Synthesis and characterization of poly(phenylene oxide) graft copolymers by atom transfer radical polymerizations

European Polymer Journal 45 (2009) 2348–2357

Contents lists available at ScienceDirect

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

Synthesis and characterization of poly(phenylene oxide) graft copolymers by atom transfer radical polymerizations Mong Liang *, Yong-Jhao Jhuang, Chun-Fu Zhang, Wei-Jhuan Tsai, Hui-Chuan Feng Department of Applied Chemistry, National Chia-Yi University, Chia-Yi 600, Taiwan

a r t i c l e

i n f o

Article history: Received 23 January 2009 Received in revised form 27 April 2009 Accepted 3 May 2009 Available online 12 May 2009

Keywords: PPO-graft copolymer Atom transfer radical polymerization (ATRP) Couchman equation Compatible

a b s t r a c t A series of comb-like poly(phenylene oxide)s (PPO) graft copolymers with controlled grafting density and length of grafts were synthesized by atom transfer radical polymerization (ATRP). The a-bromo-poly(2,6-dimethyl-1,4-phenylene oxide)s (BPPO) were used as macroinitiators to polymerize vinyl monomers and the graft copolymers carrying polystyrene (PS), poly(p-acetoxystyrene) (PAS), and poly(methyl methacrylate) (PMMA) as side chains were synthesized and characterized by NMR, FTIR, GPC, DSC and TGA. The compositiondependent glass-transition temperatures (Tg) of PPO-g-PS exhibited good correlation with theoretical curve in Couchman equations except for the cases of low PS content (<40 mol%) copolymers in which a positive deviation was observed due to enhanced molecular interactions. The increase in monomer/initiator ratio led to the increase of degree of polymerization and the decrease of polydispersity. Despite the immiscibility nature between PPO and PMMA, the PPO-g-PMMA exhibited enhanced compatibilization as apparent single Tg in a wide temperature window throughout various compositions revealing an efficient segmental mixing on a molecular scale due to grafting structure. Ó 2009 Elsevier Ltd. All rights reserved.

1. Introduction Recently, preparation of new graft copolymers have attracted considerable attention as their potential use in membrane separation technology, modification of conducting polymers [1], surface modification [2,3], and brush-type copolymers [4,5]. Through proper design or modifications, new polymeric structures with tailor-made properties are accessible for different applications. For the past decades, various grafting techniques including chemical, radiation, photochemical, plasma-induced and enzymatic have been developed to design new materials with optimal thermo, physical, mechanical and optical properties. Among them, atom transfer radical polymerization (ATRP) is one of the successful and convenient meth-

* Corresponding author. Address: Department of Applied Chemistry, National Chia-Yi University, 300 University Rd., Chia-Yi 600, Taiwan. Tel.: +886 5 2717952; fax: +886 5 2717901. E-mail address: [email protected] (M. Liang). 0014-3057/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.eurpolymj.2009.05.008

ods for new graft polymer synthesis because it can be carried out under relatively mild conditions and allows a variety of vinyl monomers to be polymerized in a controlled fashion with well-defined structures. So far three types of grafting methods employing ATRP have been used, which are grafting-through [6], grafting-to [7–9] and grafting-from [10–12], concerning the synthesis and application of graft copolymers. The grafting-from technique which involves the preparation of a multi-functional macroinitiator with a predetermined number of pendent reactive sites on polymer backbone has been widely used for molecular brushes synthesis due to its controllable grafting density, and adjustable amphiphilicity [13–15]. Despite the extensive effort in grafting-from method, most of the resulting copolymers consist of ethylene structure as backbone, whereas few of non-ethylene main chain skeletons such as cellulose [16,17], vinyl chloride [18] and vinylidene fluoride [19,20] were also reported. Styrene, acrylate and methacrylate are most commonly used monomers to graft onto various polymer backbones [14,21–23] to modify

M. Liang et al. / European Polymer Journal 45 (2009) 2348–2357

their surface properties and compatibility; however, the weak thermal stability largely limited the usage in high performance applications due to the inherent flexible backbone nature. Although recent progress in surface-initiated ATRP have shown examples in grafting PET [24], PC [25] and nylon [26] membrane surfaces with N-isopropyl acryl amide and 2-hydroxyethyl methacrylate in a heterogeneous fashion, little work has been conducted homogeneously with better thermal stability, solubility and processibility of the resulting copolymers, and to the best of our knowledge, reports regarding the synthesis and thermal properties of polyphenylene oxide (PPO) graft copolymers have yet to be explored. This is probably due to the poor solubility of the engineering plastics in common organic solvents and the difficulty in obtaining a complete initiation process on the backbone with a well-characterized architecture. In the present study, we report the use of grafting-from method to synthesize a series of comb-like PPO-graft copolymers and terpolymers carrying polystyrene (PS), poly-4-acetoxystyrene (PAS) and poly(methyl methacrylate) (PMMA) as side chains, starting from well-defined a-brominated poly(phenylene oxide) (BPPO) macroinitiators obtained in our previous work [27]. Poly(phenylene oxide) (PPO) is a well-known engineering plastic with a low dielectric constant of 2.58 (at 23 °C 60 Hz) and a high glass transition temperature (Tg) of approximately 212 °C. It is one of the potential materials to satisfy the demand of high frequency substrates in the electronics industry and its alloys are widely used in electrical appliance due to the balanced physical, chemical and electrical properties. PPO has been blended with various materials to improve its brittleness and processibility, however, except for PPO/PS system, most blends are immiscible because of their unfavorable intermolecular interactions which, in turn, attribute to the deterioration in mechanical properties at phase boundaries. Much of the work is concerned with the enhancement of interfacial adhesion between phases by incorporating the compatibilizers, nevertheless, the decrement of glass transition temperature or thermal decomposition temperature arising from the low molecular weight compatibilizers are common drawbacks of the polymer blends. The key aspect of this study was to graft polymer from PPO with different miscibility segments using ATRP method and investigated their compatibility and thermal properties.

