Dendritic organic–inorganic hybrid polyphenol and branched benzoxazine monomers with low curing temperature

Abstract

A novel dendritic organic–inorganic hybrid polyphenol (T2) based on cyclotriphosphazene was synthesized by the condensation reaction of T1 and phenol catalysed by phosphotungstic acid. The dendritic polyphenol T2 comprising twelve phenolic hydroxyl groups exhibited excellent solubility in common organic solvents. Three branched benzoxazine monomers with different oxazine ring content were synthesized via Mannich condensation reaction. These branched benzoxazine monomers showed low initial curing temperature. The properties of corresponding polybenzoxazines, such as thermal stabilities, mechanical properties, dielectric properties and gel contents significantly depended on the oxazine ring content of T3. With the increase of the oxazine ring content for T3, these properties could be greatly improved due to their high crosslinking density.

---

1. Introduction

Polyphenols are extensively distributed in renewable natural resources, such as lignin, Chinese lacquer, and tannin, and recently they have become rather unique and intriguing products because of their diverse range of physicochemical properties and unusual structures.^1–4 Plant polyphenols display a rich and complex spectrum of physical and chemical properties,⁵ giving rise to broad chemical versatility, including the absorption of UV radiation, radical scavenging, and metal ion complexation. In addition, human health has been associated with the consumption of foods and beverages rich in plant polyphenols.⁶ Tea, chocolate, and wine-inspired polyphenols can act to be colorless multifunctional coatings.⁷ In addition to the naturally occurring polyphenols, man-made polyphenol (phenol resins) were also developed 100 years ago and they have been used to be one type of an engineering material after curing. Because of the high reactivity of phenolic hydroxyl groups, polyphenols act to be reactive intermediates for many functional polymeric materials. The modification of phenolic hydroxyls can endow them with particular desirable abilities. For example, Ching-Fong Shu and coworkers⁸ have reported that the hyperbranched poly(aryl ether oxazole)s with terminal phenolic groups were easily functionalized and the glass transition temperature and the solubility of the hyperbranched poly(aryl ether oxazole)s significantly depended on the nature of the chain ends, which increases with increasing chain-end polarities. Polyphenols are also the amorphous molecular materials used in advanced photoresists because of their excellent solubility in alkali.^9–12 The branched polyphenol-based molecular glass resists had high selectivity to radiation and could form photoresist images as small as 60 nm.⁹ However, no reports concerning dendritic polyphenols and their derivatives was published, and only a few studies on branched polyphenols have been reported due to the inherent difficulties in the synthesis and purification.^13,14

One of the most interesting applications of polyphenols is as the key starting raw materials of polybenzoxazine, which is an ideal alternative of traditional phenol–formaldehyde thermosets.^15,16 Polybenzoxazines are also proved to be an excellent carbon precursor, low-surface-energy and high-thermal-stable material, novel electrolyte material and microelectronic material.^17–19 However, unsatisfied processability and high polymerization temperature limit their application in high-tech fields.^20–22 The feasible structural design of phenols or amines can overcome these problems to certain extent.^15,16 To improve the curing process, active hydrogen containing groups such as carboxylic²³ and phenolic groups^24,25 were incorporated into benzoxazine monomers, which could effectively decrease curing temperature due to their catalytic effect on ring-opening polymerization.

Cyclotriphosphazene (CP) derivatives are a typical class of organic–inorganic hybrid compounds,^26–28 and the incorporation of the CP ring into organic compounds or polymers yields fascinating polymer materials.^29–33 We had a beneficial attempt in this area and preliminarily achieved meaningful results.^34,35 In this article, a novel CP-based dendritic polyphenol T2 was designed and synthesized through a relatively simple two-step reaction. Then, highly branched benzoxazine monomer based on cyclotriphosphazene, T3, was obtained by the Mannich reaction of T2, aniline and paraformaldehyde. The representative monomers T3-23.6%, T3-46.6% and T3-75.2%, with oxazine ring content of 23.6%, 46.6% and 75.2%, respectively, were chosen to investigate their polymerization behaviors and the corresponding properties of polybenzoxazine.

