Microsoft Word - 476hernandez.docx

CHEMICAL ENGINEERING TRANSACTIONS

VOL. 43, 2015

A publication of

The Italian Association
of Chemical Engineering
Online at www.aidic.it/cet

Chief Editors: Sauro Pierucci, Jiří J. Klemeš
Copyright © 2015, AIDIC Servizi S.r.l.,
ISBN 978-88-95608-34-1; ISSN 2283-9216

Investigating the Properties of Fireproofing Materials
for an Advanced Design of Equipment Protection

Raffaela Moricone, Alessandro Tugnoli
Alma Mater Studiorum - Università di Bologna, DICAM – Department of Civil, Chemical, Environmental and Material
Engineering, LISES – Laboratory of Industrial Safety and Environmental Sustainability via Terracini 28, 40131 Bologna
Italy
a.tugnoli@unibo.it

Fire scenarios in process industry have a high potential to cause severe asset damage. Fireproofing is a
consolidated technique for passive fire protection for units and supporting structures. Since several materials
are available for passive fire protection, it is important to choose the best solution for the protected equipment
and critical fire scenarios. Current practice in rating fireproofing materials does not provide sufficient
information about the protection granted to process equipment: for example, the ‘time-to-failure’ of pressurized
vessels protected by fireproofing materials cannot be predicted from the results of standardized fire tests. This
study investigates the key properties (e.g. density, geometrical structure, thermal degradation and thermal
conductivity) of representative fireproofing materials, in order to better understand the elements underlying the
actual protection performance. An experimental activity was focused on the definition of fundamental models
to describe the thermo-physical properties of the materials. The investigation cast the foundations of a better
understanding of the dynamics underlying the effective design for passive fire protection, identifying the
criticalities and limits of the alternative fireproofing options. The changes in the physical properties of materials
during fire exposure were confirmed to play a major role on the protection performance. Such effects could not
be accounted for complex geometries by conventional simplified approaches alone: thus, the proposed
approach paves the way for a safer and more cost effective design of passive fire protection systems.

1. Introduction

Fire was responsible of several severe accidents in the process industry. Accidental fire may lead to the
escalation of severe secondary scenarios (domino effect) when it involves pieces of equipment containing
significant inventories of flammable materials (Landucci et al., 2009a; Demichela et al., 2004). In order to
reduce the severity of domino escalation, consequence mitigation is a key issue in plant design (Tugnoli et al.,
2008). Several active and passive strategies are commonly applied to reduce and/or avoid such events (see
e.g. SCI, 1992; API, 2007). Passive fire protection (PFP) avoids the rapid increase of temperature in the
protected items and the deterioration of the mechanical properties of the structural components. Thermal
coatings, known as fireproofing materials, are suitable solutions for the application of PFP in both mobile and
fixed installation (see e.g. Paltrienieri et al., 2009). The protection performance in fireproofing layers is
achieved by characteristics as high thermal stability and low thermal diffusivity. The PFP delays the heat-up of
fired units for a sufficient time lapse for the deployment of emergency teams and for starting fire suppression
measures (see e.g. Tugnoli et al., 2012; Di Padova et al., 2011). Several approaches are available to assess
the performance of a PFP material. While a complete assessment is provided only by real scale experiments,
where the test is carried out on vessels exposed to fire (see e.g. Landucci et al., 2009b; VanderSteen et al.,
2003), these should be avoided because of risks, complexity and costs. As an alternative, simulation analysis
and/or smaller scale tests can be effectively used to investigate the properties and to understand the
technological issues (e.g. preparation, formulation and design) of innovative materials (Argenti et al., 2014).
However, it is crucial to collect detailed information and to be able to model the thermodynamic and transport

DOI: 10.3303/CET1543399

Please cite this article as: Moricone R., Tugnoli A., 2015, Investigating the properties of fireproofing materials for an advanced design of
equipment protection, Chemical Engineering Transactions, 43, 2389-2394 DOI: 10.3303/CET1543399

DOI: 10.3303/CET1543399

2389

Table 1: Experimental test protocols
Instrument Application Test conditions

Thermogravimetric
analysis (TGA Q500 from TA
instruments)

