Synthesis, thermal properties and curing kinetics of fluorene diamine-based benzoxazine containing ester groups

European Polymer Journal 49 (2013) 2759–2768

Contents lists available at SciVerse ScienceDirect

European Polymer Journal journal homepage: www.elsevier.com/locate/europolj

Synthesis, thermal properties and curing kinetics of fluorene diamine-based benzoxazine containing ester groups Xuan-yu He a,b, Jun Wang b, Yu-dan Wang b, Chen-juan Liu a, Wen-bin Liu a,b,⇑, Lei Yang c a

Key Laboratory of Superlight Materials and Surface Technology, Ministry of Education, Harbin Engineering University, Harbin, Heilongjiang 150001, China Polymer Materials Research Center, College of Materials Science and Chemical Engineering, Harbin Engineering University, Harbin 150001, China c Key Laboratory of Forest Plant Ecology, Ministry of Education, Northeast Forestry University, Harbin 150040, China b

a r t i c l e

i n f o

Article history: Received 16 February 2013 Received in revised form 12 May 2013 Accepted 3 June 2013 Available online 27 June 2013 Keywords: Fluorene diamine-based benzoxazine Polymerization behavior Curing kinetics Thermal properties

a b s t r a c t A novel diamine-based benzoxazine monomer containing ester and bulky fluorene groups (BABPF-p) was successfully synthesized by the reaction of 9,9-bis-[4-(4-aminobenzoyloxy)phenyl]fluorene with paraformaldehyde and phenol. The chemical structure of monomer was confirmed by Fourier-transform infrared (FTIR) and 1H and 13C nuclear magnetic resonance spectroscopy (1H and 13C NMR). The polymerization behavior of monomer was analyzed by FTIR. The curing kinetics was studied by differential scanning calorimetry (DSC), and the kinetic parameters were determined. The autocatalytic model and Kamal model showed good agreement with the experimental results. The thermal and mechanical properties of poly(BABPF-p) were evaluated with DSC, dynamic mechanical thermal analysis (DMTA), and thermogravimetric analysis (TGA). The cured polymer exhibited the higher glass transition temperature (Tg) than diaminodiphenylmethane-based benzoxazine (P-ddm) and fluorene diamine-phenol-based polybenzoxazine (poly(BF-p)). The incorporation of flexible ester carbonyl linkages into the macromolecular backbone had slight effect on the thermal stability of fluorene-containing polybenzoxazine. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction Polybenzoxazines, as a class of thermosetting phenolic resins formed by the cationic ring-opening of the corresponding benzoxazine monomers without any added initiator, have demonstrated various attractive properties such as high thermal stability, high char yields, high glass transition temperature (Tg), near-zero volumetric change upon curing, good mechanical and dielectric properties, low water absorption, and low flammability. These unique characteristics make the polybenzoxazines a better candidate over epoxies and traditional phenolic resins in the electronics, aerospace, and other industries [1–6]. According to the related researches, most difunctional benzoxa⇑ Corresponding author at: Key Laboratory of Superlight Materials and Surface Technology, Ministry of Education, Harbin Engineering University, Harbin, Heilongjiang 150001, China. Tel./fax: +86 451 82589540. E-mail address: [email protected] (W.-b. Liu). 0014-3057/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.eurpolymj.2013.06.018

zines were synthesized from bisphenols, monoamines, and formaldehyde. The large varieties of aromatic bisphenols and monoamines allow for considerable molecule-design flexibility of benzoxazines [7–13]. Another series of important difunctional benzoxazines were the diaminebased benzoxazines. Allen and Ishida synthesized a series of aliphatic diamine-based benzoxazine monomers using monofunctional phenol with linear aliphatic diamines [14,15]. And this kind of polybenzoxazine possesses less brittleness than typical bisphenol-based polybenzoxazine, which incorporates with a flexible bifunctional amine in centered portion. Lin and coworkers successfully synthesized a series of aromatic diamine-based benzoxazines by the three-step, two-pot, or one-pot procedure [16–18]. Agag et al. reported a new route for high yield synthesis of aromatic diamine-based benzoxazines using the onestep procedure, in which xylenes as a nonpolar, high-boiling-point solvent, was used at 150 °C [19]. Therefore, various difficult aromatic diamine-based benzoxazines can be

2760

X.-y. He et al. / European Polymer Journal 49 (2013) 2759–2768

prepared by different synthesis method [20–25]. Moreover, the aromatic diamine-based polybenzoxazine can provide higher glass transition temperature and thermal stability than the conventional bisphenol-A-based polybenzoxazine and this kind of benzoxazine resin can be polymerized in the same manner as the bisphenol-based polybenzoxazine [26]. Molecular fluorene contains two benzene rings linked with a five-membered ring which provides high overlaps of p-orbitals [27]. Incorporation of bulky pendant groups can impart a significant increase in both Tg and thermooxidative stability by restricting segmental mobility. Therefore, the development of these polymers has attracted extensive research interests during the past few years [28–34]. Recently, researches on synthesis, polymerization behavior, curing kinetics, and thermal properties of fluorene-based benzoxazines have been reported. Liu et al. prepared a series of fluorene-containing benzoxazines from the reaction of 9,9-bis-(4-hydroxyphenyl)-fluorene, aromatic and aliphatic primary amines, and formaldehyde [35]. Liu et al. also reported the preparation of a furan-terminated and fluorene-containing benzoxazine, and studied their properties [36]. Lu et al. subsequently studied the curing kinetics of fluorene-based benzoxazines [37]. In addition, Xu et al. proved that hydrophobic characteristics of the fluorenyl-based polybenzoxazines are due to the low polarity of fluorenyl and p–p stacking interaction of fluorenyl on the surface [38]. Lin et al. reported the synthetic route of a benzoxazine based on 9,9-bis(4-aminophenyl)fluorene, 2-hydroxybenzaldehyde /phenol, and formaldehyde/paraformaldehyde by using two-pot and one-pot procedures, and studied the optical and thermal properties of the monomer and polybenzoxazines [39]. Very recently, we also prepared a series of fluorene diamine-based benzoxazine monomers from the reaction of 9,9-bis-(4-aminophenyl) fluorene with paraformaldehyde and unsubstituted or substituted phenols. The polymerization behavior, regioselectivity, and thermal properties of the monomers and the corresponding cured polymers were discussed. However, despite the advantageous properties of the fluorene-based benzoxazines, their further applications were limited considerably due to brittleness of the formed polymers [40]. To overcome this limitation, introduction of ester groups in the thermosetting materials has been used as one of the most useful strategies to improve the toughness of the thermosets [41–45]. Curing process of benzoxazine resin is very complicated, which includes gelation, vitrification, subsequent crosslinking, and origination of three-dimensional networks. Therefore, the fundamental investigation of benzoxazine curing reaction kinetics is an attractive topic for better understanding of the curing behavior as well as for the control and optimization of production processes and performance of final products [46]. DSC is a phenomenological method, in which kinetic analysis only need to calculate the heat variation for a whole reaction process with temperature or time, without distinguishing the individual reaction in the process. So DSC is the most utilized technique to determine kinetic parameters and rate equation of polymerization of benzoxazine [37,47–54].

