Effect of dihalides on the polymer linkages in the Cs2CO3-promoted polycondensation of 1 atm carbon dioxide and diols

Effect of dihalides on the polymer linkages in the Cs2CO3-promoted polycondensation of 1 atm carbon dioxide and diols

Materials Today Communications 18 (2019) 100–109 Contents lists available at ScienceDirect Materials Today Communications journal homepage: www.else...

NAN Sizes 0 Downloads 0 Views

Materials Today Communications 18 (2019) 100–109

Contents lists available at ScienceDirect

Materials Today Communications journal homepage: www.elsevier.com/locate/mtcomm

Effect of dihalides on the polymer linkages in the Cs2CO3-promoted polycondensation of 1 atm carbon dioxide and diols Shi Bian, Anastasia A. Andrianova, Alena Kubatova, Guodong Du


Department of Chemistry, University of North Dakota, 151 Cornell Street Stop 9024, Grand Forks, North Dakota, 58202, United States



Keywords: Polycarbonate CO2 utilization Diols Cs2CO3 Biodegradable polymers

Synthesis of biodegradable polycarbonates directly from carbon dioxide and diols has seen renewed interest. In the present report, the effect of a series of organic dihalides (CH2X2 and (CHX)2, where X = Cl, Br, and I) in the Cs2CO3-promoted copolymerization of 1 atm CO2 with diols has been studied. Structural analysis of the resultant polymers/oligomers by spectroscopic techniques reveals that a variety of main chain linkages can be obtained, depending on the dihalides used. Specifically, the methylene units from dibromomethane (DBM) are incorporated into the products as carbonate and ether linkages, and the methylene units from diiodomethane (DIM) are incorporated mainly as the ether linkage and without much CO2 incorporation. In comparison, all dihaloethanes examined lead to clean formation of alternating polycarbonates incorporating both the ethylene units from dihaloethanes and the diol units, and the halide is identified as the end group. The mechanistic pathways and the relative activity of the key intermediates are discussed in light of these results.

1. Introduction Using carbon dioxide as a feedstock for the production of chemicals and materials has been an active area of research. On the one hand, CO2 is an abundant, inexpensive and renewable C1 source. Chemicals produced from CO2 would be a desirable alternative to those currently derived from petroleum resources [1]. CO2 can also be used to replace some toxic C1 building blocks such as phosgene in syntheses [2]. On the other hand, high level of CO2 in the atmosphere is of major environmental concern [3]. Transforming CO2 into commodity chemicals would help its fixation and add value to the carbon capture and storage [4]. In this context, alternating copolymerization of CO2 with epoxides has been a promising route for the production of aliphatic polycarbonates, a class of biodegradable and biocompatible materials with various applications such as tissue engineering scaffolds and vehicles for drug delivery [5]. A wide range of homogeneous and heterogeneous catalytic systems have been reported and high activity and excellent regio- and/or stereoselectivity achieved [6]. In an indirect route, polycarbonates are produced from cyclic carbonates by ring opening polymerization (ROP) in a controlled fashion [7, 8]. Advances in organocatalytic ROP have provided alternatives to this field dominated by traditional metal-based catalysis [9]. However, these routes are limited to some degree by the requirement of the ring structure in the starting materials; therefore, direct routes for polycarbonates from diols and

CO2 have been pursued. Compared to the ring-opening copolymerization of CO2 and epoxides, polycondensation between CO2 and diols could potentially generate water byproduct that must be dealt with. One strategy is to couple the polycondensation with an efficient dehydration process. An elegant realization of this approach has been reported recently in which cerium oxide was employed as a dual catalyst for both the polycondensation of CO2 and diols and the hydration of 2-cyanopyridine, thereby removing the water byproduct instantaneously (Scheme 1, path a) [10]. The same system has also been applied for the synthesis of dialkyl and cyclic carbonates [11]. Another strategy is to add condensing agents that could divert the H2O formation to other pathways. In the presence of a condensing system consisting of phosphines PR3, carbon tetrahalides CX4 and an organic base, direct polycondensation of CO2 with diols to produce corresponding polycarbonates was accomplished (Scheme 1, path b) [12]. In the presence of a dihaloalkane, K2CO3-promoted polycondensation of CO2 and diols generated polycarbonates with both diol and dihaloalkane units incorporated into the polymer backbone (Scheme 1, path c) [13]. More recently, Gnanou and co-workers produced alternating polycarbonates (with Mn up to 43 kg/ mol) via a similar system promoted by Cs2CO3 (Scheme 1, path c) [14]. Our interests in (bio)degradable polymers in general [15] and polycarbonates in particular have prompted us to investigate the application of a Cs2CO3/dichloromethane (DCM) system that is effective in

Corresponding author. E-mail address: [email protected] (G. Du).

https://doi.org/10.1016/j.mtcomm.2018.11.007 Received 13 August 2018; Received in revised form 3 November 2018; Accepted 13 November 2018 Available online 22 November 2018 2352-4928/ © 2018 Elsevier Ltd. All rights reserved.

Materials Today Communications 18 (2019) 100–109

S. Bian et al.

Scheme 1. Synthesis of Polycarbonates from CO2 and Diols.

deemed advantageous that the reaction could be carried out under 1 atm of CO2 without use of high pressure apparatus. Therefore, the reactions in the present study were mostly set up with the same conditions, with consideration of dihalide loadings and reaction time as well. 1,4-Cyclohexanedimethanol was chosen as the representative aliphatic diol, as it is readily available and features a ring structure in the backbone that prevents the easy formation of cyclic carbonate. It should also be mentioned that the molecular weights of the resulting products are generally low, not uncommon for polycondensation reactions that require strict stoichiometry of monomers [18]. Since our focus is on the effect of dihalides on polymer/oligomer linkages, no particular attempts were made to improve the molecular weight here, though it has been shown that higher molecular weights (> 10 kg/mol) can be achieved by extending the reaction times [17].