2. Experimental

2.1 Chemicals

Hexachlorocyclotriphosphazene [N₃P₃Cl₆] was recrystallized from dry hexane followed by sublimation (60 °C, 0.05 mm Hg) twice before use. 4-Hydroxybenzaldehyde (98%), phenol (99%), phosphotungstic acid (99%), thioglycolic acid (99%), paraformaldehyde (95%) and aniline (99%) were purchased from Sinopharm Chemical Reagent Co., Ltd., China. Anhydrous potassium carbonate was dried at 140 °C prior to use. Tetrahydrofuran was distilled from sodium under a dry argon atmosphere.

2.2 Synthetic section

Synthesis of T1. T1 was prepared according to the reported method.³⁶ Yield: 82%. MP = 160–162 °C. ¹H NMR (DMSO-d₆, TMS, ppm): 6.74–7.78 (Ar-H, 8H, m), 9.48 (CH=N, 1H, s), 8.52 (CH=N, 1H, s). ¹³C NMR (DMSO-d₆, TMS, ppm): 156.8 (C–OH), 156.0 (C=N), 151.8 (C–O), 142.7 (C–N), 134.4 (CH), 131.9 (CH), 129.9 (CH), 123.1 (CH), 121.0 (C), 116.1 (CH), 115.5 (CH). ³¹P NMR (DMSO-d₆, ppm): 8.75.

Synthesis of dendritic polyphenol T2. T2 was prepared according to the reported method³⁷ with a few modifications. A 250 mL three-necked flask was fitted on a magnetic stirrer with a thermometer and a Dean–Stark trap. In the flask were placed T1 (5.0 g, 5.8 mmol), phenol (19.6 g, 208.8 mmol), phosphotungstic acid (1.0 g, 0.35 mmol) and 0.1 mL of thioglycolic acid and 80 mL of cyclohexane. The mixture was heated to 100 °C with the azeotropic removal of the water formed. Water produced during the reaction was separated out by azeotropic distillation with cyclohexane. The azeotropic distillation was continued until no more water came out. The volatile materials were removed by rotary evaporation and the residue was precipitated in a large amount of water, the resulting red solid was filtered off and washed with chloroform and dried in vacuum at 40 °C for 48 h. The crude product was recrystallized from ethyl alcohol/water (1/2). Light yellow powder was collected and dried in vacuum at 50 °C for 481 h (Scheme 1).

Scheme 1 Synthesis of dendritic polyphenol T2 and branched benzoxazine monomer T3 (R1 and R2 were a random distribution and the sum was 12).

Yield: 78%. ¹H NMR (DMSO-d₆, TMS, ppm): 6.67–6.84 (Ar-H, 12H, m), 5.23 (Ar-CH, 1H, s), 9.29 (Ar-OH, 2H, s). ¹³C-NMR (DMSO-d₆, TMS, ppm): 56.0 (C–OH), 155.06 (C–O), 148.42 (C), 134.85 (CH), 130.26 (CH), 120.72 (CH), 15.52 (CH), 54.21 (C–CH). ³¹P NMR (DMSO-d₆, ppm): 8.99.

Synthesis of branched benzoxazine monomers. T3 was prepared according to the reported method³⁸ with a few modifications. The typical procedure was as follows: A 250 mL three-necked flask was fitted with a magnetic stirring, a thermometer and a Dean–Stark trap. Aniline (6.69 g, 71.93 mmol) and paraformaldehyde (4.32 g, 143.64 mmol) were mixed and stirred in ice bath for 1 h. Then, 150 mL toluene, 50 mL DMAc, T2 (15 g, 7.98 mmol) were added into the mixture with stirring. The mixture was gently heated to 110 °C and refluxed for 72 h. Water produced during the reaction was separated out by azeotropic distillation with toluene. The samples for different reaction time (3 h, 24 h and 72 h) were removed with a syringe. After the solvents were removed, the residue was collected and dried in vacuum at 50 °C for 48 h (Scheme 1).

