Diisopropylamine as a single catalyst in the synthesis of aryl disulfides

2.1 Materials and general methods

Thiols used for experiments were purchased from Sigma-Aldrich Co. (Poznan, Poland) and used without further purification. Amines were purchased from Sigma-Aldrich Co. Acetonitrile (ACN, 99.8%) was purchased from J.T. Baker Company (Gliwice, Poland). All spectra of the synthesized products were collected in the supporting information file (Supplemental material). ¹H NMR (400 MHz) and ¹³C NMR (101 MHz) spectra were recorded on a Bruker Avance III HD NanoBay (Bruker, Poznan, Poland) (600 MHz) spectrometer using CDCl₃ as solvent (99.8 atom % D). Deuterated NMR solvent was purchased from Sigma-Aldrich Co. GC analyses were performed on a Varian 3400 with a Megabore column (30 m) and TCD (Varian, Walnut Creek, CA, USA). Mass spectra of the products were determined by GC-MS analysis on a Varian Saturn 2100T, equipped with a BD-5 capillary column (30 m) and Finnigan Mat 800 ion trap detector (Finnigan, San Jose, CA, USA).

2.2 General procedure for the synthesis of disulfides

To the 25 ml one-necked round-bottom flask, 0.9 mmol of thiol, 0.009 mmol of iPr₂NH, and 1 ml of ACN were added. Next the reaction mixture was stirred at room temperature for 1.5–2 h. Next, the solvent and the catalyst were evaporated and the products were purified by column chromatography on silica gel eluting diethyl ether to give the expected products (Table 3, 1–8) with 92%–98% yield.

3.1 Optimization of the reaction conditions

We initially focused on choosing and testing various amines and solvents. As the starting point, the oxidative coupling of benzenethiol in various conditions was considered. All experiments were carried out in vials (capacity=10 ml) equipped with magnetic stirring bars and stoppers. Results are summarized in Table 1. The aprotic polar solvents accelerate the oxidation of thiols in the presence of oxygen [35]. By using the measurements reported by Christian Reichardt [36], we decided to check the following polar aprotic solvents (relative polarity): acetonitrile (0.460), tetrahydrofuran (0.207), and diethyl ether (0.117). DMF (0.386) was not considered as the solvent in our tests, because of the results (95% conversion of thiols but after 24 h at rt) obtained by Ruano and co-workers [28]. Dimethylformamide, despite its relatively high polarity, is also known as toxic aprotic solvent. In addition, its boiling point is 153°C, which poses serious problems on distillation. Subsequently, various amines were tested as catalysts in the oxidative coupling of benzenethiol. The addition of these reagents permits the formation of thiolate species from thiols. In this connection, the basicity of amines is the major factor in these transformations. By using basicity measurements [37], [38] and considering the costs of nitrogen-containing reagents (www.abcr.de), we decided to check the following amines: diethylamine (Et₂NH, 2.5 l~35 EUR), triethylamine (Et₃N, 2.5 l~45 EUR), diisopropylamine (iPr₂NH, 2.5 l~40 EUR). Secondary and tertiary alkyl amines were selected and tested in the reaction; aromatic ones were not chosen because of their lower basicity and high toxicity compared with aliphatic derivatives [39].

Results showed that the acceleration of thiol oxidation caused by polar aprotic solvents was substantiated. The best results were consistently obtained for the mixture of reagents in acetonitrile. As mentioned, ACN demonstrated the highest polarity among the solvents tested. More interesting was the selection of the most suitable amine. By considering several factors, such as boiling points, relative toxicity, time of thiols’ conversion, and costs of reagents, diisopropylamine was selected to further work (Table 1, entry 7). In order to optimize the conditions of this reaction, we also studied the influence of the vessel capacity and solvent’s volume. The results are shown in Table 2.

Additional increase in the vessel capacity was found to have a positive influence on the time of conversion. As expected, larger capacity levels led to higher amounts of oxygen, which decisively decrease the time of the reaction (Table 2, entries 1–3). The amount of solvent can also be reduced to 1 ml. Finally, we decided to check the rate of stirring. When the stirring rate was increased to 1200 rpm (Table 2, entries 5 and 6), we observed our best results (Table 2, entry 6). The possibility of the over-oxidation product formation was investigated, and the findings proved that our method can be controlled to stop at the disulfide stage, without the over-oxidation of any by-product.

3.2 Synthesis of various disulfides catalyzed by diisopropylamine

With the chosen catalyst in hand, we next investigated various types of thiols under the optimized conditions. Aryl (Table 3, entries 1–8), alkyl, and heterocyclic thiols were used in this process. Results are summarized in Table 3.

Table 3:

The oxidative coupling of aryl thiols in the presence of diisopropylamine.^a

Entry	R	Time (min)	Product	Yield (%)b
1	Ph	105	Ph2S2 (1)	98
2	2-MePh	105	(2-MePh)2S2 (2)	93
3	3-MePh	105	(3-MePh)2S2 (3)	93
4	4-MePh	105	(4-MePh)2S2 (4)	95
5	4-FPh	80	(4-FPh)2S2 (5)	95
6	2-FPh	90	(2-FPh)2S2 (6)	93
7	4-ClPh	90	(4-ClPh)2S2 (7)	95
8	3-MeOPh	90	(3-MeOPh)2S2 (8)	92

1. ^aReactions conditions: thiol (1 eq., 0.9 mmol), solvent (2 ml), stirring rate (1200 rpm), flask capacity (25 ml), air atmosphere, closed system, room temperature; ^bisolated yields.

Aryl thiols were reactive in this process, and all of them gave the expected disulfides with excellent yields (92%–98%). The nature of substituents attached to phenyl ring in aryl mercaptans demonstrated a minor effect on the time of reaction. The presence of electron withdrawing groups (Table 3, entries 5–8) slightly shortens the thiols’ conversion. These groups considerably facilitate the formation of thiolate species. The selectivity of the proposed reaction was also excellent, which was proven by the absence of product over-oxidation. However, in some cases, we observed trace amounts of diaryl sulfides (<2%). Alkyl thiols are extensively used in nanotechnology [40], but they are less reactive than aryl derivatives due to their higher pK_a. Unfortunately, we observed the same dependence in the oxidative coupling catalyzed by diisopropylamine. When we used 0.1 eq. of the catalyst, we observed unsatisfying results (conversion <40%/6 h). The same feedback was also attained for heterocyclic thiols (conversion <60%/3 h) when we increased the amine amount (0.3 eq.).