TFMB was purchased from ITCHEM (Cheong-ju, South Korea); a, aʹ-bis (4-aminophenyl)-1,4-diisopropylbenzene (APIPB), 2,2-bis[4-(4-aminophenoxy)phenyl]hexafluoropropane (APPFP), 3,4ʹ-diaminodiphenylether (3,4-ODA), 6FPD, pAPS, 2,2ʹ,5,5ʹ-tetrachlorobenzidine (TCB), 4,4ʹ-diaminodiphenylmethane (MDA), 4,4ʹ-methylenebis (cyclohexylamine) (MBCHA), 4,4ʹ-diaminodiphenylether (ODA), TPC, and IPC were purchased from TCI (Tokyo, Japan); 4,4ʹ-methylenebis (2-chloroaniline) (MOCA), DMAc (anhydrous grade), and PO were purchased from Sigma-Aldrich (St. Louis, MO). All reagents were used as received.

Aramid polymers were synthesized in a 250-mL three-necked flask equipped with a reflux condenser, three-way stopcock, and magnetic bar. DMAc and diamine (10 mmol) were added to the flask and stirred until all amines were dissolved. The flask was then immersed in an ice bath and cooled to 0°C. IPC (3 mmol) was added to the flask and stirred for 10 min, and then TPC (7 mmol) was added and stirred for 2 h. Afterward, PO was added to remove the HCl generated as a byproduct, and varnish was obtained.

After casting the varnish on a glass substrate, it was dried in a convection oven at 40°C for 30 min and 200°C for 1 h to produce polymer films with thicknesses of several tens of micrometers.

The viscosities of the varnishes were measured at 25°C at a speed of 150 rpm using a DV2T viscometer (Brookfield) equipped with a cone plate spindle CPA-52Z. FTIR absorption spectroscopy was performed in ATR mode using an FTIR-4100 spectrophotometer (Jasco, Japan). ¹H-NMR analysis was performed using an AVANCE III 500 NMR spectrometer (Bruker) with DMSO-d₆ as the solvent. DSC was performed using a Q2000 DSC (TA Instruments) under a nitrogen atmosphere in a temperature range of 25°C–350°C at a heating/cooling rate of 10°C/min. TGA was performed using a Discovery SDT 650 TGA (TA Instruments) under a nitrogen atmosphere in a temperature range of 25°C–900°C at a heating rate of 10°C/min. Elastic moduli and tensile strengths were measured using UTM equipment (Yeonjin S-tech, Korea) at a tensile rate of 1 mm/s. Visible-light transmittance and yellow index were measured using a V-650 UV–Vis spectrophotometer (Jasco, Japan).

Diamine monomers serve as nucleophiles in a nucleophilic acyl substitution reaction with diacid chloride to produce polyamides along with an acid byproduct, HCl, through the low-temperature solution condensation reaction [7]. However, diamine monomers may also act as bases, so they react with the acid byproduct to form an amine salt insoluble in organic solvents, resulting in loss of reactivity and gelation. Therefore, in the low-temperature solution condensation reaction, the basicity of diamine is a crucial factor for efficient polymerization. For TFMB, a high-viscosity transparent varnish was obtained without salt formation, [5,6] attributable to TFMB having a relatively low basicity due to the inductive effect of the strong electron-withdrawing CF₃ group. In this study, for comparison with TFMB, various diamine monomers (TCB, pAPS, TFMB, MOCA, 6FPD, 3,4-ODA, APPFP, ODA, APIPB, MDA, MBCHA in Table 1) were examined for the low-temperature solution condensation reaction. Among them, TCB, pAPS, TFMB, MOCA, 6FPD, 3,4-ODA, and APPFP have electron-withdrawing groups (EWG), such as CF₃, sulfone, and atomic halogen. As expected, they have relatively low basicity (pKa < 4.80) compared with other monomers because of the inductive effect of strongly withdrawing electrons from the functional amine group. Exceptionally, the relatively low basicity of 3,4-ODA despite the absence of EWG probably results from its asymmetric structure. Meanwhile, ODA, APIPB, MDA, and MBCHA with no EWG have higher basicity (pKa > 5.20). In particular, MBCHA (an aliphatic diamine) has the highest basicity (pKa 10.97) due to the lone pair localization in the amine group. The pKa values of these diamine monomers are listed in Table 1 [8]. As is known from a comparison between PPTA (a para-type aramid) and Nomex (a meta-type aramid), the diacid chloride linkage structure may also affect the reactivity and solubility during polymerization [9,10]. In this study, this effect was neglected to examine only the basicity effect of diamine monomers. Thus, diacid chloride was used with a mixture of TPC and IPC at a fixed molar ratio of 70:30 (Figure 1).

