Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2024 Oct 16;9(43):43438–43446. doi: 10.1021/acsomega.4c04671

Polyurethane Waste Recycling: Thermolysis of the Carbamate Fraction

Marthe Nees †,*, Muhammad Adeel †, Lukasz Pazdur †, Matthew Porters †,‡, Christophe ML Vande Velde †, Pieter Billen †
PMCID: PMC11525496  PMID: 39494010

Abstract

graphic file with name ao4c04671_0008.jpg

The alcoholysis of polyurethane waste is currently being industrialized, making it one of the most advanced chemical recycling processes for polyurethanes. However, the recycling potential of the dicarbamate phase, which accounts for 10–40% of the polyurethane mass, is often disregarded, as mainly the polyol components are (partially) retrieved in many alcoholysis processes. In this study, we present a two-step recycling method in which the valuable carbamate fraction obtained in the initial alcoholysis step is transformed into an isocyanate-rich mixture through an additional thermolysis step. For this purpose, different carbamates were synthesized and thermolyzed, which showed that thermolysis with isopropyl-carbamate was the most favorable, obtaining a yield of 35%. As a result, the isocyanates obtained through thermolysis and the polyols obtained through alcoholysis can be reused as starting materials in polyurethane synthesis.

Introduction

Polyurethane (PU) is currently one of the world’s most versatile polymers, with a wide range of applications that are generally categorized into three groups: flexible foam, rigid foam, and CASEs (coatings, adhesives, sealants, and elastomers), with the first two groups dominating more than two-thirds of the PU market.2 Their versatile nature and the resulting extensive arsenal of applications are due to the different types of polyols and isocyanates that are used for the synthesis of polyurethane. Among the polyols, polyether polyols and polyester polyols are the most commonly employed. Toluene diisocyanate (TDI) and methylene diisocyanate (MDI) are the predominant isocyanates utilized. Isomeric variants of these isocyanates can be employed in polyurethane synthesis. Additionally, in the case of MDI, a polymeric form known as polymeric methylene diisocyanate (PMDI) finds a frequent use. PMDI is a liquid at room temperature and is therefore more commonly used in the synthesis of foams because it is easier to handle than MDI, which is solid at room temperature. To illustrate the tunability of polyurethane properties, when aliphatic isocyanates and long-chain polyols with a low number of functional groups are incorporated into the PU chain, the resulting PU will be flexible. If, on the other hand, aromatic isocyanates and low-molecular-weight polyols with a high number of functional groups are used, then a rigid PU is obtained. Most produced polyurethanes are thermosets, which contributes to their excellent durability. To date, most postconsumer polyurethane waste is disposed of in landfills, but as the circularity of materials and more specifically plastics becomes increasingly important, so does the research on better recycling methods.3 Three primary methodologies for recycling polyurethane (PU) currently exist: (1) mechanical, (2) thermochemical, and (3) chemical and biological recycling. In mechanical recycling, PU is mechanically fragmented into small pieces, washed, and subsequently reassembled through adhesive bonding. However, the present-day mechanical recycling of thermoset PU still fails to produce high-quality recycled PU due to the high degree of cross-linking. During thermochemical recycling of PU the urethane bonds are, generally, broken by applying high temperatures and high pressures. The main products are chemical products, energy, and fuels.4,5 Conversely, in chemical and biological recycling, the chemical bonds within PU are broken down, reverting to the fundamental building blocks of polyurethane. This process can be achieved by the use of traditional chemicals or by the use of microorganisms or isolated enzymes or a combination of both.6,7 Even though research into the chemical recycling of PU foam has been ongoing for decades and appears to be the most promising recycling technique, optimization is still required.8 Several chemical recycling techniques have already been investigated, such as hydrolysis, aminolysis, alcoholysis, and acidolysis.3,5,9 All of these chemical recycling techniques are designed to facilitate the dissociation of the urethane bond in PU, resulting in the generation of a polyol fraction and a derivatized isocyanate fraction. Until now, the primary focus of these chemical recycling approaches has predominantly centered around the isolation and utilization of the acquired polyol fraction, while the derivatized isocyanate fraction is often considered an undesired byproduct.3 However, with the growing interest in circular processes and the isocyanate fraction constituting 10–40% of the total polyurethane (PU) mass, there is a rising interest in the valorization of this fraction. An overview of the theoretical end products of each chemical recycling method is given in Figure 1.

Figure 1.

Figure 1

Open in a new tab

Schematic overview of different chemical recycling techniques of PU foam.