2. Experimental 2.1. Materials All reagents and solvents were reagent-grade. Poly(2,6-dimethyl-1,4-phenylene oxide) of intrinsic viscosity equal to 0.4  103 m3 kg1 (i.e., 0.4 dL g1) in chloroform at 25 °C was obtained from General Electric Plastics and purified by precipitation from chloroform solution into methanol before use. N-Bromosuccinimide (NBS; Acros), 2,20 -Azobis-isobutyronitrile (AIBN; Showa), 2,20 -bipyridine (Bpy; Acros), Copper chloride (CuCl; Showa) were purchased from commercial sources and used as received without purification. Chloro-

2349

benzene, chloroform, n-hexane (all from TEDIA), methyl alcohol (from Mallinckrodt) were purchased from commercial companies and used as received. Styrene (Sty; Acros) and pacetoxystyrene (PAS; TCI) were washed twice with 5% NaOH solution and then water, which was subsequently dried over anhydrous magnesium sulfate (MgSO4; J.T. Baker) and distilled under reduced pressure. Methyl methacrylate (MMA; Fluka) was freshly distilled before polymerization. Toluene (Tedia) was refluxed with calcium hydride (CaH2; Fluka) and then distilled under nitrogen. 2.2. Instrumentation 1 H and 13C NMR spectra were recorded on a 300 and 75 MHz Varian – Mercury+ 300 spectrometer, respectively, using the CDCl3 solvent and the internal standard tetramethylsilane. Fourier transform infrared spectroscopy (FTIR) spectra were recorded for KBr disks using a Shimadzu FT-IR 8400 spectrometer. The molecular weight of the polymers was determined by gel permeation chromatography (GPC) which was carried out with polymer solutions in THF. Samples were prepared at nominally 1 mg mL1 in THF and filtered through a 0.45 lm syringe filter and injected by WATERS717 Autosample. GPC system (Waters 515 HPLC pump, 1 mL/min, 40 °C) were equipped with WATERS STYRAGEL HR0.5, HR4E, HR5 and a Waters refractive index detector (Model 2410). DSC measurements were performed on a DSC Q10 Differential Scanning Calorimeter (TA Instrument). Samples of about 5 mg were weighed and hermetically sealed into aluminum pans (diameter: 5 mm, TA Instruments) for standard DSC analysis. The glass-transition temperatures of the PPO-g-PS and PPO-g-PAS were determined by heating at a rate of 20 °C/ min to 280 °C, held there for 10 min and then cooled to 40 °C in fast equilibrium to obtain amorphous state samples, then repeated the program at a rate of 20 °C/min again. For PPO-g-PMMA, the samples were treated similarly except the temperature was heated up to only 250 °C. All the Tg values were measured after the 2nd heating scan. The nitrogen gas flow was 50 mL/min. Thermogravimetric analysis was performed with a TGA Q50 (TA instrument) thermogravimetric analyzer under nitrogen atmosphere. Heating scans carried out from 30 to 800 °C at 20 °C/min.

2.3. Synthesis of macroinitiator poly(2,6-dimethyl-1,4phenylene oxide) (BPPO) To a stirred solution of 2.4 g PPO in chlorobenzene (100 mL) was added N-bromosuccinimide (0.89 g, 5 mmol), and 2,20 -azobis-isobutyronitrile (0.1 g, 0.6 mmol). The mixture was irradiated with mercury lamp (100 W) placed at a short distance (2 in., 21700 lW/cm2) from reaction vessel and heated at reflux conditions for 3 h. After cooling, the reaction mixture was added to a 10-fold excess of n-hexane to precipitate the product. The polymer was filtered and washed with methanol, which was redissolved in minimum amount of chloroform and precipitated into a 10-fold excess of methanol solution. The polymer was collected as a light yellow powder and dried under vacuum for overnight (BPPO(L), bromination ratio: 10%). By the same procedure,

