# Fileset

[2024A00282G_MDR - コピー.pdf](https://mdr.nims.go.jp/filesets/6b7f5c69-5501-48d4-ac90-4dc7ef5c978e/download)

## Creator

[Esmail Doustkhah](https://orcid.org/0000-0003-1459-1756), [Nao Tsunoji](https://orcid.org/0000-0003-4919-702X), [Shinya Mine](https://orcid.org/0000-0002-8339-0301), [Takashi Toyao](https://orcid.org/0000-0002-6062-5622), [Ken-ichi Shimizu](https://orcid.org/0000-0003-0501-0294), [Tetsuro Morooka](https://orcid.org/0000-0003-3436-7030), [Takuya Masuda](https://orcid.org/0000-0001-7462-2177), [M. Hussein N. Assadi](https://orcid.org/0000-0002-1559-3211), [Yusuke Ide](https://orcid.org/0000-0002-6901-6954)

## Rights

This document is the Accepted Manuscript version of a Published Work that appeared in final form in ACS Applied Materials & Interfaces, copyright © 2024 American Chemical Society after peer review and technical editing by the publisher. To access the final edited and published work see https://doi.org/10.1021/acsami.3c18709[In Copyright](http://rightsstatements.org/vocab/InC/1.0/)

## Other metadata

[Feeble Single-Atom Pd Catalysts for H<sub>2</sub> Production from Formic Acid](https://mdr.nims.go.jp/datasets/ed936f9d-7b56-4afb-bcc7-50ada60a88ff)

## Fulltext

1 Feeble Single Atom Pd Catalyst for H2 Production from Formic Acid Esmail Doustkhah,*,a,b,c Nao Tsunoji,d Shinya Mine,e Takashi Toyao,f Ken-ichi Shimizu,f Tetsuro Morooka,h Takuya Masuda,g,h M. Hussein N. Assadi,i Yusuke Ide*,a,j  a Research Center for Materials Nanoarchitechtonics (MANA), National Institute for Materials Science (NIMS), 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan, Email: IDE.Yusuke@nims.go.jp b Koç University Tüpraş Energy Center (KUTEM), Department of Chemistry, Koç University, Istanbul 34450, Turkey c Chemistry Department, Faculty of Engineering and Natural Sciences, Istinye University, 34396 Sarıyer, Istanbul, Turkey, Esmail: esmail.heragh@istinye.edu.tr d Department of Applied Chemistry, Graduate School of Advanced Science and Engineering, Hiroshima University, 1-4-1 Kagamiyama, Higashi-Hiroshima 739-8527, Japan e National Institute of Advanced Industrial Science and Technology (AIST), Research Institute for Chemical Process Technology, 4-2-1 Nigatake, Miyagino, Sendai 983-8551, Japan f Institute for Catalysis, Hokkaido University; Sapporo 001-0021, Japan g Graduate School of Chemical Sciences and Engineering, Hokkaido University, Sapporo, Hokkaido 060-0810, Japan h Research Center for Energy and Environmental Materials (GREEN), National Institute for Materials Science (NIMS), Tsukuba, Ibaraki 305-0044, Japan i RIKEN Center for Emergent Matter Science (CEMS), 2-1 Hirosawa, Wako, Saitama 351–0198, Japan j Graduate School of Engineering Science, Yokohama National University, 79-5 Tokiwadai, Hodogaya-ku, Yokohama 240-8501, Japan. Keywords: Single atom catalysis (SAC); Pd nanocluster; Silica-supported Pd; Pd DOS; Metal-support interaction; Catalyst reconstruction; Ligand-metal charge transfer; Density functional theory (DFT). 2 Abstract Single-atom catalysts are thought to be the pinnacle of catalysis. However, for many reactions, their suitability is yet to be unequivocally proven. Here, we demonstrate why single Pd atoms (PdSA) are not catalytically ideal for generating H2 from formic acid as a H2 carrier. We loaded PdSA on three silica substrates, mesoporous silicas functionalized with thiol, amine, and dithiocarbamate functional groups. The Pd catalytic activity on amino-functionalized silica (SiO2-NH2/PdSA) was far higher than the thiol-based catalysts (SiO2-S-PdSA and SiO2-NHCS2-PdSA), while the single-atom stability of SiO2-NH2/PdSA against aggregation after the first catalytic cycle was the weakest. In this case, Pd aggregation boosted the reaction yield. Our experiments and calculations demonstrate that PdSA in SiO2-NH2/PdSA loosely binds with amine groups. This leads to a limited charge transfer from Pd to the amine groups and causes high aggregability and catalytic activity. According to the density functional calculations, the loose binding between Pd and N causes most of Pd’s 4d electrons in aminofunctionalized SiO2 to remain close to the Fermi level and labile for catalysis. However, PdSA chemically binds to the thiol group, resulting in strong hybridization between Pd and S, pulling Pd’s 4d states deeper into the conduction band and away from the Fermi level. Consequently, fewer 4d electrons were available for catalysis.  Introduction Pd, known as a powerful noble catalyst, has taken a promising place in various types of catalysis, including oxidation, coupling,1 cycloaddition,2 carbonylation,3 hydrodeoxygenation,4 and especially in hydrogen generation. While many attempts at investigating the catalysis mechanism through Pd species have been established, some controversies on mechanistic elucidations remain to be answered. The debate over the mechanisms governing Pd catalytic activity mainly originates from the complex interdependent parameters in determining the effective yield, i.e., support’s surface chemistry5 and morphology,6 the crystallinity and the particle size of Pd, and the reaction conditions. For instance, some argue that the catalytic activity of Pd depends on the basicity of the functional groups on the support surface,7 while others have reported that the Pd nanoparticles with the ultrasmall size are the best approach that should be sought for obtaining optimal catalysis.8,9 3 In our recent investigation, we demonstrated that the thickness of Pd plate-like nanoparticles on any given support should be investigated to find optimal catalytic performance since the optimal Pd thickness to deliver maximum catalytic activity depends on the type of the support and its surface chemistry.10 However, how a functional group can influence Pd’s electronic structure in conjunction with the support material has rarely been investigated. Additionally, the interdependence of stability and catalytic activity is a further point that requires careful investigation in the case of Pd-supported catalysts possessing a functionalized surface. A vast fraction of investigations so far have also been focused