Base-assisted elemental sulfur-based multicomponent polymerizations to synthesize asymmetric polythioureas

Abstract

Polythioureas with great potential in precious metal recovery materials, dielectric materials, light refractive materials, self-healing materials, and adhesives have become a group of popular polymer materials. While various synthetic approaches have been reported for different polythiourea structures, which generally involve expensive and toxic monomers, several polythiourea structures, such as asymmetric aromatic polythioureas, still have limited access. Herein, a base-catalyzed multicomponent polymerization (MCP) of elemental sulfur, diisocyanides, and diamines was developed, which was generally applicable for both aromatic amines with low reactivity and aliphatic amines with high nucleophilicity, affording polythioureas with great structural diversity, high M_ws of up to 107,700 g/mol and high yields of up to 99%. Moreover, aromatic polythioureas with different aromatic spacers installed on each side of the thiourea moiety could be facilely synthesized. Amine exchange reactions were studied for thiourea compounds or polythioureas, indicating that the C—N bonds of the thiourea moiety on the aromatic substitute side were more labile compared with those on the aliphatic substitute side, rendering the potential controllable degradation of different polythiourea structures. The polythioureas generally possess high thermal stability, with the glass transition temperatures ranging from 121 to 169 °C. These polythioureas could efficiently absorb Hg²⁺ from polluted water, and aromatic polythioureas generally showed better performance compared with semi-aromatic and aliphatic polythioureas. This work has hence provided a general synthetic approach for various aliphatic, aromatic, and semi-aromatic polythioureas with symmetric or asymmetric thiourea structures. The dynamic covalent bond nature of thioureas endowed these polymer materials with controllable degradation, making them sustainable functional polymer materials.

Graphical abstract

1. Introduction

Sulfur-containing polymers such as polysulfides [[1], [2], [3]], polydisulfides [4,5], polythioesters [6], polythiocarbonates [7], polythioureas [8,9], polythioamides [10,11], and polythiophenes [[12], [13], [14]] have attracted great research interest from material scientists, owing to the unique properties endowed by the sulfur element. Among them, polythioureas are one of the most popular polymer materials recently, attributed to their excellent self-repairing properties [15,16], dielectric properties [17,18], metal coordination ability [9,19], degradability [20], and recyclability [21]. Of all the fascinating properties of polythioureas, the dynamic covalent bonds of thiourea moieties could provide great opportunities in self-healing materials, cross-linked network reprocessing, solid state recycling and degradation, and adjustable mechanical properties, which showed great potential for the development of sustainable polymer materials. For example, Zhang et al. reported a series of dynamically reversible crosslinked polythioureas with mechanical properties comparable to plastics, rubbers, and fibers, respectively [17]; Xie et al. reported a thiourea-based thermoset elastomer which could be reprocessed with enhanced mechanical properties [21]; Aida et al. prepared ether-thiourea polymers through the polycondensation reaction of 1,1′-thiocarbonyldiimidazole and diamines in DMF, and these amorphous polythioureas were mechanically robust yet readily repairable [8]; we have also developed scalable synthesis of polythioureas which could be self-repaired at room temperature or in ice bath in the glassy state [22]. Through the design of thiourea bonds with different chemical environments, the dynamic reversible behavior could be tuned, and unidirectional or bidirectional stretching could also significantly improve the tensile strength of crosslinked polythiourea [16]. Hence, it is crucial to develop an efficient and convenient synthetic approach for a great diversity of polythiourea structures to meet the requirements of various properties.

The commonly used synthetic approaches of polythioureas generally involve expensive and toxic sulfur-containing reagents, such as thiophosgene [23], 1,1′-thiocarbonyldiimidazole [8], thiourea [24], or diisothiocyanates [25], which could undergo polycondensations or polyadditions with diamine monomers to afford polythiourea structures limited by the monomer sources. CS₂, as a commonly used solvent, has been developed as a sulfur source monomer for the synthesis of polythioureas. For example, we have developed a robust one-step catalyst-free room temperature multicomponent polymerization (MCP) of CS₂, diamines, and commercially available monoisocyanide for the large-scale synthesis of polythioureas [22]; the polymerization of CS₂ and diamines has also been reported to synthesize polythioureas with the release of H₂S [26]. However, CS₂-based polymerization normally delivers symmetric polythioureas with the same spacer structure on both sides of the thiourea moieties, and aromatic diamine monomers could not react efficiently to generate aromatic polythioureas with satisfying yields and M_ws. Elemental sulfur, as a byproduct from the petroleum and natural gas industry, and was often used in vulcanization for crosslinking natural rubbers [27], was also an ideal monomer for the synthesis of polythioureas. We have previously reported a catalyst-free MCP of elemental sulfur, diisocyanides, and aliphatic diamines to prepare a series of different asymmetric polythioureas with versatile structures, high yields of up to 95% and high M_ws of up to 242,500 g/mol [9]. However, aromatic diamine monomers were not applicable for this catalyst-free MCP, considering their poor alkalinity and nucleophilicity, which could not activate elemental sulfur. These approaches generally have limited access to aromatic polythioureas with diversified structures. Compared with aliphatic polythioureas, aromatic polythioureas with conjugated and rigid structures usually enjoyed superior light refractivity, higher dielectric breakdown strength, lower loss at high electric fields, higher maximum electrical energy density [17,18], and most importantly, simpler upcycling or degradation conditions [21], whose facile and diverse synthesis were hence in great demand.