2350

M. Liang et al. / European Polymer Journal 45 (2009) 2348–2357

1.78 g, 10 mmol of NBS and 0.1 g, 0.6 mmol of AIBN gave the higher degree of brominating product, BPPO(H), bromination ratio: 31%. 1H NMR spectrum (300 MHz, CDCl3, ppm): 2.08 (Ar-CH3, s, 9H), 4.34 (Ar-CH2ABr, s, 2H), 6.47 (Ar-H of PPO repeat unit, s, 2H), 6.5–6.7 (Ar-H of bromomethylated PPO unit, m, 2H). IR (KBr, cm1): 3035–2862 (CAH stretching), 1612 (CAC aromatic ring), 1474 (CH2 bending), 1302 (CACH3 stretch), 1189 (Ar-O-Ar stretch), 1223 (CH2ABr, bend), 595 (CHABr, stretch). 2.4. Synthesis of PPO-g-PS (polyphenylene oxide-graftpolystyrene) copolymer The copolymers were synthesized by solution polymerization in toluene. In a typical experiment, a 100 mL dry three-neck flask equipped with a stir bar was added with BPPO(L) (0.59 g, 0.43 mmol), 2,20 -bipyridine (0.27 g, 1.74 mmol), copper chloride (0.043 g, 0.43 mmol) under nitrogen. After toluene (30 mL) was injected into the reaction system, the mixture was degassed with three freeze pump-thaw cycles and was heated in an oil bath at 110 °C for 1 h. Then, styrene (4.53 g, 4.35 mmol) was injected into the reaction system and the mixture was stirred at 110 °C for several hours under nitrogen. After cooling, the reaction mixture was poured into a 10-fold excess of methanol and the precipitated polymer was filtered, washed with methanol and purified by reprecipitation from chloroform solution into methanol. The graft copolymer was collected as a light yellow powder and dried under vacuum at room temperature for overnight. 1H NMR spectrum (300 MHz, CDCl3, ppm): 1.26–1.65 (methylene of PS repeat unit, CH2ACH), 1.73–1.98 (methylene of PS repeat unit, CH2ACH), 2.08 (Ar-CH3), 6.47 (Ar-H of PPO repeat unit), 6.50–6.89 (ortho Ar-H of PS repeat unit), 6.89–7.37 (para and meta Ar-H of PS repeat unit). 13C NMR (75 MHz, CDCl3) d (ppm): 155.7, 145.7, 133.4, 128.6, 128.3, 126.5, 115.1, 43.1, 40.6, 16.9. IR (KBr, cm1): 3094–2855 (CAH stretching), 1605 (CAC aromatic ring), 1470 (CH2 bending), 1305 (CACH3 stretch), 1191 (Ar-OAr stretch), 756 (CAH, out of plane bending of PS repeat unit), 698 (CAH, out of plane bending of PS repeat unit). The similar procedure was used for all of the PPO-g-PS polymers, except for entry 9–13 of Table 2 in which BPPO(H) was used as macroinitiator. 2.5. Synthesis of PPO(L)-g-PAS (polyphenylene oxide-graftpoly(p-acetoxystyrene)) The same procedure described above was applied for synthesis of PPO(L)-g-PAS. In a typical run, a dry three-neck flask fitted with magnetic stirring bar under nitrogen was added with BPPO(L) (0.4 g, 0.31 mmol), 2,20 -bipyridine (0.2716 g, 1.7 mmol), copper chloride (0.062 g, 0.43 mmol) and toluene (10 mL). After degassed with three freeze pump-thaw cycles, the mixture was heated in an oil bath at 110 °C for 1 h and 2 mL of p-acetoxystyrene (2.12 g, 13 mmol) was added. After stirring at 110 °C for several hours under nitrogen, the reaction was terminated by pouring the mixture into a 10-fold excess of methanol and the precipitated polymer was filtered, washed with methanol and purified by reprecipitation from chloroform solution into methanol. The graft copolymer

was collected as a light yellow powder and dried under vacuum at room temperature for overnight. 1H NMR spectrum: 2.08 (Ar-CH3 of PPO repeat unit), 2.25 (COACH3 of PAS repeat unit), 6.58 (Ar-H of PPO repeat unit), 6.68–7.02 (Ar-H of PAS repeat unit). 13C NMR (75 MHz, CDCl3) d (ppm): 169.2, 154.6, 148.7, 149.3, 142.1, 132.4, 128.3, 121.1, 114.3, 41.5, 41.8, 39.9, 21.1, 16.2. IR (KBr, cm1): 3035–2852 (CAH stretching), 1766 (C@O, stretch), 1608 (CAC aromatic ring), 1504 (C@C stretch), 1475 (CH2 bending), 1305 (CACH3 stretch), 1366 (CH3 bending), 1194 (Ar-O-Ar stretch), 908 (CAH, out of plane bending of PAS repeat unit). 2.6. Synthesis of PPO-g-PMMA (polyphenylene oxide-graftpoly methyl methacrylate) The similar method was used for the preparation of PPOg-PMMA. In a typical run, BPPO(L) (0.25 g, 1.96 mmol), 2,20 bipyridine (0.21 g, 1.38 mmol), Copper chloride (0.046 g, 0.46 mmol) were dissolved in toluene (30 mL) and methyl methacrylate (4.68 g, 46.8 mmol) was added. After degassed with three freeze pump-thaw cycles, the reaction temperature was kept at 90 °C under nitrogen for several hours and the graft copolymer was collected as a light yellow powder and dried under vacuum at room temperature for overnight. 1 H NMR spectrum: 0.82–1.03 (CACH3 of PMMA repeat unit), 1.78 (CH2ACH2AC of PMMA repeat unit), 2.08 (Ar-CH3 of PPO repeat unit), 3.61 (OCH3 of PMMA repeat unit), 6.46 (Ar-H of PPO repeat unit). 13C NMR (75 MHz, CDCl3) d (ppm): 177.8, 154.6, 145.3, 132.5, 114.4, 51.8, 44.8, 44.4, 18.6, 16.7. IR (KBr, cm1): 3062–2835 (CAH stretching), 1743 (C@O, stretch), 1603 (CAC aromatic ring), 1462 (CH2 bending), 1305 (CACH3 stretch), 1192 (Ar-O-Ar stretch), 1146 (CAO stretch). 2.7. PPO/PS blend sample preparation Mixtures of PPO/PS with various compositions (weight ratios: 0/100, 10/90, 20/80, 30/70, 40/60, 50/50, 60/40, 70/30, 80/20, 90/10 and 100/0) were codissolved in chloroform with a concentration of 1.5 g/100 mL and stirred at room temperature for 1 h for complete mixing. The samples were dried at room temperature for 1 h under high vacuum and then in a vacuum oven at 60 °C for overnight to remove the residual solvent.

