Orderedmesoporous ruthenium containing carbon (Ru-OMC) catalystwith semi-embedded uniformRu particledistributionwas synthesized by using RuCl3/SBA-15 as a hard template. The sucrose both acts as the carbon pre-cursor and Ru3+ stabilizer. The use of RuCl3/SBA-15 makes more exposed Ru surface available for reactants. Theturnover frequency of the Ru-OMC catalyst for hydrogenation of benzene reaches ca. 35,000 h−1 at 4 MPa,110 °C, which is 12 times improvement compared with that of Ru/OMC catalyst prepared by impregnation.The Ru-OMC catalyst can be recycled more than 9 times without much loss of performance and aggregation ofRu nanoparticles.
Carbon supports offermany advantages such asmechanical stability,high surface area, ideal porosity, abundant and adjustable surface func-tional groups which can be used as anchoring sites for noble metal par-ticles [1]. Impregnation is a normally used method to support noblemetal catalysts into carbonmaterials [2]. However, it can be challengingto synthesize nanoscale metal particles less than 3 nm in diameter onporous supports by impregnation because nanoparticles easily undergosintering through certain thermal treatments, such as calcination andreduction [3].
Recently, the preparation of confined noble metal catalyst has been ahot topic with new mesoporous materials appearing. Improvement inthe catalytic properties is derived from the confinement concept resultedfrom imprisonment of the substrate within the pores of the support,which leads to improved interactions between the active catalyst andthe substrate [4]. For example, Ji et al. [5] reported that Ru-containing or-dered mesoporous carbon (OMC) through an evaporation-inducedmulti-constituent co-assembly method exhibits good catalytic activityfor benzene hydrogenation. Zhao et al. [6,7] reported that Ru nanoparti-cles can be semi-embedded in a matrix of porous carbon by a chemicalvapor deposition (CVD) method and that these materials display highcatalytic activity. They also reported a thermal reduction method toprepare carbon supported Ru catalyst and that OMC can act both as
: +86 571 88320259.
rights reserved.
the support and the reducing agent for ruthenium nanoparticles, andthe intimate interfacial contact between the Ru nanoparticles and thecarbon support was believed to be responsible for the remarkablyhigh activity in the hydrogenation of benzene and toluene [8]. Liu etal. reported a controlled synthesis of highly dispersed platinum and sta-ble PtRu nanoparticles in OMCs by dispersing platinum acetylacetonatein the furfuryl alcohol and trimethylbenzene as the co-feeding carbonand Pt precursor [9,10]. Xiong et al. reported a method by using furfurylalcohol solution containing Ru(NO)(NO3)3 for the preparation of stableRu nanoparticles embedded on the OMC material for applications in Fi-scher–Tropsch synthesis [11]. Both of them use a mixture of noblemetal precursor in organic carbon source to prepare noblemetal embed-ded carbon catalysts. Although themesostructures of metal-carbon cata-lysts is uniform by using the mesostructured silica hard templates, mostof themetal particles are buried in the carbon framework due to the pre-mixing of metal ions with carbon precursors. The deeply buried status ofmetal particles caused problems such as less contact metal surface withreactants during reaction. Recently, Liu et al. [12] described a stepwisemethod for the accurately controlled growth of Pt nanoparticles sup-ported on ordered mesoporous carbons (Pt-OMC) by the nanocastingof carbon and metal precursors in the pore channels of mesoporous sil-icas functionalized with Si–H groups.
In the present work, we reported a one-pot method for prepara-tion of Ru-OMC catalyst with more exposed active sites by usingRuCl3/SBA-15 as a hard template. The catalytic performance of Ru-OMC catalysts was evaluated in the hydrogenation of aromatics,which are important industrial transformations and model hydroge-nation reaction for evaluation of supported noble metal catalysts[13,14].
30 Y. Li et al. / Catalysis Communications 20 (2012) 29–35
2. Experimental
2.1. Preparation of the materials
Mesoporous pure-silica SBA-15 was synthesized according to Zhaoet al. [15] Ru-OMC was synthesized by using sucrose as the carbon pre-cursor, and RuCl3 as Ru source. A detailed preparation of the Ru-OMCcatalyst was given in supporting information. The supported rutheniumcatalyst onOMC (donated asRu/OMC) andSBA-15 silica (donated as Ru/SBA-15)were prepared by awet impregnationmethod. The preparationfor the above samples are summarized in Fig. 1.
