Chiang Mai Journal of Science

Selective Hydrogenation of Quinoline over Ruthenium Single-Atom Supported on M-MWCNTs Catalyst

1. INTRODUCTION

     The petrochemical industry is welcome in various countries, and the refining of oil has been received high attention from many scientists and fuel engineers [1]. There are a large amount of impurities as organic compounds including N, O or S elements in the oil, which can decline the quality of products [2]. In the refining of petroleum, the poisoning and deactivation of catalyst are caused easily by these impurities [3]. The removal of these impurities is important before oil refining, and common removal method including solvent extraction and biodegradation are popular [4]. Many kinds of impurity including oxygen, halogen or phenol can be reduced and released in the hydrogenation [5]. But the cyclic organic compounds including hetero atoms as thiophene or quinoline are difficult to remove during the refining process, and only the solvent extraction or high-temperature oxidation method can be utilized for these substances [6-7].
     As the typical heterocyclic compounds including nitrogen atom, quinoline show great apply value now [8]. Quinoline is the important raw materials for the synthesis of anti malaria, dysentery and anesthetics drugs [9-10]. For the other fields as synthesis of dyes, food additives and pesticides, it can be used widely [11], and the hydrogenation reaction product as 1,2,3,4-tetrahydroquinoline (THQ-1) also show broad application prospects [12]. THQ-1 can be used for the prevention and treatment of diseases such as the arteriosclerosis, hyperlipidemia and arrhythmia, and the good ability of electron donating which can be applied as the synthetic intermediate for photosensitive dyes and pesticide molecules [13]. The current industrial production of THQ-1 main depend on the hydrogenation of quinoline, and seeking catalysts with high activity and selectivity is very important [14-15].
     The catalytic hydrogenation of quinoline is shown as Scheme.1, quinoline and hydrogen are the reactants, and the products are found in the liquid solvent including THQ-1, THQ-2 and THQ-3. THQ-1 represents 1,2,3,4-tetrahydroquinoline, THQ-2 represents 5,6,7,8-tetra hydroquinoline and THQ-3 represents decahydro quinoline. The transition metals are used for the hydrogenation of quinoline including Co, Ru, Rh, Pd, Pt, Ir, Au, etc [16]. Although the transition metal catalysts showed good hydrogenation activity, but quinoline performed strong adsorbing effect to the nano surface of catalysts easily [17]. With the rapid decrement for the activity of nano metal center, and the surface of catalyst was changed easily. Hence, the strong toxicity was the barrier for good repeatability in the hydrogenation of quinoline or derivatives [18].

Scheme 1. The catalytic hydrogenation of quinoline.

     Au/HSA-TiO₂ nanocatalysts was used for the selective hydrogenation of isoquinoline, and the conversion of quinoline and the selectivity to py-THQ were both greater than 99% with the temperature was 60 °C, the reaction time was 8.5 hours and toluene was selected as the solvent [19]. Nano Pd catalyst with the average particle size of 1.7 nm was prepared with using MgO as the carrier and NaBH₄ as the reducing agent. Nano Pd/MgO catalyst were used for the selective hydrogenation of quinoline, and the conversion of quinoline was greater than 99% since the temperature was 150 °C, the hydrogen pressure was 4.0 MPa, the reaction time was 8 hours [20]. Matthias Beller prepared nano cobalt oxide/cobalt core-shell particles, which were encapsulated and loaded on the alumina carrier with the nitrogen-containing graphene layer [21]. The yield of THQ-1 reached 90% after 48 hours of reaction with the temperature of 120 °C. The conversion of quinoline was low to 76% after six cycles of the hydrogenation reaction. CoO_x@CN was prepared by the mesoporous nitrogen doped graphene monolayer encapsulated nano Co particles and direct pyrolysis method [22]. Good yield of THQ-1 was obtained with the methanol as the solvent, but the yield of THQ-1 decreased to 57% after 6 cycles of reaction. The significant agglomeration phenomenon was found by the characterization results, and CoO_x@CN particles showed lower lossing rate in the hydrogenation process. Agglomeration and devitalization were common in the hydrogenation of quinoline over metal catalysts as Ru, Rh, Pt and Ir[23].
      Single-atom catalysts become the popular nano catalysts due to their highest atomic utilization efficiency, excellent catalytic activity and good selectivity to product [24]. The single-atom as active site could not only serve as the ideal model system for heterogeneous catalytic reactions at the molecular level, but also it become the bridge for constructing heterogeneous or homogeneous catalysis [2]. The loaded components could remain stable in the hydrogenation process, and the different catalytic mechanism was discussed over the single-atom catalysts [26]. The catalytic activity of single-atom catalysts depended on their structure highly. Single-atom catalyst could be prepared by manys, and the properties of the single-atom catalyst could be confirmed throuth characterization [27].
     In this work, the conventional solvent evaporation and induced self-assembly method was used to prepare multi-walled carbon nanotubes supported Ru single-atom catalysts [28]. Modified multi-walled carbon nanotubes (M-MWCNTs) was selected to avoid carbon shrinkage during high-temperature calcination and the catalyst support showed the similar feature as the two dimensional hexagonal structure, which could be helpful for increasing the dispersion highly and stability of nano Ru catalysts. Multi-walled carbon nanotubes (MWCNTs) was functionalized by mixed solution of HNO₃ or H₂SO₄ to promote oxygen-containing functional groups and then further load the catalyst nanoparticles onto that functionalized MWCNT surface. Incorporation of noble metal atom (Pt, Ru or Pd) on the functionalized MWCNTs which were helpfulto the catalytic reaction [29]. Base on the physical chemical characterization results of catalysts, and selective hydrogenation of quinoline was investigated by the Ru single-atom catalysts compared to the traditional impregnation catalysts. Additionally, high selectivity to main products and good repeatability of Ru single-atom catalysts were approved, which indicated great significance for the industrial utilization of quinoline.