on reactant/Pd interactions, that is,  how potent incoming molecules vs. how weak output molecules are.11–13 However, the interplay between the support and Pd is undeservedly less investigated.14–17 This latter interaction becomes more significant in determining Pd’s electronic structure as the size shrinks as the hybridization with the support strengthens.10,16 Single-atom catalysis (SAC) is a recently developed concept that is being widely investigated for several catalytically active metals on various supports for various reactions.18–22 It has been demonstrated that the support material plays a fundamental role in the catalytic activity of SAC.23,24 Unlike many other metals synthesized as sing-atom catalysts, which can readily be synthesized, stabilized, and reveal a high efficiency,25,26 Pd single-atom (PdSA), requires overly cumbersome synthesis methods and suffers lower stabilization.19 Furthermore, the ligand-metal charge transfer,23,24 and the metal-metal interactions21 are well-established phenomena that must be investigated separately for each species in catalysis, which are also crucial to understanding the catalytic behavior of Pd. As a result, a universal picture regarding the PdSA catalytic activity and mechanism has not been established so far despite a vast range of investigations.19,20,27–29  Herein, we investigate the relationship between the catalytic activity and the stability of PdSA when supported on different organofunctionalized-mesoporous silicas, i.e., -NH2, -SH, and -NHCS2H. PdSA on NH2 functionality in formic acid dehydrogenation indicates far higher activity than that of PdSA on thiol-based functionalities, especially at lower temperatures. Density functional calculations demonstrate that the PdSA binds rather weakly to the -NH2 group, resulting in an in situ aggregation of PdSA and subsequent enhancement in the metallic feature and charge transfer, and eventually, the catalytic activity improvement (Scheme 1). 4  Scheme 1. Schematic view of Pd single atom loaded samples and their catalytic behavior in formic acid dehydrogenation: SiO2-NH2/Pd, SiO2-S-Pd, SiO2-NHCS2-Pd.   Results and discussions We first functionalized mesoporous SiO2, SBA-15, with a pore diameter of 8 nm, with amine (SiO2-NH2) and thiol (SiO2-SH) functional groups to serve as anchoring points for PdSA and PdNP. 29Si magic angle spinning nuclear magnetic resonance spectroscopy (MAS NMR) and 13C cross-polarization MAS NMR (CP MAS NMR) (Figure S1A, B) confirmed the successful functionalization of aminopropyl and mercaptopropyl groups on SiO2 surface via Si-O-Si covalent bands for SiO2-NH2 and SiO2-SH, respectively. The successful functionalization, while retaining the original hexagonally ordered mesoporous structure,30 was confirmed by low-angle X-ray diffraction (XRD) and N2 adsorption/desorption (Figure S1C, D). Measuring the thermogravimetric (TG) and differential thermal analysis (DTA) showed that the ratio of functional groups on support SiO2 are 3.8 and 3.0 mmol per each g of SiO2, for SiO2-NH2 and SiO2-SH, respectively (Figure S2). 5 We then deposited Pd species on the functionalized mesoporous silicas. According to the 29Si MAS NMR, the PdSA-deposited samples (the products are named SiO2-NH2/PdSA and SiO2-S-PdSA) show that the ratios of organofunctionalized silanols (T peak, mainly at −66.5 ppm) to free unfunctionalized silanols (Q3 peak at −100 ppm) hardly changed compared to samples before PdSA loading (Figure S3A). 13C CP MAS NMR reveals the presence of propylene chains, which link the silica surface to the functional groups constantly before and after the deposition of Pd, as shown in Figure S3B (vs Figure S1B). The low angle XRD patterns and N2 adsorption-desorption isotherms analyses (surface areas, pore diameters, and pore volumes) (Figures S1C, D & S4, Table S1) also show how the functionalization and subsequent Pd loading on SiO2 support has affected the SiO2’s porosity and surface area. Accordingly, the surface area, pore volume, and pore size of SiO2-NH2 and SiO2-SH only slightly decreased after Pd deposition. This implies the anchored Pd species are very small in size or amount. According to the elemental analysis, the amounts of loaded Pd in each SiO2-NH2/PdSA and SiO2-S-PdSA are found to be 1.4 and 1.1 wt%, respectively (Table S2). Given the amount of grafted functional groups obtained from TG-DTA curves, the Pd to organic functional unit ratios are identical for both SiO2-NH2/PdSA and SiO2-S-PdSA and are 1.0:23.0. As expected, the transmission electron microscopy (TEM) images of SiO2-NH2/PdSA and SiO2-S-PdSA detected no Pd nanoparticles, confirming that the deposited Pd species were not aggregated as particles that could be observable under even a high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) (Figure 1A–D). Additionally, TEM-based elemental mapping images showed a spatially uniform Pd distribution on the substrate, similar to the nitrogen and sulfur distribution, reaffirming that it was unlikely that the Pd species had aggregated on SiO2-NH2/PdSA and SiO2-S-PdSA (Figures S5 and S6, respectively). X-ray absorption spectroscopy (XAS) was employed to investigate Pd’s coordination environment in SiO2-NH2/PdSA and SiO2-S-PdSA (Figure 1E, F). Figure 1E presents Pd K-edge X-ray absorption near-edge structure (XANES) spectra of SiO2-NH2/PdSA and SiO2-S-PdSA. Accordingly, the absorption edge position in the spectrum for both SiO2-S-PdSA and SiO2-NH2/PdSA sits between Pd foil and PdO (as references), suggesting that Pd species in both samples were to some extent oxidized, though cannot be claimed entirely.22 The EXAFS analysis for both SiO2-NH2/PdSA and SiO2-S-PdSA exhibits a single peak around r = ~ 2 Å (without phase shift correction), which is 6 different from the case of Pd foil and PdO crystal, indicating a lack of long-range periodicity and therefore, demonstrating Pd species are separately coordinated to the N and S, respectively. Curve-fitting analyses of the EXAFS results are also summarized in Table S3. The analysis revealed that SiO2-NH2/PdSA contains a Pd–N contribution, with a bond length of 2.06 Å and a coordination number (CN) of 4.0. In contrast, SiO2-S-PdSA contains a Pd–S contribution, with a slightly longer bond length of 2.30 Å and a similar CN of 4.3. These results suggest that Pd species in both samples are highly dispersed and coordinated by the grafted ligands.    