No abstract

The "Co y-radiolysis of N 2 0 adsorbed o n silica gel and zirconia has been studied. In both systems a large fraction of the energy absorbed in the solid is able to cause decon~position of the N 2 0 . For the silica gel -N 2 0 system, studies with electron scavengers have shown that on gels degassed at low temperatures, electrons are precursors for a large fraction of the yield and about 1 G unit of nitrogen is produced by other processes. The total nitrogen yield is smaller for gels degassed at high temperatures, but the yield of nitrogen produced by direct excitation stays constant at 1 G unit.Experiments using electron scavengers with the zirconia-N20 system have shown that direct excitation of the solid is responsible for most of the nitrogen yield at 7 G units on high surface zirconia. Studies of nitrogen yields o n a number of zirconias showed that the range over which energy is transferred is about 50 A.

Re.sr~orc11 Cltrtt~istry Brot~cl~. Wl~i/e.shell Nuclear Reseorcll Estublisl~tnent, Atonlic Energy of Cunud(~ Lir~~itrd. Pit1uit.u. Mut~itobu. ROE 1 LO Received May 19, 1972 Rate data for hydrogen-water deuterium isotope exchange are reported for a number of platinum on carbon catalysts whose carbon supports were graphitized to varying degrees before the metal was deposited on the support. The exchange over platinum black was studied for comparison. Over carbon supported catalysts, the rate was almost independent of hydrogen pressure and increased approximately linearly with water pressure. The rate controlling step seems to be the transfer of deuterium between chemisorbed hydrogen atoms and physisorbed HDO.Activation energies for the rate determining step decreased from 14 kcal/mol over unsupported platinum to as low as 2 kcal/mol over highly graphitized supports. Rates increased with heat treatment temperatures up to 2100 "C. These effects are ascribed to electron transfer from the carbon support to the metal, resulting in a weaker Pt-H bond, and faster exchange. However they may be partially the result of non-linear adsorption of water.Nous rapportons les donnees de vitesse d'echange isotopique hydrogene-deuterium dans l'eau pour certains catalyseurs de platine sur carbone dont les supports de carbone furent prealablement traites au graphite A differents degres, avant que le metal soit depose sur le support. Pour comparer, nous avons etudie l'echange sur noir de platine. Avec des catalyseurs sur carbone, la vitesse est a peu prts independante de la pression d'hy-

Some rate data are reported for deuterium exchange between water and hydrogen using a series of platinum-silica catalysts. Three different methods of preparing catalysts were used to give different series of metal particle sizes. Within each series, the particle size was increased by sintering batches in air, at temperatures up to 700 "C. These methods produced platinum particles ranging from 5 to 250 A in diameter. Metal surface areas and particle sizes were measured by hydrogen chemisorption, X-ray diffraction, and electron microscopy.The specific rates, i.e., rates per unit surface area of platinum varied by a factor of three within each series, reaching a maximum for sintering temperatures of about 500 "C. These maximum rates, measured at 127 "C with a water-to-hydrogen ratio of 0.31, varied from 1.9 x lo-' to 5 x lo-' mol D, cm-2 s-'. Thus no effect of particle size on rate was observed, and the differences noted are ascribed to other effects of the methods of preparing the catalysts.Certaines donnees de vitesse sont report& pour 1'Cchange du deuterium entre l'eau et I'hydrogene lorsqu'on utilise une serie de catalyseurs platine-siiice. Trois mkthodes differentes de preparation des catalyseurs ont permis d'obtenir differentes series de taille de particule metallique. Dans chaque serie, la taille de la particule a Bte augmentee par frittage dans I'air a des temperatures supkrieures a 700 "C. Ces methodes ont donne des particules de platine dont la gamme d b diametres varie de 5 250 A. Les aires de surfaces metalliques et les tailles des particules sont mesurees par chimisorption de l'hydrogene, par diffraction aux rayons-X et par microscopie Blectronique.Les vitesses specifiques, c'est-a-dire les vitesses par unite de surface du platine, varient avec un facteur de trois dans chacune des series, atteignant un maximum pour des tempkratures de frittage de l'ordre de 500 "C. Ces vitesses maximales mesurees a 127 "C avec un rapport eau-hydrogene de 0.31, varient de 1.9 x a 5 x lo-' mol D2 ~m -~ s-l. Par contre, aucun effet de la taille des particules n'a kt6 observee et les differences notees sont assignees aux autres effets des methodes de preparation des catalyseurs.

The kinetics of the pyrolysis of n-butane have been studied a t temperatures from 520' to 590" C, and a t pressures from 30 to 600 mm Hg; the rate was followed from pressure changes and by gas chron~atography. The reaction was accurately of the three-halves order; the activation energy was found t o be 59.9 kcal mole-', and the frequency factor 3.24X1015 CC''~ mole-u2 sec-l. The reaction is sensitive to surface; packing the vessel and 'conditioning' it usually led to a decrease in rate and an increase in activation energy. The reaction is concluded to be largely homogeneous, and to occur almost entirely by a free-radical mechanism; the initiation reaction is considered to be the dissociation of a butane molecule into two ethyl radicals, in its first-order region, and termination is believed to be the second-order combination of ethyl radicals. The mechanism proposed is shown to account satisfactorily for the observed behavior. The surface effect is attributed to a certain amount of initiation by abstraction, by a surface atom, of a hydrogen atom from butane, and to surface catalysis of the recombination of ethyl radicals. ISTRODUCTIONRecent papers from this laboratory have dealt with the kinetics and mechanisms of the thermal decomposition of ethane (1) and propane (2). The n-butane decomposition presents an interesting contrast in that hydrogen atoms are no longer predominant as chain carriers; the free-radical mechanisms are therefore of a different character. A considerable amount of recent work has been done on the pyrolysis of n-butane, but certain features of the mechanism have still remained obscure; in particular may be mentioned the orderaf the initiating reaction, the nature of the main chain-ending step, and the role of the surface. A reinvestigation of the reaction was therefore considered to be worthwhile, and was carried out in conjunction with the study of the inhibited reaction described in the following paper (Part 11).The main features of the previous work on this reaction may be summarized as follows. Pease (3) and Pease and Durgan (4) established that the reaction is largely homogeneous and approximately of the first order; the overall reaction could be represented mainly by the equationsSteacie and Puddington (5) found that the reaction products could initially be represented by the equationThey reported the first-order rate constant as Echols and Pease (6) found that the order was closer to 3/2 than to unity.The presence of free radicals in the reaction system was demonstrated by Rice, Johnston, and Evering (7), and Rice and Rice (8) proposed a free-radical mechanism which has been modified by Steacie (9), Semenov (lo), and Benson (11). The latter proposed termination steps of the pp type, leading to 3/2-order kinetics.