Palladium hydride system studied

This review aims to present an account of the catalytic properties of palladium and nickel hydrides as compared with the metals themselves (or their a-phase solid solutions with hydrogen). The palladium or nickel alloys with the group lb metals, known to form /8-phase hydrides, will be included. Any attempts at commenting on the conclusions derived from experimental work by invoking the electronic structure of the systems studied will of necessity be limited by our as yet inadequate knowledge concerning the electronic structure of the singular alloys, which the hydrides undoubtedly are. 

Rennard and Kokes (39) in their paper stated directly that their purpose was just to study the catalytic activity of palladium hydride in the hydrogenation of olefins, in this case ethylene and propylene. Kokes (39a) in his article recently published in Catalysis Reviews summarizes the results of studies on such catalytic systems. 

The catalytic system studied by Rennard and Kokes was in fact very complex. It can be expected that the satisfactory prolongation of the reaction should, however, result in a deviation from the formulated kinetics. Unfortunately no investigation comparable to that of Scholten and Kon-valinka has been done in the case of olefin hydrogenation. Such a study of the catalytic activity of the pure /3-phase of palladium hydride in comparison with the a- or (a + /3)-phases would supplement our knowledge concerning catalytic hydrogenation on palladium. 

Palladium hydride is a unique model system for fundamental studies of electrochemical intercalation. It is precisely in work on cold fusion that a balanced materials science approach based on the concepts of crystal chemistry, crystallography, and solid-state chemistry was developed in order to characterize the intercalation products. Very striking examples were obtained in attempts to understand the nature of the sporadic manifestations of nuclear reactions, true or imaginary. In the case of palladium, the elfects of intercalation on the state of grain boundaries, the orientation of the crystals, reversible and irreversible deformations of the lattice, and the like have been demonstrated. 

Palladium hydride is not a stoichiometric chemical compound but simply a metal in which hydrogen is dissolved and stored in solid state, in space between Pd atoms of crystal lattice of the host metal. Relatively high solubility and mobility of H in the FCC (face-centered-cubic) Pd lattice made the Pd H system one of the most transparent, and hence most studied from microstructural, thermodynamic, and kinetic points of view. Over the century that followed many metal-hydrogen systems were investigated while those studies were driven mostly by scientific curiosity. Researchers were interested in the interaction of hydrogen molecule with metal surfaces adsorption and diffusion into metals. Many reports on absorption of in Ni, Fe, Ni, Co, Cu, Pd, Pt, Rh, Pd-Pt, Pd-Rh, Mo-Fe, Ag-Cu, Au-Cu, Cu-Ni, Cu-Pt, Cu-Sn, and lack of absorption in Ag, Au, Cd, Pb, Sn, Zn came from Sieverts et al. [30-33]. 

The proposed mechanism is given in Scheme 15. Initially the dissociation of water, maybe trapped by the molecular sieve, initiates the catalytic cycle. The substrate binds to the palladium followed by intramolecular deprotonation of the alcohol. The alkoxide then reacts by /i-hydride elimination and sets the carbonyl product free. Reductive elimination of HOAc from the hydride species followed by reoxidation of the intermediate with dioxygen reforms the catalytically active species. The structure of 13 could be confirmed by a solid-state structure [90]. A similar system was used in the cyclization reaction of suitable phenols to dihydrobenzofuranes [92]. The mechanism of the aerobic alcohol oxidation with palladium catalyst systems was also studied theoretically [93-96]. 

Metal membranes. Since Thomas Graham [1866] discovered that metallic palladium absorbs a large amount of hydrogen, the metal hydride system of which palla um is an important example has been extensively studied. Hydrogen dissolved in a metal hydride system (according to the mechanism to be describ below) is considered to behave in an atomic or ionic form which is more reactive than molecular hydrogen in a gas phase. 

According to an INS study of the P- and a-palladium hydride phases formed during absorption and desorption of hydrogen on palladium foil, coarse palladium powder and finely divided palladium [33], the retention of hydrogen by bulk palladium materials depended on their mechanieal pre-treatment rather than their morphology the powdered materials retained more hydrogen than palladium foil under comparable conditions. This finding is explained by the presence of trap sites, more for the powders than for the foil. This paper contains a useful summary of INS studies of the H/Pd system which is reproduced here. Table 7.3. [34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49]. 

The results reported in this chapter refer to the palladium-hydrogen system, which, for a number of reasons, is often seen as the prototype metal-hydrogen system. Thus, as mentioned above, palladium was the first known case of a metal that dissolved hydrogen [3]. Moreover, the crystal lattice is very simple (NaCl) and the metal lattice in the hydride case is the same as in the metal (f.c.c.). This is particularly important in that it provides the opportunity to produce single crystals of the stoichiometric hydride, which makes it possible to study the individual excited states of the proton. 

