Development of a new continuous process for the production of 3,5-dimethylpiperidine

Tao Lin,Xiaoxun Ma

1 School of Chemical Engineering,Northwest University,Xi’an 710069,China

2 Kaili Catalyst &New Materials Co.,Ltd.,Xi’an 710201,China

3 Chemical Engineering Research Center of the Ministry of Education (MOE) for Advanced Use Technology of Shanbei Energy,Xi’an 710069,China

4 Shaanxi Research Center of Engineering Technology for Clean Coal Conversion,Xi’an 710069,China

5 Collaborative Innovation Center for Development of Energy and Chemical Industry in Northern Shaanxi,Xi’an 710069,China

6 International Scientific and Technological Cooperation Base of the Ministry of Science and Technology (MOST) for Clean Utilization of Hydrocarbon Resources,Xi’an 710069,China

Keywords:Trickle-bed reactor (TBR)Kettle reactor (KR)Hydrogenation 3,5-Dimethylpyridine (DPY)3,5-Dimethylpiperidine (DPI)

ABSTRACT This paper developed a new clean continuous process for the hydrogenation of 3,5-dimethylpyridine(DPY) to 3,5-dimethylpiperidine (DPI) without solvent.A series of Ru/C catalysts were prepared by impregnation method,which were characterized by the BET,ICP,CO chemisorptions,XRD,SEM,EDS,TEM and TG.The effect of active species,loading,catalyst support,reaction temperature and pressure on the catalytic performance was investigated.The influence of internal and external diffusion in the trickle-bed reactor (TBR) was basically eliminated by adjusting the particle size and dosage of the Ru/C catalyst.The reaction performance of the hydrogenation of DPY to DPI in the TBR and kettle reactor(KR) was compared,and the superiority of the TBR process was analyzed.The results show that this new continuous process developed in this study is an efficient way to realize the hydrogenation of DPY to DPI,and has a good industrial application prospect.

1.Introduction

3,5-Dimethylpiperidine (DPI),as a kind of piperidine compounds prepared by hydrogenation of 3,5-dimethylpyridine(DPY),is a very important fine chemical product used in pesticide field [1,2].In the past,the annual demand for DPI as a drug intermediate was about 200 t,mainly used for the production of insecticides and the tilmicosin.With the development of fine chemical industry,the application of DPI has been greatly expanded,including new environmentally friendly materials and the template of SSZ-39 molecular sieve,so the total annual market demand for DPI has been close to 10000 t [3–5].The tilmicosin is a kind of antibiotic used for animals or pets,while the SSZ-39 zeolite is used for the NOxreduction and the methanol-to-olefins reaction(Fig.1).At present,driven by downstream products,the DPI production in domestic can’t meet the demand of the market.The traditional hydrogenation of DPY to produce DPI is an intermittent KR process which has many defects such as long production cycle,need solvent for completely conversion,and low catalytic efficiency.This greatly limited the large-scale production of DPI [6,7].Therefore,the research on a green and efficient continuous reaction process for the hydrogenation of DPY to DPI has important practical significance.

A “white paper” created by the ACS GCI Pharmaceutical Roundtable in 2005 encouraged the conversion from batch to continuous processing as the most important priority to accomplish sustainable chemical production,and correlated studies were given the highest practical importance [8].Compared with the continuous process,the intermittent KR process is more suitable for the production process of fine chemicals in small quantities and in many varieties.The accumulation of products in the KR process may inhibit the catalytic activity,thus affecting the chemical balance of the reaction and hindering the reaction progress.Usually,in order to obtain a higher conversion,it is necessary to prolong the reaction time,add solvent,increase the reaction temperature and pressure[9].In the KR,the enhancement of mass transfer is limited by the stirring speed and the performance of pressure seals,and the heat transfer is limited by the internal surface area of the reactor.Therefore,it is difficult to achieve large-scale DPI production with traditional KR technology [10,11].Continuous production technology can overcome the defects of KR process,which is an important way to produce fine chemical products in the future [12].

Fig.1.A scheme of the hydrogenation of DPY to DPI and the main application of the DPI.

