Photocatalytically green synthesis of H₂O₂ using 2-ethyl-9,10-anthraquinone as an electron condenser

Abstract

A high level (∼9 mM) of hydrogen peroxide is photocatalytically synthesized by using O₂ as an oxidant and 2-ethyl-9,10-anthraquinone (EAQ) as an electron condenser. The photo-catalytic efficiency of the EAQ-assisted H₂O₂ production is ca. 10-fold higher than that of H₂ generation with Pt/P25.

---

Hydrogen peroxide (H₂O₂) is a clean oxidant that emits only water as a byproduct and is widely used in industries for organic synthesis,¹ pulp bleaching, waste-water treatment, and as a disinfectant.² Traditionally, H₂O₂ is produced with an anthraquinone oxidation (AO) process which combines H₂ with O₂ through a multistep route via an anthraquinone/anthrahydroquinone cycle. Recently, from the viewpoint of green and sustainable chemistry, direct synthesis of H₂O₂ from H₂ and O₂ has attracted a lot of attention.³ However, the potentially explosive nature of H₂/O₂ gas mixture brings a serious risk to the operation process; further, highly pure H₂ gas is indispensable which requires significant energy costs in both the AO process and the H₂/O₂ direct synthesis process. To circumvent the above disadvantages, photocatalytic attempts have begun to emerge nowadays.⁴ By using methanol/ethanol as a sacrificial electron donor, direct synthesis of H₂O₂ is reported to be available.⁵

However, direct photocatalytic generation of H₂O₂ from methanol is not an easy task due to the undesirable rapid reverse reactions^5c as well as the over-reduction of surface hydroperoxo species.⁶ As we know, the photocatalytic production of H₂O₂ is a multistep process, which is briefly shown below.^5c,7,8

Ti³⁺(e⁻) + O₂ → Ti⁴⁺–O₂˙⁻ (1)

Ti³⁺(e⁻)–O₂˙⁻ + H⁺ → Ti⁴⁺–⁻O₂H (2)

Ti⁴⁺–⁻O₂H + H⁺ → Ti⁴⁺ + H₂O₂ (3)

The photoelectron on the TiO₂ surface, Ti³⁺(e⁻), reduces the adsorbed O₂ and produces a superoxo radical.^5a Then the superoxo radical is transformed to a hydroperoxo specie via further reduction, and protonated to produce H₂O₂.^7,8 Unfortunately, the hydroperoxo species are easily decomposed by further reduction with e⁻ to produce hydroxyl radical,⁸ which has been frequently verified with Electron Paramagnetic Resonance (EPR) technique.^8,9

Ti³⁺(e⁻)–⁻O₂H + H⁺ → Ti⁴⁺–˙OH + HO⁻ (4)

Also, the back reaction of the hydroperoxo specie with photo-holes is highly possible.⁶

Ti⁴⁺–⁻O₂H + h⁺ → Ti⁴⁺–O₂H ↔ Ti⁴⁺–O₂˙⁻ + H⁺ (5)

So far, H₂O₂-assisted photocatalysis is inevitable,¹⁰ which limits the H₂O₂ concentration (from tens of μM to several mM) in most reported systems for directly photocatalytic H₂O₂ production. Hereinafter, we introduce an experimental attempt to photocatalytically produce H₂O₂ of high levels by innovatively employing 2-ethyl-9,10-anthraquinone (EAQ) as an electron condenser. The photogenerated electrons are trapped by EAQ during the photocatalytic process by forming 2-ethyl-9,10-anthrahydroquinone (H₂EAQ), which subsequently induces reduction of O₂ molecules into H₂O₂ by simple air bubbling in dark (eqn (6), Fig. S1†).

H₂EAQ + O₂ → EAQ + H₂O₂ (6)

Pt nanoclusters deposited P25 TiO₂ (Pt/P25, with a Pt percent of 0.4 wt%) was employed as the photocatalyst, which was prepared with a typical photocatalytic reduction method¹¹ and had a light gray color. Pt/P25 contains just similar TiO₂ lattice phase composition with that of P25 (rutile/anatase: 20/80, Fig. S2†), of which Pt nanoclusters of ca. 2 nm are well deposited on the TiO₂ grains (20–30 nm, Fig. S3†). Alternatively, some other TiO₂ composites, such as Ag/TiO₂^12a and MoS₂/TiO₂,^12b can also be used as the photocatalyst in this work.

