Abstract
2,5-Furandicarboxylic acid and itaconic acid are both important biobased platform chemicals and their terpolymer with 1,6-hexanediol (HDO) can be the starting point for a new class of reactive polyesters, with important applications. The green synthetic route developed in this study involves a biocatalytic condensation polymerization reaction of dimethyl furan-2,5-dicarboxylate (DMFDC) and dimethyl itaconate (DMI) with HDO in toluene at 80°C, using commercial immobilized lipases from Candida antarctica B. In the best conditions, the formed polymer product was isolated with more than 80% yield, containing about 85% terpolymer with average molecular mass of about 1200 (Mn, calculated from MALDI-TOF MS data) and 15% DMFDC_HDO copolymer. Considering the higher reactivity of DMFDC, the composition of the synthesized polymer can be directed by adjusting the molar ratio of DMFDC and DMI, as well as by extending the reaction time. Structural analysis by NMR demonstrated the regioselective preference for the carbonyl group from DMI adjacent to the methylene group. The biocatalyst was successfully reused in multiple reaction cycles.
Introduction
The dependence of polymer industry on fossil resources increased exponentially the interest in monomers from renewable resources [1]. Polyesters are widely used polymers as food containers, packaging, fiber and coating materials. The presently available technologies for polyester synthesis involve petroleum-based diacids and diols as raw materials and metallic alkoxides as catalyst [2]. One of the greatest challenges for academic and industrial sectors over the past few years was to obtain new bio-based and renewable materials.
Functional monomers produced from renewable biomass feedstock are promising candidates for green polyesters and provide a great opportunity for achieving future sustainability [3]. Moreover, the increasing interest in renewable, biobased polymers has resulted in the quest for new synthesis routes to produce polymer building blocks [4].
2,5-Furandicarboxylic acid (FDCA) and itaconic acid (IA) are found on the list of 12 most important bio-based monomers that received a great attention in the scientific and industrial field as starting materials for innovative chemicals and other important goods [5]. It was proved recently that FDCA has the potential to replace petroleum based terephthalic acid due to its similar properties. Although, there are many ways to obtain FDCA, the main path remains the oxidation of 5-hydroxymethylfurfural (HMF) in different mediums using various catalysts [6]. Enzymatic catalyzed routes of FDCA synthesis with high yield have been reported using TEMPO-mediated lipase [7], while Dijkman et al. have synthesized successfully FDCA using HMF oxidase [4]. The relevance of FDCA was proved by different groups that used FDCA and/or derivatives as raw material in combination with different comonomers for chemically and enzymatically polymerization. Comonomers as lactic acid [8], succinic acid [9], [10] and aromatic compounds [11] were used in polymerization processes using metal catalysts. Nevertheless, the biocatalytic route is explored more and more with substrates as polyols [12], [13], [14] and diamines [15].
Alongside FDCA and its derivatives, itaconic acid (IA) is another compelling renewable substrate for synthesis of novel functionalized polymers. Nowadays IA is obtained industrially from biomass fermentation with Aspergillus terreus [16]. The structural similarities of itaconic acid and the corresponding esters (such as dimethyl itaconate and dibutyl itaconate) with acrylic acid and methacrylic acid can lead to replacing the last-mentioned in dental materials, shape memory polymers, coatings, elastomers, drug-delivery and corrosion inhibitors [17]. Due to the possibility of modifying the vinylic double bond after the polycondensation, itaconic acid and its derivatives could promote the developing of new materials with remarkable properties. Polymerization with IA and its esters represents a challenge, since the double bond can isomerize at high temperature or can be involved in radical crosslinking even when it is not desired. Enzymatic polymerization of itaconic acid and/or derivatives with 1,4-butanediol [18] and ethylene glycol has been reported recently [19].
1,6-Hexanediol (HDO) can be considered a biobased compound as well [20] and is widely used in polymerization processes [21]. Polyurethane industry is the main field of applicability of HDO, but it is used in coatings, adhesives and plasticizers also [22]. As monomer, HDO was successfully used in enzymatic catalyzed polymerization as substrate for lipases [23], [24]. Moreover, HDO was showed to be the best comonomer for FDCA in the synthesis of FDCA-polyesters with aliphatic α,ω-diols [12], [23].
One of the most extensively studied enzymes in biocatalytic polyester synthesis is Candida antarctica Lipase B (CALB) [25], [26]. This enzyme presents a broad substrate adaptability and stable catalytic performance in a large variety of chemical media. Compared to conventional catalysts, CALB was successfully used for producing novel polyesters containing sensitive structures without toxic residuals [27]. These reaction products have promising applications in biomedical and tissue engineering [3].
The aim of this study was to investigate the synthesis and characterization of novel copolyesters derived from two dicarboxylic monomers: dimethyl furan-2,5-dicarboxylate and dimethyl itaconate, reacted with 1,6-hexanediol. Synthesis of terpolyesters has gained increasing popularity in order to obtain materials with applications in the medical field [28], [29], [30] or as coatings [31], [32]. Up to now, the number of reports on lipase-catalyzed routes for renewable biobased terpolyesters is scarce [24], [28], [33]. As far as we know, FDCA and IA were subjected to polycondensation with 1,3-propanediol and succinic acid, but using a chemical catalyst, tert-butyl peroxybenzoate [34]. Even though FDCA and IA have been previously reported as substrates for lipases, at our best knowledge this is the first study concerning a green route towards terpolymers based on these important building blocks. Additionally, a less-known commercial immobilized lipase from Candida antarctica B (GF-CalB-IM) was for the first time used as catalyst for polymer synthesis. Its efficiency was compared with Novozyme 435, the standard lipase used for polymerization reactions, while the GF-CalB-IM enzyme is an emerging immobilized lipase on the market.