Continuous trickle bed reactor (TBR) used for the three phase(gas,liquid and solid) reaction,has been successfully applied in the field of catalytic hydrogenation to produce fine chemicals,including the continuous hydrogenation of benzaldehyde,cinnamaldehyde,butylactone,benzene,toluene,nitrobenzene,glucose,sorbitol,acetophenone and glycerin [13–18],improved the safety and stability of production [19–21].The continuous hydrogenation of pyridines to piperidines and the hydrogenation of DPY to DPI in the TBR have not been reported.

Catalysts and technologies for the production of DPI(and other piperidine compounds)by catalytic hydrogenation have been studied extensively,and most of them are batch or kettle process with harsh reaction conditions and solvents[22].Zhou reported a batch technique for the production of DPI via hydrogenation(carbon supported noble metal catalysts)[23].The reaction temperature,pressure and time are 130°C,9 MPa and 24 h,respectively,and the DPY conversion should be improved.Xue et al.investigated the catalytic performance of 5% Ru/C catalyst on hydrogenation of Nheterocycles in KR.Compared with other pyridine compounds,the DPY is not easy to be fully hydrogenated even though enhanced the reaction conditions(multiply the usage of the catalyst and solvent) [24,25].Some patents revealed a DPI preparation method(kettle process),the catalysts including 5% Ru/Al2O3,5% Rh/Al2O3and Ni-Ru-Rh composite,where high temperature,high pressure and solvents are unavoidable [26–28].

Aiming at the deficiency of the existing technology,a continuous TBR was developed for the hydrogenation of DPY to DPI[2,29].The effects of reaction temperature,pressure,particle size and dosage of the catalyst on the hydrogenation performance of DPY in TBR were investigated,and a long period of catalyst stability experiment was carried out.At the same time,the reaction performance of TBR and KR for hydrogenation of DPY to DPI was compared.

2.Experimental

2.1.Catalyst preparation

In this study,commercial catalyst support such as γ-Al2O3,SiO2,HZSM-5 molecular sieve,coal based carbon,wood based carbon and coconut shell carbon were used as catalyst support.

Active carbon(AC)was treated by refluxing in diluted HNO3(6%(mass)) at 100 °C for 2 h.Then,the mixture was washed until the pH is higher than 3,and dried to obtain the catalyst support for use.These processing methods made AC easily re-disperse in polar solvent,which could be attributed to the introduction of oxygenated surface groups [30].

Ru/C catalysts with metal loadings of 1%,2%,3% and 4% (mass)were prepared on the acid treated carbon support by the impregnation method.The typical method for the preparation of Ru/C can be described as follows.10 g of carbon was impregnated with a certain amount of ruthenium chloride hydrate which was dissolved in 16 g hydrochloric acid(1.0 mol·L-1)solution.After being dried at 120 °C for 12 h,the obtained material was reduced at 300 °C for 2 h in H2(80 ml·min-1).3% Pd/C,3% Pt/C,3% Au/C,3%Ag/C,3% Rh/C and 3% Ni/C were prepared by the same method and used the same carbon support.3% Ru/γ-Al2O3,3% Ru/SiO2and 3% Ru/ZSM-5 were also prepared by the same method.All reagents used in this study were of chromatographic standard.

2.2.Catalytic tests

The catalyst performance was evaluated by TBR and KR,respectively.

(1) Evaluation process in TBR:In a typical experiment,4.0 g catalyst(mean size of 1241 μm,725 μm or 512 μm)was loaded in TBR (15 mm i.d.) [31].N2(60 ml·min-1) was used to replace the air in the reaction system for 10 min,and then the N2flow was switched to pure H2.The catalyst bed was heated up to the reaction temperature (140 °C,1 °C·min-1),the reaction pressure was 1.0–4.0 MPa.The DPY was fed by a syringe pump (Lab-alliance Series-III) to the catalyst bed.The mass hour space velocity (MHSV) was 0.4–3.4 h-1and the mol rate of H2to DPY is 5:1.

(2) Evaluation process in KR:The hydrogenation was carried out in 300 ml autoclave (equipped with a Teflon liner) with motor stirrer.4.0 g catalyst (mean size of 75 μm) and 80 g DPY (with 20 g H2O if needed) were added,the reactor was flushed 3 times with N2and H2,respectively.The mixture solution was stirred (800 r·min-1) at 140 °C under a reaction pressure of 4.0 MPa,sampling at set intervals.After reaction,the catalyst was recovered by centrifugation,and the clear supernatant liquid was decanted carefully.The catalyst was washed 3 times with ethanol,vacuum dried at 60 °C overnight,then used for subsequent recycle under the same reaction conditions.

The reactants and products were analyzed by online gas chromatography (GC-2014) equipped with a flame ionization detector(FID),and SE-54 column,30 m × 0.32 mm × 0.5 μm.GC analysis conditions:Both injector and detector temperature are 240 °C,the initial oven temperature was 60°C for 3 min,then was elevated to 200°C at a rate of 20°C·min-1,and retained at 200°C for 10 min.The carrier gas was N2(0.8 ml·min-1).Products were identified by using GC–MS(Agilent 6890)analysis and the GC–MS analysis conditions are similarly to GC measurements,where the carrier gas was He (0.5 ml·min-1).

In order to compare the reaction performance of the same catalyst in TBR and KR,the same reaction parameters were strictly controlled and the test was carried out.The catalytic performance was compared by the DPY conversion at the different FPH in the TBR and KR.The FPH is the DPY flux per gram of catalyst per hour(FPH,g·(g·h)-1).In the TBR,the FPH was changed by varying the MHSV,while in the KR it is regulated by sampling time.

Conversion of the DPY(X)was calculated on the basis of the following equation:

The superficial velocity (m·s-1) of the DPY (Vl) and the H2(Vg)are calculated according to the following equations:

where WHSV is the mass hour space velocity of the reaction (h-1),m is quality catalyst(g),ρ is the density of DPY (0.939 g·ml-1),d is the inner diameter(15 mm)of the TBR,M is the molecular weight of DPY (107.15 g·mol-1),Vmis the molar volume (22.4 L·mol-1).

2.3.Catalyst characterization

The BET surface areas of the catalyst samples were calculated from N2adsorption–desorption data acquired on a Gemini VII2390(Micromeritics)at liquid N2temperature.The specific surface areas were determined according to the Brunauer-Emmett-Teller (BET) method,total pore volumes were obtained from the volume of nitrogen adsorbed at a relative pressure of 0.99.

CO chemisorptions measurements of prepared Ru supported catalysts were carried out using an Auto Chem 2920 instrument.In a typical experiment,the catalyst sample (100 mg) was prereduced under H2/Ar (10:90,50 ml·min-1) at 300 °C for 2 h and flushed subsequently in He flow for 1 h.After cooled to room temperature,CO was injected at regular intervals until there was no more adsorption by catalyst.Ru dispersion and average particle size were calculated assuming the stoichiometric factor for CO to Ru as 1,as mentioned elsewhere [32].

Powder X-ray diffraction (XRD) patterns of the catalysts were recorded on a Miniflex (Rigaku Corporation,Japan) X-ray diffractometer using Ni filtered Cu Kα radiation (λ=0.15406 nm) with a scan speed of 5 (°)·min-1and a scan range of 10°–60° at 30 kV and 15 mA.

ICP (Perkin Elmer,Optima 2000DV) was used to determine the actual metal content of all samples with a pretreatment of microwave digestion in mixed acid.The content of Ru in the Ru/C catalyst was analyzed as follows:Ru/C catalyst (100 mg) was added in 9 ml HNO3(68%(mass))and 3 ml hydrofluoric acid,the mixture was oxygenolysis in Microwave Digestion System (Milestone ETHOS UP) at 210 °C for 40 min.After filtration,the content of Ru in the filtrate was analyzed by ICP standard curve method.The content of Ru in the solution was analyzed as follows:Activated carbon(dosage of 1%(mass)of the solution mass)was added into the solution,stirred at 50 °C for 40 min,then filtered and dried.Finally,the detection was performed according to the above oxygenolysis method in Microwave Digestion System.

Field-emission Scanning electron microscopic (FESEM) images were obtained on a Quanta 250 S (FEG),EDS was used to analyze the elements present in the different features observed in the SEM images.Transmission electron microscopy (TEM) study was carried out on a JEOL JEM-200CX microscope operated at 200 kV.

Thermogravimetric analysis (TG) was performed on a TGA/SDTA851E instrument under N2atmosphere from 40 °C to 800 °C.

3.Results and Discussion

3.1.Effect of the metallic catalyst

The results of the catalytic performance of carbon supported different metallic catalysts for the hydrogenation of DPY to DPI in TBR and KR are given in Table 1.Blank experiments showed that the hydrogenation reaction could not proceed without catalyst(Table 1,entry 1) or with carbon as the catalyst (Table 1,entry 2),indicating the catalytic activity of metallic catalysts.Under the same reaction conditions in TBR,the Pd/C,Pt/C,Au/C and Ag/C showed lower catalytic activity.The catalytic performance of the Rh/C,Ru/C and Ni/C in TBR is higher than that of in KR,and the comparison suggests that the catalytic activity of these three catalysts in both TBR and KR was in the same order:Rh/C ＞Ru/C＞Ni/C.The supported Rh catalyst exhibited the highest catalytic performance for the hydrogenation of the DPY.Rhodium,as a kind of precious metal,is very expensive and the price of that is more than 40 times higher than Ru(Table 2).However,the DPY conversion of the Rh/C catalyst is only 1.2 times higher than that of Ru/C catalyst which can be completely converted DPY to DPI when slightly reduce the WHSV to 0.4 h-1.So it would be much more economical and practical to use Ru/C instead of Rh/C.In the KR,increasing reaction pressure(from 2.0 MPa to 4.0 MPa),adding solvent (water) and extending the reaction time to 8 h helped to improve the DPY conversion (from 11.2% to 36.0%,from 36.0% to 97.5% and from 97.5% to 100%,respectively).Therefore,the catalytic efficiency in KR is lower,both the environmental pollution caused by the separation of solvents and safety risks caused by the high reaction pressure are inevitable.The TBR is more suitable for the hydrogenation of DPY to DPI,the advantages of which is solvent-free,efficient and safe.

Table 1 Comparison of the catalytic performance of catalysts with different metal components in TBR and KR

Table 2 Price list of Ru and Rh in recent years

3.2.Effect of the supports

The catalyst supports play a key role in the catalyst behavior and especially in the hydrogenation of aromatic or heterocyclic compounds.The Ru catalysts supported by γ-Al2O3,SiO2,HZSM-5(Si/Al=50) and activated carbon (coal based carbon,woodiness carbon and coconut shell carbon) were prepared by the same impregnation method with 3% Ru content.The catalytic performances and some related physical and chemical adsorption characterization results are shown in Table 3.It can be seen from Table 3,the DPY conversion of the same catalyst in TBR is higher than that of in KR,and the reaction pressure is lower,indicating the reaction conditions in TBR are milder.The catalytic activities of the catalysts prepared by activated carbon are higher than that of γ-Al2O3,HZSM-5 and SiO2.The excellent performance of Ru/C catalysts is presumably attributed to the homogeneously dispersed Ru NPs distributed on the carbon support,which provides numerous active sites for the hydrogenation of DPY.It is found that the Ru dispersion and the chemical sates of Ru/C catalysts determine the highly catalytic activity,so activated carbon is an ideal support for Ru catalysts.The catalyst prepared by woodiness carbon showed relatively higher activity,but the particle strength of coal based carbon is stronger,so it is more suitable for long time using in the TBR.The higher Ru dispersion of the catalyst,the higher the catalytic activity,but the specific surface area of the catalyst is not completely consistent with its activity.Although the specific surface area and the Ru dispersion are the main factors affecting the catalyst activity which is still influenced by many other factors,such as the acidity and alkalinity of the catalyst,pore structure and so on [33,34].

3.3.Influence of the Ru loading

The effect of Ru loading on the hydrogenation of DPY in TBR was examined in Table 4.Four Ru/C catalysts with different Ru loading(1% (mass),2% (mass),3% (mass) and 4% (mass) were prepared by the same method.The DPY conversion increased with Ru content,which was reached to 100%after more than 3%(mass)Ru loading.Only when the cis-trans ratio of DPI is 85%to 15%can it be used as raw material for the production of other high value-added products directly [3,5].The ratio of cis-trans products decreased with the growth of Ru loading from 3% (mass) to 4% (mass),which may be due to the high activity of the catalyst leading to the increase of the content of unstable trans products.Therefore,the Ru/C catalyst with 3% Ru loading is the most appropriate.The research on the ratio of cis-trans product is carried out in the follow-up work.

3.4.Internal and external diffusion elimination

In the TBR,the hydrogenation reaction of DPY on the Ru/C catalyst is the three-phase reaction (gas–liquid-solid),the catalyst may significantly affected by internal and external diffusion factors.The severity of the influence of internal and external diffusion factors in the TBR needs to be verified by experiments.Therefore,it is necessary to change the particle size and the dosage of catalyst to eliminate the influence of internal and external diffusion [35–37].Since hydrogenation of DPY to DPI is an exothermic reaction,in order to reduce the influence of the temperature build-up in the catalyst bed on the catalytic performance,the catalyst support(carbon) of the same particle size was used to dilute the catalyst.The result showed that the catalyst support is inactive (Table 1).The dilution ratio of the catalyst was investigated.The temperature build-up in the catalyst bed is 9 °C higher than the required value without diluting catalyst.After more than three times catalyst supports dilute the catalyst,the temperature build-up of catalyst bed can be controlled within 1 °C.As a result,when the catalyst is diluted by the catalyst support in 3 times,the influence of the temperature build-up of the catalyst bed can be ignored [20].In the industrial TBR,the reaction heat generated by the hydrogenation of DPY can be indirect removed through the thermal medium(heat conducting oil or water vapor).

In the elimination of external diffusion experiment,the effect of catalyst dosages on the DPY conversion at the same WHSV(0.5 h-1) and two reaction temperatures (140 °C and 150 °C) was studied (Table 5).As listed in Table 5,when 2.0 g catalyst was loaded,the superficial velocity of DPY (Vl) and H2(Vg) was 1.67 × 10-6m·s-1and 1.64 × 10-3m·s-1,respectively,the corresponding DPY conversion is 67.6% at 140 °C and 85.7% at 150 °C.After the catalyst dosage increased to 4.0 g,the WHSV (0.5 h-1)was fixed,but the superficial velocity (Vland Vg) has doubled,and the DPY conversion growth to 75.5% at 140 °C and 92.0% at 150 °C,indicating the existence influence of external diffusion.When the catalyst dosage was further raised to 8.0 g,the superficial velocity of DPY(Vl)and H2(Vg)increased to 6.70×10-6m·s-1and 6.58×10-3m·s-1,respectively,the DPY conversion(at 140°C and 150 °C) did not increased significantly,indicating that theexternal diffusion was basically eliminated.Therefore,the catalyst dosage of 4.0 g was selected.

Table 3 Comparison of the catalytic performance of the Ru catalysts with different supports in TBR and KR

Table 4 Influence of Ru loading on the hydrogenation of the DPY

Table 5 Influence of external diffusion on the 3% Ru/C catalyst

The effect of internal diffusion on the catalytic reaction is related to the particle size of the catalyst.Therefore,the 3%(mass)Ru/C catalyst was crushed and screened to obtain three particle sizes of 425–600 μm (average 512 μm),600–850 μm (average 735 μm) and 850–1632 (average 1241 μm),respectively.The catalytic performance was compared in the same reaction conditions and the results are shown in Table 6.It can be seen from Table 6,with the increase of the average particle size,the DPY conversion decreased.But the decrease is not obviously,especially when the particle size is smaller than 725 μm.The results showed that the less influence of internal diffusion which can be basically ignored when the particle size is smaller than 725 μm [20,21].Therefore,the catalysts with an average particle size of 725 μm were used in subsequent experiments.

Table 6 Influence of internal diffusion on the 3% Ru/C catalyst

3.5.Reaction performance comparison in the TBR and KR

The reaction performance of TBR and KR was compared by the DPY conversion at the same FPH (DPY flux per gram of catalyst per hour,g·(g·h)-1),catalyst dosage and reaction conditions.As shown in Fig.2,the conversion curves of DPY in the two reactors are almost identical when the FPH is under 0.7 h-1.As the FPH increases,the DPY conversion in both reactors significantly decreased,but it is worth noting that the DPY conversion in KR decreased faster than that of TBR,and the gap between them increased greatly with growth of the FPH.The result showed that the TBR can obtain higher DPY conversion and processing capacity at the same FPH.Since the catalyst and the reaction conditions(temperature,pressure and FPH) are exactly the same in the comparison experiment,a widening gap between the DPY conversions in the two reactors should be attributed to the back mixing in KR.Because the KR is close to full back mixing,as the reaction goes on,the concentration of the substrate (DPY) and the product (DPI)gradually decreases and increases,respectively,so the hydrogenation reaction rate decreases,the higher DPY conversion requires a longer reaction time.In TBR,the DPY passes through the catalyst bed in the form of liquid film,and hydrogenation reaction with hydrogen on the surface of the catalyst,which is similar to horizontal push flow without back mixing,so the reaction efficiency is higher.Therefore,in the case of high FPH,the TBR has obvious advantages.

Fig.2.The comparison of reaction performances between KR and TBR on 3% Ru/C catalyst (■) TBR;4.0 g catalyst,140 °C and 4.0 MPa,WHSV=0.4–3.4 h-1 (●) KR;140 °C,Sampling time=6–42 h,4.0 MPa,4.0 g catalyst,80 g DPY.

3.6.Effect of reaction temperature,pressure and WHSV

The effect of the reaction temperature on the DPY concentration using 3%Ru/C catalyst was examine by varying temperature 70 °C to 200 °C at constant hydrogen pressure of 1.0 MPa.As shown in Fig.3(a),the DPY conversion increased with the temperature,which can be divided into three stages:slowly increase stage(70–100 °C),rapid increase stage (110–150 °C),and stable stage(150–200°C).In Tables 5 and 6,the catalyst with different dosage and size,respectively,the DPY conversion increased with the temperature(140–150°C)belongs to the rapid increase stage,in which the temperature has a great influence on the conversion.

The effect of the reaction pressure on the DPY concentration using 3% Ru/C catalyst was examine by varying pressure 0.5 MPa to 4.0 MPa at constant hydrogen temperature of 140 °C.As shown in Fig.3(b),The DPY conversion increased rapidly when the reaction pressure increased from 0.5 MPa to 1.0 MPa,which growth slowly from 1.0 MPa to 3.0 MPa and basically stable to 100% at higher pressure (3.0–4.0 MPa).For effectively comparing the performance and activity of different catalysts,the lower reaction pressure (1.0 MPa) was selected in the experiments.In addition,the lower pressure is also beneficial to the safety and cost of the industrial TBR equipment.

The FPH in Fig.2 is equivalent to the WHSV,with the increased of WHSV,the DPY conversion decreased gradually at the reaction of 4.0 g catalyst,140 °C and 4.0 MPa.

3.7.Stability of the catalyst

The stability test results of the catalyst in the TBR are depicted in Fig.4.The DPY conversion was 100% at a WHSV of 0.4 h-1,140 °C,1.0 MPa.After 264 h,the WHSV was raised to 0.5 h-1and the corresponding DPY conversion was about 75%.At 840 h,with the temperature increased to 150 °C,the DPY conversion raised to 92%.The reaction conditions were restored to the initial level(0.4 h-1,140°C)at 1416 h,the DPY conversion returned back to 100%.The results showed that the Ru/C catalyst performance was not significant change in a long period experiment of 1536 h.

Fig.3.Effect of reaction temperature (a) and pressure (b) on DPY hydrogenation over 3% Ru/C catalyst (a):1.0 MPa,WHSV=0.5 h-1;(b):140 °C,WHSV=0.5 h-1.

The 3 %Ru/C catalyst has good stability in TBR (Fig.4),which was very critical for the industrial application.As can be seen from Fig.5,the Ru/C catalyst in KR exhibited considerable activity decay in the reuse process.The DPY conversion declined dramatically to 11.6%from 36.0%after five catalytic cycles.To explore the cause of the deactivation in KR,the used Ru/C catalysts (after five cycles in KR and after 100 h reaction in TBR) are carefully characterized by BET,ICP,SEM,EDS,TG TEM and XRD,the results are described below.

The actual loading of the Ru is determined by ICP and the results are given in Table 7.The Ru content in the fresh catalyst is 2.98% (mass),while in the TBR and KR used catalysts are 2.95%(mass) and 1.83% (mass),respectively.The liquid material after the reaction was collected and detected the Ru content by ICP.The results show that the Ru element in the liquid product of TBR almost undetectable (0.4 mg·L-1),while in the liquid product of KR,the Ru content was 67.4 mg·L-1.Indicating the Ru was seriously leached in the course of the KR process,and mechanical agitation may have caused the loss of Ru from the catalyst [38,39].

Fig.4.Time-on-stream study over the 3%Ru/C catalyst for the hydrogenation of the DPY Reaction conditions:140 °C,1.0 MPa,6.0 g catalyst,WHSV=0.4 h-1 (0–264 h and 1416–1536 h);140 °C,1.0 MPa,6.0 g catalyst,WHSV=0.5 h-1 (264–840 h);150 °C,1.0 MPa,6.0 g catalyst,WHSV=0.5 h-1 (840–1416 h).

As listed in Table 7,compared with the fresh Ru/C catalyst,the specific surface area of the used catalysts of TBR and KR decreased remarkably (from 743.6 m2·g-1to 438.2 m2·g-1and 492.2 m2·g-1,respectively),while the total pore volume also obviously declined.The results suggest that the catalysts were partly blocked after reaction.The strong adsorption of trace amounts of DPY and DPI on the catalysts is not easy to be washed by ethanol,which may cover the Ru active sites [40].Although the specific surface area and pore volume decreased,the catalyst activity remained stable in the TBR,so this may not be the main reason for the decrease of catalyst activity [41].For the further explanation,SEM-EDS,TG,XRD and TG tests were conducted.

Table 7 Textural properties of fresh and used Ru/C catalysts

SEM images (Fig.6(a)-(c)) show the formation of surface morphologies and appearance of fresh and used 3% Ru/C catalysts.Fig.6(a) shows SEM image of some planar structure with the dimensions～10 μm may correspond to the ordered natural graphite structures [42].It can be seen that the surface of the used samples(Fig.6(b)and(c))are seems to be covered by some materials,while the fresh sample looks very clear [43,44].However,there were nanoscale pore structures and grooves on the surface of the used catalysts for mass transfer,so the partial coverage of the catalyst surface may not be the main reason for the degradation of catalyst performance.

Fig.5.Reuse of the catalyst in the KR Reaction conditions:140 °C,6 h,4.0 MPa,4.0 g catalyst,80 g DPY.

Fig.6.SEM images and EDS spectrum of fresh and used Ru/C catalysts (a) Fresh;(b) KR 5 cycle used;(c) TBR 100 h used.

Fig.7.The TG-DTA results of fresh and used catalysts.

Fig.8.XRD patterns of fresh and used catalysts.

Fig.9.(a),(b) and (c) are TEM images of the fresh,KR and TBR used Ru/C catalyst,respectively.(d),(e) and (f) are the corresponding metal particle size distribution histograms.

The results obtained from EDS for fresh and used Ru/C catalysts are presented in Table 7 and Fig.6(d).The Ru content detected by EDS is slightly higher than theoretical value(3%(mass)),because of the analytical method by EDS is considered to be semi-quantitative[45,46].The surface C element increased after reaction,probably because some materials adsorbed or covered the surface of the catalyst,which were hard to wipe off.This consideration is reinforced by the results of BET where some materials contaminated the catalyst and reduced the specific surface area after reaction [47].

TG and DTA curves of the fresh and used catalysts are given in Fig.7.It has lost water or ethanol content up to 150°C in the fresh and used catalysts.A new weight loss between 150 °C and 450 °C occurred in the two used catalysts can be attributed to desorption and decomposition of adsorbed materials [48,49].The results showed that relatively more materials were adsorbed on the TBR used catalyst (the red line in Fig.7),which was consistent with the results of BET and EDS.

The XRD patterns of the fresh and the TBR and KR used catalysts are shown in Fig.8.There was no significant change in the XRD patterns of the fresh and the used catalysts.Only two broad peaks correspond to the carbon with a certain degree of graphitization around 2θ=25° and 44° were observed (JCPDS-ICDD card No.41-1487) [50,51].XRD characteristic peaks for metallic Ru0(JCPDS-ICDD card No.06-0663) are not detected,suggesting the Ru is well dispersed or exist in an amorphous form on the carbon support,and the absence of large Ru NPs after reaction[50–56].In addition,due to the mild reaction conditions,the metal sintering may not occur during the liquid-phase hydrogenation reaction.

The TEM images (Fig.9) confirmed that the Ru metal particles were well dispersed on the fresh and used samples.The Ru size on the used catalysts was not obvious aggregation or sintering.The majority of Ru particles were between 6 nm and 20 nm,the distribution is wide and the average particle size is around 11.5 nm.The catalyst performance in the TBR was stable,which was decreased in the KR.Since the Ru size on the catalyst was almost unchanged before and after the reaction,even a small amount of Ru aggregation or sintering is not the main reasons for the degradation of catalyst performance [57,58].

In summary,the characterization results of BET and TG showed that the TBR and KR used catalysts had different degree of coverage.The XRD and TEM indicated that the structure of the catalysts and the size of Ru nanoparticles did not significantly change.In the TBR,the Ru content is basically unchanged after 100 h reaction(2.95% (mass)),and the catalyst performance is stable,indicating the coverage has little effect on the performance of the catalyst.In the KR,the Ru content on the used catalyst decreased significantly after 5 catalytic cycles (1.83% (mass)),demonstrating the Ru leaching may caused by stirring,this may be the main reason for the degradation of catalyst performance in KR.

4.Conclusions

In this study,a new TBR continuous production process of catalytic hydrogenation of DPY to produce DPI was developed.This process has the characteristics of high catalytic efficiency,green reaction process and stable catalytic performance.

(1) The carbon supported Ru catalyst is an effective catalyst for the hydrogenation of DPY to DPI.

(2) In the TBR,the Ru/C catalysts with 3%Ru loading is the most appropriate.When the catalyst average particle size is 725 μm and the catalyst dosage is 4.0 g,the influence of internal and external diffusion effect on catalyst properties in the hydrogenation can be basically ignored.

(3) The FPH was used to compare the reaction performance of the TBR and the KR at the same conditions.The results indicate that the back mixing is the main reason for the low catalytic efficiency in KR,while the TBR is close to horizontal push flow reactor without back mixing,so the TBR has more advantages in the large-scale hydrogenation of DPY.

(4) After the reaction,both the specific surface area and pore volume of the used catalysts were decreased,indicating that the surface of the catalyst was covered by some materials,but there was no obvious effect on the catalyst performance.The main reason for the decline of catalyst activity may be related to the Ru leaching caused by mechanical stirring in KR.

(5) The experimental results and theoretical analysis show that the TBR has the advantages of higher DPY conversion,stable catalyst performance,milder reaction conditions and largescale DPY hydrogenation capacity,which has a good industrial application prospect.

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 word was supported by the National Natural Science Foundation of China (21536009),and the Science and Technology Plan Projects of Shaanxi Province (2017ZDCXL-GY-10-03).