In a continuation of the study on fluorene-based benzoxazine, the present work describes a successful preparation of a diamine-based benzoxazine monomer without much sacrifice of thermal properties by incorporating aryl ester and bulky fluorene groups into the polymer backbone. Chemical structure of monomer was confirmed by FTIR, 1H and 13C NMR. The polymerization behavior and thermal properties of the monomer and its cured polymer were discussed. Non-isothermal DSC was used to investigate the curing kinetics of BABPF-p. Starink method, autocatalytic model, and Kamal method were applied to calculate the kinetic parameters for cure reaction of BABPF-p. 2. Experimental 2.1. Materials 9,9-Bis-(4-hydroxyphenyl)fluorene was synthesized according to literature [55]. 4-nitrobenzoyl chloride, triethylamine, hydrazine monohydrate, phenol and paraformaldehyde were obtained from Shanghai Jingchun Reagent Co., Ltd. (China). Ten percent palladium on activated carbon (Pd-C) was purchased from National Pharmaceutical Chemical Co., Ltd. (China). All solvents were purchased from Tianjin Kermel Chemical Reagent Co., Ltd. (China), and used without further purification. 2.2. Monomer synthesis Scheme 1 illustrates the synthesis of 9,9-bis-[4-(4nitrobenzoyloxy)phenyl] fluorene, 9,9-bis-[4-(4-aminobenzoyloxy)phenyl]fluorene and fluorene diamine-based benzoxazine (BABPF-p). 2.2.1. 9,9-Bis-[4-(4-nitrobenzoyloxy)phenyl]fluorene In a three necked 500 mL flask equipped with a magnetic stirrer, a reflux condenser and a thermometer, 5.25 g vacuum dried 9,9-bis-(4-hydroxyphenyl) fluorene was dissolved in 50 mL dry chloroform; 50 mL of triethylamine was added as catalyst; 5.8 g of excess 4-nitrobenzoyl chloride dissolved in 250 mL of dry chloroform was added dropwise at room temperature. Then the reaction mixture was heated to the reflux temperature for 16 h. 150 mL of distilled water was added and a viscous yellow organic phase was separated and washed several times with 100 mL of methanol carefully in order to eliminate of unreacted 4-nitrobenzoyl chloride and also its hydrolysis derivatives. After that, chloroform was removed using a rotary evaporator. The product was dried under vacuum at 60 °C for 24 h. A yellow powder (87% yield, m.p. 289–291 °C) was obtained. FTIR (KBr, cm1): 1730 (C@O stretching), 1519 (NO2 asymmetric stretching), 1345 (NO2 symmetric stretching), 1257 (CAOAC asymmetric stretching), 842, 755 (CAH out-of-plane bending). 1 H NMR (500 MHz, CDCl3, ppm): 7.10–8.35 (m, 24H, ArAH). 13C NMR (500 MHz, CDCl3, ppm): 163.24 (2C, the ester carbonyl carbons), 120.38–150.91 (36C, the carbons of benzene ring), 64.64 (1C, quaternary carbon in the fluorene ring).

X.-y. He et al. / European Polymer Journal 49 (2013) 2759–2768

2761

Scheme 1. The synthesis route of diamine-based benzoxazine monomer containing aryl ester and bulky fluorene groups (BABPF-p).

2.2.2. 9,9-Bis-[4-(4-aminobenzoyloxy)phenyl]fluorene Hydrazine monohydrate (2.25 g) was added dropwise over a period of 1 h at 85 °C to a mixture of 9,9-bis-[4-(4nitrobenzoyloxy)phenyl]fluorene (6.48 g), ethanol (240 mL), and a catalytic amount of Pd-C (0.5 g). After the addition was completed, the reaction was continued at reflux temperature for another 24 h, and the mixture was filtered to remove Pd-C. After cooling, white-needle crystals were isolated by filtration and washed thoroughly with ethanol. The yield was 74%; m.p. 280–283 °C. FTIR (KBr, cm1): 3481, 3376 (NAH stretching), 1699 (C@O stretching), 1273 (CAOAC asymmetric stretching), 842, 742 (CAH out-ofplane bending). 1H NMR (500 MHz, CDCl3, ppm): 6.66– 7.97 (m, 24H, ArAH), 4.13 (s, 4H, ArANH2). 13C NMR (500 MHz, CDCl3, ppm): 165.07 (2C, the ester carbonyl carbons), 113.85–151.37 (36C, the carbons of benzene ring), 64.68 (1C, quaternary carbon in the fluorene ring).

