Skip to main content
Top

Open Access 16-09-2024

Flame Retarded Adhesive Tapes and Their Influence on the Fire Behavior of Bonded Parts

Authors: Vitus Hupp, Bernhard Schartel, Kerstin Flothmeier, Andreas Hartwig

Published in: Fire Technology

Activate our intelligent search to find suitable subject content or patents.

search-config
loading …

Abstract

Pressure-sensitive adhesive tapes are used in automotives, railway vehicles and construction, where flame retardancy is of major importance. This is why industrial applicants often buy, and industrial tape manufacturers often produce, flame-retardant adhesive tapes, advertised for their good flammability characteristics. Yet, how flame-retardant tapes influence the fire behavior of bonded materials is a rather open question. To investigate this issue, three different substrates were bonded, using eight double-sided adhesive tapes containing two different carriers and two different flame retardants. The bonded substrates were compared to their monolithic counterparts in terms of flammability, fire behavior and fire stability. The fire behavior of adhesive tape bonded materials differed significantly from the monolithic substrates. The usage of different adhesive tapes let to different burning behavior of the bonded materials mainly due to different carrier systems. In contrast, the implementation of flame retardant into the adhesive had rather minor or no effect on the burning behavior of the bonded substrates despite their positive effect on the flammability of the free-standing tape. The carrier changed the HRR curve in the cone calorimeter and was able to both, reduce and increase fire hazards. Using the carrier with the better fire performance can lower the fire growth rate by 20%, the peak of heat release rate by 27%, and the maximum average rate of heat emission by 30% in cone calorimeter tests. Overall, the fire behavior of bonded materials is a complex interaction between substrate, adhesive, and carrier, and depends on the fire scenario the materials are exposed to.
Notes

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

1 Introduction

The use of adhesive joints increased strongly in automobiles, railway vehicles and construction over recent decades due to advances in lightweight technology, and electric mobility and the use of bonded materials such as cross laminated timber in buildings. In these applications, the fire behavior of the bonds and adhesives is of major importance and needs to be understood and optimized to prevent danger to life and property. It is well known that blends and combinations of materials such as laminate structures have a fire behavior different from the sum of their single components [14]. This leads to the assumption that bonds, as a combination of several materials including substrate, adhesive and optional interlayers, also behave differently from the individual materials that they consist of. In the literature, many articles on flame-retardant adhesives [58] have already shown that adhesives and incorporated flame retardants can change the fire behavior of bonded materials with a high adhesive content such as plywood [9], wood particle boards [10] or flame-retardant expanded polystyrene [11]. In the glued material cross-laminated timber as well, adhesive bonds change the fire behavior drastically [12]. For laminates like structural insulated panels, adhesives between the layers can change the fire behavior, while flame retardants in the adhesives can improve the flame retardancy of the bonded material [13]. A special case of adhesives are pressure-sensitive adhesives (PSA) and PSA tapes. They are widely used in several applications due to their easy application and durability. Due to their low glass transition temperature (Tg), they are permanently sticky and adhere to a vast spectrum of surfaces. PSA consist mainly of rubber-like polymers that are intrinsically flammable with the exception of silicone-based PSA. The popular use of intrinsically flammable PSA suggests the enhancement of the fire behavior by flame retardants would be beneficial. Flame retardants can improve the burning behavior of PSA tapes [14] but there is a lack of knowledge of how these tapes act in their application in bonds. Especially in industrial applications, flame-retarded adhesive tapes are sold and advertised with UL 94 ratings, and other flame-retardant properties of the tapes are tested as free-standing material that is not bonded to any substrate. The approach to protecting an adhesive tape resembles the method of protecting thin foils to improve their UL 94 rating (V-2 or V-0) even though the end application is an entirely different one. The question of whether using flame retardants in adhesive tapes improves the burning behavior of bonded materials has not yet been answered and is of great interest to adhesive tape manufacturers and customers. In this research paper, the burning behavior of different PSA tapes and PSA-bonded materials is investigated, namely wood, polymethyl methacrylate (PMMA) and bisphenol-A polycarbonate (PC). Substrates with a wide spectrum of burning characteristics (charring or melting behavior) were chosen to examine the specific interactions between substrate and adhesive. Wood represents non-melting, charring materials; PMMA represents melting, non-charring materials, and PC represents melting, charring materials. As flame retardants, a dihydro-9-oxa-10-phosphaphenanthrene-10-oxide (DOPO)-derivate and resorcinol bis (diphenyl phosphate) (RDP) were used to protect the poly(n-butyl acrylate) polymer matrix of the PSA. DOPO(-derivates) [1518] and RDP [1922] are known for their good flame retardancy effects in the gas and condensed phases. Since recent research [23] has proven that flame retardants and carriers behave differently depending on the material combination and matrix, poly(ethylene terephthalate) (PET) and aluminum foil are used as carriers in different substrate configurations. The PSA tape bonded materials were investigated during ignition, developing fire and in a fully developed fire scenario.

