Renewable benzoxazine monomers from ... - Wiley Online Library

ARTICLE

WWW.POLYMERCHEMISTRY.ORG

JOURNAL OF POLYMER SCIENCE

Renewable Benzoxazine Monomers from “Lignin-like” Naturally Occurring Phenolic Derivatives , Virginia Ca diz Marc Comı, Gerard Lligadas, Juan C. Ronda, Marina Galia nica, Universitat Rovira i Virgili, Campus Sescelades, Marcel.lı Domingo s/n, Departament de Quımica Analıtica i Quımica Orga Tarragona 43007, Spain Correspondence to: J. C. Ronda (E - mail: [email protected]) Received 19 June 2013; accepted 15 August 2013; published online 16 September 2013 DOI: 10.1002/pola.26918

ABSTRACT: The benzoxazines of three naturally occurring phenylpropanoid phenols: ferulic, coumaric, and phloretic acids, and their esters are described. Benzoxazines with conjugated unsaturated chains exhibit unusual poor thermal stability and degrade partially at the polymerization temperature making necessary the use of a catalyst (BF3.Et2O) to low the polymerization temperature and prevent degradation. Polybenzoxazines are prepared thermally and characterized by DSC and TGA techniques.

The resulting materials have superior Tgs when compared with those prepared from an unsubstituted monofuctional benzoxazine due to the additional crosslinking through the ester and carC 2013 Wiley Periodicals, Inc. J. Polym. Sci., boxylic moieties. V Part A: Polym. Chem. 2013, 51, 4894–4903

INTRODUCTION Polybenzoxazines are a relatively new class of phenolic type thermosets developed to overcome most of the short-comings associated with the traditional phenolic resins. During the last decade, polybenzoxazines have attracted significant attention of both industry and research community because of their unique advantages over most of the known polymers,1 becoming one of a rare few new polymers commercialized in the past 30 years.

Cardanol is a monofunctional phenol present in the cashew nut shell oil and is obtained as a waste byproduct by the cashew industry. Cardanol-derived polybenzoxazines have technological potential as matrix for natural fiber-reinforced composites6 and reactive diluents in the synthesis process of bisphenol A (BPA) based benzoxazines.7 In a more recent work, the synthesis of fully renewable benzoxazines from guaiacol and furfuryl or stearylamine has been also described.8

The chemistry of benzoxazines is responsible for a number of inherent processing advantages, including low melt viscosity, no harsh catalyst required, and no volatiles release and minimal shrinkage during cure. The resulting materials have also remarkable properties such as heat resistance, superior electronic performance, low water absorption, low surface energy, and excellent dimensional stability.2 Clasically, benzoxazine monomers are synthesized either in solution or by a melt-state reaction using a combination of petroleum-based phenolic derivatives, formaldehyde, and primary amines. Nowadays, the scarcity of nonrenewable resources is encouraging the scientific community to develop and commercialize new biobased products that can alleviate the wide-spread dependence on fossil fuels and, enhance security, the environment, and the economy.3 Until now, only few examples of renewable resources-based benzoxazine materials have been reported into the literature. The most significant example is the cardanol-derived benzoxazine.4,5 C 2013 Wiley Periodicals, Inc. V

4894

JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2013, 51, 4894–4903

KEYWORDS: cationic polymerization; crosslinking; polybenzoxazines; renewable resources; thermal properties; thermosets

Recently, we have described the synthesis of new benzoxazine monomers derived from diphenolic acid (DPA).9 DPA is a condensation derivative of levulinic acid which is a cheap platform chemical produced commercially from cellulose rich biomass, especially waste biomass.10 The additional carboxylic acid functionality of these monomers makes possible additional technological applications like the preparation of low curing temperature impregnating resins and selffoaming thermostable rigid foams.11 Nature offers a rich variety of readily available phenolic and polyphenolic compounds that have not yet considered for the synthesis of monomers for polymer synthesis.12 Naturally occurring phenols are among the most widespread class of metabolites in vegetal organisms, and their distribution is almost ubiquitous.13 Most of these compounds arise from phenylalanine and tyrosine which after deamination feeds the phenylpropanoid way to give a large variety of



ARTICLE

SCHEME 1 Structure of CA, FA and PA and synthesis of the benzoxazine derivatives. (1) MeOH/CH(OCH3)3/p-CH3C6H4SO3H. (2) 1,3,5-(C6H5)3C3N3/(CH2O)n, 120–130 C. (3) 1,3,5-(C6H5)3C3N3/(CH2O)n, toluene, 110 C.

compounds which share the C6-C3 phenyl propanoid skeleton and contain one, two or more hydroxylic groups: benzoic acids, cinnamic acids, flavonoids, proanthocyanidins, coumarins, stilbenes, lignans, and lignins. Even they are present in many different plants,14 most of these compounds are produced at commercial scale in their pure form processing lignocellulosic bagasse. Lignin is a fundamental component of plants accounting for about 25–35% of the organic matrix of wood and comprises roughly 20% of the total mass of the planet biosphere. Lignin is a phenolic high molecular mass biopolymer composed of a highly branched phenylpropanoid framework based on p-coumaryl, coniferyl, and sinapyl moieties, which has long been recognized as a potential feedstock for producing chemicals.12,15 In this work, we explore trans-3-(4-hydroxyphenyl)propenoic acid (p-coumaric acid, CA), trans-3-(3-methoxy-4-hydroxyphenyl)propenoic acid (ferulic acid, FA), and 3-(4-hydroxyphenyl)propanoic acid (phloretic acid, PA) and their methyl ester derivatives as platform chemicals for producing valuable benzoxazine monomers and polymers (Scheme 1).

