Performance evaluation of Pt and Pd mono- and bi-metallic H₂-SCR catalysts for NO emission control applications

Abstract

Hydrogen-based selective catalytic reduction (H₂-SCR) offers a promising pathway for mitigating NOx emissions, particularly in hydrogen internal combustion engine (H₂-ICE) applications. This study investigates how noble-metal identity and Pd–Pt interactions influence the H₂-SCR of NO over CeO₂–ZrO₂-supported catalysts. Monometallic catalysts show that Pd achieves catalytic activity comparable to Pt while providing higher N₂ selectivity, indicating a more favourable balance between hydrogen activation and selective NO reduction. Incorporation of both metals modifies this activity–selectivity relationship. Comparison with a physical mixture of the corresponding monometallic catalysts shows that simple coexistence of Pd- and Pt-containing particles does not reproduce the behaviour of the co-impregnated bimetallic catalyst. H₂-TPR reveals modified reduction behaviour in the bimetallic system, while CO-DRIFTS shows changes in the CO adsorption environment of Pd-containing surface sites in the presence of Pt. Collectively, these findings support the presence of modified Pd/Pt surface environments, potentially involving close Pd–Pt proximity or interfacial interactions, which govern hydrogen utilization and the NO conversion–N₂ selectivity balance during H₂-SCR.

---

Introduction

50 billion tons of GHGs are emitted on a yearly basis, contributing to global warming and environmental degradation.¹ In response, regulatory frameworks have become more stringent to minimize the impact of GHG emissions and accelerate the energy transition.² Therefore, substantial technological and infrastructural efforts are required to meet this target, particularly in sectors with high carbon intensity.

Around 24% of global CO₂ emissions are produced by the transportation sector, which rely heavily on fossil fuels.³ The use of electric powertrains is a promising pathway for decarbonization; however, areas with limited access to charging stations, heavy-duty vehicles and long-distance travel represent key challenges.^4,5 A promising alternative is the hydrogen internal combustion engine (H₂-ICE), which is carbon-free and can be achieved with minimal modifications to existing engines.⁶ However, the high temperature of H₂ combustion results in the formation of nitrogen oxides (NOx), including nitric oxide (NO), nitrogen dioxide (NO₂) and nitrous oxide (N₂O). They are considered a major health concern to humans and the environment due to their ability to damage the respiratory system and cause cardiovascular disease, as well as their contribution to acid rain and ground-level ozone.⁷

Selective catalytic reduction (SCR) is an efficient NOx emission treatment technology, and several reducing agents have been investigated, including ammonia (NH₃-SCR),^8–10 hydrocarbons (HC-SCR),^11–13 and hydrogen (H₂-SCR).^14–17 Among these, NH₃-SCR is widely employed in the transportation sector. Despite its potential, NH₃-SCR faces several challenges, such as ammonia slip, air heater fouling, and reduced performance during startup, partly due to its effectiveness at high temperatures. In the context of H₂-ICE, H₂ can also be utilized as an environmentally-friendly and low temperature reductant, to selectively convert NOx into nitrogen (N₂) and water (H₂O) in the presence of oxygen (O₂). However, minimizing NOx emissions from H₂-ICE requires a high-performing aftertreatment system that can sustain both high NO conversion and N₂ selectivity under a wide range of operating conditions to satisfy the stringent emission regulations. Therefore, the development of a robust and reliable H₂-SCR system, capable of efficient operation across various operating conditions, is critical.

Noble metals, particularly Pt and Pd, supported on oxide or zeolitic materials have been widely investigated for H₂-SCR because of their ability to activate H₂ and promote NOx reduction under lean conditions.^18–20 Recent studies have shown that the identity of the noble metal strongly affects the temperature window and product selectivity. Shao et al. compared Pt-, Pd-, and Ir-based SSZ-13 catalysts and showed that these metals exhibit different effective temperature ranges in H₂-SCR, while combining noble-metal functions can broaden the overall operating window for NO reduction by H₂.²¹ This highlights the importance of noble-metal selection and motivates further comparison of Pt- and Pd-based catalysts under relevant H₂-SCR conditions.

In addition to the active metal, the catalytic performance is strongly influenced by the support and by the strength of the metal–support interaction. Recent work on Pt-based H₂-SCR catalysts has shown that support identity, promoter chemistry, metal loading, pretreatment, space velocity, and feed composition can significantly alter NOx conversion, H₂ activation, and N₂ selectivity.²² Maurer et al. demonstrated that Pt nanoparticles on suitable oxide supports are essential for low-temperature NOx conversion, while support acidity/basicity and promoter addition can shift or broaden the activity and selectivity windows.²³ Similarly, Li et al. reported that Zr modification of Pt/TiO₂ enhances NOx conversion and N₂ selectivity in a wider low-temperature range, showing that tuning the support composition can regulate the Pt electronic state, adsorption behaviour, and the balance between NOx reduction and competitive H₂ oxidation.²⁰

Among oxide supports, CeO₂–ZrO₂ mixed oxides are particularly attractive because they combine the redox properties of ceria with the enhanced oxygen mobility and thermal stability provided by zirconia.²⁴ The reversible Ce³⁺/Ce⁴⁺ redox cycle can influence oxygen vacancy formation, NOx adsorption, and surface oxygen mobility, while Zr incorporation improves the structural and redox stability of ceria-based materials.²⁵ For instance, CeO₂-based Pd catalysts investigated by Patel and Sharma showed improved H₂-SCR performance upon ZrO₂ addition, which was associated with changes in the Ce³⁺/Ce⁴⁺ ratio and oxygen-related surface properties.²⁴ Kalamaras et al. also reported that Pt/Ce_xZr_1−xO_2−δ catalysts are active for NO reduction using H₂ and C₃H₆ as reducing agents, with Pt/Ce_0.5Zr_0.5O₂ showing the highest NO conversion.²⁶

Bimetallic systems have attracted increasing attention because combining two metals can modify hydrogen activation, NOx adsorption, and reaction selectivity beyond what is achieved with monometallic catalysts. Au addition to Pd/TiO₂ has been reported to improve NO conversion, while Co promotion of Pd/TiO₂ enhanced both NO conversion and N₂ selectivity through modified reaction pathways involving H₂ activation and adsorbed nitrogen-containing intermediates.^27,28 These findings suggest that metal–metal interactions, together with the support environment, can play a decisive role in controlling the activity–selectivity balance in H₂-SCR. Therefore, investigating bimetallic interactions on CeO₂–ZrO₂ is a promising approach for understanding how metal identity, metal–metal interactions, and support redox properties collectively control the activity–selectivity balance in H₂-SCR.