Figure 1. Low-magnified and high-magnified HAADF-STEM images of (A, B) SiO2-NH2/PdSA, (C, D) SiO2-S-PdSA. Pd K-edge (E) XANES and (F) FT-EXAFS of SiO2-S-PdSA, SiO2-NH2/PdSA, and reference materials (PdO and Pd foil).  UV-vis reflectance spectra (Figure S7) also show that there is a chemical bonding between Pd and S/N, in which the Pd-S bonding has caused a significant enhancement in light absorption. Two characteristic peaks at 260 nm in SiO2-S-PdSA and 272 nm in SiO2-NH2/PdSA, are assignable to the ligand-centered transitions.31 A peak in SiO2 (205 nm), SiO2-NH2/Pd (209 nm), and SiO2-S-PdSA (213 nm) is also observed with a slight shift but an identical nature as characteristic of the SiO2 substrate structure.9,32 SiO2-S-PdSA indicates one extra and intense broadband at λmax = 387 nm, hinting at the strong coordination of -SH groups to Pd through d-d 7 transition and ligand-to-metal-charge-transition (LMCT),33,34 further demonstrating that in SiO2-S-PdSA, Pd species experience a fundamentally different coordination and bonding environment. The presence of LMCT-relevant peaks in thiol-to-Pd coordinated materials also shows that the Pd species in these materials are either in the form of clusters or single atoms since nanoparticles usually show weak-to-no peaks related to the LMCT.9,35  We also synthesized PdSA loaded on dithiocarbamate-functionalized SiO2 (SiO2-NHCS2H-PdSA,) via the treatment of the -NH2 functional groups of SiO2-NH2 with CS2 to convert NH2 into -NHCS2H groups and the subsequent Pd deposition. Likewise, 29Si MAS NMR and HAADF-STEM revealed the grafting of the NHCS2H functional group on SBA-15 and subsequent PdSA loading (Figure S8A, B). The XAS results of SiO2-NHSC2-Pd are in Figure S9. Accordingly, the absorption edge position of Pd in SiO2-NHSC2-PdSA is similar to that of the Pd species in SiO2-S-Pd, where the calculated bond length from EXAFS results is found to be 2.33 Å (Table S3). All the results described above indicate the successful loading of single-atom Pd on a mesoporous silica with only difference in organic functional groups (linkers) between the Pd and the silica surface in SiO2-NH2/PdSA, SiO2-S-PdSA, and SiO2-NHCS2H-PdSA, which must be helpful to investigate how the supporting environments of single-atom Pd affect its catalytic performance and stability.  Catalytic activity study Here, we selected formic acid (FA) dehydrogenation as a case catalytic reaction since FA is a facile H2 storing agent that can form by reacting H2 with CO2 and catalytically regenerating H2 with a metal-based catalyst such as Ir, Ru, Pd.12,36–38 We compared the catalytic activities of SiO2-NH2/PdSA and SiO2-S-PdSA in the catalytic dehydrogenation of FA under identical reaction conditions. Accordingly, SiO2-S-PdSA exhibited a lower H2 production yield during the monitored reaction time (90 min) than the SiO2-NH2/PdSA (Figure 2A). Furthermore, the turnover frequency (TOF, h−1) of SiO2-NH2/PdSA was obtained at 104.8 h−1, which is 7.6 folds higher than that of SiO2-S-PdSA (13.8 h−1) (Figure 2B) at 70 oC. As the reaction temperature downs to room temperature (22 oC), the catalytic activity gap between SiO2-NH2/PdSA (47.14 h−1) and SiO2-S-PdSA (1.6 h−1) increased to 29.5 folds higher catalytic activity (Figure S10). Further, the Pd wt% on SiO2-NH2 was also optimized by testing several loading amounts and 8 different catalyst quantities in the reaction. As shown in Figure S11, the optimal Pd wt% was 1.8 wt%, and the optimal amount of 1.8 wt% SiO2-NH2/PdSA was 10 mg to 5 mL FA (5 V/V%).   Figure 2. (A) Comparing the catalytic dehydrogenation activity of the SiO2-S-PdSA and SiO2-NH2/PdSA at 70 oC through FA (5 V/V%). (B) Comparing the TOF values of different single-atom Pd catalysts with SiO2-NH2/PdNP in FA dehydrogenation at 70 oC. (C) Recyclability study of SiO2-NH2/PdSA and SiO2-NH2/PdNP in FA acid dehydrogenation at 70 oC. HAADF-STEM image of the recovered SiO2-NH2/PdSA after FA dehydrogenation reaction. The inset shows the particle size distribution of the Pd nanoparticles. We hypothesize that this difference in the catalytic activity is because of the interplay between Pd and functional groups. To probe the suppressive role of thiol, we examined the catalytic activity of SiO2-NHCS2-PdSA under identical conditions to that of SiO2-S-PdSA. The -NHCS2H group is similar to -SH, being terminated with functional sulfur, and is expected to react with Pd like the thiol group. The observed catalytic activity (TOF = 1.7 h−1) for SiO2-NHCS2-PdSA was even lower than SiO2-S-PdSA (Figure 2B). This lower catalytic activity might be because the dithiocarbamate ligand in SiO2-NHCS2-Pd has two sulfur atoms in resonance with each other and the amine, making Pd-S bonds more covalent type and less active in the catalysis.39,40 Initial 2nd 3rd04080120160200 TOF (1/h)Cycles SiO2-NH2/PdNP SiO2-NH2/PdSA(C)0 20 40 60 80010203040506070H2 mmolSiO2-S-PdSATime (min)SiO2-NH2/PdSA04080120160SiO 2-NH 2/Pd NPSiO 2-NH 2/Pd SASiO 2-S-Pd SASiO 2-NHCS 2-Pd SATOF (1/h)(A) (B)(D)20 nm1.0 1.5 2.0 2.5051015 CountSize distribution (nm)1.8 nm9 To demonstrate if the observed superior catalytic activity in SiO2-NH2/PdSA can be correlated with the anime group’s basicity, regardless of what has been coordinated to Pd species, during the catalysis by SiO2-S-PdSA, we added propylamine compound, which corresponds to that of the functionalized propylamine in SiO2-NH2/PdSA. As shown in Figure S12, the added amine groups had a minor effect on the catalytic activity (from 13.8 h−1 to 18.5 h−1), indicating that amine basicity is not a critical factor in determining the Pd species’ catalytic activity.  