Enhanced CO oxidation via synergistic interactions of Pd and phosphotungstic acid confined within MIL-101(Cr) Authors: Fateme Abbasi, Cyrus Ghotbi Published: 2025 ABSTRACT AbstractMetal-organic frameworks (MOFs), particularly MIL-101(Cr), are promising catalytic supports due to their exceptional surface area and porosity. While noble metals like palladium (Pd) and polyoxometalates, such as phosphotungstic acid (PA), exhibit intrinsic catalytic activity, their application is limited by Pd agglomeration and PA’s low surface area and recovery issues. This study addresses these limitations by using MIL-101(Cr) to simultaneously disperse and stabilize Pd and PA, leveraging the MOF’s mesoporous architecture to confine these components, inhibiting agglomeration and leaching, and maximizing accessibility and catalytic efficiency. This integrated catalytic system, combining the redox activity of Pd and the Brønsted acidity of PA within the robust MOF framework, is designed to address the limitations of individual components and expand the scope of MOF-based catalysis, particularly for gas-phase reactions. This research explores the synthesis and structural characteristics of this novel Pd-PA/MIL-101(Cr) composite catalyst, with a specific focus on its potential for gas-phase carbon monoxide (CO) oxidation, an area where MOF-based catalysts remain comparatively underexplored. The synergistic effects arising from the integration of Pd and PA within the MIL-101(Cr) framework are anticipated to yield enhanced catalytic performance, paving the way for the development of efficient and stable catalysts for environmental remediation and gas-phase industrial processes.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-025-28202-3.Keywords: MIL-101(Cr), CO oxidation, MOF, Pd, Pt, PASubject terms: Environmental sciences, Chemistry Introduction Carbon monoxide (CO) oxidation is a critical reaction in environmental catalysis due to the need for effective removal of CO, a toxic, odorless, and colorless gas. CO is a byproduct of many incomplete combustion processes, and its accumulation poses significant health risks. To mitigate this, the catalytic conversion of CO into CO₂ under mild conditions has been widely investigated. An effective CO oxidation catalyst must offer not only high activity at low temperatures but also strong thermal stability and resistance to deactivation. In recent years, metal–organic frameworks (MOFs) have garnered attention as promising catalyst supports due to their tunable porosity1, high surface area2, and capability to host a variety of active species3,4. Among the MOFs, MIL-101(Cr) stands out due to its large pore volume, mesoporous structure, exceptional chemical and thermal stability, and the ability to stabilize well-dispersed active sites. Its well-defined cavities can effectively confine and prevent the aggregation of catalytic species, making it a suitable platform for liquid-phase reactions1,5,6 and gas-phase oxidation, including CO oxidation7,8. Noble metal nanoparticles, especially palladium (Pd), are widely used as active sites in oxidation reactions owing to their high catalytic activity. However, the performance of Pd-based catalysts strongly depends on the particle size and electronic environment of the Pd species. Polyoxometalates (POMs)—particularly Keggin-type heteropolyacids such as phosphotungstic acid (PA, H₃PW₁₂O₄₀)—have emerged as valuable additives due to their strong Brønsted acidity, redox properties, and ability to modulate the electronic properties of metal nanoparticles. Despite these advantages, POMs suffer from limited surface area, poor structural stability when unsupported, and potential leaching in polar solvents. Supporting POMs within porous frameworks such as MOFs can mitigate these drawbacks by enhancing dispersion and stability. Researches indicate that significant quantities of PA can be accommodated in the MIL cavities. For instance, PA/MIL has been utilized in various reactions such as epoxidation9, oxidative desulfurization10, oxidation reactions11, esterification, carbohydrates dehydration12, aldehyde-alcohol reactions13, etc. In certain reactions, the presence of noble metals is crucial; without them, supported POM catalysts either remain inactive or exhibit low activity14. Most studies have focused on noble metal catalysts supported on MOFs, POMs on MOFs, and M-POM/MOFs15 for liquid phase reactions9–13, with limited research on MOF-based