No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Novel Insight into the Hydrogen Absorption Mechanism at the Pd(110) Surface
The Journal of Chemical Physics
Abstract
The microscopic mechanism of low-temperature (80 K < T < 160 K) hydrogen (H) ingress into the H2 (<2.66 × 10(-3) Pa) exposed Pd(110) surface is explored by H depth profiling with (15)N nuclear reaction analysis (NRA) and thermal desorption spectroscopy (TDS) with isotope (H, D) labeled surface hydrogen. NRA and TDS reveal two types of absorbed hydrogen states of distinctly different depth distributions. Between 80 K and ∼145 K a near-surface hydride phase evolving as the TDS α1 feature at 160 K forms, which initially extends only several nanometers into depth. On the other hand, a bulk-absorbed hydrogen state develops between 80 K and ∼160 K which gives rise to a characteristic α3 TDS feature above 190 K. These two absorbed states are populated at spatially separated surface entrance channels. The near-surface hydride is populated through rapid penetration at minority sites (presumably defects) while the bulk-absorbed state forms at regular terraces with much lower probability per site. In both cases, absorption of gas phase hydrogen transfers pre-adsorbed hydrogen atoms below the surface and replaces them at the chemisorption sites by post-dosed hydrogen in a process that requires much less activation energy (<100 meV) than monatomic diffusion of chemisorbed H atoms into subsurface sites. This small energy barrier suggests that the rate-determining step of the absorption process is either H2 dissociation on the H-saturated Pd surface or a concerted penetration mechanism, where excess H atoms weakly bound to energetically less favorable adsorption sites stabilize themselves in the chemisorption wells while pre-chemisorbed H atoms simultaneously transit into the subsurface. The peculiarity of absorption at regular Pd(110) terraces in comparison to Pd(111) and Pd(100) is discussed.
... However, the surface is an additional barrier to hydrogen leaving the metal. The fact that hydrogen must dissociate and recombine at the surface complicates the description, but even for surfaces without a dissociation barrier, such as Pd with = 0, there is a finite desorption barrier [10,[13][14][15][16]. This leads to the inconvenient fact that only Pd and Pd alloys are technically feasible membrane materials; or, if other hydrogenpermeable membrane materials such as vanadium are used, they must be coated with an additional layer to facilitate hydrogen desorption from the surface, usually palladium [11,12,17,18]. ...
... Second criterion must be a finite hydrogen bulk solubility, although ultrathin overlayers may circumvent this restriction. In the 80ies, Behm et al. [13,14], and later Ohno et al. [16] associated the peculiarity of Pd for its fast hydrogen sorption kinetics with the existence of a sub-surface site with intermediate hydrogen sorption enthalpy. These sites would facilitate hydrogen sorption by a ''concerted penetration mechanism, where excess H atoms weakly bound to energetically less favorable adsorption sites stabilize themselves in the chemisorption wells while pre-chemisorbed H atoms simultaneously transit into the subsurface.'' ...
... These sites would facilitate hydrogen sorption by a ''concerted penetration mechanism, where excess H atoms weakly bound to energetically less favorable adsorption sites stabilize themselves in the chemisorption wells while pre-chemisorbed H atoms simultaneously transit into the subsurface.'' [16] In this paper we exploit this idea to rationalize the future development of membrane surfaces. We introduce an experimental concept and apply it to the archetypal hydrogen-palladium system. ...
Hydrogen selective metallic membranes made of palladium and its alloys are superior to membranes made of other metallic materials due to their unique surface properties. However, the prohibitive cost of Pd has prevented their widespread application. Although the basic mechanisms are known, the link between hydrogen at the surface and that in the bulk has been missing to explain the prominent position of Pd. To provide the missing link between surface and bulk, we measured the hydrogen concentration and depth distribution on the downstream side of a Pd membrane
at realistic permeation conditions (temperature and pressure) using a membrane electron spectroscopy setup. We find hydrogen in the vicinity of the surface with intermediate stability.
In general, hydrogen on the surface is more stable and thus more concentrated than in the bulk. Subsurface hydrogen with intermediate stability mitigates the difference between surface and bulk concentration and thereby facilitates overall hydrogen permeation. The observations link the results of UHV-compatible hydrogen adsorption experiments with observations made on hydrogen permeation in membranes under technical conditions. Confirming this link enables the application of the wealth of surface science based knowledge to the optimization of hydrogen permeable membrane materials.
... Pd is known as a hydrogen-absorbing material. In the hydrogen-absorption process at Pd surfaces, the penetration process from the surface into the subsurface region is recognized as the rate-limiting step (5,6). Since the surface governs the penetration process, the penetration kinetics are expected to be controllable by modifying the surface, such as its atomic structure (7) or by alloying with other materials. ...
... From clean Pd(110), three desorption peaks are observed at 153 (α 1 ), 220 (α 2 ), and 300 K (β). As reported previously, the α 1 peak corresponds to absorbed hydrogen (hydride phase) in a few nanometer deep subsurface region, the α 2 peak is related to the lifting of a hydrogen adsorption-induced surface reconstruction of Pd(110), and the β peak can be attributed to chemisorbed surface hydrogen (6,14,15). On the 0.3-ML Au/Pd(110) surface, the intensity of the α 1 peak increases significantly, the intensity of the β peak is almost the same compared with Pd(110), and the α 2 peak disappears. The peak temperature of the α 1 feature shifts to 173 K and that of the β peak shifts to 295 K. On the 1.8-ML Au/Pd(110) surface, ...
... The NRA profile from the clean Pd(110) surface shows a Gaussian-like feature centered around depth 0 and a low-intensity decaying feature up to 10 nm. The Gaussian-like feature corresponds to the chemisorbed surface hydrogen and the decaying feature corresponds to the absorbed hydrogen into the near-surface region (6). The NRA profile from Au/Pd(110) shows a high γ-ray yield near the surface that decays into the depth up to 10 nm, which indicates substantial H incorporation into the near-surface region as well as chemisorption of H on the surface. ...
Significance
Surfaces are gates to control the transport of energy and materials between the gas phase and bulk. For the hydrogen storage, the transport of hydrogen across the surface is recognized as the bottleneck, e.g., 1 H 2 in 1,000 impinging a Pd surface penetrates the surface. Here, we demonstrate that alloying the Pd(110) surface with submonolayer amounts of Au dramatically accelerates the hydrogen absorption, by a factor of more than 40. This discovery will lead to enhancement of hydrogen absorption kinetics, thereby improving the performance of hydrogen-purifying membranes and hydrogen-storage materials, which is a key for utilizing hydrogen as a carbon-free energy carrier.
... Thus, hydrogen atoms can easily diffuse into materials. Initially, hydrogen atoms occupy the interstitial sites of the crystalline lattice (a phase) [6]. As hydrogen absorption increases, hydrogen atoms expand the crystalline lattice (b phase) [6]. ...
... Initially, hydrogen atoms occupy the interstitial sites of the crystalline lattice (a phase) [6]. As hydrogen absorption increases, hydrogen atoms expand the crystalline lattice (b phase) [6]. ...
... The time constant of the rapid reduction t d1 was 1 ± 0.5 s and that of the subsequent slower reduction t d2 was 27 ± 5 s. The presence of two time constants, t d1 and t d2 , might be due to the difference in hydrogen desorption rates in the band a-states [6]. The bonding strength of a-PdH is higher than that of b-PdH, thus making t d2 larger than t d1 . ...
... [5][6][7] Palladium (Pd), a noble transition metal known for catalyzing hydrogen-based reduction reactions, [8,9] is particularly well-studied in this context. [10][11][12][13] Previous studies revealed that high hydrogen coverage induces a reconstruction of the least stable low-index Pd(110) surface, leading to formation of pairing-row or missing-row structures. [14][15][16][17][18][19] First-principles calculations based on density functional theory (DFT) suggest that migration of Pd atoms and subsequent reconstructions arise due to interactions with diffusing hydrogen atoms. ...
... [42] This explains the observed missingrow reconstruction of Pd(110) during hydride formation. [13,20,43] In case of PdH/Pd(100), the surface roughening is induced by lifting some surface atoms and formation of (111) facets (Figure 5c). This indicates that also the Pd(100) surface tends to transform into the most stable (111) termination upon hydride formation. ...
Designing electrocatalysts with optimal activity and selectivity relies on a thorough understanding of the surface structure under reaction conditions. In this study, experimental and computational approaches are combined to elucidate reconstruction processes on low‐index Pd surfaces during H‐insertion following proton electroreduction. While electrochemical scanning tunneling microscopy clearly reveals pronounced surface roughening and morphological changes on Pd(111), Pd(110), and Pd(100) surfaces during cyclic voltammetry, a complementary analysis using inductively coupled plasma mass spectrometry excludes Pd dissolution as the primary cause of the observed restructuring. Large‐scale molecular dynamics simulations further show that these surface alterations are related to the creation and propagation of structural defects as well as phase transformations that take place during hydride formation.
... Hydrogen atoms absorbed in Pd may form α-and β-palladium hydride phases depending on the hydrogen concentration inside the metal. Hydrogen atoms mainly occupy the interstitial sites of the crystalline lattice for the both phases [76]. At low hydrogen concentrations, the solid solution, α-phase, is formed. ...
... While the hydrogen concentration increases, phase β-palladium hydride is formed. The hydrogen molecules are dissociated into atoms and firstly adsorbed at the surface of palladium by high-symmetry hollow chemisorption sites, and then they fill the octahedral interstitial sites in the first subsurface layer and finally diffuse and penetrate the interstitial sites inside the bulk [76]. At room temperature, the pure α-phase palladium hydride corresponds to a stoichiometry x < 0.017 while pure β-phase is realised for x > 0.58 (where x is the stoichiometry of PdH x ). ...
Devices enabling early detection of low concentrations of leaking hydrogen and precision measurements in a wide range of hydrogen concentrations in hydrogen storage systems are essential for the mass-production of fuel-cell vehicles and, more broadly, for the transition to the hydrogen economy. Whereas several competing sensor technologies are potentially suitable for this role, ultra-low fire-hazard, contactless and technically simple magneto-electronic sensors stand apart because they have been able to detect the presence of hydrogen gas in a range of hydrogen concentrations from 0.06% to 100% at atmospheric pressure with the response time approaching the industry gold standard of one second. This new kind of hydrogen sensors is the subject of this review article, where we inform academic physics, chemistry, material science and engineering communities as well as industry researchers about the recent developments in the field of magneto-electronic hydrogen sensors, including those based on magneto-optical Kerr effect, anomalous Hall effect and Ferromagnetic Resonance with a special focus on Ferromagnetic Resonance (FMR)-based devices. In particular, we present the physical foundations of magneto-electronic hydrogen sensors and we critically overview their advantages and disadvantages for applications in the vital areas of the safety of hydrogen-powered cars and hydrogen fuelling stations as well as hydrogen concentration meters, including those operating directly inside hydrogen-fuelled fuel cells. We believe that this review will be of interest to a broad readership, also facilitating the translation of research results into policy and practice.
... Hydrogen atoms absorbed in Pd may form α-and β-palladium hydride phases depending on the hydrogen concentration inside the metal. Hydrogen atoms mainly occupy the interstitial sites of the crystalline lattice for the both phases [74]. At low hydrogen concentrations, the solid solution, α-phase, is formed. ...
... While the hydrogen concentration increases, the metal hydride, β-palladium hydride, is formed. The hydrogen molecules are dissociated into atoms and firstly adsorbed at the surface of palladium by the high-symmetry hollow chemisorption sites, and then filling the octahedral interstitial sites in the first subsurface layer and finally diffuse and penetrate the interstitial sites inside the bulk [74]. At room temperature, the pure α-phase palladium hydride corresponds to a stoichiometry x < 0.017 while pure β-phase is realised for x > 0. 58 (where x is the stoichiometry of PdH x ). ...