The photocatalytic reaction was carried out by using methanol (100 mL) as the solvent and the hole scavenger, EAQ as the electron condenser, and N₂ bubbling to keep the system anaerobic. The dosage of Pt/P25 was 0.2 g. Pt/P25 has significant photoabsorption at wavelength less than 406 nm (Fig. S4†), while EAQ extends its photoabsorption to 471 nm (Fig. S4†). Hence, the reaction solution was irradiated under either direct irradiation (300 W Xe lamp, PLS-SXE300) or visible irradiation (with an optical filter, λ > 420 nm). Fig. 1A shows the photocatalytic generation of H₂O₂ and the corresponding control reactions. H₂O₂ is kept absent throughout the reaction under visible irradiation with photocatalyst or under direct light irradiation without the Pt/P25 photocatalyst, but generated smoothly under direct irradiation with photocatalyst. As EAQ absorbs both UV and visible light while TiO₂ can only be aroused with UV light, it suggests that the H₂O₂ is generated via a photocatalysis-related process. Briefly, photogenerated electrons in the Pt/P25 photocatalyst reduce the EAQ molecules to give H₂EAQ, while the holes at the valence band of Pt/P25 are captured by hole scavenger, methanol. When H₂EAQ solution is air bubbled, the reaction of oxygen with H₂EAQ leads to the formation of H₂O₂. Notably, the H₂O₂ production rate is ca. 10-fold than that of H₂ with Pt/P25 according to their initial slopes as shown in Fig. 1A.

Fig. 1 (A) Photocatalytic generation of H₂O₂ under (a) direct irradiation with photocatalyst, (b) visible irradiation with photocatalyst, (c) direct irradiation without photocatalyst, and (d) photocatalytic generation of H₂ under direct irradiation. (B) H₂O₂ generation with EAQ dosages of (a) 1.0 g L⁻¹, (b) 2.0 g L⁻¹, (c) 3.0 g L⁻¹, (d) 4.0 g L⁻¹, (e) 6.0 g L⁻¹ with the Pt/P25 photocatalyst.

Fig. 1B shows the generating profiles of H₂O₂ with various EAQ dosages in this photocatalytic system. The H₂O₂ is initially produced efficiently in all cases; however, at the EAQ dosage of 1.0 g L⁻¹, the H₂O₂ production gets slower and reaches its maximum at a reaction time of ca. 1.5 h. The measured and the fitted maximal produced H₂O₂ concentrations are 2.6 mM and 2.5 mM, respectively. Further prolonging the irradiation time leads to a slow decrease in H₂O₂ concentration. The exhaustion of EAQ would be the main constraint that limits the maximal H₂O₂ level in this reaction. Meanwhile, two possible reasons can also be responsible for the adverse H₂O₂ concentration decrease, which we observed in Fig. 1B: one is the back-reaction of H₂EAQ with the photogenerated holes, which slowly becomes serious at higher H₂EAQ concentration. The other is the chemical deterioration of EAQ; as we know, the chemical structure change of EAQ, especially on its anthracene ring, would alter the redox potential of EAQ/H₂EAQ and disable the H₂O₂ production from O₂ via H₂EAQ.

Although a H₂O₂ concentration of 2.5 mM is quite high among the literature reported H₂O₂ levels with the directly photocatalytic synthesis,⁴ optimizing the reaction parameters to get a better production is preferred. Increasing the EAQ dosage to 2.0 g L⁻¹ changes much on the H₂O₂ generation. As shown in Fig. 1B, the H₂O₂ concentration further increases and reaches its maximum (5.3 mM) at an irradiation time of 3.9 h, which is ca. 2.1-fold than that with EAQ dosage of 1.0 g L⁻¹. As shown in Table 1 (also Table S1†), the maximal H₂O₂ concentrations become higher at higher EAQ dosages; a maximal H₂O₂ concentration of 9.1 mM is produced with EAQ dosage of 4.0 g L⁻¹ at a reaction time of 8.2 h, which should be ascribed to that more concentrated EAQ benefits the photoelectron scavenge and thus favors the formation of H₂EAQ. However, as the light absorption region of TiO₂ mostly overlaps with that of EAQ, increasing the EAQ dosage seems adverse to the initial generation rate of H₂O₂. EAQ of high concentrations would significantly compete with Pt/P25 on photon absorption and thus hinder some the photoexcitation of Pt/P25 photocatalyst. Nevertheless, a high initial EAQ dosage would enhance the reachable H₂EAQ level under irradiation; i.e., the generated amounts of H₂O₂ with higher EAQ dosages would slowly catch up with and even surpass those with lower EAQ dosages as shown in Fig. 1B. However, much longer irradiation times are required for reaching the maximal H₂EAQ levels with the higher EAQ dosages.