Structural transformations and stability of PdSA on functionalized silica To investigate the stability of atomically dispersed Pd within the silica matrices, we post-treated each of the SiO2-S-PdSA and SiO2-NH2/PdSA samples with different methods, trying to understand the stability of PdSA in exposure to different reactive compounds. In SiO2-NH2/PdSA, after treating with concentrated NaBH4, PdSA converted into PdNP, observable under the STEM (Figure S13). Accordingly, the TOF of SiO2-NH2/PdNP was found to be 177.0 h−1, which is even higher than that of SiO2-NH2/PdSA (Figures 2B and S14) and is close to that of the recovered SiO2-NH2/PdSA (TOF = 178.5 h−1, Figure 2C). In SiO2-S-PdSA, the sample was first treated with a relatively harsh oxidating agent (i.e., H2O2), which detached the coordinated PdSA from thiol groups through oxidating -SH to -SO3H groups (Figure S15). HAADF-STEM observation and low-angle XRD results after H2O2 treatment showed the aggregation of Pd species and considerable destruction of the mesoscopic structures. Despite the destruction and accumulation, the H2O2-caused S-Pd bond breaking led to a better activity than the pristine sample (SiO2-S-PdSA) in catalytic formic acid dehydrogenation (Figure S16). This investigation also shows that the Pd-S coordination is detrimental to the catalytic efficiency of Pd. Unlike SiO2-NH2/PdSA, when SiO2-S-PdSA was treated with NaBH4 under identical conditions, no Pd aggregation was found under HAADF-STEM observation, showing that the Pd-S is more resistant to reducing agent than the Pd-N bond.  We now consider the positive role of amine functionality in Pd catalysis. The TEM image of the recovered SiO2-NH2/PdSA from the first run (Figure 2D) shows that Pd species in the recovered SiO2-NH2/PdSA have aggregated into nanoparticles, averaging in size ~1.8 nm (inset of Figure 2D) during the FA dehydrogenation reaction at 70 °C. The extent of such aggregation depends on the type of functional group coordinating with Pd and the reaction conditions (e.g., 10 temperature). Our previous studies about Pd thickness optimization on silicate sheets show that such a particle size can exhibit excellent metallicity and catalytic activity.10 The recovered SiO2-NH2/PdSA from the first run, in which the PdSA is converted into nanoparticles, exhibits a higher activity than the initial sample. Eventually, the catalytic activity stabilizes in the further runs under identical reaction conditions. The in situ-generated PdNP (TOF = 178.5 h−1) reveals better catalytic behavior than the SiO2-NH2/PdSA (104.8 h−1). The low catalytic activity of single atom Pd species appears in the early minutes of the catalysis of SiO2-NH2/PdSA—the interval that PdSA converts into PdNP to start the catalytic reaction. Again, unlike the recovered SiO2-NH2/PdSA, the recovered SiO2-S-PdSA showed no Pd aggregation during the formic acid dehydrogenation, judging from TEM observation (Figure S17) and wide angle XRD pattern (Figure S18). XPS further confirmed the high PdSA stability on the thiol functional group for the overall structure, where the Pd’s 3d peak positions have slightly changed compared to the pristine ones (discussed later in Figure 3). It can be inferred that when the PdSA is coordinated with the thiol, the Pd-S ligand bond is more stable than the Pd-N bond, so the Pd species is more inert toward FA. In conclusion, PdSA’s instability or mobility on a functionalized surface can be crucial in determining its catalytic activity. In conclusion, the recovered SiO2-NH2/PdSA—as Pd in situ-generated nanoparticles’ representative—reveal a superior activity than SiO2-S-PdSA—assumed as the Pd single atoms’ representative. The investigated wide-angle XRD patterns of PdSA-based catalysts (before and after FA dehydrogenation reaction recovery from the reactions indicated no relevant Pd diffraction pattern peak, although the recovered SiO2-NH2/Pd is composed of nanoparticles/clusters (Figure S18). This result shows that the Pd nanoparticles, during formic acid dehydrogenation, do not reach a particle size or crystal structure that could generate a diffraction pattern. Furthermore, in the long-term exposure of the catalyst in the reaction, the catalyst surface may deal with some surface deactivation that can be readily reactivated by heat treating at 80‒90 oC for 20 min (experimental section). This reactivation method shows that the surface of Pd species is probably hindered by (side) product molecules that suppress the catalytic activity by surface blockade. Therefore, the catalytic activity can be reverted to the initial state by heating the sample and desorbing the compounds attached to the surface.  11 Ex-situ catalytic mechanism investigation with XPS We further studied the catalytic mechanisms of SiO2-NH2/PdSA and SiO2-S-PdSA with X-ray photoelectron spectroscopy (XPS) measurement of Pd 3d. Since Pd(0) binding energy (BE) does not fairly shift in different media,41 while ionic Pd’s BE significantly shifts depending on the ligand type, we assigned a constant value to Pd(0) in the PdSA-loaded samples, while the ionic Pd species were assigned to variable BEs. In the case of SiO2-NH2/PdSA and SiO2-NH2/PdNP, the BE of 334.1 eV was assigned to Pd(0), and all the spectra were fitted accordingly. In the case of SH-contained samples, the Pd(0) BE was also assigned to a constant value, and all peaks were fitted accordingly. According to the fitted results from SiO2-NH2/PdSA, a maximum of ~7% of Pd species can be attributed to the Pd(0) species. However, after reducing with NaBH4 (SiO2-NH2/PdNP), the Pd(0) fraction increases up to 31%. Moreover, the abundance of the assigned peak of Pd(0) in the recovered SiO2-NH2/PdSA, after the formic acid dehydrogenation reaction, reaches up to 75%. The reduction of ionic Pd during the reaction shows that Pd single atoms on amine functional groups are mobile enough to migrate and aggregate into nanoparticles (Figure 3). A slight shift in ionic Pd peaks for SiO2-NH2/Pd samples is because ionic Pd’s coordination with the -NH2 functional group is looser and, besides, can bond with the other species such as surface silanols or other guest molecules (e.g., formate after from the solution). In contrast, SiO2-S-PdSA shows a stronger BE than SiO2-NH2/PdSA, and the peak positions of ionic Pd do not shift remarkably after the catalytic process (Figure 3). 