• Weight loss analysis
• Development of an

apparent kinetic model

• Temperature ramps up to 800 °C
• Constant heating rates: 5÷45 °C/min
• Inert gas (N2) atmosphere
• Sample mass: 1÷45 mg

Differential Scanning
Calorimetry (DSC Q2000
from TA instruments)

 Thermal effect of reaction
 Heat capacity
 Validation of TG runs

 Temperature ramps 550 °C
 Constant heating rates: 5÷45 °C/min
 Inert gas (N2) atmosphere
 Sample mass: 1÷20 mg

Fixed bed tubular reactor
(FBR) heated by Carboline
HST 12/300 furnace equipped
with W301 controller

• Analysis of sample swelling
• Validation of TG runs
• Produce solid residue for

other tests

• Temperature ramps up selected final T
• Constant heating rates: 10 °C/min
• Inert gas (N2) atmosphere
• Sample mass: 1÷20 g

Thermal conductimeter
(TPS1500S from Hot Disk
AB)

 Thermal conductivity
 Heat capacity

 80x80 mm slab samples
 Tests from RT to 700 °C
 Time residence in furnace 150 min
 Inert gas (N2) atmosphere, 500 L/h
 Sample mass: 50÷120 g

Liquid pycnometer • Density
• Void fraction

• Sample mass: 1÷20 g
• Liquids reference: water, xylene

Table 2: Apparent thermal conductivity: epoxy intumescent coating and lightweight cement
Model Main equation Notes
Bulk thermal conductivity

= (1 − ) +

• bulk thermal conductivity
• thermal conductivity of solid fraction
• thermal conductivity of the pores
• = 1 − pore fraction

Thermal conductivity of
solid fraction = ,  , thermal conductivity of the p-th solid pseudo-material

 mass fraction of pseudo-material p
Thermal conductivity of
pores

= +

• thermal conductivity of the gas contained
inside pores

• thermal conductivity of the radiative
heat transfer

Thermal conductivity of
radiative heat transfer

= 4(2 ) − 1⁄  Stefan-Boltzmann constant  mean pore diameter
 emissivity of pore surface
 absolute temperature

properties of PFP materials when exposed to fire: only this way a reliable evaluation of PFP performance by
advanced simulation techniques is possible. In the current study, three materials were chosen as a
representative reference set of PFP materials: intumescent resin, lightweight concrete and inorganic fiber. A
methodology for investigating and modelling the properties of fireproofing materials is presented. The
approach relies on the experimental characterization of properties such as thermal conductivity, density,
porosity and thermal effect of the degradation. Constitutive equations are proposed for these data, based on
the experimental findings. The discussion of the result will show as significant variations of these properties
occur and should be accounted, by the proposed models, in the advanced design of equipment protection by
PFP systems.

2. Experimental activity

2.1 Materials
The reference materials considered in this study are an epoxy intumescent coating, a lightweight cement-
based coating and a silica fiber blanket. The epoxy intumescent coating (supplied by International Protective

2390

Coatings) is formed by an epoxy resin, ammonium polyphosphate (APP), boric acid, and fillers (e.g. glass
fibers, magnesium silicates, limestone, etc.). It is a spray-applied coating; the analysed specimens were
prepared by the conventional procedure with average thickness approximately equal to 9 mm. Intumescent
coating are defined by the American Petroleum Institute (API) as an active insulator: it was recently confirmed
(Gomez-Mares et al., 2012a) that the analysed coating undergo a significant physical and/or chemical change
when exposed to heat.
The lightweight cement coating (supplied by Cafco), is constituted principally by Portland cement, vermiculite
and other inorganic fillers. It is a spray-applied coating with a density lower than the one of common cement;
in typical industrial applications, the average thickness applied is about 24 mm. The silica blanket (supplied by
Insulcon) represents the family of synthetic insulating fibers and is formed by silicon dioxide, aluminium
dioxide (together constitute 97 % of the mass), and binders; its nominal thickness is equal to 12 mm. Both the
two latter materials are defined inactive insulators, because they are not inherently designed for modification
during fire exposure.

