English

• Pickering emulsion, acting as an emerging heterogeneous catalysis system, can provide a particular solvent environment, where chemical reactions are driven by the emulsifier with active sites adsorbed at the interface of two-phase generally made of water and organic solvent [1–3]. The unique emulsion catalysis properties enable regulating the selectivity of chemical reactions according to the relative solubility of reactants and products [1,4,5]. In addition, the selectivity and reactivity of chemical reactions can also be adjusted by the controlled active sites on the emulsion catalysts on the side of the water or oil [6]. Recently, our previous work confirmed that the water was one of the solvents participated in the oxidation reaction of alcohol, and it was found that the activity of water molecules was the key to promote the whole reaction efficiency [7,8]. Therefore, it is extremely desirable to rationally design the catalysts with a certain ability of enhancing the water-induced reaction process, which is also full of difficulties and challenges.

  It has been well proved that alkali as an electron donor and/or structural regulator can modify the metal-supported catalyst with high catalytic performance compared with the alkali-free catalyst [9,10]. Up to now, the promotion effects of alkali have been widely explored and applied in various heterogeneous reactions, such as water-gas shift reaction [11–14], Fischer-Tropsch synthesis [15–17], selective hydrogenation [9,10,18,19], CO oxidation [20], selective oxidation [21–26], and organic reaction [22,25,27]. A prevailing view is that the alkali metal can modulate the oxidation states and coordination environments of the active sites (Pd [28], Pt [12–14,20], Au [12,22], Ru [9,19], Ir [26]) and simultaneously promote the dispersion of metal nanoparticles on the carrier. Yet, the potential structure-activity relationships between the catalytic activity and the critical role of alkali species are still not clear. This is finally detrimental to the development and practical application for the strategy of alkali-modulated catalysts [29].

  Here, we recognize and decipher the effects of alkali cations on Pd catalyst for selective oxidation of benzyl alcohol (BA) to benzaldehyde (BAD), a chemical reaction system involving in water-promoted effects. The Pd(Na)/rGO-ZnCr₂O₄ catalyst with Na cations constructed by the coprecipitation-calcination method exhibits a turnover frequency (TOF) value of 469 h⁻¹ and the apparently low activation energy (E_a) of 45.6 kJ/mol calculated by reaction kinetics experiment, far superior to that of Na-free counterpart (199 h⁻¹ and 93.1 kJ/mol, respectively). The responsive experiments of water injection reveal that the water molecules are helpful for the oxidation reaction of BA and promote the conversion of BA. More importantly, the Na-modulated catalyst can further enhance the promotion effects of water in the reaction. The Na-modulated catalyst can obviously improve the dispersion of Pd nanoparticles. Also, the Pd active sites present a high oxidation state, and the Pd catalyst modulated by alkali cation can form abundant Pd−OH_x species with high activity. The abundant Pd−OH_x is helpful to enhance the water-induced reaction process. These finds can provide an avenue to enhance the chemical reactions involving water promotion.

  To simplify the synthesis of the catalyst with alkali metals, the catalyst precursors with graphite oxide were prepared using NaOH as alkali source by the coprecipitation-calcination method (Scheme S1 in Supporting information). Correspondingly, the residual Na in the washing process can be controlled, and the Pd catalysts were synthesized via the facile wet impregnation method. X-ray diffraction (XRD) patterns and the FT-IR spectra (Fig. S1 in Supporting information) show that the precursors and catalysts can be assigned to the spinel structure of ZnCr₂O₄ and the Na atoms in Pd(Na)/rGO-ZnCr₂O₄ is stabilized by Na−O bonds. Meanwhile, the weak diffraction peaks of Pd are presented, indicating the uniform distribution of Pd with an ultrafine size on the surface of (Na)rGO-ZnCr₂O₄. The scanning electron microscope (SEM) images (Fig. S2 in Supporting information) indicate that the ZnCr₂O₄ grows uniformly on the GO surface by the nucleation and conversion of the pre-adsorbed Zn²⁺ and Cr³⁺ on GO after calcination, forming (Na)rGO-ZnCr₂O₄ containing 9% content of rGO, as confirmed by thermogravimetric analysis (TGA) results in Fig. S3 (Supporting information). In addition, the surface wettability of Pd(Na)/rGO-ZnCr₂O₄ exhibits a similar affinity for toluene and water, which favor the formation of stable Pickering emulsion (Fig. S4 in Supporting information).

  To compare the reactivity of Na-free and Na-modulated catalysts, the BA oxidation was chosen as a model reaction, based on our previous work [7,30]. As shown in Fig. 1a, compared with that of the Pd/rGO-ZnCr₂O₄ (a conversion of 36% and selectivity of ~99%), the Pd(Na)/rGO-ZnCr₂O₄ exhibits a superior catalytic activity, with a conversion of 98% and selectivity of ~99% at 1 h, being an increase in TOF from 199 h⁻¹ to 469 h⁻¹. The high selectivity for benzaldehyde is consistent with our previous work [7,31]. Furthermore, the macroscopic kinetic experiments of catalysts were performed and the conversion-time plots were obtained at different reaction temperatures (65, 75, 85, 95 ℃) and shown in Figs. 1b and c. The −ln(1 − C) of catalysts and reaction time present a linear relationship with R² > > 0.99, which indicates that the oxidation of BA follows first-order kinetics. The rate constant (k) can be obtained from the slope and the corresponding k value is 0.039 min⁻¹ for Pd(Na)/rGO-ZnCr₂O₄ and 0.017 min⁻¹ for Pd/rGO-ZnCr₂O₄ at 85 ℃, respectively, which further indicates that the catalytic performance of Pd(Na)/rGO-ZnCr₂O₄ is superior of the Pd/rGO-ZnCr₂O₄ catalyst. The ln(k) and 1000/T at different temperatures also exhibit a linear relationship with R² > 0.95, and the apparent activation energy (E_a) is collected (Fig. 1d). The Pd(Na)/rGO-ZnCr₂O₄ (45.6 kJ/mol) presents a low E_a value in comparison to that of Pd/rGO-ZnCr₂O₄ (93.1 kJ/mol), which indicates that the Na cations over Pd(Na)/rGO-ZnCr₂O₄ can promote and enhance the catalytic performance for BA.