2.2.3. Fluorene diamine-based benzoxazine (BABPF-p) 9,9-Bis-[4-(4-aminobenzoyloxy)phenyl]fluorene (11.74 g, 0.02 mol), phenol (3.8 g, 0.04 mol), paraformaldehyde (3.6 g, 0.12 mol) and 50 mL mixed isomer xylenes were added to a 150 mL three neck round-bottomed flask equipped with a magnetic stirrer, reflux condenser, and thermometer. The mixture was stirred at 150 °C for 6 h. After that, the reaction mixture was cooled to room temperature and then poured into hexane. The precipitated yellowish powder were isolated by filtration and washed thoroughly with ethanol and dried under vacuum. The product was dissolved in dimethylformamide (DMF), precipitated in 1N aqueous solution of sodium hydroxide to remove any phenolic compounds, washed several times with water and finally with ethanol. The product was dried under vacuum at 60 °C for 24 h. Light yellow powder (63% yield) was provided. FTIR (KBr, cm1): 1726 (C@O stretching), 1380 (CH2 wagging), 1227 (CAOAC asymmetric stretching), 1139 (CANAC asymmetric stretching), 1067 (CAOAC symmetric stretching), 950 (CAH out-of-plane bending), 831, 747 (CAH out-of-plane bending). 1H NMR (500 MHz, CDCl3, ppm): 6.83-8.07 (m, 32H, ArAH), 5.42 (s, 4H, OACH2AN), 4.72 (s, 4H, ArACH2AN). 13C NMR (500 MHz, CDCl3, ppm): 164.77 (2C, the ester carbonyl carbons), 115.73-154.17 (48C, the carbons of benzene ring), 77.58 (2C, OACH2AN), 64.65 (1C, quaternary carbon in the fluorene ring), 49.58 (2C, NACH2AAr).

2.3. Curing of benzoxazine monomers BABPF-p was polymerized without initiator or catalyst according to the followings schedule: 180 °C/2 h, 200 °C/ 2 h, 220 °C/2 h and 240 °C/2 h in the air-circulating oven. 2.4. Characterization Fourier transform infrared (FTIR) spectra were recorded on a Perkin Elmer Spectrum 100 spectrometer in the range of 4000–650 cm1, which was equipped with a deuterated triglycine sulfate (DTGS) detector and KBr optics. Transmission spectra were obtained at a resolution of 4 cm1 after averaging two scans by casting a thin film on a KBr plate for monomers and cured samples. 1H and 13C NMR characterizations were performed on a Bruker AVANCE-500 NMR spectrometer using deuterated chloroform (CDCl3) as the solvent and tetramethylsilane (TMS) as an internal standard. The average number of transients for 1H and 13C NMR is 32 and 1024, respectively. A relaxation delay time of 1s was used for the integrated intensity determination of 1H NMR spectra. DSC measurements were evaluated on a TA Q200 differential scanning calorimeter under a constant flow of a nitrogen atmosphere of 50 mL/min. The instrument was calibrated with a high-purity indium standard, and a-Al2O3 was used as the reference material. About 5 mg of sample was weighed into a hermetic aluminum sample pan at 25 °C, which was then sealed, and the samples were scanned by the non-isothermal method from 20 to 350 °C using 5, 10, 15 and 20 °C/min heating rate under nitrogen. Thermogravimetric analysis (TGA) was performed on TA Instruments Q50 at a heating rate of 20 °C/min from 30 to 800 °C under nitrogen atmosphere at a flow rate of 50 mL/min. The dynamic mechanical thermal properties of the obtained polybenzoxazines were carried out with a TA Q800 dynamic mechanical analyzer. The rectangular samples (20 mm  5 mm  2 mm) were loaded in single cantilever mode at a temperature ramp of 3 °C/min from 30 to 300 °C with a frequency of 1 Hz under air atmosphere. 3. Results and discussion 3.1. Synthesis and structure of the benzoxazine monomers The diamine, 9,9-bis-[4-(4-aminobenzoyloxy)phenyl] fluorene, was synthesized in two steps. The first step was

2762

X.-y. He et al. / European Polymer Journal 49 (2013) 2759–2768

the esterification reaction between 9,9-bis-(4-hydroxyphenyl)fluorene and 4-nitrobenzoyl chloride. The second step was the catalytic hydrogenation of 9,9-bis-[4-(4nitrobenzoyloxy)phenyl]fluorene to the diamine. The IR spectrum of 9,9-bis-[4-(4-nitrobenzoyloxy)phenyl]fluorene showed characteristic absorptions for nitro groups at 1519 and 1345 cm1. The sharp band at 1730 cm1 presented in the spectra corresponded to the C@O stretching of the ester groups. After hydrogenation, the characteristic absorptions of nitro groups disappeared. New absorptions appeared at 3481 and 3376 (NAH stretching). Meanwhile, the very strong bands assigned to the stretching mode of C@O shifted to around 1699 cm1. Moreover, the structure of 9,9-bis-[4-(4-aminobenzoyloxy)phenyl] fluorene is further identified by NMR spectroscopy as shown in Fig. 1. These results clearly confirm that the diamine prepared herein is consistent with the proposed structure. The synthesis of BABPF-p was accomplished using the one-step procedure [19]. The formation of oligomers as a result of the benzoxazine polymerization was minimal. This ensures that the synthesized benzoxazine monomer has a high purity, which is recommended to facilitate the studies of polymerization behavior and curing kinetics. The 1H and 13C NMR spectra were recorded to confirm the structure (Fig. 1). In 1H NMR spectrum (Fig. 1A), the aromatic protons are observed at 6.83–8.07 ppm. The chemical shifts at 5.42 and 4.72 ppm are attributed to methylene (OACH2AN) and methylene (ArACH2AC) of oxazine ring, respectively. In 13C NMR spectrum (Fig. 1B), the peak at 164.77 ppm is assignable to the carbonyl carbon resonance. The resonances at 115.73–154.17 ppm are assigned to the aromatic carbon atoms. The chemical shifts at 49.58 and 77.58 ppm are ascribed to the carbon atom resonances of NACH2AAr and OACH2AN of oxazine rings, respectively. The signals located at 64.65 ppm are due to the presence of quaternary carbon atoms in the fluorene structure. The NMR results confirm the successful synthesis of BABPF-p. FTIR spectrum of BABPF-p monomer is shown in Fig. 2. The absorption band assigned to the out-of-plane bending vibrations of CAH located at 950 cm1 is due to presence of the characteristic mode of benzene with an attached oxazine ring. The absorptions located at 1227 cm1 and 1067 cm1 are corresponding to the asymmetric and symmetric stretching vibrations of CAOAC, respectively. The band at 1180 cm1 is attributed to the asymmetric stretching vibrations of CANAC. The absorption bands at 831 and 747 cm1 are attributed to the CAH out-of-plane bending mode of the para and ortho disubstituted benzene of aniline, aryl ester, and fluorene ring, respectively. The CH2 wagging in the oxazine ring is located at 1380 cm1. The strong peak centered at 1726 cm1 is attributed to the stretching vibration of the ester groups. Furthermore, the characteristic absorption bands of ortho-disubstituted benzene appear at 1490, 1455, and 750 cm1, respectively [14,40].

