Introduction

Poly(arylene ether ketone)s have received considerable attention because they are among the eminent thermoplastic super-performance engineering polymers that have excellent chemical tolerance.1, 2 However, the synthetic difficulties resulting from the poor solubility of the intermediate oligomers often restricts their use. During the course of investigations to overcome these problems, the introduction of a crankshaft-shaped moiety has proven to be one of the effective approaches. The twisted accumulated aromatic-ring structures presumably contribute to smooth polymerization without the stable stacking of aligned aromatic rings in the polymer main chains. Such modifications to the chemical structure have also afforded high-molecular-weight wholly aromatic polyketones, that is, poly(arylene ketone)s that are free from ether linkages in the main chain. We have achieved the synthesis of wholly aromatic polyketones3 via the electrophilic aromatic aroylation polycondensation using the minute molecular design of acyl-acceptant molecules composed of 2,2′-dioxy-1,1′-biphenylylene,4 2,2′-dioxy-1,1′-binaphthylylene,5 and o-terphenylylene6 moieties.

During the synthetic study of wholly aromatic polyketones, we unexpectedly observed the highly susceptible and selective aroylation at the 1- and 1,8-positions of 2,7-dimethoxynaphthalene.7, 8 Using X-ray crystal structure analyses, the resulting 1-mono- and 1,8-diaroylated naphthalene homologs with various substituents have been shown to contain an accumulated organization of aromatic rings in a non-coplanar arrangement, where the aroyl groups are perpendicularly attached to the naphthalene ring and situated in opposite directions.9, 10 These non-coplanar accumulation features are distinct from those of the corresponding β-isomers.11 The characteristic congested shapes of these molecules prompted us to trigger the incorporation of the 1,8-diaroyl-2,7-dimethoxynaphthalene moiety into the poly(arylene ether ketone) backbone. When the 1,8-diaroyl-2,7-dimethoxynaphthalene unit is incorporated into the polymer molecule, it is expected to have essentially the same topology of non-coplanar accumulated aromatic-rings spatial organization to that of the crystal of the relating diaroylnaphthalene molecule. In addition, the aromatic polymers synthesized via chain elongation at the 1,8-positions of the naphthalene ring core have rarely been reported12 in contrast to the analogous aromatic polymers prepared by chain elongation at the 2,6- and 1,5-positions.13, 14, 15, 16, 17 In this article, we briefly report and discuss the behavior of the polymerization reaction of 1,8-bis(4-fluorobenzoyl)-2,7-dimethoxynaphthalene (1) during nucleophilic aromatic substitution polycondensation and the several curious properties of the resulting polymers in connection with the unique structure of the 1,8-diaroylenenaphthalene unit.

Experimental procedure

Measurements

1H NMR spectra were recorded on a JEOL JNM-AL300 spectrometer (300 MHz) and a JEOL JNM-ECA-500 spectrometer (500 MHz) (JEOL Ltd, Tokyo, Japan). The chemical shifts are expressed in p.p.m. relative to an internal standard of Me4Si (δ 0.00). 13C NMR spectra were recorded on a JEOL JNM-AL300 spectrometer (75 MHz) and a JEOL JNM-ECA-500 spectrometer (125 MHz). The chemical shifts are expressed in p.p.m. relative to an internal standard of CDCl3 (δ 77.0). IR spectra were recorded on a JASCO FT/IR-4100 spectrometer (JASCO Corporation, Tokyo, Japan). Elemental analyses were performed on a Yanaco CHN CORDER MT-5 analyzer (Yanaco New Science Inc., Kyoto, Japan).

Gel permeation chromatography measurements were performed using a Shodex K-806M column at a flow rate of 0.75 ml min−1 at 40 °C using CHCl3 as an eluent on a HITACHI L-7100 HPLC pump (Hitachi, Ltd., Tokyo, Japan). The UV absorption was monitored at a wavelength of 254 nm using a HITACHI L-7405 UV–vis HPLC spectrophotometer. The inherent viscosities (ηinh) were determined in a NMP (0.3 g dl−1) solution at 30 °C using an Ostwald viscometer. The glass transition temperatures (Tgs) were determined based on the differential scanning calorimetry (DSC) curves. The DSC curves were recorded on a Rigaku DSC-8230 differential scanning calorimeter (Rigaku Corporation, Tokyo, Japan) at a heating rate of 10 K min−1 under nitrogen against the samples that were preheated at the temperature just below the weight-loss-starting temperature followed by quick quenching. The thermal degradation temperatures (Tds) were determined based on a thermogravimetric analysis (TGA) thermogram. The TGA thermogram was recorded on a Rigaku TG-8120 thermogravimeter at a heating rate of 10 K min−1.