  Figure 1

  

  Figure 1.  (a) Catalytic performance over Pd(Na)/rGO-ZnCr₂O₄ and Pd/rGO-ZnCr₂O₄. (b, c) Time-conversion plots at various temperatures for Pd(Na)/rGO-ZnCr₂O₄ and Pd/rGO-ZnCr₂O₄ catalyst, respectively. (d) Arrhenius plots for BA oxidation. (e) Response of water injection over the Pd(Na)/rGO-ZnCr₂O₄ and the corresponding Na-free catalyst for BA oxidation, with toluene as reaction solvent. (f) Catalytic activity test over Pd(K)/rGO-ZnCr₂O₄ and Pd/rGO-ZnCr₂O₄.

  DownLoad: Full-Size Img PowerPoint

  Further, the Na-promoted effects for BA oxidation are well extended to other supports such as Pd(Na)/ZnCr₂O₄ Pd/ZnCr₂O₄ with rGO-free catalysts. The catalytic performance of Pd(Na)/ZnCr₂O₄ (Fig. S5a in Supporting information) is distinctly superior to Pd/ZnCr₂O₄ in the oxidation process of BA, and the corresponding macroscopic kinetic experiments were also performed (Figs. S5b–d in Supporting information). The E_a values of Pd(Na)/ZnCr₂O₄ and Pd/ZnCr₂O₄ catalysts are 66.6 and 129.7 kJ/mol, respectively. This further confirms that the Na cations over the catalyst have a certain positive effect, and contribute to the oxidation of BA. Moreover, the catalytic performance of the Pd catalysts using (Na)rGO-ZnCr₂O₄ as support is better than that of Pd(Na)/ZnCr₂O₄ (Figs. S5a and S6 in Supporting information), which suggests that the rGO has a certain promotion effect on the present reaction. The corresponding details for the synergistic effects were confirmed and demonstrated in Figs. S7–S9 (Supporting information). Furthermore, the TOF values over various catalysts were calculated, which presents a linear relationship (R² = 0.99) with the molar ratio of Na to Pd (n_Na: n_Pd) in the catalysts (Fig. S10 in Supporting information). That is to say, the amount of Na cations over the catalysts is inseparable from the activity of the catalyst.

  In addition, numerous studies have also reported that water participates in catalytic reactions and can promote chemical reaction, such as CO oxidation [32–34], alcohol oxidation [7,35,36], and some hydrogenation reactions [37]. Furthermore, the water-promoted effects for BA oxidation catalyzed by Pd(Na)/rGO@ZnCr₂O₄ and Na-free counterpart catalyst were investigated. After the conversion rate of BA almost reaches a plateaus within 5 h only using toluene as a solvent, 1 mL water was injected, and the results are presented in Fig. 1e. The catalytic activities of Pd/rGO-ZnCr₂O₄ and Pd(Na)/rGO-ZnCr₂O₄ in the pure toluene system only has a reduction compared to that in Pickering emulsion within 5 h (Fig. 1a), which is more obvious for Pd(Na)/rGO-ZnCr₂O₄. When adding 1 mL water, the BA is further converted to benzaldehyde, which manifests that the water molecules promote the chemical reaction process. Also, the conversion rate of BA is faster over Pd(Na)/rGO-ZnCr₂O₄ than that on Pd/rGO-ZnCr₂O₄. That is to say, the water-promoted effects are more obvious and further promoted over catalyst with Na species, which is also one of the reasons why the catalytic activity of Pd(Na)/rGO-ZnCr₂O₄ is superior to that of Pd/rGO-ZnCr₂O₄ in Pickering emulsion reaction system. In general, the water in the system can contribute to the conversion of BA. The Na cation modified-catalyst can effectively further enhance the promotion effects of water molecules in the oxidation of BA. In addition, the Na cations in catalyst are also replaced by K cations, and the similar promotion effects are also observed (Fig. 1f), which further indicates that with the alkali cations in the catalyst, the reaction barrier of the rate-determining step is effectively reduced.

