Abstract
In this study, three differently composed polyurethane (PU) composites are developed and then compared with the pure PU foam in terms of combustion resistance and mechanical properties using the compression test, the drop-weight impact test, the horizontal burning test, and the sound absorption test. Flame-retardant fabric (FRF)-PU is composed of PU foam that is enclosed with two cover sheets of FRFs. FR-PU10 is composed of PU foam that contains 10 wt% of flame-retardant agents. FRF-PU10 is composed of PU foam containing 10 wt% of flame-retardant agents and enclosed by two covers sheets of FRFs. Based on the test results, both FR-PU10 and FRF-PU10 are structurally stabilized and have good combustion resistance. The sample using FRFs as cover sheet had the same combustion resistance property and better compression resistance as the sample using flame-retardant agent. The carbonized layer extinguishes the alighted samples and stops the fire spread right after they are out of the source of fire, suggesting the FRF-PU10 is flame retardant. In addition, FRF-PU10 is easier to process and healthier because of the low use of flame retardants. Therefore, in this study, we proposed PU foam composites have good mechanical and flame-retardant properties and are a suitable candidate for productions of vehicles, plants, construction, and staple merchandises.
Introduction
Diverse composites have been rapidly developed because they have greater strengths than that of metals, a light weight, and ease of processing. Therefore, composites have been used in the fields of industry, transportation, and staple merchandises. Due to having a light weight, composites that are used in vehicles remarkably decrease the weight of the load as well as the consumption of energy, which makes electric energy a compatible power supply. In addition to an economical demand of energy, using composites also helps decrease the exhaust emissions during the combustion of gasoline. As a result, vehicles are made of a raising ratio of composites to replace the metallic space frames and external panels as well as obtain a lightweight [1–3]. The majority of fillers used in buildings and vehicles are glass wool that can effectively improve vibration absorption, heat insulation, and noise reduction [4–7] of the interiors and decrease the transmission of an externally applied force, thereby attaining the safety and comfort. Glass wool, commonly used in construction and vehicles, has a fibrous manner, a low cost, good heat insulation, noise reduction, and vibration absorption [4,5,8,9]. It feels urticate when the human skin is in contact with glass wool, and glass wool also has a negative influence over the respiratory tract and lungs [10–14]. Therefore, there is a raising number of studies on substitutes for glass wool.
One common substitute for glass wool is PU foam. PU foam has a lighter weight and a more convenient usage than glass wool, yet it is also less mechanically strong and does not have as many functions. As of late, there have been much research and development about PU foam. PU foam can be converted into different forms based on different compositions and formulae [15–18], and thus has a broad range of applications as in industry, packaging materials, construction, and heat/sound insulation [19,20]. On the other hand, during combustion, the chemical composition and structure of PU foam would release a considerable amount of heat energy and poisonous gas, such as hydrogen cyanide and carbon monoxide [21]. The disadvantages restrict the application of PU foams. The thermal degradation mechanism of PU foam is as follows. When PU foam receives a fire source, the heat energy serves as energy for covalent bonds of molecular chains of PU foam, allowing them for complex rotation and vibration. The molecular chains then break and resolve into radical fragments and small molecular segments. The radicals and small molecular segments probably undergo either recombination or splitting and then the eventual stages of evaporation, diffusion, and carbonization [22–25].
Adding a flame-retardant (FR) agent is the principal method to improve the FR property of the rigid PU foam [26]. In addition to an FR agent, inorganic powders, such as nano-clay [27], are also used to improve the FR property of PU foam, providing PU foam with safer and more pervasive applications in transportation, plants, buildings, and staple merchandise. The FR agent is physically dispersed in the PU foam structure without creating chemical reactions. However, adding a greater amount of the FR agent may result in poor miscibility of the PU mixture that causes bad influences. For example, the resulting PU foam may have lower mechanical properties [22]. Therefore, this study proposes using PU foam as the matrices and fibers as reinforcement in order to produce novel protective PU foam composites that have greater combustion resistance, mechanical strengths, cushioning property, and noise reduction.
Experimental
Materials
PU foam solvent (Zhong Xing Chemical, Taiwan) is composed of a polyol foaming agent and an Isocyanate (MDI) hardener agent. An FR agent, MSDS FR-047, is purchased from Kuang Lung Shing Corporation, Taiwan. Flame-retardant polyester fiber (hereafter referred as FR fiber) is purchased from Far Eastern New Century Co., Ltd, Taiwan. The flame-retardant fabric (FRF) which made by FR fiber has a LOI value of 32 [28].
Preparation of pure PU foam and PU foam composites
Three FR PU foam composites are proposed, including FR-PU, FRF-PU, and FRF-PU10. First, a polyol foaming agent and an MDI harden agent (5:5) are mixed at 600 rpm, and the mixture undergoes a foam process in a sealed mold. The resulting rigid PU foam that has a thickness of 20 mm and a foam density of 60 kg/m3 serves as the control group. Next, 0, 5, 10, 15, or 20 wt% of the FR agent is added to the PU mixture (i.e., the foaming agent and then the harden agent). The mixtures are poured into the mold for foaming, forming FR-PU5, FR-PU10, FR-PU15, and FR-PU20. Moreover, an FRF is placed on the bottom of the mold, after which the PU mixture is poured, and the other FRF is placed to cover the mixture (as shown in Figure 1). The mold is sealed for foaming, forming FRF-PU. Finally, an FRF is placed in the bottom of the mold, after which the PU mixture containing 10 wt% FR agents is poured, and the other FRF is placed to cover the mixture. The mold is sealed for foaming, forming FRF-PU10. Table 1 indicates the denotation and composition of samples. Samples are evaluated using the mechanical property tests and the horizontal burning test.