2 Materials

Chemicals: 6H-dibenz[c,e][1,2]oxaphosphorin-6-propanoic acid, butyl ester, 6-oxide (DOB 11) was provided by Metadynea (Krems, Austria) and resorcinol bis(diphenyl phosphate) (RDP) by ICL Industrial Products (Tel Aviv, Israel). Disponil FES 32 and Disponil A1080 were provided by BASF (Ludwigshafen, Germany). Sodium peroxodisulphate, n-dodecyl mercaptan, acrylic acid and n-butyl acrylate were purchased from Merck (Taufkirchen, Germany).
Carriers: PET foil was provided by TESA (Hamburg, Germany) and aluminum (AL) foil in a thickness of 30 µm was purchased from VWR International GmbH (Darmstadt, Germany).
Substrates: Beech wood was purchased as planks from Kula Holz-GmbH & Co. KG, (Berlin, Germany). Extruded colorless PMMA (Plexiglas® XT) from Evonik Industries AG (Germany) and PC from Covestro AG (Germany) (Makrolon® GP) were purchased from Thyssenkrupp Plastics GmbH (Germany) in the dimensions 1000 × 2000 × 2 mm3.

3 Methods

3.1 Preparation of Adhesive Tapes and Bonded Specimen

Pressure-sensitive adhesives: The PSA was prepared by emulsion polymerization. To prepare a homogenous PSA dispersion it is necessary to prepare a pre-emulsion and an initiator solution.
Polymerization procedure: For the pre-emulsion, water (57 g), Disponil FES 32 (4.77 g), Disponil A1080 (0.94 g) and acrylic acid (0.72 g) were placed in a 500 ml beaker and stirred magnetically. Afterward, n-butyl acrylate (149.25 g), the flame retardant, and n-dodecyl mercaptan (0.15 g) were added and stirred continuously. To prepare the initiator solution, sodium peroxodisulphate (0.48 g) was dissolved in water (12.3 g) in a 50 ml beaker. Polymerization took place in a 500 ml reaction vessel in which water (82 g) and Disponil FES 32 (0.67 g) were placed and heated to 85 °C while stirring under argon atmosphere. In one shot 0.8 g initiator was added, and 4.5 g pre-emulsion were added after 15 min. Subsequently, the rest of the initiator solution was added to the pre-emulsion and this mixture was added dropwise to the reaction vessel over a period of one hour. The reaction was stirred for an additional 2 h at 85 °C. Finally, the dispersion was filtered through a 50 µm sieve. For better processability, the pH of all dispersions was adjusted to pH 8 by adding 25% ammonia solution. To achieve a good rheology for coating, Rheovis AS 1125 (1 wt.%) was added.
Coating: All double-sided PSA tapes were prepared in the following way. The dispersions were coated onto the PET or AL foil with a Zehnter automatic film applicator ZAA 2300 at a speed of 25 mm/s. The wet film thickness was adjusted to 110 µm. After 5 min drying in air at ambient conditions, the films were heated for 10 min in an oven at 110 °C. The back was coated in the same way after applying a release paper on the already coated side. Eight different double-sided adhesive tapes were prepared to obtain products with different flame retardants and carriers (Table 1).
Table 1
Specimen Names and Descriptions of the Adhesive Tapes
Name
Description
Butac_50_REF_PET
Poly (n-butyl acrylate) (PSA) coated on PET
Butac_DOB11_0.5_PET
PSA with DOB 11 and a phosphorus content of 0.5% coated on PET
Butac_DOB11_1.5_PET
PSA with DOB 11 and a phosphorus content of 1.5% coated on PET
Butac_RDP_0.5_PET
PSA with RDP and a phosphorus content of 0.5% coated on PET
Butac_50_REF_AL
Poly (n-butyl acrylate) (PSA) coated on AL
Butac_DOB11_0.5_AL
PSA with DOB 11 and a phosphorus content of 0.5% coated on AL
Butac_DOB11_1.5_AL
PSA with DOB 11 and a phosphorus content of 1.5% coated on AL
Butac_RDP_0.5_AL
PSA with RDP and a phosphorus content of 0.5% coated on AL
Adhesive bonding: To prepare bonded samples, the release liner of the adhesive tapes was removed, and the PSA tapes were placed onto one side of the substrates. Air bubbles were removed by a rubber hand-pressure roll. Subsequently, the second release liner was removed, and the other substrate was adhered to the tape. Again, the rubber roll was applied at the surface of the bonded material to remove potentially incorporated air bubbles. These sandwich-like laminates were prepared in different dimensions according to the demands of the following fire behavior investigations. The adhesive properties are not relevant and therefore not mentioned.

3.2 Flammability Tests

UL 94: The UL 94 test was performed according to the current UL 94 standard. The test was performed in an UL 94 Test chamber from Fire Testing Technology (UK).
To rate the flammability of the tapes that are not bonded to any object, the tapes were measured as a multilayer specimen to avoid shrinkage and distortion. Eight layers of the PSA tapes were stacked to obtain specimens 125 mm × 13 mm × 1 mm in size. Each adhesive tape layer was 120 µm thick.
The bonded materials (substrate/tape/substrate) were measured in the dimensions of 125 mm × 13 mm × 4.1 mm. The bonds were manufactured as a sandwich-like connection between Substrate and adhesive tape (substrate plate/tape/substrate plate) where both, substrate plates, consisted of the same material.
Oxygen index: The oxygen index was measured according to ISO 4589–2. The oxygen index of the PET carrier tapes was measured from samples that were prepared by folding the adhesive tape, resulting in a 70 mm × 7 mm × 1.5 mm specimen. A 100 mm × 80 mm piece of tape was cut from a DIN A4 sheet of double-sided tape and subsequently folded to the demanded size. The AL tapes were measured in the frame holder that is described in EN 4589–2 and normally used for thin foils or plastics that do not distort while burning. The bonded materials (substrate/tape/sample) were measured in the dimensions of 100 mm × 10 mm × 4 mm.