3. Results and discussion 3.1. Graft polymerization of styrene on BPPO The general synthetic pathway of PPO-graft copolymers is shown in Scheme 1. In our previous paper [27], the methyl-brominated PPO can be obtained in high yields with excellent selectivity and hardly any phenyl-brominated or methyl-dibrominated derivative was found after isolation. Two types of BPPO with 10% or 31% degree of bromination were prepared and the analysis was listed in Table 1 The ATRP of styrene on BPPO was carried out at reflux condition under nitrogen using CuCl/bpy as catalyst and two series of graft copolymers, PPO(L)-g-PS (using 10% brominated BPPO as macroinitiator) and PPO(H)-g-PS

2351

M. Liang et al. / European Polymer Journal 45 (2009) 2348–2357

CH3

CH3

CH3

O

n

CH 3

PPO

CH 2Br

CuCl/bpy/ styrenic monomer

h ν,NBS

O

O CH3

n

toluene,110 o C

m

BPPO

R

R

H 2C

CH3 H

H2 C

O

O n

C H

Br

H m

CH3

CH3

H2 C

C H

x

CuCl/bpy/MMA H

O

O

CH3

C

O

C CH3

Br

y

H

toluene,110 o C CH3

PPO-g-PS

H2 C

H 2C

CH3

x

O

R=H,-OCOCH 3

CH3

PPO-g-PS-b-PMMA

R=H,-OCOCH 3

Scheme 1.

Table 1 Analysis of macroinitiators. Polymer

m (%)

Mw

Mn

PDI

Tg (°C)

T5%,d (°C)

Char (%)

BPPO(L) BPPO(H)

10 31

37320 43955

11289 13183

3.3 3.3

217.3 209.4

334.9 304.3

34.7 42.0

(using 31% brominated BPPO as macroinitiator) with various molar ratios were obtained and the results are summarized in Table 2. The CuCl/bpy catalyst complex is chosen to use in conjunction with bromine terminated BPPO to form a mixed halogen system because initiation of the RABr bond generally occurs more efficiently with the ‘‘halogen exchange” method [28,29]. Shown in Fig. 1 are typical 1 H NMR spectra of original BPPO, PPO-g-PS, PPO-g-PAS, PPO-g-PMMA, and PPO-g-PS-b-PMMA. For PPO-g-PS, the

clean disappearance of the bromo-benzylic peak at d = 4.3 ppm in BPPO along with the concomitant appearance of peaks for polystyrene at d = 6.5–7.5 and 1.6– 2.0 ppm indicated a high initiating efficiency under these conditions. Attempts to carry out the reaction in the absence of BPPO resulted in no reaction which suggested the formation of graft copolymer without homopolymerization. The 13C NMR spectrum exhibited all the resonances for both PS and PPO moieties and FTIR spectrum

Table 2 Conditions and results for ATRP of vinyl monomer with BPPO. Entry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

BPPO (m %)

10 10 10 10 10 10 10 10 31 31 31 31 31 10 10 10 10 10 10 10 10 10

Monomer Type

[M]/[I]

Sty Sty Sty Sty Sty Sty Sty Sty Sty Sty Sty Sty Sty PAS PAS PAS PAS MMA MMA MMA MMA MMA

11 11 22 67 111 201 111 111 16 49 49 41 41 41 41 41 105 10 10 49 96 96

Time (h)

Graft (mol %)

Mn (by GPC)

Mw (by GPC)

PDI

Tg (°C)

T5%,d (°C)

Char (°C)

3 5 3 3 3 3 4.5 15 3 2 2.5 3 15 1 2 2.5 9 2 2.5 2 3 3.5

12.4 19.6 25.3 44.1 47.2 51.7 66.4 80.0 39.5 47.4 57.7 65.7 73.3 26.4 36.3 44.4 67.6 22.0 43.0 59.0 74.0 80.0

18979 20946 32518 30749 41123 62368 54153 74680 16651 28160 42674 53124 65202

92446 99967 102698 92244 127216 155316 135894 178806 91710 122666 146612 149145 181176

a

a

a

a

a

a

a

a

45758 52154 50601 54489 67984

154548 168904 156554 136764 156515

4.8 4.7 3.1 3 3 2.5 2.5 2.4 5.5 4.3 3.4 2.8 2.8 – – – – 3.4 3.2 3.1 2.5 2.3

212.3 198.9 190.6 161.4 158.2 147.2 134.0 112.5 161.2 152.4 135.3 128.9 117.8 206.6 182.4 166.7 141.3 202.8 175.2

331.9 332.9 332.6 341.2 324.1 344.8 341.1 368.1 252.5 285.4 284.9 321.6 345.7 306.7 327.1 332.7 348.5 290.7 282.2 215.7 205.9 204.1

32.3 29.9 30.7 21.6 29.5 26.3 19.1 9.4 25.2 24.4 19.0 20.0 10.8 27.4 33.2 29.0 13.9 30.1 29.5 22.2 19.0 12.8

All Tg values were obtained after the second heating scan. a Product is not completely soluble in THF. b Tg is immeasurable from DSC thermogram.

