Poly(monothiocarbonate)s from the Alternating and Regioselective

Sep 27, 2016 - Ming Luo was born in 1987 in China. He received his B.S. degree in Polymer Science and Engineering from Zhejiang University (ZJU) in 20...
1 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/accounts

Poly(monothiocarbonate)s from the Alternating and Regioselective Copolymerization of Carbonyl Sulfide with Epoxides Ming Luo,† Xing-Hong Zhang,*,† and Donald J. Darensbourg*,‡ †

MOE Key Laboratory of Macromolecular Synthesis and Functionalization, Department of Polymer Science and Engineering, Zhejiang University, Hangzhou 310027, China ‡ Department of Chemistry, Texas A&M University, College Station, Texas 77843, United States CONSPECTUS: Carbonyl sulfide (COS) is an air pollutant that causes acid rain, ozonosphere damage, and carbon dioxide (CO2) generation. It is a heterocumulene and structural analogue of CO2. Relevant to organic synthesis, it is a source of CO or CS groups and thus an ideal one-carbon (C1) building block for synthesizing sulfur-containing polymers through the similar route of CO2 copolymerization. In contrast, traditional synthesis of sulfur-containing polymers often involves the condensation of thiols with phosgene and ring-opening polymerization of cyclic thiocarbonates that are generally derived from thiols and phosgene; thus, COS/epoxide copolymerization is a “greener” route to supplement or supplant current processes for the production of sulfur-containing polymers. This Accounts highlights our efforts on the discovery of the selective formation of poly(monothiocarbonate)s from COS with epoxides via heterogeneous zinc−cobalt double metal cyanide complex (Zn−Co(III) DMCC) and homogeneous (salen)CrX complexes. The catalytic activity and selectivity of Zn−Co(III) DMCC for COS/epoxide copolymerization are similar to those for CO2/epoxide copolymerization. (salen)CrX complexes accompanied by onium salts exhibited high activity and selectivity for COS/epoxide copolymerization under mild conditions, affording copolymers with >99% monothiocarbonate units and high tailto-head content up to 99%. By way of contrast, these catalysts often show moderate or low activity for CO2/epoxide copolymerization. Of note, a specialty of COS/epoxide copolymerization is the occurrence of an oxygen−sulfur exchange reaction (O/S ER), which may produce carbonate and dithiocarbonate units. O/S ER, which are induced by the metal−OH bond regenerated by chain transfer reactions, can be kinetically inhibited by changing the reaction conditions. We provide a thorough mechanistic understanding of the electronic/steric effect of the catalysts on the regioselectivity of COS copolymerization. The regioselectivity of the copolymerization originates from the solely nucleophilic attack of the sulfur anion to methylene of the epoxide, and thus, the chiral configuration of the monosubstituted epoxides is retained. COS-based copolymers are highly transparent sulfur-containing polymers with excellent optical properties, such as high refractive index and Abbe number. Thanks to their good solubility and many available epoxides, COS/epoxide copolymers can potentially be a new applicable optical material. Very recently, crystalline COS-based polymers with or without chiral carbons have been synthesized, which may further expand the scope of application of these new materials.

■

INTRODUCTION

been proven to be a major source of acid rain because it can be oxidized to SO2 in the troposphere. Meanwhile, because of its relative inertness, COS can be transported from the troposphere to the stratosphere, where it undergoes photooxidation and damages the ozonosphere. In nature, COS consumption is mostly caused by the irreversible hydration process COS + H2O → CO2 + H2S, most likely catalyzed by the enzyme carbonic anhydrase (CA).4 COS is also an injurant in industry because it poisons many industrial catalysts and corrodes the production equipment.5 Therefore, it should be effectively captured and stored for possible utilization. However, currently, research on COS utilization is quite limited. For the past few decades, the chemical

Carbonyl Sulfide (COS): Origin and Utilization

COS, also known as carbon oxide sulfide or carbon oxysulfide, was first discovered in 1867, when it was prepared by Than.1 The chemistry and properties of COS were comprehensively reviewed by Ferm1 in 1957. COS is indeed an important intermediate of the atmospheric sulfur cycle and the most abundant sulfur-containing gas in the troposphere.2 As pictured in Figure 1, COS is widely released from marine plants in the ocean, volcano eruptions, the decomposition of sulfur compounds on land, and now more often from the consumption of fossil fuels.2 The most current data about the concentration of COS in the troposphere was reported to be 500 ppt (1.3 mg/m3) in 1980.3 This number is obviously increasing along with the rapid development of the modern industry. The overall emission of COS has made it an atmospheric pollutant (Figure 1). It has © 2016 American Chemical Society

Received: July 4, 2016 Published: September 27, 2016 2209

DOI: 10.1021/acs.accounts.6b00345 Acc. Chem. Res. 2016, 49, 2209−2219

Article

Accounts of Chemical Research

electrolytes, as well as optical, optoelectronic, and photochemical materials.20 The syntheses of poly(thiocarbonate)s are summarized in Scheme 2. Previously, these were synthesized by condensation of dithiols and phosgene (route A),21 which is not an environmentally friendly synthetic route due to the use of very toxic reactants. Poly(thiocarbonate)s can also be prepared by the ringopening polymerization of cyclic mono-, di-, or trithiocarbonates, which were first synthesized from either epoxides/CS2 coupling or thiophosgene/diol cyclization (route B), and thus, further monomer purification was necessary.22 The direct coupling of CS2 with epoxides to synthesize poly(thiocarbonate)s was initially reported by Adachi et al. in 1977,23 where diethyl zinc and various bases were used as the catalysts. In this instance, long reaction times were required, and the molecular weights of the obtained polymers were less than 560 g/mol. The copolymerization of ethylene sulfide and propylene sulfide (PS) with CS2 were reported by Soga, producing low molecular weight copolymers with irregular chain structure.24 In 2007, Nozaki et al. reported CS2/PS copolymerization catalyzed by the (salen)CrCl [(salen)H2 = bis(3,5-di-tert-butylsalicylidene)-1,2-diamine] complex in conjunction with bis(triphenylphosphoranylidene)ammonium chloride (PPNCl ) (route C). This process afforded a fully alternating poly(trithiocarbonate) with a high refractive index.25

Figure 1. Life of COS in the atmosphere and our method of the chemical fixation of COS via COS/epoxide copolymerization, affording poly(monothiocarbonate)s.

