The Contribution of Commercial Metal Amides to the Chemical Recycling of Waste Polyesters

Abstract

Simple and commercially available metal amides are investigated as catalysts for the chemical depolymerization of polyesters of commercial interest such as polylactide (PLA) and polyethylene terephthalate (PET) via alcoholysis. In the alcoholysis reactions performed with methanol or ethanol at room temperature, Zn, Mg, and Y amides showed the highest activities, while the amides of group 4 metals were revealed as poor catalysts. During the ethanolysis of PLA at higher temperatures and the glycolysis of PET, the good activity of the Zn amide was preserved, while for Mg and Y amides, a significant decrease was observed. The reaction temperature had an opposite effect on the performance of group 4 amides, with the Zr amide revealed to be the best catalyst in the PET glycolysis, reaching activities comparable to the best ones reported in the literature for metal catalysts (78% BHET yield within one hour at 180 °C). These studies represent new opportunities for the sustainable recycling of plastics, which are currently being used on a large scale, and provide significant contributions to the design of a circular economy model in the plastic industry.

1. Introduction

The huge volumes of plastic that are placed on the market for single-use products every year, in combination with the absence of a correct management strategy for their end-of-life, have caused dramatic problems of environmental pollution, demonstrating the unsustainability of the current linear productive model in the long term [1,2,3]. However, due to their versatility and low costs, plastics are irreplaceable materials, and the hypothesis of a plastic-free society is unrealistic [4,5].

To overcome this dichotomy, the efforts of scientific research have focused on the development of both more sustainable plastics, biodegradable and/or obtainable from renewable resources [6,7,8], and strategies for their chemical recycling [9]. In this context, aliphatic polyesters represent the most promising materials as alternatives to traditional petroleum-based polyolefins [10].

In the last two decades, the ring-opening polymerization (ROP) of lactones has received a great impetus as a method of election for the synthesis of polyesters [7,11]. Simple amides or alkoxides of several metals were tested as catalysts for ROP only in initial investigations, showing activity and control efficiency lower than those obtained with the industrial catalyst Sn (Oct)₂ [12,13,14,15,16].

Much more numerous are the studies concerning the catalysts based on coordination complexes in which the central metal is coordinated to an ancillary ligand and to one or more labile ligands that act as the initiating groups because these showed improved abilities in terms of the control of polymerization processes [17,18]. For these systems, the ancillary ligands have a crucial role in influencing the catalytic behavior of reactive centers and promoting punctual control over macromolecular parameters, namely the chain-end group structures, molecular masses and their distributions, and stereoregularity [19,20,21,22]. On the other hand, the synthesis of architectonically sophisticated ligands has a profound impact on the costs and sustainability of the whole catalytic process. Thus, the use of catalysts based on non-toxic metals with simple structures and/or commercially accessible is highly desirable.

However, to design a virtuous life cycle for plastics, in addition to greener processes for their production, adequate treatments of end-of-life plastic waste must be planned [23,24,25,26,27].

Beyond ecological considerations, an estimated economic loss of 120 billion dollars per year emanates from plastic waste, which currently cannot be recycled [28]. While the mechanical recycling of plastic waste provides recycled materials with lower quality, the chemical recycling and upcycling of polymers may offer new opportunities to contain the economic and environmental costs of this sector.

Some recent studies described zinc and magnesium complexes as efficient catalysts for the degradation of polyesters [29,30,31,32,33]. Among these, Wang reported zinc bis[bis(trimethylsilyl)amide], (Zn(HMDS)₂), as a simple and efficient ligand-free metal catalyst for the ring-opening polymerization of various lactones [34] and for the degradation of polylactide or mixed commercial plastics to obtain value-added molecules [35].

Inspired by these considerations, we explored the catalytic behavior of other commercially available amides of non-toxic and low-cost metals (Scheme 1) in the degradation reactions of polylactide (PLLA).

