Keywords
Deep eutectic solvent - Aqueous mixture - Photocatalysis - Amine oxidation - Heterogeneous catalysis
Photocatalyzed oxidative self-coupling of amine to imine in aqueous mixture of deep eutectic solvents is reported to have higher efficiency than conventional solvents. The use of modified reaction media is enabled by incorporating a dye in the catalyst assembly. The reaction required much lower levels of irradiation and the solvent system could be recycled up to five times with marginal compromise of the yield. The high selectivity of conversion simplifies the isolation of the product. The plausible mechanism and future scope is discussed.
Introduction
Selective oxidation of amines to imines is an important class of synthetic reactions because of the wide applicability of imines as precursors for a range of organic compounds such as nitrogen-containing heterocycles (e.g., functionalized piperidines, imidazopyridines, and pyrrolidinone derivatives) and alkaloids (e.g., quinolizidines and in-dolizidines) [1], [2]. The emphasis on incorporating environment-friendly strategies for the oxidation of amines has led to numerous protocols employing atmospheric oxygen as an oxidizing agent and other “green” measures [3], [4]. The use of TiO2-based photocatalytic systems for amine oxidation in acetonitrile was first reported by Lang and coworkers in 2011 and has subsequently been modified for better yields and selectivity [5], [6]. It was demonstrated that the replacement of acetonitrile with water as a reaction medium was possible by employing UV irradiation, with slightly lower conversion [7]. Subsequent reports focused on tuning the band gap of the TiO2 catalyst by sensitizing the surface with organic molecules and incorporating co-catalysts like TEMPO [(2,2,6,6-tetramethylpiperidin-1-yl)oxyl] [8]. For example, Shi et al. reported excellent yields and selectivity when TEMPO was used in conjunction with TiO2 coupled with phenol or its derivatives [9], [10]. Li and Lang coupled TEMPO with TiO2 as a “smart” photocatalyst for selective oxidation of amine to imine [11].
These improved synthetic protocols are mired by the fact that acetonitrile remains the most widely used solvent for photocatalytic oxidation to date, despite rising concerns about its potential hazard under various reaction conditions [12]. It must be noted that numerous studies reported the successful use of water or other green solvents for amine oxidation in thermal conditions [13]. Reports about the use of green solvents for photocatalytic oxidation of amines have been relatively limited. For example, Zhao and coworkers have reported excellent selectivity for photocatalyzed oxidative coupling of amines to imines in aqueous medium using bismuth oxybromide photocatalyst [14]. The scarcity of research related to the incorporation of “green” reaction media in photo-oxidative coupling of amines is even more surprising given the fact that solvents like water, ionic liquids, and deep eutectic solvents have been widely used for other photocatalyzed processes [13], [15], [16]. In particular, deep eutectic solvents (DES) have shown exceptional promise in terms of ease of handling, cost-effectiveness, and solvating ability [17], [18]. DES comprises eutectic mixtures of Lewis or Bronsted acid/base with a variety of cations/anions. Their use as reaction media is known to improve the reactivity and selectivity for a wide range of organic processes [19], [20].
Our research focuses on the use of green solvents as reaction media and investigating the underlying mechanistic principles [21], [22]. In the present work, we report our attempts toward tuning the reaction medium for the photocatalytic aerobic oxidation of amines to imines to ensure an efficient and circular synthetic process. Our results show that aqueous solutions of DES yield the most optimum reaction condition. Coupling the catalyst/co-catalyst system with aqueous DES allowed the use of a light source with lower intensity and was amenable to recycling protocols.
Results and discussion
The self-coupling reaction of benzylamine to imine was chosen as the model reaction for all the experiments ([Scheme 1]). Initial experiments used TiO2 as the photocatalyst and TEMPO as co-catalyst in various reaction media ([Table 1]). After 2h exposure to blue light irradiation (Hg lamp, ~450 nm, 30 W) with ethaline (1:2 choline chloride + ethylene glycol) as the reaction medium, excellent results were obtained with reline (1:2 choline chloride + urea) and glyceline (1:2 choline chloride + glycerol) – both yielding ~99% of imine in 2h of irradiation (entry 1 and 2). With ethaline (1:2 mixture of choline chloride and ethylene glycol), 70% of the benzyl amine was converted to imine product (entry 3). Interestingly, the yield was only 23 % after 2h of irradiation when pure acetonitrile was used as the solvent under identical conditions (entry 4). Higher yields were reported in the literature (69% conversion) when acetonitrile was used as a reaction medium but the difference could arise from the different luminous efficiency of the LED light source employed in the literature and the scale of the reaction (entry 6). It must be noted that our protocol leads to better conversions with higher selectivity at much lower irradiation intensity. The use of water led to better conversion (52%) as compared to acetonitrile under the present conditions. Previous studies of amine oxidation have also reported ~51% conversion when water was used as the medium [5] – albeit under very different reaction conditions. The previous study in water was carried out in the absence of TEMPO and required much higher irradiation (100 W, 9 h).