2.2. Measurement of catalytic activities
The evaluation of the catalytic properties of the Ru catalysts forbenzene hydrogenation was done in a 100 mL stainless-steel stirredpressure reactor. A given amount of Ru catalyst (0.085 g) and 30 mL ofbenzene (>99.5%, Aldrich)were placed in the reactor. The reaction pres-sure was generated by using H2 at the reaction temperature and keptconstant. The reaction was stopped until no uptake of H2 was observed.The productwas analyzed by using an Agilent 7890A gas chromatographequippedwith a DB-1 capillary columnwith a flame-ionization detector.To investigate the recyclability of the catalyst, the used Ru catalyst wasfiltered after the reaction and vacuum-dried at 80 °C overnight beforethe next reaction run.
2.3. Characterizations
Powder X-ray diffraction (XRD) patterns were recorded on a RigakuD/Max-2500/pcpowder diffraction systemusing CuKα radiation (40 kVand 100mA) over the range 0.5°≤2θ≤10° (low angle) and 10°≤2θ≤80° (high angle). Nitrogen adsorption isotherms were determinedat−196 °C on aMicromeritics ASAP 2020 system in staticmeasurementmode. The samples were outgassed at 350 °C for 10 h before adsorptionmeasurement. The Brunauer–Emmett–Teller (BET)methodwas used tocalculate the specific surface area. High resolution transmission electronmicroscopy (HRTEM) and scanning transmission electron microscope(STEM) images of the sampleswere obtained by a FEI Tecnai G20 instru-ment. Energy-dispersive X-ray spectroscopy (EDS) and inductively
SBA-15
Sucrose
H2 reduction
Ru/SBA-15
Impregnationmethod
RuCl3·H2O
OMC
Sucrose
Carbonization
Silicaremoval
Su:Ru-
Ru3+
Ru3+
Ru3+Ru3+
Ru3+
Ru3+
Ru3+
Ru3+
Ru3+
Ru3+Ru3+
Su
Su Su
SuSu
Su
Su S
Su
Su
Su
Su
Su
SuSu
SuSu
SuSu
Su
Su
SuSu
RuCl3·H2O
Fig. 1. The preparation for Ru-OM
coupled plasma plasma emission spectrometry mass spectrometry(ICP-MS) was conducted for the analysis of sample composition. Thedispersion of Ru was obtained by CO chemisorption method, whichwas carried out at 40 °C on a Quantachrome Autosorb-1/C chemisorbapparatus. Prior to measurements, the pre-reduced catalysts were re-duced in situ for 2 h at 450 °C in H2. The metal dispersion and particlesize were estimated based on assumption of a spherical geometry ofthe particles and an adsorption stoichiometry of one CO molecule onone Ru surface atom.
3. Results and discussion
3.1. Texture properties
The low-angle XRD patterns of various Ru catalysts and supportsare given in Fig. 2-a. A remarkable diffraction peak at 2θ=0.5–1.0°is observed for Ru/SBA-15, Ru-OMC, OMC and SBA-15, indicatingthe ordered mesoporous structures of these samples. The existenceof the Ru metal leads to decreasing of the intensity of the characteris-tic peaks of the SBA-15 and shrinkage of cell parameter (Table 1). Thisshould be attributed to the pore-filling effects that can reduce thescattering contrast between the pores and the framework of theSBA-15. Similar results have been reported previously [16,17]. In con-trast, this peak disappears for Ru/OMC, implying the Ru metal parti-cles is blocking the pore or the ordered mesostructure is somewhatdestroyed.
