Depolymerization of PET with n-Hexylamine, n-Octylamine, and with 3-Amino-1-Propanol Affording Terephthalamides

1. Introduction

The importance of chemical conversion of plastic waste such as chemical recycling (conversion to monomers), upcycling (conversion to value-added chemicals) has been pronounced in terms of circular economy [1,2,3,4]. Poly(ethylene terephthalate) (PET) has been a commodity plastic used as drink bottles, clothes, carpets etc. in our daily life, and mechanical recycling of PET bottle (collection, sorting, and re-processing after purification steps) has been known. Recently, in terms of carbon neutral concept as well as circular economy, chemical recycling of polyesters [5,6,7,8,9,10] including PET [10,11,12,13,14,15] has been recognized as the important technology for the purpose [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]. Since recycled PET resin through the above mechanical recycling has certation limitation due to their inferior quality, there are many studies concerning the chemical recycling, depolymerization of PET to monomers that are eventually converted to fresh resin which should be equivalent to the petroleum-derived one [12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43].

As shown in Scheme 1, there are two major methods for depolymerization of PET with ethylene glycol (EG) to afford bis(2-hydroxyethyl)terephthalate (BHET, so-called glycolysis process) or methanol to afford dimethyl terephthalate (DMT, called methanolysis). For example, (i) glycolysis by using a catalyst system consisting of Zn(OAc)₂—Na₂CO₃ (at 196 °C) [22] or Zn(OAc)₂—1,3-dimethylurea (at 190 °C) [27], or a certain combination of acids and base (5 mol%, 180 °C) [31], (ii) methanolysis under high temperature (e.g., 280–310 °C) and high-pressure conditions (ca. 4 MPa, the addition of K₂CO₃ reduced the harsh conditions) [10-12] have been known. Another method for the depolymerization of PET with n-butanol in the presence of [HO₃S-(CH₂)₃-NEt₃]Cl–ZnCl₂ at 205 °C [26] have also been known. As described above, most of methods reported [16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,43], however, require harsh conditions (high temperature, pressure) and/or excess base, acids, and/or inorganic salts. The process consists of depolymerization, decolorization, deionization (removal of salt), crystallization (initial purification), solid-liquid separation, concentration and vacuum distillation [10,11,12,13,14,15]. The methods, however, often face difficulty such as separation of byproduct (diethylene glycol etc.) and/or careful removal of inorganic salts (severe impurities in the final purification step by vacuum distillation upon heating) [10,11,12,13,14,15]. Therefore, development of the acid-, base-free depolymerization methods have been considered as the important technology especially in terms of simple purification process as well as of the sustainable process (total atom efficiency, no inorganic effluents, waste waters etc.).

More recently, acid-, base-free depolymerization of PET [40,41,42] as well as conventional polyesters [44,45,46,47] with alcohols have been demonstrated in the presence of La(acac)₃ (acac = acetylacetonato) [40,46], CaO [42,45], or Cp’TiCl₃ (Cp’ = Cp, Cp*) [41,44]. Moreover, FeCl₃ catalyzed depolymerization not only PET bottles, but also textile wastes consisting of PET with ethanol has been demonstrated [47]. These depolymerization, transesterification with alcohols (ethanol, methanol, cyclohexane methanol etc.), afforded the corresponding monomers exclusively (Scheme 2, >99% conversion, >99% selectivity). We also reported that PET was treated with morpholine in the presence of Cp*TiCl₃ to give the corresponding amide in high yields [48].

In this paper, we thus present reactions of PET with linear n-alkylamines such as n-hexylamine, n-octylamine, and 3-amino-1-propanol to expand the utility of this developed methods for the efficient upcycling of the PET waste. Through this research, we wish to provide a difference in depolymerization of polyesters between by the transesterification and by the aminolysis; the reaction of PET with 3-amino-1-propanol occurred exclusively with the amino group to afford the corresponding bis(3-hydroxypropyl) terephthalamide, 1,4-[HOCH₂CH₂CH₂N-C(O)]₂C₆H₄, even in the absence of catalyst (Scheme 3).

2. Results and Discussion

2.1. Reactions of PET with n-Hexylamine, n-Octylamine

Reactions of PET with n-hexylamine were conducted using a sealed tube in the presence of Cp*TiCl₃ catalyst (2.0 mol% to the monomer unit of PET) at 130 ºC (set into the alumina heating blocks of the parallel reactor) on the basis of the conditions for depolymerization with morpholine [48] (Scheme 4, details are described in the Experimental section). After the reaction, the reaction mixture was placed in vacuo to remove volatiles, and the resultant mixture was further purified with chloroform to isolate the product, di(n-hexyl) terephthalamide, by recrystallization. Samples of PET sheet, prepared by cutting the PET drink bottle, or PET powder, prepared from the commercially available fresh resin by using a grinding machine, were used in this study. The results conducted under various conditions are summarized in Table 1.