Consequently, investigations into the valorization of amines obtained after the hydrolysis of PU (Figure 1, path 1) have been initiated. Notably, the isolated amines can be transformed into isocyanates through the utilization of phosgene (Figure 1, path 5), subsequently enabling their reuse for the synthesis of “new” PU.10 While this method theoretically presents an ideal solution by using existing processes, several drawbacks need to be considered. First, addressing the carcinogenic nature of aromatics involves the preparation of extensive legal and regulatory documentation. Aromatic amines are presently not transported in bulk within the EU, based on our current understanding. Consequently, the recycler would have to perform an in-house phosgene process to form isocyanates. Second, the phosgene gas deployed for this process is highly toxic and is synthesized using Cl2, an energy-intensive molecule, which makes the phosgene route an energy-hungry process. The phosgene process also generates highly corrosive HCl, adding an additional layer of complexity to the recycling procedure. In addition to investigating the hydrolysis reaction, research has been conducted on the acidolysis reaction (Figure 1, path 4). In this process, the urethane bond is cleaved using a carboxylic acid, leading to the formation of a polyol fraction and an amide fraction.11 Despite acidolysis being identified as one of the fastest chemical depolymerization routes and research demonstrating substantial recovery of the polyol phase, the valorization of the residual amide fraction has been underexplored. As a result, complete recycling of PU by acidolysis has not been achieved. Another chemical recycling technique involves the aminolysis of PU (Figure 1, path 3), resulting in the generation of a polyol fraction and a urea fraction.12 The generated urea bond is inherently reversible and could thermally dissociate to form amine and isocyanate molecules (Figure 1, path 9).13 Via this process, the potential recovery of isocyanate for reuse in polyurethane (PU) synthesis can be obtained. However, the high reactivity of isocyanates toward amines to form ureas makes achieving the reverse reaction challenging.1 Among chemical recycling techniques, the alcoholysis reaction of PU stands out as one of the most renowned and widely employed methods (Figure 1, path 2). Typically, polyfunctional alcohols are utilized in a transcarbamation reaction, resulting in the formation of polyols and carbamates as reaction products. Extensive research has been conducted on the isolation of polyols obtained by this technique, exploring their reuse in PU synthesis. Additionally, preliminary investigations have commenced into the valorization of the carbamate fraction. For instance, Zahedifar et al. conducted research on the hydrolysis of the carbamate fraction to generate amines (Figure 1, path 7).14 This approach proposes converting the formed amines through the phosgene process into isocyanates, which can then be used in PU synthesis. However, as previously noted, this route has several drawbacks. Another study by Zhao and Semetey demonstrated that the obtained carbamate can be directly transformed into PU through transcarbamation with a polyol (Figure 1, path 8).15 While this pathway enables valorization of the carbamate fraction and the synthesis of “new” PU, the direct synthesis poses limitations for the applicability of the synthesized PU due to the requirements of the different PU producers for specific formulations. Nevertheless, this research provides an alternative method for valorizing the carbamate fraction.

While some of the abovementioned methods could provide a solution to the recycling of PU, they also bring relatively major drawbacks and/or challenges. First, the majority of recycling methods primarily focus on optimizing the recovery of polyol. Second, even when attention is given to the derivatized carbamate fraction, it typically presents several drawbacks. These include the formation of amides and ureas, which currently cannot regenerate isocyanates, the direct synthesis process, which imposes limitations on applicability, and the generation of amines, which are carcinogenic and rely on toxic compounds for the regeneration of isocyanates. This research therefore proposes an alternative method to directly synthesize isocyanates via the carbamate fraction after alcoholysis. Similarly to the urea bond, the urethane bond is also thermally breakable (Figure 1, path 6). Due to the lower nucleophilicity of alcohols compared to amines, dissociation is more readily achievable for carbamate than for urea. The thermolysis reaction of carbamates has already been extensively described in the literature as a phosgene-free synthesis route to make isocyanates.16−22 The highest yields obtained from these different thermolysis methods vary between 85 and 95%. Consequently, this method could ensure that the isocyanates can be directly recovered from the formed carbamates after the alcoholysis reaction in a relatively straightforward way. However, it is important to note that the yields reported in these studies were achieved under conditions of either very high dilution or the use of very specific catalysts or monofunctional molecules. These conditions limit the applicability of these thermolysis methods to industrial processes.

In this research, we build upon the already known alcoholysis of PU to create a short, yet completely circular, recycling pathway without the use of toxic substances, as illustrated in Figure 2. With these sustainability, circularity, and industrial requirements in mind, a two-step alcoholysis–thermolysis process has been developed.23 In this process, the alcoholysis reaction serves as the initial step to recover the valuable polyols and isolating them before the thermal decomposition. In the subsequent step, the produced carbamate undergoes thermolysis. This thermolytic process regenerates an isocyanate and the initial alcohol used in the first step, which can be recycled.

Figure 2.

Figure 2

Open in a new tab

Schematic overview of the proposed alcoholysis–thermolysis recycling method.