12  Figure 3. Pd 3d XPS spectra of SiO2-NH2/PdSA, SiO2-NH2/PdNP, recovered SiO2-NH2/PdSA, SiO2-S-PdSA, Recovered SiO2-S-PdSA, and SiO2-NHCS2-PdSA. Forest green is representative of Pd(0) and lawn green shows ionic Pd.   The XPS spectrum of SiO2-S-PdSA after the recovery from the reaction mixture still shows the ionic Pd as majorly retained (>~98%), compared to the pristine SiO2-S-PdSA. This resistivity in reducing ionic Pd during the reaction on the SH-type functional group shows that the Pd species are strongly bonded with the SH species and are not mobile-like -NH2 functional group. Therefore, by looking at the catalytic activity (Figure 2A) and the transformability of ionic Pd to Pd(0) in SiO2-NH2/PdSA and SiO2-S-PdSA through XPS (Figure 3) and TEM results (Figure 2D and Figure S13), it can be concluded that the more bond strength between Pd and the functional group is, the less catalytic activity would be obtained in the FA dehydrogenation reaction.  It is also noteworthy that when the Pd is bonded to the dithiocarbamate, the BE of ionic Pd slightly shifts to lower BE, compared to the -SH functional group, which is due to the resonance between thiocarbonyl, -SH groups, and the amine (Figure 3).40 However, both thiol-based functional groups show a minor amount of Pd(0) in the structure, while the amino-based structure shows a higher fraction of Pd(0) in the pristine form.   13 Density Functional Investigations Our TEM and XAS results indicate that, at least for the first catalytic cycle, most Pd ions on both SiO2-NH2/PdSA and SiO2-S-PdSA samples would likely be deposited as single atoms instead of agglomerating to form Pd nanoparticles. Considering that Pd ions on SiO2-NH2 and SiO2-SH behave radically differently, we infer that the interplay between Pd ions and the functional groups is critical to the catalytic performance. Here, we resort to density functional calculations to elucidate this interplay at the atomic level. Figure 4A and Figure 4B show the optimized structures of a single Pd ion bonded to C3H6-NH2 and C3H6-S- groups when attached to a silica substrate. In the SiO2-NH2/PdSA system, the Pd ion forms a bond with the N ion and relaxes to a position relatively far from the surface, taking a distance of 3.929 Å. Here, the Pd-N bond length is 2.079 Å, where there is a good agreement with the obtained bond length value from Pd K-edge XAS (2.06 Å, Table S3). Pd ion in the SiO2-S-PdSA system, however, relaxes to a different geometry. In this case, although the C3H6-S-Pd group is stretched up to 3.353 Å away from the silica surface, the Pd ion falls closer to the silica surface, eventually, forming a 2.256 Å bond with S. The calculated Pd-S bond length here is also well-matching the obtained value from Pd K-edge XAS (2.30 Å). Pd’s interaction with the silica surface in the SiO2-S-PdSA was not readily apparent at the start of our investigations. Before simulating the configuration in Figure 4B, we investigated a structure where all three outmost O on the silica surface were capped with H. However, after we realized Pd’s tendency to relax toward the silica surface, we removed the capping H atom from the O towards which Pd relaxes, as oxygen is commonly coordinated by two cations. In this instance, those would be one Pd and one Si. The initial functional group configurations, without Pd loading, are shown in Figure S19, sharing many geometrical features with the one presented here, except that one of its surface oxygens is over-coordinated. The stability of Pd’s attachment to a functional group can be assessed by its formation energy (Ef), which is calculated by subtracting the total energy of a product from the total energies of its constituents. Using density functional energies instead to calculate the formation energies would be a viable approximation if the vibrational energies can be negligible. Accordingly, we calculated Pd’s formation energy in SiO2-NH2/PdSA and SiO2-S-PdSA, which indicates the Pd bond strength with the functional group, as: 14 Ef(SiO2-NH2/PdSA) = EDFT(SiO2-NH2/PdSA) − EDFT(SiO2-NH2) − μPd, Equation 1 Ef(SiO2-S-PdSA) = EDFT(SiO2-S-PdSA) − EDFT(SiO2-SH) − μPd + μH.  Equation 2 Here, EDFT denotes the density functional energy for the system with and without Pd ions. The systems without Pd, i.e., SiO2-NH2 and SiO2-SH, are shown in Figure S19. μPd and μH are palladium’s and hydrogen’s chemical potential, indicating the availability of these chemicals during the synthesis process. For instance, μPd can take values between EDFT(Pdmetallic) < μPd < 0, where EDFT(Pdmetallic) is the DFT energy of bulk metallic palladium per atom. Here, μPd = 0 is the limit where no Pd is available, while μPd = EDFT(Pdmetallic) signifies the abundance of Pd, so any lower μPd would make metallic Pd more stable than chelated Pd. Since the synthesis conditions for both single-atom bearing samples in our experiment were identical, the chemical potentials were taken to be the same in both equations. Moreover, since we are more interested in comparing the relative stability of Pd in these two configurations, the exact μPd and μH were less critical to our goal. So, we set μPd to be equal to metallic Pd’s DFT energy and μH to be equal to half of the DFT energy of H2 gas. We obtained Ef(SiO2-NH2/PdSA) = −0.189 eV, and Ef(SiO2-S-PdSA) = −2.475 eV. The relatively high Ef value for SiO2-NH2/PdSA, which is equivalent to −18.268 KJ/mol, demonstrates that Pd is attached through weak physical adsorption (physisorption) to the amino-functionalized surface than that of the thiol-functionalized surface. In contrast, the lower Ef for SiO2-S-PdSA, which is equivalent to −238.806 KJ/mol, is at the stronger bound of chemical adsorption. Consequently, the Pd bond within the SiO2-S-PdSA is substantially stronger, therefore, remains less disturbed by subsequent chemical reactions during catalysis. The nature of Pd’s bonding to functional groups can be further elucidated by analyzing the electronic localization function η.42 η is the probability of finding more than one spin-like electron at a point. In essence, η measures the