Kostas
Parkatzidis
,
Leonardo
de Haro Amez
,
Nghia P.
Truong
and
Athina
Anastasaki
*
Laboratory for Polymeric Materials, Department of Materials, ETH Zürich, Vladimir-Prelog-Weg 5, 8093 Zürich, Switzerland. E-mail: athina.anastasaki@mat.ethz.ch
First published on 17th March 2023
In the vast majority of atom transfer radical polymerizations, alkyl bromides or alkyl chlorides are commonly employed as initiators, and minimal attention has been given to alkyl iodides. Herein, we report the room temperature Cu(0)-mediated reversible deactivation radical polymerization of acrylates utilizing alkyl iodide as an initiator. Kinetic experiments were conducted showing a linear dependence of Mn with conversion, good agreement between theoretical and experimental molecular weights, and low dispersity values (Đ ∼ 1.05), even at high monomer conversions. The high-end group fidelity of the iodide-terminated polymer was confirmed via MALDI-ToF-MS analysis and successful in situ chain extensions at near-quantitative conversions. Polymerization of methyl acrylate with various targeted degrees of polymerizations (DPn = 25–2400), resulted in the synthesis of well-defined polymers with low dispersities (Đ < 1.15), even at higher molecular weights (e.g. Mn = 200000, Đ ∼ 1.13). The compatibility of the methodology with various solvents, including acetonitrile (MeCN), dimethylformamide (DMF), acetone, and isopropanol (IPA) as well as a range of acrylic monomers was also investigated yielding polymers with excellent control over the molar mass distributions. A series of block copolymers consisting of different block segments were also synthesized in one pot without any intermediate purification steps, thus highlighting the potential of an alkyl iodide initiator in a controlled polymerization.
Arguably, ATRP remains to date the most studied controlled radical polymerization method due to its simplicity and versatility.10 ATRP operates by utilizing a transition metal (usually copper) to first trigger and then mediate the polymerization. In conventional ATRP, the metal in its low oxidation state (e.g. CuI) coordinates to an amine ligand (e.g. CuI/L) forming a complex which acts as a catalyst. The catalyst activates the carbon–halogen bond of an alkyl halide initiator generating a radical and a metal complex in a higher oxidation state (e.g. CuII/L), known as the deactivator. The generated radical, also referred to as propagating radical, can be either immediately deactivated by the catalyst in the higher oxidation state, propagate by reacting with one more vinyl monomers followed by deactivation, or undergo an irreversible reaction with another radical. The latter leads to the accumulation of CuII during the polymerization and is responsible for the control over the polymerization through the persistent radical effect mechanism. As a result, conventional ATRP necessitates a certain extent of termination prior to gaining control over the molar mass distributions. In addition, high amounts of catalyst are typically needed for a successful polymerization. To overcome both barriers, a number of alternative ATRP approaches have recently been developed, including eATRP,11,12 photoATRP,13–16 ARGET-ATRP,17,18 and, (P)ICAR-ATRP.19,20 A common feature of these methodologies is that they require a stimulus or an additive to in situ re-generate the active catalyst and as a result they can operate at significantly lower catalyst concentrations.21–25 A copper-mediated polymerization approach that has received considerable attention during the last fifteen years is Cu(0)-mediated RDRP in which a zero-valent metal, usually copper wire, is used to provide the initial source of copper in the reaction.26 Although the mechanism of this approach has been heavily debated in the literature, it is not the subject of this investigation.27–34 Irrespective of the mechanism, it can be commonly agreed that Cu(0)-mediated RDRP is a versatile and robust polymerization methodology that can polymerize various monomers with extremely high end-group fidelity.35–38 The increased popularity of Cu(0)-RDRP can be attributed to its’ simplicity as it can be conducted at mild polymerization conditions, it requires very low catalyst concentrations, and it allows for the vast majority of the catalyst to be removed by simply lifting the stirring bar at the end of the reaction. In addition, in many occasions the copper wire can be re-used several times upon polymerization.39
Regardless of the catalytic system employed and the way by which the active catalyst is being regenerated, all copper-mediated polymerizations employ alkyl halides initiators. The halogen atom affects the thermodynamic properties of the catalyst/initiator system and has a significant impact on the ATRP reaction.40–43 However, in the vast majority of studies, the common halogens employed are either bromine or chlorine. While the bromine-containing initiators are typically used in organic media, the chlorine-ones are often preferred in aqueous polymerizations due to the higher stability of the carbon chlorine bond against dissociation.44 Although theoretical calculations of iodine-containing initiators have been conducted, they have been rarely employed in ATRP. For example, Matyjaszewski and co-workers showed that alkyl iodide presents a faster activation rate when compared to its alkyl bromide analogue, while the KATRP had the opposite trend.40,41 This suggests that the kdeact in the case of alkyl iodide should be higher than the alkyl bromide. However, during the ATRP reaction of methyl acrylate using methyl 2-iodo-propionate as an initiator at 90 °C, very limited conversion (20%) was detected after 20 hours.45 An iodide haloform initiator has also been emoloyed by Percec and co-workers in Cu(0)-RDRP. Although the primary focus of this paper was the synthesis of ultra high molecular weight polymers, the authors also conducted an experiment whereby iodomethane was employed as a bi-functional initiator yielding the preparation of telechelic polymers at 80% of conversion (Đ ∼ 1.18).33 This paper suggests that Cu(0)-RDRP may be particularly advantageous to afford a controlled polymerization when conventional mono-functional iodide-containing initiators are employed and inspired our current study.