Table 1 Maximal concentrations and corresponding yields of H₂O₂ with various EAQ dosages

EAQ Concentration		Optimal reaction time (h)^a	Maximal H2O2 concentration (mM)^a	Maximal yield (%)^a
(g L^−1)	(mM)
a Fitting data in Fig. 1B.b Fitting data at 9 h.
1.0	4.23	1.4	2.5	59.0
2.0	8.46	3.9	5.3	62.4
3.0	12.70	6.5	8.3	65.7
4.0	16.93	8.2	9.1	54.0
6.0	25.39	9.0^b	8.7^b	34.4^b

---

The side reactions during the EAQ-assisted photocatalytic H₂O₂ production are undesired and adverse in this work, just as those in the industrial AO process. Therefore, the intermediates and the products of the reaction solution were monitored with HPLC (Fig. S5†) and LC-MS (Fig. S6 and S12†) techniques. Table 2 lists all the substances that were detected out in the solution. Correspondingly, a schematically photo catalytic reaction of EAQ is proposed in Fig. 2. H₂EAQ is obtained from EAQ via two successive single-electron reduction processes. However, an inverse oxidation is also possible to the semi-hydroanthraquinone intermediate, which leads to the formation of oxo-H₂EAQ1 and oxo-H₂EAQ2 intermediates. Due to the strong reducibility of H₂EAQ, oxo-H₂EAQ1 and oxo-H₂EAQ2, they would quickly react with oxygen molecules to produce H₂O₂ and transform themselves back to EAQ, oxo-EAQ1 and oxo-EAQ2, respectively. A prolonged photocatalytic process induces a deep hydrogenation of H₂EAQ, which results the formation of 2-ethyl-9-anthranol, 2-ethyl-10-anthranol, 2-ethyl-10-H-9-anthranone (EAN1) and 2-ethyl-9-H-10-anthranone (EAN2). These four deep hydrogenation products are not strange interlopers. They are unavoidable byproducts in AO process² and considered unavailable to produce H₂O₂. However, the last substance (isomer of EAQ) identified from the reaction solution suggests the formation of another EAQ (which differs only in the position of quinone groups to the original EAQ molecule), which is possibly the oxidation product from EAN1 and EAN2.

Table 2 Chemical substances detected out with LC-MS after photocatalysis (air bubbled)

Time (min)	Formula	Substance	Structure	Meas. m/z	Pred. Mw
11.24	C16H12O3	Oxo-EAQ1		253.120	252.122
11.52	C16H12O3	Oxo-EAQ2		253.120	252.122
14.98	C16H14O	2-Ethyl-9-anthranol		223.110	222.111
15.27	C16H14O	2-Ethyl-10-anthranol		223.109	222.111
17.32	C16H14O	EAN1		223.108	222.111
17.79	C16H14O	EAN2		223.108	222.111
18.83	C16H12O2	EAQ		—	—
20.41	C16H12O2	EAQ isomer	—	237.089	236.091

---

Fig. 2 Proposed photocatalytic reduction reaction pathways for EAQ.

Conclusions

In summary, a novel way is introduced to photocatalytically produce H₂O₂ of high levels by employing EAQ as an electron condenser. A concentration of H₂O₂ with 9.1 mM was obtained with 4.0 g L⁻¹ EAQ for 8.2 h irradiation, which is advantageous comparing with the directly photocatalytic route for H₂O₂ production (from tens of μM to several mM). The enhanced H₂O₂ formation is due to the electron condensing with EAQ and the “off-site” H₂O₂ generation strategy, which circumvents the adverse over-reduction or back-reaction of hydroperoxo species (the precursor of H₂O₂) during the photocatalytic process. It suggests a potential candidate route toward the green and safe H₂O₂ synthesis without producing highly pure H₂ in advance.