2.2 Experimental characterization
The experimental analysis investigated the thermo-physical properties of the materials, focusing in particular
on thermal conductivity, a key property for the insulation performance. A set of different experimental
techniques was used. Table 1 reports the main references of the small-scale tests carried out.

Table 3: Effective thermal conductivity: synthetic fibers
Contribution model Main equation Notes
Effective thermal
conductivity

= +

• effective thermal conductivity
• thermal conductivity

through fiber and gas phase
• thermal conductivity

radiation
Conductivity of gas and
solid

= (1 − ) +  thermal conductivity of fiber
 thermal conductivity of gas

Thermal conductivity of
fiber insulating

= ∗

• geometrical parameter
• ∗ thermal conductivity of the bulk

fiber material
• solid fraction ratio: 1-
• power varying between 1 and 3

Thermal conductivity of
gas phase = 2Φ + Ψ 2 − 2+ 1 1

• gas thermal conductivity
• Φ,Ψ parameters depend on

Knused number
• accommodation coefficient equal

to 1
• specific heat ratio
• Prandtl number
• Knused number

Knused number
= √2π P 4  Boltzmann constant  absolute temperature

 gas collision diameter
 pressure
 fiber diameter

Thermal conductivity
radiation = −163β • T absolute temperature • refraction index

• Stefan-Boltzmann constant
• = extinction coefficient, >2 constant parameter and

characteristics length

2391

Table 4: Apparent kinetic for epoxy intumescent coating
Model Main equation Notes
Apparent kinetic
( ) j

j
m

j
RT
E

j
j eA

dt
d

ξ
ξ

−=
−

• mass conversion in region j
• apparent pre-exponential factor for j
• activation energy of region j

Real density

1
−











⋅Σ=

jS
jjreal ρ

ωρ

• gas constant
• absolute temperature
• reaction order of region j
 real density of the solid fraction
 , density of pseudo-material j
 mass fraction of pseudo-material j

Swelling factor






 ⋅Σ= jjj a ξψ )ln(exp

• swelling factor
• swelling empirical material

Apparent density

( )
ψρ

ξα
ρ

⋅

⋅−
=


0

1
j jj

app

 apparent density
 weight loss in region j
 accommodation coefficient equal to 1
 density of virgin sample

3. Property models

The future activities of simulation of the behaviour of insulating materials exposed to fire require the definition
of reliable constitutive equations for the main material properties (density, thermal conductivity, thermal
stability, etc.). Focusing on thermal conductivity, this is mainly a function of three key physical parameters:
material porosity, thermal conductivity of the solid matrix and thermal conductivity of the gas inside the pores
(Gomez-Mares et al., 2012b). Table 2 reports the simplest bulk thermal conductivity model proposed to predict
conductivity of epoxy intumescent coating and lightweight cement, as applied by several earlier researches
(Kantorovich et al., 1999) for porous materials. The same model was applied to the reference fireproofing
materials analysed in current study.
In materials with an high gas fractions (i.e. high porosity), such as the silica blanket, the radiative contribution
to apparent thermal conductivity may pay a significant role at higher temperatures, as shown by Moricone et
al., 2014. In this case the generic model proposed above may provide inadequate results. A specific effective
thermal conductivity developed for synthetic fibers is presented in Table 3. This model embeds advanced
equations for heat transport in the gas phase (Kennard, 1938), for radiative heat transport (Siegel and Howell,
1992) and for solid conduction in a fibrous media (William et al., 1993).
Reactive materials, such as the epoxy intumescent coatings, need a kinetic model to be integrated into the
constitutive equations for physical and transport properties. Apparent kinetics can be obtained by various
techniques: Gomez-Mares et al. (2012b) used TGA runs to develop a kinetic model for this kind of material.
The swelling mechanism of the material, and therefore apparent density and pore fraction can be connected to
the conversion term by the approach described in Table 4.