The main goal of this research was to investigate the thermolysis reaction. Research has demonstrated that the choice of alcohol as a blocking agent for isocyanate impacts the deblocking temperature during thermolysis reactions. While the existing literature has explored the influence of specific blocking agents on thermolysis, there remains a gap in understanding the effect of simple alkyl chains on MDI-based carbamates, which are crucial for recycling economics. As noted by Rolph et al., comprehensive testing and comparison of various blocking agents are necessary to establish suitable thermolysis conditions.24 However, it is crucial to highlight that studies comparing different blocking agents primarily focus on determining the deblocking temperature rather than assessing conversion, selectivity, or yield. To study the effect of the alkyl group, attached to the carbamates, on the conversion, yield, and selectivity of the thermolysis reaction, model carbamates were synthesized and subsequently thermolyzed. The model carbamates were prepared through a reaction between an alcohol and an isocyanate. The selection of the isocyanate involved considering the most prevalent isocyanates used in polyurethane (PU) foam, TDI, and MDI. Ultimately, MDI was chosen due to its higher boiling point, reducing the chance of evaporation during the thermolysis reaction conducted at elevated temperatures. Given that the structure of the carbamate dictates the temperature at which the dissociation occurs, the selection of the alcohol employed in the synthesis of model carbamates will influence the thermolysis temperature. Because generally the carbonyl group of the blocked isocyanate has a partial positive charge, the bond between the carbonyl carbon and the blocking molecule is labile. The lower this charge difference between the carbonyl group and the blocking molecule, the weaker the bond. Thus, decreasing the nucleophilicity of the blocking molecule reduces the charge density, making the bond more labile.1 In addition to the electronic effect, steric effects also affect the strength of the bond. Therefore, the use of secondary alcohols can also lower the thermolysis temperature.1 Additionally, the alcohol and isocyanate products need to be separated immediately to avoid the reverse reaction. Therefore, a sufficient boiling point difference between the used alcohol and isocyanate is required, constituting a second selection criterion for the alcohols. As previously mentioned, the alcohol selected for the alcoholysis reaction does not only affect the thermolysis temperature but clearly also has an impact on the alcoholysis reaction itself. Since the alcoholysis reaction takes place at higher temperatures, a low boiling point of the alcohol will be disadvantageous during alcoholysis. Alcoholysis with alcohols with boiling points lower than 200 °C will have to take place in a high-pressure vessel. Consequently, the alcoholysis–thermolysis process requires a balance between the optimal alcohol for the alcoholysis reaction and the optimal alkyl group in the carbamate structure for the thermolysis reaction. Four different alcohols, octanol, butanol, methanol, and isopropanol, were selected to form the corresponding octyl-carbamate, butyl-carbamate, methyl-carbamate, and isopropyl-carbamate (Figure 3).

Figure 3.

Figure 3

Open in a new tab

Chemical structures of the dicarbamate used in this study.

Ethylene glycol is a difunctional alcohol and one of the most frequently used alcohols for the alcoholysis of PU. However, if the thermolysis reaction is performed with ethylene glycol as the incorporated alcohol, a repolymerization reaction may occur between ethylene glycol and the generated diisocyanate, just as during the synthesis of PU. To avoid this polymerization reaction, only monoalcohols were considered. Octanol has a similar boiling point to ethylene glycol (±195 °C) and would therefore also be suitable for the alcoholysis reaction, which is typically performed around 200 °C. Butanol was chosen because it has a lower boiling point of 117.7 °C and could therefore be more easily removed during the thermolysis. Methanol and isopropanol were chosen, as they separate even more efficiently during the thermolysis reaction owing to their lower boiling points. Additionally, using isopropanol provides the benefit of the lower dissociation temperature of secondary alcohols relative to primary alcohols.1 After testing the thermolysis reaction with the different carbamates, a preliminary test of the complete alcoholysis–thermolysis recycling process was successfully conducted using two of the four designated alcohols. The process was then applied to PU mattress foam. This initial trial demonstrated the ability to successfully recover isocyanates from commercially available polyurethane flexible foam.

Experimental Section

Chemicals

4,4′-Methylenebis(phenylisocyanate) (99%), octanol (98%), and methanol (99,8%) were obtained from Acros Organics, isopropanol (99,8%) was obtained from Honeywell, 1-butanol (99,9%) and acetone (99,8%) were obtained from Sigma-Aldrich, deuterated acetone (99,9 atom % and 0,03% TMS) and tetrahydrofuran (99%) were obtained from Thermo Scientific, n-heptane (99%) was obtained from Chemsolute, and ethyl acetate (99,8%) was obtained from Chem-lab.

Synthesis of Model Carbamates

MDI (0.02 mol, 5 g), the used alcohol (0.2 mol), and a magnetic stirring bar were brought into a round-bottom flask (100 mL). The reaction mixture was then stirred at room temperature for 2 h. Thereafter, the reaction mixture was cooled to room temperature. The solid reaction product was washed with water and filtered via vacuum filtration. The remaining water was removed using a rotary evaporator at 60 °C and 2 mbar. The dicarbamate product is analyzed by 1H NMR and 13C NMR (400 MHz) and MS.

Thermolysis Reaction

The dicarbamate (0.5 g) and a magnetic stirring bar were placed in a round-bottomed flask (10 mL). When the heating mantle reached the desired temperature, the flask was placed in the heating mantle. The reaction mixture was then stirred for 5/10/15 min. The reaction mixture was cooled by placing the flask in an ice bath. The product mixture was analyzed by FTIR and SEC.