3.3 Burning Behavior

Cone Calorimeter: The Cone Calorimeter tests were performed in a cone calorimeter from Fire Testing Technology (UK) according to ISO 5660. The distance between the cone heater coil and sample surface was adjusted to 35 mm to leave more space for the sample to expand while still ensuring a homogenous heat flux of 50 kW m−2 over the entire irradiated area (less than 10% deviation) [24]. PC, for example, is known for its expansion in cone calorimeter tests [25], making it advisable to increase the distance over the standard 25 mm. A heat flux of 50 kW m−2 was chosen to simulate a developing fire, while simultaneously considering the heat flux required by EN 45545–2 for application in railway vehicles. The bonded wood samples were measured in an aluminum tray with four wires preventing the bending of the specimen. All samples were measured in an aluminum tray using the standard stainless steel specimen holder without the retainer frame.

3.4 Fire Stability

The fire stability was measured in a small-scale test which was developed to determine the stability of the adhesive bond against a flame, simulating a fully developed fire. Thus the necessity of the time-consuming, expensive, large-scale tests that are normally performed [26] to measure the fire stability was eliminated [27]. This test is designed to generate a ranking of the effects of different flame retardants on the fire stability of the adhesive joint. Therefore, two substrates of the same material were bonded together by the tapes and a weight of 3 kg was mounted to the lower half of the substrate as shown in Fig. 1. The substrate thickness was 4 mm for wood and 1 mm for zinc plated steel. The two substrates were chosen due to their different behavior during fire exposure. Steel as an incombustible material is expected to stay unharmed during the test whereas wood is suspected to be ignited and deform during the test. Thus, for steel, only the thermal stability of the adhesive joint is relevant for the failure. For wood in contrast, the interaction between deforming, decomposing wood and the adhesive gap plays a role. The loaded sample was fixed in a clamp in a stand and exposed to a defined burner flame. The heat flux and temperature of the flame was calibrated for a certain distance between burner and the specimen surface and was kept the same for all measurements. The calibration was performed with a heat flux meter which was installed in the recess of a calcium silicate plate with a direct exposure to the burner flame. An irradiance of 75 kW m−2, as a full developed fire heat flux [28], was chosen as the irradiance level at the specimen surface. The distance was fixed at 12.5 cm and the gas flow of the burner (propane) was kept constant at 3 L min−1. The time to failure of the adhesive bond and the temperature of the back surface of the substrate plate that is exposed to the burner flame was measured to obtain information about how the single components of the adhesive tape and the substrates influence the adhesive’s stability in a fully developed fire. The back temperature of the first layer represents the temperature impact that the adhesive is exposed to.

4 Results and Discussion

4.1 Flammability Tests

4.1.1 Adhesive Tapes With Different Carrier Systems

Table 2 shows the flammability test results that were obtained in the vertical UL 94 and OI. In UL 94, the DOB 11 flame retardant had the greatest effect, and the tapes containing DOB 11 were rated V-2, extinguished by dripping and igniting the cotton wool. Dripping is an effective mechanism to achieve a V-2 rating in the UL 94 test due to the loss of material from the pyrolyzing zone and thus a pronounced cooling effect [29, 30]. All tapes that failed to achieve a UL 94 V rating burned until the flame reached the clamp. The OI was improved slightly by every flame retardant compared to the non-protected PSA. As for the carriers, AL led to a rapid burning of the outer adhesive layer on the non-combustible metal carrier. Thus, dripping was prevented, and none of the AL tapes achieved a UL 94 vertical rating.
Table 2
Results for the Free-Standing Adhesive Tapes in Flammability Tests
Sample
UL 94
OI [vol %] ± 0.2
Butac_PET
N.R
17.6
Butac_DOB11_0.5_PET
V-2
18.5
Butac_DOB11_1.5_PET
V-2
19.7
Butac_RDP_0.5_PET
N.R
19.7
Butac_50_REF_AL
N.R
23.3
Butac_DOB11_0.5_AL
N.R
24.1
Butac_DOB11_1.5_AL
N.R
24.2
Butac_RDP_0.5_AL
N.R
23.9
N.R. No vertical rating

4.1.2 Adhesive Tape Bonded Materials

Table 3 shows the flammability test results of the bonded substrates and the monolithic (consisting of one homogenous material) materials in comparison. In the UL 94 test, the adhesive tapes had no negative impact on the burning process of the specimens. All investigated samples achieved the same rating and behaved like the monolithic substrate. In UL 94, the ignition scenario is from the bottom of the specimen, climbing fast up the sides of the sample. Thus, the inner adhesive tape layer plays a negligible role in this testing method.
Table 3
UL 94 Vertical Test Results of the Adhesive Tape Bonded Substrates as Compared to the Monolithic Material
 