WWW.MATERIALSVIEWS.COM

CA and FA are present in relatively large quantities in vegetables like peanuts, tomatoes, fennel, coffee, or artichokes but, as main components of lignocellulose, they are commercially produced mainly by chemical or enzymatic transformation of corn, sugarcane, and other agricultural bagasses.16–19 PA can be obtained by hydrogenation of p-coumaric acid but it is commonly produced by chemical or enzymatic treatment of phloretin, a dihydrochalcone present in the apple tree leaves.

EXPERIMENTAL

Materials The following chemicals were obtained from the sources indicated and used as received: p-ferulic acid (Aldrich), p-coumaric acid (Aldrich), phloretic acid (Aldrich) paraformaldehyde (Mallinkrodt), trimethyl orthoformate (Fluka), p-toluenesulfonic acid monohydrate (Aldrich). 1,3,5-Triphenylhexahydro1,3,5 triazine and 3-phenyl-3,4-dihydro-2H-1,3-benzoxazine were synthesized according to a reported procedure.20 All solvents were purified by standard procedures.


4895

ARTICLE


Synthesis of 3-(3-Phenyl-3,4-dihydro-2H-1,3-benzoxazin6-yl)propenoic acid (CA-Bz), 3-[8-Methoxy-(3-phenyl-3,4dihydro-2H-1,3-benzoxazin-6-yl)] propenoic acid (FA-Bz), and 3-(3-Phenyl-3,4-dihydro-2H-1,3-benzoxazin-6yl)propanoic acid (PA-Bz; Scheme 1) 1,3,5-Triphenylhexahydro-1,3,5-triazine (0.04 mol), paraformaldehyde (0.12 mol), p-coumaric, ferulic, or phloretic acid (0.12 mol) and 25 mL of toluene were placed into 50 mL two-necked round-bottom flask equipped with a condenser and a teflon coated stirring bar. The reaction mixture was gently stirred and heated at 110 C for 16 h. The resulting light yellow to orange solution was filtered and concentrated in the rotary evaporator obtaining syrup that was dried under high vacuum for 48 h. The obtained solid mass was finely powdered, washed several times with diethyl ether, and finally dried again under vacuum. 3-(3-Phenyl-3,4-dihydro-2H-1,3-benzoxazin-6yl)propenoic acid (CA-Bz) Yield. 90%, clear yellow powder M.p.: 119–120 C. FTIR (cm21): 3463–2176 (r COOH), 1679 (r C@O), 1628 (r C@C), 1236 (r CAN), 1210, 1029 (rs CAOAC), 941(NACH2AO Bz ring). 1

H NMR (CDCl3, tetramethylsilane (TMS) d ppm): 7.98 (1H, d, J 15.6 Hz, ACH@CHACOOA), 7.63–7.12 (8H, ArAH), 6.58 (1H, d, J 15.6 Hz, ACH@CHAAr), 5,71 (2H, s, OACH2AN), 4.96 (2H, s, ArACH2AN). 13

C NMR (CDCl3, d ppm): 172.5 (s, C@O), 157.1 (s), 148.4 (s), 147.0 (s), 129,7 (d), 128.7 (d), 127.7 (d), 127.2 (d), 122.3 (s), 121.6 (s), 118.9 (d), 118.0 (d), 115.1 (d), 80.4 (t, OACH2AN), and 50.8 (t, ArACH2AN). 3-[8-Methoxy-(3-phenyl-3,4-dihydro-2H-1,3-benzoxazin6-yl)]propenoic acid (FA-Bz) Yield. 92%. White powder. M.p.: 120–121 C. FTIR (cm21): 3578–2185 (r COOH), 1680 (r C@O), 1625 (r C@C), 1253 (r CAN), 1200, 1082 (rs CAOAC), 929 (NACH2AO Bz ring). 1

H RMN (CDCl3, TMS, d ppm): 7.65 (1H, d, J 16.0 Hz, CH@CHACOO), 7.27–6.84 (7H, ArAH), 6.28 (1H, d, J 16.0 Hz, CH@CHAAr), 5,47 (2H, s, OACH2AN), 4.64 (2H, s, ArACH2AN), and 3.88 (3H, s, COOCH3).

13 C RMN (CDCl3, d ppm): 172.2 (s, C@O), 148.5 (s), 147.9 (s), 146.9 (s), 146.4 (s), 129.3 (d), 126.4 (d), 121.9 (d), 121.5 (s), 120.4 (s), 118.5 (d), 115.1 (d), 108.5 (d), 80.4 (t, OACH2AN), 55.9 (q, ArAOCH3), and 50.2 (t, ArACH2AN).

179.0 (s, C@O), 153.0 (s), 148.5 (s), 132.7 (d), 129.5 (d), 127.9 (d), 126.6 (s), 121.6 (s), 121.0 (d), 118.4 (d), 117.2 (d), 79.6 (t, OACH2AN), 50.6 (t, ArACH2AN), 36.0 (t, CH2AAr), and 30.0 (t, CH2ACO). Synthesis of Methyl 3-(4-hydroxyphenyl)propenoate (CAM), Methyl 3-(3-methoxy-4-hydroxyphenyl) propenoate (FAM), and Methyl 3-(4-hydroxyphenyl) propanoate (PAM; Scheme 1) A 250 mL round-bottom flask equipped with a condenser was charged with coumaric, phloretic, or ferulic acid (0.42 mol), p-toluenesulfonic acid monohydrate (4 3 1024 mol), trimethyl orthoformate (0.21 mol) and 100 mL of methanol. The mixture was heated under reflux 24 h and the methanol was removed under vacuum. The resulting oil was dissolved in ethyl ether and washed several times with a saturated NaHCO3 solution and with water. The organic layer was dried with anhydrous MgSO4, the solvent was removed in the rotary evaporator and the product was dried under vacuum at room temperature for 12 h. The resulting colorless oils, which crystallize after several days, were used without further purification in the next step. Yields of 98% were obtained in all cases. Methyl Coumarate (CAM) Yield. 98%. White crystalls M.p.: 130–132 C. ESI-TOF (M 1 Na)1: 201.0532. FTIR: (cm21): 3510–3058 (r OAH), 1682 (r C@O), 1631 (r C@C), 1431 (ra CAOAC), 1195– 1103 (rs CAOAC). 1

