A highly stable Ru/LaCO3OH catalyst consisting of support-coated Ru nanoparticles in aqueous-phase hydrogenolysis reactions

Bolong Li , Lulu Li and Chen Zhao *
Shanghai Key Laboratory of Green Chemistry and Chemical Processes, School of Chemistry and Molecular Engineering, East China Normal University, Shanghai, 200062, China. E-mail: czhao@chem.ecnu.edu.cn

Received 8th August 2017 , Accepted 18th September 2017

First published on 19th September 2017


Hydrothermal reduction under aqueous conditions is widely used to convert biomass into more valuable products. However, the harsh conditions inherent in the process can irreversibly alter the intrinsic structure of the support, as well as dissolve the metal ions into the aqueous solution. In this work, for the first time we have synthesized a new highly hydrothermally stable Ru/LaCO3OH catalyst mostly consisting of Ru nanoparticles partially encapsulated by the LaCO3OH support with a strong metal–support interaction (SMSI), which confers high stability and activity to the catalyst under hydrothermal reduction conditions in the hydrogenolysis of the biomass model molecules guaiacol and glycerol. During impregnation, the RuCl3·3H2O precursor initially reacts with LaCO3OH to form a LaRu(CO3)2Cl2 complex and LaOCl. XPS demonstrated that Ru was present in the oxidized state, TEM and XRD showed the absence of Ru0, and the XRD pattern showed the presence of the characteristic lattice fringe of LaOCl. While the LaRu(CO3)2Cl2 complex was resistant to H2 reduction at 350 °C, the complex underwent facile reduction to Ru0 under hydrothermal conditions at 240 °C. In the subsequent process, LaRu(CO3)2Cl2 and LaOCl underwent hydrolysis, forming crystalline LaCO3OH (confirmed by Ag+ titration and XRD patterns), Ru(OH)3, and HCl. The Ru(OH)3 was reduced in situ to Ru0 nanoparticles, as revealed by XPS and TEM analysis. The simultaneous hydrothermal reduction of Run+ species and the formation of crystalline LaCO3OH result in the formation of Ru nanoparticles encapsulated by a protective LaCO3OH layer, as evidenced by HRTEM and DRIFTS CO adsorption measurements. The preparation of catalysts with this unique structure comprising metal nanoparticles protected by the support itself, which confers additional stability, is a novel strategy to prepare hydrothermally stable catalysts.


Introduction

Bio-oils derived from the fast pyrolysis or liquefaction of sustainable biomass such as cellulose, lignin, chitin, and lipids have attracted great attention as partial replacements of traditional non-renewable fossil fuels.1–9 However, these bio-oils are highly oxygenated due to the high oxygen contents of the precursor macromolecular biomass, and hence cannot be used directly as fuel sources without prior reduction with metal catalysts.10,11 Due to the polar and hydrophilic nature of bio-liquids which contain multiple oxygen-containing functional groups, and water being ubiquitous in biomass utilization, water is an attractive solvent for the conversion of bio-liquids into useful biofuels. In addition, the acidic or basic properties of water can be adjusted by controlling the pH. Indeed, there are many reports involving metal catalyzed reactions conducted in the aqueous phase for the conversion of substrates obtained from biomass into more useful products, such as the hydrogenolysis of C–C and C–O bonds, hydrogenation, aldol condensation, isomerization, selective oxidation, and aqueous phase reforming reactions.12–16

It should be noted that the ionization constant of water (Kw) is highly dependent on temperature,17 and hence a reaction catalyzed with a supported metal catalyst at elevated temperatures leads to enhanced H+ and OH concentrations which can degrade the catalyst support structure. Thus, the stability of catalysts at elevated temperatures is an important factor that should be taken into consideration for the conversion of biomass. For example, Ravenelle et al. reported that the γ-Al2O3 support gradually hydrated to böhmite when Pt/γ-Al2O3 was treated under aqueous conditions at 225 °C, resulting in a dramatic decrease in the surface area, and an increase in the size of the Pt nanoparticles from 4.3 nm to 7.8 nm.18,19 Similarly, the surface area of SiO2 was shown to decrease after treatment under aqueous conditions at 200 °C for 12 h (from 280 cm3 g−1 to 70 cm3 g−1).20 With respect to mesoporous SBA-15 with highly ordered structures, the specific surface area decreased from 740 cm3 g−1 to 30 cm3 g−1 after treatment under similar conditions due to the high temperature induced alteration of the mesoporous structure of the catalyst support in the aqueous phase.20 These destructive effects of the catalyst structure caused by hydrothermal conditions greatly decrease the catalytic activities. Hence, there is a strong need to synthesize catalysts with high hydrothermal stabilities, which would in particular increase the efficiency of the production of biofuels or biochemicals under aqueous conditions.

Great efforts have been made recently in order to enhance the hydrothermal stability of supported metal catalysts. To protect hydrothermally-unstable catalysts, a strategy involving coating a carbon layer onto the metal surface of the support by introducing glucose as a carbonization agent was developed by Datye's group.20,21 In addition, the same group reported a chemical vapor deposition (CVD) method that formed graphitic carbon coated Pt/γ-Al2O3, which showed excellent stability for reforming and hydrogenation reactions in the liquid phase. In addition, the atomic layer deposition (ALD) method has been used to prevent the leaching/reoxidation of non-noble metal catalysts, thereby enhancing the stability of catalysts for aqueous phase reactions.22,23 These methods enhance the hydrothermal stability of supported metal catalysts by introducing an inert outer layer coating on the support surface.

