Cyrille
Boyer
,
Aurelia
Derveaux
,
Per B.
Zetterlund
and
Michael R.
Whittaker
*
Centre for Advanced Macromolecular Design (CAMD), School of Chemical Engineering, The University of New South Wales, Sydney, NSW 2052, Australia. E-mail: mikey.whittaker@unsw.edu.au
First published on 30th September 2011
In this communication, we describe a simple and highly efficient route to well-defined multi-block star copolymers based on copper(0)-mediated living radical polymerization. The technique involves a core first approach using a multi-functional initiator in connection with iterative copper(0)-mediated radical polymerization steps. Importantly, purification is not required between the successive chain extension steps as complete monomer conversion is reached before the addition of each consecutive monomer type.
In general, well-defined star polymers have been prepared via two different routes; the arm first approach or the core first approach.10–14 In the arm first approach, pre-synthesized arms are either attached to a suitable multifunctional core or linked iteratively together to form a star (usually through cross-linking chemistry to give a cross-linked core). However, in both these cases, steric effects generally come into play at high arm molecular weights and/or high numbers of arms, resulting in less than quantitative yields.10 The use of orthogonal ‘click chemistry’ strategies, e.g. Huisgen cycloaddition, thiol-ene and others, has allowed the routine preparation of MIKTO-arm stars in high yields.15–23 Multifunctional microgels or nanogels have also proved to be suitable cores for polymer stars synthesis, where residual vinyl groups are used for subsequent arm attachment. Controlled/living radical polymerization (CLRP)2,24–28 has been applied to the iterative linking together of polymer arms via cross-linking chemistry. This approach was pioneered and optimized for ATRP by Matyjaszewski,11 and later expanded on by Qiao and co-workers, both reporting star polymers with very low polydispersities (PDIs).29–31
The core first approach entails the synthesis of a suitable macro-initiator bearing a defined number of initiating groups, each of which is subsequently transformed into a polymer chain viaCLRP. The approach was first demonstrated using ionic polymerization techniques, which are ideally suited to this star synthesis procedure as bimolecular termination (especially coupling) does not occur, and therefore star-star coupling by-products are not present.32–34 On the contrary, the CLRP techniques of ATRP/transition metal-mediated radical polymerization,11,13,35–40 RAFT/MADIX12,41–51 and NMP24,27,52 all suffer from bimolecular termination reactions to various extents, thus limiting their applicability to multi-arm star polymer synthesis. Tedious optimization of the polymerization conditions is required to minimize these side reactions and ensure low polydispersities.1,53–56
Cu(0)-mediated radical polymerization (SET-LRP) has been exploited to prepare star polymervia the core first approach. For example, Whittaker et al.57 and Ding et al.58 have reported the synthesis of 4-arm poly(MA) stars with a PDI of approximately 1.4. The high PDI was attributed to star-star coupling. In the present work, we have optimized the synthesis of multi-block copolymer star polymers using a core first approach with a very high control of the polymer structure. To achieve this, we have used our recent strategy to prepare linear multi-block copolymers with a high level of structural complexity.59 The approach, inspired by work of Haddleton,60,61 Matyjaszewski35,62–65 and Percec and co-workers,66–69 entails iterative Cu(0)-mediated radical polymerization steps in the presence of initially added Cu(II). The experimental conditions have been optimized to minimize any star-star coupling reactions by adjusting the deactivator (Cu(II)) concentration. The effect of adding CuBr2 to copper mediated polymerization is still not perfectly understood. However, as a deactivator it aids in the preservation of high end-group functionality (CH–Br), as recently demonstrated by our team for the polymerization of methyl acrylate70 and other groups.60,61,66–69 The ratio [Cu(II)]/[CH–Br] is critical with regards to minimization of star-star coupling, which is usually observed at very high monomer conversion. To illustrate the versatility of our approach, we have prepared five different star polymers with predesigned multi-block arms comprising short blocks (250–1 000 g mol−1) of a variety of monomers (hydrophobic, hydrophilic and functional monomers).
