Proceedings of the 1st Iberic Conference on Theoretical and Experimental Mechanics and Materials /

11th National Congress on Experimental Mechanics. Porto/Portugal 4-7 November 2018.

Ed. J.F. Silva Gomes. INEGI/FEUP (2018); ISBN: 978-989-20-8771-9; pp. 309-320.

PAPER REF: 7385

FIRE RESISTANCE OF COMPOSITE SLABS WITH STEEL DECK:

EXPERIMENTAL ANALYSIS AND NUMERICAL SIMULATION
Paulo A. G. Piloto1(*), Lucas M.S. Prates2, Carlos Balsa3, Ronaldo Rigobello2
1
LAETA-INEGI, Department of Applied Mechanics, Polytechnic Institute of Bragança (IPB), Portugal
2
Federal Technological University of Paraná (UTFPR), Brazil
3
Department of Mathematics, Polytechnic Institute of Bragança (IPB), Portugal
(*)
E-mail: [email protected]

ABSTRACT
This work investigates the thermal behaviour of composite slabs with steel deck under
controlled test conditions corresponding to a fire from the bottom. This composite solution
consists of a concrete topping cast on the top of a steel deck. The concrete is typically
reinforced with a steel mesh and may also contain individual rebars. The deck also acts as
reinforcement and may be exposed to accidental fire conditions from the bottom. This
composite solution is widely used in every type of buildings and requires fire resistance, in
accordance to regulations and standards. Composite slabs need to meet fire-safety
requirements according to building codes. The fire assessment of this type of elements is
normally made using standard fire tests. Two samples are being prepared to be tested and
should take into account the criterion for stability (R), Integrity (E) and insulation (I). The
scope of this investigation concerns the fire rating for insulation (I). Numerical simulation
was performed through Matlab PDE toolbox for the thermal effects of standard fire exposure.
The results are also compared with the simplified method proposed by Eurocode, which
seems to be unsafe.

Keywords: Composite slabs, fire resistance, thermal performance, numerical simulation.

1 INTRODUCTION
Concrete slabs with steel decks are slabs that use steel deck as a permanent formwork and as
reinforcement to the concrete placed on top, see Figura 1. This represents one of the
advantages of this solution, because reduces the construction time, requires less concrete,
providing slender slabs.

Fig. 1 - Definition of the rib geometry for the cross section of part of the composite slab.

-309-
 Track-B: Computational Mechanics

The use of these composite slabs in buildings has become very popular, since 1980. The
overall depth (h1+h2) can vary between 100 to 170 mm. The thickness of the deck can vary
from 0.7 to 1.2 or more and this part of the structure is normally galvanized to increase
durability [1]. The composite floor is usually made with these plate elements supported by
secondary beams (linear elements) and shear studs that are responsible for the composite
action between both elements. The fire resistance of both elements is prescribed by the
building codes, but this investigation only considers the fire behaviour of the plate element.
Several studies have been conducted to evaluate the fire resistance of concrete slabs with steel
deck. In 1990 Hamerlinck et al [2] developed a numerical model that satisfactorily predicted
the fire behaviour of different slab geometries. In 1999 Bailey et al. [3] presented the results
of 2 experimental full-scale tests (complete building), demonstrating that the performance of
the structure under fire differed from that was expected from fire codes and demonstrated that
they were also conservative. Both tests also demonstrated that the element behaviour is
different from what is normally obtained from standard small-scale fire tests. In 2001 Lamont
et al [4], performed an analysis of the heat transfer in composite slabs of the Cardington
building. Four tests were performed in different floors of the building. An adaptive heat
transfer model was used to estimate the temperatures through the slab. The developed model
presented satisfactory results for most of the tests. In 2002 Lim et al [5] developed fire tests of
two-way concrete slabs at the BRANZ fire resistance furnace, six slabs were tested,
comprising three reinforced concrete flat slabs and three composite steel-concrete slabs. The
three flat slabs had different amount of reinforcing steel to investigate their effect on
controlling crack widths to insure integrity. The slabs were submitted to a live load of 3.0 kPa
and were heated from the bottom with standard fire ISO 834 during three hours. The slabs
supported the full duration of the tests without collapse. The structural fire resistance of the
slabs in the tests exceed the predictions of code recommendations.
More recently in 2017, Guo-Qiang Li et al [6], performed 4 tests in composite slabs with steel
decking, which were fire rated with 90 minutes and concluded that Eurocode 4 design
calculations are conservative and that could be used for the other geometries, beyond the
specified limit. The experiments were developed at Tongji University and the average
temperatures of the furnace were below the standard ISO 834 [7]. The temperature at the
bottom of the slabs (above the steel deck) were 100 °C on average below furnace temperature.
The temperature on the unexposed surface was less than 100 °C during the tests, being the fire
rating determined by stability. This research also presents a summary of previous experiments
developed on composite floor systems.
Composite slabs need to meet fire-safety requirements according to building codes. The fire
requirements are normally specified by fire rating periods of 30, 60, 90 min or more. The fire
assessment of this type of elements is normally made using standard fire tests [7]-[9] and
should take into account criterion for stability (R), Integrity (E) and insulation (I). These tests
are expensive and time-consuming, reason why the fire resistance can be evaluated by means
of numerical simulation or by the use of simple calculation methods. The fire behaviour of
composite slabs is generally defined with respect to standard fire exposure from below. Fire
exposure at the other side of the slab is less critical [1].
The European recommendations for composite steel and concrete slabs were introduced by
the ECCS [10] and a proposal for the assessment of the insulation criterion (I) was made,
based on the calculation of the effective thickness of concrete. At this stage, conservative
assumptions have been used, leading to uneconomic solutions [1].