Scheme 1
Table 1
Photocatalyzed oxidation of amine to imine with (TiO2 + TEMPO) catalyst and irradiation by 30 W blue light (~450 nm) for 2 h.a
|
Entry
|
Solvent
|
% Conversionb
|
|
a 250 mg TiO2, 11.7 mg TEMPO, 0.168 mL benzylamine (1.5 mmol) in 5 mL solvent.
b GC analysis with anisole as internal standard.
c Reference 11: Reaction conditions: benzylamine (0.3 mmol), TiO2 (50 mg), TEMPO (0.003 mmol), CH3CN (1 mL), 460 nm blue LED irradiation (3W × 4), air (1 atm), 50 min.
|
|
1
|
Reline
|
99
|
|
2
|
Glyceline
|
~99
|
|
3
|
Ethaline
|
70
|
|
4
|
Acetonitrile
|
23
|
|
5
|
Water
|
52
|
|
6
|
Acetonitrile
|
42c
|
A plausible mechanism for the cooperative catalysis by TiO2 + TEMPO in acetonitrile was proposed by Li and Lang [11]. Accordingly, the adsorbed benzylamine acts as a surface ligand on TiO2 and is oxidized to a radical cation, which is restored by TEMPO to generate the TEMPO+ species. It is the TEMPO+, which converts benzylamine to imine in a two-step process while getting reduced to hydroxylamine form of TEMPO (TEMPOH). The TEMPO is regenerated by O2 to complete the catalytic cycle. It was important to control if the same catalytic pathway was valid even when DES was used as reaction media. Control experiments were performed to confirm the mechanistic details and are summarized in [Table 2]. The decrease in yield on purging the reaction medium with N2 confirms the role of O2 as the oxidant. The decrease in yield in the presence of p-benzoquinone indicates the role of superoxide radical (O2
–•) as the reactive species (which is quenched by p-benzoquinone). It is also important to note that both TiO2 and TEMPO work in tandem to catalyze the reaction and the absence of either one will lead to a compromised yield. The overall mechanism that can be inferred from these studies is consistent with the previously reported mechanism in molecular solvents [11]. This pathway is completely different from the one followed in the absence of TEMPO. The existence of two independent alternative pathways is justified by the fact that the oxidation still proceeds, albeit with lower yields, in the absence of TEMPO or TiO2 (entries 4 and 5). The use of DES as reaction media do not seem to alter the reaction pathway and their role is limited to influencing the yield of imine formation.
Table 2
Control experiments for the photocatalytic oxidative coupling of amine to imine in reline.a
|
Entry
|
Control condition
|
% Conversionb
|
|
a 250 mg TiO2, 11.7 mg TEMPO, 0.168 mL benzylamine (1.5 mmol) in 5 mL solvent, irradiation by 30 W blue light (~450 nm) for 2 h.
b GC analysis with anisole as internal standard.
c 50 mg (0.3 mmol) of silver nitrate added.
d 22 mg (0.2 mmol) of benzoquinone added.
e No further increase with time.
|
|
1
|
N2 atm
|
41
|
|
2
|
AgNO3
c
|
21e
|
|
3
|
Benzoquinoned
|
20e
|
|
4
|
No TEMPO
|
24
|
|
5
|
No TiO2
|
35
|
We explored the possibility of standardizing the work-up and recyclability of the reaction. The use of reline as a reaction medium gave a comparable yield for up to five or more cycles, with only a marginal loss in the conversion ([Fig. 1]). Centrifugation of the crude reaction mixture (which caused the Titania catalyst to settle down) was followed by extraction of the organic product with ethyl acetate. The near-stoichiometric conversion precluded the need for a purification process and the crude product has reasonably good purity, as confirmed by NMR analysis of the crude product ([Figs. S4] and [S5] of Supporting Information). The reline + TiO2 mixture could be charged with TEMPO and fresh reactant for the next cycle. It can be seen that there was a marginal loss in reactivity even up to the fifth cycle. The only drawback of the protocol was the high viscosity of reline, which necessitated the use of EtOAc for the extraction. This in turn hindered the benefits of recycling the reaction media—since every cycle of work-up would mean incremental emission of VOCs.
