Carbohydrate Polymers 141 (2016) 143–150
Contents lists available at ScienceDirect
Carbohydrate Polymers journal homepage: www.elsevier.com/locate/carbpol
In situ development of self-reinforced cellulose nanocrystals based thermoplastic elastomers by atom transfer radical polymerization Juan Yu a , Chunpeng Wang a,b , Jifu Wang a,b,∗ , Fuxiang Chu a,b,∗ a Institute of Chemical Industry of Forestry Products, CAF; National Engineering Laboratory of Biomass Chemical Utilization; Key and Laboratory of Forest Chemical Engineering, SFA; Key Laboratory of Biomass Energy and Material, Nanjing, Jiangsu Province 210042, China b Institute of Forest New Technology, CAF, Beijing 100091, China
a r t i c l e
i n f o
Article history: Received 5 September 2015 Received in revised form 25 December 2015 Accepted 5 January 2016 Available online 6 January 2016 Keywords: Cellulose nanocrystals Atom transfer radical polymerization (ATRP) Thermoplastic elastomers
a b s t r a c t Recently, the utilization of cellulose nanocrystals (CNCs) as a reinforcing material has received a great attention due to its high elastic modulus. In this article, a novel strategy for the synthesis of self-reinforced CNCs based thermoplastic elastomers (CTPEs) is presented. CNCs were ﬁrst surface functionalized with an initiator for surface-initiated atom transfer radical polymerization (SI-ATRP). Subsequently, SI-ATRP of methyl methacrylate (MMA) and butyl acrylate (BA) was carried out in the presence of sacriﬁcial initiator to form CTPEs in situ. The CTPEs together with the simple blends of CNCs and linear poly(MMAco-BA) copolymer (P(MMA-co-BA)) were characterized for comparative study. The results indicated that P(MMA-co-BA) was successfully grafted onto the surface of CNCs and the compatibility between CNCs and the polymer matrix in CTPEs was greatly enhanced. Specially, the CTPEs containing 2.15 wt% CNCs increased Tg by 19.2 ◦ C and tensile strength by 100% as compared to the linear P(MMA-co-BA). © 2016 Elsevier Ltd. All rights reserved.
1. Introduction As the most ubiquitous and abundant renewable polymer resource, cellulose has received tremendous attention resulting from its intrinsic properties such as environmental friendliness, biodegradability, high thermal stability, low density and good mechanical strength (Bruce, 2014; Roy, Semsarilar, Guthrie, & Perrier, 2009). Cellulose nanocrystals (CNCs) are the crystalline residues generating from a variety of renewable sources such as wood, cotton, ramie, by acid hydrolysis with either sulfuric or hydrochloric acid (Boujemaoui, Mongkhontreerat, Malmström, & Carlmark, 2015; Lu & Hsieh, 2010; Zoppe et al., 2010). During the past several decades, CNCs have been a subject of increasing interest in the ﬁeld of nanocomposites, due to their attractive mechanical properties, unique morphology, low density, nanoscale dimensions, etc. Particularly being frequently used as a reinforcing material, nano-sized CNCs were reported to reinforce the polymer matrices including poly(oxyethylene) (Azizi Samir, Alloin, Gorecki, Sanchez, & Dufresne, 2004), polypropylene (Li, Xiao,
∗ Corresponding authors at: Institute of Chemical Industry of Forestry Products, CAF; National Engineering Laboratory of Biomass Chemical Utilization; Key and Laboratory of Forest Chemical Engineering, SFA; Key Laboratory of Biomass Energy and Material, Nanjing, Jiangsu Province 210042, China. E-mail addresses: [email protected]
(J. Wang), [email protected]
(F. Chu). http://dx.doi.org/10.1016/j.carbpol.2016.01.006 0144-8617/© 2016 Elsevier Ltd. All rights reserved.
