Preparation and Properties of Flame-Retardant Polyurethane Pressure Sensitive Adhesive and Its Application

: Using 10-(2,5-dihydroxyphenyl)-10-hydrogen-9-oxo-10-phosphine-10-oxide (DOPO-H Q), N,N-diethyl-bis(hydroxyethyl) aminomethylene phosphate diethyl (FRC-6), and (6-oxo- 6H-dibenzo[c,e][1,2]oxphosphine-6-yl) hydroxylmethyl-thiophene (DOPO-SF) as reactive ﬂame retardants, the ﬂame-retardant polyurethane pressure sensitive adhesive (FRPU-PSA) were prepared. The fourier transform infrared (FTIR), thermogravimetric analysis (TG), limiting oxygen index (LOI), vertical combustion (UL 94), 180 ◦ peeling, and inclined ball rolling were used to characterize and investigate the properties of FRPU-PSA. It was found that the LOI of PU/50mol%DOPO-HQ, PU/50mol%FRC-6, and PU/20wt%DOPO-SF were 30.7%, 29.3%, and 25.0%, respectively, the peel strength of PU/50mol%DOPO-HQ and PU/50mol%FRC-6 were 3.88N/25 mm and 3.42N/25 mm, respectively. FRPU-PSA not only had good bond strength, but also had good flame retardant performance.


Introduction
Pressure-sensitive adhesive (PSA) is a material that has adhesive properties after a short period of contact with small pressure. This adhesive does not require solvent evaporation or chemical reaction to achieve adhesion, Therefore the PSA is convenient and safe to use, and it is widely used in various fields of life in masking and protective tapes, packaging tapes, double-sided tapes, and labels [1]. The PSA is mainly divided into acrylic series, silicone series and rubber series, etc. Polyacrylic PSA has the advantages of excellent low temperature resistance, high temperature resistance, and no harmful gas volatilization, and is currently the most widely used PSA [2] Kajtna et al. synthesized a solvent-free acrylic acid UV-crosslinked PSA using acrylic acid monomers (2-ethylhexyl acrylate, acrylic acid and t-butyl acrylate), azobisisobutyronitrile, n-dodecylmercaptan and the unsaturated UV photoinitiator 4-acryloyloxybezophenone as raw materials. Their study showed that when acrylic acid was used as the copolymer monomer, the peel strength of the PSA decreased when the reduction of polymer molecular weight decreased, and also its peel increased. In contrast, the addition of tert-butyl acrylate monomer causes cohesive damage to the adhesive coating [3]. Jamaluddin et al. synthesized a UV-curable PSA using methyl methacrylate (MMA), trimethylolpropane triacrylate (TMPTA) and photoinitiator 1-hydroxycyclohexyl phenyl ketone. Their studying results showed that the strength of the PSA decreases with increasing MMA content, while the retention force shows an opposite trend. In contrast, the retention force of the PSA showed a significant increase with increasing TMPTA content and gel grade fraction under the effect of TMPTA cross-linking. The results showed that the peel strength of PSA decreases to zero when the oligomer content is greater than 40 wt%, while the retention force has a significant increase [4]. The unique properties of silicone such as high flexibility, low surface tension, and low intermolecular interactions give silicone PSA excellent electrical successfully improved the flflame retardancy [25][26][27]. However, the total amount of flame retardants still needs to be reduced, and the report on the combination of MH and organic boron IFRs was not many. Ai et al. synthesized an organic/inorganic synergistic flame retardant using CP-6B and MH and applied it to epoxy resin (EP). The CP-6B/MH synergistic flame retardant showed better flame and heat resistance compared with CP-6B andMH working separately. The ultimate oxygen index of EP with 3 wt% CP-6B and 0.5 wt% MH reached 31.9% and the vertical combustion rating reached V-0. It indicates that C-6B/MH has a good synergistic effect [28]. Ai et al. synthesized two more organic IFRs, one of which is the DOPO derivative: 6,6-((sulfonylbis(4,1-phenylene)) bis(azanediyl))bis(thiophen-2ylmethylene))bis(6H-dibenzo[c,e][1,2]oxaphosphinine 6-oxide(DOPO-N) and its synergistic flame retardant with inorganic boron flame retardant zinc borate (ZB) in PE. PE exhibited high thermal stability and flame retardant properties when 20 wt% ZB and 10 wt% DOPO-N were added. Its ultimate oxygen index reached 24.6% and its vertical combustion rating reached V-0 [29]. The other is an organic flame retardant containing phosphorus and nitrogen: melamine phenyl hypophosphite (MPHP). And it was applied to PP with silicon dioxide (SiO 2 ) in synergistic flame retardant. The PP added with 27 wt% MPHP and 3 wt% SiO 2 showed better flame retardant performance, and its ultimate oxygen index increased from 18.9% to 28.7% of pure PE, and the vertical combustion rating reached V-0. They further demonstrated that the addition of a small amount of silica to PP/MPHP not only reduced the heat dissipation rate, but also improved the mechanical properties of PP. Liu et al., [30]. prepared microencapsulated Mic-MBP by coating red phosphorus (RP) with melamine borate and applied it to PE in synergistic flame retardant with ZB. The flame retardant performance of the encapsulated Mic-MBP was significantly better than that of RP, and the ultimate oxygen index of PE reached 25.2% and the vertical combustion rating reached V-0. after the addition of 20 wt% Mic-MBP/10 wt% ZB [31]. On this basis, Lu et al. prepared a bilayer shell microcapsule (Mic-DP) using MBPand cross-linked β-cyclodextrin as the shell material to cover red phosphorus (RP) and applied it to polyamide 6 (PA6). The flame retardant properties of PA6 were greatly improved after the addition of 13 wt% Mic-DP. Its ultimate oxygen index reached 27.8% and its vertical combustion rating reached V-0. Their results further showed that Mic-DP has reduced hygroscopicity and better oxidation resistance compared to RP [32]. Recently, the adhesive and coat prepared by PU put forward the requirements of flame-retardant performance, especially PU pressure-sensitive adhesive (PU-PSA). However, PU is flammable, therefore it is necessary to prepare flame retardant polyurethane pressure-sensitive adhesive (FRPU-PSA) [33][34][35][36].