Using glycolysis, these studies were also extended to the chemical degradation of polyethylene terephthalate (PET), the most important commercial polyester whose annual global production exceeded 70 million tons, mainly used for the production of single-use products. Due to the lack of an efficient recycling capacity for this volume of consumption, PET recycling has become a priority for the preservation of resources and the environment [36].

2. Results and Discussion

2.1. Degradation of PLA via Alcoholysis

The depolymerization of PLA promoted by low-molecular-weight alcohols, such as methanol or ethanol, proceeds through transesterification reactions in which the metal center behaves as a Lewis acid activating the carbonylic groups of polymer chains for the nucleophilic attack by the alcohol (Scheme 2). In this study, using alcoholysis, we investigated the role of metal amides in the depolymerization of end-of-life PLLA plastic cups to alkyl lactates.

The alcoholysis reactions of commercial PLA plastic cups with a molecular mass of 58 KDa were performed at room temperature without additional solvents to minimize the environmental impact of the procedure (see degradation experiments in

Table 1. Methanolysis of PLLA using metal amides.

Run [a]	Cat	Time (h)	XInt [b] (%)	SMe-La [b] (%)	YMe-La [b] (%)
1	Zn	2	39	100	39
2	Mg	2	90	100	90
3	Y	19	98	100	98
4	Ti	24	100	89	89
5	Zr	8	11	69	8
6	Hf	2	10	52	5

^[a] All reactions were carried out by using 72 mg of PLA and 10 µmol of catalyst (1 mol% relative to ester linkages) at 25 °C with 1 mL of MeOH. ^[b] Determined by ¹H NMR.

).

The PLA degradations were monitored through the NMR analysis. The conversion of internal methine units (X_int), the selectivity of Me-La (SMe-La), and the yield of Me-La (YMe-La) were calculated by the integration of the diagnostic signals of the ¹H NMR spectra (Figure 1).

Among the catalysts investigated, the best performance was achieved with the magnesium amide (run 2, Table 1) for which the almost complete degradation of PLA was achieved after only two hours. Its activity was significantly higher than that of the Zn complex, which, after the same time, reached a conversion of 40% (run 2, Table 1). Moreover, the activity of Mg amide was significantly higher than that obtained with the homoleptic Mg complexes supported by phenoxy–imine amine (NNO) ligands that converted up to 93% of a sample of PLA cup dissolved in THF at 80 °C [31,32]. These results indicate that the Mg compound may be used effectively for catalytic degradation at room temperature, which provides high economic benefits.

With the yttrium amide, complete degradation was achieved only after 19 h (run 3, Table 1). Significantly lower activities were observed with group 4 amides (runs 4–6, Table 1), and the Zr compound showed the worst results.

Generally, PLA degradation occurs via a two-step process in which the polymer undergoes a random scission of polymer chains into oligomeric species that are progressively converted into methyl lactate.

Surprisingly, Zn, Mg, and Y catalysts promoted selective processes in which the polymer chain is directly converted into the final product (the percentage of methyl lactate was the same as the degraded PLA). This behavior was previously observed for other catalysts [29,37] that, in the absence of a solvent, promoted a mechanism in which the degradation occurs via the progressive aggression of the chain ends with the direct formation of methyl lactate.

Subsequently, the same metal compounds were tested in terms of the degradation of PLA with ethanol (Figure 2). The final product of the ethanolysis of PLA is ethyl lactate (EtLa), described as a biodegradable “green” solvent with low toxicity, which can be used as an alternative to acetone or toluene [38].

As expected, the degradation of PLA was definitively slower when performed in neat ethanol (runs 1–6,

Table 2. Ethanolysis of PLLA using metal amides.

