Copyright © 2025 Elsevier Ltd. All rights reserved.

This manuscript version is made available under the CC-BY-NC-ND 4.0 license
[Link] (see: [Link] ).
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

Behaviour of stainless steel reinforced concrete beams in fire

Javaria Mehwisha, Rabee Shamassa and Katherine A. Cashellb

a
Department of Civil and Environmental Engineering, Brunel University London, UK.

b
Department of Civil, Environmental and Geomatic Engineering, University College London, UK

Abstract:

Stainless steel possesses excellent mechanical and physical properties, has outstanding corrosion
resistance, and is extremely ductile. It finds extensive use in the construction industry and has
increasingly been employed in reinforced concrete structures in recent years as there is greater focus
on longevity and life cycle costing. However, there has been limited to no research on the fire
resistance of stainless steel reinforced concrete structures. In this study, a finite element model of a
stainless steel reinforced concrete (SSRC) beam is developed and verified to simulate the fire
resistance of the SSRC beams. The model is validated through comparison with available test data
before it is employed to undertake a comparative study to evaluate the fire performance of carbon-
reinforced and stainless steel-reinforced concrete beams. Various factors, such as load ratio,
reinforcement ratio, and concrete strength, are analysed for their impact on the fire resistance of
these beams. The ductile failure observed in stainless steel-reinforced concrete (SSRC) beams during
fire incidents suggests their potential as replacements for carbon steel-reinforced concrete (CSRC)
beams, particularly in fire-prone areas where structural integrity is paramount. SSRC beams can
achieve comparable fire resistance to CSRC beams with lower reinforcement ratios, thereby reducing
reinforcement requirements potentially. It is shown that stainless steel reinforced concrete beams
deform more under flexure than equivalent carbon steel beams largely owing to their greater survival
time and resistance to greater temperatures, significantly affecting the fire resistance. A proposed
critical temperature limit of 765 °C for SS reinforcing rebar under fire exposure is recommended as a
design guideline.

1. Introduction

This paper is concerned with the fire behaviour of stainless steel reinforced concrete structural
elements. Traditional reinforced concrete members comprising carbon steel rebars embedded in
concrete are vulnerable to corrosion and durability challenges, especially in harsh environments.
Increasingly, owing to a combination of aging infrastructure and more challenging environmental
conditions, reinforced concrete (RC) structures may require ongoing inspection, maintenance and
rehabilitation activities during the life cycle. An alternative to carbon steel RC is stainless steel
reinforced concrete (SSRC), which is inherently corrosion-resistance. It is most likely to be specified in
marine, industrial, or urban environments where chloride ingress may be problematic. The
increasingly common use of de-icing salts on bridges, tunnels, and even car parks also cause severe
deterioration of infrastructure due to the corrosion of reinforcement. In addition to the corrosion
issues, these structural elements are also vulnerable to fire, and therefore the elevated temperature
behaviour of SSRC is important.

The fire behaviour of SSRC beams has not previously been studied and the behaviour is shown herein
to differ substantially from traditional carbon steel members. While stainless steel's mechanical,
physical, and corrosion-resistant properties are well-documented, there is no research exists on its
fire resistance in reinforced concrete (RC) structures. This paper addresses this gap by investigating
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

SSRC beams' fire performance, an area that is underexplored but increasingly relevant for modern,
durability-focused construction. Stainless steel reinforcement is generally available only in the
austenitic and duplex grades; the most commonly available include austenitic grades 1.4301, 1.4307,
and 1.4311 and duplex grades 1.4362, 1.4462, and 1.4162 [1]. These grades offer excellent mechanical
properties, including high strength and stiffness and exceptional ductility, toughness, and fatigue
properties. Whilst the austenitic grades have traditionally been more popular in load-bearing
applications, the duplex grades are becoming most frequently-specified owing to their combination
of economic and mechanical efficiency. The mechanical response of stainless steel is fundamentally
different to that of carbon steel, which is an important consideration for assessing the behaviour of
SSRC under ambient and fire conditions. Whilst carbon steel reinforcement typically has an elastic-
plastic or elastic-linear hardening response, the stress-strain behaviour of stainless steel is
characterised by a nonlinear response with no clear yield point and very high levels of strain hardening
and ductility, as shown in Figure 1. In the absence of a clear yield point, the 0.2% proof strength (f0.2)
is typically used to define when the material behaviour becomes inelastic.

f0.2

f0.2

Figure 1 Typical stress-strain curves for carbon steel and stainless steel (adapted from [1])

Table 1 presents the mechanical properties of some of the most common grades of stainless steel
reinforcement, including the 0.2% proof strength f0.2, ultimate strength fu, Young’s modulus E, and
ultimate strain εu. For extreme and unexpected scenarios such as fire, the ductility offered by stainless
steel reinforcement can be very attractive, as it enables load re-distribution throughout the structure
even at severe levels of deformation [6-9].
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

Table 1. Mechanical properties of stainless steel and carbon steel reinforcement

Bar
f0.2 fu E εu
Reference Grade diameter
(N/mm2) (N/mm2) (kN/mm2) (%)
(mm)

[1] 1.4301 8 720 888 156000 44.6

1.4301 10 668 799 148600 38.3

1.4301 12 670 795 186800 26.7

1.4436 8 614 823 178500 36.5

1.4436 10 661 793 179300 25.6

1.4436 12 645 803 198600 25.3

Carbon
10 525 627 196000 20.1
steel

[2] 1.4307 12 562 796 210200 39.9

1.4307 16 537 751 211100 42.4

1.4311 12 480 764 202600 48.3

1.4162 12 682 874 199100 32.4

1.4162 16 646 844 195200 32.9

1.4311 16 528 717 199900 47.9

1.4362 16 608 834 171400 35.1

[3] 1.4301 10 515 790 200100 32.4

1.4301
12 715 868 184000 21.1
“grip-rib”
Carbon
10 589 661 201400 12.49
steel
Carbon
12 554 635 211000 9.21
steel
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

[4] 1.4462 6.5 595 800 141000 32.5

1.4462 12 660 830 141000 37.8

1.4462 16 640 795 151000 33.9

Carbon
12 380 530 230000 31
steel

[6] 1.4362 12 637 872 156 33

1.4362 16 532 768 156 36.4

1.4362 25 543 761 202 31.1

1.4362 28 514 743 138 39.5

1.4362 32 527 748 139 36.9

Carbon
16 477 654 202 26.8
steel
In design and analysis, stainless steel is typically represented using a modified Ramberg-Osgood
material model [10-12], which is a continuous function that captures key material characteristics,
including nonlinearity and strain hardening. It is applied using Eqs. 1 and 2, which capture the elastic
and inelastic stages of the response, respectively:
n
σ σ
ε = E + 0.002 (f ) for f ≤ f0.2 (1)
0.2

σ−f0.2 fu −f0.2 σ−f0.2 m (2)

ε = ε0.2 + E2
+ (εu − ε0.2 − E2
) (fu −f0.2
) for f0.2 < σ ≤ fu

where ε is the engineering strain and σ is the stress, E2 is the tangent modulus at the 0.2% proof
strength point, ε0.2 is the strain that corresponds to f0.2, and n and m are model constants that describe
how the grade under consideration hardens under strain.

The material behaviour including phase changes which is observed in stainless steel is quite different
to that of carbon steel due to its alloying elements and microstructure. In carbon steel, the
microstructure can change between different phases depending on the temperature and carbon
content. Ferrite is a soft, magnetic phase with a body-centred cubic (BCC) structure and low carbon
content. Pearlite, formed during slow cooling, is a mixture of ferrite and cementite, offering increased
hardness and strength. Austenite, with a face-centred cubic (FCC) structure, is stable at higher
temperatures and is more ductile and non-magnetic. In contrast, stainless steel contains chromium
and other elements that stabilize the austenitic phase, giving it enhanced corrosion resistance and
stability across a wider temperature range. [13].
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

For structural engineering applications, the key material characteristics at elevated temperatures are
thermal conductivity, thermal expansion, and the mechanical properties such as strength, stiffness,
and ductility. Stainless steel has a thermal conductivity value of around 15 W/m°C at 20°C, compared
with a value of approximately 51 W/m°C for carbon steel [14]. This means that heat is transmitted
through stainless steel at a slower rate compared with carbon steel. On the other hand, the coefficient
of thermal expansion is generally higher for stainless steel: approximately 14–17×10-6 /°C compared
with 12×10-6 /°C for carbon steel. This can be significant for RC structures which rely on a good bond
developing between the reinforcement and surrounding concrete. In addition, the variation in thermal
expansion behaviour could increase the risk of spalling in the concrete during a fire. Figure 2 presents
the variation in thermal elongation with temperature for stainless steel, carbon steel, and also
concrete with different types of aggregates [15]. This image shows that carbon steel has a plateau at
around 723°C due to a phase change, which does not occur in stainless steel.

