Phenol–TiO2 complex photocatalysis: visible light-driven selective oxidation of amines into imines in air

Ji-Long Shia, Huimin Haoa and Xianjun Lang*ab
aCollege of Chemistry and Molecular Sciences, Wuhan University, Wuhan 430072, China. E-mail:
bKey Laboratory of Advanced Energy Materials Chemistry (Ministry of Education), Nankai University, Tianjin 300071, China

Received 30th October 2018 , Accepted 7th December 2018

First published on 7th December 2018

Strong interfacial charge-transfer (ICT) has been observed between phenol and TiO2. It was demonstrated that a single Ph–O–Ti linkage could induce ICT transitions in the visible light region. By utilizing the surface complexation of phenol with TiO2, we, herein, present a new photocatalytic protocol for the selective oxidation of amines to imines in air. Our success depended on merging the phenol–TiO2 complex photocatalyst with (2,2,6,6-tetramethylpiperidin-1-yl)oxyl (TEMPO), so called cooperative photocatalysis, which improved stability of the surface-complexed phenol under the oxidative circumstance and promoted the selective conversion of amines. With 5 mol% of TEMPO as a co-catalyst and phenol–TiO2 complex (containing 0.8 mol% of phenol) as a photocatalyst, amines could be efficiently oxidized into imines with atmospheric O2 under blue light-emitting diode (LED) irradiation. In addition, a mechanism is proposed to explain the visible-light photocatalytic performance of the present system.


Single-site heterogeneous catalysts have attracted considerable research attention as they can lead to more efficient use of the catalysts through enhanced reactivity and selectivity.1 The integration of single active sites with heterogeneous catalysts represents a wise strategy which, in principle, could drastically improve the activity of the catalysts.2 Due to the highly mobile nature of single active sites, aggregates are often formed during the synthesis and catalytic reactions without the extensive promotion effect.3 One solution is to anchor the single sites on the surface of a semiconductor via strong electrostatic adsorption.4 However, there are huge challenges in loading large amounts of metal atoms on semiconductor surfaces. For this reason, it is imperative to develop a practicable strategy where molecular catalysts, such as surface ligands, are grafted onto the light-harvesting semiconductor surface through bridging ligands. The grafting requires some special modifications so that the peripheral ligand of the molecular catalyst contains the anchor groups (Fig. 1a).
image file: c8se00527c-f1.tif
Fig. 1 Schematic of the (a) integrating molecular catalysts on a semiconductor surface and (b) anchoring phenol on a semiconductor surface.

For metal oxide semiconductors (e.g., TiO2), the most common anchor groups, such as carboxyl,5 phosphonate,6 catechol,7 and salicylic acid,8 have been explored. These anchor groups can be categorized into two main types: the former two functional groups only act as anchor groups without giving rise to a surface complex state; the latter two functional groups act as anchor groups, underscoring the new surface complex state. However, investigation on phenol as a modification molecule for the metal oxide surface was relatively scarce. If one can anchor the hydroxyl group of phenol onto TiO2 and induce the formation of a visible light active surface complex, the photocatalysis field could be further flourished with this new Frontier, considering the richness of dye molecules containing phenol motifs.

The incorporation of wide bandgap semiconductor and molecular photosensitizer by coordination chemistry can adjust the light absorption without changing the semiconductor redox potential. The most common method is to achieve charge transfer to semiconductors by dye sensitization. In this regard, we uncovered that the organic dyes with catechol motifs have been frequently used to modify TiO2, so as to be utilized efficiently for visible light photocatalysis.9–12 The photosensitizer is chemically adsorbed onto the semiconductor by a suitable bridge to provide enhanced stability, resulting in increased surface coverage and improved electron transfer than those from physical adsorption. Adjacent dihydroxyl groups were used as the surface-attached ligands to modify the semiconductor surface via the formation of surface complexes. TiO2 nanoparticles were synthesized through four different phenols (i.e. salicylate, 3,4-dihydroxy-3-cyclobutene-1,2-dione, pyridine-2,3-diol and catechol)–titanium complexes, and the characterization of samples in both complex and nanoparticle states was investigated; due to their appropriate bandgaps and ability to absorb UV light, they can be used as photocatalysts.13 The surface complexation of TiO2 by catechol or salicylate could alter the nature of surface status and therefore, produced a bathochromic shift in their absorption, causing an electron transfer from the organic molecules into the conduction band of TiO2.14 The catechol–TiO2 complex was used for the photocatalytic generation of H2 or reduction of Cr(VI).15,16 Many investigations explored how the surface complexes formed between salicylic acid and TiO2 could induce electron transfer for the decomposition of p-nitrophenol.17,18 However, the application of phenol–TiO2 complex as a photocatalyst was rare.

Phenol and its derivatives are generally considered to be organic pollutants, which have been recognized as harmful substances to both the human health and the environment. Photocatalysis has demonstrated the efficiency for degrading multiple organic contaminants into biodegradable or less toxic organic compounds.19 Phenol has a well-known molecular structure and is easily degradable by photocatalysis. Fe–bipyridyl complexes,20 metal–quinoline complexes21 and TiO2 (ref. 22) were used for improving visible light photocatalytic degradation of phenol. The photooxidation of phenol by phthalocyanine complex anchored on the metal oxide supports (Al2O3 or TiO2), has been studied upon irradiation with visible light.23 However, these methods are contrary to the green chemistry philosophy of consuming as little energy as possible to produce useful substances.