Table 1 Effect of dihalomethanes in the copolymerization of CO2 and 1,4-cyclohexanedimethanola. Entry

CH2X2 (mmol)

t (h)


Mnc (g/mol)


1 2 3 4 5 6 7 8 9 10

CH2Cl2 (3.1) CH2Br2 (6.2) CH2Br2 (12.4) CH2Br2 (12.4) CH2Br2 (12.4) CH2Br2 (12.4) CH2I2 (3.1) CH2I2 (12.4) CH2ClBr (3.1) CH2ClBr (6.2)

24 48 12 36 48 72 24 24 24 24

73 mg (36%) – 43 mg (17%) 69 mg (31%) 77 mg (34%) 91 mg (40%) 51 mg (27%) 141 mg (79%) 79 mg (37%) 68 mg (31%)

6,400 – 1,100 3,310 3,294 3,600 1,600 2,200 1,500 1,300

2.2 – 1.6 1.2 1.3 1.8 1.5 1.5 2.1 2.3

a Reactions were performed with 1 mmol (144 mg) of 1,4-cyclohexanedimethanol (a trans and cis mixture) at 100 °C under 1 atm of CO2 in 1 ml of NMP with 4 mmol of Cs2CO3 for designated time, unless noted otherwise. Conversion of diol generally > 90%. b Yields were recorded as the mass of isolated products from 1 mmol of diol. Percent yields in parenthesis were estimated on the basis of ratios of main linkages calculated from 1H NMR and on assumption of halide end groups. c Determined by gel permeation chromatography calibrated with polystyrene standard.

2.1. Dihalomethanes To facilitate the comparison between different dihalomethanes, it would be helpful to first summarize the copolymerization of CO2 and 1,4-cyclohexanedimethanol promoted by Cs2CO3/dichloromethane (DCM) under the standard conditions. The polymeric product was isolated in 73 mg after purification, starting from 1 mmol (144 mg) scale of diol and 3.1 mmol (263 mg) of DCM at 100 °C under CO2 pressure of 1 atm (Table 1, entry 1). Replacing DCM with DBM under otherwise identical conditions resulted in no polymer formation, even with extended reaction time and larger loading of DBM (entry 2). After the loading of DBM was raised to 12.4 mmol, the isolation of polymers was noted with low yield (43 mg) and low molecular weight (entry 3). To achieve higher yield and molecular weight, we investigated the reaction at longer times with 12.4 equivalents of DBM. Indeed, the extended reaction time led to an increase in yields: 69 mg after 36 h; 77 mg after 48 h, and 91 mg after 72 h (entries 4–6). However, the molecular weights of the resultant polymers were not improved considerably. The highest molecular weight was only up to 3600 g/mol at 72 h, close to the result at reaction time of 36 h (3310 g/mol), but significantly higher than that of 12 h (1100 g/mol). Next, we examined the performance of diiodomethane (DIM) as a reagent in the reaction. The polymer would form with lower loading of DIM (3.1 mmol) and shorter time (24 h) (Table 1, entry 7). Higher yield (141 mg) were obtained with 12.4 equivalents of DIM, though the molecular weight of the resultant polymer was only slightly higher, 2200 g/mol (entry 8). A mixed dihahomethane, bromochloromethane (BCM), was also employed in the reaction. A yield of 79 mg was achieved at 24 h with 3.1 mmol of BCM and the molecular weight of the product was only about 1500 g/mol (entry 9). Increase of BCM loading had minor effect on the reaction outcome (entry 10).

generating dialkylcarbonates from CO2 and alcohols [16]. Polymeric products were isolated with up to 75% yield and moderately high molecular weight (up to 11,100 g/mol) [17]. Structural analysis of the products revealed that a new type of polymers with alternating ethercarbonate linkages was generated with the incorporation of methylene units from DCM (Scheme 1, path d). Because of the −OCH2O- unit, it may also be viewed as an alternating copolymer of acetals and carbonates. It thus appears that the nature of dihaloalkanes may play a crucial role in the formation of polycarbonate backbone. In the present study, we focus our efforts on the effect of various dihalides, specifically a series of dihalomethanes CH2X2 and dihaloethanes (CH2X)2 on the copolymerization of CO2 with diols in the presence of Cs2CO3. 2. Results and discussion In our previous study, the activity of Cs2CO3/DCM in CO2-diols copolymerization was investigated under the standard conditions (100 °C reaction temperature; 1 atm of CO2). Variations of dihalide loadings and reaction time were shown to have a significant impact on the yield of the copolymerization reaction. Other factors such as the pressure of CO2 and reaction temperature might come to interplay with different dihalides, but we kept the CO2 pressure constant since it was 101

Materials Today Communications 18 (2019) 100–109

S. Bian et al.

Fig. 1. (a)1H NMR spectra of polymer from 1,4-cyclohexanedimethanol synthesized with DBM. (b)13C NMR spectra of polymer from 1,4-cyclohexanedimethanol synthesized with DBM.

to the methylene protons (a and b) between two oxygen atoms. This observation indicated two different -OCH2O- linkages in the main chain. In the earlier study with DCM, the dihalide-derived methylene proton signals appeared at 5.3 ppm and were attributed to a -OCH2Ounit between a carbonate linkage and an ether linkage [17]. The comparison of their 1H NMR spectra (Fig. 2) clearly showed that neither of the two new resonances belong to the methylene between a

2.2. Microstructure The microstructure of the products from the above reactions were investigated by the NMR and FT-IR spectroscopic techniques, which revealed some notable features. For the polymer/oligomer obtained with DBM (entry 4, Table 1), there were two singlet peaks at 5.70 ppm and 4.58 ppm in the 1H NMR spectrum (Fig. 1a) that could be assigned 102

Materials Today Communications 18 (2019) 100–109

S. Bian et al.

Fig. 2. Comparison of 1H NMR spectra of polymer synthesized with different dihalomethanes.