Preparation of pure PU foam and PU foam composites.
Denotation and specification of the samples.
Tests
Compression test
As specified in ASTM D1621-10 (the standard test method for compressive properties of rigid cellular plastics), the PU foam composites are cut into 50 mm × 50 mm× 20 mm pieces. The testing speed is 2 mm/min, and eight samples for each specification are used. The samples are compressed into 25% thickness of the original.
Drop-weight impact test
Drop-weight impact testing is conducted according to ASTM D4168-95 (2015). The impactor has a weight of 8.5 kg, and the impact load is 9000 N. Six samples of 100 mm × 100 mm for each specification are used for this test. The impactor is released from a height of 4 cm above the sample and vertically hit the surface. The residual load is used to characterize the cushioning property of PU foam composites.
Sound absorption coefficient test
As specified in ASTM E1050-12, a two-microphone impedance tube (Automotive Research & Testing Center, Taiwan) is used to measure the sound absorption coefficient of PU foam composites at a frequency between 125 Hz and 4000 Hz. The circular samples have a diameter of 38 mm. Three samples for each specification are used for this test.
Horizontal burning test
Samples of 150 mm × 50 mm are tested for the horizontal burning test as specified in ASTM D4986. The samples have contact with a flame for 60 s in order to observe their after-flame time, melt drop, and the length of the carbonized part (as shown in Figure 2).

Schematic diagram and test image of the horizontal burning test.
Results and discussion
Effects of compositions on the mechanical properties of PU foam composites
Figure 3 shows that the mechanical properties of FR PU foam composites as related to the content of an FR agent of 0 (i.e., the control group), 5, 10, 15, and 20 wt%. Through the sample photograph shown in Figure 5, it can be found that the addition of the FR has no effect on the foamed cell size. It means that the addition of the FR does not have much influence on the production of carbon dioxide during the foaming reaction.

