Reaction-Driven Restructuring of Rh-Pd and Pt-Pd Core-Shell Nanoparticles

See allHide authors and affiliations

Science  07 Nov 2008:
Vol. 322, Issue 5903, pp. 932-934
DOI: 10.1126/science.1164170


Heterogeneous catalysts that contain bimetallic nanoparticles may undergo segregation of the metals, driven by oxidizing and reducing environments. The structure and composition of core-shell Rh0.5Pd0.5 and Pt0.5Pd0.5 nanoparticle catalysts were studied in situ, during oxidizing, reducing, and catalytic reactions involving NO, O2, CO, and H2 by x-ray photoelectron spectroscopy at near-ambient pressure. The Rh0.5Pd0.5 nanoparticles underwent dramatic and reversible changes in composition and chemical state in response to oxidizing or reducing conditions. In contrast, no substantial segregation of Pd or Pt atoms was found in Pt0.5Pd0.5 nanoparticles. The different behaviors in restructuring and chemical response of Rh0.5Pd0.5 and Pt0.5Pd0.5 nanoparticle catalysts under the same reaction conditions illustrates the flexibility and tunability of the structure of bimetallic nanoparticle catalysts during catalytic reactions.

The development of bimetallic nanoparticles (NPs) with controlled size, composition, and structure opens enormous possibilities for engineering catalysts with enhanced activity and selectivity. Important technological areas, including catalytic reforming (1, 2), pollution control (1), alcohol oxidation (3), and electrocatalysis in fuel cells (4), are based on bimetallic catalytic systems (5, 6). Although it is known that the structure and composition of the surface of materials can be modified in response to changing reaction conditions (1, 7, 8), when the size of the material is at the nanometer scale, the changes can be much more dramatic as the distinction between surface and bulk regions fades away. In the case of NP catalysts, these changes can transform the material so that it has distinctive reactivity.

We demonstrated that bimetallic NP catalysts can undergo profound structural and chemical changes in response to reactive environments at ambient pressures. This was made possible by the use of an ambient-pressure x-ray photoelectron spectroscopy (APXPS) apparatus (9, 10) that can be used to obtain XPS data from materials exposed to gas pressures much higher than the usual limits imposed by high-vacuum conditions, recently up to ∼5 to 10 torr. With it, we could study in situ the structure and composition of core-shell Rh0.5Pd0.5 and Pt0.5Pd0.5 bimetallic NPs during catalytic reactions in different gas environments. The composition and distribution of the constituent elements within the shell of Rh0.5Pd0.5 NPs were found to change dramatically in response to changes in reactant gas composition, a result that demonstrates the structural flexibility of NPs and the interplay of structure and reactivity.

Rh0.5Pd0.5 and Pt0.5Pd0.5 bimetallic NPs with diameters of 15 ± 2 nm were synthesized by means of colloidal chemistry methods [see supporting online material (SOM) (11)] and characterized by transmission electron microscopy and x-ray diffraction. NPs (Rh0.5Pd0.5 or Pt0.5Pd0.5) were deposited on the oxidized surface of silicon wafers to form model catalysts for XPS studies. The spectra were obtained at x-ray photon energies of 645 and 850 eV at the Advanced Light Source at the Lawrence Berkeley National Laboratory (beamline 9.3.2), the Berkeley Synchrotron Facility, and in ultrahigh vacuum (UHV) using Al Kα x-rays at 1486.6 eV in a laboratory Phi 550 instrument. The mean free paths (MFPs) of Rh3d and Pd3d photoelectrons excited at these three x-ray energies were approximately 0.7, 1.0, and 1.6 nm, respectively (12). The structure of the Pt0.5Pd0.5 NPs was studied at photon energies of 630, 880, and 1486.6 eV for Pd3d, and at 350, 630, and 1486.6 eV for Pt4f. The MFPs of the generated Pd3d and Pt4f photoelectrons were also 0.7, 1.0, and 1.6 nm, respectively (12).

Before their use in catalytic reactions, the as-synthesized Rh0.5Pd0.5 NPs were found to be Rh-rich in the surface layers (Fig. 1A). At the lowest photon energy [hν = 645 eV, kinetic energy (KE) = ∼335 eV], corresponding to a MFP of ∼0.7 nm, the measured atomic fraction of Rh was 0.93 ± 0.03. The determination of atomic fractions of Rh and Pd is described in the SOM (11). The average Rh atomic fraction decreased to 0.86 ± 0.03 and 0.52 ± 0.03 at depths corresponding to MFPs of 1.0 nm (hν = 850 eV, KE = ∼540 eV) and 1.6 nm (hν = 1486.6 eV, KE = ∼1180 eV), respectively. Here we use the term “shell” to refer to the three to four atomic layers within 0.7 nm of the surface. The section between 0.7 and 1.6 nm is referred to as the “intermediate layer,” and the section from 1.6 nm to the center of a NP is referred to as the “central core.” The volume ratio of material from the surface to the boundary of each of these three regions is respectively 25, 35, and 51%.