each cure stage are shown in Fig. 2. As can be seen from Fig. 2, with increasing the curing temperature, the characteristic absorption bands of the oxazine ring at 950 cm1, CH2 wagging at 1380 cm1, the asymmetric and symmetric stretching vibrations of CAOAC at 1227 and 1067 cm1 and CANAC at 1180 cm1 gradually decrease, respectively. At 240 °C, the characteristic peaks completely disappear, indicating the completion of ring-opening in this stage. Meanwhile, the very strong band assigned to the symmetric stretching modes of CANAC at around 915 cm1 appears [35]. These suggest that the crosslinking reaction further performs and the crosslink density of polybenzoxazine increases by the ring-opening polymerization at elevated temperature. The absorptions at 1490 and 1455 cm1 assigned to the ortho-disubstituted benzene ring of benzoxazine monomer shift to around 1475 cm1 due to the formation of tetrasubstituted benzene ring after thermal curing [40]. This result indicates that the crosslinking reaction of BABPF-p might also occur at para position on the phenol besides at ortho position during thermal curing. In the high-wave-number regions, the broad peaks at 3387 cm1 with a shoulder show that the overlapped intramolecular (AOH  N and AOH  O) and intermolecular (AOH  O) hydrogen bonds exist in the network structure of crosslinked polybenzoxazine, which can endow polybenzoxazine with amazing properties, such as high glass transition temperature, high modulus, and low moisture absorption [3,9,35,56].

3.2. Polymerization behavior of BABPF-p

kðTÞ ¼ A exp

The polymerization behavior of BABPF-p was investigated by FTIR. The obtained FTIR spectra of BABPF-p at

where A is the pre-exponential factor. Ea is the activation energy, which can be determined by Arrhenius law

3.3. Curing kinetics The curing kinetics analysis is based on the rate Eq. (1) [57]:

da da ¼b ¼ kðTÞf ðaÞ dt dT

ð1Þ

where k(T) is a temperature-dependent reaction rate constant, f(a) is the differential conversion function depending on the reaction mechanism, and b = dT/dt is a constant heating rate. The DSC curves were analyzed on the basis of the following assumption: the exothermic heat evolved during curing is proportional to the conversion of monomer to polymer. The degree of curing, a, is determined by the following equations:

a¼

DH i DHo

DH i ¼

Z 0

ð2Þ t

dH dt dt

ð3Þ

where DHi is the total amount of heat evolved by the reaction from the beginning to time t. DHo is the total heat of curing reaction. The temperature dependence on the rate constant is typically represented through the Arrhenius equation:

  Ea RT

ð4Þ

X.-y. He et al. / European Polymer Journal 49 (2013) 2759–2768

Fig. 1. 1H NMR (A) and

13

C NMR (B) spectra of BNBPF, BABPF and BABPF-p.

without assuming the kinetic model function. R is the universal gas constant, and T is the absolute temperature. Combining Eq. (1) with Eq. (4) it results:

  da da Ea f ðaÞ ¼b ¼ A exp dt dT RT

2763

ð5Þ

Non-isothermal method, more accurate to determine the curing kinetic parameters, is carried out at different heating rates. In addition, this method is very attractive be-

cause the kinetic data can be obtained in a relatively short period of time [49]. In this work, a multiple-heating-rate method for dynamic mode was used to evaluate the polymerization reaction kinetic parameters. Fig. 3 shows the variation of the degree of conversion (a) as a function of curing temperature obtained from dynamic DSC. All curves are in typical sigmoid shape, which indicates that the curing reaction of BABPF-p follows autocatalytic mechanism.

2764

X.-y. He et al. / European Polymer Journal 49 (2013) 2759–2768

the activation energy, the use of the integral isoconversional methods appears to be a safer alternative [59]. For nonisothermal conditions, the activation energy can be calculated by various integral isoconversional methods such as Ozawa, Kissinger, Starink methods. Compared to these methods, Straink method offers a significant improvement in the accuracy of the Ea values [57]. In this work, the average activation energy value obtained from Starink method was used for the determination of the reaction order of the BABPF-p. This method is based on Eq. (6) [62].

ln

Fig. 2. FT-IR spectra of BABPF-p at each cure stage.