The porosity of the SBA-15, Ru/SBA-15, OMC, Ru-OMC, Ru/OMC andRu-Carbon was measured by N2 sorption. Fig. 3 gives the isotherms (a)and pore size distributions (b) of various samples. For the sample SBA-15 and Ru/SBA-15, the isotherms are typical type-IV with an H1 hystere-sis loop, which is a typical characteristic for mesoporous materials with2D-hexagonal structure. The specific surface area of SBA-15 decreasesfrom 904 to 645 m2/g and pore size decreases from 6.4 to 5.6 nm (calcu-lated from the desorption branch of the isotherm) after Ru metal wassupported (Table 1). This indicates that the impregnation of Ru metalblocks parts of the pores of SBA-15 support. The isotherms of Ru-OMCandOMCare typical type-IVwith anH2hysteresis loop,which is a typicaladsorption for OMC materials with an ordered mesoporous structure[18]. Nevertheless, the isotherm of the Ru/OMC shows an obvious
Two-steps
Carbonization
Silicaremoval
Ru-OMC
Ru/OMC
H2 reduction
Sucrose(Su)x: Ru3+-sucrose
Semi-imbedded Ru NPs
Carbon
Silica
Supported Ru NPs
Su
Su Su
Su
Su
Su Su
Su
Su
Su
Su
Ru-(Su)x
Ru-(Su)x
Ru-(Su)x
Ru-(Su)x
Ru-(Su)x
SuSu
Ru 3+
Ru 3+
Ru 3+
Ru 3+
Ru 3+
Ru 3+Ru 3+
Ru 3+
Ru 3+
Ru 3+Ru
RuRu
Ru
Ru
RuRu
Ru
Ru
Ru
C, Ru/OMC and Ru/SBA-15.
1 2 3 4 5 6
Ru/SBA-15
Ru-OMCRu/OMC
OMC
SBA-15
Inte
nsity
2 Theta (degree)
a
10 20 30 40 50 60 70 80
&:silica
&
*:Ru$: graphtic carbon
$
***
*
* *
Ru-OMC
Ru/OMC
Inte
nsity
2 Theta (degree)
Ru/SBA-15
*
$
b
Fig. 2. Low-angle XRD patterns (a) of SBA-15, Ru/SBA-15, OMC, Ru-OMC, and Ru/OMCand high-angle XRD patterns (b) for Ru/OMC, Ru-OMC, and Ru/SBA-15.
0.0
Ru-Carbon
Ru-OMC
Ru/OMC
OMC
SBA-15
Ru/SBA-15
Relative Pressure (P/P )0
Vol
ads
orbe
d (c
c/g
ST
P)
a
0
Ru-Carbon
Ru-OMC
Ru/OMC
OMC
SBA-15
dV/d
D
Pore Diameter (nm)
Ru/SBA-15
b
0.2 0.4 0.6 0.8 1.0 10 20 30 40
Fig. 3. Nitrogen adsorption–desorption isotherms (a) and pore size distribution (b) forthe SBA-15, Ru/SBA-15, OMC, Ru/OMC, Ru-OMC and Ru-Carbon.
31Y. Li et al. / Catalysis Communications 20 (2012) 29–35
hysteresis of the H3 type at P/Po>0.8 as well, corresponding to capillarycondensation in the interparticle pores or some impurity phases, whichindicates that some larger pores appear and the pores of OMC are filledwith the Ru particles. The decrease in pore volume of Ru/OMC as com-paringwith that of OMC ismore probably the indication of pore blockage.The isothermal curve for Ru-Carbon is typical type-I without hysteresisloop, which indicates this sample only has micropores. This result con-firms the role of mesoporous SBA-15 templates in making of mesoporesfor OMC.
3.2. Status of the ruthenium nanoparticles
The status of ruthenium nanoparticles was characterized in detailby high angle XRD, HR-TEM, STEM, CO chemisorption, and XPS tech-niques (XPS results and related discussions are given in supportinginformation). The high-angle XRD patterns of various Ru catalystsare given in Fig. 2-b. No diffraction peak assigned to Ru metal is ob-served for Ru/SBA-15 and Ru-OMC, indicating the absence of big Runanoparticles on Ru-OMC and Ru/SBA-15. Well-resolved peaksassigned to Ru metal are observed for the Ru/OMC catalyst preparedby an impregnation method, indicating large Ru particles are formed.There are two broad peaks at 22° and 43° appear for both Ru/OMCand Ru-OMC samples which are corresponding to the (002) and(101) crystal planes from graphic structure of carbon supports,which indicates that the graphic structure is not well resolved [19].Thedistribution of Ru nanoparticles on different catalystswas character-ized by HRTEM, which is given in Fig. 4(a–f). The size distributions ofnanoparticles for various samples are given in supporting information(Figure S1). It can be observed that high dispersed Ru nanoparticlesare uniformly distributedwithin the support and no particle aggregation
Table 1Textural properties of SBA-15, OMC, and various Ru catalysts.