research on COS has mostly focused on the removal of COS from industrial processes. The principal methods include hydrogenation, hydrolysis, photolysis, and oxidation.5 COS is often used as a synthon for synthesizing thio-acids, sulfursubstituted carbinols, thiazoles, and carbamic acids.3 It is also employed to synthesize an agricultural fumigant serving as a substitute for more widely used fumigants, such as methyl bromide and phosphine. It is more effective and less toxic than the other two compounds.3 As an analogue of carbon dioxide (CO2) and carbon disulfide (CS2), COS is a heterocumulene containing CO or CS groups and can be regarded as an asymmetric form of CO2 and CS2. It could be an ideal one-carbon (C1) monomer for synthesizing sulfur-containing polymers via a similar route to CO2 copolymerization (Scheme 1). Motivated by the achieve-

Oxygen−Sulfur Exchange Reaction (O/S ER)

Polythiocarbonates

From 2008 to 2015, we have performed the coupling of CS2 with PO,26 cyclohexene oxide (CHO),27,28 cyclopentene oxide (CPO),29 and oxetane30 by using either heterogeneous zinc− cobalt double cyanide complex [Zn−Co(III) DMCC] catalyst or homogeneous (salen)CrCl complex (Scheme 3) (route D). Unfortunately, oxygen/sulfur scrambling was observed in both the polymeric and cyclic products, resulting in irregular polymer chain structure and poor selectivity. Consider CS2/PO copolymerization catalyzed by Zn−Co(III) DMCC for instance,26 different cyclic products and complicated polymer linkages including ether, thioether, carbonate, and thiocarbonate were determined by nuclear magnetic resonance (NMR) and Fourier transform infrared spectroscopy (FT-IR). Such complex products resulted from the randomly crossed coupling of the coexistent species, i.e., CS2, PO, COS, CO2, and PS (inferred from the existent thioether linkages). Zn−Co(III) DMCC could not only catalyze the copolymerization but also act as the exchanging site for the scrambling in the reaction, as shown in Figure 2. However, this mechanism still remains as conjecture. Moreover, when (salen)CrCl complexes were employed, similar results were observed.27,29 Importantly, COS was captured as a key intermediate resulting from O/S ER, and it was proven to be enriched during CS2/ epoxide copolymerization.26,30 Moreover, the monothiocarbonate linkage is always the major component in the polymer chain confirmed by 13C NMR in all of these reported CS2/epoxide coupling systems.26 These discoveries inspired us to utilize COS as a C1 monomer for synthesizing poly(monothiocarbonate)s with well-defined structure and investigating the O/S ER mechanism.

Sulfur-containing polymers are important because they possess good or excellent electrical, mechanical, and optical features as well as good properties such as adhesion to metals, resistance to heat, chemicals, radiation, and bacteria, and biocompatibility.20 Consequently, they have the potential to be utilized as high performance engineering plastics, chemically stable ion-exchange membranes in electromembrane processes, proton-conducting

COS/EPOXIDE COPOLYMERIZATION Since we first reported the completely alternating COS/PO copolymerization catalyzed by the binary (salen)CrCl/PPNCl system in 2013,15 various catalysts have been utilized for the COS/epoxide copolymerization, and these are summarized in Scheme 3. Numerous epoxides including aliphatic, alicyclic, and

Scheme 1. Synthesis of Polycarbonates from CO2/Epoxide Copolymerizationa6−14

a

R1 and R2 represent substituent groups.

ments of CO2/epoxide copolymerization in the past decade,6−14 we have undertaken research on COS/epoxide copolymerization for synthesizing poly(monothiocarbonate) (Figure 1).15 Herein, we discuss this new method to prepare sulfur-containing polymers by using COS as a monomer in which sulfur atoms can be introduced into the polymer chain in an atom-economy synthetic manner. 16−19 Moreover, COS can be simply synthesized by heating carbon monoxide and sulfur and thus has significant potential as a low-cost monomer for coupling with various epoxides.1

■

2210

DOI: 10.1021/acs.accounts.6b00345 Acc. Chem. Res. 2016, 49, 2209−2219

Article

Accounts of Chemical Research Scheme 2. Reported Methods for Synthesizing Various Poly(thiocarbonate)s

Scheme 3. Catalysts and Cocatalysts Reported for COS/Epoxide Copolymerization15−19,31,32a

a

The ground state of the catalytic center of a was proposed as a tetrahedral zinc ion with 1 Zn−OH bond (or Zn−Cl bond); CA is a complexing agent.33,34

the asymmetric molecular structure of COS. When COS copolymerizes with a monosubstituted epoxide, e.g., PO in Scheme 4, if O/S ER occurred, CO2 and PS would be produced similar to the CS2/PO copolymerization process.26 The mutual copolymerization of the coexistent species (PO, COS, CO2, and PS) would produce a copolymer with possible linkages P1−P6 and cyclic products C1−C6. Our research results showed that

aromatic substituted ones have been investigated to successfully copolymerize with COS, as shown in Figure 3. Special Chemistry of COS/Epoxide Copolymerization

Being different from CO2/epoxide copolymerization, the chemistry of COS/epoxide copolymerization is of interest and more complicated due to the possible occurrence of O/S ER and 2211

DOI: 10.1021/acs.accounts.6b00345 Acc. Chem. Res. 2016, 49, 2209−2219

Article

Accounts of Chemical Research

linkage. When COS copolymerized with epoxide, possessing an electron-donating group such as PO15 or phenyl glycidyl ether (PGE),18 the T-H content of the resulting copolymer was determined to be around 99.0%. Such excellent regioselectivity results from the sequential attack of the sulfur anion at the less hindered methylene carbon of the epoxide, indicating the a-a route (Scheme 4) was the predominant pathway for the propagation. Catalytic Activity and Polymer Selectivity