Run [a]	Cat	T (°C)	XInt [b] (%)	SEt-La [b] (%)	YEt-La [b] (%)
1	Zn	25	100	83	83
2	Mg	25	93	98	91
3	Y	25	100	91	91
4	Ti	25	75	39	29
5	Zr	25	3	51	2
6	Hf	25	5	0.2	1
7	Zn	80	100	100	100
8	Mg	80	57	50	29
9	Y	80	17	100	17
10	Ti	80	100	100	100
11	Zr	80	100	100	100
12	Hf	80	89	54	48
13	Y	60	57	58	33
14	Zr	60	41	68	28

^[a] All reactions were carried out by using 500 mg of PLLA sample and 7 × 10⁻⁵ mol of catalyst (1 mol% relative to ester linkages) with 10 mL of EtOH for 24 h. ^[b] Determined by ¹H NMR.

) than in methanol because the more sterically hindered alkyl group disfavors the approach toward the carbonyl group [39]. At room temperature, almost quantitative conversions were obtained for the amides of Zn, Mg, and Y after 24 h. Additionally, in this case, lower activities were observed with group 4 metals, especially with Zr and Hf.

At higher temperatures, the reactivity trend was significantly different: The Zn amide preserved its activity, while an important downfall was observed for Mg and Y catalysts (runs 7–12, Table 2). The Mg amide showed a performance significantly poorer than the homoleptic complexes reported by Jones and Wood [39]. This could suggest that the presence of ancillary ligands is important to stabilize the metal center when drastic reaction conditions are applied.

Unexpectedly, at higher temperatures, group 4 metal compounds improved their performances. The additional experiments performed at the intermediate temperature of 60 °C (runs 13 and 14, Table 2) confirmed the opposite effect of the temperature on the catalytic activity of yttrium and zirconium catalysts (Figure 3).

These results showed that the trend in the performances of the explored catalysts strongly depended on reaction conditions, likely because of the different thermal stability of the active species involved in the polymerization process.

To exclude the presence of epimerization side reactions through the deprotonation of the beta-CH group by amides, the ethylactate samples obtained through degradation processes were analyzed using optical rotation measurements in methylene dichloride at 22 °C.

The full coherence between the [α]₂₂^D value of the ethylactate sample obtained in run 2 of Table 2 ([α]₂₂^D = −0.2169) and that of the commercial product (−0.2919) confirmed the absence of side reactions of epimerization.

2.2. Glycolysis of PET

Following PLA degradation success, we decided to extend our studies to the alcoholysis reaction of other polyesters of high commercial interest such as polyethylene terephthalate (PET). Currently, PET accounts for about 23% of plastic use for short-time packaging; thus, improvement in the strategies for its recycling is of fundamental importance.

The degradation of PET via glycolysis is an efficient procedure to convert PET waste into a mixture of bis(2-hydroxyethyl)terephthalate (BHET) and ethylene glycol (EG), which are the monomeric units for the synthesis of PET via polycondensation (Scheme 3).

A variety of metals [40,41] and organic catalysts [42,43,44] have been reported for the chemical depolymerization of PET. Recently, bicomponent catalysts formed by organic bases with Lewis acidic metal salts have also been explored [45]. Usually, because of its scarce solubility and high stability, the alcoholysis of PET requires severe reaction conditions such as high temperature and pressure and elevated percentage of catalyst.

All the metal amides reported in Scheme 1 were investigated in the degradation of a commercial sample of PET (M_n = 42.000 g mol⁻¹, Ð = 1.7). The reactions were performed in the absence of additional solvents at 180 °C and were stopped when the complete dissolution of solid PET was observed (runs 1–6,

Table 3. Glycolysis of PET under solvent-free conditions at 180 °C.

Run [a]	Cat	Conv (%)	Sel BHET (%)	Yield BHET (%)
1	Zn	92	26	24
2	Mg	96	24	22
3	Y	96	58	26
4	Ti	52	15	8
5	Zr	86	78	65
6	Hf	73	28	21

^[a] Reaction conditions: 200 mg of PET bottle (M_n ≈ 42.000 g mol⁻¹), 0.8 mL of EG (27.8 equivalents relative to the ester bonds), and catalyst (0.013 equivalents relative to the ester bonds) at 180 °C for 1 h.

