Process for the production of a high molecular weight polyester (CO)polymer

Abstract

The present invention relates to a process for the production of a polyester (co)polymer, comprising the steps of a) adding a dialkyl oxalate ester to a starting polyester (co)polymer at least comprising carboxylic acid derived monomer units and 1,2-diol derived alcohol end groups, 5 wherein the alkyl group in the dialkyl oxalate ester is an aliphatic alkyl group wherein the carbon in the alkyl group that is attached to the oxalate is a primary carbon; and b) in case the temperature is not already higher than 220 °C, elevating the temperature of the mixture resulting from step (a) to at least 220 °C, and reducing the pressure, for a period of time sufficient to obtain a polyester (co)polymer product having a higher molecular weight than that of the starting 10 polyester (co)polymer, wherein the amount of oxalate units that remain in the (co)polymer is less than the amount added in the form of the dialkyl oxalate ester.

Description

PROCESS FOR THE PRODUCTION OF A HIGH MOLECULAR WEIGHT POLYESTER (CO)POLYMERField of the invention

[0001] The present invention relates to a (catalyst free) process for the production of a high molecular weight polyester (co)polymer.Description of the background art

[0002] An important goal of polymerization processes is to obtain polymers with a molecular weight high enough for the desired application(s). This is important as the molecular weight of the polymer relates to polymer performance such as strength, toughness and durability. The use of polymers with insufficient molecular weight may lead to application failures. Therefore, many studies concerning polymerization processes and the conditions used in those processes relate to realizing the target (high) molecular weight.

[0003] Polyesterification is a reversible reaction with a relatively low equilibrium constant. As a consequence, removal of condensation product has an impact on the molecular weight that can be achieved. Melt polycondensation at reduced pressure is commonly used in polyesterification processes for removal of condensation product. However, the increase of molecular weight of the polymers during that process also increases the viscosity of the melt material, which complicates the removal of condensation product. This may eventually become a limiting factor. Removal of condensation product can be improved, for example by using higher temperatures, longer reaction times, catalysts and improved reactor designs. However, under melt conditions, limited mass transfer due to high viscosity of the melt material, in combination with longer residence times and (potential) chemical degradation, may limit the possibilities to obtain higher molecular weights. For example, to obtain high molecular weights of polyethylene terephthalate (PET) necessary for bottles (intrinsic viscosity (IV) of 0.73 - 0.85 dL / g) and industrial yarns (IV >1.2 dL / g), an additional solid-state polymerization (SSP) step may be required. In SSP, polymer pellets are heated below the melting point while being under a nitrogen flow or vacuum, optionally being rotated. A drawback of SSP is that due to the low mobility of the end groups and condensate in the solid state, this is time and energy consuming and therefore an expensive process.

[0004] When less reactive diols such as isosorbide are introduced, it becomes even more difficult to obtain sufficiently high molecular weights. Incorporation of isosorbide is interesting due to its additional benefits on thermomechanical stability and mechanical performance, which opens new possibilities for applications. Isosorbide is less reactive due to its secondary alcohol groups, and melt polycondensation becomes considerably more difficult with increasing isosorbide content. Furthermore, the crystallinity of the polymer is strongly reduced if isosorbide contents are above around 15%. This makes it not possible to use SSP as an amorphous polymer would clump together.

[0005] An alternative route for producing high molecular weight polymers could be by utilizing a so-called chain extender after melt polycondensation (see e.g. P. Raffa et al., Reactive & Functional Polymers 72 (2012) 50-60). Chain extenders are very reactive molecules which react with the remaining alcohol and / or acid chain ends to increase the molecular weight. Only little chain extender is needed, as already a considerable chain length is usually obtained after melt polycondensation. Due to the high reactivity of the chain extender, considerably less time and less harsh conditions are required to obtain high molecular weight. This can reduce costs by cutting down on polymerization conditions (temperature, time, catalyst, reactor), or by eliminating theneed for SSP. Various chain extenders have been used for production of high molecular weight polyesters, for example ethylene carbonate, bis-oxazolines, pyromellitic dianhydride, organic phosphites, di-isocyanates, di-epoxides, carbonyl biscaprolactam, diphenyl carbonate, diphenyl terephthalate, bisketenimines and bislactams.

[0006] However, the use of chain extenders also comes with drawbacks. Chain extenders become incorporated in the polymer and thus become part of the molecular structure of the polyester. The groups that are incorporated into the polymer backbone inevitably influence the properties of the material. Further, often side reactions occur, such as crosslinking or chain scission, which also change the physical properties of the polymer. Moreover, some chain extenders are considered toxic, on their own or as residue in polymer, thereby excluding use for food-grade applications. These drawbacks are amongst the reasons why today chain extenders are rarely used in commercial polyester production. In addition, as a consequence of the use of the currently known chain extenders, when the produced polymer is to be recycled, the chemical structure of the resulting polymer is further altered after each recycling session. This complicates the end life of the polymer and limits the amount of possible recycling sessions.

[0007] WO 2024 / 094621 and WO 2024 / 094633 describe processes for the production of a molecular weight polyester (co)polymer comprising adding a diphenyl oxalate ester to a starting polyester (co)polymer comprising alcohol end groups and units derived from 1,2-diols, wherein the phenyl group in the diphenyl oxalate ester is optionally substituted, the substituent being selected from one or more of o-, m- and p- C1-C6 alkoxy, and o-, m- and p- C1-C6 alkyl. This step is followed by in case the temperature is not already higher than 220 °C, elevating the temperature of the mixture resulting from step (a) to at least 220 °C, and reducing the pressure, for a period of time sufficient to obtain a polyester (co)polymer product wherein the amount of oxalate units that remain in the (co)polymer is less than initially added in the form of the diphenyl oxalate ester; and no or less than 1 mole % oxalate units are present, the percentage relative to the total amount of monomer units, the product having a higher molecular weight than that of the starting polyester (co)polymer.

[0008] There is a desire for a process for the production of a high molecular weight polyester (co)polymer in which condensation products have at most low toxicity.

[0009] Furthermore, there is a desire for processes for the production of high molecular weight polyesters which do not require the presence of a metal catalyst. For example, the majority of PET is currently produced using antimony (Sb) catalysts. In medical applications, metal catalysts are often undesired because of possible toxicity.

[0010] Therefore, there is a desire for alternative, improved processes for the production of high molecular weight polyesters which do not require high energy consumption and preferably do not have one or more of the drawbacks of the use of the commonly known chain extenders.Brief summary of the invention

[0011] According to the present invention, such an improved process is provided. The present invention relates to a process for the production of, preferably high molecular weight, polyester (co)polymer, comprising a) adding a dialkyl oxalate ester to a starting polyester (co)polymer at least comprising carboxylic acid derived monomer units and 1,2-diol derived alcohol end groups, wherein the alkyl group in the dialkyl oxalate ester is an aliphatic alkyl group wherein the carbon in the alkyl group that is attached to the oxalate is a primary carbon; b) in case the temperature is not already higher than 220 °C, elevating the temperature of the mixture resulting from step (a)to at least 220 °C, and reducing the pressure, for a period of time sufficient to obtain a polyester (co)polymer product having a higher molecular weight than that of the starting polyester (co)polymer, wherein the amount of oxalate units that remain in the (co)polymer is less than the amount added in the form of the dialkyl oxalate ester. Preferably, wherein the amount of oxalate units that remain in the (co)polymer is less than 50% of the amount added in the form of the dialkyl oxalate ester, more preferably wherein in the (co)polymer product no or less than 1 mol% oxalate units are present, the percentage relative to the total amount of monomer units. The expression (co)polymer polyester indicates that the polyester can be either a polymer polyester wherein two types of monomer units are joined in the same polymer main chain or a copolymer polyester in which case three or more types of monomer units are joined in the same polymer main chain.

[0012] In this process, the dialkyl oxalate ester is used as a "molecular weight booster" under relatively mild conditions instead of a so-called "chain extender" according to prior art processes. As discussed above, commonly used chain extenders cope with drawbacks, as they are incorporated into the polymer chain and alter the physical properties of the polymer and may leave toxic residues. Advantageously, the dialkyl oxalate ester used according to the process of the invention acts as a molecular weight booster rather than a chain extender. The molecular weight booster tends not to be incorporated into the polymer chain of the final product. As a consequence, it does not influence the polymer properties apart from the molecular weight and properties related to the molecular weight only, and does not leave toxic residue in the polymer end product.

[0013] Furthermore, by using the molecular weight booster according to the invention the use of a catalyst, especially metal catalyst, in the process may be avoided altogether. Polyester chains of low molecular weight can be formed via autocatalysis in the absence of metal catalyst and subsequently connected by using the highly reactive dialkyl oxalate which is subsequently removed from the polymer chain upon heating.