The reported copolymerization of COS with various epoxides are summarized in Table 1. We initially studied the COS/PO copolymerization in 2013 using b1−b3 (Scheme 3) in conjunction with PPNCl (entries 1−3 in Table 1).15 These studies revealed high turnover frequencies (TOF) of 288−332 h−1. The resulting poly(monothiocarbonate) possessed an Mn of 25.3 kg/mol and alternating degree of up to 100% because O/S ER was totally suppressed at 25 °C. The T-H linkage was determined to be 98−99% based on the 13C NMR results. The ligand difference in b1−b3 had no observable effect on the activity or selectivity of copolymerization. Note that O/S ER occurred and more cyclic thiocarbonate was produced when the temperature was higher than 50 °C in this system because high reaction temperature and water favored the chain transfer reaction, which caused the generation of Cr−OH species. In contrast, when these binary (salen)CrCl catalysts were used to catalyze CO2/PO copolymerization, this process has been plagued by low catalyst activity and the concomitant production of propylene carbonate.8,35 We carried out comparative temperature-dependent kinetic studies of CO2/PO relative to CO2/ CHO coupling reactions utilizing b3 as the catalyst, which showed that at low temperature (30 °C) only polymer product was produced, whereas upon raising the temperature (80 °C), cyclic carbonate was afforded.35 Moreover, bifunctional (salen)CrNO3 catalyst d1 (Scheme 3) showed remarkable TOF (25700 h−1) for the COS/PO copolymerization (entry 4) in the absence O/S ER even at an elevated temperature of 80 °C, providing a fully alternating

Figure 2. O/S ER in CS2/PO copolymerization via Zn−Co(III) DMCC catalysis.

the O/S ER could be completely depressed in most cases by providing a dry reaction system and low reaction temperature. In this manner, various poly(thiocarbonate)s with 100% polymer selectivity and 100% P1 linkage selectivity were synthesized.15 In addition, the asymmetric structure of COS results in special regioselectivity in COS/epoxide copolymerization processes. As shown in Scheme 4, when COS is inserted into the propagating center, the nucleophilic attack can proceed in two routes: A and B. It has been clearly shown that route A is the pathway selected for COS insertion because the nucleophilicity of sulfur atom to the metal center is stronger than that of the oxygen atom of COS.15 Furthermore, the CO bond is thermodynamically stronger than the CS bond. After the insertion of COS, the sulfur anion of the propagating center could attack the methylene or methine of PO, and after two insertions of COS and PO, four consecutive monothiocarbonate diads might be produced: headto-tail (H-T), tail-to-head (T-H), tail-to-tail (T-T), and head-tohead (H−H). Herein, the T-H diad is different from the H-T diad due to the asymmetric structure of the monothiocarbonate

Figure 3. Timeline of the syntheses of COS-based copolymers from different epoxides.15−19,31,32 2212

DOI: 10.1021/acs.accounts.6b00345 Acc. Chem. Res. 2016, 49, 2209−2219

Article

Accounts of Chemical Research

Scheme 4. (Upper) If O/S ER Occurs, P1−P6 and Cyclic Products C1−C6 Will Be Formed; (Bottom) Four Possible Diads of the COS/PO Copolymer Formed by Various Regioselective Attacks in Route A

copolymer with an Mn of 220.0 kg/mol.31 However, the regioselectivity was not investigated. When the Cr3+ was replaced by Co3+, i.e., d2 in Scheme 3, rather low activity was exhibited at ambient temperature (entry 5). The elevation of the reaction temperature to 80 °C could improve TOF but caused a decrease of the polymer selectivity (entry 6). These results indicated the Co(III)-based catalyst is less suitable for COS/PO copolymerization. However, d2 was a highly active and selective catalyst for CO2/PO copolymerization. To date, the Zn−Co(III) DMCC catalyst is the only heterogeneous catalyst for the COS/epoxide copolymerization.14,16 It displays similar selectivity for both the CO2/epoxide and the COS/epoxide copolymerization.16,36 Utilizing Zn− Co(III) DMCC catalyst for the COS/PO copolymerization provided a copolymer with an irregular and complicated chain structure, indicating the occurrence of severe O/S ER. However, this catalyst performed better for the COS/CHO copolymerization.16 That is, COS/CHO copolymerization afforded copolymer with an Mn of 24.7 kg/mol, an alternating degree of 93%, and a small amount of ether linkages (entry 7), which was similar to CO2/CHO copolymerization by the same catalyst.36 However, Zn−Co(III) DMCC afforded lower productivity of 940 g of polymer/g of catalyst for COS/CHO copolymerization (110 °C, 5 h) than that for CO2/CHO copolymerization (8400 g of polymer/g of catalyst, 5 h, 80 °C).36 O/S ER could be largely suppressed at high reaction temperature (100 °C) and when tetrahydrofuran was added, provied a copolymer with 90% monothiocarbonate linkages and 2% carbonate linkages. The virtue of Zn−Co(III) DMCC for preparing COS-based copolymer is that the copolymer is often highly transparent

and colorless even without purification, whereas the poly(monothiocarbonate) from (salen)CrCl is always a little pale yellow even when washed repeatedly. The COS/CHO copolymerization via the b2/PPNCl catalysis produced a fully alternating copolymer with an Mn of 12.3 kg/ mol and PDI of 1.13 at 40 °C for 3 h without the observation of cyclic product and the occurrence of O/S ER (entry 8). The TOF was up to 325 h−1. When PO was introduced as a third monomer, the activity and molecular weight was improved. The COS/CHO/PO terpolymerization catalyzed by b2/PPNCl system afforded terpolymers with a 100% polymer selectivity and a TOF of 323 h−1 at 40 °C for 3 h (entry 11).17 The terpolymer possessed a Mn of 22.3 kg/mol and narrow PDI of 1.16. The same phenomenon was reported by Ren et al. using the bifunctional (salen)CrNO3 (d1) catalyst.31 d1 showed a TOF of 260 h−1 for COS/CHO copolymerization (entry 9). Of interest, the addition of a small amount of PO resulted in rapid conversion of CHO with a TOF of 15800 h−1 (entry 10). It was similar for COS/CPO copolymerization (entry 23), where the addition of a small amount of PO accelerated the reaction, affording a TOF of 1360 h−1. The alternating and regioselective COS/PGE copolymerization was investigated using the b3/PPNCl system,18 affording copolymers with >99% polymer selectivity, 100% monothiocarbonate linkages, and 96−98% T-H content (entries 12 and 13). Through comparisons with our early work,15−17 this copolymerization process exhibited unprecedented activity at room temperature, where within 6 min, the conversion of PGE was up to 73% (TOF: 7300 h−1). The copolymer had a Mn of 40.8 kg/ mol and narrow PDI of 1.2. Remarkably, d1 exhibited a TOF as 2213