Currently, the most commonly used kinetic method is a phenomenological model that uses various empirical model equations. On the basis of different fitting mechanisms, the cure kinetic methods can be divided into two types: the model fitting kinetic method and the model-free kinetic (MFK) method. The MFK method, unlike the model fitting method, avoids errors caused by improper models and parameters. This method can obtain more reliable activation energy values, Ea, without specifying a kinetic model [58]. The best known representatives of the model-free approach are the isoconversional methods [59]. These methods are based on the isoconversional principle which states that the reaction rate at a constant extent of conversion is only a function of the temperature [60,61]. They can generally be split in two categories: differential and integral methods. The differential isoconversional method presents the most straightforward way to evaluate the effective activation energy, Ea, as a function of the extent of reaction. This method can be applied to integral data (e.g. TG data) only after their numerical differentiation. Because this procedure may lead to erroneous estimation of

Fig. 3. Conversion a as a function of temperature for the BABPF-p in different heating rates.

b T 1:92 f

! ¼ Const  1:0008

Ea RT f

ð6Þ

where Tf is the temperature at an equivalent (fixed) state of transformation in different heating rate. From Eq. (6), a plot of ln(b/Tf1.92) vs. 1/Tf values at the same fractional extent of conversion from a series of dynamic DSC experiments at different heating rates would result in a straight line with a slope of 1.0008Ea/R. Repeating this procedure, the Ea values corresponding to different a from the DSC curing curves at different heating rates can be obtained. Fig. 4 represents the plots of ln(b/Tf1.92) vs. 1/Tf for various values of a (a = 0.05, 0.10, 0.15,. . ., 0.90, 0.95) covering the experimental range. Making fitted linear regression lines, and then the Ea values were obtained for each value of a. The plot of activation energy as a function of conversion is shown in Fig. 5. It can be clearly seen that the activation energy values tend to increase with the degree of conversion. And, the average value of the activation energy of BABPF-p is 125.6 kJ/mol. Currently, according to different cure mechanisms (i.e. reaction mechanism functions f(a)), the cure kinetic model can be divided into n-order model, the autocatalytic model, and the Kamal model (a superposition of n-order and autocatalytic model) [46,58,63]. For the n-order kinetics, the reaction rate is proportional to the concentration of the unreacted materials and has its maximum cure rate at t = 0. For the autocatalytic model, the impurities and the reaction products catalyze further reactions and the reaction rate increases with increasing time at the initial stage of the curing, undergoes a maximum and then decreases. Since there may be more than one chemical reaction

Fig. 4. Plots of lnðb=T 1:92 ) vs. 1/Tf. f

X.-y. He et al. / European Polymer Journal 49 (2013) 2759–2768

occurring during cure, the kinetic model may represent an overall process if these chemical reactions occur simultaneously [64]. These models can be expressed as Eqs. (7)– (9), respectively, where m and n are the reaction orders:

The n-order model

  da Ea ð1  aÞn ¼ A exp dt RT

The autocatalytic model

ð7Þ

  da Ea m ¼ A exp a ð1  aÞn dt RT ð8Þ

The Kamal model

da ¼ ðk1 þ k2 am Þð1  aÞn dt

ð9Þ

In Eq. (9), k1 and k2 are the temperature-dependent reaction rate constants given by Arrhenius relationship. k1 and k2 correspond to the non-catalytic and autocatalytic rate constant, respectively. Reaction order is used to indicate the reaction mechanism. Eq. (7) could be rewritten as follows:

ln

  da Ea þ ¼ n lnð1  aÞ þ ln A dt RT

ð10Þ

According to the experimental data and the average activation energy value obtained from Starink method, we can produce the plots of ln da/dt + Ea/RT vs. ln(1  a) as shown in Fig. 6. In Fig. 6, the peak points of the lines correspond to the peak point of the DSC curve. It is observed that there is a nonlinear increase before the peak points and a linear decrease after them. In the n-order model, the reaction rate is the highest at the beginning of curing, and it decreases as the reaction proceeds. For autocatalytic process, the plot would show a maximum of ln(1  a) approximately around 0.51 to 0.22, which is equivalent to degree of curing a of about 0.2–0.4. This is due to the autocatalytic nature that shows the maximum reaction rate at 20–40% conversion [53,54,65]. Since ln da/dt + Ea/ RT and ln(1  a) are not linearly related and evidently show a maximum in the range of the degree of conversion mentioned above, this suggests that the curing reaction is autocatalytic in nature. According to other works reported, the autocatalytic nature of the reaction kinetics of benzox-

Fig. 5. Variation of Ea vs. a by Starink method.

2765

azine can be explained by the generation of free phenol groups while the benzoxazine ring starts to open. These groups can actually accelerate further ring opening [47,48]. In all cases, the thermograms obtained at different heating rates were simultaneously correlated with the same set of kinetic parameters. In this way a single set of kinetic parameters is obtained to fit all the experimental data and no variation of these parameters at the heating rate is required. Theoretically, for the autocatalytic model, Eq. (8) could be solved by multiple nonlinear regressions. Because the cure rate was an exponential function of the reciprocal of the absolute temperature, it was difficult to get a good solution. The fitting results showed that large errors exist for the kinetic parameters obtained by such a method. By taking the logarithm of both sides of Eq. (8), a linear expression for the logarithm of cure rate can be obtained:

ln

    da da ¼ ln b dt dT   Ea þ m lnðaÞ þ n lnð1  aÞ ¼ ln A  RT

ð11Þ

Eq. (11) can be solved by multiple linear regression using the several different heating rates of data at the same time, in which the dependent variable is ln[b(da/dT)], and the independent variables are ln a, ln(1  a), and 1/T. Therefore, the values of A, m, and n can be obtained using the average activation energy from Starink method. The degree of curing is chosen to range between a = 0.05 and 0.95. For the Kamal model, a nonlinear regression method is used to obtain the kinetic constant and reaction orders. The parameters of the Kamal model are evaluated by using MATLAB. The results are listed in Table 1. The experimental results are compared with those predicted from two models, as shown in Fig. 7. It is clearly seen that the calculated data from the model are in good agreement with the experimental results. 3.4. Thermal and mechanical properties of polybenzoxazines Thermal properties of BABPF-p after polymerization were investigated in comparison to the typical fluorene

Fig. 6. The plots of ln da/dt + Ea/RT vs. ln(1a) for different heating rates.

