Reactions of Saturated Hydrocarbons

Of the reactions closely related to those discussed above, hydrogenolysis of various bonds (C-C, C-O, C-N) is most important. Upon hydrogenolysis addition of hydrogen is accompanied by splitting of a strong bond (with the molecule in the adsorbed state). Metals are the most active catalysts of these reactions. 

The C-X can be split thermally, but this step requires a high activation energy. Catalytic splitting is easier. In the latter case, multiple bonding with the surface occurs (see the right side) and for most of the metals of Groups 8-10, this requires that the site of the scheme above comprises several contiguous metal atoms. 

Hydrogenolytic reactions can also be suppressed by carbonaceous layer deposition or by modification of the active surfaces by — for example — sulphur. Suppression of hydrogenolysis gives more chance to a reaction the rate of which decreases only proportionally to the active metal surface concentration, like hydro/dehydrogenation, or some forms of isomerisation (5C cyclic mechanism see the chapter on elementary steps). 

A dehydrocyclization (incl. aromatization) hydrogenolysis O isomerization. Right The same catalysts after a standard sulphurization procedure. Arrows show results obtained with a physical mixture of Pt/S/ AkOaand Pt/Re/S/ AI2Q3 catalysts. T = 620 K, 1 bar total pressure. 

Pt-Re-S/Si02 catalysts under the same conditions are inactive. The Pt-Ir-S/Si02 catalysts behave slightly differently a larger proportion of the total conversion is due to the sulphur-free metallic surface. The catalysts containing Ir are faster than the Re-containing ones but the selectivity to Cs+ is lower. 

---

This realization led me to study related possible intermolecular electrophilic reactions of saturated hydrocarbons, Not only protolytic reactions but also a broad scope of reactions with varied electrophiles (alkylation, formylation, nitration, halogenation, oxygenation, etc.) were found to be feasible when using snperacidic, low-nucleophilicity reaction conditions. 

The important role of LU MO in the nucleophilic reactions of saturated hydrocarbons bearing nucleophilic substituents (halogens, alkoxy-, acyloxy-, RSO2O-, etc.) in the molecule has been pointed out 122,123). 

Interaction of chlorine with methane is explosive at ambient temperature over yellow mercury oxide [1], and mixtures containing above 20 vol% of chlorine are explosive [2], Mixtures of acetylene and chlorine may explode on initiation by sunlight, other UV source, or high temperatures, sometimes very violently [3], Mixtures with ethylene explode on initiation by sunlight, etc., or over mercury, mercury oxide or silver oxide at ambient temperature, or over lead oxide at 100°C [1,4], Interaction with ethane over activated carbon at 350°C has caused explosions, but added carbon dioxide reduces the risk [5], Accidental introduction of gasoline into a cylinder of liquid chlorine caused a slow exothermic reaction which accelerated to detonation. This effect was verified [6], Injection of liquid chlorine into a naphtha-sodium hydroxide mixture (to generate hypochlorite in situ) caused a violent explosion. Several other incidents involving violent reactions of saturated hydrocarbons with chlorine were noted [7],... 

Considerable interest in the subject of C-H bond activation at transition-metal centers has developed in the past several years (2), stimulated by the observation that even saturated hydrocarbons can react with little or no activation energy under appropriate conditions. Interestingly, gas phase studies of the reactions of saturated hydrocarbons at transition-metal centers were reported as early as 1973 (3). More recently, ion cyclotron resonance and ion beam experiments have provided many examples of the activation of both C-H and C-C bonds of alkanes by transition-metal ions in the gas phase (4). These gas phase studies have provided a plethora of highly speculative reaction mechanisms. Conventional mechanistic probes, such as isotopic labeling, have served mainly to indicate the complexity of "simple" processes such as the dehydrogenation of alkanes (5). More sophisticated techniques, such as multiphoton infrared laser activation (6) and the determination of kinetic energy release distributions (7), have revealed important features of the potential energy surfaces associated with the reactions of small molecules at transition metal centers. 

AE Shilov, GB Shulpin. Activation and Catalytic Reactions of Saturated Hydrocarbons in the Presence of Metal Complexes. Dordrecht Kluwer Academic, 2000. 

