Depolymerization process of pet coming from bottle and textile material waste

Abstract

Depolymerization process of PET of pure waste coming from bottles in the form of flakes or from textile material in the form of fibres containing PET in quantities of at least 80- 85% by glycolysis with an ethylene glycol selected from ethylene glycol, diethylene glycol and / or triethylene glycol at temperatures lower than or equal to 270 °C in a single stage a) in which feeding provides for the following steps: a1) adding the reagents in batch mode with ethylene glycol in excess with respect to the PET to guarantee the complete wettability of the PET, a2) when the solid fraction is completely depolymerized, the PET is added in semi-batch mode so as to guarantee a complete wetting in the ethylene glycol in the formed BHET and in the oligomers formed until a prefixed molar ratio of ethylene glycol / PET comprised between 0 and 1.5, preferably between 0.1 and 1.5 is obtained; a3) at the end of step a2) PET and possibly ethylene glycol are added according to CSTR mode in the aforementioned molar ratio comprised between 0 and 1.5, preferably between 0.1 and 1.5 to obtain oligomeric products with unimodal distribution having average molecular weight comprised between 500 and 5000 Da in the case the waste material comes from bottles or comprised between 300 and 450 Da, in the case the PET comes from waste textile materials.

Description

[0001] DEPOLYMERIZATION PROCESS OF PET COMING FROM BOTTLE AND TEXTILE MATERIAL WASTE

[0002] TECHNICAL FIELD

[0003] The present invention relates to a PET depolymerization process starting from PET -based waste materials by single-stage glycolysis but characterized by batch modes, followed by semi-batch and continuous modes (CSTR, Continuous Stirred-Tank Reactor), which allows to obtain oligomers with unimodal distribution of the weight average molecular weight comprised between 500 and 5000 Da.

[0004] STATE OF THE ART

[0005] Polyesters represent a significant fraction of the plastic material produced globally, and PET is one of the main exponents of this type of plastics in terms of volumes produced. In fact, thanks to its chemi cal -physical properties and its low cost, it is the most widely used plastic material in the field of packaging, fibres, and food packaging.

[0006] Given the enormous production of plastics (millions of tons per year), in recent years the world population has developed a strong awareness of the problems related to the environment and pollution caused by processing waste and, above all, by the materials themselves at the end of their lives. For this reason, many national governments and supranational bodies are increasingly discouraging the production and use of single-use plastic objects, preferring instead on the one hand bio-compatible materials, and on the other hand the reuse and recycling of used and waste plastics.

[0007] Currently, there are several methodologies to recycle PET, and they can be classified as chemical, mechanical and thermal methods, but currently all these processes have in common the inability to uncompromisingly meet the following aims: cost-effectiveness, flexibility, simplicity and the possibility of using mixed and low-value waste (such as coloured bottles and waste from the textile sector).

[0008] In particular, the chemical recycling processes are capable of treating contaminated and low-value waste, but they are much more complex, energy-intensive, and require large amounts of solvent, which means that none of them is currently operational on an industrial scale. Furthermore, at the state of the art, all the processes proposed for the chemical recycling of PET by depolymerization lead to the formation of monomers (TP A, DMT, BHET) and possibly the oligomers formed are seen as by-products and eliminated in the purification processes downstream of the depolymerization reaction. These oligomers instead would be materials that can be used for the synthesis of other higher value-added polymers, such as the polyurethanes, which in this case would therefore be composed of a high fraction of recycled material.

[0009] In fact, once the total depolymerization has been carried out, the PET oligomers can be synthesized through a monomer oligomerization reaction, but this 2-stage process is particularly inconvenient if it is compared to a single-stage process of partial depolymerization that allows the PET oligomers to be formed directly from the recovered polymer.

[0010] Currently, however, there is no process capable of operating a partial depolymerization to obtain oligomers instead of monomers.

[0011] SUMMARY OF THE INVENTION

[0012] The Applicant has now found a process that allows obtaining oligomers with unimodal distribution from waste material consisting of PET or comprising PET in quantities of at least 80-85% by weight on the total weight of said waste material.