catalysts for gas phase reactions16–20. CO oxidation reaction has been studied on a few MOF-based catalysts such as Cu/MIL-53(Al)16, PdO2/amorphized CuBTC17,18, Au/ZIF-8 19, Pd/10-mixMOF20, Pd/CuBTC, Pd/MNH2, and Pd/MIL7. However, there is significant interest in developing catalysts that exhibit higher activity, lower Tx=100 1, and enhanced stability, which can be achieved by improving the electronic structure of small Pd NPs. Incorporating both Pd nanoparticles and PA into the MIL-101(Cr) framework has the potential to yield a synergistic catalytic system, combining the high activity of Pd with the redox and acidic properties of PA, while benefiting from the structural advantages of the MOF. In this study, MIL-101(Cr) was selected as a robust support to effectively disperse small Pd nanoparticles and suppress their growth and agglomeration. To investigate the influence of noble metal type on catalytic performance, Pt/MIL was also synthesized for comparison with Pd/MIL. Furthermore, phosphotungstic acid (PA) was introduced as a functional additive aimed at modulating the electronic structure of Pd. A series of Pd–PA/MIL catalysts with varying PA loadings was systematically prepared. To elucidate the individual and combined contributions of Pd, PA, and MIL to the overall catalytic behavior, the performance of these catalysts was benchmarked against PA, PA/MIL, Pd/PA, and Pd/MIL. This approach enables a comprehensive understanding of the structure–function relationship and the potential synergistic interactions within the Pd–PA/MIL system. Experimental section Methods MIL-101 (Cr) was synthesized utilizing a hydrothermal technique based on the procedure outlined by Ferey et al. 21 Briefly, a mixture of terephthalic acid (4.17 g, 25 mmol), chromium (III) nitrate nonahydrate (10 g, 25 mmol), de-ionized water (125 ml), and hydrofluoric acid (5 ml of 5 M) was heated with a ramp of 5 °C/min up to 220 °C and kept for eight hours. The resulting mixture was permitted to gradually cool to room temperature. The resulting solid was then treated to remove unreacted linkers and further purified by solvothermal treatments in N, N-dimethylformamide, ammonium fluoride solution (1 M), and ethanol. Finally, it was dried at 70 °C and then at 150 °C under vacuum conditions. 100 mg of MIL was dispersed in deionized water and stirred for 3 h, after which the required amounts of aqueous palladium (II) nitrate and PA solutions were introduced. The mixture was stirred for an additional 16 h. The as-synthesized suspension was then separated by centrifugation and washed twice with water and methanol. Subsequently, the impregnated samples were dried at 80 °C overnight and packed into a quartz tube with an inner diameter of 6 mm before being further dried at 150 °C for 12 h. Finally, the samples were reduced under a flow of reduction gas containing 10% H2 and 90% He at a rate of 100 sccm to obtain 1Pd-xPA/MIL catalysts, where 1 and x are the nominal contents of Pd and PA in mg, respectively, per 100 mg of MIL-101. x = 0, 0.1, 0.56, 2, 10, 15, 27, and 111 are investigated in this work. 1Pt/MIL and 10PA/MIL were synthesized in a procedure similar to that for 1Pd/MIL. 1 Pd/PA was synthesized similarly to the procedure reported for 1% Pd/MIL, except that the impregnated sample was dried without centrifugation and washing steps because the PA is soluble in water. The CO oxidation reaction was conducted in a plug flow reactor at ambient pressure. A feed gas comprising 1% carbon monoxide, 20% oxygen, and 79% helium was introduced at a flow rate of 100 sccm at a temperature of 75 °C, which was gradually increased at a rate of 4 °C/min. The compositions of reaction products were measured by an online Servomex gas analyzer equipped with IR detection for CO2 measurement. The catalytic stability of the samples was evaluated under cyclic reaction conditions. In each cycle, the catalyst was exposed to the CO oxidation feed gas (1% CO, 20% O₂, 79% He; total flow = 100 sccm) at the Tx = 100 of each catalyst for 4 h. After each run, the system was purged with pure He and cooled to room temperature before starting the next cycle. This heating–cooling procedure was repeated for 100 consecutive cycles, corresponding to a total reaction time of 400 h. Stability tests lasting 100–300 h are generally sufficient to assess structural robustness and deactivation behavior22–24. This cyclic approach was intentionally selected to impose more rigorous