Synthesis of Poly(ethyl acrylate) by Single Electron Transfer-Degenerative Chain Transfer Living Radical Polymerization in Water Catalyzed by Na2S2O4

Synthesis of Poly(ethyl acrylate) by Single Electron Transfer-Degenerative Chain Transfer Living Radical Polymerization in Water Catalyzed by Na2S2O4 JORGE F. J. COELHO,1,2 ERICA Y. CARVALHO,1 DINA S. MARQUES,1 ANATOLIY V. POPOV,3 VIRGIL PERCEC,4 PEDRO M. F. O. GONC¸ ALVES,2 M. H. GIL1 1Chemical Engineering Department, University of Coimbra, Po´lo II, Pinhal de Marrocos, 3030-290 Coimbra, Portugal 2CIRES SA – Companhia Industrial de Resinas Sinte´ticas, Apartado 20, Samoqueiro – Avanca, 3864-752 Estarreja, Portugal 3Department of Radiology, University of Pennsylvania, Philadelphia, Pennsylvania 4 Roy and Diana Vagelos Laboratories, Department of Chemistry, University of Pennsylvania, Philadelphia, Pennsylvania Received 20 July 2007; accepted 18 September 2007 DOI: 10.1002/pola.22393 Published online in Wiley InterScience (). ABSTRACT: Living radical polymerization of ethyl acrylate was achieved by single-electron-transfer/degenerative-chain transfer mediated living radical polymerization in water catalyzed by sodium dithionite. The plots of number-average molecular weight versus conversion and ln[M]0/[M] versus time are linear, indicating a controlled polymerization. This method leads to the preparation of a,x-di(iodo)poly(ethyl acrylate) (a,x-di(iodo)PEtA) macroinitiator that can be further functionalized. The molecular weight distributions were determined using a combination of three detectors (TriSEC): right-angle light scattering, a differential viscometer and refractive index. The method studied in this work represents a possible route to prepare well-tailored macromolecules made of ethyl acrylate in environmental friendly reaction medium. To the best of our knowledge there is no previous report dealing with the synthesis of PEtA by any LRP approach in aqueous medium. Furthermore, the method described in this article was successfully applied in pilot scale reactions under industrial production conditions. VC 2007 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 46: 421–432, 2008 Keywords: degenerative transfer; ethyl acrylate; kinetics; living polymerization; poly(ethyl acrylate); single electron transfer; tacticity; telechelics INTRODUCTION Living Radical Polymerization is one of most powerful tools to prepare well tailored architectures that fit emerging applications. This strategy combines the exceptional flexibility and potential of the living approaches with the radical polymerization advantages. The most common strategies include reversible addition fragmentation chain transfer (RAFT),1 nitroxide-mediated living radical polymerization (NMP)2 and metalcatalyzed living radical polymerization.3 Because of the remarkable interest from the academia and industrial world, the different methods have witnessed important developments over the last decade. The reaction conditions become more attractive and easy to perform in the industrial Correspondence to: J. F. J. Coelho (E-mail: jcoelho3@ ) Journal of Polymer Science: Part A: Polymer Chemistry, Vol. 46, 421–432 (2008) VC 2007 Wiley Periodicals, Inc. 421 environment. On this matter, the development of a new strategy based on reversible activationdeactivation step required to accomplish LRP by combination of competitive single-electron-transfer (SET) and degenerative-chain transfer4 is a clear example. Discovered by Percec, Popov and coworkers,4–6 this strategy has proved to be effective in the polymerization of activated7–9 and non-activated monomers.4–6,8–11 The incorporation of acrylic based polymers was commonly used to the rubber toughening of glassy polymers, exerting their modifying influence in improving the toughness of the blended systems.12 The acrylic monomers are particularly useful to tailor the properties of other materials due to their characteristics, such as water repellence,13 good filmability,13 transparency,13 chemical resistance, and low cost. There are several strategies reported to incorporate PEtA segments, such as blending,12,14 grafting,13 and radical copolymerization.15,16 The main problems associated to the above approaches are concerned to the lack of control over the final macrostructures that typically depend on the relative amount of monomers used and reactivities. For that reasons, the strategies that allow the preparation of telechelic polymers are highly demanded. Regarding the synthesis of PEtA via a living route there is little information available in the literature. Wu et al.17 synthesized