Materials and methods
Materials
Dimethyl furan-2,5-dicarboxylate (DMFDC) (99.9%) was purchased from Apollo Scientific, dimethyl itaconate (DMI) (97%), 1,6-hexanediol (HDO) (97%), (trans-2-[3-(4-t-butyl-phenyl)-2-methyl-2-propenylidene]malononitrile (DCTB), potassium trifluoroacetate (KTFA), tin (II)-2-ethylhexanoate (SnOct2) (92.5–100%), chloroform (>99%) and toluene (>99%) were purchased from Sigma Aldrich. Immobilized Candida antarctica lipase B on acrylic resin (Novozyme 435) was from Novozymes, while the immobilized Candida antarctica lipase B on microporous ion exchange resin (GF-CalB-IM) was purchased from GenoFocus Inc. (Korea Republic).
Methods
Polyester synthesis
The appropriate amounts of dimethyl furan-2,5-dicarboxylate (18–37 mg, 0.1–0.2 mmol), dimethyl itaconate (15–32 mg, 0.1–0.2 mmol) and 1,6-hexanediol (29–42 mg, 0.25–0.35 mmol) were solubilized in 1 mL toluene. The molar ratios of DMFDC and DMI raw materials were selected as 1:1, 1:2 and 2:1, while the ratio of the biocatalyst was set to 3%, 5% and 10% (wt.%), respectively. Since both biocatalysts were Candida antarctica B lipases immobilized by adsorption on ion exchange resins, the same quantities of immobilized preparates were used in parallel experiments. In the absence of reliable data concerning the amount of loaded lipase in both these commercial products, the biocatalyst ratio was considered the weight of the immobilized enzyme related to the total weight of co-monomers. Alternatively, the reaction was carried out with a chemical catalyst, 5% (wt.%) tin (II)-2-ethylhexanoate. In all experiments, a slight molar excess of HDO (0.5 M) was used. The polymerizations were carried out in 2 mL Eppendorf vials at 80°C, 1000 rpm for 24 h, by using a Comfort Thermomixer (Eppendorf). Because partial evaporation of the solvent during the reaction could not be prevented, during the time-course experiment toluene was added at every 24 h interval, to restore the initial total volume in the vial. At the end of the reaction, the resulted mixture was dissolved in chloroform and the enzyme was removed by filtration. The organic solvents, together with the unreacted raw materials, were removed by vacuum evaporation overnight at 50°C and the resulting white solid product was weighed and analyzed by MALDI-TOF MS, NMR, FT-IR, TG and DSC techniques. The total polymer yields were calculated by division of the isolated polymer weight to the total monomer amount introduced in the reaction and were between 70 and 85% at 2:1 DMFDC:DMI molar ratio, 60–70% at equimolar DMFDC:DMI ratio and between 50 and 60% at 1:2 DMFDC:DMI molar ratio, indicating that some itaconate monomer remained unreacted. All reactions were performed in duplicate.
Characterization of the polymerization products
MALDI-TOF MS analysis were carried out by using an UltrafleX extreme Bruker spectrometer with FlexControl and FlexAnalysis software packages for acquisition and processing of the data (Bruker Daltonics, Germany), at an acceleration voltage of 25 kV, using DCTB as matrix and KTFA as ionization agent. The sample preparation and analysis was performed as previously described [35]. The number average molecular weight (Mn), weight average molecular weight (Mw) and dispersity (ĐM) have been calculated as described elsewhere [36].
NMR spectra of the isolated product were recorded on a Bruker AVANCE III spectrometer operating at 500.0 MHz (1H) and 125.0 MHz (13C). The samples were dissolved in deuterated chloroform and the chemical shifts δ are given in ppm from TMS.
Fourier Transform Infrared (FT-IR) spectra of the samples were obtained in attenuated total reflectance (ATR) mode on a Bruker Vertex 70 (BrukerDaltonics GmbH, Germany) spectrometer equipped with a Platinium ATR, Bruker Diamond Type A225/Q. Spectra were collected in the range 4000–400 cm−1 with a resolution of 4 cm−1 and with 64 co-added scans.
The thermogravimetric analyses of the polyesters were performed by using TG 209 F1 Libra thermogravimetric analyzer (Netzsch, Germany) under nitrogen atmosphere, heating rate 10°C min−1 in the temperature range 20–500°C.
The differential scanning calorimetry characterization was accomplished by a DSC 204 F1 Phoenix differential scanning calorimeter (Netzsch, Germany) under nitrogen atmosphere, heating rate 10°C min−1, in the temperature range from −70 to 500°C. The DSC parameters such as initiation temperature (Ti), final temperature (Tf), their difference (ΔT=Tf−Ti), peak temperature (T pk) and enthalpy (ΔH) were collected.