It was reported that interfacial charge transfer transitions can be induced by a single linkage between TiO2 and phenol (Fig. 1b).5 It was also reported that a charge transfer complex forms between surface Ti atoms and organic mono-hydroxyl compounds.24 In view of previous success, strong ICT has been observed in surface complexes between the TiO2 nanoparticles and the aromatic compounds with at least one hydroxyl group of phenol. Interfacial charge transfer complex formation between TiO2 powder and variety of para-substituted phenol derivatives was achieved and the bathochromic shift of optical absorption was observed.25 In the above examples, phenol and its derivatives function as a convenient choice to mimic the behavior of complexes on TiO2 that only exhibit ICT. To achieve the application of phenol–TiO2 complex in the photocatalytic field is of considerable significance. Compared with the photocatalytic organic reduction reactions, oxidation reactions are a more challenging task as highly oxidative conditions might result in the degradation of the coordination structures. Chlorophenols can be photodegraded on anatase TiO2 via ICT under visible light irradiation.26 Surface complexes could also not be stable between 2,4,5-trichlorophenol and anatase TiO2, thus the degradation of 2,4,5-trichlorophenol failed to be initiated.27 Thus, the substituted groups on phenol could substantially determine the ICT process, which supported our design principle.

Imines are widely applied as valuable intermediates in pharmaceutical syntheses and organic transformations. Several very nice examples of heterogeneous photocatalysts aiming at the selective oxidation of amines to imines have been disclosed.28–34 Therefore, it would be possible to develop a system that utilizes the surface complexation of phenol derivatives to achieve selective and efficient amine oxidation under irradiation by advanced blue light emitting diodes (LEDs), generating value-added products. At the same time, it solved the environmental pollution caused by phenol. Unfortunately, related reports remain extremely rare. Prior to this report, we demonstrated the photocatalytic conversion of amine using a catechol–TiO2 complex photocatalyst under visible light irradiation with the aid of TEMPO co-catalyst.35 Herein, we attempted to conduct further studies to clarify the visible-light-induced photocatalysis for amines oxidation assisted by the phenol–TiO2 surface complexes. It is essential to explore the range of organic molecules that are suitable for the interfacial charge transfer complex formation with TiO2, from diol compounds to mono-hydroxy compounds. TEMPO, a so-called smart catalyst, has been widely used in catalytic chemistry because it functions as a co-catalyst for the cooperative photocatalysis.36 Inspired by our prior successes,9–11 TEMPO was added to the reaction system as a cooperative redox catalyst to improve the conversions of substrates and selectivity of products by ensuring the stability of the phenol–TiO2 complex in the oxidation environment. We successfully transformed the phenol–TiO2 complex into a functional photocatalyst for the selective oxidation of amines. Our success depends on the combination of the phenol–TiO2 complex photocatalyst with TEMPO, so called cooperative photocatalysis, which improves the stability of the surface-complexed phenols under the oxidative environment for the conversion of amines.

Results and discussion

The phenol–TiO2 surface complex was formed between phenol and TiO2 in CH3CN at room temperature without any additional conditions. The phenol–TiO2 complex was directly used to test the photocatalytic activity of phenol and its derivatives for the selective oxidation of amines under a blue LED irradiation. The influence of the substituted groups of phenol on the selective aerobic oxidation of amines was investigated first (BET specific surface areas: phenol–TiO2 complex, 251 m2 g−1; 4-methoxylphenol–TiO2 (4-OCH3–phenol–TiO2) complex, 220.9 m2 g−1; 4-trifluoromethylphenol–TiO2 (4-CF3–phenol–TiO2) complex, 248.7 m2 g−1; PXRD characterization, see Fig. S1). Previously, it was discovered that the introduction of a moderately electron-withdrawing substituent, such as an acetyl group or a cyano group, onto the catechol led to an increase in the efficiency of the catechol–TiO2 complex for dye-sensitized solar cells.37 Here, the introduction of different substituents, either a strong electron-donating one or a strong electron-withdrawing one on the phenol units of the phenol–TiO2 surface complex, failed to enhance the photocatalytic activity for the selective oxidation of amines to imines under blue LED irradiation compared with the scenario of phenol (entries 1–3, Table 1). We speculated that this might be related to the properties of the surface complex itself. However, the specific reasons for this observation need to be further studied to explain the substitution effect.
Table 1 The influence of different phenol derivatives on the visible light-driven selective oxidation of benzylamine to iminea

image file: c8se00527c-u1.tif

Entry Surface ligand Conv.b [%] Sel.b [%]
a Reaction conditions: benzylamine (0.3 mmol), phenol (2.4 × 10−3 mmol), TiO2 (50 mg), TEMPO (0.015 mmol), LED irradiation (3 W × 4), air (1 atm), CH3CN (1 mL), 30 min.b Determined by GC-FID using chlorobenzene as the internal standard, conversion of benzylamine, selectivity of N-benzylidenebenzylamine.
1 Phenol 83 99
2 4-Methoxylphenol 61 97
3 4-Trifluoromethylphenol 63 99