b

128.4 125.6

2352

M. Liang et al. / European Polymer Journal 45 (2009) 2348–2357

where d are the integral of the signals originated from corresponding resonances and A is the average number of amethyl hydrogen per repeat unit based on the degree of bromination of macroinitiator which can be calculated using the formula A = (6 * n + 5 * m)/(m + n). It was observed that the resulting polymers showed increases in number-average molecular weight (Mn) with reaction time and monomer concentration demonstrating the effectiveness of ATRP and formation of graft copolymer. With the increase in monomer/initiator ratio ([M]/[I]), the degree of polymerization increases and the polydispersity decreases as shown in Table 2. It should be noted that the high PDI value was due to the intrinsic nature of macroinitiator. Also, the increase in molecular weight and the amount of styrene incorporated in the grafts measured by NMR and GPC supports the formation of graft copolymer with a controlled process. The glass transition behaviors of graft copolymers were examined with differential scanning calorimetry and all of them exhibited a single Tg indicating a single-phase nature as shown in Figs. 2 and 3. When PS content increases, the Tg of the copolymer decreases linearly suggesting the formation of homogeneous phase upon chain growing. For binary miscible polymer blends, many empirical formulas have been used to describe the dependence of glass transition temperature on composition [30–34]. We found the Couchman equation [34] was ideal one in evaluating the relation between the Tg and composition of the PPO-g-PS system with an adjustable parameter k in fitting Eq. (2) to experimental data:

a. BPPO b. PPO-g-PS c. PPO-g-PAS d. PPO-g-PMMA e. PPO-g-PS-b-PMMA

e

d

c

b

a

9

8

7

6

5

4

3

2

1

0

ln T g ¼

ppm Fig. 1. 1H NMR spectra of PPO-graft copolymers.

also confirmed the presence of copolymers. The molar content of PS side chain is measured by 1H NMR spectroscopy by comparing the average integral of aromatic protons of styrene at d = 6.89–7.37 ppm (denoted as h, g, respectively, in Eq. (1)) to the total area of benzylic protons at d = 2.08 ppm (denoted as a, a0 , respectively) according to the equation:

w1 ln T g1 þ kw2 ln T g2 w1 þ kw2

ð2Þ

where Tg1, Tg2 are the glass-transition temperatures of component 1 (PS) and component 2 (PPO), respectively; w1 and w2 represent the weight fractions of components 1 and 2. Fig. 4 shows composition dependence of Tg for PPO/PS blends, PPO(L)-g-PS PPO(H)-g-PS system and the theoretical curve of Couchman equation. The short dot line represents values predicted by the Couchman equation. The experimental values of Tg in PPO/PS blends fit well to the

12.4%

h g a'

a c H

O CH 3

H2 C

H2 C

CH 3

b n

b'

C H

e O

m

19.6%

f

d

25.3%

Br

44.1%

x

47.1%

H 51.7%

CH3

66.4% 80.0% PPO-g-PS-b-PMMA

XðPS mol%Þ ¼ f½ðdg þ dh Þ=3=½ðdg þ dh Þ=3 þ ðda þ da0 Þ=Ag

40

60

80

100

120

140

160

180

200

220

240

ð1Þ Fig. 2. DSC heating curves of PPO(L)-g-PS at various PS molar content.

M. Liang et al. / European Polymer Journal 45 (2009) 2348–2357

39.5% 47.5% 57.7%

65.7% 73.3%

100

150

200 o

Temperature ( C) Fig. 3. DSC heating curves of PPO(H)-g-PS at various PS molar content.

500 PPO/PS blend

480

Couchman curve

Temperature (K)

PPO(H)-g-PS

460

PPO(L)-g-PS

440

420

400

380 0.0

0.2

0.4

0.6

0.8

1.0

PS weight fraction Fig. 4. The compositional variation of Tg for PPO/PS blend, PPO(H)-g-PS, PPO(L)-g-PS and theoretical curve of Couchman equation when k = 0.85.

theoretical curve when k = 0.85. Given that k is equal to DCp2/DCp1, where DCp is the specific heat capacity change at Tg of polymer 1 and polymer 2, the corresponding k value for the matrix is calculated as 0.895 [35,36], which is comparable with our data. The good agreement between the experimental data and the theoretical curve revealed accurate estimations between structural compositions and Tg measurements in this work. In the case of PPO(L)g-PS system, the higher PS compositions (40–80 wt.%) basically follow the Couchman equations whereas a positive deviation can be observed in the lower range (0–40 wt.%) of PS, which implies that there might exist some enhanced molecular interactions when the PS content is low. Presumably the ATRP of BPPO can allow an even distribution of PS on PPO skeleton and provide more effective interactions between the phenyl groups of PS and phenyl rings of PPO [37], or between the methyl groups of PPO and the phenyl rings of PS than those of conventional PPO/PS blend. This effect was pronounced especially when degree of polymerization of PS is low but became less significant with increasing PS content due to the considerable intermolecular interactions associating with alloy formation. Evidence for intermolecular interactions between PPO and PS could be obtained by FTIR spectroscopy. It is known