2766

X.-y. He et al. / European Polymer Journal 49 (2013) 2759–2768

Table 1 Model fitting results by autocatalytic model and Kamal model. Parameters

Autocatalytic

Kamal

Ea (kJ/mol)

125.6

ln A

28.23

m n Correlation coefficients (R2)

0.79 1.26 0.991

Ea1 = 112.1 Ea2 = 140.4 ln A1 = 22.58 ln A1 = 31.59 1.03 1.37 0.995

Fig. 8. Temperature dependence curves of storage modulus and tan delta for poly(BABPF-p) and poly(BF-p).

Fig. 9. TGA thermograms of the cured poly(BABPF-p) and poly(BF-p) under nitrogen.

Fig. 7. Comparison of experimental values (symbols) and calculated values (lines) using autocatalytic model (A) and Kamal model (B).

diamine-based polybenzoxazine (poly(BF-p)) derived from 9,9-bis-(4-aminophenyl)fluorene/phenol bifunctional benzoxazine (BF-p) without aryl ester [40]. The Tg of poly(BABPF-p) was obtained by DSC and DMTA. Fig. 8 shows the DMTA thermogram of poly(BABPF-p) and poly(BF-p). We can see that the Tg value of poly(BABPF-p) reaches 243 °C by DSC and 265 °C by DMTA, respectively, and is 9 °C higher than that of poly(BF-p). It is much higher than that of the diaminodiphenylmethane-based benzoxazine (P-ddm) thermoset (200 °C by DMTA) and bisphenol fluorene-aniline-based polybenzoxazine (poly(B-pbf), 229 °C by DSC) [18,35]. This is mainly attributed to the backbone rigidity of polybenzoxazine molecule. Bulky fluorene units have

the high rigidity in the chain backbone, which restrain the internal rotations and thermal motion of polymer segments. In addition, the obtained high Tg value of poly(BABPF-p) can be attributed to abundant intermolecular and intramolecular hydrogen bonds, especially between ester carbonyl group (C@O) and phenolic hydroxyl group (AOH) of the poly(BABPF-p). Furthermore, the storage modulus of poly(BABPF-p) is about 1.7 GPa, which is much lower than that of poly(BF-p) (2.6 GPa). The result shows that the stiffness of polybenzoxazine observably decreases due to the introduction of flexible ester carbonyl linkages and the brittleness of fluorene-containing polybenzoxazine is greatly improved. Thermal stability for poly(BABPF-p) was evaluated by TGA under nitrogen atmosphere. As can be seen in Fig. 9, the temperatures corresponding to 5% and 10% weight loss (T5 and T10) are 380 and 399 °C, respectively, which are lower than that of poly(BF-p) (401 °C for T5, 432 °C for T10), but much higher than that of poly(B-pbf) (334 and 364 °C, respectively). The char yield (Yc) at 800 °C is 47%, which is the same as that of poly(BF-p), and slightly below 4% that of poly(B-pbf) [35]. The high char yield of poly(BABPF-p) can be ascribed to their high aromatic

X.-y. He et al. / European Polymer Journal 49 (2013) 2759–2768

content. This result indicates that the incorporation of flexible aryl ester linkages into macromolecular backbone has not significant influence on the thermal stability of fluorene-containing polybenzoxazine. 4. Conclusions The fluorene diamine-based benzoxazine with aryl ester linkages was successfully prepared and characterized. The polymerization behavior, mechanical and thermal properties were studied by FT-IR, DMTA, and TGA. With the introduction of bulky fluorenyl moieties, poly(BABPFp) exhibits both high glass transition temperature and thermal stability. Compare with poly(BF-p), the incorporation of flexible ester carbonyl linkages into fluorene-based polybenzoxazine has minor effect on the thermal stability, however, the brittleness of polybenzoxazine is greatly improved. The curing kinetics of BABPF-p was investigated using non-isothermal DSC methods. Autocatalytic model and Kamal model were used to describe the cure kinetic phenomena of BABPF-p and kinetic parameters were estimated. The theoretically calculated curves show a good agreement with the experimental data. Acknowledgements This work has been funded by the financial supports from National Natural Science Foundation of China (Project No. 50973022), Specialized Research Fund for the Doctoral Program of Higher Education (Project No. 20122304110019), Fundamental Research Funds for the Central Universities (Project Nos. HEUCFT1009, HEUCFR1227, and 201310006), and Natural Science Foundation of Heilongjiang Province (Project No. E200921). References [1] Ning X, Ishida H. Phenolic materials via ring-opening polymerization: synthesis and characterization of bisphenol-A based benzoxazines and their polymers. J Polym Sci Part A: Polym Chem 1994;32:1121–9. [2] Ning X, Ishida H. Phenolic materials via ring-opening polymerization of benzoxazines: effect of molecular structure on mechanical and dynamic mechanical properties. J Polym Sci Part B: Polym Phys 1994;32:921–7. [3] Ishida H, Allen DJ. Physical and mechanical characterization of nearzero shrinkage polybenzoxazines. J Polym Sci Part B: Polym Phys 1996;34:1019–30. [4] Laobuthee A, Chirachanchai S, Ishida H, Tashiro K. Asymmetric mono-oxazine: an inevitable product from Mannich reaction of benzoxazine dimers. J Am Chem Soc 2001;123:9947–55. [5] Ghosh NN, Kiskan B, Yagci Y. Polybenzoxazines – New high performance thermosetting resins: synthesis and properties. Prog Polym Sci 2007;32:1344–91. [6] Spontón M, Lligadas G, Ronda JC, Galià M, Cádiz V. Development of a DOPO-containing benzoxazine and its high-performance flame retardant copolybenzoxazines. Polym Degrad Stabil 2009;94:1693–9. [7] Agag T, Takeichi T. Novel benzoxazine monomers containing pphenyl propargyl ether: polymerization of monomers and properties of polybenzoxazines. Macromolecules 2001;34:7257–63. [8] Agag T, Takeichi T. Synthesis and characterization of novel benzoxazine monomers containing allyl groups and their high performance thermosets. Macromolecules 2003;36:6010–7. [9] Su YC, Chang FC. Synthesis and characterization of fluorinated polybenzoxazine material with low dielectric constant. Polymer 2003;44:7989–96.

