Confirmation of crosslinking
FTIR was performed on PEI/PSP-coated papers to confirm crosslinking. Glutaraldehyde crosslinks the amine groups in the film, sourced from PEI, by forming a Schiff base (Yang et al. 2011), which is evidenced by the disappearance of the primary amine peak. Fig. 2 shows that the primary amine stretch (N-H) (3500 cm−1) is present in the PEI/PSP film, but after crosslinking this peak is no longer detected. There is also a broad hydroxyl stretch ranging from 3500-3200 cm−1 in the PEI/PSP film that is not present in the crosslinked film. This stretch can be attributed to significant water uptake of the film, which results in a noticeable tackiness of the PEI/PSP coating. After GA treatment, the coating is no longer tacky, resulting in the disappearance of the hydroxyl peak for the crosslinked film. Along with the reduction in tackiness, crosslinking can be observed through an obvious color change following exposure to GA, which can be seen in Fig. 3. In Fig. 3a, the original color of the paper is brown, which is maintained after application of the PEI/PSP film, as shown in Fig. 3b. Following GA crosslinking, the coating changes from colorless to opaque (Fig. 3c), which suggests a chemical change has taken place.
Surface morphology
SEM images reveal that uncoated cellulosic paper is extremely porous, as shown in Fig. 3d. Following the application of the PEI/PSP coacervate, the pores of the paper are filled, yielding the smooth, conformal surface shown in Fig. 3e. The deposition of the coacervate onto the cellulosic paper results in a 64% weight gain, which is likely due to the high surface area of the paper and particulates of polymeric material on the surface of the coating, as shown in Fig. 3e. After glutaraldehyde crosslinking, the PEI/PSP particulates have been washed away, leaving a uniformly crosslinked film, while simultaneously reducing the weight gain to 35%. Fig. 3f also reveals a rougher surface texture compared to the non-crosslinked film, with small craters appearing throughout the film.
Gas barrier properties
Based on the smooth surface due to pore-filling, and previous coacervate studies (Chiang et al. 2021; Haile et al. 2017; Li et al. 2021), it was hypothesized that the PEI/PSP coating on cellulosic paper could have excellent gas barrier properties. The oxygen transmission rate (OTR) was measured for uncoated, coated, and crosslinked paper and it was found that the deposition of the PEI/PSP coacervate, which sufficiently clogs the pores in the substrate, yields a 90% reduction in the OTR compared to the uncoated paper. After crosslinking, the film only results in a 30% reduction in OTR. This worsened barrier for the crosslinked film is likely due to a combination of surface morphology and weight gain (i.e., thickness and coverage). As previously discussed, SEM images of the crosslinked film show microscopic craters throughout the film, which act as weak points for oxygen and other molecules to more easily penetrate the film. Additionally, the weight gain of the crosslinked film is half that of the PEI/PSP coated film, which suggests poorer surface coverage. With its relatively smooth texture, the uncrosslinked PEI/PSP coacervate film has an optimal surface morphology suitable for significant reduction in OTR.
Flame retardant behavior
Vertical flame testing was used to evaluate the flame retardant behavior of the uncoated, coated, and crosslinked paper by subjecting each sample to a methane flame for 12 s. As shown in Fig. 4a, the uncoated paper is completely consumed by the flame, leaving virtually no residue in the exposed area. With the addition of the PEI/PSP coacervate coating, the flame still travels up the entirety of the sample, shown in Fig. 4b, but the resulting charred area has a much higher residue compared to the uncoated paper. While all of the necessary flame retardant components are present in the PEI/PSP coacervate coating, the high amount of polymeric material, which is flammable, could be hindering the flame retardant performance. After crosslinking the PEI/PSP coated sample with GA, the paper achieves self-extinguishing behavior with nearly half the weight gain of the coacervate coated sample. As shown in Fig. 4c and Table 1, the ignition of the crosslinked paper results in an average char length of only 3.4 in. While the char of the PEI/PSP coated film, as shown in Fig. 4d, is completely black. The charred area of the crosslinked sample appears to have several different colored sections surrounding the blackened char. These colorful areas are likely due to a variance in the coating’s heat exposure. The black char of the crosslinked paper appears to be very layered compared to the coated residue without crosslinking, as shown in Fig. 4e. SEM images of the secondary discolored area (Fig. 4f) reveal that the coating that was exposed to the tip of the flame is still intact, with intumescent bubbling beginning to appear in an attempt to extinguish the flame. As a result of the self-extinguishing behavior, the area directly above the char had minimal exposure to the flame, which results in only a slight discoloration of the paper substrate.