  As we know, the particle size of active metal has some effects on the activity of catalysts [38,39]. The transmission electron microscopy (TEM) was performed to detect the morphology and size distribution of Pd nanoparticles on Pd(Na)/rGO-ZnCr₂O₄ and Pd/rGO-ZnCr₂O₄, and the results were shown in Fig. S11 (Supporting information). It was found that the Pd nanoparticles with ultrafine size are evenly dispersed on the surface of the support without the distinct lattice fringe, which were further analyzed by high-angle annular dark-field scanning TEM (HAADF-TEM). The Pd average size for Pd(Na)/rGO-ZnCr₂O₄ and Pd(Na)/rGO-ZnCr₂O₄ is statistically 1.3 nm and 2.5 nm, respectively, as shown in Fig. 2 and Fig. S12 (Supporting information). It is indicated that the Na cation can induce the dispersion of Pd nanoparticles on the surface of Pd(Na)/rGO-ZnCr₂O₄. In fact, the analogous behavior was also found and confirmed in the past [13]. The corresponding elemental mapping images also reveal that the produced catalyst presents the uniform distribution of Pd and Na over the surface of catalyst. To better present the promotion effects of Na for Pd dispersion, CO pulse experiments were conducted to reveal the dispersion of metal, the ratio of active metal atoms to total metal atoms on the catalyst (Figs. S13a and b in Supporting information). The metal dispersion of Pd(Na)/rGO-ZnCr₂O₄ (54%) is superior to that of Pd/rGO-ZnCr₂O₄ (39%) and similar results can be found for the corresponding rGO-free catalyst, which further confirms that the alkali cations in catalyst can promote the dispersion of Pd atoms. In addition, the metal dispersions of Pd(Na)/rGO-ZnCr₂O₄ and Pd/rGO-ZnCr₂O₄ are superior to that of the corresponding rGO-free counterpart catalysts (Figs. S13c and d in Supporting information), respectively, maybe resulting from the synergistic interaction rGO and ZnCr₂O₄. The above results disclose that the Na cations in the catalyst indeed enable the dispersion of Pd active sites, which can expose the abundant active sites to enhance the oxidation of BA.

  Figure 2

  

  Figure 2.  HAADF-STEM images of (a) Pd/rGO-ZnCr₂O₄ and (b) Pd(Na)/rGO-ZnCr₂O₄ and the corresponding curves of Pd size distribution (inset). (a1, b1 and a2, b2) EDS element maps of Pd and Na for Pd(Na)/rGO-ZnCr₂O₄ and Pd/rGO-ZnCr₂O₄, respectively.

  DownLoad: Full-Size Img PowerPoint

  Additionally, the O 1s region for X-ray photoelectron spectroscopy (XPS) spectra for all catalysts (Fig. 3a and Fig. S14 in Supporting information) reveals three types of oxygen species (denoted as O1, O2, and O3), which can be attributed to lattice O^* (529.0–530.0 eV), lattice OH bonded with metal atoms (530.5–531.5 eV) and crystalline H₂O (532.1–533.0 eV) [22,40], respectively. Compared with Pd/rGO-ZnCr₂O₄, the O1 and O3 peaks of Pd(Na)/rGO-ZnCr₂O₄ have a shift to low and high binding energy, respectively. The electronic properties of the Pd(Na)/rGO-ZnCr₂O₄ surface can be altered. In addition, the O2 region area for Pd(Na)/ZnCr₂O₄ and Pd(Na)/rGO-ZnCr₂O₄ catalysts is larger than that of the corresponding Na-free catalyst. This suggests that the appearance of Na cations significantly increases the density of lattice OH, forming the active Pd−OH_x structure analogously described in the literature [11,12]. The abundant surface OH can also serve as the adsorption sites for BA to contribute to the chemical oxidation reaction [41]. In addition, the Na KLL Auger peak signal can be seen at the 534–536 eV toward the Pd(Na)/rGO-ZnCr₂O₄ and Pd(Na)/ZnCr₂O₄ catalysts (Fig. 3a and Fig. S14). The Na 1s XPS spectra (Fig. 3b) reveal that there is only one major peak centered at 1071.6 eV, implying that the sodium species bind to the surrounding substrate through –O–Na linkages [14], which is consistent with the results of FTIR spectra (Fig. S1b in Supporting information). Nevertheless, the relatively weak Na 1s signal for Pd/rGO-ZnCr₂O₄ is presented, which derives from a small amount of remained Na species within the catalyst. The ICP results (Table S1 in Supporting information) further reveal that the content of Na in Pd/rGO-ZnCr₂O₄ and Pd(Na)/rGO-ZnCr₂O₄ are 0.9 and 3.3 wt%, respectively.

  Figure 3

  

  Figure 3.  The XPS spectra of (a) O 1s, (b) Na 1s and (c) Pd 3d for Pd(Na)/rGO-ZnCr₂O₄ and the corresponding Na-free catalyst. (d) In situ DRIFTS of CO chemisorption over Pd(Na)/rGO-ZnCr₂O₄ and the corresponding Na-free catalyst.

  DownLoad: Full-Size Img PowerPoint

  To eliminate the effects of residual Na in the catalyst, the precursor of catalyst was also synthesized by NH₃·H₂O as a coprecipitation reagent instead of NaOH, and meanwhile the catalysts with the different molar ratios of Na to Pd were also prepared. The corresponding catalytic performances were shown in Fig. S15 (Supporting information). It is observed that the activity gradually improves with an increase of Na content in the catalyst, indicating that the performance of catalyst is closely correlated with the content of Na species in the catalyst. The TOF values for the various catalysts were calculated, showing a linear relationship (R² = 0.92) with the molar ratio of Na to Pd within the catalysts. Furthermore, the XPS spectra (Fig. S16 in Supporting information) reveal that the O1 and O3 peaks of the catalysts with an increase of Na content have a shift to low and high binding energy, respectively. The O2 (a symbol of Pd−OH_x structure) region area gradually increases with an increase of Na content in the catalyst, reaching the peak value at the molar ratio of Na to Pd of 5:1 and keeps constant. The above results further reveal that the Na cations in the catalyst can change the surface electron properties of the catalyst, creating the highly active Pd−OH_x species. The activities of catalysts are enhanced with increased molar ratio of Na to Pd and almost reaches the maximum at 5:1. Further, at 7:1, the activity only has a slight change. The corresponding changes are consistent with the O2 region area of the catalysts. This further confirms that the alkali-stabilized Pd−OH_x species contribute to enhancing the activity of catalyst. The increased content of Na in the catalyst can be also supported by the Na 1s XPS spectra (Fig. S16b).