Wood
PMMA
PC
Sample name
UL 94 V
OI (vol %) ± 0.3
UL 94 V
OI (vol %) ± 0.3
UL 94 V
OI (vol %) ± 0.3
Monolithic
N.R
27.7
N.R
17.7
V-2
27.1
Butac_PET
N.R
27.5
N.R
19.5
V-2
27.3
Butac_DOB11_0.5_PET
N.R
27.5
N.R
18.9
V-2
26.3
Butac_DOB11_1.5_PET
N.R
27.6
N.R
18.8
V-2
26.9
Butac_RDP_0.5_PET
N.R
29.3
N.R
20.0
V-2
27.2
Butac_50_REF_AL
N.R
27.7
N.R
18.6
V-2
28.7
Butac_DOB11_0.5_AL
N.R
28.3
N.R
18.8
V-2
28.5
Butac_DOB11_1.5_AL
N.R
28.3
N.R
18.4
V-2
28.9
Butac_RDP_0.5_AL
N.R
27.3
N.R
18.3
V-2
28.1
The OI results show that the tapes had different impacts on the flammability of the substrates. In wood, all tapes with different flame retardants had no impact beyond the standard deviation, except for the RDP flame-retarded tape. It increased the OI by about 1.5 vol % compared to the other tapes. This increase by the flame retardant active in the condensed phase is explained by improved charring, which is relevant in the OI. The high OI values are typically for untreated bonded beech wood [31] and other wood species (pine) [32] due to their charring properties. In PMMA, there is a trend toward an increased OI for the bonded materials. Comparing the OI of the tape materials and the OI of PMMA, the OI of PET and aluminum (non-flammable) are higher than that of PMMA. This leads to a small increase. In PC, the flammability of the PET-taped samples resembled the flammability of the monolithic material. When the AL carrier tapes were used, a trend toward increased OI was observed.

4.2 Cone Calorimeter Measurements

1.
Wood
 
Figure 2 shows the HRR curve of the different wood samples. Monolithic wood is compared with the tape bonded wood. The same adhesives are compared with different carriers (PET and aluminum). The burning behavior differs between these three samples. For monolithic wood, there are two local maxima of heat release rate, whereas for the bonded wood there are three local maxima at different positions. The local maxima can either be interpreted as a peak due to the following decrease in HRR. Here, the local maxima are described as PHRR. Monolithic wood has the typical shape of two PHRR described in the literature [33, 34] where the peak heights and forms depend on the wood species and material dimension. First, the surface heats up, ignites, and builds up the first PHRR. Then a char layer is built up, which causes a plateau with a HRR minimum (after ignition) due to its insulating effect [35]. After the char layer cracks, pyrolysis progresses and the wood beneath the char layer burns so that the second PHRR emerges. After the fire load is exhausted, the HRR drops to the afterglow level.
For the bonded wood, the first layer of wood ignited and subsequently the char layer built up (first peak/shoulder); then, after a small minimum at around 75 s, the char layer cracked and the second PHRR emerged. After the first layer of wood was burned, the HRR decreased, and the all-time minimum after ignition appeared. This minimum is caused by the char layer of the first wood layer, the insulation by the adhesive layer, and the char layer of the second wood layer, which prevent heat conduction within the substrate. Furthermore, the adhesive joint displayed a weak spot within the sample for mechanical impact. The wood samples deformed and shrank heterogeneously due to fiber orientation and natural in homogeneities within the material. The first layer shrank and created small gaps between the first and second substrate layers, which act as an insulation layer in the cone calorimeter measurement. The HRR reached its all-time maximum after ignition at around 250 s as soon as the adhesive gap breached and the char layer of the second wood layer cracked. The typical shoulder or separation into two PHRR of the second layer was missing, which led to the assumption that the char layer of the second wood layer already built up during burning of the first layer.
The carriers, PET, and AL, differ in their ability to insulate the second wood layer from the impact of the flame. The aluminum, as a metal foil, blocked the irradiation and the pyrolysis for a longer time than the burnable PET carrier. The non-combustible AL foil protected the surface of the second wood layer over the entire time of the measurement whereas there was no protection for the second wood layer in the PET tape-bonded specimens after the PET tape was consumed. This result already agrees with the literature, which shows that AL interlayers delay the burning of the second layer of material [36]. This led to an earlier PHRR (third peak) of the PET-carrier bonded wood, which automatically led to a higher maximum average rate of heat emission (MARHE)3, as can be seen in Table 4. The subscript numbers describe the chronology of events in the burning process. The PHRR3 of the aluminum carrier tapes were higher, but later than for the PET-carrier taped samples. The cone calorimeter measurements of the glued wood samples were similar to those of bonded multilayer arrangements such as plywood [37] or wood foam core sandwich panels [2] where the insulating effect of the adhesive layer(s) leads to a decrease in HRR and changes the curve compared to a monolithic material.
Table 4
Characteristic Values of the Cone Calorimeter Wood Measurement of Monolithic and Bonded Wood
Sample
tig (s) ± 3
FIGRA (kW m−2s−1) ± 0.3
PHRR (kW m−2) ± 50
THE (MJ m−2) ± 2
MARHE1 (kW m−2) ± 10
MARHE2
(kW m−2) ± 10
MARHE3 (kW m−2) ± 10
Monolithic
37
4.0
589
66
99
229
 