Synthesis of 1,8-bis(4-fluorobenzoyl)-2,7-dimethoxynaphthalene (1)

2,7-dimethoxynaphthalene (5 mmol, 0.941 g), 4-fluorobenzoic acid (11 mmol, 1.54 g) and P2O5–MsOH18 (22 ml) were added to a dried 30 ml one-necked flask. After the reaction mixture had been stirred for 1 h at 60 °C, it was poured into ice-cold water (40 ml), and the mixture was extracted using CHCl3 (30 ml × 3). The combined extracts were washed with aqueous 2 M sodium hydroxide, saturated NaCl (aq) and dried over anhydrous magnesium sulfate. The solvent was removed under reduced pressure to yield powdery product 1 (quant.).

1,8-bis(4-fluorobenzoyl)-2,7-dimethoxynaphthalene (1)

1H NMR δ (300 MHz, CDCl3): 3.69 (6H, s), 7.02 (4H, dd, J=8.6 Hz), 7.21 (2H, d, J=9.3 Hz), 7.71 (4H, dd, J=8.2, 5.5 Hz), 7.95 (2H, d, J=9.3 Hz) p.p.m.; 13C NMR δ (75 MHz, CDCl3): 56.28, 111.12, 115.06 (d, JC-F=21 Hz), 125.46, 129.75, 131.51, 131.64, 132.25, 135.18, 156.24, 165.49 (d, JC-F= 254 Hz), 195.38 p.p.m.; IR (KBr) 1663, 1613, 1596, 1567, 1512, 1504, 1270 cm−1; melting point=196 °C; C26H18O4F2 anal. calcd. for C, 72.22%; H, 4.20%. Found C, 72.05%; H, 4.20%.

Typical procedure for the nucleophilic aromatic substitution polymerization of monomer 1 using diol 2a

1,8-Bis(4-fluorobenzoyl)-2,7-dimethoxynaphthalene (1, 0.2 mmol, 86.5 mg), diol 2a (0.2 mmol, 45.6 mg), potassium carbonate (1.0 mmol, 138 mg), and freshly distilled DMAc (0.25 ml) were placed into a 10 ml one-necked flask that was equipped with a condenser. After the reaction mixture had been stirred at 150 °C for 48 h, it was poured into methanol (20 ml) to separate the pale yellow precipitate. The precipitate was collected by filtration with suction and dried in vacuo at 80 °C for 8 h to yield a pale yellow powder of poly(ether ketone) 3a.

Other reactions were performed using essentially the same procedure as described above.

Poly(ether ketone) 3a

1H NMR δ (500 MHz, CDCl3): 1.65 (6H, s), 3.67 (6H, s), 6.81 (4H, brs), 6.96 (4H, brs), 7.11–7.23 (6H, m), 7.61 (4H, brs), 7.88 (4H, brs) p.p.m.; 13C NMR δ (125 MHz, CDCl3): 31.10, 42.39, 111.29, 113.43, 116.68, 119.90, 121.64, 125.53, 128.31, 131.52, 132.02, 133.47, 146.74, 153.38, 156.16, 161.77, 195.22 p.p.m.; IR (KBr): 1668, 1609, 1592, 1497, 1237, 1269 cm−1; C41H32O6 anal. calcd. for C, 79.34%; H, 5.20%. Found C, 77.56%; H, 5.16%.

Poly(ether ketone) 3b

1H NMR δ (500 MHz, CDCl3): 3.68 (6H, s), 6.84 (4H, brs), 6.92–7.08 (8H, m), 7.17 (2H, brs), 7.62 (4H, brs), 7.89 (2H, brs) p.p.m.; 13C NMR δ (125 MHz, CDCl3): 56.59, 111.29, 116.35, 120.14, 121.55, 121.95, 125.57, 129.61, 131.57, 132.08, 133.54, 150.97, 154.00, 156.18, 161.98, 195.26 p.p.m.; IR(KBr): 1664, 1595, 1488, 1219 cm−1; C38H26O7 anal. calcd. for C, 76.76%; H, 4.41%; found C, 74.87%; H, 4.61%.