2.3 Measurements

The structure of the compounds were verified by solution-state proton (¹H), carbon (¹³C) and phosphorus (³¹P) nuclear magnetic resonance spectroscopy (NMR) using Bruker AV400NMR spectrometer at a proton frequency of 400 MHz, as well as the corresponding phosphorus frequency at room temperature. Chemical shifts were reported in ppm downfield from SiMe₄. Thermal transitions were monitored with a differential scanning calorimeter (DSC), Model 204F1 from NETZSCH Instruments, and a scan rate of 10 °C min⁻¹ over a temperature range of 30–300 °C and nitrogen flow rate of 20 mL min⁻¹ were used in DSC experiments. Thermogravimetric analysis (TGA) was performed with a NETZSCH Instruments' High Resolution STA 409 PC thermogravimetric analyzer that was purged with nitrogen at a flow rate of 70 mL min⁻¹. A heating rate of 20 °C min⁻¹ was used and scanning range was from RT to 1000 °C. Infrared spectra were recorded using a Bruker VERTEX 70 Fourier transform infrared spectrometer (FT-IR) under ambient condition. The potassium bromide disks were prepared by compressing the powder. Mechanical properties were measured using a dynamic mechanical thermal analysis (DMA) apparatus (PerkinElmer, Diamond DMA). Specimens (50 × 10 × 1.0 mm) were tested in 3 point bending mode. The thermal transitions were studied in the scope of 20–300 °C at a heating rate of 4 °C min⁻¹ and at a fixed frequency of 1 Hz.

2.4 Preparation of polybenzoxazines

Samples for dynamic mechanical analysis (50 × 10 × 1.0 mm), dielectric measurement (20 × 10 × 1.0 mm) and humidity absorption test (50 × 10 × 1.0 mm) were prepared as follows: three benzoxazine monomers were added to the tinfoil mold and gently melted at 150 °C under vacuum for 1 h. Subsequently, the samples were heated stepwise and cured in a temperature-controlled oven at 160 °C and 180 °C each for 8 h, then post-cured at 200 °C and 220 °C each for 2 h. Thereafter, the samples were allowed to slowly cool to room temperature to prevent cracking. Finally, the samples were polished to be regularly shaped.

2.5 Gel content of polybenzoxazines

The gel fraction of the samples (50 × 10 × 1.0 mm) was determined by standard extraction method³⁹ using chloroform. The process involved continuous extraction with chloroform in a 500 mL round bottom flask for 72 h until constant weight was obtained. After the extraction, the samples were dried and the gel content was calculated according to the formula (1) as follows:where m_t and m₀ represent sample weights after and before extraction, respectively.

2.6 Dielectric measurements of polybenzoxazines

Dielectric constant and dielectric loss were measured at room temperature under atmospheric air by the two parallel plate modes at 125 Hz–18 MHz using Agilent 4294A Precision Impedance Analyzer.^40,41 A sample (about 20 mm × 10 mm × 2 mm) was placed between the two copper electrodes to form a parallel plate capacitor. Prior to each measurement, the sample was dried under vacuum at 100 °C for 3 h.

2.7 Humidity absorption

In the humidity absorption measurements, the cured samples were conditioned under vacuum at 90 °C for 20 h before they were placed in air (75% and 33% RH).⁴² All these experiments were conducted at room temperature. Then, the weight percentages of humidity absorption of the cured samples were calculated according to the formula (2) as follows:where the W_t and W₀ represent the sample weights after and before (dry sample) humidity absorption, respectively. Humidity absorption content was measured for 5 samples of each material group and the average values were recorded.

3. Results and discussion

3.1 Synthesis and characterization of polyphenol T2

Branched polyphenol T2 was facilely prepared via the condensation of T1 with phenol catalysed by phosphotungstic acid and thioglycolic acid (Scheme 1). Initially, we performed the reaction in the solvent of toluene, but ¹H NMR analysis showed that the purity of the product was poor due to high azeotropic temperature in toluene–water system. To reduce side reactions, which might result from the oxidation of phenolic hydroxyls and the deactivation of catalysts, cyclohexane was used in the reaction. Main impurities (phenol and catalysts) in the crude product were easy to remove through repeated dissolution in ethanol and precipitation in water. T2 with high purity were obtained by recrystallization from ethyl alcohol/water. The dendritic polyphenol exhibited excellent solubility in common organic solvents such as methanol, ethanol, acetone, 1,4-dioxane, THF, ethyl acetate, DMSO and DMF.

Fig. 1 shows the ¹H NMR spectra of T2. Peaks at 9.29 ppm (Ar-OH) and 5.23 ppm (Ar-CH) verified the successful introduction of phenol to the T1 structure. Signals at 6.67–6.84 ppm for Ar-H were also observed. In the corresponding ¹³C NMR spectrum in Fig. 2, resonance appearing at 54.21 ppm was assigned to the methyne carbon of Ar-CH. Other chemical shifts (ppm) were assigned to the aromatic carbon resonances: 156.0 (C–OH), 155.06 (C–O), 148.42 (C), 134.85 (CH), 130.26 (CH), 120.72 (CH), 115.52 (CH). The ³¹P NMR is shown in Fig. 3, and a single peak appeared at 8.98 ppm, which indicated that the substituted reaction of phenol onto T1 was complete.