Butac_PET
35
4.9
493
70
106
206
237
Butac_DOB11_0.5_PET
35
5.1
567
69
111
223
244
Butac_DOB11_1.5_PET
29
5.2
560
70
119
221
252
Butac_RDP_0.5_PET
37
4.3
494
66
96
199
217
Butac_50_REF_AL
34
5.2
634
69
107
221
207
Butac_DOB11_0.5_AL
30
5.1
651
68
118
230
226
Butac_DOB11_1.5_AL
35
4.8
618
69
108
220
220
Butac_RDP_0.5_AL
36
5.1
637
70
106
218
218
tig Time to ignition, FIGRA fire growth rate, THE total heat evolved
Figure 3a shows a comparison of all PET carrier tapes. The same tapes that differ in UL 94 rating and OI have the same burning behavior in cone calorimeter evaluations. Figure 3b shows the AL carrier tapes where no flame retardant had a significant impact on the burning behavior. All deviations are within the range of error.
Table 4 shows the characteristic values of the different wood samples in the cone calorimeter. The time to ignition (tig) tended to lower ignition times for the bonded materials due to the reduced thermal thickness of the first layer compared to the monolithic material (the dimensions of the specimen are the same). For thermally thin materials, the ignition time reduces with the thickness of the material [28].
The PHRR for the monolithic wood is determined by the peak in the final, gradual decay phase. This peak was higher than in PET-taped samples. Aluminum samples, in contrast, had a more pronounced PHRR once the aluminum barrier was overcome. The all-time MARHE is determined by the last peak (peak 3) for most of the samples. Comparing bonded and monolithic material, the MARHE was similar for all samples, even though the determining HRR peaks were at different times and different heights. Comparing the THE, all samples released the same level of heat over the entire test period, which amounts to the same level of combustion at flameout. Discussing MARHE1, MARHE2, and MARHE3, Table 5 shows that the MARHE1, which is determined by the first peak/shoulder of HRR, was higher for the bonded samples. The first layer of material tended to ignite earlier and had a steeper slope in HRR, which leads to an earlier peak and thus a higher MARHE. MARHE3 existed only for the bonded specimen and was similar for AL- and PET-bonded materials. MARHE3 is determined by the last peak for almost all the samples. Monolithic samples had a higher PHRR than the PET samples, but the peak appeared later, so that the MARHE2 was lower than the MARHE3 of the bonded samples. AL-bonded samples had the highest peak 3, but the AL insulation shifted it to a significantly later time so that the MARHE3 is reduced, and they had the lowest MARHE3.
Table 5
Characteristic Values of the Cone Calorimeter Measurement of Monolithic and Bonded PMMA
Sample
tig (s) ± 2
FIGRA (kW m−2 s−1) ± 0.2
PHRR (kW m−2) ± 90
MARHE (kW m−2) ± 20
THE (MJ m−2) ± 2
Monolithic
38
8.1
1181
590
116
Butac_PET
39
7.9
1150
594
119
Butac_DOB11_0.5_PET
37
7.6
1109
594
121
Butac_DOB11_1.5_PET
42
7.6
1140
577
120
Butac_RDP_0.5_PET
37
8.4
1139
608
122
Butac_AL
39
8.3
870
413
120
Butac_DOB11_0.5_AL
36
9.0
846
430
118
Butac_DOB11_1.5_AL
38
8.9
854
423
117
Butac_RDP_0.5_AL
37
10.1
920
425
116
2.
PMMA
 
Figure 4 shows the comparison between monolithic PMMA and the Butac_DOB11_1.5 taped samples with PET and AL carriers. The monolithic PMMA behaved as expected with the typical shape of the HRR curve described in the literature [38, 39]. The differences are due to the carriers interrupting the heat transfer by conduction and convection within the sample. After ignition, the HRR of the taped samples tended to rise more rapidly than the HRR of the monolithic PMMA. The interrupted heat conduction led to a reduced sample thickness and thus to a faster heating up of the first layer. After 100 s, the first layer of PMMA was consumed; in the PET-taped material, the adhesive gap was eliminated by melting the adhesive and the carrier. The PET tape, with its higher melting point, delayed the burning process, which led to a reduced slope of the HRR compared to the monolithic PMMA. Monolithic PMMA and the taped sample with PET carrier tape behaved very similarly after the first layer of PMMA and the PET tape were burned. The AL-carrier taped samples behaved quite differently: After the first layer of PMMA was consumed, Butac_DOB11_1.5_AL showed a strong decrease in HRR, which was caused by the AL as a non-combustible material protecting the second layer from irradiation and preventing heat transfer by convection from the first layer. A HRR minimum was observed at 160 s. After this minimum was overcome, the second layer started to burn with a lower PHRR than the first layer due to the remaining incombustible AL layer.
Figure 5 shows the comparison between the different adhesives used to bond PMMA with a) PET carriers and b) AL carriers. As in wood, the amount of flame retardant and adhesive is too small to make a difference in burning behavior. All adhesives behave the same considering the uncertainties of the samples. The carriers, in contrast, have a remarkable effect on the burning behavior and completely transform the HRR curve. As shown in Table 5, PHRR, FIGRA and MARHE changed significantly, and the fire risk was clearly reduced when an aluminum carrier tape was used for the bond. The FIGRA was higher for the aluminum tapes due to the early peak of heat release rate. For the PET tapes and the monolithic PMMA, the PHRR took place at the end of the burning process. In the AL-taped samples, the PHRR shifted to the front and happened shortly after ignition. The PHRR of AL tapes was reduced by around 25% compared to the PET tapes, which can be explained by the heat conduction into the second sample layer by the AL tape. In the PET tape, the PHRR took place at the end of the burning process as the samples got thinner and thinner and heated up faster because no more heat was conducted away from the sample surface. The sample became thermally thinner, so that more material was volatilized at once and the PHRR was reached. In the AL sample, the first layer of PMMA was responsible for the PHRR, and the same effect happened as in the monolithic sample/PET samples. The first layer got thinner and heated up faster, creating the PHRR. The difference is that in this case, the heat was conducted into the remaining layer of PMMA, which prevented the first layer from heating up as fast and subsequently prevented it from forming the same PHRR as the monolithic sample. Due to the shift of the PHRR, also the MARHE is smaller. In all samples, the complete PMMA is consumed and only the AL layer remains which leads to the same THE values.
3.
PC
 