degree of electron localization and provides a way to visualize the distribution of electrons.43 η values closer to zero indicate regions of low electron delocalization, and values closer to one indicate areas of high electron localization. As in Figure 4C and Figure 4E, η takes a global maximum near the anion and a global minimum at the bond center in Pd-N, Pd-S. This η profile indicates ionic bonding,44 whose strength depends on the transferred charge. Bader charge analysis shows that Pd has 15.92 valance electrons in SiO2-NH2/Pd and 15.48 e in SiO2-S-Pd. Note that the Pd pseudopotential contained 16 electrons, 15 4p64d105s0. Lower Pd charge in SiO2-S-Pd demonstrates a more significant charge transfer to S and O, resulting in stronger ionic bonding and higher chemical stability. According to the Sabatier principle, a good catalyst must be able to bond to the adsorbate, formic acid in our case, and be able to release the catalytic products relatively easily. The first half of the Sabatier principle is directly related to the transition metal’s ability to fill the adsorbate’s antibonding orbitals, shown in Figure 4G.45 To achieve this, a transition metal must have labile electrons in narrow d bands, not too far from the Fermi level.20 To assess the electronic lability of Pd in SiO2-NH2/PdSA and SiO2-S-PdSA, we integrated the Pd 4d states, shown in Figure 4D and Figure 4F, within the highlighted 3 eV window below the Fermi level. Electrons in this energy window are more likely to participate in electronic hybridization during the catalytic reactions.10,20,46–48 The SiO2-NH2/PdSA system had a total of 5.624 e labile electrons, while the SiO2-S-PdSA system had 2.335 e, less than half of SiO2-NH2/PdSA, explaining the better catalytic performance observed. The larger labile electronic population in SiO2/NH2/PdSA can be explained by the lack of hybridization with N 2s and 2p states, as filled N 2s and 2p states are concentrated at energies lower than −4 eV and empty N 2p states are far above the Fermi level at ~2 eV. This lack of hybridization leaves the Pd 4d states metallic-like, gravitating to the vicinity of the Fermi level.49 In SiO2-S-PdSA, Pd 4d states hybridize strongly with filled s and p states from both O and S in the region −7 eV < E < −3 eV, evidenced by many overlapping peaks. Consequently, the 4d states distribution over the energy range becomes more like a chalcogenide compound where a great deal of 4d states is pulled to lower energies.50 16   Figure 4. (A) and (B) The optimized structure of the SiO2-NH2/PdSA and SiO2-S-PdSA. (C) The electronic localization function (η) along the Pd-N bond in SiO2-NH2/PdSA. (D) The partial density of states of Pd 4d and its coordinating N 2s and 2p in SiO2-NH2/PdSA. (E) The electronic localization function (η) along the Pd-S bond in SiO2-SPdSA. (F) The partial density of states of Pd 4d and coordinating S 3s and 3p along with 2p in SiO2-S-PdSA. (G) Charge transfer mechanism from Pd catalyst to formic acid’s lowest unoccupied orbital facilitating the catalytic reaction.  Conclusion The electronic interaction between single atom Pd (PdSA) and different functional groups on mesoporous silica was investigated to determine the effect of a functional group on the catalytic activity of PdSA in formic acid dehydrogenation reaction. Comprehensive spectral analyses (synchrotron XAS and XPS) and TEM observations showed that the oxidation state of PdSA loaded on the amine functional group can readily change from ionic to Pd(0), while in the case of thiol-based functional groups, the oxidation state of Pd barely changes. Additionally, we observed an inverse relationship between PdSA’s catalytic activity and the stability of its oxidation state caused by the aggregation into nanoparticles during the reaction. The position of Pd’s 4d states with respect to the Fermi level can explain the catalytic behavior. When Pd’s 4d states concentrate closer to the Fermi level, they become more labile for catalytic activity. This occurs when the Pd bond with the functional group involves little to no charge transfer or when 17 Pd ions aggregate to form metallic nanoparticles. Both these scenarios manifest in the case of Pd atoms loaded on amine functional groups. In contrast, when the bond between Pd and the functional group is more ionic, more of Pd’s 4d states are pulled to deeper energy levels away from the Fermi level. In such a case, the deeper 4d states do not participate in catalysis. Stronger Pd-ligand bonding also prevents Pd aggregation into metallic nanoparticles. The latter case describes the behavior of PdSA loaded in thiol groups. Associated content The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsami.XXX.   Further details about the experimental section, including chemical and apparatus, XPS measurement, EXAFS/XANES measurements, synthesis of thiofunctionalized SiO2 (SiO2-SH), synthesis of aminofunctionalized SiO2 (SiO2-NH2), Synthesis of single atom PdSA, synthesis of dithiocarbamate-functionalized SiO2 (SiO2-NHCS2H), post-modification SiO2-NH2/PdSA and SiO2-S-PdSA, catalytic tests, and computational methods; and further information about 29Si MAS NMR and 13C CP MAS NMR spectra, TG-DTA curves, low-angle XRD patterns, N2 adsorption-desorption isotherms, HAADF-STEM images (including TEM-based elemental mappings) and UV-Vis spectra of SiO2-S-PdSA and SiO2-NH2/PdSA, characterizations of post-treated SiO2-NH2/PdSA and SiO2-S-PdSA, further catalytic dehydrogenation of FA under different conditions, characterizations of recovered samples, and the optimized model structures without Pd are available in the Supporting Information.   Acknowledgments This work was supported by JSPS KAKENHI (Grant Numbers 21H02034). This work was also supported by the Joint Usage/Research Center for Catalysis. The XAS measurements were performed at BL14B2 of SPring-8 (proposal No. 2022A1827). E. D. acknowledges the TÜBİTAK and Horizon-2020 Marie Skłodowska Curie for providing financial support in Co-Funded Brain Circulation Program (Project No. 120C057) framework.  