Procedure for Quantitative Analysis

After the thermolysis reaction, 5 mL of dry acetone was added to dissolve the solid mixture and 2 mL of methanol/isopropanol was added to the reaction mixture to protect the isocyanate groups present in the mixture. The reaction mixture was stirred for 15 min, and solvents were evaporated with a rotary evaporator.

Alcoholysis of Flexible Mattress PU Foam

An autoclave with a pressure capacity of up to 200 bar was used in this experiment. Twenty grams of isopropanol and 6.0 g of flexible foam were charged to the autoclave reactor, and the reactor was sealed. The autoclave was purged with argon for 15 min. The pressure of the autoclave was increased to 40 bar with argon. The reaction mixture was heated to 200 °C for 2 h, making the pressure rise to 80 bar. After the reaction, the reaction mixture was cooled to room temperature, and pressure was released. The remaining 2-propanol was removed using a rotary evaporator. The reaction mixture is then analyzed by GPC and HPLC.

Results and Discussion

The decomposition of the synthesized model carbamates was investigated by TGA to estimate the optimal thermolysis temperature. When the urethane bond of the carbamate is broken, the released alcohol evaporates, and a decrease in mass is observed. The TGA plot in Figure 4 illustrates that the methanol-based carbamate, specifically methyl-carbamate, starts to decompose at slightly lower temperatures. This can be explained by the electronic effects of the methyl chain of methanol on the dissociation reaction. With a shorter alkyl chain, the inductive electron-donating effect of the alkyl chain is smaller than in longer alkyl chains, and so, the stabilizing effect on the urethane bonding between the carbonyl carbon and the oxygen decreases. Consequently, this relatively small decrease in stabilization can cause a notable decrease in the dissociation temperature. However, steric effects also contribute to this behavior, as the extended chain of octanol introduces heightened steric hindrance, causing a lower dissociation temperature for octyl-carbamate than for butyl-carbamate, for instance. Because methyl-carbamate dissociates at a lower temperature than the other carbamates, the liberated isocyanate has more time to react with the other compounds present and forms a thermoset polymer. The TGA curves show a weight loss commencing at approximately 250 °C, with a significant decline around 300 °C, which is around the boiling point of MDI (314 °C). Consequently, three distinct temperatures, namely, 250, 275, and 300 °C, were chosen for the thermolysis reaction, each with durations of 5, 10, and 15 min. The thermolysis procedure was executed in a round bottom flask without solvent or catalyst, maintaining a pressure around 15 mbar to enhance the removal of the alcohol, while making sure that the pressure is not too low to vaporize the generated MDI. The analysis of the thermolysis products remaining in the round-bottom flask was performed using gel permeation chromatography (GPC) and Fourier transform infrared spectrometry (FTIR) (Figures S12–S15, Supporting Information). FTIR analysis of the thermolysis mixture, as depicted in Figure 4, revealed that even after 5 min at 250 °C, isocyanate groups could be identified in the thermolysis mixture, as evidenced by the prominent absorption peak in the range of 2250 to 2270 cm–1. The GPC chromatograms display three distinct peaks corresponding to the starting carbamate, the intermediate carbamate-isocyanate, and the MDI. Additional peaks observed at lower retention times are presumed to represent oligomeric compounds, which will be discussed in subsequent sections.

Figure 4.

Figure 4

Open in a new tab

TGA curves of octyl-carbamate, butyl-carbamate, methyl-carbamate, and isopropyl-carbamate and their FTIR spectrum after a thermolysis reaction at 250 °C for 5 min.

To determine the yield of the thermolysis reaction, a quantitative analysis protocol using an external standard HPLC method was established. The analysis was performed on the entire reaction mixture, as isolating MDI at this scale and stage of the study would be too challenging. However, prior to performing the HPLC method, derivatization of the product mixture was necessary to prevent the reactive isocyanate groups from reacting with the used eluent, the column, or the moisture present in the air during sample preparation. As shown in Figure 5, the derivatization was achieved by adding an alcohol back into the reaction mixture to block the reactive isocyanate group by forming a new carbamate. For the thermolysis with octyl-carbamate, butyl-carbamate, and isopropyl-carbamate, methanol was used as a protecting group, and in the case of methyl-carbamate, isopropanol was used. The HPLC chromatogram in Figure 5 shows (1) one peak corresponding to the protected MDI, (2) peaks that corresponds to the intermediate, which has one protected NCO functionality and one original carbamate functionality, and (3) peak that corresponds to the starting carbamate, and the remaining peaks are assumed to be oligomers formed during the reaction. However, the exact structure of these oligomers has yet to be determined.

Figure 5.

Figure 5

Open in a new tab

General blocking reaction of the isocyanate groups and HPLC chromatograms of thermolysis mixtures of isopropyl-carbamate.