2353

that the IR spectroscopy can provide useful information in studying PPO/PS polymer blends and such analysis has already been applied to explain the conformational changes brought by dipole–dipole coupling of the benzene rings of two homopolymers [38]. Also, Benedetti et al. have demonstrated the influence of temperature changes on intermolecular interactions and the sensitivity of IR frequency on band width between the PPO and PS [39] under different environments. Illustrated in Fig. 5 is the comparison of band shape in the region 1100–1240 cm1 of PPO(L)-g-PS and PPO/PS blend at 10%, 20%, 40% and 80% of PS content, respectively. For PPO(L)-g-PS, the band at 1190 cm1 is generally assigned to the asymmetric CAOAC backbone stretching, which is narrower than that of the PPO/PS blend at the same composition. The narrowing of band width under same environments might be explained by specific molecular interactions, or the less conformational freedom caused by structural differences between graft copolymer and blend of two homopolymers. Because of its restricted backbone conformation, PPO(L)-g-PS has a narrower band width and higher value of Tg than PPO/PS blend. However, the extent of band narrowing in Fig. 5d became barely distinguishable which was coincident with the fact that the contributions might have been diluted to negligible extent due to the increase of PS content, and therefore presented no difference in Tg values in Couchman equations. Another possible explanation for the increments of Tg could be due to the increased bulkiness of side chains of PPO-g-PS. Given that Tg is a function of rotational freedom, the growth of PS on side chain would restrict the rotational mobility and chain conformation of polymer to certain extent and lead to the increase of Tg. Related to this result was found by Kuo et al. [40], who obtained higher Tg values of PPOblock-PS than those of the PPO/PS blends using low molecular weight PPO as macroinitiator. Thermal stabilities of polymers were examined by thermogravimetric analysis (TGA) with the 5 wt.% decomposition temperature, Td,5% and the results are summarized in Table 2. All polymers displayed essentially a two-stage degradation wherein the degradation temperatures between 240 and 280 °C could be ascribed to the terminal chain dissociation of the weaker CABr bond and the second stage degradation occurring at 420–500 °C revealed a backbone scission of PS and PPO. Interestingly, although the pure PS generally exhibits lower thermal stability than that of PPO, the more PS content in graft copolymer generally leads to the higher Td,5% value. We attribute the trend to the relative content of the halogen-capped end group per molecular chain in polymer. The reduction in the bromine content is usually associated with the growth of PS chain per unit mass of polymer and consequently, displays improved initial thermal stability with relatively less char yield. Consistent with this explanation, the PPO(L)-g-PS which contains less CABr terminus is found to exhibit higher Td,5% values than those of PPO(H)-g-PS at the nearly same compositions. A general feature of ATRP is the ability to form multiblock copolymers polymer by sequential addition of different monomers. Employing the same methodology, we may also produce the PPO-g-PS-b-PMMA grafted block terpolymers by ATRP of methyl methacrylate using PPO-g-PS-X (X = Br

2354

M. Liang et al. / European Polymer Journal 45 (2009) 2348–2357

b

a

PPO(L)-g-PS PPO/PS blend

PPO(L)-g-PS PPO/ PS blend

1240

1220

1200

1180

1160

cm

1140

1120

1240

1100

1220

1200

1180

-1

cm

c

1160

1140

1120

1100

-1

d PPO(L)-g-PS PPO/PS blend PPO(L)-g-PS PPO/PS blend

1240

1220

1200

1180

cm

1160

1140

1120

1100

-1

1240

1220

1200

1180

cm

1160

1140

1120

1100

-1

Fig. 5. Comparisons of FTIR spectra in the region 1100–1240 cm1 of PPO(L)-g-PS and PPO/PS blend at the same molar composition. (a) PS = 10 mol%, (b) PS = 20 mol%, (c) PS = 40 mol% and (d) PS = 80 mol%.

or Cl due to halogen exchange) as macroinitiator and CuCl/ bpy complex as a catalyst. Fig. 6 illustrates the representative FTIR spectra of BPPO, PPO-g-PS, PPO-g-PMMA and PPO-g-PS-b-PMMA. In Fig. 6b, the characteristic absorptions at 1468, 1188 and 1022 cm1 of the FTIR spectrum were assigned to the aromatic stretching vibration, the asymmetric CAOAC backbone stretching vibration and the methyl rocking vibrations of PPO, respectively, whereas the absorptions at 698, and 756 cm1 corresponded to the two CAH out of plane bending vibration of mono-substituted ring of PS. After ATRP of methyl methacrylate, a carbonyl stretch was observed at 1734 cm1 indicating the formation of PPO-gPS-b-PMMA grafted block terpolymers. Further structural verifications of terpolymers were performed by NMR spectroscopy, which was also used to determine the molar composition ratio. The 1H NMR spectrum in Fig. 1e disclosed three typical singlets at d = 3.61 ppm (ACOOCH3) and d = 1.03, 0.82 ppm (ACCH3) for repeated MMA unit and other signals such as the methyl protons at d = 2.1 ppm (Ar-CH3) of PPO segments as well as the aromatic protons at d = 6.88–7.45 ppm on styrene repeat unit also agreed with the structure. 13C NMR spectrum exhibited all the characteristic resonances; five for PPO (16.8, 115.0, 133.4, 146.3, 154.7), three for aromatic ring carbons of PS (126.4, 128.2,

128.8) and two for the methoxy and carbonyl carbons of PMMA (52.1, 178.75). The composition of the PPO-g-PS-bPMMA was analyzed by integration ratio of the characteristic peak area and the molar ratio of PPO:PS:PMMA was calculated approximately 36.3:23.7:56.7. The apparent glass transition was detected at 125 °C and the Td,5% value was measured at 299.2 °C. 3.2. Graft polymerization of p-acetoxystyrene on BPPO For the ATRP of p-acetoxystyrene using the aforementioned technique, polymers of varying PAS chain length were obtained and thermal properties were illustrated in Table 2 and Fig. 7. The glass transition temperature of PPOg-PAS decreases linearly with the increasing PAS component whereas the Td,5% value increases inversely, which is consistent with the observation in styrene copolymer. The molar ratio of PAS is calculated from integral ratio of the NMR signals originated from the PAS block at 2.25 ppm (CH3COA) and the PPO block at 2.08 ppm (Ar-CH3), respectively. Likewise, the grafted block terpolymer, PPO-g-PAS-b-PMMA could be subsequently prepared by ATRP with MMA, indicating that significant portion of PPO-g-PAS could be reactivated and a molar ratio of PPO:PAS:PMMA = 0.2:0.49:0.31

