Oxidation mechanism of molecular oxygen over cyclohexene catalyzed by a cobalt L-glutamic acid complex

Abstract

Herein, we introduce a new strategy towards the cyclohexene oxidation reaction, which is driven by molecular oxygen with a cobalt(II)/amine acid complex as the catalyst without any solvents. The as-prepared products include cyclohexene oxide, 2-cyclohexene-1-ol and 2-cyclohexene-1-one. Under optimized conditions, the oxidation of cyclohexene with molecular oxygen yields a conversion rate of 82.5%. Notably, the catalytic oxidation system has the advantages of being clean, safe and operationally simple. Our experiment indicates that the oxidation follows a radical-chain mechanism rather than a catalytic mechanism, which reveals that 2-cyclohexene-1-hydroperoxide could play a key role in the free-radical chain reaction through a complicated but necessary product analysis.

---

1. Introduction

The allylic oxidation of cyclohexene is of great importance in both the bulk chemical industry and syntheses of valued fine chemicals.¹ In particular, the products of the allylic oxidation cyclohexene (2-cyclohexene-1-ol and 2-cyclohexene-1-one) are promising starting materials in the fragrance industry and organic synthesis.² Numerous studies have revealed that the allylic oxidation of cyclohexene is catalyzed by homogenous and heterogeneous catalysts.^3,4 Cao⁵ synthesized carbon and nitrogen-doped nanotubes as metal-free catalysts in the selective allylic oxidation of cyclohexene using molecular oxygen as the oxidant in the liquid phase. Liu⁶ investigated the aerobic oxidation of cyclohexene over TiZrCo catalysts with O₂ as the oxidant, in which 2-cyclohexene-1-one was produced with a high selectivity of 57.6% at a conversion of 92.2%. Baiker⁷ discovered that the Cu-MOF [Cu(bpy)(H₂O)₂(BF₄)₂(bpy)] (bpy: 4,40-bipyridine) is an active and highly selective catalyst in the allylic oxidation of cyclohexene with molecular oxygen under mild, solvent-free conditions. The main product, cyclohexene hydroperoxide, is produced in 85–90% selectivity and epoxidation is a minor side reaction. Salavati-Niasari⁸ synthesized square-planar bis(macrocyclic)dinickel(II) {[Ni([16]aneN₅)]₂R}(ClO₄)₄ complexes, which have been found to be effective catalysts for the selective oxidation of cyclohexene to 2-cyclohexene-1-one and 2-cyclohexene-2-ol in the absence of solvents at 70 °C with molecular oxygen as the oxidant.

In spite of the numerous studies, which have revealed that homogenous and heterogeneous Fe, Mn, Cu and Cr catalysts play a key role in the allylic oxidation reaction of cyclohexene,^9,10 chemists continue to find it difficult to reveal the detail pathways for some oxidation reactions. In addition, the separation of objective products from reaction mixtures is still tricky, and conversion rates are usually not satisfactory for classical transition metal catalysis. The development of a rational approach for the design of simple but effective catalysis is highly desirable.

Oxidation driven by molecular oxygen is particularly attractive for applications in economic and environmental fields.¹¹ However, the high activation energy of molecular oxygen limits the direct involvement of O₂ in the oxidation of organics. Currently, researchers have successfully synthesized many catalysts to activate molecular oxygen in the oxidation of organics.^12,13 A number of transition metal complexes, which are environmentally friendly, have been identified as catalysts for various reactions due to their great potential in the activation of molecular oxygen under mild conditions.^14–19 A series of amino acid metal complexes were synthesized and characterized through various methods.^20,21 Inspired by the pioneering research that amino acid metal complexes could serve as the activated center of proteins,²² innovative efforts have been undertaken to obtain insights into the relationship between the binding pathways of amino acid/metal ion complexes and their catalytic properties, thus further exploring the mechanism of metabolism. For instance, as mentioned by Burk,²³ cobalt(II) complexes of L-histidine have the ability to absorb molecular oxygen in aqueous solution. As a result, the oxygenation of amino acid cobalt(II) complexes has been extensively studied over the past decades. Our previous report^24,25 confirmed that with different rates of aging or auto-oxidation, many amino acid cobalt(II) complexes exhibit novel reversible oxygenation performances in aqueous solution.^26,27 Despite the fact that there are a few reports on their reversible oxygenation properties, particularly in aqueous solution,²⁸ there has been less attention on the use of oxygenated amino acid cobalt(II) complexes as catalysts for the activation of molecular oxygen.