To solve the problem of elemental sulfur-based polymerization with aromatic amine monomers, inorganic bases were designed to activate elemental sulfur, and a K₂CO₃-assisted MCP of elemental sulfur, dibenzyl acids, and aromatic diamines was succeed to afford aromatic polythioamides [10]; a KF-assisted MCP of elemental sulfur, aromatic diynes, and aromatic diamines was developed to afford aromatic poly(thioamide-thiourea)s [28]. A similar strategy was adopted for the synthesis of aromatic symmetric polythioureas. For example, we have developed a KF-assisted MCP of elemental sulfur, dichloromethane and aromatic diamines to realize large-scale synthesis of aromatic symmetric polythioureas, which could be chemically recycled through aminolysis [29], and a t-BuOK-assisted MCP of elemental sulfur, chloroform, and aromatic diamines was also reported to generate diisocyanides in situ and then polymerize with diamines to afford aromatic symmetric polythioureas [30]. In these MCPs, a large amount of base was generally required, and only aromatic amines were applicable to avoid side reactions. Hence, only symmetric aromatic polythioureas could be accessed. The asymmetric aromatic polythioureas with unique and versatile structures could bring more variation to the polymer chain structures and hence widely tunable functionalities of these promising polythiourea materials, which still lack efficient synthetic approaches.

In this work, we adopted a base-activated sulfur strategy in the MCP of elemental sulfur, diisocyanides, and diamines to furnish polythioureas covering symmetric and asymmetric, aromatic and aliphatic structures (Fig. 1). Unlike the reported reaction mechanism of aliphatic amine, elemental sulfur, and isocyanide, in this base-assisted MCP, elemental sulfur was activated by base instead of aliphatic amine [9,31]. The base-assisted ring-opening reaction of the S₈ ring first took place to afford polysulfide species, which were then attacked by isocyanide to form zwitterionic polysulfide A. Then A was transformed to highly reactive isothiocyanate intermediate B under alkalic conditions, followed by the efficient reaction between isothiocyanate and amine, including less reactive aromatic amine to furnish thiourea product C (Scheme S1a). The control experiment between 2,6-dimethylphenyl isocyanide and sulfur at 100 °C in the presence of K₂CO₃ suggested the rapid generation of 2,6-dimethylphenyl isothiocyanate in 10 min (Scheme S1b). Without base, sulfur, 2,6-dimethylphenyl isocyanide and p-toluidine barely react even at 100 °C. K₂CO₃ was utilized to activate sulfur, and the MCP of elemental sulfur, diisocyanides, and diamines generally went smoothly, affording 15 polythioureas with well-defined and diversified structures, high M_ws (up to 107,700 g/mol), and excellent yields (up to 99%). The MCP was generally applicable for aliphatic and aromatic diisocyanides, aromatic and aliphatic diamines, primary and secondary diamines, generating diverse polythioureas structures, making it a powerful synthetic approach for polythioureas. Moreover, a series of small molecular model reactions were designed to reveal the dynamic covalent nature of different thiourea structures and their amine exchange reactions under basic condition. The aromatic polythioureas were found to be more labile upon thermal treatment, more facile to undergo amine exchange reaction, and showed higher sensitivity in mercury removal application, compared with the aliphatic polythioureas.

Fig. 1.

Synthetic methods for the elemental sulfur-based multicomponent synthesis of polythioureas.

2. Materials and methods

Detailed materials and instruments, synthetic procedures and characterization data, effect of solvent, loading ratio, base, temperature, monomer concentration, reaction time, and monomers on the MCP are provided in Supporting information.