  It has also been reported that the alkali, as catalyst promoters, can affect the oxidation state of active metals [9,10]. Furthermore, the high-resolution spectra of the Pd 3d region for Pd(Na)/rGO-ZnCr₂O₄ and Pd/rGO-ZnCr₂O₄ (Fig. 3c) can be deconvoluted with binding energies of 335.9, 337.4 eV and 341.2, 342.7 eV, which correspond to Pd⁰ and Pd²⁺, respectively. The fraction of Pd²⁺/Pd⁰ for Pd(Na)/rGO-ZnCr₂O₄ (1.52) on the basis of the peak area has a dramatic increase in comparison to that of Pd/rGO-ZnCr₂O₄ (0.76), which indicated that the Na cation in catalysts can tune the electronic structure of (Na)rGO-ZnCr₂O₄ surface and the Pd species exist in the oxidation states resulting from the effects of Pd−OH_x species. This can be confirmed by the increased fraction of Pd²⁺/Pd⁰ with the increase of Na content in the prepared catalysts using NH₃·H₂O with different molar ratios of Na to Pd (Fig. S17 in Supporting information), where the peak corresponding to Pd²⁺ shows a shift toward a higher binding energy than that of Na-free catalyst.

  To further confirm the effects of Na cations on Pd active sites, in situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) of CO chemisorption was performed. The four peaks are exhibited for Pd/rGO-ZnCr₂O₄, as presented in Fig. 3d. The bands at 2079 and 2050 cm⁻¹ are assigned to linear adsorption of CO on catalyst [42]. The band at 1967 cm⁻¹ is μ₂-bridge adsorption of CO on the step and facet edges. The broad band around 1760–1944 cm⁻¹ is considered as μ₂-bridge adsorption of CO on (111) facet of Pd nanoparticles [43]. In the case of Pd(Na)/rGO-ZnCr₂O₄ with Na species, the absorption peaks shift toward low wavenumber, changing from 2079 cm⁻¹ to 2073 cm⁻¹, 1967 cm⁻¹ to 1957 cm⁻¹, and 1760–1944 cm⁻¹ to 1722-1944 cm⁻¹, indicative of the back-donation of Pd d-electrons to the 2π^* antibonding orbital of CO [44]. The promotion of Pd electron-donation ability may result from the electronic effects of Pd−OH_x structure triggered by Na cation. Furthermore, the positive effects of Na cations on Pd active sites are well extended to other catalysts such as Pd(Na)/ZnCr₂O₄ and Pd/ZnCr₂O₄, and the results were exhibited in Fig. S18 (Supporting information). Only the wide peak for μ₂-bridge-bonded CO on (111) facets of Pd nanoparticles appears and shifts from 1757–1944 cm⁻¹ to 1724–1944 cm⁻¹. The reasons may be resulted from the low content of Na for Pd(Na)/ZnCr₂O₄ and Pd/ZnCr₂O₄ compared with Pd(Na)/rGO-ZnCr₂O₄ (Table S1). Overall, the Na cations in catalyst mainly modulate the electronic configuration of Pd active sites and further form the Pd−OH_x on the surface of catalyst. Therefore, the Pd−OH_x in Pd(Na)/rGO-ZnCr₂O₄ is closely related to and responsible for the enhanced catalytic activity.

  The concentration of base sites on the catalyst surface is also a significant factor to affect the oxidation of BA [45], in which benzyl alcohol can be adsorbed, and converted into benzaldehyde. Therefore, it is necessary to investigate the effects of Na cations on the base sites of the catalyst surface. As a result, CO₂-TPD was performed and the results are shown in Fig. 4a and Fig. S9 (Supporting information). For CO₂-TPD curves of supports, the wide peak at a low temperature of 60–200 ℃ can be found, which is due to the weak base sites of rGO-ZnCr₂O₄ or (Na)rGO-ZnCr₂O₄ and the weak physical interaction between CO₂ and supports. Also, a weak and similar CO₂ desorption signal corresponding to the strong base sites can also be found at high temperature of 200–450 ℃. This implies that the base sites on the surface of supports are unaffected by alkali cations. After anchoring Pd on (Na)rGO-ZnCr₂O₄ and rGO-ZnCr₂O₄, the peak at a low temperature still appears and strong peak of CO₂ desorption corresponding to the strong base sites is observed at high temperature region. It can be inferred that the strong base sites are mainly attributed to the supported Pd nanoparticles. In addition, the total concentration of base sites for Pd(Na)/rGO-ZnCr₂O₄ (3.43 mmol/g) has a remarkable increase in comparison to that of Pd/rGO-ZnCr₂O₄ (2.34 mmol/g). The increased concentration of strong base sites is mainly attributed to the facts that the Na cations substantially enhance the dispersion of Pd nanoparticles and meanwhile facilitate the formation of the surface Pd−OH_x series. The strong base sites on the Pd(Na)/rGO-ZnCr₂O₄ surface are beneficial to the realization of excellent catalytic performance for BA oxidation.

  Figure 4

  

  Figure 4.  (a) CO₂ temperature-programmed desorption (CO₂-TPD) profiles on Pd(Na)/rGO-ZnCr₂O₄ and the corresponding Na-free catalyst. (b) CO-TPR profiles of Pd(Na)/rGO-ZnCr₂O₄ and the corresponding Na-free catalyst, which corresponds to MS signal: (c) H₂ and (d) CO₂.

  DownLoad: Full-Size Img PowerPoint

  In the process of BA oxidation, the −OH on the catalyst surface can be helpful for the dissociation −O−H bonds of the BA and decrease the energy barrier of α-C−H bonds of the ensuing alkoxide intermediate to achieve the conversion toward benzaldehyde [41]. In the past, it has been reported that the introduction of alkali cation in Au and Pt catalysts can induce the production of surface hydroxyl groups [12,13]. To further confirm the presence of surface hydroxyl groups, the CO temperature-programmed reduction (TPR) coupled with mass spectrometry to track the off-gas, an effective way of identifying the surface −OH and active lattice [O] over catalysts (2CO + 2[OH] → 2CO₂ + H₂, CO + [O] → CO₂), was performed and the results were exhibited in Fig. 4b. Compared to the Pd/rGO-ZnCr₂O₄, the Pd(Na)/rGO-ZnCr₂O₄ catalyst exhibits a similar TCD signal peak region at 85–310 ℃. However, the Na-containing catalyst has a significant increase toward CO consumption from 3.54 mmol/g to 7.26 mmol/g. Meanwhile, the CO-TPR experiments over the Pd/rGO-ZnCr₂O₄ and Pd(Na)/rGO-ZnCr₂O₄ catalysts were also carried out and the off-gas was traced by mass spectrometry, of which the results were presented in Figs. 4c and d. At a relatively low temperature (< 225 ℃), the produced H₂ and CO₂ molecules are mainly attributed to the reaction between CO and −OH species of catalyst surface. At high temperature of 225–350 ℃, it can also be found that the partial pressure of H₂ decreases while that of CO₂ increases. In this process, the decrease of H₂ partial pressure results from the gradual consumption of −OH over the catalyst surface with an increase of temperature. The increased CO₂ partial pressure comes from the reaction between CO and the active lattice [O] species on the catalysts surface. This indicates that the surface −OH over the catalysts can feature a higher activity than surface lattice [O] species at low temperature. It can be easily concluded that surface −OH has a stronger activity than surface reactive oxygen species, which manifests that the oxidation reaction of BA mainly follows the Langmuir-Hinshelwood mechanism rather than the Mars-van Krevelen mechanism involved in the active lattice [O] species. In addition, the Na-modified Pd(Na)/rGO-ZnCr₂O₄ shows a high partial pressure of H₂ and CO₂, and the peak position shifts to low temperature in comparison to Pd/rGO-ZnCr₂O₄ catalyst. This further indicates that the Na cations in catalyst benefit to provide the abundant −OH species. Further, the surface −OH species over Pd(Na)/rGO-ZnCr₂O₄ with high activity will easily bind the Pd species.

  In addition, according to the results of water injection experiment, the water promotes the oxidation reaction of BA. Thus, it is necessary to investigate the adsorption of catalysts toward water. To decipher the relationship between H₂O molecule and the Na cation modified Pd(Na)/rGO-ZnCr₂O₄, H₂O pulse adsorption experiments were carried out over Pd(Na)/rGO-ZnCr₂O₄ and the counterpart catalyst, and the results were presented in Fig. 5a. The Pd/rGO-ZnCr₂O₄ almost reaches adsorption saturation for H₂O after suffering from H₂O vapor pulse for three times. Comparably, the Na-modified Pd(Na)/rGO-ZnCr₂O₄ catalyst undergoes H₂O vapor pulse for seven times and can reach adsorption saturation. This suggests the enhanced adsorption of H₂O molecules in the presence of Na cations, which maybe result from the hydrogen bond interaction between H₂O molecules and the new appearing surface Pd−OH_x species. Further, the adsorption characteristics of BA on catalyst surface were investigated by UV-vis absorption experiments. The BA in the filtrate adsorbed by Na-modified Pd(Na)/rGO-ZnCr₂O₄ shows a lower absorbance than that of the Pd/rGO-ZnCr₂O₄ (Fig. 5b), which shows that the Na cations in catalyst can improve the absorption capacity of catalyst toward BA. To gain the detailed adsorption information of BA over catalyst surface, in situ Raman experiments were performed (Fig. 5c). For Pd(Na)/rGO-ZnCr₂O₄ and Pd/rGO-ZnCr₂O₄ without BA, the similar peaks at 193 (δ(Zn−O)), 1344 (D bond), and 1600 cm⁻¹ (G band) are observed. When BA is adsorbed on the Pd(Na)/rGO-ZnCr₂O₄, it can be found that there are four characteristic peaks of BA around 400 cm⁻¹ to 900 cm⁻¹. The peak at 485 cm⁻¹ corresponding to out-of-plane deformation of the phenyl ring for BA is almost unchanged. While the peak at 620 cm⁻¹, being in-plane bending vibration of the phenyl ring, shifts to low wavenumber of 601 cm⁻¹. Also, the −C−H bending vibration at the phenyl ring shifts from 750 cm⁻¹ to 784 cm⁻¹ and 807 cm⁻¹ to 827 cm⁻¹, respectively. The shifted peaks are from the distorted/twisted BA that is adsorbed on the catalyst, thus the corresponding reactive barrier of BA to benzaldehyde is reduced. For Pd/rGO-ZnCr₂O₄ catalyst with BA, the peak at 485 cm⁻¹ belonging to out-of-plane distorted phenyl ring can be recognized. The other peaks assigned to BA molecules are relatively weak in comparison to that of Pd(Na)/rGO-ZnCr₂O₄. It is easy to conclude that the adsorption of Pd/rGO-ZnCr₂O₄ for BA is relatively weak than that of Pd(Na)/rGO-ZnCr₂O₄ modified by the Na species. In other words, the Na-modified Pd(Na)/rGO-ZnCr₂O₄ catalyst can enhance the adsorption of H₂O molecule on account of the Pd−OH_x structure induced by Na cations compared to the corresponding Na-free sample. Also, the changed adsorption due to the distorted structure of BA and improved adsorption capacity for BA on the catalyst with Na species are synergistically beneficial for the conversion of BA. This is also visualized and shown in Fig. 5d.

  Figure 5

  

  Figure 5.  (a) H₂O pulse curves over the Pd(Na)/rGO-ZnCr₂O₄ and the corresponding Na-free sample. (b) UV-vis absorption spectra for aqueous solutions after BA is adsorbed over the Pd(Na)/rGO-ZnCr₂O₄ and the corresponding Na-free sample. (c) Raman spectra of BA, the various catalysts (Pd(Na)/rGO-ZnCr₂O₄ and Pd/rGO-ZnCr₂O₄) after adsorbing BA. (d) Adsorption kinetics models of BA and H₂O over the Pd(Na)/rGO-ZnCr₂O₄ and the corresponding Na-free sample.

  DownLoad: Full-Size Img PowerPoint

  For the oxidation of BA to benzaldehyde, the reaction process involves the cleavage of α−C−H and −O−H bonds of BA. It was confirmed that H₂O molecules in the reaction system feature the multiple effects during BA oxidation as follows (Fig. 6a): (ⅰ) Enhancing adsorption towards BA derived from being distorted and twisted BA on the surface of catalyst [7], (ⅱ) enhancing adsorption and activation of O₂ [46], (ⅲ) generating active ^*OH or ^*OOH formed by the reaction of H₂O and O₂, where the ^*OH or ^*OOH can further react with the hydrogen of α−C−H [36,41]. Also, when the catalyst is modified with a reasonable amount of alkali cations, the surface of the catalyst can feature the abundant high active Pd−OH_x species compared with Na-free catalyst, which has been confirmed by the previous experiment. Further, the catalyst with abundant high active Pd−OH_x exhibits more excellent catalytic performance for BA oxidation than Na-free catalyst under the same reaction conditions. It can be imagined that the −OH in alkali-stabilized Pd−OH_x would function as ^*OH or ^*OOH to some degree. In this case, the reaction can be further promoted, as discussed in the results. Based on this, the corresponding mechanism of alkali-promoted alcohol oxidation can be proposed following the Langmuir-Hinshelwood mechanism and the detailed reaction is schemed in Fig. 6b. The BA and H₂O linked by hydrogen bond interaction (^*H₂O···HO−CH₂−C₆H₅) are easily adsorbed on the surface of the catalyst [7]. The H₃O^* formed by the −O−H cleavage of ^*H₂O···HO−CH₂−C₆H₅ on active sites of catalyst surface combines with the activated oxygen decomposed by O₂ on the catalyst surface to form the H₂O molecule, ^*OH or ^*OOH species, and intermediate (IMT) C₆H₅−CH₂O^* [32,36]. While the hydrogen in the α−C−H of IMT would efficiently combine with the abundant high active Pd−OH_x or ^*OH and ^*OOH to form C₆H₅−CHO and H₂O. Moreover, it was reported that the consumed −OH for Pd−OH_x catalyst in the reaction can be supplemented via ^*OOH or ^*OH species derived from the reaction of H₃O^* and the activated oxygen decomposed by O₂. The roles of alkali in the catalyst are mainly to provide the abundant high active Pd−OH_x, resulting in excellent catalytic performance.

  Figure 6

  

  Figure 6.  (a) Schematic diagram for water-induced oxidation of BA. (b) Scheme of the detailed reaction for BA oxidation on the surface of abundant Pd−OH_x. For a better view, only one −OH on the Pd surface is drawn in the schematic.

  DownLoad: Full-Size Img PowerPoint

  In conclusion, the positive effects of alkali cations on Pd catalyst for oxidation of BA are investigated and deciphered in this work. The macroscopic kinetic experiments confirm that rational Na-containing catalyst can decrease the apparent activation energy of BA oxidation reaction from 93.1 kJ/mol to 46.3 kJ/mol resulting from the enhanced water-induced reaction process by the Na-modified catalyst. It is well recognized and identified that the Na cations over the catalyst can decrease the size of Pd nanoparticles and promote the dispersity of Pd active sites. The Na species are present over the catalyst in the form of Na−O linkage. Also, the Pd active sites show a high oxidation state compared with that of Na-free catalyst, and the surface of Na-modified catalyst presents the abundant −OH species by the formation of Pd−OH_x. This is helpful for the adsorption of water and BA, further leading to the distorted BA with high reaction activity. The abundant highly active Pd−OH_x species derived from the effects of Na cations can efficiently combine with the hydrogen of α−C−H for C₆H₅−CH₂O^* intermediate, which is responsible for excellent catalytic performance for BA oxidation. It is expected that the strategy of alkali cations-modified catalysts would provide an avenue to promote and enhance the chemical reactions involved in water-induced effects including CO oxidation, alcohol oxidation, hydrogenation reaction, etc.

  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

  Acknowledgments

  This work was supported by the National Natural Science Foundation of China (NSFC, Nos. 51872035 and 22078052) and the Innovation Program of Dalian City of Liaoning Province (No. 2019RJ03).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2022.107939.

  
  
• 1. [1]

     S. Crossley, J. Faria, M. Shen, D.E. Resasco, Science 327 (2010) 68–72. doi: 10.1126/science.1180769

  2. [2]

     D. Tian, X. Zhang, H. Shi, et al., J. Am. Chem. Soc. 143 (2021) 16641–16652. doi: 10.1021/jacs.1c07482

  3. [3]

     X. Zhang, Y. Hou, R. Ettelaie, et al., J. Am. Chem. Soc. 141 (2019) 5220–5230. doi: 10.1021/jacs.8b11860

  4. [4]

     P. Zhou, Q. Wang, C.L. Zhang, et al., Chin. Chem. Lett. 26 (2015) 657–661. doi: 10.1016/j.cclet.2015.04.006

  5. [5]

     Y. Hao, Y. Liu, R. Yang, et al., Chin. Chem. Lett. 29 (2018) 778–782. doi: 10.1016/j.cclet.2018.01.010

  6. [6]

     Y. Zhang, R. Ettelaie, B.P. Binks, H. Yang, ACS Catal. 11 (2021) 1485–1494. doi: 10.1021/acscatal.0c04604

  7. [7]

     Q. Wei, C. Yu, X. Song, et al., J. Am. Chem. Soc. 143 (2021) 6071–6078. doi: 10.1021/jacs.0c10618

  8. [8]

     X. Yang, X. Wang, C. Liang, et al., Catal. Commun. 9 (2008) 2278–2281. doi: 10.1016/j.catcom.2008.05.021

  9. [9]

     R. Qin, L. Zhou, P. Liu, et al., Nat. Catal. 3 (2020) 703–709. doi: 10.1038/s41929-020-0481-6

  10. [10]

      H. Wei, Y. Ren, A. Wang, et al., Chem. Sci. 8 (2017) 5126–5131. doi: 10.1039/C7SC00568G

  11. [11]

      B. Zugic, S. Zhang, D.C. Bell, F. Tao, M. Flytzani-Stephanopoulos, J. Am. Chem. Soc. 136 (2014) 3238–3245. doi: 10.1021/ja4123889

  12. [12]

      M. Yang, S. Li, Y. Wang, et al., Science 346 (2014) 1498–1501. doi: 10.1126/science.1260526

  13. [13]

      Y. Zhai, D. Pierre, R. Si, et al., Science 329 (2010) 1633–1636. doi: 10.1126/science.1192449

  14. [14]

      M. Yang, J. Liu, S. Lee, et al., J. Am. Chem. Soc. 137 (2015) 3470–3473. doi: 10.1021/ja513292k

  15. [15]

      T. Wang, Y. Xu, Y. Li, et al., ACS Catal. 11 (2021) 3553–3574. doi: 10.1021/acscatal.1c00169

  16. [16]

      J. Xie, P.P. Paalanen, T.W. van Deelen, et al., Nat. Commun. 10 (2019) 167. doi: 10.1038/s41467-018-08019-7

  17. [17]

      Z. Li, L. Zhong, F. Yu, et al., ACS Catal. 7 (2017) 3622–3631. doi: 10.1021/acscatal.6b03478

  18. [18]

      Q. Zhang, J. Bu, J. Wang, et al., ACS Catal. 10 (2020) 10350–10363. doi: 10.1021/acscatal.0c02730

  19. [19]

      S. Cao, J.R. Monnier, J.R. Regalbuto, J. Catal. 347 (2017) 72–78. doi: 10.1016/j.jcat.2017.01.009

  20. [20]

      S. Cao, Y. Zhao, S. Lee, et al., Sci. Adv. 6 (2020) eaba3809. doi: 10.1126/sciadv.aba3809

  21. [21]

      C. Wang, Y. Li, L. Zheng, et al., ACS Catal. 11 (2020) 456–465.

  22. [22]

      S. Cao, M. Yang, A.O. Elnabawy, et al., Nat. Chem. 11 (2019) 1098–1105. doi: 10.1038/s41557-019-0345-3

  23. [23]

      X. Chen, M. Chen, G. He, et al., J. Phys. Chem. C 122 (2018) 27331–27339. doi: 10.1021/acs.jpcc.8b07160

  24. [24]

      C. Zhang, F. Liu, Y. Zhai, et al., Angew. Chem., Int. Ed. 51 (2012) 9628–9632. doi: 10.1002/anie.201202034

  25. [25]

      D. Kiani, S. Sourav, J. Baltrusaitis, I.E. Wachs, ACS Catal. 11 (2021) 10131–10137. doi: 10.1021/acscatal.1c01392

  26. [26]

      Y. Li, X. Chen, C. Wang, C. Zhang, H. He, ACS Catal. 8 (2018) 11377–11385. doi: 10.1021/acscatal.8b03026

  27. [27]

      P. Zhai, C. Xu, R. Gao, et al., Angew. Chem., Int. Ed. 55 (2016) 9902–9907. doi: 10.1002/anie.201603556

  28. [28]

      Y. Li, C. Zhang, H. He, J. Zhang, M. Chen, Catal. Sci. Technol. 6 (2016) 2289–2295. doi: 10.1039/C5CY01521A

  29. [29]

      A. Li, X. Chang, Z. Huang, et al., Angew. Chem., Int. Ed. 55 (2016) 13734–13738. doi: 10.1002/anie.201605666

  30. [30]

      L. Ni, C. Yu, Q. Wei, J. Chang, J. Qiu, Green Chem. 22 (2020) 5711–5721. doi: 10.1039/d0gc01855d

  31. [31]

      C. Yu, L. Fan, J. Yang, Y. Shan, J. Qiu, Chem. Eur. J. 19 (2013) 16192–16195. doi: 10.1002/chem.201300949

  32. [32]

      J. Saavedra, H.A. Doan, C.J. Pursell, L.C. Grabow, B.D. Chandler, Science 345 (2014) 1599–1602. doi: 10.1126/science.1256018

  33. [33]

      C. Wang, X.K. Gu, H. Yan, et al., ACS Catal. 7 (2017) 887–891. doi: 10.1021/acscatal.6b02685

  34. [34]

      G. Croft, M.J. Fuller, Nature 269 (1977) 585–586. doi: 10.1038/269585a0

  35. [35]

      Z. Liu, E. Huang, I. Orozco, et al., Science 368 (2020) 513–517. doi: 10.1126/science.aba5005

  36. [36]

      C. Shang, Z.P. Liu, J. Am. Chem. Soc. 133 (2011) 9938–9947. doi: 10.1021/ja203468v

  37. [37]

      Z. Zhao, R. Bababrik, W. Xue, et al., Nat. Catal. 2 (2019) 431–436. doi: 10.1038/s41929-019-0257-z

  38. [38]

      H. Wang, X.K. Gu, X. Zheng, et al., Sci. Adv. 5 (2019) 6413. doi: 10.1126/sciadv.aat6413

  39. [39]

      C.M. Olmos, L.E. Chinchilla, A. Villa, et al., J. Catal. 375 (2019) 44–55. doi: 10.1016/j.jcat.2019.05.002

  40. [40]

      H. Huang, C. Yu, X. Han, et al., Energy Environ. Sci. 13 (2020) 4990–4999. doi: 10.1039/d0ee02607g

  41. [41]

      B.N. Zope, D.D. Hibbitts, M. Neurock, R.J. Davis, Science 330 (2010) 74–78. doi: 10.1126/science.1195055

  42. [42]

      J. Chen, J. Zhong, Y. Wu, et al., ACS Catal. 10 (2020) 10339–10349. doi: 10.1021/acscatal.0c03111

  43. [43]

      J. Lu, B. Fu, C. Kung Mayfair, et al., Science 335 (2012) 1205–1208. doi: 10.1126/science.1212906

  44. [44]

      M. Xu, S. Yao, D. Rao, et al., J. Am. Chem. Soc. 140 (2018) 11241–11251. doi: 10.1021/jacs.8b03117

  45. [45]

      J. Liu, S. Zou, L. Lu, et al., Catal. Commun. 99 (2017) 6–9. doi: 10.1016/j.catcom.2017.05.015

  46. [46]

      J. Saavedra, T. Whittaker, Z. Chen, et al., Nat. Chem. 8 (2016) 584–589. doi: 10.1038/nchem.2494

• 

  Figure 1  (a) Catalytic performance over Pd(Na)/rGO-ZnCr₂O₄ and Pd/rGO-ZnCr₂O₄. (b, c) Time-conversion plots at various temperatures for Pd(Na)/rGO-ZnCr₂O₄ and Pd/rGO-ZnCr₂O₄ catalyst, respectively. (d) Arrhenius plots for BA oxidation. (e) Response of water injection over the Pd(Na)/rGO-ZnCr₂O₄ and the corresponding Na-free catalyst for BA oxidation, with toluene as reaction solvent. (f) Catalytic activity test over Pd(K)/rGO-ZnCr₂O₄ and Pd/rGO-ZnCr₂O₄.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  HAADF-STEM images of (a) Pd/rGO-ZnCr₂O₄ and (b) Pd(Na)/rGO-ZnCr₂O₄ and the corresponding curves of Pd size distribution (inset). (a1, b1 and a2, b2) EDS element maps of Pd and Na for Pd(Na)/rGO-ZnCr₂O₄ and Pd/rGO-ZnCr₂O₄, respectively.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  The XPS spectra of (a) O 1s, (b) Na 1s and (c) Pd 3d for Pd(Na)/rGO-ZnCr₂O₄ and the corresponding Na-free catalyst. (d) In situ DRIFTS of CO chemisorption over Pd(Na)/rGO-ZnCr₂O₄ and the corresponding Na-free catalyst.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) CO₂ temperature-programmed desorption (CO₂-TPD) profiles on Pd(Na)/rGO-ZnCr₂O₄ and the corresponding Na-free catalyst. (b) CO-TPR profiles of Pd(Na)/rGO-ZnCr₂O₄ and the corresponding Na-free catalyst, which corresponds to MS signal: (c) H₂ and (d) CO₂.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) H₂O pulse curves over the Pd(Na)/rGO-ZnCr₂O₄ and the corresponding Na-free sample. (b) UV-vis absorption spectra for aqueous solutions after BA is adsorbed over the Pd(Na)/rGO-ZnCr₂O₄ and the corresponding Na-free sample. (c) Raman spectra of BA, the various catalysts (Pd(Na)/rGO-ZnCr₂O₄ and Pd/rGO-ZnCr₂O₄) after adsorbing BA. (d) Adsorption kinetics models of BA and H₂O over the Pd(Na)/rGO-ZnCr₂O₄ and the corresponding Na-free sample.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  (a) Schematic diagram for water-induced oxidation of BA. (b) Scheme of the detailed reaction for BA oxidation on the surface of abundant Pd−OH_x. For a better view, only one −OH on the Pd surface is drawn in the schematic.

  下载: 全尺寸图片 幻灯片
•