Figure 2. Variation in thermal elongation for different materials in RC as a function of temperature

[15]
There has been considerable research into the elevated temperature mechanical properties of
structural stainless steel, but much less research on stainless steel reinforcement. Gardner et al. [2]
carried out one of the few experimental studies into stainless steel reinforcement, conducting 164
isothermal and anisothermal tensile tests on four different grades of stainless steel rebar, including
austenitic grades 1.4307 (304L) and 1.4311 (304LN), and duplex grades 1.4162 (LDX 2101) and 1.4362
(2304). Both plain and ribbed rebars were studied with a diameter of either 12 or 16 mm. The test
results were presented in terms of proposed reduction factors, kiθ, where i is the mechanical property
under consideration (e.g., 0.2% proof strength, f0.2), and k is the ratio of the value of that property at
temperature θ to the corresponding value for that property at 20°C (ambient temperature). Gardner
et al. [2] demonstrated that the reduction factors for the 0.2% proof strength, the strength at 2%
strain, and the ultimate strength derived for stainless steel plate and strip can also be applied to
stainless steel reinforcing bars [2]. However, new reduction factors were proposed for the ultimate
and fracture strain values at elevated temperatures for rebar. This is likely due to the different
manufacturing processes for reinforcement compared with flat plates. In addition, the proposed
reduction factors for stainless steel rebar were shown to be different for stainless steel reinforcement
compared with those recommended for carbon steel reinforcement.
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

Gardner et al. [2] also examined the applicability of the modified Ramberg-Osgood material model for
stainless steel reinforcement at elevated temperatures and calibrated the model constants for
different grades. For stainless steel reinforcement in the elastic phase, up to stress values equal to the
0.2% proof strength f0.2p,θ, the strain at a temperature θ, θ is related to the stress at θ, σθ, by:
nθ
σθ σθ (3)
εθ = + 0.002 ( ) for σθ ≤ f0.2p,θ
Eθ f0.2p,θ

where nθ is the strain hardening exponent at θ. For the inelastic phase, using the expression given in
Eq. 4, which passes through the strength at 2% strain fy,θ, provides an accurate depiction of the
elevated temperature behaviour of stainless steel reinforcement:
mθ
σ−f0.2p,θ σ−f0.2p,θ (4)
εθ = + εu ( ) + ε0.2,θ for f0.2p,θ < σ ≤ fu,θ
E0.2,θ fu,θ −f0.2p,θ

In this expression, E0.2,θ is the tangent modulus at θ, ε0.2,θ is the total strain at 0.2% proof strength f0.2p,θ,
and nθ and mθ are the elevated temperature strain hardening exponents. Different values for nθ and
mθ are proposed for austenitic and duplex stainless steel rebars.

Almost no performance information is available in the literature on the fire behaviour of SSRC
structural members. Preliminary research at the University of Liège [16] compared the performance
of SSRC and CSRC beams numerically and found that fewer stainless steel reinforcing bars are
needed compared with traditional carbon steel reinforcement to get the same level of fire
resistance. This is due to the combined effects of stainless steel retaining greater strength at
elevated temperatures compared with carbon steel and the excellent ductility of stainless steel,
which enables large deformations in the members before failure occurs. Accordingly, the motivation
for the current research was to study the structural analysis and design implications of replacing the
CS rebars with SS rebars, particularly in the fire load case. The paper proceeds with a description of
the numerical model which was developed to analyse the behaviour, and this is followed by the
parametric study and design guidance for SSRC beams at elevated temperatures.

2. Development of the numerical model

A numerical model was developed using the ABAQUS software [17] to simulate the response of SSRC
beams, at both ambient and elevated temperatures. The model is employed herein to understand the
key behavioural characteristics and the most influential parameters and to compare them with the
response of traditional CSRC elements. In ABAQUS, the simulation is conducted using sequentially
coupled thermal-stress analysis, comprising a heat transfer analysis followed by thermal-stress
analysis, using the data from the heat transfer analysis as the input heating load. The following
subsections describe the development and validation of the model in greater detail.

2.1 General
The concrete is modelled using 3D solid elements with reduced integration (i.e., the C3D8R elements
from the ABAQUS [17] library), and the reinforcement is defined using linear 3D truss elements (T3D2).
The reinforcing bars are embedded in the concrete to simulate a perfect bond relationship between
the reinforcement and concrete. Due to symmetry, only half of the beam is modelled. After studying
the sensitivity of the results to the mesh density, 25 mm elements were employed as these were found
to provide the optimal solution in terms of accuracy and computational speed. Load is applied to the
beam through concentrated point loads operated in displacement control at user-defined locations.
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

In Abaqus, the contact position is where the concrete beam meets the loading mechanism. A "tie
contact" is defined at this boundary, ensuring the surfaces move together as one during the analysis.

Figure 3. FE model for sample beam with boundary conditions in Abaqus [17].

Figure 3 shows the boundary conditions for the sample beam. In ABAQUS, the host region is a region
within a model that an embedded region constraint is applied to. The software offers different
solution strategies for structural problems. The implicit dynamic solution procedure uses an implicit
time integration scheme to calculate the transient dynamic or quasi-static response. This can be used
for quasi-static applications by introducing an inertia effect that regularises unstable behaviour in
analyses where the main focus is a final static response [17]. The implicit dynamic solution method
provides good convergence behaviour for all specimens assessed in the current study and is therefore
adopted.

2.2 Material modelling

2.2.1 Concrete
The concrete is simulated using the concrete damaged plasticity (CDP) model, available in ABAQUS.
The CDP model is selected because it considers the inelastic behaviour of concrete both in tension and
compression at elevated temperatures. This approach assumes that the two main failure mechanisms
are tensile cracking and compressive crushing, and the behaviour is defined in terms of the full
compressive and tensile stress-strain properties. In compression, the stress-strain response at
ambient temperature is represented using the material model provided in Eurocode 2 [14], as shown
in Figure 4. The compressive damage parameter dc, which ranges from a value of zero for undamaged
concrete to unity for a state where the material can sustain no more load, is defined as the ratio
between the inelastic and total strain, and the following expression is used to calculate dc:
fcm − σc (5)
dc = when εc ≥ εc1
fcm
where σc and εc are the compressive stress and strain in the concrete, respectively, and fcm is the
maximum compressive stress corresponding to a strain of εc1, as shown in Figure 4.
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

Figure 4. Stress-strain response for concrete in compression in accordance with EN 1992-1-1 [14]

In tension, the ambient temperature behaviour is simulated as a linear response up to the ultimate
tensile strength σcr, which is followed by a gradually descending branch that implicitly accounts for
tension stiffening and the effect of the bond between the rebar and the concrete. Different tension-
stiffening models are available in the literature, and the model proposed by Kaklauskas et al. [18] is
adopted herein as it applies to both reinforced and fibre-reinforced concrete with only minor
alterations. The initial stress-strain relationship for concrete in tension which was derived from beam
tests conducted by Clark and Speirs [19] and Figarovskij [20] provides the foundation for the tension-
stiffening model. The tensile response is shown in Figure 5, in which σt and εt refer to tensile stress
and strain, σcr and εcr are the stress and strain at which cracking occurs, and β is a parameter that
defines when the concrete no longer carries any tensile strength and thus, in effect, captures the effect
of tension stiffening and bond strength. In addition to the compressive and tensile constitutive
relationships, other parameters required in the CDP model include the dilation angle, eccentricity,
ratio of the strength in the biaxial state to the strength in the uniaxial state (fb0/fc0), parameter K, and
a viscosity parameter. The dilation angle was given a value of 38°, and this value is recommended in
ABAQUS to simulate concrete's lateral expansion under stress, affecting crack patterns and failure
behaviour. The other parameters were set following the ABAQUS manual [17] and therefore given
values of 0.1, 1.16, and 0.667 respectively. The viscosity parameter was selected as 0.0003 . The
viscosity stabilization parameter was introduced to improve numerical stability in the FE model by
controlling convergence during simulations of concrete cracking.