Monomers methyl acrylate (MA, Aldrich, 99%), n-butyl acrylate (nBA, Sigma-Aldrich, 99%), ethyl acrylate (EA, Sigma-Aldrich, 99%), tert-butyl acrylate (tBA, Aldrich, 99%), 2-ethyl hexylacrylate (2-EHA, Sigma-Aldrich, 99%) were de-inhibited by percolating over a column of basic alumina (Ajax, AR).
For the subsequent chain extension (i.e. the next iterative polymerization), a further 0.4 mL of a degassed monomer was then carefully added via gas tight syringe and again the solution was allowed to polymerize at room temperature for another 24 h with stirring.
The above iterative polymerization-sampling-extension procedure was repeated as required.
Other [Cu(II)]0: [CH–Br]0 ratios (0.08:1 and 0.16: 1) were tested using the similar recipe.
DMAc GPC analysis of the polymers was performed in N,N-dimethylacetamide [DMAc; 0.03% w/v LiBr, 0.05% 2, 6-di-butyl-4-methylphenol (BHT)] at 50 °C (flow rate = 1 mL min−1) using a Shimadzu modular system comprised of an SIL-10AD auto-injector, a PL 5.0-mm bead-size guard column (50 × 7.8 mm) followed by four linear PL (Styragel) columns (105, 104, 103, and 500 Å) and an RID-10A differential refractive-index detector. Calibration was achieved with commercial polystyrene standards ranging from 500 to 106 g/mol.
Scheme 1 Synthesis of 5-arm star copolymersvia iterative Cu(0)-mediated radical polymerization. |
In preliminary experiments, we followed our previous work on the synthesis of linear multi-block copolymers,44,59 using a similar [Cu(II)]: [CH–Br] ratio of 0.04: 1, and the same amount of Cu(0) in the presence of the penta-functional initiator. To simplify the characterization, methyl acrylate was initially used as a model acrylate monomer, with a targeted block length of Mn = 500 g/mol. GPC analysis of the polymer formed after the first iteration (Fig. 1A) reveals a highly multimodal molecular weight distribution (MWD) with a prominent high molecular weight peak due to star-star coupling. Further successive chain extensions (by addition of degassed MA/DMSO mixture (50/50 V-%) without polymerpurification) resulted in the MWDs gradually shifting to higher molecular weights as anticipated (Fig. 1A), but the broadness of the MWDs increased (PDI > 1.5 and Mn > Mn, th; Fig. 2) indicating the formation of star-star coupled products.
Fig. 1 Molecular weight distributions obtained in cycles no. 1–5 during synthesis of multi-block 5 arm stars by Cu(0)-mediated radical polymerization of methyl acrylate in the presence of different initial amounts of Cu(II): (A) [Cu(II)]: [CH–Br] = 0.04: 1, (B) [Cu(II)]: [CH–Br] = 0.08: 1 and (C) [Cu(II)]: [CH–Br] = 0.16: 1. |
Fig. 2 (A) Evolution of molecular weight (Mn, GPC) and PDI versus number of cycles during synthesis of multi-block 5 arm stars by Cu(0)-mediated radical polymerization of methyl acrylate in the presence of different initial amounts of Cu(II). Mn, th corresponds to the theoretical values calculated by the following equation: Mn, th. = (5 × [M]/[Initiator]) × MWMonomer + MWInitiator, with [M], [Initiator], MWMonomer and MWInitiator corresponding to monomer and initiator concentrations, and molar masses of monomer and initiator, respectively; (B) Evolution of molecular weight determined by NMR analysis (Mn, NMR) using the following equation: Mn, NMR = (∫CH3O3.6 ppm/3 × ∫CH6.3 ppm) × MWMA + MWInitiatorversus the number of cycles. |
Increasing the amount of initially added Cu(II) by a factor of two ([Cu(II)]: [CH–Br] ratio = 0.08: 1) resulted in a significant improvement of the control over the MWD with the PDI remaining lower than 1.3 (Fig. 2A). However the occurrence of star-star coupling during the first and second cycle is still evidenced by high molecular weights shoulders in the MWDs (Fig. 1B). A further increase in the amount of Cu(II) by a factor of two led to excellent control over the MWDs throughout all five polymerization cycles with PDI < 1.10 (Fig. 1C) and Mn, GPC ≈ Mn, th (Fig. 2), confirming the formation of a well-defined 5 arm star polymer after 5 successive chain extensions (Fig. 1C). 1H NMR analysis of each iteration, (without purification) reveals monomer conversions >95%, indicating that Cu(II) does not cause any marked reduction in the rate of the polymerization in agreement with results of Haddleton and co-workers.60,61
After purification of the star polymers, 1H NMR analysis was performed to determine the end group fidelity by use of the characteristic signal at 4.3 ppm (i.e., –CH–Br). The presence of the initiator was confirmed by the signals at 6.3, 5.7, 5.0–5.5 and 4.1 ppm attributed to the carbohydrate ring (Fig. 3A). Poly(MA) signals at 3.6 and 1.0–2.5 ppm are attributed to –OCH3, –CH– and –CH2–, respectively.
