Introduction

Atom transfer radical polymerization (ATRP) is a widely used technique for synthesizing well-defined polymers. 1,2 In ATRP, control over the polymerization is based upon a reversible redox process using catalysts that can efficiently sustain activation and deactivation of polymer chain ends. [3][4][5] In this regard, copper complexes are one of the most efficient and widely used ATRP catalysts with L/Cu I and L/Cu II -X (X: Br or Cl) species being the activator and deactivator, respectively. [6][7][8] In addition, other transition metal-based catalysts such as iron, 9,10 ruthenium, 11 iridium 12 and also organic photoredox catalysts [13][14][15] are capable of catalyzing ATRP with high efficiency. 16 In ATRP reactions, the bond dissociation energy of the carbon-halogen chain ends and affinity of the halogen atom to the catalyst are important criteria for efficient atom transfer processes to succeed. As a result, Br and Cl chain end functionalities are effective in affording well-controlled polymers in the presence of atom transfer catalysts, whereas F chain ends are not suitable due to the high energy required to cleave the C-F bonds. 17 Interestingly, C-I bonds are weaker and therefore can be cleaved easily. However, previous reports have shown low affinity of the I atom to Cu catalysts rendering inefficient ATRP catalysis in the presence of alkyl iodides. 18 Iodine-based polymerizations typically occur via a degenerative transfer process in the presence of iodinated chain transfer agents. [19][20][21][22][23][24][25] The transfer and dynamic exchange of iodine between propagating radicals and dormant species affords moderate control over polymerization. However, in the presence of compounds such as amines or iodide salts that form complexes with alkyl iodides, the polymerization becomes catalytic undergoing a similar mechanism to ATRP (also referred as reversible complexation mediated polymerization, RCMP 26 ) to catalyze iodine atom transfer, and consequently improves control over polymer chains.

Alkyl iodides are generally light/heat sensitive compounds that require special care for storage and handling. However, for iodine-mediated ATRP reactions, alkyl iodide initiators can be generated in situ from more stable alkyl bromides using iodide salts. 27 Indeed, the light sensitivity of alkyl iodides can be advantageous for developing photocontrolled iodinemediated polymerization strategies. Scheme 1 illustrates the general mechanism of iodine-mediated photoATRP via in situ generation of the alkyl iodide initiator. In the presence of iodide salts that also act as a catalyst, the C-I bond can be photochemically cleaved to form initiating/propagating radicals and also generating an iodine-based radical. 26,[28][29][30][31] The iodine radical complexes with an iodide anion to form an iodine radical anion (I 2

•-

) species, which serve as the deactivator for propagating radicals. Furthermore, in an iodine-based polymerization, a degenerative transfer process may also be involved between the growing radicals and iodine-capped dormant species to impart control over the growth of polymer chains.

Water is considered a green solvent for polymerizations and is compatible with different reversible-deactivation radical polymerizations (RDRP) including ATRP, RAFT and other techniques. [32][33][34][35] However, iodine-mediated polymerizations have not been fully investigated in aqueous media. Moreover, RDRP methods have been recently advanced to enable well-controlled polymerizations in the presence of oxygen. 36 Indeed, oxygen can be consumed in situ by applying various (photo)chemical techniques without the need for performing conventional deoxygenation processes, thereby simplifying the synthesis of polymers with potential utility for new applications. [36][37][38][39][40][41][42] In this paper, iodine-mediated photoATRP was investigated in aqueous media, which enabled fast and well-controlled polymerizations that were also tolerant to residual oxygen without the need for degassing procedures. The alkyl iodide initiator was generated in situ by an exchange reaction of Br to I using ethyl α-bromophenylacetate (EBPA) and various iodide salts, which also acted as the catalyst for the polymerization. Polymerizations were conducted under a wide range of visible light as external stimuli enabling temporal control over the polymerization.

Results and discussion

Iodine-mediated photoATRP was initially attempted using poly (ethylene glycol) methyl ether methacrylate (average M n 300, PEGMA 300 ) as a water-soluble monomer and ethyl α-bromophenylacetate (EBPA) under blue light irradiation in water (Scheme 2). Tetrabutylammonium iodide (TBAI) was used to generate alkyl iodide initiator by exchanging EBPA's Br to I. Table 1 shows the results of the polymerization of PEGMA 300 and the effect of concentration of different components involved in the polymerization. High monomer conversion was achieved in less than 2 h under blue light irradiation. Increasing the concentration of TBAI resulted in a well-controlled process providing polymers with low dispersity (Đ) values. For example, in the presence of 1 equiv. of TBAI with respect to EBPA, the polymerization of PEGMA 300 reached high monomer conversion showing molecular weights close to theoretical values, but with a high Đ of 1.88. However, increasing concentration of TBAI to 2 equiv., with respect to initiator, improved control over the polymerization and provided polymers with Đ of 1.43 (entries 1 and 2, Table 1). These observations suggest that in the presence of high concentration of the iodide salt control over the polymerization was achieved mainly through an ATRP mechanism with iodide salts acting as the catalyst under photochemical conditions. In the absence of excess iodide salts, control may be attributed predominately to a degenerative mechanism.

Furthermore, under more dilute conditions, the rate of polymerization slightly decreased but resulted in well-controlled polymers, showing lower Đ values. For instance, in the presence of 2 equiv. of TBAI with respect to initiator, decreasing concentration of the monomer from 50 to 33 and 25 vol% in water resulted in decreasing Đ from 1.43 to 1.34 and 1.24, respectively (entries 2-4, Table 1). In the presence of 4 equiv. of TBAI and 25 vol% monomer, the polymerization was wellcontrolled giving a low Đ of 1.18 (entry 5, Table 1). Importantly, in all these polymerizations experimental molecular weights agreed well with theoretical values, indicating high initiation efficiency. In addition, alkali metal iodide salts including potassium, sodium and lithium iodides were also used to perform and efficiently control the polymerization of PEGMA 300 in water (entries 6-9, Table 1 and Fig. S4-S6 †). Notably, these organic and alkali metal iodide salts were readily soluble in the reaction medium and therefore were used without additional solubilizing agents, as was previously required for the in situ exchange of Br-I when conducting polymerization in organic media. 30 Control experiments performed in the absence of either EBPA, iodide salts, or light did not lead to initiation of a polymerization, indicating the importance of the contributing components for a successful iodine-mediated photoATRP under visible light (entries 10-12, Table 1).

Effect of polymerization medium

Using water as a solvent resulted in a well-controlled iodinemediated photoATRP of PEGMA 300 monomer under blue light irradiation. The efficiency of polymerization in aqueous media was further demonstrated by conducting the polymerization in bulk or in an organic solvent, N,N-dimethylformamide (DMF). Tetrabutylammonium iodide (TBAI) was used to exchange the initiator's Br to I. Kinetics of the polymerization in bulk showed a decrease in the rate of the polymerization over time yielding 87% monomer conversion in 6 h (Fig. 1A). Size exclusion chromatography (SEC) analysis of the polymer obtained in bulk showed a broad molecular weight distribution with a high Đ of 2.16 (Fig. 1B). The uncontrolled polymerization in bulk can be related to the poor solubility of TBAI in the monomer, and consequently inefficient Br-I exchange and hence poor catalysis. Under these conditions, using DMF as an organic solvent, a decrease in the rate of polymerization was observed as the polymerization progressed with 70% monomer conversion in 6 h. A relatively high Đ of 1.43 was obtained in DMF (Fig. 1C). Fast generation of radicals under a strong blue light irradiation and relatively slow rate of propagation in DMF led to the termination of polymer chains and hence a decrease in the rate of polymerization. However, conducting the polymerization of PEGMA 300 monomer in water resulted in a fast and well-controlled process ensured by a rapid formation of the radicals under a strong blue light and a high rate of propagation in water. Linear semi-logarithmic kinetics were observed with the polymerization reaching 94% monomer conversion in 90 min providing polymers with low Đ of 1.22 (Fig. 1D and S1 †). Notably, SEC traces of the polymers obtained in water showed a shoulder appearing at higher molecular weights as the reaction reached high monomer conversions. The appearance of this shoulder can be attributed to the presence of di-functional impurities in the monomer, resulting in branching at higher conversions, which can be suppressed by conducting the polymerizations under dilute conditions (Fig. S1-S3 †).

Effect of light sources on polymerization

Iodine-mediated photoATRP in water was successfully initiated and controlled with a wide range of visible light sources. Fig. 2A shows the kinetics of the polymerization of PEGMA 300 monomer in the presence of TBAI under blue, green and yellow light LEDs. Irradiation under blue light LEDs (460 nm) resulted in 94% monomer conversion in 2 h (M n = 15 100, Đ = 1.17), Fig. 2B and Fig. S3. † Under a green light (520 nm) irradiation, polymerization was slower and reached ∼90% monomer conversion in 6 h and provided polymers with well-controlled properties (M n = 13 300, Đ = 1.20), Fig. 2C. Polymerizations under blue and green lights showed linear semi-logarithmic kinetics with a lower rate of polymerization observed under green LEDs. Importantly, wellcontrolled polymerizations were achieved in both cases with molecular weights in agreement with theoretical values and Đ < 1.2. Under yellow light LEDs (595 nm), a slow polymerization of PEGMA 300 monomer was observed giving ∼77% conversion in 24 h, and polymers showed a Đ of 1.43, Fig. 2D. These results demonstrate the efficiency of iodine-mediated photoATRP in aqueous media achieved over a wide range of visible light in the presence of simple iodide salts.

Notably, iodine-mediated photoATRP was successfully performed in the absence of any added conventional photoinitiator/photocatalyst. Therefore, in the presence of iodide salts, direct photolysis of the labile C-I bond by visible light may be responsible for the formation of radicals and therefore successful polymerizations.

Temporal control

Kinetics of the polymerization was mediated by switching the light on and off. As shown in Fig. 3, iodine-mediated photoATRP of PEGMA 300 in water was activated only under blue light irradiation and was switched off by removing the light. Importantly, control over the polymerization was maintained during multiple light on/off switches with molecular weights in line with theoretical values and low Đ < 1.2.

View Article Online

Polymerizations in the presence of residual oxygen Iodine-mediated photoATRP was controlled in the presence of residual oxygen without performing typical deoxygenation processes. Oxygen tolerant polymerizations of PEGMA 300 were conducted under blue light irradiation using TBAI or KI as the catalyst and targeting degrees of polymerization (DP t ) of 50-200. Results of the polymerizations showed well-controlled polymers obtained in the presence of residual oxygen reaching high monomer conversions and providing polymers with low Đ < 1.2 (Table 2). As shown in Fig. 4, the polymers synthesized in the presence of residual oxygen showed monomodal, narrow molecular weight distributions, indicating a high level of control over polymers achieved under these conditions.

Conclusions

In summary, iodine-mediated photoATRP was investigated in aqueous media. Polymerizations were catalyzed by iodide salts that were also used to initially generate alkyl iodide initiator in situ. Use of water as a solvent resulted in fast polymerizations, providing high monomer conversions with polymers showing well-controlled properties. Importantly, iodinemediated photoATRP was initiated and controlled under a wide range of visible light irradiation including blue, green, and yellow lights in the absence of conventional photocatalysts. Furthermore, polymerizations were successfully controlled in the presence of residual oxygen signifying the potential of iodine-mediated photoATRP for use in applications without the need for performing deoxygenation processes.