Acknowledgements

This work was supported by the National Nature Science Foundations of China (21177039) and the Innovation Program of Shanghai Municipal Education Commission (13ZZ042).

Notes and references

1. K. Sato, M. Aoki and R. Noyori, Science, 1998, 281, 1646.
2. J. M. Campos-Martin, G. Blanco-Brieva and J. L. Fierro, Angew. Chem., Int. Ed., 2006, 45, 6962.
3. (a) J. H. Lunsford, J. Catal., 2003, 216, 455; (b) J. K. Edwards, B. Solsona, E. Ntainjua, A. F. Carley, A. A. Herzing, C. J. Kiely and G. J. Hutchings, Science, 2009, 323, 1037; (c) G. Blanco-Brieva, E. Cano-Serrano, J. M. Campos-Martin and J. L. Fierro, Chem. Commun., 2004, 1184; (d) P. Landon, P. J. Collier, A. J. Papworth, C. J. Kiely and G. J. Hutchings, Chem. Commun., 2002, 2058; (e) S. Melada, R. Rioda, F. Menegazzo, F. Pinna and G. Strukul, J. Catal., 2006, 239, 422; (f) V. R. Choudhary, A. G. Gaikwad and S. D. Sansare, Angew. Chem., Int. Ed., 2001, 40, 1776.
4. (a) H. Goto, Y. Hanada, T. Ohno and M. Matsumura, J. Catal., 2004, 225, 223; (b) Y. Yamada, A. Nomura, T. Miyahigashi and S. Fukuzumi, Chem. Commun., 2012, 48, 8329; (c) M. Fukushima, K. Tatsumi, S. Tanaka and H. Nakamura, Environ. Sci. Technol., 1998, 32, 3948.
5. (a) M. Teranishi, S. Naya and H. Tada, J. Am. Chem. Soc., 2010, 132, 7850; (b) D. Tsukamoto, A. Shiro, Y. Shiraishi, Y. Sugano, S. Ichikawa, S. Tanaka and T. Hirai, ACS Catal., 2012, 2, 599; (c) T. Hirakawa and Y. Nosaka, J. Phys. Chem. C, 2008, 112, 15818; (d) V. Maurino, C. Minero, G. Mariella and E. Pelizzetti, Chem. Commun., 2005, 2627.
6. R. Nakamura and Y. Nakato, J. Am. Chem. Soc., 2004, 126, 1290.
7. T. Hirakawa, T. Daimon, M. Kitazawa, N. Ohguri, C. Koga, N. Negishi, S. Matsuzawa and Y. Nosaka, J. Photochem. Photobiol., A, 2007, 190, 58.
8. R. Nakamura, A. Imanishi, K. Murakoshi and Y. Nakato, J. Am. Chem. Soc., 2003, 125, 7443.
9. (a) X. Li, C. Chen and J. Zhao, Langmuir, 2001, 17, 4118; (b) F. Herrera, J. Kiwi, A. Lopez and V. Nadtochenko, Environ. Sci. Technol., 1999, 33, 3145.
10. R. Cai, Y. Kubota and A. Fujishima, J. Catal., 2003, 219, 214.
11. P. Chowdhury, H. Gomaa and A. K. Ray, Int. J. Hydrogen Energy, 2011, 36, 13442.
12. (a) H. Li, W. Lu, J. Tian, Y. Luo, A. M. Asiri, A. O. Al-Youbi and X. Sun, Chem.–Eur. J., 2012, 18, 8508; (b) Q. Liu, Z. Pu, A. M. Asiri, A. H. Qusti, A. O. Al-Youbi and X. Sun, J. Nanopart. Res., 2013, 15, 2057.

---

Footnote

† Electronic supplementary information (ESI) available: Synthesis and characterizations of the Pt/P25 catalyst, photocatalytic H₂O₂ synthesis, determination of H₂O₂, hydrogen evolution, by-products analysis with HPLC and LC-MS, mechanism of H₂O₂ generation, reduction kinetics of EAQ and the corresponding reaction rate constants. See DOI: 10.1039/c4ra09702e

---