star-shaped polymer (polystyrene)n-[poly(ethyl acrylate)m] from a hydrobrominated PS macroinitiator via atom transfer radical polymerization (ATRP) in toluene at 90 8C. Shi18 proposed the synthesis of diblock copolymers PS-b-PEtA from a PS macroinitiator with an active bromine in the x-end of the chain by ATRP at 90 8C. Jianying19 studied the random copolymerization of styrene and ethyl acrylate at 125 8C using TEMPO as a mediator. Finally, the PEtA was also prepared by ATRP20 at 90 8C following a procedure developed for other vinylic monomers21 several years ago. The rigorous reaction conditions used to prepare the macroinitiator and final structures are incompatible with any attempt to prepare those structures in large scale. The aim of the present work is to study the synthesis of PEtA by single electron transfer-degenerative chain transfer mediated living radical polymerization (SET-DTLRP) in water. To the best of our knowledge there are no reports about synthesis of a,x-di(iodo)PEtA macroinitiators in aqueous medium that can be further modified. The modification and functionalization of the polyacrylate macroinitiators is extremely useful, aiming to the formation of new families of materials with improved properties. Moreover, the synthesis of polymers containing segmented blocks has noticed an increasing attention from both scientific and technological points of view.16 From the mechanistic standpoint, the possibility of synthesizing an acrylate with short alkyl chain in aqueous medium is also extremely interesting due to low steric stabilization resonance expected from the short alkyl side chain over the iodine dormant species. The SET-DTLRP approach have presented promising results in its industrial implementation, leading to flexible materials that are able to replace some commercial products that are made from a thermoplastic blended with free plasticizers,11,22–24 being foreseeable in its commercialization, in large production in short time period. The development and improvement of methods that involve commercially available compounds at the same time inexpensive and easy to handle is extremely important to foresee the industrial implementation of the LRP processes. Preliminary experiments leading to the activation of the a,x-di(iodo) chain ends were successfully carried out. The polymer solutions in THF were also characterized by multi-detector size chromatography (TriSEC), determining by this way the relationship between the intrinsic viscosity versus molecular weight and radius of gyration versus molecular weight. EXPERIMENTAL Materials THF HPLC-grade uninhibited, basic alumina, iodoform (99%), sodium dithionite (85%), sodium bicarbonate (99%), and EtA were purchased from Sigma-Aldrich. EtA was purified through a basic Al2O3 column just before polymerization. p-Toluenesulfinic acid, sodium salt, hydrate (pTsNa; 98%) was purchased form Acros Organics. The Polystyrene standards for TriSEC measurements were purchased from Polymer Laboratories. Poly (vinyl alcohol) (PVA) (average Mw 85,000– 124,000; 87–89% hydrolyzed) was purchased from Sigma-Aldrich. Hydroxypropyl methylcellulose—Methocel F50 (MF50) was purchased from Dow Chemical Company. The other compounds 422 COELHO ET AL. Journal of Polymer Science: Part A: Polymer Chemistry DOI 10.1002/pola were ordered from Sigma-Aldrich and used as received. Polymerization of EtA via SET/DTLRP The typical procedure was performed as described (example ratio [EtA]0/[CHI3]0 ¼ 100). A 50 mL Ace Glass 8645#15 pressure tube equipped with bushing and plunger valve was charged with 9 mL of deionized water, 48.3 mg of a 3% PVA solution (490 ppm), and 33.4 mg of a 1.86% MF50 solution (210 ppm). The content was stirred and bubbled with nitrogen for 10 min. Then other compounds were added: catalyst (Na2S2O4, 191.57 mg, 1.10 mmol), initiator (CHI3, 108.30 mg, 0.28 mmol), buffer (NaHCO3, 33.51 mg, 0.40 mmol), additive (pTsNa, 107.94 mg, 0.55 mmol) and EtA (3 mL, 27.75 mmol). The tube was closed, frozen in MeOH/dry ice and degassed through the plunger valve by applying cycles of reduced pressure followed by filling the tube with inert gas for 20 times at
40 8C. The valve was closed and the reaction was kept in a water bath for 1 h at 35 8C agitated with a magnetic stirring bar. At the end of the reaction, a small aliquot was taken for TriSEC measurements. The polymer was placed in a preweighted vial and the remaining part of the tube was carefully washed with THF and placed in a different vial. After drying in a vacuum oven until the weight was constant, both vials were weighted to determine the final conversion (68%). Characterization Techniques The chromatography parameters of the samples were determined using a HPSEC, Viscotek (Dual detector 270, Viscotek, Houston, USA) with a differential viscosimetry (DV), right angle laser light-scattering (RALLS; Viscotek), and RI (Knauer K-2301). The column set consisted of a PL 10 lm guard column (50 mm 3 7.5 mm) followed by two MIXED-B PL columns (300 mm 3 7.5 mm, 10 lm). HPLC pump (Knauer K-1001) was set with a flow rate of 1 mL/min. The eluent (THF) was previously filtered through a 0.2 lm filter. The system was also equipped with a Knauer on-line degasser. The tests were done at 30 8C using an Elder CH-150 heater. Before the injection (100 lL), the samples were filtered through a PTFE membrane with 0.2 lm pore. The system was calibrated with narrow polystyrene standards. The differential refractive index of PEtA for 670 nm was determined (dn/dc ¼ 0.061). The analysis of light scattering data by Viscotek’s software was done assuming that the second virial coefficient was zero, considering the low solution concentrations used in this work. The 1H NMR spectra (500 MHz) were recorded in a Bruker DRX 500 spectrometer at 32 8C in CDCl3 with tetramethylsilane as internal standard. Diad tacticities of the polymer were determined from 1H NMR as it was described elsewhere.25 Dynamical mechanical thermal analysis (DMTA) of thick specimens (15.20 mm 3 7.45 mm 3 1.20 mm) were performed using a Triton Tritec 2000 in the constrain layer damping mode using two frequencies (1 Hz and 10 Hz), with a standard heating rate of 2 8C min
1 . The Tg was determined as the peak in tan d (Tan d ¼ E@/E0 ) where E@ and E0 are the loss and storage modulus, respectively. RESULTS AND DISCUSSION The kinetics experiments carried out in this work are summarized in Table 1. The points in the kinetics shown in Figure 1 were obtained gravimetrically according to the procedure described (vide infra). Figure 1(a) shows the kinetic plots for Na2S2O4/NaHCO3-catalyzed LRP of EtA at 35 8C for DP ¼ 100. The PVA 88 and MF50 were used as suspending agents to stabilize the ethyl acrylate droplets. Table 1. Molar Ratios of Reagents and Amounts of Surfactants and Water Studied in the Kinetic Experiments Carried Out by SET-DTLRP No. EtA CHI3 Na2S2O4 NaHCO3 pTsNa PVA88 MF50 H2O mol mol mol mol mol ppm ppm mL 1 100 1 4 1. 2 250 1 4 1. 3 500 1 4 1. . SYNTHESIS OF POLY(ETHYL ACRYLATE) BY SET-DTLRP 423 Journal of Polymer Science: Part A: Polymer Chemistry DOI 10.1002/pola The kinetic data exhibit two different slopes on the ln[M]0/[M] versus polymerization time. The first slope kp1 ¼ 1.24 h
1 represents the region where monomer diffusion is not limited by the viscosity of the reaction mixture. Because of the increase of conversion the reaction medium becomes more viscous and eventually leads to the formation of a solid phase. This process leads to the appearance of a second stage that presents a kinetic constant (kp2) several times lower than kp1. The described trend was also observed in the cases of SET-DTLRP of other monomers such as vinyl chloride,4–6,9,26 2-ethylhexyl acrylate,9 tert-butyl acrylate,9 and butyl acrylate.8 The results presented in Figure 2, for DP ¼ 500 and DP ¼ 1000 show only one kinetic constant, contrarily to the results observed for DP ¼ 100 and DP ¼ 250. The reason for this result is not fully understood at the moment. However, such observation could be related to slower reaction during the first hours of polymerization, which leads to a slower increase of viscosity inside the droplet that is known to decrease the reaction rate. The termination reactions can be neglected, since the kinetic results show a first-order kinetic of the reaction relatively to the monomer concentration. Such behavior is observed for the kinetics presented in this article regardless the ratio [monomer]0/[initiator]0 considered (Figs. 1 and 2). Furthermore, the results suggest a linear dependence of the molecular weight determined by TriSEC (Mn,TriSEC) versus the theoretical molecular weight (Mn,th). These two features of the SETDTLRP of EtA support the living polymerization Figure 1. Na2S2O4/NaHCO3-catalyzed LRP of EtA initiated with iodoform in H2O in the presence of SA Methocel F50 and PVA 88, [EtA]0/[H2O]0 ¼ 1/3 (v/v): (a) [EtA]0/ [CHI3]0/[Na2S2O4]0/[pTsNa]0/[NaHCO3]0 ¼ 100/1/4/2/1.45 (mol/mol/mol/mol); [MethocelF50]/[PVA 88] ¼ 210/490 (ppm/ppm, w/w relative to EtA), 35 8C; (b) [EtA]0/[CHI3]0/ [Na2S2O4]0/[pTsNa]0/[NaHCO3]0 ¼ 250/1/4/2/1.45 (mol/mol/mol/mol); [Methocel F50]/ [PVA 88] ¼ 210/490 (ppm/ppm, w/w relative to EtA), 35 8C. 424 COELHO ET A