2. MATERIALS AND METHODS

2.1 Materials
     Phenol, formaldehyde, hydrochloric acid, sulphuric acid, toluene, sodium hydroxide and ethanol were purchased from Sigma Aldrich limited company (Analytical grade, Guangzhou, China). RuCl₃·3H₂O and ethyl orthosilicate (TEOS) were purchased from Macklin reagent limited company (Analytical grade, Shanghai, China). Surface active agent (F127) was purchased from Sinopharm chemical reagent limited corporation (Analytical grade, Shanghai, China). Quinoline and 3-methoxy-propyl-trimeth-oxysilane (MPTMS) were purchased from Beijing yiruokai technological limited corporation (Analytical grade, Beijing, China). Multi-walled carbon nanotubes (MWCNTs) was provided by Suzhou ruoensi carbon material technological limited corporation (Analytical grade, Suzhou, China). Nitrogen and hydrogen gas (99.99%) were purchased from Mingxing gas limited company (Maoming, China).

2.2 Methods
2.2.1 Preparation of catalyst
     Phenolic resin was prepared as the following method: 6.0 g of phenol was transferred in a round flask and 1.5 g of sodium hydroxide solution (5.0 mol·L^-1) was impregnated slowly in the water bath with continuous stirring at 45 °C. 12.0 g of formaldehyde solution (1.5 mol·L^-1) was transferred into this mixture with refluxing for 120 min at the same temperature. Hydrochloric acid solution (2.0 mol·L^-1) was impregnated for the neutralization after cooling down to room temperature. Vacuum distillation was used for evaporating the water, and the product was dissolved by ethanol after remove the sodium chloride crystal by filtration several times.
     Cosider the defects of the multi-walled carbon nanotubes (MWCNTs) was few and it was difficult for supporting some metal species, acidification was selected to optimize the surface of this material. The acidification process was as following, 75 mL sulphuric acid was transferred into the flask, 3.0 g MWCNTs was added into the acid liquid and stirred for 6 h at 90 °C. After cooled to the room temperature, the black powder was filtrated and washed by the distilled water. Then, this material was dried in air overnight at 120 °C and calcined at 400 °C for 6 h and it was named as M-MWCNTs.
     Both 1.0 g of TEOS and 1.0 g MPTMS were mixed in a round flask and 5.0 g of ethanol was transferred into this mixture with continuous stirring at 40 °C for 60 min. The certain mass of M-MWCNs powder was impregnated in 5.0 g of ethanol solution and 1.5 g of F127 was transferred into this mixture with continuous stirring at 40 °C for 30.0 min. These two mixtures were mixed quickly, and the certain volume of RuCl₃ solution (35.0 mmol·L^-1) was impregnated slowly in the water bath with continuous stirring at 40 °C. 5.0 g of phenolic resin solution was impregnated slowly with continuous stirring at 40 °C for 120 min, and the mixture was transferred into glass petri dish for the evaporation of the solvent. The white solid membranous substance was transferred to the vacuum drying oven and dried continuously at 80 °C for 600 min, and the solid powder was obtained by grind repeatedly.
     1.0 g of this material was transferred into a flask with hot refluxing by sulfuric acid solution (3.0 mol·L^-1) at 70 °C, and the mixture was maintained for 720 min in order to remove inorganic impurities. After cooled to the room temperature, the black solid was washed with distilled water and filtration several times. Finally, the black powder was transferred into the ceramic tube furnace for programmed heating and calcination after sufficient drying (heat up to 350 °C with the rate of 1 °C·min^-1 and maintain for 180 min under nitrogen gas, heat up to 600 °C with the rate of 3 °C·min^-1 and maintain for 180 min under hydrogen gas). After cooling, black carbon supported ruthenium catalyst was obtained and labeled as x%-Ru-m/M-MWCNs (x% represent the mass percentage of Ru in the catalyst).
     Traditional catalysts are also prepared by wet impregnation method which were labeled as x%-Ru/M-MWCNs (x% represent the mass percentage of Ru in the catalyst) for comparison. The detailed method as following, the certain volume of RuCl₃ solution (35.0 mmol·L^-1) was transferred into the round flask, and the certain amount of M-MWCNs powder was placed into this solution with continuous stirring at 40 °C for 480 min. The mixture was transferred into the drying oven for dried continuously at 110 °C for 12 h. After cooled to the room temperature, the black powder was transferred into the ceramic tube furnace for programmed heating and calcination (heat up to 350 °C with the rate of 1 °C·min^-1 and maintain for 180 min under nitrogen gas, heat up to 600 °C with the rate of 3 °C·min^-1 and maintain for 180 min under hydrogen gas). After cooling to the room temperature, black carbon supported ruthenium catalyst was obtained successfully.