(a) The compression load (N) and (b) residual force (N) of the flame-retardant PU composites.
Figure 3(a) shows the compression load of the compression test, and Figure 3(b) shows the residual force of the drop-weight impact test. Based on the test results, the content of the FR agent has a greater influence on the compression load than on the residual force of the PU foam composites, which is ascribed to the fact that the FR agent is physically dispersed within the porous PU structure. The FR hinders the formation of the molecular network structure during the PU curing reaction.
Dispersing 5 wt% of the FR agent provides the PU foam with some elasticity, which slightly improves the compression load. Conversely, using FR agent that is more than 10 wt% causes a low miscibility of the PU mixture, which has a negative influence on the structural strength of the PU foam composites and the resistance against a compression force. As a result, the content of the FR agent is inversely proportional to the compression load of the PU foam composites. Specifically, using 20 wt% FR agent causes the lowest compression load.
The drop-weight (9000 N) impact test (Figure 3(b)) also shows a similar trend to that of the results of compression test. The rigidness of the PU foam decreases as a result of increasing FR agent, and the structure thus fails to support, transmit and disperse an impact load. Namely, the residual force increases when PU foam is composed of a greater amount of FR agent, indicating a low impact force resistance.
Effects of compositions on the combustion resistance of PU foam composites
Figure 4 shows that the status of PU foam composites after combustion. For FR-PU20, the drastic decrease in structural strength leads to a presence of deformation, and this group is thus excluded from the experiment. Figure 4(a) shows the combusted samples of control group that does not contain an FR agent. The rigid PU foam does not exhibit dripping or a significant deformation. Nevertheless, the flame spreads along the bottom of the control group, suggesting that the control group is not FR. Figure 4(b) to (d) shows that for FR-PU series that use 5, 10, or 15 wt% FR agents, the contact end with the flame is covered by a carbonized layer. In addition, the greater the content of FR agent, the smaller the area of sample can be alighted.

Combustion resistance of PU foam composites of (a) the control group, (b) FR-PU5, (c) FR-PU10, and (d) FR-PU15. The front and back sides of combusted samples are respectively shown in odd and even rows.
According to the figure, we can observe the internal combustion of the sample. The test results and Figure 5 show there is only the surface of the sample was burned by the flame. And the flame was spread on the surface of the sample. The flame immediately extinguishes when sample is out of the fire source, indicating that FR-PU series are highly FR. As a result, the addition of an FR agent affects the viscosity and curing of the PU mixtures and the corresponding mechanical properties of the products. The optimal parameter is examined to be FR-PU10 which is used for the following discussions.

Combustion resistance of PU foam composites of (a) the control group, (b) FR-PU5, (c) FR-PU10, (d) FR-PU15, (e) FR-PU20, and (f) FRF-PU. The cross-section of combusted samples.
Effects of compositions on compression load of PU foam composites
Figure 6 shows the influences of composition on the compression load of the PU composites, including pure PU foam (i.e., control group), FR-PU-10, and FRF-PU, and FRF-PU10. The test results suggest that using an FR agent has a negative influence on the compression load of FR-PU-10 and FRF-PU10. The addition of the FR agent affects the viscosity of the PU mixture, foaming reaction, pore size, formation and curing property of the foam.

Compression load of PU foam composites as related to different compositions.
The FR agent is physically dispersed within the porous PU structure and it hinders the formation of the molecular network structure during the PU curing reaction which has a negative influence on the structural strength of the PU foam composites and the resistance against a compression force. Therefore, poor formation and low rigidness are responsible for a low-compression load. On the other hand, FRF-PU is composed of two cover sheets of FRFs. The reticular fabrics are saturated with the PU mixture and become highly dense and rigid surfaces, which efficiently improve the elasticity and toughness of FRF-PU. As a result, FRF-PU and FRF-PU10 have the optimal compression load.
Effects of compositions on residual force of PU foam composites
Figure 7 shows the effects of compositions on the residual force of the PU foam composites, including pure PU foam (i.e., control group), FR-PU-10, and FRF-PU, and FRF-PU10. Based on the test results, FRF-PU and FRF-PU10 exhibit the lowest residual impact force. Both FRF-PU and FRF-PU10 are composed of two cover sheets of FRFs. The reticular fabrics are then saturated with PU mixture, which the blowing agent is impregnated by pressure in the foaming reaction and impregnated into the fabric, forming highly dense and rigid surfaces. As a result, FRF-PU and FRF-PU10 have optimal cushioning effect because they have good structural support to distribute the impact force of 9000 N and the impact energy.