DOI: 10.1021/acs.accounts.6b00345 Acc. Chem. Res. 2016, 49, 2209−2219

Article

Accounts of Chemical Research Table 1. Representative Reported Successful Copolymerization of COS with Epoxidesa entry

epoxideb

catalytic systemc

T (°C)

TOFd (h−1)

polymer selectivitye (%)

Mnf (kg/mol)

PDIf

T-Hg (%)

Tgh (°C)

ref

1 2 3 4 5 6 7i 8 9 10j 11k 12l 13 14 15 16 17i 18 19 20 21 22 23 24 25 26m

1 1 1 1 1 1 2 2 2 2 1+2 3 3 3 4 4 4 4 4 5 6 7 8 9 9 9

b1, PPNCl b2, PPNCl b3, PPNCl d1 d2 d2 a b2, PPNCl d1 d1 b2, PPNCl b3, PPNCl b3, PPNCl d1 b3, PPNCl b4, PPNCl a c1, PPNCl c2, PPNCl d1 d1 d1 d1 b3, TBD d3 d3

25 25 25 80 25 80 110 40 80 80 40 20 20 80 20 20 50 30 30 80 80 80 80 25 25 −25

288 310 332 25700 34 103 940 325 260 15800 323 7300 500 48800 83 83 394 36 40 25700 21500 53900 1360 ND ND ND

99 99 99 >99 99 60 >99 >99 >99 >99 >99 >99 >99 >99 >99 >99 >99 84 93 >99 >99 >99 >99 75 98 99

24.4 21.9 25.3 220.0 11.5 4.3 24.7 12.3 12.3 45.5 22.3 22.6 40.8 123.0 77.2 41.2 11.5 32.8 41.6 124.0 190.0 135.0 66.8 0.5 0.8 3.1

1.26 1.35 1.41 1.25 1.21 1.33 2.10 1.10 1.30 1.23 1.16 1.12 1.20 1.36 1.18 1.14 2.01 1.15 1.18 1.34 1.38 1.38 1.28 2.13 1.88 1.37

99 99 98 ND ND ND

ND ND 22.4 30.0 ND ND 112.4 114.6

15 15 15 31 31 31 16 17 31 31 17 18 18 31 19 19 19 19 19 31 31 31 31 32 32 32

98 99 ND 78 78 84 52 50 ND ND ND ND ND ND

105.5 55.9 ND 44.9 14.1 78.4 ND ND ND ND 15.0 2.9 13.7 55.8 ND ND ND

a

These results are selected from the reported articles without occurrence of O/S ER based on 13C NMR spectra; monothiocarbonate linkage >99% from 1H NMR spectra. ND = not determined. bIf not mentioned, the epoxide is racemic (Figure 3). cSee Scheme 3. dTurnover frequency (TOF) = (mol epoxide consumed)/(mol metal hour). eDetermined by 1H NMR spectroscopy. fDetermined by gel permeation chromatography in THF calibrated with polystyrene standards. gDetermined by 13C NMR spectroscopy. hGlass transition temperature determined by differential scanning calorimetry. iProductivity (g of polymer/g of catalyst). jMolar ratio of PO/CHO (or CPO) = 1/100. kMolar ratio of CHO/PO = 1/1. lReaction time of 0.1 h. m(S)-Epichlorohydrin.

high as 48800 h−1 for COS/PGE copolymerization at 80 °C, providing a fully alternating copolymer with Mn of 123.0 kg/mol (entry 14).31 For styrene oxide (SO), which contains an electron-withdrawing phenyl group, the binary b3/PPNCl system exhibited high activity and selectivity at 20 °C. The resultant poly(styrene monothiocarbonate) had a Mn of 77.2 kg/mol and PDI of 1.18 (entry 15).19 Meanwhile, the polymer selectivity was greater than 99%, and the O/S ER was totally suppressed. Other catalysts, i.e., catalyst a and tetramethyltetraazaannulene chromium chloride [(tmtaa)CrCl] (c1 anf c2 in Scheme 3),37−39 were used for COS/SO copolymerization (entries 18 and 19). Catalysts c1 and c2 were less active than b3 with the copolymer selectivity being 84 and 93% for c1 and c2, respectively. Similar to catalyst b3, O/ S ER did not occur. Heterogeneous catalyst a was also used to promote this reaction (entry 17). In the a-catalyzed COS/SO copolymerization process, a polymer selectivity of >99% with no O/S scrambling was observed.19 In comparison, the copolymerization of SO and CO2 via the catalyst of a provided a nearly completely alternating poly(styrene carbonate) with high copolymer selectivity.33 In addition, the bifunctional catalyst d1 was reported to successfully catalyze the alternating copolymerization of COS with the monosubstituted epoxides 5−7 (entries 20−22, Table 1) with a high TOF and resulting in copolymers of high molecular weights.

Regioselectivity

When COS is copolymerized with epoxides possessing an electron-donating group such as PO catalyzed by the binary b1− b4/PPNCl catalysts, the anionic sulfur of the growing chain predominantly attacked the less sterically crowded methylene carbon of the epoxide (Scheme 5). That is, the regioselectivity is Scheme 5. Regioselectivity of the Copolymerization of COS and an Epoxide with an Electron-Donating Group

driven by the steric hindrance around the methine carbon center of the epoxide, resulting in a well-defined poly(monothiocarbonate) with 99% tail-to-head linkages.15,18 However, when COS copolymerizes with SO, the regioselectivity was simultaneously driven by both steric hindrance and electron-withdrawing effects of the phenyl group of SO as shown in Scheme 6. These two reaction pathways afford two isomeric linkages 1 and 2, respectively, which can be differentiated by 1H NMR spectroscopy. When the b3/PPNCl system was used 2214

DOI: 10.1021/acs.accounts.6b00345 Acc. Chem. Res. 2016, 49, 2209−2219

Article

Accounts of Chemical Research Scheme 6. Copolymerization of COS with Styrene Oxide (Pn is Growing Polymer Chain)

Figure 4. 1H NMR spectra of the COS/SO copolymer resulting from catalysts a, b3, and c1.

