Glycolysis process for polyurethane waste recycling.

Resumen

Polyurethane (PU) is nowadays one of the most relevant polymers in the plastic market due to its versatility and suitability for a large number of applications, which makes its consumption to grow continuously. In fact, polyurethane is the 6th most used polymer all over the world with a production of approximately 18 million tons per year. As a direct consequence of its commercial success and its resistance to biodegradation, an increasing quantity of polyurethane waste is being disposed in landfilling becoming a relevant environmental matter. Such waste comprises not only post-consumer products but also scraps from slabstock manufacturing, which can reach the 10% of the total foam production. In the past, landfilling was the solution to the problem; however, the massive enforcement of the environmental laws, following the mandates of the European policies about waste management, is pointing out a new route in the polymer waste removal sector based in the polymer recycling. This fact has brought the polymers waste treatment as a key research topic, engaging the attention of the scientific and the industrial world, with the global goal of maintaining the natural resources, diminishing the quantity of waste disposed in landfills and enhancing the sustainability for forthcoming generations. Physical recycling processes such as rebonding, powdering or compression molding are successfully applied to thermoplastics polyurethanes, but they are useless for the majority of the polyurethane specialties due to their thermostable nature. An alternative way for recycling includes chemical treatment to convert the PU back into its starting raw materials, mainly polyols. Hydrolysis, treatment with esters of phosphoric acid, aminolysis with low weight alkanolamines and glycolysis have been described as suitable procedures to break down the polyurethane chain, converting the PU back into its starting raw materials. Among them, glycolysis is the one that presents the highest development grade in terms of research and technological maturity. However, its development level depends on the process type considered: single-phase or split-phase, being the single-phase glycolysis processes the ones implanted in the industry up to now, in spite of the low quality of the recovered polyol that these processes yield. On contrary, split-phase glycolysis processes have been only developed up to pilot plant scale due to the uncertainty that exists with respect to its economic viability. The problems for the economic feasibility are: the high price of the cleavage agent, which has to be used in a considerable molar excess to ensure phases separation, and the uncertain possibility of joint recycling of any kind of PU scraps without performing a selective and costly waste collection. Therefore, the purpose of this research work is trying to overcome these two main impediments of the split-phase glycolysis process. For this purpose the efforts have been focused first in demonstrating the feasibility of the process to any kind of flexible PU foam waste, regardless the polyol type used in the assembling of the original foam; and, second, in the necessary improvement of the recycling treatment economy, looking for a more convenient glycolysis agent. It is important to note that flexible foams are categorized as conventional (containing or non-containing polymeric polyols) (≈60%), viscoelastic (≈20%) and High Resilience (HR) (≈20%). A series of works about the split-phase glycolysis of a conventional flexible foam based on a flexible polyether polyol of Mn 3500 Da and hydroxyl number 48 mgKOH/g were previously carried out by the research group and represented the starting point of the present PhD Thesis, providing the optimal conditions to carry out the glycolysis process (catalyst concentration in the glycolysis agent = 1.3%; mass ratio of glycolysis agent to PU foam = 1.5:1; reaction temperature = 190ºC). Nevertheless, currently, flexible polyurethane production is mainly based on high performing foams, such as conventional foams containing polymeric polyols and viscoelastic foams, since they provide outstanding mechanical properties in comparison to conventional ones. Hence, the first efforts of this research work were focused in the extension of the split-phase glycolysis process to these flexible foams types. In this regard, split-phase glycolysis reactions were performed with flexible polyurethane foams waste containing graft polymeric polyols. From the obtained results, it was stated that it is possible to perform the glycolysis of these special foams using the same reaction conditions optimized for conventional polyurethane foams. In this study, an upper phase mainly formed by the recovered polyol (close to a 69 wt%) was obtained and after a purification step, consisting in a liquid-liquid extraction with slightly acidified water, the recovered polyols could be valorized in a further foaming process replacing part of a raw flexible polyether polyol. The purification process was optimized in order to maximize the yield and purity of the recovered polyol. The recipe assayed for the synthesis of polyurethane foams with the recovered polyols was the same that had been used with the recovered polyol from conventional foam synthesized without graft polymeric polyol adjusting just the isocyanate proportion depending on the recovered polyol hydroxyl number in order to maintain the desired isocyanate index (105). Therefore, PU waste scraps coming from conventional foams could be treated together, regardless the presence of polymeric polyol, obtaining a similar final product susceptible of substituting partially the flexible polyether polyol in a further foaming process. With the same aim, next experiments were oriented to the extension of the glycolysis process to viscoelastic flexible PU foam scraps, due to the sharply growing that is undergoing the manufacture of these foams in recent years. The high production increase of the foam type is mainly resulting from its advantageously pressure distribution performance as a consequence of the employment of two polyalcohols, a high molecular weight polyether polyol and a short chain one, that provide the breaking of the polyurethane matrix symmetry, achieving low resiliency properties. Transesterification reactions were successfully carried out using the same reaction conditions optimized in previous works with flexible conventional PU foam waste. An upper phase mainly formed by the recovered polyol (≈61 wt%) was obtained confirming that a proper split-phase glycolysis took place. Moreover, after the same purifying process that commented above, a recovered polyol with optimal properties to be foamed was achieved (Mn = 2555.86, f = 2.983 y OH = 64.56 mg KOH/g). This way, it was demonstrated that the joint recycling treatment of conventional and viscoelastic PU foams blends