α,ω-Epoxide, Oxetane, and Dithiocarbonate Telechelic Copolyolefins: Access by Ring-Opening Metathesis/Cross-Metathesis Polymerization (ROMP/CM) of Cycloolefins in the Presence of Functional Symmetric Chain-Transfer Agents

Epoxide- and oxetane-α,ω-telechelic (co)polyolefins have been successfully synthesized by the tandem ring-opening metathesis polymerization (ROMP)/cross-metathesis (CM) of cyclic olefins using Grubbs’ second-generation catalyst (G2) in the presence of a bifunctional symmetric alkene epoxide- or oxetane-functionalized chain-transfer agent (CTA). From cyclooctene (COE), trans,trans,cis-1,5,9-cyclododecatriene (CDT), norbornene (NB), and methyl 5-norbornene-2-carboxylate (NBCOOMe), with bis(oxiran-2-ylmethyl) maleate (CTA 1), bis(oxetane-2-ylmethyl) maleate (CTA 2), or bis(oxetane-2-ylmethyl) (E)-hex-3-enedioate (CTA 3), well-defined α,ω-di(epoxide or oxetane) telechelic PCOEs, P(COE-co-NB or -NBCOOMe)s, and P(NB-co-CDT)s were isolated under mild operating conditions (40 or 60 °C, 24 h). The oxetane CTA 3 and the epoxide CTA 1 were revealed to be significantly more efficient in the CM step than CTA 2, which apparently inhibits the reaction. Quantitative dithiocarbonatation (CS2/LiBr, 40 °C, THF) of an α,ω-di(epoxide) telechelic P(NB-co-CDT) afforded a convenient approach to the analogous α,ω-bis(dithiocarbonate) telechelic P(NB-co-CDT). The nature of the end-capping function of the epoxide/oxetane/dithiocarbonate telechelic P(NB-co-CDT)s did not impact their thermal signature, as measured by DSC. These copolymers also displayed a low viscosity liquid-like behavior and a shear thinning rheological behavior.


Introduction
Telechelic polymers, i.e., polymers with functional end-groups, attract much interest due to their unique properties. They are commonly used as precursors to block copolymers, as cross-linking agents, or as intermediates in the formation of polymeric networks [1][2][3][4]. Polyolefins (POs) are the most commercially produced polymers, featuring great resistance to harsh chemical environments and broad-ranging mechanical and thermal characteristics, resulting in a wide variety of applications, such as in packaging, including food wrapping, electrical wires protection or piping systems, fabrics, Chemical shifts (δ) are reported in ppm and were referenced internally relative to tetramethylsilane (δ 0 ppm) using the residual 1 H and 13 C solvent resonances of the deuterated solvent.
The molar mass values of PCOE samples were determined by a 1 H NMR analysis in CDCl 3 (M n,NMR ) from the integral value ratio of the signals of end-groups' hydrogens (typically δ ca. 3.07 ppm (H c )) to internal olefin hydrogens (δ ca. 5.35 (H 1 )) ( Figure 1). For copolymers, the molar mass values were determined by a 1  The average molar mass (M n,SEC ) and dispersity (Ð M = M w /M n ) values of the freshly prepared polymer samples (at most within one week unless otherwise stated) were determined by size exclusion chromatography (SEC) in THF at 30 • C (flow rate = 1.0 mL·min −1 ) on a Polymer Laboratories PL50 apparatus that was equipped with a refractive index detector and a set of two ResiPore PLgel 3 µm MIXED-E 300 × 7.5 mm columns. The polymer samples were dissolved in THF (2 mg·mL −1 ). All elution curves were calibrated with 12 monodisperse polystyrene standards (M n range = 580-380,000 g·mol −1 ). M n,SEC values of polymers were uncorrected for their possible difference in hydrodynamic volume in THF versus polystyrene. The SEC traces of the polymers all exhibited a monomodal and symmetric peak.
ASAP (Atmospheric Solid Analysis Probe) high-resolution mass spectrometry (HRMS) data were recorded at the CRMPO-Scanmat (Rennes, France) with a Bruker MicrOTOF-Q II mass spectrometer that was equipped with an APCI (Atmospheric Pressure Chemical Ionisation) source in positive mode by direct introduction at 370 • C.
MALDI-ToF mass spectra of polymers were recorded at the CESAMO (Bordeaux, France) on a Voyager mass spectrometer (Applied Biosystems) that was equipped with a pulsed N 2 laser source (337 nm, 4 ns pulse width) and a time-delayed extracted ion source. Spectra were recorded in the positive-ion mode using the reflectron mode and with an accelerating voltage of 20 kV. A freshly prepared solution of the polymer sample in THF (HPLC grade, 10 mg·mL −1 ) and a saturated solution of trans-2-[3-(4-tert-butylphenyl)-2-methyl-2-propenylidene]-malononitrile (10 mg, DCTB) in THF (1 mL, HPLC grade) were prepared. A MeOH solution of the cationizing agent (NaI, 10 mg·mL −1 ; Na + ions interact with the polymer's heteroatoms enabling the observation of DF PCOE) was also prepared. The solutions were combined in a 10:1:1 v/v/v of matrix-to-sample-to-cationizing agent. The resulting solution (1-2 µL) was deposited onto the sample target and vacuum-dried.
FTIR spectra of the polymers were acquired (16 scans) with a resolution of 4 cm −1 on a IRAffinity-1 (Shimadzu, Champs sur Marne, France) that was equipped with an ATR module.
Apparent viscosity was measured with an ARES G2 viscosimeter that was equipped with a plate-plate geometry, at a speed gradient of 0.01 s −1 over a shear rate range from 0.01 to 100·s −1 . At each imposed shear rate, the apparent viscosity was determined in the steady state regime. The temperature was fixed at 23 ± 0.3 • C. For each sample, the viscosimetric test duration was 5 min.