2.2.2 Hydrogenation tests
     Hydrogenation of quinoline was performed in a 50 mL Teflon-lined stainless steel autoclave with a magnetic stirrer, the electric temperature controller and pressure instrument. The reactant solution (30 g of 10 wt% quinoline toluene solution) and the certain amount of catalyst were mixed carefully. The reactor was sealed and purged with hydrogen gas to exclude air five times, and it was pressurized to the definite pressure until reached the required temperature with the vigorous stirring.
     In the process of reaction, the autoclave was controlled intelligence, both the high temperature and low temperature were prevented. In the inner of autoclave, the gas pressure may was increased apparently with the elevation of the reaction temperature, but the gas pressure always under the design hydrogen gas pressure. Hence, the inner mixtures could not escaped from the autoclave under the external design high hydrogen gas pressure and the pressure could be stable.
     After the reaction, the catalysts were separated by filtration carefully from the liquid phase system. The quantitative determination of the liquid products was done by the internal standard method using GC/MS analytic technology. After the liquid phase products were analyzed accurately, the content of the selected products were determined by gas chromatography (Agilent Technologies, 8870 A) equipped with a HP-88, 60 m × 0.25 mm capillary column and the flame ionization detector (FID). The liquid phase products included 1,2,3,4-tetrahydroquinoline, 5,6,7,8-tetra hydroquinoline, decahydro quinoline and unreacted quinoline. The conversion of quinoline and the selectivity to major products was calculated with the following equations [30].

     In order to measure the catalytic activity, turn over frequency (TOF) was selected to evaluate the catalytic efficiency. Since the value of TOF was great by the following calculate formula (metal exposure rate was obtained by CO chemical adsorption), high efficiency of catalyst was confirmed.

     The initial reaction rate and the rate constant were calculated at different temperature since the conversion within 20%. The activation energy of this hydrogenation of quinoline also were obtained over x%-Ru-m/M-MWCNTs or x%-Ru/M-MWCNTs catalysts by the arrhenius equation (ln k = ln A- Ea/RT) [31]. The Entropy change of reaction was calculated by the equation (△S = ∑S_(products)-∑S_(reactants)), and the entropy value were obtained from Lange’s Handbook of Chemistry.

3. RESULTS AND DISCUSSION

     The textural properties of several catalysts are shown by the Table S1 in the supplementary file. The surface area of M-MWCNTs, 1%-Ru-m/M-MWCNTs and 1%-Ru/M-MWCNTs are 226.3, 220.3 and 194.8 m²·g^-1 respectively. It can be seen from Table S1 that the BET surface area decreases with the addition of Ru species, which are ascribed to the Ru species may loaded on the inner surface of the pore and blocked some pore channels. The pore volume and average pore size are the same trend with the increment of Ru species. But the micro pore ratio is increased to 65.47% with the increment of the Ru species. The Ru loading amount of the catalysts by the ICP characterization method were closer to the designed amount of Ru in the catalysts, which demonstrated the relative error was within the acceptable range.
     N₂ adsorption and desorption isotherms are shown by Figure S1 in the supplementary file. All samples exhibited the type H3 hysteresis loop according to IUPAC classification, whereas 1%-Ru-m/M-MWCNTs catalyst by Figure S1 in the supplementary file revealed the type I langmuir isotherms feature. The pore size distributions of the samples are displayed by Figure S2 in the supplementary file, and the pore size of 1%-Ru-m/M-MWCNTs catalyst were localized at 3.2 nm.
     The FT-IR spectra of the samples are shown by Figure S3 in the supplementary file. The bands around 1150, 1283, 1713, 2080, and 3415 cm^-1 in the curve of 1%-Ru-m/M-MWCNTs catalyst are the bending and stretching vibration peaks of S-O, -S-H, S=O in the -SO₃, C-C of M-MWCNTs, and O-H, respectively. The stretching vibration peak of Ru-O is not found in the curve, which is ascribed to the strong reduction in the adequate H₂ gas. The similar signals are existed in the curve of 1%-Ru/M-MWCNTs catalyst, and the shift of the wave number are distinct. These different signals indicated that the properties of the functional groups on the surface of catalyst were different.
     The XRD spectra of the samples are shown in Figure 1, it exhibits the characteristic diffraction peaks of the different crystal forms of carbon and Ru atomic lattice. Obviously, the characteristic diffraction peak at 2θ = 26.2° and 2θ = 42.5° are ascribed to the C(002) and C(100) for the multi-walled carbon nanotubes [32]. The characteristic diffraction peak at 2θ = 44.1° is ascribed to the (101) crystalline plane of Ru (JCPDS 88-1734), and which is consistent with the literature [33]. Compared with the 1%-Ru/M-MWCNTs, the peaks of C(002) and Ru(101) in the 1%-Ru-m/M-MWCNTs catalyst become weaker [29,34]. The average crystal particle size of Ru(101) was centered at 2.2 nm in the 1%-Ru-m/M-MWCNTs catalyst, but the average crystal particle size of Ru(101) was centered at 3.3 nm in the 1%-Ru/M-MWCNTs catalyst using the Debye-Scherrer formula. Better dispersion of Ru species was obtained in the Ru single-atom catalyst with the introduction of phenolic resin, and the average crystal particle size of Ru species became smaller.
     The hydrogen chemsorption data of some catalysts are summarized in Table 2. The 1%-Ru-m/M-MWCNTs catalyst demonstrated better dispersion, larger hydrogen uptake quantity and larger metal surface area from Table S2. Also, the H₂ uptake of ruthenium atom can be improved effectively with the increment of Ru species, but too much Ru species may plug in the channel and reduce the contact area between metal atoms and H₂, this is consistent with the characterization results of BET. Under the same content of Ru species in the catalyst, the apparent advantage of Ru single-atom catalyst was certified than the traditional impregnation catalysts.