The oxides of the rare earth element lanthanum, such as La2O3, LaO2CO3 and LaCO3OH, have been frequently used as supports in metal catalyzed ammonia synthesis, dry reforming, and oxidation reactions.24,25 Importantly, due to the rather harsh aqueous conditions used during its preparation, LaCO3OH exhibits high hydrothermal stability. Hence, LaCO3OH could be an appropriate hydrothermally stable support for the metal catalyzed hydrogenolysis of the C–O bonds of biomass derived materials under aqueous conditions.

In our previous work, Ru/C was found to be more selective than Pd/C and Pt/C for the hydrogenolysis of the C–O bond of lignin model compounds in the aqueous phase.13,14 Herein, we report a new Ru/LaCO3OH hydrothermally stable catalyst formed as a result of successive procedures consisting of incipient impregnation, air calcination, and hydrothermal reduction, which consists of protected Ru nanoparticles encapsulated by LaCO3OH with a strong metal–support interaction (SMSI), and this catalyst is highly stable and leads to high hydrogenolysis rates for the conversion of the model bio-molecules guaiacol and glycerol to the corresponding reduced products. In order to understand the relationship between catalytic activities and structural properties, extensive characterization of the catalyst and studies on the hydrogenolysis reaction were carried out; in addition, the two processes involved in the impregnation of the Cl containing Ru precursor into the support and the hydrothermal reduction of the as-formed Ru complex were investigated in depth.

Results and discussion

Three separate catalysts were prepared to study and compare their morphology and catalytic properties, and were initially prepared by loading a Ru metal precursor with a metal content of 5 wt%, as determined by inductively coupled plasma spectroscopy (ICP-AES), via the impregnation method to different supports (SiO2, ZrO2 and LaCO3OH) with Brunauer–Emmett–Teller (BET) surface areas of 231.5, 114.0 and 8.2 m2 g−1, respectively. After impregnation, the high temperatures involved during air calcination and hydrogen reduction processes led to a decrease in the BET surface areas of the catalysts to 160.0, 69.4, and 3.5 m2 g−1, respectively (as shown in Fig. S1). The XRD diffractograms of the reduced samples are shown in Fig. 1. While the characteristic Ru0 peaks (JCPDS 06-0663) appeared for the Ru/SiO2 catalyst, these peaks were absent for the Ru/ZrO2 and Ru/LaCO3OH catalysts, probably due to the presence of smaller Ru nanoparticles on the surface of the ZrO2 and LaCO3OH supports. Notably, in the Ru/LaCO3OH catalyst, apart from the intense peaks assigned to LaCO3OH (JCPDS 26-0815), two broad peaks appeared at 25.2° and 33.9° which were assigned to the characteristic LaOCl species (JCPDS 08-0477). The formation of lanthanide oxychloride (LnOCl, Ln = La, Ce, Pr, Nd, Sm, Eu, Gd, Dy, and Er) in lanthanide-supported catalysts has been previously reported in the literature when chloride salts have been used as the metal precursors. For example, Basińska et al.26 detected the presence of the tetragonal LaOCl phase after conducting the water–gas shift reaction at 350 °C using a Ru/La2O3 catalyst prepared with a chlorine containing precursor and the corresponding rare earth element oxide. Similarly, Le Normand et al.27 found that using a palladium chloride salt as the metal precursor and rare earth element oxides (La2O3, Pr2O3, and Tb2O3) during the impregnation step led to the formation of a stable oxychloride phase (LaOCl, PrOCl, and TbOCl). In our case, the LaOCl phase was formed from the RuCl3·3H2O precursor and the LaCO3OH support after the sequential impregnation and calcination procedures. For the Ru/LaCO3OH sample, the intensity of the peaks corresponding to LaCO3OH decreased, while new broad peaks corresponding to the LaOCl crystalline phase appeared. This suggests that LaOCl forms on the external surface of the support as the RuCl3·3H2O precursor reacts with the LaCO3OH support.
image file: c7gc02414b-f1.tif
Fig. 1 XRD patterns of Ru/SiO2, Ru/ZrO2, and Ru/LaCO3OH catalysts.

Transmission electron microscopy (TEM) was used to obtain the particle information of the catalysts (Ru/SiO2, Ru/ZrO2, and Ru/LaCO3OH) after the H2 reduction of the calcined samples at 350 °C (Fig. 2). In Ru/SiO2, the catalyst has Ru particles with a mean size of 7.0 nm (Fig. 2a). The TEM image of the Ru/ZrO2 catalyst showed little contrast between Ru and ZrO2, and only a few Ru particles (d = ca. 3.2 nm) can be seen at the edge of the ZrO2 (Fig. 2b). The LaCO3OH support in the Ru/LaCO3OH catalyst consisted of large crystals with irregular shapes (Fig. 2c). In addition, the HRTEM image shows that while LaOCl (001) with visible lattice fringes (0.68 nm) was present (Fig. 4d), no Ru0 particles were found in this catalyst, which is consistent with the XRD result (Fig. 1). It is suggested that the Cl ions from RuCl3 may help to disperse Ru on the LaCO3OH support, or that RuCl3 in the metal precursor may react with the LaCO3OH support to form a new Ru complex species that is highly dispersed on the external surface of the support. The energy dispersive spectroscopy (EDS) mapping on the Ru/LaCO3OH catalyst demonstrated that Ru, Cl, and La were evenly dispersed throughout the support (Fig. 3a–c). The high angle annular dark field-scanning TEM (HAADF-STEM) image showed that no Ru nanoparticles were present on the detected regions of the catalyst (Fig. 3d).