Figure 5. Stress-strain response for concrete in tension following Kaklauskas et al. [18]
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

At elevated temperatures, the CDP model is also employed to model the concrete response, employing
the values given for the reduction in mechanical properties as given in Eurocode 2 Parts 1–2 [14]. Figure
5 illustrates the variation in the stress-strain response of concrete in compression at different levels of
elevated temperature based on the material model from Eurocode 2 Part 1–2 [14]. Achieving numerical
convergence of the solution is challenging as a result of the nonlinear behaviour of the concrete at
elevated temperatures, but this can be resolved using the viscosity stabilisation parameter μ, as
recommended elsewhere [21]. The parameter μ is determined using an iterative calibration process,
and a value of μ = 0.0003 is adopted in this analysis.

Figure 6. Elevated temperature stress-strain response of concrete in compression in accordance

with EN 1992-1-2 [14]

The tensile strength of the concrete also reduces with increasing temperature, and the reduction factors
provided in Eurocode 2 Part 1–2 [14] are employed in the finite element model. The propagation of
cracks in the concrete under tension is modelled using a fracture energy GF approach. In the absence of
experimental data, the value of GF (in N/m) for ordinary normal-weight concrete is estimated using the
expression given in the Fib Model Code 2010 [22], given as:
GF = 73 × fcm 0.18 (10)

In the current work, the value of GF is assumed to be constant at all temperatures. Although recent
research suggests a slight variation in fracture energy at elevated temperatures, it was not shown to
noticeably impact the behaviour of RC beams under fire [23]. The tensile damage parameter dt is
defined as the ratio of the cracking strain to the total strain and is determined using Eq. 11:
ft − σt (11)
dt =
ft
2.2.2 Stainless steel reinforcement
The stainless steel reinforcement is represented using the modified Ramberg-Osgood material model,
which is commonly employed for stainless steel as it captures the nonlinear and ductile constitutive
behaviour across all grades. At ambient temperature, the Ramberg-Osgood expressions, as presented
in Eqs. 1–2, are employed to define the relationship between engineering stress (σ) and engineering
strain (ε). In the ABAQUS model, the engineering stress-strain relationship must be converted into a
true stress σtrue versus true strain εtrue, following Eqs. 12 and 13:

σtrue = σ(1 + ε) (12)

εtrue = ln(1 + ε) (13)
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

At elevated temperatures, the modified Ramberg-Osgood expressions given in Eqs. 3 and 4 are
employed to represent the stress-strain response of stainless steel reinforcement. The reduction
factors proposed by Gardner et al. [2] for austenitic and duplex stainless steel reinforcement
considered herein include reduction factors for the Young’s modulus (kE = Eθ/E), 0.2% proof strength
(k0.2p = f0.2p,θ/f0.2p), ultimate strength (ku = fu,θ/fu), ultimate strain (kεu = εu,θ), and fracture strain (kεf =
εf,θ/εf, where εf and εf,θ are the fracture strains at elevated and ambient temperature, respectively).

Figure 7 [12] presents the variation in kE, k0.2p, and ku with elevated temperature from the isothermal
and anisothermal tests on 16 mm grade 1.4307 and 1.4311 austenitic stainless rebars, together with
grade 1.4362 duplex stainless steel. Two different types of bars were tested, representing D16 as
ribbed bars and R16 as plain bars, each with a diameter equal to 16mm.

Following the tests and analysis, Gardner et al. [24] recommended that the existing reduction factors
for structural stainless steel [24] can be safely employed for the Young’s modulus, 0.2% proof strength,
2% proof strength and strain, and ultimate strength for the analysis of stainless steel reinforcement,
while new reduction factors were proposed for ultimate strain and fracture strain. The reduction
factors recommended in that work, together with the elevated temperature modified Ramberg-
Osgood material model, are employed in the current work to represent the response of stainless steel
reinforcement at elevated temperatures. Two different grades of austenitic stainless steel are studied
(EN 1.4307 and EN 1.4311), and duplex stainless steel grade 1.4362 is also examined. The values for
the nθ and mθ constants at elevated temperature are taken as 7.9 and 7.2, respectively, for austenitic
stainless steel reinforcement and 6.3 and 7.9 for duplex stainless steel rebars at elevated temperature.

1 1
0.9 0.9
Reduction factors, k0.2p

0.8 0.8
Reduction factors, kE

0.7 0.7
0.6 0.6
0.5 0.5
0.4 0.4
Duplex 1.4362D16 Austenitic 1.4307R16
0.3 0.3
0.2 Austenitic 1.4307R16 0.2 Austenitic 1.4311D16
0.1 Austenitic 1.4311D16 0.1 Duplex 1.4362D16
0 0
0 100 200 300 400 500 600 700 800 900 1000 0 100 200 300 400 500 600 700 800 900 1000
Temperature (°C) Temperature (°C)
(a) (b)
1
0.9
0.8
Reduction factors, ku

0.7
0.6
0.5
0.4
Austenitic 1.4307R16
0.3
0.2 Austenitic 1.4311D16
0.1 Duplex 1.4362D16
0
0 100 200 300 400 500 600 700 800 900 1000
Temperature (°C)

(c)

Figure 7. Influence of elevated temperature on the mechanical properties of austenitic and duplex
stainless steel rebars including (a) modulus of elasticity, (b) 0.2% proof strength and (c) ultimate
strength [2]
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

2.2.3 Thermal analysis

A nonlinear transient heat transfer analysis is conducted to simulate the thermal response and
temperature histories throughout the RC beam at each time step. A user-defined fire model is
employed, in accordance with the time-temperature response specified in ASTM E119 [25]. This fire
model was used in experimental tests which are reported in the literature, and employed later in this
paper for validating the numerical model. The time-temperature response is given as follows:

Tf = To + 750(1 − exp(−3.79553√t h ) + 170.41√t h (14)

where T0 and Tf are the initial and final temperatures in °C, respectively, and th is the time in hours.

The ASTM E119 [25] fire curve is presented in Figure 8, together with the ISO 834 standard fire curve
[26], which is employed later in this paper for the parametric study.

Figure 8. Fire curves for ISO 834 and ASTM E119

This fire load is applied in the model in the form of an amplitude at the bottom and side surfaces of
the beam only. The temperature transfer through convection is modelled using a surface film
condition (ensures heat transfer between surface and surrounding fluid), and conduction between the
steel and concrete regions is simulated with the tie constraint as they are physically connected [17].
Following EN 1991-1-2 [14], the convective heat transfer coefficient for concrete is taken as
25 W/m2 °C and 9 W/m2 °C for the fire-exposed and unexposed surfaces, respectively. The concrete
elements are modelled using three-dimensional, eight-node solid linear heat transfer elements known
as DC3D8 in the ABAQUS library, and reinforcing steel bars are modelled using two-node heat transfer
link elements (DC1D2). Both of these element types have a temperature degree of freedom. C3D8R
elements were selected for concrete due to their efficiency and reduced computational cost, while
T3D2 elements were chosen for rebar to effectively model axial behavior in reinforcement [17].

The thermal properties, including thermal conductivity, specific heat capacity, and the coefficient of
thermal expansion, are taken from the guidance given in Eurocode 2 Part 1-2 for concrete and carbon
steel reinforcement [14] and the Design Manual for Stainless Steel [27] for stainless steel rebars. The
thermal properties of concrete depend on several factors, including the aggregate type, moisture
content, and concrete weight. The bond between steel and concrete is assumed to be perfect, and it
is implicitly modelled by incorporating the tension-stiffening effect into the constitutive material
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

model of the concrete. In addition, at elevated temperatures, concrete spalling is not considered in
the model, and high-temperature creep is also not explicitly modelled but is implicitly accounted for
in the concrete constitutive material model. The thermal structural stress analysis is performed after
the heat analysis to predict the behaviour of the RC beam at elevated temperatures.

2.3 Validation of the numerical model

No experimental data are available on the response of SSRC beams to fire, and therefore the numerical
model developed in this paper is validated in two separate steps. First, the available test data from
Mokhtar et al. [28] on SSRC beams are employed to examine the accuracy of the model at ambient
temperatures. Then, the elevated temperature modelling of RC beams is examined by analysing the
tests conducted by Dwaikat and Kodur [29]. The accuracy of the key behavioural aspects of both SSRC
beams, including ductility and the modelling of RC elements in fire conditions, are carefully examined
in the validation.