     The hydrogen TPR profiles of the samples are given in Figure S4. The acute peak near 322 °C in the curve of 1%-Ru-m/M-MWCNTs catalyst is ascribed to the reduction of Ru species. However, the reduction peak of Ru is closer to 368 °C in the curve of 1%-Ru/M-MWCNTs catalyst and there are two weak reduction peaks between 400 °C to 620 °C. For the traditional impregnation catalysts, the reduction process of RuCl₃ may be achieved by some steps in the H₂ gas and the vary valence state of Ru metal species are existed at the different temperature. It was amazing that the Ru single-atom catalyst reflected the relative easier of reduction, and the uniform zero-valent metal atoms were existed on the surface of this catalyst [35].
     The TEM images of the Ru particles distribution of 1%-Ru-m/M-MWCNTs and 1%-Ru/M-MWCNTs catalysts are shown in Figure 2. For the Ru single-atom catalyst, good dispersion and the Ru particle sizes are centered at 1.0-2.0 nm, and Ru metal atomic agglomeration is not discovered. On the contrary, metal atomic agglomeration is existed in Figure 2(d) clearly and Ru particle sizes are centered at 1.5-3.0 nm. The uneven distribution of Ru metal atoms is confirmed effectively for the traditional impregnation catalyst.

     The aberration corrected TEM (AC-TEM) images of 1%-Ru-m/M-MWCNTs and 1%-Ru/M-MWCNTs catalysts are shown in Figure 3. Compared to the traditional impregnation catalyst, Ru metal atoms were dispersed on the surface of 1%-Ru-m/M-MWCNTs catalyst highly, and the vast majority metal existed in the form of single atoms. High dispersed atom catalysts exhibited significant advantages at the resistance for agglomeration or sintering in the selective hydrogenation process of quinoline.