and realistic conditions on the catalyst compared to a single continuous test. Repeated heating and cooling between cycles can induce mechanical and structural stresses, as well as redox and thermal fluctuations, that often accelerate catalyst deactivation. Therefore, maintaining stable activity under such cyclic conditions provides a stronger indication of the intrinsic durability and structural robustness of the catalyst than a conventional continuous long-term test. Characterization methods The samples were evacuated at 155 °C for 12 h and characterized by a Micromeritics instrument. N2 adsorption/desorption isotherms measured at 77 K were utilized to determine the BET surface area, pore volume, and pore size. The samples’ structural characteristics were analyzed using a PANalytical XPERT-PRO X-ray diffraction. The angular step size and 2θ range were 0.026 ° and 2–80 °, respectively. Inductively coupled plasma optical emission spectrometry (ICP-OES) was performed to determine the exact loading of Pd and M metals. X-ray photoelectron spectroscopy (XPS) was carried out to reveal the electronic state of the metals. The positions of the XPS peaks were corrected by the contaminated carbon peak at 284.6 eV as a reference. TEM was performed on FEI Tecnai G2 F20 Super-Twin transmission. Results and discussions Table 1 presents the final loading values of Pd and PA on MIL. The final Pd content in the 1Pd/MIL sample closely aligns with the nominal value, signifying a strong interaction between the active sites of the MIL support and Pd NPs. In the case of the 1Pd-111PA sample, the final loadings of Pd and PA are approximately 0.007. Maximum loading of 45.5 PA on MIL indicates that MIL sites can adsorb or interact with up to 45.5 PA, equivalent to 0.158 mmolPA/gMIL. The moles of PA used in the synthesis were nine times that of Pd, which decreases the possibility of interaction between Pd NPs and MIL sites, resulting in lower Pd loading. While a high PA loading may be advantageous for certain reactions, it proves detrimental for CO oxidation due to its inhibitory effect on Pd loading, as will be demonstrated subsequently. In the 1Pd-27PA/MIL sample, the final loadings of Pd and PA are 0.81 m and 24.94, respectively. Therefore, the ratio of total moles of Pd and PA species to grams of MIL is 0.137 mmol/gMIL, which is roughly comparable to the 0.158 mmol/gMIL in the 1Pd-111PA/MIL sample. Although the nominal amount of PA in the 1Pd-27PA/MIL sample (0.094 mmolPA/gMIL) is lower than 0.158 mmol/gMIL and there are sufficient MIL sites to interact with PA, the final PA value is less than its nominal value. This could be attributed to the lower PA/Pd ratio (versus moles) in the 1Pd-27PA/MIL sample than 1Pd-111PA/MIL, which increases the probability of Pd being exposed to MIL sites, thereby enhancing Pd loading. However, comparing the final and nominal loadings of Pd and PA reveals that Pd and PA have been loaded at 81% and 92.4% of their nominal values, respectively, indicating the somewhat stronger interaction between PA and MIL. The final loadings of Pd and PA in the 1Pd-15PA/MIL, 1Pd-10PA/MIL, 1Pd-2PA/MIL, 1Pd-0.56PA/MIL, and 1Pd-0.1PA/MIL samples, as well as the final loading of PA in 10PA/MIL, are very close to their nominal values. This is because, in these samples, the total nominal contents of Pd and PA are below 0.158 mmol/gMIL, providing ample MIL sites for interaction with both Pd and PA species. Additionally, since both Pd and PA interact effectively with MIL sites, only minimal amounts are lost during the washing process. Table 1ICP-OES results of different samples. The nominal Pd value in the samples is 0.0434 mmolPd/gMIL.SamplesNominal PA (mmolPA/gMIL)Nominal PA/Pd ratio (mmolPA/mmolPd)Final Pd loading (mg/100 mg MIL)Final PA loading (mg/100mgMIL)1Pd-111PA/MIL0.3858.880.00745.5321Pd-27PA/MIL0.0942.160.80824.9371Pd-15PA/MIL0.0521.200.84614.8091Pd-10PA/MIL0.03470.80.9889.9701Pd-2PA/MIL0.00690.160.9871.9911Pd-0.56PA/MIL0.00190.0450.9880.5991Pd-0.1PA/MIL0.000350.0080.9900.9991Pd/MIL-00.991-10 PA/MIL0.0347--9.979 ICP-OES results of different samples. The nominal Pd value in the samples is 0.0434 mmolPd/gMIL. Figure 1 illustrates X-ray diffraction patterns for MIL, 1Pd/MIL, and 1Pd-10PA/MIL samples. The X-ray diffraction pattern of MIL closely resembles that reported by Zhu et al. 25, indicating its high crystallinity. The 1Pd/MIL and 1Pd-10PA/MIL samples show no significant loss of crystallinity, suggesting that the host matrix maintains its crystalline structure during the reduction step and that the Cr3+ cations within MIL were not reduced to metallic Cr. Although all diffraction peaks of MIL appear in the X-ray diffraction patterns of both 1Pd/MIL and 1Pd-10PA/MIL, some peaks exhibit reduced intensity in the range of 2θ = 2 ° to 10 °. This can be attributed to the incorporation of Pd and PA within the MIL structure, either within the pores or on the external surface, as the interaction between the electrophilic surfaces of Pd and PA with the atoms of the MIL framework alters charge distribution and electrostatic fields. Moreover, any distinct peaks for Pd and PA are not detected, suggesting that their sizes are sufficiently small to be confined within the MIL cavities. XRD patterns of MIL, 1Pd/MIL, and 1Pd-10PA/MIL. Figure 2 shows N2 adsorption-desorption isotherms (a) and pore size distributions (b) of MIL, 1Pd/MIL, and 1Pd-10PA/MIL. MIL demonstrated a BET pore volume of 1.139 cm3/g and a BET surface area of 2071.8 m2/g, which are consistent with those reported by Kempe et al. 26. The isotherms exhibit two distinct uptake steps around P/P₀ = 0.1 and 0.2, which indicate the presence of two types of pores in the structure. This is further supported by the pore size distribution curve (Fig. 2b), where MIL exhibited two pore sizes of 1.76 nm and 2.45 nm with a mean diameter of 2.35 nm, placing it within the mesoporous range. The 1Pd/MIL showed a BET surface area of 1951 m2/g and a pore volume of 0.97 cm3/g. The lower surface area and pore volume in 1Pd/MIL compared to MIL suggest that Pd NPs occupy the MIL cavities. The pore dimensions in 1Pd/MIL are below 1.7 nm and 2.17 nm, which are smaller than those in MIL, further supporting the incorporation of Pd NPs into the MIL pores. Following the incorporation of PA and Pd NPs, there was a significant reduction in the surface area of host MIL, with a BET surface area of 1860 m2/g obtained for 1Pd-10PA/MIL, revealing that most of the PA and Pd have been effectively confined within the MIL cavities. N2 adsorption-desorption isotherms (a) and pore size distributions (b) of MIL, 1Pd/MIL, and 1Pd-10PA/MIL. Figure 3 shows the XPS spectra for the samples of 1Pd/MIL, 10PA/MIL, and 1Pd-10PA/MIL. These spectra not only reveal the elements of the MIL framework (Cr, C, O) but also confirm the presence of Pd and W. Two dominant peaks located at 579.1 and 588.2 eV are attributed to Cr (III) 2p3/2 and Cr(III) 2p1/2, respectively27,28. Any peaks indicative of Cr (II) or Cr (0) were not detected, suggesting that the Cr3 + ions of the host framework remained unreduced throughout the reduction step, thereby preserving the structure of the MIL, which aligns with the XRD findings. The C 1s peaks in the samples appear at 284.85 eV (Fig. 3), which aligns with sp2 carbons. In the XPS analysis of 1Pd/MIL, peaks at 335.7 eV and 340.9 eV are assigned to Pd (0) 3d5/2 and 3d3/2, respectively, while those at 337.3 eV and 342.4 eV are attributed to Pd (II) 3d5/2 and 3d3/2. Concerning the standard reduction potential of Pd/Pd (II) (0.951 V), Pd (II) ions can readily be reduced to Pd (0). Thus, the Pd (II) species observed in the XPS spectrum likely formed during catalyst preparation, resulting from strong interactions between Pd precursors and the oxygen atoms within MIL, or from the oxidation of Pd NPs in ambient air due to their high reactivity. XPS spectra of 1Pd/MIL, 10PA/MIL, and 1Pd-10PA/MIL: (a) total, (b) Pd 3d, and (c) w 4 f. The XPS spectrum for the 10PA/MIL catalyst shows W 4f binding energies at 35.3 eV (4f7/2) and 37.4 eV (4f5/2), which are indicative of W6+ ions within the PA structure. This suggests that the tungsten’s oxidation state remains unchanged, and PA has not undergone reduction during the reduction process. In the high-resolution spectra for Pd 3d and W 4f, the Pd(0) 3d peaks in the 1Pd-10PA/MIL sample shift to higher binding energies by approximately 0.3 eV compared to the 1Pd/MIL, while the W 4f peaks in the 1Pd-10PA/MIL sample shift to lower binding energies by about 0.3 eV relative to the 10 PA/MIL. These considerable shifts demonstrate a notable change in the electronic environment of both W and Pd. The increase in binding energy for Pd and the decrease for W suggest an electronic transfer from Pd to PA, highlighting a strong interaction between PA and Pd. It should be noted that in the 1Pd-10PA/MIL sample, the nominal molar amounts of Pd and PA are 0.0434 and 0.035 mmol/gMIL, respectively, totaling 0.078 mmol/gMIL—significantly lower than the estimated site capacity of 0.158 mmol/gMIL. This indicates that sufficient MIL sites are available to accommodate both species without substantial competition. As a result, both Pd and PA were effectively loaded, with minimal loss during the washing process. Although PA may exhibit a slightly stronger interaction with MIL compared to Pd, this does not play a critical role in this case due to the excess availability of MIL sites and the moderate molar ratios. These favorable interactions with the support contribute to the final loadings being close to their nominal values. As mentioned above, the final PA loading in both 10PA/MIL and 1Pd-10PA/MIL samples is very close to their respective nominal values, indicating that PA was fully loaded in both cases. Consequently, there seems to be no notable difference in the PA–MIL interaction between these two samples. However, the significant difference observed in the binding energy of tungsten between 10PA/MIL and 1Pd-10PA/MIL further supports the presence of an interaction between Pd and PA, which alters the electron density of each component. TEM analysis was performed to investigate the microscopic structure of the 1Pd/MIL and 1Pd-10PA/MIL samples. As shown in Fig. 4a, b, no significant aggregation of Pd or PA species was observed, indicating that both Pd nanoparticles and PA clusters are well confined within the MIL cavities. For the Pd/MIL sample, the TEM images reveal that Pd nanoparticles exhibit an average diameter of 2.32 ± 0.83 nm, suggesting that most particles are sufficiently small to be accommodated within the pores of the MIL support. In the case of the 1Pd-10PA/MIL sample, the average nanoparticle diameter is measured to be 2.69 ± 1.09 nm, a value slightly larger but comparable to that of Pd/MIL. Given that both Pd and PA species are simultaneously present and uniformly distributed on the support, as confirmed by elemental mapping (STEM-EDX), and considering the similar electron contrast of these species in TEM imaging, this measured size distribution is attributed to the combined population of Pd and PA nanoparticles. For further structural insight, the HRTEM image of the 1Pd-10PA/MIL sample (Fig. 4c) displays interplanar spacings of 0.201 nm and 0.232 nm, corresponding to the [100] and [111] planes of face-centered cubic (fcc) Pd, respectively. TEM images of 1Pd-10PA/MIL (a), 1Pd/MIL (b), and HR-TEM images of Pd-10PA/MIL (c). Figure 5 shows a STEM image along with elemental mapping of the 1Pd-10PA/MIL sample, demonstrating a uniform distribution of elements including C, O, Cr, P, W, and Pd. The EDX spectra in Fig. 6 confirm the presence of Pd and PA. The region marked as point 1 demonstrates the presence of W and Pd (Fig. 6b), suggesting that Pd and PA are spatially close. This proximity implies a potential interaction between Pd and PA, corroborating the XPS findings. The EDX analysis reveals a W/P ratio of approximately 12.01, indicating that PA reduction did not occur during the synthesis and reduction processes. STEM image and elemental mapping of 1Pd-10PA/MIL. STEM image and EDX spectra of 1Pd-10PA/MIL corresponding to surface 1 (a) and point 1 (b). Figure 7 shows the catalytic activities of MIL, PA, 10PA/MIL, 1Pd/PA, 1Pd/MIL, and 1Pt/MIL. The MIL exhibits no catalytic activity in the whole studied temperatures. Although the Lewis acid sites of MIL readily bind CO, activating O2 necessitates higher temperatures, meaning that higher temperatures are required for the dissociative adsorption of oxygen. 1Pd/MIL starts to show activity at 140 °C, with CO conversion significantly increasing at 180 °C and achieving complete CO conversion at 185 °C. This highlights the crucial role of Pd sites in the activation of oxygen molecules. To further investigate how the characteristics of noble metal sites influence catalytic activity, a 1 Pt/MIL catalyst was also synthesized. The 1Pt/MIL exhibited only an 11% conversion of carbon monoxide (CO) at 225 °C. The significantly lower activity of the 1Pt/MIL compared to the 1Pd/MIL can be attributed to the reduced final concentration of Pt in the 1Pt/MIL sample. Inductively Coupled Plasma (ICP) analysis revealed that the final concentrations were 0.005 for Pt and 0.991 for Pd. This indicates that a significant quantity of Pt was lost during the washing step, suggesting that the interaction between Pd2+ and MIL is stronger than that between Pt2+ and MIL. To prevent the loss of Pt, a 1Pt/MIL was created without a washing step after the impregnation of Pt2+, and its performance is illustrated in Fig. 7. This catalyst shows better activity than the Pt/MIL synthesized with a washing step due to its higher Pt concentration. However, it underperforms compared to 1Pd/MIL, explaining that the interaction of NPs with MIL has a significant effect on the catalytic activity, as it influences the size and distribution of the NPs. CO conversion over MIL, PA, PA/MIL, Pt/MIL, Pd/PA, and Pd/MIL. As shown in Fig. 7, both PA and PA/MIL were inactive in the CO oxidation reaction, indicating that the presence of Pd NPs is necessary for O2 activation. The 1Pd/PA catalyst achieved a temperature of 191 °C for complete conversion in the first catalytic run (Fig. 7). After cooling the catalyst under He flow to room temperature, it was retested in the CO oxidation reaction, achieving 100% CO conversion at 210 °C These findings suggest that the catalytic performance of Pd/PA is irreversible, whereas the catalytic performance of Pd/MIL remained consistent across over 100 reaction cycles, demonstrating MIL’s significant role in enhancing catalytic stability by confining NPs within its cavities, thereby preventing their growth and agglomeration. Figure 8 illustrates the catalytic behavior of 1Pd–xPA/MIL catalysts, where x = 0.1, 0.56, 2, 10, 15, 27, and 111. The catalytic performances of 1Pd-0.56PA/MIL and 1Pd-2PA/MIL are very close to each other, with only minor differences across the studied temperature range. To provide a clearer comparison, the exact CO conversion values of both samples at all tested temperatures are reported in Supplementary Figure S1, confirming their nearly identical behavior. The similarity of these two catalysts is attributed to their low PA/Pd molar ratios (0.045 and 0.16, respectively), where the amount of PA is insufficient to induce a significant electronic modification of Pd. For the same reason, the catalyst with an even lower PA/Pd ratio (1Pd-0.1PA/MIL, PA/Pd = 0.008) exhibits weaker activity than both 1Pd-0.56PA/MIL and 1Pd-2PA/MIL, with Tx=100 values of 175 °C versus 170 °C for the latter two. At higher PA/Pd ratios (> 2), the catalytic activity decreases again compared to 1Pd-10PA/MIL. For example, in 1Pd-27PA/MIL (PA/Pd = 2.16), the performance is lower than that of 1Pd-10PA/MIL despite a relatively high PA loading. According to ICP-OES results, the final Pd content in 1Pd-27PA/MIL (0.808 mg/100 mg MIL) is reduced compared to the nearly complete Pd loading in 1Pd-10PA/MIL (0.988 mg/100 mg MIL). This lower Pd content, together with partial occupation of MIL pores by excess PA, decreases the accessibility of Pd active sites and weakens the beneficial Pd–PA interaction. At even higher PA content (1Pd-111PA/MIL, PA/Pd = 8.88), the activity further decreases, with Tx=100 observed at 210 °C. This reduced performance is consistent with ICP-OES results showing a dramatic reduction in final Pd loading (0.007 mg/100 mg MIL) and BET analysis indicating decreased surface area and pore volume due to severe pore blocking. CO conversion over Pd/MIL and 1Pd-xPA/MIL. A systematic trend emerges when considering the entire 1Pd–xPA/MIL series. At intermediate PA/Pd ratios (0.8–1.2, corresponding to 1Pd-10PA/MIL and 1Pd-15PA/MIL), the catalysts exhibit the best performance with Tx=100 of 160 °C. In this range, Pd and PA are effectively accommodated within the MIL framework, enabling strong electronic interaction. For the 1Pd-10PA/MIL catalyst, which exhibited the best catalytic performance, the enhanced activity can be attributed to the efficient incorporation of both Pd and PA into the MIL framework’s cavities and the uniform dispersion of Pd nanoparticles. Structural characterization (Fig. 4) confirmed the homogeneous distribution of Pd species, while XPS analysis (Fig. 3) revealed distinct ~ 0.3 eV shifts in the Pd 3d and W 4f binding energies, evidencing electron transfer from Pd to PA. This electronic interaction modifies the electronic structure of Pd, thereby facilitating O₂ activation and accelerating CO oxidation. TEM/EDX mapping further demonstrated the close spatial proximity of Pd and PA species, which promotes direct electronic and structural interactions essential for cooperative catalysis. Taken together, these features indicate that at the optimal Pd-to-PA weight ratio of 0.1, a favorable electronic environment is established, leading to superior catalytic activity compared to other catalysts. Overall, these results demonstrate that the catalytic activity of Pd–PA/MIL depends non-monotonically on the PA/Pd molar ratio. Both insufficient PA (inadequate electronic modification, as in 1Pd-0.1PA/MIL) and excessive PA (loss of Pd loading and pore blocking, as in 1Pd-27PA/MIL and 1Pd-111PA/MIL) are detrimental. In contrast, an intermediate PA/Pd ratio (0.8–1.2) provides the most favorable balance, yielding a pronounced synergistic effect and the highest catalytic activity. The significantly enhanced catalytic activity of 1Pd-10PA/MIL, compared to the individual Pd/MIL and PA/MIL systems, strongly suggests that a synergistic interaction occurs between Pd and PA species within the MIL framework. This synergistic behavior can be primarily attributed to electronic interactions and enhanced adsorption/desorption properties. Supporting this, X-ray photoelectron spectroscopy (XPS) analysis revealed distinct shifts in the binding energies of Pd and W in the Pd-PA/MIL sample relative to those in the Pd/MIL and PA/MIL systems. These shifts indicate a redistribution of electron density caused by the interaction between Pd and PA, which likely alters the activation of CO and O₂ molecules and thereby facilitates the catalytic process. Moreover, this conclusion is reinforced by TEM-EDS mapping, which confirmed the uniform dispersion of Pd and PA species throughout the MIL support and demonstrated their close spatial proximity. Such proximity promotes direct electronic or structural interactions between the two components—an essential condition for cooperative catalytic activity. Furthermore, the co-impregnation synthesis method used in this study enabled the simultaneous incorporation of Pd and PA into the MIL-101(Cr) structure. This method promotes intimate contact and integration of the active components, stabilizing the catalytic sites and enhancing their cooperative function beyond the individual capabilities of Pd or PA alone. The thermal stability of the synthesized MIL-101(Cr) and Pd/MIL samples was evaluated by thermogravimetric analysis (TGA) under an air atmosphere at a heating rate of 5 °C/min. As shown in Figure S2, MIL-101(Cr) exhibited a gradual weight loss below 100 °C, which corresponds to the removal of physically adsorbed water and residual solvents. The main decomposition of the framework structure began at approximately 275 °C, indicating the onset of thermal degradation. In the case of Pd/MIL, a similar thermal behavior was observed, with a slightly higher decomposition temperature, suggesting that incorporation of Pd nanoparticles may enhance the thermal stability of the framework to some extent. Notably, during CO oxidation testing, the catalyst samples were exposed to a maximum temperature of 210 °C, which is well below the decomposition point observed in the TGA curves. These results demonstrate that the MIL-based catalysts used in this study maintain their structural and thermal stability during the entire catalytic process. The CO oxidation performance of the catalysts was evaluated as a function of temperature, and the results are presented in Figs. 7 and 8. As the temperature increased, the CO conversion gradually rose and then exhibited a sharp increase as it approached the Tx=100. This behavior is characteristic of MOF-based catalysts and has been widely reported in the literature16,17,18, 19,20,7. The observed sharp rise in conversion near Tx = 100 can be attributed to the temperature-activated dissociative adsorption of O₂, which becomes more efficient above a critical threshold. Combined with the high surface area of MIL-101(Cr) and the presence of well-dispersed Pd and PA active sites, the reaction proceeds rapidly once the minimum activation temperature is reached. The performance of 1Pd-10PA/MIL was further examined at temperatures beyond Tx=100. As shown in Figure S2, once full conversion was achieved at 160 °C, the CO conversion remained constant at X = 100% across the extended temperature range up to 210 °C, confirming the catalyst’s stability and the reliability of the measurement. Additionally, 1Pd-10PA/MIL catalyst demonstrated sustained performance over multiple catalytic cycles (more than 100), where it maintained its activity at 160 °C for 4 h during each cycle (Figure S3). To evaluate the catalyst after the stability test, XRD (Fig. 9) and TEM (Fig. 10) analyses were conducted. The XRD data indicated that the structural integrity of the Pd-10PA/MIL catalyst was preserved, as its post-test structure was comparable to its pre-test state. This observation underscores the structural stability of the MIL framework under testing conditions and suggests that the MIL structure was unaltered after hours of catalytic operation, emphasizing the chemical and thermal stability of this MOF. TEM analysis determined an average particle size of 4.3 nm and indicated a minimal growth in nanoparticle size after the prolonged stability test in comparison to the initial catalyst. This indicates the absence of significant nanoparticle agglomeration and highlights the ability of the MIL framework to confine nanoparticles, preventing their agglomeration and growth. Consequently, the catalyst’s activity was very satisfactorily maintained throughout the stability. XRD patterns of 1Pd-10PA/MIL after stability test compared to before one. TEM image of tested 1Pd-10PA/MIL and its corresponding histogram. Notably, employing the cost-effective PA not only improves catalytic performance but also minimizes the required amount of costly palladium, thereby enhancing the catalyst’s economic viability and fostering its practical applications. In addition, the introduction of PA and Pd into the catalyst structure was readily achieved via an impregnation method, which is highly reproducible; this enhances the industrial applicability of the catalyst. Conclusion In conclusion, this study successfully synthesized and characterized a series of Pd-PA catalysts supported on MIL-101(Cr) using a facile wet impregnation method, achieving tunable Pd and PA loadings as confirmed by ICP-OES. XRD analysis verified the preservation of the MIL-101(Cr) framework structure upon catalyst preparation. Crucially, catalytic evaluation for CO oxidation revealed a pronounced synergistic effect between Pd and PA when supported on MIL-101(Cr). While Pd/MIL demonstrated good activity (Tx=100 = 185 °C), the incorporation of PA significantly enhanced performance, with the 1Pd-10PA/MIL catalyst exhibiting optimal activity, achieving complete CO conversion at a notably lower temperature of 160 °C. Furthermore, MIL-101(Cr) conferred enhanced stability to the Pd nanoparticles, contrasting with the deactivation observed for unsupported Pd/PA. These findings underscore the potential of rationally designed MOF-supported catalysts incorporating synergistic components for efficient and stable CO oxidation, offering a promising avenue for future catalyst development in gas-phase reactions. In addition, a comparison between Pd- and Pt-based MIL catalysts revealed that the type of noble metal significantly influences catalytic performance. The 1Pt/MIL catalyst showed only ~ 11% CO conversion at 225 °C, much lower than that of 1Pd/MIL, highlighting the superior activity of Pd due to its stronger interaction with the MIL framework and better nanoparticle dispersion. These results emphasize that the intrinsic nature of the noble metal and its interaction with the support are key factors governing catalytic efficiency. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Acknowledgements The authors would like to acknowledge the support and resources provided by the lab at Sharif University of Technology, which facilitated this research. Author contributions F. A. synthesized the catalysts, prepared the experimental setup, conducted the reaction tests, and processed the data. Also, she performed physical characterizations (XRD, BET, TEM, TPD, XPS, HRTEM, and SEM) as well as wrote and edited the manuscript. S. G. supervised and directed the research, and both authors collaborated in the analysis of the results and discussions. Funding The author declares that no funding was received for the conduct of this research or the preparation of this article. Data availability The datasets generated or analyzed during the current study are included in this article and are also available from the corresponding author on reasonable request. Declarations Competing interests The authors declare no competing interests. Footnotes The minimum temperature in which the 100% CO conversion is obtained. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement The datasets generated or analyzed during the current study are included in this article and are also available from the corresponding author on reasonable request.