The nickel(o)-catalysed oligomerization of butadiene has been further studied. Cyclo-octa-1,5-diene is a major product in hydroxylic solvents, and a minor product in the presence of secondary amines." The oligomerization of butadiene in the presence of nickel(ii)acetylacetonate, triethylamine, alkyl cyanamides, and miscellaneous heterocycles has also been studied, and catalyst systems causing an almost quantitative conversion into trans,trans,trans-, cis,trans,trans-, or cis,cis,trans-cyclododeca-l,5,9-triene were found." The ability of some hydrides, trans-MHX-(PR3)2 (M = Ni or Pd), to act as catalysts for butadiene oligomerization has been found to depend upon the substituent, X, and the solvent the reactions of some cationic palladium hydrides with butadiene was also examined. Unlike other organo-cobalt catalyst systems, CoCUPPhjjj has been found to catalyse the dimerization... 

A short survey of information on formation, structure, and some properties of palladium and nickel hydrides (including the alloys with group IB metals) is necessary before proceeding to the discussion of the catalytic behavior of these hydrides in various reactions of hydrogen on their surface. Knowledge of these metal-hydrogen systems is certainly helpful in the appreciation, whether the effective catalyst studied is a hydride rather than a metal, and in consequence is to be treated in a different way in a discussion of its catalytic activity. 

As has been shown by the X-ray diffraction method the parent metals (i.e. Pd or Ni), the a-phase, and /3-phase all have the same type of crystal lattice, namely face centered cubic of the NaCl type. However, the /9-phase exhibits a significant expansion of the lattice in comparison with the metal itself. Extensive X-ray structural studies of the Pd-H system have been carried out by Owen and Williams (14), and on the Ni-H system by Janko (8), Majchrzak (15), and Janko and Pielaszek (16). The relevant details arc to be found in the references cited. It should be emphasized here, however, that at moderate temperatures palladium and nickel hydrides have lattices of the NaCl type with parameters respectively 3.6% and 6% larger than those of the parent metals. Within the limits of the solid solution the metal lattice expands also with increased hydrogen concentration, but the lattice parameter does not depart significantly from that of the pure metal (for palladium at least up to about 100°C). 

Scholten and Konvalinka (9) in 1966 published the results of their studies on the kinetics and the mechanism of (a) the conversion of para-hydrogen and ortho-deuterium and (b) hydrogen-deuterium equilibration. At first the a-phase of the Pd-H system was used as catalyst, and then the results were compared with those obtained when the palladium had previously been transformed into its /3-hydride phase. 

The results used for a subsequent comparison of catalytic activity of all group VIII metals are related by Mann and Lien to palladium studied at a temperature of 148°C. At this temperature the appearance of the hydride phase and of the poisoning effect due to it would require a hydrogen pressure of at least 1 atm. Although the respective direct experimental data are lacking, one can assume rather that the authors did not perform their experiments under such a high pressure (the sum of the partial pressures of both substrates would be equal to 2 atm). It can thus be assumed that their comparison of catalytic activities involves the a-phase of the Pd-H system instead of palladium itself, but not in the least the hydride. 

Insertion of the alkyne into the Pd-H bond is the first step in the proposed catalytic cycle (Scheme 8), followed by insertion of the alkene and /3-hydride elimination to yield either the 1,4-diene (Alder-ene) or 1,3-diene product. The results of a deuterium-labeling experiment performed by Trost et al.46 support this mechanism. 1H NMR studies revealed 13% deuterium incorporation in the place of Ha, presumably due to exchange of the acetylenic proton, and 32% deuterium incorporation in the place of Hb (Scheme 9). An alternative Pd(n)-Pd(iv) mechanism involving palladocycle 47 (Scheme 10) has been suggested for Alder-ene processes not involving a hydridopalladium species.47 While the palladium acetate and hydridopalladium acetate systems both lead to comparable products, support for the existence of a unique mechanism for each catalyst is derived from the observation that in some cases the efficacies of the catalysts differ dramatically.46... 

The system is not a homogeneous one, as the organometallic compounds and alkali metal hydrides are insoluble in benzene, the solvent used however, the rate of reaction was found not to be affected by the addition of platinum or palladium catalysts. The reaction is formally similar to the base-catalyzed activations of hydrogen studied by Wilmarth this becomes clear if Eq. (45) is written as... 

---