4. Results and discussion

4.1 Thermal stability during heating
As for Table 1, the thermal decomposition of the representative PFP materials was studied by TG and DSC
analysis. Figure 1-A shows the weight losses determined in nitrogen atmosphere. The silica blanket (curve c)
presents moderate weight losses (about 10 %) mostly due to water evaporation and degradation of binders,
so the material was considered thermally invariant up to temperatures as high as 800 °C. The lightweight
concrete (curve b) shows some weight loss, which becomes relevant (about 27% of the initial weight) only at
temperatures above 600 °C. These can be interpreted as losses of carbon dioxide and water (Alarcon-Ruiz et
al., 2005; Villain et al., 2007), and correspond to endothermic effects in DSC analysis (Figure 1-B). The epoxy
intumescent coating (curve a) clearly demonstrate the need of a kinetic model in order to track property
variation: degradation starts at low temperatures with dehydration of boric acid (completed at 260 °C, with
losses about 12 % of initial weight) and continues in several steps up to total weight losses of about 65 % of
initial weight at 800 °C. The DSC data show as the dehydration of boric acid is an endothermic process (about
202 mJ/mg, indicated in Figure 1-B with point 1a), while the weight loss between 260 °C and 550 °C,

2392

associated to the degradation of epoxy resin and APP (point 2a of the DSC signal) is an exothermic process
(about 142 mJ/mg). In the latter step, the degradation generates the swelling phenomenon with char
formation. These phenomena are in agreement with the results of studies on similar materials (Jimenez et al.,
2009; Jimenez et al., 2006) and they were confirmed by test conducted in FBR. The char residue at 800°C is
equal to 35% of the initial mass. These deep modifications of the material and the thermal phenomena
associated are expected to have a significant role in the definition of the fireproofing performance.

4.2 Fireproofing materials properties
The experimental characterization according to the protocols defined in Table 1 allowed the identification of
the parameters of the models discussed in section 3. The results for the major properties are shown in Figure
2-A and Figure 2-B, considering a temperature ramp at constant heating rate (10°C/min).
The apparent density and porosity of silica blanket remains almost constant with temperature (around 95 %)
due to the lack of degradation phenomena. On the other hand, the lightweight cement coating shows a density
reduction at high temperatures due to the increasing of pore fraction as a consequence of calcination
processes. The epoxy intumescent coating suffers a dramatic change in density due to the swelling and
significant weight loss; the porosity changes from initial 0.13 to 0.93 when the material is exposed to final
temperature.
Figure 2-B shows the results obtained from experimental measuring, using TPS method, and thermal
conductivity models for three different fireproofing materials. As shown in the figure, a good agreement is
present between experimental data and model predictions for all the reference materials. The thermal
conditions affect the properties of materials, it has been observed that the drastic increase of porosity, in the
case of active fireproofing materials , leads to a decrease of effective conductivity, favouring the its protective
action. This increase, however, is linked to the degradation of the material that leads to problems of failure of
the structure. In inert materials, instead, as silica blanket, a high degree of porosity allows maintaining a low

Figure 1: (A) Results of the constant heating rate (10°C/min) in TG runs on the reference samples (a) epoxy
intumescent coating, (b) lightweight cement and (c) silica blanket. (B) Results of DSC test.

Figure 2: (A) Apparent density and Pore fraction in thermal trend for the reference samples (a) epoxy
intumescent coating, (b) lightweight cement and (c) silica blanket ; (B) Thermal conductivity results as a
function of the temperature during constant heating rate (10 °C/min) .

2393

thermal conductivity in the range of temperatures considered. In the case of the fibrous material has also been
noted that the contribution of the radiation is not negligible for temperatures higher than 500 ° C, while it is for
the intumescent material.

5. Conclusions

The thermal behaviour of three different commercial fireproofing materials was investigated. The key
properties defining the performance of the fireproofing were experimentally measured by an array of
laboratory-scale techniques. Constitutive equations were proposed for modelling these properties, in view of
future application in advanced simulation of the fireproofing system (e.g. FEM). The behaviour of the material
during simulated fire exposure evidenced large changes in the values of key parameters such as thermal
conductivity and apparent density. The role of these changes, frequently unrecognized or inadequately
modelled, is expected to pay a primary role in the effectiveness of the passive fire protection system. The
investigation method proposed in current paper, avoiding expensive large-scale tests, paves the way to an
improved analysis of PFP based small-scale tests and performance simulation.