carbonate linkage and an ether linkage. We thus assigned the new resonances to the methylene protons between two carbonate linkages (5.7 ppm), and two ether linkages (4.6 ppm), respectively. In the 13C NMR spectrum (Fig. 1b), the two peaks around 85 ppm and 96 ppm were assignable to the methylene (a, b) carbons, respectively, as confirmed by an HSQC NMR experiment. Furthermore, the incorporation of CO2 was supported by the carbonate carbon signal at 154 ppm in the 13 C NMR and by the absorbance at 1740 cm−1 in the FT-IR spectrum. Thus the methylene unit from DBM was mainly incorporated between carbonate linkages, same as that observed in the literature with α,ωdihalides, but different from the case of DCM. A minor part of the methylene unit from DBM was incorporated as ether linkages in the backbone. The ratio of the carbonate linkage and ether linkage was 1.9:1, as indicated by the integrations of the peaks at 5.7 ppm (methylene protons between carbonates) and 4.6 ppm (methylene protons between ethers). Consistent with these assignments, the majority of the diol unit ended up between two carbonates, and a minor part between two ethers linkages: the peaks around 4.0 ppm could be assigned to the diol-derived methylene moiety (c) connected to a carbonate, and the peaks around 3.4 ppm assigned to the methylene group (d) connected to an ether linkage. It should be pointed out that the signals around 4.0 and 3.4 ppm are greater than the corresponding methylene units at 5.7 and 4.6 ppm, which could be accounted for by the presence of additional minor linkages, e.g. diol unit between two carbonates (would appear at 4.0 ppm) and diol-derived ethers (would appear at 3.4 ppm), as well as the end groups such as -OCH2Br (vide infra). The carbonate and ether linkages in the structure were further investigated with an HMBC NMR experiment (Fig. 3). The connections of C1-C2 and C2-C3 were established by the cross peaks of C2-H1 and C2H3, and the connection of C5-C4 was established by the cross peak of C4-H5. Analysis using electrospray with time of flight mass spectrometry (ESI-ToF-MS) revealed the presence of these two repeating units (∼244.1 Da for carbonate linkage -OC8H14-CO2-OCH2-CO2- in blue and ∼156.1 Da for ether linkage –OC8H14-OCH2- in black), thus supporting the existence of the two different linkages (Fig. 4). The somewhat unsystematic appearance of these linkages suggested they were randomly connected in the backbone. It should be pointed out that the MS data might be not representative of the main components obtained.

Fig. 3. HMBC NMR of polymer from 1,4-cyclohexanedimethanol synthesized with DBM.

However, in combination of NMR characterization, it does provide corroborating evidence for the linkages. With these assignments in mind, it was relatively straightforward to figure out the linkages in the polymers/oligomers obtained from DIM and BCM by comparing their 1H NMR spectra (Fig. 2). In case of DIM, it was clear that the diol monomer was mostly (∼84%) incorporated as ether (peaks at 3.3 ppm) and only a small portion (∼16%) as carbonates (peaks around 4.0 ppm). In keeping with this observation, the DIM-derived methylene unit showed up mostly (∼98%) in ether range (4.6 ppm) with a minor part (∼2%) in the carbonate range (5.7 ppm). Thus, polyethers between DIM and diol were predominately generated without much incorporation of CO2. When a mixed dihalide BCM was employed, the product was similar to the polymers synthesized from DCM in that the majority (∼55%) of dihalide-derived methylene unit appeared at around 5.3 ppm, indicative of an ether-carbonate linkage. Notably there was appreciable amount of ether linkages (around 4.6 ppm) but no carbonate linkages (5.7 ppm). These assignments and 103

Materials Today Communications 18 (2019) 100–109

S. Bian et al.

Fig. 4. ESI-ToF-MS of polymer from 1,4-cyclohexanedimethanol synthesized with DBM (Mn = 3600 g/mol, Mw/Mn = 1.8, entry 6, Table 1). The mass accuracy error of the repeating units was 3–30 ppm.

Scheme 2. The linkages of polymers obtained from different dihalides.

we replaced dihalomethanes with dihaloethanes to further study the effect of halides (Table 2). Remarkably, the combinations of Cs2CO3 with dihaloethanes, such as 1,2-dichloroethane, promoted the polycondensation of CO2 and diols exhibiting even higher activity, consistent with the observation in literature [14a]. High isolated yields and high molecular weight of the products were achieved with only 1.5 mmol of 1,2-dibromoethane (DBE) under the optimized reaction condition (12.4 mmol were needed in case of DBM), and the reaction time was shorter than with DBM (Table 2, entry 2). At DBE loadings of 3.1 mmol and 6.2 mmol, even higher yields were isolated though the

their compositions were summarized in Scheme 2. Obviously, the results here demonstrated that the formation of polymer/oligomer linkage was significantly dependent on the dihalides used in the reaction.

2.3. Dihaloethanes It is of note that α,ω-dihalides were typically employed in the literature reports, whereas dihalomethanes are quite different from α,ωdihalides in that both halides are connected to the same carbon. Thus, 104

Materials Today Communications 18 (2019) 100–109

S. Bian et al.

spectrum of the polymer/oligomer synthesized from trans-1,4-cyclohexanedimethanol and DBE was shown in Fig. 5. The two main resonances at 4.37 ppm and 3.99 ppm of equal intensity were assigned to the two different methylene groups, one derived from DBE (a) and the other derived from diol (b), directly connected to the carbonate linkages, while the two minor resonances at 4.44 ppm and 3.55 ppm were assignable to the end groups derived from dihalides (e, f). These assignments were supported by the 13C and HSQC NMR spectra as well as the ESI-MS of the isolated products. The equal intensity of the 4.37 and 3.99 ppm peaks suggested a likely alternating arrangement of diol units and dihalide units in the polymer linkage. The corresponding C12H18O6 unit with ∼258.1 Da was also detected in the ESI-ToF-MS. Together these results indicated an exclusive formation of alternating –CO2OC8H14-CO2-OCH2CH2- linkage in the polycarbonate. When 1,2-dichloroethane (DCE) was used as the dihalide, the resultant polymer featured the same two main resonances in the carbonate linkage region in the 1H NMR spectrum. The resonances representing the end group connected with chloride now shifted to 3.73 ppm (vs 3.55 ppm with bromide), which was expected given that chlorine is more electronegative than bromine. In agreement with the alternating nature of the structure, the ESI mass spectrum of the polymer (Fig. 6) indicated a series of peaks at 258n +98 + 18 with a charge of +1, which can be assigned to n(C12H18O6) +C2H4Cl2+NH4+. These data also agree with the assignment of the end groups of -CH2CH2Cl on both sides. Notably, the same set of signals appeared in the 1H NMR spectrum of the product with BCE as that of the polymer produced with DCE, but different from the DBE-derived polymers. It suggested that only chloride remained in the polymer chain end while bromide were all gone. On the basis of these observations with the chain ends resulting from dihaloethanes, it was reasonable to assume that the polymer products from dihalomethanes might have analogous chain ends, i.e., that derived from dihalides –OCH2X. When 1,2-diiodoethane was used as the dihalide, formation of the ether linkage -OC8H14-OCH2CH2dominated the product with minimum incorporation of CO2, similar to the results obtained with DIM.