image file: c7gc02414b-f2.tif
Fig. 2 TEM images of (a) Ru/SiO2, (b) Ru/ZrO2, and (c) Ru/LaCO3OH and (d) HRTEM image of Ru/LaCO3OH samples with a fringe lattice of LaOCl after hydrogen reduction at 350 °C.

image file: c7gc02414b-f3.tif
Fig. 3 TEM-EDS mapping for the elements in Ru/LaCO3OH (a) La, (b) Cl, and (c) Ru and (d) HAADF-STEM image.

image file: c7gc02414b-f4.tif
Fig. 4 (a) XPS spectra of Ru/SiO2, Ru/ZrO2 and Ru/LaCO3OH after hydrogen reduction at 350 °C with H2; (b) XPS spectra of Ru/LaCO3OH after air-calcination and hydrogen reduction at 350 °C; (c) XPS spectra of Cl in Ru/LaCO3OH after air-calcination and hydrogen reduction at 350 °C.

X-ray photoelectron spectroscopy (XPS) was carried out to characterize the Ru 3d oxidation state of the three Ru catalysts after hydrogen reduction at 350 °C (Fig. 4a). The Ru 3d3/2 peak and the C 1s peak at 284.8 eV partially overlap, and the binding energies were calculated taking as reference the C 1s (284.8 eV) peak of carbon contamination.28 In the XPS spectra of Ru/SiO2 and Ru/ZrO2 catalysts, clearly visible asymmetric peaks corresponding to Ru 3d5/2 were present at 280.2 eV and 280.1 eV, respectively, while the Ru 3d3/2 peaks were observed at 284.3 eV and 284.2 eV, respectively. The presence of the typical peaks corresponding to Ru0 indicates that the SiO2 and ZrO2 catalysts undergo facile reduction at 350 °C.29

With respect to the Ru/LaCO3OH catalyst, after deconvolution, a C[double bond, length as m-dash]O signal from the CO32− group was observed at 289.3 eV.28 With respect to the calcined Ru/LaCO3OH catalyst, a Ru 3d5/2 signal was observed at 282.7 eV (Fig. 4b), which can be attributed to the Ru(III) species; this value is very close to that measured by Mazzieri et al. for the Ru/Al2O3 catalyst (282.9 eV).30 In the hydrogen reduced catalyst, a Ru 3d5/2 signal appeared at 281.6 eV corresponding to an electron deficient ruthenium species (Run+), which was assigned to ruthenium(II) with chloride counterions (Fig. 4b).29,30 This result and the XRD data together demonstrate that the Ru/LaCO3OH catalyst is incompletely reduced even after hydrogen reduction, perhaps resulting from the influence of the residual chloride in the RuCl3·3H2O precursor. Indeed, the XPS experiment determined that chloride was present as evidenced by the XPS signals with a binding energy of 199.58 eV and 197.98 eV, assigned to 2p1/2 and 2p3/2, respectively, of Cl in the LaOCl species (Fig. 4c and Fig. S2).31 The intermediate Ru(II) oxidation state is likely to be highly stable in the presence of a LaCO3OH support and residual Cl ions. Moreover, it has previously been reported that chloride containing precursors may induce incomplete hydrogen reduction of Ru.32 Mazzieri et al.30 reported that the hydrogen reduction of a Ru/Al2O3 catalyst using a RuCl3 precursor led to partial reduction, with a signal at 281.5 eV assigned to oxidized Ru, and a signal at 280.0 eV assigned to Ru0 also present. After the reduction of the sample in hydrogen at a higher temperature of 500 °C for 4 h, the incompletely reduced Ru2+ still existed with the Cl residue, as confirmed by the signal at 281.5 eV in the XPS spectrum (Fig. S3). This is in accordance with the finding of Faroldi et al.33 that Ru2+ was the dominant species with the typical Ru 3d5/2 signal at 281.8 eV from the XPS spectrum, when reducing Ru/La2O3 with H2 at 550 °C using RuCl3·3H2O as the precursor.

The reducibility of the three air-calcined supported Ru catalysts was investigated by hydrogen temperature-programmed reduction (H2-TPR), as shown in Fig. 5. The H2-TPR profile of the Ru/SiO2 catalyst exhibited two reduction peaks at 149 °C and 190 °C. The two temperatures were attributed to the reduction of different particle sized RuOx to Ru0 metal, as suggested by Yan et al.34 In comparison, Ru/ZrO2 showed one reduction peak at 134 °C, assigned to the reduction of RuO2 to Ru metal. The slightly lower reduction temperature is attributed to the smaller Ru particle sizes and the higher dispersion of Ru on ZrO2, which is supported by the XRD pattern and the TEM image results as well. Surprisingly, Ru/LaCO3OH exhibited a broad reduction peak from 260 to 420 °C, with a smaller peak at 341 °C, and the main peak at 382 °C. Taking into consideration the XPS results which confirm the presence of Ru in the +2 and +3 oxidation states, and the fact that less energy is required to reduce the higher oxidation state, the lower reduction peak at 341 °C can be ascribed to the reduction of (Ru(III)) to an intermediate oxidation state (Ru(II)), while the peak at 382 °C may be ascribed to the reduction of Ru(II) species to Ru metal. The reduction temperatures of the Ru/LaCO3OH catalyst were far higher than those of the Ru/ZrO2 and Ru/SiO2 samples, indicating that Ru formed a very stable complex species rather than discrete RuOx particles after air calcination.


image file: c7gc02414b-f5.tif
Fig. 5 H2-TPR profiles of air-calcined Ru/SiO2, Ru/ZrO2, and Ru/LaCO3OH samples.