Figure 1 Percent conversion of amine with successive recycling of reline and TiO2 medium.
This led us to explore the use of co-solvent with DES as reaction media and prompted the use of water as a co-solvent in the DES reaction mixture. Previous reports have indicated that presence of water in DES may augment the reactivity and selectivity of biochemical processes [23]. On the other hand, the presence of water is bound to disrupt the hydrogen bond donor (HBD)–hydrogen bond acceptor (HBA) interaction in DES since water can itself act as an HBA/HBD. Varying amounts of water were added to the DES solvents and the % conversion was measured under conditions identical to those used for pure DES. It was observed that the effect of water was nonlinear, the reasons for which may be difficult to delineate ([Table 3]). While adding a small amount of water in reline led to a marginal change in the yield, increasing the water content beyond 50% v/v led to a sharp decrease in the conversion (entries 1, 3, and 5). The effect of water also depends on the identity of the DES component—the maximal effect on % conversion being observed in mixtures with ethaline and glyceline (entries 9 and 13). It must be noted that the presence of water had a kinetic effect and slowed down the reaction but did not alter the overall yield—all mixtures yield stoichiometric conversion after 24 h (entries 2, 4, 6, 10, and 14). The chief advantage of using 1:1 v/v mixture of DES and water was that the reaction medium was now easier for work-up and recycling primarily due to lower viscosity and hydrophobic nature of the substrate.
Table 3
Effect of composition of the reaction medium (DES + water) on the reactivity in photocatalyzed oxidation of benzylamine.a
|
Entry
|
Solvent
|
Composition of solvent (% v/v of DES)
|
% Conversionb
|
|
a 250 mg TiO2, 11.7 mg TEMPO, 0.168 mL benzylamine in 5 mL solvent, irradiation by 30 W blue light (~450 nm) for 2 h.
b GC analysis with anisole as internal standard.
c % conversion after 24 h.
|
|
1
|
Reline + water
|
95
|
90
|
|
2
|
|
|
99c
|
|
3
|
|
90
|
95
|
|
4
|
|
|
99c
|
|
5
|
|
50
|
55
|
|
6
|
|
50
|
91c
|
|
7
|
Ethaline + water
|
95
|
77
|
|
8
|
|
90
|
73
|
|
9
|
|
50
|
70
|
|
10
|
|
50
|
99c
|
|
11
|
Glyceline + water
|
95
|
91
|
|
12
|
|
90
|
92
|
|
13
|
|
50
|
60
|
|
14
|
|
50
|
99c
|
The observations showed that the presence of water at the TiO2 interface compromised the yield by possibly influencing the interaction between the semiconductor photocatalyst, co-catalyst, and substrate. One of the possibilities is that water molecules increase the local acidity of the TiO2 interface [24]. This in turn may lead to acid-promoted disproportionation of TEMPO to 4-hydroxy TEMPO (TEMPOL), which cannot be regenerated and hence, the net activity of TEMPO is limited [25]. Since the amount of “free” TEMPO (active form of the co-catalyst) is reduced in the presence of water, the reaction rate is expected to slow down ([Fig. 2]). One strategy to counter the detrimental effect of water would be to limit or prevent the access of water molecules on the interface of the TiO2 catalyst. In the current system, benzylamine is proposed to assemble on the interface and contribute to the photoactivation process. In order to overcome this hurdle, it was necessary to add a hydrophobic component that would shield the catalyst interface from the aqueous environment without affecting the light-harvesting pathway adversely.
Figure 2 Possible mechanism for effect of aqueous media on visible light-mediated selective aerobic oxidation of amine.
