General aspects of fire performance of Sandwich Panels

The purpose of this chapter is to show how the different components of a sandwich panel (exterior surfaces, core and adhesives) can influence the performance of the panel in the presence of fire and how, with careful design, these panels can meet the stringent fire performance requirements. In fact, if properly designed and used, fire-rated sandwich panels are certainly the best solution in all building applications where fire barriers are required.

It should be remembered that there is currently a tendency to assess the risk of a building, in the event of fire, on the basis of the expected performance of the building as a whole, rather than that of the individual parts of which it is composed, such as sandwich panels. An important aspect of this assessment is related to the installation of additional fire protection measures such as smoke detectors and water jet extinguishing systems.

In any case, insurance companies may require passive fire resistance, for example in the form of fire-resistant partitions.

The decision to use sandwich panels in a building may be influenced by general safety objectives, such as personnel safety, prevention of building loss, and environmental protection:

protection of personnel: building occupants, firefighters, and members of the public in the vicinity of the building may be exposed to the risks associated with the occurrence of a fire. Therefore, the building must be designed and constructed in such a way that:
the bearing capacity of the construction is guaranteed for a satisfactory period of time;
the generation and spread of fire and smoke inside the building is limited;
the spread of the fire to surrounding buildings is limited;
occupants can leave the building or be rescued by other means;
the safety of the recovery equipment is taken into consideration;
environmental protection: the factors to be considered are related to limitation:
of the effects of the fire on adjacent buildings or structures;
of gaseous combustion products and fibers released into the atmosphere;
of contamination of the water-bearing slopes after decommissioning operations;
prevention against property loss: a fire can have a significant impact on the survival of a business, so care must be taken to limit the damage:
in the building structure;
in the contents of the building itself;
in the structures vital to the continuation of business operations;
in the public image.

Damage to health related to the occurrence of fire

When organic materials, such as polyurethane foam, burn, the most obvious products of combustion are fumes and hot gases. These products can pose a serious health threat.

The nature and amount of toxic products developed by the combustion of polyurethane foam depends on the conditions of decomposition of the polyurethane foam. Low-intensity flameless fires generate mostly isocyanates and carbon monoxide (CO), while well-ventilated and fully developed flaming fires produce CO and, at temperatures above 800 °C, hydrogen cyanide (HCN) in relatively low concentrations.

Non-combustible materials generate very low levels of fumes and toxic products. Materials produced at elevated temperatures, such as rock wool, are completely inert in their base form, but the agents used to bind the fibers and the adhesives used to bond the slabs to the metal surfaces of the panel may produce small amounts of smoke and toxic products.

Factors to consider when using flammable materials in the construction industry are as follows:

toxicity and irritant capacity of combustion products: examination of fire victims reveals that the most common cause of death is related to the inhalation of toxic gases, mainly carbon monoxide and hydrogen cyanide; the main effects are eye and skin irritation, nausea, headache, dizziness, vomiting, drop in blood pressure, loss of consciousness, convulsions and symptoms of asphyxia;
temperature intensity/irradiation of combustion products: in case of fire, hot gases can threaten vital functions by damaging, due to the high temperature, the respiratory tract, and by causing, due to the effect of the radiated heat, skin pains and severe burns;
obscuration caused by the products of combustion: smoke in a burning building impedes normal vision, making egress more difficult and delaying extinguishing operations;
unexpected structural behavior: in rare circumstances, unexpected structural collapse of a sandwich panel can constitute a danger to human life. For example, fire on the exterior façade of a building can cause the separation of the exterior metal surface of a panel and its consequent fall onto the people below, just as an interior fire can lead to the delamination of a roof panel, if there is no mechanical connection of its two surfaces to the supporting steel structure.

Phases of a fire

The fire cycle is divided into three distinct phases: the initiation and growth phase, the fully developed phase, and the extinction phase.

In the first phase of initiation and growth, the heat produced is still very low. The process can occur with both slow combustion and flame, and ends when the volatile products of combustion begin to burn.

The second phase, the fully developed phase, begins with the so-called flashover,

in correspondence of which all flammable sources in the immediate vicinity improvisedly participate in the fire. Flashover is essentially a phenomenon associated with thermal instability (Fig. 9.2). This leads to an immediate increase in flue gas temperatures and in the energy emitted by irradiation. In this phase the temperature of the gases is very high and may vary within a range of 700-1200 °C.