Considering the observed experimental data, we suggest that during impregnation, Cl ions from the RuCl3·3H2O precursor participate in a surface reaction with the LaCO3OH support to form crystalline LaOCl (as detected by XRD and TEM experiments), and the Ru3+ ions react strongly with the support to form a complex-like species. According to the TEM, XPS, and H2-TPR results, ruthenium was found to be in a complex chemical environment (Scheme 1), and based on the stoichiometry of the surface reaction, the ruthenium complex species is inferred to be LaRu(CO3)2Cl2 (eqn (I)):

 
2LaCO3OH + RuCl3 → LaOCl + LaRu(CO3)2Cl2 + H2O(I)


image file: c7gc02414b-s1.tif
Scheme 1 The surface reaction between the RuCl3·3H2O precursor and the LaCO3OH support.

The high reduction temperature of ruthenium as determined by the TPR-H2 profile confirms the highly stable nature of the LaRu(CO3)2Cl2 complex. However, when the chloride counterions were removed from the LaRu(CO3)2Cl2 complex by washing with ammonia, there was a marked decrease in the reduction temperature to 166 °C (as shown in Fig. 5), further confirming the hypothesis that chloride participates in the surface reaction during impregnation to form a highly stable Ru complex species.

Hydrothermal reduction of Ru complex on LaCO3OH to Ru0 in the presence of H2

Thus far, the TPR-H2 experiment has determined the difficulty of reducing the Ru/LaCO3OH catalyst after calcination, while other experimental data (TEM, XPS, and XRD) suggest that RuCl3 reacts with LaCO3OH to form a new La species (LaOCl) and a complex Ru species (LaRu(CO3)2Cl2) on the external surface of LaCO3OH (Scheme 1). Next, we attempted to reduce the calcined Ru/LaCO3OH catalyst under hydrothermal conditions (240 °C, under H2 at 2 bar pressure for 3 h). During the hydrothermal treatment, the pH value gradually decreased from 6.89 to 4.45 (180 min), suggesting that hydrolysis may release HCl (Fig. 6a). The addition of Ag+ ions led to the precipitation of white AgCl (inset of Fig. 6a), showing that chloride ions are released into the aqueous phase during the hydrothermal treatment. This suggests that the harsh conditions during the hydrothermal treatment induce the hydrolysis of chloride containing species in the solid phase, releasing chloride ions into the aqueous phase as HCl (Scheme 2).
image file: c7gc02414b-f6.tif
Fig. 6 (a) The variation of pH values during hydrothermal treatment at 240 °C in the presence of H2, (b) XRD patterns of LaCO3OH and Ru/LaCO3OH after H2 reduction, and Ru/LaCO3OH after hydrothermal reduction in the presence of hydrogen, (c) XPS spectra of Ru/LaCO3OH after H2 reduction, and Ru/LaCO3OH after hydrothermal reduction in the presence of H2.

image file: c7gc02414b-s2.tif
Scheme 2 The evolution of Ru/LaCO3OH under hydrothermal treatment in the presence of H2.

In addition, after the hydrothermal treatment, XRD analysis determined that the Ru/LaCO3OH catalyst contained a highly crystalline LaCO3OH phase (JCPDS: 26-0815, Fig. 6b) while the LaOCl lattice disappeared. This indicated that the LaOCl phase was no longer present after the hydrolytic removal of surface chloride species, and the La3+ was reconstituted and recrystallized into LaCO3OH under hydrothermal conditions. Notably, no Ru species were observed in the XRD pattern, perhaps due to the overlap with the strong crystal peak of LaCO3OH. Subsequently, XPS was used to characterize the electronic character of Ru after the hydrothermal treatment of Ru/LaCO3OH in the presence of hydrogen (Fig. 6c). Under hydrothermal conditions, the Ru 3d5/2 signal shifted from 281.6 eV (H2 reduction at 350 °C) to the well-known signal indicative of Ru0 at 280.1 eV. Hence, under hydrothermal conditions after the elimination of surface chloride species on the Ru/LaCO3OH catalyst, there is facile reduction of oxidized ruthenium in the calcined catalyst to Ru0.

Therefore, while in the LaRu(CO3)2Cl2 complex the reduction of ruthenium in the presence of H2 to Ru0 in flowing H2 at 350 °C was minimal, under hydrothermal conditions, the reduction occurs under much milder conditions (240 °C). During the hydrothermal process, Cl ions in both LaRu(CO3)2Cl2 and LaOCl are removed from the catalyst surface into the aqueous solution, and LaCO3OH is formed by recrystallization (confirmed by Ag+ titration and XRD patterns). Under hydrothermal conditions, the Run+ species in the LaRu(CO3)2Cl2 complex is hydrolyzed to Ru(OH)3, which is then subsequently reduced to Ru0 nanoparticles (confirmed by XPS). The overall process for the hydrolysis of LaRu(CO3)2Cl2 and LaOCl, and the in situ reduction of Ru(OH)3 occurring on the surface of the catalyst is summarized by eqn (II) and (III):

