Hiroaki
Shimomoto
*,
Takumi
Kubo
,
Kazunari
Nishigawa
,
Hazuki
Okuda
,
Makoto
Ishimoto
,
Tomomichi
Itoh
and
Eiji
Ihara
*
Department of Applied Chemistry, Graduate School of Science and Engineering, Ehime University, 3 Bunkyo-cho, Matsuyama 790-8577, Japan. E-mail: shimomoto.hiroaki.mx@ehime-u.ac.jp; ihara@ehime-u.ac.jp; Fax: +81-89-927-9949; Fax: +81-89-927-8547; Tel: +81-89-927-9949, +81-89-927-8547
First published on 6th January 2025
A series of diazoacetates with an ester-type dendron unit of 1st to 3rd generation as an ester substituent were newly prepared and their Pd-initiated polymerization was conducted to afford dendronized C1 polymers with the SEC-estimated number-average molecular weight (Mn,SEC) of up to 16700 and a relatively narrow molecular weight distribution (Đ = 1.12–1.37). The glass transition temperature (Tg) of these dendronized polymers increased with the increasing generation of the dendron unit, where the degree of increase (ΔTg) was much larger than that observed with their vinyl polymer counterparts (polymethacrylates with the same dendron substituents). The diazoacetate with the 3rd generation dendron unit was successfully copolymerized with benzyl diazoacetate to afford a copolymer, whose dendron-derived units were converted to those with multiple hydroxy groups via deprotection of acetonide-protecting groups at the peripheral positions.
As initiators for the polymerization, some Rh and Pd complexes have been reported to be effective, yielding high number-average molecular weight (Mn) polymers in good yield. Our group has been focusing on the development of Pd-based initiating systems for the diazoacetate polymerization, reporting some effective systems such as (NHC)Pd(nq)/NaBPh4 (NHC = N-heterocyclic carbene, nq = naphthoquinone),29 π-allylPdCl/NaBPh4,30,31 and (nq)2Pd/NaBPh4.32 Although these initiating systems can afford high Mn polymers in good yield, it is difficult to control Mn of the product, and the controlled polymerization of diazoacetate with Pd-based initiating systems is one of our current important objectives. (Some Pd-based initiators have been reported to be effective for the controlled polymerization of diazoacetate by other research groups.33–35)
Meanwhile, in the course of our research, we have found and reported that, when the monomer has a sterically bulky ester substituent such as cyclotriphosphazene13,36 and benzyl ether-type dendrons,26 the π-allylPdCl/NaBPh4-initiated polymerization of such diazoacetates proceeds in a controlled manner to give polymers with a narrow molecular weight distribution (MWD). In particular, for the latter monomer, we have revealed that the use of the dendron unit with a higher generation leads to a narrower MWD of the C1 polymer products.26 In addition, we introduced protected carboxyl groups at the peripheral positions of the benzyl ether-type dendron unit of the monomer, and after its polymerization followed by deprotection we observed enhanced acidity of the resulting C1 polymer in comparison to its vinyl polymer counterpart.
Herein, we attempted to introduce ester-type dendron units37–40 in place of the above-described benzyl ether-type dendron units as an ester substituent of the monomers. Whereas the ester-type dendron has been utilized for preparation of dendronized polymers via ring-opening metathesis polymerization (ROMP) and radical polymerization so far,41–45 the results obtained here will be the first example of a dendronized C1 polymer with an ester-type dendron. In addition, because the C1 polymer with a benzyl-ether type dendron described above is the only example of a dendronized C1 polymer, this attempt to prepare new diazoacetates with dendron units and to investigate their polymerization behavior should be worthwhile for extending the scope of dendronized polymers46–51 as well as developing the promising chemistry of C1 polymerization. Diazoacetate with an ester-type dendron substituent with a higher generation is expected to undergo controlled polymerization in a similar manner to that with a benzyl ether-type dendron26 because of the steric bulkiness of the substituent as well. In addition, because the acetonide-protecting groups at the peripheral positions of the dendron unit can be removed to liberate multiple hydroxy groups, the incorporation of the ester-type dendron unit via copolymerization with other diazoacetates can be an effective general method for introducing highly hydrophilic substituents into the C1 polymer backbone. We report the synthesis of new diazoacetates with a series of ester-type dendron substituents and their polymerization behavior with π-allylPdCl/NaBPh4 as an initiator. In addition, copolymerization of a dendron-containing monomer with other less bulky diazoacetates and deprotection of the acetonide-protecting group to yield C1 copolymers with multiple hydroxy groups are described.