In order to achieve not only an effective photocatalytic oxidation at the target site but also a good conversion rate and selectivity for the phenol–TiO2 complex, our next investigation began with the evaluation of a series of reaction parameters for the selective oxidation of benzylamine (Table 2). We were pleased to find out that the phenol–TiO2 complex could serve as an effective photocatalyst for the desired transformation under blue LED irradiation (entry 1, Table 2). Visible light played a crucial role in our reaction system because the reaction stopped immediately without it (entry 2, Table 2). TEMPO is an excellent cooperative catalyst. Both the conversion rate and the selectivity of the reaction are significantly lower in the absence of TEMPO (entry 3, Table 2). The phenol–TiO2 complex can effectively enhance the selective oxidation of amines compared to pure TiO2 under visible light irradiation. In the case of pure TiO2, the addition of TEMPO does not significantly improve the reaction process (entries 4–5, Table 2). Because the phenol–TiO2 complex can be excited by visible light, electrons can be transferred to the conduction band of TiO2, which promotes the conversion of oxygen to superoxide radicals (O2˙) and thus, positively affects the photocatalytic process. TiO2 is an essential component in the present photocatalytic reactions (entries 6–8, Table 2). In the absence of the photocatalytic components, the reaction did not proceed at all (entry 9, Table 2).

Table 2 Control experiments for the photocatalytic selective oxidation of benzylamine to imine under visible light irradiationa

image file: c8se00527c-u2.tif

Entry Conditions Conv.b [%] Sel.b [%]
a Reaction conditions: benzylamine (0.3 mmol), phenol (2.4 × 10−3 mmol), TiO2 (50 mg), TEMPO (0.015 mmol), LED irradiation (3 W × 4), air (1 atm), CH3CN (1 mL), 20 min.b Determined by GC-FID using chlorobenzene as the internal standard, conversion of benzylamine, selectivity of N-benzylidenebenzylamine.
1 Standard 55 96
2 Dark 0 0
3 NO TEMPO 23 97
4 TiO2 8 99
5 TiO2, TEMPO 9 99
6 TEMPO 0 0
7 Phenol, TEMPO 0 0
8 Phenol 0 0
9 Blank 0 0

Thus, different factors, like the amount of TEMPO (Table S1), the type of LED (Table S2), and different solvents (Table S3) were examined to identify the critical issues in the selective oxidation of amines to imines. A prominent subset of photocatalytic reactions is the selective aerobic oxidations, which utilize semiconductor materials, such as pure TiO2, to drive difficult organic transformations, such as the production of imines from benzylamines. An increased amount of TiO2 led to significantly advance the conversion for benzylamine and selectivity for imine (Fig. 2). Furthermore, the influence of different TiO2 on the selective aerobic oxidation of benzylamine to imine under visible light irradiation was investigated (Table S4).

image file: c8se00527c-f2.tif
Fig. 2 The influence of the amount of TiO2 on the selective aerobic oxidation of benzylamine to imine under visible light irradiation. Reaction conditions: benzylamine (0.3 mmol), phenol (2.4 × 10−3 mmol), TEMPO (0.015 mmol), blue LED irradiation (3 W × 4), air (1 atm), CH3CN (1 mL), 30 min. Conversions and selectivity were determined by GC-FID using chlorobenzene as the internal standard, for conversion of benzylamine, and selectivity of N-benzylidenebenzylamine.

Having obtained the optimum reaction conditions for benzylamine oxidation, the scope of substrates for different types of amines was explored. Various primary amines, including aromatic, heteroaromatic and di-substituted derivatives were smoothly oxidized with high conversions and selectivity (Table 3). The photocatalytic performance of the phenol–TiO2 complex for the selective aerobic oxidation of amines to imines with the aid of TEMPO under visible light irradiation is much better in terms of both selectivity and efficiency than that of the photocatalytic selective oxidation of amines with TiO2 under UV irradiation with a 100 W of Hg lamp or visible light irradiation with a 300 W Xe lamp.38,39 The light source plays a decisive role in ensuring high efficiency of the photocatalytic process for the selective oxidation of amines. More efficient blue LED was used for the photocatalytic selective oxidation of amines, achieving the same conversion results under the same experimental parameters within a shorter timespan of 0.6 h.

Table 3 Visible light-driven selective oxidation of primary amines into imines in air by combining the phenol–TiO2 complex photocatalyst with TEMPOa

image file: c8se00527c-u3.tif

Entry Substrate Product T (h) Conv.b [%] Sel.b [%]
a Reaction conditions: primary amine (0.3 mmol), phenol (2.4 × 10−3 mmol), TiO2 (50 mg), TEMPO (0.015 mmol), blue LED irradiation (3 W × 4), air (1 atm), CH3CN (1 mL).b Determined by GC-FID using chlorobenzene as the internal standard for the conversion of amine and the selectivity of the corresponding imine.c Benzylamine (1.2 mmol), phenol (4.8 × 10−3 mmol), TiO2 (100 mg), TEMPO (0.06 mmol), blue LED irradiation (3 W × 4), air (1 atm), CH3CN (mL).
1c image file: c8se00527c-u4.tif image file: c8se00527c-u5.tif 1.6 95 98
2 image file: c8se00527c-u6.tif image file: c8se00527c-u7.tif 0.6 95 99
3 image file: c8se00527c-u8.tif image file: c8se00527c-u9.tif 0.5 97 99
4 image file: c8se00527c-u10.tif image file: c8se00527c-u11.tif 0.8 93 99
5 image file: c8se00527c-u12.tif image file: c8se00527c-u13.tif 1.0 90 99
6 image file: c8se00527c-u14.tif image file: c8se00527c-u15.tif 0.5 95 99
7 image file: c8se00527c-u16.tif image file: c8se00527c-u17.tif 0.5 96 99
8 image file: c8se00527c-u18.tif image file: c8se00527c-u19.tif 0.5 89 99
9 image file: c8se00527c-u20.tif image file: c8se00527c-u21.tif 0.5 91 99
10 image file: c8se00527c-u22.tif image file: c8se00527c-u23.tif 0.8 93 99
11 image file: c8se00527c-u24.tif image file: c8se00527c-u25.tif 0.5 93 92
12 image file: c8se00527c-u26.tif image file: c8se00527c-u27.tif 1.2 98 97
13 image file: c8se00527c-u28.tif image file: c8se00527c-u29.tif 0.5 86 98

A wide arrange of benzylamines with electron-withdrawing or electron-donating substituted groups can be successfully oxidized to their corresponding imines under the standard conditions (Table 3). To scale up, 0.102 g of N-benzylidenebenzylamine can be afforded in 87% isolated yield, in which the initial concentration of benzylamine is 1.2 mol L−1 (entry 1, Table 3). We have listed the detailed 1H and 13C NMR and HRMS data for the specific reaction in S5. With a reaction time of 30 min, the phenol–TiO2 complex photocatalyst was recycled three times. There was only a slight decrease in conversion from 83% to 73% upon recycling. Obviously, substituted benzylamine with electron-donating groups reacts faster than benzylamine within the same time (entries 3, 4 and 9 vs. entry 2, Table 3). Careful observations were repeated for the methoxy (CH3O–)-substituted benzylamine, which exhibited a different conversion rate at the standard conditions. This is probably due to the existence of an electronic effect since benzylamine with CH3O– substitutions at different positions on the phenyl ring showed reaction rates in the order of ortho- < meta- < para- substituted benzylamines (entries 3, and 9–10, Table 3). When comparing with the different para-substituted benzylamines, the amines with electron-withdrawing substituents on the phenyl ring exhibited decreased conversions than those bearing electron-donating ones (entries 3 vs. 6–8, Table 3). However, methyl- and 4-tert-butyl-substituted benzylamines have the opposite photocatalytic performances (entries 4–5, Table 3), and the specific reasons need further study. To our delight, the phenol–TiO2 complex was able to oxidize di-substituted benzylamine and heteroaromatic amines into the corresponding imines with excellent conversion rates and selectivity (entries 11–13, Table 3).

It is more challenging to oxidize the secondary amines than the primary amines under the same experimental parameters.40,41 Secondary amines also afforded the corresponding imines with sluggish conversion and moderate selectivity over the phenol–TiO2 complex under visible light irradiation (Table 4). However, in the case of the selective oxidation of secondary amines conversion was lower and a much longer reaction time as well as a lower concentration (0.2 mmol) of substrates were required for conversion. Dibenzyl amine and its derivatives could be successfully oxidized to their desired imine products with the aid of the phenol–TiO2 complex under visible light irradiation (entries 1–7, Table 4). Dibenzyl amine substituted with electron-donating groups or electron-withdrawing groups were suitable substrates for the reaction, affording the corresponding imines in considerable yields. Furthermore, electronic properties of the substituents on the phenyl ring have some effects on the efficiency of the reaction: the former were more favored (entries 2–4, Table 4) to be oxidized than the latter (entries 5–7, Table 4). Moreover, higher reaction rates for para-substituted substrates, relative to the meta isomers, revealing the presence of a steric effect (entry 6 vs. entry 7, Table 4). However, to verify the oxidative performance of our photocatalytic system, we investigated different-substituted N-t-butyl benzylamines, which were demonstrated to have poor reactivity in our systems because steric hindrance around the nitrogen atom is a key factor for determining the reaction rate (entries 8–13, Table 4). For N-isopropyl benzylamine, since the Cα–H bond is away from the benzene ring, the selectivity is lowered (entry 14, Table 4).