Devices enabling early detection of low concentrations of leaking hydrogen and precision measurements in a wide range of hydrogen concentrations in hydrogen storage systems are essential for the mass-production of fuel-cell vehicles and, more broadly, for the transition to the hydrogen economy. Whereas several competing sensor technologies are potentially suitable for this role, ultra-low fire-hazard, contactless and technically simple magneto-electronic sensors stand apart because they have been able to detect the presence of hydrogen gas in a range of hydrogen concentrations from 0.06% to 100% at atmospheric pressure with the response time approaching the industry gold standard of one second. This new kind of hydrogen sensors is the subject of this review article, where we inform the academic physics, chemistry, material science and engineering communities as well as industry researchers about the recent developments in the field of magneto-electronic hydrogen sensors, including those based on magneto-optical Kerr effect, anomalous Hall effect and Ferromagnetic Resonance with a special focus on Ferromagnetic Resonance (FMR) based devices. In particular, we present the physical foundations of magneto-electronic hydrogen sensors and we critically overview their advantages and disadvantages for applications in the vital areas of the safety of hydrogen-powered cars and hydrogen fuelling stations as well as hydrogen concentration meters, including those operating directly inside hydrogen-fuelled fuel cells. We believe that this review will be of interest to a broad readership, also facilitating the translation of research results into policy and practice.
... 62 When it comes to the rate-determining step of hydrogen absorption, various studies have shown different outcomes. Some studies argued that the rate-determining step is the hydrogen dissociation, 63,64 while others claimed it is the filling of subsurface sites 60,63,65 or bulkdiffusion. [66][67][68] It is worth noting that those studies were done for different film thickness/particle size, surface (facets), temperatures and hydrogen pressure, which might explain the different outcomes. ...
... 62 When it comes to the rate-determining step of hydrogen absorption, various studies have shown different outcomes. Some studies argued that the rate-determining step is the hydrogen dissociation, 63,64 while others claimed it is the filling of subsurface sites 60,63,65 or bulkdiffusion. [66][67][68] It is worth noting that those studies were done for different film thickness/particle size, surface (facets), temperatures and hydrogen pressure, which might explain the different outcomes. ...
Plasmonic metal nanoparticles and polymer materials have independently undergone rapid development during the last two decades. More recently, it has been realized that combining these two systems in a hybrid or nanocomposite material comprised of plasmonically active metal nanoparticles dispersed in a polymer matrix leads to systems that exhibit fascinating properties, and some first attempts had been made to exploit them for optical spectroscopy, solar cells or even pure art. In my thesis, I have applied this concept to tackle the urgent problem of hydrogen safety by developing Pd nanoparticle-based “plasmonic plastic” hybrid materials, and by using them as the active element in optical hydrogen sensors. This is motivated by the fact that hydrogen gas, which constitutes a clean and sustainable energy vector, poses a risk for severe accidents due to its high flammability when mixed with air.
Therefore, hydrogen leak detection systems are compulsory in the imminent large-scale dissemination of hydrogen energy technologies. To date, however, there a several unresolved challenges in terms of hydrogen sensor performance, whereof too slow sensor response/recovery times and insufficient resistance towards deactivation by poisoning species are two of the most severe ones.
In this thesis, I have therefore applied the plasmonic plastic hybrid material concept to tackle these challenges. In summary, I have (i) developed hysteresis-free plasmonic hydrogen sensors based on PdAu, PdCu and PdAuCu alloy nanoparticles; (ii) demonstrated ultrafast sensor response and stable sensor operation in chemically challenging environments using polymer coatings; (iii) introduced bulk-processed and 3D printed plasmonic plastic hydrogen sensors with fast response and high resistance against poisoning and deactivation.
... On the other hand, some recent experiments suggest that exposure of the surface to high-pressure H 2 urges H penetration into the inside. [16][17][18] These facts indicate that exploring the interaction between LH 2 or high-density H 2 gas and materials is inevitable to advance H storage technology. ...
... 3(a) and 3(b) corresponds to the H penetration from the surface into near-surface region due to highpressure effects. 16,17 Taking into account that the H diffusion in bulk is rate-limiting process when H 2 pressure is sufficiently high, 11 Thermal hopping and QT are the two possible diffusion modes. In thermal diffusion, H atoms in fcc octahedral Pd sites must acquire the energy to surmount the activation barrier E diff $ 226 meV. ...
We report real-time detection of hydrogen (H) absorption in metallic palladium (Pd) nano-contacts immersed in liquid H2 using inelastic electron spectroscopy (IES). After introduction of liquid H2, the spectra exhibit the time evolution from the pure Pd to the Pd hydride, indicating that H atoms are absorbed in Pd nano-contacts even at the temperature where the thermal process is not expected. The IES time and bias voltage dependences show that H absorption develops by applying bias voltage 30 ∼ 50 mV, which can be explained by quantum tunneling. The results represent that IES is a powerful method to study the kinetics of high density H on solid surface.
... In contrast, AFM analysis indicates that the surface concentration of hydrogen remains largely unchanged until ∼135 s into the scan and then depopulates almost instantaneously. These observations are consistent with studies by Ohno et al. 30 that document the presence of multiple populations of hydrogen upon interaction with Pd. Further discussion of this concept follows. ...
... The virtual step change of the AFM phase-angle observation (Figure 4b) may complement recent work to elucidate the still ill-defined physical processes during sorption of hydrogen from the gas phase into Pd. Using 15 N nuclear reaction analysis (NRA) and thermal desorption spectroscopy (TDS) with surfaces labeled with isotopes (H and D), Ohno et al. 30 report evidence of two different hydrogen populations at or near the Pd surface. If distinct populations of hydrogen exist in the hydrogensaturated state of Pd, then hydrogen would be expected to vacate Pd in a nonuniform fashion. ...
Characterization of the interactions of hydrogen with catalytic metal surfaces and the mass transfer processes involved in heterogeneous catalysis are important for catalyst development. Although a range of technologies exists to study catalytic surfaces, much of it relies on high-vacuum conditions that preclude in-situ research. In contrast, atomic force microscopy (AFM) provides an opportunity for direct observation of surfaces at or near actual reaction conditions. Tapping-mode AFM was explored here since it expands AFM beyond the usual topographic information towards speciation and other more subtle surface information. The work presented here describes using phase-angle information from tapping-mode AFM to follow the interactions of hydrogen with palladium, polycarbonate, and iron. Real-time AFM phase-angle information allowed for the observation of multi-phase mass transfer to and from the surface of palladium at atmospheric pressure and room temperature without the need for complex sample preparation. The AFM observations are quantitatively benchmarked against and confirm mass transfer predictions based on bulk hydrogen diffusion data. Additionally, they support recent studies that demonstrate the existence of multiple hydrogen states during interactions with palladium surfaces.
... [23][24][25][26] Pd-based systems have previously been studied through both experimental techniques such as low-energy electron diffraction (LEED), [27][28][29] electron energy loss spectroscopy (EELS), [30,31] thermal desorption spectroscopy (TDS), [32] X-ray diffraction (XRD), [33] X-ray photoelectron spectroscopy (XPS); [34] and theoretical approaches such as semiempirical approaches like embedded-cluster methods [35] or effective-medium models [36] and ab initio methods based on density functional theory (DFT). [37,38] These reports, among others, have identified the preferred adsorption/absorption -sites, -geometries, and -energies, surface diffusion, and electronic structures of H chemisorption on Pd(111), Pd(100), and Pd(110) surfaces. More interestingly, experimentalists observed an anomalous generation of excess heat in the Pd-catalyzed electrochemical hydrogen evolution reaction (HER), which later became known as the highly debated "cold fusion". ...
The structure–activity relationship is a cornerstone topic in catalysis, which lays the foundation for the design and functionalization of catalytic materials. Of particular interest is the catalysis of the hydrogen evolution reaction (HER) by palladium (Pd), which is envisioned to play a major role in realizing a hydrogen‐based economy. Interestingly, experimentalists observed excess heat generation in such systems, which became known as the debated “cold fusion” phenomenon. Despite the considerable attention on this report, more fundamental knowledge, such as the impact of the formation of bulk Pd hydrides on the nature of active sites and the HER activity, remains largely unexplored. In this work, classical electrochemical experiments performed on model Pd(hkl) surfaces, “noise” electrochemical scanning tunneling microscopy (n‐EC‐STM), and density functional theory are combined to elucidate the nature of active sites for the HER. Results reveal an activity trend following Pd(111) > Pd(110) > Pd(100) and that the formation of subsurface hydride layers causes morphological changes and strain, which affect the HER activity and the nature of active sites. These findings provide significant insights into the role of subsurface hydride formation on the structure–activity relations toward the design of efficient Pd‐based nanocatalysts for the HER.
... This limits the technique's versatility in situations when the target material contains both 1 H and 2 D, such as in the study of hydrogen isotope exchange processes in surface layers. The latter are of profound interest in the context of plasma-surface interactions occurring in thermonuclear fusion devices [6,7] or for the elucidation of hydrogen transportation mechanisms across metal surfaces related to hydrogenation catalysis, hydrogen storage and purification [8][9][10]. Depth profiling of deuterium ( 2 D) is typically performed with 2 D( 3 He,p) 4 He NRA using 3 He ion beams of much lower energy [11], which is in turn selective for only 2 D. Synchronous analysis of 1 H and 2 D could hitherto only be achieved by elastic recoil detection (ERD) [12][13][14]. ...
The cross section for the combined ²D(¹⁵N,p)¹⁶N and ²D(¹⁵N,nγ)¹⁶O nuclear reactions of deuterium (²D) with incident ¹⁵N ions is evaluated and found to increase by two orders of magnitude between 3.3 and 7.0 MeV. Detecting the γ-rays at 6.1 and 7.1 MeV from these reactions allows quantifying the depth-integrated ²D content in materials in parallel with nanometer-resolved quantitative hydrogen (¹H) depth profiling in surface layers through resonant ¹H(¹⁵N,αγ)¹²C nuclear reaction analysis (NRA) using ¹⁵N ion beams. The information depth and sensitivity of ¹⁵N-²D NRA is estimated through energy loss simulations of the ¹⁵N projectiles in the analyzed material. Good agreement of the integral ²D quantitation through ²D(¹⁵N,p)¹⁶N and ²D(¹⁵N,nγ)¹⁶O NRA with ²D depth profiles obtained independently via ²D(³He,p)⁴He NRA is demonstrated at the example of ²D plasma-exposed tungsten samples. The simultaneous quantitation of ²D and depth-resolved ¹H profiling with a single ¹⁵N ion beam promises application potential to investigate hydrogen-isotope exchange and diffusion processes in surface and thin buried layers.
... The possibility to lower the activation energy by a concerted motion of atoms has been demonstrated recently for hydrogen absorption at the Pd (110) surface. [33] Unravelling the atomistic details of TiH2-mediated H2 recombination is a challenge that may be tackled in a future investigation. In a purely SL process, the transformed fraction ( ) should be linear until completion. ...
We study the kinetics of hydrogen sorption in Mg‐Ti‐H nanoparticles prepared by gas phase condensation of mixed Mg‐Ti vapors under a H2‐containing atmosphere. Four samples with different Ti contents from 14 to 63 at.% Ti are examined in the 100–150 °C range. The hydrogen absorption kinetics coupled with the formation of MgH2 can be described by a nucleation and growth model. The activation energy is in the range 43-52 kJ/mol and the rate constant (at 150 °C) increases from 27·10-3 s⁻¹ to 92·10-3 s⁻¹ with increasing Ti content. Hydrogen desorption is well modeled by a sequence of surface‐limited and contracting‐volume kinetics, except at the highest Ti content where nucleation and growth is observed. The activation energy of surface‐limited kinetics is ∼32kJ /mol. The rate constant (at 150 °C) increases from 0.5·10-3 s⁻¹ to 1.2·10-3 s⁻¹ with the Ti content. These results open an unexplored kinetic window for Mg‐based reversible hydrogen storage close to ambient temperature.
... The experiments were performed under ultrahigh vacuum conditions. The Pd(210) surface was cleaned by repeated cycles of Ar ion sputtering, annealing at 1000 K, annealing at 750 K under an O 2 pressure of 5 × 10 −5 Pa followed by cooling in H 2 of 5 × 10 −5 Pa, and final flashing at 600 K until a clear low-energy electron diffraction pattern was observed [25]. After exposing the clean Pd(210) surface to H 2 (D 2 ) at 50 K, TPD data were taken with a ramp rate of 2.8 K s −1 . ...