In this work, we investigated the effect of bimetallic systems supported on CZ on NOx reduction using H₂. Pd–Pt catalysts with varying metal compositions (Pd/CZ, 0.7Pd–0.3Pt/CZ, 0.5Pd–0.5Pt/CZ, 0.3Pd–0.7Pt/CZ, Pt/CZ) were prepared to investigate how the interaction of Pd and Pt influences the catalytic performance of H₂-SCR.

Experimental

Catalyst preparation

An aqueous solution of 2.5 g cerium(III) nitrate hexahydrate [Ce(NO₃)₃] (Sigma Aldrich) and 3.5 g zirconyl(IV) nitrate hydrate [ZrO(NO₃)₂·xH₂O] (Thermo Scientific) was prepared by dissolving the salts in deionized water and stirred for 2 h at room temperature. A separate recipient containing 20 mL of ammonium hydroxide (NH₄OH, pH = 10.5) was used for co-precipitation. The mixed metal nitrate solution was loaded into a syringe pump and injected simultaneously with NH₄OH (Sigma-Aldrich, 97%) into the vessels using a Gilson pump under continuous stirring. During the co-injection, the formation of brownish/yellowish precipitates was observed. The resulting precipitates were stirred for an additional 30 min at 450 rpm, and then separated by centrifugation at 7500 rpm for 7 min. The precipitate was washed three times with deionized water, with 10 min of sonication between each wash. The cleaned precipitate was dried at 100 °C overnight, ground into a fine powder, and then calcined at 500 °C for 6 h at a heating rate of 2 °C min⁻¹, resulting in a final composition of 50 wt% of each metal oxide. This co-precipitation method ensures homogeneous and well-mixed oxide formation by maintaining a constant pH during the synthesis.¹⁸

The active metal was introduced into the oxide support using the conventional incipient wetness impregnation method. 17.2 mg of palladium(II) chloride (PdCl₂) (Alfa Aesar) and 29.7 mg of sodium hexachloroplatinate(IV) (Na₂PtCl₆) (Thermo Fisher Scientific) were dissolved in a mixture of deionized water and hydrochloric acid (HCl) for Pd/CZ and Pt/CZ, respectively. For bimetallic catalysts, appropriate amounts of Pd and Pt precursors were dissolved corresponding to each Pd/Pt ratio. The resulting solution was added dropwise onto the oxide support under continuous stirring to ensure uniform metal dispersion. The impregnated samples were freeze-dried overnight to remove moisture. The synthesized catalysts have a total metal loading of 0.5 wt% and are denoted as Pd/CZ, 0.7Pd–0.3Pt/CZ, 0.5Pd–0.5Pt/CZ, 0.3Pd–0.7Pt/CZ, and Pt/CZ, reflecting their respective mass Pd/Pt compositions. Additionally, a physical mixture of 0.5%Pd/CZ and 0.5%Pt/CZ was prepared by mechanically mixing the corresponding monometallic catalysts to compare with the impregnated 0.5Pd–0.5Pt/CZ catalyst.

Catalyst characterization

Nitrogen physisorption was used to determine the specific surface area, pore volume, and pore size distribution of the samples using an ASAP 2420 surface area and porosity analyzer (Micromeritics). Prior to measurement, all samples were degassed at 350 °C for 4 h using a heating rate of 10 °C min⁻¹. The Brunauer–Emmett–Teller (BET) method was utilized to calculate the specific surface area over a relative pressure range (P/P₀) of 0.05–0.30. The pore volume and pore size distribution were determined using the Barrett–Joyner–Halenda (BJH) method.

X-ray diffraction (XRD) analysis was performed using a Bruker D8 Advance diffractometer equipped with a Cu Kα radiation source (λ = 0.15406 nm). The measurements were conducted over a 2θ range of 10–90°, with a scanning rate of 5° min⁻¹ and a step size of 0.02°.

H₂ temperature-programmed reduction (H₂-TPR) was used to investigate the effect of platinum group metal addition on the reduction behaviour of the catalysts, using an AutoChem II Chemisorption Analyzer (Micromeritics) equipped with a thermal conductivity detector (TCD). Prior to analysis, 0.1 g of catalyst was pretreated under a He flow (50 mL min⁻¹) at 350 °C for 1 h, followed by cooling to 35 °C. A gas mixture of 10% H₂/Ar (50 mL min⁻¹) was then introduced, and the temperature was increased to 700 °C at a heating rate of 10 °C min⁻¹.

CO diffuse reflectance infrared Fourier transform spectroscopy (CO-DRIFTS) measurements were performed to investigate the nature of the accessible surface metal sites. Prior to CO adsorption, the catalyst was treated under a N₂ flow (50 mL min⁻¹) at 300 °C for 1 h using a heating rate of 10 °C min⁻¹. The sample was then reduced under a 6% H₂/Ar flow (50 mL min⁻¹) at 300 °C for 45 min, followed by purging with N₂ at the same temperature for 45 min. Subsequently, the sample was cooled to room temperature under a N₂ flow, and a background spectrum was recorded. CO adsorption was carried out by exposing the catalyst to 3% CO/N₂ (30 mL min⁻¹) for 45 min. After adsorption, the sample was purged with N₂ (50 mL min⁻¹) for 45 min to remove weakly adsorbed and gas-phase CO species, after which the DRIFTS spectrum was collected.