Fig. 3 (A) 1H NMR spectrum of poly(MA)5 star polymer obtained by Cu(0)-mediated radical polymerization of methyl acrylate in five successive iterations in the presence of different initial amounts of Cu(II); (B) Chain end functionality versus the number of cycles for different initial amounts of Cu(II). |
The CH–Br functionality (i.e. livingness) was calculated based on the equation: %CH–Br = (∫CH–Br4.3 ppm/(5 × ∫CH 6.3 ppm)) × 100, with the CH intensity corresponding to the CH of the initiator at 6.3 ppm (Fig. 3A and Figure S1 in the ESI†). Fig. 3B shows the thus obtained CH–Br functionalities versus number of cycles. As already inferred from the GPC data above, it is immediately apparent that an increase in the amount of initially added Cu(II) greatly increases the CH–Br functionality. At the highest Cu(II) content, i.e. [Cu(II)]: [CH–Br] = 0.16: 1, the CH–Br functionality after the first cycle is as high as 98%. As pointed out in our previous work,59 this is a remarkable feature of this technique considering the high level of monomer conversion. It appears that the amount of Cu(II) plays a crucial role in preserving the CH–Br functionality. At lower concentrations of Cu(II), the functionality decreases markedly with increasing number of cycles, reaching as low as 65% after 5 cycles using the lowest Cu(II) concentration. At the highest Cu(II) concentration, the CH–Br functionality is maintained above 95% after 5 cycles. This is consistent with the corresponding MWDs (Fig. 1C), which are narrow and do not feature any prominent low or high molecular weight shoulders, indicative of minimal presence of nonfunctional arms and limited, if any, star-star coupling.
The molecular weights determined by GPC for the first two blocks (for all [Cu(II)]: [CH–Br] ratios) are lower than the theoretical molecular weights (Mn, th) calculated by the equation: Mn, th. = (5 × [M]/[Initiator]) × Monomerconversion × MWMonomer + MWInitiator, with [M], [Initiator], MWMonomer and MWInitiator corresponding to monomer and initiator concentrations, and molar masses of monomer and initiator, respectively (Fig. 2A). This discrepancy is attributed to the lower hydrodynamic volume of a star polymer than a linear polymer (i.e. the GPC standards) of the same molar mass. To obtain more accurate Mn values, 1H NMR analysis was used. The molecular weight (Mn, NMR) was calculated using the equation: Mn, NMR = (∫CH3O3.6 ppm/3 × ∫CH6.5 ppm) × MWMA + MWInitiator, with ∫CH3O3.6 ppm and ∫CH6.3 ppm corresponding to the intensities from methyl acrylate (methyl ester) and initiator (CH in adjacent position of CH2O), respectively. This approach is based on the assumption that all chains contain one initiator moiety, i.e. the occurrence of bimolecular termination reactions by coupling (star-star coupling) would cause the thus obtained value of Mn to be underestimated. A better agreement between Mn, NMR and Mn, th. was obtained (Fig. 2B).