General Procedure for the Polymerization of COE
All polymerizations were performed according to the following typical procedure ( Table 1, entry 1). The only differences lie in the nature of the solvent, the catalyst loading, the CTA, and the CTA's initial concentration ([CTA] 0 ). Under an argon atmosphere, a Schlenk flask that was equipped with a magnetic stir bar was charged sequentially with dry CH 2 Cl 2 (5.0 mL), COE (1.53 mL, 1.29 g, 11.7 mmol), and CTA 2 (66.5 mg, 0.234 mmol). The initial concentration of COE was kept at 1.8 mol·L −1 . The resulting solution was heated at 40 • C, and the polymerization was started upon addition, via a cannula, of a freshly prepared CH 2 Cl 2 solution (2.0 mL) of G2 (9.9 mg, 11.7 µmol). The reaction mixture turned highly viscous within 2 min. The viscosity then slowly decreased over the following 10 min. After the desired reaction time (typically 24 h; reaction time was not necessarily optimized), volatiles were removed under vacuum. The polymer was recovered upon precipitation in methanol (50 mL) (thereby allowing for removal of the catalyst), filtration, and drying at 25 • C under vacuum. All polymers were recovered as white powders, which are readily soluble in CHCl 3 and THF, and insoluble in MeOH (Table 1). All experiments were at least duplicated. The isolated polymers were characterized by 1 H NMR spectroscopy, MALDI-ToF MS, and DSC analyses (Figures 1 and 2, Table 5).