Table 2 Effect of dihaloethanes in copolymerization of CO2 and 1,4-cyclohexanedimethanola. Entry

(CH2X)2 (mmol)


Mnc (g/mol)


1 2 3 4 5 6 7 8 9

(CH2Br)2 (6.2) (CH2Br)2 (1.5) (CH2Br)2 (3.1) (CH2Cl)2 (6.2) (CH2Cl)2 (1.5) (CH2Cl)2 (3.1) CH2ClCH2Br (1.5) CH2ClCH2Br (3.1) CH2ClCH2Br (1.0)

283 mg (98%) 244 mg (93%) 219 mg (80%) 87.4 mg (31%) 52.3 mg (19%) 255 mg (93%) 235 mg (89%) 246 mg (90%) 153 mg (57%)

2,900 9,800 3,500 1,100 1800 1,900 5,000 1,600 2,400

1.4 1.7 1.5 1.4 1.2 1.4 2.0 1.5 1.6

a Reactions were performed with 1 mmol of trans-1,4-cyclohexanedimethanol at 100 °C under 1 atm of CO2 in 1 mL of NMP with 4 mmol of Cs2CO3 for 24 h. Conversions were more than 95%. b Isolated products by mass in mg. Data in parenthesis were percent yields calculated assuming GPC-based Mn and halide end groups. c Determined by GPC calibrated with polystyrene standard.

molecular weight of the product decreased significantly from 9.8 kg/ mol to 2.9 kg/mol (Table 2, entries 1, 3). When 1,2-dibromoethane was replaced by 1,2-dichloroethane (DCE), a substantial drop in activity was observed (Table 2, entries 4 & 5). Even when a high yield was observed with 3.1 mmol of 1,2-dichloroethane, the molecular weights obtained with 1,2-dichloroethane was much lower than with 1,2-dibromoethane (Table 2, entry 6). The activity of 1-bromo-2-chloroethane (BCE) in the polycondensation was somewhat in between: higher than 1,2-dichloroethane but lower than 1,2-dibromoethane (Table 2, entries 7–9). Another noteworthy observation was that the best performance of the reaction with 1,2-dichloroethane and 1-bromo-2-chloroethane was not achieved by either the highest or lowest loadings of dihalides. Instead, the highest yields were achieved with 3.1 equivalent of dihalides while the highest molecular weights were achieved with 1.5 equivalent of dihalides. The characterization of the products by the NMR and FT-IR spectroscopies indicated that there were exclusively carbonate linkages for all three dihaloethanes examined, and both diol and dihalide units were incorporated between two carbonates in the main chain. These results were different from that of dihalomethanes but agreed with the results from other α,ω-dihalides in the literature [13,14]. The 1H NMR

2.4. Mechanistic Considerations It is evident that the choice of dihalides employed, as well as the nature of the diol monomers, may have a significant effect on the

Fig. 5. 1H NMR spectrum of polymer from 1,4-cyclohexanedimethanol (trans) with DBE. 105

Materials Today Communications 18 (2019) 100–109

S. Bian et al.

Fig. 6. ESI-ToF-MS of polymer from 1,4-cyclohexanedimethanol synthesized with DCE (Mn = 1800 g/mol, Mw/Mn = 1.2, entry 5, Table 2).

Scheme 3. Proposed pathways for the copolymerization of the diols and CO2 with dihalides.

either alkoxide Int-A or carbonate Int-B. Even though the alkoxide IntA is a better nucleophile, the concentration of the carbonate Int-B is likely higher such that only its nucleophilic attack on Int-D (via site a) leads to chain growth with the methylene unit between two carbonate linkages. At the same time, carbonate Int-B is not nucleophilic enough and only alkoxide Int-A can react with Int-C, leading to the methylene unit between two ether linkages. The preference of methylene incorporation into carbonate linkage was indicated by the 1H NMR spectra of the product (ratio of ether/carbonate is about 1.0/1.9). In both scenarios, the cross reactions between Int-C and carbonate Int-B or between Int-D and alkoxide Int-A are not significant, and no or only small amount of ether-carbonate linkages were generated, which was in contrast with the results observed with DCM [17]. In case of DIM, iodide is a better leaving group than bromide such that the alkoxide Int-A reacts preferentially with DIM over CO2. As a result, not much carbonate Int-B is formed in the presence of DIM and polyether –OCH2OC8H14- is generated through reaction between Int-C (X = I) and alkoxide Int-A, and without much incorporation of CO2. In case of DCM, Int-D (X = Cl) is less active than Int-D (X = Br) and reacts preferably with the more nucleophilic alkoxide Int-A, leading to the ether-

incorporation of CO2 and the main chain linkages produced. The Cs2CO3-based systems promoting organic carbonates synthesis from CO2 have been studied in the literature in detail [19]. In conjunction with our results here and in previous study, we can envisage a reaction scheme that reasonably accommodates most observations, at least in a qualitative sense (Scheme 3). The reaction without CO2 under otherwise identical conditions afforded small amount of carbonate linkage (< 10%), suggesting the carbonate linkage mostly comes from the gaseous CO2. Similar results have been observed previously [14a]. The deprotonation of a hydroxyl group of the diol by Cs2CO3 generates an alkoxide Int-A, which then acts as a nucleophile to attack CO2 giving rise to a carbonate Int-B. Since excess CO2 is present under the reaction conditions, the majority of alkoxide Int-A is likely converted to carbonate Int-B. Both the alkoxide Int-A and carbonate Int-B would react with dihalide, generating intermediates Int-C and Int-D, respectively. These two species bear electrophilic sites (Int-D has two such sites, a and b) and can further react with alkoxide Int-A or carbonate Int-B. Under the previous assumption, Int-D would likely be in higher concentration than Int-C. This process repeats and generates oligomeric and polymeric products. In case of DBM, Int-D (X = Br) can react with 106

Materials Today Communications 18 (2019) 100–109

S. Bian et al.

carbonate linkage. Alternatively, reaction between Int-C and carbonate Int-B can also give rise to the ether-carbonate linkage. The above arguments can be reasonably applied to the case of BCM. Obviously bromide is a better leaving group than chloride, so with BCM the attack by alkoxide Int-A or carbonate Int-B would leave the chloride end intact, generating Int-C (X = Cl) or Int-D (X = Cl). Both will react with alkoxide Int-A to form ether linkage and ether-carbonate linkage, respectively. For the dihaloethane series (CH2X)2, because the carbon center of the ethylene unit now only has one halide connected, it becomes less nucleophilic than the case of CH2X2 and the formation of Int-C type intermediate is minimal compared to the formation of carbonate Int-B. Similarly site b of the ensuing Int-D type intermediate becomes the preferred position for nucleophilic attack by Int-B, leading to nearly exclusive formation of alternating carbonate linkages, regardless of the different dihaloethanes employed.

CH2X2 and dihaloethanes (CH2X)2 have been examined as the reagent in the Cs2CO3-promoted copolymerization of CO2 with diols under 1 atm pressure of CO2. The isolated polymeric products have been fully investigated by spectroscopic techniques such as IR, NMR and GPC, from which the structures of the polymers/oligomers are derived. While dihaloethanes invariably lead to the formation of alternating polycarbonates incorporating both the ethylene units from dihaloethanes and the diol units, regardless of the identity of the halides, dihalomethanes lead to a wider range of backbone structures depending on the nature of the halides. Specifically, the methylene units from dibromomethane (DBM) are incorporated into the polymeric products as carbonate and ether linkages, and the methylene units from diiodomethane (DIM) are incorporated mainly as the ether linkage and without much CO2 incorporation. These observations are different from the previous results with dichloromethane (DCM), in which an alternating ether-carbonate linkage is identified as the main product. The variation in the main chain linkages may be attributed to the subtle differences in reactivity of key intermediates involved in the reaction network and their interplay with concentrations. The insights learned here might point to further directions in controlling the reactivity of the system as well as the resulting structures.

2.5. Diols with DBM Given the uniqueness of dihalomethanes compared to α,ω-dihalides in general, we next investigated a variety of primary diols in the direct copolymerization with CO2 promoted by the Cs2CO3-DBM system under the conditions identified above (Table 3), and the products were characterized by the NMR and IR spectroscopies. For regular aliphatic primary diols, both carbonate and ether linkages derived from DBM were observed in the backbone, as judged by the presence of singlets around 5.7 ppm and 4.6 ppm assignable to the DBM-derived methylene groups in the 1H NMR. The reaction using trans-1,4-cyclohexanedimethanol (Table 3, entry 2) had slightly lower isolated yield than the reaction using a trans and cis mixture (Table 3, entry 1), and the molecular weights were similar (in the range of 3.0–4.0 kg/mol). When a linear aliphatic primary diol, 1,6-hexanediol, was employed in the reaction, relatively low yield was obtained (Table 3, entry 3). It was possible that a higher percentage of shorter chain oligomers or cyclic products could form with such linear diols when compared with diols bearing a more rigid backbone; these components were not readily precipitated out from the mixture during the isolation. This idea was supported by the low number average molecular weight of the product from 1,6-hexanediol (Mn ∼1.5 kg/mol). When a benzylic diol, 1,3benzenedimethanol, was used (Table 3, entry 4), the polymer/oligomer products contained predominantly carbonate linkages (> 90%) with no dihalide methylene incorporation. Unlike the cases with aliphatic diols, DBM and DCM here behaved similarly and led to the same type of backbone linkages. In other words, the benzylic diols are more selective towards polycarbonate formation than the aliphatic diols in these reactions.

4. Experimental section 4.1. Materials and methods Reagent grade organic compounds and solvents including 1,4-cyclohexanedimethanol, 1,6-hexanediol, 1,3-benzenedimethanol, dichloromethane, dibromomethane, diiodomethane, bromochloromethane, 1,2-dichloroethane, 1,2-dibromoethane, 1-bromo-2chloroethane and N-methyl-2-pyrrolinone (NMP) were purchased from MilliporeSigma and trans-1,4-cyclohexanedimethanol was purchased from TCI. The diols were dried under vacuum overnight and dihalides were dried over activated molecular sieves (4 Å) overnight prior to use. The pre-dried Cs2CO3 and NMP were used directly. All compounds were after drying stored in a glove box under nitrogen atmosphere. The NMR spectra were recorded on a Bruker AVANCE-500 NMR spectrometer (1H, 13C and 2D). The FT-IR spectra were recorded on a Perkin-Elmer Spectrum 400 FT-IR spectrometer. Gel permeation chromatography (GPC) analysis was performed on a Varian Prostar instrument calibrated with polystyrene standards, using a PLgel 5 μm MixedD column, a Prostar 355 RI detector, and THF as eluent at a flow rate of 1 mL/min (20 °C). High resolution time-of-flight mass spectrometry (HR ToF MS) with electrospray ionization (ESI) was employed (G1969 A, Agilent Technologies, Santa Clara, CA) in a positive ionization mode. The electrospray (capillary) and fragmentor voltages were set to 5500 and 225 V, respectively. Nitrogen at a flow rate of 4 L/min was used as a nebulizing gas. A 5 μg/mL solution of an analyte in acetonitrile-water 1:1 (v/v) containing 2.5 mM of ammonium acetate as an electrolyte was

3. Conclusions In summary, a series of organic dihalides including dihalomethanes Table 3 Copolymerization of CO2 and diols with Cs2CO3/DBMa. Yield (mg)

Mnb (g/mol)






















AB (> 90%)



a Reactions were performed with 1 mmol of diols at 100 °C under 1 atm of CO2 in 1 ml of NMP with 4 mmol of Cs2CO3 and 12.4 mmol of DBM for 72 h. Conversions of diols generally greater than 95%. b Determined by gel permeation chromatography. c The ratio of AB, AC, BB, and CC type linkages (see Scheme 2 for their structures) estimated by 1H NMR.


Materials Today Communications 18 (2019) 100–109

S. Bian et al.

introduced using direct infusion at 5 μL/min. The differential scanning calorimetry (DSC) data were collected on a Perkin Elmer Pyris DSC using 10.0 °C/min heating rate from -30 °C to 150 °C with 20 mL/min nitrogen flow.