     In order to investigate the effects of the reaction temperature, many experiments at the different temperature were performed over two kinds of catalyst, and the result was listed in Table S3 and Table S4. The reaction temperature played an important role in the hydrogenation of quinoline. The conversion of quinoline showed the rapid upward trend since the reaction temperature increased from 90 °C to 120 °C, and the conversion reached 100% and the selectivity to THQ-1 reached 99.67% over 1.0%-Ru-m/M-MWCNTs catalyst. As the reaction temperature was raised, the change of the conversion was not obvious and the selectivity to THQ-1 was decreased slightly, and the suitable reaction temperature was optimized at 120 °C. We also observed the different tendency from the data of Table S4 for the 1.0%-Ru/M-MWCNTs catalyst. The conversion of quinoline showed slow upward trend when the reaction temperature increased from 90 °C to 150 °C, and the selectivity to THQ-1 always was closer to 65% and the selectivity to THQ-3 showed apparent upward trend with the prolong of the reaction temperature. Hence, the higher temperature was not suitable for the high selectivity to main product. The content of aromatic compounds and benzene was decreased over the noble metals applied on various carriers in the hydrogenation [36], and which were consist with the hydrogenation result of quinoline over 1.0%-Ru/M-MWCNTs catalyst.
     H₂ pressure maybe is the important factor for this hydrogenation of quinoline, and the results are listed in the Table S5 and Table S6. The conversion of quinoline displayed the rapid upward trend when the H₂ pressure increased from 1 MPa to 4 MPa, and the conversion reached 100% and the selectivity to THQ-1 reached 99.67% over 1.0%-Ru-m/M-MWCNTs catalyst. But this phenomenon was different since the catalyst was replaced as 1.0%-Ru/M-MWCNTs in the Table S6. The conversion of quinoline displayed the slow upward trend when the H₂ pressure increased from 1 MPa to 6 MPa, and the selectivity to THQ-1 and THQ-3 displayed the slight uptrend. But the selectivity to THQ-2 was always above 10% and the selectivity to THQ-1 only reached 66.37% since the H₂ pressure was 4 MPa, and the higher H₂ pressure was not suitable for the high selectivity to main product in the hydrogenation of quinoline.
     The effects of reaction time over the 1%-Ru-m/M-MWCNTs catalyst was shown by the Figure S5. The conversion of quinoline was increased continually with the increment of the reaction time until 180 minutes, and the conversion of quinoline reached 100% and it was no longer changed with the prolong of the reaction time. The selectivity to THQ-1 was always closer to 100% with the prolong of the reaction time, and this phenomenon was belonged to the typical characteristics of the zero order reaction [37]. Combined the CO-TPD characterization result, the value of TOF reached 164.6 h^-1 since the conversion reached 20%. In the products of this reaction, the selectivity to THQ-2 and THQ-3 were closer to 0.1% with the prolong of the reaction time. Hence, the adsorption of the benzene ring for the Ru single-atom catalyst was nonexistent [34]. For the traditional impregnation catalysts, the adsorption of the benzene may was not ignorable, and the selectivity to THQ-2 or THQ-3 were distinct.
     Many different initial rates were obtained with the various initial concentration of quinoline, and the results are shown in Figure S6. The different initial reaction rates were always closer to 30 mol·L^-1·min^-1 with the increment of initial quinoline concentration. Hence, the reaction rate was independent of the initial reactant concentration, and which was consisted with the regularity of the zero order reaction.
     The values of TOF with the cyclical order over the 1%-Ru-m/M-MWCNTs catalyst are displaced in the Figure S7. The values of TOF are always closer to 160 h^-1 before the sixth cyclical order, the stable feature was obvious which indicated the great stability for the hydrogenation of quinoline. From the TEM and ACTEM characterization, high dispersed Ru metal atom catalysts exhibited significant advantages at the resistance for agglomeration. Good reusability of 1%-Ru-m/M-MWCNTs catalyst was confirmed for the nitrogen-containing hetero cyclic substrates and the metal active particles was not lost easily. The surface properties of the catalyst remain unchanged after several cycle testing.
     The effect of cyclical order for quinoline hydrogenation over various catalysts are shown by the Figure 4. For the Ru singe atom catalyst, the conversion of quinoline and the selectivity to THQ-1 always above 90% until the fifth used (Figure 4(a)). The catalytic performance decreased slightly until the fifth recycle and the catalyst could be recycled easily for this hydrogenation of quinoline. For the traditional impregnation catalyst, both the conversion of quinoline and the selectivity to THQ-1 showed the rapid decline with the cyclical used (Figure 4(b)). The conversion of quinoline was 55.12% and the selectivity to THQ-1 was 43.67% at the fifth used, which indicated the loss of Ru metal particles was serious. From the TEM and AC-TEM characterization of this catalyst, agglomeration or devitalization were occurred easily. On the contrary, the selectivity to THQ-3 was increased to 33.46% slightly, which was attributed to the continued catalysis of intermediate compound or products adsorbed on the surface of this catalyst. Ru single-atom catalyst displayed no adsorption properties for the intermediate products, so the good selectivity to THQ-1 were maintained well.
     After the reaction was completed, the reaction time was further extended to 8 hours. The significant increment of THQ-3 was not found over the 1%-Ru-m/M-MWCNTs catalyst. The desorption of the products contained nitrogen ring from the active sites was difficult, and which it was adsorbed on the surface of catalyst after hydrogenation [34,37]. In the competitive adsorption for the phenyl or nitrogen groups, and the nitrogen group was more adsorbed easily. The poisoning of catalyst was performed easily, and the reaction was prone to stop. The hydrogenation may required greater reliance on the lost metal atoms, and the selectivity to THQ-3 was increased over the 1%-Ru/M-MWCNTs catalyst since the reaction time was further extended to 8 hours. Hence, the repetitive stability and advantage of the Ru single-atom catalyst was confirmed for this hydrogenation of quinoline.
     The plot of initial reaction rates at the different temperature over various catalyst are shown by the Figure S8, and the activation energy and entropy change were obtained through the fitting and calculation. The good linear relationship reflected the stability of the initial reaction rate at the different temperature. The activation energy, entropy changes and the values of TOF over different catalysts are listed in the Table S7. The adsorption energy of quinoline on the surface of the 1%-Ru/M-MWCNTs catalyst was low relatively, but the desorption energy of the products on the surface of this catalyst was very high, and these were the main reasons for the elevated activation energy (69.82 kJ·mol^-1). The nitrogen-containing heterocycles promoted the improvement of catalytic activity by itself, which was also the reason for the initial reaction rate of 1%-Ru/M-MWCNTs catalyst was higher than the Ru single-atom catalyst [38]. Moreover, the greater the entropy change was helpful for the adsorption of intermediates on the surface of catalyst. The stronger ability for adsorbing nitrogen groups of the Ru single atom catalyst may be the reason for the decrement of react activity and the high selectivity to main product.