Fig. 1 ¹H NMR spectrum of T2 (solvent: DMSO-d₆).

Fig. 2 ¹³C NMR spectrum of T2 (solvent: DMSO-d₆).

Fig. 3 ³¹P NMR spectrum of T2 (solvent: DMSO-d₆).

3.2 Synthesis of branched benzoxazine monomers

T3 was synthesized using phenol (T2), aniline, and formaldehyde according to Scheme 1. As shown in Scheme 1, the benzoxazine and phenol groups were randomly distributed as branches of benzoxazine monomer T3. The exact position and numbers of the benzoxazine and phenol groups in T3 molecule were time-dependent and could not be determined. However, the sum of the benzoxazine and phenol groups in T3 molecule was 12. As the reaction proceeds, the phenol groups gradually turned into benzoxazine groups through Mannich reaction. The formation of oxazine rings was monitored by ¹H NMR and the results are presented in Fig. 4. Resonances appearing at 4.50 ppm and 5.27 ppm are assigned to the methylene protons of O–CH₂–N of the benzoxazine ring, respectively. The multiplets at 6.66–6.80 ppm are assigned to the aromatic protons. The singlet peak at 9.28 ppm is assigned to the unreacted phenolic hydroxy proton. The single peak at 5.73 ppm is assigned to the methine proton connected with three phenyl rings.

Fig. 4 ¹H NMR spectra of branched benzoxazine monomers (solvent: DMSO-d₆). The different oxazine ring contents at given time interval are marked in the spectra.

The oxazine ring content for benzoxazine monomers at a given time interval was determined from the ¹H NMR spectra using the eqn (3) as follows:

where C is the oxazine ring content at given time interval, R the ratio of the integral area of methylene protons for O–CH₂–N and the integral area of all aromatic protons. The equation was established in the condition where the opening of oxazine ring was not considered. Different oxazine ring contents for all monomers at a given time interval is listed in Table S1.† The estimated oxazine ring content after 3 h, 24 h and 72 h were 23.6%, 46.6% and 75.2%, respectively. The peak assigned to the phenolic hydroxyl still exists after 72 h; thus, it is not necessary to prolong the reaction time.

3.3 DSC studying for the curing behavior of branched benzoxazine monomers

It is known that the high polymerization temperature is a major drawback that limits the application of polybenzoxazine. The polymerization temperature of the three hydroxyl-containing benzoxazine monomers was examined by DSC and the results are shown in Fig. 5 and summarized in Table 1. A wide exothermic peak corresponding to the ring-opening polymerization was observed for T3-23.6%, in which the onset and maximum temperatures of the exotherm were 97.2 °C and 201.5 °C, respectively. With the increase of oxazine ring content, the exothermic peaks obviously changed from the DSC results. For T3-46.6%, the onset and maximum temperatures of the exotherm were increased to 169.7 °C and 222.4 °C, respectively; for T3-75.2%, the onset and maximum temperatures of the exotherm were separately increased to 170.8 °C and 206.4 °C. Comparing to T3-46.6% and T3-75.2%, the polymerization temperature for T3-23.6% was obviously reduced. The results indicated that the hydroxyl groups could reduce the onset polymerization temperature of benzoxazine monomers, which was in accordance with other report.²⁴ Nevertheless, the maximum temperature of the exotherm gave the ascending order of T3-23.6%, T3-75.2% and T3-46.6%. The reason for the higher maximum temperature of T3-46.6% than T3-75.2% may be due to the different oxazine ring distribution in T3-46.6% and T3-75.2%. It was reported that the side groups of the CP derivatives were arranged approximately perpendicular to the CP ring forming parallel triplets that were pointing upward and downward to give a calamitic superstructure capable of being organized in a nematic phase.^43,44 This meant that the oxazine rings and hydroxyl groups were separately distributed in the top and bottom faces of the CP plane. In addition, the position of hydroxyl group relative to benzoxazine structure played a significant role in accelerating the polymerization.⁴⁵ For T3-46.6%, the oxazine ring content was close to 50%, and the oxazine rings and hydroxyl groups were almost distributed at far intervals as the branches of T3-46.6%, while for T3-75.2% with far more oxazine ring content than that of T3-46.6%, the high concentration of oxazine rings were closer to the hydroxyl groups, which benefited the catalytic efficiency of the hydroxyl groups.