The rotational state and ortho-para conversion of H2 on a Pd(210) surface is investigated with rotational-state-selective temperature-programmed desorption (RS-TPD) and theoretical calculations. The isotope dependence of TPD shows a higher desorption energy for D2 than that for H2, which is ascribed to the rotational and zero-point vibrational energies. The RS-TPD data show that the desorption energy of H2(J=1) (J: rotational quantum number) is higher than that of H2(J=0). This is due to the orientationally anisotropic potential confining the adsorbed H2, which is in agreement with theoretical calculations. Furthermore, the H2 desorption intensity ratio in J=1 and J=0 indicates fast ortho-para conversion in the adsorption state, which we estimate to be of the order of 1 s.
... 61,62 On clean Pd surfaces multiple TDS features have been observed depending on the hydrogen exposure and temperature. Several distinct features have been identied as originating from the hydrogen chemisorbed at Pd surfaces and most of them desorb above 300 K. 62 The present samples have been pre-treated under H 2 followed by outgas under vacuum at 300 K; consequently, the chemisorbed hydrogen atoms at the surface were not removed. Therefore, all TDS experiments probe only the hydrogen desorption from the bulk Pd hydride powder and "bulk-like" core of 1 nm hydrogen loaded Pd clusters neglecting the surface contribution. ...
We report here the unprecedented modification of the hydrogen absorption/desorption properties of Pd clusters with 1 nm average size relative to bulk and nanoparticles down to 2-3 nm. These metal clusters have been synthesized by a facile double solvent impregnation method. They contain on average 33 atoms and are confined/stabilized into a Metal-Organic-Framework with different metal loadings (5-20 wt%). Such ultra-small nanoparticles are crystalline with the archetypical fcc structure of bulk metal, as confirmed by both HR-TEM and in situ EXAFS. This is the first time, to the best of our knowledge, that 1 nm Pd clusters are effectively confined into a MOF for high metal loadings. Hydrogen absorption/desorption properties of 1 nm Pd clusters have been characterized by both laboratory and synchrotron facilities. At ambient conditions, 1 nm Pd clusters absorb hydrogen forming solid solutions instead of hydride phase, as usually encountered for bulk and Pd nanoparticles down to 2-3 nm. This can be understood by a decrease of the critical temperature of the two-phase region in the Pd-H phase diagram below room temperature. Moreover, the activation energy of hydrogen desorption from Pd clusters strongly decreases relative to bulk Pd. This suggests a change in the rate limiting step from surface recombination or β → α phase transformation usually encountered in bulk Pd to hydrogen diffusion into α and β phases in 1 nm clusters.
... Accordingly, the magnetic behaviors of 10-and 20-nm-thick Co-Pd alloy films were monitored with variations in preservation days. Pd materials are an efficient catalyst for hydrogen gas dissociation and are suitable for the solid storage of hydrogen [26][27][28][29][30][31][32][33]. Many Pdrelated magnetic alloys, multi layers, and nanostructures have been reported to exert a pronounced hydrogenation effect on the magnetic behavior [17,19,21,22,[34][35][36][37]. ...
In this study, the self-assembly of surface nanoclusters on 10-20-nm-thick Co50Pd50 (Co-Pd) alloy thin films deposited on the Al2O3(0001) substrate was systematically investigated. The time-dependent evolution of the nanocluster size and magnetic properties was monitored using an atomic force microscope (AFM) and the magneto-optical Kerr effect. When the Co-Pd alloy films were stored in an ambient environment, small nanodots gradually gathered to form large nanoclusters. Approximately 30 days after growth, a nanocluster array formed with an average lateral size of 100 ± 20 nm and average height of 10 ± 3 nm. After 100 days, the average lateral size and average height had increased to 140 ± 20 and 25 ± 5 nm, respectively. The AFM phase image exhibited a structured contrast on the nanocluster surface, indicating the nonuniform stiffness distribution of the nanoclusters. A microscopic Auger spectroscopy measurement suggested that in contrast to the Pd-rich signal in the flat area, the nanoclusters were cobalt- and oxygen-rich areas. Cross-sectional investigation through transmission electron microscopy coupled with energy dispersive spectroscopy showed that the nanoclusters were mostly composed of Co oxide. A uniform Pd-rich underlayer had ben maintained underneath the self-assembled Co-oxide nanoclusters. With the formation of a Co-oxide nanocluster array and Pd-rich underlayer, the magnetic easy axis of the Co-Pd film gradually altered its direction from the pristine perpendicular to in-plane direction. Because of the change in the magnetic easy axis, the hydrogenation-induced spin-reorientation transition was suppressed with the evolution of the surface Co-oxide nanoclusters.
... Recent experimental and theoretical works on the absorption of H in a reconstructed surface of Pd(110) produced interesting results. 11,12) These works revealed that the facile absorption of H in the subsurface occurs through the assistance of other H atoms. We extended this investigation by focusing on the adsorption of CO because of its importance in realizing hydrogen storage based on Pd surfaces. ...
The adsorption of CO molecule on (110)-(1 × 2) missing-row reconstructed surfaces of Pd, Au, and Pd3Au with 87.5% Au segregation in the topmost layers were investigated by first-principles calculations based on density functional theory. On a pure Pd surface, CO is preferencially adsorbed on pseudo-fcc hollow site near the ridge sites. On pure Au and alloy surfaces, generally the ridge sites are the most stable for adsorption. The charge accumulation between the C atom and the surface atoms explains the calculated adsorption energies, with higher charge accumulation yielding a higher binding energy. The interaction of CO with the surfaces and the adsorption energies were also analyzed using the crystal orbital overlap population (COOP) technique and by considering the d-band model. CO vibrational modes were experimentally investigated for comparison. The vibrational frequencies are strongly influenced by the number of nearest neighbors of surface atoms as well as the type of atoms that constitute the nearest neighbors. In general, fewer nearest neighbors for the CO adsorption site results in higher vibrational frequencies. For adsorption sites with a certain number of nearest neighbors, the presence of Pd slightly decreases the vibrational frequencies. This observation is attributed to the broadening and the population of the derived 2Π+ orbital of CO upon its interaction with the surface.
... 33) On the other hand, E g in the present experiment is much larger than the value of ∼0.1 eV obtained in previous experiments in Pd-H systems. 34,35) The surface condition of the sample would play an important role in the origin of this difference. Further measurements by the VW under various temperatures and surface conditions will give accurate information on E g . ...
A vibrating wire (VW) method was applied to investigate the hydrogen absorption and desorption properties of palladium. At room temperature, a considerable shift in resonance frequency was successfully observed in VW spectra under H2 gas exposure. The shift is reversible in the initial stage of the exposure and is attributed to changes in the density and Young's modulus of the VW sensor. Irreversibility of the shift because of embrittlement is detected after a sufficient exposure time. H absorption is slowed down enormously at T = 200K owing to suppression of the thermal activation process.
... The quick response and the delayed part in recovery might be due to the hydrogen desorption from the b-state and a-state, respectively, because of the different bonding strength in b-PdH and a-PdH. 30 In Fig. 4(a), MOKE hysteresis loops corresponding to the different H 2 pressures are plotted. The reversibility of the effect of hydrogenation on the magnetic properties can be clearly observed. ...
The effect of hydrogenation on a 14 nm Co14Pd86/Al2O3(0001) thin film was investigated on the basis of the magnetooptical Kerr effect. After exposure to H2 gas, the squareness of the hysteresis loop showed a large transition from approximately 10% to 100% and the saturation Kerr signal was reduced to nearly 30% of the pristine value. The reversibility of the transition was verified and the response time was within 2–3 s. These observations indicate that the hydride formation transformed the short-range coupled and disordered magnetic state of the Co14Pd86
film to a long-range-ordered ferromagnetic state and induced appreciable decrease in the magnetic moment. The enhanced long-range-ordering and the reduction of the magnetic moment were attributed to the change of electronic structure in Co14Pd86 with hydrogen uptake.
Hydrogen permeation through pure and oxidised bulk chromium membranes was measured by the classical gas technique to get insight into oxide as a hydrogen permeation barrier (HPB). An additional palladium-coated reference chromium membrane was tested to avoid the influence of native Cr oxide. Key parameters for Cr permeability: P0 = 3.23 × 10⁻⁷ mol H2/s/m/Pa0.5 and Ea = 0.68 eV and Cr diffusivity D0 = 9.0 × 10⁻⁵ m²/s and Ea = 0.59 eV. In the sample preparation stage, a thin ∼2 nm thick oxide was formed. Additional oxidation in pure oxygen at 400 °C increased the thickness from 20 to 50 nm. At this temperature, its efficiency as HPB was evaluated by comparing permeation rates to the reference chromium membrane. The highest permeation reduction factor of ∼3900 corresponded to only a ∼28 nm thick Cr oxide layer. Surface morphology and oxide thickness were investigated by SEM, while the thickness and type of chromium oxide by XPS.
The fabrication of a hydrogen isotope enrichment system is essential for the development of industrial, medical, life science, and nuclear fusion fields, and therefore, efficient enrichment techniques with a high separation factor and economic feasibility are still being explored. Herein, we report a hydrogen/deuterium (H/D) separation ability with polymer electrolyte membrane electrochemical hydrogen pumping (PEM-ECHP) using a heterogeneous electrode consisting of palladium and graphene layers (PdGr). By mass spectroscopic analysis, we demonstrate significant bias voltage dependence of the H/D separation factor with a maximum of ∼25 at 0.15 V and room temperature, which is superior to those of conventional separation methods. Theoretical analysis demonstrated that the observed high H/D factor stems from tunneling of hydrogen isotopes through atomically thick graphene during the electrochemical reaction and that the bias dependence of H/D results from a transition from the quantum tunneling regime to the classical overbarrier regime for hydrogen isotopes transfer through the graphene. These findings will help us understand the origin of the isotope separation ability of graphene discussed so far and contribute to developing an economical hydrogen isotope enrichment system using two-dimensional materials.
Exploiting the high surface-area-to-volume ratio of nanomaterials to store energy in the form of electrochemical alloys is an exceptionally promising route for achieving high-rate energy storage and delivery. Nanoscale palladium hydride is an excellent model system for understanding how nanoscale-specific properties affect the absorption and desorption of energy carrying equivalents. Hydrogen absorption and desorption in shape-controlled Pd nanostructures does not occur uniformly across the entire nanoparticle surface. Instead, hydrogen absorption and desorption proceed selectively through high-activity sites at the corners and edges. Such a mechanism hinders the hydrogen absorption rates and greatly reduces the benefit of nanoscaling the dimensions of the palladium. To solve this, we modify the surface of palladium with an ultrathin platinum shell. This modification nearly removes the barrier for hydrogen absorption (89 kJ/mol without a Pt shell and 1.8 kJ/mol with a Pt shell) and enables diffusion through the entire Pd/Pt surface.
Ortho-para conversion of molecularly chemisorbed H2 on a Pd(210) surface at a surface temperature of 50 K is reported. A combination of a pulsed molecular beam, photostimulated desorption, and resonance-enhanced multiphoton ionization techniques was used for probing the change in the rotational states of molecularly chemisorbed H2 on the surface. Our result shows that fast ortho-para conversion of chemisorbed H2 occurs. The conversion time was experimentally determined to be about 2 s, which is in good agreement with a previous theoretical calculation. This agreement supports that the ortho-para conversion mechanism of the molecularly chemisorbed H2 on Pd(210) is a two-step process based on the hyperfine-Coulomb excitation.
The data collected in the present work extend the measured phase diagram for palladium hydride and palladium deuteride to a region that has been sparsely reported in open literature. Absorption isotherms were measured using a 2.5-g bed of palladium powder at temperatures between 130 and 393 K and pressures less than 1.3 × 10⁵ Pa. Such low-pressure and low-temperature measurements are useful for characterizing palladium beds used for tritium pumping and storage. For tritium storage, pressures are kept below a few millibars for safety reasons. Low temperatures increase the tritium storage capacity of palladium.
The measured absorption isotherms show the well-documented, two-phase behavior for this system: two solubility regions and a mixed, hydride-forming region. The isotherms show that an increased quantity of hydride is formed at lower temperatures, as marked by an increase in the hydride-forming region. This region exceeds hydrogen-to-metal ratios of 0.75 for T ≤ 273 K. Equilibrium pressures in the mixed region decrease with decreasing temperatures until a critical temperature is reached for each isotope. Below these critical temperatures, the rate of pressure decrease with decreasing temperature is significantly reduced. This change in trend suggests hydrogen isotopes are adsorbed onto the palladium surface, rather than forming a hydride. Using the equilibrium pressures recorded at temperatures between 236 and 393 K for protium and between 211 and 354 K for deuterium, the van’t Hoff constants were calculated to be A = −36 ± 1 kJ/mol and B = 88 ± 3 J/K for protium and A = −32 ± 2 kJ/mol and B = 88 ± 9 J/K for deuterium. These constants agree favorably with literature in the range where the temperatures of the measured isotherms overlap.