The versatility of this synthetic route to multi-block star polymers was further demonstrated by preparing two 5 arm star polymers (P(MA)5) using the same initiator as above and [Cu(II)]: [CH–Br] = 0.16:1, comprising blocks of targeted Mn of 250 and 1000 g mol−1, respectively. In both cases, well-defined star polymers were obtained after 5 cycles as evidenced by good agreement between Mn, NMR and Mn, th. and narrow MWDs (Fig. 4A–C). The CH–Br functionality decreases with number of cycles, but remains greater than 90% after 5 cycles for both target Mns (Fig. 4D). The functionality is slightly higher for the shorter block lengths as anticipated; the shorter the chains generated, the shorter is the cumulative time that each chain end spends in the active state, thus minimizing side reactions (e.g. bimolecular termination). We are currently exploring the preparation of material comprised of higher molecular weight blocks (Mn > 10 000 g mol−1) using this copper mediated polymerization technique. Theoretically, this technique can easily be extrapolated to higher molecular weight block materials, however, the ratios of the components need to be optimized.77
Fig. 4 Molecular weight and end group functionality data for multi-block star poly(MA) obtained via iterative Cu(0)-mediated radical polymerization: (A) MWDs vs. the number of cycles (block length 1000 g mol−1); (B) Mn and PDI vs. the number of cycles (block length 1000 g mol−1); (C) Mn and PDI vs. number of cycles (block length 250 g mol−1); D- End group functionality (% CH–Br chain ends at α-end) vs. number of cycles: (red dot) block length 250 g mol−1, (black square) block length 1000 g/mol. |
Finally, the synthetic methodology was applied to short multi-block star copolymers comprising a range of different monomers. We decided to use various hydrophobic and hydrophilic acrylates: hydrophobic: ethyl acrylate (EtA), tert-butyl acrylate (tBA), 2-ethylhexyl acrylate (HEA) and n-butyl acrylate (nBA), dodecyl acrylate (DA); hydrophilic: 2-dimethylaminoethyl acrylate (DMAEA) and hydroxyl acrylate (HEA). Two different star copolymers were synthesized, namely poly(PMA-b-PHEA-b-PDMEA-b-PDA) and poly(PMA-b-PnBA-b-PEA-b-PEHA-b-PtBA) (Table S1 in the ESI†). In both cases, the targeted block length was a molecular weight of approximately 500 g/mol. In the case of poly(PMA-b-PnBA-b-PEA-b-PEHA-b-PtBA), GPC analysis reveals increasing molecular weights with increasing number of cycles (Figure S4 in the ESI†), with exceptionally good control of the MWD (PDI well below 1.10) (Fig. 5A–B and Figure S2 in the ESI†). Excellent control over the polymer structure was also obtained for poly(PMA-b-PHEA-b-PDMEA-b-PDA) (Table S1 and Figure S3 in the ESI†). The Mn,th values are significantly lower than the molecular weights obtained by GPC in both cases, attributed to GPC error associated with linear poly(styrene) standards as discussed above. Therefore, the molecular weights were again estimated by 1H NMR, although the task is more complex in the case of copolymers. For example, Mn of poly(MA-b-nBA) was calculated via the equation: Mn (NMR) = (∫CH3O3.6 ppm/3 × ∫CH6.3 ppm) × MWMA + (∫CH24.1 ppm/2 × ∫CH6.3 ppm) × MWnBA + MWInitiator, where ∫CH3O3.6 ppm, ∫CH31.3 ppm, ∫CH6.3 ppm, MWMA, MWnBA and MWInitiator correspond to intensities from MA (methyl ester), CH2–O (ester) and CH of the initiator (at 6.3 ppm), molar mass of MA, nBA and initiator, respectively. For both multi-block star copolymers, the Mn values calculated by 1H NMR are in good accord with the theoretical values (Table 1 in the ESI, and Figure S2 and S3 in the ESI†). These results clearly illustrate the versatility of the present approach, which provides a simple and efficient route to multi-block star copolymers with excellent control over the polymer microstructure for a wide range of hydrophobic and hydrophilic acrylates.
Fig. 5 M n and PDI values (A) for poly(PMA-b-PnBA-b-PEA-b-PEHA-b-PtBA) and (B) poly(PMA-b-PBA-b-PEA-b-PEHA-b-PtBA) multi-block star copolymers obtained via iterative Cu(0)-mediated polymerization (approximate molecular weight of each block = 500 g mol−1, degree of polymerization ≈ 4–5). |
Footnote |
† Electronic supplementary information (ESI) available: Additional molecular weights and NMR of poly(MA) homopolymer star and block-copolymer. See DOI: 10.1039/c1py00384d |
This journal is © The Royal Society of Chemistry 2012 |