3. Results and discussion

3.1. MCP of sulfur, diisocyanides and aromatic diamines

Aromatic diamine 3a was selected as the monomer to polymerize with sublimed sulfur S₈ 1, diisocyanide 2a, and the MCP was first carried out in toluene in the presence of K₂CO₃ at 100 °C under nitrogen for 3 h, with the monomer loading ratio of 1/8[S₈]: [2a]: [3a] = 4: 1: 1 and monomer concentration of [3a] = 0.3 M. A polymer product P1 with a M_w of 15,700 g/mol was obtained in 42% yield due to poor solubility of polythioureas in toluene (Table S1). Considering the reaction temperature, solvents with high boiling points and potential good solubility for polythioureas such as DMSO, DMF, and NMP were selected to replace toluene. The polymerization efficiency was significantly improved, and the best polymerization result was obtained in DMF with a M_w of 28,100 g/mol and 92% yield. Excess amount of sulfur was found to improve both yields and M_ws of products, and the best result was obtained when 1/8[S₈]: [2a]: [3a] = 4: 1: 1 (Table S2). The type of base and its concentration were also essential to the MCP. When K₂CO₃ was adopted as the base, the concentration mattered: when [K₂CO₃]/[2a] gradually increased from 0.2 to 0.8, the yield and M_w first increased, and then started to generate insoluble product when [K₂CO₃]/[2a] was larger than 0.6. Moreover, other organic or inorganic bases such as t-BuOK, DABCO, KF, or Cs₂CO₃, could also facilitate the polymerization and generate polymers with satisfying yields and M_ws (Table 1). Increasing polymerization temperature from 30 to 100 °C was generally beneficial to the polymerization (Table S3). Although the optimal result was obtained at 100 °C, polymer product with a M_w of 15,100 g/mol and 89% yield could also be accessed at 50 °C. Further increasing the polymerization temperature at 110–130 °C caused decreased yields and M_ws, probably due to the polymer degradation at high temperature. The effect of monomer concentration was also studied. When [2a] was increased from 0.2 to 0.3 M with the fixed monomer loading ratio, the yield and M_w of the product increased; however, gelation occurred after 35 min when [2a] was increased to 0.4 M. Satisfying polymerization result could be obtained at 30 °C when [2a] = 0.5 M, and soluble polymer with a M_w of 19,300 g/mol could be obtained in 90% yield after 10 h (Table S4). Moreover, neat polymerization under solvent-free condition could still produce polymer with a M_w of 15,600 g/mol.

Table 1.

Optimization of multicomponent polymerization of elemental sulfur 1, diisocyanide 2a, and aromatic diamine 3a.^a

Entry	Base	[base]: [2a]: [3a]	t (min)	yield (%)b	Mw (g/mol)c	Mw/Mnc
1	KF	1.0: 2.5: 2.5	180	94	18,300	1.54
2	t-BuOK	1.0: 2.5: 2.5	180	82	20,600	1.63
3	Cs2CO3	1.0: 2.5: 2.5	180	88	22,900	1.82
4	DABCO	1.0: 2.5: 2.5	180	95	20,200	1.47
5	K2CO3	1.0: 2.5: 2.5	180	92	28,100	2.03
6	K2CO3	1.5: 2.5: 2.5	180	86	26,100d	1.96
7	K2CO3	2.0: 2.5: 2.5	180	85	insoluble	insoluble
8	K2CO3	0.5: 2.5: 2.5	180	85	19,700	1.63
9	K2CO3	1.0: 2.5: 2.5	60	93	32,000	2.00
10	K2CO3	1.0: 2.5: 2.5	30	90	25,700	1.60
11	K2CO3	1.0: 2.5: 2.5	10	91	19,500	1.62
12e	K2CO3	1.0: 2.5: 2.5	10	89	16,300	1.58

^aCarried out at 100 °C under nitrogen in DMF in 10 mL sealed polymerization tube for indicated times. 1/8[S₈]: [2a]: [3a] = 4.0: 1.0: 1.0. [2a] = [3a] = 0.3 M.

^bThe yield was calculated based on the diisocyanide monomer 2a.

^cM_w and M_w/M_n were determined by GPC in DMF based on polystyrene standard samples.

^dPartially soluble.

^eThe polymerization was conducted in air.

Kinetic study of the MCP of 1, 2a and 3a was then conducted by in situ IR spectrometry, and the polymerization was carried out under nitrogen at 100 °C in DMF. When the polymerization time increased, it was obvious that two new peaks emerged at 1540 and 1339 cm⁻¹ in the stacked in situ IR profiles collected at different time intervals of the polymerization solution of 1, 2a, and 3a at 100 °C (Fig. 2a), which were assigned to the stretching vibration of C=S bond and C—N bonds of thiourea group, respectively, through the comparison with the in situ IR spectra of DMF solutions of 2a, 3a, and P1 (Fig. 2b). The time-dependent peak intensity and the three-dimensional Fourier transform IR profiles of the peak at 1540 cm⁻¹ reached saturation after 10 min (Fig. 2c,d), suggesting that the MCP was fast and efficient. Time course study also indicated that the MCP proceeded rapidly and efficiently under optimal conditions. After 10 min, the polymerization yield was above 91%, and the M_w could reach 19,500 g/mol, and polymerization in air did not show a significant influence on the results, indicating the MCP was robust. Interestingly, although the M_w of the product was gradually increased from 10 to 60 min and reached 32,000 g/mol, a slight decrease in the M_w was observed after 1 h, indicating possible thermo-degradation of aromatic polythioureas with elongated polymerization time (Table 1).