We have developed a novel nonevaporable getter named oxygen-free Pd/Ti. After activation at 133℃, oxygen-free Pd/Ti evacuates H2 and CO. Its pumping speeds do not decrease even after repeated cycles of activation and exposure to the air. Surface analysis by synchrotron radiation X-ray photoelectron spectroscopy showed that the carbon contamination decreased to an extent on heating in UHV, but decreased considerably on heating under an O2 pressure. The partial pressures of H2, CO, H2O, and CH4 in an oxygen-free-Pd/Ti coated chamber reduced to some degree after baking in UHV, but reduced remarkably after baking under an O2 pressure (O2 baking). Catalytic chemical reactions which remove carbon and hydrogen adsorbed on Pd surfaces appear to be responsible for the reduction of the partial pressures of H2, CO, H2O, and CH4. The pumping speeds of the oxygen-free Pd/Ti coated chamber for H2 and CO improved remarkably after O2 baking.
Fullsize Image
Applying a density functional approach to slab models of planar, (111), and rough, (110), Pd surfaces, we determined the isomerization free energy barriers of 1-butene to be significantly lower than the hydrogenation barriers. Microkinetic modeling allows one to mirror the kinetic experiments on conversions of 1-butene at the corresponding single-crystal surfaces in a qualitative fashion. Despite the inherent limitations of such kinetic modeling, theoretical predictions are fully supported by experimental data using Pd model catalysts: i.e., Pd(111) and Pd(110) surfaces. The isomerization mechanism was calculated to proceed via an initial dehydrogenation of 1-butene to 1-buten-3-yl as an intermediate—in contrast to the commonly proposed 2-butyl intermediate, associated with the Horiuti–Polanyi mechanism. Our modeling results rule out the original assumption that isomerization has to start with a hydrogenation step to rationalize the dependence of isomerization on hydrogen. However, this hydrogen dependence may arise in the second step, after an initial dehydrogenation, as suggested by the experimental data under hydrogen-deficient conditions.
Hydrogenation has recently been proposed for use in modulating the magnetic properties of Pd-rich ferromagnetic (FM) alloy films. This study successfully fabricated Pd-rich Mn/MnPd/Fe antiferromagnetic/FM films with an established exchange bias field at RT, and determined that hydrogenation can enhance the exchange bias coupling in such films. Specifically, magnetic hysteresis loops revealed that the magnetic state of the Pd-rich Mn/MnPd/Fe films was gradually changed from an unbiased state to an exchange-biased state upon exposure to a hydrogen environment; moreover, hydrogenation engendered a larger magnitude of the exchange bias field than conventional field cooling. This phenomenon can be attributed to the enhanced long-range antiferromagnetic ordering of the Pd-rich MnPd film upon uptake of hydrogen atoms.
Dynamic observation of hydrogen on catalytic metal surfaces is a challenging aspect of studying liquid-phase heterogeneous catalysis. Current methods suffer from one or more of the following limitations: the requirement to observe the surface in high vacuum, the inability to provide nanometer-level spatial resolution, the inability to deal with opaque catalysts and/or liquid immersion phase, the lack of real-time scanning of the surface area, and the inability to assess pronounced topographies or mixed materials. Atomic force microscopy (AFM) phase-shift imaging remedies these issues and provides an opportunity for dynamic direct observation of catalyst surfaces at or near actual reaction conditions immersed in liquid.
Hydrogen was delivered to a palladium surface immersed in water by diffusion through a support film of dense polycarbonate. The palladium surface was continuously probed by tapping-mode AFM. The theoretically predicted time-dependent appearance of hydrogen on the water-covered palladium surface matched the experimental observation reasonably well. The technique demonstrated here is unique in that the appearance of hydrogen is dynamically detected in real time on a catalyst surface immersed in water with nanometer-scale spatial resolution. The results presented here supply a new level of information for heterogeneous catalysis that is not available with existing techniques. This work opens new avenues in the study of heterogeneous catalysis, a field with tremendous practical importance and serious analytical challenges.
A newly developed high-resolution elastic recoil detection analysis (HERDA) system installed at the 1 MV Tandetron in UTTAC at the University of Tsukuba is introduced. The effective solid angle of detector, energy resolution and detection limit for hydrogen are, for the first time, determined quantitatively by the measurements on an a-C:H (and D) film deposited on a Si substrate. In the case of a 500 keV ¹⁶O⁺ as the incident beam, an energy resolution of ∼0.45 keV and a detection limit of ∼3.8 × 10²⁰ atoms/cm³ (∼0.18 at.%) with a data acquisition time of ∼310 s are derived.
Because of the Pd-catalyzed hydrogen dissociation and absorption, magnetic Pd-alloys provide a model
system for the investigation of the critical hydrogenation effect on magnetism. In this study, Co50Pd50
(CoPd) alloy thin films were fabricated by e-beam-heated co-evaporation on Al2O3(0001) substrates.
These films exhibited a thickness-dependent spin reorientation transition (SRT) from perpendicular
direction to in-plane direction with increase of thickness. For 10-30 nm-thick CoPd alloy films with
perpendicular magnetic anisotropy (PMA), hydrogenation triggered a SRT to an in-plane anisotropy. The
reversibility of SRT was demonstrated by cyclicly changing the hydrogen gas pressure. Furthermore,
hydrogenation-induced SRT randomized the magnetic domain orientation. The surface morphology of
the CoPd thin films was composed of nanoclusters, which may play a crucial role in hydrogen dissociation
and affect PMA. In comparison with a bare CoPd film, a stronger PMA and a less pronounced
hydrogenation-induced SRT were observed in a Pd-capped CoPd film. These observations suggest that
the hydrogen content in CoPd alloy films can drastically and reversibly modify PMA, inferring the
possible hydrogenation-induced charge transfer and modulation of electronic structure in CoPd.
A microcalorimetric method has been combined with a potentiostatic method to measure simultaneously the rate of heat evolution and the electrical current in a powdered sample of palladium during thermokinetic oscillations accompanying the sorption of deuterium in the metal. Deterministic chaos has been confirmed in the temporal variations in current (of ca. 1-4 mA) on the onset of both the sorption and the desorption of deuterium from Pd. It has been found that the first derivative of the current in time, dI/dt, turns out to be correlated precisely with the periodicity of thermokinetic oscillations. The dI/dt curves consist of regularly timed outbursts of aperiodic, high frequency (HF) fluctuations, interlinked by calm periods. The calm periods correlate with the descending slopes of thermokinetic oscillations (i.e., with decrease in the rate of heat evolution) and their lengths depend on the frequency of thermokinetic oscillations. In turn, the outbursts of aperiodic HF fluctuations in the dI/dt derivatives correlate with the ascending slopes of thermokinetic oscillations (i.e., with increasing rate of heat production), but their length is practically constant, irrespective of the thermokinetic frequency. We propose a periodic mechanism of sorption including a collective action of adsorbed deuterium taking place on the Pd surface. The periodicity of this mechanism arises from the temporal separation of its two sub-processes. The sub-process (1) involves only the adsorption of molecular D2 on the Pd surface and proceeds with little heat evolution until a critical coverage of D2 is achieved. The sub-process (2) initiates the dissociation of the adsorbed D2 and the penetration of the dissociated atomic D species into the Pd lattice. It is the more energetic of the two, but it only begins after a threshold coverage of D2 on the Pd surface has been achieved. We suggest that these sub-processes occurring alternatingly may provide a kernel for the oscillatory behavior observed in Pd/H(D) systems.
Catalytic hydrogen evolution plays a significant role in hydrogen production and utilization. The combinative desorption of hydrogen (Tafel step, i.e., 2 H⁎→H2) from metal catalysts has been extensively reported as the rate-determining step. However, a full atomic-level understanding on how the H-Metal binding strength affects on the elementary Tafel steps is still lacking. In the current study, H2 evolution over Pd catalysts was investigated by combining theoretical and experimental techniques. Density functional theory calculations revealed that H2 evolution was governed by either the combination barriers of 2 H⁎ or the desorption barriers of molecular H2 from the surface of the palladium catalyst, which was strongly dependent on the size of Pd particles: the rate-limiting step of H2 evolution for large nanoparticles (NPs) is diffusive combination of H⁎ across the metal surface, while both 2 H⁎ combination and H2 desorption are difficult for subnanometer-sized Pd clusters. By tuning the combined effect of H adatom combination and H2 desorption, a highly performance Pd catalyst for hydrogen evolution both for temperature-programmed palladium hydride decomposition and catalytic dehydrogenation of formate was designed and synthesized. TiO2-supported Pd NPs that were 2 nm in size exhibited excellent activity for formate dehydrogenation with an TOF value that was as high as 2184 h⁻¹ at 298 K.
Among ion-beam-based analytical methods, the direct observation of nuclear reactions induced by highly energetic (MeV regime) charged particles is dedicated to the quantitative determination of volume distributions of light elements from Z = 1 (H) to Z = 41 (Ga) in the near surface region of solids. In most cases, discrimination between their isotopes is enabled up to 37Cl. The incident ions are generally protons, deuterons, helium-3, or helium-4 ions. Nuclear reactions induced by heavier ions are sometimes also used, mostly for hydrogen depth profiling. All these reactions are characterized by the prompt emission of charged particles (protons or helium-4 ions) and/or γ-rays. Nuclear reaction analysis (NRA), performed in ion millibeam or microbeam modes, is an efficient complement to charged particle-induced X-ray emission, Rutherford backscattering spectrometry, and elastic recoil detection methods. Its applications are all intended for either absolute quantification or tracing experiments. They cover a broad panorama from life sciences to cultural heritage artifacts, including metallurgy, Earth sciences, nanotechnology, and material science.
Hydrogen is involved in a variety of chemical processes on surfaces. While hydrogen exhibits vibrational and rotational dynamics in its adsorption state, it in some cases undergoes diffusion into the substrate as well as on the surface, and participates in chemical reactions. Furthermore, hydrogen exchanges an electron with surfaces having a significant effect on the surface electronic structure. In this personal account, we review our recent studies on surface nuclear dynamics of hydrogen, hydrogen transport across surfaces, catalytic hydrogenation/isotope exchange reactions, and charge transfer between the surface and hydrogen by using a depth-resolved technique of nuclear reaction analysis and a quantum-state-selective detection of resonance enhanced multiphoton ionization in combination with surface science techniques. As a future prospect, we refer to ultraslow μ spin rotation spectroscopy for a direct probe of the hydrogen charge state at surfaces.
This chapter surveys the interaction of hydrogen with solid (metallic) surfaces. After some introductory remarks concerning the chemistry and physics of hydrogen and an overview of the terminology of H-surface interaction, a more detailed description of processes such as associative and dissociative adsorption including kinetics and energetics, surface diffusion, hydrogen phase formation, and two-dimensional phase transitions is provided and illustrated by means of experimental examples. In addition, the issues of H-induced surface reconstruction, the possible formation of subsurface hydrogen, surface hydrides, and absorbed hydrogen are touched upon. Then follow short excurses on hydrogen surface vibrations and some considerations regarding the electronic interaction between the H2 molecule and a metallic surface including UV photoemission and work function effects. The chapter ends with a brief summary of current quantum chemical theories that have proven useful to appropriately describe the H-surface interaction.
Creating a vacuum is to pump gas molecules in a vacuum chamber. Since outgassing properties of chamber walls and vacuum parts are of critical importance for vacuum technology, knowledge on the interaction of molecules with solid surfaces is required. We review properties of typical molecules familiar to vacuum technology and their interaction with well-defined solid surfaces.