Fig. 5 DSC curves of polybenzoxazines by curing monomers T3-23.6%, T3-46.6% and T3-75.2%.

Table 1 Thermal stability of polybenzoxazines by curing T3-23.6%, T3-46.6% and T3-75.2%

Samples	T0^a (°C)	Tmax^b (°C)	T5%^c (°C)	T10%^d (°C)	Tmax^e (°C)	Yc^f (%)
a Onset temperature of exothermic peak.b Maximum of the polymerization exotherm.c The temperature for 5% weight loss.d The temperature for 10% weight loss.e Maximum weight loss temperature.f Char yields at 850 °C.
T3-23.6%	97.2	201.5	375	407	403	46.8
T3-46.6%	169.7	222.4	408	437	435	51.9
T3-75.2%	170.8	206.4	427	474	498	56.2

---

3.4 Gel contents of the polybenzoxazines

The gel contents of the polybenzoxazines cured by monomers T3-23.6%, T3-46.6% and T3-75.2% were determined using the methodology described in the experimental part, and the results are shown in Fig. 6. The gel contents of PT3-23.6%, PT3-46.6% and PT3-75.2% were 90.8%, 95.5% and 99%, respectively. The results indicated that the gel contents of all polybenzoxazines depended on the different oxazine ring content of the monomers. The increase of the oxazine ring content in the monomer resulted into high gel content and high crosslinking density. For the polybenzoxazine PT3-23.6% cured by monomer T3-23.6%, in which the oxazine ring content was 23.6%, the gel content reached 90.8%, which was the lowest among the three monomers. This meant that even small amount of benzoxazine units attached on the middle molecular weight polyphenol T2 (mol. W_t = 1882 g mol⁻¹) could act to be effective crosslinkers to form thermosets with high crosslinking density. When the oxazine ring content exceeded 50%, the gel content of the corresponding polybenzoxazine was above 95%.

Fig. 6 Gel contents of the polybenzoxazines PT3-23.6%, PT3-46.6% and PT3-75.2%.

3.5 Thermal properties of cured polymers

Thermal stability of the three polybenzoxazines was investigated by TGA. The TGA profiles are shown in Fig. 7, and the details are shown in Table 1. The 5% and 10% weight loss temperatures (T_5% and T_10%) for the polybenzoxazine PT3-23.6% were 375 and 407 °C, respectively, while for PT3-46.6% and PT3-75.2%, the T_5% and T_10% were all higher than that of 23.6%: for PT3-46.6%, they were 408 and 437 °C, respectively; for PT3-75.2%, they were 427 and 474 °C, respectively. Similarly, the maximum weight loss temperature (T_max) of the polybenzoxazine PT3-75.2% was also lower than those of PT3-23.6% and PT3-46.6%. The Y_c of PT3-75.2% was as high as 56.2%, and that of PT3-23.6% and PT3-46.6% were considerably lower than that of PT3-75.2%; the PT3-23.6% sample held the lowest yield, only 46.8%.

Fig. 7 TG curves of polybenzoxazines. (a) PT3-75.2%, (b) PT3-46.6%, (c) PT3-23.6%.

From the TGA results, it can be deduced that the organic–inorganic hybrid polybenzoxazines, by curing the above monomers T3-23.6%, T3-46.6% and T3-75.2%, showed outstanding thermal stability. It was considered that the rigid inorganic CP ring enhanced the thermal stability of the polybenzoxazines, which provide thermally stable sites compared to the traditional main-chain type polybenzoxazine such as bisphenol-based polybenzoxazines.²³ Furthermore, the number of benzoxazine moieties on the CP ring also played an important role in the thermal stability of polybenzoxazines by improving cross-linked density and minimizing dangling side groups. Therefore, the higher the oxazine ring content, the better was the thermal stability of polybenzoxazine.

3.6 Mechanical properties of polybenzoxazines

The dynamic mechanical properties of the CP-based polybenzoxazines PT3-23.6%, PT3-46.6% and PT3-75.2% were measured from room temperature to the rubbery plateau of each material. The storage moduli (E′) of these polybenzoxazines are presented in Fig. 8(A). The E′ value of PT3-23.6% at room temperature was 0.319 GPa, while they were 1.03 GPa and 2.16 GPa for PT3-46.6% and PT3-75.2%, respectively. The E′ value of the polybenzoxazine were influenced by the number of the benzoxazine moieties, which led to more spatial crosslinking sites. Therefore, the high oxazine ring content was greatly in favor of the increase of the E′ value for the corresponding polybenzoxazine.

Fig. 8 Storage moduli (A) and tan δ (B) of polybenzoxazines PT3-23.6%, PT3-46.6% and PT3-75.2%.

The glass transition temperature (T_g) is also an important property of the polybenzoxazines, which could be deduced from the corresponding peak temperature of tan δ value, as shown in Fig. 8(B). The T_g of PT3-23.6% was 173.3 °C, while they were 175.1 °C and 179.2 °C for PT3-46.6% and PT3-75.2%, respectively. The highest cross-linked degree for PT3-75.2% could constrain the move of the cross-linked chains under heat and move them at higher temperature. Thus, the T_g of PT3-75.2% was higher than that of PT3-23.6% and PT3-46.6%. In addition, the PT3-75.2% sample produced a broader glass transition than PT3-23.6% and PT3-46.6%, which meant that the composition distribution in PT3-75.2% was not considerably narrow as that in PT3-23.6% and PT3-46.6%.

3.7 Dielectric analysis of polybenzoxazines

Dielectric property of polymers is one of the key characters for its application in microelectronic devices.⁴⁰ The dependence of the dielectric constant and dielectric loss of the three polybenzoxazine samples PT3-23.6%, PT3-46.6% and PT3-75.2% were studied in the frequency range of 125 Hz to 10⁶ Hz at room temperature. Fig. 9 shows that both the dielectric constant (ε) and the dielectric loss (tan δ) largely depended on the oxazine ring content. Because the dielectric constant and dielectric loss were directly related to the polarizability of material, which could be increased by increasing the polarizability, the dielectric constant and dielectric loss were strongly dependent on its chemical structure.⁴¹ The dielectric constants of the polybenzoxazine PT3-23.6% at 125 Hz was about 5.74, for PT3-46.6% and PT3-75.2%, the value changed to 4.57 and 4.45, respectively. In the frequency range of 125 Hz to 10⁶ Hz, the dielectric constant for each sample showed a small decrease. When the frequency of the applied field increased to 10⁶ Hz, the dielectric constants of PT3-23.6%, PT3-46.6% and PT3-75.2% declined to 4.85, 4.09 and 3.75, respectively. This was because of the fact that PT3-75.2% was a highly cross-linked polymer and most of its components linked together by chemical bonds, thus it showed lowest polarizability. However, for PT3-23.6% and PT3-46.6%, which were cured by the monomers with the oxazine ring content of 23.6% and 46.6%, respectively, the uncrosslinked groups acted to be polar pendant groups linked to the network by chemical bond and increased the polarizability. The dielectric loss of PT3-23.6%, PT3-46.6% and PT3-75.2% at 125 Hz were 0.061, 0.042 and 0.040, respectively. These values showed small changes in the frequency range of 125 Hz to 10⁶ Hz. The reason for these changes was the same as that of the dielectric constant.

Fig. 9 Frequency dependence of the dielectric constant (A) and dielectric loss (B) of the polybenzoxazines PT3-23.6%, PT3-46.6% and PT3-75.2%.

3.8 Humidity absorption of the polybenzoxazines

The humidity absorption of the three polybenzoxazine samples PT3-23.6%, PT3-46.6% and PT3-75.2% were measured at different relative humidity (RH = 75% & 33%) at room temperature. As shown in Fig. 10, all the polybenzoxazine samples exhibited a low water uptake value (<1.0 wt%) at room temperature after 10 days regardless of the RH value of the air was high or low due to the complete hydrogen-bond network formed between phenolic OH and nitrogen atoms, which hinder the water absorption and transportation.⁴² Moreover, the water uptake ability of the samples regularly changed with the increase of the oxazine ring content. The final water absorption values of the three polybenzoxazines decreased in the order of PT3-23.6%, PT3-46.6% and PT3-75.2%. The value of PT3-23.6% was about 0.95 wt%, and the value decreased to about 0.79 wt% and 0.66 wt% for PT3-46.6% and PT3-75.2%, respectively. The lower water uptake ability of PT3-75.2% was attributed to the higher crosslinking density in the network.

Fig. 10 Humidity adsorption of the polybenzoxazines PT3-23.6%, PT3-46.6% and PT3-75.2% at different relative humidity (RH) at room temperature. A. RH = 75%; B. RH = 33%.

4. Conclusions

We have synthesized a novel dendritic organic–inorganic hybrid polyphenol (T2) based on cyclotriphosphazene. The dendritic polyphenol possessed good solubility in common solvents. Based on the dendritic phenol, three branched benzoxazine monomers T3-23.6%, T3-46.6% and T3-75.2% with different content of phenolic hydroxyl groups were synthesized via Mannich condensation reaction. The polymerization behaviors of the monomers and properties of the polybenzoxazines largely depended on the oxazine ring content of corresponding monomers. These monomers T3-23.6%, T3-46.6% and T3-75.2% showed low initial polymerization temperature due to the catalytic effect of their phenolic hydroxyl groups. The more phenolic hydroxyl groups the monomer had, more easily it underwent ring-opening polymerization under heat. The thermal stability, mechanical property, dielectric property and humidity absorption of the polybenzoxazines relied on the oxazine ring content in corresponding monomers. The char yields at 850 °C were near to 60%, all T_g were higher than 170 °C, and dielectric constants were about 4 at 1 MHz, and corresponding humidity absorptions of them were below 1.0 wt% (the lowest one was about 0.66 wt%). Combined with the excellent performances of polybenzoxazines, we believe that the facile synthesis and versatility of dendritic polyphenol are attractive potentials that will lead to many additional applications beyond our description.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (no. 21274049), the Opening Project of Key Laboratory of Optoelectronic Chemical Materials and Devices of Ministry of Education, Jianghan University (no. JDGD-2013-06) and the Fundamental Research Fund for Central Universities (2013QN159). All authors sincerely thank Analysis and Testing Center of HUST for NMR and DMA test.

Notes and references

1. B. Lochab, S. Shukla and I. K. Varma, RSC Adv., 2014, 4, 21712–21752.
2. W. Jiang, K. Lai, K. Liu, R. Xia, F. Gao, Y. Wu and Z. Gu, Nanoscale, 2014, 6, 1305–1310.
3. D. G. Barrett, T. S. Sileika and P. B. Messersmith, Chem. Commun., 2014, 50, 7265–7268.
4. J. A. Nichols and S. K. Katiyar, Arch. Dermatol. Res., 2010, 302, 71–83.
5. E. Haslam, Practical polyphenolics: from structure to molecular recognition and physiological action, Cambridge University Press, 1998.
6. V. R. Preedy, Tea in health and disease prevention, Elsevier, London, 2013.
7. T. S. Sileika, D. G. Barrett, R. Zhang, K. H. A. Lau and P. B. Messersmith, Angew. Chem., Int. Ed., 2013, 52, 10766–10770.
8. Z. H. Gong, C. M. Leu, F. I. Wu and C. F. Shu, Macromolecules, 2000, 33, 8527–8533.
9. D. Bratton, R. Ayothi, H. Deng, H. B. Cao and C. K. Ober, Chem. Mater., 2007, 19, 3780–3786.
10. A. D. Silva, L. K. Sundberg, H. Ito, R. Sooriyakumaran, R. D. Allen and C. K. Ober, Chem. Mater., 2008, 20, 7292–7300.
11. T. Nakayama and M. Ueda, J. Mater. Chem., 1999, 9, 697–702.
12. J. Liu, Y. Qiao, Z. Liu and L. Wang, RSC Adv., 2014, 4, 21093–21100.
13. J. Zhang, H. Wang and X. Li, Polymer, 2006, 47, 1511–1518.
14. H. Wu, X. Huang, M. Gao, X. Liao and B. Shi, Green Chem., 2011, 13, 651–658.
15. N. Ghosh, B. Kiskan and Y. Yagci, Prog. Polym. Sci., 2007, 32, 1344–1391.
16. Y. Yagci, B. Kiskan and N. N. Ghosh, J. Polym. Sci., Part A: Polym. Chem., 2009, 47, 5565–5576.
17. C. F. Wang, Y. C. Su, S. W. Kuo, C. F. Huang, Y. C. Sheen and F. C. Chang, Angew. Chem., Int. Ed., 2006, 45, 2248–2251.
18. C. H. Lin, S. L. Chang, T. Y. Shen, Y. S. Shih, H. T. Lin and C. F. Wang, Polym. Chem., 2012, 3, 935–945.
19. S. K. Kim, T. Ko, K. Kim, S. W. Choi, J. O. Park, K. H. Kim, C. Pak, H. Chang and J. C. Lee, Macromol. Res., 2012, 20, 1181–1190.
20. P. Yang, X. Wang, H. Fan and Y. Gu, Phys. Chem. Chem. Phys., 2013, 15, 15333–15338.
21. T. Agag and T. Takeichi, Polymer, 2000, 41, 7083–7090.
22. W. C. Chu, J. G. Li and S. W. Kuo, RSC Adv., 2013, 3, 6485–6498.
23. W. H. Hu, K. W. Huang and S. W. Kuo, Polym. Chem., 2012, 3, 1546–1554.
24. R. Andreu, J. Reina and J. Ronda, J. Polym. Sci., Part A: Polym. Chem., 2008, 46, 6091–6101.
25. W. Men, Z. Lu and Z. Zhan, J. Appl. Polym. Sci., 2008, 109, 2219–2223.
26. H. R. Allcock, Chemistry and applications of polyphosphazenes, Wiley-Interscience, 2003.
27. S. Sudhakar and A. Sellinger, Macromol. Rapid Commun., 2006, 27, 247–254.
28. J. Y. Chang, H. J. Ji, M. J. Han, S. B. Rhee, S. Cheong and M. Yoon, Macromolecules, 1994, 27, 1376–1380.
29. P. Mohanty, L. D. Kull and K. Landskron, Nat. Commun., 2011, 2, 401.
30. C. Y. Yuan, S. Y. Chen, C. H. Tsai, Y. S. Chiu and Y. W. Chen-Yang, Polym. Adv. Technol., 2005, 16, 393–399.
31. M. Harmjanz, B. L. Scott and C. J. Burns, Chem. Commun., 2002, 1386–1387.
32. Y. W. Cho, J. R. Lee and S. C. Song, Bioconjugate Chem., 2005, 16, 1529–1535.
33. P. Jia, J. Xu, J. Ma and X. Lu, Eur. Polym. J., 2009, 45, 772–778.
34. X. Wu, S. Z. Liu, D. T. Tian, J. J. Qiu and C. M. Liu, Polymer, 2011, 52, 4235–4245.
35. X. Wu, Y. Zhou, S. Z. Liu, Y. N. Guo, J. J. Qiu and C. M. Liu, Polymer, 2011, 52, 1004–1012.
36. G. A. Carriedo, L. Ferńndez-Catuxo, F. J. G. Alonso, P. G. Elipe, P. A. Gonźlez and G. Śnchez, J. Appl. Polym. Sci., 1996, 59, 1879–1885.
37. J.-Z. Xu, N.-Z. Jiang, J. Zhang and R.-S. Jiang, Chem. Res. Chin. Univ., 2005, 21, 65–68.
38. S. W. Choi, H. Y. Sun, M. J. Lee and W. S. Jeon, European Patent, 1760110 A1, 2006.
39. M. Claudino, J. M. Mathevet, M. Jonsson and M. Johansson, Polym. Chem., 2014, 5, 3245.
40. H. C. Chang, H. T. Lin and C. H. Lin, Polym. Chem., 2012, 3, 970–978.
41. K. D. Demir, B. Kiskan, S. S. Latthe, A. L. Demirel and Y. Yagci, Polym. Chem., 2013, 4, 2106–2114.
42. J. Sundberg, G. Toriz and P. Gatenholm, Polymer, 2013, 54, 6555–6560.
43. K. Moriya, T. Suzuki, S. Yano and S. Miyajima, J. Phys. Chem. B, 2001, 105, 7920–7927.
44. J. Barbera, M. Bardaji, J. Jimenez, A. Laguna, M. P. Martinez, L. Oriol, J. L. Serrano and I. Zaragozano, J. Am. Chem. Soc., 2005, 127, 8994–9002.
45. M. Baqar, T. Agag, R. Z. Huang, J. Maia, S. Qutubuddin and H. Ishida, Macromolecules, 2012, 45, 8119–8125.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra07035f

---