Table 1
Vertical flame test results for uncoated, coated, and crosslinked paper.
Sample | Weight Gain (%) | Afterburn (s) | Char Length (in) |
Uncoated | - | 57.0 ± 3.6 | Entire sample |
PEI/PSP | 63.9 ± 3.8 | 25.8 ± 15.6 | Entire sample |
PEI/PSP crosslinked | 35.3 ± 3.6 | 0 ± 0 | 3.4 ± 0.5 |
Microcombustion calorimetry measures the oxygen consumption rate of a given material exposed to a heat source and can be used to assess flame retardant efficacy (Lyon et al. 2013). Fig. 5 shows MCC results for uncoated, coated, and crosslinked paper. The uncoated cellulosic paper exhibits one distinct peak, with an average peak heat release rate (pkHRR) of 152 W/g, as shown in Table 2. The coacervate-coated paper and the crosslinked paper also display one distinct peak with a significant decrease in the pkHRR compared to the uncoated substrate. The similarity between the PEI/PSP and PEI/PSP crosslinked HRR, despite vastly different behaviors in the vertical flame test, is likely due to the crosslinked intumescent system primarily working in the condensed phase rather than the gas phase. As previously stated, MCC measures the oxygen consumption of the system, which can limit the detection of other flame retardant mechanisms (Leistner et al. 2015; Lyon and Walters 2004). As shown in Fig. 4, charring is a crucial component of the self-extinguishing crosslinked system.
Table 2
MCC and DSC data for coated, crosslinked, and uncoated paper.
Sample | MCC | DSC |
pKHRR (W g−1) | THR (kJ g−1) | Energy Balance (J g−1) |
Uncoated | 152 ± 3 | 8.5 ± 0.1 | +0.21 |
PEI/PSP | 73 ± 4 | 6.1 ± 0.2 | -0.65 |
PEI/PSP crosslinked | 85 ± 8 | 6.9 ± 0.1 | -12.85 |
Differential scanning calorimetry (-20 °C-500 °C) was performed to determine the role of the condensed phase energetic effects of the coated and crosslinked paper samples during pyrolysis. DSC reveals that the crosslinked film exhibits stronger endothermic behavior than the PEI/PSP coated paper. As shown in Table 2, the thermal degradation of uncoated paper appears as an exothermic peak (0.21 J/g). When the PEI/PSP coacervate was applied to the substrate, the coated paper degrades slightly endothermically (-0.65 J/g), resulting in an overall cooling effect during pyrolysis. After crosslinking the PEI/PSP film, additional endothermic reactions occur upon heating, which may be attributed to initiation of combustion (-12.85 J/g) (Gliko-Kabir, et al. 1999). This significant cooling effect reaffirms that the crosslinked coating acts in the condensed phase through system cooling and charring to self-extinguish the flame.
Conditioning of the paper
The PEI/PSP crosslinked film is able to withstand various conditioning tests, including water soaking and manual conditioning (i.e., paper crumpling). Following a 30-minute water soak, the integrity of the PEI/PSP crosslinked film is maintained, which can be confirmed by the minimal weight loss after water exposure (-3.4 wt%) and unaltered flame performance (average char length of 3.8 in.), as shown in Table 3 and Fig. 6a. To further investigate the durability of the crosslinked coating, the paper was manually conditioned by crumpling the paper into a ball to mimic the possible manipulation that the coating may be subjected to in practical applications. After the paper was conditioned, the coating did have some small cracks, but as shown in Fig. 6b and Table 3, the crosslinked film maintained the ~35% weight gain and was still able to self-extinguish with a char length of 3.8 in. The same conditioning tests were performed on uncrosslinked paper. The uncrosslinked paper had a 15% weight loss after water soaking and an unaltered weight following crumpling. Both types of conditioned samples burned the length of the paper in a vertical flame test.
Table 3
Vertical flame test results for crosslinked coating on paper after conditioning tests.
Sample | Conditioning | Weight Loss (%) | Afterburn (s) | Char Length (in) |
PEI/PSP crosslinked | Water soak | 3.40 ± 0.66 | 0 ± 0 | 3.8 ± 0.4 |
PEI/PSP crosslinked | Crumple | 0.23 ± 0.21 | 0 ± 0 | 3.8 ± 0.6 |