4. CONCLUSIONS

     The Ru single-atom catalyst was prepared by the conventional solvent evaporation and induced self-assembly method successfully, which exhibited great advantage in the selective hydrogenation of quinoline. The high conversion and high selectivity to THQ-1 were obtained for the Ru single-atom catalyst with the reaction temperature of 120 ℃ and the H₂ pressure of 4 MPa. The characterization results of catalysts, the initial reaction rate, TOF value, activation energy, entropy changes were obtained.
     Compared to the traditional impregnation catalyst, Physical characterization revealed the better dispersion of Ru atoms on the surface of the 1%-Ru-m/M-MWCNTs catalyst. The Ru single-atom catalyst showed the larger metal surface area, larger hydrogen uptake quantity and reduction of Ru species was easy relatively. 100% quinoline conversion and 99.67% selectivity to 1,2,3,4-tetrahydro quinoline (THQ-1) were obtained over the 1%-Ru-m/M-MWCNTs catalyst successfully. The hydrogenation rate of quinoline was belonged to the zero order reaction by the different initial reaction rate testing, and the value of TOF reached 164.6 h^-1. The activation energy was low as 58.13 kJ·mol^-1 and the entropy changes was -541.32 J·mol^-1·K^-1 combined experiments and calculations.
     The Ru single-atom catalyst showed good stability, reusability and high selectivity to THQ-1 over the Ru single-atom catalyst may be attributed to no adsorption of the benzene ring, easier dissociation step of H₂, higher resistance to nitrogen-containing heterocycles and no loss of Ru atoms. However, loss of Ru metal atom was occurred easily for the traditional impregnation catalyst, and the lost Ru nanoparticles exerted catalytic activity for the intermediates by many hydrogenation experiments.
     The activation energy, entropy changes and TOF value over different catalysts were calculated by the initial reaction rate testing. Lower activation energy and larger TOF value were confirmed by the Ru single-atom catalyst by the thermodynamics experiments. All these works exhibited the great advantage of the Ru single-atom catalyst which could be applied in the industrial utilization of quinoline.

ACKNOWLEDGEMENTS

     The authors are grateful for the financial support from the “sailing plan” applied research project of maoming green chemical industry research institute (MMGCIRI-2022YFJH-Y-036).

AUTHOR CONTRIBUTIONS

     Wei Long: Conceptualization, Methodology, Resources, Project administration, Funding acquisition, Experimental data, Characterization analysis, Writing-Reviewing and Editing. Yongxiang Zhu: Experimental data, Formal analysis, Validation, Writing-Original draft preparation.

CONFLICT OF INTEREST STATEMENT

     The authors declare that there are no conflicts of interest regarding the publication of this study.     
     All data, analyses, and conclusions presented in this research are based on objective findings without any financial, personal, or professional influences that could affect the integrity of the work.

DECLARATION OF GENERATIVE AI IN PREPARATION OF MANUSCRIPT

     During the preparation of this work, the author(s) used "ChatGPT-4o and Grammarly" to improve the readability and language of the manuscript. After using this tool, the author(s) reviewed and edited the content as necessary and take(s) full responsibility for the publication’s content.

FUNDING

     This research was financially supported by the “sailing plan” applied research project of maoming green chemical industry research institute (MMGCIRI-2022YFJH-Y-036).

REFERENCES

[1] Treitl S., Nolz P.C. and Jammernegg W., Incorporating environmental aspects in an inventory routing problem: A case study from the petrochemical industry. Flexible Services & Manufacturing Journal, 2014; 26: 143–169. DOI 10.1007/s10696-012-9158-z.

[2] Willing A., Lubricants based on renewable resources an environmentally compatible alternative to mineral oil products. Chemosphere, 2001; 43: 89-98. DOI 10.1016/S0045- 6535(00) 00328-3.

[3] Banda R., Sohn S.H. and Lee M.S., Process development for the separation and recovery of Mo and Co from chloride leach liquors of petroleum refining catalyst by solvent extraction. Journal of Hazardous Materials, 2012; 213-214: 1-6. DOI 10.1016/j.jhazmat. 2011.12.078.

[4] Martínez A.T., Speranza M. and Del-Río J.D., Biodegradation of lignocellulosics: Microbial, chemical, and enzymatic aspects of the fungal attack of lignin. International Microbiology, 2005; 8: 195-204. DOI 10.1016/j.ibiod.2005.05.002.

[5] Murakami Y., Kuboyama H., Tani H., Asami K. and Fujimoto K., Technical report effect of low-pressure secondary treatment in the development of a low-pressure synthesis process for biodiesel. Journal of the Japan Petroleum Institute, 2025; 68: 61-67. DOI 10.1627/jpi.68.61.

[6] Michael J.P., Quinoline, quinazoline and acridone alkaloids. ChemInform, 2004; 21: 650-668. DOI 10.1039/b310691h.

[7] Lu S.M., Wang Y.Q., Han X. and Zhou Y., Asymmetric hydrogenation of quinolines and isoquinolines activated by chloroformates. Angewandte Chemie International Edition, 2006; 45: 2260-2263. DOI 10.1002/anie.200503073.