Then, as shown in Scheme 2, the 1st generation diazoacetate monomer G1-D was successfully prepared by diazoacetylation of the precursor alcohol G1-OH with the standard procedure reported by Fukuyama and coworkers.52 In addition, as summarized in Scheme 2, starting from G1-Si, the systematic divergent syntheses of G2-D and G3-D were also realized by repeating (1) the acetonide-deprotection with CuCl2·2H2O, (2) esterification of the hydroxy groups with 1, (3) silyl deprotection, and (4) diazoacetylation. Experimental procedures and characterization data for the isolated new compounds are described in the ESI.†
Run | Monomer (M) | Temp. | [M]/[Pd] | Yieldb (%) | M n | Đ | M n,MALS | Đ MALS |
T
g![]() |
---|---|---|---|---|---|---|---|---|---|
a In THF for 15 h, [monomer] = ca. 0.1 M, [NaBPh4]/[Pd] = ca. 1.2. b Determined by gravimetry after purification with preparative SEC. c Determined by SEC calibrated using PMMA standards. d M n,MALS values were determined by SEC-MALS. The average degree of polymerization (DP) values were calculated from Mn,MALS. Apparent initiator efficiency (IE) values were calculated from Mn,MALS and polymer yield. e Determined by DSC measurements. | |||||||||
1 | G1-D | RT | 50 | 48 | 4900 | 1.29 | |||
2 | G1-D | −20 °C | 50 | 56 | 6500 | 1.29 | 10![]() |
1.34 | 5 |
3 | G1-D | −20 °C | 100 | 61 | 10![]() |
1.32 | |||
4 | G1-D | −20 °C | 200 | 64 | 16![]() |
1.37 | |||
5 | G1-D | −20 °C | 300 | 51 | 13![]() |
1.30 | |||
6 | G2-D | RT | 50 | 35 | 4200 | 1.12 | |||
7 | G2-D | −20 °C | 50 | 49 | 8700 | 1.19 | 12![]() |
1.21 | 10 |
8 | G2-D | −20 °C | 100 | 50 | 10![]() |
1.14 | |||
9 | G2-D | −20 °C | 200 | 42 | 14![]() |
1.31 | |||
10 | G3-D | −20 °C | 50 | 30 | 7000 | 1.16 | 39 |
As for the polymerization of G2-D, the results of polymerization at −20 °C are again better than those at room temperature (runs 6 and 7), and with the increase of the [G2-D]/[Pd] feed ratio to 100 and 200 at −20 °C (runs 8 and 9), the Mn,SEC increased up to 14200. Compared to the results of G1-D, MWDs of p(G2-D)s obtained at −20 °C were narrower in the range of Đ = 1.14–1.31; these low Đ values cannot be considered as higher controllability of G2-D than that of G1-D but indicated the same level of controllability, on the basis of the comparison with the data reported for their vinyl polymer counterparts as discussed later in this paper. Meanwhile, the polymer yield became slightly lower (42–50%), suggesting that the sterically larger ester substituent in G2-D brought about lower reactivity because of its steric demand during the polymerization. Fig. 1 shows 1H NMR spectra of G2-D and p(G2-D)′ of a sample obtained in run 6, where we can observe broad signals in the spectrum of p(G2-D)′ corresponding to each signal in the spectrum of the monomer except for the signal for the H on the diazo-bearing carbon atom, which was converted to a main chain CH exhibiting a broad signal at 3.0–3.8 ppm in the spectrum of p(G2-D)′.
In contrast to G1-D and G2-D affording polymers in moderate yield (ca. 50%), the polymerization of G3-D with a [G3-D]/[Pd] feed ratio of 50 afforded a polymer in a low yield of 30% (run 10), suggesting that the ester substituent of G3-D is too bulky for the polymerization to proceed effectively. Also, the narrow MWD of the p(G3-D)′ (Đ = 1.16) suggests that the polymerization of G3-D would proceed with the same level of controllability as those of G1-D and G2-D as well (see discussion below).
SEC-MALS measurements were carried out for the samples in runs 2 and 7 in Table 1. As reported in our previous publication for C1 polymers obtained from diazoacetates, the Mn,MALS values were 1.5–1.7 times higher than those estimated by standard PMMA-calibrated SEC. In addition, on the basis of the Mn,MALS and polymer yield, the apparent initiator efficiency ([Pd acting as an initiator]/[Pd employed]) was calculated for these samples and is listed in Table 1. These data indicate that while 66% of π-allylPdCl employed for the polymerization of G1-D acted as an initiator, the Pd complex almost quantitatively participated in the initiation in the case of polymerization of G2-D. These results suggest more efficient initiation of G2-D with bulkier ester substituents than G1-D.