Fig. 1.

(A) Dependence of Rh and Pd atomic fractions of as-synthesized Rh0.5Pd0.5 NPs measured at 25°C in UHV as a function of photoelectron KE and mean free path. Detected photoelectrons with a KE of ∼1152 eV originate from the shell, intermediate layer, and central core of the Rh0.5Pd0.5 NPs. The signal of photoelectrons with KE of ∼335 and ∼540 eV is mainly contributed by the atoms in surface layers because of the shorter MFP of these photoelectrons. (B) Dependence of Pd and Pt atomic fractions of the as-synthesized Pt0.5Pd0.5 measured in UHV at 25°C as a function of the KE and mean free path of the excited photoelectrons. Schematics showing the core-shell structures of the Rh0.5Pd0.5 and Pt0.5Pd0.5 NPs are included (these schematics do not represent the shape of the NPs). The y-axis data points have an associated error of ±0.03.

Pt0.5Pd0.5 NPs have a similar core-shell structure, but the shell is Pd-rich (Fig. 1B). The observed atomic fraction of Pd in these NPs was 0.84 ± 0.03, 0.67 ± 0.03, and 0.52 ± 0.03 within the MFP distances of 0.7, 1.0, and 1.6 nm, respectively.

After the initial characterization, we measured the changes in surface composition and chemical state of the Rh0.5Pd0.5 NPs under oxidizing (100 mtorr NO or O2), catalytic (100 mtorr NO and 100 mtorr CO reacting to produce N2 and CO2), and reducing (100 mtorr CO or H2) conditions using APXPS. The atomic fractions presented in Fig. 2 were obtained with an x-ray energy of 645 eV and thus represent the changes in composition in the 0.7-nm shell. The top part of Fig. 2 shows a substantial oscillation of the relative atomic fractions as the gas environment changed from oxidizing to catalytic at 300°C. After oxidation by 100-mtorr NO, the Rh in the shell was almost completely oxidized, with ∼94% of the Rh in oxide form. When 100-mtorr CO was added to the 100-mtorr NO to produce CO2 and N2 (catalytic conditions), the total Rh atomic fraction in the shell decreased from 0.92 ± 0.03 to 0.46 ± 0.02 and that of Pd increased from 0.08 ± 0.03 to 0.54 ± 0.02 (reaction 2 in top part of Fig. 2). This result indicates that a drastic restructuring of the shell and core of the NP took place, where Pd migrated to the shell and Rh migrated to the core. The production of CO2 and N2 formed in the reaction was measured by sampling gas into a mass spectrometer located in a separate UHV chamber, indicating that the NPs were active during CO and NO catalytic conversion. The ∼1.0-eV downshift in binding energy of the Rh3d core level indicates the reduction of RhOy to metal Rh0. The Rh that remained in the shell was ∼76% metallic during the catalytic reaction (reaction 2), in contrast to the ∼94% RhOy during the oxidizing reaction. The reduction of RhOy to metallic Rh shows a substantial chemical response accompanying the change in atomic distribution.

Fig. 2.

(Top) Evolution of Rh (Rh0 + Rh2y+) and Pd (Pd0 + Pd2y+) atomic fractions in the Rh0.5Pd0.5 NPs at 300°C under oxidizing conditions (100 mtorr NO or O2) and catalytic conditions (100 mtorr NO and 100 mtorr CO) denoted in the x axis. (Bottom) Evolution of the fraction of the oxidized Rh (left y axis) and Pd atoms (right y axis) in the examined region under the same reaction conditions as the top part of the figure. All atomic fractions in this figure were obtained with an x-ray energy of 645 eV for Rh3d and Pd3d, which generates photoelectrons with a MFP of ∼0.7 nm. Schematic diagrams above the top of the figure show the reversible segregation of Rh and Pd under alternating oxidizing and catalytic conditions. The y-axis data points for reactions 1, 3, and 5 have an associated error of ±0.03; for reactions 2 and 4, the error bar is ±0.02.

The observed changes in atomic distribution and chemical state are reversible and depend on the composition of the surrounding reactive gases, as shown in a sequence of five reactions in Fig. 2. If CO is removed while NO remains in the reactor chamber (reaction 3 in Fig. 2), Rh diffuses back to the shell and becomes substantially oxidized. Analysis of the XPS peaks shows that the reconstructed shell contains ∼72 ± 3% Rh (top part of Fig. 2), of which ∼90% is oxidized (bottom panel). If CO is introduced again, added to the NO gas phase (reaction 4), the spatial distribution and chemical state of the atoms are restored to those during reaction 2. The restructuring of the shell and core of the Rh0.5Pd0.5 NPs can be repeated several times by changing the gas composition from oxidizing (NO or O2) to reducing (CO) or catalytic (CO + NO) conditions as shown in Fig. 2 and fig. S3 (11). If another reducing gas, hydrogen, is used instead [reaction 2 in fig. S4 (11)], the RhOy is reduced, with the Rh atoms migrating back to the core and the Pd atoms to the shell, forming a Pd-rich shell similar to that formed in pure CO [fig. S3 (11)]. When Al Kα x-rays at 1486.6 eV were used, the Rh fraction in the Rh0.5Pd0.5 NPs measured in UHV after alternating oxidizing and reducing reactions [fig. S4 (11)] was 0.36 ± 0.03, much lower than the 0.52 ± 0.03 value of the as-synthesized NPs before reaction also measured in UHV with Al Kα x-rays (Fig. 1A). This difference indicates that the core region (at depth greater than 1.6 nm) also participates in the restructuring of the NPs.

The opposite segregation behavior of Rh and Pd under oxidizing and reducing conditions can be explained by considering the surface energy in the metals and in the oxides. The lower surface energy of Pd relative to Rh tends to drive Pd metal atoms to the surface (1315). The fact that the Rh oxide is more stable than the Pd oxide provides the driving force for the segregation and preferential oxidation of Rh at the surface (16). When a reducing gas, CO, is added to NO, the oxides are reduced to the metallic state and the oxygen atoms react with adsorbed CO to form CO2 and desorb. Because of their higher surface free energy, the Rh atoms migrate to the core, thereby decreasing the atomic fraction of Rh in the shell under reducing and catalytic conditions.

In the Pt0.5Pd0.5 NPs, synthesized with the same procedure described in the SOM, the shell region is substantially richer in Pd. Pd has lower surface energy and is a more reactive metal than Pt (15, 16). Under oxidizing conditions, it forms a shell with 92% PdOy without substantial segregation of Pt atoms as compared to the as-synthesized Pt0.5Pd0.5 NPs (Figs. 3 and 1B). During catalytic reaction and under reducing conditions, the PdOy is substantially reduced (bottom part of Fig. 3). There is no obvious segregation under these conditions. Compared to Rh0.5Pd0.5, however, the Pt0.5Pd0.5 NPs do not exhibit the strong segregation and reversibility characteristic of the Rh0.5Pd0.5 NPs as the gas composition changes sequentially from oxidizing, to catalytic, to reducing as shown in the top part of Fig. 3. Because Pt is much less easily oxidized (16) and has higher surface energy (15), the Pt atoms are not pinned to the surface by the formation of oxide. Thus, there is no substantial atomic reorganization in reactive environments.

Fig. 3.

(Top) Evolution of the Pd and Pt atomic fractions in Pt0.5Pd0.5 NPs at 300°C under oxidizing (100 mtorr NO), catalytic (100 mtorr NO and 100 mtorr CO), and reducing (100 mtorr CO) conditions. The x axis represents the different gas environments. (Bottom) Evolution of the atomic fraction of the oxidized Pd atoms in the examined region under the same reaction conditions as the top panel. All atomic fractions were obtained at an x-ray energy of 350 eV for Pt4f (KE ∼280 eV) and 630 eV for Pd3d (KE ∼ 280 eV). The y-axis error in the data is ±0.03.

The restructuring phenomenon observed in the bimetallic NPs induced by changes in reactive gas offers an interesting way of engineering the nanostructure of NPs for catalysis and other applications. One goal could be the synthesis of “smart” catalysts whose structure changes advantageously depending on the reaction environment. Our results suggest that the combination of a tunable colloid chemistry-based synthesis, followed by the controllable engineering of the structure of NPs with the use of reactive gases, opens a new door for designing new catalysts and shaping the catalytic properties of nanomaterials by structural engineering in reactive environments.

Supporting Online Material

Materials and Methods

SOM Text

Figs. S1 to S5


References and Notes

View Abstract

Stay Connected to Science

Navigate This Article