a The a0 value is calculated by a0=2×d100/ffiffiffi
3p
.b Calculated fromdesorption branch of isotherm according to the Barrett–Joyner–Halen
method.
is observed for Ru-OMC (Fig. 4a–c). From Figure S1, it can be seen thatthe particle size distribution for Ru-OMC is relatively narrow. About92% particles for Ru-OMC are in the range of 2.0–3.0 nm. However, theparticles on Ru/OMC are not as homogeneous as that of Ru-OMC. Thesize of Ru nanoparticles for Ru/OMC is around3–8 nmand small amountof large particles (>15 nm) also can be observed. This result is con-firmed by the high angle XRD results (Fig. 2–right) as discussed above.The particle size distribution for Ru/SBA-15 (Figure S1-f) is relativelynarrow and no big particle was observed, about 83% particles for Ru/SBA-15 are in the range of 6.0–8.0 nm and the filling of the SBA-15pores can be clearly observed.
Please note that the Ru-OMC was prepared by using as-preparedRuCl3/SBA-15 as a hard template. The pore of SBA-15 is filled withRu3+ before the infiltration of carbon precursor (sucrose). This ap-proach is different from references [3,8]. Zhao et al. [3] reported asandwich like Ru-OMC which is prepared by using benzene as carbonprecursor by a CVDmethod. In present approach, the sucrose has a di-rect interaction with Ru3+ ions at room temperature, the abundanthydroxyl groups can stabilize the Ru3+ ions. The carbonization isdone by a two-step method. The first stage of carbonization is thetreatment of the composite by H2SO4 at 160 °C for 6 h, which makescross linking of carbon structure before the reduction of Ru3+. Thesecond stage of carbonization was operated at 850 °C for 3 h underinert gas atmosphere. The reduction of Ru3+ and further carboniza-tion occur in this stage. The sintering of Ru nanoparticles can begreatly reduced with the confinement effect of carbon structures. Toidentify the stabilized role of sucrose, a Ru-Carbon was prepared bya direct carbonization of RuCl3-sucrose composite without usage ofSBA-15 as a template. Detailed preparation process was given in sup-porting information. The HRTEM image for Ru-Carbon is given inFig. 4d and particle size distribution is given in Figure S1. It can beseen that the particle size distribution for Ru-Carbon is similar tothat of Ru-OMC. About 90% particles Ru-Carbon are in the range of1.5–2.5 nm. From the data given in Table 2, the average particle sizeis 1.8 nm, which is even smaller than that of Ru-OMC (2.4 nm). Thisis maybe due the more efficient stabilization role of sucrose withoutexitance of SBA-15 template in the preparation. This result confirmsthe role of sucrose in making small and uniform Ru nanoparticles. Al-though the nanoparticles of Ru metal are homogeneously distributed,they cannot be exposed to reactants during reaction. Therefore, it canbe predicted that this sample is not suitable for catalytic applicationsalthough the particles are small and uniform.
From the images in Fig. 4 (a and c), it can be observed that most ofthe Ru particles are semi-embedded within the mesoporous carbonframework arising from the thermal reduction where the consumption
Fig. 4. HRTEM (a, c–f) and STEM (b) images of various catalysts: (a–c) Ru-OMC; (d) Ru-Carbon; (e) Ru/OMC; and (f) Ru/SBA-15. The bright dot in the STEM image and black dot inHRTEM is Ru as confirmed by EDS.
32 Y. Li et al. / Catalysis Communications 20 (2012) 29–35
of carbon in reducing the Ru3+ ions to Rumetals allows them to sit in orbecome semi-embedded in carbon matrix. This has been reported byZhao et al. [8]. These semi-embedded metal particles have excellentperformance in various reactions compared with traditional supportedmetal nanoparticles due to the deep interaction between the Ru nano-particles and the carbon support [11,12]. It should be noted that the pre-sent approach is also different from that reported in references [9–11]in which the carbon and the metal precursor were mixed togetherfirst, and they were impregnated into the pore of SBA-15 to makemetal-carbon catalysts. Therefore, the metals are easily fully buried inthe carbon framework during carbonization. In present approach, theRuCl3 was supported in the pore of SBA-15 templated, although partof Ru metal nanoparticles were buried in the carbon framework, mostof the Ru particles are still reachable after the silica was removed.What's more, it is reported that the mesoporous structure of OMC
[18–21] are interconnected by many micropores. This also facilitatesthe transport of reactants during reaction.
The dispersion and particle size of ruthenium nanoparticles deter-mined by CO chemisorption and comparied with that obained by TEMare given in Table 2. The level of CO chemisorption over the Ru-OMCis in the 176.0 μmol/gcat range. Based on the chemisorption data andassuming an adsorption of one CO (or H) per metal atom [22,23], onecan estimate the dispersion of the metal catalysts. The dispersion forRu-OMC by CO chemisorption is 74.1%, the highest among all the pre-sent catalysts. Such level of Ru dispersion is quite high. The averagesize of Ru particles calculated by CO-chemisorption is 1.8 nm, whichis much smaller than that estimated based on TEM images (2.4 nm).The dispersion calculated by TEM is 55.6%. As reported in references[24,25] by several groups, an overestimated of dispersion often occurswhen CO was used as a probe molecule. This is due to a quantity of
Table 2Dispersion and particle size of ruthenium nanoparticles determined by CO chemisorption and TEM.
Ru-OMC* was synthesized via method b in Fig. 5.a The average particle size was estimated by the equation: dn ¼ ∑nidi
∑nibased on TEM data.
b Ru dispersion was obtained using the equation DRu=1.33/dRu [22].
33Y. Li et al. / Catalysis Communications 20 (2012) 29–35
very small Ru particles that are produced on which CO is chemisorbed.The H2 chemisorption on Ru catalysts was also attempted to measurethe dispersion of Ru nanoparticles. However, we did not observe theH2 chemisorption. Zhao et al. [8] have reported a similar catalystprepared by a thermal reduction method, they mentioned that theyalso could not observe the H2 chemisorption. The reason for this is notclear at present. Although the dispersion determined by CO chemisorp-tion is not accurate, the parallel comparison of dispersion does makesense. From the data in Table 2, it can be seen that the amount of CO ad-sorption for Ru-OMC is onefold increases compared with Ru/OMC andRu/SBA-15 catalysts. This is also proved by TEM measurement. Whilethe particle size and dispersion determined by TEM (1.8 nm and 74.1%respectively) and CO chemisorption (9.9 nm and 13.5% respectively)has a big discrepancy for Ru-carbon. This is due to most of the rutheni-um particles are buried in the framework of carbon for Ru-Carbon,which can not be seen by CO molecule. Therefore, an overestimated ofdispersion and particle size for Ru-Carbon is understandable.
For clarity, a comparison mechinism for ruthenium containing car-bon catalysts prepared by different methods are given in Fig. 5. As faras we know, all the reported methods for the preparation of noblemetal containing carbon catalysts used amixture of noblemetal precur-sor in organic carbon source [10,11]. Although the mesostructures ofmetal-carbon catalysts is uniform by using the mesostructured silicaas hard templates, most of the metal particles are buried in the carbonframework due to the premixing of metal ions with carbon precursors.This is demostrated in Fig. 5 (b). The deeply buried status of metal par-ticles caused problems such as less contact metal surfacewith reactantsduring reactions.
