Received Jan. 28, 2007.

Abstract Under the assistance of the efficient transfer-reaction between the dithiocarbonate group and radicals in RAFT polymerization, the graft copolymerization of methyl acrylate (MA) onto chitosan with high grafting efficiency was performed. The effects of polymerization conditions, such as initiator concentration, monomer concentration, reaction time and temperature, on grafting parameters were studied. The graft copolymer was confirmed by a series of characteristic techniques including Fourier transform infrared spectroscopy (FTIR), X-ray diffraction (XRD), and thermogravimetric analysis (TGA). The experimental results have shown that the introduction of the dithiocarbonate groups onto chitosan has efficiently increased the graft efficiency, via minimizing the formation of homo-polymers in graft copolymerization.
Keywords Chitosan; Graft copolymerization; Reversible addition-fragmentation chain transfer (RAFT) polymeization

1. INTRODUCTION
Chitosan, N-deacetylated derivative of chitin, is extensively studied in recent years, because it is biodegradable, biocompatible, nontoxic cheap and abundant in nature. However, as it is well known, chitosan is a type of semi-crystal and compact macromolecule whose insolubility in water limits its application. Fortunately, the structure of chitosan contains a large number of reactive groups such as hydroxyl and amino groups; it can thus be modified by various chemical reactions ^[1-4]. A lot of researches were concentrated on the chitosan modification through copolymerization of chitosan and vinyl monomers for the preparation of new materials, which can endow chitosan with special properties and versatile applications ^[5]. Conventionally, vinyl monomers are grafted onto chitosan using various induction methods such as redox systems, radiation, microwave and so on, to conduct the polymerization ^[6-8]. Graft efficiency is the most important parameter in graft copolymerization, however，it is not mentioned or in a low level in most of the previous literatures ^[9-14]. For example, chitosan-g-polystyrene initiated with ⁶⁰Co gamma radiation and chitosan-g-polyacrylamide induced by microwave were once studied, but neither of the two literatures displayed the graft efficiency and another microwave induced reaction produced chitosan-g-poly (methylmethacrylate) with its highest graft efficiency of 40%^[10-12].
    Reversible addition-fragmentation chain transfer polymerization (RAFT), the youngest living polymerization, which appears to be the most versatile controlled radical polymerization with respect to the monomer and the reaction medium, leading to the development of novel polymeric materials. In RAFT polymerization, effective RAFT agents for this process are dithioesters [Z-C (S=) S-R] with various leaving R-groups and stabilizing Z-moieties that involves dithiobenzoates, dithioalkanoates, trithiocarbonates, dithiocarbonates (xanthates) and dithiocarbamates^[15]. The basic working principle of RAFT polymerization is that the reversible addition-fragmentation process of thiocarbonyl-thio groups in RAFT polymerization can minimize the relative rate of bimolecular termination, and keep most of the growing polymeric chains bonding to the thiocarbonyl-thio compound and still remaining potentially active throughout the polymerization process ^[16-17]. On the other hand, in the presence of conventional initiator, initial and macromolecular radicals with different molecular weight (homo-polymers) were produced and some of them easily combined with dithiocarbonyl groups. With time going on the reversible addition-fragmentation actions at last create a polymer including dithiocarbonyl moieties. These characteristics implied that if the RAFT agent moiety is first introduced into the macromolecules, such as chitosan, it might facilely create initiating sites and further synthesize the graft copolymer of chitosan and common vinyl monomers with higher graft efficiency.
   In this study, we first introduced the dithiocarbonate groups, the reversible addition-fragmentation chain transfer moiety, onto chitosan via the reaction of amino group with benzyl chloride and carbon disulfide. And then, this modified chitosan was directly applied to produce chitosan-g-polymethylacrylate (chitosan-g-PMA) with higher grafting efficiency in the presence of common initiator. In addition, effects of the initiator concentration, temperature, time, amount of water, and ratio of monomer to modified chitosan on the grafting copolymerization were investigated by determining the grafting parameters, i.e. grafting efficiency, grafting percentage and monomer conversion. It was found that the grafting efficiency was considerably enhanced under the assistance of radical transferring interaction of dithiocarbonate groups on chitosan, comparing with the conventional grafting methods.