General Procedure for the Copolymerization of Cyclic Olefins
All copolymerizations were performed according to the following typical procedure that is described thereafter for NB and CDT. Typically, under an argon atmosphere, a Schlenk flask that was equipped with a magnetic stir bar was charged sequentially with 1,2-dichloroethane (10.0 mL), NB (0.93 g, 9.87 mmol), CDT (1.80 mL, 1.60 g, 9.87 mmol), and CTA 3 (168 mg, 0.49 mmol) ( Table 3, entry 8). The resulting solution was heated at 60 • C, and the polymerization was started upon addition, via a cannula, of a dry, freshly prepared CH 2 Cl 2 solution (4.0 mL) of G2 (5.9 mg, 7.01 µmol). Note that reactions performed in CH 2 Cl 2 or THF were run at 40 or 60 • C, respectively. The reaction mixture turned highly viscous within 2 min. The viscosity then visually slowly decreased over the following 10 min. After the desired reaction time (typically 24 h; reaction time was not necessarily optimized), volatiles were removed under vacuum. The isolated copolymers were characterized by 1 H, 13 C{ 1 H} NMR spectroscopy, DSC, and viscosimetric analyses (Tables 2, 3, 5, and 6; Figures 3, 4 and S6-S11).

Quantification of Cyclic Nonfunctional (CNF) Copolymers within the Recovered Chemically Modified P(NB-co-CDT) Crude Copolymers
CNF copolyolefins were separated from the crude α,ω-bis(dithiocarbonate) P(NB-co-CDT) copolymers by eluting a mixture that was issued from the crude recovered copolymers with a multifunctional amine (Lupasol FG ® ) (1.5 equiv NH 2 versus dithiocarboante) in CH 2 Cl 2 (50 mL), through a silica column chromatography, using pentane as eluent. The recovered CNF polymers were then weighted and characterized by 1 H NMR spectroscopy ( Figure S13).

Metathesis of COE in the Presence of CTAs 1-3
The use of the symmetric epoxide CTA 1 for the preparation of the corresponding epoxide-telechelic PCOE homopolymer by ROMP/CM [37,38] of cyclooctene (COE) using G2 was already successfully established [39]. The efficiency of the ROMP/CM of COE using G2 in the presence of the new oxetane symmetric CTAs 2-3 to provide the corresponding oxetane-telechelic PCOEs was, thus, first evaluated, prior to the synthesis of epoxide-and oxetane-end-capped coPOs (vide infra).
The tandem ROMP/CM of COE catalyzed by G2 in the presence of oxetane CTAs 2 and 3 was investigated in CH 2 Cl 2 at 40 • C over 24 h (Table 1). Although the COE was always fully consumed, the conversion of CTA 2 remained sluggish and incomplete (≤20%), even decreasing with larger amounts of CTA (Table 1, entries 1-3). On the other hand, the metathesis reaction using CTA 3 showed the complete conversion of both COE and CTA; in this case, the molar mass values as determined by a 1 H NMR analysis (M n,NMR ) fairly matched the data that were calculated (M n,theo ) on the basis of the formation of only DF PCOEs (M n,theo ) (i.e., without taking into account any CNF nor any nonfunctional PCOE; refer to the Experimental Section) ( Table 1, entries 4, 5). The observed difference between the molar mass values as measured from the SEC analysis (M n,SEC ) and M n,NMR or M n,theo most likely reflects the difference in the hydrodynamic radius between the PCOEs and the polystyrene standards that were used for the calibration of the SEC apparatus, as previously observed [34][35][36][37][38][39][40][41][42]. Nevertheless, the M n,SEC values proportionaly increased with larger amounts of monomer consumed. The additional methylene group in between the C=C double bond and the ester group in CTA 3 versus CTA 2 thus significantly improved the efficiency of the tandem metathesis reaction, as similarly reported with alike epoxyde or azlactone symmetric CTAs [39,42]. The PCOE thus formed displayed the expected oxetane chain end-capping groups as evidenced by NMR spectroscopy and mass spectrometry analyses. The 1 H NMR spectrum showed the characteristic signals of COE repeating units (H 1 -H 3 ) along with the typical CH 2 O methylene groups (δ (ppm) 4.44 and 4.23; Figure 1) of the oxetane moiety. Correspondingly, the MALDI-ToF mass spectrum (MS) clearly evidenced the α,ω-dioxetane PCOE as the major population observed with a repeating unit of 110 g·mol −1 , with e.g., m/z experimental = 1465.1 g·mol −1 versus m/z simulated = 1465.3 g·mol −1 for n = 10 ( Figure 2). The oxetane-based CTAs were, thus, revealed to be stable in the presence of G2 and suitably enabled the preparation of α,ω-dioxetane PCOE, thus foreseeing the successful preparation of coPOs, more specifically using CTA 3.

Ring-Opening Metathesis Copolymerization of Cyclic Olefins in the Presence of CTAs 1 or 3
The two most efficient CTAs, namely 1 and 3, were selected for the copolymerization studies. Commercially available monomers where herein considered towards the synthesis of telechelic copolymers, namely COE, NB, and the related NB featuring a methyl ester substituent, NB COOMe . The molar ratio of comonomers was fixed at 50:50, based on previous investigations that revealed such resulting copolymers with equal composition in each comonomer to display a lower viscosity, thereby being more attractive towards the elaboration of low viscosity liquid polymer materials that are aimed at adhesive or coating applications [38−42].

Ring-Opening Metathesis Copolymerization of Cyclic Olefins in the Presence of CTAs 1 or 3
The two most efficient CTAs, namely 1 and 3, were selected for the copolymerization studies. Commercially available monomers where herein considered towards the synthesis of telechelic copolymers, namely COE, NB, and the related NB featuring a methyl ester substituent, NB COOMe . The molar ratio of comonomers was fixed at 50:50, based on previous investigations that revealed such resulting copolymers with equal composition in each comonomer to display a lower viscosity, thereby being more attractive towards the elaboration of low viscosity liquid polymer materials that are aimed at adhesive or coating applications [38−42].  The two most efficient CTAs, namely 1 and 3, were selected for the copolymerization studies. Commercially available monomers where herein considered towards the synthesis of telechelic copolymers, namely COE, NB, and the related NB featuring a methyl ester substituent, NB COOMe . The molar ratio of comonomers was fixed at 50:50, based on previous investigations that revealed such resulting copolymers with equal composition in each comonomer to display a lower viscosity, thereby being more attractive towards the elaboration of low viscosity liquid polymer materials that are aimed at adhesive or coating applications [38][39][40][41][42].

α,ω-Diepoxide Telechelic P(COE-co-NB) and P(COE-co-NB COOMe ) Copolymers
The ROMP of COE and NB or NB COOMe with G2 in the presence of CTA 1 (the most easily accessible CTA (one-step; high yield)) was similarly performed at 40 • C for 24 h in CH 2 Cl 2 (Scheme 1). An average conversion in CTA 1 of 75% was reached at lower COE/NB loadings (typically 500 mol. equiv. each versus G2), while increasing the initial content in both comonomers resulted in a lower CM efficiency (Table 2, entries 1-4). In comparison, the homopolymerization of COE under similar operating conditions was as effective [39]. On the other hand, the copolymerization of NB with CDT under similar operating conditions was revealed to be ineffective (vide infra, Table 3, entries 1-2). This suggested the beneficial presence of COE for its successful copolymerization with NB. Similarly, the COE/NB COOMe copolymerization proceeded fairly with a ca. 45% conversion of CTA 1 with a loading of comonomers as high as 1000 mol. equiv. (Table 2, entries 5-6). The experimental molar mass values determined by NMR matched well the calculated values, while the dispersity remained within the common range for coPOs (Ð M <1.5), thus highlighting a fairly controlled polymerization. The NMR characterizations of the copolymers demonstrated the successful formation of the corresponding random copolymers end-capped at both termini by the epoxide moiety, in agreement with data in the literature on homoPCOEs [39], as illustrated in Figures S6-S7.

α,ω-Diepoxide and -Dioxetane Telechelic P(NB-co-CDT) Copolymers
The one-pot simultaneous copolymerization of NB and CDT catalyzed by G2 or the Hoveyda-Grubbs catalyst HG2 in the presence of CTA 1 or 3 was performed at 40 or 60 • C for 24 h in CH 2 Cl 2 or (CH 2 Cl) 2 (Table 3, Scheme 1).  As aforementioned, attempts to copolymerize NB and CDT (50:50 mol/mol) in the presence of the catalytic system G2/CTA 1 in CH 2 Cl 2 at 40 • C failed, as evidenced by a 1 H NMR analysis that showed the presence of unreacted CTA 1 (Table 3, entries 1-2, Figure S8). Replacing G2 by the Hoveyda-Grubbs catalyst HG2 enabled us to reach a good CM efficiency (up to ca. 80% CTA 1 consumption) when the reaction temperature was raised to 60 • C and using THF as the reaction medium (Table 3, entries 3-7). Well-defined copolymers with controlled molar mass values (i.e., M n,NMR values close to the anticipated ones M n,theo , fairly narrow dispersity values) were isolated after purification by dialysis in THF (so as to remove unreacted CTA 1). Both the 1 H and 13 C NMR spectra supported the formation of NB/CDT copolymers end-capped with glycidyl moieties (Figures S9 and S10). Indeed, the spectra exhibited the main chain olefinic (NB and CDT/butadiene) hydrogens along with the typical signals of the glycidyl α,β-unsaturated carboxylate end-groups, similarly to the diepoxide telechelic PCOE homopolymers spectra [39], thereby demonstrating the formation of DF and possibly CNF P(NB-co-CDT) copolymers.
In contrast, the similar ROMP/CM of NB and CDT (50:50) with CTA 3 mediated by G2 in CH 2 Cl 2 was effective ( Table 3, entries 8-9). The CM proceeded to completion (quantitative CTA 3 conversion) affording P(NB-co-CDT) copolymers with the expected molar mass value as determined by a 1 H NMR analysis, yet with a somewhat large dispersity. Characterization of the copolymers by 1 H and 13 C{ 1 H} NMR spectroscopy supported the formation of α,ω-dioxetane telechelic P(NB-co-CDT) copolymers (Figures 3 and 4). The typical 1 H spectrum that is depicted in Figure 3  Remarkably, the oxetane end-groups are always linked to a CDT/butadiene unit, and not to NB, as also observed in the presence of CTA 1 (this was, however, not the case with alike azlactone end-capped copolyolefins [42]. This linkage mode was further evidenced by a 2D COSY 1 H-1 H NMR analysis that displayed the correlation peak between the oxetane hydrogens H a,b and H c with the CDT/butadiene hydrogen H 2 , while no cross peak was observed between H a,b and H 6 hydrogens, evidencing the absence of an oxetane/NB direct bond ( Figure S11). Correspondingly, the 13 C{ 1 H} NMR spectrum distinctively showed the typical NB (C 6 , C 7 , C 8 ; in green) and CDT/butadiene (C 1 , C 2 , C 4 ; in red) repeating units signals, along with those of the terminal oxetane function (C a -C i ; in black) ( Figure 4). However, the 13 C signals of the C=C isomerized bond (C 3isom , C 4isom , C 5isom ; in blue), and those of the C 6 carbon in the cis or trans configuration of the C=C bond, were not resolved.   (Table 3, entry 8).