Table 4 Visible light-driven selective oxidation of secondary amines into imines in air by combining the phenol–TiO2 complex photocatalyst with TEMPOa

image file: c8se00527c-u30.tif

Entry Substrate Product T (h) Conv.b [%] Sel.b [%]
a Reaction conditions: secondary amines (0.2 mmol), phenol (2.4 × 10−3 mmol), TiO2 (50 mg), TEMPO (0.015 mmol), blue LED irradiation (3 W × 4), air (1 atm), CH3CN (1 mL).b Determined by GC-FID using chlorobenzene as the internal standard for the conversion of benzylamine and selectivity of N-benzylidenebenzylamine.
1 image file: c8se00527c-u31.tif image file: c8se00527c-u32.tif 1.0 87 58
2 image file: c8se00527c-u33.tif image file: c8se00527c-u34.tif 0.9 85 62
3 image file: c8se00527c-u35.tif image file: c8se00527c-u36.tif 1.0 75 62
4 image file: c8se00527c-u37.tif image file: c8se00527c-u38.tif 1.0 79 76
5 image file: c8se00527c-u39.tif image file: c8se00527c-u40.tif 1.5 72 52
6 image file: c8se00527c-u41.tif image file: c8se00527c-u42.tif 1.5 74 56
7 image file: c8se00527c-u43.tif image file: c8se00527c-u44.tif 1.5 66 45
8 image file: c8se00527c-u45.tif image file: c8se00527c-u46.tif 3.0 77 88
9 image file: c8se00527c-u47.tif image file: c8se00527c-u48.tif 3.0 77 88
10 image file: c8se00527c-u49.tif image file: c8se00527c-u50.tif 4.0 63 90
11 image file: c8se00527c-u51.tif image file: c8se00527c-u52.tif 2.5 79 79
12 image file: c8se00527c-u53.tif image file: c8se00527c-u54.tif 3.0 69 82
13 image file: c8se00527c-u55.tif image file: c8se00527c-u56.tif 3.0 42 81
14 image file: c8se00527c-u57.tif image file: c8se00527c-u58.tif 1.0 79 59

For a deeper understanding of the impact of the electronic effect of the substituted groups, the relative rates of oxidation of para-substituted benzylamines (p-Me, p-OMe, p-H, p-F, p-Cl, and p-Br) were investigated (Fig. 3b and S2). We did not begin from time zero to make sure that the linear fitting was better. There might be an induction period for the visible-light-driven complex formation process. It can be clearly seen that the photocatalytic selective oxidation of benzylamine follows a zero-order kinetic model. Benzylamine with no substitutions displayed 96% conversion with a reaction constant kH of 0.00787 mol L−1 mol−1. Kinetically relevant elementary steps in the reaction pathways were measured by changing the benzylic H to D in benzylamine, with a reaction constant kD of 0.00488 mol L−1 mol−1; the obtained kinetic isotope effect (KIE) value was 1.63 (eqn (1)). This value indicates that it belongs to a primary KIE. Although the value is out of the range, it reveals that the cleavage of Cα–H is the rate-determining step for the photocatalytic selective oxidation of benzylamine. It also manifests that 2,2,6,6-tetramethylpiperidine-1-oxoammonium (TEMPO+) captures the hydrogen of the amines and is reduced to 2,2,6,6-tetramethylpiperidin-1-ol (TEMPOH), as observed during the analysis of our control experiments and ESR characterization data. The Hammett plot can be used to interpret the electronic effect of para-substituents on the reaction rate (Fig. 3b), suggesting the participation of free radicals in the reaction.

image file: c8se00527c-f3.tif
Fig. 3 Reaction kinetics studies on the visible light-driven selective oxidation of amines in air by combining the phenol–TiO2 complex photocatalysis with TEMPO catalysis of (a) benzylamine (rounds), and benzyl-α,α-d2-amine (squares). (b) Hammett plot for the oxidation of para-substituted benzylamines.

The effect of different quenchers was investigated to recognize the reactive oxygen species (ROS) and possible intermediates (Table 5). In the case of nitrogen environment, the reaction efficiency was significantly slower (entry 1, Table 5), while O2 could significantly accelerate the progress of the photocatalytic reaction (entry 1, Table 1). The conversion rate of 10% under this O2-free photocatalytic condition was attributed to the fact that TEMPO acted as a stoichiometric oxidant for the oxidation of benzylamine. When 0.2 equiv. of para-benzoquinone was added to the reaction solution, the conversion sharply decreased, which was evidence of O2˙ is the pivotal ROS (entry 2, Table 5). Application of AgNO3 as a quencher revealed the presence of electron transfer (entry 3, Table 5). Next, we performed the reaction using CD3CN as a solvent. The reaction constants in CH3CN and CD3CN were almost identical except for some experimental variations (entry 4, Table 5). We believe that the energy transfer of singlet oxygen did not lead to the formation of imine. Since the lifetime of singlet oxygen is significantly longer in deuterated solvents, the oxidation rate in CD3CN should be much faster than that in CH3CN. This result suggests that electron transfer was the only pathway for the formation of imine. With adding 2 equiv. of isopropanol (IPA) into the system, the conversion of benzylamine was not influenced (entry 5, Table 5), indicating that ˙OH was not involved in the formation of the product.

Table 5 Quenching experiments to determine the ROS for the visible light-driven photocatalytic aerobic oxidation of benzylaminesa

image file: c8se00527c-u59.tif

Entry Quencher (equiv.) Roles Yieldb [%]
a Reaction conditions: benzylamine (0.3 mmol), phenol (2.4 × 10−3 mmol), TiO2 (50 mg), TEMPO (0.015 mmol), LED irradiation (3 W × 4), air (1 atm), CH3CN (1 mL), 30 min.b Determined by GC-FID using chlorobenzene as the internal standard for the conversion of benzylamine, and selectivity of N-benzylidenebenzylamine.c p-BQ is para-benzoquinone.d CD3CN (1 mL).
1 N2 (—) O2 replacement 10
2c p-BQ (0.2) O2˙ scavenger 7
3 AgNO3 (1) Electron scavenger 9
4d CD3CN (—) Singlet oxygen maintainer 80
5 IPA (2) ˙OH scavenger 75

The long-lived superoxide radical has been discovered in the irradiated TiO2 suspensions using a continuous-flow chemiluminescence detection system.42 In order to further confirm that O2˙ participated in the reaction, ESR technology was applied to affirm the free radicals that included in the reaction process (Fig. 3a). We observed the results of DMPO tracking O2˙, wherein the signal gradually increased with the reaction time changes. This can be clearly explained by the existence of O2˙ in the progress of photocatalytic oxidation of amines. Furthermore, we investigated the influence of initial O2 pressure on the selective aerobic oxidation of benzylamine to imine under visible light irradiation (Fig. 4b). With the elevation of initial O2 pressures, the conversion to imine increased. When the initial oxygen pressure increased to the limit that our reaction vessel could withstand, the trend gradually disappeared, suggesting that the prompting effect of O2 on the oxidation of benzylamine was quite inadequate. To sum up, O2˙ was an overwhelmingly critical ROS in the formation of imine under visible light irradiation.

image file: c8se00527c-f4.tif
Fig. 4 (a) The ESR signals of superoxide radicals captured by DMPO during the phenol–TiO2 complex photocatalysis. (b) The influence of initial O2 pressure on the oxidation of benzylamines.

TEMPO, which is a long-life radical that is persistent at room temperature, has a wide range of uses owing to its exceptional redox behavior, which gives rise to its latest prominence in catalysis. To gain a deeper understanding of the role of TEMPO on the amines oxidation process, a series of experiments and ESR characterization were performed (Fig. 5). When the same catalytic amount of TEMPOH was added to the reaction mixture instead of TEMPO under the same experimental parameter, the same photocatalytic performance as TEMPO could be achieved (eqn (2)), indicating that TEMPOH was oxidized to TEMPO by O2˙. We analyzed the change in TEMPO during the oxidation of benzylamine to imine by the strength of its electron spin resonance (ESR) signal. The TEMPO ESR signal gradually weakened in 4 minutes of illumination, and the signal immediately restored in one minute after the irradiation stopped. Through the above experiments, the synergistic effect of TEMPO on the photocatalytic selective oxidation of amines can be clearly demonstrated. The photocatalytic cleavage of Cα–H bond of benzylamine played an important role in the selective oxidation of amines under visible light irradiation. This agrees well with Fig. 8, in which the abstraction of the hydrogen in amines by TEMPO+ to TEMPOH is the key step.

image file: c8se00527c-f5.tif
Fig. 5 The ESR signals of TEMPO during the phenol–TiO2 photocatalysis.

The diffuse reflectance UV-visible spectra of TiO2 and surface ligand-TiO2 have been listed in Fig. S3. UV-visible spectroscopy showed that the optical properties of TiO2 were improved by the surface modification of phenol and its derivatives, expanding the visible light absorption range by forming a titanium(IV) charge transfer complex (Fig. 6a and b). Furthermore, absorbance area of phenol–TiO2 overlapped with the emission range of blue LED (Fig. 6c), which further improved the overall efficacy of the photocatalytic system for the chemical transformation reaction. No matter whether it is benzylamine or phenol or TEMPO, it cannot absorb visible light (Fig. S3a–c). The absorption by the phenol–TiO2 complex in the visible region underpinned the visible light-photocatalytic performance for the selective oxidation of amines into imines in air.

image file: c8se00527c-f6.tif
Fig. 6 Diffuse reflectance UV-visible spectra of (a) phenol–TiO2, 4-OCH3–phenol–TiO2, 4-CF3–phenol–TiO2, and TiO2; (b) UV-visible light absorbance of the phenol–TiO2 surface complex; (c) relative spectrum distribution of the blue LED.

In addition, we detailed the electron transfer process of the phenol–TiO2 complex (Fig. 7). Compared with dark conditions, it could effectively promote the electron transfer process to the TiO2 conduction band under visible light irradiation. Further increasing the intensity of illumination did not increase the signal further (Fig. 7, 3 min). However, in the presence of visible light irradiation, the signal returned to its initial state after an exposure with O2 (Fig. 7, O2 1 min), which indicated the electron transfer process from the adsorbed phenol to the conduction band of TiO2 and finally to O2.

image file: c8se00527c-f7.tif
Fig. 7 The ESR spectra of conduction band electrons of the phenol–TiO2 complex under visible light irradiation (77 K).