79 mg, 1H NMR (CDCl3, 298 K): δ 5.28 (s, OCH2O, 2 H), 4.82 (m, OCH2O, 1 H), 4.66 (m, OCH2O, 1 H), 4.00 (m, COOCH2CH, 4 H), 3.45 (m, OCH2CH, 6 H), 1.84 (s, CHCH2CH2, 6 H), 1.68 (s, CH2CHCH2, 2 H), 1.56 (s, CH2CHCH2, 2 H), 1.44 (s, CH2CHCH2, 2 H), 1.02 (s, CHCH2CH2, 6 H). 13C NMR (CDCl3, 298 K): δ 154.98 (C=O), 95.81 (OCH2O), 92.80 (OCH2O), 76.25 (OCH2CH), 73.24 (OCH2CH), 70.93 (OCH2CH), 68.73 (OCH2CH), 40.61 (CH2CHCH2), 38.42 (CH2CHCH2), 37.37 (CH2CHCH2), 35.64 (CH2CHCH2), 34.83 (CH2CHCH2), 30.89 (CHCH2CH2), 29.04 (CHCH2CH2), 25.57 (CHCH2CH2). Polycarbonate from 1,4-cyclohexanedimethanol with 1,2-dichloroethane, Yield: 255.3 mg, 1H NMR (CDCl3, 298 K): δ 4.40 (m, COOCH2CH2Cl, 1 H), 4.36 (s, COOCH2CH2, 4 H), 3.98 (m, COOCH2CH, 4 H), 3.73 (m, CH2CH2Cl, 1 H), 1.85 (d, CHCH2CH2, 4 H), 1.68 (s, CH2CHCH2, 2 H), 1.04 (m, CHCH2CH2, 4 H). 13C NMR (CDCl3, 298 K): δ 155.25 (C=O), 73.43 (OCH2CH), 67.32 (CH2CH2Cl), 65.44 (OCH2CH2), 41.42 (CH2CH2Cl), 37.20 (CH2CHCH2), 28.75 (CHCH2CH2). Polycarbonate from 1,4-cyclohexanedimethanol with 1,2-dibromoethane, Yield: 243.7 mg, 1H NMR (CDCl3, 298 K): δ 4.44 (m, COOCH2CH2Cl, 1 H), 4.37 (s, COOCH2CH2, 11 H), 3.99 (m, COOCH2CH, 11 H), 3.55 (m, CH2CH2Cl, 1 H), 1.86 (d, CHCH2CH2, 12 H), 1.67 (s, CH2CHCH2, 7 H), 1.04 (m, CHCH2CH2, 12 H). Polycarbonate from 1,4-cyclohexanedimethanol with 1-bromo2-chloroethane, Yield: 246 mg, 1H NMR (CDCl3, 298 K): δ 4.41 (m, COOCH2CH2Cl, 1 H), 4.37 (s, COOCH2CH2, 10 H), 3.99 (m, COOCH2CH, 10 H), 3.72 (m, CH2CH2Cl, 1 H), 1.86 (d, CHCH2CH2, 10 H), 1.68 (s, CH2CHCH2, 6 H), 1.04 (m, CHCH2CH2, 10 H). 13C NMR (CDCl3, 298 K): δ 155.25 (C=O), 73.42 (OCH2CH), 67.33 (CH2CH2Cl), 65.45 (OCH2CH2), 41.43 (CH2CH2Cl), 37.16 (CH2CHCH2), 28.69 (CHCH2CH2).

4.2. Typical procedure for polymerization In a dry nitrogen filled glove box, a 100 mL pre-dried Schlenk flask was loaded with 1 mmol of a diol, 4 mmol of Cs2CO3, 12.8 mmol of dibromomethane and 1 mL of N-methyl-2-pyrrolidone (NMP). The closed flask was taken out and cooled with liquid N2. After the reaction mixture was frozen, the N2 atmosphere was replaced with CO2 through three evacuation-refill cycles. The reaction mixture was heated at 100 °C for 24 h with stirring. The conversion of diol was checked by 1H NMR of the crude mixture. Two different purification procedures were applied depending on the diol substrates. For aliphatic diols, the reaction mixture was filtered and washed with DCM (5 mL x 3). The combined filtrate was concentrated to ∼ 1 mL under vacuum. Methanol (5 mL) was then added, and the precipitate was collected by centrifugation and further washed with methanol (1 mL x 3). For benzylic diol, the reaction mixture was treated with water (10 mL), and the insoluble solid was collected by filtration or centrifugation, and washed with methanol (2 mL x 3). Finally, the solid product was dried under vacuum to constant weight to determine the yield. Given the multiphasic nature of the reaction system and laborious isolation procedure, most reactions were performed multiple times, which showed fairly consistent and reproducible yields and linkages. Polycarbonate from 1,4-cyclohexanedimethanol with DBM, Yield: 91 mg, 1H NMR (CDCl3, 298 K): δ 5.70 (s, OCH2O, 2 H), 4.58 (s, OCH2O, 1 H), 4.04 (d, COOCH2CH, 1 H), 3.96 (d, COOCH2CH, 3 H), 3.89 (d, COOCH2CH, 0.5 H),3.48 (s, OCH2CH, 0.5 H), 3.40 (d, OCH2CH, 1 H), 3.27 (s, OCH2CH, 2 H), 1.86 (s, CHCH2CH2, 1 H), 1.77 (s, CHCH2CH2, 6 H), 1.61 (s, CH2CHCH2, 2 H), 1.49 (s, CH2CHCH2, 3 H), 1.37 (s, CH2CHCH2, 3 H), 0.94(m, CHCH2CH2, 5 H). 13C NMR (CDCl3, 298 K): δ 154.16 (C=O), 95.80 (OCH2O), 84.85 (OCH2O), 74.04 (OCH2CH), 73.84 (OCH2CH), 73.40 (OCH2CH), 71.64 (OCH2CH), 68.65 (OCH2CH), 38.48 (CH2CHCH2), 37.40 (CH2CHCH2), 37.09 (CH2CHCH2), 34.57 (CH2CHCH2), 29.37 (CHCH2CH2), 25.55 (CHCH2CH2). Polycarbonate from 1,3-benzendimethanol with DBM, Yield: 62 mg, 1H NMR (CDCl3, 298 K): δ 7.35 (m, ArH, 8 H), 5.16 (s, ArCH2OCOO, 8 H), 4.70 (s, ArCH2O, 1 H), 4.64 (s, ArCH2O, 1 H). 13C NMR (CDCl3, 298 K): δ 155.18 (C=O), 135.79, 129.17, 128.96, 128.67, 128.35, 127.92, 127.78, 127.31, 127.03, 126.49, 126.31, 125.81, 69.90 (ArCH2OCOO), 69.63 (ArCH2OCOO), 69.14 (ArCH2OCOO), 65.44 (ArCH2O), 65.22 (ArCH2O). Polycarbonate from 1,6-hexanediol with DBM, Yield: 35 mg, 1H NMR (CDCl3, 298 K): δ 5.77 (s, OCH2O, 1 H), 4.67 (s, OCH2O, 2 H), 4.21 (m, COOCH2CH2, 5 H), 3.66 (m, OCH2CH2, 2 H), 3.53 (m, OCH2CH2, 4 H), 1.70 (s, OCH2CH2CH2, 6 H), 1.62 (s, OCH2CH2CH2, 11 H), 1.41 (s, OCH2CH2CH2, 12 H). 13C NMR (CDCl3, 298 K): δ 154.03 (C=O), 95.45 (OCH2O), 84.76 (OCH2O), 69.14 (OCH2CH2), 68.62 (OCH2CH2), 68.02 (OCH2CH2), 63.10 (OCH2CH2), 32.90 (OCH2CH2CH2), 32.76 (OCH2CH2CH2), 29.90 (OCH2CH2CH2), 29.78 (OCH2CH2CH2), 28.85 (OCH2CH2CH2), 28.75 (OCH2CH2CH2), 28.63 (OCH2CH2CH2), 28.55 (OCH2CH2CH2), 26.24 (OCH2CH2CH2), 26.09 (OCH2CH2CH2), 25.76 (OCH2CH2CH2). Polyether from 1,4-cyclohexanedimethanol with DIM, Yield: 141 mg, 1H NMR (CDCl3, 298 K): δ 4.79 (s, OCH2O, 1 H), 4.66 (s, OCH2O, 3 H), 3.47 (m, OCH2CH, 1 H), 3.34 (d, OCH2CH, 8 H), 1.83 (m, CHCH2CH2, 13 H), 1.54 (s, CH2CHCH2, 4 H), 0.99 (m, CHCH2CH2, 10 H). 13C NMR (CDCl3, 298 K): δ 95.75 (OCH2O), 74.33 (OCH2CH), 73.47 (OCH2CH), 73.18 (OCH2CH), 68.89 (OCH2CH), 40.77 (CH2CHCH2), 38.43 (CH2CHCH2), 37.45 (CH2CHCH2), 29.65 (CHCH2CH2). Polyether from 1,4-cyclohexanedimethanol with BCM, Yield:

Data availability The processed data required to reproduce these findings are available to download from [https://data.mendeley.com/submissions/ evise/edit/p2bwrgbhsg?submission_id=S0014-3057(18)31532-5& token=10b54d05-ff55-41c3-a952-a72508c67f32]. Acknowledgements This work is supported by the National Science Foundation under Grant No. NSF EPSCoR Award IIA-1355466. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation. References [1] (a) Z. Li, J. Wang, Y. Qu, H. Liu, C. Tang, S. Miao, Z. Feng, H. An, C. Li, ACS Catal. 7 (2017) 8544–8548; (b) C. Das Neves Gomes, O. Jacquet, C. Villiers, P. Thuéry, M. Ephritikhine, T. Cantat, Angew. Chem., Int. Ed. 51 (2012) 187–190; (c) O. Jacquet, C. Das Neves Gomes, M. Ephritikhine, T. Cantat, J. Am. Chem. Soc. 134 (2012) 2934–2937. [2] (a) D. Pati, X. Feng, N. Hadjichristidis, Y. Gnanou, J. CO2 Util. 24 (2018) 564–571; (b) M. Aresta, A. Dibenedetto, Dalton Trans. (2007) 2975–2992; (c) T. Sakakura, J.C. Choi, H. Yasuda, Chem. Rev. 107 (2007) 2365–2387; (d) M. North, R. Pasquale, C. Young, Green Chem. 12 (2010) 1514–1539. [3] C. Federsel, R. Jackstell, M. Beller, Angew. Chem. Int. Ed. 49 (2010) 6254–6257. [4] (a) M. Tamura, M. Honda, Y. Nakagawa, K. Tomishige, J. Chem. Technol. Biotechnol. 89 (2014) 19–33; (b) D. Chaturvedi, Tetrahedron 68 (2012) 15–45; (c) B. Yu, L.N. He, ChemSusChem 8 (2015) 52–62; (d) M. Cokoja, C. Bruckmeier, B. Rieger, W.A. Herrmann, F.E. Kühn, Angew. Chem., Int. Ed. 50 (2011) 8510–8537; (e) L. Zhang, Z. Hou, Chem. Sci. 4 (2013) 3395–3403; (f) M. Aresta, A. Dibenedetto, A. Angelini, Chem. Rev. 114 (2014) 1709–1742. [5] (a) J. Xu, E. Feng, J. Song, J. Appl. Polym. Sci. 131 (2014) 39822; (b) W. Chen, F. Meng, R. Cheng, C. Deng, J. Feijen, Z. Zhong, J. Control. Release 190 (2014) 398–414; (c) F. Suriano, R. Pratt, J.P.K. Tan, N. Wiradharma, A. Nelson, Y.-Y. Yang,


Materials Today Communications 18 (2019) 100–109

S. Bian et al.


[7] [8]



