Michel Coulon

Morphology and crystalline order in catalytic carbons

Abstract The influence of temperature and of catalyst composition on the morphology and crystalline order has been studied for a series of catalytic carbons prepared by carbon monoxide disproportionation (2CO→CO2+C) over an iron-cobalt, an iron-nickel, and a nickel catalyst, and by methane decomposition (CH4→2H2+C) over a nickel catalyst. The morphology was studied by TEM in a bright field mode. The products are filamentous when prepared at low temperature and granular when prepared at high temperature. The ranges of temperature corresponding to each morphology depend both on the alloy composition and on the reacting gas. The structure of the catalytic carbons, as studied by TEM in SAD mode and by X-ray diffraction, is turbostratic (2D, i.e. biperiodic) when the carbons are prepared at low temperature and crystalline (3D, i.e. triperiodic) when prepared at high temperature. When the particle size is large, the temperature of transition is about 500°C, whatever the composition of the alloy and of the reacting gas are. However, small particles are always 2D. The carbon layers, as observed by TEM in lattice fringe mode, are short and curved for turbostratic, and large and stiff for crystalline phase. Biperiodic products are not graphitizable.

Formation de fibres de carbone a partir du methane: I Croissance catalytique et epaississement pyrolytique

Abstract Carbon fibers are grown from a methane-hydrogen mixture (30 moles CH4 percent; 200 cm3/min; room pressure) flowing over a catalyst (whose precursor is an alcoholic solution of ferric nitrate) deposited on a reconstituted graphite support progressively heated up to 1150°C (Fig. 2). Densities of about 300 fibers (mean length 2 mm) per square millimeter of the support are obtained. Less fibers but of higher length are obtained when preheating (up to 950°C) is carried out under pure hydrogen. In our usual process, the final temperature is kept constant at 1150°C during half an hour under the gas flow in order to get fibers thick enough to be handled. Diameters measurements (by transmission optical microscopy) of fibers obtained after different durations at 1150°C allowed calculation of mean rate of carbon deposition that was found to decrease with time (Fig. 3) from 7.4 to 2 μg/cm2/mint-1. Moreover, some thin fibers are wrapped together by pyrolytic carbon into a thicker one (Fig. 5). Fibers are not perfectly cylindrical but show an enlargement (twice) of diameter from one end to the other. Just before the thickening stage at 1150°C, mean diameter is already larger than 1 μ. In order to examine fibers at intermediate stages of growth, some experimental runs were interrupted at lower temperatures. After stopping at 1100°C, fibers are thinner and some of them can only be observed by transmission electron microscopy. Figure 6 shows the two types of carbon, catalytic and pyrolytic, composing the fibers. The diameters of the catalytic particles (Fig. 7) inside these fibers are between 9 and 14.5 nm if preheating has been carried out under pure hydrogen and between 4 and 10 nm if hydrogen-methane mixture has been used during the whole heating. When heating is stopped at 1050°C, electron micrographs show only filaments whose structure is like the one already reported for catalytic filaments (Fig. 8). In Section 4 of this article, we confront our results (structure and shape of the fibers) with the two-stage mechanism admitted in the literature: lengthening of extremely fine catalytic filament followed by radial thickening of this precursor filament by pyrolytic deposition of carbonaceous material.

Formation de fibres de carbone a partir du methane II: Germination du carbone et fusion des particules catalytiques

Abstract This article presents the evolution of the catalyst and the beginning of carbon precipitation (germination) before the growth, carried out by a process already described[1], of carbon fibers from methane. The precursor of the catalyst is a film of ferric nitrate. The evolution of the catalyst particles is studied by interrupting the process at successive stages, quenching and observation under a TEM (Figs. 2 and 3). The mean diameter of the particles as well as the distribution of their sizes have been determined as a function of the highest temperature of treatment (Figs. 4–7). Average diameters increases with temperature; it is larger when heating has been carried out under pure hydrogen (Figs. 2, 4, and 6) than it is when the mixture H 2 CH 4 has been used (Figs. 3, 5, and 7). Size distributions at 950°C are showed on Fig. 8: curve C1 after heating under the mixture, curve C2 after heating under pure H2. In the latter, the distribution is very sensitive to carbon: introduction of the mixture and heating to 1000°C leads to a redispersion (curve C3). So, two antagonistic effects are evidenced: —Sintering particles induced by heating. —Redispersion and prevention of sintering by carbon coating; this redispersion is specially important when preheating has been carried out under pure hydrogen. Previous results have shown that no fibers are obtained from particles with diameters lower than 4 nm or higher than 14.5 nm and that catalytic lengthening of the fibers occurs between 1050 and 1100°C. The size effect on the melting point of the iron particles is derived here by a complete thermodynamic calculation taking into account the Young-Laplace law. It shows that the populations of particles that melt in the temperature range where the catalytic growth occurs correspond to the population observed in the fibers. We believe then that solely the melting of particles allows lengthening rate high enough to yield long carbon fibers before the catalyst particles are prisoned by pyrocarbon deposit. Figure 9 suggests a scheme of the mechanism leading from the precursor to fibers.

Formation and characterization of catalytic carbons obtained from CO disproportionation over an iron nickel catalyst—I: Fragmentation and rates of carbon deposition

Abstract A series of catalytic carbons has been prepared by CO disproportionation (2CO→CO2 + C) over an iron nickel catalyst. The catalyst is under the form of filings (particle sizes ranging from 10 to 160 μm) of an iron nickel alloy (75 wt % nickel, 25 wt % iron). The dependence of the rate of carbon deposition on temperature, between 400 and 650°C, and on the CO partial pressure in the reacting CO CO 2 mixtures has been investigated. Since the specific area of the catalyst is initially very small, its fragmentation by carbon deposition is necessary to obtain an appreciable rate of reaction. Experimental evidence suggest that the existence of a high density of dislocations in the catalyst is a necessary condition for its fragmentation.

Variation du potentiel d’extraction électronique du carbone par adsorption d’hydrogène moléculaire et atomique

L’hydrogene moleculaire provoque une faible variation du potentiel d’extraction electronique du carbone reversible avec la pression entre 10 -7 et 10 -4 torr.L’hydrogene atomique produit par dissociation de l’hydrogene moleculaire sur un filament chaud de tungstene, s’adsorbe irreversiblement sur le carbone. Lorsque la surface est saturee, la variation correspondante du potentiel d’extraction electronique est de — 180 ± 20 mV. Cette couche se desorbe par chauffage; a 1 200 °C la desorption est pratiquement complete.