In this work we utilize an alkyl iodide initiator to trigger the Cu(0)-RDRP of various acrylates at room temperature (Scheme 1). This is in contrast to iodine-transfer polymerization (ITP), controlled radical polymerization catalysed via different transition metals (Ruthenium, Rhenium, Iron, etc.) or reversible complexation mediated polymerization (RCMP) whereby high temperatures and/or free radical initiators are typically required to catalyze a successful polymerization.46–56 Under relatively mild conditions, we seek to employ a mono-functional initiator to target a wide range of molecular weights at near-quantiative convesions and without compromising the control over the molar mass distributions. Indeed, poly(methyl acrylate) with excellent control over the molar mass distributions (Đ as low as 1.05) could be obtained. Notably, a series of polymers with various molecular weights could be targeted (DPn = 25–2400) at high monomer conversions and the resulting polymers possess high end-group fidelity as demonstrated by in situ chain extensions and block copolymer formation. The effect of the solvent was also investigated and a range of solvents were found to be compatible with our polymerization approach.
Considering the use of iodo-compounds in the presence of vinyl monomers, the possibility of a potential degenerative transfer reaction cannot be excluded. In fact, degenerative transfer is often cited as the main reason that iodine-containing initiators have been avoided in ATRP.42,57 To explore this, we conducted a series of control experiments, as shown in Table 1. In the first control experiment, polymerization of MA was attempted at room temperature and in the absence of copper wire. However, no monomer conversion could be observed even after 24 h, thus strongly suggesting that copper wire is necessary to abstract the Iodine and generate radicals (Table 1, row 1). In a similar fashion, no polymerization took place when either the ATRP initiator or the ligand were excluded (Table 1, row 2 and 3), and as such it was concluded that all the reagents are essential for a successful RDRP at room temperature. However, the low dispersity obtained (Fig. 1) can be attributed to either sufficient ATRP deactivation, degenerative chain-transfer or a combination thereof. To investigate the possibility of degenerative chain transfer, we sought for an alternative radical source (i.e. to replace copper wire). As a room temperature free radical initiator was not available in our laboratory, we first repeated our key experiment at 80 °C and narrow molar mass distributions were also obtained (Đ ∼ 1.06) (Table S2†). In fact, regardless of the polymerization temperature (60 °C, 80 °C or 110 °C), very low dispersities (<1.09) were achieved accompanied with an increased polymerization rate at higher temperatures (Table S2†). This highlights sufficient deactivation and minimal side reactions under the studied ATRP conditions. Instead, when the identical experiment was repeated using AIBN as a radical source at 80 °C (i.e. without copper wire/ligand), significantly broader molar mass distributions were obtained (Đ ∼ 1.7), albeit with a good control over the molecular weight (Table S2†). These experiments highlight that (i) degenerative chain transfer can occur in the presence of Iodide-containing initiators (when a radical source such as AIBN is employed) leading to good control over the molecular weight and high dispersity, in line with the literature and (ii) the low dispersity obtained is primarily attributed to the ATRP deactivation rather than degenerative chain transfer. Taking into account that the synthesis of polyacrylates through RCMP typically occurs at higher temperatures (i.e. 110 °C), rather than room temperature, we also performed a control experiment at 110 °C and in the absence of any copper catalyst (but in the presence of ligand that is responsible for the halogen abstraction/radical generation). Within 1.5 hours the polymerization mixture turned viscous, with 1H NMR showing 80% monomer conversion. The SEC analysis of the sample showed a monomodal molecular weight distribution with controlled molecular weight, albeit with a very high dispersity (Table 1, entry 4), thus confirming an RCMP mechanism at higher temperatures with poor control over the molecular weight distributions. Taken altogether, these control experiments indicate that the low dispersity obtained is due to the Cu(0)-RDRP mechanism although the possibility of degenerative chain transfer cannot be excluded.