would be possible, avoiding, this way, the necessity of a selective collection or a previous separation step of PU scraps. Furthermore, glycolysis bottom phase obtained from viscoelastic flexible PU foam scraps was put in value as a replacement of up to 75 % of a raw rigid polyether polyol in the synthesis of new rigid PU foams, achieving a global valorization of the glycolysis phases. Next, in order to accomplish the second main aim of the thesis by finding an economical, sustainable and environmental-friendly cleavage agent that allowed to improve significantly the split-phase glycolysis economy; crude glycerol (purity 80%) was used for the split-phase glycolysis of viscoelastic and high resilience polyurethane foams. Crude glycerol is a subproduct of the biodiesel production, reason why its price is at least 10 times lower than DEG price. Besides, its use as cleavage agent would solve, in some cases, waste generation problems of the biodiesel industry and would give value to this waste. The crude glycerol molecule contains three hydroxyl groups that allow the interchange reaction with the ester group of the urethane, giving as a result the recovered polyol contained in the original foam and a low weight carbamate ending in hydroxyl groups. First, transesterification reactions with viscoelastic PU foam waste were carried out using a molar excess of crude glycerol and provided a split-phase product with lower content of byproducts and transesterification agent than in the case of using the best transesterification agent described until this moment (DEG), since crude glycerol presents a higher dielectric constant. Moreover, also as a result of crude glycerol higher dielectric constant, glycolysis bottom phases free of polyol were obtained, increasing this way the net yield of the glycolysis process in terms of polyol recovery content (≈70 wt% vs 61wt%). Simultaneously, the feasibility of the glycolysis process of high resilience (HR) flexible polyurethane foam waste containing PU dispersion polyol (PIPA polyol) using crude glycerol was demonstrated. For this foam type, three different cleavage agents were assayed: DEG, glycerol 99% PS and crude glycerol. This way, the results obtained with crude glycerol were compared to those from DEG, which is the most common cleavage agent studied up to now; and, on the other hand, the use of glycerol 99% PS allowed to check the influence of the glycerol purity on the final results. As expected, two phases were obtained as a result of the chemolysis reactions of the HR foam. The upper phases were mainly constituted by the PIPA HR base polyol, whereas the bottom phases consisted of several glycolysis byproducts and the excess of the transesterification agents. Crude glycerol provided an upper phase with a lower concentration of byproducts and glycolysis agent and with a higher proportion of the HR recovered polyol (89 wt% vs 71 wt% when DEG was used), which presented a similar hydroxyl number to the raw HR base polyol without need of purification. Additionally, crude glycerol avoided the solubilisation of the HR recovered polyol in the bottom phase. Furthermore, even after carrying out a purification process of the upper phase obtained from DEG as cleavage agent, the upper phase directly obtained from the glycolysis reaction developed with crude glycerol presented a higher concentration of the HR recovered polyol (89 wt% vs 79 wt%). These results pointed out crude glycerol as the best alternative from a technical, environmental and economical point of view to carry out the degradation process of any kind of flexible PU foam waste. Moreover, flexible and rigid PU foams were successfully synthesized, regardless the cleavage agent employed and the type of PU waste treated, by using the glycolysis upper phases (refined product in the case of the reaction with DEG) and the bottom ones, respectively. However, in spite of the promising properties that had presented the recovered polyols (OH number, functionality, molecular weight, water content) it was considered a crucial task to carry out a study of influence of the percentage of these polyols on the properties of the flexible foams synthesized in order to confirm the feasibility of the replacement of virgin polyol by the recovered ones in the synthesis of new polyurethane foams. Hence, a detailed analysis of the physical, mechanical and structural properties of the polyurethane foams synthesized by means of recovered polyols coming from different types of flexible polyurethane waste scraps was carried out. The analyses of the cell size and morphology, apparent density, remaining deformation 50%, tensile strength and elongation at break; confirmed that it is possible to obtain flexible PU foams with physical and structural properties similar to those of a PU foam obtained with a raw flexible polyether polyol by means of employing mixtures of the recovered polyols with the raw ones. In fact, some properties (remaining deformation and tensile strength) experienced an improvement in comparison with the raw polyol based foam, depending on the PU foam recycled and the cleavage agent employed. Recovered polyols from viscoelastic foam yielded flexible PU foams with better compression set behaviour, regardless the cleavage agent used; crude glycerol let to obtain flexible foams with lower average cell size and new flexible PU foams with higher average cell size were achieved when the recovered polyol from graft polymeric conventional foam waste was the one employed. Moreover, an increase in the recovered polyol content provoked an improvement in the tensile properties of the PU foams synthesized, regardless the cleavage agent employed and the kind of PU foam recycled. Density and elongation values were nearly maintained constant, regardless the type of PU foam recycled and the cleavage agent employed. Therefore, high quality flexible PU foams were synthesized by means of employing recovered polyols from different kinds of flexible PU foams waste employing as cleavage agents DEG or crude glycerol. Finally, the economic assessment of the split-phase glycolysis process for the recycling of flexible polyurethane foams waste employing crude glycerol as cleavage agent, showed that the proposed process is economically feasible. This fact, together with the environmental and sustainability advantages associated to the process, demonstrated that the developed split-phase glycolysis process presents great improvements with respect to the glycolysis processes industrially established at present. In summary it can be concluded that, the split-phase glycolysis process proposed in this work is sustainable, environmental-friendly, suitable for the recycling of practically any kind of polyurethane flexible foam waste and economically viable, obtaining valuable products of both glycolysis phases, encouraging the industrial implantation of the split-phase glycolysis in the near future.