rate in helium to that in nitrogen would be 1.4. The observed ratio, 51/34 or 1.5, agrees as closely as one could expect in view of all the assumptions and measurements involved. However, on the same basis the predicted rate of distillation in steam is only 10% greater than that in nitrogen. T h e observed rate in steam is 300y0 of that in nitrogen. Clearly some factor besides diffusional resistance must be at work in the case of steam.

ARLIER studies of hydrocarbon pyrolysis (12, 14, 4 6 ) were E usually conducted at temperatures below 650" C. (1202" F.), or, when higher temperatures were used, at extremely short reaction times, t o obtain data suitable for kinetic interpretation. Because of the emphasis given to these relatively mild pyrolysis conditions, the interpretation of results generally stressed dissimilarity in the behavior of individual or groups of hydrocarbons. However, except in the production of acetylene, practical operating conditions for hydrocarbon pyrolysis fall in the temperature range of 1200" t o 2000" F. at reaction time well above 0.1 second.In this range of pyrolysis conditions the gaseous product distribution for most feed hydrocarbons other than methane shows a striking similarity, only slightly affected by the composition, structure, or molecular weight of the feed hydrocarbon when cracked at equivalent temperatures, reaction times, and partial pressures (14, 28).I n the first part of this study, empirical relationships for the prediction of gaseous product distribution are developed for a wide range of feed hydrocarbons by the use of a simple and general approach t o the analysis of the behavior of the complex gaseous reaction systems encountered in high-temperature, vapor-phase cracking. Pyrolysis data show similiar gaseous product distributionsPrevious studies have shown that hydrocarbons of equal carbon-hydrogen weight (C/H) ratio will yield approximately equal quantities of gas of similar composition (26-28, $3) independently of their molecular weight or structure as long as the severity of cracking is sufficient to eliminate the effects of the primary decornnosition reactions. The composition, although not the total yield, of the nongaseous products was shown to be affected by feed hydrocarbon properties other than carbon-hydrogen ratio because of the tendency of feed hydrocarbons of higher molecular weight t o form nongaseous cracking products through primary reactions-that is, those specific to the feed hydrocarbons, such as dehydrogenation, cyclization and dehydrocyclization, and polymerization.As illustrated in Table I, the influence of carbonhydrogen ratio on gaseous product distribution is small for natural gas liquids, and for petroleum and shale oils, so that only minor changes in cracking conditions will compensate for the tendency of hydrocarbons of higher carbon-hydrogen ratio to form gases of higher hydrogen content and higher paraffinolefin ratio (26, 28). This similarity in gaseous product distribution tends to be masked by the relatively slow rates of the primary cracking reactions for hydrocarbons of low molecular weight such as methane, ethane, and ethylene.From the data on pyrolysis obtained in this and earlier studies, any reaction schemes attempting t o interpret gaseous product distributions-by the assumption of dehydrogenation or carbon-carbon bond scission reactions specific to a given hydrocarbon (14, 38) did not appear applicable for reaction temperatures above 1300' F. and reaction ti...

No abstract

An empirical model of natural gas discovery and production in theVr?i[ed Sta!eshas bewndeveloped to express rhe over-all eflect of~he various (echno!ogic, geologic and economic forces, Validity of this model is supported by the good fit of historical da[a, conformance wilh reasonable boundary conditions and applicability to the closely related crude oil discovery and production his(oty, Forecasts from this model are compared wirh plausible estimates of gas demand and corresponding supply requirernerrts to the year 2000, In no care are the ultimately recoverable reserves a iimiting factor. However, it isckarly shown that the ratio of proved reserves to annual production (RIP ratio) cannot be maintained near present levels without requiring unrealistically high discovery rates to meet even conservative esriinates of demand. Only by allowing a decline to about a IO-year R/p ratio by the year 1990, anda6-to8-year Rlp ratio by theyear 2000, could demand be met without a major swp-up in discovery rate. These declining reserve levels are expected to require development of economic sources of suuplernental gas to stabilize the Rlp ratio for natural gas. Depending on the criteria used ro define need forsupplementalg ar, 0.4 to I trillion cu ftlyem wordd have to b-ecome available ar early as 1978, but not later than 1990. Capital requirements for .supplemenial ga.r capacity, either as imported LNG or as synthetic pipeline gas from coal, would be on the order tif $1 billion per year.

No abstract