References

Alarcon-Ruiz L., Platret G., Massieu E., Ehrlacher A., 2005, The use of thermal analysis in assessing the
effect of temperature on a cement paste. Cement and Concrete Research 35, 609.

API 2007. API 14G-2007, Recommended practice for fire prevention and control on fixed open-type offshore
production platforms (4th ed.), American Petroleum Institute, API, Washington, D.C.

Argenti F., Landucci G., 2014, Experimental and numerical methodology for the analysis of fireproofing
materials, Journal of Loss Prevention in the Process Industries, 28, 60-71.

Demichela M., Piccinini N., Poggio A., 2004, Analysis of an LPG accidental release, Process Safety and
Environmental Protection 82 (2 B), 128-131.

Di Padova A., Tugnoli A., Cozzani V., Barbaresi T., Tallone F., 2011, Identification of fireproofing zones in Oil
& Gas facilities by a risk-based procedure, J Hazard Mater., 191(1-3), 83-93.

Gomez-Mares M., Tugnoli A., Landucci G., Barontini F., Cozzani V., 2012a, Behavior of intumescent epoxy
resins in fireproofing applications, Journal of Analytical and Applied Pyrolysis, 97, 99–108.

Gomez-Mares M., Tugnoli A., Landucci G., Cozzani V., 2012b, Performance assessment of passive fire
protection materials, Industrial and Engineering Chemistry Research, 51 (22) , 7679-7689.

Kantorovich I., Bar-Ziv E., 1999, Heat transfer within highly porous chars: a review, Fuel, 78, 279.
Kennard E.H., 1938, Kinetic Theory of Gases, McGraw-Hill, New York, NY, 163-318.
Jimenez M., Duquesne S., Bourbigot S., 2009, Kinetic analysis of the thermal degradation of an epoxy-based

intumescent coating, Polymer Degradation and Stability, 94, 404.
Jimenez M., Duquesne S., Bourbigot S., 2006, Characterization of the performance of an intumescent fire

protective coating”, Surface & Coatings Technology, 201, 979.
Landucci G., Molag J., Cozzani V., 2009a, Modeling the performance of coated LPG tanks engulfed in fires,

Journal of Hazardous Materials, 172, 447-450
Landucci G., Molag M., Reinders J., Cozzani V., 2009b, Experimental and analytical investigation of thermal

coating effectiveness for 3m3 LPG tanks engulfed by fire, Journal of Hazardous Materials, 161, 1182.
Moricone R., Tugnoli A., Villa V., Cozzani V., 2014, Analysis and testing of fibrous passive fire protection

materials for fireproofing of equipment units, Chemical Engineering Transactions, 36, 331-336.
Paltrinieri N., Landucci G., Molag M., Bonvicini S., Spadoni G., Cozzani V., 2009, Risk reduction in road and

rail LPG transportation by passive fire protection, Journal of Hazardous Materials, 167, 332-344.
SCI Steel Construction Institute 1992, Availability and properties of passive and active fire protection systems

(OTI 92 607). United Kingdom: Health and Safety Executive.
Siegel R., Howell J.R., 1992, Thermal Radiation Heat Transfer, Taylor & Francis, London, UK.
Tugnoli A., Cozzani V., Di Padova A., Barbaresi T., Tallone F., 2012, Mitigation of fire damage and escalation

by fireproofing: A risk-based strategy, Reliab Eng Sys Safe., 105, 25-35.
Tugnoli A., Khan F., Amyotte P., Cozzani V., 2008, Safety assessment in plant layout design using indexing

approach: implementing inherent safety perspective - Part 2 – Domino Hazard Index and case study,
Journal of Hazardous Materials, 160, 110–121.

VanderSteen J.D.J., Birk A.M., 2003, Fire tests on defective tank-car thermal protection systems, Journal of
Loss Prevention in the Process Industries, 16, 417-425.

Villain G., Thiery M., Platret G., 2007, Measurement methods of carbonation profiles in concrete:
thermogravimetry, chemical analysis and gammadensimetry. Cement and Concrete Research, 37, 1182.

Williams S.D., Curry D.M., 1993, Prediction of Rigid Silica Based Insulation Conductivity, NASA TP-3276.

2394