The final phase is the extinction phase, which is characterized by a drop in flue gas temperature.

It is necessary to consider that real fires have very variable characteristics, since their duration and the maximum temperatures reached depend mainly on the amount of oxygen available, the quantity and porosity of the flammable material, and the thermal insulation characteristics of the structure in which the fire develops. For these reasons, the standard time-temperature curves used in the standards are arbitrary and can only marginally represent a real fire, but they are essential tools for comparing the performance offered by the products.

Reaction to fire and fire resistance concepts

The reaction to fire is the degree of participation of a sandwich panel in a fire to which it is subjected; therefore, it is a certain behavior that assumes extreme importance in the initial phases of a fire.

In the growth phase of a fire, the following important fire reaction properties can be distinguished:

flammability: this property determines the difficulty with which a material can be ignited by a small flame or a small heat source;
non-combustibility: this property determines the difficulty with which a material burns, develops heat, and produces smoke and toxic/irritating substances;
rate of heat produced: the rate of heat produced influences the amount of combustion products generated and the rate at which these products are transported through the building;
flame propagation: this property indicates the speed with which the flame propagates along the surface of a material;
liquefaction and shrinkage: some materials, such as polystyrene, shrink with the heat source and release droplets of molten material, which ignite on contact with the fire, contributing to the spread of the flame front;
corrosiveness: many of the materials used in the construction industry produce, after combustion, gases that are corrosive to metals, such as hydrogen chloride; some polystyrene foams, loaded with flame-retardant agents, and PVC, widely used in the insulation of electrical cables, can have corrosive effects.

If the reaction to fire properties are extremely important in the initial growth phase of a fire, the fire resistance behavior is important once the flashover phenomenon occurs. Fire resistance, as shown in Figure 9.3, represents the ability of an element for the construction of:

resist structural collapse;
resist the penetration of flames and hot gases (while maintaining structural integrity);
keep the surface not exposed to fire sufficiently cool so as not to initiate any combustion in the materials that may be in contact with it.

The fire resistance is expressed as the time, in minutes, during which the tested specimen satisfies these requirements.

Reaction to fire

Classification in accordance with EN 13501-1

The objective of EN 13501-1 is to define a harmonized procedure for the classification of the reaction-to-fire performance of construction products and building elements. The classification is based on the following test methods:

non-combustibility test (prEN ISO 1182): this test method identifies products that do not contribute, or do not contribute significantly, to fire;
higher calorific value (prEN ISO 1716): this test method determines the maximum heat production in the unit of time by an element subjected to combustion;
– Single Burning Item test (EN 13823): this test assesses the potential contribution of a panel to the development of a fire, in a scenario simulating a single burning item in a corner of the room;
– flammability test (prEN ISO 11925-2): this test assesses the flammability of a product subjected to a small flame.

Once these tests are performed, the results are used to issue a classification list, the purpose of which is to provide a harmonized method of classifying the tested product. Based on the test results, the product is assigned a specific Euroclass according to the following table 9.1.

Single Burning Item (SBI) Test

Most of the building products sold in Europe must be tested and classified by a test method called Single Burning Item (SBI), i.e., insulated burning element, carried out according to EN 13823 – “Reaction to fire tests for building products – Building products, excluding floors, exposed to thermal stress from an insulated burning element”.

The smoldering burning insulated element (SBI) test is a method to determine the fire performance of a panel in the corner of a room (excluding floors), when exposed to thermal stress from a smoldering insulated element (a propane-fired burner) placed in that corner (Fig. 9.4).

Isolated burning

The test apparatus (Fig. 9.5) consists of a chamber with a length of 3 m, width of 3 m, and height of 2.4 m, and provided with 2 windows that allow a view of the sample during the test.

The chamber is provided with an opening through which a movable support can slide, in which the test sample is placed; the movable support, upon entering the chamber, is placed inside a frame, positioned near one of the walls of the test chamber.

The frame supports a rectangular hood through which the combustion products are collected with a suitable suction system. This system, consisting of a duct with an internal diameter of 315 mm, incorporates a measuring section, equipped with thermocouples and sensors through which parameters such as temperature and differential pressure, necessary to calculate the capacity of the combustion products, are measured; the measuring section also contains a probe for measuring the rate of heat produced, and the relative guides to reduce the turbulence of the flow.