Zheng, & Xiao, 2011), polyester (Hu, Lin, Chang, & Huang, 2015), polyurethane (Yao et al., 2014), starch-based polymers (Lu, Weng, & Cao, 2006), poly(caprolactone) (Habibi et al., 2008), and poly(lactic acid) (Arrieta, Fortunati, Dominici, Lopez, & Kenny, 2015; Fortunati et al., 2012). Thermoplastic elastomers (TPEs) are a class of multi-functional polymeric materials generally possessing thermoplastic and elastomeric properties, which have been widely utilized in our daily life as engineering plastics. Polystyrene-b-polybutadiene-bpolystyrene (SBS) and polystyrene-b-polyisoprene-b-polystyrene (SIS) triblock copolymers are mostly used commercial TPEs (Jerome et al., 1987; Tong et al., 2000). Due to the stringent condition in manufacturing via ionic polymerization, and the presence of double bonds in the diene parts that are prone to thermal and oxidative degradation, linear or star shape “all acrylic” TPEs, including block copolymers of poly(methyl methacrylate-bbutyl acrylate-b- methyl methacrylate) and random copolymers of poly(methyl methacrylate-co-butyl acrylate) (P(MMA-co-BA)) and poly(butyl acrylate-co-acrylonitrile), were developed by controlled radical polymerization such as atom transfer radical polymerization (ATRP)(Jiang, Wang, Qiao, Wang, & Tang, 2013; Nese et al., 2010; Shin, Lee, Tolman, & Hillmyer, 2012) and reversible additionfragmentation chain transfer (RAFT) polymerization(Luo, Wang, Zhu, Li, & Zhu, 2010; Mahanthappa, Bates, & Hillmyer, 2005; Satoh, Lee, Nagai, & Kamigaito, 2014). These polymers offer an improvement of their thermal properties. Recently, a novel architecture of
J. Yu et al. / Carbohydrate Polymers 141 (2016) 143–150
TPEs on a grafting strategy was reported (Jiang et al., 2013; Liu et al., 2014). These novel TPEs were prepared by the random copolymerization of methyl methacrylate (MMA) and butyl acrylate (BA) from cellulose via electron transfer for atom transfer radical polymerization (ARGET ATRP) in homogeneous conditions. With the existence of 0.9−3.4 wt% cellulose in copolymers, these TPEs exhibited desired mechanical properties that were superior to those of linear copolymer counterparts (P(MMA-co-BA)) under the same condition. With the similar strategy, copolymers based on rosin and fatty acid-derived monomers were also grafted to cellulose with the aim to fabricate sustainable TPEs. However, the relatively poor mechanical properties of these acrylic based TPEs limited the scope of their application. Inspired by the reinforcing application of CNCs and the multiarm star shape acrylic based TPEs obtained by using a graft random copolymer architecture (Dufour et al., 2008; Nese et al., 2010; Yu et al., 2015), we herein report a strategy to fabricate a novel self-reinforced CNCs based thermoplastic elastomers (CTPEs) by a facile surface-initiated ATRP (SI-ATRP) of MMA and BA on the ATRP initiators immobilized surface of CNCs with the aid of sacriﬁcial initiator (ethyl 2-bromoisobutyrate). After the polymerization, the poly(MMA-co-BA) copolymer grafted CNCs (CNCs-g-P(MMAco-BA)) are in situ mixed with linear P(MMA-co-BA) (grown from sacriﬁcial initiator) and this mixture is directly evaluated as CTPEs. This is a simple and convenient procedure to prepare CTPEs, by which grafted polymer would not be puriﬁed by the process of extraction of linear polymers and not be blended with another kind of polymer matrix (Bruce, 2014; Li et al., 2011). The effects of CNCs on the mechanical properties of CTPEs were investigated. To the best of our knowledge, this is the ﬁrst report to synthesize CTPEs, which provides a convenient, simple and economic way for the reinforcement of polymers. 2. Experimental
2-bromoisobutyryl bromide (2.38 mL, 0.0185 mol) was added dropwise and then stirred at room temperature for 24 h. The resulting suspension was centrifuged followed by continuously washed with saturated sodium bicarbonate, ethanol, dichloromethane, hexane, and dried at 50 ◦ C under vacuum for 12 h until constant weight (CNCs-Br2). 2.3. Preparation of CTPEs The synthesis of CTPEs was carried out using a similar procedure as previous reported (Jiang et al., 2013). A mixture of MMA (3.05 mL, 27.8 mmol), BA (4.95 mL, 34.0 mmol), PMDETA (13.0 L, 0.0619 mmol), CNCs-Br2 (15.4 mg), ethyl 2bromoisobutyrate (EBiBr) (9.2 L), THF (9 mL) and DMF (1 mL) were charged into a round bottom ﬂask. After three freeze-pump-thaw cycles, the mixture was transferred into a Schlenk ﬂask that contained Cu(I)Br (9.0 mg, 0.0619 mmol) under nitrogen. The Schlenk ﬂask was placed in an oil bath set at 80 ◦ C. The conversion of monomers was determined by 1 H NMR using DMF as an internal reference. After the reaction reached the designed conversion, the polymerization was stopped by diluting the reaction mixture with THF. The suspension was precipitated in methanol at least three times to remove Cu catalyst and unreacted monomers. The resulting products (CTPEs) were collected and dried under vacuum at 60 ◦ C until constant masses. A series of CTPEs with different contents of PMMA or PBA were prepared in a similar way, so were the linear pure copolymers P(MMA-co-BA) which were prepared by just using ethyl 2-bromoisobutyrate (EBiBr) as an initiator. For comparative purpose, the grafted CNCs (CNCs-g-P(MMA-coBA) and linear P(MMA-co-BA) copolymers were separated by the following procedure. CNCs-g-P(MMA-co-BA was obtained by the centrifugation of CTPEs in THF for many times and Soxhlet extracted with dichloromethane to remove remaining P(MMA-co-BA) grafts. The centrifuged suspension was then precipitated in methanol to obtain pure P(MMA-co-BA) copolymers.
2.1. Materials CNCs were prepared according to a published procedure (Morandi, Heath, & Thielemans, 2009) and the characterization of CNCs was shown in Fig. S1. N,N,N ,N ,N pentamethyldiethylenetriamine (PMDETA, 99%), methyl methacrylate (stabilized, 99%), n-butyl acrylate (stabilized, 99%), Cu(I)Br (99.999%), were purchased from Aldrich and used as received. Ethyl 2-bromoisobutyrate (EBiBr, 98%), 2bromoisobutylryl bromide (98%) and dimethylaminopyridine (DMAP, 99%) were obtained from Shanghai Orgpharma Chemical Co., Ltd and used as received. Tetrahydrofuran (THF) and N,N-Dimethylformamide (DMF) were dried over 4A molecular sieves and then distilled before use. All the other solvents were AR reagents and purchased from Nanjing Reagent Chemical Co., Ltd. 2.2. Preparation of CNCs initiators (CNCs-Br) The procedure used for synthesis of CNCs initiators was divided into two steps according to the previous report (Majoinen, Walther, McKee, Kontturi, & Aseyev, 2011). Firstly, CNCs (2.0 g) were placed in 50-mm culture dish which was put into a 2 L saturate tank. 4 mL of 2-bromoisobutyryl bromide was then added. The tank was closed and protected from light for 24 h at 30 ◦ C. The mixture was sequentially washed by different kinds of solvents (distilled water, ethanol, dichloromethane, hexane) and dried at 40 ◦ C for use (CNCsBr1). Second, the dried CNCs-Br1 (1.0 g) obtained in the ﬁrst step was immersed in 100 mL DMF and mild sonicated for 10 min. The suspended CNCs-Br1 in DMF was cooled to 0 ◦ C in an ice/water bath and DMAP (2.33 g, 0.0185 mol) were added. Subsequently,
2.4. Characterization FT-IR analysis was performed using a Nicolet iS10 FT-IR spectrometer by an Attenuated Total Reﬂectance method. 1 H NMR analysis was carried out on a Bruker DMX 300 NMR spectrometer, and CDC13 was used as solvent. Gel permeation chromatography (GPC) was performed at 40 ◦ C on Malvern Viscotek 3580 System equipped with Viscotek GPC2502 refractive detector and a GPC1007 pump with THF at 1 mL/min, the columns were T6000M, General Mixed Org 300 mm × 7.8 mm (CLM3009). Monodispersed polystyrene (PS) was used as the standard to generate the calibration curve. All samples were ﬁltered over a microﬁlter with a pore size of 0.22 m (Nylon, Millex-HN 13 mm Syringes Filters, Millipore). Elemental analysis were performed at ThermoFisher Flash 2000 HT Elemental Analyzer. The Zetasizer (Nano ZS, Malvern Instruments Ltd., UK) was used to measure the size of CNCs in diluted suspension. TEM was performed on a JEOL-1010 transmission electron microscope at an acceleration voltage of 100 kV. The samples for TEM were prepared by placing the copolymer solutions on copper grids previously coated with carbon and dried one day in air. The samples were directly observed by TEM. Atomic force microscopy (AFM) images were collected with a SPM-9600 AFM (Shimadzu, Japan) using AC240 (OLYMPUS) probes with a nominal resonant frequency at 70 kHz and a nominal spring constant at 2 N/m. All scans were performed in air with on silicon wafers or mica under the phase mode. For the observation of CNCs-g-P(MMA-co-BA), solutions in THF with the concentration of 0.1 mg/mL were spin-cast on fresh mica surface or silicon wafer, For the investigation of the surface
J. Yu et al. / Carbohydrate Polymers 141 (2016) 143–150
topography of CTPEs, suspension in THF with the concentration of 1 mg/mL was spin-cast on silicon wafer, then dried at 40 ◦ C for 24 h. Differential Scanning Calorimetry (DSC) was run on Perkin Elmer Diamond Differential Scanning Calorimeter under nitrogen atmosphere. The temperature was increased at a rate of 10 ◦ C/min and decreased at same rate during the ﬁrst scan cycle. The data were collected from the second heating scan with a heating rate of 10 ◦ C/min for measurement of glass transition temperature (Tg ). Thermogravimetric analysis (TGA) was performed on NETZSCH STA 409 PC ranging from 30 to 600 ◦ C at a rate of 10 ◦ C/min under nitrogen gas at a ﬂow rate of 100 mL/min. Dynamic mechanical analysis, as well as tensile creep and recovery measurement were performed on a dynamic mechanical analyzer (DMA Q800, TA Instruments) in tension mode and in tensile creep and recovery mode, respectively. In dynamic mechanical analysis, the dumbbell samples were heated from −50 to 60 ◦ C at a heating rate of 2 ◦ C/min. In tensile creep and recovery test, the test specimens were prepared in the shape of rectangular ﬁlm strips, typically 6 mm × 25 mm with the thickness ranging between 0.30 mm and 0.50 mm from the ﬁlms. The experiments were conducted at a frequency of 1 Hz in tensile creep and recovery mode and subjected to a constant stress level of 0.02 MPa for 10 min. During this period, the strain in response to the stress was recorded. The tensile stress was then released after 10 min and the strain recovery was recorded for at least another 100 min. The strain-time curve is converted into a compliance-time by the following equation: J(t) = ε(t)/, where ε(t) is the time dependent recoverable strain, and is the stress applied in the creep zone. Then the steady state part of the curve was linear ﬁtted and the intercept of ﬁtted line represented the elastic creep compliance JS0 . The percentage of elastic strain recovery () is calculated as = [(Jmax − Jnr )/Jmax ] × 100%, where Jmax represents the compliance when the stress is removed and Jnr represents the compliance at the end of the test. The mechanical tests were done at room temperature using CMT7504 universal testing machine with the crosshead speed of 50 mm/min and the load cell was 250N. The ﬁlm samples were prepared by casting the suspension of copolymers on the polytetraﬂuoroethylene dish and dried at 80 ◦ C under vacuum for 24 h. Dumbbell specimens of 75 mm × 25 mm with the thickness ranging from 0.25 to 0.45 mm were cut from the cast ﬁlm. The results were based on ﬁve independent measurements of each sample performed at the same condition. The tensile cyclic processing was conducted as follows: in each step, once the sample reached the targeted maximum strain, the crosshead direction was reversed and the sample strain was decreased at the same nominal strain rate (∼0.02 s−1 ) until zero stress was achieved. After that, the crosshead was immediately reversed, and the sample was then extended again at the same constant strain rate until it reached the next targeted maximum strain.
3. Results and discussion 3.1. Preparation of CNCs initiator (CNCs-Br) The preparation of CNCs initiator was conducted in two steps. As shown in FT-IR spectra (Fig. 1(a)), after the introduction of 2bromoisobutyryl groups on the surface of CNCs in the ﬁrst step, the characteristic absorption peak at 1730 cm−1 corresponding to stretching vibrations of the C O in the spectrum of CNCs-Br1 was observed. However, the intensity of peak of C O at 1730 cm−1 was relatively weak in the ﬁrst step (Fig. 1(a) CNCs-Br1), indicating that only a portion of hydroxyl groups on the surface of CNCs were substituted by the 2-bromoisobutyryl groups. Therefore, the second step of esteriﬁcation of CNCs and 2-bromoisobutyryl bromide was imperative to perform and resulted in CNCs-Br2 with more
Fig. 1. (a) FT-IR spectra of CNCs, grafted CNCs and CTPEs (Table 1, Entry 2); (b) XRD spectra of CNCs and grafted CNCs; (Inset) state of suspension of CNCs-Br2(1)and CNCs-g-P(MMA-co-BA)(2)in THF (1 mg/ml).