The results of the quantitative analysis are listed in Table 1. The conversion was calculated by dividing the amount of MDI by the initial amount of the starting carbamate, and the yield was calculated by dividing the obtained MDI by the maximum theoretical MDI that could be obtained and the selectivity was calculated by dividing the amount of MDI by the total amount of obtained products. The results of the thermolysis reaction conducted at 275 °C for 15 min, as well as those at 300 °C for 10 and 15 min, are not presented, as these conditions produced a thermoset material that could not be dissolved. A first general observation is the large gap between conversion and yield in all thermolysis reactions. This gap may be due to the reactivity of the NCO group that allows many side reactions to occur.25 A second trend is that conversion increases at higher temperatures and longer reaction times. For the thermolysis of octyl-carbamate, butyl-carbamate, and isopropyl-carbamate, this also applies to the yield. However, for the thermolysis of methyl-carbamate, longer reaction times do not increase the yield. This can be explained by the fact that methyl-carbamate has a lower dissociation temperature and therefore dissociates faster, giving the released NCO groups more time to react with other components present in the reaction mixture.26 The highest yields are obtained when isopropyl-carbamate is used in the thermolysis reaction. This can be explained by the favorable electronic effects of the short alkyl chain, the favorable steric effects of the secondary alcohol, and the efficient elimination of the low-boiling IPA from the reaction mixture.1 The highest yield is observed when isopropyl-carbamate is thermolyzed at 275 °C for 10 min. These observations are grounded in the general consensus within the field regarding the deblocking temperature of alcohols. However, a comprehensive study providing data on the effect of various alcohols on the thermolysis reaction is currently lacking. This gap in the literature may be attributed to the complexity of the thermolysis process, where multiple variables can influence the reaction. Only some general guidance may be drawn from the published comparisons available.26,27

Table 1. Results of HPLC Quantification Method for the Different Thermolysis Reactionsa.

carbamate temp. (°C) time (min) conv. (%) yield (%) select. (%)
octyl-carbamate 250 5 44 0.2 0.2
  250 15 56 0.7 0.6
  275 5 51 1.4 1.4
  275 10 69 14 9.6
  300 5 65 6.9 5.2
butyl-carbamate 250 5 31 0.5 1.1
  250 15 46 5.1 7.0
  275 5 41 3.9 6.1
  275 10 71 19 17
  300 5 63 12 12
methyl-carbamate 250 5 50 13 21
  250 15 89 4.9 4.3
  275 5 85 9.6 8.9
  275 10 98 7.0 5.7
  300 5 94 20 17
isopropyl-carbamate 250 5 24 0.5 1.5
  250 15 85 14 11
  275 5 54 9.4 12
  275 10 91[a] 35[b] 26[c]
  300 5 88 11 8.7
Open in a new tab
a

Average value of four measurements with a standard deviation of [a] ± 3.3, [b] ± 5.4, and [c] ± 3.9.

Due to the high reactivity of the isocyanate groups and the high temperature during the thermolysis reaction, the formation of byproducts during the thermolysis reaction is inevitable. Figure 5 shows the HPLC chromatogram, which reveals several peaks in addition to those of isopropylcarbamate (starting compound) and methylcarbamate (blocked MDI). The first identifiable peak (peak 2) corresponds to the intermediate urethane isocyanate (Figure 6A), which has a higher thermal stability than the carbamate and is therefore a more abundant product than MDI. The stability of the intermediate is evidenced by the strong peak in the HPLC chromatogram and confirmed in the literature.1 However, the main limitation of the thermolysis reaction is the formation of various byproducts resulting from the generated isocyanates at high temperatures (Figure 6B). One possible byproduct is polycarbodiimide, which forms due to polycondensation of diisocyanates with the elimination of carbon dioxide. The presence of the carbodiimide functionality can be confirmed by FTIR, with a characteristic peak at 2100 cm–1 originating from the R–N=C=N–R stretch.28 Isocyanurate, resulting from trimerization of isocyanates, is a common byproduct formed at higher temperatures (>250 °C) and is generally considered to be thermally stable.1 Isocyanates are highly reactive to molecules containing acidic hydrogen, such as starting carbamate. The formed isocyanate can react with the carbamate to form an allophanate, which, in turn, can react with an isocyanate group to form an oligomeric structure. In addition to the abovementioned side reactions, several other reactions are possible, which contribute to the complexity of the thermolysis mixture. In the case where other impurities would be present such as additives present in the PU foam or residual byproducts of alcoholysis, the number of possible byproducts increases significantly.

Figure 6.

Figure 6

Open in a new tab

(A) Two reaction steps of the thermolysis reaction and (B) possible side reactions of MDI during the thermolysis reaction.1

After the investigation of thermolysis of model carbamates, a proof of concept of the entire two-step recycling process was performed, starting from PU foam to isocyanates and polyols, that consists of two steps and an intermediate separation: the alcoholysis reaction using isopropanol, isolation of the formed dicarbamate, and thermolysis of the dicarbamate. The results of this proof-of-concept run were analyzed using HPLC, as seen in Figure 7. The alcoholysis was tested on 5.42 g of flexible PU foam.

Figure 7.

Figure 7

Open in a new tab

Recycling of flexible PU mattress foam with the alcoholysis–thermolysis process analyzed via HPLC with (a1) the reaction products after the alcoholysis reaction, (a2) the HPLC chromatogram of the reaction mixture after the alcoholysis reaction, (b1) the reaction products after the thermolysis and blocking reaction with MeOH, and (b2) the HPLC chromatogram of the reaction mixture after the thermolysis and blocking reaction with MeOH.