The drop in visibility caused by the smoke is determined by a white light lamp and a photocell system.

The test specimen consists of two vertical wings, arranged at a right angle; in correspondence with this angle, two C-shaped steel elements are used to fix the panels (Fig. 9.6).

The dimensions of the wings of the test specimen are:

(495 ± 5) mm wide by (1500 ± 5) mm high;
(1000 ± 5) mm wide by (1500 ± 5) mm high.

The sample is mounted on a mobile support (Fig. 9.7) placed below the fume and combustion product aspiration system; the mobile support is equipped with a triangular propane burner whose sides have a length of 250 mm, and is capable of generating a power of 30 kW.

The reaction of the sample to the action of the burner is controlled both visually and with the appropriate instruments. The parameters measured, both during and after the burner action, are as follows (Fig. 9.8):

temperature increase;
O2 and CO2 concentrations;
attenuation of light caused by smoke;
Lateral Flame Spread (LFS), i.e., whether the flame spread reaches the edge of the sample farthest from the corner;
extent of the damaged area of the sample;
droplets or particles of ignited material falling;
any other relevant aspects of the sample are noted during the test.
The SBI component of the Euroclassification is based on the following parameters:
Fire Growth Rate Index (FIGRA);
Total heat produced in the first 10 minutes (THR600s) after burner ignition;
Smoke Growth Rate Index (SMOGRA);
Total smoke production in the first 10 minutes (TSP600s) after burner ignition.
The Commission has defined the criteria for the assignment of classes to the panels tested; these are shown in Table 9.2 below.

Other classifications refer to:

smoke production – classes S1, S2, S3;
release of inflamed droplets – classes D0, D1, D2,

with the highest numbering classes assigned to materials with a higher tendency to produce fumes and release flaming droplets.

Therefore, recalling these classification criteria and Table 9.1, the following types of panels should be included in the corresponding classes indicated:

Mineral wool panels (also acoustic) A2 S1 D0
Firemet B S1 D0 Panels
PIR panels B S2 D0
Panels PUR-B2 B S3 D0
PUR-B3 C S3 D0 Panels

In Figures 9.9 and 9.10, which show two samples subjected to SBI testing, it can be seen how the reaction to fire of the PIR foam panel is lower than that of the B3 foam panel.

Loss Prevention Certification Board (LPCB) and Factory Mutual (FM) Approvals

As already mentioned in Chapter 6, the sandwich panel industry in recent years has been subjected to increasingly stringent fire performance requirements, not only because of the great attention paid to safety issues in the construction field, but also because of the ever-increasing pressure exerted by insurance companies, which are looking for ever greater guarantees.

Consequently, some insurance companies, such as Factory Mutual (USA), and Lloyds (UK), have set up in-house engineering departments, the most important of which are the FMRC (Factory Mutual Research Corporation), and the LPCB (Loss Prevention Certification Board), developing in-house test methods for fire performance assessment.

The two main approvals, which make it possible to assess the degree of involvement of a given panel system in a fire to which it is subjected, are:

Approval in LPS 1181 of the Loss Prevention Certification Board (LPCB), and
Factory Mutual Standard 4880 Factory Mutual (FM) approval.

Approval in the LPS 1181 of the LPCB

The LPS 1181 test assesses the performance of a sandwich panel system to determine its contribution to the development of a fire. It allows testing not only of panels, but also of joining methods, providing a more accurate assessment of actual fire performance than that indicated by the small-scale reaction-to-fire tests used by the usual certifying bodies. Therefore, sandwich panels, which meet the requirements imposed by LPS 1181, do not offer a significant contribution to fire growth, when used in wall/covering combinations, consistent with the test configuration provided by the same regulation.

The LPS 1181 test uses a small room with an open front wall, built on a solid floor of commentary and other material, having a length equal to 10 m, width equal to 4.5 m, and height equal to 3 m (Fig. 9.11 and 9.12).

On the open side of the room, a 750 mm high partition wall was used to recreate an opening with a height equal to 2250 mm. In addition, the room is provided with an air intake and an observation window on the B wall.

A wooden pyre is placed asymmetrically in the right corner of the test chamber on the side opposite to that on which the air intake is located. The pyre should be placed on a support so that the base of the pyre is 760 mm above the pavement (Fig. 9.13).