-310-
 Proceedings TEMM2018 / CNME2018

The current version of Eurocode 4 [11] proposes a simple calculation method, in Annex D, to
define the fire resistance (I), which depends linearly in a set of geometric parameters, but that
seems to be over conservative as well.

2 SIMPLIFIED METHOD
t
According to Annex D of Eurocode 4 [11], the fire resistance i , of both simply supported
and continuous concrete slabs with profiled steel decks, when submitted to standard fire, may
calculated according to equation (1).

t i = a0 + a1 .h1 + a2 .φ + a3 . A Lr + a4 .1 l3 + a5 . A Lr 1 l3 (1)

where

[
A Lr = h2 (l1 + l2 ) / 2 / l2 + 2 h22 + ((l1 − l2 ) / 2)2 ] (2)

The partial factors ai are proposed for normal weight concrete (NC), according to Table 1.

Table 1 - Partial factors used for the calculation of fire resistance (NC).
a0 a1 a2 a3 a4 a5
[min] [min/mm] [min] [min/mm] [min.mm] [min]
-28.8 1.55 -12.6 0.33 -735 48

In a previous work [12], authors concluded that the fire resistance is also independent of the
steel deck thickness and present a quadratic dependence on concrete depth above the deck h1.
These observations are summarised in Table 2.

Table 2 - Fire resistance in completed minutes (insulation criterion).

Geometry h1 [mm] 40 50 60 70 80 90 100 110
L1/L2=84/40 ti [min] 34 50 65 81 96 112 127 143
L1/L2=105/60 ti [min] 38 53 69 84 100 115 131 146

This study intends to analyse the model with h1=40 mm and L1/L2=105/60, with an expected
fire resistance of 38 min, according to the simplified method.

3 NUMERICAL SIMULATION METHOD

3.1 Heat transfer equation
A two dimensional model was used for the numerical simulations. The cross section of the the
slab is meshed to solve a nonlinear transient thermal analysis. The finite element method
requires the solution of equation (3) in the domain of the cross section (Ω) and equation (4)
for the boundary conditions exposed to fire (∂Ω).

∇(λ(T ) .∇T ) = ρ(T ) .Cp(T ) . ∂T ∂t (Ω) (3)

-311-
 Track-B: Computational Mechanics

ρ
λ(T ) .∇T.n = αc (Tg − T ) + φ.ε m.ε f .σ .(Tg4 − T 4 ) (4)

In these equations: T represents the temperature of each material; ρ (T ) defines the specific
mass; Cp(T ) defines the specific heat; λ(T ) defines the thermal conductivity; α c specifies the
convection coefficient; Tg represents the gas temperature of the fire compartment, using
standard fire ISO 834 [6] to be applied to the lower part of the slab, φ specifies the view
factor; ε m represents the emissivity of each material (in both cases equals 0.7); ε f specifies
the emissivity of the fire; σ represents the Stefan-Boltzmann constant.