Isopropanol was selected because of the high yields in the previously studied thermolysis on model carbamates. The alcoholysis of PU foam was carried out at 200 °C for 2 h under an argon atmosphere. Since the boiling point of isopropanol is 82 °C, this alcoholysis was performed under 40 bar of argon pressure in an autoclave, increasing to 80 bar during the reaction. The composition of the reaction mixture was analyzed by means of GPC and HPLC, and the corresponding HPLC chromatogram is shown in Figure 7a2. The first peak at 2.9 min corresponds to compound 1, methylene diphenyl diamine (MDA), a side product formed during the alcoholysis reaction. This side product is formed due to the presence of urea compounds in the PU foam. The peak at 5.4 min, compound 2, is an MDI derivative where both an amine group and carbamate group are formed due to the presence of both urea and carbamate compounds in the foam. The peak at 7.7 min is compound 4, isopropyl-carbamate. Quantitative analysis showed that 0.573 g of isopropyl-carbamate was present in the reaction mixture, which corresponds to 10.3 mass % of the reaction mixture. Besides these three constituents, the alcoholysis mixture also includes the polyol. However, as these polyols, amounting to 70% of the mixture, are not UV active, the DAD detector coupled to HPLC was unable to detect their presence. The rather low amount of isopropyl-carbamate is caused by two different factors. A first factor is that the foam used was synthesized using polymeric MDI (PMDI) instead of MDI. PMDI consists usually of only 30–80% monomeric MDI supplemented by higher-molecular-weight MDI homologues,29 while we mainly quantified on the basis of monomeric MDI that can be isolated for the thermolysis reaction. The second factor is the blowing agent used during the foaming process; if a foam is formed by using water, it creates urea compounds in the foam that are converted to amines rather than carbamates during alcoholysis.

After the formation and quantity of the carbamate were confirmed, a separation using flash chromatography was established where 0.26 g of the isopropyl-carbamate was obtained and used in the thermolysis reaction. The isolated isopropyl carbamate was analyzed via 1H NMR, which confirmed its identity and purity. Given that the obtained yield is below the customary 0.5 g of standard employed in the previous thermolysis reactions, the sample containing the isolated carbamate after the alcoholysis reaction was augmented to 0.5 g by incorporating pure isopropyl carbamate. This decision was made since scaling up the alcoholysis reaction is not possible due to the limited capacity of the reactor and ongoing investigations into a more efficient separation method. After the isolation of isopropyl-carbamate, it was subjected to a thermolysis reaction. This reaction is carried out at 275 °C for 10 min and with a pressure below 15 mbar. The analysis of the thermolysis products is performed by means of GPC, FTIR (Supporting Information), and HPLC (Figure 7b2), which confirmed the presence of MDI in the reaction mixture by the formation of methyl-carbamate after the reaction of MDI with methanol, indicated by the peak at 1.7 min. Quantitative analysis showed that 0.095 g of MDI was present in the reaction mixture, which corresponds to 34 mass % of the reaction mixture. Despite being able to recover MDI from the PU foam, several important considerations must be made in light of this outcome: first, as previously mentioned, the foam contains PMDI instead of monomeric MDI. While this results in a lower quantity of monomeric MDI after the process, it does not necessarily imply a lower isocyanate yield after the alcoholysis–thermolysis process. An initial FTIR analysis of a thermolysis mixture of PMDI-based carbamates after a 10 min thermolysis reaction at 275 °C revealed a significant isocyanate peak. A second point to consider in this conclusion is the presence of urea linkages in the PU foam, which during the alcoholysis reaction contribute to the formation of amines. The quantity of produced amines is consequently dependent on the amount of urea compounds in the PU foam, which is in turn dependent on the quantity of water used during foam blowing. Therefore, foams with a substantial amount of urea compounds may be less or not suitable for the alcoholysis–thermolysis process when compared to foams with little or no urea compounds. A third consideration is the optimization of the separation technique from flash chromatography to, ideally, liquid–liquid extraction.