After the pyre has been ignited (Fig. 9.14), the timer and all temperature recording instruments are started simultaneously. Detailed observations are made during the test regarding the general behavior of the panel structure, including the time of initiation, duration, and position of each flare, any possible deformation of the panels, and delamination of the metal surfaces; in general, a photographic and recorded record of the test is also made for the entire duration of the test.

The test lasts approximately 30 minutes, after which the performance of the panel system is evaluated according to the following criteria (Fig. 9.15 and 9.16):

flashover: no flashover must occur on the ceiling, which means that the average temperature inside the room must not exceed the flashover temperature of 600 °C;
flame propagation on the inner surface: no flames* should occur on the inner surface beyond 1.5m from the perimeter of the pyre in both horizontal directions;
flame propagation on the outside surface: no flame propagation* should occur at any point on the outside surface of the test chamber. There must be no flame penetration to the outside through the joints;
hidden combustion (insulating layer): compliance with this requirement can be determined during the visual post-test analysis and can be assessed on the basis of the extent of damage described in point 6;
Falling pieces of flaming material: no pieces of flaming material should fall from the ceiling to the outside of the area around the pyre defined in Fig. 9.16;
damage entity: compliance with this criterion is certified by the laboratory and checked by the LPCB on the basis of existing data.

A panel system that passes such a test is approved as LPS 1181 Grade B. In either case, if the panel system is also tested for fire resistance (discussed later in this chapter), in accordance with LPS 1208, and has achieved at least 30-minute integrity and 15-minute insulation, it may be approved as LPS 1181 Grade A.

LPCB approval mark

The LPCB approval mark for use on badges, literature, advertising, packaging and other graphic forms is shown in Fig. 9.17.

Factory Mutual (FM) Approval

In contrast to LPS 1181, Factory Mutual has created a different classification for the reaction-to-fire performance of sandwich panels which, since they are to be used in the construction industry, need to be included in Class 1.

The requirements that sandwich panels must meet in order to be classified as Class 1 according to the FM reaction to fire classification are indicated in the following standards:

Factory Mutual Standard 4880, which includes approval requirements for interior wall and ceiling panels, as well as fire test requirements for exterior wall panels;
Factory Mutual Standard 4881, which includes approval requirements for exterior wall panels;
Factory Mutual Standard 4471, which includes approval requirements for cover panels; a Class 1 cover panel meets the criteria defined by this standard for resistance to fire, wind, footfall, hail, and water tightness.

The reaction to fire tests prescribed by the above-mentioned standards to assign class 1 classification to the panels are as follows:

corner test with height 25 ft (7.6 m);
corner test with height 50 ft (15.2 m);
room test, carried out in accordance with ISO 9705.

Description of tests

The structure used for the corner test with a height of 25 ft (7.6 m) consists of a steel frame on which the wall and roof/roof panels to be tested are mounted.

The samples should be mounted to cover the full height of both walls, in the area from the corner to 6.10 m in both horizontal directions, and the upper half of the walls from 6.10 m to 11.58 m on the south wall, and to 15.24 m on the east wall (Fig. 9.18); plaster tables, with a thickness of 16 mm, should be used to cover the remaining non-paneled sections of these walls. To obtain a continuous test surface, specimens should be installed on the inside wall of the steel frame.

After having installed the thermocouples for temperature detection in the positions indicated by the standard, a pyre of 336 kg of wood pieces, placed near the corner, is ignited.

The duration of the test is 15 minutes, during which the signals derived by the thermocouples are recorded at intervals of no more than 10 seconds.

In addition:

a recorded record of the test is made from the moments of preparation to its completion;
black and white and color photographs are taken before the test, during the test at intervals not exceeding one minute, and after completion of the test once the smoke has cleared and the test structure has cooled;
detailed observations are noted or recorded before the test, when significant events occur during the test, and after the test as soon as visibility and temperature conditions permit.
During the test there should be no significant instances of air venting through the wall and cover/roof panels. These purges cause a significant cooling of the affected area and a reduction in flame propagation, with a consequent significant reduction in the accuracy of the results obtained.
The corner test structure with a height of 50 ft (15.2 m) is a steel frame consisting of horizontal and vertical supports on which the wall and cover/roof panels are mounted.
Both frame walls are 6.10 m long and form a 90-degree angle at the intersection point. The distance between the concrete floor and the roof framing is equal to 50 feet (15.24 m). The roof truss forms an isosceles triangle whose sides have a length of 6.10 m (Fig. 9.19).
The test is carried out by lighting a pyre of 336 kg of wood pieces, placed near the test angle. The duration of the test is 15 minutes, during which the same procedures for temperature detection and visual data as described for the 25-ft corner test are followed.
The ISO 9705 room test assesses the reaction-to-fire characteristics of a building product by igniting a fire in the corner of a small room with a single door-like opening in one of its walls with a shorter length.
The test apparatus is a small room constructed of non-combustible material, such as cinder blocks, with dimensions of 2.4 m wide, 3.6 m long and 2.4 m high (Fig. 9.20).
The sample to be tested is obtained by mounting the panels both on the ceiling and on the walls of the room, except on the wall with the opening.
A propane gas burner is placed in one of the corners of the room and produces a power output of 100 kW for the first 10 minutes, and 300 kW for the next 10 minutes; the total test time is equal to 20 minutes.
The combustion gases are collected with an aspiration system, whose hood is placed outside the room in front of the entrance door, and in which the heat produced in the unit of time, the total heat produced, and the quantity of fumes released are measured. The flame propagation along the walls and roof is analyzed and described based on purely visual observation. If the flames come out of the entrance door, this means that the flashover phenomenon has occurred and the test is concluded (Fig. 9.21).

Requirements

As far as corner tests are concerned, the standard prescribes that:

for a Class 1 approval at a maximum height of 30 feet (9.1 m), the panel assembly shall not support any self-propagating fire that reaches even a single edge of the test structure at a height of 25 feet (7.6 m), with obvious fire and material damage;
for a Class 1 approval at a maximum height of 50 feet (15.2 m), the panel assembly shall satisfy the requirements for Class 1 approval at a maximum height of 30 feet (9.1 m) and shall not support any self-propagating fire that reaches even a single edge of the test structure at a height of 50 feet (15.2 m), with obvious fire and material damage;
for a Class 1 approval with no height bond, the panel assembly shall satisfy the requirements for Class 1 approval at a maximum height of 30 feet (9.1 m), shall not support any self-propagating fire that reaches even a single edge of the test structure at a height of 50 feet (15.2 m), with obvious fire and material damage, and shall not initiate fire in the roof panels in the angle test at a height of 50 feet (15.2 m).

Regarding the room test (ISO 9705), a panel assembly:

shall not encourage any self-propagating fire out of the test room within 20 minutes of the test duration, with obvious fire and material damage;
shall not generate an excessive amount of fumes during the test, and
shall sustain the applied load (if present) for the duration of the test.

FM approval marks

The diamond-shaped FM approval mark for use on plates, literature, advertising, packaging and other graphic forms is shown in Fig. 9.22; when reproduction of the mark is impossible, a modified version of the diamond is suggested (Fig. 9.23).

Italian reaction-to-fire classification (Ministerial Decree 26.6.84)

In Italy, the determination of the reaction-to-fire characteristics of materials is regulated by Ministerial Decree 26.6.84 “Classification of reaction to fire and approval of materials for fire prevention”.

The standard prescribes that sandwich panels, which are combustible because of the organic materials they are made of (polyurethane foam, binding agents for mineral wool fibers, paints) are classified according to the results obtained in the following tests:

CSE RF2, small flame combustibility test, performed by applying a small flame to the edge of the test specimen in such a way as to simulate the initiation phase of a fire (Fig. 9.24);

CSE RF3, radiant panel test (Fig. 9.25), performed by subjecting the test specimen to a small pilot flame and a radiant panel, so as to simulate a fully developed fire acting on the same specimen.

The rating method assigns the panel a number that can vary from 0 to 5 (0,1,2,3,4,5); the lowest numbers indicate the best reaction-to-fire characteristics, and takes into consideration the final application of the panel under test: sandwich panels can be considered either as pure structural elements or as isolated structural elements.

A sandwich panel, considered as a pure structural element, is classified by testing only the outer metallic surface; since the paint present on this surface is essentially an organic material, it can generate a small flame spread on the surface of the panel itself, making the panel Class 1.