18 Reference (1) Koga, Y.; Kaneda, T.; Saito, Y.; Murakami, K.; Itami, K. Synthesis of Partially and Fully Fused Polyaromatics by Annulative Chlorophenylene Dimerization. Science 2018, 359 (6374), 435–439. https://doi.org/10.1126/science.aap9801. (2) Trost, B. M.; Huang, Z.; Murhade, G. M. Catalytic Palladium-Oxyallyl Cycloaddition. Science 2018, 362 (6414), 564–568. https://doi.org/10.1126/science.aau4821. (3) Kiss, G. Palladium-Catalyzed Reppe Carbonylation. Chem. Rev. 2001, 101 (11), 3435–3456. (4) Motokura, K.; Ding, S.; Usui, K.; Kong, Y. Enhanced Catalysis Based on the Surface Environment of the Silica-Supported Metal Complex. ACS Catal. 2021, 11 (19), 11985–12018. https://doi.org/10.1021/acscatal.1c03426. (5) Martis, M.; Mori, K.; Fujiwara, K.; Ahn, W.-S.; Yamashita, H. Amine-Functionalized MIL-125 with Imbedded Palladium Nanoparticles as an Efficient Catalyst for Dehydrogenation of Formic Acid at Ambient Temperature. J. Phys. Chem. C 2013, 117 (44), 22805–22810. https://doi.org/10.1021/jp4069027. (6) Zhang, J.; Liao, W.; Zheng, H.; Zhang, Y.; Xia, L.; Teng, B.-T.; Lu, J.-Q.; Huang, W.; Zhang, Z. Morphology-Engineered Highly Active and Stable Pd/TiO2 Catalysts for CO2 Hydrogenation into Formate. J. Catal. 2022, 405, 152–163. https://doi.org/10.1016/j.jcat.2021.11.035. (7) Wen, M.; Mori, K.; Kuwahara, Y.; Yamashita, H. Plasmonic Au@Pd Nanoparticles Supported on a Basic Metal–Organic Framework: Synergic Boosting of H2 Production from Formic Acid. ACS Energy Lett. 2017, 2 (1), 1–7. https://doi.org/10.1021/acsenergylett.6b00558. (8) Wang, Z.; Liang, S.; Meng, X.; Mao, S.; Lian, X.; Wang, Y. Ultrasmall PdAu Alloy Nanoparticles Anchored on Amine-Functionalized Hierarchically Porous Carbon as Additive-Free Catalysts for Highly Efficient Dehydrogenation of Formic Acid. Appl. Catal., B 2021, 291, 120140. https://doi.org/10.1016/j.apcatb.2021.120140. (9) Wang, N.; Sun, Q.; Bai, R.; Li, X.; Guo, G.; Yu, J. In Situ Confinement of Ultrasmall Pd Clusters within Nanosized Silicalite-1 Zeolite for Highly Efficient Catalysis of Hydrogen Generation. J. Am. Chem. Soc. 2016, 138 (24), 7484–7487. https://doi.org/10.1021/jacs.6b03518. (10) Doustkhah, E.; Tsunoji, N.; Assadi, M. H. N.; Ide, Y. Pd Thickness Optimization on Silicate Sheets for Improving Catalytic Activity. Adv. Mater. Interfaces 2023, 10 (4), 2202368. https://doi.org/10.1002/admi.202202368. (11) Lim, B.; Xia, Y. Metal Nanocrystals with Highly Branched Morphologies. Angew. Chem., Int. Ed. 2011, 50 (1), 76–85. https://doi.org/10.1002/anie.201002024. (12) Doustkhah, E.; Hasani, M.; Ide, Y.; Assadi, M. H. N. Pd Nanoalloys for H2 Generation from Formic Acid. ACS Appl. Nano Mater. 2020, 3 (1), 22–43. https://doi.org/10.1021/acsanm.9b02004. (13) Gao, F.; Goodman, D. W. Pd–Au Bimetallic Catalysts: Understanding Alloy Effects from Planar Models and (Supported) Nanoparticles. Chem. Soc. Rev. 2012, 41 (24), 8009–8020. https://doi.org/10.1039/C2CS35160A. (14) Chen, P.; Khetan, A.; Yang, F.; Migunov, V.; Weide, P.; Stürmer, S. P.; Guo, P.; Kähler, K.; Xia, W.; Mayer, J.; Pitsch, H.; Simon, U.; Muhler, M. Experimental and Theoretical Understanding of Nitrogen-Doping-Induced Strong Metal–Support Interactions in Pd/TiO2 Catalysts for Nitrobenzene Hydrogenation. ACS Catal. 2017, 7 (2), 1197–1206. https://doi.org/10.1021/acscatal.6b02963. (15) Weerachawanasak, P.; Mekasuwandumrong, O.; Arai, M.; Fujita, S.-I.; Praserthdam, P.; Panpranot, J. Effect of Strong Metal–Support Interaction on the Catalytic Performance of Pd/TiO2 in the Liquid-Phase Semihydrogenation of Phenylacetylene. J. Catal. 2009, 262 (2), 199–205. https://doi.org/10.1016/j.jcat.2008.12.011. (16) Ament, K.; Köwitsch, N.; Hou, D.; Götsch, T.; Kröhnert, J.; Heard, C. J.; Trunschke, A.; Lunkenbein, T.; Armbrüster, M.; Breu, J. Nanoparticles Supported on Sub-Nanometer Oxide Films: Scaling Model Systems to Bulk Materials. Angew. Chem., Int. Ed. 2021, 60 (11), 5890–5897. https://doi.org/10.1002/anie.202015138. 19 (17) Du, L.; Qian, K.; Zhu, X.; Yan, X.; Kobayashi, H.; Liu, Z.; Lou, Y.; Li, R. Interface Engineering of Palladium and Zinc Oxide Nanorods with Strong Metal–Support Interactions for Enhanced Hydrogen Production from Base-Free Formaldehyde Solution. J. Mater. Chem. A 2019, 7 (15), 8855–8864. https://doi.org/10.1039/C8TA12019F. (18) Saptal, V. B.; Ruta, V.; Bajada, M. A.; Vilé, G. Single-Atom Catalysis in Organic Synthesis. Angewandte Chemie International Edition 2023, 62 (34), e202219306. https://doi.org/10.1002/anie.202219306. (19) Xu, H.; Zhao, Y.; Wang, Q.; He, G.; Chen, H. Supports Promote Single-Atom Catalysts toward Advanced Electrocatalysis. Coord. Chem. Rev. 2022, 451, 214261. https://doi.org/10.1016/j.ccr.2021.214261. (20) Greiner, M. T.; Jones, T. E.; Beeg, S.; Zwiener, L.; Scherzer, M.; Girgsdies, F.; Piccinin, S.; Armbrüster, M.; Knop-Gericke, A.; Schlögl, R. Free-Atom-like d States in Single-Atom Alloy Catalysts. Nat. Chem. 2018, 10 (10), 1008–1015. https://doi.org/10.1038/s41557-018-0125-5. (21) Shan, J.; Ye, C.; Jiang, Y.; Jaroniec, M.; Zheng, Y.; Qiao, S.-Z. Metal-Metal Interactions in Correlated Single-Atom Catalysts. Sci. Adv. 2022, 8 (17), eabo0762. https://doi.org/10.1126/sciadv.abo0762. (22) Finzel, J.; Sanroman Gutierrez, K. M.; Hoffman, A. S.; Resasco, J.; Christopher, P.; Bare, S. R. Limits of Detection for EXAFS Characterization of Heterogeneous Single-Atom Catalysts. ACS Catal. 