As shown in Figure 1, ¹H- and ¹³C-NMR spectra, the reaction product, isolated as white solids, was di(n-hexyl) terephthalamide as a pure form especially when the reaction was conducted for 48 h (runs 1,2). As shown in Figure S1a in Supplementary Materials (SM), the ¹H NMR spectrum (in tetrachloroethane-d₂) of the reaction mixture after removal of volatiles shows that the observed resonances were assigned as the desired products, suggesting that the reaction proceeded exclusively in this catalysis. It seems that the reactions completed after 16 h (runs 3,4), since the products yields isolated were already high (runs 3,4), very close to those conducted after 48 h (runs 1,2). As observed in the transesterification of PET with ethanol [41,47], the catalyst performances were not strongly affected by the nature of PET samples (sheet or powder, runs 1-4).

In contrast, resonances at δ= 4.73 and 8.13 ppm were observed in addition to the terephthalamide in the ¹H NMR spectrum of the reaction mixture after 6 h (run 6, Figure 2b). Interestingly, the reaction also proceeded in the absence of Cp*TiCl₃ (run 7, Figure 2c), but the intensities of the additional peaks were apparently high. These resonances became decreasing upon increasing the catalyst (Ti 3.0 and 5.0, respectively, Figure S1, SM). These resonances disappeared when the reaction mixture was passing through a Celite Pad before recrystallization; this caused a difficulty for identification of the byproduct by GC, GC-MS. In order to isolate the intermediate, the reactions were conducted at 100 ºC in the presence of Cp*TiCl₃ (runs 8,9), and the byproduct was precipitated from the reaction mixture by addition of methanol. On the basis of NMR spectra (Figure 2d,e) and DSC thermogram of the resultant solid, the melting temperature (237 ºC) was relatively close to that in PET (244.8 C, Figure S2 in SM). It thus seems likely that these resonances are due to PET oligomers (although we were not able to measure the molecular weight due to their insolubility in THF nor chloroform even for the GPC measurement). The results clearly explain that the PET was depolymerized to PET oligomers and gradually converted to the corresponding to terephthalamide, as observed in the depolymerization with alcohols by transesterification [41,42]. These results also clearly indicate that Cp*TiCl₃ play a role as the catalyst that facilitates the aminolysis under these conditions.

Similarly, reactions of PET with n-octylamine were conducted at 130 ºC in the presence of Cp*TiCl₃ (2.0 mol%), and the results are summarized in Table 2. The reactions afforded the corresponding amide, di(n-octyl) terephalamide, exclusively without contamination of the other byproduct, as shown in Figure S3 (NMR spectra) in SM. Compared to the reaction with n-hexylamine, isolation of the n-octylamide seemed rather difficult due to difficulty of removing n-octylamine in vacuo and the product by repetitive washing with chloroform was thus necessary to remove the amine completely. As observed in the transesterification of PET with ethanol [41,47] as well as in the reactions with n-hexylamine (Table 1), the catalyst performances were not strongly affected by the nature of PET samples (sheet or powder, runs 10,11). As shown in Figure S4 (SM), the isolated product contained impurity (PET oligomer) in trance amount in the ¹H NMR spectrum (reaction 16 h); it thus seems that longer reaction time was necessary for obtainment of the terephthalamide in the exclusive yield.

2.2. Reaction of PET with 3-Amino-1-Propanol

Reactions of PET with 3-amino-1-propanol were conducted at 100 or 130 ºC in the presence/absence of Cp*TiCl₃ catalyst, and the results are summarized in Table 3. It was revealed that the reaction product was bis(3-hydroxy) terephthalamide, 1,4-[HOCH₂CH₂CH₂NC(O)]C₆H₄, exclusively as shown in Figure 3 (¹³C NMR spectrum) as well as Figure S5, SM. The pure isolation of the amide seemed rather difficult because of separation of the product with 3-amino-1-propanol remained even after removal of volatiles in vacuo. Therefore, repetitive precipitation with chloroform was required (reason for rather low yields compared to those in the reaction with n-hexylamine). Reason for exclusive obtainment of amide instead of ester would be due to a stability of the product (amide is more stable thermodynamically compared to esters).

It should be noted that the reactions completed without catalyst and the yields in the absence of Ti catalyst were highly close to those in the presence of Ti catalyst. Moreover, no significant differences were seen when the reactions were conducted at 100 ºC after 3 h (run 17 vs run 21). The reactions were rather affected by the temperature employed, since the reaction mixture conducted at 80 ºC remained PET slurry after 6 h (runs 18,22).

3. Materials and Methods

All experiments were conducted under a nitrogen atmosphere using a glove box. CpTiCl₃, Cp*TiCl₃ (Tokyo Chemical Industry, Co., Ltd., Tokyo, Japan), n-hexylamine, n-octylamine, and 3-amino-1-propanol (Tokyo Chemical Industry, Co., Ltd., Tokyo, Japan) were used as received. Poly(ethylene terephthalate (PET) resin (IV = 0.80 ± 0.02 dL/g) were received from companies (by donation for research purpose only), and were used as PET powers after griding using 0.25 mm mesh using grinding machine. PET sheets were prepared by cutting the PET drink bottle. All ¹H and ¹³C{¹H} NMR spectra (in tetrachloroethane-d₂ at 100 °C or methanol-d₄ at 25 ºC) were measured on a JEOL JNM ECS400 spectrometer (399.8 MHz for ¹H and 100.5 MHz for ¹³C, JEOL Ltd., Tokyo, Japan) using SiMe₄ as the reference at 0.00 ppm (chemical shifts were reported in parts per million).