CO chemisorption was carried out using 0.1 g of catalyst in an AutoChem III Chemisorption Analyzer (Micromeritics) coupled with a TCD. Prior to CO pulse adsorption, the catalyst was reduced under a 10% H₂/Ar flow (75 mL min⁻¹) at 225 °C for 30 min, followed by cooling to 35 °C. CO pulse injections were then performed using 10% CO/He at a flow rate of 75 mL min⁻¹ until saturation was reached. For the monometallic Pd/CZ and Pt/CZ catalysts, metal dispersion and average particle size were determined based on the amount of chemisorbed CO, assuming metal-to-CO stoichiometric ratios (M/CO) of 2.0 and 1.0 for Pd/CZ and Pt/CZ, respectively. Quantitative metal dispersion and particle size were not determined for the bimetallic catalysts because the CO adsorption stoichiometry cannot be reliably assigned for mixed Pd–Pt surface sites. Metal dispersion was calculated using the following equation:

where V_m is the cumulative quantity (cm³ g⁻¹ STP), W_a is the metal molar mass (g mol⁻¹), F_s is the M/CO ratio, V_mol is the standard molar volume (cm³ mol⁻¹) and M_loading is the total metal loading.

The metal loading and support composition were determined using inductively coupled plasma optical emission spectroscopy (ICP-OES). Transmission electron microscopy (TEM) and TEM-energy-dispersive X-ray spectroscopy (TEM-EDX) were employed as complementary techniques to examine the catalyst morphology and detect the supported noble-metal elements using a Titan 60-300 instrument (FEI Co., Netherlands). High-resolution X-ray photoelectron spectroscopy (XPS) was performed using an Axis Ultra spectrometer to investigate the surface chemical states. The resulting spectra were further analyzed by peak fitting using CasaXPS software.

Catalytic activity measurements

H₂-SCR catalytic activity measurements were conducted in a quartz packed-bed reactor with an internal diameter of 10 mm, operated at atmospheric pressure over a temperature range of 150–450 °C. In each experiment, the reactor was loaded with 250 mg of catalyst (35–60 mesh) positioned between two layers of quartz wool. The total flow rate of the reaction feed was maintained at 100 mL min⁻¹, corresponding to a gas hourly space velocity (GHSV) of 50 000 h⁻¹. The feed composition consisted of 450 ppm NO, 0.5% H₂, 2.5% O₂, and balance N₂. The H₂-SCR experiments were conducted using a Microactivity Reference system from PID Eng&Tech. The PID system includes an internal thermocouple for temperature monitoring, and Bronkhorst mass flow controllers (MFCs) connected to the PID system for precise control of the reaction conditions. Reactants and products were analyzed using MultiGas™ 2030 FTIR Gas Analyzers (MKS) and a TRACE™ 1600 Series Gas Chromatograph (Thermo Fisher Scientific). FTIR was employed for the analysis of NO, NO₂, N₂O, NH₃, and H₂O, while GC was used to quantify H₂ and O₂. N₂ selectivity was calculated using eqn (3) based on the outlet concentrations of NO, NO₂, N₂O, and NH₃ measured by FTIR. Prior to each H₂-SCR measurement, the catalyst was pretreated in a N₂ flow of 200 mL min⁻¹ for 2 h at 450 °C, followed by a reduction step of 20 mL min⁻¹ of 4% H₂/N₂ gas flow for 1 h at 225 °C. The NO conversion, N₂ selectivity and H₂ conversion were calculated as follows:

Results and discussion

Catalytic activity measurements

Fig. 1 shows the NO conversion (Fig. 1a) and N₂ selectivity (Fig. 1b) for Pd/CZ, 0.7Pd–0.3Pt/CZ, 0.5Pd–0.5Pt/CZ, 0.3Pd–0.7Pt/CZ, and Pt/CZ catalysts in the 150–450 °C range at a GHSV of 50 000 h⁻¹. Pt/CZ exhibits a superior NO conversion of over 92% in the low temperature range (150–200 °C) with a maximum NO conversion of 98% at 175 °C. This correlates with the H₂ conversion shown in Fig. 1c, suggesting that Pt/CZ exhibits strong H₂ activation below 200 °C, which contributes directly to the high NO reduction efficiency under these conditions. However, its catalytic activity drops above 200 °C, most likely due to the complete depletion of H₂ by the unselective oxidation with O₂. Pt/CZ also exhibits a significant decline in N₂ selectivity at low temperature, dropping to 45% at 200 °C, despite achieving complete H₂ conversion.

Fig. 1 (a) NO conversion, (b) N₂ selectivity and (c) H₂ conversion over xPd–(1 − x)Pt/CZ catalysts, where x = 0–1. Reaction conditions: 450 ppm NO, 0.5%H₂, 2.5% O₂, balance N₂, total flow rate = 100 mL min⁻¹, GHSV = 50 000 h⁻¹.

In contrast, Pd/CZ maintains an excellent NO conversion of over 80% in the range of 150–250 °C and relatively high N₂ selectivity. A comparison between the temperatures of peak NO conversion and the onset of full H₂ conversion reveals that the decline in NO conversion at higher temperatures is due to increased competitive H₂ combustion, which consumes the reducing agent (H₂) needed for the H₂-SCR reaction.

On the other hand, bimetallic catalysts display a different trend, with Gaussian-shaped NO conversion profiles peaking in the 200–250 °C range. Among them, 0.7Pd–0.3Pt/CZ shows the highest NO conversion with roughly 80% in the 175–275 °C range. Bimetallic catalysts also present high overall N₂ selectivity, with the 0.5Pd–0.5Pt/CZ sample showing the best performance, maintaining over 75% N₂ selectivity between 150 and 350 °C. All catalysts exhibit a decline in both NO conversion and N₂ selectivity at higher temperatures. This decrease is attributed to the depletion of available H₂ due to its preferential reaction with oxygen, leading to the formation of H₂O rather than N₂. At elevated temperatures, H₂ depletion leads to NO₂ formation originating from NO oxidation as it can be seen in Fig. S1b. Interestingly, complete H₂ consumption over the monometallic catalysts begins at temperatures of 100–150 °C lower than those observed for their bimetallic counterparts. This shift indicates that the existence of a Pt–Pd system delays H₂ oxidation, suggesting a synergistic interaction between the two noble metals that modifies the surface properties and redox behaviour of the catalyst.²⁷

Considering the distinct catalytic behaviour of the Pd–Pt-containing catalysts, we further examined whether the simultaneous presence of Pd and Pt in separate catalyst particles could reproduce the performance of the co-impregnated bimetallic catalyst. For this purpose, a physical mixture of the same mass of 0.5 wt% Pd/CZ and 0.5 wt% Pt/CZ was tested alongside the co-impregnated 0.5Pd–0.5Pt/CZ catalyst under identical reaction conditions (Fig. 2). However, the physical mixture exhibits conversion and selectivity profiles that differ from those of the individual monometallic catalysts, indicating that the coexistence of Pd- and Pt-containing particles influences the overall H₂-SCR behaviour.