[0013] This process is conducted in a single stage and involves the reaction of glycolysis of the PET in the presence of a catalyst with at least one ethylene glycol selected from ethylene glycol, diethylene glycol and triethylene glycol in a single stage a) at temperatures below or equal to 270 °C in which feeding provides for the following steps: al) adding the reagents in batch mode with an excess of at least one ethylene glycol with respect to the PET to guarantee the complete wettability of the PET, a2) when the solid fraction is completely depolymerized, adding in semi-batch mode the PET so as to guarantee a complete wetting in the at least one ethylene glycol in the formed BHET and in the oligomers formed in the depolymerization until a desired ethylene glycol / PET molar ratio comprised between 0 and 1.5, preferably between 0.1 and 1.5 is obtained; a3) at the end of a2) PET and possibly the at least one ethylene glycol are added according to CSTR mode in the ethylene glycol / PET molar ratio comprised between 0 and 1.5, preferably between 0.1 and 1.5.

[0014] When the waste material consists of PET, with this process oligomers with weight average molecular weight with unimodal distribution comprised between 500 and 5000 Da are obtained, while when a waste material comprising at least 80-85% by weight of PET is used, based on the total weight of the waste material, oligomers with weight average weight comprised between 300 and 450 Da are obtained. The process of the invention has the following advantages:

[0015] • provides for a single reactive step, which can be easily operated continuously.

[0016] • During the same depolymerization reaction, the chemical structure of the PET oligomers can be modified by replacing part of the ethylene glycol used with another diol (such as DEG, TEG, ...)

[0017] • The output of the process is a mixture of unimodal oligomers, normally distributed over an average molecular weight selectable in the range 500-5000 Dalton, in the case the waste material consists of PET as a function of the reaction temperature and the PETZEG ratio fed in steps a2) and a3) of the process of the invention;

[0018] • The process can use both transparent and coloured PET waste, which cannot be treated mechanically, and does not require special purification stages.

[0019] • The oligomers obtained can be used as diols in the formulation of higher value- added materials such as the polyurethanes (operating an up-cycling), and the variability of the molecular weights and compositions of the oligomers are reflected on the properties of the polyurethanes obtainable subsequently.

[0020] • The process operates with extremely small amounts of solvent (glycol in stoichiometric and sub-stoichiometric amounts with respect to PET), under mild conditions, at atmospheric pressure and temperatures below 270 °C, so it is not particularly energy-intensive compared to the alternative chemical recycling processes and has a low environmental impact.

[0021] • This process allows the treatment of any type of PET waste, from bottle flakes (pure and clean) to the mixed textiles (polycotton), in the latter case when the PET content present in the treated fibres is >80-85%.

[0022] DESCRIPTION OF THE FIGURES

[0023] Figure 1 shows the bimodal distribution of the molecular masses of the PET partial depolymerization products by operating: (a) in excess of Ethylene Glycol and (b) in defect of Ethylene Glycol, i.e. with EG / PET molar ratio <1.

[0024] Figure 2 shows the temperatures involved in the PET glycolysis reaction.

[0025] Figure 3 demonstrates the non-linearity of the dependence between EG / PET molar ratio fed to the reactor and average molecular weight of the glycolyzed product.

[0026] Figure 4 shows the results of the GPC analysis of the glycolyzed product obtained using decreasing EG / PET molar ratios. Figure 5 shows the appearance of the oligomers obtained in some of the experiments shown in Table 4 (comparison between colour of the starting material (a), presence of nitrogen during the reaction and pre-treatment in the oven (b), type of catalyst used (c). Figure 6 shows the results of the GPC analyses of the oligomers obtained in some of the experiments shown in Table 4 (in the presence and absence of inert environment (nitrogen) (a), with different catalyst (b), different final target molecular weight (c).

[0027] Figure 7 shows the appearance of the PET flakes having different characteristic dimension (XL, L, M, S, XS as shown in Table 1).

[0028] Figure 8 shows the result of the depolymerization obtained with the process of the invention during the start-up of the reactor according to the differences in dimension of the S, M, L, and XL flakes.