 
LaOCl + LaRu(CO3)2Cl2 + 4H2O → 2LaCO3OH + 3HCl + Ru(OH)3(II)
 
2Ru(OH)3 + 3H2 → Ru + 6H2O(III)

Since the reduction of Run+ species to Ru0 and the recrystallization of LaCO3OH occurred simultaneously, it is possible that the LaCO3OH could be interacting with the hydrothermally reduced Ru nanoparticles. To confirm the specific nature of the interaction and the morphology of the hydrothermally reduced catalyst, HRTEM and diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) of CO adsorption experiments were carried out. As direct evidence, the TEM and HRTEM images of the Ru/LaCO3OH catalyst after hydrothermal hydrogen reduction are shown in Fig. 7a and b. From these images it is clear that the lattice fringe for LaOCl and the LaRu(CO3)2Cl2 species both disappeared after hydrothermal H2 reduction, while newly formed Ru appeared as dark spots on the LaCO3OH crystal surface. Ru particles with crystal facets were observed with a mean diameter of 2.8 ± 0.3 nm (Fig. 7a). After hydrothermal treatment, the newly formed LaCO3OH layer partially coated the Ru particles with a fringe lattice of Ru (100) (d spacing: 0.234 nm), as clearly shown in Fig. 7b and c. Such structural protection may significantly impede the particle growth and enhance the stability of the Ru/LaCO3OH catalyst under the harsh hydrothermal conditions.


image file: c7gc02414b-f7.tif
Fig. 7 (a) TEM image and particle distribution and (b) the HRTEM image of hydrothermally H2-reduced Ru/LaCO3OH, (c) the Ru (100) fringe lattice of hydrothermally H2-reduced Ru/LaCO3OH.

It is expected that the presence of a LaCO3OH coating surrounding the Ru particles on the hydrothermally hydrogen-reduced Ru/LaCO3OH catalyst would change the surface adsorption characteristics, which can be confirmed by diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) measurements of CO adsorption. The IR spectra for CO desorption on the Ru catalysts with three different supports under a purge of N2 at room temperature are shown in Fig. 8a–c. The strong bands at 2171 cm−1 and 2117 cm−1 on the three Ru catalysts disappeared after a prolonged purge, and were attributed to gaseous CO rather than the adsorption of multicarbonyl species on Run+.33 In the initial stage of the purge, four bands are detected on the Ru/SiO2 catalyst at 2171 cm−1, 2125 cm−1, 2117 cm−1 and 2065 cm−1. After purging for 1500 seconds, the bands at 2171 cm−1 and 2117 cm−1 disappeared. The band at 2125 cm−1 is assigned to the multicarbonyl species bound to Run+ (Run+(CO)n), which lies in the high-frequency 1 (HF1) region (2020–2156 cm−1), while the band at 2065 cm−1 corresponds to the CO bound linearly to Run+, and lies in the HF2 region (2060–2110 cm−1).36 No band corresponding to the bridge-bonded CO was observed in Ru/SiO2. The CO desorption pattern on the Ru/ZrO2 catalyst was quite similar to that of Ru/SiO2, except that the bands in the HF1 (2127 cm−1) and HF2 (2071 cm−1) regions are slightly shifted to higher wavenumbers. The blue shift of the bands may be due to the electron deficient character of Ru on Ru/ZrO2, which may be induced by the Lewis acidic Zr atoms on ZrO2. However, in the Ru/LaCO3OH catalyst, the bands in the region of CO adsorption on Ru (2040–2069 cm−1) and the gaseous CO peak rapidly decreased to near baseline levels. This might be caused by the partial exposure of the Ru particles and the weakly adsorbed CO on the Ru particles, which is in line with the HRTEM image (Fig. 7b) showing that the Ru nanoparticles are partially coated by a LaCO3OH support layer. This result was similar to that observed for Au/HAP-500, where the IR spectra of CO adsorption confirmed that Au nanoparticles were wrapped by hydroxyapatite (HAP).35 Therefore, the characterization of the catalyst with both HRTEM and DRIFT of the adsorbed CO suggests the presence of a LaCO3OH support layer partially coated around the Ru nanoparticles showing a strong metal–support interaction (SMSI).


image file: c7gc02414b-f8.tif
Fig. 8 The DRIFT spectra of CO adsorption on (a) H2-reduced Ru/SiO2, (b) H2-reduced Ru/ZrO2, and (c) hydrothermally-reduced Ru/LaCO3OH in the presence of hydrogen.

Combining the results obtained thus far (XRD patterns, changes in pH, XPS, HRTEM, and DRIFT of adsorbed CO) allows the determination of the various species formed during the hydrothermal reduction of the Ru/LaCO3OH calcined catalyst. Under hydrothermal conditions, the surface species LaOCl and LaRu(CO3)2Cl2 are hydrolyzed (as confirmed by XRD), and the chloride ions that are released in the process lead to a reduction of the pH via the formation of HCl, while the lanthanum side-product LaOCl is recrystallized into LaCO3OH (as determined by XRD). As shown in Scheme 2, hydrolysis of LaRu(CO3)2Cl2 also leads to the formation of the Ru intermediate Ru(OH)3, which was sequentially reduced to Ru0 nanoparticles in the presence of H2 (as demonstrated by XPS). We propose that during the hydrothermal reduction, the Ru nanoparticles are partially coated with a support layer of LaCO3OH, since the formation of reduced Ru nanoparticles occurs simultaneously with the release of LaCO3OH, formed via the hydrolysis of the initially present LaRu(CO3)2Cl2. The HRTEM image and DRIFT of the adsorbed CO further validated the idea of the presence of a LaCO3OH support partially coating the Ru nanoparticles. This support-wrapped metal catalyst is expected to exhibit high stability and activity in aqueous phase hydrogenolysis reactions at high temperatures.