For the seven diamine monomers with lower basicity, an amine salt was not formed when diacid chloride was added to the solution, and the viscosity increased while the reaction system remained transparent (Case 1 in Figure 1). Notably, the rate of viscosity increase differed for the various monomers. The TFMB solution came close to gelation within approximately 10 min after adding the diacid chloride, showing the fastest increase in viscosity, attributable to the high rigidity of the growing polymer chain. Meanwhile, the other six monomers showed a gradual increase in viscosity. Among them, the rate of increase in viscosity for TCB was noticeably slow, so the viscosity increased to reach equilibrium over several hours, attributable to its very low nucleophilicity. The four atomic chlorine groups likely caused a strong inductive effect, significantly lowering the nucleophilicity of the monomer. All seven monomers could produce varnishes and films using the same method as that reported for TFMB in the literature (Case 1 in Figure 1). After adding diacid chloride to the diamine monomer solutions, when the reaction system reached an appropriate viscosity, PO was added to remove HCl, and the reaction solution was used as a varnish without purification to obtain films with a superb appearance and several tens of micrometer thickness. Consequently, diamine monomers with relatively low basicity could easily and conveniently provide aramid films through the one-pot reaction with diacid chloride in the same vessel. Meanwhile, when the other aromatic diamine monomers, ODA and MDA, which have relatively higher basicity, were reacted in the same manner, amine salts were formed immediately after the addition of diacid chloride, making the reaction system opaque and insoluble. As expected, when these reaction solutions were poured into a mixed solvent of water and alcohol, almost no polymer was obtained. The cycloaliphatic diamine MBCHA also showed a similar result. This is because these highly basic monomers formed salts through an acid–base reaction with the acid byproduct during polymerization and lost their reactivity (Case 2 in Figure 1). In addition, the formation of an insoluble gel probably occurred because the terminal amine groups of the growing polymer chains reacted with HCl to become positively charged. Consequently, strong ion–dipole interactions occurred with the amide groups in other polymer chains, ultimately resulting in salt bridge gelation (Figure 2). Once the amine salt was formed, it did not dissolve again, even if more solvent was added or the temperature was increased. Notably, APIPB provided a fairly high-viscosity reaction solution without amine salt formation, despite having almost the same basicity as MDA. Therefore, APIPB belongs to the Case 1 monomers in this study. The reason is thought to be as follows. Unlike MDA with one methylene linkage (two Csp²–Csp³ single bonds), APIPB has two methylene linkages (four Csp²–Csp³ single bonds), which could have increased the main chain flexibility. As a result, the amine ends of the growing polymer chain were probably given great mobility, and the condensation reaction proceeded quickly enough to surpass the salt formation reaction with HCl. This result suggests that not only the diamine monomer basicity but also the growing polymer chain flexibility can greatly affect reactivity. According to the literature, although ODA has a fairly high basicity, when reacted with IPC alone, a solution with a fairly high viscosity is obtained without salt formation, from which a polymer film can be easily obtained, but it appears quite yellow [11]. However, because the polymer chain flexibility and coloring do not meet the purpose of this study, aramid film manufacturing using IPC alone was not examined.

Because the obtained polymers were all insoluble in THF, gel permeation chromatography analysis was impossible using this solvent as the eluent. Instead, the molecular weight and chain rigidity were inferred by measuring the viscosity of the polymer solutions in DMAc. The varnishes used in film production showed a high viscosity in the range of 160–2520 cP, even at a low concentration of 12 wt% (Table 1). Particularly, polymers from TFMB, MOCA, and APPFP had an extremely high viscosity over 1000 cP. This indicates that the polymers have a high molecular weight and/or chain rigidity. Because of the high viscosity, casting on a glass substrate was very easy, and thick films with a thickness of approximately 50–60 μm were produced through a typical convection drying process. The films had a smooth surface and a superb appearance and did not tear or break when pulled or bent by hand. Figure 3 shows images of the film manufactured from TFMB.