Poly(ether ketone) 3c

1H NMR δ (500 MHz, CDCl3): 3.72 (6H, s), 6.96 (4H, brs), 7.12–7.20 (6H, m), 7.71 (4H, brs), 7.79 (4H, brs) 7.93 (2H, brs) p.p.m.; 13C NMR δ (125 MHz, CDCl3): 56.54, 111.26, 113.68, 118.26, 118.85, 121.15, 125.58, 129.65, 131.61, 132.39, 133.23, 134.77, 156.29, 159.94, 163.23, 194.27, 195.53 p.p.m.; IR (KBr): 1661, 1589, 1497, 1238 cm−1; C39H26O7 anal. calcd. for C, 77.22%; H, 4.32%; found C, 75.21%; H, 4.54%.

Poly(ether ketone) 3d

1H NMR δ (500 MHz, CDCl3): 3.69 (6H, s), 6.85 (4H, brs), 7.09 (4H, brs), 7.18 (2H, brs), 7.65 (4H, brs), 7.91 (2H, brs) p.p.m.; 13C NMR δ (125 MHz, CDCl3): 56.57, 111.28, 116.54, 121.50, 121.92, 125.58, 129.62, 131.54, 132.08, 133.69, 151.98, 156.18, 161.73, 195.27 p.p.m.; IR (KBr): 1666, 1597, 1493, 1226 cm−1; C32H22O6 anal. calcd. for C, 76.48%; H, 4.41%; found C, 76.08%; H, 4.39%.

Poly(ether ketone) 3e

1H NMR δ (300 MHz, CDCl3): 3.68 (6H, s), 6.81–7.02 (4H, m), 7.17 (4H, brs), 7.30 (2H, brs), 7.59–7.73 (4H, m), 7.81 (2H, brs), 7.90 (2H, brs) p.p.m.; 13C NMR δ (125 MHz, CDCl3): 56.56, 111.29, 115.12, 115.76, 117.13, 119.86, 121.48, 125.58, 128.00, 130.09, 131.56, 132.12, 133.85, 135.44, 154.12, 156.23, 161.31, 195.32 p.p.m.; IR (KBr): 1666, 1634, 1594, 1510, 1498, 1228, 1268 cm−1; C36H24O6 anal. calcd. for C, 78.25%; H, 4.38%; found C, 76.92%; H, 4.58%.

Results and Discussion

The bis(fluorobenzoyl)naphthalene monomer, 1,8-bis(4-fluorobenzoyl)-2,7-dimethoxynaphthalene (1), was prepared by the P2O5–MsOH-mediated direct condensation of 2,7-dimethoxynaphthalene with 4-fluorobenzoic acid according to the previous study (see Supplementary Charts S1–S3 show 1H and 13C NMR spectra of monomer 1).7 Based on the X-ray crystal structure analysis, the two aroyl groups of the bis(4-fluorobenzoyl)naphthalene monomer 1 are asymmetrically bent (Figure 1; Supplementary Figure S1 shows the X-ray crystal structure with the detailed crystal information).19 The interplanar angles between the naphthalene ring and the two benzene ring planes of monomer 1 are 66.88(7)° and 88.09(6)°, respectively. To our best knowledge,9, 10, 20, 21 the 1,8-diaroylated naphthalene molecules that have been reported in this study have larger interplanar angles than other homologous molecules, for example, 1,8-diaroylnaphthalene without 2,7-methoxy groups.22 These results indicate that the steric environment around the two carbonyl groups in monomer 1 is highly congested.

Figure 1
figure 1

X-ray crystal structure of 1,8-diaroylnaphthalene 1. A full color version of this figure is available at Polymer Journal online.

Full size image

Table 1 presents the results of the nucleophilic aromatic substitution polymerization of bis(4-fluorobenzoyl)naphthalene monomer 1 with bisphenol A (2a) performed in aprotic polar solvents of DMF, NMP and DMAc. In DMF, the polymerization proceeded slowly (Entry 1). Decreasing the volume of DMF and elongating the reaction time slightly accelerated the chain elongation reaction (Entries 2 and 3). In NMP, the polymerization was retarded, even at rather high reaction temperatures (Entry 4). However, the use of smaller amounts of DMAc achieved smooth polymerization (Entries 5, 6 and 7). In 0.25 ml of DMAc against 0.2 mmol of bis(4-fluorobenzoyl)naphthalene monomer 1, the reaction solution gradually became viscous, and gelatinous curds were finally formed within 1.5 h (Entries 7 and 8). After reprecipitation, the precipitated mass of polymer 3a was demonstrated to be insoluble not only in organic solvents but also in conc. H2SO4 (Entry 8). Polymer 3, which was obtained in Entry 7, was partially soluble in chloroform. The molecular weights of the soluble parts were estimated at approximately 10 000 (n=16) from the 1H NMR spectra. However, the insoluble parts easily yielded a transparent flexible film by heat casting. These results led us to conjecture that the chain elongation reaction proceeded satisfactorily and yielded an insoluble high-molecular-weight polymer. Therefore, diols 2b–e were employed as monomers with bis(4-fluorobenzoyl)naphthalene monomer 1 under similar reaction conditions of Entry 8 (Entries 9–12). The corresponding polymerization proceeded smoothly. The products also showed poor solubility in organic solvents and good film-forming ability by heat casting.