On the basis of the above results, we propose a possible reaction mechanism with respect to the conversion of amines into imines by phenol–TiO2 photocatalyst under visible light irradiation (Fig. 8). We venture that the ESR active species (confirmed by TEMPO) on the photocatalysts played a core role to that of the reported co-photocatalyst, which could facilitate the electron transfer to control the conversion of the transfer imidization reaction. The phenol–TiO2 surface complex is a type II sensitizer that can, upon excitation with visible light, induce the direct transfer of electrons from the organic substrate into the conduction band of TiO2 during the photocatalytic process. TEMPO interacts with the radical cations generated on the surface of the phenol–TiO2 complex photocatalyst to avoid its degradation by ROS. Then, TEMPO was into TEMPO+, which is a two-electron oxidant that could conduct the selective oxidation of benzylamine to benzenemethanimine, forming TEMPOH in the meantime. Benzenemethanimine couples with unreacted benzylamine to get the final product of N-benzylidenebenzylamine. TEMPOH will eventually be oxidized to TEMPO by O2˙ produced by O2 accepting conduction band electrons from TiO2.

image file: c8se00527c-f8.tif
Fig. 8 Plausible mechanism for the visible light-driven selective oxidation of benzylamine into imine in air by merging the phenol–TiO2 complex photocatalysis with TEMPO catalysis.


In summary, we have documented the phenol–TiO2 complex as an efficient photocatalyst for the selective oxidation of amines to imines with the aid of TEMPO as the co-catalyst under visible light irradiation. By applying this complex in the selective oxidation of amines, both primary and secondary amines can be oxidized to their corresponding imines with atmospheric O2 as the oxidant. Besides, we discovered that TEMPO, so-called smart catalyst, improves the stability of surface-complexed phenols under the oxidative environment as well as ameliorates the conversion of amines. This work laid a firm foundation for the visible light-driven selective organic transformations over the surface-complex photocatalysis.

Conflicts of interest

There are no conflicts to declare.


This work was financially supported by the National Natural Science Foundation of China (grant numbers 21503086 and 21773173), the Fundamental Research Funds for the Central Universities (grant number 2042018kf0212), the 111 project (B12015) and the start-up fund of Wuhan University.