[MA]![]() ![]() ![]() ![]() ![]() ![]() |
Temp. (°C) | Time | Conv. (%) | M n (SEC) (Da) | Đ |
---|---|---|---|---|---|
50![]() ![]() ![]() ![]() ![]() ![]() |
RT | 24 h | — | — | — |
50![]() ![]() ![]() ![]() ![]() ![]() |
RT | 24 h | — | — | — |
50![]() ![]() ![]() ![]() ![]() ![]() |
RT | 24 h | — | — | — |
50![]() ![]() ![]() ![]() ![]() ![]() |
110 | 1.5 h | 80 | 7600 | 1.66 |
50![]() ![]() ![]() ![]() ![]() ![]() |
110 | 5 min | 82 | 4700 | 1.09 |
Since we validated the controlled behaviour of Cu(0)-RDRP of MA utilising alkyl iodide as an initiator, we then sought to further characterize the synthesized polymers and examine their end-group fidelity. To do so, low molecular weight PMA (targeted DP = 25) was initially synthesized and then analyzed via matrix assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-ToF-MS). Fig. 2a shows the MALDI spectrum whereby it can be confirmed that every polymer chain is initiated by the expected initiator and has a living iodide at the ω-end. Notably, no other end-groups were observed, highlighting that very high end-group fidelity can be maintained. To further highlight the excellent end-group functionality of the polymers, as well as the stability of the iodine end-group, we subsequently conducted an in situ chain extension experiment. PMA with DP 200 was first synthesized and upon near-quantitative monomer conversion (i.e. 96%), a second aliquot of a deoxygenated MA and DMSO solution was added, and the polymerization was allowed to proceed for 3 further hours. 1H NMR and SEC analysis indicated a successful chain extension, with the SEC traces shifting to higher molecular weights and a final dispersity of 1.09 (Fig. 2b).
In order to expand the scope, we also conducted the synthesis of a series of PMAs with various targeted DPs (Fig. 3 and Table S3). Starting with DP 25, the synthesis resulted in a well-defined PMA with Mn of 3200 and Đ of 1.06. Increasing the targeted DP up to 2400 did not influence the control over the polymerization yielding PMA with Mn of 200000 and Đ of 1.13. Overall, very good control was maintained for all the targeted DPs, with high monomer conversions and low dispersities in all cases (Fig. 3 and Table S3). Next, a range of solvents was investigated. For this purpose, five different solvents were employed for the polymerization of PMA, under the following conditions [MA]
:
[EIP]
:
[Me6TREN] = [200]
:
[1]
:
[0.18] using 5 cm of Cu(0) wire and 1
:
1 monomer to solvent volume ratio. Irrespective of the chosen solvent, well-controlled polymers were obtained with good control over the molecular weight and low dispersity values (Đ < 1.2). However, the polymerization rate was significantly affected (Fig. 4), in agreement with the literature.58 In particular, the fastest polymerization took place in DMSO (kapp = 4.3 × 10−4 s−1), whereby 90% monomer conversion was reached within 2 hours. The next solvent with the fastest polymerization rate was DMF (kapp = 1.3 × 10−4 s−1) followed by MeCN, acetone and finally IPA.59–61 A range of different acrylic monomers were also polymerized including the hydrophilic (EGA), the functional (tBA), the semi-fluorinated (TFEA) and the macromonomer (PEGA480). All the polymerizations proceeded efficiently resulting in low dispersity polymers (Đ < 1.1) (Fig. 5 and Table S4†). Finally, the in situ synthesis of diblock copolymers with different block segments was attempted (Fig. 6). A PMA200 macro-initiator was first synthesized and upon reaching near-quantitative monomer conversion, the second monomer was added in one pot. In all the cases, the successful formation of block copolymers was confirmed by both 1H NMR and SEC analysis yielding low dispersity final materials.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d2py01563c |
This journal is © The Royal Society of Chemistry 2023 |