Figure 6 shows the comparison between monolithic PC and the taped samples. There are significant discrepancies between the HRR curves and characteristic values. After ignition at the same time, the HRR rises to its peak as it is common for PC and other charring plastics. The PHRR differs between the monolithic, PET-carrier taped, and the AL-carrier taped samples. The increase of the PHRR is due to the reduced layer thickness and the stronger separation: In monolithic PC, there is just one block of polymer that can melt and burn, and no disturbance of heat conductivity and char formation takes place. The typical HRR with a high heat release rate after ignition the PHRR shortly afterwards and the subsequent decrease in HRR with a strong afterglow is typical for PC [4042]. In the bonded materials, the first layer of polymer can deform and create gaps that hinder the heat transfer into the underlaying layer. This leads to a faster heating of the first layer and thus a higher PHRR. The influence of PET and AL tapes is significant. The separating effect of aluminum is even stronger pronounced and leads to an even higher PHRR. After the PHRR, all samples build up a char layer that is typical for PC. The char layer differs from taped to monolithic samples. The first layer is rapidly consumed, forming a char layer which afterwards protects the second layer as a barrier. The whole second layer tends to burn with a plateau-like HRR curve shape for PET. For AL-carrier tapes, a tendency toward a valley of HRR is observed due to its strong barrier effect and because it is not consumed during the burning process as the PET carrier tapes. The monolithic PC doesn’t have that fast consumption of the first layer and chars continuously over the whole burning process leading to a fading HRR.
Figure 7 shows the HRR curve of different tape bonded PC specimens. The key results (Table 6) of the cone calorimeter measurements resemble each other and for AL-tapes as well as for PET-tapes, there are only slight differences depending on which adhesive and flame retardant is used. The discrepancies are mainly due to a high variation of char formation after the PHRR which is typical for PC. Reviewing the comparison over all materials investigated, the cone calorimeter measurements show that rather the carrier than the adhesive or flame retardant in the adhesive makes the difference contemplating the fire behavior of these bonded substrates.
Table 6
Characteristic Values of the Cone Calorimeter PC Measurement of Monolithic and Bonded PC
Sample
tig (s) ± 2
FIGRA (kW m−2 s−1) ± 0.2
PHRR (kW m−2) ± 90
MARHE (kW m−2) ± 20
THE (MJ m−2) ± 2
Monolithic
79
3.5
420
207
66
Butac_PET
72
4.5
538
188
76
Butac_DOB11_0.5_PET
83
3.9
504
178
69
Butac_DOB11_1.5_PET
87
3.9
498
179
68
Butac_RDP_0.5_PET
76
4.5
475
237
76
Butac_AL
88
5.3
654
200
74
Butac_DOB11_0.5_AL
79
5.2
634
209
75
Butac_DOB11_1.5_AL
85
5.4
669
189
72
Butac_RDP_0.5_AL
86
6.2
735
244
76

4.3 Fire Stability Test

The test, developed and built by us, serves as a qualitative measurement method which can be used to investigate the influence of flame retardants or other changes in adhesive formulations on the fire stability of the adhesive joint. The fire stability test shows that different flame retardants influence the thermal stability of the adhesive.
RDP, as a more condensed phase active flame retardant, improved the fire performance of the adhesive bond with different carrier and substrate materials. Table 7 shows the time to failure and the back-surface temperature of the bonded substrate at that time. Comparing zinc-plated steel and wood, of course, zinc-plated steel had the lower time to failure, because it has higher thermal conductivity, and the adhesive temperature rises faster. The temperatures on the back surface at which the adhesive joints failed were similar. Comparing the adhesives, those with RDP as a flame retardant that is predominantly active in the condensed phase, showed the longest time to failure in beech wood (47 s PET and 51 s AL carrier) and in zinc-plated steel (23 s PET and 26 s AL carrier). It withstood the highest back-surface temperatures in beech wood (87 °C and 84 °C) and in zinc-plated steel (108 °C and 117 °C). While this fire stability test doesn’t replace large scale testing, it provides insight into certain material combinations and helps ranking the tapes. If the results can be transferred to real applications and large scale (must be investigated), the test save resources that would be otherwise spent on testing unpromising PSA-bonded materials.
Table 7
Fire Stability of Different Substrates Bonded by PSA Tapes
Adhesive tape
Wood
Zinc-plated steel
Time (s)
Temperature (°C)
Time (s)
Temperature (°C)
Butac_PET
26 ± 8
81 ± 17
20 ± 3
86 ± 24
Butac_DOB11_0.5_PET
34 ± 5
79 ± 4
18 ± 2
97 ± 16
Butac_DOB11_1.5_PET
24 ± 8
68 ± 16
14 ± 4
69 ± 9
Butac_RDP_0.5_PET
51 ± 6
84 ± 11
26 ± 3
117 ± 18
Butac_AL
27 ± 2
75 ± 20
17 ± 1
75 ± 5
Butac_DOB11_0.5_AL
16 ± 7
57 ± 17
14 ± 5
78 ± 16
Butac_DOB11_1.5_AL
25 ± 6
65 ± 15
18 ± 2
81 ± 8
Butac_RDP_0.5_AL
47 ± 10
87 ± 9
23 ± 1
108 ± 17