The key difficulty stems from the unclear oxidation mechanism of molecular oxygen species. Moreover, little attention has been paid to the comprehensive and intensive study of the reaction mechanism concerning amino acid complexes as catalytic activators of molecular oxygen. Therefore, a study on the binding properties of amino acid–cobalt(II) complexes is of great significance to reveal the mechanism of oxygenation and activation of molecular oxygen.

Herein, we introduce the facile method of using a cobalt(II) complex of L-glutamic to effectively drive the oxidation of cyclohexene, which has received special attention due to the high application value of its oxidation products.^29,30 The oxidation pathway of cyclohexene driven by L-Glu–Co(II) is deduced through the analysis of the main products. It is notable that our method is very simple and lacks the involvement of any organic solvents throughout the entire reaction process, which strongly adheres to the fundamental requirements of green chemistry. Insights into the reversible oxygenation–deoxygenation mechanism of our products should prove useful in the design of more effective Co-complexes to drive the oxidation of organics with molecular oxygen.

2. Experimental

2.1 Materials

Cyclohexene was obtained from Sinopharm Chemical Reagent Co., Ltd. Analytical-reagent-grade L-glutamic acid and cobalt acetate were purchased from Aladdin, whereas cyclohexene oxide, 2-cyclohexene-1-ol, and 2-cyclohexene-1-one were purchased from Sigma. Moreover, ferulic acid was purchased from Xiya Reagent. Triphenylphosphine was obtained from Energy Chemical. Acetonitrile was of gas chromatography grade. All other chemicals and solvents were of AR grade.

2.2 Catalyst preparation and characterization

Cobalt acetate was ground before application. L-Glutamic acid (0.01 mmol) and the ground cobalt acetate (0.01 mmol) were mixed and ground together for 10 min in a dinitrogen atmosphere at room temperature. Furthermore, UV-vis spectra of the synthesized catalysts in 10.0 mL cyclohexene solution were recorded with a UV-2550 spectrophotometer, whereas UV-vis spectra of the solid catalysts were recorded with a UV-4802s spectrophotometer over the range from 200 to 800 nm.

2.3 Catalytic oxidation of cyclohexene with molecular oxygen

The cobalt complex of L-glutamic acid was used as a catalyst for the oxidation of cyclohexene without solvents. The catalyst (0.02 mmol) was mixed with cyclohexene (1.0 mL), and then added into a 35.0 mL test tube reactor, which was equipped with a magnetic stirring bar and connected to a balloon filled with sufficient molecular oxygen. Subsequently, the mixture was vigorously stirred at 70 °C for 24 h, and at the same time the oxidation products were identified by gas chromatography. Furthermore, the structural features of the products were also determined by a gas chromatograph connected to a mass spectrometer (GC-MS).

2.4 Physical measurements

The GC results were recorded with a GC-14B gas chromatograph fitted with an FID detector via an RTX-5 fused silica capillary column (30 m length, 0.32 mm internal diameter and 0.25 μm film thickness). The parameters involved in the analysis are as follows: flow rate of the carrier gas N₂ was 1 mL min⁻¹, injector temperature = 200 °C and detector temperature = 220 °C. Moreover, mass spectra were recorded with a GC-MS QP2010 (a RTX-5 column), and UV-vis spectra of the catalysts were recorded on a UV-2550 spectrophotometer. Finally, the UV-vis spectra of the catalysts in the solid state were recorded on a UV-4802s spectrophotometer.

3. Results and discussion

3.1 Characterization of catalysts

The visible light absorption spectra of L-glutamic (curve 1 in Fig. 1) and cobalt acetate (curve 2 in Fig. 1) solutions alone show little absorption. Moreover, as shown in curve 3 in Fig. 1, the UV-vis spectrum of the catalyst–cyclohexene mixture exhibits a small, but distinct absorption in the N₂ atmosphere. In the case of the involvement of dioxygen, the absorption intensity of L-Glu–Co(II) experiences an evident increase, as shown by curve 4 in Fig. 1, which thereby indicates the formation of L-Glu–Co(II) and the consequent adsorption of molecular oxygen.

Fig. 1 Formation of the L-Glu–Co(II) complex as characterized by UV-vis spectra (curve 1: L-Glu; 2: Co(II); 3: L-Glu–Co(II) complex in N₂; and 4: L-Glu–Co(II) complex in O₂).

