No data available.
Please log in to see this content.
You have no subscription access to this content.
No metrics data to plot.
The attempt to load metrics for this article has failed.
The attempt to plot a graph for these metrics has failed.
Clean‐Up and Pressure Effects in Low Pressure Mercury Vapor Discharges: A Reversible Electrical Clean‐Up of Mercury
1.Summaries of many important papers on this subject will be found in books on high vacuum technique, of which may be mentioned those of S. Dushman, L. Dunoyer, F. H. Newman and G. W. C. Kaye.
1.See also W. V. Meyeren, Zeits. f. Physik 84, 531 (1933);
1.W. V. Meyeren, 91, 727 (1934).
2.The Hg is thought of as going on the wall as ions, and coming off probably at least in large part as neutral atoms (see below).
3.R. C. Mason, J. App. Phys. 9, 131 (1938), has recently described the clean‐up of Hg ions in an ionization gauge. His observations agree well with those here reported. (His reference 4, to my work, was meant to apply to the present paper.)
4.S. Dushman and Clifton G. Found, Phys. Rev. 17, 7 (1921).
4.(a) Kenty, J. App. Phys. 9, 705 (1938).
5.Because of lag in the gauge, the true pressure drop was probably greater than indicated.
6.L. Hamburger, Proc. Amsterdam Soc. 20, 1043 (1917).
7.F. Skaupy, Verh. d. D. Phys. Ges. 19, 264 (1917).
7.For additional work on the subject see A. Rüttenauer, Zeits. f. Physik 10, 269 (1922).
8.I. Langmuir, J. Frank. Inst. 196, 75 (1923).
9.Druyvesteyn and Warmholtz, Phil. Mag. 17, 1 (1934), have observed the effect in a Na vapor discharge at pressures of the same order as here used. They refer to such phenomena as electrophoretic effects.
10.J. Stark, Boltzmann Festschrift, p. 399 (1904).
11.A. Wehnelt and J. Franck, Verh. d. D. Phys. Ges. 12, 444 (1910).
12.I. Langmuir and H. M. Mott‐Smith, Gen. Elec. Rev. 27, 762, 770 (1924).
13.L. Tonks, Trans. Electrochem. Soc. 72, 167 (1937), has given an admirable discussion of various causes which may operate in low pressure gas discharge apparatus to produce surges.
14.The value to which the current could be jumped from 20 ma without putting the arc out, varied by a factor of 2 or more on different occasions (from 0.5 to 1 amp. at ). This suggests that the behavior may depend on the initial state of saturation of the walls with Hg (see below).
15.I. Langmuir and H. M. Mott‐Smith, Gen. Elec. Rev. 27, 762 (1924). Run 32, Table XIV.
16.Langmuir and Mott‐Smith, Gen. Elec. Rev. 27, 538 (1924), Run 36‐a, Table III. was calculated from and as given according to
17.L. Tonks and I. Langmuir, Phys. Rev. 34, 876 (1929).
18.T. J. Killian, Phys. Rev. 35, 1238 (1930).
19.The mean free path of a Hg atom at 1.1 bar and (say) 100 C as calculated from viscosity measurements (Loeb, Kinetic Theory of Gases) is about 4 cm. An individual atom moving in the gas with an energy of the order of 1 volt, corresponding in speed to a Hg ion under the radial field (reference 17) would have a mean free path greater than this, first by a factor of on account of its high speed relative to the other atoms and second by a factor of 4 to 8 because of the Sutherland correction (see Carl Kenty, J. App. Phys. 9, 53, 58 (1938) and references therein cited). Since little is known of the actual free path of an ion in its parent gas, it will be assumed, as is commonly done, that it is of the same order as that of a neutral atom. If so, an ion in the present case would only rarely collide with an atom on its way to the wall.
20.S. Dushman, High Vacuum (1922).
21.Langmuir (reference 8, p. 756) suggests statistical fluctuations of the positive ion space charge density in the sheath as a possible cause of the diffuseness of reflection. Other possible contributing factors which suggest themselves are (a) reflection of the electrons from the walls themselves (see reference 17) and (b) secondary electron emission from the waits if appreciable.
21.Killian’s results (reference 18, p. 1252) would, however, indicate the reflection to be more nearly specular.
22.The location, length and general behavior and appearance of these regions depend on the Hg pool temperature, the location of condensed Hg in the tube, if any, and on the state of the walls as regards saturation with cleaned up Hg. The last depends on the previous history of the tube: whether it was shut off when hot, how long it stood inactive, and whether any special heat treatments were performed, locally or otherwise. At low pool temperatures (0 °C) the region usually appears to extend over the whole length of the tube, the current at no time rising very high (see preceding paper). At ordinary temperatures the region would not be expected to appear near the cathode end because of proximity to the source of Hg vapor.
23.The first maximum in curve A, Fig. 4, indicated that the net amount of Hg cleaned up during the surge period due to over‐all conditions is greater than if no surges were present. This is not incompatible with the present suggestion because the speed of the ions during an actual surge may be of a higher order than that between surges, the latter favoring a large clean‐up. It was pointed out by Dr. Langmuir that the grazing nature of the ionic collisions with the walls during a surge may be an influencing factor. The maximum referred to is probably not due to Hg driven off the walls by rising wall temperature because curve A, Fig. 2, where no surges were present does not show such a maximum.
24.At the negative end evidently a negative pressure effect might be said to exist.
25.In an earlier tube this trap was inserted between the appendix and the arc tube. When the trap was closed and the current was suddenly raised from a low value to a high value the pressure decreased tremendously throughout the tube and there was no building up of pressure again as in Fig. 2; rather the pressure continued to decrease till the arc went out.
26.These currents were sufficiently small so as not to cause an appreciable positive pressure effect.
27.H. Alterthum and A. Lompe, Zeits. f. Tech. Physik 17, 407 (1936), observed a large clean‐up of Ne (and also He and A) in Fe cathodes in a glow discharge tube which they attributed to ions being driven into the metal.
28.If is the concentration of atoms in the gauge and n that in the tube it follows at these low pressures that where is the temperature of the gauge and T that of the tube. Now the rate at which atoms strike a surface (or a hole) will be proportional to and therefore to Also if the volume of the gauge is small compared with that of the tube, the amount of gas held on the right of the trap at any given gauge reading will be inversely proportional to therefore, for a given difference in gauge readings on the two sides of the trap, the rate of decrease of reading or “leak” as shown by the gauge reading on the right will be proportional to Measurements of the leak at 60° and at 365° confirmed this conclusion.
29.S. Tolansky, Proc. Phys. Soc. 42, 556 (1930).
30.The experiments of von Meyeren (reference 1) indicate that also bombardment by 800–900 volt electrons is able to release electrically cleaned up atoms of A and other gases.
Article metrics loading...
Full text loading...
Most read this month
Most cited this month