The Pd-catalyzed hydrogenation of C=C double bonds is one of the most important synthetic routes in organic chemistry. This catalytic surface reaction is known to require hydrogen in the interior of the Pd catalyst, but the mechanistic role of the Pd-dissolved H has remained debated controversially. To shed new light into this fundamental problem, we visualized the H distribution near a Pd single crystal surface charged with absorbed hydrogen during a typical catalytic conversion of butene (C4H8) to butane (C4H10), using H depth profiling via nuclear reaction analysis. This has revealed that the catalytic butene hydrogenation (1) occurs between 160 and 250 K on a H-saturated Pd surface, (2) is triggered by the emergence of Pd bulk-dissolved hydrogen onto this surface, but (3) does not necessarily require large stationary H concentrations in subsurface sites. Even deeply bulk-absorbed hydrogen proves to be reactive, suggesting that Pd-dissolved hydrogen chiefly acts by directly providing reactive H species to the surface after bulk diffusion rather than by indirectly activating surface H through modifying the surface electronic structure. The chemisorbed surface hydrogen is found to promote hydrogenation reactivity by weakening the butene-Pd interaction and by significantly reducing the decomposition of the olefin.
Nuclear reaction analysis (NRA) via the resonant (1)H((15)N,αγ)(12)C reaction is a highly effective method of depth profiling that quantitatively and non-destructively reveals the hydrogen density distribution at surfaces, at interfaces, and in the volume of solid materials with high depth resolution. The technique applies a (15)N ion beam of 6.385 MeV provided by an electrostatic accelerator and specifically detects the (1)H isotope in depths up to about 2 μm from the target surface. Surface H coverages are measured with a sensitivity in the order of ~10(13) cm(-2) (~1% of a typical atomic monolayer density) and H volume concentrations with a detection limit of ~10(18) cm(-3) (~100 at. ppm). The near-surface depth resolution is 2-5 nm for surface-normal (15)N ion incidence onto the target and can be enhanced to values below 1 nm for very flat targets by adopting a surface-grazing incidence geometry. The method is versatile and readily applied to any high vacuum compatible homogeneous material with a smooth surface (no pores). Electrically conductive targets usually tolerate the ion beam irradiation with negligible degradation. Hydrogen quantitation and correct depth analysis require knowledge of the elementary composition (besides hydrogen) and mass density of the target material. Especially in combination with ultra-high vacuum methods for in-situ target preparation and characterization, (1)H((15)N,αγ)(12)C NRA is ideally suited for hydrogen analysis at atomically controlled surfaces and nanostructured interfaces. We exemplarily demonstrate here the application of (15)N NRA at the MALT Tandem accelerator facility of the University of Tokyo to (1) quantitatively measure the surface coverage and the bulk concentration of hydrogen in the near-surface region of a H2 exposed Pd(110) single crystal, and (2) to determine the depth location and layer density of hydrogen near the interfaces of thin SiO2 films on Si(100).
The quantum states of adsorbed hydrogen atom on Pd(1 1 0) surface are investigated in this work. From the calculated potential energy surface (PES) of hydrogen atom on Pd(1 1 0), the wave functions and eigenenergies in the ground and few excited states of protium (H) and deuterium (D) are calculated. Localized wave functions of hydrogen atom exist on pseudo-threefold and long bridge sites of Pd(1 1 0). The short bridge site is a local minimum from the result of PES, however, quantum behavior of hydrogen revealed that its vibration would allow it to hop to other pseudo-threefold site (that crosses the short bridge site) than to stay on the short bridge site. Exchange of ordering of the wave functions between H and D is attributed to the difference in their masses. The calculated eigenenergies are found to be in fair agreement with experimental data based from the identified vibrations of hydrogen with component perpendicular to the surface. The activation barriers measured from the eigenenergies are in better agreement with experimental findings in comparison to the data gathered from PES.
The hydrogen absorption behavior of the Pd NPs has been investigated by the noble techniques. The Pd NPs with the clean surface have been fabricated by the gas evaporation method. The P-C isotherm of the hydrogen absorption of the Pd NPs has been obtained using the QCM without the exposure to the air. The P-C isotherms have shown the clear size dependent absorption behavior for the Pd NPs. Further absorption and desorption cycles decrease the solubility of H signicantly in the Pd NPs. [DOI: 10.1380/ejssnt.2015.343]
The underlying mechanism of H atom absorption in the Pd(1 1 0) (1 × 2) missing-row reconstructed surface is investigated by performing density functional theory based calculations. The stronger binding energy of H on ridge than on trough site of the missing-row surface is due to the more pronounced creation of derived bonding state as had been depicted from the electronic structure of the system. Hydrogen absorption takes place with the involvement of other incoming H atoms through an assisted absorption process that is facilitated by the repulsion between the incoming H and the absorbing H. The geometry of the missing-row surface enables the Pd atoms to accommodate the H atoms efficiently leading to H absorption as well as H2 dissociation.
To assess the possibility to control the desorption temperature of palladium-absorbed hydrogen (Habs) through surface structural manipulation, we investigated co-adsorption systems of H and CO on Habs-charged Pd(110) surfaces through temperature-programmed desorption (TPD), low energy electron diffraction (LEED), and H-depth profiling by nuclear reaction analysis (NRA). A CO coverage of 0.5 ML lifts the H-induced (1×2) pairing-row (PR) reconstruction on Habs pre-charged Pd(110), and, as on clean Pd(110), heating Pd(110) (1×1) holding 0.3 - 1.0 ML CO gives rise to a missing-row (MR) structure. Whereas Habs desorbs through surface defects of clean, PR-reconstructed Pd(110) at 160 K, CO co-adsorption onto Habs-loaded Pd(110) gives rise to three new high-temperature shifted Habs desorption modes at 200, 270, and 375 K that are assigned to different exit sites for resurfacing Habs atoms at regular terraces of the individual Pd(110) structures, i.e., the (1×2) PR, the bulk-terminated (1×1), and the (1×2) MR reconstruction, respectively. Our results thus manifest the ability to control the Habs desorption temperature through surface restructuring in well-defined CO coverage regimes. The long-speculated transfer of chemisorbed H into the Pd interior upon CO co-adsorption is furthermore confirmed directly by NRA, revealing also that all Habs diffuses into the Pd bulk at 200 K.
Step edges and kinks, abundant on multi-faceted nanoparticles, are catalytically active sites. Weakly-bound atomic H, at either topmost surface or subsurface sites, would be important for low-temperature hydrogenation in platinum-based catalysts. Here we report experimental results for such H atoms on Pt(111). Saturation-adsorbed atomic H from molecular H2 on the defect-free Pt(111) surface indeed gave only a single-peaked H2 desorption (β2) at 285 K. Instead, defected Pt(111) surfaces rendered triple peaks (β1 to β3) including a prominent feature (β1) at as low as 205 K in addition to another desorption (β3) at 360 K. This β1–H state was inhibited and created by pre- and post-adsorbed CO, respectively. We attribute the β1–H2 desorption to H atoms trapped at interstitial sites beneath surface defects on the basis of: (1) its desorption at a very low temperature in addition to two other peaks from terrace- and defect-adsorbed H; (2) its and total H uptakes by far larger than the surface defect density; (3) its desorption amount up to ~ 3.6 times that of the β3 desorption from defects; (4) its complete inhibition by a small pre-coverage of CO; and (5) the complete β3-to-β1 H conversion, while the β1–H state remaining intact, by postdosed CO. Our proposed mechanism is that the derelaxation (upward lifting) of the H- or CO-bound Pt lattice atoms at (step) defects, as a result of strong H–H and even stronger H–CO lateral repulsions under (near) saturation surface coverages, opens a low-barrier path for H diffusion into the subsurface.
This review introduces hydrogen depth profiling by nuclear reaction analysis (NRA) via the resonant 1H(15N,αγ)12C reaction as a versatile method for the highly depth-resolved observation of hydrogen (H) at solid surfaces and interfaces. The technique is quantitative, non-destructive, and readily applied to a large variety of materials. Its fundamentals, instrumental requirements, advantages and limitations are described in detail, and its main performance benchmarks in terms of depth resolution and sensitivity are compared to those of elastic recoil detection (ERD) as a competing method. The wide range of 1H(15N,αγ)12C NRA applications in research of hydrogen-related phenomena at surfaces and interfaces is reviewed.
Special emphasis is placed on the powerful combination of 1H(15N,αγ)12C NRA with surface science techniques of in-situ target preparation and characterization, as the NRA technique is ideally suited to investigate hydrogen interactions with atomically controlled surfaces and intact interfaces. In conjunction with thermal desorption spectroscopy, 15N NRA can assess the thermal stability of absorbed hydrogen species in different depth locations against diffusion and desorption. Hydrogen diffusion dynamics in the near-surface region, including transitions of hydrogen between the surface and the bulk, and between shallow interfaces of nanostructured thin layer stacks can directly be visualized. As a unique feature of 15N NRA, the analysis of Doppler-broadened resonance excitation curves allows for the direct measurement of the zero-point vibrational energy of hydrogen atoms adsorbed on single crystal surfaces.
A novel H2 molecular adsorption state on metal surfaces has been detected by temperature-programmed desorption and electron energy loss spectroscopy experiments of the H2/Pd(210) system. The molecular nature of this state has been verified by isotope exchange experiments. This molecular state leads to a decrease of the surface work function while atomic hydrogen on Pd(210) causes an increase. Ab initio total-energy calculations have confirmed all experimental findings. Through these calculations the microscopic nature of this novel molecular adsorption state could be identified; it turns out that this state is stabilized by the presence of atomic hydrogen on the Pd(210) surface.
Substantial hydrogen absorption in nanometer-sized palladium crystallites on a thin film alumina support is observed at 90–350 K under low H2 pressure conditions (<2×10−3 Pa) by 1H(15N,αγ)12C nuclear reaction analysis. The enthalpy of H solution varies with the absorbed H content and with −(0.28±0.02) eV∕H exceeds that of bulk Pd at H∕Pd ratios below 0.20. By resolving absorbed H from surface-adsorbed H, this enhanced H stability is identified as an intrinsic volume property of the small crystallites rather than being associated with particular binding sites on or near their surfaces.
We employ classical molecular dynamics calculations based on density-functional theory molecule-surface interaction potentials to study H-coverage effects on H2 dissociative adsorption on Pd(100). In contrast with one of the basic assumptions of the widely used Langmuir model, we have found that a single isolated H-vacancy is enough to spontaneously dissociate low-energy H2 molecules on H-covered Pd(100). We also show that for a given initial coverage (e.g., Theta=1/2 ), the dissociative adsorption probability of low-energy H2 molecules can vary by a factor of five depending on the particular arrangement of the H adatoms on the surface.
The interaction of hydrogen with palladium surfaces represents one of the model systems for the study of the adsorption and absorption at metal surfaces. Theoretical gas-surface dynamics studies have usually concentrated on the adsorption dynamics on clean surfaces. Only recently it has become possible, based on advances in the electronic structure codes and improvements in the computer power, to address the much more complex problem of the adsorption dynamics on precovered surfaces. Here, I present ab initio molecular dynamics (AIMD) simulations based on periodic density functional theory (DFT) calculations of the adsorption of H(2) on hydrogen-precovered Pd(100) for a broad variety of different hydrogen coverage structures. The stability of the adsorbate structures and the adsorption dynamics are analyzed in detail. Calculated sticking probabilities are larger than expected for pure site-blocking consistent with experimental results. It turns out that the adsorption dynamics on the strongly corrugated surfaces depends sensitively on the dynamic response of the substrate atoms upon the impact of the impinging H(2) molecules. In addition, for some structures the adsorption probability was evaluated as a function of the kinetic energy. Adsorbate structures corresponding to the same coverage but with different arrangements of the adsorbed atoms can lead to a qualitatively different dependence of the adsorption probability on the kinetic energy changing also the order of the preferred structures, as far as the adsorption is concerned, as a function of the kinetic energy. This indicates that dynamical effects such as steering and dynamical trapping play an important role in the adsorption on these precovered substrates.