The UV-vis absorption spectra of solid-state L-Glu (curve 1) and Co(II) (curve 2) alone are shown in Fig. 2. There are small absorptions in the range of 250–350 nm for L-Glu and 400–600 nm for Co(II). In the dinitrogen atmosphere, the solid-state L-Glu–Co(II) complex showed a moderate intensity spectrum with the main absorption peak at λ = 520 nm (curve 3). With regard to dioxygen instead of dinitrogen, the absorption intensity of the L-Glu–Co(II) complex experienced a sharp increase, which further presents a strong absorption with absorption peaks at λ = 260 nm and 540 nm, as shown in curve 4 in Fig. 2. This indicates that the synthesized catalyst could uptake O₂. Thus, the L-Glu–Co(III)–O₂ species could undergo oxidation due to the transfer of the unpaired electron of cobalt(II) to molecular oxygen.^31,32

Fig. 2 Formation of the solid-state L-Glu–Co(II) complex as characterized by UV-vis spectra (curve: L-Glu; 2: Co(II); 3: L-Glu–Co(II) in N₂; 4: L-Glu–Co(II) in O₂).

3.2 Catalytic oxidation products

Based on both spiking and comparison with standards, three catalytic oxidation products of cyclohexene were identified with the employment of gas chromatography, as shown in Table 1. It can be seen that there were no products when L-Glu and Co(II) were employed alone as catalysts in the reaction, or without any catalysts involved. Therefore, these results indicate that L-Glu–Co(II) complexes have favorable catalytic properties toward cyclohexene, which confirm that the synthesized catalyst can activate adsorbed molecular oxygen, thereby enabling it to cause a catalytic effect. However, the peaks of the oxidation products detected were cyclohexene oxide, 2-cyclohexene-1-ol and 2-cyclohexene-1-one. The structural features of the products were determined by mass spectrometry, which showed an M-1 peak at 98 m/z for cyclohexene oxide and cyclohexene-ol and an M⁺ peak at 96 m/z for cyclohexene-1-one.

Table 1 Results of the oxidation of cyclohexene with different catalysts

Catalyst	Conversion (%)	Total selectivity (%)
Bank	0.8	1.1
L-Glutamic acid	0.9	1.2
Cobalt acetate	1.2	1.9
Prepared catalyst	82.5	93.2

---

To confirm that activated O₂ was involved in the oxidation reaction, ¹⁸O₂ instead of ¹⁶O₂ was employed to perform the catalytic oxidation of cyclohexene under the same conditions. The GC-MS results indicate that the main products were the same as the three former types of product in the ¹⁸O₂ source. Furthermore, the oxygen isotope in the products was ¹⁸O, thereby indicating that the oxygen atom included in the products was derived from activated molecular oxygen, as shown in Fig. 3. In summary, in terms of the basic reaction mechanism of catalysts for the oxidation of cyclohexene, three steps are involved, namely the uptake of molecular oxygen by the synthesized catalyst, the activation of adsorbed molecular oxygen, and the involvement of activated molecular oxygen in the oxidation reaction. However, there has to be a further intensive study related to the reaction mechanism.

Fig. 3 Mass spectra of the main products oxidized by ¹⁶O₂ and ¹⁸O₂. Cyclohexene oxide (a) and (b), 2-cyclohexene-1-ol (c) and (d), and 2-cyclohexene-1-one (e) and (f).

The effect of four reaction parameters namely solvent selection, reaction temperature, cyclohexene amount and reaction time were studied to determine the optimized working conditions. Detailed computational procedures and results can be found in the ESI.†

3.3 Possible reaction pathway

The synthesized catalysts have marked catalytic activity mainly due to oxygenated complexes, which play the role of O₂ carriers and activators. In regard to the activated molecular oxygen, it can be involved in the catalytic oxidation of cyclohexene. It is suggested that the key step of this reaction is the transfer of activated molecular oxygen to cyclohexene, thereby resulting in the formation of 2-cyclohexene-1-hydroperoxide (ROOH).

Hence, revealing the existence of R-OOH is of fundamental importance to support our proposal. During experimentation, a peak with a retention time of 4.8 min in the total ion chromatograph can be observed, as shown in Fig. 6a, and the area of this peak decreased with an increase in temperature and reaction time. Herein, we propose that this peak is attributed to 2-cyclohexene-1-hydroperoxide due to the instability of this compound. Moreover, the –O–O– bond can decompose upon heating, which is in agreement with the mass spectrum, as shown in Fig. 4. The study of the decomposition process via mass spectrometry can confirm the presence of ROOH, as shown Fig. 5. Furthermore, there is another route to confirm the presence of ROOH, which is to add triphenylphosphine (PPh₃) to the mixture solution since PPh₃ can be easily oxidized to PPh₃ oxide with an oxygen atom, whereas at the same time, ROOH can be directly converted to 2-cyclohexene-1-ol.^33,34 If this peak is produced by the formation of ROOH, the objective peak should be eliminated after the addition of PPh₃. At the same time, the peak area of 2-cyclohexene-1-ol (ROH) should increase when PPh₃ is added. The expected result is displayed in Fig. 6.