(2016) 24038. [11] (a) A. Bansode, A. Urakawa, ACS Catal. 4 (2014) 3877–3880; (b) M. Honda, M. Tamura, Y. Nakagawa, K. Nakao, K. Suzuki, K. Tomishige, J. Catal. 318 (2014) 95–107; (c) M. Honda, M. Tamura, Y. Nakagawa, S. Sonehara, K. Suzuki, K. Fujimoto, K. Tomishige, ChemSusChem 6 (2013) 1341–1344; (d) M. Honda, M. Tamura, K. Nakao, K. Suzuki, Y. Nakagawa, K. Tomishige, ACS Catal. 4 (2014) 1893–1896. [12] (a) J. Kadokawa, H. Habu, S. Fukamachi, M. Karasu, H. Tagaya, K. Chiba, Macromol. Rapid Commun. 19 (1998) 657–660; (b) J. Kadokawa, S. Fukamachi, H. Tagaya, K. Chiba, Polym. J. 32 (2000) 703–706. [13] S. Oi, K. Nemoto, S. Mastuno, Y. Inoue, Macromol. Rapid Commun. 15 (1994) 133–137. [14] (a) Z. Chen, N. Hadjichristidis, X. Feng, Y. Gnanou, Polym. Chem. 7 (2016) 4944–4952; (b) D. Pati, Z. Chen, X. Feng, N. Hadjichristidis, Y. Gnanou, Polym. Chem. 8 (2017) 2640–2646; (c) Z. Chen, N. Hadjichristidis, X. Feng, Y. Gnanou, Macromolecules 50 (2017) 2320–2328. [15] (a) S. Abbina, V.K. Chidara, G. Du, ChemCatChem 9 (2017) 1343–1348; (b) S. Vijjamarri, S. Streed, E.M. Serum, M.P. Sibi, G. Du, ACS Sustainable Chem. Eng. 6 (2018) 2491; (c) S. Vijjamarri, M. Hull, E. Kolodka, G. Du, ChemSusChem 11 (2018) 2881–2888; (d) S. Abbina, G. Du, ACS Macro Lett. 3 (2014) 689–692; (e) S. Vijjamarri, V.K. Chidara, G. Du, ACS Omega 2 (2017) 582–591; (f) S. Bian, S. Abbina, Z. Lu, E. Kolodka, G. Du, Organometallics 33 (2014) 2489–2495. [16] Y. Yamazaki, K. Kakuma, Y. Du, S. Saito, Tetrahedron 66 (2010) 9675–9680. [17] S. Bian, C. Pagan, A.A. Andrianova, G. Du, ACS Omega 1 (2016) 1049–1057. [18] N. Nomura, K. Tsurugi, T.V. RajanBabu, T. Kondo, J. Am. Chem. Soc. 126 (2004) 5354–5355. [19] (a) S.-I. Kim, F. Chu, E.E. Dueno, K.W. Jung, J. Org. Chem. 64 (1999) 4578–4579; (b) G. Dijkstra, W.H. Kruizinga, R.M. Kellogg, J. Org. Chem. 52 (1987) 4230; (c) M.R. Reithofer, Y.N. Sum, Y. Zhang, Green Chem. 15 (2013) 2086.

P. Dubois, J.L. Hedrick, Biomaterials 31 (2010) 2637–2645; (d) P. Alagi, R. Ghorpade, Y.J. Choi, U. Patil, I. Kim, J.H. Baik, S.C. Hong, ACS Sustain. Chem. Eng. 5 (2017) 3871–3881. (a) S. Paul, Y. Zhu, C. Romain, R. Brooks, P.K. Saini, C.K. Williams, Chem. Commun. (Camb.) 51 (2015) 6459–6479; (b) M.I. Childers, J.M. Longo, N.J. Van Zee, A.M. LaPointe, G.W. Coates, Chem. Rev. 114 (2014) 8129–8152; (c) B. Li, G.-P. Wu, W.-M. Ren, Y.-M. Wang, D.-Y. Rao, X.-B. Lu, J. Polym. Sci. Part A: Polym. Chem. 46 (2008) 6102; (d) S.I. Vagin, R. Reichardt, S. Klaus, B. Rieger, J. Am. Chem. Soc. 132 (2010) 14367; (e) X.-B. Lu, W.-M. Ren, G.-P. Wu, Acc. Chem. Res. 45 (2012) 1721; (f) M.W. Lehenmeier, S. Kissling, P.T. Altenbuchner, C. Bruckmeier, P. Deglmann, A.K. Brym, B. Rieger, Angew. Chem. Int. Ed. 52 (2013) 9821; (g) S. Abbina, G. Du, Organometallics 31 (2012) 7394–7403; (h) S. Abbina, V.K. Chidara, S. Bian, A. Ugrinov, G. Du, ChemistrySelect 1 (2016) 3175–3183; (i) L. Jin, H. Zeng, A. Ullah, Polym. Chem. 8 (2017) 6431–6442. S.M. Guillaume, J.-F. Carpentier, Catal. Sci. Technol. 2 (2012) 898–906. (a) V.W.L. Ng, J.P.K. Tan, J. Leong, Z.X. Voo, J.L. Hedrick, Y.Y. Yang, Macromolecules 47 (2014) 1285–1291; (b) W. Guerin, A.K. Diallo, E. Kirilov, M. Helou, M. Slawinski, J.-M. Brusson, J.F. Carpentier, S.M. Guillaume, Macromolecules 47 (2014) 4230–4235; (c) J. Feng, R.-X. Zhuo, X.-Z. Zhang, Prog. Polym. Sci. 37 (2012) 211–236; (d) P. Brignou, S.M. Guillaume, T. Roisnel, D. Bourissou, J.F. Carpentier, Chem. Eur. J. 18 (2012) 9360–9370. (a) A.P. Dove, ACS Macro Lett. 1 (2012) 1409; (b) F. Nederberg, E.F. Connor, M. Möller, T. Glauser, J.L. Hedrick, Angew. Chem. Int. Ed. 40 (2001) 2712; (c) B.G.G. Lohmeijer, R.C. Pratt, F. Leibfarth, J.W. Logan, D.A. Long, A.P. Dove, F. Nederberg, J. Choi, C. Wade, R.M. Waymouth, J.L. Hedrick, Macromolecules 39 (2006) 8574–8583; (d) S.E. Felder, M.J. Redding, A. Noel, S.M. Grayson, K.L. Wooley, Macromolecules (51) (2018) 1787–1797. M. Tamura, K. Ito, M. Honda, Y. Nakagawa, H. Sugimoto, K. Tomishige, Sci. Rep. 6