(entry 15), the percentages of linkages 1 and 2 were 88 and 12% (T-H content was 78% based on 13C NMR, Figure 4), respectively, indicating that steric hindrance was the dominant effect on the ring-opening process of SO.19 Similarly, b4, a (salen)CrCl complex with the salen ligand having the opposite chirality diamine backbone as b3, was employed for COS/SO copolymerization (entry 15), providing a copolymer with the same regioselectivity as b3. Moreover, when the less bulky (tmtaa)CrCl catalysts (c1 and c2) were employed (entries 18 and 19), 24% of linkage 2 was produced (Figure 4), indicating an enhancement of the anionic sulfur attacking the sterically hindered methine center. The T-H

contents of 50−52% indicates regio-irregular copolymers. In contrast, a possessed the least steric hindrance among all catalysts employed because the central metal zinc exhibits little steric hindrance (Scheme 3). When a was employed for COS/SO copolymerization (entry 17), 92% of linkage 2 resulted, indicating that the anionic sulfur attacked primarily at the methine carbon (Figure 4). Clearly, the regioselectivity predominantly resulted from the electronic effect of the phenyl group of SO and the H-T content accounted for 84% in the polymer chain.19 Of note, a exhibited the highest regioselectivity of all studied catalysts (entries 15−19). Moreover, it is the only 2215

DOI: 10.1021/acs.accounts.6b00345 Acc. Chem. Res. 2016, 49, 2209−2219

Article

Accounts of Chemical Research

Scheme 7. Mechanism of Fixation of COS by the CA Compound [L3ZnOH]+, with L as Histidine (CA Center Structure), or Ligand (Generally, Amine, CA Models)42

Scheme 8. Proposed O/S ER Mechanism for COS/Epoxide and CS2/Oxetane Copolymerization for (A) Metal−OH Catalysis15and (B) Rearrangement of Tetrahedral Intermediate (Take Cr3+ as an Example)30

■

MECHANISTIC ASPECTS OF THE COS/EPOXIDE COPOLYMERIZATION PROCESS A special mechanistic consideration of the COS/epoxide metalcatalyzed coupling reaction is the possible occurrence of oxygen/ sulfur atom exchange, which leads to a decrease in the number of monothiocarbonate units in the copolymer backbone. The metal catalysts depicted in Scheme 3 can afford metal−OH intermediates during the copolymerization reaction resulting from hydrolysis or chain transfer processes. These metal−OH species can mimic reactions as seen in the ubiquitous human carbonic anhydrase II (HCAII) enzyme. A pathway as illustrated in Scheme 7 can convert COS into CO2 quantitatively.42 In general, it has been found that low reaction temperatures and well-dried polymerization systems can depress the occurrence of O/S ER in these processes. For the copolymerization of CS2/ epoxides or CS2/oxetanes, O/S ER are much more difficult to control. The proposed mechanism of CS2/epoxide copolymerization undergoes a Cr−OH intermediate (Scheme 8A), whereas the process of CS2/oxetane copolymerization likely involves a

reported catalyst that provides a copolymer with predominant HT linkage.15−19 In 1975, Inoue et al. reported that ring-opening of SO took place predominantly (96%) at the methine of SO during its copolymerization with CO2 catalyzed by the diethyl zinc/water system.40 On the contrary, Lu and co-workers reported that (salen)Co complex-catalyzed CO2/(S)-SO (98% ee) copolymerization gave only 11% occurrence of ring-cleavage at the methine.41 Considering all of these results involving SO copolymerization with CO2, we can conclude that when less bulky catalysts are used the regioselectivity is predominantly affected by the strong electron-withdrawing ability of the phenyl group of SO. Thus, the propagating species attacks the methine carbon center, predominantly resulting in a high percentage of H-T linkage. On the other hand, when catalysts containing bulky ligands are used, the steric hindrance environment around the metal center will lead to a ring-opening site at the less hindered methylene carbon of SO, and thus, a high percentage of H-T linkage will be produced. 2216

DOI: 10.1021/acs.accounts.6b00345 Acc. Chem. Res. 2016, 49, 2209−2219

Article

Accounts of Chemical Research tetrahedral intermediate containing two sulfur atoms (Scheme 8B).43

■

PROPERTIES OF COS/EPOXIDE COPOLYMER The glass transition temperatures (Tgs) of the reported poly(monothiocarbonate)s vary from 2.9 to 114.6 °C (Figure 5). The substituents on the epoxides have a strong impact on Tgs

Figure 6. Refractive indices (n) of COS/CHO/PO terpolymers via the wavelength from 400 to 800 nm (nd is refractive index at wavelength 587.6 nm).

thiocarboxylate species attacks the chloromethylene site of ECH, resulting in chain termination at 25 °C. This in turn sharply lowers the molecular weights and results in amorphous copolymers. Recently, we synthesized a semicrystalline poly(trimethylene monothiocarbonate) (PTMMTC) from a perfectly alternating COS/oxetane copolymerization via the catalysis of b1−b4/ PPNCl system (Scheme 9). Fortunately, O/S ER did not occur,

Figure 5. Tgs of the poly(monothiocarbonate)s from COS/epoxide copolymerization.

of the COS/epoxide copolymers, which makes these copolymers potentially useful for a wide temperature range and for various purposes. The lowest reported Tg (2.9 °C) is for the copolymer derived from COS and 1,2-epoxyhexane resulting from the long butane group.31 The COS/CHO copolymer presented the highest Tg of 114.6 °C due to its rigid alicyclic structure and high initial decomposition temperature of 244 °C.17 By terpolymerization, such as COS/PO/CHO terpolymerization, Tgs of the terpolymers could be linearly tuned from 22.4 to 114.6 °C by varying the feeding ratio of CHO and PO. Poly(monothiocarboante)s from COS/epoxide copolymerization have excellent optical properties. These copolymers are highly transparent and colorless solids. The refractive index (nd) of the COS/PO15 and COS/CHO copolymers17 were 1.63 (1.524) and 1.70 (1.548), respectively, by Abbe’s refractometer (spectroscopic ellipsometer), and higher than some typical commercial optical plastics, such as poly(methyl methacrylate) (PMMA) (1.49). Of importance, the Abbe numbers (Vds) of COS/PO/CHO terpolymers could be tuned from 32.1 to 43.1 by terpolymerization (Figure 6). Because a small Vd causes stronger chromatic dispersion, the Vd of a visual optical material should be around 30 to 60; COS-based copolymers will be good substitutes for traditional polymers, such as PS and PC, which have Vds of only 31 and 30, respectively. Because poly(monothiocarboante)s are soluble in common organic solvents and possess balanced thermal/mechanical properties and tailored optical properties, they are potential optical materials.