5 Conclusions

The influence of different flame retarded PSA tapes on the flammability, fire behavior and fire stability of bonded materials was investigated. The flame retardants DOB 11 as well as RDP showed improvement in the flammability of the free-standing adhesive tapes. Using DOB 11 in PET carrier adhesive tapes led to a V-2 rating, in contrast to the non-flame-retarded tape and the RDP tape, which failed the vertical UL 94 test. No AL carrier tape passed the UL 94 test with a rating because extinguishing via dripping was not possible due to the non-combustible AL carrier. In OI investigations, both flame retardants increased the OI slightly, from 17.6 vol % to 19.7 vol % for the PET carrier tapes, and from 23.3 vol % to 24.2 vol % for the AL tapes. These improvements were not measured in the flammability of bonded materials which showed the same OI as their corresponding monolithic materials. In the fire scenario of a developing fire, depicted by the cone calorimeter, the behavior of bonded materials was significantly different from the monolithic materials. The adhesive tapes influenced the way heat was transported within the sample. They created gaps between the individual layers and thus hindered heat transfer by conduction or affected the convective heat transport within the melt due to barrier effects. These barrier effects are dominated by the choice of carrier in the adhesive tape, which leads to individual changes in the HRR curves of all substrates. PET carrier tapes, as a thermoplastic material, can melt and lose their barrier effects as soon as the pyrolysis/melt front approaches and overcomes the first substrate layer, as was the case for bonded PMMA. AL carriers, in contrast, did not lose their barrier effect over the whole burning process and remained until the end of the test. This is why AL carriers reduce MARHE and PHRR in PMMA by 25 and 30% compared to the monolithic material and the PET-carrier taped PMMA. In PC as a charring material, the usage of an AL carrier tape created an insulating barrier that generated a gap between the layers during burning. This resulted in a PHRR after ignition 30% higher, and a FIGRA 30% higher than for the monolithic material. For wood, the HRR shape observed was entirely different between that of the monolithic and taped materials. As for PC, the adhesive joint displayed a weak spot in the sample, such that the first layer loosened and deformed due to the impact of the cone heater. This generated an additional peak in the HRR curve and led to a 20% increase in the FIGRA.
The performed cone calorimeter investigations are not only academic investigations but serve as release relevant tests for the interior of railway vehicles (EN 45545, 2020). Also, the UL 94 tests are applied in industry as release relevant test for electric applications.
Considering the burning behavior of the different bonded substrates, the influence of adhesive tapes leads to burning behavior significantly different than that of the monolithic material and yields new fire hazards. But in contrast to the current state of the art, the solution is not to protect the adhesive tape so that it can pass certain tests like UL 94. Instead, it is advisable to adapt the adhesive tape and tailor the carrier to the application and substrate it is used in.

Acknowledgements

The research was based on an IGF Project. The IGF Project (20762 N) of the Research Association DECHEMA (Deutsche Gesellschaft für Chemische Technik und Biotechnologie e. V., 60486 Frankfurt am Main, Germany) was supported by the AiF within the framework of the program “Förderung der Industriellen Gemeinschaftsforschung (IGF)” of the German Federal Ministry for Economic Affairs and Climate Action, based on a decision of the Deutschen Bundestag. I would like to thank Fernanda Romero who supported the UL 94 and OI measurements, Yin Yam Chan for the support in fire stability measurements and Michael Schneider and Detlev Rätsch for the sample manufacturing.

Declarations

Conflict of interests

The authors declare no conflict of interests.