The energetic, kinetic and structural properties of hydrogen chemisorbed on a Pd(100) surface were studied by means of thermal desorption, work function and LEED measurements. Under the applied conditions no interference with bulk dissolution occurs and dissociative adsorption gives rise to a continuous increase of the work function by up to 0.20 eV. The dipole moment of the adsorbate complex is constant up to θ ≈ 0.9 and then increases until saturation at θ ≈ 1.35 (at 170 K) is reached. The formation of a second adsorbed state at high coverages manifests itself also by a low-temperature shoulder in the thermal desorption spectra and in the variation of the isosteric heat of adsorption, E ad , with coverage: E ad remains practically constant ( 24.5 kcal mole ) up to θ ≈ 0.9 and then decreases. The sticking coefficient is initially rather high ( s 0 ≈ 0.5) and varies with coverage in a way which can be described by a precursor-state model. The preexponential factor for desorption is about 10 −2 cm 2 atom −1 s −1 . Desorption follows second order kinetics only at very low coverages, at high θ it exhibits quasifirst order. This effect is attributed to the existence of lateral interactions between adsorbed hydrogen atoms which manifest themselves also in the appearance of a c(2 × 2) LEED pattern at low temperatures. The “extra” diffraction spots attain their maximum intensity at θ = 0.5, and a structural model is proposed whereafter in this phase the H atoms occupy next-nearest neighboring adsorption sites with local fourfold symmetry. Order-disorder transitions were followed by recording the intensity of the half-order spots as a function of temperature at various coverages. The resulting phase diagram exhibits a critical temperature T c = 260 K at θ = 0.5 and is slightly asymmetric with respect to this coverage. The data are analysed in terms of a lattice gas model and estimates for the pairwise interaction energies yield repulsion between nearest neighbors ( w 1 = 0.5 kcal mole ) and attraction between next-nearest neighbors ( w 2 = −0.3 kcal mole ). The additional operation of non-pair-wise interactions is made responsible for the asymmetric shape of the phase diagram. Whereas the adsorbed layer is obviously localized at T ⩽ 270 K , a detailed analysis of the adsorption entropy reveals that for T ⩾ 370 K a rather good description can be obtained with a model of delocalized two-dimensional translation.
Ab initio investigation based on density functional theory is performed
to determine the behavior of H atom diffusion in Pd(110) surface to the
first and second subsurface layers. Potential energy surface is
constructed to determine the local minima and activation barriers of H
pathways. Contribution of the relaxation of surface atoms in the binding
energies of H and activation barriers along the diffusion paths, as well
as the zero point energy corrections are also included in this work. The
binding energies of H in the second subsurface layer are lower compared
to its binding energies in the first subsurface layer and this is
attributed to the interaction of H with the surface atoms and the
differences in interlayer spacing of the surface layers. Comments on the
adsorbate induced Pd(110) (1× 2) missing/adding-row reconstruction
phenomenon is also given with reference to the observed results in this
work as H is absorbed from the surface to the first subsurface layer.
Periodic, self-consistent DFT-GGA(PW91) calculations are used to study the interaction of hydrogen with different facets of seventeen transition metals—the (100) and (111) facets of face-centered cubic (fcc) metals, the (0001) facet of hexagonal-close packed (hcp) metals, and the (100) and (110) facets of body-centered cubic (bcc) metals. Calculated geometries and binding energies for surface and subsurface hydrogen are reported and are, in general, in good agreement with both previous modeling studies and experimental data. There are significant differences between the binding on the close-packed and more open (100) facets of the same metal. Geometries of subsurface hydrogen on different facets of the same metal are generally similar; however, binding energies of hydrogen in the subsurface of the different facets studied showed significant variation. Formation of surface hydrogen is exothermic with respect to gas-phase H2 on all metals studied with the exception of Ag and Au. For each metal studied, hydrogen in its preferred subsurface state is always less stable than its preferred surface state. The magnitude of the activation energy for hydrogen diffusion from the surface layer into the first subsurface layer is dominated by the difference in the thermodynamic stability of these two states. Diffusion from the first subsurface layer to one layer further into the bulk does not generally have a large thermodynamic barrier but still has a moderate kinetic barrier. Despite the proximity to the metal surface, the activation energy for hydrogen diffusion from the first to the second subsurface layer is generally similar to experimentally-determined activation energies for bulk diffusion found in the literature. There are also some significant differences in the activation energy for hydrogen diffusion into the bulk through different facets of the same metal.
We calculate the corresponding two-dimensional (2D) potential energy surfaces (PESs) for one fixed lateral H2 center of mass position and one H2 orientation on Mg(0 0 0 1), Ti(0 0 0 1) Ni(1 1 1), Pd(1 1 1) and La(0 0 0 1) surfaces within the density functional theory. From the results, on the Ti, Ni, Pd and La surfaces, the energy barriers for H2 dissociative adsorption are either small or negligible. On the other hand, on the Mg surface, a high energy barrier exists. Furthermore, we can practically explain these differences among the surfaces by considering the differences in the valence electron configurations of the substrate atoms.
Sum frequency generation (SFG) vibrational spectroscopy was carried out in conjunction with thermal desorption spectroscopy, low-energy electron diffraction, and Auger electron spectroscopy to examine the coadsorption of CO and H2 on Pd(111). Sequential dosing as well as various CO/H2 mixtures was utilized to study intermolecular interactions between CO and H2. Preadsorbed CO effectively prevented the dissociative adsorption of hydrogen for CO coverages >=0.33 ML. While preadsorbed hydrogen was able to hinder CO adsorption at low temperature (100 K), hydrogen was replaced from the surface by CO at 150 K. When 1:1 mixtures of CO/H2 were used at 100 K, hydrogen selectively hindered CO adsorption on on-top sites, while above ~125 K no blocking of CO adsorption was observed. The observations are explained in terms of mutual site blocking, of a CO-H phase separation, and of a CO-assisted hydrogen dissolution in the Pd bulk. The temperature-dependent site blocking effect of hydrogen is attributed to the ability (inability) of surface hydrogen to diffuse into the Pd bulk above (below) ~125 K. Nonlinear optical SFG spectroscopy allowed us to study these effects not only in ultrahigh vacuum but also in a high-pressure environment. Using an SFG-compatible ultrahigh vacuum-high-pressure cell, spectra of 1:10 CO/H2 mixtures were acquired up to 55 mbar and 550 K, with simultaneous gas chromatographic and mass spectrometric gas phase analysis. Under reaction conditions, CO coverages >=0.5 ML were observed which strongly limit H2 adsorption and thus may be partly responsible for the low CO hydrogenation rate. The high-pressure and high-temperature SFG spectra also showed indications of a reversible surface roughening or a highly dynamic (not perfectly ordered) CO adsorbate phase. Implications of the observed adsorbate structures on catalytic CO hydrogenation on supported Pd nanoparticles are discussed.
The notion of “active sites” is fundamental to heterogeneous catalysis. However, the exact nature of the active sites, and hence the mechanism by which they act, are still largely a matter of speculation. In this study, we have presented a systematic quantum chemical molecular dynamics (QCMD) calculations for the interaction of hydrogen on different step and terrace sites of the Pd (332) surface. Finally the dissociative adsorption of hydrogen on step and terrace as well as the influence of surface hydrogen vacancy for the dissociative adsorption of hydrogen has been investigated through QCMD. This is a state-of-the-art method for calculating the interaction of atoms and molecules with metal surfaces. It is found that fully hydrogen covered (saturated) step sites can dissociate hydrogen moderately and that a monovacancy surface is suitable for significant dissociative adsorption of hydrogen. However in terrace site of the surface we have found that dissociation of hydrogen takes place only on Pd sites where the metal atom is not bound to any pre-adsorbed hydrogen atoms. Furthermore, from the molecular dynamics and electronic structure calculations, we identify a number of consequences for the interpretation and modeling of diffusion experiments demonstrating the coverage and directional dependence of atomic hydrogen diffusion on stepped palladium surface.
The adsorption of deuterium on Pd(110) has been studied by means of low‐energy electron diffraction, work function measurements (Δϕ), thermal desorption spectroscopy, nuclear reaction analysis (NRA), and Rutherford backscattering (RBS). These measurements show that, upon D2 adsorption at low temperature, (2×1) and (1×2) phases are formed in sequence. The RBS data show that the (2×1) structure is unreconstructed, while the (1×2) is reconstructed, with an entire monolayer (ML) of Pd atoms displaced laterally from their bulklike locations by ≳0.01 nm. The NRA data demonstrate that depending on the method of preparation, the (2×1) phase can be associated with a total coverage of 1 or 1.5 ML, while the (1×2) phase can be produced with total coverages of 1.5 or 2 ML. The additional 0.5 ML, associated with the high coverage form of each phase, is due to subsurface adsorbed species.
It is shown that desorption energies may be extracted from thermal desorption spectra without any assumption on the preexponential, ν, regardless of coverage. This allows a determination of ν, at least at low coverages. Experimental TD spectra of CO adsorbed at are evaluated to illustrate the applicability of the procedure.
The present article focusses on the chemosorptive and physisorptive behavior of hydrogen interacting with solid surfaces. Depending on the electronic structure of the solid hydrogen will either physically interact and form a weak van der Waals type of bond, or the H2 molecule will dissociate and the H atoms will form a chemical bond with the surface atoms. In some cases the interaction is not just restricted to the surface but may lead to absorption or even hydride formation. Also, reversible or irreversible changes of the structure of the solid may occur as the hydrogen is interacting. This leads to relaxation, reconstruction and facetting phenomena.After some introductory remarks recent results on H2 physisorption and H chemisorption phenomena will be presented and compiled whereby emphasis is put on metal single crystal work. Particular attention will be paid to the energetics and kinetics of the adsorption process, and to the structural properties of hydrogen adsorbed layers. In this context, order-disorder transitions within chemisorbed phases will be considered as well as structural phase transformations involving the underlying solid surfaces (H-induced surface reconstruction). Thereafter a section will be devoted to theoretical model calculations describing the mechanisms of the H chemisorption and the bonding.
The formation of ordered structures of hydrogen on Pd(311) at a substrate temperature of 110 K has been investigated with low-temperature He beams. We found, before the completion of a saturated c(1 × 1) phase, the formation of three low-coverage (2 × 1) phases (the notation refers to the centered rectangular unit cell of the substrate). The first (2 × 1)H structure is completed after an exposure of just 0.09 L, suggesting - with an initial sticking coefficient near unity - a coverage of 0.25 ML where hydrogen atoms adsorb along every second close-packed row. Upon additional H2 exposure, this phase evolves into the (2 × 1)2H phase, which reaches its optimum order after 0.20 L and shows a glide symmetry plane along [233]. Progressive filling of the close-packed rows leads to the appearance of a (2 × 1)3H phase and finally to the formation of the c(1 × 1). Diffraction spectra for this phase show only slight changes upon H2 exposure in the range 0.5–10 L. The existence of three low-coverage hydrogen overlayers for coverages < 1 ML suggests an interaction mechanism between hydrogen atoms which is very different to the one present on Pd(110). Possible structural models for the reported phases are discussed.
The structures of the (1×2) reconstructions of Ni(110) and Pd(110), induced by Had at T<200 K, were determined in a LEED I/V analysis. Characteristic features of their formation and stability-nonactivated reconstruction kinetics, creation of additional adsorption sites as energetic driving force, thermodynamically metastable-are explained on the basis of the resulting "row pairing" structure. Mechanistic arguments and experimental evidence for the assignment of "missing row" structure to the thermodynamically stable "streaked" (1×2) phase, induced by Had at T>200 K in an activated process, are given. © 1987 Elsevier Science Publishers B.V. (North-Holland Physics Publishing Division).
The adsorption of hydrogen on a Pd(110) surface was investigated using Video-LEED, work function (Δφ) and thermal desorption mass spectroscopy (TDS) techniques. At 120 K, dissociative adsorption occurs readily with a sticking probability near unity. The formation of a “true” ordered (2×1) hydrogen phase corresponding to θH=1.0 is followed by a (1×2) phase with intense LEED spots indicating a H-induced surface reconstruction. The work function of the clean surface increases continuously with H exposure, up to a saturation value of 325 mV. TDS reveals a total of four hydrogen binding states. Two high-temperature states β1 and β2 are produced by chemisorbed H, whereas the low-temperature states α1 and α2 only occur on the reconstructed surface and are tentatively attributed to H atoms that have moved to subsurface sites located below the topmost Pd atom layer. The energy splitting of the α states possibly indicates two different kinds of subsurface sites or two different desorption mechanisms; the α1 site is progressively populated as the surface reconstruction proceeds.