[8] Zhao X., Fang R., Wang F. and Li Y., Coordination-mediated atomic metal catalysts for cascaded transformation of β-O-4 dimeric model compound into quinoline. AIChE Journal, 2024; 70. DOI 10.1002/aic.18432.

[9] Tanwar D., Mahala S., Ahluwalia D., Bhuvanesh N., Joshi H. and Kumar U., Nickel complexes bearing quinoline derived NNS donor ligands as catalytic activators for N-Alkylation of anilines with alcohols. Chemistry-An Asian Journal. 2024; 19: 557-569. DOI 10.1002/ asia.202400557.

[10] Munir R., Athar M.M., Zia-ur-Rehman M. and Javid N., Synthesis of 6/8-methyl-2-(piperdin-1-yl)quinoline-3-carbaldehydes: A facile CTAB catalyzed protocol. Chiang Mai Journal of Science, 2020; 47(1): 175-180.

[11] Lv K-H., Chen L. and Yan S.J., Cu-catalyzed decarboxylative annulation of nphenylglycines with maleimides: Synthesis of 1H-pyrrolo [3,4-c] quinoline-1,3(2h)-diones. Journal of Organic Chemistry, 2023; 88: 2358-2366. DOI 10.1021/acs.joc.2c02757.

[12] Lin X.F., Cui S.L. and Wang Y.G., A highly efficient synthesis of 1,2,3,4-tetrahydroquinolines by molecular iodine-catalyzed domino reaction of anilines with cyclic enol ethers. Tetrahedron Letters, 2006; 47: 4509-4512. DOI 10.1016/j.tetlet.2006.03.123.

[13] Kapitonov A.V., Ryzhkin S.A., Kurbatova S.V. and Zemtsova M.N., Thermodynamics of the sorption of quinoline derivatives, 1,2,3,4-tetrahydroquinoline and pyridazino[4,5-c] quinoline under conditions of RP HPLC. Russian Journal of Physical Chemistry A, 2024; 98: 2828-2837. DOI 10.1134/S00360 2442470208X.

[14] Ryzhkin S.A., Kapitonov A.V., Kurbatova S.V. and Zemtsova M.N., Sorption of 1,2,3,4- tetrahydro quinoline and pyridazino[4,5-c]quinoline derivatives under conditions of reversed-phase high-performance liquid chromatography. Russian Journal of Physical Chemistry, 2024; 98: 2346-2354. DOI 10.1134/S00360244247015 16.

[15] Ahmad S. and BühI M., Revisiting the reviewed: A meta-analysis of computational studies on transition metal-catalyzed hydrogenation reactions. ChemCatChem, 2025; 17(1): e202401053. DOI 10.1002/cctc.202401053.

[16] Lu S.M., Han X.W. and Zhou Y.G., An efficient catalytic system for the hydrogenation of quinolines. ChemInform, 2007; 692: 3065-3069. DOI 10.1016/j.jorganchem.2007.03.032.

[17] Bommannavar A.S. and Montano P.A., In situ study of the hydrogenation of quinoline over a Ni-Mo supported catalyst. Applications of Surface Science, 1984; 19: 250-266. DOI 10.10 16/0378-5963(84) 90065-5.

[18] Zhang L., Bo H.U., Chen H., Li X.J. and Li R.X., Catalytic performance of porous SiO₂·xH₂O supported RuB nanoparticles for the hydrogenation of quinoline. Acta Physico-Chimica Sinica, 2010; 26: 2422-2428. DOI 10.3866/PKU.WHXB20100901.

[19] Ren D., He L., Yu L., Ding R.S., Liu Y.M. and Fan K.N., An unusual chemoselective hydrogenation of quinoline compounds using supported gold catalysts. Journal of the American Chemical Society. 2012; 134: 17592-17598. DOI 10.1021/ja3066978.

[20] Rahi R., Fang M.F., Ahmed A. and Sánchez-Delgado R.A., Hydrogenation of quinolines, alkenes, and biodiesel by palladium nanoparticles supported on magnesium oxide. Dalton Transactions. 2012; 41: 14490-14497. DOI 10.1039/c2dt31533e.

[21] Chen F., Surkus A-E., He L., PohI M-M., Radnik J., Topf C., et al., Selective catalytic hydrogenation of heteroarenes with N-graphene-modified cobalt nanoparticles (Co₃O₄-Co/NGr@α-Al₂O₃). Journal of the American Chemical Society, 2015; 137: 11718-11724. DOI 10.1002/chin. 201609047.

[22] Wei Z.Z., Chen Y.Q., Wang J., Su D.F. and Wang Y., Cobalt encapsulated in N-doped graphene layers: An efficient and stable catalyst for hydrogenation of quinoline compounds. ACS Catalysis, 2016; 6(9): 5816-5822. DOI 10.1021/acscatal.6b01240.

[23] Vít Z., Cinibulk J. and Gulková D., Promotion of Mo/Al₂O₃ sulfide catalyst by noble metals in simultaneous hydrodesulfurization of thiophene and hydrodenitrogenation of pyridine: A comparative study. Applied Catalysis A: General. 2004; 272: 99-107. DOI 10.1016/j.apcata.2004.05.037.