Fig. 2.

(a) Stacked in situ IR profiles collected at different time of the polymerization solution of 1, 2a, and 3a at 100 °C. (b) The in situ IR spectra of 2a, 3a, and P1 in DMF, and the polymerization solutions of 1, 2a, and 3a after reaction at 100 °C for 1 h. (c) The time-dependent peak intensity at 1540 cm⁻¹. (d) Three-dimensional Fourier transform IR profiles of the peaks at ∼1540 cm⁻¹ for the MCP at 100 °C.

3.2. General applicability of aromatic and aliphatic diamines for the MCP

To explore the general applicability of this MCP to access various polythioureas, a series of aliphatic and aromatic diisocyanides 2a-d were synthesized according to the literature [32], and four aromatic diamines 3a-d were selected as the monomers to polymerize with elemental sulfur (Fig. 3). When aliphatic diisocyanides 2a and 2b were polymerized with aromatic diamines and sulfur, polythioureas P1-P3 could be obtained in 85–93% yields with the M_ws in the range of 15,900–32,000 g/mol. In this reaction, both aliphatic isocyanides and aromatic isocyanides, even isocyanide with large steric hindrance showed high reactivity. For example, multi-substituted aromatic diisocyanide 2c was designed as a monomer to improve the solubility of the resulting aromatic polythiourea products by introducing bulky alkyl chains and inhibit strong intermolecular hydrogen bonds. When 2c was reacted with diamines and sulfur, aromatic polythioureas P4-P6 with high M_ws (31,400–107,700 g/mol) could be obtained in 67–82% yields, and these asymmetric aromatic polythioureas could hardly be accessed through other sulfur-based MCPs (Figs. 3 and S1, Table S5).

Fig. 3.

Multicomponent polymerizations of sulfur 1, diisocyanides 2a-c and aromatic diamines 3a-d. Carried out at 100 °C under nitrogen atmosphere in DMF for 1 h. Chemical structures, yields, and M_ws of P1-P6. M_ws were measured with polystyrenes as standard.

Furthermore, the MCP of aliphatic diamines could also proceed smoothly under these basic conditions without side reactions being observed. When aliphatic primary diamine 3e and secondary diamine 3f were selected as the monomers to polymerize with sulfur and aliphatic or aromatic diisocyanides, polythioureas P7-P9 were afforded with the M_ws of 28,300–46,100 g/mol in 88–99% yields. In addition, primary diamines 3e and 3g were selected to polymerize with aliphatic diisocyanides 2a/2b and sulfur to afford aliphatic polythioureas P10 and P11, respectively (Figs. 4 and S1, Table S5).

Fig. 4.

Multicomponent polymerizations of sulfur 1, diisocyanides 2a, 2b, 2d and aliphatic diamines 3e-g. Carried out at 100 °C under nitrogen atmosphere in DMF for 1 h. Chemical structures, yields, and M_ws of P7-P11. M_ws were measured with polystyrenes as standard.

To expand the scope of asymmetric aromatic polythioureas, diisocyanides 2e and 2f were also studied as the monomers to polymerize with sulfur and diamine 3a, affording polythioureas P12 and P13 with limited solubility in decent yields (Scheme S2). Moreover, electron-rich aromatic diamine 3h and electron-deficient aromatic diamine 3i were studied as monomers of this MCP to afford P14 and P15, respectively. Electron-rich monomers showed higher reactivity and better polymerization results than electron-deficient diamine monomers. The base-assisted MCP was generally applicable for aromatic and aliphatic diisocyanide/diamine monomers, providing access to diversified polythioureas. Of all the synthesized polythioureas in this work, P1-P7 and P12-P15 have not been synthesized from sulfur before, and P4-P6, P12, and P13 could only be accessed from the current approach. For P8-P11 with aliphatic thiourea structures which could also be synthesized from base-free MCP of sulfur, with the assistance of K₂CO₃, the polymer yields were similarly high compared with the literature [9], and the M_ws were generally decreased, considering the possible K₂CO₃-catalyzed degradation of polythiourea upon heating in alkalic solution.

The chemical structures of these polythioureas were carefully characterized by ¹H and ¹³C NMR spectra, and IR spectra. Taking P1 for example, in its ¹H NMR spectrum, the proton resonance of −NH₂ of 3a at δ 4.78 was disappeared, and two −NH− proton resonances of thiourea moieties emerged at δ 8.08 and δ 9.54 (Fig. 5a–c); in its ¹³C NMR spectrum, the typical carbon resonance of −NC from 2a at δ 156.71 was disappeared, and the characteristic C=S carbon peak at δ 181.11 appeared (Fig. 5d–f). Similarly, the characteristic −NH− proton resonances at δ 7.31–10.02 were observed from the ¹H NMR spectra of P2-P15, and C=S carbon resonances at δ 180.03–183.03 were observed from the ¹³C NMR spectra of P2-P11, proving the existence of their thiourea moieties (Fig. S2–S9). In the IR spectra of polythioureas P1-P11, the stretching vibration peaks of C=S bonds at 1496–1551 cm⁻¹ and the stretching vibration peaks of N—H bonds at 3232–3351 cm⁻¹ were observed, proving the formation of thiourea moieties (Fig. S10–S14).