2355

M. Liang et al. / European Polymer Journal 45 (2009) 2348–2357

PPO and PMMA, would show any improved compatibility due to the graft structure. Similar reaction conditions were applied to the grafting of BPPO(L) with MMA, except the reaction temperature was kept at 90 °C. Variation of reaction time would basically afford copolymers with same backbone length and graft-density but different in the PMMA graft length. 1H NMR and FTIR spectroscopes are depicted in Figs. 1d and 6c and other data are presented in Table 2. The composition of the graft copolymers is determined according to the following equation:

d c

b

k

a

a'

a c

CH3

b

H 2C

O

CH3

j

C

O

H2 C

C CH3

H

O CH3

2000

1800

1600

1400

1200

1000

800

600

400

O

n

b'

m

H

Br

y

i

CH 3

YðPMMA mol%Þ ¼ fðdk =3Þ=½ðdk =3Þ þ ðda þ db0 þ dc Þ=2g

cm -1

ð3Þ

Fig. 6. FTIR spectra of PPO-graft copolymers. (a) BPPO(L), (b) PPO(L)-g-PS, (c) PPO(L)-g-PMMA and (d) PPO(L)-g-PS-b-PMMA.

with single sharp glass transition at 136.4 °C and Td,5% value was obtained at 330.9 °C. The structure of graft copolymer was also confirmed by IR, 1H and 13C NMR. 3.3. Graft polymerization of methyl methacrylate on BPPO Considering the effect of molecular interactions observed in PPO(L)-g-PS, We were tempted to investigate whether two completely immiscible components, such as

Except for entry 20 whose Tg was immeasurable from DSC thermogram, the Tg values decreased almost linearly with increasing length of PMMA graft and Td,5% values decreased correspondingly due to the inherent thermal stability of PMMA. Fig. 8 shows DSC heating curves of PPOg-PMMA copolymers. As shown in Fig. 8a, the mutually immiscible PPO/PMMA blend (50 mol% of each) exhibits two unperturbed glass transitions at 212 and 120 °C, respectively, indicating its incompatible nature or some phase separation of the components, whereas for PPO-gPMMA containing various PMMA contents, each shows single glass transition, except for entry 20 whose Tg is immeasurable from DSC thermogram. As the miscibility

Fig. 7. DSC heating curves of PPO(L)-g-PAS at various PAS molar content.

2356

M. Liang et al. / European Polymer Journal 45 (2009) 2348–2357

a

PMMA

b c

PPO

d

BLEND

e

22%

f

43% 59% 74%

80% 60

80

100

120

140

160

180

200

220

240

o

Temperature ( C) Fig. 8. DSC heating curves of PPO/PMMA blends and PPO(L)-g-PMMA at various PMMA molar content.

of incompatible polymers depends on specific interactions in the blends including hydrogen bonding, dipole–dipole interaction [41], the single and broadening in glass transitions reveals enhanced molecular interactions between PPO and PMMA. Since the miscibility in polymer blends is strongly dependent on the chain microstructure, it may be deduced by the architecture of PPO-g-PMMA when PMMA is uniformly distributed along the PPO backbone, the dipole–dipole interaction sites are large, thus the great extent of chain entanglement may have enhanced segmental mixing on a molecular scale and resulting in a significant effect on its phase boundaries despite the solubility parameter difference between PPO and PMMA [42,43]. The degree of intermixing can also be displayed by the breadth of glass transition DTg, which is quantified by the difference between onset and end point temperatures of DSC curves [44]. If the transition breadth is used as a measure of heterogeneity, the observed broadening of DTg in Fig. 8d and e is considered to indicate that the level of miscibility is less than that of other compositions. Such broadening effect is not observed in PPO-g-PS cases due to the fully miscible nature between PPO and PS. Nevertheless, it is the first time, to the best of our knowledge, that PMMA can be directly incorporated into PPO backbone to produce a soluble and processable copolymer throughout such a wide composition window. Without compatibilizers, the PPO-g-PMMA demonstrates good ability in preparing compatible graft copolymers at 22, 43 and 80 mol% of PMMA whereas only partial miscibility was obtained for 59% and 74%. The detailed investigations of film property and applications are ongoing and will be reported later. 4. Conclusions Employing ATRP methodology on the grafting technique is of great interest since it provides access to new polymeric materials, or surface modifications which could lead to useful applications for which the known polymers are not adequate. From industrial standpoint, it is impor-

