Ion Identification in Flames by Mass Spectrometry

1460 NATURE frequency of the F2-la.yer at any point changes by only a small a.mount during this period. At 1800 G.M.T. on September 11, 1957, when the record was obtained, the critical frequencies of the F2-layer at Slough and Ottawa were 10·5 Mc./s. and 9·9 Mc./s. respectively. Interpolation using the predicted contours of electron density gives a critical frequency of 10 ·3 Mc./s. at the mid-point of the pa.th, whereas the critical frequency deduced from the record, using the Appleton and Beynon equations, is 10·6 Mc./s. The slight difference can be readily explained as a normal layer fluctuation. Fig. l was obtained using the critical frequency of 10·6 Mc./s. and the true layer height of 360 km. The broken lines show the calculated trace, while the full lines show the observed trace superimposed to give the best 'fit'. The correspondence between the two sets of traces is sufficiently close to show that the application of the parabolic-layer equations to this path enables the maximum usable frequencies for ea.ch mode of propagation, and the pa.th delay times, to be found with good accuracy. The calculated angles of elevation of the rays in each mode of propagation are also shown on Fig. 1. This information is of obvious value in the design of aerials at the terminal points. The solution of the Appleton and Beynon equations has also been applied to propagation over a number of other long-distance paths, and close agreement has been obtained between observation and theory. A graphical technique has been evolved for applying the equations to any path, and should prove of considerable value in solving many problems of long-distance propagation. This work was carried out as part of the programme of the Radio Research Board, and is published by permission of the Director of Radio Research of the Department of Scientific and Industrial Research. F. KrFT Radio Research Station, Ditton Park, Slough, Bucks. March 11. ' Warren, E., and Hagg, E. L., Nature, 181, 34 (1958). • Appleton, E. V., and Beynon, W. 1. G., Proc. Ph111. Soc., 62, 518 (1940). Ion Identification in Flames by Mass Spectrometry IN a recent comrnunica.tion1 , P. F. Knewstubb and T. M. Sugden reported some prelimina.ry results on ion identification in flames by mass spectrometry. A different method has been developed in our Laboratory• for the extraction of ions from a flame for identification. Our first results were reported earlier•. However, after considerable improvement in the vacuum efficiency of our mass analyser (up to 10-• mm. mercury), we have modified the interpretation of our measurements due to a different distribution of the peak intensities. It seems, therefore, interesting to compare our observations with the results of Knewstubb and Sugden. All our experiments were made on acetylene/oxygen/nitrogen flames (nitrogen, 75; acetylene, 8 ·75; oxygen, 16 ·25 per cent) burning at a pressure of 40 mm. mercury. The following important peaks were observed. (1) The most important peak (- 10-11 amp.) corresponds to mass 19. Its identity, H 80 +, is shown by its splitting into four peaks with partially deuterated acetylene : 19, 20, 21 and 22 (the 22 peak May 24, 1958 vo1-. ,a, comes next to the Na peak a t a distance of one mass unit when sodium salt is also introduced in the flame). (2) Next to mass 19 a weaker peak (- 10-16 amp.) appears at mass 18 (H 2 0 + ).Using partially deuterated acetylene its splitting overlaps with the splitting of peak 19. But considered together they split into five different peaks : 18, 19, 20, 21 and 22. (3) Two weak pea.ks ( ,.._, 10-14 a.mp.) are also observed at masses 28 and 30 ; the corresponding ions do not contain hydrogen, and are most probably CO+ and NO+. The intensity of the CO+ peak seems to be enhanced in richer flames. Another interesting observation concerns some experiments which were made on a hydrogen/nitrous oxide/nitrogen flame : this was chosen instead of a hydrogen/oxygen/nitrogen flame because being yellow it was easier to run and to control than a colourless flame. As was expected for such an ion-poor flame, we observed no important peaks, except a weak maximum (- lo-u amp.) corresponding to mass 30 (NO +). The eventual dependence of relative peak intensity on the applied electric field also needs to be studied before it will be possible to decide whether the ions which have been identified are really formed in the flame or whether they result from charge transfer. This work has been sponsored in part by the Aeronautical Research Laboratory, Wright Air Development Center of the Air Research and Development Command, U.S. Air Force, through its European Office (Contract No. AF61(514)-1099). J. DECKERS .L A. VAN T1GGELEN Laboratory of Inorganic and Analytical Chemistry, University of Louvain. April 10. 'Nature, 181, 474 (1958). 'Oombwtion 0114 Flama, 1, 281 (1 957). • Bull. Soc. Ohim. Belg., 88, 664 (1957). Measurement of Low Vapour Concentrations by Collision with Excited Rare Gas Atoms THE ionization of gas or vapour molecules by collision with excited rare gas atoms has been studied in detail by Jesse and Saduski.s1 • Ionization occurs when the excitation potential of the rare gas is equal to, or greater than, the ionization potential of the colliding molecule. The vapours of nearly all organic substances have ionization potentials lower than the excitation potentials of the first three noble gases so that ionization by collision with excited rare gas a.toms forms the basis of a method for measuring low concentrations of organic vapours. In practice, high, steady concentrations of excited atoms can be established in a rare gas by the simultaneous application of a high-intensity electrical field and a source of free electrons. The acceleration of the electrons in the field raises their temperature sufficiently to excite the rare gas atoms by collision. The concentration of excited atoms is related exponentially to the electrical field intensity and directly to the concentration of free electrons ; the concentration of rare gas ions remains low until the field intensity approaches 2,000 V./cm. at 760 mm. gas pressure. The introduction of vapour molecules into a volume of rare gas containing excited atoms leads to an increase of the ion concentration. If the © 1958 Nature Publishing Group  (...truncated)