Palladium surfaces are well-suited model systems for hydrogen absorption and diffusion studies. In this paper we present - after reviewing previous results on Pd(110) - new He-diffraction and thermal desorption spectroscopy (TDS) results on Pd(311). At 120 K, dissociative adsorption occurs with an initial sticking coefficient near unity. With increasing exposure, the formation of (2×1)H, (2×1)2H, (2×1)3H and c(1×1) phases was observed, with coverages 0.25 ML, 0.50 ML, 0.75 ML and 1 ML, respectively. In all four phases, long-range order disappears at 170 K. These order-disorder transitions were found to be completely reversible for the (2×1)H and (2×1)2H phases upon heating to 220 K and subsequent cooling to 120 K. For the (2×1)3H phase, however, a very different behaviour was observed: heating to 220 K leads to the migration of 0.25 ML chemisorbed H-atoms into subsurface sites, and the (2×1)2H phase is recovered after cooling down to 120 K. The fact that no evidence for a substrate reconstruction has been found in any of the three (2×1) phases suggests that the mechanism involved in the population of subsurface sites on Pd(311) is very different from the one present on Pd(110), where it is driven by a pairing-row reconstruction.
Conclusive evidence is provided for the first time that through adsorbate-induced surface-structure changes, effective channels open for hydrogen to move from the surface to the interior of metals. In Pd(110), all available subsurface sites between the first and the second metal layers can be selectively populated by thermal activation from a specific type of chemisorption place, which is associated with a substrate reconstruction at high hydrogen surface coverage.
Scitation is the online home of leading journals and conference proceedings from AIP Publishing and AIP Member Societies
The energetic, kinetic and structural properties of hydrogen chemisorbed on a Pd(100) surface were studied by means of thermal desorption, work function and LEED measurements. Under the applied conditions no interference with bulk dissolution occurs and dissociative adsorption gives rise to a continuous increase of the work function by up to 0.20 eV. The dipole moment of the adsorbate complex is constant up to θ ≈ 0.9 and then increases until saturation at θ ≈ 1.35 (at 170 K) is reached. The formation of a second adsorbed state at high coverages manifests itself also by a low-temperature shoulder in the thermal desorption spectra and in the variation of the isosteric heat of adsorption, Ead, with coverage: Ead remains practically constant () up to θ ≈ 0.9 and then decreases. The sticking coefficient is initially rather high (s0 ≈ 0.5) and varies with coverage in a way which can be described by a precursor-state model. The preexponential factor for desorption is about 10−2 cm2 atom−1 s−1. Desorption follows second order kinetics only at very low coverages, at high θ it exhibits quasifirst order. This effect is attributed to the existence of lateral interactions between adsorbed hydrogen atoms which manifest themselves also in the appearance of a c(2 × 2) LEED pattern at low temperatures. The “extra” diffraction spots attain their maximum intensity at θ = 0.5, and a structural model is proposed whereafter in this phase the H atoms occupy next-nearest neighboring adsorption sites with local fourfold symmetry. Order-disorder transitions were followed by recording the intensity of the half-order spots as a function of temperature at various coverages. The resulting phase diagram exhibits a critical temperature Tc = 260 K at θ = 0.5 and is slightly asymmetric with respect to this coverage. The data are analysed in terms of a lattice gas model and estimates for the pairwise interaction energies yield repulsion between nearest neighbors () and attraction between next-nearest neighbors (). The additional operation of non-pair-wise interactions is made responsible for the asymmetric shape of the phase diagram. Whereas the adsorbed layer is obviously localized at T ⩽ 270 K, a detailed analysis of the adsorption entropy reveals that for T ⩾ 370 K a rather good description can be obtained with a model of delocalized two-dimensional translation.
The 15N hydrogen profiling method is discussed together with its use in a number of applications. It is shown that the width of the nuclear resonance used in this method is 0.4 keV (c.m.) which is less than half the published value. It is shown that many aspects of glass hydration are in agreement with the interdiffusion model of Doremus, that hydrogen surface contaminationscan be important in neutron “bottles”, and that there is a correlation between hydrogen content and the transition temperature, Tc, for Nb3 Ge superconductors.
In a recent study of the selective hydrogenation of butadiene to butenes, the reactivity of palladium (in the form of single crystals and of supported catalysts) was compared with that of platinum. The reactivity of single-crystal faces with (111) orientation is comparable to that of SiO{sub 2}-supported catalysts. The argument put forward to explain the comparatively lower activity of platinum was its lower hydrogen coverage under a given pressure, in the presence of the hydrocarbons, arising from a heat of adsorption smaller than that on palladium (respectively 67 and 85 kJ/mol). Concerning its poorer selectivity toward butenes, the reason invoked was a similar heat of adsorption for butadiene and butenes: the olefinic intermediate compound remains partly on the surface and the reaction proceeds to butane. A powerful kinetic method which is able to provide direct evidence for this different adsorption of dienes and butenes would be the competitive hydrogenation of these two molecules. However, to distinguish between the ethylenic hydrocarbon formed during the reaction and the molecules already present in the reactants, the experiments were performed according to the sequence: competitive hydrogenation of a butadiene-propene mixture followed by the hydrogenation of a 1-butene-propene mixture. This paper describes a study which confirms that the reason for the selectivity of palladium (after stabilization of the catalyst) can be attributed unambiguously to its capacity of adsorbing a diene more strongly than an alkene molecule.
The (110) surface of many transition metals undergoes reconstructions either on the clean substrate or induced by adsorbates. The reconstructions can be induced by a variety of strongly bound adsorbates, e.g., H, O, S, and alkali-metal atoms. In this work, H/Pd(110) is chosen as a prototype of such reconstructions. The pairing-row and missing-row reconstructions are studied in a wide range of coverages, θ=0.5–1.5ML, by using density-functional theory with the local-density approximation and the generalized gradient approximation. The driving force for the reconstructions is also analyzed in detail. The pairing-row reconstruction is driven essentially by the repulsion between the hydrogen atoms adsorbed in the same trough while the driving force for the missing-row reconstruction is the better adsorbate-substrate interaction on the reconstructed surface.
We present self-consistent density-functional calculations of the adsorption of atomic and molecular hydrogen on the (210) surface of palladium using a plane-wave basis set with optimized ultrasoft pseudopotentials. The layer relaxations of the (210) surface and the preferred adsorption sites for atomic hydrogen adsorption are determined. Furthermore, we show that on the rather open Pd(210) surface a molecular H2 adsorption state becomes stabilized by the presence of atomic hydrogen on the surface. This provides a consistent explanation of recent experiments. An analysis of the bonding situation in terms of the local density of states is also presented.
The conditions enabling the transition of surface-adsorbed hydrogen atoms into bulk metals are explored by comparing the response of chemisorbed H on Pd(100) and Ti(0001) single crystals to thermal activation in vacuum. Thermal desorption spectroscopy and 1H(15N,αγ)12C nuclear reaction analysis reveal that heating causes H2 desorption from Pd(100), whereas H atoms on Ti(0001) reversibly exchange between the surface and the Ti bulk with negligible desorption loss of H2. A general model is proposed in which the competition between desorption and bulk absorption of surface hydrogen is, on one hand, kinetically determined by the activation energy for associative H2 desorption relative to the effective energy barrier for H absorption. On the other hand, the thermodynamic possibility to dissolve the surface H atoms into the metal must be considered to consistently explain the opposite behavior of H on Pd(100) and Ti(0001). The first experimental estimation of the energy of surface adsorption, εs=−0.92 eV, for H/Ti(0001) is presented, in good agreement with theoretical calculations.
The full three-dimensional potential energy surface of H on Pd(111) has been calculated with periodic band-structure computations using the generalized gradient approximation of density-functional theory. The fcc hollow site was found to be most stable, followed by the hcp hollow site. Excellent agreement with experimental values of the adsorption energy and the vibrational frequencies was achieved. Subsurface occupation at low coverages and low temperatures is ruled out by our results, but there are no or very low barriers for hydrogen reaching the subsurface region from the molecular gas phase, thus direct absorption is feasible at high coverages. Different ordered structures of the adsorbed hydrogen were considered, and we found that two structures with √3×√3R30° symmetry are most stable at low temperatures, in agreement with experiment. Results for the adsorption energies and effective hydrogen-hydrogen interactions showed that only fcc hollow sites were occupied on the ordered structures, also in agreement with experiment.
The activated transition of chemisorbed hydrogen atoms into subsurface sites on Pd(311) has been investigated by means of He-atom scattering, high resolution electron energy loss spectroscopy (HREELS), thermal desorption spectroscopy (TDS) and work function measurements. At 120 K, hydrogen exposure leads to the formation of (2×1)H, (2×1)2H, (2×1)3H and c(1×1) 2H phases, with coverages of 0.25, 0.50, 0.75, and 1 monolayers (ML), respectively. The TDS data show three desorption states: α at ∼170 K, β1 at ∼285 K and β2 at ∼310 K. Chemisorbed H atoms forming the ordered layers desorb in the β2 state, whereas the β1 is originated by H atoms located at subsurface sites. The α state is originated by decomposition of layers of Pd hydride near the surface. In all four phases, long-range order disappears at ∼170 K. Heating to 220 K leads to the migration of 0.25 ML H atoms into subsurface sites only if the coverage of the disordered layer is greater than 0.5 ML. The HREELS data demonstrate that this behavior is caused by the occupation of different adsorption sites as a function of coverage: only fourfold coordinated sites are occupied in the (2×1)H and (2×1)2H phases, whereas threefold coordinated sites are also occupied for Θ>0.5 ML. A surprising result is that the HREELS peaks of the surface hydrogen vibrations still exhibit significant changes once all surface sites are occupied, and saturate only after saturation of the subsurface sites. This effect presumably results from strong repulsion between H atoms adsorbed on threefold coordinated sites and subsurface H atoms located in octahedral sites. © 1999 American Institute of Physics.
The adsorption and initial absorption of hydrogen (H-2) by Pd(1 0 0) at T = 100 - 160 K are investigated by thermal desorption spectroscopy (TDS) and nuclear reaction analysis, which enables a depth-resolved determination of the hydrogen concentration at and below the Pd surface. Under 3 x 10(8) mbar H-2 and T < 130 K hydrogen is first chemisorbed at the surface and the coverage saturates at 1.0 ML. A small fraction of further dosed hydrogen penetrates the surface and accumulates in a region of up to approximate to4 nm depth, contrasting previous assignments of a corresponding 'alpha (1)'-TDS peak at 180 K to exclusive occupation of first-layer subsurface sites. At T > 130 K H-migration from such a nearsurface hydride phase to the deeper bulk of the metal occurs and a solution of H in Pd is formed by means of facile diffusion. This bulk dissolved hydrogen emerges as a broad structure in TDS at T = 200-260 K exhibiting a tailing at high temperatures characteristic of diffusion-controlled desorption kinetics.
Surface behavior of thin Pd film in the course of palladium hydride (PdHx) formation under H2 pressure of 101 kPa at 298 K and its decomposition in Ar or O2 atmospheres was studied “in situ” by combined methods of a video technique and atomic force microscopy. The main effort was concentrated on the determination of the role of local defects on the surface in the process of PdHx formation and its catalytic reaction with oxygen. The formation of PdHx within a thin Pd film causes drastic local changes of the surface corrugation as a result of stress creation and its relaxation within the film. Local surface changes around active sites and evaluation of these changes occurring during successive cycles of H2/O2 titration are distinguished and discussed. A comparison of the studied phenomena observed on two significantly different substrates, glass with random distribution of local defects and well-ordered mica with steps and terraces, exhibits differences resulting from the influence of the substrate on the generation of surface defects within the palladium film. Consequently the distribution of active sites in the reaction with hydrogen on thin Pd films reflects the distribution of local point defects on the substrate surface. The applied experimental procedure gives the possibility to distinguish between active sites in nanometric resolution and their evaluation in the course of the studied processes.
We investigate the quantum states of hydrogen atom motion on Pd(111) surface and in its subsurface by calculating the wavefunctions and the eigenenergies for hydrogen atom motion within the framework of the variation method on an adiabatic potential energy surface (PES), obtained through first-principles calculations, for the hydrogen atom motion. The calculated results show that the ground-state wavefunction for the hydrogen atom motion localizes on the face-centered cubic (fcc) hollow site of the surface. The higher excited state wavefunctions are distributed between the first and second layers, and subsequently delocalized under the second atom layer. These suggest that an effective diffusion path of the hydrogen atom into the subsurface area passes through the fcc hollow site to the octahedral sites in the subsurface. Moreover, activation energies for diffusion of H and D atoms over the saddle point of the PES between the fcc hollow site and the first (second) octahedral site are estimated as 598 (882) meV and 646 (939) meV, respectively. Furthermore, the activation energies for diffusion of H and D atoms over the saddle point of the PES between the first (second) octahedral site and the fcc hollow site are estimated as 285 (483) meV and 323 (532) meV, respectively.
Any technologically important chemical reaction typically involves a number of different elementary reaction steps consisting of bond-breaking and bond-making processes. Usually, one assumes that such complex chemical reactions occur in a step-wise fashion where one single bond is made or broken at a time. Using first-principles calculations based on density functional theory we show that the barriers of rate-limiting steps for technologically relevant surface reactions are significantly reduced if concerted reaction mechanisms are taken into account.
Mit Hilfe von Übertragungskatalysatoren (UH3-, UD3-, Cu-Pulver) wurden an kompaktem Pd (0,3 und 0,15 mm Folien) Aufbau- und Abbauisothermen mit H2 und D2 gemessen; Temperaturen + 75 bis −78 °C, Drucke 760 bis 10⁻³ Torr, Atomzahlverhältnisse n (= H/Pd bzw. D/Pd) von 0,001 bis 0,83. Aus den Messungen im Bereich der α-Phase wurden die chemischen Potentiale des gelösten H bzw. D im Grenzzustand n 0 und ihre auf eine Attraktionswechselwirkung zurückgehenden, der Konzentration proportionalen Zusatzanteile berechnet. Im Bereich der β-Phase wurden die entsprechenden Zusatzanteile ermittelt. Sie treten hier formal als Repulsionswechselwirkung auf; die Desorptionsenthalpien nehmen linear mit n ab. extrapoliert auf n 1 bis nahezu auf 0. – Eine eingehende Diskussion der im Bereich des Zweiphasengebietes auftretenden Hysterese führt zu dem Schluß, daß die Zersetzungsdrucke nahezu den Gleichgewichtsdrucken entsprechen. Aus ihrer Temperaturabhängigkeit ergaben sich die Zersetzungsenthalpien zu ΔHH2 = 9,32 ± 0,1; ΔHD2 = 8,88 ± 0,1 (kcal/mol), die Zersetzungsentropien zu ΔSH2⁰ = 21,8 ± 0,2, ΔSD2⁰ = 23,4 ± 0,2 (cal/grd.mol). Messungen an Pd-Mohr lieferten innerhalb der Fehlergrenzen dieselben Werte. - Für die Trennfaktoren von H2/D2-Gemischen 1/1 an Pd-Mohr wurde α = 2,25 (50 °C) bis 3,7 (−78 °C) gefunden.
The desorption of deuterium from the low temperature (α2) state on a Pd(110) surface was studied by TDS, LEED and Δφ. Simultaneous measurements of the isothermal desorption of the α2 state, intensity of a half-order beam of the (1 × 2) phase and work function change show clearly that the desorption of this low temperature state is associated with the phase transition (1 × 2) to (2 × 1). This transition correlation with a work function change of ∼ 70 mV. The desorption follows nearly zero-order kinetics with an activation energy of ∼ 71 kJ mol−1. The desorption kinetics are thus closely analogous to those on Ni(110).
The adsorbed states of hydrogen on Pd(1 1 0) surfaces at 100–300 K have been studied using high-resolution electron energy loss spectroscopy (EELS) and low-energy electron diffraction (LEED). The observed losses are at 96, 118 and ∼ 165 meV for the (2 × 1)-H surface; 96 and ∼ 125 (shoulder) meV for the (1 × 2)-H surface formed at 100 K; ∼ 90 and 118 meV for the streaky (1 × 2)-H surface formed at 300 K. Tentative assignments of the losses are made. The adsorbed states of hydrogen are discussed correlating the EELS-LEED results with the thermal desorption and molecular beam diffraction studies which have previously been reported.
It is now established by experiments that the chemisorption of hydrogen on Pd(110) surface can induce reconstructions. Under different conditions of coverage and temperature, two types of reconstruction have been found: pairing-row and missing-row. Although the way in which the substrate reconstructs can be clearly deduced from experiments, the way in which H atoms are arranged on the reconstructed surfaces is not so well established. This is due to the difficulty of directly “seeing” H atoms by experimental methods. In this work, we report a systematic study of the reconstructions on Pd(110) by ab initio calculations based on density functional theory with local density approximation (LDA), generalized gradient approximation (GGA) and ultrasoft pseudopotentials. A variety of models are studied in detail. We show that only particular arrangements of H atoms can induce the surface reconstructions considered. The driving force for the reconstructions is analysed.
The mechanism of olefin hydrogenation on a supported noble-metal catalyst requires the presence of weakly bound hydrogen atoms absorbed in the volume of the metal particle (see picture). Co-adsorbed carbonaceous deposits affect the hydrogen distribution in the metal clusters and critically control their activity and selectivity in olefin conversions. (Figure Presented)
Surface-structure models for the 21 and 12 hydrogen chemisorption phases formed on Pd(110) at 100 K have been derived from He-diffraction data. The respective coverages correspond to 1 and 1.5 monolayers (ML). Upon heating to 200 K, the 12 saturation phase transforms back into the 21, and 0.5 ML hydrogen moves subsurface. Based on structural arguments, we suggest that only the first available subsurface sites, i.e., the octahedral interstitials between topmost and second layers are populated by thermal activation. The subsurface movement is eased since H-chemisorption sites on top of the second Pd layer can be occupied in the 12 owing to substrate reconstruction. Structural considerations also explain that exactly 1 ML H can be accommodated subsurface by thermal cycling. New TDS measurements corroborate these notions: only the 2 desorption state, probably associated with Hin subsurface sites between first and second Pdlayers, is selectively filled by the thermal-cycling processes. The 1 state remains empty upon thermal cycling, and is very likely connected with hydrogen deeper in the bulk.
A method for the measurement of the concentration of hydrogen versus depth in solids using the 1H+15N resonant nuclear reaction is discussed. This method has a typical depth resolution of 50–100 Å, can be used to a depth of several microns, and can measure hydrogen in concentrations of one part per thousand or greater.
Chemisorption of hydrogen on metal surfaces requires the dissociation of the H2 molecule in the first place; this process has been experimentally investigated and theoretically described in terms of multi-dimensional potential energy diagrams. The adsorption of atomic hydrogen is frequently accompanied by displacements of the metal surface atoms leading to phenomena such as layer relaxation or surface reconstruction. Especially surface reconstruction may be regarded as a precursor stage for a progressive chemical attack of the hydrogen atoms also on the bulk metal, leading to the occupation of so-called “subsurface” sites, to bulk diffusion and, finally, to hydride compound formation. All these processes depend sensitively on the crystallographic structure of the surface, and some examples for H on Rh, Co and Pd surfaces will demonstrate the general correlation between the hydrogen surface concentration and the metal's cohesive energy, surface crystallography, and its tendency to reconstruct.
The absorption of hydrogen at Pd(100) during exposure to H2 is studied by thermal desorption and high-resolution electron energy-loss spectroscopies. At the temperatures investigated (i.e. 105–200 K), absorption occurs by a mechanism in which the hydrogen molecules impinging on the H-covered surface are dissociated and penetrate at surface defect sites. Neither the penetration of prechemisorbed hydrogen atoms nor the direct absorption of gaseous hydrogen at terrace sites is observed under our experimental conditions. Below 120 K, H atoms penetrate via quantum tunneling and remain at the subsurface sites to form palladium hydride, which gives rise to a molecular hydrogen desorption peak at 155 K. The absorption energy for the subsurface site is estimated to be 3.4 kcal mol-H−1. At higher temperatures, the hydrogen atoms diffuse thermally into the Pd bulk during exposure and desorb at around 300 K, exhibiting a broad structure. An isotopic difference is observed in the absorption state which is associated with the higher mobility of D. The activation energies of penetration into the subsurface sites are determined to be 1.1 kcal mol−1 for H and 0.8 kcal mol−1 for D. The structure of the absorption site is discussed.
High precision nuclear reaction resonance width measurement techniques were applied to the 429 keV resonance in yielding a result which may be stated at present as: Γ=120±30) eV.Work is in progress to improve precision further. The measurements were performed using a 2 MV Van de Graaff equipped with both slit and capacitive pick-off plate feedback stabilization and fitted with an automatic energy scanning system. Energy resolutions as low as ∼40 eV fwhm were reached. Thick, nearly stoichiometric niobium nitride targets highly enriched in 15N were prepared on silicon backings by reactive sputtering in a 15N2 and argon mixture. The measurements were carried out using a ∼1μA beam, at room temperature, in 10−6 Torr liquid nitrogen trapped vacuum. The characterization of the targets as well as the fitting procedures used to extract Γ from the thick target yield curves (including Doppler effects) are presented.
The dissociative adsorption of a hydrogen molecule on the Pd(110) surface is studied by ab initio total energy calculations. These calculations are based on the density functional theory (DFT) with generalized gradient approximation (GGA), plane wave basis, and ultra-soft pseudo-potentials. A variety of dissociation pathways are analyzed in detail. We have found non-activated as well as some activated pathways. Moreover, a precursor state and transition states are identified. In order to understand the dissociation mechanism, the electronic structure along different reaction paths is analyzed in detail by examining the projected density of states onto bonding and antibonding orbitals. This has allowed us to understand how the adsorbate–substrate bond is formed, and the role played by the orientation of the H–H bond. The energetic order of the different pathways is essentially determined by the interaction between the Pd orbitals and the bonding orbital of the hydrogen molecule.
The adsorption of hydrogen on clean Pd(110) and Pd(111) surfaces as well as on a Pd(111) surface with regular step arrays was studied by means of LEED, thermal desorption spectroscopy and contact potential measurements. Absorption in the bulk plays an important role but could be separated from the surface processes. With Pd(110) an ordered 1 × 2 structure and with Pd(111) a 1 × 1 structure was formed. Maximum work function increases of 0.36, 0.18 and 0.23 eV were determined with Pd(110), Pd(111) and the stepped surface, respectively, this quantity being influenced only by adsorbed hydrogen under the chosen conditions. The adsorption isotherms derived from contact potential data revealed that at low coverages θ ∞ √pH2, indicating atomic adsorption. Initial heats of H2 adsorption of 24.4 kcal/mole for Pd(110) and of 20.8 kcal/mole for Pd(111) were derived, in both cases Ead being constant up to at least half the saturation coverage. With the stepped surface the adsorption energies coincide with those for Pd(111) at medium coverages, but increase with decreasing coverage by about 3 kcal/mole. D2 is adsorbed on Pd(110) with an initial adsorption energy of 22.8 kcal/mole.
The interaction of atomic hydrogen with the Pd(111), Pd(100) and Pd(110) surfaces is studied by ab-initio density functional calculations within the generalized gradient approximation (GGA). For the three surfaces, we have determined the preferred adsorption sites, the adsorption structures, the work function changes and the surface diffusion barrier, including relaxation effects. This comparative study allows some common features to be seen, in particular in the adsorption energies and geometries for both surface and subsurface H-atoms, and some significant differences such as the surface diffusion and the dispersion of the H-induced surface state. The origin of these differences is explained by a detailed analysis of the electronic structures of both clean and hydrogen-covered surfaces. Our study leads to an interesting correlation between the hydrogen diffusion barrier and the surface roughness since it plays an important part in the catalytic activity of the respective surfaces.
Temperature programmed desorption (TPD) subsequent to various hydrogen exposure conditions indicates the formation of chemisorption, solid solution, and hydride phases of hydrogen in the near surface region of Pd(111). Variation of the sample exposure temperature (Te) between 80 and 300 K has a strong effect on the subsequent TPD spectra. At Te = 80 K a single desorption peak, β, appears at 310 K. Coverage variation of the β peak is consistent with second-order recombinative desorption of chemisorbed hydrogen. For Te between 90 and 140 K a slight saturate, exhibits near-zeroth-order desorption kinetics, and is assigned to the decomposition of a near surface palladium hydride phase. Population of the α peak is thermally activated with a maximum at Te≈115 K. For Te greater than 140 K, α disappears while the total amount of absorbed hydrogen increases significantly. At these temperatures, the concentration of absorbed hydrogen decreases significantly if the sample is held in vacuo at Te after completion of the hydrogen exposure. At all exposure temperatures there is also a broad desorption feature near 800 K which is enhanced by higher Te and is associated with hydrogen in solid solution.