3.2 Matlab PDE toolbox

The PDE toolbox from Matlab was used for the analysis of this thermal model, using the
finite element method [13]. The maximum size of the finite element mesh is 0.01m, see
Figure 2. The thermal properties (specific heat, density and conductivity) of the materials
(concrete and steel) are temperature dependent. The exposed side is submitted to a heat flux
by convection and radiation, using different view factors and a bulk temperature following the
standard fire. The unexposed side is submitted to a convective heat flux (including the
radiation heat flux), using a constant bulk temperature of 20ºC. The model considers 1.2 mm
for the thickness of the steel deck, the geometric ratio L1/L2 is equal to 105/60 and the
concrete depth above the deck measures h1=40 mm. The mesh uses triangular finite elements
with 3 nodes and one degree of freedom per node (temperature). The interpolating functions
are linear. The time increment is smaller than 1 s. The convergence criterion is based on the
heat flow calculation, for an absolute tolerance of 10-6, a relative tolerance of 10-3, a residual
tolerance of 10-4, using a maximum number for iterations equal to 25.

Fig. 2 - Finite element mesh used for the slab (L1/L2=105/60mm/mm, h1=40 mm, SDT=1.2mm).

3.3 View Factor

The view factor ( φ ) specified in the equation (4), quantify the geometric relation between the
surface emitting radiation and the receiving surface, that is dependent of the surfaces areas
and orientations, as the distance between them [14].
The view factor at the lower flange of the composite slab is given as φinf = 1 . The view factor
of the web and of the upper flange of the steel deck are smaller than one, due to the

-312-
 Proceedings TEMM2018 / CNME2018

obstruction caused by the ribs of the steel deck. This values can be calculated by Hottel´s
crossed-string method [15]. This method is also used by the Eurocode 4. The resulting
equations for the web ( φweb ) and upper flange ( φupper) view factors are calculated according to
equations (5) and (6), being the geometric parameters represented in the Figure 3.

2 2
 l −l  l −l 
h22 +  l3 + 1 2  − h22 +  1 2 
ad + cb − ab − cd  2   2  (5)
φupper = =
2ab l3

2 2
l −l   l −l 
h22 +  1 2  + (l3 + l1 − l2 ) − h22 +  l3 + 1 2 
ac + cd − ad  2   2 
φ web = = (6)
2ac l −l 
2

2 h + 1 2 
2
2
 2 

Fig. 3 - Symbols for trapezoidal sheeting (adapted from [11])

3.4 Material Properties

The thermal proprieties are temperature dependent and vary according the standards used for
composite slabs, steel and concrete [11], [16], [17]. Both properties are depicted in Figura 4
and Figure 5. The conductivity of the steel decreases with temperature and the specific heat
has a strong variation due to the allotropic phase transformation. The specific mass and the
conductivity of the concrete decrease with temperature, being the upper value used for these
simulations. The specific heat of concrete presents a peak value related with 3% in moisture
content of concrete weight.

Fig. 4 - Thermal properties for carbon steel. Fig. 5 - Thermal properties for concrete.

-313-
 Track-B: Computational Mechanics

3.5 Boundary Conditions

An initial uniform temperature is applied to all the nodes (20ºC). The lower part of the deck is
submitted to standard fire conditions, using a convection coefficient of 25 [W/m2K] and an
emissivity of the fire equal to 1. These parameters are depicted in the Figure 6. The upper part
of the slab is submitted to a convective coefficient of 9 [W/m2K] to include the radiation
effect [18].

Fig. 6 - Definition of the slab geometry and display of the boundary conditions and different view factors, that
have effect in each part of the slab.

4 EXPERIMENTAL METHOD
Two composite steel-concrete slab specimens are to be tested. Both samples represent only
one part of normal slab dimensions. This specimens allow for the verification of the fire
insulation behaviour. Each slab presents 1.15 m wide and 1.2 m long. The thicknesses of the
slabs were fixed to 40 mm. The slabs used the same proportion and quantity of reinforcement
steel as used for the normal slab dimensions. The slab model H60 is presented in Figure 7,
where one can observe the geometry of this model.

Fig. 7 - Model of the slab and the respective geometry.

Normal weight concrete is used for the specimens. The compressive strength of the concrete
is 30 MPa. The compressive strengths of the concrete is to be determined by cylinder crushing
tests conducted 7 days and 28 days after the concrete is cast. After the material
characterization, the specimens need casted, prepared and instrumented. The tests should be
conducted to a maximum of one hour or before the final time, if the slab is deformed to a
point impend of structural collapse, which may lead to damage of the furnace. The test should
also be stopped before this period if the insulation criteria is achieved, and critical time can be
determined [5].

4.1 Furnace
The fire tests are predicted to be conducted in a natural gas furnace with maximum power of
360 kW. The furnace has 4 burners located in different planes and positions. The geometry of

-314-
 Proceedings TEMM2018 / CNME2018

the furnace is depicted in Figure 8, as well as the relative position of the slab. This slab is
mounted in a special frame. This furnace is running with standard fire ISO834 [7].

Fig. 8 - Furnace geometry.

4.2 Thermocouples
The thermocouple layout was based on standards for testing EN 1363-1 [8] and the EN 1365-
2 [9], with additional thermocouples for numerical validation. More thermocouples were
included through the depth of the slab to obtain more results during the test duration. Sensors
are identified in Figure 9 and Figure 10 for all 21 thermocouples.

Fig. 9 - Top view of the thermocouples locations.

-315-
 Track-B: Computational Mechanics

Fig. 10 - Thermocouples locations in the section A-A.

4.3 Fire resistance criteria

To prevent fire propagation into adjacent compartments, slabs must meet the requirements for
fire resistance, preventing the propagation of fire and limiting the temperature of the
unexposed surface in the fire compartment. The insulation criterion (I) for fire resistance of
this construction element depends on the temperature evolution at the unexposed surface. The
performance level used to define insulation shall be the average temperature rise on the
unexposed surface limited to 140 °C above the initial average temperature, or, with the
maximum temperature rise at any point limited to 180 °C above the initial average
temperature [8]. A temperature increase of 140 °C at the unexposed side is usually taken as
the limiting insulation criterion [10], but the other condition for the maximum temperature
can also be a limiting condition.
For concrete slabs with steel decks, the integrity criterion (E) is easily verified, because
concrete slab is cast in situ, assuring that joints are correctly sealed. Possible cracks that may
occur during the tests due to fire exposure are protected by the steel deck, preventing the
penetration of flames and hot gases through the slab.

5 RESULTS
The time history for the temperature evolution was calculated in the some expected locations
for the experimental measurements. The results are plotted into two separated graphs;
according to the locations for measurements (see Figure 11 for the unexposed locations and
Figure 12 for intermediate locations). The temperature of the unexposed side is characterized
by T1, T2, T3, T4, T5, T6, T7, T8, T9, T10 and T11, while the temperature for intermediate
measurements are characterized by T12, T13, T14, T16, T18, T19 T21, T15, T20 and T17.
The average and the maximum temperature rise on the unexposed surface is based on
measurements obtained from disk thermocouples, located at or near the centre of the section
and at or near the centre of each quarter section [8], see thermocouples T1, T2, T6, T10 and
T11. The maximum temperature was determined by the highest temperature registered by any
of the unexposed thermocouples. Figure 13 represents the average (T_AVE) temperature and
maximum temperature (T_MAX) evolution in the composite slab.

-316-
 Proceedings TEMM2018 / CNME2018

Fig. 11 - Temperature evolution of the unexposed. Fig. 12 - Temperature evolution for intermediate
side locations.

Fig. 13 - Unexposed temperature evolution (T_MAX and T_AVE).

Figure 13 also presents the performance criteria for the fire insulation rating
(TMAX_C=180+20 ºC or TAVE_C=140+20 ºC). The maximum temperature is obtained
through the time history of the thermocouple T6 that is the one that presents the highest
temperature during the simulation. Five thermocouples were used for the calculation of
T_AVE (T1, T2, T6, T10 and T11). The expected fire resistance is equal to 1528 s that
corresponds to 25 min complete minutes. This value is smaller than 38 min, which is the
value determined by the simplified model proposed in the Eurocode (see Table 2).
The temperature filed is plotted for time equal to 6, 12, 19 and 25 minutes, see graphs from
Figure 14 to Figure 17.

Fig. 14- Temperature field after 6 min, for the slab geometry (L1/L2=105/60, h1=40 mm, SDT=1.2mm).

-317-
 Track-B: Computational Mechanics

Fig. 15 - Temperature field after 12 min, for the slab geometry (L1/L2=105/60, h1=40 mm, SDT=1.2mm).

Fig. 16 - Temperature field after 19 min, for the slab geometry (L1/L2=105/60, h1=40 mm, SDT=1.2mm)

Fig. 17- Temperature field for the critical time (25min), for the slab geometry (L1/L2=105/60,
h1=40 mm, SDT=1.2mm).

6 CONCLUSIONS
The numerical simulation of the thermal effects caused by a fire on a composite concrete slab
with steel deck is presented. This simulation allows to determine the fire resistance of this
structural element with regard to the insulation criterion. The numerical simulation predicts
lower fire resistance (I) when compared to actual standards, based on the simple calculation
method. This fire resistance was defined by the average temperature rise in the unexposed
side of the slab. The fire resistance obtained with the simple calculation method, proposed in
the Eurocode, seems to be unsafe because it gives a critical time value quite higher to the one
obtained with the numerical simulation. Experimental results are important to validate the
numerical results, by carrying out experimental tests according to standards, as specified in
this work.

-318-
 Proceedings TEMM2018 / CNME2018

REFERENCES
[1]-A. F. Hamerlinck, “The behaviour of fire-exposed composite steel/concrete slabs,”
Eindhoven University of Technology, 1991.
[2]-R. Hamerlinck and J. W. B. Stark, “A numerical model for fire-exposed composite steel /
concrete slabs,” in 10th International Specialty Conference on Cold-Formed Steel Structures -
International Specialty Conference on Cold-Formed Steel Structures. 5., 1990, pp. 115-130.
[3]-C. G. Bailey, T. Lennon, and D. B. Moore, “The behaviour of full-scale steel-framed
buildings subjected to compartment fires,” Struct. Eng., vol. 77, no. 8, pp. 15-21, 1999.
[4]-S. Lamont, A. S. Usmani, and D. D. Drysdale, “Heat transfer analysis of the composite
slab in the Cardington frame fire tests,” Fire Saf. J., vol. 36, no. 8, pp. 815-839, Nov. 2001.
[5]-L. Lim and C. Wade, “Experimental Fire Tests of Two-Way Concrete Slabs - Fire
Engineering Research Report 02/12,” Christchurch, New Zealand, 2002.
[6]-G.-Q. Li, N. Zhang, and J. Jiang, “Experimental investigation on thermal and mechanical
behaviour of composite floors exposed to standard fire,” Fire Saf. J., vol. 89, no. November,
pp. 63-76, Apr. 2017.
[7]-International Organization for Standardization, “ISO 834-1: Fire Resistance Tests -
Elements of Building Construction - Part 1: General Requirements.” International
Organization for Standardization, Switzerland, p. 25, 1999.
[8]-CEN- European Committee for Standardization, EN 1363-1: Fire resistance tests Part 1 :
General Requirements, CEN-Europ. Brussels: CEN- European Committee for
Standardization, 2012.
[9]-CEN - European Committee for Standardization, 1365-2: Fire resistance tests for
loadbearing elements - Part 2: Floors and roofs, CEN-Euro., vol. 44, no. 0. CEN - European
Committee for Standardization, 1999.
[10]-ECCS - Committee T3 - Fire Safety of Steel Structures, “Calculation of the fire
resistance of composite concrete slabs with profiled steel sheet exposed to the standard fire,”
ECCS: Publication 32. ECCS, Committee T3 - Fire safety of steel structures, technical note,
p. 48, 1983.
[11]-CEN- European Committee for Standardization, EN 1994-1-2: Design of composite steel
and concrete structures. Part 1-2: General rules - Structural fire design. Brussels: CEN-
European Committee for Standardization, 2005.
[12]-P. A. G. Piloto, L. M. S. Prates, C. Balsa, and R. Rigobello, “Numerical simulation of the
fire resistance of composite slabs with steel deck,” Int. J. Eng. Technol., vol. 7, no. 2.23, pp.
83-86, 2018.
[13]-I. Mathworks, “Partial Differential Equation Toolbox TM User’s Guide R2017b,” pp. 1-
1406, 2017.
[14]-J. Jiang, J. A. Main, F. Sadek, and J. M. Weigand, “Numerical Modeling and Analysis of
Heat Transfer in Composite Slabs with Profiled Steel Decking, NIST TN-1958,” 2017.
[15]-Y. A. Cengel and A. J. Ghajar, Heat and Mass Transfer: Fundamentals and Applications.
New York: McGraw-Hill Education, 2011.

-319-
 Track-B: Computational Mechanics

[16]-CEN- European Committee for Standardization, EN 1993-1-2: Design of steel structures

- Part 1-2: General rules - Structural fire design Eurocode. Brussels: CEN - European
Committee for Standardization, 2005.
[17]-CEN- European Committee for Standardization, EN 1992-1-2: Design of concrete
structures - Part 1-2: General rules - Structural fire design, vol. BS EN 1992. Brussels: CEN -
European Committee for Standardization, 2004.
[18]-CEN- European Committee for Standardization, EN 1993-1-2, Eurocode 1: Actions on
structures - Part 1-2: General actions - Actions on structures exposed to fire, vol. 2, no. 2005.
Brussels: CEN- European Committee for Standardization, 2005.

-320-