[0014] The high reactivity of the dialkyl oxalate ester provides flexibility in its use in polymerization processes. It can be used as alternative to SSP in the case of crystalline polymers, or it may open up routes to produce polymers starting from less reactive or unstable monomers at less severe reaction conditions while still obtaining acceptable molecular weights. For example, for polymers containing isosorbide-derived units it is important to obtain high molecular weight directly after melt polycondensation as they are amorphous if the isosorbide content is over about 10-15 mole% and SSP is not possible. The process of the present invention may further allow for more effective recycling of certain polymers, as the process allows to increase the number of recycle sessions without altering the physical properties of the polymers that are produced in each of the recycle sessions.

[0015] In addition to the high reactivity, the dialkyl oxalate ester molecular weight boosters, in particular dimethyl oxalate (DMO) and diethyl oxalate (DEO), allow for facile evaporation of the alkanol byproducts (methanol and ethanol, respectively). In addition, DMO and DEO have low toxicity and may be produced in efficient and effective processes known to the skilled worker.

[0016] The present invention provides an advantageous process for the preparation of polyester (co)polymers while obtaining high number average molecular weights of the polyester end product. For example, advantageously, according to the presently claimed process, it may be possible to produce very high molecular weight polyethylene furanoate (PEF) or polyethylene terephthalate (PET) without the addition of a metal catalyst. The absence of a metal catalyst canbe advantageous from multiple perspectives. For example, the majority of PET currently is produced using antimony (Sb) catalysts. However, there are concerns about depletion of Sb reserves. Furthermore, in medical applications metal catalysts are often undesired because of toxicity.

[0017] The high molecular weight polyester (co)polymers produced according to the invention can advantageously be used in a broad range of commercial applications, such as in fibers, injection blow molded parts and bottles, 3D printing and packaging materials.Detailed description

[0018] The following is a description of certain embodiments of the invention given by way of example only.

[0019] The present invention relates to a process for the production of a polyester (co)polymer, in particular for the production of a high molecular weight polyester (co)polymer.

[0020] By a "polyester" herein is understood a polymer comprising a plurality of monomer units linked via ester functional groups in its main chain. An ester functional group can be formed by reacting a hydroxyl group (-OH) with a carboxyl / carboxylic acid group (-C(=O)OH). Typically, a polyester is a synthetic polymer formed by the reaction of one or more bifunctional carboxylic acids with one or more bifunctional hydroxyl compounds. Polyesters may also comprise units derived from monomers carrying both a hydroxyl group and a carboxylic acid group, such as hydroxycarboxylic acids, like lactic acid (LA) and glycolic acid (GA), and hydroxyalkanoates (HA), and the like. By the term "starting polyester (co)polymer" herein is understood the initial polyester (co)polymer that is used as a starting point to increase its molecular weight.

[0021] By a "monomer unit" is herein understood a unit as included in a polyester (co)polymer or oligomer, which unit can be obtained after polymerization of a monomer, that is, a "monomer unit" is a constitutional unit contributed by a single monomer or monomer compound to the structure of the polymer or oligomer, herein in particular the smallest diol or di-acid repeating unit.

[0022] By a "monomer" or "monomer compound" is herein understood the smallest building block used as the starting compound to be polymerized, such as a diol or di-acid compound, but also a hydroxycarboxylic acid.

[0023] By an "oligomer" or "oligomer compound" is herein understood a molecular structure comprising an average number of monomer units of in the range from equal to or more than 2 to equal to or less than 50 monomer units, and preferably at least 25 monomer units. Next to diol and di-acid derived monomer units, also other monomer units may be part of the oligomer, such as hydroxycarboxylic acid derived monomer units, in particular derived from a-hydroxycarboxylic acids, such as glycolic acid, lactic acid, mandelic acid and 3-alkoxy carbonic acid.

[0024] The present invention relates to a process for the production of a preferably high molecular weight, polyester (co)polymer, comprising the use of a dialkyl oxalate ester as molecular weight booster. The term "molecular weight booster" herein relates to a chemical that is used to increase the molecular weight of a polymer by assisting chain extension of the polymer, thereby forming a polymer product with higher molecular weight without becoming part of the polymer product itself, i.e. the chemical is not incorporated as a unit into the polymer chain.

[0025] The present process relates to the production of a polyester (co)polymer, comprising adding a dialkyl oxalate ester as molecular weight booster to a starting polyester (co)polymer at least comprising alcohol end groups and units derived from 1,2-diols. Particularly advantageousare starting polyester (co)polymers comprising glycol derived end groups. The term "glycol derived end groups" herein means: end groups derived from monomers wherein two hydroxyl (-OH) groups are attached to different carbon atoms, such as is the case when the end group is derived from for example mono ethylene glycol, 1,2-propanediol, 2,3-butanediol, and the like. A preferred glycol derived end group is the group derived from mono ethylene glycol namely the 2- hydroxyethyl end group.

[0026] In the current process only a small amount of the dialkyl oxalate ester is needed to boost the molecular weight. Amounts as small as 0.2 mol % of the dialkyl oxalate ester may be suitable, more preferably 0.4 mol %; in particular, an amount of up to 25 mol % is used, preferably up to 15 mol %, more preferably up to 10 mol %, the percentages being relative to the total amount of carboxylic acid-derived monomer units in the starting polyester (co)polymer. Ideally, the amount of dialkyl oxalate ester used is about half the amount of the total amount of alcohol end groups present in the starting polyester (co)polymer. With a theoretical stoichiometric ratio of the dialkyl oxalate ester to the reactive alcohol end groups, it is likely that very high molecular weight can be obtained within a very short time. Thus, ideally, and if appropriate, the amount of moles of the dialkyl oxalate ester used in the process is about equal to half the amount of moles of glycol derived end groups present in the starting polyester (co)polymer. In practice, due to the volatility of the dialkyl oxalate ester, typically a larger amount of the dialkyl oxalate is added to allow more alcohol end groups to react. The dialkyl oxalate ester may be added either in portions or all at once, depending on the circumstances and the desired product.

[0027] The molecular weight booster of the present invention is based on alkyl diesters of oxalate (the oxalate forms two alkyl-ester bonds), wherein the alkyl group is an aliphatic alkyl group wherein the carbon in the alkyl group that is attached to the oxalate is a primary carbon. Herein, a primary carbon means that the carbon is attached to one carbon atom only or to zero carbon atoms (in the case of DMO). In other words, the primary carbon in the dialkyl oxalate ester is a CHj group that is attached to an oxygen atom in the ester bond and to either a third hydrogen (DMO) or to a carbon atom such as in diethyl oxalate ester or dipropyl oxalate ester. In one embodiment, the alkyl chain is substantially linear, which herein means that counting from the ester group, at least the first three, preferably at least the first four, more preferably at least the first five carbon atoms form a linear alkyl chain without alkyl side chains. In one embodiment, the alkyl group is an aliphatic linear alkyl group without any branching. Examples of such alkyl groups include methyl, ethyl, propyl, butyl and pentyl groups. Especially good results were obtained with dimethyl oxalate (DMO) and diethyl oxalate (DEO).

[0028] Dialkyl oxalate esters possess a high reactivity towards alcohol groups, even to such an extent that in the process an additional catalyst may not be needed. The reactivity comes from the good leaving group ability of the alkyl groups and the volatility of the alkanol condensation products (in case of DMO and DEO: methanol and ethanol) in combination with the structure of oxalate where the carboxyl groups are directly connected to each other.

[0029] Without being bound to any theory, it is hypothesized that in the course of the process of the invention an oxalate ring product is formed from an intermediate oxalate species and the 1,2- diol derived end groups, for example glycol-derived or other suitable diol derived units that were originally present in the starting polyester (co)polymer. It is believed that chain ends couple, leading to an increase of the molecular weight of the polyester, and the oxalate ring product is removed at elevated temperature and low pressures, preferably so-called vacuum, while further removing the related alkanol, more especially methanol or ethanol, that is also released duringthe process. Furthermore, in the course of the reaction, it is possible that also other suitable diol derived units within an intermediate polymer product react with an intermediate oxalate species to form an oxalate ring product that can be removed at elevated temperature and vacuum thereby boosting the molecular weight. In such case, reactions between chains take place. While the oxalate ring product and alkanol are removed from the process, high molecular weight polymers are formed without incorporating a substantial amount of oxalate derived units in the polymer chain. Only minor, hardly detectable or indetectable, amounts of oxalate derived units were found to have remained in the polymer chain of the product.