The presence of alizarin S (ARS) was intended to achieve the same effect with the additional benefits of augmenting the light-harvesting efficiency of the system ([Scheme 2]). In previous reports, ARS-TiO2 systems have been used for amine oxidation with encouraging results [26]. The reported amount of catalyst required was lower than that required in experiments without ARS. For the present report, yields for the benzylamine oxidation carried out using (TiO2 +ARS + TEMPO) photocatalyst in aqueous DES mixtures is shown in [Table 4]. It can be seen that the addition of ARS has the desired effect and is instrumental in improving the rate of conversion. The conditions yield complete conversion in approximately 2 h or less for all the DES mixtures.
Scheme 2
Table 4
Effect of composition of the reaction medium (DES + water) on the reactivity in photocatalyzed oxidation of benzylamine.a
|
Entry
|
Solvent
|
Composition of solvent (% v/v of DES)
|
% Conversionb
|
|
a 125 mg TiO2, 1.14 mg of ARS dye, 3.12 mg TEMPO, 0.109 mL benzylamine in 5 mL solvent, irradiation by 30 W blue light (~450 nm) for 2 h.
b GC analysis with anisole as internal standard.
c irradiation by 3 W blue light (~450 nm) for 8 h.
|
|
1
|
Reline + water
|
95
|
98
|
|
2
|
|
90
|
99
|
|
3
|
|
50
|
99
|
|
4
|
|
50
|
89c
|
|
5
|
Ethaline + water
|
95
|
96
|
|
6
|
|
90
|
96
|
|
7
|
|
50
|
96
|
|
8
|
|
50
|
86c
|
|
9
|
Glyceline + water
|
95
|
94
|
|
10
|
|
90
|
94
|
|
11
|
|
50
|
96
|
|
12
|
|
50
|
92c
|
|
13
|
Acetonitrile
|
–
|
39
|
Given the excellent conversion efficiency for all the solvent systems studied, we explored the possibility of reducing the energy consumption of the reaction by reducing the power of the irradiation source. The 30 W light source was replaced by a 3 W lamp with the same wavelength. The results show that although the reaction takes longer to reach completion, the conversion is almost stoichiometric after 8 h (entries 4, 8, and 12). It may be noted that when acetonitrile was used as the solvent with the (TiO2 + ARS + TEMPO) photocatalyst, the yield was 39%, in comparison to the 23% reported without ARS ([Table 1]). The addition of ARS ensures comparable yields but in a lower time and with exposure to low-intensity light sources. For example, it is possible to obtain comparable yields after 24h photocatalyzed oxidation in the absence of ARS and using a 30 W irradiation source (entry, 10 and 14 of [Table 3]). But the energy cost of the 2h exposure to a 3 W source (24 W-h) is lower than the ~24h exposure to a 30 W light source (720 W-h, [Table 3]). More importantly, photocatalysts enabling the use of low-intensity light sources are a step toward harnessing ambient sunlight and should be considered as the first choice. Clearly, the dye sensitization of the TiO2 catalyst is more effective in aqueous DES media than organic solvents.
The origin of this enhanced reactivity is the adsorption of the ARS dye on the catalyst surface and improve its photosensitivity. The presence of water in the reaction medium will thus attenuate the process by increased adsorption of the hydrophobic dye. This is evident from the fact that faster conversion is seen in reaction systems with a higher proportion of water. It may also be possible that ARS may initiate an alternative catalytic pathway, independent of the one facilitated by TEMPO. Detailed kinetic and mechanistic studies would be necessary to confirm the detailed reaction process. To confirm that the effect of ARS is in alignment with our hypothesis, we studied the adsorption of ARS on the surface of TiO2. Briefly, a mixture was ARS and TiO2 was allowed to equilibrate in the 1:1 v/v DES + water system. The supernatant solution was then analyzed using UV–visible spectroscopy to determine the amount of dye adsorbed.
Our results show that 26 μg of ARS was adsorbed per gram of catalyst in a 1:1 mixture of ethaline + water. The amount was slightly higher in the glyceline + water solvent system (30 μg/g of TiO2). This is comparable to the dye adsorbed when pure water was used for comparison (32 μg/g of TiO2). It must be noted that the least amount of dye was adsorbed when reline + water solvent mixture was used—the amount being only 6.5μg/g of TiO2 and this could be due to the “salting-in” effect of urea in aqueous solutions. Further discussions about the mechanism would require mechanistic studies, which are currently underway.