2.3.1 Ambient temperature validation

The test programme on SSRC beams, as conducted by Mokhtar et al. [28], included four large-scale
specimens denoted as B1-B4. Beams B2, B3, and B4 are simulated here using the FE model to analyse
the load-deflection responses. B1 was omitted from the validation exercise because the test results
showed some significant anomalies. The beams were simply supported and loaded through a single
point load at the top surface of the mid-span in each beam. The members had an overall length of
2440 mm, a clear span between the supports of 2200 mm, a width of 250 mm, and a height of 400
mm. Two different types of stainless steel reinforcement were studied, namely austenitic grade
1.4406 (also known as grade 316LN) and duplex stainless steel grade 1.4462 (also known as grade
2205). All beams also had 10 mm-diameter shear reinforcement bars at 200 mm intervals along the
length. The cover distance between the bottom of the beams and the rebars was 35 mm, and the
concrete had a compressive strength of 37 N/mm2. Table 2 gives the details of the specimens,
including the reinforcement arrangements, which are described as 3M30, for example, which
represents three bars with a diameter of 30 mm.

Figure 9 presents the load versus deflection responses of the three beams; the dashed lines are the
simulated responses from the FE model and the solid lines are the test data. The FE model accurately
captures the overall load-deflection response. In the case of beams B2 and B4, which feature austenitic
stainless steel reinforcement, all of the key performance characteristics, such as stiffness and load
capacity, are well represented. Despite a slight disparity in the initial stiffness for B3 between the FE
model and the test data, the two curves converge effectively in the later stages of the test. In
summary, these comparisons confirm that the FE model accurately predicts the behaviour of SSRC
beams at ambient temperature.

Table 2: Details of the sample beams [28]

Name Type of main steel bars Main steel bars Secondary Shear
steel bars reinforcement

B2 Austenitic EN 1.4406 3M30 2M10 10 mm bars at 200

mm centres

B3 Duplex EN 1.4462 5M20 2M10 10 mm bars at 200

mm centres

B4 Austenitic EN 1.4406 5M30 2M10 10 mm bars at 200

mm centres
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

(a) (b)
Figure 9. Comparison of the numerical results with the test data including mid-span deflection
versus time (a) B3 (duplex stainless steel) (b) B2, & B4 (austenitic stainless steel) [28]

2.3.2 Elevated temperature validation

The elevated temperature response predicted by the FE model is validated through a comparison of
the results with test data from traditional CSRC beams in fire. Several researchers have conducted
these types of tests [e.g. 29–33], and the test specimen B1 examined by Dwaikat and Kodur [29] is
replicated herein as a sample beam. This beam had an overall length of 3952 mm and a clear span of
3660 mm between two simple supports. The cross-section of the beam was 254mm wide and 406mm
deep. Middle two third length of the beam equal to 2440mm was uniformly heated during experiment.
The beam was reinforced with three longitudinal reinforcing bars, each 19 mm in diameter, and shear
reinforcement with a 6 mm diameter, placed at 150 mm centre-to-centre spacing. The reinforcement
ratio was 0.595%. The concrete had a compressive strength of 58 N/mm2, and the main reinforcing
and stirrup steel yield strengths were 420 N/mm2 and 280 N/mm2, respectively.

During testing, firstly, two identical point loads of 50 kN were applied to the beam; this was held
constant for 30 minutes, after which no further deformation occurred. This was considered to be the
starting deflection for the beam when the fire was started using an electric furnace. The middle two-
thirds of the span was directly exposed to a time-temperature response as given in ASTM E119 [25].
The applied mechanical loading remained constant throughout the heating phase. After three hours
of exposure to fire, the beam experienced failure in both bending and shear.

50 kN 50 kN 2𝜙 13mm rebar
250 mm
600 mm 600 mm
50mm cover to
reinforcement at
top side
406 mm

𝜙 6mm stirrups
150mm c/c
2440 mm

3658 mm

3952 mm 38 mm (cover to
reinforcement at 3𝜙 19mm
bottom and sides)

(a) (b)
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

Figure 10. Cross-sectional view of the sample beam of CS subjected to fire loading including (a)
elevation view and (b) cross-sectional details. [29]

Figure 11(a) presents the beam deflection versus time relationship during the fire exposure, from both
the physical test and the numerical simulation. The model can capture the general behaviour and the
time at which failure occurs reasonably well. Moreover, Figure 11(b) presents a comparison of the
temperature development at several key locations of the beam, measured during the tests using
thermocouples and simulated using the experimental model. The model can replicate the
temperature development very well, and the results are very similar. As expected, the locations
closest to the fire exposure (bottom of the beam) reach the greatest temperatures, with the rebar
reaching 600°C by the end of the test.

(a) (b)
Figure 11. Comparison of the numerical results with the test data including (a) mid-span
deflection versus time and (b) temperature development versus time at several key locations

During the experimental testing, the beam failed in flexure, with longitudinal cracks at the tension
bar level and spalling at the bottom. Figure 12 compares the observed failure mode (a) from the
experimental results with the FE model predictions for (b) tensile plastic strain and (c) compressive
plastic strain. Figure 12(a) shows images from the test specimen, with cracks in the tension zone and
significant spalling at the bottom of the beam, indicating failure due to excessive tensile and
compressive stresses. Figure 12(b) displays the FE model results for tensile equivalent plastic strain,
which represents irreversible deformation in regions under tensile stress. As the beam bends, the
tension zone experiences plastic deformation, leading to crack formation. The model accurately
predicts these cracks, correlating with the observed failure in the experimental results. Figure 12(c)
shows the FE model results for compressive equivalent plastic strain, indicating areas of plastic
deformation in the compressive region. These regions correspond to the spalling observed at the
bottom of the beam, where compressive stresses cause material crushing. The comparisons confirm
that the FE model effectively captures the failure behavior, accurately predicting both tensile
cracking and compressive spalling, and replicating the deflected shape and crack patterns observed
in the experimental results. Therefore, as the key behavioural aspects of both SSRC beams at
ambient temperature and traditional carbon steel RC beams at elevated temperatures are well
depicted by the numerical model, it is employed hereafter to further investigate the response of
SSRC beams in fire.
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

(a)

(b)

(c)
Figure 12. Comparisons of the deflected shape and crack patterns including (a) an image of the
test specimen following testing [29], (b) the tensile equivalent plastic strain as predicted by the FE
model and (c) the compressive equivalent plastic strain as predicted by the FE model

2.3.3 Behaviour of SSRC beams at elevated temperature

In this section, the validated FE model under elevated temperature conditions is employed to simulate
the behaviour of SSRC beams under the same conditions. The beam examined in the previous section,
i.e., beam B1 [29], is adopted as the benchmark case for which the carbon steel reinforcement is
replaced with identical arrangements of stainless steel rebars, either in austenitic grades 1.4307 or
1.4311 or duplex stainless steel grade 1.4362. Table 1 presents the ambient temperature mechanical
properties for these grades, and Figure 7 presents the elevated temperature reduction factors.

Figure 13(a) presents the comparison between the numerical predictions for the carbon steel RC
(CSRC) beam and the corresponding simulations with stainless steel reinforcement. The beams have
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

identical cross-sectional geometry, concrete strength, and reinforcement ratio. It is clear that the SSRC
beams last for significantly longer compared with the CSRC beam at all stages of the fire, but they last
for a considerably longer fire duration. The SSRC beams tend to deflect more that the CSRC beams,
although only to a minor extent. In this simulation, the CSRC beam fails after 175 minutes, whereas
the SSRC beams with grade 1.4307, 1.4311, or 1.4362 stainless steel rebars fail after 290, 284 and 290
minutes, respectively.