If the sandwich panel is considered as a structural element with insulating properties, the standard provides for a double number classification, with the first number referring to the sandwich panel as a whole (for sandwich panels in general it is equal to zero after self-certification), and the second referring only to the inner insulating layer.

As an example, the Italian classification is given for the following types of panel (for which the classification based on EN 13501-1 was given earlier in this chapter):

Mineral wool panels (also acoustic) Class 1 / Class 0-0
Firemet Class 1 / Class 0-2 Panels
Class 1 / Class 0-2 PIR panels
PUR-B2 Class 1 / Class 0-2 Panels
PUR-B3 Class 1 / Class 0-4 Panels

Fire resistance

The determination of the fire resistance foresees the exposure of the element to a fire with characteristics regulated by standard, and the resistance time determined in this way is an important property of the construction elements, since it can represent the time interval sufficient for people to escape from a fire.

Sandwich panels are tested by European countries according to the following harmonized standards:

EN 13501-2, Fire classification of construction products and building elements – Part 2: Classification using data derived from fire resistance tests;
EN 1363-1, Fire resistance test – Part 1: General requirements;
EN 1364-1, Fire resistance test for non load-bearing elements – Part 1: Walls;
EN 1364-2, Fire resistance test for non load-bearing elements – Part 2: Ceilings;
EN 1365-2, Fire resistance test for load-bearing elements – Part 2: Flooring and coverings.

According to these standards, the sandwich panels are mounted on the structure of a kiln, so as to obtain an assembly whose dimensions are typically 3 x 3 meters for walls, and 4 meters long by 3 meters wide in the case of roofs and ceilings. Two different types of ovens are used, a horizontal one for roofs and ceilings, and a vertical one for wall panels (Fig. 9.26).

Once the mounting of the sample panels on the furnace structure has been completed, a series of thermocouples are placed on the outer metal surfaces of the same panels to determine the increase in maximum and average temperatures during the test (Fig. 9.27).

Figure 9.28 shows the assembly scheme of the wall sandwich panels for the fire resistance test; in the figure the squares indicate the thermocouple application points for the maximum temperature, while the circles indicate the thermocouple application points for the average temperature. In addition, the curvature of the panel system is measured at regular time intervals in correspondence of the points marked with the black dots.

Finally, the fire resistance test is carried out by increasing over time the temperature of the cavity behind the panel system according to a standardized curve.

The final result of the test consists of a classification of the element by three symbols:

R for bearing capacity (only in case of coverings/roofs);
E for integrity;
I for insulation,

which express the three different fire resistance requirements, each of them linked to the corresponding time of detected resistance (minutes); these time intervals can assume one of the following values: 15, 20, 30, 45, 60, 90,

The symbols are used as follows:

Load-bearing elements (coverings/surfacing):
REI (time) minimum time during which all criteria are satisfied;
RE (time) minimum time during which two criteria, bearing capacity and integrity, are satisfied;
R (time) minimum time during which the bearing capacity criterion is satisfied.
Non-bearing elements (walls/ceilings):
EI (time) minimum time during which two criteria of integrity and isolation are satisfied;
E (time) minimum time during which the completeness criterion is satisfied.

Thus, a sandwich panel with a bearing capacity of 155 minutes, an integrity of 80 minutes, and a thermal insulation of 42 minutes, is classified as REI 30, RE 60, or R 120 (times are rounded down to the nearest values provided by the standard); similarly, a building element with a bearing capacity of 70 minutes, and an integrity of 35 minutes is classified as R 60, or RE 30.

It should be recognized that a sandwich panel exposed to fire rapidly loses its flexural strength (Fig. 9.29). In any case, if the metal surfaces of the panel are solidly connected to the structure, satisfactory periods of fire resistance can be achieved.

With regard to insulation characteristics, fire resistances in excess of 120 minutes are possible in the case of mineral wool, whereas, because of the rapid deterioration of the insulating properties, most panels with a foam insulation layer can achieve only short periods of fire resistance, typically up to 15 minutes for a polyurethane foam panel.

As an example, the following PanelSadnwich.ORG panels are rated for fire performance and fire resistance:

Roca Roof 80 mm wool
Fire resistance : REI 60
Reaction-to-fire : A2 S1 D0
80 mm refrigerator
Fire resistance : REI 30
Reaction-to-fire : B S1 D0