Conclusions
Substitution of organic solvent with an aqueous mixture of DES along with the use of the dye–TiO2–TEMPO catalyst complex results in substantial improvement of the yield for amine oxidation. Stoichiometric conversion (~99 %) is obtained using a TiO2 + TEMPO + ARS catalyst system with 1:1 water + reline solution in ~2 h. Using 3W irradiation also leads to similar yields, but requires longer exposure time (8 h). The use of (1:1 DES + water) as reaction media allows for better work-up conditions and lower energy requirements. Future efforts would include detailed mechanistic studies and substrate scope for the current protocol. The present work successfully demonstrates that (DES + water) can be a promising alternative to conventional reaction media—providing the means to improve the yields without compromising the environmental aspects.
Experimental section
Materials
More than 99% purity anatase TiO2, alizarin red dye, and TEMPO (2,2,6,6-tetramethyl-1-piperidinyloxy) were used as obtained. The purity of benzylamine was confirmed using GC-MS (>98%). High-purity choline chloride, urea, ethylene glycol, and glycerol were purchased and used without any purification for the preparation of DES. HPLC-grade water was used to make water fractions and LR-grade organic solvents were used for work-up and purification, if required.
Preparation of DES
All three DES used in the present work were prepared by mixing a 1:2 mol ratio of choline chloride with hydrogen bond donors in a round bottom flask. The mixture was stirred at 80°C till colorless homogeneous liquid formed. The resulting DES was dried in a vacuum oven at 60 °C for 8 h and stored in a desiccator till required. The three DES (reline, ethaline, and glyceline) were characterized by using FT-IR (see [Figs. S1–S3] of Supplementary Information).
General procedure for amine oxidation
First, 250 mg of anatase TiO2, 11.7 mg TEMPO, and 0.168 mL benzylamine (1.5 mmol) were added to a round bottom flask containing 5 mL solvent media [11]. The reaction mixture was sonicated for 5 min and then stirred under dark for 30 min to allow equilibration. The reaction was initiated by irradiating with blue light (~450 nm, 3W or 30W) and stirred at 1500 rpm. The progress of the reaction was monitored using TLC (5% EtOAc in n-hexane). Ethyl acetate and water (if not already present) were added to the reaction mixture for work-up. The pure product was isolated as a yellow viscous liquid and characterized using NMR and GC-MS ([Figs. S4 and S5] of Supplementary Information).
For the dye-sensitized experiments, 1.14 mg of ARS dye was added to 125 mg of anatase TiO2, 3.12 mg TEMPO and 0.109 mL benzylamine (1.0 mmol) in 5 mL of reaction medium [26]. The reaction conditions and work-up procedures were the same as above.
All reactions were carried out in a custom-made photoreactor, constructed as per previous literature [27]. The performance of the photoreactor was confirmed by checking the reproducibility of the previous reports.
GC analysis
GC analysis was conducted on Young Lin Autochro-3000 (6500 GC system) equipped with Elite 1701 column (15 m × 0.53 mm × 1 μm) and flame ionization detector (FID) using N2 as a carrier gas, with slight modifications to the reported method [11]. For analysis of the imine product, the injector temperature was maintained at 250 °C and detector temperature was 280 °C. The column temperature was programmed as 80 °C (hold 2 min) and raised up to 280 °C (at a rate of 40 °C min−1 and finally held for 2 min). Anisole was used as the internal standard to determine the % conversion and yield.
Adsorption studies
0.125 g of TiO2 and 1.14 mg ARS dye (3.33 × 10−3 mmol) were added in 5 mL solvent and allowed to equilibrate. The solution was centrifuged at 35,000 rpm for 10 min to separate the adsorbent from the solution. The supernatant solution was analyzed using a Shimadzu UV-1800 UV–visible spectrophotometer to determine the concentration of the dye. The amount of ARS dye absorbed on the TiO2 was determined using the following formula:
q
=
C
0
−
C
f
m
V
where q is the amount of ARS dye adsorbed on anatase TiO2 (mg g−1), C
0 is the initial strength of the ARS dye (mg L−1), Cf
is the dye strength after adsorption (mg L−1), V is the volume of the ARS dye (L), and m is the mass of the adsorbent (g).
Bibliographical Record
Varsha S. Kare, Shraeddha S. Tiwari. Aqueous Mixtures of Deep Eutectic Solvents: Tuning the Reaction Medium and Photocatalyst for Highly Selective Amine Oxidation. Sustainability & Circularity NOW 2025; 02: a25056356.
DOI: 10.1055/a-2505-6356