Fig. 5.

¹H NMR spectra of (a) 2a, (b) 3a, and (c) P1. ¹³C NMR spectra of (d) 2a, (e) 3a, and (f) P1 in DMSO‑d₆. The solvent peaks were marked with asterisks.

3.3. Dynamic reversible covalent exchange reactions in thioureas and polythioureas

Dynamic covalent bonds which could reversibly break and form under certain conditions have attracted great attention recently [33], and dynamic reversible covalent exchange reactions, such as transesterification [34], sulfide and mercapto exchange [35], alkoxyamine exchange [36], and olefin metathesis reactions [37], have been designed and utilized in a series of important materials preparation [[38], [39], [40], [41], [42]]. The dynamic reversible covalent exchange reactions of thiourea derivatives have also been revealed in the literature. For example, when using organic bases as catalysts, thiourea molecules could react in THF at 60 °C to dissociate into a dynamic mixture composed of amines and isothiocyanates [20]. The C—N bonds of thiourea molecules could hence be broken under basic conditions to form −NH₂ and −NCS, respectively, which would react again to form thiourea products under the same conditions. When the original thiourea structure was asymmetric, there would be different C—N bond broken pathways, different combinations of amines and isothiocyanates, and therefore different thiourea structures. To study the amine exchange reactions of thioureas, excess elemental sulfur 1, tert‑butyl isocyanide 7 (7.5 mmol) and p-toluidine 8 (5 mmol) were reacted at 100 °C for 4 h with K₂CO₃, three different thiourea structures 4 (0.55 mmol), 5 (1.40 mmol), and 6 (1.25 mmol) were isolated (Fig. 6a), suggesting the dynamic covalent nature of the C—N bonds in thiourea moiety.

Fig. 6.

(a) Amine exchange of thiourea under base catalysis. Carried out at 100 °C under nitrogen atmosphere in DMF for 4 h, K₂CO₃ = 0.4 M. (b) Possible mechanism of reaction of asymmetric thiourea 9.

A series of orthogonal experiments were conducted to verify the amine exchange reaction and investigate the amine exchange reactivity. Considering that the possible exchange product tert‑butylamine with low boiling point might evaporate at elevated temperature, an asymmetric aromatic-aliphatic thiourea 9 synthesized from n-hexylamine was selected as the model compound, which could undergo C—N bond breakage to form aromatic amine 8 and aliphatic thioisocyanide 12 through pathway a, or undergo C—N bond breakage to form aliphatic amine 11 and aromatic thioisocyanate 10 through pathway b (Fig. 6b), upon heating in the presence of base. In the mixed reaction solution, thioisocyanate 12 could react with aliphatic amine 11 to afford aliphatic thiourea 13, and similarly, thioisocyanate 10 could react with aromatic amine 8 to afford aromatic thiourea 6.

Firstly, asymmetric thiourea 9 was stirred in the presence of K₂CO₃ at 120 °C for 6 h in DMSO‑d₆, and the ¹H NMR spectra of the solution before and after the reaction indicated that besides the remaining thiourea 9, large signal indicating aromatic amine 8 emerged, with the yield of 30% calculated based on the internal standard mesitylene (Fig. 7a). It should be noted that the chemical shifts of thioureas in the presence of K₂CO₃ would shift compared with those without base (Fig. S15). Secondly, when equivalent amount of aromatic thiourea 6 and aliphatic thiourea 13 were stirred at the same condition in the presence of K₂CO₃ at 120 °C for 6 h in DMSO‑d₆, large peak of 13 remained in the ¹H NMR spectrum, while the peak of aromatic thiourea 6 decreased significantly, accompanying with the emergence of large peak of aromatic amine 8 with the calculated yield of 54%, indicating that aromatic thiourea 6 underwent bond breakage while aliphatic thiourea 13 was rather stable in the reaction condition (Fig. 7b). Alternatively, when an equivalent amount of aliphatic thiourea 13 and aromatic amine 8 were stirred at the identical hot basic solution of DMSO‑d₆, the ¹H NMR spectrum remained almost unchanged after 6 h reaction at 120 °C, again proving the stability of aliphatic thiourea under such conditions (Fig. 7c). Last but not least, when the same amount of aromatic thiourea 6 and aliphatic n-hexylamine 11 were stirred at the same condition, after 6 h reaction at 120 °C in the presence of K₂CO₃ in DMSO‑d₆, the characteristic proton peaks for thiourea 6 almost disappeared, and large peak representing aromatic amine 8 emerged with the calculated yield of 63% (Fig. 7d), suggesting that aromatic thiourea would be consumed and transformed to aromatic amine, asymmetric thiourea, or other side products. When aromatic thiourea 6 and aliphatic n-hexylamine 11 were stirred without K₂CO₃ under similar conditions, the peak of aromatic amine 8 could also emerge, but the yield was decreased to 36% (Fig. S16). During the polymerization and degradation process, the reactive isothiocyanate intermediate was difficult to isolate, considering its high reactivity and rapid transformation to thiourea or other side products upon heating in the presence of K₂CO₃ (Fig. S17 and S18). These systematic investigations had revealed that aromatic thioureas were more labile in alkaline conditions upon heating, and aliphatic thioureas were more stable, which hardly underwent bond breakage reactions or exchange reactions. These differences in bond strength could help design future smart polymer materials with controlled bond breakage and reformation.

