“Living”/Controlled Radical Polymerization. Transition-Metal-Catalyzed Atom Transfer Radical Polymerization in the Presence of a Conventional Radical Initiator Jin-Shan Wang and Krzysztof Matyjaszewski* Department of Chemistry, Carnegie-Mellon University, 4400 5th Avenue, Pittsburgh, Pennsylvania 15213 Received June 20, 1995 Revised Manuscript Received September 1, 1995polymerization at 130 °C initiated by AIBN combined ‘with or without Cul'Cl, complexed by bpy. The molar ratio of Cul'CL, versus AIBN represents a critical parameter for the controlled polymerization while keep- ing constant (bpylo[Cut"Cle}, of 21. In the absence of CuCl the polymerization ie an illcontolid proces, as evidenced by a much higher molecular weight than expected and broader molecular weight distribution, ‘Mu/Me ~ 3.0. However, the polymerization becomes ‘uch better controlled when only 2 molar equiv of Cu! Cy versus initiator is added. ‘The initiator efficiency, 7 was inereased from 22 to 65% and MwMy reduced from 3.0 0 1.8. An merease in (CulICL AIBN ratio further improves the degree of the control of the polymerization. A 10-fold molar excess of CullClz ‘results in ¢ weli-controlled radical polymerization with 95% initiator efficiency and the polydispersity as low as MIM, ~ 1.30. It must be pointed out here that 1 molar equiv of Gull species versus growing radicals should be sufficient to promote a “living’/eontrelles ATRP of St? The present results thus suggest that in these heterogeneous polymerization systems the real concentration of Cul! species in the solution is much lower than CuCl, added, due to its low salubility in the aystem, and the rate of solubilization is lower than ‘that of decomposition of AIBN.Table 1. Characterization Data of Styrene Bulk Polymerization Initiated with AIBN/CuCli/bpy at 130 °C [AIBN], [CuClela [bpyl time conv, (moVL) (moVL) (moVL) (h) _(%) Mas? Mnszc Mu/Mp 0.0230 0 15 85 1680) 75600 3.00 0.084 0.068 0.185 21 9012.00 9 0 18 500 0 0.0081 0.09 0.18 110 91 45200 41200 1.45 0.0065 0.080 0.12 119 88 59500 56900 145 * Detected by GC using THF as an internal reference. > Calcu- lated based on eq 1.° Bimodal. Ma = 104 x {ATMI/(2[AIBN],)} x conv (1)0 20 40 60 80 100 Conversion, % Figure 1. Evolution of molecular weight, M,, and molecular weight distribution, M,/M,, with monomer conversion for the bulk polymerization of styrene initiated with AIBN/Cu"Cl,/ bpy at 130 °C. (St}) = 8.75 M, [AIBN}p = 0.025 M, [Cu"Clz} = 0.25 M, [bpyls = 0.50 M.It is known that CuCl, is a strong and efficient inhibitor/retarder for radical polymerization under usual conditions.® However, at high temperatures and in the presence of a coordinative ligand such as bpy, a revers- ible activation of the resulting alkyl chlorides occurs.” By analogy to ATRP initiated with the R-X/M"L. system,? the present results can be interpreted in terms, ofan alternative ATRP cycle (Scheme 1). Initially, the highly oxidized transition-metal species, M.'"#X, do- nates the halogen atom X to the initiating or propagat- ing radicals, I) or I-P*, with the formation of the reduced transition-metal species, M;", and the dormant species, I-X and P,-X. In the subsequent steps, the ‘transition-metal species, M.", promotes exactly the same ATRP process as the one initiated by R-XIM//L; as described previously.” The only difference between these two ATRPs is the following, In the case of using the R-X/M//Ly initiating system, ATRP is first pro- moted by the activation of a dormant species, RX, with Me’? On the other hand, using I-UM¢~X/L,, ATRP originates from the deactivation reaction between radi- eals, For Pi-T, and My".Initiation: a ae MX Se eK eM ws [om +" Py + MO XS 1 Pe ME Propagation: Pao + MMTX S=* PAX + My”igure 2 presents the 1H NMR spectrum of PSt pro- duced using [AIBNJy[ CuCl (1/1020) at 190°C. ‘Two broad triplots at ea 4.4 ppm are attributed to the end group ¢-CH.-C(Ph)H.-CI of two different stereciso- ters (mr) in Clendcapped PSt. Moreover, compar- son of the integration of the signals of end group c with ‘those of group «oF in the main chain of PS gives a soleculas weight, Mp;xux ~ 2950, close to the one from SEC based on PSt standards, My sxc ~ 3140 (My/My ® 1.30). This strongly suggests that the PSt radicals are efficiently deactivated by CuCl, Hes 4 d Figure 2. :HNME of polystyrene initiated with AIBN/Cu!- Clyboy 1020) at 330°C,However, an extension of the AIBN/Cu"/opy-catalyzed ATRP to other monomers reveals some interesting phenomena. For instance, by using the same initiating system, AIBN/Cu"Ciy/opy (1/10/20), and the same ex- perimental conditions, sn balk and at 190 °C, as ATRP of St, the MA polymerization is not a controlled process. ‘The polymerization was very fast and led to gelation at cca, 40% monomer conversion, SEC analysis of the THF- soluble portion gave a much higher molecular weight than ealeulated according to eq 1, Masec ~ 28 500 against M,cuc ~ 6800, and a very broad multimodal molecular ‘weight distribution, Mw/M, ~ 3.7. This jests that the deactivation of growing radicals by Cull is much less efficient in MA polymerization than in St polymerization smitiated with AIBN/Cul(Clsfopy (1) 10/20) at 180 °C, leading to a very high concentration of growing radicals. This is consistent with the fact that tlectrophilie PMA radieals are less efficiently scavenged ty electrophilic CuX, than nucleophilic PSt radicals, due to the different polar properties of the transition states in these reactions Moreover, ft is also well -known that the rate constant of radical propagation for ‘MA\is ca, 15 times higher than that for St.° Therefore, the ratio of the deactivation rate of the growing radicals ‘to the dormant species to the propagation rate must be higher in St ATRP than in MA ATRP. This, in addition to differences in solubilities of Cu"X., may be respon sible for tho broader molecular weight distribution observed in MA ATRP.In conclusion, the present work again suggests that our recently developed ATRP process offers a versatile route toward the “living’/controlled radical polymeri--- zation of alkenes with radical stabilizing substituents such as styrenes and (meth)acrylates. The presence of a catalytic amount of growing radicals and a fast equilibration between growing radicals and dormant species are mandatory for a successful ATRP.