In present strategy as demonstrated in Fig. 5 (a), the RuCl3 was in-troduced to in the pores of SBA-15 templates first and sucrose was in-troduced after metal precursors. The Ru ions can be stablized byabundant hydroxly groups of sucrose and thusmost of the Ru nanopar-ticles are located near the surface of the carbon framework after beenreduced during pyrolysis of sucrose. Therefore, most parts of Ru metal
Carbon precursor
SBA-15 (Template)
Ru3+/SBA-15(Template)
Ru 3+ Ru 3+Ru 3+
Carbon support(OMC)
Ru3+Ru3
Ru3+-Carbon precursor
RuCl3·3H2O
Ru3+ Ru3
Ru 3+Ru 3
a
b
c
Fig. 5. Scheme for the mechanisms of ruthenium containing carbon catalysts prepared by(NPs) with more exposed Ru surface; (b): Strategy reported in references [10,11] to prepar
particles can be exposed and act as acitve sites after the silica templatesbeen removed. Carbon supported Ru nanoparticles can be prepared byimmpregnation method with all the Ru nanoparticles exposed at thesurface of carbon framework as demonstrated in Fig. 5 (c). However,the Ru nanoparticles can easily undergo aggeragation during reductionor in the reaction systems. The present prepared Ru-OMC catalyst withsemi-embedded nanoparticles has more exposed Ru surface. It has theadvantages of both supported metal catalysts and embedded catalysts.We have prepared a compared Ru-OMC* sample by premixing RuCl3and sucrose together and then impregnated the mixture into the poreof SBA-15. the other steps are the same with that of Ru-OMC. It isfound that the dispersion determined by CO adsorption is quite lowcompared with that of Ru-OMC, only 11.7%. but the particle size mea-sured via TEM is in the same range with that of Ru-OMC. This couldbe due to the deeply buried status of Ru particles in Ru-OMC* samples.Further characterizations of this sample is under progress. This resultfurther comfirms that the preparation of Ru-OMC sample by usingRuCl3/SBA-15 as hard template can improve Ru-OMC sample withmore exposed active site.
3.3. Catalytic properties
The catalytic properties of the above Ru catalysts for benzene hy-drogenation toward cyclohexane are summarized in Table 3. TheOMC support and RuCl3 are inactive or less active in hydrogenationof benzene. The turnover frequency (TOF) for Ru/SBA-15 gives 974(h−1). The TOF for Ru/OMC gives 2685 (h−1). As is known fromTable 2, the particle size and dispersion of Ru are similar for bothRu/SBA-15 and Ru/OMC. The activity difference may be due to the dif-ferent nature of carbon and silica support. The carbon supportednoble metal catalysts often have better activity than the silica due tothe better metal support interaction and good electron transferringability of carbon than silica [3,8]. Ru-OMC catalyst exhibits a remark-ably higher catalytic activity (entry 6) than other catalysts. The TOF
Reduction+ Ru3+
Carbonization
Silica removal
Carbonization
Silica removal+ Ru3+
+Ru 3+
Ru-OMC*
Ru-OMC
Ru/OMC
different methods. (a): Present strategy to prepare semi-embedded Ru nanoparticlese semi-embedded Ru NPs; (c): Impregnation method to prepare supported Ru NPs.
Ru-OMC* was synthesized via method b in Fig. 5.a Molar ratio of benzene over Ru.b Loading of Ru was measured by EDS and confirmed by inductance coupling plasma emission spectroscopy analysis.c The activity was calculated as the conversion of moles of benzene per mole of Ru per hour.d Data from reference [3].e Data from reference [8].
34 Y. Li et al. / Catalysis Communications 20 (2012) 29–35
for Ru-OMC catalyst reaches 35112 (h−1) which is around 12 timeshigher than the OMC supported Ru catalyst (Ru/OMC) with the simi-lar Ru loading. The TOF value is the highest among all the datareported in the references for hydrogenation of benzene as far aswe know [8,26,27]. From the data given in Table 1, it can be seenthat Ru-OMC and Ru/OMC have similar surface area and pore volume,which excludes the effect of the textural properties on the catalyticperformance. The dispersion of Ru metal nanoparticles of Ru-OMC ishigher than that of Ru/OMC as discussed in previous part, which contrib-utes to higher performance of Ru-OMC samples prepared by the one-potapproach. However, the high dispersion and uniformly distributedmetalparticles are not the only reason for so high performance of Ru-OMCcompared to that of Ru/OMC catalysts. The metal nanoparticles aresemi-embedded in the carbon framework for Ru-OMC while the metalparticles distribute on the surface of OMC supports for Ru/OMC. Thesemi-embedded status of Ru nanoparticles creates more Ru-carbonpoint contacts. The concave Ru-carbon contact of semi-embedded sam-ples may also strongly anchor and immobilize the Ru particles on thepore surface and prevent the Ru nanoparticles from leaching and mi-grating across the pore, thus agglomerating to form larger particles isprevented. Themore intensive Ru-carbon surface contacts also promotethe transport of adsorbed hydrogen from the Ru particles to the supportfor hydrogen spillover and hydrogenation reactions [28,29].
In summary, the high activity of Ru-OMC can be attributed to highefficient metal-support interaction when the metal is semi-embeddedin the carbon framework. This phenomenon has been discussed in detailin references [8,9]. In addition, the activity of Ru-OMC catalyst (entry 4)at 110 °C and 4 MPa is around 6 times than that of the sandwiched Ru(RuC2, entry 9) prepared by the CVD method [3]. The activity of Ru-OMC at 110 °C, 4 and 8 MPa (entry 4 and 11) is compared with that ofthe Ru/TMC (entry 10 and 12) prepared by a thermal reductionmethodreported in the reference [8], which is almost twice higher. This may bedue to a higher dispersion of Ru nanoparticles prepared by the presentsucrose stabilized approach. The catalytic performance for Ru-Carbonprepared without SBA-15 templates is given in Table 3. It can be seenthat the TOF value for Ru-Carbon is only 2406 h−1, which is quite lowcompared with that of Ru-OMC although the nanoparticles of Ru aresmall and homogeneous. This is due to the lack of mesopores of Ru-Carbon and thus the amount of exposed active sites is much lowerthan that of the Ru-OMC. However, the activity is in the same level ofRu/OMC. This is due to the a few active sites located in the external sur-face of Ru-Carbon catalysts. The TOF value for Ru-OMC* is 14610 h−1,which is higher than that of Ru-Carbon. This can be explained by theexisting of mesopores. However, the activity is still much lower than
that of Ru-OMC. This is due to less exposed active site of Ru-OMC* com-pared with that of Ru-OMC. The compared catalytic performances andall the above characterizations results for Ru-Carbon, Ru-OMC* andRu-OMC samples further confirmed that the semi-embedded Ru nano-particles with more exposed sites in Ru-OMC are responsible for itshigh catalytic performance. Further study on this effective methodfor preparation of metal embedded carbon materials is under progress.
The reuse of Ru-OMC was studied and data are given in Table 3.The cycling performance of the Ru-OMC has no obvious loss of activ-ity even after nine consecutive runs (entry 7 and 8 in Table 3). TheHRTEM and STEM images and the particle size distribution for the 9threcycled catalysts are given in supporting information (Figure S3). Theparticle size and distribution did not change much comparing withthat of the fresh Ru-OMC. This result proves that the Ru-OMC is quitestable in the present reaction conditions.
4. Conclusion
We have developed a one-pot method for the preparation of Ru-OMC catalyst with uniform Ru nanoparticles in 2–3 nm size rangesemi-embedded in the carbon structure. The RuCl3/SBA-15 was usedas the hard template and the sucrose was used as the carbon precur-sor. The usage of sucrose as the carbon precursor not only plays animportant role in stabilizing Ru3+ but also acts as reductant duringfurther carbonization. The interaction of sucrose with Ru3+ ions anda two-step carbonization can significantly avoid the sintering ofnoble metal particles during high temperature thermal treatmentthus facilitating the high dispersion of metal particles. The use ofRuCl3/SBA-15 as hard template instead of premixing of RuCl3 with car-bon precursors makes most exposed ruthenium nanoparticles surfaceavaible for reactants during catalytic reactions. This Ru-OMC catalysthas remarkably high activity and stability in the heterogenous hydroge-nation reactions. This easy, environmentally benign approach can bepotentially extended to prepare carbon supported nano catalysts suchas Pt, Pd and Au catalysts.
Acknowledgments
The financial support from the Natural Science Foundation of China(NSFC Grant No. 20803064), Natural Science Foundation of ZhejiangProvence (Y4090348) andQianjiang Talent Project in Zhejiang Province(2010R10039) are gratefully acknowledged.
35Y. Li et al. / Catalysis Communications 20 (2012) 29–35
Appendix A. Supplementary data
Supplementary data to this article can be found online at doi:10.1016/j.catcom.2011.12.037.
References
[1] C.D. Liang, Z.J. Li, S. Dai, Angewandte Chemie International Edition 47 (2008)3696–3717.
[2] H.Y. Wang, Y. Wan, Journal of Materials Science 44 (2009) 6553–6562.[3] F.B. Su, F.Y. Lee, L. Lv, J.J. Liu, X.N. Tian, X.S. Zhao, Advanced Functional Materials
17 (2007) 1926–1931.[4] F. Goettmann, C. Sanchez, Journal of Materials Chemistry 17 (2007) 24–30.[5] Z.H. Ji, S.G. Liang, Y.B. Jiang, H. Li, Z.M. Liu, T. Zhao, Carbon 47 (2009) 2194–2199.[6] J.H. Zeng, F.B. Su, Y.F. Han, Z.Q. Tian, C.K. Poh, Z.L. Liu, J.Y. Lin, J.Y. Lee, X.S. Zhao,
Journal of Physical Chemistry C 112 (2008) 15908–15914.[7] J.J. Liu, X.N. Tian, X.S. Zhao, Australian Journal of Chemistry 62 (2009) 1020–1026.[8] F.B. Su, L. Lv, F.Y. Lee, T. Liu, A.I. Cooper, X.S. Zhao, Journal of the American Chemical
Society 129 (2007) 14213–14233.[9] S.H. Liu, R.F. Lu, S.J. Huang, A.Y. Lo, S.H. Chien, S.B. Liu, Chemical Communications
Materials 20 (2008) 1622–1628.[11] K. Xiong, J. Li, K. Lie, X. Zhang, Applied Catalysis A: General 389 (2010) 173–178.[12] S.H. Liu, S.C. Chen, Journal of Solid State Chemistry 184 (2011) 2420–2427.[13] L.D. Rogatis, M. Cargnello, V. Gombac, B. Lorenzut, T. Montini, P. Fornasiero,
ChemSusChem 3 (2010) 24–42.
[14] A. Roucoux, J. Schulz, H. Patin, Chemical Reviews 102 (2002) 3757–3778.[15] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D. Stucky,
Science 279 (1998) 548–552.[16] W. Zhang, C. Liang, H. Sun, Z. Shen, Y. Guan, P. Ying, C. Li, Advanced Materials 14
(2002) 1776–1778.[17] Y. Li, Z.C. Feng, Y.X. Lian, K.Q. Sun, L. Zhang, G.Q. Jia, Q.H. Yang, C. Li, Microporous
and Mesoporous Materials 84 (2005) 41–49.[18] S. Jun, S.H. Joo, R. Ryoo, M. Kruk, M. Jaroniec, Z. Liu, T. Ohsuna, O. Terasaki, Journal
of the American Chemical Society 122 (2000) 10712–10713.[19] Y. Li, J. Zhong, Z. Yang, G.J. Lan, H.D. Tang, New Carbon Materials 26 (2011)
123–129.[20] T.W. Hansen, J.B. Wagner, P.L. Hansen, S. Dahl, H. Topsøe, C.J.H. Jacobsen, Science
294 (2001) 1508–1510.[21] O. Thomas, G. Flavien, S.H. Marie-Anne, G. Antoine, Carbon 47 (2009) 2352–2357.[22] M.L. Toebes, F.F. Prinsloo, J.H. Bitter, K.P. Dillen, A.J. de Jong, Journal of Catalysis
(2004) 384–396.[24] M. Kantcheva, S. Sayan, Catalysis Letters 60 (1999) 27–28.[25] R. Ilenia, P. Nicola, F. Lucio, Applied Catalysis A 248 (2003) 97–103.[26] S. Miao, Z. Liu, B. Han, J. Huang, Z. Sun, J. Zhang, T. Jiang, Angewandte Chemie,
International Edition 45 (2006) 266–269.[27] X. Zhou, T. Wu, B. Hu, T. Jiang, B.X. Han, Journal of Molecular Catalysis A: Chemical
306 (2009) 143–148.[28] S.D. Lin, M.A. Vannice, Journal of Catalysis 143 (1993) 539–553.[29] S.D. Lin, M.A. Vannice, Journal of Catalysis 143 (1993) 554–562.