FTIR analysis of the polymers was performed (Figure 4a). For the fabricated polymers, absorption peaks were observed at ~1,650 and 1,575 cm⁻¹, attributable to the amide C=O and C–N stretching, respectively. A broad absorption peak due to amide N–H stretching was observed within 3,500–3,100 cm⁻¹, suggesting that the amide groups strongly interacted with each other through hydrogen bonding. In addition, absorption peaks due to specific functional groups (e.g., CF₃ and sulfone) present in each monomer were observed. ¹H-NMR analysis was performed using DMSO-d₆ as a solvent (Figure 4b). All polymers showed peaks due to aromatic protons around 7–9 ppm, and the peak due to amide protons shifted further downfield to appear at around 10–11 ppm. These results confirm that the aramid polymers were successfully synthesized as desired.

The thermal decomposition stability of the polymers was evaluated using TGA (TGA curves in Figure 5a, Table 1). The thermal decomposition temperature (T_d5) at a weight loss of 5 wt% under a nitrogen atmosphere was greater than 400°C for all polymers, indicating that the polymers have very high thermal decomposition resistance. This thermal stability is attributable to the coplanar resonance structure comprising aromatic and amide groups in main chain [12,13]. DSC analysis was also performed to evaluate the thermodynamic stability of the polymers (Figure 5b). The polymers did not show any endothermic peak due to enthalpy change over a wide temperature range from room temperature to 350°C, indicating that they do not melt even when the temperature is increased to near the decomposition temperature. Moreover, the TFMB-, pAPS-, and TCB-based polymers did not show any significant phase change due to glass transition. This is attributable the polymers having a rigid rod-like chain structure and strong hydrogen bonds between the chains, thereby requiring a very low entropy change and a high enthalpy change for the phase transition. For the APIPB-, APPFP-, 3,4-ODA-, 6FPD-, and MOCA-based polymers, which are thought to be relatively flexible, peaks due to glass transition were clearly observed around 121°C, 207°C, 214°C, 257°C, and 176°C, respectively.

The stress–strain curves were analyzed using UTM (Figure 6, Table 1). All polymer films exhibited an elastic modulus of more than 4.1 GPa and a tensile strength of more than 52.6 MPa. Among them, the TFMB-based film had the highest elastic modulus and tensile strength of 8.2 GPa and 119 MPa, respectively, indicating a very rigid chain. The pAPS- and 6FPD-based films exhibited relatively high elastic moduli of 5.26 and 5.07 GPa, respectively, and high tensile strengths of 52.6 and 83.0 MPa, respectively. Meanwhile, the APPFP-based film had a lower elastic modulus and tensile strength of 4.1 GPa and 81 MPa, respectively, and the highest value of elongation at break (38%), indicating a less rigid chain. These values are comparable to those of the colorless and transparent polyimide and polyamideimide films reported so far [14,15,16]. For example, fluorinated polyimide obtained from TFMB and 4,4′-(hexafluoroisopropylidene)diphthalic anhydride (6FDA) has an elastic modulus of 3.1 GPa and a tensile strength of 120 MPa. The polyamideimide film synthesized by reacting 0.4-eq 6FDA and subsequently 0.6-eq TPC with TFMB has an elastic modulus of 5.1 GPa and a tensile strength of 110 MPa.

In fluorinated polyimides, a strong EWG, such as CF₃, effectively suppresses intramolecular charge transfer to prevent yellowing [17,18]. This was true for all aramid films obtained in this study. The TFMB-based film had a high transmittance of 95.1% at 420 nm and a low yellow index of 2.4 (Figure 7, Table 1). The 6FPD-based film is optically superior with a slightly higher transmittance (96.2%) and lower yellow index (1.91). The pAPS-based film is also quite colorless and transparent; the transmittance and yellow index were 91.3% and 4.3, respectively. Although the pAPS-based film is somewhat inferior to the TFMB- and 6FPD-based films in terms of optical performance, these three films are colorless and transparent enough to be used as a transparent substrate or cover window for electronic devices. Aramid films obtained from diamines other than TFMB, 6FPD, and pAPS appeared slightly yellow and had relatively low light transmittance in the visible region. For the MOCA-based film, the optical transmittance was 71.9% and the yellow index was 10.9; this chlorinated aromatic polymer is vulnerable to attack by radicals or anions because of the relatively low dissociation energy of the Csp²–Cl bond, so they probably formed unexpected color species during polymerization or convection drying. Similarly, the 3,4-ODA-based film showed the lowest light transmittance (54.5%) and the highest yellow index (17.0).