2767

[10] Ishida H, Ohba S. Synthesis and characterization of maleimide and norbornene functionalized benzoxazines. Polymer 2005;46:5588–95. [11] Liu HC, Su WC, Liu YL. Self-assembled benzoxazine-bridged polysilsesquioxanes exhibiting ultralow-dielectric constants and yellow-light photoluminescent emission. J Mater Chem 2011;21:7182–7. [12] Kawaguchi AW, Sudo A, Endo T. Synthesis of highly polymerizable 1,3-benzoxazine assisted by phenyl thio ether and hydroxyl moieties. J Polym Sci Part A: Polym Chem 2012;50:1457–61. [13] Zúñiga C, Lligadas G, Ronda JC, Galià M, Cádiz V. Self-foaming diphenolic acid benzoxazine. Polymer 2012;53:3089–95. [14] Allen DJ, Ishida H. Effect of phenol substitution on the network structure and properties of linear aliphatic diamine-based benzoxazines. Polymer 2009;50:613–26. [15] Allen DJ, Ishida H. Polymerization of linear aliphatic diamine-based benzoxazine resins under inert and oxidative environments. Polymer 2007;48:6763–72. [16] Lin CH, Chang SL, Hsieh CW, Lee HH. Aromatic diamine-based benzoxazines and their high performance thermosets. Polymer 2008;49:1220–9. [17] Lin CH, Lin HT, Sie JW, Hwang KY, Tu AP. Facile, one-pot synthesis of aromatic diamine-based phosphinated benzoxazines and their flame-retardant thermosets. J Polym Sci Part A: Polym Chem 2010;48:4555–66. [18] Chang SL, Lin CH. Facile, one-pot synthesis of aromatic diaminebased benzoxazines and their advantages over diamines as epoxy hardeners. J Polym Sci Part A: Polym Chem 2010;48:2430–7. [19] Agag T, Jin L, Ishida H. A new synthetic approach for difficult benzoxazines: Preparation and polymerization of 4,40 diaminodiphenyl sulfone-based benzoxazine monomer. Polymer 2009;50:5940–4. [20] Spontón M, Larrechi MS, Ronda JC, Galià M, Cádiz V. Synthesis and study of the thermal crosslinking of bis(m-aminophenyl) methylphosphine oxide based benzoxazine. J Polym Sci Part A: Polym Chem 2008;46:7162–72. [21] Chernykh A, Agag T, Ishida H. Novel benzoxazine monomer containing diacetylene linkage: an approach to benzoxazine thermosets with low polymerization temperature without added initiators or catalysts. Polymer 2009;50:3153–7. [22] Men WW, Lu ZJ, Zhan ZR. Synthesis of a novel benzoxazine precursor containing phenol hydroxyl groups and its polymer. J Appl Polym Sci 2008;109:2219–23. [23] Lin CH, Chang SL, Lee HH, Chang HC, Hwang KY, Tu AP, et al. Fluorinated benzoxazines and the structure-property relationship of resulting polybenzoxazines. J Polym Sci Part A: Polym Chem 2008;46:4970–83. [24] Liu YL, Hsu CW, Chou CI. Silicon-containing benzoxazines and their polymers: copolymerization and copolymer properties. J Polym Sci Part A: Polym Chem 2007;45:1007–15. [25] Yang P, Gu Y. A novel benzimidazole moiety-containing benzoxazine: synthesis, polymerization, and thermal properties. J Polym Sci Part A: Polym Chem 2012;50:1261–71. [26] Rimdusit S, Tiptipakorn S, Jubsilp C, Takeichi T. Polybenzoxazine alloys and blends: some unique properties and applications. React Funct Polym 2013;73:369–80. [27] Fakis M, Zacharatos F, Gianneta V, Persephonis P, Giannetas V, Nassiopoulou A. Photoluminescence properties of porous silicon/ fluorene dye composites. Mater Sci Eng B 2009;165:252–5. [28] Lin SC, Pearce EM. Epoxy resins. II. The preparation, characterization, and curing of epoxy resins and their copolymers. J Polym Sci: Polym Chem Ed 1979;17:3095–119. [29] Liu W, Qiu Q, Wang J, Huo Z, Sun H. Curing kinetics and properties of epoxy resin–fluorenyl diamine systems. Polymer 2008;49:4399–405. [30] Reddy TA, Srinivasan M. Preparation and properties of cardopolyimides containing phenoxaphosphine units. J Polym Sci Part A: Polym Chem 1989;27:1419–24. [31] Hsiao SH, Li CT. Synthesis and characterization of new fluorenebased poly(ether imide)s. J Polym Sci Part A: Polym Chem 1999;37:1403–12. [32] Sagar AD, Shingte RD, Wadgaonkar PP, Salunkhe MM. Polyamides containing s-triazine rings and fluorene ‘‘cardo’’groups: synthesis and characterization. Eur Polym J 2001;37:1493–8. [33] Nishikawa M. Design of polyimides for liquid crystal alignment films. Polym Advan Technol 2000;11:404–12. [34] Srinivasan PR, Mahadevan V, Srinivasan M. Preparation and properties of some cardopolyamides. J Polym Sci: Polym Chem Ed 1981;19:2275–85.

2768

X.-y. He et al. / European Polymer Journal 49 (2013) 2759–2768

[35] Wang J, Wu MQ, Liu WB, Yang SW, Bai JW, Ding QQ, et al. Synthesis, curing behavior and thermal properties of fluorene containing benzoxazines. Eur Polym J 2010;46:1024–31. [36] Liu YL, Chang CY, Hsu CY, Tseng MC, Chou CI. Preparation, characterization, and properties of fluorene-containing benzoxazine and its corresponding cross-linked polymer. J Polym Sci Part A: Polym Chem 2010;48:4020–6. [37] Lu Y, Li M, Zhang Y, Hu D, Me L, Xu W. Synthesis and curing kinetics of benzoxazine containing fluorene and furan groups. Thermochim Acta 2011;515:32–7. [38] Fu Z, Liu H, Cai H, Liu X, Ying G, Xu K, et al. Synthesis, thermal polymerization, and properties of benzoxazine resins containing fluorenyl moiety. Polym Eng Sci 2012;52:2473–81. [39] Chang HC, Lin CH, Tian YW, Feng YR, Chan LH. Synthesis of 9,9-bis(4aminophenyl)fluorene-based benzoxazine and properties of its high-performance thermoset. J Polym Sci Part A: Polym Chem 2012;50:2201–10. [40] Wang J, He XY, Liu JT, Liu WB, Yang L. Investigation of the polymerization behavior and regioselectivity of fluorene diamineBased benzoxazines. Macromol Chem Phys 2013;214:617–28. [41] Tuzun A, Kiskan B, Alemdar N, Erciyes AT, Yagci Y. Benzoxazine containing polyester thermosets with improved adhesion and flexibility. J Polym Sci Part A: Polym Chem 2010;48:4279–84. [42] Baqar M, Agag T, Ishida H, Qutubuddin S. Poly(benzoxazine-courethane)s: a new concept for phenolic/urethane copolymers via one-pot method. Polymer 2011;52:307–17. [43] Yeganeh H, Jangi A. Thermally curable polyurethanes containing naphthoxazine groups in the main chain. Polym Int 2010;59:1375–83. [44] Kiskan B, Yagci Y, Ishida H. Synthesis, characterization, and properties of new thermally curable polyetheresters containing benzoxazine moieties in the main chain. J Polym Sci Part A: Polym Chem 2008;46:414–20. [45] Jubsilp C, Takeichi T, Rimdusit S. Property enhancement of polybenzoxazine modified with dianhydride. Polym Degrad Stabil 2011;96:1047–53. [46] Wang J, Fang X, Wu MQ, He XY, Liu WB, Shen XD. Synthesis, curing kinetics and thermal properties of bisphenol-AP-based benzoxazine. Eur Polym J 2011;47:2158–68. [47] Ishida H, Rodriguez Y. Curing kinetics of a new benzoxazine-based phenolic resin by differential scanning calorimetry. Polymer 1995;36:3151–8. [48] Ishida H, Rodriguez Y. Catalyzing the curing reaction of a new benzoxazine-based phenolic resin. J Appl Polym Sci 1995;58:1751–60. [49] Jubsilp C, Damrongsakkul S, Takeichi T, Rimdusit S. Curing kinetics of arylamine-based polyfunctional benzoxazine resins by dynamic differential scanning calorimetry. Thermochim Acta 2006;447:131–40.

[50] Liu Y, Yue Z, Gao J. Synthesis, characterization, and thermally activated polymerization behavior of bisphenol-S/aniline based benzoxazine. Polymer 2010;51:3722–9. [51] Andronescu C, Garea SA, Deleanu C, Iovu H. Characterization and curing kinetics of new benzoxazine monomer based on aromatic diamines. Thermochim Acta 2012;530:42–51. [52] Chow WS, Grishchuk S, Burkhart T, Karger-Kocsis J. Gelling and curing behaviors of benzoxazine/epoxy formulations containing 4,40 -thiodiphenol accelerator. Thermochim Acta 2012;543:172–7. [53] Ke L, Hu D, Lu Y, Feng S, Xie Y, Xu W. Copolymerization of maleimide-based benzoxazine with styrene and the curing kinetics of the resultant copolymer. Polym Degrad Stabil 2012;97:132–8. [54] Jubsilp C, Punson K, Takeichi T, Rimdusit S. Curing kinetics of benzoxazine–epoxy copolymer investigated by non-isothermal differential scanning calorimetry. Polym Degrad Stabil 2010;95:918–24. [55] Liu W, Wang J, Qiu Q, Ji L, Wang C, Zhang M. Synthesis and characterisation of 9, 9-bis (4-hydroxyphenyl)-fluorene catalysed by cation exchanger. Pigm Resin Technol 2008;37:9–15. [56] Kasapoglu F, Cianga I, Yagci Y, Takeichi T. Photoinitiated cationic polymerization of monofunctional benzoxazine. J Polym Sci Part A: Polym Chem 2003;41:3320–8. [57] Vyazovkin S, Burnham AK, Criado JM, Pérez-Maqueda LA, Popescu C, Sbirrazzuoli N. ICTAC kinetics committee recommendations for performing kinetic computations on thermal analysis data. Thermochim Acta 2011;520:1–19. [58] Zhang C, Binienda WK, Zeng L, Ye X, Chen S. Kinetic study of the novolac resin curing process using model fitting and model-free methods. Thermochim Acta 2011;523:63–9. [59] Vyazovkin S. Computational aspects of kinetic analysis: Part C. The ICTAC kinetics project — the light at the end of the tunnel? Thermochim Acta 2000;355:155–63. [60] Vyazovkin S, Sbirrazzuoli N. Isoconversional kinetic analysis of thermally stimulated processes in polymers. Macromol Rapid Commun 2006;27:1515–32. [61] Wang J, Wang H, Liu JT, Liu WB, Shen XD. Synthesis, curing kinetics and thermal properties of novel difunctional chiral and achiral benzoxazines with double chiral centers. J Therm Anal Calorim 2013. http://dx.doi.org/10.1007/s10973-013-3081-8. [62] Starink MJ. The determination of activation energy from linear heating rate experiments: a comparison of the accuracy of isoconversion methods. Thermochim Acta 2003;404:163–76. [63] Kamal MR, Sourour S. Kinetics and thermal characterization of thermoset cure. Polym Eng Sci 1973;13:59–64. [64] Boey FYC, Qiang W. Experimental modeling of the cure kinetics of an epoxy-hexaanhydro-4-methylphthalicanhydride (MHHPA) system. Polymer 2000;41:2081–94. [65] Hsieh TH, Su AC. Cure kinetics of an epoxy-novolac molding compound. J Appl Polym Sci 1990;41:1271–80.