Ethical approval

Some materials and discussion were supplied by industrial partners within the IGF project as their contribution to the pre-competitive and independent research, following the strict rules of IGF framework to ensure compliance.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creativecommons.​org/​licenses/​by/​4.​0/​.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Literature
5.
go back to reference Wang Z-H, Liu B-W, Zeng F-R, Lin X-C, Zhang J-Y, Wang X-L, Wang Y-Z, Zhao H-B. (2022) Fully recyclable multifunctional adhesive with high durability, transparency, flame retardancy, and harsh-environment resistance. Sci Adv 8(50):eadd8527. https://doi.org/10.1126/sciadv.add8527. Wang Z-H, Liu B-W, Zeng F-R, Lin X-C, Zhang J-Y, Wang X-L, Wang Y-Z, Zhao H-B. (2022) Fully recyclable multifunctional adhesive with high durability, transparency, flame retardancy, and harsh-environment resistance. Sci Adv 8(50):eadd8527. https://​doi.​org/​10.​1126/​sciadv.​add8527.
17.
go back to reference Artner J, Ciesielski M, Ahlmann M, Walter O, Doring M, Perez RM, Altstadt V, Sandler JKW, Schartel B (2007) A novel and effective synthetic approach to 9,10-dihydro-9-oxa-10-phosphaphenanthrene-10-oxide (DOPO) derivatives. Phosphorus Sulfur Silicon Relat Elem 182(9):2131–2148. https://doi.org/10.1080/10426500701407417CrossRef Artner J, Ciesielski M, Ahlmann M, Walter O, Doring M, Perez RM, Altstadt V, Sandler JKW, Schartel B (2007) A novel and effective synthetic approach to 9,10-dihydro-9-oxa-10-phosphaphenanthrene-10-oxide (DOPO) derivatives. Phosphorus Sulfur Silicon Relat Elem 182(9):2131–2148. https://​doi.​org/​10.​1080/​1042650070140741​7CrossRef
19.
go back to reference Pawlowski KH, Schartel B (2007) Flame retardancy mechanisms of triphenyl phosphate, resorcinol bis(diphenyl phosphate) and bisphenol bis(diphenyl phosphate) in polycarbonate/acrylonitrile-butadiene-styrene blends. Polym Int 56(11):1404–1414. https://doi.org/10.1002/pi.2290CrossRef Pawlowski KH, Schartel B (2007) Flame retardancy mechanisms of triphenyl phosphate, resorcinol bis(diphenyl phosphate) and bisphenol bis(diphenyl phosphate) in polycarbonate/acrylonitrile-butadiene-styrene blends. Polym Int 56(11):1404–1414. https://​doi.​org/​10.​1002/​pi.​2290CrossRef
22.
go back to reference Dukarski W, Krzyżanowski P, Gonsior M, Rykowska I (2021) Flame retardancy properties and physicochemical characteristics of polyurea-based coatings containing flame retardants based on aluminum hydroxide, resorcinol bis(diphenyl phosphate), and tris chloropropyl phosphate. Materials 14(18):5168CrossRef Dukarski W, Krzyżanowski P, Gonsior M, Rykowska I (2021) Flame retardancy properties and physicochemical characteristics of polyurea-based coatings containing flame retardants based on aluminum hydroxide, resorcinol bis(diphenyl phosphate), and tris chloropropyl phosphate. Materials 14(18):5168CrossRef
25.
go back to reference Zhang J, Koubaa A, Xing D, Godard F, Li P, Tao Y, Wang X-M, Wang H (2021) Fire retardancy, water absorption, and viscoelasticity of borated wood—polycarbonate biocomposites. Polymers 13(14):2234CrossRef Zhang J, Koubaa A, Xing D, Godard F, Li P, Tao Y, Wang X-M, Wang H (2021) Fire retardancy, water absorption, and viscoelasticity of borated wood—polycarbonate biocomposites. Polymers 13(14):2234CrossRef
27.
go back to reference Hull TR (2008) 11—Challenges in fire testing: reaction to fire tests and assessment of fire toxicity. In: Horrocks AR, Price D (eds) Advances in Fire Retardant Materials. Woodhead Publishing, pp 255–290CrossRef Hull TR (2008) 11—Challenges in fire testing: reaction to fire tests and assessment of fire toxicity. In: Horrocks AR, Price D (eds) Advances in Fire Retardant Materials. Woodhead Publishing, pp 255–290CrossRef
31.
go back to reference Borysiuk P, Jaskolowski W, Boruszewski P, Jenczyk-Tolleczko I, Jablonski M, Bylinski D (2011) Ignitability of wood impregnated with fireproof agent based on diammonium hydrogen phosphate, citric acid and sodium benzoate. Forest Wood Technol 73:181–185 Borysiuk P, Jaskolowski W, Boruszewski P, Jenczyk-Tolleczko I, Jablonski M, Bylinski D (2011) Ignitability of wood impregnated with fireproof agent based on diammonium hydrogen phosphate, citric acid and sodium benzoate. Forest Wood Technol 73:181–185
34.
go back to reference Paál M, Rychlý J, Vykydalová A, Šurina I, Lisý A, Brezová V, Nemčeková K, Labuda J (2023) Burning and thermal degradation of wood under defined conditions: a route of preparation of carbonaceous char and its characterization for potential applicability in evaluation of real fire. Fire Technol 59(5):2733–2749. https://doi.org/10.1007/s10694-023-01422-7CrossRef Paál M, Rychlý J, Vykydalová A, Šurina I, Lisý A, Brezová V, Nemčeková K, Labuda J (2023) Burning and thermal degradation of wood under defined conditions: a route of preparation of carbonaceous char and its characterization for potential applicability in evaluation of real fire. Fire Technol 59(5):2733–2749. https://​doi.​org/​10.​1007/​s10694-023-01422-7CrossRef
39.
go back to reference Babrauskas V (2002) Chpt. 3-1. Heat release rates. In: Walton D (eds) The SFPE handbook of fire protection engineering. 3rd ed. National FireProtection Association Inc., USA Babrauskas V (2002) Chpt. 3-1. Heat release rates. In: Walton D (eds) The SFPE handbook of fire protection engineering. 3rd ed. National FireProtection Association Inc., USA
Metadata
Title
Flame Retarded Adhesive Tapes and Their Influence on the Fire Behavior of Bonded Parts
Authors
Vitus Hupp
Bernhard Schartel
Kerstin Flothmeier
Andreas Hartwig
Publication date
16-09-2024
Publisher
Springer US
Published in
Fire Technology
Print ISSN: 0015-2684
Electronic ISSN: 1572-8099
DOI
https://doi.org/10.1007/s10694-024-01637-2