[0030] For instance, when using DEO as molecular weight booster, an intermediate oxalate species is thought to be formed, that is possibly a part of the polymer chain, which species forms a six membered ring with the 1,2-diol derived units. Thus, for example, in the case of adding DEO to a polymer with ethylene glycol derived end groups, an ethylene oxalate ring is formed (1,4- dioxane-2, 3-dione). The produced ethylene oxalate and ethanol can be removed from the polymer at elevated temperature and reduced pressure, while leaving behind a high molecular weight polymer without any substantial residue of the molecular weight booster in the chain.

[0031] Advantageously, no or essentially no residue of units of the molecular weight booster remains in the polymer chain. Specifically, the polyester (co)polymer contains less than 1 mol% oxalate units relative to the total amount of monomer units in the polyester. Such amount does not alter the physical and chemical properties of the polymer produced at the end of the process. This is a significant advantage, especially when a polymer product is subjected to repeated recycling. Consequently, the polymer product properties are not changed by the current process, even after recycling. It is further advantageous for recycling processes to have no catalyst present in the polymer to avoid build-up of metals. Favorably, according to the process of the invention the use of a catalyst may be avoided.

[0032] A further advantage of the current process is that the cyclic oxalate species and the alkanol may suitably be recycled. Advantageously, building blocks like oxalate, ethylene glycol and ethanol or methanol may be re-introduced into the process for re-use.

[0033] The current process relates to production of a high molecular weight polyester (co)polymer, particularly wherein at least one of the units derived from a 1,2-diol is derived from an aliphatic diol selected from mono ethylene glycol, 1,2 propanediol, 1,2-butanediol, 2,3- butanediol and 1,2-cyclohexanediol.

[0034] Additional diols that may be used in the current process, other than the linear diols as defined herein above, may be any suitable diol, primary and secondary, and may preferably be selected from cis- and / or trans- 2,2,4,4-tetramethyl-l,3-cyclobutanediol, and 1, 4:3,6- dianhydrohexitols, in particular isosorbide. Thus, the starting polyester (co)polymer may comprise additional diol derived units, preferably selected from cis- and / or trans- 2,2,4,4-tetramethyl-l,3- cyclobutanediol, and l,4:3,6-dianhydrohexitols, in particular isosorbide.

[0035] Suitably, the starting polyester may be any (co)polymer comprising alcohol end groups and units derived from 1,2-diols. Preferably, the starting polyester (co)polymer comprises dicarboxylic acid derived units, the dicarboxylic acids or any esters thereof selected from aromatic dicarboxylic acids which can contain one or more heteroatoms or any esters thereof, and C2-C18 aliphatic dicarboxylic acids or any esters thereof, which may be linear, cyclic or branched dicarboxylic acids, such as, but not limited to, 1,4-cyclohexanedicarboxylic acid, diglycolic acid, and especially linear dicarboxylic acids of the formula HOOC(CH2)nCOOH wherein n is an integer of 1 to 20. The aromatic dicarboxylic acid preferably is selected from terephthalic acid, a terephthalicacid monoester, a terephthalic acid diester, a furandicarboxylic acid, a furandicarboxylic acid monoester and a furandicarboxylic acid diester.

[0036] In a preferred embodiment, the starting polyester (co)polymer contains dicarboxylic acid units derived from terephthalic acid and / or a furandicarboxylic acid (in particular 2,5- furandicarboxylic acid) and mono ethylene glycol derived units, and optionally isosorbide derived units. The composition of the starting polyester (co)polymer depends on the desired properties of the final polyester (co)polymer product. More preferably, the starting polyester (co)polymer consists of dicarboxylic acid units derived from terephthalic acid and / or 2,5- furandicarboxylic acid and mono ethylene glycol derived units, and optionally isosorbide derived units.

[0037] The process of the invention is preferably performed as follows, comprising (i) providing or producing the starting polyester (co)polymer, wherein the process for producing the polyester (co)polymer comprises esterification or transesterification, and polycondensation; (ii) adding the dialkyl oxalate ester to the starting polyester (co)polymer; (iii) elevating the temperature of the mixture resulting from step (ii) to at least 220 °C, or, in case the temperature is already higher than 220 °C, maintaining the temperature for a period of time; (iv) reducing pressure, preferably to at most 5 mbar, more preferably to equal to or below 1 mbar, thereby removing at least the alkanol (coming from the dialkyl oxalate ester) and producing a (co)polymer product; and (v) optionally repeating steps (ii) to (iv) one or more times with as starting material the (co)polymer with the higher molecular weight of the previous step (iv) until a final polyester (co)polymer product is produced wherein no or less than 1 mole% oxalate units are present, the percentage being relative to the total amount of monomer units, and with the desired high molecular weight properties.

[0038] A preferred step (i) comprises producing the starting polyester (co)polymer by reacting a dicarboxylic acid or ester thereof with a diol and / or polyol and / or oligomer with diol derived end groups such as bis(2-hydroxyethyl) terephthalate (BHET), and other terephthalate oligomers such as PET glycolysis products, see e.g. T. Spychaj in "Handbook of thermoplastic polymers", 2002 Wiley, Chapter 27, p 1259-61, optionally in the presence of a catalyst, wherein at least one of the diols and / or polyols and / or oligomers with diol derived end groups comprises a vicinal diol group. Preferably, the diol and / or polyol and / or oligomer with diol derived end groups is selected from a linear diol selected from saturated C2-C12 aliphatic diol compounds, and BHET, and preferably from mono ethylene glycol, 2,3-butanediol and 1,2-cyclohexanediol and at least comprises adjacent hydroxy groups in other words vicinal diols.

[0039] In a preferred embodiment of step (i), the dicarboxylic acid or ester thereof is reacted with an excess of the diol and / or polyol and / or oligomer with diol derived end groups. The use of an excess increases the relative amount of alcohol end groups in the starting polyester (co)polymer and decreases the relative amount of carboxylic end groups. The excess may range between 1.005 and 1.50 equivalents of diol relative to 1 equivalent of the dicarboxylic acid or ester groups, for example in the range between 1.01 and 1.10 or between 1.05 and 1.40, for example around 1.25 equivalents. A too low amount of diol results in an increased number of carboxylic acid or its ester end groups, whereas a too high amount of diols may result in a starting polyester (co)polymer with a very low molecular weight.

[0040] As indicated, the starting polyester (co)polymer in step (a) of the process may be readily prepared for this purpose in step (i), as described herein above, but may also advantageously be a recycled polymeric material or derived therefrom. Preferably, such polyester (co)polymer fromrecycling sources contains an aliphatic 1,2-diol derived unit, the diol selected from mono ethylene glycol, 1,2-propanediol, 1,2-butanediol, 2,3-butanediol and 1,2-cyclohexanediol.

[0041] The temperature in step (iii) depends on the type of the polyester (co)polymer that is produced. Suitably, the temperature is high enough when after addition of the dialkyl oxalate ester, the material melts or is still in a molten state and the mixture can properly be stirred. For more flexible polyester (co)polymers with low melting points, the temperature may for example be at least 220 °C, for more rigid polyester (co)polymers, such as PET or PEF polymers, the temperature preferably needs to be at least 230 °C,

[0042] In a scaled up process (kilogram scale), the molecular weight booster may advantageously be added at the end of the polycondensation (see step (ii) above) or in an extruder, together with the (co)polymer that is to be extruded.

[0043] The molecular weight boosting strategy of the present invention was shown to be relatively easy to scale up. The synthesis of dialkyl oxalate esters such as, for example, DMO and DEO may be performed in methods known to the skilled worker. In addition, DMO and DEO are commercially available chemicals. Advantageously, only a relatively small amount of DEO is required to have the desired molecular weight boosting effect. For example, for larger scale production of PEF only around 20 to 100g of DEO per kg of PEF may be needed. This makes this strategy economically attractive, but also easy to apply in existing reactor setups. The dialkyl oxalate ester can for example be fed to the reaction vessel via a catalyst addition funnel. Further, by increasing the polycondensation time or adding a catalyst, the required amount of the dialkyl oxalate ester can be reduced.

[0044] In a ton scale polymer production, the result may still be the production of kilograms of, for example, ethanol (in case DEO is used) and ethylene oxalate ring as condensates. To make the process more economic and sustainable, and commercially attractive, the condensates should have a designated use. Ethanol could be recycled and reused in the DEO synthesis process. Ethylene oxalate may also be reused for DEO synthesis as well, however it is probably more efficient to use it directly as a monomer. Ethylene oxalate may for example be used to produce polyethylene oxalate by ring opening polymerization, or it may be combined with other ring opening monomers, such as lactide and glycolide to produce PLA and PGA polymers comprising oxalate units.

[0045] In a further aspect, the invention relates to (co)polyesters obtainable by, or obtained by, the currently claimed process. The process allows the preparation of a range of existing and novel polyester (co)polymers with high molecular weights that conventionally would not be obtainable. In an embodiment of the invention, the process comprises a step (i) wherein the starting polyester (co)polymer was produced using a catalyst, followed by steps (ii) to (v). Such a process combines the favorable features of the present invention and features of existing polyester production methods. In another preferred embodiment no catalyst is used in the entire process, allowing the production of metal catalyst free polyester (co)polymers, that may be advantageous for certain uses requiring the absence of any catalyst, such as medical uses of polyesters.

[0046] Metal catalyst free polyester is considered to be polyester comprising less than 100 ppm, preferably less than 10 ppm, preferably less than 1 ppm by weight of metals based on total amount of polyester.

[0047] Thus, in an embodiment, preferably the polyester that is produced is a metal catalyst free polyester (co)polymer, which was produced without addition of a metal catalyst, and preferably with Mn of 20kDa or more, the polyester being selected from:- poly(ethylene furan-2,5-dicarboxylate);- poly(ethylene co-isosorbide furan-2,5-dicarboxylate);- poly(ethylene co-isosorbide co-cyclohexanedimethylene furan-2,5-dicarboxylate);- poly(ethylene co-isosorbide terephthalate) with isosorbide content of 15% or higher;- poly(ethylene co-isosorbide co-cyclohexanedimethylene terephthalate);- poly(ethylene furan-2,5-dicarboxylate co-terephthalate);- poly(ethylene co-isosorbide furan-2,5-dicarboxylate co-terephthalate); wherein the Mn is measured using gel permeation chromatography with poly(methyl methacrylate) standards as reference material.

[0048] In a preferred embodiment, the polyester (co)polymer product that is produced with the process according to the invention has a molecular weight (Mn) of 20 kDa or more, preferably 30 kDa or more, even more preferably 40 kDa or more.

[0049] The weight average molecular weight (Mw) and the number average molecular weight (Mn) can be determined by means of gel permeation chromatography (GPC) at 35 °C, using for the calculation poly(methyl methacrylate) standards as reference material, and using hexafluoro- 2-propanol as eluent.

[0050] The glass transition temperature (Tg) of the polyester copolymer can be measured by conventional methods, in particular by using differential scanning calorimetry (DSC) with a heating rate of 10 °C / minute in a nitrogen atmosphere. All glass transition temperatures herein are determined as described under the analytical methods section of the examples. In a preferred embodiment, the polyester (co)polymer that is obtainable with the process according to the invention has a Tg of at least 80 °C, preferably at least 90 °C, more preferably at least 100 °C.

[0051] The polyester (co)polymer obtainable by or obtained by the process of the invention can suitably be combined with additives and / or other (co)polymers and therefore the invention further provides a composition comprising said polyester copolymer and in addition one or more additives and / or one or more additional other (co)polymers. Such composition can for example comprise, as additive, nucleating agents. These nucleating agents can be organic or inorganic in nature. Examples of nucleating agents are talc, calcium silicate, sodium benzoate, calcium titanate, boron nitride, zinc salts, porphyrins, chlorin and fluorine.

[0052] The composition according to the invention can also comprise, as additive, nanometric (i.e. having particles of a nanometric size) or non-nanometric and functionalized or nonfunctionalized fillers or fibres of organic or inorganic nature. They can be silicas, zeolites, glass fibres or beads, clays, mica, titanates, silicates, graphite, calcium carbonate, carbon nanotubes, wood fibres, carbon fibres, polymer fibres, proteins, cellulose fibres, lignocellulose fibres and nondestructured granular starch. These fillers or fibres can make it possible to improve the hardness, the stiffness or the permeability to water or to gases. The composition can comprise from 0.1% to 75% by weight, for example from 0.5% to 50% by weight, of fillers and / or fibres, with respect to the total weight of the composition. The composition can also be of composite type, that is to say can comprise large amounts of these fillers and / or fibres.

[0053] The composition can also comprise, as additive, opacifying agents, dyes and pigments. They can be chosen from cobalt acetate and the following compounds: HS-325 Sandoplast Red BB, which is a compound carrying an azo functional group also known under the name SolventRed 195, HS-510 Sandoplast Blue 2B, which is an anthraquinone, Polysynthren Blue R and Clariant RSB Violet. Sandoplast and Clariant are trademarks of Clariant. Polysynthren is a trademark of the Heubach group.

[0054] The composition can also comprise, as additive, a processing aid for reducing the pressure in the processing device. A mould-release agent, which makes it possible to reduce the adhesion to the equipment for shaping the polyester, such as the moulds or the rollers of calendering devices, can also be used. These agents can be selected from fatty acid esters and amides, metal salts, soaps, paraffins or hydrocarbon waxes. Specific examples of these agents are zinc stearate, calcium stearate, aluminium stearate, stearamide, erucamide, behenamide, beeswax or Candelilla wax.

[0055] The composition can also comprise other additives, such as stabilizers, etc. as mentioned herein above.

[0056] In addition, the composition can comprise one or more additional polymers other than the one or more polyester (co)polymers according to the invention. Such additional polymer(s) can suitably be chosen from the group consisting of polyamides, polystyrene, styrene copolymers, styrene / acrylonitrile copolymers, styrene / acrylonitrile / butadiene copolymers, polymethyl methacrylates, acrylic copolymers, poly(ether / imide)s, polyphenylene oxides, such as poly(2,6- dimethylphenylene oxide), polyphenylene sulfide, poly(ester / carbonate)s, polycarbonates, polysulphones, polysulphone ethers, polyetherketones and blends of these polymers.

[0057] The composition can also comprise, as additional polymer, a polymer which makes it possible to improve the impact properties of the polymer, in particular functional polyolefins, such as functionalized polymers and copolymers of ethylene or propylene, core / shell copolymers or block copolymers.

[0058] The compositions according to the invention can also comprise, as additional polymer(s), polymers of natural origin, such as starch, cellulose, chitosan, alginates, proteins, such as gluten, pea proteins, casein, collagen, gelatine or lignin, it being possible or not for these polymers of natural origin to be physically or chemically modified. The starch can be used in the destructured or plasticized form. In the latter case, the plasticizer can be water or a polyol, in particular glycerol, polyglycerol, isosorbide, sorbitans, sorbitol, mannitol or also urea. Use may in particular be made, in order to prepare the composition, of the process described in the document WO 2010 / 010282A1.

[0059] These compositions can suitably be manufactured by conventional methods for the conversion of thermoplastics. These conventional methods may comprise at least one stage of melt or softened blending of the polymers and one stage of recovery of the composition. Such blending can for example be carried out in internal blade or rotor mixers, an external mixer, or single-screw or co-rotating or counter-rotating twin-screw extruders. However, it is preferred to carry out this blending by extrusion, in particular by using a co-rotating extruder. The blending of the constituents of the composition can suitably be carried out at a temperature ranging from 220 to 300°C, preferably under an inert atmosphere. In the case of an extruder, the various constituents of the composition can suitably be introduced using introduction hoppers located along the extruder.

[0060] The invention also relates to an article comprising a polyester (co)polymer according to the invention or a composition comprising a polyester (co)polymer according to the invention and one or more additives and / or additional polymers. The polyester (co)polymer may conveniently be used in the manufacturing of films, fibers, injection molded parts and packaging materials, suchas for example receptacles. The use of the polyester (co)polymer is especially advantageous where such films, fibers, injection molded parts or packaging materials need to be heat-resistant or coldresistant.

[0061] The article can also be a fibre for use in for example the textile industry. These fibres can be woven, in order to form fabrics, or also nonwoven. The article can also be a film or a sheet. These films or sheets can be manufactured by calendering, cast film extrusion or film blowing extrusion techniques. These films can be used for the manufacture of labels or insulators.

[0062] This article can be a receptacle especially for use for hot filling and reuse applications. This article can be manufactured from the polyester (co)polymer or a composition comprising a polyester (co)polymer and one or more additives and / or additional polymers using conventional conversion techniques. The article can also be a receptacle for transporting gases, liquids and / or solids. The receptacles concerned may be baby's bottles, flasks, bottles, for example sparkling or still water bottles, juice bottles, soda bottles, carboys, alcoholic drink bottles, medicine bottles or bottles for cosmetic products, dishes, for example for ready-made meals or microwave dishes, or also lids. These receptacles can be of any size. The article may for example be suitably manufactured by extrusion-blow moulding, thermoforming or injection-blow moulding.

[0063] The present invention therefore also conveniently provides a method for manufacturing an article, comprising the use of one or more polyester (co)polymers according to the invention and preferably comprising the following steps: 1) the provision of a polyester (co)polymer obtainable by or obtained by the process of this invention; 2) melting said polyester (co)polymer, and optionally one or more additives and / or one or more additional polymers, to thereby produce a polymer melt; and 3) extrusion-blow moulding, thermoforming and / or injection-blow moulding the polymer melt into the article.

[0064] The article can also be manufactured according to a process comprising a stage of application of a layer of polyester in the molten state to a layer based on organic polymer, on metal or on adhesive composition in the solid state. This stage can be carried out by pressing, over moulding, lamination, extrusion-lamination, coating or extrusion-coating.

[0065] Advantageously, high molecular weight polyester (co)polymers produced according to the process of the invention can be used in 3D printing. In case very high molecular weights are desired, the use of alternative types of reactors could potentially be a solution, such as extruders and compounders that are known to be used with polymers produced with several types of chain extenders. For example, a spinning disk reactor is highly suitable for processing highly viscous polymers.

[0066] The invention is further illustrated by the following non-limiting examples.Examples

[0067] List of abbreviationsMonomersEG ethylene glycolFDCA 2,5-furandicarboxylic acidISO isosorbideBoostersDEO diethyl oxalateDMO dimethyl oxalateDtBO di-tert-butyl oxalateOA oxalic acidChemicalsTEA hydroxide tetra ethyl ammonium hydroxideTCE-d2 deuterated tetrachloroethaneTFA-dl deuterated trifluoroacetic acidPolymersPEF polyethylene furanoate) PEIF poly(ethylene-co-isosorbide furanoate) PET poly(ethylene terephthalate) PEIT poly(ethylene-co-isosorbide terephthalate)Analysis methodsNMR Nuclear Magnetic Resonance DSC Differential Scanning Chromatography GPC Gel Permeation ChromatographyGeneral terms eq- equivalent h hourRH relative humidity RT room temperature TE / PC transesterification or polycondensationMaterials and reagents

[0068] Triethylamine (99.5%), TFA-dl (99.5%), diethyl oxalate (>99%), di-tert-butyl oxalate (99%), oxalic acid (98%), ethylene glycol (99.5%), Ti(IV)isopropoxide (97%), and tetraethylammonium hydroxide 35% w / w aq. soln, were supplied by Sigma Aldrich. Isosorbide (>99.5%) was supplied by Roquette. Dimethyl oxalate (>99%) was bought from TCI chemicals. TCE- d2 (99.5%) was ordered from Thermo Scientific. 2,5-Furandicarboxylic acid (99%, polymer grade) was supplied by Avantium, RNP. All chemicals were used as received.CharacterizationNMR

[0069] 1H-NMR spectra were recorded at appropriate frequencies on a Bruker AV300-II (1H, 300.10 MHz) spectrometer with a 5 mm BBOF ATMA probe, and a Bruker AV400 (1H, 400.13 MHz) spectrometer with a 5 mm BBO ATMA probe. Chemicals shift are referenced to the residual proton in the specified solvent.XH NMR sample preparation: ~7 mg of polymer was dissolved in 0.6 mL TCE-d2 or TFA-dl.DSC

[0070] Differential scanning calorimetry thermograms were acquired using a Mettler Toledo DSC 3 STAResystem. Approximately 5 mg of the sample was measured in a conventional aluminum crucible (40 pl). The analysis was conducted in three stages, employing a Nj flow of 50 ml*min-1. Initially, the sample was stabilized at 20 °C for 5 minutes and then subjected to analysis, with a temperature increase from 20 to 250 °C at a rate of 10oC*min-1. Subsequently, the sample was cooled down to the initial temperature of 20 °C at a cooling rate of 50oC*min-1. Finally, the initial step was repeated, and the Tsvalue from the second cycle is used for reporting.GPC

[0071] GPC analysis was conducted using a Hitachi Chromaster 5450 equipped with an Agilent HPLC system featuring two PFG 7 micrometer (pm) Linear M (300x7.5 mm) columns. PMMA standards were used for calibration. Each sample, comprising 10 mg, was dissolved in 10 mL HFIP. The mobile phase, a combination of HFIP and potassium trifluoroacetate (0.02 mol) to prevent bacterial growth in the HFIP, maintained a consistent flow rate of 1 mL / min at a temperature of 35 °C. Analysis was facilitated with the refractive index detector (Chromaster 5450), and subsequent data processing was accomplished using ASTRA 6.1 software (Wyatt Technology).Compression molding films

[0072] Barrier measurement films were crafted through compression molding using a thermal press (Carver Auto Four / 3015-NE,H). Polymer granulates were dried in a vacuum oven for two days at 60 °C (PEF) and 85 °C (PEIF) at <1 mbar. An aluminum foil circular press shape (3 layers, 010 cm, "'0.75 mm thickness) was pre-pressed (10 Force metric tons) in between two Teflon sheets (20*20*0.14 cm) and two aluminum plates (20*20*1 cm). The sandwich was opened and 1.2 g of material was carefully placed in the center of the press shape. The polymer was pre-molten by positioning the sandwich in the hot press at 245°C (PEF) and 255°C (PEIF) for a maximum of 1.5 minutes without applying pressure. Subsequently, the sandwich was pressed at 0.5 tons for 1 minute, 1 tons for 30 seconds, 2 tons for 30 seconds, 5 tons for 30 seconds, and 10 tons for 30 seconds. The sandwich was then removed from the press. The Teflon sheets with the polymer and press shape was separated from the sandwich and left to cool at a flat, cold surface. Once cooled, the pressing shape and Teflon sheets were removed, resulting in films with a thickness of approximately 85 pm. Multiple films were produced and visually inspected for bubbles and defects, with only the highest quality films selected for subsequent barrier measurements.Barrier measurements

[0073] The barrier measurements were performed on a Totalperm (Permtech s.r.l) instrument. The water barrier measurements were measured according to ASTM E96 / E96M-15 standards. The endpoint was reached when equilibrium was obtained with a tolerance level of 0.5%.

[0074] For both PEF and PEIF the water vapor transmission rate (WVTR) measurements were performed at 38 °C with 90% relative humidity (RH). For PEIF the carbon dioxide transmission rate (COTR) measurements were performed at 23 °C with 0% RH. The oxygen transmission rate (OTR) measurements were performed at 23 °C with 0% + 65% RH for PEF and for PEIF at 23 °C and 30 °C with 0% + 50% RH.Injection molding

[0075] Tensile bars were prepared using a Thermo Scientific HAAKE Minijet II apparatus equipped with an ISO-527-2-A5 mold. The apparatus was configured with specific parameters, including cylinder temperature (255 °C PEF, 265 °C PEIF), and mold temperature (50 °C). The air compressor, Stanley Fatmax (2. OHP, 5Lt, lObar), was used for each run to press down the piston into the melting chamber to move the molten polymer into the mold (950 bar, 6 seconds). Next, a rest phase allowed the polymer to expand if needed (300 bar, 6 seconds). Before each run, thorough attention was given to ensuring the cleanliness of the cylinder, eliminating remnants from prior operations. To facilitate the molding process, the mold was coated with a water-based silicon mold release agent, allowed to dry, and then placed in the holder. Approximately 2.5-2.7 grams ofpolymer were weighed and transferred into the cylinder. The polymer underwent a melting phase within the cylinder for a maximum duration of 2 minutes. Subsequently, the cylinder, containing the molten polymer, was positioned atop the mold, the door securely closed, and the injection program initiated. Upon completion of the injection process, the mold was disassembled, allowing for the careful removal of the sample from the mold. The samples were examined for any potential defects such as bubbles or fractures. Any residual polymer was discharged, and the finalized samples were used for further analysis.Tensile testing

[0076] The tensile bars were analyzed using an Instron 5565 machine, featuring a lkN load cell and an Instron strain gauge extensometer. The sample dimensions were set to a width of 4.0 mm, thickness of 1.91 mm and parallel length of 25 mm. The test speed used was 5 mm / min. Upon reaching the maximum elongation of the extensometer (100%, 20 mm), the extension frame was used to ascertain the elongation at the point of breakage.PEFEXAMPLE 1: Production of PEF starting polyester without a metal catalyst

[0077] A 2 L autoclave was charged with FDCA (624.4 g, 4.0 mol, 1.0 eq.), EG (310.4 g, 5.0 mol, 1.25 eq.) and the ether suppressant TEA hydroxide 35% w / w aq. soln. (0.43 mL, 200 ppm). The oil temperature was set to 220-230 °C with a Nj bleed of 2 L / h. The set temperature was reached in about 15 min. (internal ~215 °C). As soon as the internal temperature reached 175 °C, stir speed was set to 125 RPM (anchor stirrer). After 3 h of transesterification the Nj flow was halted. At this stage 92.7 g of condensate was collected. Next, vacuum was applied and the oil temperature was set to 260-265 °C. Full vacuum (<1 mbar) was reached in about 30 min. and maintained for 1 h. The colorless to slightly yellowish PEF starting polyester was collected on a plate covered with Teflon sheets (707.7 g, 82% yield).

[0078] As can be seen in Table 1, GPC analysis provided Mn, Mwand PDI values of 7.1 kDa, 15.0 kDa and 2.1, respectively for the PEF starting polyester from Example 1. DSC analysis provided a Tgof 70.0 °C and1H-NMR analysis shows the presence of 13.6% EG-end groups, the percentage being relative to total amount of FDCA-derived monomer units, which is attributed to the excess use of EG in the reaction.

[0079] The PEF starting polyester were used as the starting polyester in the following examples for the processes according to the invention for the production of a polyester.EXAMPLE 2: 2ndstage booster reactions PEF without a metal catalyst

[0080] The PEF starting polyester of Example 1 (20.0 g, 105.7 mmol, 1.0 eq.) was reacted with the respective booster (6.5 mmol, 6.1 mol%, relative to the total amount of FDCA-derived monomer units in the starting polyester) at an oil temperature set to 260 °C and a Nj bleed of 50 mL / min. PEF starting polyester from the same batch obtained in Example 1 was used for all reactions in Example 2. First the mixture was melted for 2 min., and afterwards, the mixture was stirred for 5 min. at 100 RPM. Next, vacuum was applied, rapidly reaching 400 mbar and subsequently dropping to <1 mbar in 15 min. Samples were taken for GPC every 10, 30 and 60 minutes. After 1 h at full vacuum, the polymer was taken out of the flask under a continuous flow of Nj.

[0081] A blank reaction was conducted without booster, where the PEF starting polyester were reacted for 60 minutes under full vacuum. The properties of the resulting PEF material are shown in Table 1. Comparing these results with the PEF starting polyester of Example 1, an increase of 7.2 kDa and 7.9 °C is observed in Mnand Ts. The EG-end content reduced with 6.6% (relative to amount of FDCA-derived monomer units). The blank reaction sets a benchmark for comparison with subsequent booster reactions.

[0082] As can be seen in Table 1, the use of acid boosters (FDCA or OA) yielded Mn values similar to the blank reaction, while oxalate boosters di-ethyl oxalate and dimethyl oxalate reveal Mnvalues of 20.2 and 21.0 kDa, which is an increase of more than 13.1 kDa compared to the blank reaction. The slightly superior GPC results for DM0 may be attributed to its less sterically hindered methyl groups, enhancing the kinetic favorability of the transesterification step. The DM0 and DEO booster reactions result in increased Tsvalues of 82.6 °C and 82.4 °C, and decreased EG-end contents of 3.5% and 3.6%. Importantly,1H NMR indicated an absence of ethyl / methyl-end groups in the obtained PEF, thereby revealing that all booster has reacted with fully, i.e. there is no unreacted booster left, neither are there alkyl end groups in the resulting polyester, which would indicate that the booster molecule has only reacted with one alcohol end group.

[0083] The DtBO oxalate booster, being a branched dialkyl oxalate ester, distinctly exhibits the lowest reactivity among the oxalate boosters, resulting in an Mnof only 10.4 kDa - 3.9 kDa lower than the blank reaction. This outcome is likely tied to the steric hindrance introduced by the tertbutyl groups, which impedes the transesterification.Table 1. Overview GPC(a|, DSC(b|and1H NMR(c|data including the Mn, Mw, PDI and Tsvalues (average of 2 spectra) & the corrected quantity (%) of the EG-end contents, relative to the total amount of FDCA-derived monomer units, of the PEF starting polyester (Example 1), blank reaction and booster reactions (Example 2)Example 3: PEF optimized with DEO (18.0 mol%), 50 gram scale without metal catalyst

[0084] The PEF starting polyester of Example 1 (50.0 g, 264 mmol, 1.0 eq.) was reacted with DEO (3.1 g, 20.9 mmol, 6.8 mol%, relative to the total amount of FDCA-derived monomer units in the starting polyester) at an oil temperature set to 260 °C and a Nj bleed of 50 mL / min. First the mixture was melted for 2 min., and afterwards, the mixture was stirred for 5 min. at 100 RPM. Next, vacuum was applied, rapidly reaching 400 mbar and subsequently dropping to <1 mbar in 15 min. After 1 h at full vacuum, additional DEO was added six more times: 2.3, 0.68, 0.50, 0.51, 0.50 and 0.54 g (total 5.0 g, 39 mmol, 11.2 mol%, relative to the total amount of FDCA-derived monomer units in the starting polyester). After each addition, the reaction was stirred 3 min. under a continuous Nj flow, followed by a 30-minute vacuum application, before proceeding withthe next addition. The reaction was monitored with1H NMR to verify the quantity of EG-end groups. The final torque was 20.9 Ncm at 30 RPM, as measured with an IKA Lab stirrer MINISTAR 80 control. The skilled person knows that the measured torque is the power that is required to rotate / stir at a given RPM (rounds per minute) for a given stirrer apparatus. Next, the brown polymer was taken out of the flask under a continuous flow of Nj.Example 4: PEF optimized with DEO (11.6 mol%), 50 gram scale without metal catalyst

[0085] A mixture of the PEF starting polyester (50.0 g, 264 mmol, 1.0 eq.) and DEO (4.1 g, 27.8 mmol, 9.0 mol%, relative to the total amount of FDCA-derived monomer units in the starting polyester) was slowly heated from RT to 260 °C at a Nj bleed of 30 mL / min. The mixture started melting at 180 °C and was slowly stirred at 30 RPM. After heating for 30 min. the oil setpoint of 260 °C was reached and the mixture was fully melted and stirred at 100 RPM. Next, vacuum was applied, rapidly reaching 400 mbar and subsequently dropping to <1 mbar in 15 min. After 1.5 h at full vacuum, additional DEO was added two more times: 0.94 and 0.24 g (total 1.2 g, 81 mmol, 2.6 mol%, relative to the total amount of FDCA-derived monomer units in the starting polyester). After each addition, the reaction was stirred 3 minutes under a continuous Nj flow of 50 mL / min, followed by a 2-hour vacuum application, before proceeding with the next addition. The reaction was monitored with1H NMR to verify the quantity of EG-end groups. The final torque was 18 Ncm at 30 RPM, as measured with an IKA Lab stirrer MINISTAR 80 control. Next, the brown polymer was taken out of the flask under a continuous flow of Nj.Example 5: PEF optimized with DEO (7.2 mol%), 50 gram scale with 25 ppm Ti catalyst

[0086] A mixture of PEF starting polyester (50.0 g, 264 mmol, 1.0 eq.), DEO (3.1 g, 20.9 mmol, 6.8 mol%, relative to the total amount of FDCA-derived monomer units in the starting polyester) and Ti(IV)isopropoxide (7.4 mg, 26.1 pmol, 25 ppm Ti) was slowly heated from RT to 260 °C at a Nj bleed of 30 mL / min. The mixture started melting at 180 °C and was slowly stirred at 30 RPM. After heating for 20 min. the oil setpoint of 260 °C was reached and the mixture was fully melted and stirred at 100 RPM. Next, vacuum was applied, rapidly reaching 400 mbar and subsequently dropping to <1 mbar in 15 min. After 2 h at full vacuum, additional DEO was added one more time: 170 mg (1.2 mmol, 0.4 mol%, relative to the total amount of FDCA-derived monomer units in the starting polyester). After the addition, the reaction was stirred 3 min. under a continuous Njflow of 50 mL / min, followed by a 2-hour vacuum application. The reaction was monitored with1H NMR to verify the quantity of EG-end groups. The final torque was 22.1 Ncm at 30 RPM, as measured with an IKA Lab stirrer MINISTAR 80 control. Next, the dark brown polymer was taken out of the flask under a continuous flow of Nj.Example 6: PEF optimized with DMO (16.2 mol%), 50 gram scale without metal catalyst

[0087] A mixture of the PEF starting polyester (50.0 g, 264 mmol, 1.0 eq.) and DMO (4.0 g, 34.0 mmol, 11.0 mol%, relative to the total amount of FDCA-derived monomer units in the starting polyester) was slowly heated from RT to 260 °C at a Nj bleed of 30 mL / min. The mixture started melting at 180 °C and was slowly stirred at 30 RPM. After heating for 30 min. the oil setpoint of 260 °C was reached and the mixture was fully melted and stirred at 100 RPM. Next, vacuum was applied, rapidly reaching 400 mbar and subsequently dropping to <1 mbar in 15 min. After 2.5 h at full vacuum, additional DMO was added two more times: 1.8 and 0.07 g (total 1.9 g, 16 mmol, 5.2 mol%, relative to the total amount of FDCA-derived monomer units in the starting polyester).After each addition, the reaction was stirred 3 minutes under a continuous Nj flow of 50 mL / min, followed by a 2-hour vacuum application, before proceeding with the next addition. The reaction was monitored with1H NMR to verify the quantity of EG-end groups. The final torque was 17 Ncm at 30 RPM, as measured with an IKA Lab stirrer MINISTAR 80 control. Next, the brown polymer was taken out of the flask under a continuous flow of Nj.

[0088] Table 2 shows the properties of the resulting PEF polymers obtained in Examples 3-6.Table 2. Overview GPC(a|, DSC(b|and1H NMR(c|data including the Mn, Mw, PDI and Tsvalues (average of 2 spectra) & the corrected quantity (%) of the EG-end contents of the PEF polymers obtained in Examples 3-6.

[0089] Two optimization reactions without metal catalyst were conducted on a 50 gram scale incorporating PEF starting polyester from the autoclave production of Example 1 with distinct quantities of DEO (Examples 3 and 4). The primary distinction between the two reactions lies in the initial amount of DEO used, and the temperature at start. It was discovered that employing a higher initial mol% DEO, combined with starting at RT, substantially reduced the need for subsequent titrations. Yet, this process demands a nuanced approach as adding too much DEO risks a 1:1 reaction with EG-end groups, instead of the desired 1:2 ratio. Conversely, as DEO evaporates, adding too little necessitates multiple titrations leading to prolonged reaction times and potential risks of coloration and degradation.

[0090] In Example 3 (PEF optimization with 18.0 mol% DEO), an initial 6.8 mol% DEO was employed, and the reaction commenced directly at 260 °C. On the contrary, in Example 4 (PEF optimization involving 11.6 mol% DEO), the initial amount of DEO was increased to 9.0 mol%, and the reaction initiated at room temperature, reaching 260 °C in 30 minutes. As a result, the first scenario necessitated six additional titrations and a total reaction time of 6 hours, while the second approach achieved comparable results with only two titrations and a shorter reaction time of one hour.

[0091] The outcomes from the two DEO boosted PEF optimization reactions (Examples 3 and 4) reveal marginal differences. A slight improvement of the 11.6 mol% DEO boosted PEF (Example 4) was observed, yielding an Mnvalue of 44.9 kDa, which is 4.2 kDa higher than the 18 mol% DEO boosted PEF (Example 3). The materials exhibit similar Tsvalues of 84.2 °C and 84.6 °C, along with total end (EG-end + ethyl-end) contents of 1.57 and 1.63.

[0092] If FDCA-EG-end groups react with oxalate ethyl-end groups, the outcome is the formation of a PEF unit and EtOH and l,4-dioxane-2, 3-dione. Subsequently, EtOH could react with this PEF- unit, leading to the formation of less reactive FDCA-ethyl esters. Without being bound to theory, the inventors believe that EG-end groups display lower reactivity towards FDCA-ethyl esters compared to oxalate ethyl-end groups. This heightened susceptibility of oxalates to nucleophilic attacks over esters could account for an observed lack of reactivity between the quantities ofethyl-end and EG-end groups. While, in theory, these end groups could continue to react with each other, prolonged reaction time did not lead to a reduction in the overall amount of end content.

[0093] Example 5, using 7.2 mol% DEO and a titanium catalyst revealed an increase in Mnof 5.9 kDa, reaching 50.8 kDa, compared to DEO optimized PEF without metal catalyst (Example 4). In contrast the Tsis almost similar at 84.6 °C.

[0094] Example 6 (PEF optimization with 16.2 mol% DMO) revealed no significant differences in PEF polymer characteristics compared to DEO.

[0095] The optimized PEF samples (Examples 4 and 5) were tested for their barrier and tensile properties which are shown in Table 3.Table 3. Barrier properties of optimized PEF obtained in examples 4 and 5: oxygen, carbon dioxide and water transmission rates and permeability at 85 pm thickness.

[0096] The optimized PEF samples were also tested for their mechanical properties. To examine the mechanical properties, the optimized PEF materials were processed into five tensile bars. The tensile strength, Young's modulus, stress at yield and elongation at break were analyzed for three samples per example. The Young's moduli for all PEF samples had mean values ranging between 3.5-3.7 GPa, an ultimate tensile strength ranging between 103-110 MPa and an extension at break ranging between 39-290%.PEIF

[0097] The efficacy of DEO as a boosting agent for achieving high molecular weight PEIF material was also explored. In PEIF, the secondary diol isosorbide is incorporated. Similar as for PEF, first a PEIF starting polyester was prepared, followed by reactions to optimize the molecular weight with DEO as boosting agent.Example 7: Production of PEIF starting polyester without a metal catalyst

[0098] FDCA (200.0 g, 1.3 mol, 1.0 eq.), ISO (60.3 g, 0.41 mol, 0.32 eq.), EG (73.0 g, 1.2 mol, 0.92 eq.) and TEA hydroxide 35% w / w aq. soln. (150 mg, 200 ppm) were charged to a 500 mL 3-neck flask equipped with an overhead stirrer, condenser, N2(50 mL / min) inlet and thermometer. Theoil temperature was set to 230 °C, which was reached in about 15 min. As soon as the oil temperature reached 190°C, stir speed was set to 125 RPM. After 3.5 h of transesterification a clear melt was obtained and the Nj flow was halted. At this stage 38.1 g of condensate was collected. Next, vacuum was applied and the oil temperature was set to 260 °C. Full vacuum (<1 mbar) was reached in about 30 min, and maintained for 2 h. The torque after polycondensation increased from 3.8 to 4.4 Ncm. The colorless to slightly yellowish PEIF starting polyester was collected on a plate covered with Teflon sheets (227 g, 1.1 mol, 97% yield).

[0099] This PEIF starting polyester was used as the starting polyester in the following examples for the processes according to the invention for the production of a polyester.Example 8: PEIF optimized with DEO (20.6 mol%), 20 gram scale without a metal catalyst

[0100] The PEIF starting polyester of Example 7 (20.0 g, 98.4 mmol, 1.0 eq.) was reacted with DEO (575 mg, 6.8 mmol, 4.0 mol%, relative to the total amount of FDCA-derived monomer units in the starting polyester) by an oil temperature set to 260 °C and a Nj bleed of 50 mL / min. First the mixture was melted for 2 min. and, and afterwards, the mixture was stirred for 5 min. at 100 RPM. Next, vacuum was applied, rapidly reaching 400 mbar and subsequently dropping to <1 mbar in 15 min. After 1 h at full vacuum, additional DEO was added fifteen more times: 123.9, 137.2, 162.5, 171.6, 157.6, 164.4, 255.4, 167.4, 140.0, 159.0, 152.0, 155.1, 153.0, 155.0 and 129.0 mg (total 2.4 g, 16.3 mmol, 16.6 mol%). After each addition, the reaction was stirred 3 min. under a continuous Nj flow, followed by a 30-minute vacuum application, before proceeding with the next addition. The reaction was monitored with1H NMR to verify the quantity of end groups. The final torque was 18.4 Ncm at 30 RPM. Next, the brown polymer was taken out of the flask in about 10 min. under a continuous flow of Nj.Example 9: PEIF optimized with DEO (10.7 mol%), 20 gram scale, with 50 ppm Ti catalyst

[0101] The PEIF starting polyester of Example 7 (20.0 g, 98.4 mmol, 1.0 eq.) was reacted with DEO (575 mg, 6.8 mmol, 4.0 mol%, relative to the total amount of FDCA-derived monomer units in the starting polyester) and Ti(IV)isopropoxide (5.9 mg, 20.8 pmol, 50 ppm Ti) by an oil temperature set to 260 °C and a Nj bleed of 50 mL / min. First the mixture was melted for 2 min., and afterwards, the mixture was stirred for 5 min. at 100 RPM. Next, vacuum was applied, rapidly reaching 400 mbar and subsequently dropping to <1 mbar in 15 min. After 1 h at full vacuum, additional DEO was added four more times: 133.4, 74.0, 72.0 and 84.2 mg (total 936 mg, 64.0 mmol, 6.7 mol%, relative to the total amount of FDCA-derived monomer units in the starting polyester). After each addition, the reaction was stirred 3 min. under a continuous Nj flow, followed by a 30-minute vacuum application, before proceeding with the next addition. The reaction was monitored with1H NMR to verify the quantity of end groups. The final torque was 18.4 Ncm at 30 RPM. Next, the brown polymer was taken out of the flask in about 10 min. under a continuous flow of Nj.Example 10: PEIF optimized with DEO (6.9 mol%), 50 gram scale, with 50 ppm Ti catalyst

[0102] The PEIF starting polyester (50.0 g, 246 mmol, 1.0 eq.) was reacted with DEO (2.48 g, 16.9 mmol, 6.9 mol%, relative to the total amount of FDCA-derived monomer units in the starting polyester) and Ti(IV)isopropoxide (14.8 g, 52.2 mmol, 50 ppm Ti) by an oil temperature set to 260 °C and a Nj bleed of 50 mL / min. First the mixture was melted for 2 min., and afterwards, the mixture was stirred for 5 min. at 100 RPM. Next, vacuum was applied, rapidly reaching 400 mbar and subsequently dropping to <1 mbar in 15 min. After 3 hours at full vacuum the reaction wascompleted with a final torque of 28 Ncm at 30 RPM. The dark brown polymer was taken out of the flask in about 15 min. under a continuous flow of Nj.Table 4. Overview GPC(a|and DSC(b|and1H NMR(c|data including the Mn, Mw, PDI and Tsvalues (average of 2 spectra) & the amount of isosorbide in the PEIF polymers obtained in Examples 7- 10.

[0103] Given the use of 0.32 eq. ISO during the production of the PEIF starting polyester in Example 7 , the theoretical maximum ISO content in the polymer was anticipated to be 32%. However, the1H NMR analysis showed the actual presence of 24.8% ISO as a repeating unit in the polymer, 5.7% ISO-endo-end groups, and an additional 0.9% free ISO. Consequently, the total ISO content in the polymer is 27.7%, with the total amount of ISO-end groups (ISO-endo-end + 2 * free ISO) calculated at 7.5%.

[0104] In Examples 8 and 9, a 20 gram scale reaction was carried out respectively with and without a Ti catalyst. In contrast to Example 8, only 6.7 mol% DEO was added in four incremental titrations in Example 9. Furthermore, a significant 21.1 kDa increase in Mnand 4.6 °C in Tgwas observed, compared to Example 8 without metal catalyst, reaching 41.9 kDa and 110.8 °C.

[0105] Example 10, featuring 50 ppm Ti from Ti(IV)isopropoxide, was conducted on a 50 gram scale. Following a series of test reactions, it was determined that an initial 6.9 mol% DEO was sufficient, eliminating the need for additional titrations, and therefore resulting in less degradation and coloration. Similar to the previous reaction, this reaction concluded after 3 hours. The Mnof the boosted PEIF material slightly increased with 1.3 kDa, reaching 43.2 kDa, compared to the 20 gram scale reaction with Ti-catalyst (Example 9). The Tgshows a value of 111.4 °C, which is a slight improvement of 0.6 °C.

[0106] In stark contrast to PEF, using a small amount of a low-toxic, abundant, and cost-effective Ti catalyst significantly improves PEIF's composition and physicochemical properties, resulting in more than a twofold increase in Mn. The advantages of employing 50 ppm Ti, including a significant 60% decrease in reaction time and 70% reduction in required mol% DEO, could arguably outweigh the drawbacks, such as undesired side reactions and coloration.

[0107] Without being bound to theory, the inventors believe that the pronounced disparity in the greater influence of Ti on increasing the Mnof PEIF compared to PEF, is likely associated with the steric ISO units in PEIF. These groups render the reversible reaction more challenging for EtOH, thereby preventing the formation of FDCA-ethyl esters (chain stoppers).

[0108] The optimized PEIF of Example 10 was tested for its barrier and tensile properties. The results are indicated in Table 5 below.Table 5. Barrier properties of optimized PEIF obtained in example 10: oxygen and water transmission rates and permeability at 85 pm thickness.

[0109] The optimized PEIF samples obtained in Example 10 was also tested for their mechanical properties. The results are indicated below. To examine the mechanical properties, the optimized PEF materials were processed into five tensile bars. The tensile strength, Young's modulus, stress at yield and elongation at break were analyzed for three samples per example.Table 6. Tensile data for optimized PEIF obtained in Example 10.

[0110] For all examples according to the invention, wherein PEF is obtained,1H-NMR analysis revealed that the normalized FDCA-derived proton integrals were approximately equal to the normalized glycol-derived proton integrals. In addition, there was no indication of ethylene glycol - oxalate ester units in the PEF. This indicates that the oxalate has reacted to form 1,4-dioxane- 2,3-dione and has not been incorporated into the polyester chain.

Claims

CLAIMS1. A process for the production of a polyester (co)polymer, comprising the steps of a) adding a dialkyl oxalate ester to a starting polyester (co)polymer at least comprising carboxylic acid derived monomer units and 1,2-diol derived alcohol end groups, wherein the alkyl group in the dialkyl oxalate ester is an aliphatic alkyl group wherein the carbon in the alkyl group that is attached to the oxalate is a primary carbon; b) in case the temperature is not already higher than 220 °C, elevating the temperature of the mixture resulting from step (a) to at least 220 °C, and reducing the pressure, for a period of time sufficient to obtain a polyester (co)polymer product having a higher molecular weight than that of the starting polyester (co)polymer, wherein the amount of oxalate units that remain in the (co)polymer is less than the amount added in the form of the dialkyl oxalate ester.

2. A process according to claim 1, wherein in step b) the period of time is sufficient to obtain a polyester (co)polymer product wherein the amount of oxalate units that remain in the (co)polymer is less than 50% of the amount added in the form of the dialkyl oxalate ester.

3. A process according to claim 1 or claim 2, wherein in step b) the period of time is sufficient to obtain a polyester (co)polymer product wherein no or less than 1 mol% oxalate units are present, the percentage relative to the total amount of monomer units.

4. The process according to any of the preceding claims, wherein in step a) an amount of 0.2 to 25 mole% of the dialkyl oxalate is added to the starting polyester (co)polymer, the percentages being relative to the total amount of carboxylic acid-derived monomer units in the starting polyester (co)polymer.

5. The process according to any of the preceding claims, wherein the dialkyl oxalate ester is dimethyl oxalate or diethyl oxalate.

6. The process according to any of the preceding claims, wherein at least one of the units derived from a 1,2-diol is derived from an aliphatic diol selected from mono-ethylene glycol, 1,2 propanediol, 1,2-butanediol, 2,3-butanediol and 1,2-cyclohexanediol.

7. The process according to any of the preceding claims, wherein the starting polyester (co)polymer comprises additional diol-derived units, preferably selected from cis- and / or trans- 2,2,4,4-tetramethyl-l,3-cyclobutanediol, and l,4:3,6-dianhydroxohexitols, in particular isosorbide.

8. The process of any of the preceding claims, wherein the starting polyester (co)polymer comprises dicarboxylic acid derived units, the dicarboxylic acids or any esters thereof selected from (hetero)aromatic dicarboxylic acids or any esters thereof and C2-C18 aliphatic dicarboxylic acids or any esters thereof, which may be linear, cyclic or branched.

9. The process of any of the preceding claims, comprising the steps of(i) providing or producing the starting polyester (co)polymer, wherein the process for producing the polyester (co)polymer comprises esterification or transesterification, and polycondensation;(ii) adding the dialkyl oxalate ester to the starting polyester (co)polymer;(iii) elevating the temperature of the mixture resulting from step (ii) to at least 220 °C, or, in case the temperature is already higher than 220 °C, maintaining the temperature for a period of time;(iv) reducing pressure, preferably to at most 5 mbar, more preferably to equal to or below 1 mbar, thereby producing a (co)polymer product; and(v) optionally repeating steps (ii) to (iv) one or more times with as starting polyester (co)polymer the (co)polymer product with the higher molecular weight of the previous step (iv) until a final polyester (co)polymer product is produced wherein no or less than 1 mole % oxalate units are present, the percentage relative to the total amount of monomer units, and with the desired high molecular weight properties.

10. The process of claim 9, wherein step (i) comprises producing the starting polyester (co)polymer by reacting dicarboxylic acids or esters thereof with diols and / or polyols and / or oligomers with diol derived end groups, optionally in the presence of a catalyst, wherein at least one of the diols and / or polyols and / or oligomers with diol derived end groups comprises a vicinal diol group.

11. The process of claim 9 or 10, wherein the starting polyester (co)polymer is a recycled polyester material or derived from a recycled polyester material.

12. A process according to any of the preceding claims, wherein step a) of the process is caried out in the absence of a metal catalyst.

13. A process according to any of the preceding claims, wherein the process is carried out in absence of a metal catalyst.

Images