Post-Polymerization Chemical Modification of α,ω-Di(epoxide) Telechelic P(NB-co-CDT) into α,ω-Dithiocarbonate Telechelic P(NB-co-CDT)
α,ω-Diepoxide telechelic P(NB-co-CDT) copolymers were next chemically modified into the analogous copolymers featuring dithiocarbonate termini according to our previously established 1 H and 13 C{ 1 H} NMR and FTIR analyses of the α,ω-bis(dithiocarbonate) P(NB-co-CDT) copolymers supported the successful formation of dithiocarbonate chain ends. The NMR signals corresponding to the NB and CDT repeating units were clearly observed in the spectra of the prepolymer ( Figure 5 and Table 6 versus Figures S9 and S10; Table S1). In addition, the typical signals of the dithiocarbonate moiety corresponding to the CH 2 -S (H a , δ (ppm) 3.59) and CH 2 -OC(O) (H c , δ (ppm) 4.49), the former one being clearly deshielded relative to the related signal in the prepolymer (H a , δ (ppm) 2.67), were unambiguously identified. Note that the dithiocarbonate methine signal (H b , δ (ppm) 5.37) could not be used to evidence the successful thiocarbonatation reaction as it overlaps with the main chain signals. The dithiocarboante moiety was clearly evidenced in the 13 C{ 1 H} NMR spectrum by the C=S signal (C g δ (ppm) 210.9, Figure 6). Note that the diagnostic signal of CS 2 (δ 193.1 ppm) was not observed in the spectrum, within the NMR detection limits. Finally, the ATR FTIR spectrum also clearly displays the dithiocarboante C=S and C=O absorption bands at ν = 1192 and 1520 cm −1 , respectively ( Figure S12).  Table 4, entry 1) copolymer sample that was prepared upon dithiocarbonatation of an α,ωbis(dioxetane) P(NB-co-CDT) copolymer with CS2/LiBr (Table 3, Table 4, entry 1) copolymer sample that was prepared upon dithiocarbonatation of an α,ωbis(dioxetane) P(NB-co-CDT) copolymer with CS2/LiBr (Table 3, entry 5).
Evaluation of the amount of cyclic nonfunctional polymer (CNF) within the recovered chemically modified P(NB-co-CDT) crude copolymers was performed upon eluting with pentane a CH2Cl2 mixture of the latter samples with a multifunctional amine (Lupasol FG ® ; 1.5 equiv. NH2 versus dithiocarbonate) through a silica column. Reaction of the dithiocarbonate groups with the  Table 4, entry 1) copolymer sample that was prepared upon dithiocarbonatation of an α,ω-bis(dioxetane) P(NB-co-CDT) copolymer with CS 2 /LiBr (Table 3, entry 5).
Evaluation of the amount of cyclic nonfunctional polymer (CNF) within the recovered chemically modified P(NB-co-CDT) crude copolymers was performed upon eluting with pentane a CH 2 Cl 2 mixture of the latter samples with a multifunctional amine (Lupasol FG ® ; 1.5 equiv. NH 2 versus Polymers 2018, 10, 1241 15 of 19 dithiocarbonate) through a silica column. Reaction of the dithiocarbonate groups with the amine resulted in cross-linking of the α,ω-bis(dithiocarbonate) P(NB-co-CDT), while eluting the apolar CNF copolymer. A 1 H NMR analysis of the eluted sample did not show the signals of the dithiocarbonate moiety (H a -H c in Figure 5) or related functional groups obtained upon reaction with amines, but displayed only signals from NB and CDT repeating units, in agreement with the elution of only CNF copolymers ( Figure S13). Weighting of the thus recovered CNF gave an average weight content of <ca. 10wt%. This value falls within the range of CNF contents previously reported during the alike ROMP/CM metathesis preparation of telechelic POs [38][39][40][41][42]. The alternative quantification of the CNF by 1 H NMR spectroscopy using an internal standard gave the same amounts of CNF.