2023, 6462–6473. https://doi.org/10.1021/acscatal.3c01116. (23) Li, R.; Wang, D. Understanding the Structure–Performance Relationship of Active Sites at Atomic Scale. Nano Research 2022, 15 (8), 6888–6923. https://doi.org/10.1007/s12274-022-4371-x. (24) Liu, J.; Zou, Y.; Cruz, D.; Savateev, A.; Antonietti, M.; Vilé, G. Ligand–Metal Charge Transfer Induced via Adjustment of Textural Properties Controls the Performance of Single-Atom Catalysts during Photocatalytic Degradation. ACS Appl. Mater. Interfaces 2021, 13 (22), 25858–25867. https://doi.org/10.1021/acsami.1c02243. (25) Gan, T.; Wang, D. Atomically Dispersed Materials: Ideal Catalysts in Atomic Era. Nano Research 2023. https://doi.org/10.1007/s12274-023-5700-4. (26) Bajada, M. A.; Di Liberto, G.; Tosoni, S.; Ruta, V.; Mino, L.; Allasia, N.; Sivo, A.; Pacchioni, G.; Vilé, G. Light-Driven C–O Coupling of Carboxylic Acids and Alkyl Halides over a Ni Single-Atom Catalyst. Nature Synthesis 2023. https://doi.org/10.1038/s44160-023-00341-3. (27) Zhang, Z.; Zhu, J.; Chen, S.; Sun, W.; Wang, D. Liquid Fluxional Ga Single Atom Catalysts for Efficient Electrochemical CO2 Reduction. Angew. Chem., Int. Ed. 2023, 62 (3), e202215136. https://doi.org/10.1002/anie.202215136. (28) Singh, B.; Sharma, V.; Gaikwad, R. P.; Fornasiero, P.; Zbořil, R.; Gawande, M. B. Single-Atom Catalysts: A Sustainable Pathway for the Advanced Catalytic Applications. Small 2021, 17 (16), 2006473. https://doi.org/10.1002/smll.202006473. (29) Vilé, G.; Albani, D.; Nachtegaal, M.; Chen, Z.; Dontsova, D.; Antonietti, M.; López, N.; Pérez-Ramírez, J. A Stable Single-Site Palladium Catalyst for Hydrogenations. Angew. Chem., Int. Ed. 2015, 54 (38), 11265–11269. https://doi.org/10.1002/anie.201505073. (30) Lettow, J. S.; Han, Y. J.; Schmidt-Winkel, P.; Yang, P.; Zhao, D.; Stucky, G. D.; Ying, J. Y. Hexagonal to Mesocellular Foam Phase Transition in Polymer-Templated Mesoporous Silicas. Langmuir 2000, 16 (22), 8291–8295. https://doi.org/10.1021/la000660h. (31) Xu, Y.; Claiden, P.; Zhu, Y.; Morita, H.; Hanagata, N. Effect of Amino Groups of Mesoporous Silica Nanoparticles on CpG Oligodexynucleotide Delivery. Sci. Technol. Adv. 2015, 16 (4), 045006. https://doi.org/10.1088/1468-6996/16/4/045006. (32) Jin, M.; Guo, Z.; Lv, Z. Immobilization of Tungsten Chelate Complexes on Functionalized Mesoporous Silica SBA-15 as Heterogeneous Catalysts for Oxidation of Cyclopentene. J. Mater. Sci. 2019, 54 (9), 6853–6866. https://doi.org/10.1007/s10853-019-03361-7. (33) Potter, M. E.; Le Brocq, J. J. M.; Oakley, A. E.; McShane, E. B.; Mhembere, P. M.; Carravetta, M.; Vandegehuchte, B. D.; Raja, R. Thiol Functionalised Supports for Controlled Metal Nanoparticle 20 Formation for Improved C–C Coupling. Chem. Asian J. 2021, 16 (22), 3610–3614. https://doi.org/10.1002/asia.202100732. (34) Yang, Z.; Smetana, A. B.; Sorensen, C. M.; Klabunde, K. J. Synthesis and Characterization of a New Tiara Pd(II) Thiolate Complex, [Pd(SC12H25)2]6, and Its Solution-Phase Thermolysis to Prepare Nearly Monodisperse Palladium Sulfide Nanoparticles. Inorg. Chem. 2007, 46 (7), 2427–2431. https://doi.org/10.1021/ic061242o. (35) Creighton, J. A.; Eadon, D. G. Ultraviolet–Visible Absorption Spectra of the Colloidal Metallic Elements. J. Chem. Soc., Faraday Trans. 1991, 87 (24), 3881–3891. https://doi.org/10.1039/FT9918703881. (36) Guan, C.; Zhang, D.-D.; Pan, Y.; Iguchi, M.; Ajitha, M. J.; Hu, J.; Li, H.; Yao, C.; Huang, M.-H.; Min, S.; Zheng, J.; Himeda, Y.; Kawanami, H.; Huang, K.-W. Dehydrogenation of Formic Acid Catalyzed by a Ruthenium Complex with an N,N′-Diimine Ligand. Inorg. Chem. 2017, 56 (1), 438–445. https://doi.org/10.1021/acs.inorgchem.6b02334. (37) Himeda, Y. Highly Efficient Hydrogen Evolution by Decomposition of Formic Acid Using an Iridium Catalyst with 4,4′-Dihydroxy-2,2′-Bipyridine. Green Chem. 2009, 11 (12), 2018–2022. https://doi.org/10.1039/B914442K. (38) Hull, J. F.; Himeda, Y.; Wang, W.-H.; Hashiguchi, B.; Periana, R.; Szalda, D. J.; Muckerman, J. T.; Fujita, E. Reversible Hydrogen Storage Using CO2 and a Proton-Switchable Iridium Catalyst in Aqueous Media under Mild Temperatures and Pressures. Nat. Chem. 2012, 4 (5), 383–388. https://doi.org/10.1038/nchem.1295. (39) von Wrochem, F.; Gao, D.; Scholz, F.; Nothofer, H.-G.; Nelles, G.; Wessels, J. M. Efficient Electronic Coupling and Improved Stability with Dithiocarbamate-Based Molecular Junctions. Nat. Nanotech. 2010, 5 (8), 618–624. https://doi.org/10.1038/nnano.2010.119. (40) Sauter, E.; Nascimbeni, G.; Trefz, D.; Ludwigs, S.; Zojer, E.; von Wrochem, F.; Zharnikov, M. A Dithiocarbamate Anchoring Group as a Flexible Platform for Interface Engineering. Phys. Chem. Chem. Phys. 2019, 21 (40), 22511–22525. https://doi.org/10.1039/C9CP03306H. (41) Moulder, J. F.; Stickle, W. F.; Sobol, P. E.; Bomben, K. D. Handbook of X-Ray Photoelectron Spectroscopy; Perkin-Elmer Corporation: Eden Prairie, MN, 1992; 1992. (42) Becke, A. D.; Edgecombe, K. E. A Simple Measure of Electron Localization in Atomic and Molecular Systems. J. Chem. Phys. 1990, 92 (9), 5397–5403. https://doi.org/10.1063/1.458517. (43) Savin, A.; Nesper, R.; Wengert, S.; Fässler, T. F. ELF: The Electron Localization Function. Angew. Chem., Int. Ed. 1997, 36 (17), 1808–1832. https://doi.org/10.1002/anie.199718081. (44) Fuentealba, P.; Chamorro, E.; Santos, J. C. Chapter 5 Understanding and Using the Electron Localization Function. In Theoretical and Computational Chemistry; Toro-Labbé, A., Ed.; Elsevier, 2007; Vol. 19, pp 57–85. https://doi.org/10.1016/S1380-7323(07)80006-9. (45) Nilsson, A.; Pettersson, L. G. M.; Hammer, B.; Bligaard, T.; Christensen, C. H.; Nørskov, J. K. The Electronic Structure Effect in Heterogeneous Catalysis. Catal. Lett. 2005, 100 (3), 111–114. https://doi.org/10.1007/s10562-004-3434-9. (46) Nørskov, J. K.; Abild-Pedersen, F.; Studt, F.; Bligaard, T. Density Functional Theory in Surface Chemistry and Catalysis. Proc. Natl. Acad. Sci. 2011, 108 (3), 937–943. https://doi.org/10.1073/pnas.1006652108. (47) Assadi, M. H. N.; Shigeta, Y. The Effect of Octahedral Distortions on the Electronic Properties and Magnetic Interactions in O3 NaTMO2 Compounds (TM = Ti–Ni & Zr–Pd). RSC Adv. 2018, 8 (25), 13842–13849. https://doi.org/10.1039/C8RA00576A. (48) Yeo, B. C.; Nam, H.; Nam, H.; Kim, M.-C.; Lee, H. W.; Kim, S.-C.; Won, S. O.; Kim, D.; Lee, K.-Y.; Lee, S. Y.; Han, S. S. High-Throughput Computational-Experimental Screening Protocol for the Discovery of Bimetallic Catalysts. Npj Comput. Mater. 2021, 7 (1), 137. https://doi.org/10.1038/s41524-021-00605-6. (49) Todorova, M.; Reuter, K.; Scheffler, M. Oxygen Overlayers on Pd(111) Studied by Density Functional Theory. J. Phys. Chem. B 2004, 108 (38), 14477–14483. https://doi.org/10.1021/jp040088t. 21 (50) Blanco-Rey, M.; Jenkins, S. J. Methane Dissociation and Methyl Diffusion on PdO{100}. J. Chem. Phys. 2009, 130 (1), 014705. https://doi.org/10.1063/1.3046683.   Graphical abstract: