The preparation of cyclic ethylene and butylene 2,5-furandicarboxylate oligoesters and their conversion to furan-based polyesters poly(ethylene furanoate) (PEF) and poly(butylene furanoate) (PBF) by ring-opening polymerization (ROP) are described. The cyclic oligoesters were obtained in high yields by both high dilution condensation and thermal cyclodepolymerization methods, and they consisted of mixtures of small size species. Cyclic dimer, trimer and tetramer oligoesters were isolated by semipreparative chromatography and found to be crystalline compounds melting within the 140–200 °C range. ROP catalyzed by Sn(Oct)2 of both mixtures and individual species afforded PEF and PBF with weight-average molecular weights between 50,000 and 60,000 g mol-1. Polymerization rate was found to be higher for butylene than for ethylene cyclic oligofuranoates, and also to increase slightly as cycle size decreased. The thermal properties of PEF and PBF prepared by ROP were in full agreement with those reported for these polyesters obtained by melt polycondensation.Peer ReviewedPostprint (author's final draft

In this paper, the preparation of PBS-ran-PCL copolyesters by enzymatic ring opening polymerization is presented for the first time. The copolyesters were produced in a wide composition range and free of metallic contaminants, so they may be regarded as potential biomaterials. The copolymers have been characterized by proton and carbon nuclear magnetic resonance (1 H and 13 C NMR), gel permeation chromatography (GPC), thermogravimetric analysis (TGA), differential scanning calorimetry (DSC), polarized light optical microscopy (PLOM) and wide angle X-ray scattering (WAXS). The PBS-ran-PCL copolyesters were able to crystallize in the entire composition range and displayed a pseudo-eutectic region. Most copolymers away from the pseudo-eutectic region exhibited a single crystalline phase (PBS-rich or PCL-rich crystalline phase), while within the pseudoeutectic region the copolymers were double crystalline. Observations by PLOM, during isothermal crystallization showed that both nucleation density and spherulitic growth rate of the copolyesters are determined by the component that constitutes the majority phase. WAXS studies revealed that d spacings of selected crystallographic planes depend on composition. Therefore, both DSC and WAXS results suggest that the copolymers are probably isodimorphic, as the PBS-rich crystalline phase may contain small inclusions of PCL co-units, while the PCL-rich crystalline domains may also contain a minor quantity of PBS co-units inside.

Poly(butylene 2,5-furandicarboxylate-co-succinate) copolyesters (coPBFxSy) have been synthesized by\ud ring opening polymerization (ROP). Cyclic butylene 2,5-furandicarboxylate and butylene succinate\ud oligomer mixtures, to be used as monomers for ROP, were prepared by high dilution condensation\ud and enzymatic cyclization reactions, respectively. Two different catalytic systems, tin dioctanoate and\ud supported Candida antarctica lipase B (CALB), were used for polymerization. Thus two series of copolyesters\ud covering the whole range of compositions were prepared and their properties comparatively\ud examined. In both cases, random copolyesters with compositions close to those used in their respective\ud feeds were obtained. The influence of composition on reaction kinetics with respect to time and\ud temperature was evaluated for the two series. Chemically catalyzed ROP rendered copolyesters with Mw\ud in the ~50 000–65 000 g mol-1 range, whereas values between 15 000 and 45 000 g mol-1 were attained\ud when the ROP reaction was assisted by CALB. The thermal behavior of coPBFxSy obtained by ROP was\ud similar to that reported for such copolymers prepared by melt polycondensation. They all start to decompose\ud above 300 °C and display melting enthalpy and temperatures that decrease with copolymerization,\ud attaining minimum values when the comonomer contents are approximate to balance. On the\ud contrary, the glass-transition temperature increased almost linearly with the content of butylene furandicarboxylate\ud units, covering the whole range of values between those of the two parent homopolyesters.\ud Small deviations in thermal properties observed between the two series could be attributed to their differences\ud in molecular weights. Hydrolytic and enzymatic degradation studies revealed that coPBFxSy\ud became more degradable with increasing content of succinic units, whereas the homopolyester PBF\ud remained practically unaffected when incubated under similar conditions.Peer ReviewedPostprint (author's final draft

A series of poly­(butylene 2,5-furan­dicarboxylate-co-terephthalate) copolyesters coPBF x T y covering the whole range of compositions has been prepared via ring opening polymerization (ROP). Cyclic oligomers of butylene 2,5-furan­dicarboxylate c(BF) n and butylene terephthalate c(BT) n , both mainly consisting of a mixture of dimer, trimer, and tetramer species, were synthesized by the high dilution condensation method. Random copolyesters with the targeted compositions and weight-average molecular weights within the 55 000–80 000 g·mol–1 range were obtained by ROP of c(BF) n /c(BT) n mixtures in periods of time much shorter than those required by melt polycondensation. The thermal properties of these copolyesters were consistent with their compositions and comparable to their isocompositional analogs obtained by polycondensation. A comparative kinetics study of the isothermal crystallization of the homopolyesters and copolyesters differing in composition revealed that the presence of the 2,5-furan­dicarboxylate units decreased the crystallization rate of PBT. Nevertheless, coPBF x T y copolyesters with moderate contents in BF units were still able to crystallized rapidly from the melt.

Cyclic oligo(butylene 2,5‐furandicarboxylate) and ɛ‐caprolactone were copolymerized in bulk at 130–150 °C by enzymatic ring opening polymerization using CALB as catalyst. Copolyesters within a wide range of compositions were thus synthesized with weight‐average molecular weights between 20,000 and 50,000, the highest values being obtained for equimolar or nearly equimolar contents in the two components. The copolyesters consisted of a blocky distribution of the ɛ‐oxycaproate (CL) and butylene furanoate (BF) units that could be further randomized by heating treatment. The thermal stability of these copolyesters was comparable to those of the parent homopolyesters (PBF and PCL), and they all showed crystallinity in more or less degree depending on composition. Their melting and glass‐transition temperatures were ranging between those of PBF and PCL with values increasing almost linearly with the content in BF units. The ability of these copolyesters for crystallizing from the melt was evaluated by comparative isothermal crystallization and found to be favored by the presence of flexible ɛ‐oxycaproate blocks. These copolyesters are essentially insensitive to hydrolysis in neutral aqueous medium but they became noticeably degraded by lipases in an extend that increased with the content in CL units. © 2017 Wiley Periodicals, Inc. J. Polym. Sci., Part A: Polym. Chem. 2018, 56, 290–299

Cyclic oligomers of isomannide 2,5-furandicarboxylate were synthesized using the high dilution condensation method. A mixture of dimer, trimer and tetramer species largely enriched in the dimer was obtained. These cyclic oligomers were copolymerized with those made of butylene 2,5-furandicarboxylate in bulk at 220 º C by ring opening polymerization using Sn(Oct) 2 as catalyst. A series of random poly(butylene 2,5-furandicarboxylate) copolyesters containing isomannide in a range of 5 to 50 %-mole and with weight-average molecular weights between 30,000 and 50,000 g•mol-1 were prepared. These copolyesters started to decompose above 300 ºC and only those containing less than 10 %-mole of isomannide showed signs of crystallinity. They displayed glass-transition temperatures in the 40-100 ºC range with values increasing steadily with the content in isomannide. At difference with the poly(butylene 2,5-furandicarboxylate) homopolyester that is reluctant to hydrolysis, the isomannide-containing copolyesters were noticeably degraded by water, much more rapidly when exposed to the presence of lipases.

A mathematical model for pretreatment and hydrolysis of lignocellulosic biomasses in twin-screw extruders is presented. The model allows for calculation of the concentrations of extracted hemicellulose and sugars (xylose and glucose) at different operating conditions (pressure, temperature, and NaOH and H2SO4 concentrations, as well as screw rotational speed). Parameter sensitivity analyses on the effects of process variables on product concentrations are reported. Regarding the pretreatment case, the effects of NaOH concentration, liquid/solid ratio, and screw rotational speed on the extracted hemicellulose are studied. In the case of hydrolysis, the effects of NaOH concentration, screw rotational speed, and barrel temperature on the final concentrations of glucose and xylose are addressed. Reasonably good agreement between experimental data from different pretreatment/hydrolysis systems and laboratories and the calculations obtained with our model is observed.

A recent mathematical model describing the pretreatment and alkaline hydrolysis of lignocellulosic biomasses in twin screw extruders (TSEs) is used to study the deconstruction and degradation of blue agave bagasse (BAB), an important waste biomass from the Tequila industry, in a laboratory-scale extruder. This pretreatment process using TSEs is technologically important. Experiments using NaOH for the delignification, extraction, and partial hydrolysis of hemicellulose and cellulose were performed. The effects of using different configurations of the screw elements, temperatures, and operational conditions on biopolymer degradation were assessed. NaOH concentration, residence time, and temperature are found to be determining factors in the extraction of cellulose and hemicelluloses. Partial hydrolysis of the carbohydrates is observed at some of the studied conditions. Reasonably good agreement between experimental data and model predictions is observed, which confirms that the key aspects of the actual process are captured well with the model. This modeling frame may be used in the future for the analysis and study (e.g., scaling up) of industrial processes.