Fig. 4 Mass spectrum of 2-cyclohexene-1-hydroperoxide.

Fig. 5 Decomposition process of 2-cyclohexene-1-hydroperoxide in terms of mass spectrometry results.

Fig. 6 Gas chromatograms of samples before (a) and after (b) the addition of triphenylphosphine.

There are two proposed mechanisms for the catalytic oxidation of cyclohexene with the employment of molecular oxygen, the catalytic mechanism and the radical-chain mechanism. To gain insight into the mechanism of this reaction of amino amide complex, the free-radical scavenger ferulic acid was added during the oxidation reaction. This can further determine whether there are reaction products. Furthermore, it is evident that there are hardly any products that could completely be restrained after the addition of sufficient ferulic acid to scavenge all of the free radicals, thus clearly indicating that the reaction follows a radical-chain mechanism instead of a catalytic mechanism, as demonstrated in Table 2.

Table 2 Effects of the addition of ferulic acid

Catalyst	Conversion (%)	Total selectivity (%)
Bank	0.8	1.1
Prepared catalyst + ferulic acid	0.7	0.9
Prepared catalyst	82.5	93.2

---

If this reaction follows the radical-chain mechanism, ROOH would be produced in the presence of L-Glu–Co(III)OOCo(III)–Glu-L and cyclohexene solution through an addition reaction. ROOH is unstable in the free-radical reaction, thus the –OOH group may experience O–O bond homolytic cleavage, which results in one cyclohexenyl oxy radical (RO˙) and one HO˙ radical. Simultaneously, Co(II) would be oxidized to Co(III), and 2-cyclohexene-1-one (ROH) and 2-cyclohexene-1-one (R=O) are generated upon the attack on another RO˙ radical. Moreover, cyclohexene oxide (CyO) and ROH are generated upon the attack on the cyclohexene solution. Furthermore, a schematic of this product transformation process during the oxidation of cyclohexene in the presence of L-Glu–Co and O₂ oxidant is presented in Fig. 7.

Fig. 7 Schematic of the reaction pathways of cyclohexene oxidation with O₂ in the presence of L-Glu–Co(II) catalyst.

It is also notable that the selectivity of CyO was merely 0.6% at the end of the reaction, whereas the selectivity of ROH and R=O was 36.9% and 55.7%, respectively. It can be seen that the CyO species was formed in a very minute quantity in this reaction. Hence, this proves that the reaction follows the radical-chain mechanism. This result is consistent with the result drawn in the case of the addition of the free-radical scavenger ferulic acid. In terms of the oxidation experiments performed during the initiation period, a buildup of ROOH could be observed. This is then followed by a gradual decrease in ROOH selectivity, which is caused by the conversion to mainly ROH and R=O. However, the ROH species was oxidized to R=O in the presence of oxygen atoms as well as an increase in reaction time. The reaction conditions favored the formation of R=O with an increase in reaction time and a decrease in ROH selectivity for the reaction. These results are consistent with the results of GC-MS analysis, which further explains why small amounts of ROOH and CyO could be observed at the end of the oxidation reaction.

4. Conclusions

In summary, we have developed a facile protocol to drive the oxygenation of cyclohexene with molecular oxygen in the presence of the simple and green L-Glu–Co(II) complex. It is notable that the as-prepared L-Glu–Co(II) complex exhibits novel properties to activate molecular oxygen to carry out the catalytic oxidation of cyclohexene. Under optimized conditions, the oxidation of cyclohexene with molecular oxygen yields a conversion rate of 82.5%. The possible oxidation mechanism was deduced through additional oxidation experiments involving a radical scavenger, which indicates that the oxidation follows a radical-chain mechanism instead of a catalytic mechanism and 2-cyclohexene-1-hydroperoxide plays a key role in the free-radical chain reaction. Furthermore, hydroperoxide has been proposed as a key intermediate in the reaction and the ketone is produced as the major product in all cases.

Acknowledgements

The authors are grateful for the financial support from the National Natural Science Foundation (NSFC-21261022, 21162027).

