JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 92, NO. D2, PAGES 2195-2199, FEBRUARY 20, 1987 Carbon Kinetic Isotope Effect in the Reactionof CH, With HO J. A. DAVIDSON,C. A. CANTRELL, S.C. TYLER, R. E. SHETTER, R. J. CICERONE, AND J. G. CALVERT National Center for Atmospheric Research, Boulder,Colorado The carbon kinetic isotope effect in theCH,, + HO reaction can enrich atmospheric methane in 13C relative to its sources. In the only previousmeasurement of the carbon kinetic isotope effect in this reaction, a valueof k•2/k•3 of 1.003 was reported. In the present experimental determination we have found a significantly larger kinetic isotopeeffectof 1.010+ 0.007 (95% confidence interval).This report discusses the experimental details of the carbon kinetic isotopeeffect determination in the CH,, + HO reaction and the possible atmospheric implications in the useof the present result. Tropospheric methane is known to have a 13C/12C ratio of -47%o (partsper thousand) relativeto Peedee belemnite, and most known methanesources for which measurements of the stablecarbon isotoperatio have beenperformed aredepleted in •3C relative to the atmosphere. A k•2/k•3 ratio of 1.010 leads to an expected value for the weighted average of the 13C/12C ratio of the methane sources which is about 10%o more depleted than the atmosphere (approximately -57%o). This result is fundamentally important to thoseresearchers who usestablecarbon isotoperatios to study atmospheric methane. INTRODUCTION Atmospheric methane concentrations are increasing glo- bally. This temporal increase began 150-200 years ago, and the rate of increase has accelerated in recent decades [Graedel and McRae, 1980; Blake et al., 1982; Stauffer et al., 1985; Rinsland et al., 1985]. It is important to deducethe reasons for these global changes, if for no other reason than to permit well-basedpredictions of future CH 4 concentrations and their climatic and chemical effects [-Ramanathanet al., 1985; Dickin- son and Cicerone, 1986]. Since complex chemical, physical, and microbiological processes exert control over atmospheric CH4, it is difficult to interpret these global changes. One tech- nique involves measuring the ratio of [13CHa.-I to [12CH4-I from various sources and comparing these results to the 13C/12Cratio in atmospheric methane. The resultsof these studiesare usually expressed as a per mil (parts per thousand) difference: (• -- {(R x/Rstd) -- 1} x 1000 where R x is the 13C/12C ratioin a given sample andRst dis the ratio in some standard material. The overwhelming majority of sources of methane in the atmosphere which have been measured are depleted in 13CH4 relative to the atmosphere [Tyler, 1986; Stevensand Rust, 1982]. The carbon isotope ratio for atmospheric methane has been measured and found to be -47.0%0 [Stevens and Rust, 1982]. This value, corrected for atmospheric fractionation, should equal the weighted average per mil value of the sources, provided all the sources of methane in the atmosphere are known. The correction for fractionation can in principle be easily applied, if the only sink of tropospheric methane is its reaction with hydroxyl radicals. CH• + HO-• CH 3 q- H20 (1) Also, implicit in the use of stable carbon isotope ratios as a tool for determining the budget of tropospheric methane are the assumptions that methane is globally uniform and that its concentrationis, to a first approximation,constant.If kl2 is therateconstant for •2CHa. -t-HO andk•3 is therateconstant for the corresponding •3CHa. reaction, a positive carbon iso- tope effect, k12/ki3 > 1, in this reaction fractionates atmo- spheric methane, causing an enrichment in 13CH4 relative to methanesources. Sincethe observed •3C/•2C ratio in atmo- spheric CH 4 must be correctedby this fractionation, it is es- sentialto know the true value of ki2/k • 3' Only one determination of the carbon kinetic isotope effect in this reaction has been previously reported. Rust and Stevens [1980] found the ratio of the rate constantsk•2/k•3 to be 1.003. In this communication we report results of a determi- nation of the carbon kinetic isotope effect in (1). EXPERIMENTAL PROCEDURES The measurement of the carbon kinetic isotope effect in (1) is a difficult problem for several reasons. First, one would expect that the rate constant for the reaction of 12CHa. would exceed that of •3CHa. by at mosta fewpercent, because of the small changes in the energy partition functions in methane and the correspondingmethane-hydroxyl transition state due to the substitutionof 13C for •2C. On the other hand, a frac- tion of a percent difference in the rate constants is critical in the use of carbon isotope ratios as a diagnostic for under- standing sources of atmosphericmethane. The determination of rate constant ratios to a fraction of a percent requires ex- tremely accurate and precise analytical techniques. Second, the reaction of CH• with HO is rather slow (k298= 7.7 x 10-•5 cm 3 molecule- • s- • [DeMore et al., 1985]), so species which react more rapidly with CH4 can interfere with the determination even though they have a much lower steady state concentration. Third, many compoundsreact much more rapidly with HO than does methane. In order to allow HO to react effectivelywith methane and thus to obtain a reasonable methane conversion in a tractable amount of time, the con- centration of these compounds must be kept to a minimum. The experimental conditions and proceduresof this study can best be described by division into three sections describing the chemical,photolysis, and analytical systems independently. Copyright 1987 by the American Geophysical Union. Paper number 6D0651. 0148-0227/87/006D-0651 $05.00 Chemical System After consideringa number of possible HO sources, the O(ID) + H20 reaction was chosen as the most viable, since 2195