). The conversion of PET was evaluated as a ratio of the weight of the decomposed PET (i.e., the difference between the initial weight of PET and the weight of residual PET) and the weight of the initial sample of PET. The yield and selectivity in the production of BHET were estimated as described in the experimental section.

The degradation products were extracted with distilled water and crystallized from cold water. The purity of the BHET obtained via degradation was evaluated with ¹H NMR spectroscopy (Figure 4) and mass spectrometry using either electrospray (ESI) or Matrix Assisted Laser Desorption Ionization—Time of Flight (MALDI-ToF) methods (Figure 5).

In the ESI-MS spectra, three peaks were evident: two were related to [BHET + Na⁺] m/z = 277.06 and [BHET + H⁺] m/z = 255.08 ions and the other one corresponding to the [BHET − H₂O + H⁺] ion. (Figure 5)

In the MALDI-ToF spectra, a single peak was evident for the [BHET − H₂O + H⁺] ion m/z = 255.08.

Among the investigated catalysts, Zn, Mg, and Y were able to almost quantitatively convert the whole PET amount after one hour. However, the best results were obtained with Zr amide, which reached a good percentage of degradation (86%) and very high selectivity (78%) in the production of BHET, with values that are comparable with the best results obtained with metal catalysts.

Since, in the presence of a large amount of alcohol, the metal amides are reasonably converted into the related metal alkoxides, a comparison experiment was performed by using Zr(OEt)₄ as a catalyst, under the same reaction conditions. In this case, under the same reaction conditions, a much lower percentage of the degradation of PET was achieved (19%).

The very good performances shown by Zn amide, in comparison to the related alkoxide, could be a consequence of the nature of its active species and/or of the co-presence of the free amine produced in situ that could activate the alcohol compound through the H bond interaction, as proposed for bicomponent catalysts formed by metal halides and amines described by Dove [45].

To the best of our knowledge, these studies represent the first examples of a systematic exploration of commercial metal amides in PET glycolysis.

3. Experimental Section

3.1. Materials and Methods

All the manipulations of air- and/or water-sensitive compounds were carried out under a dry nitrogen atmosphere using a Braun LabMaster glovebox or standard Schlenk line techniques. The glassware and vials used in the polymerization were dried in an oven at 120 °C overnight and exposed three times to vacuum–nitrogen cycles.

Methanol and ethanol were refluxed over Na and distilled under nitrogen. Benzene, hexane, and tetrahydrofuran were distilled under nitrogen over sodium benzophenone. (THF).

Deuterated solvents, CDCl₃, and C₆D₆ were purchased from Eurisotop were dried over molecular sieves. All the other reagents and solvents were purchased from Aldrich and used without further purification. The metal amides were purchased from Aldrich and used as received. L-lactide was purchased from Aldrich and crystallized using dry toluene and afterward stored at −20 °C in a glovebox. All the other chemicals were commercially available and used as received unless otherwise stated.

3.2. NMR Analysis

The NMR spectra were recorded on Bruker Advance 300, 400, and 600 MHz spectrometers (¹H: 300.13, 400.13, 600.13 MHz) at 25 °C unless otherwise stated. Chemical shifts (δ) are expressed as parts per million and coupling constants (J) in hertz.

The resonances are reported in ppm (δ) and the coupling constants in Hz (J) and were referred to with the residual solvent peaks at δ = 7.16 ppm for C₆D₆ and δ = 7.27 for CDCl₃.

The ¹³C NMR spectra were referred to using the residual solvent peaks at δ = 128.06 for C₆D₆ and δ = 77.23 for CDCl₃. Spectra recording was performed using Bruker’s TopSpin v2.1 software. Data processing was performed using TopSpin v2.1 or MestReNova v6.0.2 software.