  1. J. Jones, H. F. Xiong, A. T. Delariva, E. J. Peterson, H. Pham, S. R. Challa, G. S. Qi, S. Oh, M. H. Wiebenga, X. I. P. Hernández, Y. Wang and A. K. Datye, Science, 2016, 353, 150–154 CrossRef CAS PubMed.
  2. B. Zhang, H. Asakura, J. Zhang, J. G. Zhang, S. De and N. Yan, Angew. Chem., Int. Ed., 2016, 55, 8319–8323 CrossRef CAS PubMed.
  3. R. Lang, T. B. Li, D. Matsumura, S. Miao, Y. J. Ren, Y. T. Cui, Y. Tan, B. T. Qiao, L. Li, A. Q. Wang, X. D. Wang and T. Zhang, Angew. Chem., Int. Ed., 2016, 55, 16054–16058 CrossRef CAS PubMed.
  4. C. Gao, J. Wang, H. X. Xu and Y. J. Xiong, Chem. Soc. Rev., 2017, 46, 2799–2823 RSC.
  5. D. I. Won, J. S. Lee, J. M. Ji, W. J. Jung, H. J. Son, C. Pac and S. O. Kang, J. Am. Chem. Soc., 2015, 137, 13679–13690 CrossRef CAS PubMed.
  6. S. L. Esarey and B. M. Bartlett, Langmuir, 2018, 34, 4535–4547 CrossRef CAS PubMed.
  7. G. Odling, A. Ivaturi, E. Chatzisymeon and N. Robertson, ChemCatChem, 2018, 10, 234–243 CrossRef CAS.
  8. D. Finkelstein-Shapiro, S. H. Petrosko, N. M. Dimitrijevic, D. Gosztola, K. A. Gray, T. Rajh, P. Tarakeshwar and V. Mujica, J. Phys. Chem. Lett., 2013, 4, 475–479 CrossRef CAS PubMed.
  9. X. J. Lang, J. C. Zhao and X. D. Chen, Angew. Chem., Int. Ed., 2016, 55, 4697–4700 CrossRef CAS PubMed.
  10. Y. C. Zhang, Z. Wang and X. J. Lang, Catal. Sci. Technol., 2017, 7, 4955–4963 RSC.
  11. Z. Wang and X. J. Lang, Appl. Catal., B, 2018, 224, 404–409 CrossRef CAS.
  12. X. Li, J. L. Shi, H. M. Hao and X. J. Lang, Appl. Catal., B, 2018, 232, 260–267 CrossRef CAS.
  13. Y. Absalan, E. A. Fortalnova, N. N. Lobanov, E. V. Dobrokhotova and O. V. Kovalchukova, J. Organomet. Chem., 2018, 859, 80–91 CrossRef CAS.
  14. J. Moser, S. Punchihewa, P. P. Infelta and M. Graetzel, Langmuir, 1991, 7, 3012–3018 CrossRef CAS.
  15. K. L. Orchard, D. Hojo, K. P. Sokol, M. J. Chan, N. Asao, T. Adschiri and E. Reisner, Chem. Commun., 2017, 53, 12638–12641 RSC.
  16. P. Karthik, R. Vinoth, P. Selvam, E. Balaraman, M. Navaneethan, Y. Hayakawa and B. Neppolian, J. Mater. Chem. A, 2017, 5, 384–396 RSC.
  17. S. X. Li, F. Y. Zheng, W. L. Cai, A. Q. Han and Y. K. Xie, J. Hazard. Mater., 2006, 135, 431–436 CrossRef CAS PubMed.
  18. L. Li, Y. J. Feng, Y. Z. Liu, B. Wei, J. X. Guo, W. Z. Jiao, Z. H. Zhang and Q. L. Zhang, Appl. Surf. Sci., 2016, 363, 627–635 CrossRef CAS.
  19. C. C. Wang, J. R. Li, X. L. Lv, Y. Q. Zhang and G. S. Guo, Energy Environ. Sci., 2014, 7, 2831–2867 RSC.
  20. D. Wang, Y. N. Sun, Q. K. Shang, X. Y. Wang, T. T. Guo, H. Y. Guan and Q. Lu, J. Catal., 2017, 356, 32–42 CrossRef CAS.
  21. Y. W. Huang, Q. K. Shang, D. Wang, S. Yang, H. Y. Guan, Q. Lu and S. C. Tsang, Appl. Catal., B, 2016, 187, 59–66 CrossRef CAS.
  22. L. Mino, A. Zecchina, G. Martra, A. M. Rossi and G. Spoto, Appl. Catal., B, 2016, 196, 135–141 CrossRef CAS.
  23. V. Iliev, J. Photochem. Photobiol., A, 2002, 151, 195–199 CrossRef CAS.
  24. J.-i. Fujisawa, S. Matsumura and M. Hanaya, Chem. Phys. Lett., 2016, 657, 172–176 CrossRef CAS.
  25. D. N. Sredojevic, T. Kovac, E. Dzunuzovic, V. Dordevic, B. N. Grgur and J. M. Nedeljkovic, Chem. Phys. Lett., 2017, 686, 167–172 CrossRef CAS.
  26. S. Kim and W. Choi, J. Phys. Chem. B, 2005, 109, 5143–5149 CrossRef CAS PubMed.
  27. A. G. Agrios, K. A. Gray and E. Weitz, Langmuir, 2003, 19, 1402–1409 CrossRef CAS.
  28. B. Chen, L. Y. Wang, W. Dai, S. S. Shang, Y. Lv and S. Gao, ACS Catal., 2015, 5, 2788–2794 CrossRef CAS.
  29. S. Biswas, B. Dutta, K. Mullick, C. H. Kuo, A. S. Poyraz and S. L. Suib, ACS Catal., 2015, 5, 4394–4403 CrossRef CAS.
  30. Y. P. Bhoi and B. G. Mishra, Eur. J. Inorg. Chem., 2018, 2018, 2648–2658 CrossRef CAS.
  31. F. Z. Su, S. C. Mathew, L. Möhlmann, M. Antonietti, X. C. Wang and S. Blechert, Angew. Chem., Int. Ed., 2011, 50, 657–660 CrossRef CAS PubMed.
  32. L. Ye and Z. H. Li, ChemCatChem, 2014, 6, 2540–2543 CrossRef CAS.
  33. Y. Zheng, L. H. Lin, B. Wang and X. C. Wang, Angew. Chem., Int. Ed., 2015, 54, 12868–12884 CrossRef CAS PubMed.
  34. H. Liu, C. Y. Xu, D. D. Li and H. L. Jiang, Angew. Chem., Int. Ed., 2018, 57, 5379–5383 CrossRef CAS PubMed.
  35. J. L. Shi, H. M. Hao, X. Li and X. J. Lang, Catal. Sci. Technol., 2018, 8, 3910–3917 RSC.
  36. X. J. Lang and J. C. Zhao, Chem.–Asian J., 2018, 13, 599–613 CrossRef CAS PubMed.
  37. Y. Ooyama, K. Yamaji and J. Ohshita, Mater. Chem. Front., 2017, 1, 2243–2255 RSC.
  38. X. J. Lang, H. W. Ji, C. C. Chen, W. H. Ma and J. C. Zhao, Angew. Chem., Int. Ed., 2011, 50, 3934–3937 CrossRef CAS PubMed.
  39. X. J. Lang, W. H. Ma, Y. B. Zhao, C. C. Chen, H. W. Ji and J. C. Zhao, Chem.–Eur. J., 2012, 18, 2624–2631 CrossRef CAS PubMed.
  40. S. Furukawa, Y. Ohno, T. Shishido, K. Teramura and T. Tanaka, ACS Catal., 2011, 1, 1150–1153 CrossRef CAS.
  41. B. Yuan, R. F. Chong, B. Zhang, J. Li, Y. Liu and C. Li, Chem. Commun., 2014, 50, 15593–15596 RSC.
  42. D. B. Wang, L. X. Zhao, D. Wang, L. Yan, C. Y. Jing, H. Zhang, L. H. Guo and N. Tang, Phys. Chem. Chem. Phys., 2018, 20, 18978–18985 RSC.


Electronic supplementary information (ESI) available. See DOI: 10.1039/c8se00527c

This journal is © The Royal Society of Chemistry 2019