Conclusions

The thermolysis of various carbamates was studied for incorporation into a two-step recycling process for PU, in combination with alcoholysis. The reactions were investigated using TGA, GPC, FTIR, NMR, and HPLC to determine the optimal thermolysis conditions and the yield and conversion of the reactions. TGA provided insight into the optimal thermolysis temperature as a starting point for the research. FTIR and GPC were used to analyze the reaction mixture, and after protection of the isocyanate groups with an alcohol, HPLC analysis allowed for the determination of the yield of the reaction. The results showed that the conversion and yield generally increased with higher temperatures and longer reaction times. The highest yield was observed for isopropyl-carbamate thermolysis at 275 °C for 10 min, and this reaction was repeated four times with an average yield of 35 ± 5.4%. While the obtained yields may appear lower compared to other studies in the field, it is worth noting that the latter often relies heavily on solvent consumption and catalyst use, significantly compromising feasibility and environmental impact. Moreover, the promising yields achieved are noteworthy, especially considering that the functionality/reactivity of the isocyanate is a more critical factor than the concentration of pure MDI. After all, these recyclates will likely end up as a component for either adhesives or sealants, which offer more degrees of freedom,30 and the strategy is thus completely different from those in commodities in manufacturing. The study also revealed the presence of various byproducts, including polycarbodiimide, which was confirmed by FTIR. The investigation of thermolysis of model carbamates has led to the first trial run of the two-step recycling process from mattress PU foam to isocyanates consisting of an alcoholysis reaction, the isolation of formed dicarbamate, and a thermolysis of the dicarbamate. The analysis of the thermolysis products confirmed the formation of MDI. While the yield from the conducted trial run was rather limited, it can be significantly improved by selecting foams that are a better fit for this recycling method and further optimizing the existing process. The envisioned recycling process is expected to be a sustainable approach for the recycling of PU foam waste that valorizes the carbamate fraction without having to use an energy-intensive process with toxic reagents.

Acknowledgments

This work was supported by the University of Antwerp by BOF BOF-DOCPRO4 Cyclo2PUR (nr. 44630) and IOF POC CycloPUR (nr. 44253).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.4c04671.

  • Used chemicals and instruments, experimental procedures, NMR spectra, IR spectra, MS spectra, and HPLC and GPC chromatograms (PDF)

The authors declare no competing financial interest.

Supplementary Material

ao4c04671_si_001.pdf (1.9MB, pdf)