In the 1970s, Brouwer and Kifflin reported the reactions of saturated hydrocarbons with aliphatic aldehydes and ketones in superacidic media. Analysis of the products from these reactions suggested that the protonated aldehydes and ketones (carboxonium ions) were reacting at the carbon-hydrogen o-bonds of the alkanes. This was a surprising observation because carboxonium... 

Shilov AE, Shul pin GB (2000) Activation and catalytic reactions of saturated hydrocarbons. Kluwer, Dordrecht... 

The main facts about the mechanism of the exchange reactions of saturated hydrocarbons are as follows ... 

Protolytic reactions of saturated hydrocarbons in superacid media21 were interpreted by Olah as proceeding through the protonation (protolysis) of the covalent C—H and C—C single bonds. The reactivity is due to the electron donor ability of the <7 bonds via two-electron, three-center bond formation. Protolysis of C—H bonds leads via five-coordinate carbocations with subsequent cleavage of H2 to trivalent ions, which then themselves can further react in a similar fashion ... 

These methods appear rather simple, yet they were the starting point of a long evolution. Gilles Klopman, whose research interests at Case-Western Reserve University later turned to modeling bioactive molecules, was the first to use Sandorfy s methods. Kenichi Fukui made extensive use of them in his well-known work on the structures and reactions of saturated hydrocarbons and their derivatives. Fukui added his frontier orbital considerations. Around 1959 the milieu of developments in quantum chemistry contributed to inspire William N. Lipscomb to conceive the extended Hiickel method, which was subsequently implemented by Lawrence L. Lohr and Roald Hoffmann.83 Soon thereafter, John Pople and his coworkers introduced self-consistent field methods based on the zero-differential overlap approximation.815... 

A similar mechanism (Eq. 4) is operative in reactions of saturated hydrocarbons with closed-shell oxidizing electrophiles (E = Haln+, NOz+, etc.) where the H-trans-fer from a C-H bond is accompanied by an ET through the linearly H-coupled fragment (H-coupled electron transfer) [13]. The hydrocarbon part of the transition structures resembles the respective radical cation (that for the reaction of adamantane with Cl7+ is shown (6) in Scheme 2) [13]. 

The product mixture does not contain equal numbers of moles of the dichloroethanes so we do not show a stoichiometricaUy balanced equation. Because reactions of saturated hydrocarbons with chlorine can produce many products, the reactions are not always as useful as might be desired. 

Recently we observed the effect which supports the conclusion about the substantial role of the radical reaction outside of the catalyst grains. When a very efficient OCM oxide catalyst (10% Nd/MgO) was placed into the reactor together with an inactive metal filament (Ni-based alloy) the sharp increase of conversion accompanied by the selectivity shift from oxidative coupling to the formation of CO and H2 was observed [19]. Since the metal component has a low activity also with respect to ethane oxidation, this behavior is not due to successive oxidation or decomposition of C2 hydrocarbons on the metal surface, but should be attributed to the reactions of methane oxidation intermediates. Almost total disappearance of ethane (which is a product of CH3 radicals recombination) and acceleration of the apparent reaction rate by the addition of an "inert material indicate that the efficiency of methane oxidative transformations can be substantially increased if the radicals have a chance to react outside the zone where they formed and the role of reaction (-1) decreases. Although the second (metal) surface is not active enough to conduct the reaction of saturated hydrocarbon molecules (methane and ethane), the radicals generated by the oxide can react further on the metal surface. As a result, the fraction of the products formed from methane activated in the reaction (1) increases, and the formation of the final reaction mixture of different composition takes place. 

The second hypothesis seems to be a straightforward conclusion (granted that protons are available) as far as olefin reactions are concerned. However, it is not so obvious with respect to reactions of saturated hydrocarbons. Nevertheless, there is experimental evidence that small amounts of olefins will greatly accelerate certain acid-catalyzed reactions of saturated hydrocarbons. Pines and Wackher (51) showed that the isomerization of pure n-butane with AICI3 + HC1 does not proceed in the absence of added olefins (or other carbonium ion former) except under conditions where olefins or their equivalent are probably produced. 

---