High stability of hydrothermally reduced Ru/LaCO3OH catalyst for aqueous phase hydrogenolysis reactions

To test the hydrogenolysis activity and stability of the novel hydrothermally prepared reduced Ru/LaCO3OH catalyst, the reduction of guaiacol and glycerol (substrates widely available from bio-feedstocks) under aqueous conditions was chosen as the model reaction, and their activities were compared with the conventionally prepared Ru/SiO2 and Ru/ZrO2 catalysts. The product distributions resulting from guaiacol hydrogenolysis under aqueous conditions at 240 °C and 0.2 MPa H2 pressure with the three catalysts are summarized in Fig. 9a. The conversion of guaiacol reached 38.6%, 95.4%, and 95.6%, while the yield of benzene was 2.4%, 30.6%, and 75.8%, for the Ru/SiO2, Ru/ZrO2, and Ru/LaCO3OH catalysts, respectively. Fig. 9b shows the kinetics of guaiacol hydrogenolysis on hydrothermally reduced Ru/LaCO3OH. Accompanied by the rapid decrease of the guaiacol reactant, the maximum phenol yield reached 60.0% at 30 min with an initial hydrogenolysis rate of 3.24 g g−1 h−1, and then gradually decreased. The yield of benzene gradually increased to 75.8% at the reaction time of 150 min. This result suggested that the hydrogenolysis of the methoxy group in guaiacol occurs as the first step in the reduction process, and is followed by the hydrogenolysis of phenol to benzene as the second step, which was determined to be the rate-limiting step according to kinetics modeling (Fig. S4). This result indicates that the newly developed Ru/LaCO3OH catalyst shows high activity and selectivity in the hydrogenolysis of guaiacol to benzene under hydrothermal conditions.
image file: c7gc02414b-f9.tif
Fig. 9 (a) Hydrogenolysis of guaiacol over Ru/SiO2, Ru/ZrO2, and hydrothermally-reduced Ru/LaCO3OH, conditions: 1.0 g guaiacol, 0.75 g catalyst, H2O (150 mL), 0.8 MPa N2 and 0.2 MPa H2, 150 min; (b) kinetics of the hydrogenolysis of guaiacol on hydrothermally-reduced Ru/LaCO3OH.

Glycerol hydrogenolysis in the aqueous phase was conducted over these three Ru catalysts at 200 °C and 3 MPa H2 pressure. The glycerol conversion reached 10%, 21%, and 62%, while the selectivity to 1,2-propanediol reached 2%, 5%, and 46% with the Ru/SiO2, Ru/ZrO2, and Ru/LaCO3OH catalysts, respectively (Fig. 10a).


image file: c7gc02414b-f10.tif
Fig. 10 (a) Hydrogenolysis of glycerol over Ru/SiO2, Ru/ZrO2, and hydrothermally-reduced Ru/LaCO3OH, conditions: 0.4 g glycerol, 0.1 g catalyst, H2O (80 mL), 3 MPa H2, 3 h; (b) kinetics of the hydrogenolysis of glycerol on hydrothermally-reduced Ru/LaCO3OH.

A kinetics curve for glycerol conversion (Fig. 10b) over hydrothermally reduced Ru/LaCO3OH revealed that the primary product was 1,2-propanediol with an initial hydrogenolysis rate of 2.0 g g−1 h−1. When the reaction time was prolonged, the C–C cleavage products of ethylene glycol and methanol slightly increased to 5%, while the yield of 1,2-propanediol increased to nearly 40%. These results further confirm the superiority of the Ru/LaCO3OH catalyst for achieving high activity and selectivity for aqueous phase hydrogenolysis reactions.

In the recycling tests for guaiacol hydrogenolysis in water, Ru/LaCO3OH showed a much higher durable capability than Ru/SiO2 and Ru/ZrO2, as displayed in Fig. S5 and S6. Ru/LaCO3OH still maintained high conversion (>95%) and benzene selectivity (>70%) even after eight runs, but Ru/ZrO2 and Ru/SiO2 only attained 70% and 10% conversion after four runs with 22% and 1% benzene selectivity after four runs, respectively. A dramatic activity loss was observed for the unprotected Ru nanoparticles on Ru/ZrO2 and Ru/SiO2 catalysts when recycling in the high-temperature water phase.

In order to test the influence of the protective layer towards the hydrogenolysis reaction, Ru/LaCO3OH without a support layer (5 wt% Ru loading, identical to the metal loading of Ru/LaCO3OH with Cl) was synthesized subsequently. To obtain the Cl free catalyst, RuCl3·3H2O metal salt was impregnated with a basic agent NH3·H2O solution. Thus, the Ru(OH)3 precursor was formed after impregnation, and NH4Cl stayed in the aqueous phase as confirmed by Ag+ titration. In contrast to the Ru/LaCO3OH catalyst with Cl ions, in the Cl free catalyst, Ru nanoparticles were not protected by a support layer, as demonstrated by the HRTEM image in Fig. S7. Accordingly, the as-formed unprotected Ru nanoparticles were comparatively easily reduced with H2 at a relatively low temperature (166 °C), as shown in the TPR-H2 profile in Fig. 5.