[24] Shen Q.K., Jin H.Q., Li P.P., Yu X.H., Zheng L.R., Song W.G., et al., Breaking the activity limitation of iridium single-atom catalyst in hydrogenation of quinoline with synergistic nanoparticles catalysis. Nano Research, 2022; 15: 5024-5031. DOI 10.1007/s12274-022-4235-4.

[25] Rebelli J., Detwiler M., Ma S., Williams C.T. and Monnier J.R., Synthesis and characterization of Au-Pd/SiO₂ bimetallic catalysts prepared by electroless deposition. Journal of Catalysis, 2010; 270: 224-233. DOI 10.1007/s12274-022-4235-4.

[26] Liu P.X., Zhao Y., Qin R.X., Mo S.G. and Zheng N.F., Photochemical route for synthesizing atomically dispersed palladium catalysts. Science, 2016; 352: 797-800. DOI 10.1126/science.aaf5251.

[27] Pei G.X., Liu X.Y., Wang A.Q., Li L., Huang Y.Q., Zhang T., et al., Promotional effect of Pd single atoms on Au nanoparticles supported on silica for the selective hydrogenation of acetylene in excess ethylene. New Journal of Chemistry, 2014; 38: 2043-2051. DOI 10.1039/C3NJ01136D.

[28] Wei H.S., Liu X.Y., Wang A.Q., Zhang L.L., Qiao B.T., Yang X.F., et al., FeO_x-supported platinum single atom and pseudo-single-atom catalysts for chemoselective hydrogenation of functionalized nitroarenes. Nature Communications, 2014; 5: 5634-5641.DOI 10.1038/ncomms6634.

[29] Wongthep P., Weankeaw P., Sarakonsri T., Lapanantnoppakhun S. and Saipanya S., Successive reduction for preparation of various Pd_xPt_y-MWCNTs catalysts for formic acid oxidation. Chiang Mai Journal of Science. 2016; 43(3): 600-609.

[30] Jiang H.Y. and Zheng X.X., Tuning the chemoselective hydrogenation of aromatic ketones, aromatic aldehydes and quinolines catalyzed by phosphine functionalized ionic liquid stabilized ruthenium nanoparticles. Catalysis Science & Technology, 2015; 5: 3728-3734. DOI 10.1039/C5CY00293A.

[31] Park I.S., Kwon M.S. and Kang K.Y., Rhodium and Iridium nanoparticles entrapped in aluminum oxyhydroxide nanofibers: Catalysts for hydrogenations of arenes and ketones at room temperature with hydrogen balloon. Advanced synthesis & Catalysis, 2007; 349: 2039-2047. DOI 10.1002/adsc.200600651.

[32] Tessonnier J.P., Pesant L., Ehret G., Ledoux M.J. and Pham-Huu C., Pd nanoparticles introduced inside multi-walled carbon nanotubes for selective hydrogenation of cinnamaldehyde into hydrocinnamaldehyde. Applied Catalysis A General, 2005; 288(1-2): 203-210. DOI 10.1016/j.apcata.2005.04.034.

[33] Tang M.H., Deng J., Li M., Li X. and Yong W., 3D-interconnected hierarchical porous N-doped carbon supported ruthenium nanoparticles an efficient catalyst for toluene and quinoline hydrogenation. Green Chemistry, 2016; 18: 6082-6090. DOI 10.1039/C6GC01858K.

[34] Yu X.L., Nie R.F., Zhang H.F., Lu X.H., Zhou D. and Xia Q.H., Ordered mesoporous N-doped carbon supported Ru for selective adsorption and hydrogenation of quinoline. Microporous and Mesoporous Materials, 2018; 256: 10-17. DOI 10.1016/j.micromeso.2017.07.048.

[35] Tang W., Zhang H. and Yang X., Ru single atom catalyst with dual reaction sites for efficient Fenton-like degradation of organic contaminants. Applied Catalysis, B. Environmental, 2023; 320: 121952-121611. DOI 10.1016/j.apcatb.2022.121952.

[36] Sassykova L. and Aubakirov Y., Catalytic hydrogenation of gasoline fractions under elevated pressure. Chiang Mai Journal of Science, 2018; 45(1): 474-483.

[37] Zuo Y.H., Li R., Hua J.F., Hao S.Y., Xie J., Xu L.X., et al., Preparation of Co_0.5Cu_0.5/CNR catalyst and its performance in hydrogen production by hydrolysis of ammonia borane. Journal of Fuel Chemistry and Technology, 2024; 52: 1307-1317. DOI 10.1016/S1872-5813(24)60442-1.

[38] Alvarado Y., Busolo M. and Lopez-Linares F., Regioselective homogeneous hydrogenation of quinoline by use of pyrazolylborate ligands and transition metal complexes as precatalysts. Journal of Molecular Catalysis A: Chemical, 1999; 142: 163-175. DOI 10.1016/s1381-1169(98)00297-0.