Scheme 9. Syntheses of PTMC from Oxetane with CO245 and PTMMTC from Oxetane with COS via the (Salen)Cr Complexes44

and only small amounts of the cyclic thiocarbonate were produced over a wide range of reaction temperatures from 40 to 130 °C. Of importance, PTMMTC afforded crystallization behaviors similar to that of polyethylene. It crystallizes rapidly, affording typical spherulite in bulk, exhibiting a Tm value of 127.5 °C.44 In contrast, poly(trimethylene carbonate) (PTMC) from CO2/oxetane copolymerization using the similar catalysts was totally amorphous.45 It is of interest that PTMMTC, which has one oxygen atom in the carbonate unit replaced by a sulfur atom, can be crystallized. Specifically, PTMMTC is a biodegradable crystalline polymer synthesized in the absence of a chiral catalyst and monomers. Clearly, this process is also different from the regioselective copolymerization of COS with chiral epoxides32 and enantioselective copolymerization of CO2 with epoxides by using chiral catalysts.46 Because COS is a readily accessible and potentially low-cost C1 building block, this preliminary study should widen the scope of biodegradable materials.

■

CRYSTALLINE COS/EPOXIDE COPOLYMER The synthesis of biodegradable crystalline C1-based polymers is a significant challenge. A semicrystalline poly(thiocarbonate) from the regioselective copolymerization of enantiopure ECH and COS by catalyst d1 (Scheme 3) has been reported.32 The resulting copolymer had a Tm of 96.7 °C and a ΔHm of 49.8 J/g. However, a rather low reaction temperature (−25 °C) had to be employed to obtain crystalline COS/ECH copolymer (Mn: 3130 g/mol). This is necessary because the propagating mono-

■

CONCLUDING REMARKS The production of sulfur-containing polymers from the copolymerization of COS and epoxides represents an atomeconomical and significantly “greener” route to these thermoplastics. Our efforts on the COS/epoxide copolymerization disclosed the specialty of COS polymerization, which include the unit selectivity caused by O/S ER and regioselectivity induced by 2217

DOI: 10.1021/acs.accounts.6b00345 Acc. Chem. Res. 2016, 49, 2209−2219

Article

Accounts of Chemical Research the asymmetrical structure of COS. Of importance, COS can copolymerize with numerous epoxides, whether they contain electron-donating or -withdrawing groups, displaying high catalytic activities and regioselectivity in a well-controlled process carried out under mild conditions. Moreover, the structure− property relationship of these polythiocarbonates could be tailored over a wide range, thereby broadening the applications possible for these polymers. The resulting poly(monothiocarbonate)s are often highly transparent with high refractive index and Abbe number. Their excellent optical property is made more attractive and applicable due to their solubility properties in many common organic solvents. With regard to future directions for synthesizing COS-based copolymers, we hope to broaden the scope of catalysts capable of catalyzing COS with epoxides, affording various poly(monothiocarbonate)s as well as their block copolymers and crystalline polymers. Moreover, epoxides derived from biomass are expected to be investigated, thereby making useful polymers via a nonpetroleum route. With regard to better understanding mechanistic aspects of the process, further detailed studies of O/ S exchange reactions are needed to develop synthetic methods to enhance or depress these processes. These mechanistic studies of COS/epoxide copolymerization should also be beneficial to advance our understanding of CO2/epoxide copolymerization. We also hope to realize the controlled copolymerization of CS2 with epoxides, affording well-defined sulfur-rich polythiocarbonates. Ultimately, these initial research findings on the COS copolymerization reaction summarized herein will provide guidance for synthesizing new well-defined sulfur-containing polymers from CS2 and epoxides. Further studies will hopefully lead to an enhanced utility of this copolymerization process, including the production of sulfur-containing polymers with functional groups.

■

Following a nine month period at the Texaco Research Center in Beacon, NY, he was on the faculties of the State University of New York at Buffalo from 1969 to 1972 and Tulane University from 1973 to 1982. He has been at Texas A&M University since 1982, where he currently is a Distinguished Professor. Among his current interests are the utilization of C1 monomers in copolymerization reactions and the ring-opening polymerization of renewable monomers such as lactides.

■

ACKNOWLEDGMENTS M.L. and X.-H.Z. gratefully acknowledge the financial support of the Distinguished Young Investigator Fund of Zhejiang Province (LR16B040001) and the National Science Foundation of the People’s Republic of China (21274123). We are most grateful to the graduate students, Jialiang Yang, Hailin Wu, Yang Li, and Yingying Zhang for their ongoing work on COS copolymerization. D.J.D. is thankful to the Robert A. Welch Foundation (A0923) for financial support.

