highly efficient cobalt doped carbon nitride polymers for solvent free selective oxidation of cyclohexane

Accepted Manuscript Highly Efficient Cobalt-doped Carbon Nitride Polymers for Solvent-Free Selective Oxidation of Cyclohexane Yu Fu, Wangcheng Zhan, Yanglong Guo, Yun Guo, Yunsong Wang, Guanzhong Lu PII: S2468-0257(17)30005-5 DOI: 10.1016/j.gee.2017.01.006 Reference: GEE 51 To appear in: Green Energy and Environment Received Date: January 2017 Revised Date: 23 January 2017 Accepted Date: 26 January 2017 Please cite this article as: Y Fu, W Zhan, Y Guo, Y Guo, Y Wang, G Lu, Highly Efficient Cobaltdoped Carbon Nitride Polymers for Solvent-Free Selective Oxidation of Cyclohexane, Green Energy & Environment (2017), doi: 10.1016/j.gee.2017.01.006 This is a PDF file of an unedited manuscript that has been accepted for publication As a service to our customers we are providing this early version of the manuscript The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain ACCEPTED MANUSCRIPT Highly Efficient Cobalt-doped Carbon Nitride Polymers for Solvent-Free Selective Oxidation of Cyclohexane Yu Fu, Wangcheng Zhan*, Yanglong Guo, Yun Guo, Yunsong Wang, Guanzhong Lu* Key Laboratory for Advanced Materials and Research Institute of Industrial Catalysis, and Technology, Shanghai 200237, P R China RI PT School of Chemistry and Molecular Engineering, East China University of Science *Corresponding author, Email: zhanwc@ecust.edu.cn; gzhlu@ecust.edu.cn Abstract: Selective oxidation of saturated hydrocarbons with molecular oxygen has SC been of great interest in catalysis, and the development of highly efficient catalysts for this process is a crucial challenge A new kind of heterogeneous catalyst, M AN U cobalt-doped carbon nitride polymer (g-C3N4), was harnessed for the selective oxidation of cyclohexane X-ray diffraction, Fourier transform infrared spectra and high resolution transmission electron microscope revealed that Co species were highly dispersed in g-C3N4 matrix and the characteristic structure of polymeric g-C3N4 can be retained after Co-doping, although Co-doping caused the incomplete polymerization to some extent Ultraviolet–visible, Raman and X-ray photoelectron TE D spectroscopy further proved the successful Co doping in g-C3N4 matrix as the form of Co(II)–N bonds For the selective oxidation of cyclohexane, Co-doping can markedly promote the catalytic performance of g-C3N4 catalyst due to the synergistic effect of Co species and g-C3N4 hybrid Furthermore, the content of Co largely affected the EP activity of Co-doped g-C3N4 catalysts, among which the catalyst with 9.0 wt% Co content exhibited the highest yield (9.0%) of cyclohexanone and cyclohexanol, as AC C well as a high stability Meanwhile, the reaction mechanism over Co-doped g-C3N4 catalysts was elaborated Keywords: Selective oxidation of cyclohexane; Oxygen oxidant; Carbon nitride; Co-doping Introduction C–H activation is always a spotlight for the development of chemical industry owing to the ubiquity of C–H bonds in organic molecules [1-3] However, saturated hydrocarbons consist of only strong and localized single bonds, C–C and C–H bonds, so that they generally have no empty orbitals of low energy or filled orbitals of high ACCEPTED MANUSCRIPT energy that can be readily stimulated to react with other molecules [4-6] What is worse is that the reaction products are generally far more reactive than saturated hydrocarbons and thus are prone to further transform to by-products Among various selective oxidation reactions of hydrocarbons, selective oxidation of cyclohexane is a very attractive reaction because its oxidation products, cyclohexanone (K) and RI PT cyclohexanol (A), are the key intermediates for producing nylon-6 and nylon-66 [7] To date, a large number of heterogeneous catalysts have been reported for the selective oxidation of cyclohexane, including transition metal oxides [8], carbon nanotubes [9,10], molecular sieves [11-13], etc Even though significant progress has SC been achieved, it is still a significant challenge to control the selectivity for the target products while obtaining a high conversion of cyclohexane because of the trade-off between selectivity and high conversion in selective oxidation reactions On the other M AN U hand, in order to reconcile the demand of economy and environment, there is an unremitting drive to pursue a solvent free system using molecular oxygen as the oxidant to exploit KA oil [14, 15] Graphitic carbon nitride (g-C3N4) has attracted increasing attention owing to its excellent physicochemical stability, as well as unique electronic structure with a large TE D band gap of 2.7 eV So far, most of the research efforts have been focused on its application in photocatalysis, such as water splitting for H2 and O2 generation [16-18], reduction of CO2 to hydrocarbon fuels [19, 20] and degradation of pollutants [21] However, there are a relatively limited number of reports on g-C3N4 as one kind of EP metal-free heterogeneous catalysts [22-24] Wang et al reported boron and fluorine co-doped mesoporous carbon nitride as a metal-free catalyst for selective oxidation of AC C cyclohexane with H2O2 as oxidant, the catalyst showed moderate activity but remarkably high selectivity to cyclohexanone [25] In addition, Yang and co-workers designed Fe–Co doped g-C3N4 by taking advantage of the electron-rich structure and found that it is highly effective for the selective oxidation of cyclohexene to 2-cyclohexene-1-one with O2 [26] Despite these efforts, there has never been an effort to reveal the catalytic activity of metal-doped carbon nitride polymers for solvent-free selective oxidation of cyclohexane Interestingly, metalloporphyrins possessing the similar structure with metal-doped carbon nitride have been found to be effective for the selective oxidation of cyclohexane with air as homogeneous catalysts [27] Since homogeneous catalysts definitely suffer from the problems of separation and recycling, metal-doped carbon nitride polymers provide us the ACCEPTED MANUSCRIPT opportunity to develop scalable and effective heterogeneous catalysts for selective oxidation of cyclohexane with air or O2 In the present work, a series of Co-doped graphitic carbon nitride (g-C3N4) polymers with different Co content were synthesized and their catalytic performances for selective oxidation of cyclohexane in a solvent free system were investigated RI PT After Co-doping, g-C3N4 composites showed superior performance in the selective oxidation of cyclohexane with excellent conversion of cyclohexane and selectivity to KA oil simultaneously Meanwhile, controlled experiments were made to elucidate SC the mechanism of this catalytic system Experimental section 2.1 Synthesis of the catalysts M AN U In a typical synthesis of Co-doped g-C3N4, g of dicyandiamide was dissolved in mL of CoCl2·9H2O aqueous solution with a certain concentration, followed by continuous stirring at 80 °C until the water was evaporated Consequently, the as-prepared light blue solid was heated at 550 °C in N2 flow for h at a heating rate of °C/min The sample obtained was ground into powder and washed thoroughly with TE D hot water to remove the residual Co precursor on the surface Finally, the solid was dried at 120 °C for 12 h, and the obtained samples with different Co contents were denoted as Co–g-C3N4(x), in which x stands for the mass ratio of CoCl2·9H2O and dicyandiamide in the synthesis mixture The actual Co content in samples was EP detected by energy disperse spectroscopy and shown in Table Pure g-C3N4 was also prepared with the same procedure as Co–g-C3N4, but no AC C CoCl2·9H2O was added Similarly, the corresponding precursor (MnCl2·4H2O, CuCl2·2H2O and CeCl2·7H2O) was added instead of CoCl2·9H2O to synthesis Mn–g-C3N4, Cu–g-C3N4 and Ce–g-C3N4, respectively 2.2 Characterization of catalysts X-ray diffraction (XRD) data were collected on a Bruker D8 Focus diffractometer using Cu Kα radiation (40 kV, 40 mA) at room temperature The composition of Co-doped g-C3N4 catalysts was measured on a NOVA Nano SEM450 scanning electron microscope equipped with TEAMEDS energy disperse spectroscopy (EDS) The Fourier transform infrared (FT-IR) spectra of samples were carried out on a Nicolet Nexus 670 FT-IR spectrometer, and the samples were ground with anhydrous KBr and pressed into thin wafers The Ultraviolet–visible (UV-Vis) spectra of ACCEPTED MANUSCRIPT samples were performed on a Varian Cary 500 spectrometer by using the diffuse reflectance technique in the range of 200-800 nm, and BaSO4 was used as the reference, while the Raman spectra of samples was measured on a Renishaw spectrometer Thermal gravimetric (TG) analysis was performed at a heating rate of 10 °C/min from 40 °C to 800 °C in air flow of 30 mL/min on a PerkinElmer Pyris RI PT TGA thermal analyzer The structure information and elemental mapping images of the samples were measured on a high resolution transmission electron microscope (HR-TEM) (JEM-2100, JEOL) at an accelerating voltage of 200 kV X-ray photoelectron spectroscopy (XPS) was analyzed on a Kratos Axis Ultra-DLD system SC with Al Kα radiation and the XPS results were calibrated using the C 1s line at 284.8 eV 2.3 Testing of catalytic activity M AN U The selective oxidation of cyclohexane was carried out in a 50 mL autoclave lined with polytetrafluoroethylene (PTFE) and equipped with an explosion-proof pressure sensor g of cyclohexane, 20 mg of catalyst and 10 µL TBHP as initiator were added into the reactor After purging with O2 for three times, the pressure of O2 was adjusted to 1.0 MPa, and then the reactor was heated to a certain temperature under stirring at TE D 300 rpm After the reaction, the reactor was quenched with ice water to avoid the losses due to evaporation of volatile organic compounds The reaction mixture was diluted with ethanol to thoroughly dissolve the side products After the catalyst was separated by centrifugation, an excessive amount of triphenyphosphine (Ph3P) was EP added to completely convert the intermediate cyclohexylhydroperoxide (CHHP) to cyclohexanol The reaction products were analyzed by Agilent gas chromatograph AC C 7890B equipped with an HP-5 capillary column and a flame ionization detector Methylbenzene was used as an internal standard substance In addition, the side-products were identified with Agilent 7890A-5975C gas chromatograph-mass spectrometry (GC-MS) The recycling experiments of the catalyst were carried out under the same conditions mentioned above and the catalyst was repeatedly optimized for five times in the selective oxidation reaction After each run, the catalyst was recovered from the reaction solution by centrifugation, washed with ethyl alcohol for three times, and then dried at 100 °C to constant weight in the air Results and discussion ACCEPTED MANUSCRIPT 3.1 XRD The powder XRD patterns of pure g-C3N4 and Co–g-C3N4 catalysts are shown in Fig Pure g-C3N4 exhibits a strong diffraction peak corresponding to the (002) facet at 2θ = 27.4°, which is associated with the typical interlayered stacking of the aromatic systems Another diffraction peak can also be discerned at 2θ = 13.1°, RI PT assigned to the in-plane structural packing motif of tri-s-triazine units Compared with pure g-C3N4, the intensity of both diffraction peaks for the Co–g-C3N4 catalysts sharply decreases with the increase in the content of cobalt, in agreement with other metal doped g-C3N4 hybrids [28] The reason is that cobalt species can inhibit SC polymeric condensation of dicyandiamide during the synthesis process of the catalysts Meanwhile, no diffraction peaks of any crystalline phase of Co species appear in the patterns of all the Co–g-C3N4 catalysts, indicating that Co are chemically coordinated Intensity (a.u.) M AN U to the g-C3N4 matrix in the form of Co–N bonds g-C3N4 Co−g-C3N4(5) TE D Co−g-C3N4(10) 10 20 30 40 Co−g-C3N4(15) Co−g-C3N4(20) 50 60 70 80 2Theta (Degree) AC C 3.2 TEM EP Fig XRD patterns of pure g-C3N4 and Co–g-C3N4(x) catalysts with different Co contents To further probe the structure of the Co–g-C3N4 catalysts, HRTEM and elemental mappings were carried out and shown in Fig The Co–g-C3N4(15) catalyst exhibits a sheet-like structure (Fig 2a) and no agglomerated Co species can be observed Furthermore, elemental mapping results confirm that Co and N elements are homogeneously distributed in the g-C3N4 matrix, which is clearly in consistent with XRD results SC RI PT ACCEPTED MANUSCRIPT M AN U Fig HRTEM (a, b) and elemental mapping (c, d, e) images of Co−g-C3N4(15) catalysts 3.3 FT-IR The FT-IR spectra of g-C3N4 and Co–g-C3N4 catalysts are shown in Fig The broad absorption peak centered at 3130 cm−1 is assigned to the stretching vibration of primary and secondary amines Compared with the spectrum of pure g-C3N4, the TE D intensity of this absorption peak slightly decreases for the Co–g-C3N4 catalysts, due to the acceleration of deamination by cobalt salts during the self-polymerization process of dicyandiamide precursors On the contrary, the intensity of absorption peak at ~2170 cm−1, ascribed to nitrile species, increases with the increase in the Co content EP of the Co–g-C3N4 catalysts, due to the incomplete polymerization caused by Co-doping, which was in accord with XRD results [29, 30] In addition, a group of AC C strong absorption peaks between 1100 and 1700 cm−1 and a sharp peak at 805 cm−1 can be observed for pure g-C3N4 and Co-doped catalysts, which are ascribed to the characteristic stretching modes of aromatic CN heterocycles in the polymeric melon network and the triazine units, respectively [31, 32] On the whole, there is no significant difference between the spectra of pure g-C3N4 and Co–g-C3N4 catalysts, revealing that the typical structure of the g-C3N4 can be retained after Co-doping ACCEPTED MANUSCRIPT Transmittance (a.u.) Co−g-C3N4(20) Co−g-C3N4(15) Co−g-C3N4(10) Co−g-C3N4(5) 4000 3500 3000 2500 2000 1500 1000 -1 Wavenumber (cm ) RI PT g-C3N4 500 SC Fig FT-IR spectra of pure g-C3N4 and Co–g-C3N4(x) catalysts with different Co contents 3.4 Thermal analysis TG curves of all the catalysts are present in Fig All the catalysts exhibit a M AN U remarkable weight loss after 450 °C, attributed to the oxidation of g-C3N4 matrix However, the temperature corresponding to the weight loss for the Co–g-C3N4 catalysts shifts significantly to the lower temperature compared with that for pure g-C3N4 These results reveal that Co-doping can affect the thermal stability of g-C3N4 matrix due to the incomplete polymerization of the matrix as shown in XRD and TE D FT-IR results 100 g-C3N4 60 Co-g-C3N4(5) Co-g-C3N4(10) 40 EP Conversion (%) 80 Co-g-C3N4(15) Co-g-C3N4(20) AC C 20 100 200 300 400 500 600 700 800 o Temperature ( C) Fig TG curves of pure g-C3N4 and Co–g-C3N4(x) catalysts with different Co contents 3.5 UV-vis and Raman The UV-vis and Raman spectra were carried out to confirm the state of Co and investigate the effect of Co-doping on the electronic structure of g-C3N4 As shown in Fig 5a, pure g-C3N4 exhibits the typical semiconductor absorption band at 384 nm However, all the Co–g-C3N4 catalysts show a significant red shift and the higher ACCEPTED MANUSCRIPT intensity in the visible light region (410-500 nm) in comparison with pure g-C3N4, confirming a host-guest interaction between Co and g-C3N4 matrix Meanwhile, a broad absorption band between 500 nm and 700 nm, attributed to the 4A2(F)→4T1(P) energy transition of tetrahedral Co(II) species, is existed in UV-vis spectra of all Co–g-C3N4(x) catalysts [33-36] RI PT Raman patterns of the catalysts are presented in Fig 5b Neither peaks due to cobalt nor characteristic A1g, F2g, F2g and Eg phonon modes of cobalt oxides can be observed, further revealing that Co species are coordinated with the g-C3N4 matrix [37] (a) (b) g-C3N4 Co−g-C3N4(5) Co−g-C3N4(20) Co−g-C3N4(10) 300 400 500 600 Wavelength (nm) 700 SC M AN U Co−g-C3N4(20) Intensity (a.u.) Absorbance (a.u.) Co−g-C3N4(15) 800 500 1000 Co−g-C3N4(15) Co−g-C3N4(10) Co−g-C3N4(5) g-C3N4 1500 2000 -1 Raman Shift (cm ) Fig UV-vis spectra (a) and Raman patterns (b) of pure g-C3N4 and Co–g-C3N4(x) catalysts 3.6 XPS TE D with different Co contents Fig shows the survey scan, N 1s and Co 2p XPS spectra of Co–g-C3N4(x) EP catalysts There are three distinguishable peaks of N1s located at 398.8, 400.1 and 401.2 eV (Fig 6b), assigned to pyridinic N, pyrrolic N and graphitic N respectively, indicating the typical tri-s-triazine (melem) repeating building blocks of graphitic AC C carbon nitride materials [38-40] As shown in Fig 6c, the Co 2p3/2 peak can be deconvoluted into three peaks at 781.7, 785.2 and 788.0 eV The main peak at 781.7 eV can be assigned to Co2+, while the latter two peaks are the satellite peaks of Co 2p [41,42] Notably, there is no peak related to Co3+ and metallic Co in the Co–g-C3N4 catalysts These results can approve that Co is chemically coordinated to the g-C3N4 matrix in the form of Co–N bonds [43] ACCEPTED MANUSCRIPT (a) (b) N1s O1s 400.1 398.8 N 1s g-C3N4 g-C3N4 Co−g-C3N4(5) Intensity (a.u.) Co−g-C3N4(5) Intensity (a.u.) Co2p 401.2 C1s Co−g-C3N4(10) Co−g-C3N4(15) Co−g-C3N4(10) Co−g-C3N4(15) Co−g-C3N4(20) 1200 1000 800 600 400 200 404 402 Binding Energy (eV) 781.7 Co 2p 788.0 785.2 Co−g-C3N4(5) Intensity (a.u.) 400 398 396 394 Binding Energy (eV) (c) SC Co−g-C3N4(10) Co−g-C3N4(15) M AN U Co−g-C3N4(20) 810 RI PT Co−g-C3N4(20) 800 790 780 Binding Energy (eV) Fig Survey scan (a), N 1s (b) and Co 2p (c) XPS spectra of pure g-C3N4 and Co–g-C3N4(x) TE D catalysts with different Co contents 3.7 Catalytic performance of the catalysts for selective oxidation of cyclohexane 3.7.1 Effect of Co content The catalytic performances of Co–g-C3N4 catalysts for the selective oxidation of EP cyclohexane were evaluated and the results are shown in Table In the blank testing, 3.5% conversion of cyclohexane and 89.1% selectivity to KA oil are AC C achieved Using pure g-C3N4 catalyst, 7.2% conversion of cyclohexane and 89.8% selectivity to KA oil could be obtained, because g-C3N4 materials can reductively adsorb O2 [22,44] When Co-doped g-C3N4 hybrids are employed as the catalysts, the higher conversion of cyclohexane could be obtained than that over pure g-C3N4 and CoCl2, as well as the physical mixture of CoCl2 and pure g-C3N4 (7.8% conversion of cyclohexane), indicating synergistic effect of Co species and g-C3N4 hybrids Furthermore, the conversion of cyclohexane monotonically increases with the increase in the Co content of the Co–g-C3N4 catalysts For example, the conversion of cyclohexane gradually increases from 9.3% to 11.2%, when the content of Co increases from 3.2 wt% to 11.9 wt% On contrary, the selectivity to ACCEPTED MANUSCRIPT KA oil decreases with the increase in the Co content of the Co–g-C3N4 catalysts, due to the deep oxidation of products KA oil Furthermore, the ratio of cyclohexanol to cyclohexanone exhibits the similar trend as the selectivity to KA oil Comparing with pure g-C3N4, Co-doped g-C3N4 catalysts show an obvious decrease in the ratio of cyclohexanol to cyclohexanone and the lowest ratio is obtained over RI PT Co–g-C3N4(20) In summary, the Co–g-C3N4(15) catalyst exhibits the highest yield (8.97%) of KA oil among the Co-g-C3N4 catalysts, accompanying with 10.6% conversion of cyclohexane and 84.6% selectivity to KA oil On the other hand, different metal-doped g-C3N4 catalysts were also synthesized and employed in the SC selective oxidation of cyclohexane under the same reaction conditions for comparison As shown in Table 1, transition metal (Mn and Cu) or rare earth (Ce) doped g-C3N4 catalysts also show the higher conversion of cyclohexane than that of M AN U pure g-C3N4, but lower than that of Co-doped g-C3N4 catalysts Table Catalytic oxidation of cyclohexane over different catalystsa Conversion (wt%) (%) A K Others Blank test - 3.5 39.6 49.5 10.9 1.28 g-C3N4 - 7.2 39.7 50.1 10.2 1.29 Co–g-C3N4(5) 3.2 9.3 33.5 57.4 9.1 1.75 Co–g-C3N4(10) 6.3 9.8 31.4 57.6 11.0 1.87 Co–g-C3N4(15) 9.0 10.6 29.4 55.2 15.4 1.92 Co–g-C3N4(20) EP Selectivity (%) Co content 11.9 11.2 27.2 51.1 13.2 1.92 - 8.4 48.3 39.1 12.6 0.83 - 7.8 44.9 45.1 10.0 1.03 Co–g-C3N4(15)c 9.0 trace - - - - Mn–g-C3N4 8.8 8.0 36.1 37.5 26.4 1.06 Cu–g-C3N4 9.2 8.8 36.6 49.7 13.7 1.39 Ce–g-C3N4 8.7 9.1 32.8 56.3 10.9 1.75 CoCl2 AC C CoCl2+g-C3N4b TE D Catalyst a K/A Reaction conditions: g cyclohexane, 20 mg catalyst, 10 µL TBHP as initiator, initial p(O2) = 1.0 MPa, at 140 °C for h By-products mainly include diacids (succinic acid and adipic acid) and esters (dicyclohexyl adipate and hexanolactone) b c Physical mixture of mg CoCl2 and 16 mg g-C3N4 as catalyst 120 mg of hydroquinone was added ACCEPTED MANUSCRIPT 3.7.2 Effect of reaction temperature Fig 7a shows the effect of reaction temperature on the selective oxidation of cyclohexane over the Co–g-C3N4(15) catalyst It is clear that the conversion of cyclohexane sharply increases from 3.0% to 10.6% when the reaction temperature RI PT increases from 120 to 140°C However, continually increasing the reaction temperature to 160 °C, the conversion of cyclohexane remains steady On the contrary, the selectivity to KA oil gradually decreases from 93.4% to 78.8% with the increase in the reaction temperature In summary, the reaction temperature of 140 °C is KA oil (a) 30 (b) 100 100 60 15 40 10 20 0 120 130 140 150 160 Conversion & Selectivity (%) 20 80 Conversion of cyclohexane Selectivity to Cyclohexanol Selectivity to Cyclohexanone M AN U 80 Selectivity (%) Conversion (%) 25 SC optimal, with an attractive conversion of cyclohexane and a moderate selectivity to 60 40 20 10 o TE D Temperature ( C) 10 15 20 25 Amount of Catalyst (mg) Fig The effect of reaction temperature (a) and amount of catalyst (b) on cyclohexane oxidation over the Co–g-C3N4(15) catalyst EP 3.7.3 Effect of catalyst amount Fig 7b shows the effect of catalyst amount on the selective oxidation of AC C cyclohexane over the Co–g-C3N4(15) catalyst With the increase in catalyst amount from mg to 25 mg, the conversion of cyclohexane increases firstly and gradually comes to a standstill, while the selectivity to KA oil decreases Meanwhile, the highest ratio of cyclohexanone to cyclohexanol can be obtained when 20 mg of Co–g-C3N4(15) catalyst are used in the reaction The results above suggest that 20 mg is an optimal amount of catalyst under applied reaction conditions 3.7.4 The stability of the Co–g-C3N4(15) catalyst To investigate the stability of the Co–g-C3N4(15) catalyst for the selective oxidation of cyclohexane in solvent-free reaction system, its reusability was examined ACCEPTED MANUSCRIPT and the results are shown in Fig The Co–g-C3N4(15) catalyst can maintain the high conversion of cyclohexane and selectivity to KA oil in runs without obvious deterioration, indicating that the Co–g-C3N4(15) catalyst possesses a high stability for the selective oxidation of cyclohexane Fig shows FT-IR spectra of fresh and used Co–g-C3N4(15) catalyst after runs of cyclohexane oxidation Compared with the RI PT fresh catalyst, FT-IR spectrum of the used catalyst is hardly changed after the reaction, indicating that the typical structure of g-C3N4 hybrids can be retained after runs of cyclohexane oxidation Furthermore, the Co content of the used Co–g-C3N4(15) catalyst is 8.8 wt%, which is very close to that of the fresh catalyst (9.0 wt%) 100 Cyclohexane conversion Selec tivity to KA oil SC 80 70 60 50 M AN U Conversion & Selectivity (%) 90 10 5 Cycle (time) s-triazine ring modes (805) Used -C-N str (1243) AC C EP Transmittance (a.u.) TE D Fig Recycling runs of the Co–g-C3N4(15) catalyst for cyclohexane oxidation -C=N str (1637) Fresh 1800 1600 1400 1200 1000 800 600 400 -1 Wavenumber (cm ) Fig FT-IR spectra of fresh and used Co–g-C3N4(15) catalyst after runs of cyclohexane oxidation 3.7.5 The reaction mechanism It has been widely accepted that a free radical mechanism is implicated in liquid-phase free-solvent oxidation of cyclohexane with molecular oxygen [45-47] To further probe the reaction pathway, controlled experiment was carried out With ACCEPTED MANUSCRIPT the addition of hydroquinone as the radical scavenger (Table 1, Entry 9), the reaction is suppressed and no products are obtained, confirming the existence of free radicals in reaction system Co salt has been widely used as homogenous catalyst in the conventional industrial processes for selective oxidation for cyclohexane Under the reaction conditions in RI PT our work, CoCl2 also exhibits a high conversion of cyclohexane (8.4%) If the physical mixture of CoCl2 and g-C3N4 is used as catalyst, the conversion of cyclohexane decreases to 7.8%, due to a relatively low activity of pure g-C3N4 (7.2%) On the contrary, the conversion of cyclohexane can be obviously improved over SC Co-doped g-C3N4 catalysts due to the synergistic effect of Co species and g-C3N4 hybrids According to the characterization results, Co species is chemically coordinated to the g-C3N4 matrix in the form of Co(II)–N bonds, which are more M AN U active for cyclohexane oxidation than C–N bonds The related sequence of reactions can be expressed in Eqs (1) – (6) Initially, oxygen can insert into C–H bonds of cyclohexane to generate the primary chain-propagation product and the most important intermediate, i.e cyclohexyl hydroperoxide (CHHP) (Eq (1)) The g-C3N4 materials can reductively adsorb O2 and TE D then impels the generation of CHHP, as shown in scheme N N N N N N N N N N N e- N N N N N N N N N N O2 N N N N N O2 - + N e- EP OOH AC C Scheme Reductively adsorption of O2 on the g-C3N4 materails and the transfer to cyclohexane [44] Subsequently, the cleavage of CHHP can proceed based on a so-called Haber–Weiss catalytic cycle (Eq (2)) accompanied with Co2+↔ Co3+ transformation, and this process is faster than thermal decomposition [48-50] As shown in Eqs (3) and (4), the propagation process involves H abstraction from cyclohexane by CyOO• (cyclohexyl peroxy radicals) and CyO• to form CHHP and cyclohexanol, respectively, accompanying with the formation of Cy• (cyclohexyl radicals) Meanwhile, molecular oxygen can rapidly react with Cy• to rapidly generate CyOO• (Eq (5)) [51] Finally, ACCEPTED MANUSCRIPT the chain-termination process involves mutual destruction of two CyOO• radicals to form cyclohexanol and cyclohexanone (Eq (6)) OOH (1) O2 OOH O OH - (2) Co3+ Co2+ RI PT + OOH SC OO H OO M AN U OOH + (3) OH O TE D + (4) OO (5) + O2 EP O OO AC C OO OH (6) + O2 Conclusions Co-doped g-C3N4 catalysts with different Co content have been successfully synthesized and their catalytic performances for the selective oxidation of cyclohexane with molecular oxygen was investigated Based on varied characterizations, Co ions were evenly distributed on g-C3N4 matrix in the form of Co(II)–N bonds, which is similar to metalloporphyrins structure As a result, Co-doping can obviously improve the catalytic performance of g-C3N4 for the ACCEPTED MANUSCRIPT selective oxidation of cyclohexane 10.6% conversion of cyclohexane and 84.6% selectivity to KA-oil were obtained over the Co–g-C3N4(15) catalyst under the optimized reaction conditions Meanwhile, a Haber–Weiss catalytic cycle can be proposed to explain the promotion of the activity for cyclohexane oxidation over Co-doped g-C3N4 catalysts These results can be extended to the selective oxidation of RI PT sp3 C–H bonds with molecular oxygen to produce ketones and alcohols Acknowledgements This project was supported financially by the National Natural Science Foundation of SC China (91545103, 21103048) References M AN U [1] J.M Thomas, R Raja, G Sankar, R G Bell, Nature 398 (1999) 227–230 [2] J Zhang, X Liu, R Blume, A Zhang, R Schlögl, D S Su, Science 322 (2008) 73–77 [3] L Kesavan, R Tiruvalam, M.H.A Rahim, M.I.b Saiman, D.I Enache, R.L Jenkins, N Dimitratos, J.A Lopez-Sanchez, S.H Taylor, D.W Knight, C.J Kiely, G.J Hutchings, Science 331 (2011) 195–199 [4] J.A Labinger, J.E Bercaw, Nature 417 (2002) 507–514 TE D [5] M.S Chen, M.C White, Science 327 (2010) 566–571 [6] K Kamata, K Yonehara, Y Nakagawa, K Uehara, N Mizuno, Nat Chem (2010) 478–483 [7] A.K Suresh, M.M Sharma, T Sridhar, Ind Eng Chem Res 39 (2000) 3958–3997 [8] M.Z Wu, W.C Zhan, Y.L Guo, Y Guo, Y.S Wang, L Wang, G.Z Lu, Appl Catal A- Gen EP 523 (2016) 97–106 [9] Y Cao, H Yu, J Tan, F Peng, H Wang, J Li, W Zheng, N.B Wong, Carbon 57 (2013) AC C 433–442 [10] H Yu, F Peng, J Tan, X Hu, H Wang, J Yang, W Zheng, Angew Chem Int Ed 50 (2011) 3978–3982 [11] R Raja, G Sankar, J M Thomas, J Am Chem Soc 121 (1999) 11926–11927 [12] A Sakthivel, P Selvam, J Catal 211 (2002) 134–143 [13] Y Fu, W Zhan, Y Guo, Y Wang, X Liu, Y Guo, Y Wang, G Lu, Micropor Mesopor Mater 214 (2015) 101–107 [14] T Punniyamurthy, S Velusamy, J Iqbal, Chem Rev 105 (2005) 2329−2363 [15] M.A Patel, F Luo, M.R Khoshi, E Rabie, Q Zhang, C.R Flach, R Mendelsohn, E Garfunkel, M Szostak, H He, ACS Nano 10 (2016) 2305−2315 [16] X Wang, K Maeda, A Thomas, K Takanabe, G Xin, J.M Carlsson, K Domen, M Antonietti, Nat Mater (2009) 76−80 ACCEPTED MANUSCRIPT [17] J Zhang, X Chen, K Takanabe, K Maeda, K Domen, J.D Epping, X Fu, M Antonietti, X Wang, Angew Chem Int Ed 49 (2010) 441−444 [18] J Liu, Y Liu, N Liu, Y Han, X Zhang, H Huang, Y Lifshitz, S.T Lee, J Zhong, Z Kang, Science 347 (2015) 970−974 [19] P Niu, Y Yang, J.C Yu, G Liu, H.M Cheng, Chem Commun 50 (2014) 10837−10840 Y Huang, Y Wang, Y Bi, J Jin, M.F Ehsan, M Fu, T He RSC Adv (2015) 33254−33261 RI PT [20] [21] J Yu, S Wang, J Low, W Xiao, Phys Chem Chem Phys 15 (2013) 16883−16890 [22] F Su, S.C Mathew, G Lipner, X Fu, M Antonietti, S Blechert, X Wang, J Am Chem Soc 132 (2010) 16299−16301 Y Wang, H Li, J Yao, X Wang, M Antonietti, Chem Sci (2011) 446–450 [24] X.H Li, X Wang, M Antonietti, ACS Catal (2012) 2082−2086 [25] Y Wang, J Zhang, X Wang, M Antonietti, H Li, Angew Chem Int Ed 49 (2010) SC [23] [26] M AN U 3356−3359 D Yang, T Jiang, T Wu, P Zhang, H Han, B Han, Catal.Sci.Technol (2016) 193–200 [27] C Guo, M Chu, Q Liu, Y Liu, D Guo, X Liu Appl Catal A- Gen 246 (2003) 303–309 [28] X Wang, X Chen, A Thomas, X Fu, M Antonietti, Adv Mater 21 (2009) TE D 1609–1612 [29] Q Liu, J Zhang, Langmuir 29 (2013) 3821−3828 [30] A Thomas, A Fischer, F Goettmann, M Antonietti, J O Müller, R Schlögl, J.M Carlsson, J Mater Chem 18 (2008) 4893−4908 X Wang, K Maeda, X Chen, K Takanabe, K Domen, Y.D Hou, X Fu, M Antonietti, EP [31] J Am Chem Soc 131 (2009) 1680–1681 [32] F Goettmann, A Fischer, M Antonietti, A Thomas, Angew Chem Int Ed 45 (2006) AC C 4467–4471 [33] A.B.P Lever, Inorganic Electronic Spectroscopy, Elsevier, Amsterdam, 1968, Chapter 9, p 323 [34] L Poul , N Jouini, F Fiévet, Chem Mater 12 (2000) 3123–3132 [35] M.S Hamdy, A Ramanathan, T Maschmeyer, U Hanefeld, J.C Jansen, Chem Eur J 12 (2006) 1782–1789 [36] A.P Katsoulidis, D.E Petrakis, G.S Armatas, P.N Trikalitis, P.J Pomonis, Micropor Mesopor Mater 92 (2006) 71–80 [37] V.G Hadjiev, M.N Iliev, I.V Vergilov, J Phys C: Solid State Phys 21 (1988) 199–201 [38] M Kawaguchi, Adv Mater (1997) 615–625 ACCEPTED MANUSCRIPT [39] H.M Jeong, J.W Lee, W.H Shin, Y.J Choi, H.J Shin, J.K Kang, J.W Choi, Nano Lett 11 (2011) 2472–2477 [40] Z.S Wu, W Ren, L Xu, F Li, H.M Cheng, ACS Nano (2011) 5463–5471 [41] S Todorova, H Kolev, J.P Holgado, G Kadinov, C Bonev, R Períguez, A Caballero, Appl Catal B: Environ 94 (2010) 46–54 [42] D Schanke, S Vada, E.A Blekkan, A.M Hilmen, A Hoff, A Holmen, J Catal 156 [43] RI PT (1995) 85–95 M Yuasa, A Yamaguchi, H Itsuki, K Tanaka, M Yamamoto, K Oyaizu, Chem Mater 17 (2005) 4278−4281 [44] V.V Strelko, N.T Kartel, I.N Dukhno, V.S Kuts, R.B Clarkson, B.M Odintsov, Surf Sci 548 (2004) 281–290 I Hermans, T.L Nguyen, P.A Jacobs, J Peeters, ChemPhysChem (2005) 637–645 [46] I Hermans, P Jacobs, J Peeters, Chem Eur J 13 (2007) 754–761 [47] B.P.C Hereijgers, B.M Weckhuysen, J Catal 270 (2010) 16–25 [48] J.F Black, J Am Chem Soc 100 (1978) 527–535 [49] D.L Vanoppen, D.E Devos, M.J Genet, P.G Rouxhet, P.A Jacobs, Angew Chem Int Ed 34 (1995) 560–563 [50] L Vereecken, T.L Nguyen, I Hermans, J Peeters, Chem Phys Lett 393 (2004) 432–436 [51] M AN U SC [45] B Modén, B.Z Zhan, J Dakkab, J.G Santiesteban, E Iglesia, J Catal 239 (2006) AC C EP TE D 390–401