Thermal Properties of the Polymers
The nature of the end-capping group of telechelic PCOEs as well as of P(NB-co-CDT) copolymers, either epoxide, oxetane, or dithiocarbonate, was found not to impact the thermal transition temperatures (Table 5). PCOEs terminated with epoxide or oxetane did not display a clear glass transition, similarly to other reported telechelic PCOEs, such as those functionalized with glycidyl alkenoate (M n,NMR = 6500 g·mol -1 , T g = not observed, T m = 57 • C, T c = 45 • C) [39], carboxylate (M n,NMR = 5000 g·mol -1 , T g = not observed, T m = 62 • C, T c = 51 • C) [15], dithiocarbonate (M n,NMR = 6500 g·mol -1 , T g = not observed, T m = 56 • C, T c = 43 • C) [39], or azlactone (M n,NMR = 5900 g·mol -1 , T g = not observed, T m = 56 • C, T c = 47 • C) [42] α,ω-end-functional PCOEs. Also, the T m value of PCOEs is known to inform on the cis/trans configuration of the C=C bonds along the PO backbone. T m values of PCOEs ranging from 10-54 • C are indicative of 99%-20% of cis C=C bonds, respectively [49]. Correlation of the experimentally observed T m values of the telechelic PCOEs that are gathered in Table 5 thus suggested ca. 20% of cis C=C bonds along the polymer's backbone. All of the P(NB-co-CDT)s end-capped with epoxide, oxetane, or dithiocarbonate moieties showed a very similar thermal signature, similar to the one that was reported for azlactone end-capped PCOEs, although a glass transition was not observed in the latter copolymers (M n,NMR = 6900 g·mol -1 , T g = not observed, T m = 7 • C, T c = −18 • C) [42]. It is noteworthy that PCOE (M n,NMR = 2900 g·mol -1 , T g = −78 • C, T m = 52 • C, T c = 45 • C) and P(NB-co-CDT) (M n,NMR = 2900 g·mol -1 , T g = not observed, T m = 19 • C, T c = −3 • C) α,ω-end-functionalized with trimethoxysilyl groups exhibited a thermal profile that was distinct from these telechelic coPOs [40].

Viscosity of the Copolymers
The viscosity properties of the epoxide, oxetane, and dithiocarbonate telechelic copolymers were investigated at ambient temperature in simple shear flows (Table 6). Epoxide telechelic P(COE-co-NB) and P(COE-co-NB COOMe ) copolymers showed a rather high, and similar, viscosity regardless of the presence of the methyl ester substituent on the NB units ( Table 6, entries 1-2). This suggested that the functional NB COOMe did not alter the linearity of the P(COE-co-NB) copolymer (a longer alkyl group-e.g., hexyle-on the ester may affect the viscosity more significantly), and that COE most likely enabled an increase in the viscosity of NB segments. In comparison, the viscosity of P(NB-co-CDT) copolymers showed, regardless of the terminal epoxy, oxetane, or dithiocarbonate functionality, a much lower viscosity, most likely imparted by the CDT segments. The latter copolymers also exhibited the same viscosity as the related azlactone telechelic P(NB-co-CDT) copolymers [42]. All copolymers displayed a shear thinning rheological behavior. Table 6. Newtonian viscosity of telechelic P(NB-co-CDT) copolymers.

Conclusions
Both α,ω-diepoxide and -dioxetane telechelic (co)POs have been straightforwardly and selectively prepared from the tandem ROMP/CM polymerization of COE, of COE with NB or NB COOMe , or of CDT with NB, catalyzed by G2 in the presence of the symmetrical epoxide or oxetane functional alkene CTA, respectively. The CM was found to be more effective for CTA 3 > 1 >> 2; this further highlights the key importance of the central C=C moiety in these chain-transfer agents, where both electronic and steric considerations must be taken into account. Well-defined α,ω-di(epoxide or oxetane) telechelic PCOEs, P(COE-co-NB or -NB COOMe )s, and P(NB-co-CDT)s were, thus, obtained under mild operating conditions (solvent, 40-60 • C, 24 h). The post-metathesis polymerization quantitative chemical modification (CS 2 /LiBr, 40 • C, THF) of α,ω-di(epoxide) telechelic P(NB-co-CDT) successfully afforded a convenient approach to the corresponding α,ω-bis(dithiocarbonate) telechelic P(NB-co-CDT). All of the (co)polymers display spectroscopic (NMR, FTIR) and spectrometric (MS) data evidencing the polyolefinic backbone along with the respective chain-end functions. Cross-metathesis of CTAs takes place selectively on COE or CDT segments, and not on NB units. The DSC signature and the low viscosity liquid behavior of the epoxide/oxetane/dithiocarbonate telechelic P(NB-co-CDT)s were not influenced by the nature of the termini. Overall, this work establishes the first reported preparation of both α,ω-di(oxetane) and α,ω-bis(dithiocarbonate) telechelic copolyolefins.