2-bromoisobutyryl groups. The peak at around 1730 cm−1 was clearly present after second step reaction. The bromine contents of CNCs-Br1 and CNCs-Br2 were determined by elemental analysis to be 4.02 wt% and 8.57 wt %, respectively. To further conﬁrm that the esteriﬁcation of CNCs took place on the surface, X-ray Diffraction (XRD) was used to measure the main crystal planes of CNCs and grafted CNCs. Fig. 1(b) shows the principal peaks with different intensities from XRD patterns with 2 values at 14.96◦ , 16.70◦ , 20.69◦ , 22.89◦ , 34.70◦ , which represent ¯ 0 2 1, 0 0 2, 0 4 0) of CNCs (Hon & ﬁve lattice planes (1 0 1, 1 0 1, Yan, 2001). After the immobilization of 2-bromoisobutyryl groups on the surface of CNCs, CNCs-Br1 obtained in step1 and CNCsBr2 obtained in step2 still retained crystal structures, as indicated by the ﬁve diffraction peaks with similar shape and intensity on XRD proﬁles. These results veriﬁed that chemical modiﬁcation just occurred on the surface of CNCs and did not signiﬁcantly alter the crystallite region of CNCs. 3.2. Preparation of CTPEs The grafting of P(MMA-co-BA) from CNCs was performed by SI-ATRP with the aid of CNCs-Br. As demonstrated in previous reports (Carlmark & Malmström, 2002; Lee et al., 2004; Zhou et al., 2005), SI-ATRP of cellulose (ﬁlter paper, jute ﬁbers, CNCs, etc.) was carried out in heterogeneous media, where a sacriﬁcial initiator
J. Yu et al. / Carbohydrate Polymers 141 (2016) 143–150
Table 1 Reaction conditions and results for the preparation of CTPEs. Entry
Mn/PDI 1 2 3 4 5 6 7 8 9
P(MMA450 -co-BA) CNC0.53 -MMA450 -2 CNC1.07 -MMA450 -2 CNC2.15 -MMA450 -2 CNC0.70 -MMA450 -2 CNC0.47 -MMA450 -1 P(MMA400 -co-BA) CNC0.44 -MMA400 -2 CNC0.97 -MMA400 -2
0/1/450/550 0.24/1/450/550 0.48/1/450/550 0.96/1/450/550 0.48/1/450/550 0.13/1/450/550 0/1/400/600 0.24/1/400/600 0.48/1/400/600
44,000/1.22 43,000/1.24 42,000/1.32 32,000/1.31 36,000/2.00 39,000/1.59 47,000/1.27 44,000/1.25 41,000/1.19
47.6 40.6 40.3 40.0 61.4 46.1 36.8 48.7 44.0
0 0.53 1.07 2.15 0.70 0.47 0 0.44 0.97
50.26 50.69 51.08 46.94 47.21 47.32 47.54 43.17 44.63
1.01 1.04 1.07 0.93 0.91 0.91 0.91 0.77 0.82
19.4 21.0 22.1 16.5 14.0 6.3 11.0 0.5 1.5
a Chemical formula is represented by a generic nomenclature of CNCX -MMAY -Z, in which X stands for wt% of CNCs in CTPEs, Y stands for MMA feed molar number, and Z of 1or 2 refers to CTPEs initiated with CNCs-Br1 or CNCs-Br2. b Molar feed ratio of IC /I/MMA/BA, IC stands for CNCs-Br. c Mn and PDI of P(MMA-co-BA) in CTPEs. d Measured by gravimetry and 1 H NMR spectroscopy. e r = mMMA /mBA ; mMMA and mBA are mass compositions of MMA and BA in copolymers, respectively.
(such as ethyl 2-bromoisobutyrate, EBiBr) was usually added to produce a sufﬁcient concentration of radical species in order to achieve a good control of ATRP, due to low concentrations of initiating sites on the surface of CNCs (Tizzotti, Charlot, Fleury, Stenzel, & Bernard, 2010). Generally, the linear polymers initiated by a sacriﬁcial initiator are removed by centrifugation followed by Soxhlet extraction with different solvents to obtain pure grafted cellulose copolymers (Li et al., 2011; Majoinen, Walther, McKee, Kontturi, Aseyev, Malho, et al., 2011; Morandi et al., 2009). Consequently, EBiBr (sacriﬁcial initiator) was employed in our work to aid the performance of SI-ATRP for the grafting of P(MMA-co-BA) from the surface of CNCs. The linear P(MMA-co-BA) generated from EBiBr was retained in polymer matrix to form the CTPEs. This approach is a more convenient, simple and economic method for in situ preparation of CTPEs compared with previously published methods (Li et al., 2011; Ma, Zhang, Meng, Anusonti, & Wang, 2015). A series of CTPEs with different concentrations of initiating sites of CNCs were then prepared by SI-ATRP of MMA and BA under the same molar ratios of ([MMA]/[BA]/[CuBr]/[Sacriﬁcial initiator]) and the same reaction conditions. The GPC traces (Fig. S2) of copolymers with or without CNCs-Br, and the molecular weight information (Table 1) showed that the utilization of sacriﬁcial initiator can facilitate a good control of molecular weight with molecular weight distribution (PDI) values