to extend the benefits of the project among academy, professionals, and national and international normalisation committees.