The same analysis is repeated but with a reduced reinforcement ratio of 0.584% for the SSRC beams,
with the CSRC beam remaining at a reinforcement ratio of 0.824%. Figure 13(b) presents the results,
which show that the beams with stainless steel reinforcement (either austenitic or duplex) behave
similarly to each other, reaching similar levels of deflection and duration throughout the fire exposure.
All of the SSRC beams still last significantly longer compared with the CSRC beam, despite containing
less reinforcement. The CSRC beam fails at around 175 minutes; the beams with grade 1.4307, 1.4311,
or 1.4362 stainless steel rebars fail after 224, 230, and 237 minutes, respectively. This is owing to their
improved ductility and greater retention of mechanical properties at elevated temperatures. In
addition, the SSRC beam with duplex rebars has a stiffer response compared with the members with
austenitic reinforcement, with lower deflection after a given fire duration. Concerning failure mode,
the SSRC beams fail due to the yielding of the steel bars, whereas the CSRC beam fails due to a
combination of concrete crushing and steel rupture. The thermal elongation of austenitic stainless
steel is notably higher compared to both carbon and duplex stainless steel, as shown in Figure 2.
Thermal elongation refers to the extent to which a material expands or contracts in response to
changes in temperature. In this context, consequently, when subjected to elevated temperatures,
austenitic stainless steel exhibits a greater degree of expansion, resulting in higher levels of deflection.
Figure 13 visually depicts this phenomenon.

(a) (b)
Figure 13. Comparison of the numerical results of CSRC and SSRC-EN 1.4307,1.411,1.462 (a)
Similar steel ratio (b) reduced steel ratio for SSRC

The enhanced load-carrying capacity of SSRC under fire conditions primarily depends on the inherent
strength properties of stainless steel rebar. This assertion is based on the observation that the
temperature distribution within the stainless steel rebar is comparable to that of carbon steel rebars.
Figure 14 presents a depiction of the temperature distribution through the cross-section for (a) carbon
steel reinforced beam (b) SSRC beam, respectively. The results show a negligible difference in the heat
transfer performance for identical beams with different rebars.
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

(a) (b)
Figure 14: Temperature distribution after 240 minutes of fire for a (a) CSRC beam (b) SSRC
(austenitic grade 1.4307) beam

Investigating the distribution of stresses and strains in carbon steel (CS) rebar and austenitic
SS rebar shows that the strain hardening of an SS rebar enables SSRC to withstand fire for a
longer time compared to a CSRC beam and therefore develop significantly higher stresses and
strains. Figure 15 illustrates the stress and strain profiles of CS and SS rebars, with time, from
the current analysis

The stress development pattern for SS and CS rebars is quite similar. However, CS rebars reach
their maximum stress in about half the time it takes SS rebars to do so. Additionally, SS rebars
generate significantly higher strains—Figure 15(b) shows a strain y-axis an order of magnitude
greater than that in Figure 15(a)—reflecting the greater ductility of SSRC members. The fire
duration at which strains begin to increase substantially is also longer for the SSRC beam than
for the CSRC beam. The key difference lies in how each material sustains maximum stress and
strain: while both CS and SS rebars reach peak stress and strain within a certain time, the
strain in CS rebars escalates sharply afterward, leading to CSRC beam failure. In contrast, SS
rebars sustain their maximum strain much longer before rising toward failure. This strain
durability enables SSRC beams to endure fire exposure significantly longer than CSRC beams.

(a) (b)
Figure 15: Distribution of stress and strain in (a) CS rebar (b) SS rebar
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

Overall, this study shows that although the SSRC beams exhibit higher levels of deflection during a
fire, they also survive much longer with a very ductile response. Even with a significant reduction in
the reinforcement ratio from 0.84% to 0.59%, the SSRC beams still offer almost 60 minutes more of
fire resistance compared with the CSRC beam. With appropriate and reliable research, this finding
may lead to SSRC beams being employed to achieve greater fire resistance with less material usage,
making them a promising alternative to CSRC beams in areas where fire protection is critical.

Although the austenitic SSRC beam demonstrates a significantly higher deflection than the CSRC
beam, the failure mode of the SSRC beam remains highly ductile. This ductility is of paramount
importance in practical scenarios, ensuring a more gradual and predictable failure mode that
enhances safety, particularly in structures subjected to varying loads and conditions. Figure 16 (a &
b), (c & d), and (e & f) show differences in the damage profiles of similar SSRC beams reinforced with
austenitic (1.4307 & 1.4311) and duplex stainless steel at varying reinforcement ratios, as derived
from the FE model. Examination of these damage profiles indicates an important correlation:
increasing the area of stainless steel reinforcement results in higher stress within the concrete. This
suggests that simply adding more stainless steel does not always enhance structural performance;
instead, it can increase the level of stress in the concrete, potentially affecting durability and
performance under load. Therefore, careful consideration is required when increasing stainless steel
reinforcement, balancing both the material properties of concrete and reinforcement area to optimize
the beam's overall strength and fire resistance while avoiding excessive concrete stress and brittle
failure.

(a) (b)

(c) (d)
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

(e) (f)
Figure 16. Damage profile of SSRC beams with (a) austenitic grade 1.4307 reinforcement with a
diameter of 16 mm, (b) austenitic grade 1.4307 reinforcement with a diameter of 19 mm, (c)
austenitic grade 1.4311 reinforcement with a diameter of 16 mm, (d) austenitic grade 1.4311
reinforcement with a diameter of 19 mm, (e) duplex grade 1.4362 reinforcement with a diameter
of 16 mm, and (f) duplex grade 1.4362 reinforcement with a diameter of 19 mm.

3. Parametric study
The previous section highlighted the potential for SSRC beams reinforced with austenitic and duplex
stainless steel to provide better fire resistance than equivalent members made from CS
reinforcement, and even better resistance for fewer material requirements. As a result, the validated
FE model is used in this section for a detailed parametric analysis to investigate how the most
important parameters affect fire performance. A number of salient variables are examined including
the strength of the concrete (fc = 40 MPa or 50 MPa), cover distance (c = 30 mm or 40 mm), type of
reinforcement (austenitic grade 1.4311 or duplex grade 1.4362), diameter of the reinforcement
(diameter = 16 mm or 19 mm, corresponding to reinforcement ratios of 0.584% or 0.824%,
respectively), and load level (P = 0.3Pu or 0.67Pu, where Pu is the ultimate load of the section in normal
conditions). These load levels relate to the service loading applied on the beam. These load levels were
selected based on previous studies by other researchers who utilized the same load parameters [34-
37]. All of the beams studied in this section are simply supported and are 3954 mm in length, 254 mm
in width and 406 mm in depth.

The inclusion of shear reinforcement on the behaviour of RC beams under fire is first assessed in
order to determine the importance of this variable for inclusion in the beams to be examined. Figure
17 presents the deflection versus time results for a series of identical RC beams with either CS or
duplex SS longitudinal rebars. In all cases, the beam is reinforced with three 16 mm longitudinal
reinforcement bars at a cover distance of 40 mm, resulting in a reinforcement ratio of 0.584%. The
concrete has a compressive strength of 30 MPa. In the cases where stirrups are included, these
comprise 6 mm bars at 150 mm centres made from CS reinforcement of yield strength 280 N/mm2. It
is evident from the results that the influence of stirrups is rather minor, which is expected as the
failure mode in all cases is dominated by flexural failure rather than shear failure, as discussed
earlier in this paper. There are slightly greater deflections for the members without stirrups, but the
failure time is almost identical. As a result, the following parametric study focuses on the primary
bottom bars within the beam configurations under investigation without stirrups.
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

Figure 17: Comparison of deflection failure curves for CS and DS RC beams with and without
stirrups