Fortunately for us, A. Hult and coworkers reported the preparation and physical properties of polymethacrylates with the same ester-type dendron as ours with 1st to 4th generation via atom transfer radical polymerization (ATRP).41,42 Coincidentally, as shown in Chart 1, the oxyethylene spacer unit between the dendron and the main chain of the polymethacrylates is exactly the same as ours, making the comparison here more relevant. In their report, they prepared polymethacrylate with a 1st generation dendron via ATRP and polymers with higher generations via divergent transformation (post-polymerization modification) therefrom. Importantly to us, the SEC-determined Đ value gradually decreased with the increase of the dendron generation, from 1.24 to 1.11 (Chart 1), even though the polymethacrylate backbone is identical for these dendronized polymers. These results indicated that probably because of the higher rigidity and bulkiness of the polymer chain with the increase of the dendron generation, the SEC-determined Đ value became lower. Accordingly, the results of the apparent narrower MWD with the increase of dendron generation in the C1 polymerization of G(1–3)-D cannot be considered as improved controllability of the polymerization as well. However, these results at least indicate that the relatively high controllability for p(G1-D)′ is retained for p(G2-D)′ and p(G3-D)′.
![]() | ||
Chart 1 Comparison of Mn, Đ, and Tg values between p[G(1–3)-D]′ and the corresponding polymethacrylates. |
Fig. 2 shows the charts of DSC measurements of p[G(1–3)-D]′ samples obtained in runs 2, 7, and 10. It is clearly demonstrated that the glass transition temperature (Tg) increased with the increase of the dendron generation, indicating that the increased side chain bulkiness would restrict the main chain mobility. In particular, the higher Tg (39 °C) of p(G3-D)′ despite its lower Mn indicates that the steric demand caused by the ester-type dendron G3 unit led to severe restriction for the main chain mobility.
With respect to the observed Tgs, the results reported by A. Hult and coworkers41,42 are again quite helpful for us to conduct reasonable comparison of the C1 polymers with their vinyl polymer counterparts. They reported that the Tg increase was 0.7 °C with the dendron generation increase from G2 to G3 as listed in Chart 1. In contrast, the Tgs of our polymers increased from 10 to 39 °C, whose ΔTg is a much larger value of 29 °C. These results clearly demonstrate that because of the denser packing of the ester-type dendron in our dendronized C1 polymers without any CH2 unit in the main chain, the restriction of the main chain mobility caused by the generation increase from G2 to G3 is enhanced to a much greater extent compared to their vinyl polymer counterparts.
![]() | ||
Fig. 3 1H NMR spectra of (a) p[(G3-D)′-co-BDA′] (run 1 in Table 2) and (b) its deprotected product. |
Run | Feed ratio ([G3-D]/[BDA]) | Yieldb (%) | Compositionc ([G3-D′]/[BDA′]) | M n | Đ | M n,MALS | Đ MALS | T g (°C) |
---|---|---|---|---|---|---|---|---|
a In THF for 15 h; [NaBPh4]/[Pd] = ca. 1.2. b Determined by gravimetry after purification with preparative SEC. c Determined by 1H NMR measurements. d Determined by SEC calibrated using PMMA standards. e Determined by SEC-MALS. f Determined by DSC measurements. | ||||||||
1 | 1![]() ![]() |
52 | 1![]() ![]() |
12![]() |
1.44 | |||
2 | 1![]() ![]() |
41 | 1![]() ![]() |
13![]() |
1.42 | 27![]() |
1.23 | 16 |
3 | 0![]() ![]() |
67 | 11![]() |
1.48 | 18![]() |
1.33 | 26 |
Run 3 in Table 2 shows the results of homopolymerization of BDA under the same conditions as those in runs 1 and 2. Although the comparison is not straightforward because the molecular weight of G3-D is much higher than that of BDA, it is at least confirmed that the use of G3-D as a comonomer did not prevent the progress of the polymerization of BDA.
Next, we attempted to remove the acetonide-protecting groups of the copolymer sample obtained in run 1 in Table 2 with aqueous HCl-treatment in THF. As shown in Fig. 3(b), disappearance of the Me–H signals derived from the acetonide group after the HCl-treatment was clearly observed in comparison to the precursor polymer in Fig. 3(a). Although the presence of many hydrophilic hydroxy groups at the peripheral positions of the dendron unit could change the polymer solubility,53,54 the product after the deprotection was still soluble in organic solvents such as CHCl3, probably because of the relatively low content of the G3-D unit in the copolymer. These results indicate that the copolymerization of G3-D with other diazoacetates can be an effective method for introducing the highly hydrophilic repeating unit in the C1 polymer after the deprotection.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4py01324g |
This journal is © The Royal Society of Chemistry 2025 |