2. EXPERIMENTAL
2.1 Materials    
Chitosan (deacetylation degree > 85%, Mn=4.6×10⁵) used in this study was prepared from indigenous waste shrimp shell ^[18]. Methyl acrylate (MA) was washed successively with aqueous sodium hydroxide and distilled water to remove the inhibitors, and distilled finally. The grafting initiator, potassium persulfate (KPS) was recrystallized from water. Other chemicals were of analytical grade and were used without further treatment.
2.2 Measurements
Fourier transform infrared measurements were used to confirm the structure of chitosan-g-PMA with a Bruker EQUINOX55 (Bruker, Germany) in the potassium bromide pellets. The X-ray diffraction analysis was measured with a Yaa 900 X-ray diffraction instrument (Dandong Ray Apparatus Corp., China). The TGA curves were completed with a Pyris6 DTATG apparatus (PerkinElmer, America) in atmospheric N₂ at a heating rate of 10°C/min.
2.3 Graft copolymerization of MA onto modified chitosan                        
A glass tube containing a magnetic bar, deionized water, KPS and modified chitosan prepared according to the literature ^[19] was first pretreated with standard cycles of evacuation with nitrogen gas and then charged with required amount of monomer by a syringe. The reaction mixture was thermostatically controlled at the required temperature and magnetically stirred. After the completion of the reaction, the reaction mixture was cooled and poured into methanol. The precipitated product was filtered through a weighted, sintered glass funnel and was washed several times with methanol. Then, the crude graft copolymer was dried to a constant weight in vacuum at 50°C. The homo-polymer of PMA was removed from the crude graft copolymer by exhaustive Soxhlet extraction with acetone for 48 h. The final graft copolymer was dried under vacuum at 50°C to a constant weight.
   Grafting parameters used in this study were defined and calculated as follows:

3. RESULTS AND DISCUSSION
3.1 Formation of graft copolymer                            
The graft polymerization of MA onto modified chitosan under the assistance of dithiocarbonate moiety was shown in Scheme 1. Some initial radicals from KPS initiate homo-polymerization of MA, while another great part of them bond to the chitosan by the transfer-reaction with dithiocarbonate moiety to produce new radicals. In the same way, the macroradicals of homo-PMA formed in the system is apt to anchoring onto chitosan backbone, producing the titled graft copolymer. It is worthwhile to state that, similar to the conventional grafting method, a part of hydroxyl and amino groups can also be converted to macroradicals through the direct transfer-reaction of KPS radicals, and leading to the initiation of monomer MA and chain-grow of PMA. In the experiment for comparison, the grafting efficiency is 40.2% and 67.7% for pure chitosan and modified chitosan used as matrix, implying that introducing dithiocarbonate moiety onto chitosan have great effect on increasing the graft efficiency.

Scheme 1. A proposed mechanism for the graft polymerization

3.2.3 Effect of water
When the other conditions were invariable (MA: 0.42mL and chitosan: 0.15g), the effect of water amount on graft parameters is shown in Fig.3. It was found that E % was slightly affected by the amount of water, and reached its maximum value 78.4%, when water was 1.0mL. P % and C % decreased in the whole process due to the dramatically lowering radicals concentration in water phase as water increased. According to the curve of E %, it was concluded that when the amount of water was at lower level relative to chitosan, the reaction system could not be dispersed effectively, therefore the probability of chitosan contacting with PKS radicals decreased. Furthermore, excessive water has the same effect because of the reduction of initiator concentration.

Fig. 3 Effect of water on graft parameters
Chitosan: 0.15g; MA: 0.42mL; KPS: 0.013g; temperature: 60°C; time: 1h.

Fig. 6 FTIR spectra of chitosan (A), modified chitosan (B) and chitosan-g-PMA (C)

3.3.2 XRD analysis
The graft copolymerization of MA onto chitosan was also supported by XRD measurement of pure chitosan and chitosan-g-PMA (G%=197%). As shown in Fig.7, The chitosan-g-PMA with high grafting percentage exhibited a single broad absorption at 2q = 21.5^o, appeared to be due to extensive chain coverage of PMA, because of the lacks complete stereoregularity of PMA macromolecular chains. Contrast to this, two major peaks at about 10^o and 20^o ascribed to 020 and 110 reflections, respectively, were observed in the XRD pattern of pure chitosan with more crystallinity. The result elucidates that the crystalline structure was changed to some extent owing to introduction of PMA onto chitosan

Fig. 7 XRD patterns of chitosan-g-PMA (A) and chitosan (B)

3.3.3 Thermograms of the grafted chitosan
Fig.8 displays the thermograms for graft copolymer (G %=197%), modified chitosan and chitosan. Both chitosan and modified chitosan showed a weight loss with two stages under heating in an inert atmosphere. The first weight-losing stage spanned over 55 °C to 230 °C and exhibited about 5% loss in weight, which was ascribed to the loss of adsorbed and bond water. The second stage of weight loss started at 230 °C and continued up to 500 °C, due to the degradation of chitosan and modified chitosan. Different from chitosan and modified chitosan, chitosan-g-PMA demonstrated only one-stage degradation without the weight loss related to adsorbed and bond water. This can be explained by the following reason that hydrophobic property of graft copolymer was considerably increased duo to the attachment of hydrophobic PMA onto chitosan surface. Chitosan-g-PMA showed a higher thermostability with weight loss of 20% at about 350 °C as compared to chitosan and modified chitosan, which displayed about 35% and 50% weight loss respectively. As an accessorial proof, the remarkable difference in thermal properties among chitosan, modified chitosan and chitosan-g-PMA illustrated that PMA chains has been grafted onto chitosan surface.

Fig. 8 Thermogravimetric trace of graft copolymer, modified chitosan and chitosan

4. CONCLUSIONS
In order to increase the grafting efficiency, a new method of grafting MA onto chitosan by aiding of RAFT agent has been developed in this work. The effect of various experimental conditions on the grafting parameters has been systematically evaluated and the FTIR, TGA and XRD measurements elucidated that methyl acrylate was efficiently grafted onto chitosan surface. The highest grafting efficiency in this study is found to be about 90% due to introducing dithiocarbonate groups onto chitosan under this condition-chitosan: 0.15g; KPS concentration: 0.013g/mL; temperature: 55 °C; reaction time: 1 hour.

REFERENCE
[1] Huang M F, Fang Y E, Biopolymers, 2006, 82: 597.
[2] Chen T H, Kumar G, Harris M. T. et al. Biotechnology and bioengineering, 2000, 70: 564.
[3] Yu H J, Wang W S, Chen X S et al. Biopolymers, 2006, 83: 233.
[4] J. Fangkangwanwong, Mitsuru Akashi, Toshiyuki Kida et al. Biopolymers, 2006, 82: 580.
[5] Kang H. -M. Cai Y-L. Liu P-S Carbohydrate Research, 2006, 341(17), 11: 2851.
[6] E. Yilmaz , T. Adali , O. Yilmaz, Reactive & Functional Polymers 2007, 67: 10.
[7] Wang J P, Chen Y Zh, Ge X W et al. Chemosphere 2007, 66: 1752.
[8] Ge H C, Pang W, Luo D K, Carbohydrate Polymers 2006, 66: 372.
[9] Yao P J, Wan H Y, Jiang L B et al. Technology&Development of Chemical Industry,2005, 34: 9.
[10] Vandana Singh, Ashutosh Tiwari, Devendra Narayan Tripathi et al Polymer 2006, 47: 254.
[11] Vandana Singh, Devendra Narayan Tripathi, et al, Carbohydrate Polymers 2006, 65: 35.
[12] Liu P F, Zhai M L, Wu J L, Radiation Physics and Chemistry, 2001, 61:149.
[13] Zhang J, Yuan Y L, Shen J et al. European Polymer Journal 2003, 39: 847.
[14] Cao L Y, HuangJ F, Cao J J et al. Speciality Petrochemicals, 2001, 3: 19.
[15] Hideharu Mori, Motonobu Matsuyama, Kazuhiko Sutoh et al. Macromolecules 2006, 39: 4315.
[16] Michelle L. Coote, David J. Henry, Macromolecules 2005, 38: 5774.
[17] Christine Schilli, Michael G. et al. Macromolecules 2002, 35: 6819.
[18] Pourjavadi A, Mahdavinia GR, Zohuriaan-Mehr MJ et al. J Appl Polym Sci 2003; 88: 2048.
[19] Tao L R, Mei G Z, Ming CH T et al. Chemical Journal of Chinese Universities, 1999, 20: 1897.