The initial mechanical load applied to an RC beam plays a pivotal role in determining its performance
under fire conditions. In the parametric analysis, the analytical model proposed and validated
elsewhere [37] is used to calculate the load-carrying capacity of the beams at ambient temperature.
Then, a proportion (either 30% or 67%) of the full load capacity is applied in the thermo-mechanical
finite element model to observe the fire behaviour of the specimens. A total of 48 different RC beams
are examined, with parameters varied individually from a control specimen. The control specimen has
a compressive strength of 58.2 MPa at ambient temperature and is reinforced with three 19 mm bars
with a cover distance of 38 mm to the bottom surface of the beam. The stainless steel mechanical
properties are adopted as given in Table 1 at ambient temperature whilst the CS bars have a yield
strength of 420 N/mm2 and a Young’s modulus of 200,000 N/mm2, in accordance with the tests of
Dwaikat and Kodur [29]. All of the beams are subjected to an ISO 834 [26] standard fire. The key factors
used to study the performance of the beams in fire are the time to failure, the maximum temperature
of the reinforcement, the deflection at failure, and the failure mode.
Table 3 presents the results of the analyses, including the theoretical ultimate load section capacity at
ambient temperature Pu, the maximum stresses at failure measured in the reinforcement (sf) and the
concrete (cf), the failure time in minutes and the failure mode. Two different failure modes are
observed (i.e. rebar rupture or concrete failure) and these are assessed based on the maximum
stresses measured in the respective materials at each load step. The failure mode is also discernable
from the shape of the stress-strain curve whereby a ductile failure mode is characterised by a smooth
response whilst a brittle failure is typically accompanied by a sudden reduction in load-carrying
capacity.
Table 3 also includes the maximum deflection at the time of failure (δu) with the maximum
temperature in the rebar (θmax, rebar). The specimens as given in Table 3 are denoted using the following
notation: the initial term represents the reinforcement type (CS, AS and DS for carbon steel, austenitic
stainless steel, or duplex stainless steel, respectively); the second letter indicates the diameter of the
reinforcing bars (D16 or D18, representing bars with a diameter of 16 mm or 19 mm, respectively);
the third term denotes the concrete strength (fc30 or fc40, representing compressive strengths of 30
MPa or 40 MPa); then is the load level (0.3P or 0.67P representing 30% or 67% of Pu); and finally the
last term identifies the concrete cover ( C30 for 30 mm of cover and C40 for 40 mm cover distance).
Table 3. Key data from the parametric study
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

Failure
Time θmax, rebar δu Failure sf cf
Beam Sample Name Pu
(Minute (oC) (mm) mode (MPa) (MPa)
(kN)
s)

CS-D16-fc40-0.3P-C30 125.7 166 505 45 Rebar 215 16.5

CS-D19-fc40-0.3P-C40 122.1 150 418 48 Rebar 314 17.7

CS-D19-fc40-0.67P-C30 125.7 121 395 62 Rebar 375 27.8

CS-D19-fc40-0.67P-C40 122.1 131 345 58 Rebar 371 30.0

CS-D16-fc50-0.3P-C30 126.6 168 508 45.5 Rebar 223 19.2

CS-D16-fc50-0.3P-C40 122.9 150 418 45 Rebar 315 21.3

CS-D16-fc50-0.67P-C30 126.6 121 395 60 Rebar 375 30.7

CS-D16-fc50-0.67P-C40 122.9 134 379 60 Rebar 375 29.1

CS-D19-fc40-0.3P-C30 173.5 174 515 62 Rebar 244 17.0

CS-D19-fc40-0.3P-C40 168.4 171 519 55 Rebar 280 22.4

CS-D19-fc40-0.67P-C30 173.0 123 393 66 Rebar 371 34.1

CS-D19-fc40-0.67P-C40 167.9 150 411 66 Rebar 347 35.8

CS-D19-fc50-0.3P-C30 175.1 172 510 54 Rebar 236 20.6

CS-D19-fc50-0.3P-C40 170.0 172 460 53 Rebar 281 23.2

CS-D19-fc50-0.67P-C30 175.1 127 404 72 Rebar 377 37.0

CS-D19-fc50-0.67P-C40 170.0 156 423 72 Rebar 349 37.6

AS-D16-fc40-0.3P-C30 177.7 318 760 95 Rebar 256 16.0

AS-D16-fc40-0.3P-C40 172.1 350 759 84 N/A 245 15.7

AS-D16-fc40-0.67P-C30 177.7 162 475 86.8 Concrete 434 37.12

AS-D16-fc40-0.67P-C40 172.1 136 388 64 Concrete 450 35.3

AS-D16-fc50-0.3P-C30 181.5 313 760 100 Rebar 260 23.2

AS-D16-fc50-0.3P-C40 175.8 350 759 100 N/A 246 19.5

AS-D16-fc50-0.67P-C30 181.5 142 450 76 Concrete 446 41.4

AS-D16-fc50-0.67P-C40 175.8 222 565 100 Rebar 430 38.7

Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

AS-D19-fc40-0.3P-C30 237.5 325 768 97 Rebar 246 19.4

AS-D19-fc40-0.3P-C40 229.7 350 749.6 87 N/A 252 18.2

AS-D19-fc40-0.67P-C30 237.5 172 510 110 Concrete 438 39.7

AS-D19-fc40-0.67P-C40 229.7 232 576 111.6 Rebar 423 45.6

AS-D19-fc50-0.3P-C30 243.9 324 765 98 Rebar 249 23.6

AS-D19-fc50-0.3P-C40 236.0 350 749.6 85.3 N/A 252 21.3

AS-D19-fc50-0.67P-C30 243.9 76 256 48 Concrete 476 18.8

AS-D19-fc50-0.67P-C40 236.0 108 304 54.5 Concrete 453 47.3

DS-D16-fc40-0.3P-C30 206.7 258 683 83.6 Rebar 282 18.5

DS-D16-fc40-0.3P-C40 199.9 308 709 86 Rebar 253 18.2

DS-D16-fc40-0.67P-C30 206.7 101 340 58 Concrete 530 38.7

DS-D16-fc40-0.67P-C40 199.9 172 566 81.7 Rebar 499 40.3

DS-D16-fc50-0.3P-C30 212.7 250 672 81.8 Rebar 295 20.2

DS-D16-fc50-0.3P-C40 205.8 300 698 81.4 Rebar 264 19.5

DS-D16-fc50-0.67P-C30 212.7 105 352 58 Concrete 535 39.6

DS-D16-fc50-0.67P-C40 205.8 110 317 52 Concrete 511 38.7

DS-D19-fc40-0.3P-C30 270.6 264 687 86 Rebar 275 22.7

DS-D19-fc40-0.3P-C40 261.2 300 692 78.5 Rebar 263 20.0

DS-D19-fc40-0.67P-C30 270.6 65.5 224 50 Concrete 497 45.1

DS-D19-fc40-0.67P-C40 261.2 145 402 77 Concrete 514 39.9

DS-D19-fc50-0.3P-C30 281.1 260 680 82 Rebar 279 23.8

DS-D19-fc50-0.3P-C40 271.6 300 692 80 Rebar 265 22.1

DS-D19-fc50-0.67P-C30 281.1 143 448 85.5 Rebar 515 40.2

DS-D19-fc50-0.67P-C40 271.6 155 425 75.8 Concrete 509 47.1

The results shown in Table 3, which focus on rebar stresses, temperature, and beam deflections,
provide a better understanding of the behaviour of SSRC beams under elevated temperatures. The
findings indicate that SSRC beams maintain their load-carrying capacity even when the temperature
in the stainless steel rebar reaches as high as 765°C. In addition, Figure 18(a) illustrates the beam mid-
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

span deflections versus temperature for all of the studied beams. It is evident firstly that the beams
with SS rebars achieve much greater rebar temperatures and therefore survival durations compared
with the CSRC beams. Moreover, that at very high temperatures, the beam deflections remain below
the failure limit of L/20=200 mm for all SSRC beams. This observation emphasises the structural
robustness and resilience of these members, highlighting their ability to withstand elevated
temperatures while still adhering to stringent failure criteria for deflection limits. Figure 18(b)
illustrates the stresses in the rebar versus the rebar temperature for all members and shows that the
CSRC beams experience a loss of strength within the temperature range of 300–500 °C. In contrast,
the beams with austenitic and duplex stainless steel maintain approximately 80% of their strength
even up to 600°C. The SSRC beams exhibit resilience, capable of retaining 30% of their strength even
at temperatures exceeding 650°C. This highlights the superior fire performance of austenitic and
duplex stainless steel reinforced concrete compared to traditional CSRC.

Figure 18. Comparison of (a) deflection and (b) rebar stresses versus temperature for all examined
beams

3.1 Effect of load ratio

The load ratio is an important parameter in the assessment of structural performance in fire. When
this value remains constant, the fire resistance of the RC beam is enhanced with higher concrete
strength and/or greater reinforcement ratio. Conversely, the load ratio has a detrimental effect on
the fire resistance of the RC beam. Two load levels are examined corresponding to 30% and 67% of
the ultimate load capacity at ambient temperature (Pu) and the results are presented in Figure 19. It
is shown that, as expected, beams subjected to lower load ratios exhibit significantly prolonged fire
resistance compared with more heavily loaded specimens. For comparison, beams with varying
concrete strengths and reinforcement ratios, all with a concrete cover of 40 mm, are shown in Figure
19(a). Notably, the deflection curves exhibit similar trends under both load ratios. The majority of the
beams experience ductile failure, which is characterised by failure of the reinforcing steel. However,
the duplex SSRC beams exhibit concrete failure when relatively low-strength concrete is used.
Although employing high-strength steel enhances its fire resistance, the nature of failure remains
consistent. This is owing largely to the excellent ductility of the stainless steel reinforcement even
after exposure to very high temperatures. On the other hand, the SSRC with austenitic rebars
demonstrate rebar rupture as the failure mode when low-strength concrete is employed but shift
towards concrete failure when higher strength concrete is used. As higher-strength concrete increases
the ambient load-carrying capacity of the beam, even a slight increase in the initial load adversely
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

affects the performance of the RC beam under fire. This behaviour underscores the significance of
selecting the appropriate material strength to fully harness the advantages of each structural element
material, ensuring the proper capacity of the RC beam under fire conditions.

(a)

(b)

(c)
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

(d)
Figure 19: Comparison of fire resistance for RC beams with different types of reinforcement
highlighting the influence of (a) load ratio, (b) concrete cover, (c) reinforcement ratio and (d)
concrete strength
3.2 Concrete cover
The level of cover provided by the concrete to the reinforcement plays a crucial role in determining
the load-carrying capacity of RC beams, both under normal conditions and during exposure to
elevated temperatures, such as in a fire. A smaller concrete cover results in a greater moment capacity
due to an increased effective depth, whereas a larger concrete cover helps mitigate the impact of high
temperatures on the reinforcing steel, allowing the RC beam to withstand fire for a longer period.
However, a higher concrete cover reduces the effective depth, leading to a decrease in the sectional
capacity and thus a lower load ratio compared to beams with smaller covers. The impact of concrete
cover is illustrated in Figure 19(b). It is evident that RC beams with higher concrete covers demonstrate
extended fire resistance compared to those with smaller covers. This influence becomes more
pronounced at higher load ratios, as the reinforcement can sustain greater stress levels. This effect is
particularly significant for SSRC beams, where the enhanced ductility and thermal resistance of
stainless steel rebars contribute to improved fire performance.

3.3 Reinforcement ratio

Figure 19(c) illustrates the influence of changes in reinforcement ratio on the analysed RC beams. It is
obvious that under ambient temperature conditions, increasing the size of the reinforcement leads to
an improved load-carrying capacity for the RC beam. However, when the beam is exposed to elevated
temperatures, a change in reinforcement ratio has a much lesser impact, particularly for members
with a high load ratio. At higher load ratios, increasing the area of reinforcement tends to reduce the
endurance of the beam under fire due to the heightened initial load effect. Furthermore, Figure 19(c)
illustrates that the influence reinforcement ratio is often negligible in the current study as many of the
SSRC beams fail by concrete crushing, and therefore the full capacity of the rebars is not necessarily
employed. Overall, SSRC beams with a higher reinforcement ratio tend to exhibit greater deflections
and duration of fire survival.

3.4 Concrete strength

Figure 19(d) shows the effect of concrete strength on the fire resistance of SSRC beams at elevated
temperatures. The results show that an increase in concrete strength does not offer any difference
for the CS and stainless steel (SS) beams at either a 30% or 67% load ratio. The reason is that an
increase in concrete strength also increases the sectional capacity of the RC beam, which adversely
affects the fire resistance of the RC beam. So, the contribution of a small increment of concrete
 Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

strength for the RC beam in fire resistance is only to avoid its brittle failure rather than increase its
load-bearing time under fire.

4. Design commentary
The performance of RC beams under fire is assessed based on three criteria defined in EN 1992-1-2
[14]:

• R (load-bearing resistance)
• E (structural integrity)
• I (insulation, if required)

The primary failure criterion according to EN 1992-1-2 [14] is based on the strength limit, which is
reached when the moment capacity of the beam falls below the applied moment. BS 476 [38] also
sets deflection-based failure criteria, specifying that failure occurs if the deflection exceeds L/20 or if
the deflection rate surpasses L2/9000d (mm/min), where L is the beam length and d its effective depth.

Additionally, the critical temperature of the reinforcing steel is crucial in determining the fire
resistance of RC beams. It represents the maximum temperature at which a beam can sustain its load
without failure during a fire. While EN 1992-1-2 [14] provides guidelines for determining this
temperature based on the fire resistance rating, it does not specify a single universal threshold. In
contrast, ASTM E119 [25] sets a critical temperature limit of 593°C for reinforcing steel.

The FE analysis conducted in this study generates key outputs like temperature, stress, strain, and
deflection at specific fire exposure times for SSRC beams. The results indicate that SS rebars can reach
temperatures as high as 765°C without causing catastrophic failure. This suggests that SSRC beams
have a higher critical temperature limit compared to CSRC beams.

From the comprehensive analysis in Table 3, it is recommended that the critical temperature limit for
SS rebars be set above 765°C. The sample beam (AS-D16-fc50-C40) demonstrates failure at maximum
stress levels in both concrete and reinforcing steel. Consequently, design practices should prioritize
selecting appropriate grades of stainless steel and concrete to fully leverage the strengths of both
materials. This approach ensures an optimal balance of mechanical properties, enhancing the
structural performance and fire resistance of RC beams.

5. Conclusions
A numerical simulation study was conducted to investigate the fire resistance of SSRC beams using
sequentially coupled thermomechanical stress analysis in ABAQUS software [14]. The accuracy of the
simulation method was validated by comparing the results with existing fire test data for CSRC beams.
A material model was developed for stainless steel rebars at elevated temperatures, which accounts
for different grades. The model uses the Ramberg-Osgood expression and incorporates reduction
factors obtained from Gardner's experimental work. The material model was successfully
implemented in the finite element analysis of RC beams under fire. Based on this, a parametric analysis
was conducted to identify the factors that affect the fire resistance of RC beams and determine their
respective influences on the fire resistance and behaviour response of the CS and SSRC beams.

In conclusion, the findings of parametric study specifically pertain to fire scenarios where flexural
failure was observed as the dominant mode of failure in our FE models. While the inclusion of stirrups
was found to have a minimal impact on deflection and failure time under the simulated conditions,
we acknowledge that this may not fully reflect the behavior of RC beams under more complex loading
scenarios. The decision to exclude stirrups in certain parametric studies was made to streamline the
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

analysis; however, this is recognized as a simplification. It is recommended that future research should
investigate cases where shear interactions and the presence of stirrups may become more critical to
ensure a comprehensive understanding of the fire performance of RC beams.

Indeed, this study offers valuable insights into the behaviour of RC beams under fire through the use
of informative graphs. These graphs enable a clear understanding of how different factors, such as
concrete cover, load ratio, and steel reinforcement, affect the fire resistance of RC beams. The findings
of this study can be summarised as follows:

1. The Ramberg-Osgood equation, combined with Gardner's reduction factors, offers a viable
material model applicable for finite element analysis of SSRC beams when subjected to fire.
2. The ductile failure of SSRC beams under fire makes them a suitable replacement for CSRC
beams in fire-prone regions where structural integrity is the primary concern.
3. SSRC beams can achieve similar fire resistance levels to CSRC beams with a lower
reinforcement ratio, making it simpler to reduce the reinforcement ratio in the former.
4. The SSRC beam with a greater concrete cover provides better fire resistance than a beam
with a smaller cover or a CSRC beam. This impact is more pronounced at smaller load ratios.
5. Factors that impact the fire resistance of SSRC include the load ratio, material strength, and
concrete cover. The load ratio is a critical factor that greatly impacts the fire resistance
performance of SSRC beams. Specifically, for identical specimens, as the load ratio increases
from the lowest serviceable applied load (30%) to the maximum serviceable applied load
(67%), the decrease in the use of the beam under fire is substantial.
6. The considerably high coefficient of thermal expansion present in stainless steel material
results in increased expansion of the SSRC beams, leading to additional bending moments in
SSRC beams when exposed to temperature changes. This effect significantly impacts the fire
resistance performance of SSRC beams under serviceability criteria, particularly due to the
large deflections caused by the additional bending moments.
7. This study proposes a new critical temperature limit of 765 oC for SS reinforcing rebar under
fire exposure.

Future research should prioritize the practical fire testing of SSRC beams to validate the numerical
findings and assess their real-world performance under varying fire scenarios. This will help confirm
the observed behaviours, such as deflections and failure modes, and ensure the reliability and
applicability of SSRC beams in fire-prone environments.

Acknowledgments

I would like to express my sincere gratitude to acknowledge the financial support from the Engineering
and Physical Sciences Research Council (EPSRC), which made this research possible.

References

[1] M. Rabi, K.A. Cashell, R. Shamass, P. Desnerck, Bond behaviour of austenitic stainless steel
reinforced concrete, Eng. Struct., 221 (2020), Article 111027.
[Link]

[2] L. Gardner, Y. Bu, P. Francis, N.R. Baddoo, K.A. Cashell, F. McCann, Elevated temperature
material properties of stainless steel reinforcing bar, Constr. Build. Mater., 114 (2016), pp.
977-997. [Link]
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

[3] Rabi, M., Shamass, R. and Cashell, K.A., Experimental investigation on the flexural behaviour
of stainless steel reinforced concrete beams, Structure and Infrastructure Engineering, 0 (in
press), pp. 1-13. doi: 10.1080/15732479.2022.2065687.

[4] Qingfu Li, Wei Guo, Chenhui Liu, Yihang Kuang, Huitao Geng, Experimental and theoretical
studies on flexural performance of stainless steel reinforced concrete beams, Adv. Civ. Eng.,
2020 (2020), pp. 1-13. [Link]

[5] Q. Li, Y. Cui, J. Wang, Basic mechanical properties of duplex stainless steel bars and
experimental study of bonding between duplex stainless steel bars and concrete, Materials,
14 (11) (2021), p. 2995. [Link]

[6] C. Xu, M.L. Nehdi, M.A. Youssef, T. Wang, L.V. Zhang, Seismic performance of RC beam-column
edge joints reinforced with austenite stainless steel, Eng. Struct., 232 (2021), Article 111824.
[Link]

[7] Pardeshi, R., Thakur, B. and Parghi, A., March, Seismic investigation of longitudinally aligned
shape memory alloy-stainless steel reinforced concrete column, In IOP Conference Series:
Materials Science and Engineering (Vol. 814, No. 1, 2020), p. 012012. DOI 10.1088/1757-
899X/814/1/012012.

[8] M. Panagiotou, W. Trono, G. Jen, P. Kumar, C.P. Ostertag, Experimental seismic response of
hybrid fiber-reinforced concrete bridge columns with novel longitudinal reinforcement
detailing, Journal of Bridge Engineering, Volume 20, Issue 7, July 2015.
[Link]

[9] G. Zhang, Z. Zhang, C. Wang, Seismic performance of stainless steel reinforced concrete
columns, Adv. Sci. Lett., 4 (8–9) (2011), pp. 3119-3123.
[Link]

[10] Ramberg, W. & Osgood, W.R., Description of stress-strain curves by three parameters, (1943).

[11] K.J.R. Rasmussen, Full-range stress–strain curves for stainless steel alloys, J. Constr. Steel Res.,
59 (1) (2003), pp. 47-61. [Link]

[12] E. Mirambell, E. Real, On the calculation of deflections in structural stainless steel beams: an
experimental and numerical investigation, J. Constr. Steel Res., 54 (1) (2000), pp. 109-133.
[Link]

[13] Zaki Ahmad, Chapter 4 - Types of Corrosion: Materials and Environments, Editor(s): Zaki
Ahmad, Principles of corrosion ngineering and corrosion control, Butterworth-Heinemann,
2006, pp. 120-270, ISBN 9780750659246. [Link]
6/50005-2.

[14] EN 1992-1-2, Eurocode 2: Design of concrete structures-Part 1-2: General rules-Structural fire
design, European Committee for Standardization (CEN), (2004).

[15] Rabi, M., Shamass, R., Cashell, K.A., Structural performance of stainless steel reinforced
concrete members: A review, Construction and Building Materials, Volume 325, 28 March
2022, 126673. [Link]

[16] Gernay T., Franssen J.M., Economic analysis of stainless steel reinforced concrete members
designed for a fire resistance of 2 hours, Study performed for ISSF International Stainless Steel
Forum, University of Liège. [Link]
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

[17] ABAQUS, User Assistance Manual, Simulia, Dassault Systèmes, (2020).

[18] Kaklauskas, G.A., New stress-strain relationship for cracked tensile concrete in flexure, ISSN
1392-1525, Statyba - Civil Engineering, 1999, V Tomas, Nr. 6.
[Link]

[19] Clarks A., Speirs D.M., Tension stiffening in reinforced concrete beams and slabs under short-
term load, Technical Report No 42, Cement and Concrete Association, London, 1978.

[20] V.V. Figarovskij, Experimental investigation of stiffness and cracking of reinforced concrete
flexural members subjected to short-term and long-term loading (in Russian), Ph.D. Thesis,
Moscow, 1962.

[21] P. Kmiecik, M. Kamiński, Modelling of reinforced concrete structures and composite

structures with concrete strength degradation considered, Archives of Civil and Mechanical
Engineering, Volume 11, Issue 3, 2011, pp. 623-636, ISSN 1644-9665.
[Link]

[22] fib, fib model code for concrete structures 2010, Berlin: Ernst & Sohn, 2013.

[23] Farshad Dabbaghi, Hossein Fallahnejad, Sepideh Nasrollahpour, Mehdi Dehestani, Hossein
Yousefpour, Evaluation of fracture energy, toughness, brittleness, and fracture process zone
properties for lightweight concrete exposed to high temperatures, Theoretical and Applied
Fracture Mechanics, Volume 116, 2021. [Link]

[24] Gardner L., Insausti A., Ng K.T., Ashraf M., Elevated temperature material properties of
stainless steel alloys, Journal of Constructional Steel Research, 66, pp. 634-647, 2010. DOI:
10.1016/[Link].2009.12.016

[25] ASTM E119, Standard methods of fire test of building construction and materials, American
Society for Testing and Materials, West Conshohocken, PA, 2008.

[26] ISO 834, Fire resistance test element – Elements of building construction – Part 1: General
requirements, International Standard (1999).

[27] Design manual for structural stainless steel, 4th edition, SCI publication P413.

[28] Mokhtar K., Alhadid M.M., Youssef M.A., Early age thermal expansion of stainless-steel
reinforced concrete sections, International Conference on Short and Medium Span Bridges,
Quebec City, Quebec, Canada, 2018.

[29] M.B. Dwaikat, V.K.R. Kodur, Response of restrained concrete beams under design fire
exposure, Journal of Structural Engineering, vol. 135, no. 11, pp. 1408–1417, 2009.
[Link]

[30] Ellingwood B., Lin T.D., Flexure and shear behavior of concrete beams during fire, J. Struct.
Eng. ASCE, 117(2), pp. 440-458, 1991. [Link]
9445(1991)117:2(440)

[31] Wu H.J., Lie T.T., Hu J.Y., Fire resistance of beam-slab specimens – experimental studies,
Internal Report No. 641, Institute for Research in Construction, National Research Council
Canada, Canada, 1993. [Link]
Behaviour of stainless steel reinforced concrete beams in fire - ScienceDirect

[32] Kumar, A., Kumar, V., Behaviour of RCC beams after exposure to elevated temperature,
Journal of the Institution of Engineers. India, 84(3), pp. 165-170, 2003.

[33] Rafi, M.M., Nadjai, A., Ali, F., Analytical modeling of concrete beams reinforced with carbon
FRP bars, Journal of Composite Materials, 41(22), pp. 2675-2690. doi:
10.1177/0021998307078728.

[34] F. Abed, A.R. Alhafiz, Effect of basalt fibers on the flexural behavior of concrete beams
reinforced with BFRP bars, Compos Struct, 215 (2019), p. pp
[Link]

[35] A. El-Nemr, E.A. Ahmed, B. Benmokrane, Flexural behavior and serviceability of normal-and
high-strength concrete beams reinforced with glass fiber-reinforced polymer bars, ACI Struct
J, 110 (6) (2013) [Link]

[36] Rabee Shamass, K.A. Cashell, Experimental investigation into the flexural behaviour of basalt
FRP reinforced concrete members, Engineering Structures, Volume 220,2020,110950,ISSN
0141-0296,[Link]

[37] M. Rabi, K. Cashell, R. Shamass, Flexural analysis and design of stainless steel reinforced
concrete beams, Eng. Struct., 198 (2019), Article 109.

[38] BS 476, Fire tests on building materials and structures – Part 20: Method for determination of
the fire resistance of elements of construction (general principles), BS 476-3:1987, B