Results and discussion
In this work the enzymatic polymerization of three biobased monomers: dimethyl furan-2,5-dicarboxylate, dimethyl itaconate, and 1,6-hexanediol in organic solvent was investigated, to yield new terpolymers.
The possible products of the polymerization reactions are presented in Fig. 1. Linear and cyclic terpolymers with DMFDC, DMI and HDO can be synthesized as main reaction products, but the formation of the secondary reaction products consisting of DMFDC and HDO or DMI and HDO cannot be ignored. For this reason, the effects of different reaction parameters were studied, to favor the formation of the reaction products containing a high terpolymer amount.
Reaction scheme of the enzymatic synthesis of terpolymers and copolymers from dimethyl furan-2,5-dicarboxylate, dimethyl itaconate and 1,6-hexanediol.
The formation of terpolymers was demonstrated based on MALDI-TOF MS spectra. In Fig. 2, a typical MALDI-TOF MS spectrum of the reaction products, obtained at 2:1 DMFDC:DMI molar ratio and GF-CalB-IM (5%, wt.%) as biocatalyst, is presented. For example, the peaks at 1560.7, 1799.0 and 2037.3 m/z correspond to the K+ adducts of the polymeric chains consisting (DMFDC)5−7-(DMI)1-(HDO)8−9. Polymeric chains containing two DMI moieties were also identified. The peak series 1362.4, 1601.7 and 1839.9 m/z correspond to adducts of (DMFDC)4−6-(DMI)2-(HDO)5−8. As main secondary products, copolymers with DMFDC and HDO moieties were also detected, e.g. the peaks at 1500.6, 1738.9, and 1977.3 m/z correspond to (DMFDC)6−8-(HDO)6−8.
MALDI-TOF MS spectrum (1350–2050 m/z range) of the synthesized polyesterification product, catalyzed by GF-CalB-IM (5%, wt.%), at 2:1 DMFDC:DMI molar ratio. The molecular masses of selected polymeric chains, with different number of 2,5-furandicarboxylate (F), itaconate (I), and 1,6-hexanedioxy (A) units, are indicated by arrows.
Terpolyester synthesis using a chemical catalyst
The synthesis of the terpolyesters was investigated by using two catalytic routes: chemical and enzymatic. For the chemical route a commonly used initiator, tin (II) 2-ethylhexanoate (SnOct2), was tested. Previously, this catalyst was successfully used by our group for the synthesis of polyesters based on hydroxy fatty acids [37], as well as by Morales Huerta for synthesis of furanic cyclic oligomers [38], [39]. The reactions were carried out at 80°C, by using toluene as reaction medium at different molar ratios of the co-monomers.
The average molecular weights and the composition of each product (linear/cyclic terpolymers and copolymers) were assessed based on the MALDI-TOF MS spectra. The results (Table 1) indicate that average molecular weights values did not exceed 750 m/z and were not influenced by the relative molar ratio of the monomers. The highest terpolymer content of the reaction product (39%) was obtained when a higher amount of dimethyl itaconate was added, but in this case the formation of cyclic oligomers was favored against the linear form, as well. The DMFDC_HDO copolymer was in all cases the main reaction product, while DMI_HDO copolymer was not detected.
Influence of molar ratio of the monomers on the composition of the polymerization products.
| Entry | Molar ratioa | Mn [Da] | Mw [Da] | Đ M | Composition of the product [%] |
DPmaxb | |||
|---|---|---|---|---|---|---|---|---|---|
| Terpolymers |
Copolymers (linear and cyclic) |
|
|||||||
| Linear | Cyclic | DMFDC_HDO | DMI_HDO | ||||||
| 1 | 1:1:2.5 | 722 | 726 | 1.01 | 9.4 | 13.8 | 76.8 | 0 | 7 |
| 2 | 1:2:3.5 | 689 | 689 | 1.00 | 13.1 | 26.1 | 60.8 | 0 | 5 |
| 3 | 2:1:3.5 | 703 | 706 | 1.00 | 9.0 | 6.9 | 84.1 | 0 | 7 |
-
aMolar ratio of the co-monomers DMFDC, DMI and HDO, respectively; bmaximal polymerization degree of the terpolymer. The experiments were carried out with 5% tin (II)-2-ethylhexanoate, at 80°C, 24 h.
Biocatalytic polymerization. Influence of the reaction parameters
The second catalytic route investigated for the synthesis of terpolymers was the biocatalytic transesterification, by using two commercially available immobilized lipases. The lipase from Candida antarctica B immobilized on acrylic resin (Novozyme 435) was selected based on previous reports on successful utilization of this catalyst for polyester synthesis from itaconic acid derivatives and various diols [40]. The effect of the biocatalyst concentration (wt.%, related to the total amount of monomers) and the DMFDC:DMI molar ratio were investigated, targeting an increased terpolymer amount in the polymerization product. Three DMFDC:DMI molar ratios were considered: 1:1, 1:2 and 2:1, at three enzyme concentrations, 3%, 5% and 10% (wt.), respectively. The results are presented in Table 2.