■

REFERENCES

(1) Ferm, R. J. The Chemistry of Carbonyl Sulfide. Chem. Rev. 1957, 57, 621−640. (2) Rasmussen, R. A.; Khalil, M. A. K.; Dalluge, R. W.; Penkett, S. A.; Jones, B. Carbonyl Sulfide and Carbon Disulfide from the Eruptions of Mount St. Helens. Science 1982, 215, 665−667. (3) Svoronos, P. D. N.; Bruno, T. J. Carbonyl Sulfide: A Review of Its Chemistry and Properties. Ind. Eng. Chem. Res. 2002, 41, 5321−5336. (4) Notni, J.; Schenk, S.; Protoschill-Krebs, G.; Kesselmeier, J.; Anders, E. The Missing Link in COS Metabolism: A Model Study on the Reactivation of Carbonic Anhydrase from Its Hydrosulfide Analogue. ChemBioChem 2007, 8, 530−536. (5) Zhao, S.; Yi, H.; Tang, X.; Jiang, S.; Gao, F.; Zhang, B.; Zuo, Y.; Wang, Z. The Hydrolysis of Carbonyl Sulfide at Low Temperature: A Review. Sci. World J. 2013, 2013, 739501−739501. (6) Inoue, S.; Koinuma, H.; Tsuruta, T. Copolymerization of Carbon Dioxide and Epoxide. J. Polym. Sci., Part B: Polym. Lett. 1969, 7, 287− 292. (7) Coates, G. W.; Moore, D. R. Discrete Metal-Based Catalysts for the Copolymerization of CO2 and Epoxides: Discovery, Reactivity, Optimization, and Mechanism. Angew. Chem., Int. Ed. 2004, 43, 6618−6639. (8) Darensbourg, D. J.; Mackiewicz, R. M.; Phelps, A. L.; Billodeaux, D. R. Copolymerization of CO2 and Epoxides Catalyzed by Metal Salen Complexes. Acc. Chem. Res. 2004, 37, 836−844. (9) Darensbourg, D. J. Making Plastics from Carbon Dioxide: Salen Metal Complexes as Catalysts for the Production of Polycarbonates from Epoxides and CO2. Chem. Rev. 2007, 107, 2388−2410. (10) Klaus, S.; Lehenmeier, M. W.; Anderson, C. E.; Rieger, B. Recent Advances in CO2/Epoxide CopolymerizationNew Strategies and Cooperative Mechanisms. Coord. Chem. Rev. 2011, 255, 1460−1479. (11) Kember, M. R.; Buchard, A.; Williams, C. K. Catalysts for CO2/ Epoxide Copolymerisation. Chem. Commun. 2011, 47, 141−163. (12) Lu, X. B.; Ren, W. M.; Wu, G. P. CO2 Copolymers from Epoxides: Catalyst Activity, Product Selectivity, and Stereochemistry Control. Acc. Chem. Res. 2012, 45, 1721−1735. (13) Nakano, R.; Ito, S.; Nozaki, K. Copolymerization of Carbon Dioxide and Butadiene Via a Lactone Intermediate. Nat. Chem. 2014, 6, 325−331. (14) Luo, M.; Li, Y.; Zhang, Y. Y.; Zhang, X. H. Using Carbon Dioxide and Its Sulfur Analogues as Monomers in Polymer Synthesis. Polymer 2016, 82, 406−431. (15) Luo, M.; Zhang, X. H.; Du, B. Y.; Wang, Q.; Fan, Z. Q. Regioselective and Alternating Copolymerization of Carbonyl Sulfide with Racemic Propylene Oxide. Macromolecules 2013, 46, 5899−5904. (16) Luo, M.; Zhang, X. H.; Du, B. Y.; Wang, Q.; Fan, Z. Q. Alternating Copolymerization of Carbonyl Sulfide and Cyclohexene Oxide Catalyzed by Zinc−Cobalt Double Metal Cyanide Complex. Polymer 2014, 55, 3688−3695.

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Notes

The authors declare no competing financial interest. Biographies Ming Luo was born in 1987 in China. He received his B.S. degree in Polymer Science and Engineering from Zhejiang University (ZJU) in 2011 and received his Ph.D. degree in Polymer Chemistry and Physics from ZJU in 2016. He worked as a visiting scholar at Texas A&M University guided by Dr. Donald J. Darensbourg from September 2014 to September 2015. His research interest is the synthesis of COS(CS2)based polymers. Xinghong Zhang was born in 1977 in Lu’an, China. He received his B.S. degree from Fuyang Teachers College in 2000, M. Eng. degree at Shantou University in 2003, and Ph.D. degree at ZJU in 2006. He became an Assistant Professor in 2006 and was promoted to Associate Professor in 2009. In 2012-2013, he worked as a visiting scholar at the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana & Champaign under the supervision of Dr. Jeffrey S. Moore. His current research interests include the catalysis of C1-involved copolymerization and polymer mechanochemistry. Donald D. Darensbourg was born in Baton Rouge, LA in 1941 and received his B.S. and Ph.D. degrees from California State University at Los Angeles and the University of Illinois at Urbana, respectively. 2218

DOI: 10.1021/acs.accounts.6b00345 Acc. Chem. Res. 2016, 49, 2209−2219

Article

Accounts of Chemical Research

Vs Cyclic Carbonate Production. J. Am. Chem. Soc. 2003, 125, 7586− 7591. (36) Sun, X. K.; Zhang, X. H.; Wei, R. J.; Du, B. Y.; Wang, Q.; Fan, Z. Q.; Qi, G. R. Mechanistic Insight into Initiation and Chain Transfer Reaction of CO2/Cyclohexene Oxide Copolymerization Catalyzed by Zinc-Cobalt Double Metal Cyanide Complex Catalysts. J. Polym. Sci., Part A: Polym. Chem. 2012, 50, 2924−2934. (37) Mountford, P. Dibenzotetraaza[14]Annulenes: Versatile Ligands for Transition and Main Group Metal Chemistry. Chem. Soc. Rev. 1998, 27, 105−116. (38) Darensbourg, D. J.; Fitch, S. B. An Exploration of the Coupling Reactions of Epoxides and Carbon Dioxide Catalyzed by Tetramethyltetraazaannulene Chromium(III) Derivatives: Formation of Copolymers Versus Cyclic Carbonates. Inorg. Chem. 2008, 47, 11868− 11878. (39) Darensbourg, D. J.; Fitch, S. B. Copolymerization of Epoxides and Carbon Dioxide. Evidence Supporting the Lack of Dual Catalysis at a Single Metal Site. Inorg. Chem. 2009, 48, 8668−8677. (40) Hirano, T.; Inoue, S.; Tsuruta, T. Stereochemistry of the Copolymerization of Carbon Dioxide with Optically Active Phenylepoxyethane. Makromol. Chem. 1975, 176, 1913−1917. (41) Wu, G. P.; Wei, S. H.; Lu, X. B.; Ren, W. M.; Darensbourg, D. J. Highly Selective Synthesis of CO2 Copolymer from Styrene Oxide. Macromolecules 2010, 43, 9202−9204. (42) Schenk, S.; Notni, J.; Koehn, U.; Wermann, K.; Anders, E. Carbon Dioxide and Related Heterocumulenes at Zinc and Lithium Cations: Bioinspired Reactions and Principles. Dalton Trans. 2006, 38, 4191− 4206. (43) Diebler, J.; Komber, H.; Häußler, L.; Lederer, A.; Werner, T. Alkoxide-Initiated Regioselective Coupling of Carbon Disulfide and Terminal Epoxides for the Synthesis of Strongly Alternating Copolymers. Macromolecules 2016, 49, 4723−4731. (44) We will report the catalytic synthesis of the semicrystalline COS/ oxetane copolymer in the near future. (45) Darensbourg, D. J.; Moncada, A. I.; Choi, W.; Reibenspies, J. H. Mechanistic Studies of the Copolymerization Reaction of Oxetane and Carbon Dioxide to Provide Aliphatic Polycarbonates Catalyzed by (Salen)CrX Complexes. J. Am. Chem. Soc. 2008, 130, 6523−6533. (46) Wu, G. P.; Xu, P. X.; Lu, X. B.; Zu, Y. P.; Wei, S. H.; Ren, W. M.; Darensbourg, D. J. Crystalline CO2 copolymer from Epichlorohydrin Via Co(III)-Complex-Mediated Stereospecific Polymerization. Macromolecules 2013, 46, 2128−2133.