Table 1 Nucleophilic aromatic substitution polycondensation of 1,8-diaroylnaphthalene 1 with diol 2a
Full size table

Table 2 presents the results of polymerization of bis(4-fluorobenzoyl)naphthalene monomer 1 with several diols 2 performed under gentle reaction conditions, in 0.5 ml of DMAc at 150°C for 24 h. The products are easily soluble in CHCl3 (see Supplementary Charts S4–18 show 1H and 13C NMR spectra of the resulting polymers). The numbers of polymerization (n) are determined from the 1H NMR spectra (The 1H NMR spectra are available from Supplementary Information) and calibrated by gel permeation chromatography. The n values that were calibrated by gel permeation chromatography are smaller than those estimated from the 1H NMR spectra. This apparently contradictory observation indicates that the products give small exclusion volumes in CHCl3, although they have large molecular weights. This result can be explained because the polymer chains are tightly folded by the 1,8-diaroylenenaphthalene units. In other words, the structures of 1,8-diaroylenenaphthalene units provide a greater influence to the spatial organization of the polymer than that of the diol units.

Table 2 Numbers of polymerization estimated by 1H NMR spectra and molecular weights calibrated by GPC of poly(ether ketone)s 3a
Full size table

Table 3 summarizes the Tgs and Tds of the poly(ether ketone)s 3 determined using DSC and TG, respectively. The Tgs rise in the order of poly(ether ketone)s 3a, 3b<3c<3e<3d. In addition, the Tds rise in the order of the poly(ether ketone)s 3a<3c<3b<3e<3d. The difference in the Tgs is relatively small. The Tgs are presumably governed by the restricted rotation of the two adjacent aroylene moieties in the 1,8-diaroylenenaphthalene unit. Flexible diol units such as 2a and 2b might permit unrestricted rotation of the bonds. The Tds likely depend on the structure of the diol units. The sp3 carbon in the polymer main chain might lower the Td.

Table 3 Thermal properties of poly(ether ketone)s 3a
Full size table

The polymerization behavior and the solubility/thermal elasticity properties of the aromatic poly(ether ketone)s are well explained based on the structural features of the non-coplanar accumulated aromatic-ring repeating units. The polymerization likely proceeds with the folding of the resulting polymer chains at the 1,8-diaroylenenaphthalene moieties. The benzene rings in the poly(ether ketone)s (3) are coagulated to give accumulation of the mass of the benzene rings such as polystyrene. The concentration of the randomly assembled benzene rings in the poly(ether ketone) (3) and polystyrene might be almost equal. However, the mobility of the polymer chain coagulation in the poly(ether ketone)s (3) at room temperature is far less than that of polystyrene. During the solution polymerization of monomer 1, the solvated solvent molecules might be imperfectly squeezed out from the polymer chains until the formation of the less mobile mass of the yielded polyketone (3) finally results in the appearance of gelatinous curds. The solvation of the polymer chain of the polyketone (3) should be slow to present low solubility after the complete desolvation by reprecipitation. However, the good heat-cast film-forming ability of the polyketone (3) is interpreted as follows: Heating of the polymer solid should make the restricted spatial organization of the polymer loosen slightly by rotation of the bonds that are connected to the carbonyl group of the two adjacent aroyl groups in the 1,8-diaroylenenaphthalene moiety, which enables the polymer mass to behave as mobile chains of benzene rings, just as in polystyrene melt.

Conclusions

The nucleophilic aromatic polycondensation of bis(4-fluorobenzoyl)naphthalene monomer 1, which is a non-coplanar accumulated aromatic-ring molecule, with bisphenols 2 has been successfully achieved in aprotic polar solvents under mild reaction conditions, yielding sufficient amounts of a gelatinoid solid. The solid masses of poly(ether ketone)s 3 obtained via reprecipitation show poor solubility in organic solvents and high film-forming ability through heat casting.