tant to generate graft copolymer in situ that can bind the different miscibility components with controlled length or density of grafts with good processability. In this work, we have successfully synthesized a new class of graft copolymer consisting of poly(2,6-dimethyl-1,4-phenylene oxide) as backbone by introducing styrene, acetoxystyrene and methyl methacrylate monomers onto the side chain sequentially. In general, incorporation of various extents of low Tg component yields completely miscible copolymers with decrement of glass transition temperature and thermal stability. The experimental DSC data of glass transition temperature as a function of PS compositions fit quite well for theoretical curve in Couchman equations and the FTIR analysis indicates that enhanced molecular interactions exist in PPO-g-PS when PS molar content is below 40%, which in turn results in higher Tg values than those of polymer blends at the same compositions. In the cases of PPO-g-PMMA, the Tg values decreased almost linearly with increasing length of PMMA graft and the Td,5% values decreased correspondingly due to the inherent thermal stability of PMMA. Despite the immiscibility nature between PPO and PMMA, the graft copolymers exhibit enhanced miscibility of these two components in the amorphous phase over a wide composition range. It is proposed that the chain microstructure of PPO-g-PMMA allows an enhanced segmental mixing on a molecular scale and demonstrates good ability in preparing compatible graft copolymers directly. Acknowledgement Financial supports from National Science Council (NSC 95-2113-M-415-005- MY2) of Taiwan, ROC is gratefully acknowledged. References [1] Bhattacharya A, Misra BN. Prog Polym Sci 2004;29(8):767–814. [2] Hu S, Brittain WJ. Macromolecules 2005;38(15):6592–7. [3] Edmondson S, Osborne VL, Huck WTS. Chem Soc Rev 2004;33(1):14–22. [4] Sheiko SS, Sun FC, Randall A, Shirvanyants D, Rubinstein M, Lee H, et al. Nature 2006;440:191. [5] Neugebauer D, Zhang Y, Pakula T, Sheiko SS, Matyjaszewski K. Macromolecules 2003;36(18):6746–55. [6] Yamada K, Miyazaki M, Ohno K, Fukuda T, Minoda M. Macromolecules 1999;32(2):290–3. [7] Fredrickson GH. Macromolecules 1993;26(11):2825–31. [8] Subbotin A, Saariaho M, Ikkala O, Brinke G. Macromolecules 2000;33(9):3447–52. [9] Muchtar Z, Schappacher M, Deffieux A. Macromolecules 2001;34(22):7595–600. [10] Carlmark A, Malmstrom EJ. Am Chem Soc 2002;16(6):900–1. [11] Borner HG, Beers K, Matyjaszewski K, Sheiko SS, Moller M. Macromolecules 2001;34(13):4375–83. [12] Zhao B, Brittain WJ. Prog Polym Sci 2000;25(5):677–710. [13] Xie HQ, Xie D. Prog Polym Sci 1999;24(2):275–313. [14] Peng D, Zhang X, Huang X. Polymer 2006;47(17):6072–80. [15] Hu D, Cheng Z, Zhu J, Zhu X. Polymer 2005;46(18):7563–71. [16] Shen D, Yu H, Huang YJ. Polym Sci A Polym Chem 2005;43(18):4099–108. [17] Vlcek P, Janata M, Latalova P, Krız J, Cadova E, Toman L. Polymer 2006;47(8):2587–95. [18] Paik H, Gaynor SG, Matyjaszewski K. Macromol Rapid Commun 1998;19(1):47–52. [19] Chen Y, Liu D, Deng Q, He X, Wang X. J Polym Sci A Polym Chem 2006;44(11):3434–43. [20] Kim YW, Lee DK, Lee KJ, Kim JH. Eur Polym J 2008;44(3):932–9.

M. Liang et al. / European Polymer Journal 45 (2009) 2348–2357 [21] Cheng G, Boker A, Zhang M, Krausch G, Muller AHE. Macromolecules 2001;34(34):6883–8. [22] Beers KL, Gaynor SG, Matyjaszewski K, Sheiko SS, Moller M. Macromolecules 1998;31(26):9413–5. [23] Neugebauer D, Zhang Y, Pakula T, Matyjaszewski K. Polymer 2003;44(22):6863–71. [24] Friebe A, Ulbricht M. Langmuir 2007;23(20):10316–22. [25] Lokuge I, Wang X, Bohn PW. Langmuir 2007;23(1):305–11. [26] Xu FJ, Zhao JP, Kang ET, Neoh KG, Li J. Langmuir 2007;23(16):8585–92. [27] Huang CC, Yang MS, Liang M. J Polym Sci A Polym Chem 2006;44(20):5875–86. [28] Matyjaszewski K, Shipp DA, Wang JL, Grimaud T, Pattern TE. Macromolecules 1998;31(20):6836–40. [29] Shipp DA, Wang JL, Matyjaszewski K. Macromolecules 1998;31(23):8005–8. [30] Fox TG. Bull Am Phys Soc 1956;1:23. [31] Gordon M, Taylor JS. J Appl Chem 1952;2:493. [32] Kwei TK. J Polym Sci Polym Lett Ed 1984;22:307. [33] Simha R, Boyer RF. J Chem Phys 1962;37:1003–7.

2357

[34] Couchman PR. Macromolecules 1978;11(6):1156–61. [35] Couchman PR, Karasz FE. Macromolecules 1978;11(1):117–9. [36] Kroschwitz JI, Mark HF. Encyclopedia of polymer science and technology, vol. 12. 3rd ed. New York: Wiley-Interscience; 2003– 2004. p. 168–74. [37] Djordjevic MB, Porter RS. Polym Eng Sci 1983;23:650. [38] Koenig JL, Tovar Rodriquez MJM. Appl Spectrosc 1981;35(6): 543–8. [39] Benedetti E, Moscatelli D, Vergamini P. Macromol Chem Phys 2002;203(10–11):1497–500. [40] Kuo SW, Huang CF, Tunga PH, Huang WJ, Huang JM, Chang FC. Polymer 2005;46(22):9348–61. [41] Zhang Z, Mo Z, Zhang H, Zhang Y, Na T, An Y, et al. J Polym Sci Part B Polym Phys 2002;40(17):1957–64. [42] Picchioni F, Casentini E, Passaglia E, Ruggeri G. J Appl Polym Sci 2003;88(11):2698–705. [43] Brandrup J, Immergut EH, Grulke EA, editors. Polymer handbook, 4th ed.. New York: Wiley; 1999. p. VII/684. [44] Song M, Hammiche A, Pollock HM, Houston DJ, Reading M. Polymer 1996;37(25):5661–5.