H NMR (DMSO d6, TMS, d ppm): 10.00 (1H, s, ArAOH), 7.56 (1H, d, J 13.9 Hz, CH@CHACOO), 7.54–6.77 (7H, ArAH), 6.38 (1H, d, J 13.9 Hz, CH@CHAAr), 3.68 (3H, s, COOCH3). 13C NMR (DMSO d6, d ppm): 167.1 (s, C@O), 159.0 (s),144.8 (s), 130.3 (d), 125.1 (d), 115.8 (d), 113.9 (d), and 51.2 (q, COOCH3). Methyl Ferulate (FAM) Yield. 98%. White crystalls M.p.: 62–64 C. ESI-TOF (M 1 Na)1: 231.0631. FTIR: (cm21): 3572–3103 (r OAH), 1697 (r C@O), 1671 (r C@C), 1512, 1431 (ra CAOAC), 1250–1159 (rs CAOAC). 1

H NMR (CDCl3, TMS, d ppm): 7.61 (1H, d, J 16.0 Hz, CH@CHACOOA), 7.26–6.90 (7H, m, ArAH), 6.28 (1H, d, J 16.0 Hz, ACH@CHAAr), 6.09 (1H, s, ArAOH), 3.92 (3H, s, ACOOCH3), and 3.79 (3H, s, ArOCH3).

13

C NMR (CDCl3, d ppm): 167.9 (s, C@O), 148.2 (s), 146.9 (s), 145.2 (d), 127.0 (d), 123.2 (d), 115.2 (d), 114.9 (d), 109.6 (d), 56.1 (q, COOCH3), 51.2 (q, ArAOCH3).

3-(3-Phenyl-3,4-dihydro-2H-1,3-benzoxazin-6yl)propanoic acid (PA-Bz) Yield. 95%. Yellow powder M.p.: 121–122 C. FTIR (cm21): 3574–2158 (r COOH), 1692 (r C@O), 1597 (r C@C), 1498 (ra CAOAC), 1252 (r CAN), 1039 (rs CAOAC), and 921 (NACH2AO Bz ring).

Methyl Phloretate (PAM) Yield 90% colourless liquid. ESI-TOF (M 1 Na)1: 203.0680. FTIR (cm21) 3572–3080 (r OAH), 1708 (r C@O), 1438 (ra CAOAC), and 1261–1148 (rs CAOAC).

1

1

H NMR (CDCl3, TMS, d ppm): 7.28–6.73 (8H, ArAH), 5.33 (2H, s, OACH2AN), 4.60 (2H, s, ArACH2AN, 2.85 (2H, t, CH2ACOO), 2.62 (2H, t, CH2AAr). 13C NMR (CDCl3, d ppm):

4896



H NMR (CDCl3, TMS, d ppm): 7.02–6.74 (4H, ArAH), 3.66 (3H, s, COOCH3), 2.86 (2H, t, CH2ACOO), and 2.60 (2H, t, ACH2AAr).



ARTICLE

13 C NMR (CDCl3, d ppm): 174.3 (s, C@O), 154.3 (s), 131.9 (s), 129.2 (d), 115.3 (d), 51.77 (q, COOCH3), 35.9 (t, ACH2AAr), and 29.9 (t, ACH2ACO).

H NMR (CDCl3, TMS, d ppm): 7.19–6.65 (8H, ArAH), 5.26 (2H, s, OACH2AN), 4.53 (2H, s, ArACH2AN), 3.59 (3H, s, COOCH3), 2.70 (2H, t, ACH2ACOO), 2.50 (2H, t, CH2AAr).

Synthesis of Methyl 3-(3-phenyl-3,4-dihydro-2H-1,3benzoxazin-6-yl) propenoate (CAM-Bz), Methyl-3-[8methoxy-(3-phenyl-3,4-dihydro-2H-1,3-benzoxazin-6-yl)] propenoate (FAM-Bz), and 3-(3-Phenyl-3,4-dihydro-2H1,3-benzoxazin-6-yl) propanoate (PAM-Bz; Scheme 1) A 100 mL round-bottom flask, was charged with 1,3,5-triphenylhexahydro-1,3,5-triazine (0.04 mol), paraformaldehyde (0.12 mol), and (0.04 mol) of methyl coumarate, methyl phloretate, or methyl ferulate. The reaction mixture was heated for 2 h at 130 C in the case of methyl coumarate, 4 h at 120 C in the case of methyl ferulate and methyl phloretate. The resulting light yellow solid was dissolved in ethyl ether, filtered, and washed several times with 20% NaOH. The organic layer was dried with anhydrous MgSO4 and concentrated under vacuum to obtain syrup that was subsequently dried under high vacuum. The resulting solid was powdered and recrystallized using a heptanes/1,2-dichloroethane 7:3 mixture.

13

3-(3-Phenyl-3,4-dihydro-2H-1,3-benzoxazin-6-yl) propenoate (CAM-Bz) Yield. 91%. White powder M.p.: 110–111 C. FTIR (cm21): 1687 (r C@O), 1641 (r C@C), 1232 (r CAN), 1035 (rs CAOAC), and 968 (NACH2AO Bz ring). 1

H NMR (CDCl3, TMS, d ppm): 7.41 (1H, d, J 13.9 Hz, ACH@CHACOO), 7.14–6.77 (7H, ArAH), 6.62 (1H, d, J 13.9 Hz ACH@CHAAr), 5,22 (2H, s, OACH2AN), 4.46 (2H, s, ArACH2AN), and 3.61 (3H, s, COOCH3). 13

C NMR (CDCl3, d ppm): 168.1 (s, C@O), 156.4.2 (s), 148.2 (s), 144.6 (s), 129.6 (d), 128.1 (d), 127.7 (d), 127.2 (s), 122.1 (s), 121.5 (s), 118.7 (d), 117.8 (d), 115.7 (d), 80.2 (t, OACH2AN), 51.8 (q, AOCH3), 50.7 (t, ArACH2AN).

Methyl 3-[8-methoxy-(3-phenyl-3,4-dihydro-2H-1,3benzoxazin-6-yl)] propenoate (FAM-Bz) Yield. 93%. Yellow powder M.p.: 166–167 C. IR (cm21): 1692 (r C@O), 1632 (r C@C), 1233 (r CAN), 1033 (rs CAOAC), and 941 (NACH2AOA Bz ring). 1

H NMR (CDCl3, TMS, d ppm): 7.56 (1H, d, J 16.0 Hz, ACH@CHACOO), 7.26–6.82 (7H, ArAH), 6.28 (1H, d, J 16.0 Hz, ACH@CHAAr), 5,47 (2H, s, OACH2AN), 4.64 (2H, s, ArACH2AN), 3.88 (3H, s, ACOOCH3), and 3.79 (3H, s, ArAOCH3). 13

C NMR (CDCl3, d ppm): 167.8 (s, C@O), 148.6 (s), 148.2 (s), 146.2 (s), 144.9 (s), 129.5 (d), 126.9 (d), 122.1 (d), 121.6 (s), 120.1 (s), 118.7 (d), 115.8 (d), 108.5 (d), 80.5 (t, OACH2AN), 56.1 (q, AOCH3), 51.8 (q, ArAOCH3), and 50.2 (t, ArACH2AN).

3-(3-Phenyl-3,4-dihydro-2H-1,3-benzoxazin-6-yl) propanoate (PAM-Bz) Yield. 97% White solid Mp: 156–157 C. FTIR (cm21): 1728 (r C@O), 1226 (r CAN), 1023 (rs CAOAC), and 943 (NACH2AOA Bz ring). WWW.MATERIALSVIEWS.COM

1

C NMR (CDCl3, d ppm): 173.7 (s, C@O), 154.8 (s), 148.5 (s), 132.9 (d), 129.4 (d), 127.8 (d), 126.6 (s), 121.5 (s), 121.3 (d), 118.3 (d), 117.0 (d), 79.5 (t, OACH2AN), 51.8 (t, ArACH2AN), 50.5 (a, OCH3), 36.2 (t, CH2AAr), and 30.2 (t, CH2ACO). Curing Curing was performed in teflon coated molds under pressure following a preset temperature cycle. Samples of 50 3 8 3 3 mm3 were prepared starting from finely powdered monomer degassed at 60 C under vacuum for 1 h. Catalyzed samples were prepared adding 3% (w/w) of BF3.Et2O to a dichloromethane solution of the monomer and evaporated to dryness prior to the degassing step. Instrumentation 1 H NMR (400 MHz) and 13C NMR (100.6 MHz) spectra were recorded using a Varian Gemini 400 spectrometer. Chemical shifts were reported in ppm relative to TMS or CHCl3 as internal standards. FTIR spectra were recorded on a Bomem Michelson MB 100 FTIR spectrophotometer with a resolution of 4 cm21 in the absorbance mode. An attenuated total reflection (ATR) device with thermal control and a diamond crystal (Golden Gate heated single-reflection diamond ATR, Specac-Tknokroma) was used. Calorimetric studies were carried out on a Mettler DSC821 differential scanning calorimeter using N2 as a purge gas (20 mL/min). Thermal stability studies were carried out on a Mettler TGA/SDTA851e/LF/1100 with N2 as purge gas. The studies were performed in the 30–800 C temperature range at a scan rate of 10 C/min. Degradation studies were carried out in a Carbolite TZF 12/38/400 oven connected to a condenser cooled by liquid nitrogen. GC-MS measurements were carried out using an HP 6890 gas chromatograph with an Ultra 2 capillary column (crosslinked 5% PH ME siloxane) and an HP 5973 mass detector. RESULTS AND DISCUSSION

As commented in the introduction, the main aim of this work is to explore the synthesis of new benzoxazine monomers and polymers derived from some 4-hydroxycynamic acid derivatives of renewable origin: CA, FA, and PA. These building blocks share a 3-phenylpropanoic acid skeleton with a phenolic group in the para position and a double bond in trans configuration in the case of CA and FA. One of the main drawbacks in the synthesis of benzoxazine monomers is the extensive formation of oligomers, which reduces the yield of benzoxazine and makes the purification difficult. This undesired side-reaction has been reported to be favoured mostly by water, polar solvents and high temperatures.21–25 One of the synthetic approaches described to


4897

ARTICLE



ylether the final product. By this procedure, benzoxazine monomers CA-Bz, FA-Bz, and PA-Bz were obtained in high yield. (Scheme 1). In the case of methyl ester derivatives, the lower melting point of the CAM, FAM, and PAM allowed to carry out the reaction in bulk. The reaction was complete in 3 h working a 110 C in the case of FAM-Bz and PAM-Bz but required higher temperatures and longer times in the case of CAM-Bz due to the higher melting point of CAM. Oligomeric traces were removed washing with 10% NaOH and the crude monomers were purified by recrystallization to yield essentially pure CAM-Bz and FAM-BZ. The formation of the benzoxazine ring is detected by the FTIR characteristic band at about 945 cm21 due to the out-of-plane CAH vibration of the benzene ring to which the oxazine ring is attached and by the 1H NMR methylene singlet signals at c.a. 5.2 ppm (OACH2AN) and at about 4.5 ppm (ArACH2AN). The structure of all benzoxazine monomers was confirmed by FTIRATR and 1H and 13C NMR spectroscopy. As example, Figure 1 shows the 1H NMR spectra of some benzoxazines.

FIGURE 1 1H NMR spectra of (a) PA-Bz, (b) CAM-Bz, and (c) CA-Bz.

minimize the formation of oligomers consist of the use of 1,3,5-triphenylhexahydro-1,3,5-triazine and paraformaldehyde as reagents in bulk conditions or with the help of solvents like toluene or xylenes.20,26 The high melting point of CA, FA, and FA, prevent the synthesis of the corresponding benzoxazines in bulk conditions and toluene had to be added as solvent. The reactions required 16h to be complete. Attempts to shorten the reaction time by increasing the temperature using xylenes, gave products with high oligomeric content. Usually, phenolic oligomers are eliminated by washing with 2M sodium hydroxide but in this case the presence of acid functionalities prevents this treatment. Purification was done by filtering solid impurities and rinsing with dieth-

Thermal behavior and stability were characterized by DSC and TGA (Table 1). By DSC all benzoxazine monomers show a melting endotherm (with low associated enthalpy) in most cases followed by a relatively narrow exotherm at higher temperatures. Usually, this exothermic peak is attributed to the thermal benzoxazine ring opening polymerization. However, for all unsaturated monomers TGA data clearly indicates that this process is overlapped with thermal decomposition of the monomer, which involves weight losses around 20–25% of the initial mass. A different behaviour can be inferred for PA-Bz and PAM-Bz, which have a saturated C3 substituent. For these monomers, the weight loss is negligible and the exothermic process can consequently be attributed to the benzoxazine polymerization. This behavior is represented in Figure 2 in the case of CA and PA benzoxazines. Most benzoxazine monomers reported in the literature are stable at the polymerization temperature 150–250 C allowing preparing well defined thermosetting resins. At higher

TABLE 1 DSC and TGA Characterization of the Benzoxazine Monomers Mon.

Tm ( C)a

Texoo ( C)b

Texom ( C)c

Tdo ( C)d

Tdm ( C)e

Wexp (%)f

Wtheo.(%)g

CA-Bz

117

131

137

137

169

14,5

44.8

FA-Bz

121

132

130

130

168

19.2

40.5

PA-Bz

122

131

212

212

263

2.8

44.5

CAM-Bz

109

213

216

216

252

26.5

42.6

FAM-Bz

166

224

203

203

246

23.5

38.7

PAM-Bz

156

180

197

197

278

3.8

42.3

a

f

b

g

Maximum temperature of the melting endotherm by DSC. Onset temperature of the exotherm by DSC. c Maximum temperature of the exotherm by DSC. d Onset temperature of the decomposition by TGA. e Temperature of the maximum decomposition rate for the first decomposition step by TGA.

4898


Percentage of weight loss after the first decomposition step by TGA. Theoretical weight percent loss, supposing the elimination of one molecule of aniline and one molecule of formaldehyde.



ARTICLE

ester, and methoxy groups were observed in all cases, indicating that these moieties remain unaffected at these temperatures. The spectra of the polymeric residues show the characteristic signals of the aniline ring, indicating that a part of the aniline groups remain in the polymer and are not volatilized. This can be confirmed by the TGA results collected in Table 1. As can be seen, the percentage of weight loss after the first degradation step is always lower than the theoretical expected for the loss of one molecule of aniline and one molecule of formaldehyde. Aniline has been described as one of the initial decomposition products in aniline-derived benzoxazine resins, which are detected at temperatures above 270 C.31 However, this compound is detected at temperatures as low as 130–150 C in the case of the carboxylic acid benzoxazines CA-Bz and FA-Bz indicating a lower thermal stability of these monomers. These results suggest that the presence of a double bond conjugated to the benzoxazine phenolic ring causes unstabilization to benzoxazine ring. FIGURE 2 DSC and TGA plots for CA-Bz and PA-Bz monomers. (Melting of CA-Bz is not detected in this DSC plot). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

temperatures polybenzoxazine resins undergo well known thermal degradation processes, which have been the matter of several studies.27–31 The unsaturated monomers studied in this work show an unusual low thermal stability, which must be related with some structural characteristics. Precedent studies show an important influence of the Mannich base and the phenolic structure in the thermal degradation of polybenzoxazines, especially in the initial degradation steps.27 To gain more insight about these degradation processes, an analysis of the volatile components at the maximum temperature of the degradation exotherm was carried out by pyrolysis-GC-MS. Unsaturated benzoxazine monomers produced few volatile products, collected as a white solid and yellowish oil, that were identified, respectively, as paraformaldehyde and aniline. Also traces of phenol and methylphenols were detected. This seems to indicate that thermal decomposition of these monomers proceeds mainly through the ring-opening and subsequent fragmentation of the benzoxazine ring. The orange-red pyrolysis residue was partially soluble in DMSOd6 and CDCl3 especially in the case of the ferulic acidderived benzoxazines. The 1H NMR spectra of this soluble fractions present broad signals at 4.4–4.0, 3.8–3.4, and 2.2– 2.0 ppm. Polymerization of benzoxazines have been reported to proceed through the formation of N,O-ketalic bridges (methylene signals at 4.25 ppm) that rearrange to Mannich bridges (methylene signals at 3.80 ppm) at high temperature.32 Thus, the structure of the pyrolisis residues seems to correspond to oligomeric fractions with both Mannich and N,O-ketalic bridges. In addition the peak 2.1 ppm can be attributed to the formation of methylphenolic groups.20 Moreover, signals corresponding to double bond, methyl


Taking into account, the above evidences and the generally accepted benzoxazine polymerization mechanisms,33,34 we propose a possible degradation mechanism which is consistent with the formation of the detected products from the free and polymerized monomer (Scheme 2). According to the results collected in Table 1, only monomers PA-Bz and PAM-Bz can be thermally polymerized without significant decomposition. For the rest of the monomers the polymerization proceeds with partial decomposition preventing the preparation of useful materials. It must be pointed out that, as expected, the temperature of the polymerization and degradation exotherm shows a clear influence of the benzoxazine structure being almost 100 C lower for the monomers with a free carboxylic acid group due to its catalytic effect on the benzoxazine ring opening.35–37 It is well known that, apart from protic acids several other substances catalyze the benzoxazine ring-opening lowering considerably the polymerization temperature.33,38,39 As boron trifluoride complexes have been shown as effective catalysts for the benzoxazine curing,36,40 we used BF3.Et2O as catalyst in an attempt to polymerize the unsaturated benzoxazine monomers at lower temperatures than the decomposition temperature. In this way, all monomers were heated in presence of 1, 2, and 3 mol % of BF3.Et2O and the polymerization behaviour was followed by DSC. The results, collected in Table 2, indicate that catalyst activates the polymerization reducing the reaction temperature for all monomers. According to TGA data, in the case of CA-Bz, CAM-Bz, FA-Bz, and FAM-Bz, the polymerization proceeds without significant degradation when using 2 or 3% of BF3.Et2O. The most relevant case is FAM-Bz, which is shown in Figure 3. Without catalyst, this monomer starts to polymerize at about 210 C with extensive degradation but in presence of 1% catalyst a double exotherm starting just after the monomer


4899

ARTICLE



SCHEME 2 Proposed mechanism for the thermal degradation of the unsaturated benzoxazine monomers and polymers.

melting at about 160 C can be observed. As the amount of catalyst increase, the enthalpy of the high temperature exotherm (associated to the degradation processes) progressively decreases and the peak completely disappears when using 3% BF3.Et2O. The exotherm at low temperature only can be observed in presence of catalyst and increases its enthalpy with the amount of catalyst and is the only peak observed when using 3% BF3.Et2O. TGA analysis indicates that in presence of catalyst the polymerization is favoured

and the decomposition phenomena are delayed or completely suppressed. According to the above results, PA-Bz and PAM-Bz were polymerized without catalyst and all the unsaturated monomers CA-Bz, CAM-Bz, FA-Bz, and FAM-Bz were polymerized in presence of 3% of BF3.Et2O. 3-phenyl-3,4-dihydro-2H-1,3benzoxazine Ph-Bz was also polymerized in absence of catalyst as reference material. In Table 2 are collected the curing

TABLE 2 Onset (Texoo) and Maximum (Texom) Temperatures of the Exothermic Peaks in the Polymerization of Benzoxazine Monomers Without Catalyst and Catalyzed with 1, 2, and 3% of BF3.Et2O Mon. CA-Bz FA-Bz PA-Bz

Neat Texoo/Texom

1% Texoo/Texom

2% Texoo/Texom

3% Texoo/Texom

136

130

129

126

125 (3 h)

140

139

137

136

135 (1 h)

132

100

96

88

100 (3 h)

152

143

142

138

135 (1 h)

151

–

–

–

160 CAM-Bz FAM-Bz PAM-Bz

220

170

159

147

155 (3 h)

246

215

206

202

200 (1 h)

224

169

167

165

175 (3 h)

244

204/239

207/237

205

200 (1 h)

210

–

–

–

215 (3 h)

–

–

–

259

245 (1 h)

262 Curing and postcuring conditions selected.

4900

150 (3 h) 160 (1 h)

244 Ph-Bz

Curing conditions


150 (3 h) 260 (1 h)


ARTICLE


of the double bond protons are absent indicating probably some polymerization of the unsaturated chains. Noteworthy, no signals over 4.0 ppm can be observed confirming that after the used curing cycles, no N,O-ketalic links remain in the structure of these resins, which are composed basically by Mannich type bridges. Polybenzoxazine resins were characterized by DSC and TGA. The Tg values and the thermal stability under air and nitrogen is collected in Table 3. All monofunctional benzoxazinebased materials are characterized by low crosslinking degrees. In our monomers, the presence of substituents in some of the active aromatic positions, were expected to reduce even more the crosslinking points, especially in the case of ferulic derivatives.

FIGURE 3 DSC plots of FAM-Bz (————-) (a) and its mixtures with 1% (- - - -) (b), 2% (– – –) (c), and 3% (– - – -) (d), of BF3.Et2O. Upper plot is the TGA trace of FAM-Bz with 3% of BF3.Et2O. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

conditions for each monomer, which were established according to the DSC behavior. In this way, curing was performed for 3 h at about 10 C over the onset of the polymerization exothem, and postcuring heating for 1 h at the maximum temperature of the polymerization exothem. The resulting thermosets were hard and fragile orange to red solids which were completely insoluble in solvents such as DMSO, DMF, or NMP indicating their crosslinked structure. The only exception was the material obtained from FAM-Bz, which was partially soluble in DMSO. 1 H NMR analysis of this soluble fraction shows two main broad peaks: one at 7.5–6.5 ppm, corresponding to the aromatic protons, and a second at 4.0–3.6 ppm corresponding to the Mannich bridges, which appear overlapped with the signals of the COOCH3 and OCH3 groups. Also small signals at 2.9 and 2.0 ppm are observed, the first one attributable to the formation of diphenyl-methane bridges. Moreover, signals

On the contrary, the measured Tg’s of the obtained materials were in the order of 120–150 C and higher than the Tg of the substituent free Ph-BZ. Moreover, in all cases the methylester-derived benzoxazine materials have higher Tg values than their carboxylic acid counterparts. These high Tg values can be attributed to the occurrence of additional crosslinking reactions through the propanoic or propenoic units. In the case of DPA derived benzoxazines, we have recently demonstrated the existence of esterification and specially transesterification reactions between the phenolic hydroxylic groups resulting from the benzoxazine ring opening and the free carboxylic or the methyl ester groups.9 Analysis by FTIR-ATR of the obtained materials show new peaks at 1740–1760 cm21 for both the carboxylic acid and the methylester-derived polybenzoxazines. In fact this higher frequency peak constitutes the most intense carbonyl peak in all the thermosets indicating the extensive formation of phenolic ester linkages as extra crosslinking links. Referring to the thermal stability, all materials resulted less stable than the unsubstituted polibenzoxazine Ph-BZ with the exception of PAM-Bz. In addition CA and FA based benzoxazines are less stable than PA-based benzoxazines. Moreover, materials with free carboxylic acid groups are less stable than the corresponding methyl esters. These behaviors can be related with the low stability of unsaturated chains and the occurrence of decarboxylation processes. The presence of free carboxylic groups and unsaturated chains also

TABLE 3 Thermal Properties of the Benzoxazine Thermosets DSC Monom.

1/2DCp ( C)

TGA Nitrogen

TGA Air

T5% ( C)

Tmax ( C)

R800 (%)

T5% ( C)

Tmax ( C)

R800 (%)

CA-Bz

119

220

235, 414

30

221

235, 392, 451, 622

0

FA-Bz

120

200

200, 341

34

190

191, 265, 388, 552

0

PA-Bz

130

293

281, 455

43

295

277, 449, 650

0

CAM-Bz

154

280

300, 432

43

255

268, 468, 640

0

FAM-Bz

135

259

260, 330, 440

45

259

260, 344, 400, 554, 649

0

PAM-Bz

143

340

328, 423

43

352

330, 431, 652

0

Ph-Bz

115

320

395, 469

46

326

335, 470, 650

0



4901

ARTICLE



FIGURE 4 First derivative TGA plots of (a) PolyPh-Bz (. . . .) and the acid containing polybenzoxazines (——-): 1 PolyCA-Bz, 2 PolyFA-Bz, and 3 PolyPA-Bz and (b) PolyPh-Bz (. . . .) and ester containing polybenzoxazines (——-): 4 PolyCAM-Bz, 5 PolyFAM-Bz, and 6 PolyPAM-Bz. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

seems to influence the char residue, which is clearly lower (30–34%) when compared with the ester containing and PA derived benzoxazines (over 40%). Figure 4 shows the TGA curves and the fist derivative recorded under nitrogen. As can be observed in Figure 4, degradation behavior changes with the structure of the starting benzoxazines. CA, CAM, PA, and PAM thermosets present two main degradation stages, one with maximum degradation rate at 230–330 C and another, more important with maximum degradation rate at 410–460 C. Both stages occur at lower temperatures in the case of the acid containing materials. The first stage is also more significant in these materials indicating that probably corresponds to decarboxylation processes. PA and PAM-derived materials show the highest stability whereas FA and FAM derived materials seems to have a more complex degradation showing at least three maxima in nitrogen and a more complex pattern under air (see Table 2). The only structural difference between CA/CAM and FA/ FAM derivatives is the presence of a methoxy substituent in the last ones. FA/FAM derivatives only have two aromaticfree positions for the formation of Mannich bridges during the polymerization and meta positions are known to be not very active. This is consistent with a low crosslinking degree of the FA/FAM materials, their poor thermal stability, and their complex degradation pattern. CONCLUSIONS

Natural phenols derived from lignin (coumaric, ferulic, and phoretic acids) can be used to produce a palete of benzoxazine monomers. These monomers can be polymerized in

4902


presence of catalyst to produce materials with comparable or even improved properties when compared with the corresponding petroleum-based analogues.

ACKNOWLEDGMENTS

Financial support by the MICINN (Ministerio de Ciencia e Innovacion; MAT2011-24823) is gratefully acknowledged.

REFERENCES AND NOTES 1 H. Ishida, In Handbook of Benzoxazine Resins; H. Ishida, T. Agag, Eds.; Elsevier: Amsterdam, 2011; Chapter 1, pp 3–81. 2 H. Y. Low, H. Ishida, Macromolecules 1997, 30, 1099–1106. 3 J. J. Bozell, M. Patel, Eds. In Feedstocks for the Future: Renewables for the Production of Chemicals and Materials; ACS Symposium Series 921; American Chemical Society: Washington, DC, 2006. 4 A. Minigher, E. Benedetti, O. De Giocomo, P. Campaner, V. Aroulmoji, Nat. Prod. Comm. 2009, 4, 521–528. 5 O. A. Attanasi, M. S. Behalo, G. Favi, D. Lomonaco, S. E. Mazzetto, G. Mele, I. Pio, G. Vasapollo, Curr. Org. Chem. 2012, 16, 2613–2621. 6 E. Calo, G. Maffezzoli, F. Mele, S. E. Martina, A. Mazetto, A. Tarzia, C. Stifani, Green Chem. 2007, 9, 754–759. 7 B. Lochab, I. K. Varma, J. J. Bijwe, Therm. Calorim. 2010, 102, 769–774. 8 C. Wang, J.-Q. Sun, X.-D. Liu, A. Sudo, T. Endo, Green Chem. 2012, 14, 2799–2806. 9 C. Zuniga, M. S. Larrechi, G. Lligadas, J. C. Ronda, M. Galia, V.Cadiz, J. Polym. Sci. Part A: Polym. Chem. 2011, 49, 1219–1227. 10 J. J. Bozell, L. Moens, D. C. Elliot; Y. Wang, G. G. Neuenscwander, S. W. Fitzpatrick, R. J. Bilski, J. L. Jarnefeld, Resour. Conserv. Recy. 2000, 28, 227–239.



11 C. Zuniga, G. Lligadas, J. C. Ronda, M. Galia, V. Cadiz, Polymer 2012, 53, 3089–3095.

ARTICLE

12 A. Gandini, Biocatalysis in Polymer Chemistry; Katja Loos, Ed.; Wiley-VCH verlag GmbH & Co, KGaA: Weinheim, 2011.

26 T. Agag, L. Jin, H. Ishida, Polymer 2009, 50, 5940–5944. 27 H. Y Low, H. Ishida, J. Polym. Sci. Part B: Polym. Phys. 1998, 36, 1935–1946. 28 H. Y. Low, H. Ishida, Polymer 1998, 40, 4365–4376.

13 D. M. Pereira, P. Valentao, J. A. Pereira, P. B. Andrade, Molecules 2009, 14, 2202–2211.

29 K. Hemvichian, A. Laobuthee, S. Chirachanchai, H. Ishida, Polym. Degrad. Stab. 2002, 76, 1–15.

14 B. Dimitrios, Trends Food Sci. Technol. 2006, 17, 505–512.

30 K. Hemvichian, H. D. Kim, H. Ishida, Polym. Degrad. Stab. 2005, 87, 213–224. 31 S. B. Fam, T. Uyar, H. Ishida, J. Hacaloglu, Polym. Int. 2012, 61, 1532–1541. 32 A. Sudo, R. Kudoh, H. Nakayama, K. Arima, T. Endo Macromol. 2008, 41, 9030–9034. 33 C. Liu, D. Shen, R. M. Sebastian, J. Marquet, R. Shonfeld, Macromolecules 2011, 44, 4616–4622. 34 A. Sudo, R. Kudoh, T. Endo, J. Polym. Sci. Part A: Polym. Chem. 2011, 49, 1724–1729. 35 J. Dunkers, H. Ishida, J. Polym. Sci. Part A: Polym. Chem. 1999, 37, 1913–1921. 36 M. A. Espinosa, V. Cadiz, M. Galia, J. Appl. Polym. Sci. 2003, 90, 470–481.

15 M. Kleinert, T. Barth, Chem. Eng. Technol. 2008, 31, 736– 745. 16 H.-D. Shin, S. Mcclendon, T. Le, F. Taylor, R. R. Chen, Biotechnol. Bioeng. 2006, 95, 1108–1115. 17 S. I. Mussatto, G. Dragone, L. C. Roberto, Ind. Crop. Prod. 2007, 25, 231–237. 18 A. Tilay, M. Bule, J. Kishenkumar, U. Annapure, J. Agric. Food. Chem. 2008, 56, 7644–7648. 19 S. Y. Ou, Y. L. Luo, C. H. Huang, M. Jackson, Innov. Food Sci. Emerg. 2009, 10, 253–259. 20 R. Andreu, J. A. Reina, J. C. Ronda, J. Polym Sci. Part A: Polym. Chem. 2008, 46, 3353–3366. 21 X. Ning, H. Ishida. J. Polym. Sci. Part B: Polym. Phys. 1994, 32, 921–927. 22 X. Ning, H. Ishida. J. Polym. Sci. Part A: Polym. Chem. 1994, 32, 1121–1129. 23 H. Ishida, Y. Rodriguez, Polymer 1995, 36, 3151–3158. 24 H. Ishida, U.S. Patent 5,543,516, 1996. 25 Z. Brunovska, J. P. Liu. H. Ishida, Macromol. Chem. Phys. 1999, 200, 1745–1752.


37 R. Andreu, J. A. Reina, J. C. Ronda, J. Polym. Sci. Part A: Polym. Chem. 2008, 46, 6091–6101. 38 Y.-X. Wang, H. Ishida, Polymer 1999, 40, 4563–4570. 39 A. Sudo, S. Hirayama, T. Endo, J. Polym. Sci. Part A: Polym. Chem. 2010, 48, 479–484. 40 J. A. Cid, Y.-X. Wang, H. Ishida, Polym. Polym. Comp. 1999, 7, 409–420.


4903