Fig. 7.

Control experiments of the thiourea exchange reactions. (a) Asymmetric thiourea 9. (b) Aromatic thiourea 6 and aliphatic thiourea 13. (c) p-toluidine 8 and aliphatic thiourea 13. (d) n-hexylamine 11 and aromatic thiourea 6. Carried out at 120 °C in DMSO‑d₆ for 6 h, K₂CO₃ = 0.2 M, mesitylene = 1 mmol. The solvent peaks were marked with asterisks. The mesitylene peaks were marked with triangles.

To study the amine exchange reaction in polythioureas, P1 was reacted with excess tert‑butylamine 14 in DMF with K₂CO₃ at 100 °C (Fig. 8 and Table S6). The M_w of P1 was continuously decreased from 44,900 g/mol to 12,400 g/mol along with the reaction time. The ¹H NMR spectra of the product suggested that besides the polythiourea backbone structure, new signals for terminal amine groups emerged at both aromatic and aliphatic regions, indicating the polymer chain of P1 was cut to short chains (Fig. S19). When tert‑butylamine was replaced by n-hexylamine, the M_w of P1 also decreased significantly to 5600 g/mol. Compared with thiourea small molecules, these polythioureas were more stable, and the corresponding amine exchange reaction reactivity was lower. To compare the degradability of different polythiourea structures, asymmetric aliphatic-aromatic polythiourea P2, asymmetric aromatic polythiourea P4, and aliphatic polythiourea P11 were investigated as examples. Each polymer was reacted in the presence of NH₃•H₂O at 120 °C in DMSO for 4 h, respectively (Fig. 8b). After the treatment with NH₃•H₂O, the reaction solution was dropped into methanol to precipitate the polymer residue. For P2, only 10% of precipitate was collected, suggesting that most of the polymer was degraded; no precipitate was collected for P4, suggesting complete degradation of aromatic polythioureas. The ¹H NMR spectra of their reaction solutions indicated that aromatic diamines 3a and 3h emerged after 4 h of reaction (Fig. S20). On the contrary, almost no change was observed for aliphatic polythiourea P11, suggesting good stability of aliphatic thiourea moieties. Moreover, the degradation reactions of polythiourea P2, thiourea model compounds 6 and 9, respectively, were conducted with NH₃•H₂O in DMSO at 120 °C for 4 h. The HR-MS spectra were measured with the degradation solution (Fig. S21). From the degradation solution of P2, the presence of aromatic diamine monomer 3a was observed (found at 201.1017, calcd. at 201.1022); from the degradation solution of thioureas 6 or 9, the presence of p-toluidine 8 was detected from the HR-MS spectrum (found at 108.0810 for 6, and found at 108.0814 for 9, calcd. at 108.0808). These evidences suggested that the aromatic amines were the degradation products.

Fig. 8.

(a) The degradation of P1 after reacting with t-butylamine and n-hexylamine respectively for 1 h. Carried out at 100 °C under nitrogen in DMF, K₂CO₃ = 0.4 M. (b) Degradation results of P2, P4 and P11 in air. Carried out at 120 °C in air in DMSO for 4 h, V_NH3·H2O = 0.15 mL, V_DMSO = 0.10 mL.

3.4. Solubility and thermal stability

The polythioureas generally possessed good solubility in polar solvents such as DMSO and DMF despite the existence of abundant hydrogen bonds. Thermogravimetric analysis (TGA) suggested that their decomposition temperatures (T_d) at 5 wt% weight loss under nitrogen ranged from 210 to 286 °C (Fig. 9a), and the T_d of aromatic polythioureas were generally lower compared with those of the aliphatic polythioureas in the literature [9]. Such difference in thermal stability again indicated that the aromatic thiourea bonds were more labile compared with aliphatic thiourea bonds upon thermal treatment. Furthermore, DSC analysis suggested that the glass transition temperature (T_g) of these polythioureas was influenced by the polymer backbone structures. P1, P4, and P6 with mainly aromatic structure possessed T_g above 160 °C, while P2, P7 and P8 with flexible alkyl chains embedded in the mainchain possessed T_g below 130 °C (Fig. 9b).