The hydrogenolysis of guaiacol was conducted over four recycling runs to determine the stability of two Ru/LaCO3OH samples. In Ru/LaCO3OH with Cl ions, a constant conversion of 95% and a benzene yield of 76% were obtained for four recycling runs (Fig. 11). The leaching of Ru ions into the solution in the recycling tests was not detected by ICP measurement. Under identical conditions, the hydrogenolysis of guaiacol with the Cl free catalyst took place with 96% conversion and a benzene yield of 35% in the first run. The successive recycling tests led to gradual decreases in activity and selectivity, with a conversion of 94% and a benzene yield of 12.1% in the fourth run. The slightly higher conversion in the first run of the Cl free catalyst may be due to the presence of a greater amount of unprotected Ru active sites. The marked difference in the capacity to undergo hydrodeoxygenation between these two catalysts demonstrates the importance of the protective coating around Ru with SMSI, and suggests that the Cl containing species may hold great potential as a robust catalyst for reactions conducted under hydrothermal conditions.


image file: c7gc02414b-f11.tif
Fig. 11 Four-run recycling tests on the hydrogenolysis of guaiacol over hydrothermally-reduced Ru/LaCO3OH with and without Cl ions. Conditions: 1.0 g guaiacol, 0.75 g catalyst, H2O (150 mL), 0.8 MPa N2 and 0.2 MPa H2, 150 min.

Conclusions

We have synthesized a new highly hydrothermally-stable Ru/LaCO3OH catalyst with a structure comprising a support partially coating Ru nanoparticles via the sequential incipient impregnation, air-calcination, and hydrothermal reduction procedures. This catalyst exhibited superior performance in terms of stability and activity compared to Ru/SiO2 and Ru/ZrO2 catalysts as determined by the conversion and yield obtained in the hydrogenolysis of the biomass model molecules guaiacol and glycerol under hydrothermal conditions. The structure of the hydrothermally reduced Ru/LaCO3OH catalyst (the support coating the Ru nanoparticles) was maintained even after multiple reaction runs, and the leaching of Ru ions was negligible.

In contrast to the Ru/SiO2 and Ru/ZrO2 catalysts, under standard (non-aqueous) conditions, H2 reduction of the Ru/LaCO3OH sample at 350 °C did not lead to the formation of Ru0 species. The reduction of the Ru/LaCO3OH catalyst is hindered by the presence of the stable RuLa(CO3)2Cl2 complex and LaOCl, which are formed instead of RuOx during the impregnation procedure when the RuCl3·3H2O precursor reacts with LaCO3OH. However, upon hydrothermal H2 reduction at 240 °C, LaOCl and RuLa(CO3)2Cl2 were hydrolyzed into highly crystalline LaCO3OH and Ru0 particles. The HRTEM image and DRIFT CO adsorption experiments suggested that the Ru nanoparticles were partially encapsulated by LaCO3OH layers.

The strong interaction between the Ru nanoparticles and the partially coated LaCO3OH support layers impeded the further growth of the metal nanoparticles under harsh reduction conditions. Our newly developed methodology, wherein the metal center is partially protected by support layers themselves during an in situ hydrothermal hydrogen reduction process, is a novel strategy that offers the potential for synthesizing hydrothermally-stable catalysts.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We acknowledge the financial support from the National Key Research and Development Program of China (Grant No. 2016YFB0701100), the Recruitment Program of Global Young Experts in China, the National Natural Science Foundation of China (Grant No. 21573075), and the Foundation of Key Laboratory of Low-Carbon Conversion Science & Engineering, Shanghai Advanced Research Institute, Chinese Academy of Sciences.