Residual impact force of PU foam composites as related to different compositions.
Effects of compositions on sound absorption coefficient of PU foam composites
Figure 8 shows the effects of compositions on the sound absorption coefficient of the PU foam composites, including pure PU foam (control group), FR-PU-10, and FRF-PU, and FRF-PU10. The test results show that either adding the FR agent or using two cover sheets of FRFs does not have a significant influence on the sound absorption coefficient of the PU foam composites. Adding the FR agent affects the viscosity, foaming reaction, pore size, and formation of the PU foam. Nevertheless, it also decreases the rigidness of the PU foam, allowing the acoustic energy to be debilitated via the contraction and vibration of the PU foam. On the other hand, two cover sheets of FRFs have a high areal density and high hardness due to the saturation of PU mixture. The solid surfaces contribute to the reflection of sound waves, thereby decreasing the sound absorption coefficient. Similarly, FRF-PU10 is composed of two cover sheets and a PU mixture with 10 wt% FR agents. The cover sheets are strengthened to have a high areal density and high rigidity, whereas an FR agent in the PU mixture has an adverse effect on the rigidity of the cover sheets.

Sound absorption coefficient of PU foam composites as related to the compositions.
Effects of compositions on combustion resistance of PU foam composites
Figure 9 shows the effects of compositions on the horizontal burning results of the PU composites, including pure PU foam (control group), FR-PU-10, and FRF-PU, and FRF-PU10. The cross-section of combusted samples is shown in Figure 5. Figure 9(a) shows that during the combustion of the pure PU foams, melt dripping and apparent deformation is absent, but carbonization cover layer is present. Moreover, the fire burns along the bottom of the pure PU foam, indicating a limited combustion resistance. Figure 9(b) shows that when in contact with fire, FR-PU10 that is PU foam made of PU mixture and 10 wt% FR agent has a carbonized layer which prevents the fire from spreading. Moreover, when FR-PU10 is out of the source of fire, the alighted sample immediately extinguishes, suggesting that FR-PU10 is FR. Similarly, Figure 9(c) shows that FRF-PU also generates the carbonized layer and the alighted sample extinguishes when the source of the fire is removed. There is no FR added to the FRF-PU, but it has the same good flame resistance. One of the reasons is because the surface-composited FR non-woven fabric has good FR properties, and another reason is that the non-woven fabric and the PU foam composite form a high-density interface layer. Therefore, it needs to take longer times and higher temperatures to reach the ignition point when burning. Beside of this, the sandwich composite structure of FRF-PU can achieve all-purpose FR protection effect. Figure 9(d) shows that FRF-PU10 also has good combustion resistance, demonstrated by the occurring of a carbonized layer and absence of spread fire in its bottom. To sum up, PU foam that is enclosed with FRFs as cover sheets (i.e., FRF-PU) is good at retarding the fire spread. Therefore, FRF-PU without 10 wt% FR agent has a comparable combustion resistance to that of FRF-PU10.

Combustion resistance of PU foam composites of (a) the control group, (b) FR-P10, (c) FRF-PU, and (d) FRF-PU10 as related to different compositions. The front sides are shown in the odd rows, while the back sides are shown in the even rows.
Conclusion
This study successfully develops PU foam composites which have greater combustion resistance and mechanical properties. For FR-PU series, including FR-PU5, FR-PU10, FR-PU15, and FR-PU20, using excessive FR agent has a negative influence on the foaming reaction and mechanical properties, which determines the optimal content of FR agent is 10 wt%. Furthermore, pure PU foam, and PU foam composites (i.e., FR-PU, FRF-PU, and FRF-PU10) are then compared in terms of combustion resistance and mechanical properties. The test results show that FRF-PU and FRF-PU10 are both mechanically stronger and have greater combustion resistance to prevent fire spread. In particular, FRF-PU has 10–20% greater compression strength, 50% lower residual impact force, and sound absorption performance that is as good as that of other PU foam composites.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: The authors would especially like to thank Ministry of Science and Technology of Taiwan for financially supporting this research under Contract MOST 105–2221-E-035–031 and MOST 106–2632-E-035–001.