Fig. 2 (a) NO conversion and N₂ selectivity and (b) H₂ conversion over 0.5Pd–0.5Pt/CZ catalyst and physical mixture of 0.5%Pd/CZ and 0.5%Pt/CZ. Reaction conditions: 450 ppm NO, 0.5%H₂, 2.5% O₂, balance N₂, total flow rate = 100 mL min⁻¹, GHSV = 50 000 h⁻¹. Squares with solid lines represent the impregnated 0.5%(0.5Pd–0.5Pt)/CZ catalyst and circles with dashed lines present the physical mixture of 0.5%Pd/CZ and 0.5%Pt/CZ catalysts. Black and red lines in (a) represent NO conversion and N₂ selectivity, respectively.

This observation prompted further investigation of whether co-impregnation alters the reduction behaviour, oxidation state and surface adsorption properties of the noble-metal species. Accordingly, various characterization analyses, including H₂-TPR and CO-DRIFTS, were used to probe the differences between the physical mixture and the co-impregnated Pd–Pt/CZ catalyst and to obtain further insight into the nature of the Pd–Pt interaction.

Textural properties

Table 1 summarizes the textural properties of the mesoporous catalysts, including the BET surface area (56–79 m² g⁻¹), pore volume (0.058–0.074 cm³ g⁻¹) and pore diameter (3.7–4.3 nm). All catalysts exhibit type IV isotherms with N₂ adsorption–desorption hysteresis loops, consistent with mesoporous structures according to the IUPAC classifications.²⁹ The N₂ adsorption–desorption isotherm of the CeO₂–ZrO₂ support remains unchanged after the introduction of Pd and Pt via impregnation, suggesting that the mesostructure is preserved. This finding, further supported by XRD results in Fig. 3, confirms that the structural integrity is maintained, ensuring a reliable basis for comparison among the catalysts.

Table 1 Textural properties of xPd–(1 − x)Pt/CZ catalysts, where x = 0–1

Catalyst	SBET (m^2 g^−1)	Vp (cm^3 g^−1)	DP (nm)
Pd/CZ	70	0.069	3.77
0.7Pd–0.3Pt/CZ	66	0.059	3.69
0.5Pd–0.5Pt/CZ	56	0.061	4.26
0.3Pd–0.7Pt/CZ	56	0.058	3.93
Pt/CZ	79	0.074	3.72
CZ	74	0.063	3.53

---

Fig. 3 XRD patterns for CeO₂, ZrO₂ and xPd–(1 − x)Pt/CZ catalysts, where x = 0–1.

All catalysts display mesoporous structures (D_P = 3.7–4.3 nm), typical of CeZrO_x-based materials.^26,30 The monometallic catalysts exhibit relatively higher BET surface areas (70–79 m² g⁻¹) compared to the bimetallic ones (56–66 m² g⁻¹), likely due to partial pore blockage by the metal nanoparticles. All catalysts exhibit close textural properties including surface area, pore size, and pore volume. As a result, the variations observed in the catalytic performance cannot be ascribed to differences in diffusional properties.

Fig. 3 shows the XRD patterns for CeO₂, ZrO₂, CZ, and xPd–(1 − x)Pt/CZ catalysts (where x = 0–1). No distinct ZrO₂ diffraction peaks are observed, suggesting successful incorporation into the ceria framework to form a solid solution. All samples display a single cubic CeO₂–ZrO₂ crystalline phase consistent with the standard reference (PDF #04-005-9597).³¹ The main diffraction peaks of the crystal structure (CeO₂) appear at 2θ = 28.5°, 33.1°, 47.6°, 56.4°, and 69.4° corresponding to the (111), (200), (220), (311), and (400) planes, respectively. In contrast, the peaks for the CeO₂–ZrO₂ solid solution are slightly shifted to higher angles (2θ = 29.6°, 34.0°, 49.1°, 58.5° and 70.9°), indicating the successful incorporation of Zr⁴⁺ into the CeO₂ lattice. This shift is attributed to the smaller ionic radius of Zr⁴⁺ (0.84 Å) compared to Ce⁴⁺ (0.97 Å).²⁴ No diffraction peaks corresponding to Pd or Pt are detected, likely due to the low metal loading or high dispersion of these metals.

The influence of redox properties is investigated by H₂-TPR experiments, and the resulting reduction profiles are presented in Fig. 4. For all metal loaded CZ catalysts, a single main hydrogen consumption is consistently observed and attributed to the reduction of palladium oxide (PdO_x) and/or platinum oxide (PtO_x) species interacting with the CZ support.^24,32

Fig. 4 H₂-TPR profiles of CZ, xPd–(1 − x)Pt/CZ catalysts (where x = 0–1) and the physical mixture of Pd/CZ and Pt/CZ.