References

1. U. Neuenschwander, F. Guignard and I. Hermans, ChemSusChem, 2010, 3, 75–84.
2. G. Smeureanu, A. Aggarwal, C. E. Soll and C. M. Drain, Chem.–Eur. J., 2009, 15, 12133–12140.
3. S. Khare and S. Shrivastava, J. Mol. Catal. A: Chem., 2004, 217, 51–58.
4. S. T. Castaman, S. Nakagaki, R. R. Ribeiro, K. J. Ciuffi and S. M. Drechsel, J. Mol. Catal. A: Chem., 2009, 300, 89–97.
5. Y. H. Cao, H. Yu, F. Peng and H. J. Wang, J. Am. Chem. Soc., 2014, 5, 1617–1625.
6. T. Liu, H. Y. Cheng, W. W. Lin, C. Zhang, Y. C. Yu and F. Y. Zhao, Catalysts, 2016, 6, 3390–3398.
7. D. Jiang, T. Mallat, D. M. Meier, A. Urakawa and A. Baiker, J. Catal., 2010, 270, 26–33.
8. M. Salavati-Niasari, J. Mol. Catal. A: Chem., 2007, 272, 207–212.
9. S. Khare and S. Shrivastava, J. Mol. Catal. A: Chem., 2004, 217, 51–58.
10. S. T. Castaman, S. Nakagaki, R. R. Ribeiro, K. J. Ciuffi and S. M. Drechsel, J. Mol. Catal. A: Chem., 2009, 300, 89–97.
11. Z. Z. Shi, C. Zhang, C. H. Tang and N. Jiao, Chem. Soc. Rev., 2012, 41, 3381–3430.
12. T. Matsushita, K. Ebitani and K. Kaneda, Chem. Commun., 1999, 265–266.
13. Y. Ishii, S. Sakaguchi and T. Iwahama, Adv. Synth. Catal., 2001, 343, 393–427.
14. J. Haber, L. Matachowski, K. Pamin and J. Poltowicz, J. Mol. Catal. A: Chem., 2000, 162, 105–109.
15. Z. Li, Z. Q. Jing and C. G. Xia, Chin. J. Org. Chem., 2007, 27, 34–44.
16. E. M. Serwicka, J. Poltowicz, K. Bahranowski, Z. Olejniczak and W. Jones, Appl. Catal., A, 2004, 275(1–2), 9–14.
17. T. Abe and M. Kaneko, Prog. Polym. Sci., 2003, 28, 1441–1488.
18. N. Sehlotho and T. Nyokong, J. Mol. Catal. A: Chem., 2004, 209, 51–57.
19. M. Biesaga, K. Pyrzyńska and M. Trojanowicz, Talanta, 2000, 51, 209–224.
20. R. M. Wang, C. J. Hao, Y. P. Wang and S. B. Li, J. Mol. Catal. A: Chem., 1999, 147, 173–178.
21. S. M. Seyedi, R. Sandaroos and G. H. Zohuri, Chin. Chem. Lett., 2010, 21, 1303–1306.
22. J. Du, H. J. Zhang and B. Fang, Prog. Chem., 2003, 15, 288–294.
23. D. Burk, J. Hearon, L. Caroline and A. L. Schade, J. Biol. Chem., 1946, 165, 723–724.
24. X. C. Zhang, F. Yue, Y. Huang, P. Fu, X. Cheng and J. D. Wang, Chem. J. Chin. Univ., 2012, 33, 1370–1375.
25. H. M. Wen, X. Zhang, H. Li, F. Yue and J. D. Wang, Chem. J. Chin. Univ., 2013, 34, 2262–2269.
26. J. H. Fu, J. F. Li, P. Fu, F. Yue, X. L. Zhang and J. D. Wang, Chem. J. Chin. Univ., 2011, 32, 1483–1487.
27. P. Fu, J. H. Fu, J. F. Li, F. Yue, X. L. Zhang and J. D. Wang, Chin. J. Inorg. Chem., 2012, 28, 360–1364.
28. X. C. Zhang, Bachelor Degree thesis, Xinjiang University, 2012.
29. H. Kameyama, F. Narumi and T. Hattori, J. Mol. Catal. A: Chem., 2006, 258, 172–177.
30. B. P. C. Hereijgers, R. F. Parton and B. M. Weckhuysen, ACS. Catal., 2011, 1, 1183–1192.
31. M. J. Carter, D. P. Rillema and F. Basolo, J. Am. Chem. Soc., 1973, 96, 392–400.
32. M. R. Tiné, Coord. Chem. Rev., 2012, 256, 316–327.
33. Z. F. Duan, R. M. Wang, Y. F. He, Y. T. Xie, Y. P. Wang and C. G. Xia, J. Mol. Catal., 2005, 19, 383–387.
34. Q. Tang, P. Z. Sheng and J. Feng, Chem. Ind. Eng. Prog., 2004, 23, 219–228.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra23598k

---