Liquid-phase epoxidation of propylene with molecular oxygen by chloride manganese meso-tetraphenylporphyrins

Xian-Tai Zhou,Ling-Ling Wang,Yang Li,Hong-Bing Ji,3,*

1 Fine Chemical Industry Research Institute,School of Chemical Engineering and Technology,Sun Yat-sen University,Zhuhai 519082,China

2 Fine Chemical Industry Research Institute,School of Chemistry,Sun Yat-sen University,Guangzhou 510275,China

3 School of Chemical Engineering,Guangdong University of Petrochemical Technology,Maoming 525000,China

Keywords:Propylene Epoxidation Dioxygen Manganese porphyrins Liquid phase

ABSTRACT Propylene molecule owns two active sites,the direct epoxidation of propylene by dioxygen is still a challenge due to the limitation of selectivity.In this work,the direct liquid-phase propylene aerobic epoxidation protocol by chloride manganese meso-tetraphenylporphyrin (MnTPPCl) was developed.The conversion of propylene was 12.7%,and the selectivity towards PO (propylene oxide) reached up to 80.5%.The formation of PO was attributed to the mechanism via high-valent Mn species,which was confirmed by means of in situ UV-vis spectrum.

1.Introduction

Propylene oxide (PO) is a key intermediate chemical largely produced for manufacturing numerous commodity chemicals such as polyurethanes,polyester resins,and other products [1-4].Chlorohydrin and organic peroxide processes are the current industry producing PO routes,which have the drawbacks of generating waste,high cost or the risk of security [5-8].Therefore,developing direct propylene epoxidation with molecular oxygen as oxidant is still a significant challenge [9-11].

Inspired by ethylene epoxidation process,silver catalysts are deemed to be the most promising catalyst in the gas-solid catalytic system.Numerous studies around the modified Ag catalysts were reported [12-15].Extra metal catalysts,such as Au catalysts [16-18],copper or copper-manganese mixed metal oxide catalysts[18-20],and MoO3or Mo-Bi catalysts [21-23],have been developed for the direct epoxidation of propylene.However,due to the thermodynamic limitations,the allylic C-H bonds in propylene tend to be oxidized,which results in the low selectivity towards PO[24-26].Meanwhile,the reaction conditions for such gas-solid phase epoxidations are usually rigid.

Therefore,epoxidation of propylene involving homogeneous catalytic systems is an alternative way to improve the selectivity of PO under mild conditions.Efforts have been dedicated to the liquid-phase epoxidation of propylene,in which the engaged catalysts mainly focused on modified titanosilicate [27-30].A tungsten-containing catalyst showed excellent efficiency for PO production when coupled with the 2-ethylanthraquinone/2-ethy lanthrahydroquinone redox process forin situH2O2generation[31].Promising results were obtained for the direct PO production based on palladium nanoparticles supported onto nanocrystalline titanium silicalite zeolite,in which hydrogen peroxide wasin situgenerated from hydrogen and oxygen with supercritical CO2as solvent [32].

Aerobic oxidation of hydrocarbons catalyzed by metalloporphyrins has attracted much attention due to the efficient and friendly conditions [33-36].Recently,we reported the liquidphase epoxidation of propylene with molecular oxygen catalyzed by manganese porphyrins,ruthenium porphyrins and Ce(SO4)2[37,38].Although high efficiency and selectivity of PO were obtained,the large consumption of benzaldehyde as co-substrate and cerium (IV) sulfate were unfavorable for industrial.

As our continuous research in pursuing more efficient and simpler epoxidation process of propylene,the liquid-phase epoxidation catalyzed by metalloporphyrins without any additives was developed in this work.The influence of various reaction parameters such as solvent,catalyst and reaction temperature on the activity and selectivity towards PO was evaluated.The protocol also provides a possible approach for manufacturing PO instead of chlorohydrin and organic peroxide routes.

2.Experimental

2.1.General methods

Chemicals were of analytical grade and purchased from Aldrich without further purification unless indicating.Propylene and molecular oxygen were purchased from Zhuozheng Gas Co.,Ltd.Solvents were of analytical purity and used as received.Metalloporphyrins were prepared according to our previously described methods [39].

Mass spectra were obtained on a Shimadzu LCMS-2010A.MALDI-TOF-MS spectrum was obtained on a Bruker REFLEX III.Elemental analyses were carried out with an Elementar vario EL elemental analyzer.1HNMR was recorded on a Bruker AVANCE 400 spectrometer (500 MHz).FT-IR spectra were recorded on a Bruker 550 FT-IR spectrometer.Thein situUV-vis spectra of metalloporphyrins were recorded on the AvaSpec-2048 spectrometer,which was equipped with a high-pressure,high-temperature probe and connected to the stainless-steel reactor.

2.2.Epoxidation of propylene

In a typical experiment,the stainless steel reactor was sealed after liquid reagents and metalloporphyrins solution were added in,subsequently,propylene was introduced by weighing,and dioxygen was added successively inside the reactor and kept for 20 min to achieve the gas-liquid balance.The mixture was heated at 100 °C for 6 h.Then,the reactor was quenched in ice for sampling.Further,the products of each reaction were determined by gas chromatography (GC,Shimadzu GC-2010 plus) and GC-MS(Shimadzu GCMS-QP2010 plus) systems.Each catalytic reaction was repeated three times to secure reproducibility.Thein situUV-vis spectrum of catalyst was conducted in a stainless steel reactor,which connected to an AvaSpec-2048 spectrometer.The spectrophotometer was programmed for acquiring UV-vis spectrum.

3.Results and Discussion

Catalytic activity of metelloporphyrins catalysts for propylene epoxidation was measured.The chemical equation of propylene epoxidation was presented in Fig.1.In metalloporphyrinscatalyzed direct liquid oxidation of propylene,main products are propylene oxide (PO),acetaldehyde and acrolein.

3.1.Effect of catalyst on propylene epoxidation

The effects of different metals on the epoxidation of propylene have been investigated.First,in control experiment without catalyst,only 1.4% conversion of propylene was achieved (Entry 1 in Table 1).However,both propylene conversion and PO selectivity increased obviously by adding metalloporphyrins catalysts(Entries 2-5 in Table 1).Comparing catalytic activities of catalysts,manganese porphyrin was the most effective,in which propylene could be converted in 12.7% conversion and the selectivity towards PO was 80.5% (Entry 2 in Table 1).The catalytic activity of different metal ions is probably influenced by their electric potential and the stability of metal atoms valence [40].In addition,the higher activity of manganese porphyrin was mainly the lower charge density around central manganese ion.The fewer positive charges of MnTPPCl would favor the combination of Mn2+and oxygen molecule or active oxygen species [41].In addition,the products of propylene oxidation were influenced by the different metals in some extent.It seemed that cobalt and copper complexes favored the selectivity of acetaldehyde that generated from C═C bond cleavage oxidative [42,43].

Subsequently,using MnTPPCl as catalyst,effect of catalyst amount on propylene epoxidation was examined.It could be known that the efficiency was close related with the amount of catalyst.Lower usage of catalyst presented less efficient (Entry 6 in Table 1).The conversion of propylene increased as catalyst amount rose (Entries 6-8 in Table 1).But no significant efficiency difference was observed when the amount of catalyst was up to 0.01% (mol) (based on propylene).It also could be known that the selectivity of PO was not significantly affected by the amount of catalyst.But when the amount of catalyst was increased contin-uously,the selectivity of acrolein generated from the α-H oxidation increased slightly (Entry 8 in Table 1).

Table 1 Effect of catalyst on the propylene epoxidation

Fig.1.Direct epoxidation of propylene catalyzed by metalloporphyrins.

3.2.Effect of solvent on propylene epoxidation

The effect of solvent on the propylene oxidation with MnTPPCl as catalyst was also examined.As shown in Table 2,the efficiency and PO selectivity is closely related with the polarity of solvents.Ethyl acetate,as one of the moderate polarity solvents was most favorable to the propylene epoxidation.The efficiency differences among solvents could be attributed to two aspects.One is the solvation affects,another side is ethyl acetate is favorable for the solubility of propylene (Entry 3 in Table 2).

In addition,the selectivity of product was related with the polarity of solvents.The increased selectivity of acetaldehyde in nonpolar solvents was observed(Entries 4-6 in Table 2).Nonpolar solvent such as benzotrifluoride was favorable for the stability of free radicals during the oxidation [44].Under the reaction conditions,no products from cyclohexane oxidation was observed(Entry 5 in Table 2).And in the blank experiment,no reaction occurred when cyclohexane was used as substrate under the same conditions.Hence,the effect of solvent consumption on cyclohexane oxidation could be excluded.Due to the two active sites of propylene molecule,the oxidized products like PO,acetaldehyde and acrolein were generatedviadifferent parallel reaction pathways.Hence,the solvent with moderate polarity and good solubility for propylene was favorable for the selectivity of PO.

Table 2 Effect of solvent on the propylene epoxidation

3.3.Effect of temperature and pressure on propylene epoxidation

In epoxidation of propylene,the conversion and selectivity are related to reaction temperature.As presented in Fig.2(a),the conversion of propylene was greatly influenced by the reaction tem-perature.With increasing temperature,propylene conversion increased as the temperature rose.Furthermore,increased reaction temperature from 70°C to 120°C,the reaction rate increased substantially,where conversion improved remarkably from 1.1% to 14.5%.It indicates that the arising temperature promotes the formation of free radicals in the catalytic system,resulting in the increasing reaction rate.The selectivity towards PO decreased under higher temperature,especially when the reaction temperature was higher than 100 °C.Meanwhile,the selectivity towards both acetaldehyde and acrolein increased.

The effect of oxygen pressure on propylene epoxidation was also investigated (Fig.2(b)).Propylene conversion generally increased with the oxygen pressure,but the increase rate was not significant when the oxygen pressure was higher 1.6 MPa.In addition,the oxygen pressure had a great influence on the selectivity of products in the catalytic protocol.PO selectivity gradually decreased with oxygen pressure.The higher oxygen concentration at higher pressures promoted the generation of acetaldehyde.It seemed that oxygen pressure had little effect on the selectivity of acrolein.

3.4.Mechanistic insight into the propylene epoxidation

The profiles of propylene epoxidation catalyzed by MnTPPCl and molecular oxygen is shown in Fig.3.In the first 2 h,propylene conversion slowly increased.Followed,the reaction rate accelerated rapidly.There is an obvious induction period in the catalytic epoxidation system.Propylene epoxidation exhibited the features of the radical-involved reaction.To test the free-radical mechanism,free-radical inhibitor (2,6-di-tert-butylphenol,0.5 mmol)was added in the solution.It was observed that the epoxidation was subsequently quenched.

From the reaction profile,the selectivity of PO,acrolein and acetaldehyde hardly fluctuated during the reaction process.High reaction temperature and nonpolar solvents are both favorable for the generation of acetaldehyde.It is necessary to consider the mechanism involving PO as the intermediates.Oxidation using PO as substrate was carried out under the same reaction conditions as epoxidation of propylene.However,no acetaldehyde or formaldehyde was obtained (Fig.4).Therefore,the probability of acetaldehyde formation from propylene oxide could be excluded.

Bhan group reported that propylene,acrylic acid and acrolein were the precursors of acetaldehyde [45].Acrylic acid was not found in the liquid phase.Therefore,the control experiment was carried out to determine whether acetaldehyde was generated from acrolein oxidation under the same reaction conditions as propylene oxidation.However,no acetaldehyde was determined(Fig.4).Therefore,acetaldehyde was produced from the deep oxidation of acrolein could also be excluded.It also indicated that the three products were generated by different parallel reaction pathways.

Fig.2.Effect of temperature (a) and oxygen pressure (b) on the epoxidation of propylene.

Fig.3.The reaction profile for the propylene epoxidation catalyzed by MnTPPCl.

Fig.4.Oxidation of propylene oxide and acrolein catalyzed by MnTPPCl.

It is well known that high valent metal-oxo species is the active oxidant in the epoxidation catalyzed by metalloporphyrins biomimetic catalysts [46].The epoxidation of propylene was conducted in a stainless-steel reactor,which connected to an AvaSpec-2048 spectrometer to testin situUV-vis spectra of manganese porphyrin catalyst.The spectrophotometer was programmed for acquiring UV-vis spectrum (Fig.5).

From Fig.5,as the reaction proceeded,the formation of a new intermediate with an electronic absorption band at 527 nm was observed.Meanwhile,the Soret band gradually decreased at 477 nm.Such changes forin situUV-vis spectra of manganese porphyrin could be attributed to the generation of the active species(PorMnIV=O) during the epoxidation.

Fig.5. In situ UV-vis spectra of the epoxidation of propylene catalyzed by manganese porphyrins in the presence of molecular oxygen.

Fig.6.Plausible mechanism of the direct propylene epoxidation by MnTPPCl.

Based on above experiments,a plausible mechanism of the direct epoxidation of propylene catalyzed MnTPPCl was proposed as shown in Fig.6.The α-hydrogen atom of alkene contains allyl group is easily abstracted in high temperatures.High-valent metal species were generated through free radical propagation initiated allylic radical[47].One of the allyl C—H bonds of propylene dissociates with the H atom to generate allylic radical.The reaction was proposed to be initiated by the generation of allylic radical(a).Ally peroxyl radical (b) rapidly formed by the combination of one oxygen molecule with allylic radical.Subsequently,Ally peroxyl radical reacted with manganese porphyrins,generating propylene alkoxyl radical (c) and high-valent Mn porphyrin intermediate(d).The formation of PO was attributed to the Jacobsen epoxidation mechanism through the high-valent Mn species (pathway A).Propylene alkoxyl radical,which picked up a hydrogen atom to generate acrolein.Acetaldehyde could be generated from the C═C bond oxidative cleavage mechanism.Radical (e) is formed by addition of ally peroxyl radical to propylene molecule.Acetaldehyde and acrolein could be generated through decomposition of the combination between radical (e) and oxygen molecule (pathway B).

Variety of scavengers such as DMPO(5,5-dimethyl-1-pyrroline-N-oxid),PBN (N-tert-butyl-alpha-phenylnitrone) were used to test series free radicals on EPR(electron paramagnetic resonance)spectrometer (JESFA-200,JEOL,Japan).But it is difficult to capture the propylene-based free radicals due to the excellent activity of radicals.Further studies on the detailed mechanism on metalloporphyrins-catalyzed direct epoxidation of propylene are in progress.

4.Conclusions

In summary,with molecular oxygen as oxidant,the liquidphase epoxidation of propylene catalyzed by metalloporphyrins was developed.The influence of various reaction parameters such as catalyst,solvent and reaction temperature on the activity and selectivity towards PO were investigated.Under the optimized reaction conditions,the conversion of propylene and selectivity towards PO was 12.7% and 80.5%,respectively.Based onin situUV-vis spectrum studies,a plausible mechanism for the direct epoxidation of propylene was proposed.In this protocol,the rate of propylene oxide production was 0.42 mol·m-3·min-1,which could provide one possible competitive approach for manufacturing PO instead of chlorohydrin and organic peroxide routes.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was financially supported by the National Key Research and Development Program of China (2020YFA0210900),the National Natural Science Foundation of China (No.21938001 and 21878344),Research and Innovation Team Construction Project of Guangdong University of Petrochemical Technology.