(17) Luo, M.; Zhang, X. H.; Du, B. Y.; Wang, Q.; Fan, Z. Q. WellDefined High Refractive Index Poly(Monothiocarbonate) with Tunable Abbe’s Numbers and Glass-Transition Temperatures Via Terpolymerization. Polym. Chem. 2015, 6, 4978−4983. (18) Luo, M.; Zhang, X. H.; Darensbourg, D. J. Highly Regioselective and Alternating Copolymerization of Carbonyl Sulfide with Phenyl Glycidyl Ether. Polym. Chem. 2015, 6, 6955−6958. (19) Luo, M.; Zhang, X. H.; Darensbourg, D. J. An Examination of the Steric and Electronic Effects in the Copolymerization of Carbonyl Sulfide and Styrene Oxide. Macromolecules 2015, 48, 6057−6062. (20) Kultys, A. Sulfur-Containing Polymers in Encyclopedia of Polymer Science and Technology, 4th ed.; John Wiley & Sons, Inc.: Hoboken, 2010. (21) You, Y. Z.; Hong, C. Y.; Pan, C. Y. A New Strategy for the Synthesis of Polytrithiocarbonates Using a Polymeric Support System. Macromol. Rapid Commun. 2002, 23, 776−780. (22) Nemoto, N.; Sanda, F.; Endo, T. Controlled Cationic RingOpening Polymerization of Monothiocarbonate. Macromolecules 2000, 33, 7229−7231. (23) Adachi, N.; Kida, Y.; Shikata, K. Copolymerization of Propylene Oxide and Carbon Disulfide with Diethylzinc−Electron-Donor Catalyst. J. Polym. Sci., Polym. Chem. Ed. 1977, 15, 937−944. (24) Soga, K.; Sato, M.; Imamura, H.; Ikeda, S. Copolymerization of Carbon Disulfide and Episulfides. J. Polym. Sci., Polym. Lett. Ed. 1975, 13, 167−171. (25) Nakano, K.; Tatsumi, G.; Nozaki, K. Synthesis of Sulfur-Rich Polymers: Copolymerization of Episulfide with Carbon Disulfide by Using [PPN]Cl/(Salph)Cr(III)Cl System. J. Am. Chem. Soc. 2007, 129, 15116−15117. (26) Zhang, X. H.; Liu, F.; Sun, X. K.; Chen, S.; Du, B. Y.; Qi, G. R.; Wan, K. M. Atom-Exchange Coordination Polymerization of Carbon Disulfide and Propylene Oxide by a Highly Effective Double-Metal Cyanide Complex. Macromolecules 2008, 41, 1587−1590. (27) Darensbourg, D. J.; Andreatta, J. R.; Jungman, M. J.; Reibenspies, J. H. Investigations into the Coupling of Cyclohexene Oxide and Carbon Disulfide Catalyzed by (Salen)CrCl. Selectivity for the Production of Copolymers Vs. Cyclic Thiocarbonates. Dalton Trans. 2009, 8891− 8899. (28) Zhang, X. H.; Huang, Y. J.; Liu, F.; Sun, X. K.; Fan, Z. Q.; Qi, G. R. Copolymerization of Carbon Disulfide and Cyclohexene Oxide with a Double-Metal Cyanide Complex Catalyst. Acta. Polym. Sin. 2009, 1, 546−552. (29) Darensbourg, D. J.; Wilson, S. J.; Yeung, A. D. Oxygen/Sulfur Scrambling During the Copolymerization of Cyclopentene Oxide and Carbon Disulfide: Selectivity for Copolymer Vs Cyclic [Thio]Carbonates. Macromolecules 2013, 46, 8102−8110. (30) Luo, M.; Zhang, X. H.; Darensbourg, D. J. An Investigation of the Pathways for Oxygen/Sulfur Scramblings During the Copolymerization of Carbon Disulfide and Oxetane. Macromolecules 2015, 48, 5526−5532. (31) Ren, W. M.; Liu, Y.; Xin, A. X.; Fu, S.; Lu, X. B. Single-Site Bifunctional Catalysts for Cox (X= O or S)/Epoxides Copolymerization: Combining High Activity, Selectivity, and Durability. Macromolecules 2015, 48, 8445−8450. (32) Yue, T. J.; Ren, W. M.; Liu, Y.; Wan, Z. Q.; Lu, X. B. Crystalline Polythiocarbonate from Stereoregular Copolymerization of Carbonyl Sulfide and Epichlorohydrin. Macromolecules 2016, 49, 2971−2976. (33) Wei, R. J.; Zhang, X. H.; Du, B. Y.; Sun, X. K.; Fan, Z. Q.; Qi, G. R. Highly Regioselective and Alternating Copolymerization of Racemic Styrene Oxide and Carbon Dioxide Via Heterogeneous Double Metal Cyanide Complex Catalyst. Macromolecules 2013, 46, 3693−3697. (34) Zhang, X. H.; Wei, R. J.; Zhang, Y. Y.; Du, B. Y.; Fan, Z. Q. Carbon Dioxide/Epoxide Copolymerization Via a Nanosized Zinc−Cobalt(III) Double Metal Cyanide Complex: Substituent Effects of Epoxides on Polycarbonate Selectivity, Regioselectivity and Glass Transition Temperatures. Macromolecules 2015, 48, 536−544. (35) Darensbourg, D. J.; Yarbrough, J. C.; Ortiz, C.; Fang, C. C. Comparative Kinetic Studies of the Copolymerization of Cyclohexene Oxide and Propylene Oxide with Carbon Dioxide in the Presence of Chromium Salen Derivatives. In Situ Ftir Measurements of Copolymer 2219

DOI: 10.1021/acs.accounts.6b00345 Acc. Chem. Res. 2016, 49, 2209−2219