Fig. 9.

(a) TGA curves of polythioureas P1-P11 with a heating rate of 20 °C/min under nitrogen. (b) DSC curves of polythioureas with a heating rate of 10 °C/min from 30 to 200 °C under nitrogen.

3.5. Mercury absorption and fluorescence sensing

The detection and removal of toxic mercury ion in aqueous media have received extensive attention [43]. Great efforts have been made to develop porous materials, such as graphenes [44], molecular sieves [45] and metal-organic framework/covalent organic framework (MOF/COF) [46,47] for mercury adsorption. Taking advantage of the strong coordination between the sulfur atom in thiourea and mercury ion, the aliphatic polythioureas have shown great potential in the detection and removal of Hg²⁺ in our previous work [9], which could serve as adsorbents for the efficient removal of Hg²⁺ from polluted water.

To compare the mercury absorption performance of aliphatic and aromatic polythioureas, P2, P6 and P11 were selected as the examples (Table 2). The initial concentration [Hg²⁺]₀ was 60 mg/L in acidic aqueous solution, and the DMF solution of polythiourea was added to 5 mL of the mercury solution. Precipitate was formed, which was removed by centrifugation after stirring for 1 h, and the remaining concentration [Hg²⁺] was measured by atomic fluorescence spectrometer (AFS) to calculate the removal efficiency. When the added DMF solutions of P2, P6 and P11 possessed identical ratio of [thiourea moiety]: [Hg²⁺] = 4, the removal efficiencies of P2, P6 and P11 were 83.55%, 99.99% and 76.55%, respectively, suggesting aromatic polythiourea P6 possessed higher efficiency compared to semi-aromatic polythiourea P2 or aliphatic polythiourea P11. To further study the sensitivity of mercury removal, different amounts of P6 were added and [thiourea moiety]: [Hg²⁺] was tuned from 1 to 16, suggesting that an efficiency of 99.05% could already be achieved when the ratio was 2. The remaining [Hg²⁺] was only 1.9 µg/L when the ratio was increased to 8, reaching the drinking water standard limit (< 2 µg/L).

Table 2.

The mercury removal efficiencies of P2, P6 and P11.^a

Entry	Polymer	[Hg2+]: [Polymer]	[Hg2+]0 (µg/L)	[Hg2+] (µg/L)	Efficiency (%)
1	P2 (2.4 mg)	1: 4	60,000	9870	83.55
2	P6 (1.1 mg)	1: 1	60,000	8430	85.95
3	P6 (2.2 mg)	1: 2	60,000	570	99.05
4	P6 (4.4 mg)	1: 4	60,000	2.587	>99.99
5	P6 (8.8 mg)	1: 8	60,000	1.913	>99.99
6	P6 (17.6 mg)	1: 16	60,000	0.129	>99.99
7	P11 (1.9 mg)	1: 4	60,000	14,070	76.55

^aP2, P6 and P11 were prepared by Table S5 and polymers were dissolved in DMF. V_Hg²⁺ = 5 mL, [Hg²⁺]₀ = 60 mg/L.

Moreover, P6 possessed a tetraphenylethene (TPE) moiety, which could show aggregation-induced emission (AIE) property when forming nanoparticles in a poor solvent. In DMF solution, P6 absorbed at 345 nm (Fig. S22), and showed weak emission at ∼500 nm; adding poor solvent water in DMF solution could enhance its emission. The photoluminescence spectra of P6 in DMF/water mixtures with different water fractions were measured, showing a continuous fluorescence enhancement when the water fraction was increased, and the maximum emission intensity was observed in a 90 vol% aqueous mixture, which is 45 times higher compared with that in DMF solution (Fig. 10a). The fluorescence quantum efficiencies of the DMF/water solution with 90 vol% water and DMF solution of P6 were 5.2% and 0.4%, respectively (Fig. 10b). P6 with AIE property could serve both as an adsorbent of Hg²⁺ and a fluorescent indicator for the simultaneous detection and removal of Hg²⁺. When fixed amount of DMF solution of P6 was added to aqueous solution of Hg²⁺ with different initial concentrations [Hg²⁺]₀ (0–165 mg/mL), respectively, the remaining [Hg²⁺] after agitating for 1 h and centrifugation to remove the precipitates were measured by AFS, which ranged from 1.6 × 10^-3 to 1.3 mg/mL with the removal efficiency generally above 99%. Meanwhile, the emission at ∼500 nm was significantly decreased when [Hg²⁺]₀ was higher, suggesting the remaining amount of fluorescent polymer in the solution after centrifugation was dramatically decreased (Fig. 10c,d).

Fig. 10.

(a) PL spectra of P6 in DMF/water mixtures with different water fractions (f_w). (b) Plot of relative emission intensity (I/I₀) versus water fraction of the aqueous mixtures. Excitation wavelength: 395 nm, [P6] = 8 mg/L. Φ_soln = fluorescence quantum yield in DMF solution, Φ_aggr = fluorescence quantum yield in DMF/water mixtures with 90 vol% water contents. (c) Fluorescence emission spectra were obtained by adding 73 µL (0.01 mmol) DMF solution of P6 (10 mg/mL) to 2 mL of mercury chloride aqueous solution with an initial concentration range of 0–165 mg/mL. (d) The relationship between the relative fluorescence intensity (I/I₀) and the residual mercury ion concentration [Hg²⁺]. Excitation wavelength: 395 nm. Insert figure: Fluorescence image of the solution after mercury ion adsorption under ultraviolet irradiation.

4. Conclusion

In this work, simply assisted by inorganic base, the MCP of elemental sulfur, diisocyanides, and diamines could be widely applicable to less reactive aromatic diamine monomers to polymerize efficiently to afford 10 polythioureas with satisfying yields and high M_ws, and to highly reactive aliphatic diamine monomers to polymerize to afford 5 polythiourea products selectively in excellent yields and high M_ws, mitigating side reactions. The general applicability of aromatic and aliphatic diamines, aromatic and aliphatic diisocyanides, combined with the diversified monomer combinations of MCPs, has endowed a great number of possible polythiourea structures which have never been reported before, especially asymmetric aromatic polythioureas. The rigid skeleton and conjugated structure of aromatic polythioureas have brought them unique chemical and material properties. While the thiourea bond showed a dynamic covalent feature, the aromatic thiourea bonds were more labile compared with aliphatic thiourea bonds, bringing opportunity for future design of dynamic polymer networks and sustainable polymer materials. Moreover, on account of the strong coordination of thiourea moieties and Hg²⁺, the polythioureas could be utilized to remove Hg²⁺ from polluted water, and the removal efficiency of different polythioureas was compared, revealing that the aromatic polythioureas possessed superior performance compared with aliphatic or semi-aromatic polythioureas. This work demonstrated the potential application of these polythioureas for mercury ion removal, showcasing their practical relevance. These aromatic polythioureas could also be designed with unique dielectric properties, enabling potential applications in flexible electronics, energy storage devices, and high-temperature dielectric materials. Furthermore, this unique synthetic approach could also bring access to asymmetric electron-deficient aromatic polythioureas with versatile and tunable catalytic property, enabling future design of polythiourea catalysts. The development of generally applicable polymerization approaches for the synthesis of versatile polythiourea structures and materials was hence anticipated to accelerate the exploration of functionality and application of these promising, sustainable sulfur-containing polymer materials.

CRediT authorship contribution statement

Wenxia Cao: Data curation. Rui Luo: Data curation, Writing – original draft. Rongrong Hu: Conceptualization, Supervision, Writing – review & editing. Ben Zhong Tang: Project administration, Resources, Supervision.

Biographies

Wenxia Cao received her B.S. degree in materials science and engineering from Wuhan University of Technology in 2017, and later received her master degree in materials science and engineering from South China University of Technology in 2020. Her research is focused on the development of elemental sulfur-based multicomponent polymerization.

Rui Luo received his B.S. degree in materials science and engineering from Fuzhou University in 2022. He is currently a master student at the school of materials science and engineering at South China University of Technology, under the supervision of Prof. Rongrong Hu. His research is focused on the development of multicomponent reactions and polymerizations with CS₂ and elemental sulfur.

Rongrong Hu received her B.S. degree from Peking University in 2007 and Ph.D. degree from the Hong Kong University of Science and Technology (HKUST) in 2011. She conducted her research associate study at HKUST during 2012–2014. She joined the Stake Key Laboratory of Luminescent Materials and Devices in South China University of Technology and started her independent work as associate professor in 2014 and was promoted to full professor in 2016. Her current research interests are focused on the development of multicomponent polymerizations and exploration of sulfur or selenium-containing functional polymer materials.

Ben Zhong Tang received his B.S. and Ph.D. degrees from South China University of Technology and Kyoto University in 1982 and 1988, respectively. He conducted postdoctoral research at University of Toronto in 1989–1994. He joined the Hong Kong University of Science and Technology in 1994 and was promoted to Chair Professor in 2008. He was elected to the Academician of Chinese Academy of Sciences in 2009 and the World Academy of Sciences for the Advancement of Science in Developing Countries in 2020. In 2021, he joined the Chinese University of Hong Kong, Shenzhen, as Dean of the School of Science and Engineering, with a concurrent appointment of X.Q. Deng Presidential Chair Professor. He is mainly engaged in the study of materials science, macromolecular chemistry and biomedical theranostics. He coined the concept of aggregation-induced emission (AIE), and his labs are spearheading the AIE research in the world.