Characterization
The Fourier transform infrared (FTIR) spectra was recorded using a Nicolet FTIR spectrometer (Madison, WI, USA). Thermogravimetric analysis (TG) was perfor using a Netzsch 209 F3 thermal analyzer (Selb, Germany) at a heating rate of 10 °C/ and the nitrogen flow rate was 40 mL/min. The limiting oxygen index (LOI) was meas using a Fire Testing Technology instrument (FTT, East Grinstead, UK), accordin ASTM D2863. The dimensions of each sample were 80 × 10 × 4 mm. According to ASTM 3801 guidelines, vertical burning (UL 94) tests were performed with a FTT U instrument by using samples with dimensions of 125 × 12.7 × 3.2 mm. Depending o self-extinguishing time and dripping, burning grades were classified as either V-0, V-2, or no rating (NR). The initial adhesion test was according to GB/T4852-2002. The peel test was conducted with a BT1-FR010TH.A50 universal material testing mac (zwick, Germany) by using samples with dimensions of 125 × 50 × 1.5 mm and peel a descent rate of 300 mm/min, which according to GB/T2792-1998.

Preparation of I-FRPU-PSA
Polyetherdiol was weighed and added into a reactor to remove water by vac distillation. Afterwards, TDI was added into the reactor when the temperature was ra to 60 °C and then mixed and stirred under nitrogen atmosphere. PU prepolymer (P was obtained after 3h, and then TMP and flame-retardant monomer (DOPO-HQ or 6 or DOPO-SF) was added into PU-P. I-FRPU-PSA was obtained after 30 min. And 30 min of vacuum defoaming and was poured into a preheated standard dumbbell p tetrafluoroethylene mold and step temperature rise at 60 °C, 70 °C and 80 °C for 1 h After cooling to room temperature, I-FRPU-PSA samples were obtained. Table 1 sh the feeding ratio of I-FRPU-PSA samples. The molecular structure of DOPO-HQ and 6 contains dihydroxyl groups, which are involved in the reaction of preparing flam tardant PU (FRPU), The molecular structure of DOPO-SF contains only one hydr group, which is partially involved in the reaction of preparing FRPU (Table 2).

Characterization
The Fourier transform infrared (FTIR) spectra was recorded using a Nicolet 6700 FTIR spectrometer (Madison, WI, USA). Thermogravimetric analysis (TG) was performed using a Netzsch 209 F3 thermal analyzer (Selb, Germany) at a heating rate of 10 • C/min, and the nitrogen flow rate was 40 mL/min. The limiting oxygen index (LOI) was measured using a Fire Testing Technology instrument (FTT, East Grinstead, UK), according to ASTM D2863. The dimensions of each sample were 80 × 10 × 4 mm. According to the ASTM 3801 guidelines, vertical burning (UL 94) tests were performed with a FTT UL 94 instrument by using samples with dimensions of 125 × 12.7 × 3.2 mm. Depending on the selfextinguishing time and dripping, burning grades were classified as either V-0, V-1, V-2, or no rating (NR). The initial adhesion test was according to GB/T4852-2002. The 180 • peel test was conducted with a BT1-FR010TH.A50 universal material testing machine (zwick, Germany) by using samples with dimensions of 125 × 50 × 1.5 mm and peeled at a descent rate of 300 mm/min, which according to GB/T2792-1998.

Preparation of I-FRPU-PSA
Polyetherdiol was weighed and added into a reactor to remove water by vacuum distillation. Afterwards, TDI was added into the reactor when the temperature was raised to 60 • C and then mixed and stirred under nitrogen atmosphere. PU prepolymer (PU-P) was obtained after 3h, and then TMP and flame-retardant monomer (DOPO-HQ or FRC-6 or DOPO-SF) was added into PU-P. I-FRPU-PSA was obtained after 30 min. And then 30 min of vacuum defoaming and was poured into a preheated standard dumbbell polytetrafluoroethylene mold and step temperature rise at 60 • C, 70 • C and 80 • C for 1 h each. After cooling to room temperature, I-FRPU-PSA samples were obtained. Table 1 shows the feeding ratio of I-FRPU-PSA samples. The molecular structure of DOPO-HQ and FRC-6 contains dihydroxyl groups, which are involved in the reaction of preparing flame retardant PU (FRPU), The molecular structure of DOPO-SF contains only one hydroxyl group, which is partially involved in the reaction of preparing FRPU (Table 2).

Preparation of I-FRPU-PSA Tapes
A 50 µm PET film was stretched and tiled in a vacuum coater, and then I-FRPU-PSA was evenly coated with a 50 µm diameter wire rod coater. The thickness of I-FRPU-PSA was 50 µm. After coating, it was heated at 100-120 • C for 5-10 min. After 24 h of curing in a closed and clean space, the PET sheet containing I-FRPU-PSA film was cut into 25 mm wide strips.

FTIR Spectra
The FTIR spectra of PU-P after chain extension was showed in Figure 1. The absorption peak of -NCO was very weak at about 2280 cm −1 , which indicated that -NCO reaction was almost complete after chain extension. The peak at 3425 cm −1 was caused by the overlap of stretching vibration absorption peak of -OH (remaining in excess chain extender) and C-N (in carbamate group). Absorption peaks at 2920 cm −1 and 2880 cm −1 were -CH 3 Figure 2 showed the thermogravimetric (TG) and derivative thermogravimetric (DTG) curves of PU, PU/DH, PU/F, PU/DH/F and PU/DS, and their thermal decomposition data are summarized in Table 3. The temperature at 5% thermal mass loss (Tonset) of PU/DH, PU/F, PU/DH/F, and PU/DS was lower than that of PU. This was because DOPO-  Figure 2 showed the thermogravimetric (TG) and derivative thermogravimetric (DTG) curves of PU, PU/DH, PU/F, PU/DH/F and PU/DS, and their thermal decomposition data are summarized in Table 3. The temperature at 5% thermal mass loss (T onset ) of PU/DH, PU/F, PU/DH/F, and PU/DS was lower than that of PU. This was because DOPO-HQ (DH), FRC-6 (F) and DOPO-SF (DS) contain phosphorus. When PU/DH, PU/F, PU/DH/F and PU/DS were burned, the C-P bond broke first and then pyrolyzed to produce acid substances such as polyphosphate and metaphosphoric acid, which catalyzed the dehydration of PU to carbon. Figure 1. The FTIR spectra of PU-P after chain extension. Figure 2 showed the thermogravimetric (TG) and derivative thermogravim (DTG) curves of PU, PU/DH, PU/F, PU/DH/F and PU/DS, and their thermal decomp tion data are summarized in Table 3. The temperature at 5% thermal mass loss (Tons PU/DH, PU/F, PU/DH/F, and PU/DS was lower than that of PU. This was because DO HQ (DH), FRC-6 (F) and DOPO-SF (DS) contain phosphorus. When PU/DH, P PU/DH/F and PU/DS were burned, the C-P bond broke first and then pyrolyzed to duce acid substances such as polyphosphate and metaphosphoric acid, which cataly the dehydration of PU to carbon.    Based on the DTG curve, PU/DH, PU/F, PU/DH/F and PU/DS showed two weight loss stages corresponding to two different DTG peaks. The first stage of thermal weight loss of PU/FRC-6 was at 200-320 • C, corresponding to a DTG peak (Tmax1). In this stage, FRC-6 was heated to produce ammonia and water, which generated acidic substances, and catalyzed the early decomposition of PU. The second stage was at 320-410 • C, corresponding to a strong DTG peak (Tmax2), which was the further degradation of PU. The first stage of thermal weight loss of PU/DOPO-HQ was at 220-350 • C, corresponding to a DTG peak (Tmax1). In this stage, DOPO-HQ was pyrolyzed to produce acidic substances, which catalyzed PU decomposition in advance. The second stage was at 330-450 • C, corresponding to a strong DTG peak (Tmax2), which was the further degradation of benzene ring and PU. This indicated that the higher the DOPO-HQ content was, the higher was the Tmax1 of PU/DH. This was because the benzene and phenanthrene rings in DOPO-HQ made the sample to have higher thermal and chemical stabilities than those of FRC-6. At 320-450 • C, the thermal weight loss rate of PU was the largest.

Thermal Performance Analysis
In comparison, the thermal weight loss rates of PU/DOPO-HQ and PU/FRC-6 were reduced, and the higher the FRC-6 content was, the smaller was the thermal weight loss rate. Compared with DOPO-HQ, FRC-6 contained elements of P and N, which could not only generate acidic substances to promote the dehydration of the substrate into carbon but also generate noncombustible gases to dilute the oxygen concentration in the combustion environment, thus inhibiting the combustion.
Compared with PU/DOPO-HQ and PU/FRC-6, PU/DOPO-SF had three weight loss stages. The results showed that the thermal weight loss at 150-210 • C may be due to the dehydration and pyrolysis of the DOPO-SF. When DOPO-SF was added, the maximum thermal weight loss rate of PU/DOPO-SF decreased obviously, that is to say, the DOPO-SF containing samples are thermally unstable, which indicated that DOPO-SF had a certain flame-retardant effect on PU, but it was not as good as DOPO-HQ and FRC-6.

LOI and Vertical Combustion
When R (n (NCO):n (OH)) was 2.5, Polyetherdiol and TDI were used as raw materials to prepare PU-P, and DOPO-HQ and FRC-6 were introduced to extend the chain of PU-P. The influence of mole fraction (TMP/DOPO-HQ/FRC-6) on the flame retardancy of PU was studied under the premise of unchanged hydroxyl content of the chain extender, and the results are shown in Table 4. Table 5 shows the influence of DOPO-SF content on flame retardancy of PU. When 10 mol%, 30 mol%, and 50 mol% of DOPO-HQ were added, the LOI of PU/DOPO-HQ were 23.0%, 26.2%, and 30.7%, respectively, which were 25%, 42%, and 67% higher than those of PU, and the vertical combustion test reached V-2, V-1, and V-1 grades, which indicated that with the increase of DOPO-HQ in the chain extender, the flame retardancy of PU was enhanced. When 10 mol%, 30 mol%, and 50 mol% of FRC-6 were added, the LOI of PU/FRC-6 were 23.0%, 26.6%, and 29.3%, respectively, which were 25%, 45%, and 59% higher than those of PU. The vertical combustion test was V-2, V-1, and V-1, which indicated that the flame retardancy of PU was enhanced with the increase of FRC-6 in the chain extender. When the molar ratios of DOPO-HQ/FRC-6 were 4:1, 1:1, and 1:4, the LOI of PU/DOPO-HQ/FRC-6 were 28.8%, 28.7%, and 28.5%, respectively, and when DOPO-HQ was further added, the LOI of PU/DOPO-HQ/FRC-6 were at around 28.6%. Even so, the LOI of PU/DOPO-HQ/FRC-6 was lower than that of PU/DOPO-HQ and PU/FRC-6, which indicated that DOPO-HQ and FRC-6 had no synergistic flame-retardant effect. This was because DOPO-HQ contained P element, which would release free radicals such as PO· and PO2· during combustion, and PO· and PO2· could quench H·and HO·from the pyrolysis of PU. Thus, the combustion chain reaction was interrupted.
In addition, due to its low bond energy, the C-P bond in DOPO-HQ broke first during combustion and then pyrolyzed to generate acid substances such as polyphosphoric acid and metaphosphoric acid, which catalyzed the dehydration of PU into carbon and formed a flame-retardant carbon layer on the surface of the substrate. The carbon layer further prevented the combustion heat and combustible gas from diffusing into the interior of the substrate, acting as a condensed-phase flame-retardant. Compared with small molecular phosphonates or acyclic organic phosphates, DOPO-HQ had higher thermal and chemical stabilities due to the high rigidity of the benzene and phenanthrene rings in its molecular structure, and the P-containing group was formed by the side joining the ring O=P-O bond. This also delayed the catalytic decomposition of PU by the acidic substances produced by DOPO-HQ pyrolysis to some extent. In the combustion process, FRC-6 contained P and N elements generated acid substances, promoted the dehydration of the substrate into carbon, formed carbon layer for heat insulation and oxygen isolation, and played the role of condensed-phase flame-retardant. Moreover, phosphorus-free radicals could be generated to inhibit the combustion chain transfer, and the noncombustible gases were formed to dilute the oxygen concentration in the combustion environment, which played as a P-N gas-phase synergistic flame-retardant.
It could be seen from Table 5 that the flame retardancy of PU/DOPO-SF was higher than that of PU. With the increase of DOPO-SF content, the LOI of PU/DOPO-SF increased at first and then decreased and reached the maximum when 20 wt% DOPO-SF was added. When DOPO-SF content increased from 5% to 20%, LOI values were 22.4% and 25.0%, which were 22% and 36% higher than those of PU. PU/50mol%FRC-6 carbon residue was at 1238 cm and 1138 cm and PU/20wt%DOPO-SF carbon residue was at 1250 cm −1 and 1140 cm −1 . The bending and sile vibration absorption peaks of P-O was at 1028 cm −1 and 490 cm −1 in the FTIR of PU/50mol%DOPO-HQ, 1011 cm −1 and 493 cm −1 in the PU/50mol%FRC-6 carbon resi and 950 cm −1 and 493 cm −1 in the PU/20wt%DOPO-SF carbon residue.  Table 6 and Figure 4 show the adhesive properties of I-FRPU-PSA tapes and c mercial PSA tapes. From Table 6, it can be seen that PU/50mol%DOPO-HQ PU/50mol%FRC-6 tapes had similar adhesive properties with those of PU tape. It wa dicated that the reactive flame-retardant did not reduce the adhesion of PU-PSA. C pared with that of PU tape, the peel strength of PU/50mol%DOPO-HQ tape (3.88 N mm) was slightly higher, and the initial viscosity was also slightly higher. This was cause DOPO-HQ contained aromatic rings, which enhanced the cohesion of PSA to a tain extent. In addition, when TMP was replaced by 50% DOPO-HQ, the chain flexib were improved, which could enhance the wettability with the substrate. Compared w that of PU tape, the initial viscosity of PU/50mol%FRC-6 tape increased from 7 steel to 10 steel ball, and the peeling strength was slightly lower, which was 3.42N/25mm. C pared with that of PU/50mol%DOPO-HQ, the initial viscosity of PU/50mol%FRC-6 also higher, but the peeling strength was lower. This was because FRC-6 had better f bility than that of DOPO-HQ. In addition, when FRC-6 replaced TMP, the crosslink point were fewer and the initial viscosity was higher.  Table 6 and Figure 4 show the adhesive properties of I-FRPU-PSA tapes and commercial PSA tapes. From Table 6, it can be seen that PU/50mol%DOPO-HQ and PU/50mol%FRC-6 tapes had similar adhesive properties with those of PU tape. It was indicated that the reactive flame-retardant did not reduce the adhesion of PU-PSA. Compared with that of PU tape, the peel strength of PU/50mol%DOPO-HQ tape (3.88 N/25 mm) was slightly higher, and the initial viscosity was also slightly higher. This was because DOPO-HQ contained aromatic rings, which enhanced the cohesion of PSA to a certain extent. In addition, when TMP was replaced by 50% DOPO-HQ, the chain flexibility were improved, which could enhance the wettability with the substrate. Compared with that of PU tape, the initial viscosity of PU/50mol%FRC-6 tape increased from 7 steel ball to 10 steel ball, and the peeling strength was slightly lower, which was 3.42 N/25 mm. Compared with that of PU/50mol%DOPO-HQ, the initial viscosity of PU/50mol%FRC-6 was also higher, but the peeling strength was lower. This was because FRC-6 had better flexibility than that of DOPO-HQ. In addition, when FRC-6 replaced TMP, the crosslinking point were fewer and the initial viscosity was higher.   Compared with those of PU tape, the initial adhesion and peel strength of I-PSA tape prepared by PU/20wt%DOPO-SF decreased. It can be seen from Table 6 t initial adhesion range of I-FRPU-PSA tape prepared in this study was 0-10 steel ba the peel strength range was 0.05 N/25 mm-3.98 N/25 mm. Under the same test cond four kinds of commonly used commercial tapes, such as Scotch tape, electrical tap tapeand post-it note, were tested. The values of initial adhesion (4-11 steel balls) an strength (0.10N/25mm-4.15N/25mm) (yellow part in Figure 4) were obtained. The no residual adhesive in the peeling process. By comparison, the I-FRPU-PSA coul the requirements of initial adhesion and peel strength of commercial PSA product

Conclusions
DOPO-HQ, FRC-6and DOPO-SF were used as reactive flame retardants to pre FRPU-PSA. The LOI values of PU/50mol%DOPO-HQ, PU/50mol%FRC-6, and PU/ OPO-SF were 30.7%, 29.3%, and 25.0%, respectively, which passed UL 94 V-1 grad peel strength of PU/50mol%DOPO-HQ and PU/50mol% FRC-6 were 3.88 N/25 m 3.42 N/25 mm, respectively, and the initial adhesion was also good. Compared wi Compared with those of PU tape, the initial adhesion and peel strength of I-FRPU-PSA tape prepared by PU/20wt%DOPO-SF decreased. It can be seen from Table 6 that the initial adhesion range of I-FRPU-PSA tape prepared in this study was 0-10 steel ball, and the peel strength range was 0.05 N/25 mm-3.98 N/25 mm. Under the same test conditions, four kinds of commonly used commercial tapes, such as Scotch tape, electrical tape, duct tapeand post-it note, were tested. The values of initial adhesion (4-11 steel balls) and peel strength (0.10 N/25 mm-4.15 N/25 mm) (yellow part in Figure 4) were obtained. There was no residual adhesive in the peeling process. By comparison, the I-FRPU-PSA could meet the requirements of initial adhesion and peel strength of commercial PSA products.

Conclusions
DOPO-HQ, FRC-6 and DOPO-SF were used as reactive flame retardants to prepare I-FRPU-PSA. The LOI values of PU/50mol%DOPO-HQ, PU/50mol%FRC-6, and PU/20wt% OPO-SF were 30.7%, 29.3%, and 25.0%, respectively, which passed UL 94 V-1 grade. The peel strength of PU/50mol%DOPO-HQ and PU/50mol% FRC-6 were 3.88 N/25 mm and 3.42 N/25 mm, respectively, and the initial adhesion was also good. Compared with PSA products such as commercial transparent tape, the prepared FRPU-PSA tape could meet the adhesive strength of different tapes and achieve practical application.