[0029] Figure 9 shows the volume occupied by flakes of size L, M, and S for the same mass (1.5 [g])

[0030] Figure 10 shows in graph the materials fed and the temperature trend over time in a startup operation according to the process of the invention of the reactor (depolymerization reaction of PET from mixed textile waste).

[0031] DETAILED DESCRIPTION OF THE INVENTION

[0032] For the purposes of the present invention, the definition “comprising” does not exclude other components, in addition to those explicitly stated after said definition.

[0033] For the purposes of the present invention, the definition “consists of’ excludes the presence of further components, than those specified after said definition.

[0034] The expression “feeding in batch mode” or “discontinuous mode” before triggering the reaction means that the reagents in step al) are fed to the reactor in a single solution.

[0035] By the definition “feeding in semi-batch mode” in step a2) it is meant that the solid PET is added periodically in portions after the previous portion of PET has been completely wetted in the solvent.

[0036] By the definition of “feeding with CSTR mode” it is meant the continuous stirring reactor feeding or continuous feeding of PET and of at least one ethylene glycol.

[0037] By ethylene glycol it is meant a glycol of formula OH(CH2CH2O)nH wherein n is an integer between 1 and 3.

[0038] For the purposes of the present invention, unimodal distribution means a maximum frequency distribution in which only one modal value appears. The average molecular weight of the oligomeric products obtained by the process of the invention is determined by methods known to the person skilled in the art, for example by GPC (Gel Permeation Chromatography) according to the method shown below.

[0039] The solid samples are dissolved in HFIP (hexafluoro isopropanol) and analysed by means of a GPC Agilent 1100 equipped with a UV detector set at 290 nm, using HFIP also as mobile phase, with a flow rate of 1 mL / min. Separation is carried out in a GPC Agilent column commercially available (PFG columns) based on silica (PSS (Polymer Standards Service), 8 x 300 mm, particle size of 7 pm), injecting a volume of 60 pL. A calibration curve is then built to correlate the elution time of the sample with its molecular weight using PET standards with average molecular weight between 254 and 50,000 g / mol.

[0040] Preferably the waste material consisting of PET alone comes from transparent or coloured bottles and is in the form of flakes.

[0041] Flakes generally have dimensions >5mm; 2-5mm; l-2mm; 0.25 -1mm; <0.25 mm.

[0042] According to a preferred alternative, in the process according to the present invention in the feeding step al) the ethylene glycol / PET ratio is comprised between 3 and 9.

[0043] Preferably when the process is conducted from waste material consisting of PET wherein the ratio of at least one ethylene glycol / PET at the end of feeding a2) and in a3) we can only add PET, when ethylene glycol is added the molar ratio is preferably between 0.1 and 1.5 even more preferably between 0.2 and 1.

[0044] Preferably when the process of the invention is conducted on waste material comprising PET in concentrations of at least 80-85% by weight on weight, this material is textile material preferably it is polycotton and it is preferably in the form of fibres.

[0045] Preferably in this case at the end of feeding a2) and in the feeding step a3) the molar ratio at least one ethylene glycol / PET is comprised between 0.5 and 1.25.

[0046] Although the process of the invention can be conducted in air, it is preferably conducted in an inert environment, preferably under nitrogen.

[0047] Preferably, the catalyst is selected among sodium bicarbonate, potassium bicarbonate, sodium carbonate, potassium carbonate, zinc acetate, titanium oxide, zinc oxide, calcium oxide, aluminium oxide, magnesium acetate, manganese acetate, sodium hydroxide, potassium hydroxide, and mixtures of the foregoing, but it is preferably selected among sodium carbonate and zinc acetate.

[0048] If the process of the invention is conducted using zinc acetate as a catalyst, then it is preferable to conduct the process of the invention in an inert environment, under nitrogen. The following experimental part that allowed to highlight the innovative characteristics of the process subject matter of the present invention is shown below for illustrative, but not limiting purposes.

[0049] Most of the PET glycolysis technologies and processes currently present in the literature propose to operate the depolymerization in large excess of ethylene glycol, which is necessary in order to ensure a complete and uniform wettability of the PET flakes used. In particular, the amount of ethylene glycol needed to ensure this depends on the average dimension of the PET flakes used, as can be seen in Table 1, and it is evident that it is not possible to fall below an EG / PET weight ratio ~ 2.

[0050] Table 1 - amount of ethylene glycol required for wetting 1 [g] of PET as a function of the dimension of the PET flake

[0051] On the other hand, with regard to the reaction stoichiometry, it would be sufficient to work with an EG / PET molar ratio = 1 to obtain (if the depolymerization was irreversible and a complete conversion could be achieved) a complete depolymerization to the BHET monomer.

[0052] That said, it is evident that the limiting factor for the described reaction is the minimum amount of ethylene glycol necessary for the complete wettability of the solid fraction of PET, and in the best case (considering PET flakes ‘S’, i.e. of the dimension comprised between 0.25 [mm] and 1 [mm]), the conventional reaction of glycolysis of PET to BHET is conducted in very large stoichiometric excess of ethylene glycol (EG / PET ~ 5.5), as can be seen from Table 1 and Table 2. In this case, the dimension of the flakes ‘XS’ (< 0.25 [mm]) is not taken into account as it is extremely difficult to be obtained with shredding machinery, both industrially and on a laboratory scale.) Table 2 - mass and molar ratios between PET and ethylene glycol typical of a PET glycolysis reaction.

[0053] Consequently, the products of this type of reaction will be rich in ethylene glycol, which must be separated and purified, before being reused for another glycolysis reaction. This involves high costs, as the operations of extracting the BEET monomer from the EG solvent (repeated crystallizations in water) are tedious and in particular the regeneration of a viscous and high boiling solvent such as ethylene glycol is particularly problematic. Indeed, attempts to conduct a PET glycolysis in a conventional manner under quasi- stoichiometric conditions (mol(EG) / mol(PET) — > 1) tend to lead to the formation of a fraction of BEET monomer, but the remaining portion of polymer (solid at the reaction temperature) that has not been sufficiently wetted is not depolymerized and remains in the solid phase. Consequently, GPC analysis of such a product will highlight a multimodal (or at least bimodal) distribution of the molecular weights. The low molecular weight peak will indicate the fraction of PET that, having been well wetted by ethylene glycol, will have depolymerized to monomer (BEET), while the high molecular weight peak will indicate the fraction of PET that, having not been in contact with ethylene glycol, will have remained solid and not completely depolymerized. This bimodal distribution is represented in Figure 1, in the case of a reaction conducted in excess and in defect of EG.

[0054] In the latter case, the partially glycolyzed product will therefore consist of a non- homogeneous mixture of high molecular weight oligomers and BEET monomer, which are not easily separable and being thus heterogeneous, are not attractive for subsequent applications such as for example the synthesis of polyurethanes or other polyesters. In fact, the final properties of the latter polymers will depend largely on the molecular weight of the polyol (in this case: BHET and oligomers), and the latter being very heterogeneous, the control of the final properties of the polymer will be random and uncontrollable.

[0055] Furthermore, the control and reproducibility of the relative amounts of monomer and high molecular weight oligomers is also impossible, as is the control of the average molecular weight of the fraction of high molecular weight oligomers, since both these values are strongly influenced by the wetting degree that can be ensured to the PET flakes, and this is extremely unpredictable when such a small amount of ethylene glycol is used, with which it is not possible to ensure a complete and uniform wetting.

[0056] To solve these problems of material transport and wetting, the idea behind the proposed process is to change the method of feeding the reagents into the reactor (operating the start-up in semi-batch mode and operating at full capacity in CSTR mode) by exploiting the mixture of glycolyzed product (i.e. monomer (BHET) and low molecular weight oligomers (< 4 units)) as a co-solvent for the PET glycolysis reaction, so as to allow a complete wetting of the solid PET fed throughout the duration of the glycolysis reaction, and to be able to work under stoichiometric or even sub-stoichiometric conditions of ethylene glycol with respect to PET (EG / PET molar ratio <1).

[0057] This strategy is possible by considering that the glycolysis reaction is conducted at 200 °C, which is higher than the melting temperature of the BHET monomer (104 °C), as can be seen from Figure 2.

[0058] Operating from this perspective, the reactor is started up as follows: initially the reactor is loaded in batch mode with PET flakes and with ethylene glycol in large excess, in order to ensure a complete wetting of the flakes. Subsequently, once the solid fraction of PET is completely depolymerized, the feeding continues in semi-batch mode, and other PET flakes are added in order to maintain a complete wetting of the solid flakes throughout the start-up step, and this is possible since the BHET monomer and the low molecular weight oligomers are liquid at the reaction temperature.

[0059] Once the desired conditions have been reached (i.e. the desired EG / total PET molar ratio has been fed), the reactor can continue to be operated in CSTR mode, whereby a stoichiometric or sub-stoichiometric mixture of liquid ethylene glycol and solid PET flakes continue to be fed, no longer having any wetting problem of the solid fraction, since the same monomers and small oligomers act as a co-solvent in the reactor.

[0060] In addition, the presence of the monomer and of these small oligomers is also very useful from the point of view of the reaction: in fact, this mixture behaves as a reactive co- solvent. The BHET monomer, in fact, is terminated by 2 OH groups, and can depolymerize the PET polymer chain by attacking the ester bond.

[0061] This feature thus makes it possible to obtain a unimodal distribution of oligomers of the desired molecular weight, thus obtainable by working at different EG / PET molar ratio and by changing the temperature to ensure that the final glycolyzed product having the desired molecular weight (500-5000 Da) remains in the liquid phase throughout the duration of the process.

[0062] In fact, if the target molecular weight is higher than 1000 Da, it is necessary to raise the reaction temperature to keep the oligomer mixture in the liquid phase, but the operation of the reactor remains unchanged.

[0063] In this case, the higher temperatures involved also make the polymerization reaction (oligomerization) among the chains of small oligomers present favourable, and this means that the relationship between the EG / PET molar ratio and the average molecular weight of the glycolyzed product is no longer linear, as can be seen from the results of the tests shown in Figure 3.

[0064] In particular, Table 3 below shows all the correspondences between molecular weights of the glycolyzed final product, temperatures, EG / PET molar ratios and PETZEG weight ratios obtained in tests carried out in a non-inert environment, on XL-sized flakes, without any pre-treatment in the oven, and with sodium carbonate as a catalyst.

[0065] Table 3 - correspondence between PETZEG weight ratio, EG / PET molar ratio, temperatures, and average molecular weight of the glycolyzed product for different PET partial glycolysis tests under the same all other operating conditions.

[0066] Figure 4 shows the results of a GPC analysis of the liquid mixture of oligomers obtained during the start-up step of the reactor continued until an average molecular weight of 5000 Da was obtained. Subsequently, further tests were conducted, keeping the molecular weight target constant and consequently the temperature and the molar ratio of EG / PET fed, but changing operating conditions such as: type of catalyst (zinc acetate and sodium carbonate), reactor inertisation (N2 / no N2), pre-treatment in the oven to dry the PET flakes (oven / no oven), and type of flake used (clear / blue & green). A graphic diagram of the tests described is represented in Table 4.

[0067] The following Table 4 shows the tests conducted at 230 °C and molar ratio in steps a2) and a3) respectively equal to 0.2 and 0.4.

[0068] Table 5 shows the same examples as Table 4 but conducted at 260 °C and molar ratio in steps a2) and a3) respectively equal to 0.1 and 0.2.

[0069] Table 4 - partial glycolysis of PET: experiments conducted at the temperature of 230 °C and with constant EG / PET molar ratio in steps a2) and a3), respectively equal to 0.2 and 0.4, but varying the other operating conditions.

[0070] Table 5 - partial glycolysis of PET: experiments conducted at the temperature of 260 °C and with constant EG / PET molar ratio in steps a2) and a3), respectively equal to 0.1 and 0.2, but varying the other operating conditions.

[0071] These latest tests confirm the reproducibility of the reaction operated in the manner described, and highlight in particular the following aspects:

[0072] - the difference between the type of material treated (transparent or coloured flakes) has no significant impact on the reaction, but it is noted that the glycolyzed product maintains a slight colouring corresponding to the type of starting material, as shown in Figure 5.

[0073] The presence or absence of the pre-treatment in the oven does not appear to have a perceptible effect on the reaction and on the type of glycolyzed product obtained. Conducting the reaction in an inert environment is very useful in order to obtain a glycolyzed product of higher quality, in fact by limiting oxidation reactions, a more pure product is obtained, as can be seen from Figure 5b and Figure 6a, where the presence of the tail at low molecular weights from the GPC analysis is attributable to the presence of oxidation by-products.

[0074] The type of catalyst has an equally evident effect on the glycolysis reaction described, with zinc acetate allowing faster reaction times than sodium carbonate, but on the other hand leading to greater oxidation of the glycolyzed product (albeit slight), as seen in Figure 5c and Figure 6b.

[0075] The target molecular weight of the oligomers obtained has a very marked effect on the amount of degraded product, as can be clearly seen from Figure 6c, and only by conducting the reaction in an inert environment it is possible to obtain a glycolyzed product poor in degraded by-products, as seen in Figure 6a.

[0076] Note that all the tests described above (summarized in Table 4) were conducted maintaining a constant EG / PET molar ratio (equal to 0.5) and consequently a constant target of the weight average molecular weight (MWw) of the oligomers of 1200 [g / mol]. Despite this, for some of the operating combinations shown in the table, additional tests were conducted maintaining the operating conditions unchanged, but using a different EG / PET ratio and consequently obtaining a different final molecular weight (for example in the case of exp_5 and exp_5’, in which the final weight average molecular weights are equal to 1200g / mol and 3200g / mol, respectively).

[0077] In addition, it is worth underlining that all the tests described were conducted using post-consumer flakes of PET bottles commonly intended for mechanical recycling processes obtained from an industrial supplier, without any further mechanical treatment being applied to them. In particular, as shown in Figure 7, the flakes used have a characteristic dimension > 5 [mm], and have been labelled as XL flakes.

[0078] Subsequently, the other flakes (with the dimensions previously shown in Table 1) were also tested according to the proposed start-up process, in particular by feeding fresh PET in such an amount as to always guarantee an excellent wetting and effective mixing, and by feeding a new aliquot of fresh PET at the moment when the solid fraction previously fed was completely depolymerized (and dissolved) in the liquid phase.

[0079] It is worth underlining that the same mass of PET was added to each feeding of fresh PET in all the tests, despite the same mass of PET in all the tests, despite the fact that in the case of smaller flakes (M and S size) it was possible to feed a greater amount of PET, while maintaining an excellent wetting and effective mixing as shown in Table 1. In fact, the aim of these tests was to compare the depolymerization rates typical of flakes of different dimensions. As can be seen from Figure 8, industrial flakes (XL dimensions) take more than twice as long (125 vs. 58 min) to complete the depolymerization process to liquid (i.e. soluble in the reagent liquid) monomer and oligomers.

[0080] On the other hand, the flakes with intermediate dimension (L and M) require practically the same time to complete depolymerization, while a further slight reduction in reaction time is observed in the case of the reduction of the flakes (S size). In particular, these differences demonstrate that diffusive limitations have an important impact on the process and can be explained with three considerations.

[0081] - Firstly, the slight difference between the depolymerization speed between M and L flakes is explained by considering that the typical geometry of this type of flakes is lamellar, so a slight difference in terms of the characteristic diameter of the lamellas (from [2 - 5 ][mm] to[ 1 - 2] [mm]) will result in a minimal variation in the surface area of the flake useful for the reaction.

[0082] Turning to the S dimension, on the other hand, there is a more significant increase in the depolymerization speed, due to the fact that at this characteristic diameter we are dealing with a now pseudo-spherical geometry, and a reduction in the characteristic diameter has a significant impact on the surface area useful for the reaction.

[0083] The greatest difference in depolymerization times, which occurs by passing from XL flakes to L flakes, can instead be explained by considering that, to obtain the latter, the XL flakes went through a shredding process in a blender, which supposedly caused cuts and abrasions to the surface of the flakes, so in this way it greatly increased the surface of the flake useful for the reaction.

[0084] In any case, the different characteristic dimension of the PET flakes (especially between L, M, and S flakes) has a very marked effect on the maximum amount of PET that can be fed in order to allow and maintain a good wetting and efficient mixing, as already discussed and highlighted in Table 1.

[0085] In fact, given the irregular shape of the flakes, it occurs that a reduction in the characteristic diameter, for example from [2 - 5] [mm] to [1 - 2] [mm], turns out to have a considerable impact on the volume of the PET, as seen in Figure 9, where in order to obtain an equivalent mass (1.5 [g]), the L-sized flakes occupy a volume almost twice that of the M-sized flakes, and almost four times that of the S-sized flakes. In conclusion, this operating procedure allows an excellent mixing of the reagents and ensures that an effective wetting between solid and liquid phase is maintained. The reactor, operated in the manner described (semi-batch and CSTR), also allows to operate easily in an inert atmosphere and so as to drastically increase the quality of the final glycolyzed product.

[0086] The same approach can be used for the partial depolymerization of the PET fraction contained in mixed textile waste, for example polycotton (the mixture of cotton and PET, i.e. the two most widely used materials in the textile industry).

[0087] In this case, the very nature of the starting material (fibrous, very bulky, and absorbent) makes it even more evident the need to use a very large excess of ethylene glycol in order to be able to carry out a conventional depolymerization of the PET fraction present in the starting mixed textile material. With the approach described (start-up in semi-batch mode and feeding in CSTR mode) it is instead possible to drastically reduce the amount of ethylene glycol necessary for depolymerization, provided that the percentage of PET in the starting polycotton is such that a sufficient liquid (EG + RETET) flow rate to completely and effectively wet the fed textile material can be maintained in the reactor (operating in steady state). This occurs in particular when the PET fraction present in the starting textile material is >80-85%.

[0088] In this case, unlike the case described above (when 100% pure PET bottle flakes were used as the starting material), the aim is no longer so much to obtain oligomers but rather to obtain the monomer using the minimum possible amount of ethylene glycol (i.e., using an EG / PET molar ratio as close as possible to 1). Indeed, the purification process of the glycolyzed product coming from recovered textile material is much more complex and important than that to be carried out on material coming from pure waste (such as bottle flakes) and is much easier if carried out on the monomer rather than on the oligomers.

[0089] Table 6 shows the operating conditions in terms of EG / PET molar ratio and the molecular weight of the product obtained following a series of tests conducted with the same procedure described above but by feeding 90% PET textile fibres into the reactor. All reactions listed were conducted in a non-inert environment, without pretreatment in the oven, and using sodium carbonate as a catalyst. Table 6 - operating conditions and average molecular weight of the glycolyzed product coming from mixed textile fibres

[0090] Evidently, the described approach continues to work also for the depolymerization of textile material with high PET content operating under stoichiometric conditions or little more than EG stoichiometric conditions with respect to PET.

[0091] In the case of reactions operated under stoichiometric conditions (exp fl, exp_f2, and exp_f5), a glycolyzed product is obtained with a molecular weight slightly higher than that of the monomer (i.e. 300-330 [g / mol] compared to the 254 g / mo] typical of the BHET monomer), but this is understandable considering the fact that under the operating conditions in question (absence of ethylene glycol and temperatures of 230 °C), the oligomerization reactions begin to be slightly favoured, at least with regard to the formation of dimers and possibly trimers.

[0092] Finally, Figure 10 depicts the feeding trend of the reagents (PET fibres and ethylene glycol) during the start-up step of the described depolymerization reaction, which aimed to work with an EG / PET molar ratio = 1.25. The reaction was conducted in a reactor equipped with a reflux condenser and heated with a heating mantle, and the temperature trend is explained considering that in the steps in which the EG / PET molar ratio is high, the reaction temperature remains at 200 °C (boiling temperature of ethylene glycol), while in the steps in which this ratio is lowered, the reaction temperature rises, since the thermal power provided by the heating mantle remains constant.

[0093] Unlike what is commonly done in the literature, using such a small amount of ethylene glycol, it is possible to avoid the enormous technological and above all economic problems that an excess (unreacted) ethylene glycol recovery and purification unit would entail.

[0094] In addition, the described process (CSTR) allows to operate easily in an inert atmosphere, so as to drastically reduce the amount of degraded products and to obtain a good quality glycolyzed product.

[0095] Finally, with this type of configuration, reaction times are extremely low, especially in the case of fibres (10 minutes, as can be seen in Figure 9) but also in the case of S, M, and L flakes (15-20 minutes, as can be seen in Figure 8). In fact, the limitations on the diffusion of mass and heat are extremely low in both cases.

Claims

Claims1. Depolymerization process of PET coming from waste material consisting of PET or containing it in quantities of at least 80-85% by glycolysis of PET in the presence of catalyst and at least one ethylene glycol selected from ethylene glycol, diethylene glycol and triethylene glycol at temperatures below or equal to 270 °C conducted in a single stage a) in which feeding provides for the following steps: al) adding the reagents in batch mode by adding the at least one ethylene glycol in excess with respect to the PET to ensure the complete wettability of the PET, a2) when the solid fraction is completely depolymerized, adding the PET in semibatch mode so as to ensure a complete wetting within the at least one ethylene glycol, in the formed BEET and in the oligomers formed in the depolymerization until reaching a desired ethylene glycol / PET molar ratio comprised between 0 and 1.5, preferably between 0.1 and 1.5; a3) adding the PET and possibly the at least one ethylene glycol in CSTR mode in the aforementioned ethylene glycol / PET molar ratio comprised between 0 and 1.5, preferably between 0.1 and 1.5 to obtain oligomeric products with unimodal distribution having average molecular weight comprised between 500 and 5000 Da in case the waste material consists only of PET, comprised between 300 and 450 Da in the case the PET comes from waste textile materials.

2. Process according to claim 1 wherein said waste material consisting of PET comes from coloured and transparent beverage bottles in the form of flakes.

3. Process according to claim 2 wherein said flakes have dimensions selected from: >5mm; 2-5mm; l-2mm; 0.25 -1mm; <0.25 mm.

4. Process according to any one of claims 1-3, wherein the molar ratio of at least one ethylene glycol / PET in the feeding step al) is comprised between: 3 and 9.

5. Process according to any one of claims 1-3 wherein the ratio of at least one ethylene glycol / PET at the end of feeding a2) and in a3) is comprised between 0.2 and 1.

6. Process according to any one of claims 1-5 conducted in an inert environment.

7. Process according to any one of claims 1-6, wherein the glycolysis catalyst is selected among sodium bicarbonate, potassium bicarbonate, sodium carbonate, potassium carbonate, zinc acetate, titanium oxide, zinc oxide, calcium oxide,aluminium oxide, magnesium acetate, manganese acetate, sodium hydroxide, potassium hydroxide, but preferably it is selected among sodium carbonate and zinc acetate.

8. Process according to claim 7, wherein the catalyst is selected among sodium bicarbonate, potassium bicarbonate, sodium carbonate, potassium carbonate, zinc acetate, titanium oxide, zinc oxide, calcium oxide, aluminium oxide, magnesium acetate, manganese acetate, sodium hydroxide, potassium hydroxide, but preferably it is selected among sodium carbonate and zinc acetate, and the glycolysis is conducted under nitrogen.

9. Process according to any one of claims 1-4, wherein when the waste material comprises at least 80-85% PET and the textile material is in the form of fibres.

10. Process according to claim 9, wherein at the end of feeding a2) and in the feeding step a3) the molar ratio at least one ethylene glycol / PET is comprised between 1 and 1.25.

Images