3.3. GPC Size Exclusion Chromatography

The number-average molecular weights (M_n) and molecular weight distributions of polymers (dispersity, Ð) were evaluated through size exclusion chromatography (SEC), using Agilent 1260 Infinity Series GPC (ResiPore 3 μm, 300 × 7.5 mm, 1.0 mL min⁻¹, UV (250 nm) and a refractive index (RI, PLGPC 220) detector. All the measurements were performed with THF as the eluent at a flow rate of 1.0 mL/min at 35 °C. Monodisperse poly(styrene) polymers were used as calibration standards.

MALDI-ToF-MS analysis was performed on a Waters MALDI Micro MX equipped with a 337 nm nitrogen laser. An acceleration voltage of 25 kV was applied. The polymer sample was dissolved in THF with Milli-Q water containing 0.1% formic acid at a concentration of 0.8 mg mL⁻¹. The matrix used was 2,5-dihydroxybenzoic acid (DHBA) (Pierce) and was dissolved in THF at a concentration of 30 mg mL⁻¹.

3.4. General Procedure for the Depolymerization of Polylactide

The depolymerization reaction was carried out under an inert atmosphere. In a Braun LabMaster glovebox, a magnetically stirred reactor vessel (10 mL) was charged with polylactide. In a vial (4 mL), the metal complex was dissolved in MeOH or EtOH and added to the polymer. The reaction mixture was stirred at room temperature. At desired times, small aliquots of the reaction mixture were sampled, dissolved in CDCl₃ or C₆D₆, and analyzed using ¹H NMR spectroscopy. At the end of the depolymerization, the reaction was stopped with CH₂Cl₂ and dried under vacuum. The conversion of PLA, methyl lactate, and oligomers were calculated from ¹H NMR, by the following equations:

Xint (%) = 1 − ([Int])/([Int]₀) × 100

S Me-La (%) = ([Me-La])/([Int] − [Int]₀) × 100

Y Me-La (%) = Xint × S_( Me-La)

3.5. General Procedure of PET Glycolysis

For each experiment, 0.200 g PET particles, the opportune volume of ethylene glycol, and a predicted amount of the catalyst were added to a 25 mL reaction tube with a magnetic stirrer. Glycolysis reactions were carried out at 180 °C for one hour. After this time, the reaction mixture was cooled to room temperature, and about 12 mL of distilled water was added; the resulting mixture was stirred and then filtered to eliminate the residual PET that was dried at 60 °C in vacuo to constant weight.

The water solution was concentrated and stored at 4 °C for the night. The crystals of BHET were recovered via filtration, dried, and weighted.

The conversion of PET was calculated by the following equation:

conv % = (W₀ − W_r)/W₀ × 100

where W₀ is the initial weight of PET, and W_r is the weight of residual PET.

The water solution was concentrated by using a vacuum rotary evaporator at 70 °C and then refrigerated at 0 °C for 12 h to obtain white crystals of pure BHET.

The selectivity and yield of BHET were calculated according to the following equations:

Sel BHET % = (mol BHET crystals)/(mol PET soluble) × 100

Yield BHET % = (mol BHET crystals)/(PET initial) × 100

4. Conclusions

In this work, a study of the catalytic behavior of commercial metal amides in the chemical degradation of polylactide (PLA) and polyethylene terephthalate (PET) via alcoholysis was described.

In the degradation of PLA performed with methanol or ethanol at room temperature, the magnesium amide showed the highest activity, followed by Zn and Y amides, while the amides of group 4 metals were scarcely efficient. In the ethanolysis of PLA and the glycolysis of PET conducted at higher temperatures, the good activity of the Zn amide was preserved, while for Mg and Y amides, a significant decrease was observed. Surprisingly, for the glycolysis of PET, the Zr amide was revealed to be the best catalyst, enabling 78% BHET yield within one hour at 180 °C, which is a result that is comparable to the best ones described in the literature for metal catalysts. These studies represent the first examples of PET glycolysis mediated by homogeneous zirconium and yttrium catalysts.

The obtained results highlighted that simple non-toxic metal compounds can represent a valid tool to efficiently promote the sustainable recycling of exhausted polymeric materials into chemical products of synthetic interest.