Influence of the molar ratio of co-monomers on the composition of the polymerization products, at different Novozyme 435/monomer ratios (wt.%, related to the total monomer amount).
| Entry | Enzyme [%] | Molar ratioa | Mn [Da] | Mw [Da] | Đ M | Composition of the product [%] |
DPmaxb | |||
|---|---|---|---|---|---|---|---|---|---|---|
| Terpolymers |
Copolymers (linear and cyclic) |
|||||||||
| Linear | Cyclic | DMFDC_HDO | DMI_HDO | |||||||
| 1 | 3 | 1:1:2.5 | 843 | 878 | 1.04 | 67.9 | 0.1 | 32.0 | 0 | 14 |
| 2 | 3 | 1:2:3.5 | 823 | 852 | 1.04 | 72.0 | 0 | 26.8 | 1.3 | 12 |
| 3 | 3 | 2:1:3.5 | 1093 | 1223 | 1.12 | 32.1 | 0.1 | 67.8 | 0 | 23 |
| 4 | 5 | 1:1:2.5 | 938 | 997 | 1.06 | 53.2 | 1.1 | 42.7 | 3.0 | 14 |
| 5 | 5 | 1:2:3.5 | 843 | 877 | 1.04 | 61.7 | 0.8 | 35.5 | 2 | 12 |
| 6 | 5 | 2:1:3.5 | 935 | 969 | 1.04 | 71.5 | 0.5 | 26.1 | 1.9 | 16 |
| 7 | 10 | 1:1:2.5 | 1041 | 1121 | 1.08 | 74.8 | 0.7 | 23.5 | 1.0 | 17 |
| 8 | 10 | 1:2:3.5 | 836 | 868 | 1.04 | 72.1 | 0.1 | 27.3 | 0.5 | 12 |
| 9 | 10 | 2:1:3.5 | 1205 | 1343 | 1.11 | 71.0 | 0.5 | 28.4 | 0.1 | 22 |
-
aMolar ratio of the co-monomers DMFDC, DMI and HDO, respectively; bmaximal degree of polymerization (of the terpolymer). The reactions were carried out at 80°C, 24 h.
Using Novozyme 435 as biocatalyst, at equimolar DMFDC:DMI amounts a slight increase of the average molecular weight with increasing enzyme concentration was observed and the highest terpolymer content of the reaction product (more than 74%) was obtained at 10% Novozyme 435. At DMI molar excess (1:2 DMFDC:DMI molar ratio), the average molecular weight, the maximal polymerization degree and the terpolymer content of the product were lower but not harshly affected, indicating that the enzyme is less selective towards itaconic acid, compared to 2,5-furandicarboxylic acid. This observation is confirmed by the highest molecular weights and terpolymer contents obtained at DMFDC molar excess (2:1 molar ratio), regardless to the enzyme concentration employed. Our data are in concordance with the decrease of the average molecular weights at higher itaconic anhydride molar ratios reported by Yamaguchi et al. for the synthesis of terpolymers from itaconic anhydride, 1,6-hexanediol and succinic/glutaric anhydride, with Novozyme 435 as catalyst [40]. Using Novozyme 435 as biocatalyst, the optimal reaction conditions were 2:1 DMFDC:DMI molar ratio and 10% biocatalyst concentration, leading to average molecular weights higher than 1300 Da, maximal polymerization degree above 20, and more than 70% terpolymer content of the product. The low amounts of DMI_HDO copolymer detected in the polymerization product demonstrate the much higher reactivity of DMFDC compared to DMI in this polycondensation reaction.
The same experiments were performed using another commercially available immobilized lipase from Candida antarctica B (GF-CalB-IM), used for the first time for polyester synthesis. The purpose was to assess the catalytic efficiency of this less-known biocatalyst, compared to Novozyme 435, the most widely used immobilized lipase. The results are presented in Table 3.
Influence of molar ratio of co-monomers on the composition of the polymerization products, at different GF-CalB-IM/monomer ratios (wt.%, related to the total monomer amount).
| Entry | Enzyme [%] | Molar ratioa | Mn [Da] | Mw [Da] | ĐM | Composition of the product [%] |
DPmaxb | |||
|---|---|---|---|---|---|---|---|---|---|---|
| Terpolymers |
Copolymers (linear and cyclic) |
|||||||||
| Linear | Cyclic | DMFDC_HDO | DMI_HDO | |||||||
| 1 | 3 | 1:1:2.5 | 978 | 1055 | 1.08 | 55.7 | 0.6 | 42.6 | 1.1 | 17 |
| 2 | 3 | 1:2:3.5 | 824 | 855 | 1.04 | 72.0 | 0.3 | 27.6 | 0.2 | 12 |
| 3 | 3 | 2:1:3.5 | 1078 | 1198 | 1.11 | 39.6 | 0.4 | 60.0 | 0 | 23 |
| 4 | 5 | 1:1:2.5 | 1028 | 1112 | 1.08 | 79.7 | 0.4 | 19.4 | 0.5 | 18 |
| 5 | 5 | 1:2:3.5 | 916 | 974 | 1.06 | 85.2 | 0.2 | 12.9 | 1.7 | 15 |
| 6 | 5 | 2:1:3.5 | 1278 | 1444 | 1.13 | 63.7 | 0.7 | 35.5 | 0.1 | 27 |
| 7 | 10 | 1:1:2.5 | 930 | 990 | 1.06 | 87.0 | 0.1 | 12.5 | 0.4 | 16 |
| 8 | 10 | 1:2:3.5 | 795 | 817 | 1.03 | 69.8 | 1.8 | 6.5 | 21.9 | 9 |
| 9 | 10 | 2:1:3.5 | 1160 | 1303 | 1.12 | 65.9 | 1.3 | 32.8 | 0 | 20 |
-
aMolar ratio of the co-monomers DMFDC, DMI and HDO, respectively; bmaximal degree of polymerization (of the terpolymer). The reactions were carried out at 80°C, 24 h.
With GF-CalB-IM as biocatalyst and at equimolar DMFDC:DMI ratio, the highest average molecular weight was obtained at 5% enzyme concentration, but the total terpolymer content in the reaction product (more than 87%) was higher at 10% enzyme concentration. At DMI molar excess (1:2 molar ratio) remained the same tendency concerning the average molecular weight, but the maximum of terpolymer was also found at 5% enzyme concentration. As in the case of Novozyme 435, the best polymerization results were obtained at DMFDC molar excess (2:1 molar ratio), with average molecular weights about 1200–1300 Da and maximal degrees of polymerization more than 20. However, the terpolymer content of the polymerization product decreased to less than 70%. Therefore, the best reaction conditions, leading to both high average molecular mass and high terpolymer yield, were selected as 5% biocatalyst concentration and 2:1 DMFDC:DMI ratio. It can be also concluded that GF-CalB-IM was a more efficient biocatalyst than Novozyme 435 for this process, leading to a polymer product with average molecular weight higher than 1100 Da and terpolymer content in the reaction product higher than 80% at lower (5%) biocatalyst concentration. Therefore, this enzyme was used in the further studies. Compared to the chemically catalyzed reaction, the biocatalytic process was clearly more efficient, but this conclusion is valid only for the tested stannous catalyst.
Time-course of the polymerization reaction
The previous experiments were carried out for 24 h based on some preliminary data, but a thoroughly investigation of the time course of this reaction was also essential, particularly to show whether a longer reaction time could improve or not the polymerization degree. The terpolymer formation and the average molecular weights were assessed (based on the MALDI-TOF MS spectra) at predetermined reaction periods, up to 72 h (Fig. 3). After 12 h reaction time a steep increase of the relative terpolymer content can be observed, in the same time with the decrease of the copolymer content. The DMFDC_HDO copolymer was the major product in the early stages of the reaction, representing 96% of the formed polymer product after 4 h, but dropped to 35% at 24 h reaction time. The highest value for polymerization degree of terpolymer (not shown in Fig. 3) was also achieved after 24 h. After 48 h and 72 h reaction time the relative content of terpolymer still increased by inclusion of the less reactive itaconate units in the polymeric chain, but the average molecular weights were lower compared to 24 h. Therefore, 24 h was considered the optimal reaction time, but a longer time could be also considered for further applications if the terpolymer content will be emerge as main issue.
Time course of the polymerization reaction, at DMFDC:DMI:HDO molar ratio of 2:1:3.5, using GF-CalB-IM (5%, wt.%, related to the total monomer amount).
Reuse of the biocatalyst
The operational stability of GF-CalB-IM was tested in five consecutive synthesis cycles. The selected molar ratio was DMFDC:DMI:HDO=2:1:3.5, while the amount of the biocatalyst was 5% (wt.) related to the total monomer mixture. The results, calculated based on the MALDI-TOF MS spectra (Fig. 4), indicate that after 3 cycles the reaction mixture still contained about 50% of terpolymer with average molecular weights around 1300 Da and maximal polymerization degrees beyond 20 (not shown in Fig. 5). After the third reaction cycle a decrease of the medium molecular weights and of the terpolymer content was observed, due to a possible inhibitory effect of the itaconate derivative. To overcome this problem, a very attractive alternative could be the utilization of thin-film reaction systems, as it was reported by Pellis et al. for the polycondensation of diethyl adipate and 1,4-butanediol, catalyzed by immobilized CALB [41]. Due to the improved mass transfer, such reaction systems allow carrying out the reaction in a solvent-free process characterized by higher viscosity of the reaction mixture, and will be tested in our forthcoming experiments, particularly for the scaling-up of the process.
The efficiency of GF-CalB-IM biocatalyst (5%, wt.%) in multiple polymerization reaction cycles (DMFDC:DMI:HDO molar ratio of 2:1:3.5).
The HMBC NMR spectrum of the synthesized terpolymer by using GF-CalB-IM (5%, wt.%), at DMFDC:DMI:HDO molar ratio of 2:1:3.5.
Structural analysis of the reaction product
The functional group changes during polymerization was monitored by FT-IR spectroscopy (Fig. 1S, Supplementary material). The ester formation was proved by the shifting of the C=C skeletal-in-plane band from 1583 cm−1 in DMFDC to 1574 cm−1 in terpolymer. The broad adsorption peak in terpolymer spectrum at 3433 cm−1 corresponds to -OH stretching. The peak at 1722 cm−1 corresponds to C=O stretching of the ester units, while the peak at 818 cm−1 corresponds to the C=CH2 stretching of itaconate units.
The chemical structures of the terpolymers were determined by 1H NMR, 13C NMR, and by NMR techniques such as HMBC, COSY, 135-DEPT and HSQC. The HMBC NMR spectrum is presented in Fig. 5 (other NMR spectra are shown in Figs. 2S–6S, Supplementary material). Proton peaks (Hg) of the furan rings were observed at 7.20–7.19 ppm while the itaconate units Hc1, Hc2 were identified at 6.33 ppm and 5.71 ppm. The furanoate:itaconate ratio was 2:1.
Using 2D HMBC the chemical ester group among itaconate and furanoate units was demonstrated by the correlation signals among carbonylic carbon atoms 1-C (170.8 ppm) and 2-C (158.2 ppm) with methylene protons Hd and Hf. This is a distance coupling over three bonds (3J) among the carbonylic carbon and the methylene protons of the 1,6-hexanedioxy unit. The ratio between methylene protons Hg (near furanoate carbon) and Hd (near itaconate carbon) is 4:1, confirming the proposed structure presented in Fig. 5. The protons ratio 9.62:1 from the furan ring Hg (7.20 ppm) and the methylene protons Hi (3.93 ppm) confirms that 10 furanoate units are included in the terpolymer chain and one is terminal. In almost all cases the itaconate unit was identified at the end of the terpolymer chain proved by the 1:1.1 protons ratio between Hc1 (6.33 ppm), Hc2 (5.71 ppm) and Ha (3.77 ppm).
The coupled signal among the ester methylene protons Hd (4.10 ppm) and itaconate carbonylic carbon 1-C (170.8 ppm) confirms the preferential transesterification of the carbonylic carbon closest to the methylene group (Hb). The Hb protons ratio indicate that 85% of the itaconate moiety is transesterified at the 1-C carbonylic atom and 15% at 3-C. Moreover, the coupling of methylene protons Hb (4.16 ppm) and the farthest carbonylic carbon 3-C can be identified. Such a regioselectivity between the two carbonyl groups of IA was already demonstrated by Yamaguchi et al. [40]. In our case, the NMR structural data show that the initiation step of the polycondensation reaction mostly occurs with transesterification of DMI with HDO at the carbonyl group adjacent to the methylene group which is more reactive than the carbonyl group adjacent to the vinylidene group. During the polycondensation steps, further itaconate units can be inserted in the polymeric chain and in the polymeric chains initiated by the transesterification of DMFDC and HDO.
Thermal properties
The thermal properties of synthesized terpolymer, together with starting materials were investigated by TG and DSC analysis in nitrogen atmosphere. Figure 6 presents the nonisothermal TG curves of the terpolymer synthesized at DMFDC:DMI:HDO=2:1:3.5 molar ratio using GF-CalB-IM (5%, wt.%) and of the raw materials. The presence of one inflexion point on each thermogram indicate that the decomposition takes place in one single step.
TG curves of the terpolymer (purple) compared to the raw materials (DMFDC – pink; DMI – blue; HDO – green).
For the raw materials it can be observed an abrupt weight loss starting from 100°C, except the itaconic derivative, and more than 99.5% of the mass is lost 200°C. The weight loss of terpolymer starts at about 72°C, but up to 300°C only 21.10% of the mass is lost. The weight loss behavior of terpolymer is partially similar to previously reported terpolymer with itaconic acid [34]. Both show a slower weight loss up to 300°C, which may be attributed to the high molecular weight of the terpolymer. The highest weight loss of terpolymer, about 70%, was observed in the temperature range of 300–400°C. The results are clearly indicating a higher thermal stability of the terpolymer compared to the starting materials.
The DSC parameters are summarized in Table 4 (the thermograms are presented in Fig. 7S, Supplementary material). The terpolymer shows an endothermic peak at 390°C, while the monomers have lower melting temperatures. Also, DMI shows an exothermic peak at −58.8°C due to its liquid state at room temperature. The terpolymer has a higher melting temperature compared to the previously reported melting temperature (144.5°C) of the copolymer consisting of furan and HDO units [42].
DSC parameters of the DMFDC:DMI:HDO terpolymer in nitrogen atmosphere.
| Compound | Ti [°C] | Tf [°C] | ΔT [°C] | Tpk [°C] | ΔH [J/g] |
|---|---|---|---|---|---|
| DMI | −65.1 | −50.5 | 14.6 | −58.8 | 126.5 |
| 23.0 | 54.0 | 31.0 | 44.3 | −233.7 | |
| HDO | 21.1 | 59.0 | 37.9 | 48.0 | −267.8 |
| DMFDC | 103.4 | 122.4 | 19.0 | 113.6 | −209.1 |
| Terpolymer | 337.2 | 417.0 | 79.8 | 390.2 | −339.6 |
Conclusions
Copolyesters with average molecular weights up to 1300 m/z containing dimethyl furan-2,5-dicarboxylate, dimethyl itaconate and 1,6-hexanediol moieties were successfully synthesized, using commercial immobilized lipases. A less-known Candida antarctica lipase B immobilized on microporous ion exchange resin (GF-CalB-IM) performed even better than Novozyme 435, showing excellent potential for higher molecular weight terpolymer synthesis reactions. The optimum molar ratio, catalyst concentration, and reaction time were determined, demonstrating that both polymerization degree and terpolymer composition can be controlled by fine tuning of these parameters. The stability of the GF-CalB-IM biocatalyst was investigated during five reaction cycles, showing a decrease of the average molecular mass after three reuses. Based on MALDI-TOF MS spectra, polyesters containing more than 90% terpolymer in the reaction product were synthesized at prolonged reaction time. The NMR analysis offered important structural information concerning the reactivity and regioselectivity of this reaction, while the thermal analysis confirmed the high thermal stability of the terpolymer and its availability for various applications.
Article note
A collection of papers presented at the 17th Polymers and Organic Chemistry (POC-17) conference held 4–7 June 2018 in Le Corum, Montpelier, France.
Aknowledgement
This work was partially supported by a grant of the Romanian Authority for Scientific Research and Innovation, CNCS/CCCDI – UEFISCDI, project number PN-III-P22.1-PED-2016-0168, within PNCDI III.
References
[1] Y. Zhu, C. Romain, C. K. Williams. Nature540, 354 (2016).10.1038/nature21001Search in Google Scholar PubMed
[2] J. Scheirs, T. E. Long (eds), Modern Polyesters: Chemistry and Technology of Polyesters and Copolyesters, John Wiley & Sons, New York (2005).Search in Google Scholar
[3] Y. Jiang, A. J. J. Woortman, G. O. R. Alberda van Ekenstein, D. M. Petrović, K. Loos. Biomacromolecules15, 2482 (2014).10.1021/bm500340wSearch in Google Scholar PubMed
[4] W. P. Dijkman, D. E. Groothuis, M. W. Fraaije. Angew. Chem. Int. Ed.53, 6515 (2014).10.1002/anie.201402904Search in Google Scholar PubMed
[5] T. Werpy, G. Petersen. Top Value Added Chemicals From Biomass, Volume 1 – Results of Screening for Potential Candidates From Sugars and Synthesis Gas, Report No. NREL/TP-510–35523; National Renewable Energy Laboratory, Golden, CO (2004).10.2172/15008859Search in Google Scholar
[6] A. F. Sousa, C. Vilela, A. C. Fonseca, M. Matos, C. S. R. Freire, G.-J. M. Gruter, J. F. J. Coelho, A. J. D. Silvestre. Polym. Chem.6, 5961 (2015).10.1039/C5PY00686DSearch in Google Scholar
[7] M. Krystof, M. Perez-Sanchez, P. D. de María. ChemSusChem.6, 826 (2013).10.1002/cssc.201200954Search in Google Scholar PubMed
[8] M. Matos, A. F. Sousa, A. C. Fonseca, C. S. R. Freire, J. F. J. Coelho, A. J. D. Silvestre. Macromol. Chem. Phys.215, 2175 (2014).10.1002/macp.201400175Search in Google Scholar
[9] N. Jacquel, R. Saint-Loup, J.-P. Pascault, A. Rousseau, F. Fenouillot. Polymer24, 234 (2015).10.1016/j.polymer.2014.12.021Search in Google Scholar
[10] C. Zeng, H. Seino, J. Ren, K. Hatanaka, N. Yoshie. Macromolecules46, 1794 (2013).10.1021/ma3023603Search in Google Scholar
[11] A. F. Sousa, M. Matos, C. S. R. Freire, A. J. D. Silvestre, J. F. J. Coelho. Polymer54, 513 (2013).10.1016/j.polymer.2012.11.081Search in Google Scholar
[12] A. Cruz-Izquierdo, L. A. M. van den Broek, J. L. Serra, M. J. Llama, C. G. Boeriu. Pure Appl. Chem.87, 59 (2015).10.1515/pac-2014-1003Search in Google Scholar
[13] K. Haernvall, S. Zitzenbacher, H. Amer, M. T. Zumstein, M. Sander, K. McNeill, M. Yamamoto, M. B. Schick, D. Ribitsch, G. M. Guebitz. Biotechnol. J.12, 1600741 (2017).10.1002/biot.201600741Search in Google Scholar PubMed
[14] Y. Jiang, A. J. J. Woortman, G. O. R. Alberda van Ekenstein, K. Loos. Polym. Chem.6, 5198 (2015).10.1039/C5PY00629ESearch in Google Scholar
[15] Y. Jiang, D. Maniar, A. J. J. Woortman, G. O. R. Alberda van Ekenstein, K. Loos. Biomacromolecules16, 3674 (2015).10.1021/acs.biomac.5b01172Search in Google Scholar PubMed
[16] M. Winkler, T. M. Lacerda, F. Mack, M. A. R. Meier. Macromolecules48, 1398 (2015).10.1021/acs.macromol.5b00052Search in Google Scholar
[17] T. Robert, S. Friebel. GreenChem.18, 2922 (2016).Search in Google Scholar
[18] L. Corici, A. Pellis, V. Ferrario, C. Ebert, S. Cantone, L. Gardossi. Adv. Synth. Catal.357, 1763 (2015).10.1002/adsc.201500182Search in Google Scholar
[19] C. Hoffmann, M. C. Stuparu, A. Daugaard, A. Khan. J. Polym. Sci. A53, 745 (2015).10.1002/pola.27498Search in Google Scholar
[20] J. He, S. P. Burt, M. Ball, D. Zhao, I. Hermans, J. A. Dumesic, G. W. Huber. ACS Catal.8, 1427 (2018).10.1021/acscatal.7b03593Search in Google Scholar
[21] S. P. Burt, K. J. Barnett, D. J. McClelland, P. Wolf, J. A. Dumesic, G. W. Huber, I. Hermans. Green Chem.19, 1390 (2017).10.1039/C6GC03606FSearch in Google Scholar
[22] J. He, K. Huang, K. J. Barnett, S. Krishna, D. M. Alonso, Z. J. Brentzel, S. P. Burt, T. Walker, W. F. Banholzer, C. T. Maravelias, I. Hermans, J. A. Dumesic, G. W. Huber. Faraday Discuss.202, 247 (2017).10.1039/C7FD00036GSearch in Google Scholar PubMed
[23] A. Pellis, J. W. Comerford, A. J. Maneffa, M. H. Sipponen, J. H. Clark, T. J. Farmer. Eur. Polym. J.106, 79 (2018).10.1016/j.eurpolymj.2018.07.009Search in Google Scholar
[24] Y. Jiang, A. J. J. Woortman, G. O. R. Alberda van Ekenstein, K. Loos. Biomolecules3, 461 (2013).10.3390/biom3030461Search in Google Scholar PubMed PubMed Central
[25] S. Kobayashi. Proc. Jpn. Acad. Ser. B Phys. Biol. Sci.86, 338 (2010).10.2183/pjab.86.338Search in Google Scholar PubMed PubMed Central
[26] S. Weinberger, A. Pellis, J. W. Comerford, T. J. Farmer, G. M Guebitz. Catalysts8, 369 (2018).10.3390/catal8090369Search in Google Scholar
[27] A. Pellis, E. Herrero Acero, V. Ferrario, D. Ribitsch, G. M. Guebitz, L. Gardossi. Trends Biotechnol.34, 316 (2016).10.1016/j.tibtech.2015.12.009Search in Google Scholar PubMed
[28] D. G. Barrett, T. J. Merkel, J. C. Luft, M. N. Yousaf. Macromolecules43, 9660 (2010).10.1021/ma1015424Search in Google Scholar
[29] S. Brännström, M. Finnveden, M. Johansson, M. Martinelle, E. Malmström. Eur. Polym. J.103, 370 (2018).10.1016/j.eurpolymj.2018.04.017Search in Google Scholar
[30] B. Guo, Y. Chen, Y. Lei, L. Zhang, W. You Zhou, A. B. M. Rabie, J. Zhao. Biomacromolecules12, 1312 (2011).10.1021/bm2000378Search in Google Scholar PubMed
[31] S. Brännström, E. Malmström, M. Johansson. J. Coat. Technol. Res.14, 851 (2017).10.1007/s11998-017-9949-ySearch in Google Scholar
[32] J. Dai, S. Ma, X. Liu, L. Han, Y. Wu, X. Dai, J. Zhu. Prog. Org. Coat.78, 49 (2015).10.1016/j.porgcoat.2014.10.007Search in Google Scholar
[33] J. C. Morales-Huerta, A. Martínez de Ilarduya, S. Muñoz-Guerra. J. Polym. Sci. A56, 290 (2018).10.1002/pola.28895Search in Google Scholar
[34] J. Dai, S. Ma, N. Teng, X. Dai, X. Shen, S. Wang, X. Liu, J. Zhu. Ind. Eng. Chem. Res.56, 2650 (2017).10.1021/acs.iecr.7b00049Search in Google Scholar
[35] A. Todea, L. G. Otten, A. E. Frissen, I. W. C. E. Arends, F. Peter, C. G. Boeriu. Pure Appl. Chem.87, 51 (2015).10.1515/pac-2014-0716Search in Google Scholar
[36] H. J. Rader, W. Schrepp. Acta Polym.49, 272 (1998).10.1002/(SICI)1521-4044(199806)49:6<272::AID-APOL272>3.0.CO;2-1Search in Google Scholar
[37] D. Aparaschivei, A. Todea, I. Păuşescu, V. Badea, M. Medeleanu, E. Şişu, M. Puiu, A. Chiriţă-Emandi, F. Peter. Pure Appl. Chem.88, 1191 (2016).10.1515/pac-2016-0920Search in Google Scholar
[38] J. C. Morales-Huerta, A. Martínez de Ilarduya, S. León, S. Muñoz-Guerra. Macromolecules51, 3340 (2018).10.1021/acs.macromol.8b00487Search in Google Scholar
[39] J. C. Morales-Huerta, A. Martínez de Ilarduya, S. Muñoz-Guerra. Polymers10, 483 (2018).10.3390/polym10050483Search in Google Scholar
[40] S. Yamaguchi, M. Tanha, A. Hult, T. Okuda, H. Ohara, S. Kobayashi. Polym. J.46, 2 (2014).10.1038/pj.2013.62Search in Google Scholar
[41] A. Pellis, L. Corici, L. Sinigoi, N. D’Amelio, D. Fattor, V. Ferrario, C. Ebert, L. Gardossi. Green Chem.17, 1756 (2015).10.1039/C4GC02289KSearch in Google Scholar
[42] G. Z. Papageorgiou, D. G. Papageorgiou, Z. Terzopoulou, D. N. Bikiaris. Eur. Polym. J.83, 202 (2016).10.1016/j.eurpolymj.2016.08.004Search in Google Scholar
Supplementary Material
The online version of this article offers supplementary material (https://doi.org/10.1515/pac-2018-1015).
©2018 IUPAC & De Gruyter. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. For more information, please visit: http://creativecommons.org/licenses/by-nc-nd/4.0/