General procedure for reaction of poly(ethylene terephthalate (PET) with amines. An oven-dried reaction apparatus (100 mL scale screw-cap glass tube) was charged with the prescribed amount of CpTiCl₃, 500 mg of PET (powder or sheet, shown in Scheme 4), and 5.0 mL of amine (n-hexylamine, n-octylamine, or 3-aminopropanol) under a nitrogen atmosphere. The reaction mixture was stirred at prescribed temperature using an alumina heating blocks in parallel reaction apparatus (ChemiStation^TM, Tokyo Rikakikai Co., Ltd., PPS-2511). After the reaction, the mixture was cooled to room temperature and transferred to a round-bottom flask for removal of volatiles, and the products, terephthalamide, were isolated by precipitation or recrystallization by ethanol or toluene. The reaction mixture containing impurities (PET oligomer) was separated by filtration using Celite pad after dissolving the products by chloroform-methanol mixed solution.

Di(n-hexyl) terephthalamide: ¹H NMR (tetrachloroethane-d₂): δ 7.82 (s, 4H, Ar-H), 6.09 (s, 2H, -NH-), 3.48 (q, J = 6.8 Hz, 4H, -NHCH₂-), 1.71-1.64 (m, 4H, -NHCH₂CH₂-), 1.47-1.37 (m, 12H, -CH₂-), 0.97 (t, J = 7.1 Hz, 6H, -CH₃). ¹³C{¹H} NMR (tetrachloroethane-d₂): δ 166.2(-CO-), 137.4 (Ar), 126.9 (Ar), 40.2, 31.2, 29.5, 26.4, 22.2, 13.6.

Di(n-octyl) terephthalamide: ¹H NMR (tetrachloroethane-d₂): δ 7.82 (s, 4H, Ar-H), 6.06 (s, 2H, -NH-), 3.49 (q, J = 6.7 Hz, 4H, -NHCH₂-), 1.71-1.64 (m, 4H, -NHCH₂CH₂-), 1.47-1.35 (m, 20H, -CH₂-), 0.95 (t, J = 6.9 Hz, 6H, -CH₃). ¹³C{¹H} NMR (tetrachloroethane-d₂): δ 166.2 (-CO-), 137.4 (Ar), 126.9 (Ar), 40.2, 31.5, 29.5, 29.0, 28.9, 26.8, 22.3, 13.7.

Di(3-hydroxypropyl) terephthalamide: ¹H NMR (DMSO-d₆): δ 8.56 (t, J = 5.5 Hz, 2H, -NH-), 7.89 (s, 4H, Ar-H), 4.50 (t, J = 5.0 Hz, 2H, -OH), 3.46 (q, J = 6.0 Hz, 4H, -CH₂OH-), 3.32 (q, J = 6.5 Hz, 4H, -NHCH₂- ), 1.71-1.65 (m, 4H, -CH₂-). ¹³C{¹H} NMR (methanol-d₄): δ 169.4 (-CO-), 138.4 (Ar), 128.4 (Ar), 60.6 (-CH₂OH), 38.2(-NHCH₂-), 33.2(-CH₂-).

4. Conclusions

As described in the introduction, conversions of PET into fine chemicals called upcycling have been one of the important key technologies in terms of establishment of circular economy. We have demonstrated preparations of three amides in the depolymerization of PET with n-hexylamine, n-octylamine and with 3-amino-1-propanol, affording the corresponding terephthalamides in high yields (without by-production of side products). Cp*TiCl₃, the effective catalyst for aminolysis of PET with morphoiline [48] plays a role to facilitate the reaction but these reactions proceeded even in the absence of Ti catalyst. Indeed, no significant differences in the product yields were observed in the depolymerization of PET with 3-amino-1-propanol affording the corresponding bis(3-hydroxy) terephthalamide in high yields even at 100 ºC. The results are significant contrast to those in the depolymerization with alcohols by transesterification, because presence of Cp’TiCl₃ (Cp’ = Cp, Cp*) was prerequisite to proceed the reaction as catalyst. The difference would be probably explained as due to the stability of the product, amide, thermodynamically compared to esters, as well as amine would probably play a role as catalyst in the aminolysis of PET, on the basis of results in the depolymerization with 3-amino-1-propanol. Since the method could be effective for chemical conversion of the other PET wastes (textile waste), as demonstrated previously [47]. The observed fact should be potentially important for catalytic chemical conversion of polyester to fine chemicals. We shall explore more possibilities with various polyesters and will introduce in the near future.