References

  1. Delebecq E.; Pascault J. P.; Boutevin B.; Ganachaud F. On the Versatility of Urethane/Urea Bonds: Reversibility, Blocked Isocyanate, and Non-isocyanate Polyurethane. Chem. Rev. 2013, 113 (1), 80–118. 10.1021/cr300195n. [DOI] [PubMed] [Google Scholar]
  2. ″in Europe, Middle East and Africa (EMEA) - 16th ed. (including Raw Materials Vol.),″ 2023, doi: 10.1016/0012-1606(69)90033-5. [DOI] [Google Scholar]
  3. Fonseca L. P.; et al. Reducing the carbon footprint of polyurethanes by chemical and biological depolymerization: Fact or fiction?. Curr. Opin. Green Sustainable Chem. 2023, 41, 100802 10.1016/j.cogsc.2023.100802. [DOI] [Google Scholar]
  4. Kemona A.; Piotrowska M. Polyurethane Recycling and Disposal: Methods and Prospects. Polymers 2020, 12 (8), 1752. 10.3390/polym12081752. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Rossignolo G.; Malucelli G.; Lorenzetti A. Recycling of polyurethanes: where we are and where we are going. Green Chem. 2024, 26 (3), 1132–1152. 10.1039/D3GC02091F. [DOI] [Google Scholar]
  6. Liu J.; et al. Biodegradation and up-cycling of polyurethanes: Progress, challenges, and prospects. Biotechnology advances 2021, 48, 107730 10.1016/j.biotechadv.2021.107730. [DOI] [PubMed] [Google Scholar]
  7. Gallorini R.; et al. Subcritical Hydrothermal Liquefaction as a Pretreatment for Enzymatic Degradation of Polyurethane. ACS omega 2022, 7 (42), 37757–37763. 10.1021/acsomega.2c04734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Simón D.; Borreguero A. M.; de Lucas A.; Rodríguez J. F. Recycling of polyurethanes from laboratory to industry, a journey towards the sustainability, Waste Management. Waste Manage. 2018, 76, 147–171. 10.1016/j.wasman.2018.03.041. [DOI] [PubMed] [Google Scholar]
  9. Zia K. M.; Bhatti H. N.; Bhatti I. A. Methods for polyurethane and polyurethane composites, recycling and recovery: A review. React. Funct. Polym. 2007, 67 (8), 675–692. 10.1016/j.reactfunctpolym.2007.05.004. [DOI] [Google Scholar]
  10. Six C.; Richter F.. Isocyanates, Organic, In Ullmann’s Encyclopedia of Industrial Chemistry, 2003, doi: 10.1002/14356007.a14_611. [DOI] [Google Scholar]
  11. Grdadolnik M.; et al. Insight into chemical recycling of flexible polyurethane foams by acidolysis. ACS sustainable chemistry & engineering 2022, 10 (3), 1323–1332. 10.1021/acssuschemeng.1c07911. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Grdadolnik M.; et al. Chemical recycling of flexible polyurethane foams by aminolysis to recover high-quality polyols. ACS sustainable chemistry & engineering 2023, 11 (29), 10864–10873. 10.1021/acssuschemeng.3c02311. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Honorien J.; Fournet R.; Glaude P. A.; Sirjean B. Theoretical Study of the Thermal Decomposition of Urea Derivatives. J. Phys. Chem. A 2022, 126 (36), 6264–6277. 10.1021/acs.jpca.2c04291. [DOI] [PubMed] [Google Scholar]
  14. Zahedifar P.; Pazdur L.; Vande Velde C. M.; Billen P. Multistage chemical recycling of polyurethanes and dicarbamates: A glycolysis–hydrolysis demonstration. Sustainability 2021, 13 (6), 3583. 10.3390/su13063583. [DOI] [Google Scholar]
  15. Zhao L.; Semetey V. Recycling polyurethanes through transcarbamoylation. ACS omega 2021, 6 (6), 4175–4183. 10.1021/acsomega.0c04855. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Wang P.; Liu S.; Deng Y. Important Green Chemistry and Catalysis: Non-phosgene Syntheses of Isocyanates – Thermal Cracking Way. Chin. J. Chem. 2017, 35 (6), 821–835. 10.1002/cjoc.201600745. [DOI] [Google Scholar]
  17. Grzegorz Lewandowski E. M. Thermal decomposition of methylene-4,4′-di(ethylphenyl-carbamate) to methylene-4,4′-di(phenylisocyanate). J. Hazard. Mater. 2005, 119 (1–3), 19–24. 10.1016/j.jhazmat.2004.11.016. [DOI] [PubMed] [Google Scholar]
  18. Wang Q.; Kang W.; Zhang Y.; Yang X.; Yao J.; Chen T.; Wang G. Solvent-free thermal decomposition of methylenediphenyl di(phenylcarbamate) catalyzed by nano-Cu2O. Chin. J. Catal. 2013, 34 (3), 548–558. 10.1016/S1872-2067(11)60494-4. [DOI] [Google Scholar]
  19. Hyun M. J.; Shin M.; Kim Y. J.; Suh Y. W. Phosgene-free decomposition of dimethylhexane-1,6-dicarbamate over ZnO. Res. Chem. Intermed. 2016, 42, 57–70. 10.1007/s11164-015-2224-x. [DOI] [Google Scholar]
  20. Dai Y.; et al. Phosgene-free synthesis of phenyl isocyanate by catalytic decomposition of methyl N-phenyl carbamate over Bi 2 O 3 catalyst. Catalysis letters 2008, 123, 307–316. 10.1007/s10562-008-9424-6. [DOI] [Google Scholar]
  21. Zhao X.; Wang Y.; Wang S.; Yang H.; Zhang J. Synthesis of MDI from dimethyl carbonate over solid catalysts. Industrial & engineering chemistry research 2002, 41 (21), 5139–5144. 10.1021/ie020084f. [DOI] [Google Scholar]
  22. Uriz P.; Serra M.; Salagre P.; Castillon S.; Claver C.; Fernandez E. A new and efficient catalytic method for synthesizing isocyanates from carbamates. Tetrahedron Lett. 2002, 43 (9), 1673–1676. 10.1016/S0040-4039(02)00094-1. [DOI] [Google Scholar]
  23. Pieter B.; Lukasz P.; Christophe V. V.. Two-stage chemical recycling of polyurethanes,″ Belgium, 2023. [Google Scholar]
  24. Rolph M. S.; Markowska A. L.; Warriner C. N.; O’Reilly R. K. Blocked isocyanates: From analytical and experimental considerations to non-polyurethane applications. Polym. Chem. 2016, 7 (48), 7351–7364. 10.1039/C6PY01776B. [DOI] [Google Scholar]
  25. Prenveille T.; Garreau C.; Matner M.; Dijkstra D.; Oppermann W.; Johannsmann D. Reactivity of urethanes at high temperature: Transurethanization and side reactions. J. Polym. Sci., Part A: Polym. Chem. 2019, 57 (5), 621–629. 10.1002/pola.29301. [DOI] [Google Scholar]
  26. Wicks D. A.; Wicks Z. W. Jr Blocked isocyanates III: Part A. Mechanisms and chemistry. Prog. Org. Coat. 1999, 36 (3), 148–172. 10.1016/S0300-9440(99)00042-9. [DOI] [Google Scholar]
  27. Guo Z.; Ding X.; Wang Y. How To Get Isocyanate?. ACS omega 2024, 9 (10), 11168–11180. 10.1021/acsomega.3c10069. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. da Silva J. E. E.; Alarcon R. T.; Gaglieri C.; Magdalena A. G.; da Silva-Filho L. C.; Bannach G. New thermal study of polymerization and degradation kinetics of methylene diphenyl diisocyanate. J. Therm. Anal. Calorim. 2018, 133, 1455–1462. 10.1007/s10973-018-7211-1. [DOI] [Google Scholar]
  29. 4,4′-Methylene diphenyl isocyanate (MDI) and polymeric MDI″ (PMDI) [MAK Value Documentation, 1997], In The MAK-Collection for Occupational Health and Safety: Annual Thresholds and Classifications for the Workplace, 2012. pp 66–96 10.1002/3527600418.mb10168stae0008. [DOI]
  30. Kovacs A.; et al. Hybrid substitution workflows should accelerate the uptake of chemical recyclates in polymer formulations. Current Opinion in Green and Sustainable Chemistry 2023, 41, 100801 10.1016/j.cogsc.2023.100801. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao4c04671_si_001.pdf (1.9MB, pdf)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES