Efficient Plastic Waste Recycling to Value‐Added Products by Integrated Biomass Processing

Abstract The industrial production of polymeric materials is continuously increasing, but sustainable concepts directing towards a circular economy remain rather elusive. The present investigation focuses on the recycling of polyoxymethylene polymers, facilitated through combined catalytic processing of polymer waste and biomass‐derived diols. The integrated concept enables the production of value‐added cyclic acetals, which can flexibly function as solvents, fuel additives, pharmaceutical intermediates, and even monomeric materials for polymerization reactions. Based on this approach, an open‐loop recycling of these waste materials can be envisaged in which the carbon content of the polymer waste is efficiently utilized as a C1 building block, paving the way to unprecedented possibilities within a circular economy of polyoxymethylene polymers.

The industrial production of polymeric materials is continuously increasing, but sustainable concepts directing towards ac ircular economy remainr ather elusive. The present investigation focuseso nt he recyclingo fp olyoxymethylene polymers, facilitated through combined catalytic processing of polymer waste and biomass-derived diols. The integrated concept enablest he production of value-added cyclic acetals, whichc an flexibly functiona ss olvents, fuel additives, pharmaceutical intermediates, and even monomeric materials for polymerization reactions. Based on this approach, an open-loop recycling of these waste materials can be envisaged in which the carbon content of the polymer waste is efficiently utilized asaC1 building block, pavingt he way to unprecedented possibilities within a circulareconomy of polyoxymethylene polymers.
Most plastics are made of synthetic polymers that are produced by the repetitive linkage of versatile small monomers. The continuous optimization of manufacture processes and tailoring of material properties has paved the way to mass productiono fabroad diversity of plastics. Thus, highly versatile consumer products with low weight, high strength,a nd extreme durability are broadlya vailablea tl ow costs. Consequently,p lastics have become crucial in av ariety of strategic industrial sectorsa nd represent essential materials for construction, transportation,a nd packaging. The production of plastic materials is steadily growing and since 1950a round 8300 million tons of this materials have been synthesized. [1] As aresult,worldwidegeneration of plastic waste has dramatically increased and is currently around 150 million tons per year. However,7 9% of this plastic waste ends up in landfills, resulting in up to 2.41 million tons of robustw aste entering our environmente very year. [2] Consequently,t he reuse or substitution of the plastic materials is currently strongly fostered, ase fficient and sustainable recycling strategies remainr ather elusive. [3] Moreover,available recycling concepts are not cost competitivea nd produce polymer materials of lower quality,h ampering the development of ac irculare conomy.T herefore, the development of effective recycling processes within ac ircular economy approach is of utmost importance,i deally producing not only monomers for new plastics, but value-added products or intermediates for other supply chains. [4] Herein, as ustainable concept is introduced, enabling efficient plastic recycling through combined catalytic processing with biomass-derived chemicals, yielding high-value platform chemicals.
The present investigation focuses on the recycling of polyoxymethylene polymers (POM, also knowna sp olyacetal), a thermoplastic material produced by the homo-or copolymerization of formaldehyde in ap roduction capacity of around 1.7 Mtons per year. [5] POMp lasticsh ave been known for more than 40 years and are used in precision parts requiring high stiffness, low friction,a nd excellent dimensional stability in automotive interiors. [6] Currently,P OM can be repurposed via injectionm olding processes, which are limited by material degradationa nd releaseo ff ormaldehyde. Chemical recycling has not been intensively investigated and few approaches focusing on the transformationo fP OM into formaldehyde or trioxane have been reported. [7] In this study,w eh ave demonstrated effectiver ecycling of POM plastic materials, which is facilitated by the combined catalytic processing of polymerw aste and biomass-derived diols. In detail, POM reacts with either biomass-derived [8] or recycled [4c] glycols to produce chemicalproducts such as 1,3-dioxolane,1 ,3-dioxane,o r1 ,3-dioxepane (Scheme 1). These useful cyclic acetal products can serve as solvents, fuel additives, pharmaceutical intermediates, and even monomeric materials for polymerization reactions. [9] The investigation was initiatedw ith ac atalytic reactiono f granules of commercial POM homopolymer together with 1,3propanediol. The biomass-derived substrate, 1,3-propanediol, can be industrially obtained via fermentation processes using the renewable resources corn syrup or glycerol. [10] For the first reactions, 1,4-dioxane wasc hosen as solvent and av ariety of Brønsted acid catalystsw eres creened, promoting the depoly- KGaA. This is an openaccessarticleunder the termsoft he Creative Commons AttributionL icense, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. merization of POM to the formaldehydem onomer and fostering the subsequent condensation to the cyclic acetal (Table 1). In the absence of any acid, the envisaged formation of 1,3-dioxane product was not observed and the starting polymerr emained undissolved in the reactionm ixture (Table 1, entry 1). Gratifyingly,w ith hydrochloric acid as catalystt he POM polymer started to dissolve and 1,3-dioxane product was formed in 31 %y ield after 2h (Table 1, entry 2), corroborating the viability of the envisaged cascade transformation.
When p-toluenesulfonic acid (p-TsOH) was tested as catalyst, only trace amount of the 1,3-dioxane product could be detected, whereas the stronger acid trifluoromethanesulfonimide (HNTf 2 )a fforded the cyclic acetal in 20 %y ield (Table 1, entries 3a nd 4). However,w ith triflic acid (TfOH), 1,3-dioxane was obtainedi n44% yield after 2hand 99 %yield after 8hof reaction time (Table1,e ntries 5a nd 6). From thesei nitial results, the effectiveness of Brønsted acids for the catalytic depolymerization of the POM polymer and subsequent condensation with 1,3-propanediol could be demonstrated. In then ext development step, metal triflates with bifunctional Brønsteda nd Lewis behavior were investigated as catalysts. [11] After screening severalL ewis acids (Table 1, entries 7a nd 8; see also the Supporting Information), Bi(OTf) 3 catalyst demonstrated superior activity and afforded the 1,3-dioxane product in 41 %y ield after 2h,a nd 99 %y ield after 8h (Table 1, entries 9a nd 10). The comparable reactivity observed with Bi(OTf) 3 and TfOH could be attributed to partially hidden Brønsted acid properties of the Lewisa cid upon partial alcoholysis (by 1,3-propandiol) or hydrolysis (by water byproduct).T hus, the dissociation of Bi(OTf) 3 Lewis acid unlockst he hidden Brønsted acidity, while maintaining the Lewis acidic behavior.M oreover,t he Lewis acidity of Bi(OTf) 3 can additionally promote activation of the hydroxy groups of the diol substrate. In this manner,t he formation of aB i-metallacycle intermediate has been previously reportedt of acilitatet he condensation of glycols. [11] Interestingly,w hen only 1.2 equivalents of 1,3-propandiol were used, the product could be obtained in up to 96 %y ield in only 2h (Table 1, entry 11). With additional 1,4-dioxane, the solvation of the polymeric substrate could be facilitated, resulting in 99 % yield of the 1,3-dioxane product ( Table 1, entry 12).
We then focusedo no ptimization of the solvation-depolymerization process by studying the influence of temperature on the possible increase of the reactivity in presence of 1,3propanediol ( Table 2). As expected, the dissolving/depolymerization of the starting POMp olymer slowed down at lower temperatures of 60a nd 70 8C, affording the desired product in 5a nd 26 %y ield compared to 96 %y ield at 80 8C( Ta ble 2, entries 1-3). On increasing the reaction temperature to 100 8C, the POM polymer granules were completely dissolved in the solution after only 20 min and the 1,3-dioxane product was obtained in 97 %y ield, comparedt oo nly 12 %y ield at 80 8C ( Table 2, entries 4a nd 5). This higher reactivity at 100 8Ct emperaturel ed to the study of the effect of lower catalystl oadings. When the reaction time was extendedt o4 0min, the startingm aterial was completely converted and the desired product was obtained in 99 %y ield at low catalyst loading of 1mol %( Ta ble 2, entry 6). Interestingly,w hen the amount of startings ubstrate was increasedf ourfold, we obtained the 1,3dioxanep roduct in 95 %y ield at al ow catalyst loading of 1mol %a nd in just 90 min of reactiontime (Table 2, entry 7).
In the subsequent investigation, 1 HNMR experiments should provide information on the rates of depolymerization and condensation within the cascade transformation. Thus, we carried out as tudy on the time profile of the selective synthesis of 1,3-dioxane from POM polymer/1,3-propanediol mixtures by using Bi(OTf) 3 as catalyst (Figure 1; see also the Supporting Information). The reactionr evealed ac onstant rate of formation  Based on optimized conditions with 1,3-propanediol, we next evaluated the scope of this reactionw ith structural variations on the biomass-derived diols. [8a-h] In the first set of experiments 1,2-butanediol, 1,2-pentanediol, and 1,2-hexanediol were used as substrates together with POMp olymer.T he corresponding cyclic acetal products were obtained in excellent yield of 93-98 %( Ta ble 3, entries 1-3). Furthermore, when sterically constrained 2,3-butanediola nd 2,4-pentanediol were used, lower yields of 65 %a nd 57 %w ere achieved ( Table 3, entries 4a nd 5). Interestingly,t he use of 1,3-butanediol as substrate resulted in the formation of 4-methyl-1,3-dioxanea cetal in an excellent yield of 98 %( Ta ble3,e ntry 6). Subsequently, we attempted the challenging synthesis of seven-membered cyclic acetalss tarting from biomass-derived 1,4-diols. When 1,4-butanediol and 1,4-pentanediolw ere employed as substrates, the corresponding seven-membered cyclic acetal products (1,3-dioxepanes) were formed in 26 %a nd 59 %y ield ( Table 3, entries 7a nd 8) in addition to the formationo faset of different linear glycolic acetals as by-products (see Supporting Information). The higher reactivity of 1,4-pentanediol in comparison to 1,4-butanediol can be attributed to the presence of as econdary alcohol functionality,i mproving the cyclization-acetalization step towards am ore stable seven-membered-ring acetal product.
Grounded on the principles of green chemistry and previous investigations from our group on polymer recycling, ar eaction withouts olventu se wast argeted. [12] Thus, at ransformation only with the application of the starting materials 1,3-propanediol and POM polymer was performed. In the initial phase of the reaction the absence of adapted solventr educes the solvation and subsequent depolymerization rates of POM. However, the accumulation of the produced 1,3-dioxane product in the reactionm ixture continuouslye nhances the POMs olvation step in as elf-breeding (autosolvation)m anner.M ore specifically,s tarting with neat conditions, the reaction resulted in the formation of 1,3-dioxane with an overall yield of 86 %w ith only 0.2 mol %o fa cidc atalyst( Scheme2). More interestingly, when the self-breeding approachi ss upported with an initial addition of 1,3-dioxane product (0.5 mL), the solvation-depolymerization of POM was enhanced and the yield of the produced 1,3-dioxane increased to 93 %( Scheme 2). This important finding strongly facilitates the development of an adapted process concept, as final purification only requires the distillative separation of water and the cyclic acetal product. HNMR spectroscopy using mesitylene as an internal standard. In addition to the biomass-derived diols, as et of these diols can also be derived from the recycling of polymeric plastic waste. [4c, 8a,e] Twos pecific examples of the latter case are represented by ethylene glycola nd propylene glycol,w hichc an be obtained from waste polyethylene terephthalate (PET) andp olylactic acid (PLA). [4c] Moreover,t he expected cyclic acetal products of these glycol substrates have already been introduced (mainly 1,3-dioxolane) as commercial solvents for polar polymers, in paint stripping formulations, and as ag eneral cleanup solventf or epoxy and urethane systems, as well as for the synthesis of selected pharmaceutical intermediates. [13] Remarkably,w hen ethylene glycol and propylene glycol were employed in the reaction with POM substrate, the corresponding cyclic acetal products,1 ,3-dioxolane and 4-methyl-1,3-dioxolane, were obtainedi n6 6% and 93 %y ield, respectively (Scheme3), clearly validating the versatility and substrate flexibility of the developed concept.
In the last decade, market demand for POM polymerh as doubled, leadingt oa ni ncrease of the production capacity to 1.7 million tons per annum in 2015. POM polymers are widely used in the production of av ariety of commercial plastic products, including consumer goods such as toys, zippers, clips, cosmetic containers, and pens, which are typical items found in contaminated shore and sea areas. Moreover,P OM has also been used for complicated engineering applications, mainly in the automotive industry for production of around 3000 different components used for external&internal auto parts. Consequently, in af inal set of experiments, our approach of utilizing POM polymer as an alternative C1 source for the synthesis of cyclic acetals was applied on some commercial consumer products,s uch as small gears, baggage clips, and disposable lighters, as well as old laboratory joint clips ( Figure 2). These commercial POMp lastic wastes were shredded to smallf ragments (< 3mmi ns ize) and treated under the developed con-ditions with 1,3-propanediol. Astonishingly,t he waste material selectively forms the envisaged 1,3-dioxane products in very good yield (Figure 2a nd Table S10 in the Supporting Information). Interestingly,t he POM polymer content of these plastics was fully converted, whereas the dyes and additives could be precipitated andr emovedb yf iltration. More specifically,1 ,3-dioxane was formed in 87, 90, 88, 92, and 88 %y ield, starting from baggage clips, joint clips,d isposable lighters, and plastic gears,r espectively,u sing only 0.2 mol %o fB i(OTf) 3 catalyst.
Finally,ascale-up of the reactions tartingf rom approximately 6g of am ixture of commercial POM waste under neat conditions was performed. After product distillation, 91 %y ield of pure 1,3-dioxanep roduct could be isolated. The detailed steps of the straightforward recycling concept on al arger scale are shown in Figure 3, clearly corroborating the potentialo ft he facile approach. Moreover,t he performance of the reaction under neat conditions and the ease of separation of the product disclosethe future potential of this method.
The efficiency of this recycling approach of POM plastic waste can be evaluated by the converted carbonc ontent of the used plastic waste. Interestingly,t he complete carbon content of the POM homopolymer is efficiently converted into the acetal unit of the cyclic product drivenb yt he excellent yields obtained for these products. Moreover,the addressed recycling methodo fP OM is associated with av ery low E-factor, [14] where water is produced as the only byproduct of the acetalization reaction.
In the presentw ork, the development of as ustainable concept for effective polymerr ecycling was targeted, focusing on the transformation of polyoxymethylene waste materials. The basic concept was grounded on acombined catalytic processing of polymer waste materiala nd biomass derived diols. The respective integrated concept enabled as elective cascade reaction, encompassing effective depolymerization and condensation, leading to cyclic acetals. The robustness of this approacha llowed the flexible formation of various cyclic products in high yield by facile modification of biomassderived diol. Based on this approach, an efficient open-loop recycling of these waste materials can be envisaged,p aving the way to unprecedented possibilities within ac ircular economy of polyoxymethylene plastic polymers.

Experimental Section
General procedure for the synthesis of 1,3-dioxane from homo-POM plasticpolymer and 1,3-propanediol All experiments were conducted in 5mLs ealable heavy-walled glass vials equipped with am agnetic stir bar.A fter weighing Bi(OTf) 3 (0.058 g, 0.088 mmol), homo-POM (51-54 mg, 1.7-1.8 mmol;c alculated based on the formaldehyde H 2 CO repetition unit), and diol (0.156 g, 2.05 mmol) in the vial, 1,4-dioxane (2 mL) was added and the vial was sealed with a2 0mma luminum seal equipped with septa. The reaction mixture was stirred and heated to 100 8Cb yu sing ac ustomized aluminum heating block. After 2h,t he vial was cooled to room temperature and NMR samples were prepared by using [D 6 ]DMSO solvent and mesitylene as internal standard.