The presence of a single hydrogen consumption contribution in all bimetallic catalysts suggests an interaction between Pd and Pt species. This interaction likely facilitates the H₂ spillover phenomenon, where Pd acts as the primary site for H₂ dissociation, and the resulting hydrogen could migrate to adjacent Pt sites, enabling the reduction of the catalysts' metals at lower temperature. In other words, this cooperative effect results in a unified reduction event indicating an interaction between Pd and Pt metals.³³

The H₂-TPR reduction temperature increases with a higher Pt/Pd ratio, ranging from 115 °C for Pd/CZ to 240 °C for the Pt-only/CZ catalyst. This trend is attributed to the stronger Pt–O–Ce/Zr bonds, which stabilize oxidized Pt species and make them more resistant to reduction. Table 2 summarizes the H₂ uptake for each catalyst, showing a general increase in H₂ consumption with increasing Pt content. The Pt/CZ sample exhibits the highest H₂ uptake, indicating a greater oxidation state of the Pt species. A broad contribution appears at around 545 °C, corresponding to the surface reduction of Ce⁴⁺ to Ce³⁺ in the CeZrO_x support. For the Pd- and Pt-loaded CZ catalysts, this reduction peak shifts to around 500 °C, indicating that the presence of noble metals facilitates the surface reduction of the ceria–zirconia support.^34,35

Table 2 H₂-uptake using H₂-TPR, and metal dispersion and particle size by CO-chemisorption over xPd–(1 − x)Pt/CZ catalysts, where x = 0–1

Catalyst	H2-TPR		CO-chemisorption
	Temperature at maximum (°C)	Exp. H2 quantity (mmol g^−1)	Metal dispersion (%)	Particle size (nm)
Pd/CZ	115	0.16	67	1.6
0.7Pd–0.3Pt/CZ	98	0.15	—	—
0.5Pd–0.5Pt/CZ	128	0.18	—	—
0.3Pd–0.7Pt/CZ	151	0.19	—	—
Pt/CZ	243	0.71	54	2.0

---

To further investigate the interaction of Pt and Pd in the bimetallic system, H₂-TPR of the physically mixed Pd/CZ and Pt/CZ samples was performed. In contrast to the bimetallic systems, the physical mixture of Pd/CZ and Pt/CZ exhibits a distinctly different H₂-TPR profile. There are two H₂ consumption events at low temperature and another one at higher temperature, corresponding to the reduction of Pd and Pt, respectively. The presence of multiple, separated reduction peaks suggests that Pd and Pt species have a lower interaction compared to the bimetallic samples.

The cumulative amount of CO adsorbed during pulse chemisorption measurements was used to determine the metal dispersion and average particle size of the catalysts. For the monometallic samples, metal : CO stoichiometric ratios (M : CO) of 2 : 1 for Pd/CZ and 1 : 1 for Pt/CZ were assumed, in accordance with literature reports for Pd and Pt surfaces.^36,37 Based on these assumptions, both monometallic catalysts exhibit high metal dispersion, which is consistent with their high catalytic performance in the H₂-SCR reaction. An accurate determination of metal dispersion and average particle size in these bimetallic systems is complicated by the coexistence of two metals with potentially different CO adsorption behaviour and surface compositions.

Fig. 5 shows CO-DRIFTS for Pd/CZ, 0.5Pd–0.5Pt/CZ and Pt/CZ, which was performed to obtain qualitative insight into the surface metal sites and the interaction between Pt and Pd species. The spectra of Pt/CZ show an intense band at 2059 cm⁻¹ with shoulders at 2075 and 2019 cm⁻¹, which can be assigned to linearly adsorbed CO on Pt sites.^38,39 In contrast, Pd/CZ exhibits a weaker linear CO band at 2042 cm⁻¹ together with pronounced lower-frequency bands at 1945, 1861, and 1772 cm⁻¹, characteristic of bridged and multi-coordinated CO adsorbed on Pd ensembles.⁴⁰

Fig. 5 CO-DRIFTS of Pd/CZ, 0.5Pd–0.5Pt/CZ and Pt/CZ.

The presence of these intense low-frequency features indicates that Pd/CZ contains contiguous Pd surface sites accessible for multi-coordinated CO adsorption. The Pd–Pt/CZ catalyst displays a distinct CO adsorption profile compared with both monometallic catalysts. A Pt-like linear CO band is observed at 2059 cm⁻¹, while the Pd-related bridged CO region is substantially modified, showing broader and weaker features around 1882 and 1828 cm⁻¹.⁴⁰ This change suggests that the bimetallic catalyst does not behave as a simple combination of isolated Pd and Pt sites. Instead, the lower intensity and shift of Pd-like bridged CO features are consistent with the modification of the Pd surface environment, possibly through Pd ensemble dilution, electronic interactions with Pt, or formation of Pt–Pd interfacial sites. Thus, these CO-DRIFTS results should be interpreted as evidence of Pt–Pd interactions. Additional bands at 1628, 1587, and 1428 cm⁻¹ are attributed to carbonate/carboxylate-type species associated with CO interactions with the CeZrOx support or metal–support interfacial oxygen species.^41,42

TEM and TEM-EDX analyses were conducted to further examine the metal distribution on the support. Although Pd and Pt signals were clearly detected in the EDX spectra, indicating metal incorporation, no distinct metal nanoparticles were observed in the TEM images. The absence of visible particles is most likely attributed to the low metal loading (0.5 wt%), which reduces the contrast between the metallic phase and the ceria–zirconia support.

Therefore, direct visualization of the metallic phase under standard imaging conditions remains challenging. The corresponding TEM images and EDX spectra are provided in Fig. S2–S11 in the SI.

The surface chemical state of selected synthesized catalysts after calcination was investigated using XPS. Fig. 6a shows the Ce 3d core level XPS spectra for all fresh catalysts. Eight distinct peaks are observed, representing two different spin–orbit components which represent Ce 3d_3/2 and Ce 3d_5/2.^43–45 The peaks at 884.5 and 902.3 eV are assigned to Ce³⁺ species, while the peaks at 882.1, 889.2, 898.2, 900.7, 907.3, and 916.5 eV are attributed to Ce⁴⁺.^43,46

Fig. 6 XPS spectra of (a) Ce 3d, (b) O 1s and (c) Pt 4f over Pd/CZ, 0.5Pd–0.5Pt/CZ and Pt/CZ.

The deconvolution of the XPS spectra peaks confirms the coexistence of both Ce³⁺ and Ce⁴⁺ oxidation states in the CeO₂-based catalysts. The relative Ce³⁺ content was quantified by calculating the area of Ce³⁺ peaks to the total Ce 3d peak area. The Ce³⁺ concentrations were found to be 32% for Pd/CZ, 24% for 0.5Pd–0.5Pt/CZ, and 35% for Pt/CZ catalysts. Notably, the fraction of Ce³⁺ in the Pt/CZ and Pd/CZ samples is relatively higher compared to the 0.5Pd–0.5Pt/CZ sample, which correlates with the enhanced NO conversion.

Fig. 6b displays the O 1s spectra for all fresh catalysts, deconvoluted into two peaks representing lattice oxygen (O_β) at 529.4 eV and surface-adsorbed oxygen (O_α) at 531.0 eV.⁴⁴ The relative proportion of surface oxygen, expressed as O_α/(O_β + O_α), follows the trend Pt/CZ > Pd/CZ > 0.5Pd–0.5Pt/CZ, as summarized in Table 3.

Table 3 The relative surface concentration of Ce³⁺, O_α and Pt²⁺ from XPS results

Catalyst	Ce^3+/(Ce^3+ + Ce^4+) (%)	Oα/(Oα + Oβ) (%)	Pt^2+/(Pt^2+ + Pt^4+) (%)
Pd/CZ	32	42	—
0.5Pd–0.5Pt/CZ	24	36	83
Pt/CZ	35	55	84

---

Oxygen vacancies play a crucial role in de-NOx performance by enhancing oxygen mobility within the catalyst system, which reflects the catalytic performance trend of Pd/CZ, 0.5Pd–0.5Pt/CZ and Pt/CZ.^47,48 The Ce³⁺ content is well-correlated with the presence of surface oxygen, which is favourable for the adsorption and activation of oxygen species.²⁴

The Pt 4f XPS spectra of both 0.5Pd–0.5Pt/CZ and Pt/CZ catalysts show oxidized Pt species, namely Pt²⁺ and Pt⁴⁺, which is consistent with the fact that the samples were not subjected to any pre-reduction treatment prior to analysis. The Pt 4f_7/2 and Pt 4f_5/2 peaks appear at 72.5 eV and 75.9 eV representing Pt²⁺ while the Pt 4f_7/2 and Pt 4f_5/2 peaks appear at 74.5 eV and 77.2 eV representing Pt⁴⁺. These binding energies match reported values for Pt²⁺ and Pt⁴⁺,^49,50 confirming the presence of oxidized platinum on the catalyst surface. The shift in Pt 4f binding energies between Pt/CZ and Pd–Pt/CZ correspond to changes in Pt's electronic structure induced by Pd. This shift in binding energy evidences an electronic interaction between Pd and Pt, leading to redistribution of electron density on Pt.⁵¹

Notably, the Pt²⁺/(Pt²⁺ + Pt⁴⁺) ratio in Fig. 6c is similar for both catalysts, indicating that the overall oxidation state distribution of platinum on the surface remains essentially unchanged in the bimetallic catalyst. Therefore, based strictly on the XPS results, both catalysts present comparable surface populations of Pt²⁺ and Pt⁴⁺ species, with no evidence of metallic Pt⁰ under the analysed conditions.

Unfortunately, XPS analysis of Pd oxidation states via the 3d core-level was not feasible for the Pd/CZ and 0.5Pd–0.5Pt/CZ catalysts due to significant spectral interference from Zr 3p orbitals. Specifically, the Pd 3d_5/2 peak and Pd 3d_3/2 peaks typically located at ∼335.0 eV and ∼340.0 eV, respectively, overlap energetically with the Zr 3p_3/2 (∼330 eV) and Zr 3p_1/2 peak (∼344 eV) signals.⁵² This energetic overlap complicates peak deconvolution, making it challenging to reliably quantify the Pd oxidation states.

Structure–activity correlations

The results show that catalytic performance in the H₂-SCR of NO is controlled by the balance between hydrogen activation and its selective utilization for NO reduction. Under the studied conditions, Pt/CZ promotes strong low-temperature activity but also favours rapid H₂ consumption, which limits N₂ selectivity. Pd/CZ provides a more favourable balance between activity and selectivity, indicating that metal identity plays a central role in controlling the competition between selective NO reduction and non-selective H₂ oxidation.

The bimetallic catalysts display a modified activity–selectivity balance rather than a simple enhancement of NO conversion. In particular, the delayed H₂ consumption and improved N₂ selectivity suggest that incorporating Pd and Pt within the same supported catalyst moderates hydrogen utilization and suppresses, to some extent, pathways leading to non-selective H₂ oxidation. The comparison with the physical mixture indicates that this behaviour cannot be fully reproduced by the simple coexistence of separate Pd- and Pt-containing catalyst particles. Instead, the H₂-TPR results indicate that co-impregnation produces a different reduction environment for the noble-metal species, exhibiting the presence of Pd–Pt interactions.

Further evidence of a modified Pd/Pt surface environment is provided by CO-DRIFTS. Although Pd/CZ exhibits CO adsorption features associated with contiguous Pd ensembles, these features are weakened and altered in the bimetallic catalyst. This indicates that the surface arrangement and/or electronic environment of Pd sites is modified by the presence of Pt. Such changes are consistent with disruption of contiguous Pd ensembles, electronic interactions between Pd and Pt, or the formation of Pt–Pd interfacial sites. Importantly, these observations support the presence of modified Pd/Pt surface environments but do not provide definitive evidence for the formation of a Pt–Pd alloy.

Overall, the catalytic behaviour of the bimetallic catalysts can be attributed to the combined influence of the intrinsic properties of Pd and Pt and the modified surface environment generated by co-impregnation on the CeO₂–ZrO₂ support. The interaction between these factors alters hydrogen activation and utilization, thereby shifting the balance between NO conversion and N₂ selectivity. Determining whether these modified sites correspond to alloyed particles, closely associated segregated domains, or specific Pt–Pd interfacial structures would require additional atomic-scale or operando characterization.

Conclusion

This study establishes the influence of noble-metal identity and Pd–Pt interactions on the H₂-SCR performance of CeO₂–ZrO₂-supported catalysts under lean conditions. Pt/CZ exhibited the highest low-temperature activity, reaching 98% NO conversion at 175 °C; however, its rapid H₂ consumption was accompanied by a pronounced decrease in N₂ selectivity. In contrast, Pd/CZ maintained high NO conversion over a broader low-temperature range while providing higher N₂ selectivity, indicating that Pd achieves a more favourable balance between hydrogen activation and selective NO reduction.

The incorporation of Pd and Pt into bimetallic catalysts modified this activity–selectivity relationship rather than simply increasing the NO conversion. Among the bimetallic formulations, 0.7Pd–0.3Pt/CZ exhibited the highest NO conversion, whereas 0.5Pd–0.5Pt/CZ showed the most favourable selectivity behaviour, maintaining more than 75% N₂ selectivity between 150 and 350 °C. The delayed H₂ consumption observed for the bimetallic catalysts indicates a modification of hydrogen utilization, which appears to limit low-temperature conversion but contributes to suppressing non-selective H₂ oxidation and preserving N₂ selectivity over a wider temperature interval.

The comparison with the physical mixture demonstrated that the behaviour of the co-impregnated bimetallic catalyst cannot be reproduced simply by combining Pd/CZ and Pt/CZ in the reactor bed. H₂-TPR showed that co-impregnation modifies the reduction environment of the noble-metal species, while CO-DRIFTS revealed changes in the CO adsorption behaviour of Pd-containing surface sites in the presence of Pt. In particular, the modification of Pd-associated multi-coordinated CO features is consistent with disruption of contiguous Pd ensembles and/or the formation of modified Pd–Pt interfacial environments.

Author contributions

Baqer Aljaman: conceptualization, methodology, data curation, formal analysis, writing – original draft, and writing – review & editing. Sarah Komaty: methodology and writing – review & editing. Nawaf M. Alghamdi: investigation, project administration, funding acquisition, and writing – review & editing. Christos Kalamaras: investigation, project administration, funding acquisition, and review & editing. S. Mani Sarathy: supervision, project administration, and writing – review & editing. Javier Ruiz-Martinez: supervision, project administration, and writing – review & editing.

Conflicts of interest

There are no conflicts to declare.

Data availability

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary information (SI).

Supplementary information: the SI contains the experimental procedures for TEM and TEM-EDX characterization, as well as the corresponding TEM and TEM-EDX images. See DOI: https://doi.org/10.1039/d6cy00260a.

Acknowledgements

The authors acknowledge the financial support from Saudi Aramco under project #FuelCom4. The authors also thank Eganathan Kaliyamoorthy for assistance with ICP-OES measurements, Mark Meijerink for TEM measurements and Ildar Mukhambetov for technical support.

References

1. H. Ritchie, Sector by sector: where do global greenhouse gas emissions come from?, https://ourworldindata.org/ghg-emissions-by-sector, (accessed 3 November 2024).
2. P. Kamboj, M. Hejazi, Y. Qiu, P. Kyle and G. Iyer, Energy Strategy Rev., 2024, 55, 101537.
3. D. POPP, Euro 7: Deal on new EU rules to reduce road transport emissions, https://www.europarl.europa.eu/news/en/press-room/20231207IPR15740/euro-7-deal-on-new-eu-rules-to-reduce-road-transport-emissions, (accessed 22 March 2025).
4. E. M. Szumska and R. S. Jurecki, Energies, 2021, 14(16) DOI:10.3390/en14164821.
5. M. De Gennaro, E. Paffumi, G. Martini, U. Manfredi, H. Scholz, H. Lacher, H. Kuehnelt and D. Simic, in SAE Technical Papers, SAE International, 2014, vol. 1.
6. Z. Stępień, Energies, 2021, 14, 6504.
7. L. Cox, J. Eddinger, D. Grano, W. Vatavuk and R. Strivastava, Nitrogen Oxides (NOx): Why and How They Are Controlled, U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards, Research Triangle Park, NC, 1999.
8. L. Chen, S. Ren, L. Liu, B. Su, J. Yang, Z. Chen, M. Wang and Q. Liu, J. Environ. Chem. Eng., 2022, 10(2) DOI:10.1016/j.jece.2022.107167.
9. S. Komaty, M. Andijani, N. Wang, J. C. Navarro de Miguel, S. Kumar Veeranmaril, M. N. Hedhili, C. I. Q. Silva, Y. Wang, M. Abou-Daher, Y. Han and J. Ruiz-Martinez, Environ. Sci. Technol., 2024, 58(34) DOI:10.1021/acs.est.4c01585.
10. R. Khairova, S. Komaty, A. Dikhtiarenko, J. L. Cerrillo, S. K. Veeranmaril, S. Telalović, A. A. Tapia, J. L. Hazemann, J. Ruiz-Martinez and J. Gascon, Angew. Chem., Int. Ed., 2023, 62(40) DOI:10.1002/anie.202311048.
11. P. M. More, D. L. Nguyen, P. Granger, C. Dujardin, M. K. Dongare and S. B. Umbarkar, Appl. Catal., B, 2015, 174–175, 145–156.
12. P. S. Kim, M. K. Kim, B. K. Cho, I. S. Nam and S. H. Oh, J. Catal., 2013, 301, 65–76.
13. R. Mrad, A. Aissat, R. Cousin, D. Courcot and S. Siffert, Appl. Catal., A, 2015, 504, 542–548.
14. G. G. Olympiou and A. M. Efstathiou, Chem. Eng. J., 2011, 170, 424–432.
15. M. Leicht, F. J. P. Schott, M. Bruns and S. Kureti, Appl. Catal., B, 2012, 117–118, 275–282.
16. A. Väliheikki, K. C. Petallidou, C. M. Kalamaras, T. Kolli, M. Huuhtanen, T. Maunula, R. L. Keiski and A. M. Efstathiou, Appl. Catal., B, 2014, 156–157, 72–83.
17. C. Xu, W. Sun, L. Cao, T. Li, X. Cai and J. Yang, Chem. Eng. J., 2017, 308, 980–987.
18. L. E. Gevers, L. R. Enakonda, A. Shahid, S. Ould-Chikh, C. I. Q. Silva, P. P. Paalanen, A. Aguilar-Tapia, J. L. Hazemann, M. N. Hedhili, F. Wen and J. Ruiz-Martínez, Nat. Commun., 2022, 13(1) DOI:10.1038/s41467-022-30679-9.
19. A. F. Suarez-Corredor, J. Shao, B. Westerberg, M. U. Bäbler and L. Olsson, Appl. Catal., B, 2026, 395, 126855.
20. Y. Li, Y. Huang, H. Zhao, Y. Fan, J. Li, H. Xu, J. Wang and Y. Chen, J. Colloid Interface Sci., 2025, 691, 137443.
21. J. Shao, P. H. Ho, D. Creaser and L. Olsson, Appl. Catal. O: Open, 2024, 188, 206947.
22. M. Jabłońska and A. Osorio Hernández, ChemCatChem, 2024, 16(22) DOI:10.1002/cctc.202400977.
23. M. Maurer, R. Popescu, H. Störmer, M. Casapu and J.-D. Grunwaldt, Appl. Catal., B, 2026, 386, 126347.
24. V. K. Patel and S. Sharma, Catal. Today, 2021, 375, 591–600.
25. E. Aneggi, M. Boaro, S. Colussi, C. de Leitenburg and A. Trovarelli, in Handbook on the Physics and Chemistry of Rare Earths, Elsevier B.V., 2016, vol. 50, pp. 209–242.
26. C. M. Kalamaras, G. G. Olympiou, V. I. Pârvulescu, B. Cojocaru and A. M. Efstathiou, Appl. Catal., B, 2017, 206, 308–318.
27. S. Xu, Z. Liu, C. Peng and Z. Liu, ACS Appl. Nano Mater., 2025, 8(9) DOI:10.1021/acsanm.4c07324.
28. Y. Zhang, S. Xu, J. Li, E. He and Z. Liu, J. Phys. Chem. C, 2023, 127, 7248–7256.
29. K. S. W. Sing, D. H. Everett, R. A. W. Haul, L. Moscou, R. A. Pierotti, J. Rouquerol and T. Siemieniewska, Pure Appl. Chem., 1985, 57, 603–619.
30. F. Jonas, B. Lebeau, S. Siffert, L. Michelin, C. Poupin, R. Cousin, L. Josien, L. Vidal, M. Mallet, P. Gaudin and J. L. Blin, ACS Appl. Nano Mater., 2021, 4, 1786–1797.
31. S. N. Kabekkodu, A. Dosen and T. N. Blanton, Powder Diffr., 2024, 39, 47–59.
32. K. Chen, J. Wan, T. Wang, Q. Sun and R. Zhou, J. Rare Earths, 2023, 41, 896–904.
33. M. Luo, F. Wang, Q. Liu, W. Li, C. Shao, X. Liu and B. Ai, React. Kinet., Mech. Catal., 2023, 136, 2039–2051.
34. H. Wang, Y. Chen, Q. Zhang, Q. Zhu, M. Gong and M. Zhao, J. Nat. Gas Chem., 2009, 18, 211–216.
35. H. Zhang, J. Wang, Y. Zhang, Y. Jiao, C. Ren, M. Gong and Y. Chen, Appl. Surf. Sci., 2016, 377, 48–55.
36. C. Karakaya and O. Deutschmann, Appl. Catal., A, 2012, 445–446, 221–230.
37. P. Canton, G. Fagherazzi, M. Battagliarin, F. Menegazzo, F. Pinna and N. Pernicone, Langmuir, 2002, 18, 6530–6535.
38. H. Unterhalt, G. Rupprechter and H.-J. Freund, J. Phys. Chem. B, 2002, 106, 356–367.
39. M. J. Hossain, M. M. Rahman and Md. Jafar Sharif, Nanoscale Adv., 2020, 2, 1245–1252.
40. D. S. Dolling, J. Chen, J.-C. Schober, M. Creutzburg, A. Jeromin, V. Vonk, D. I. Sharapa, T. F. Keller, P. N. Plessow, H. Noei and A. Stierle, ACS Nano, 2024, 18, 31098–31108.
41. C. Li, Y. Sakata, T. Arai, K. Domen, K. Maruya and T. Onishi, J. Chem. Soc., Faraday Trans. 1, 1989, 85, 929–943.
42. K. Föttinger, W. Emhofer, D. Lennon and G. Rupprechter, Top. Catal., 2017, 60, 1722–1734.
43. X. Wu, L. Xu and D. Weng, Appl. Surf. Sci., 2004, 221, 375–383.
44. A. E. Nelson and K. H. Schulz, Appl. Surf. Sci., 2003, 210, 206–221.
45. F. Zhang, P. Wang, J. Koberstein, S. Khalid and S. W. Chan, Surf. Sci., 2004, 563, 74–82.
46. S. Ricote, G. Jacobs, M. Milling, Y. Ji, P. M. Patterson and B. H. Davis, Appl. Catal., A, 2006, 303, 35–47.
47. C. N. Costa, A. M. Efstathiou, V. N. Stathopoulos and V. C. Belessi, J. Catal., 2001, 197, 350–364.
48. T. Nagasawa, A. Kobayashi, S. Sato, H. Kosaka, K. Kim, H. Min You, K. Hanamura, A. Terada and T. Mishima, Chem. Eng. J., 2023, 453(2) DOI:10.1016/j.cej.2022.139937.
49. L. S. Kibis, D. A. Svintsitskiy, A. I. Stadnichenko, E. M. Slavinskaya, A. V. Romanenko, E. A. Fedorova, O. A. Stonkus, V. A. Svetlichnyi, E. D. Fakhrutdinova, M. Vorokhta, B. Šmíd, D. E. Doronkin, V. Marchuk, J. D. Grunwaldt and A. I. Boronin, Catal. Sci. Technol., 2021, 11, 250–263.
50. J. Lu, Y. Li, S. Li and S. P. Jiang, Sci. Rep., 2016, 6 DOI:10.1038/srep21530.
51. Q. Zhang, T. Chen, R. Jiang and F. Jiang, RSC Adv., 2020, 10, 10134–10143.
52. M. Zhao, H. Zhang, X. Li and Y. Chen, J. Energy Chem., 2014, 23, 755–760.

---