Notes and references

  1. G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 106, 4044–4098 CrossRef CAS PubMed .
  2. C. Li, X. Zhao, A. Wang, G. W. Huber and T. Zhang, Chem. Rev., 2015, 115, 11559–11624 CrossRef CAS PubMed .
  3. C. Luo, S. Wang and H. C. Liu, Angew. Chem., Int. Ed., 2007, 46(40), 7636–7639 CrossRef CAS PubMed .
  4. N. Ji, T. Zhang, M. Y. Zheng, A. Q. Wang, H. Wang, X. D. Wang and J. G. Chen, Angew. Chem., Int. Ed., 2008, 47, 8510–8513 CrossRef CAS PubMed .
  5. J. C. Kong, M. Y. He, J. A. Lercher and C. Zhao, Chem. Commun., 2015, 51, 17580–17583 RSC .
  6. X. Chen, S. L. Chew, F. M. Kerton and N. Yan, Green Chem., 2014, 16, 2204–2212 RSC .
  7. S. H. Zhu, X. Q. Gao, Y. L. Zhu and Y. W. Li, Green Chem., 2016, 18, 782–791 RSC .
  8. S. H. Zhu, X. Q. Gao, Y. L. Zhu, W. B. Fan, J. G. Wang and Y. W. Li, Catal. Sci. Technol., 2015, 5, 1169–1180 CAS .
  9. B. Ma and C. Zhao, Green Chem., 2015, 17, 1692–1701 RSC .
  10. J. Zakzeski, P. C. A. Bruijnincx, A. L. Jongerius and B. M. Weckhuysen, Chem. Rev., 2010, 110, 3552–3599 CrossRef CAS PubMed .
  11. C. Zhao, Y. Kou, A. A. Lemonidou, X. Li and J. A. Lercher, Angew. Chem., Int. Ed., 2009, 48, 3987–3990 CrossRef CAS PubMed .
  12. H. F. Xiong, H. N. Pham and A. K. Datye, Green Chem., 2014, 16, 4627–4643 RSC .
  13. Z. C. Luo and C. Zhao, Catal. Sci. Technol., 2016, 6, 3476–3484 CAS .
  14. Z. C. Luo, Y. M. Wang, M. Y. He and C. Zhao, Green Chem., 2016, 18, 433–441 RSC .
  15. Z. C. Luo, Z. X. Zheng, Y. C. Wang, G. Sun, H. Jiang and C. Zhao, Green Chem., 2016, 18, 5845–5858 RSC .
  16. Y. B. Huang, L. Yan, M. Y. Chen, Q. X. Guo and Y. Fu, Green Chem., 2015, 17, 3010–3017 RSC .
  17. A. A. Peterson, F. Vogel, R. P. Lachance, M. Froling, J. M. Antal Jr. and J. W. Tester, Energy Environ. Sci., 2008, 1, 32–65 CAS .
  18. R. M. Ravenelle, J. R. Copeland, A. H. Van Pelt, J. C. Crittenden and C. Sievers, Top. Catal., 2012, 55, 162–174 CrossRef CAS .
  19. R. M. Ravenelle, J. R. Copeland, W. G. Kim, J. C. Crittenden and C. Sievers, ACS Catal., 2011, 1, 552–561 CrossRef CAS .
  20. H. N. Pham, A. E. Anderson, R. L. Johnson, K. Schmidt-Rohr and A. K. Datye, Angew. Chem., Int. Ed., 2012, 51, 13163–13167 CrossRef CAS PubMed .
  21. H. F. Xiong, T. J. Schwartz, N. I. Andersen, J. A. Dumesic and A. K. Datye, Angew. Chem., Int. Ed., 2015, 54, 7939–7943 CrossRef CAS PubMed .
  22. J. Lee, D. H. K. Jackson, T. Li, R. E. Winans, J. A. Dumesic, T. F. Kuech and G. W. Huber, Energy Environ. Sci., 2014, 7, 1657 CAS .
  23. B. J. O'Neill, D. H. K. Jackson, A. J. Crisci, C. A. Farberow, F. Y. Shi, A. C. Alba-Rubio, J. L. Lu, P. J. Dietrich, X. K. Gu, C. L. Marshall, P. C. Stair, J. W. Elam, J. T. Miller, F. H. Ribeiro, P. M. Voyles, J. Greeley, M. Mavrikakis, S. L. Scott, T. F. Kuech and J. A. Dumesic, Angew. Chem., Int. Ed., 2013, 52, 13808–13812 CrossRef PubMed .
  24. C. Carrara, J. Múnera, E. A. Lombard and L. M. Cornaglia, Top. Catal., 2008, 51, 98–106 CrossRef CAS .
  25. Y. J. Zhang, K. D. Han, T. Cheng and Z. Y. Fang, Inorg. Chem., 2007, 46, 4713–4717 CrossRef CAS PubMed .
  26. A. Basińska, L. Kępiński and F. Domka, Appl. Catal., A, 1999, 183, 143–153 CrossRef .
  27. F. Le Normand, J. Barrault, R. Breault, L. Hilaire and A. Kiennemann, J. Phys. Chem., 1991, 95, 257–269 CrossRef CAS .
  28. B. Faroldi, C. Carrara, E. A. Lombardo and L. M. Cornaglia, Appl. Catal., A, 2007, 319, 38–46 CrossRef CAS .
  29. C. Elmasides, D. I. Kondarides, W. Grünert and X. E. Verykios, J. Phys. Chem. B, 1999, 103, 5227–5239 CrossRef CAS .
  30. V. Mazzieri, F. Coloma-Pascual, A. Arcoya, P. C. L'Argentière and N. S. Fígoli, Appl. Surf. Sci., 2003, 210, 222–230 CrossRef CAS .
  31. Y. Lei, B. Liu, J. Lu, J. A. Libera, J. P. Greeley and J. W. Elam, J. Phys. Chem. C, 2014, 118, 22611–22619 CAS .
  32. P. G. J. Koopman, A. P. G. Kieboom and H. van Bekkum, J. Catal., 1981, 69, 172–179 CrossRef CAS .
  33. B. M. Faroldi, E. A. Lombardo and L. M. Cornaglia, Appl. Catal., A, 2009, 369, 15–26 CrossRef CAS .
  34. Q. G. Yan, T. H. Wu, W. Z. Weng, H. Toghiani, R. K. Toghiani, H. L. Wan and C. U. Pittman Jr., J. Catal., 2004, 226, 247–259 CrossRef CAS .
  35. H. L. Tang, J. K. Wei, F. Liu, B. T. Qiao, X. L. Pan, L. Li, J. Y. Liu, J. H. Wang and T. Zhang, J. Am. Chem. Soc., 2016, 138, 56–59 CrossRef CAS PubMed .
  36. S. Y. Chin, C. T. Williams and M. D. Amiridis, J. Phys. Chem. B, 2006, 110, 871–882 CrossRef CAS PubMed .

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c7gc02414b

This journal is © The Royal Society of Chemistry 2017