Gehrcke 1906
E. Gehrcke and 0. Reichenheim, Verhandl. Dtsch. Phys. Ges.8,559 (1906)
Anode Rays
E. Gehrcke and O. Beichenheim.
Communication from the Physical and Technical Institute.
(Presented at the meeting of October 19, 1906.)
(See above p. 544.)
§ 1.
While the cathode of Geissler tubes, as is well known, becomes the source of a special type of radiation, the cathode rays, when the gas is suitably diluted, the anode behaves completely differently; its shape and position in the discharge space generally have an extremely small, if not negligible, influence on the phenomena of current passage through the tube. Nevertheless, a parallel exists between the anode and the cathode; one need only recall here the anode glow and the potential jump, which is briefly referred to as the anode drop. The latter fact, the anode drop, in particular, does not dismiss the assumption that, under suitable circumstances, the anode may also be capable of becoming the seat of radiation and emitting positively charged ions. We have set ourselves the task of discovering experimental conditions under which this is the case.
The canal rays discovered by Goldstein, which according to W. Wien are positively charged, are known to be radiation originating near the cathode or on the cathode surface itself.559-1 The previously frequently expressed view that the canal rays originate from the anode is incorrect according to more recent investigations.
§ 2.
The starting point of our observations was the above-mentioned anodic glow light, which one of us had already investigated.559-2 We have now succeeded in obtaining significant results using suitably designed discharge tubes.
1 ) E. GEHRCKE, Verh. D. Phys. G81!. 3, 63-70, 1905.
To achieve the desired magnitude of the anodic glow. However, we will not go into these experiments in detail here , especially since in none of these cases did we have reason to assume that any radiation originating from the anode was present in any noticeable amount. Rather, all the phenomena we observed can be explained by the hypothesis of a gaseous cathode 1) formed secondarily in front of the anode.
We therefore abandoned the arrangements we had used and proceeded with the following experiments.
§ 3.
In a tube whose cathode was an electrically annealed platinum sheet coated with barium oxide according to Wehnelt's method , 2) , there was a platinum wire about 3 cm long and 0.3 mm thick as the anode. To our surprise , immediately after applying the voltage ( 110 volts ) , we observed sharp , yellowish rays , the point of origin of which was a small , bright spot on the anode. The rays were initially quite intense, but quickly faded and disappeared after a few seconds. The phenomenon did not recur at first , even when the current was increased to such an extent that the entire anode heated to a bright yellow glow and melted.
In the further course of the investigation of this phenomenon , it emerged that the active agent which had produced the observed yellowish rays was to be sought in traces of impurities found on the platinum anode. A very well-cleaned platinum anode never caused the phenomenon to recur, but when some salt, such as borax or table salt, was brought into contact with the anode , the yellowish rays returned , albeit with considerable intensity. Nevertheless, here too, the duration of the phenomenon was still counted in seconds; after about 30 seconds , the tube presented the usual appearance , without the anode being in any way distinctive.
§ 4.
We then proceeded to produce anodes containing larger quantities of molten salts. These anodes had the following shape (cf. Fig. 1a ): A rectangular
1 ) E. GBHRCKB, Ann. d. Phye. (4) 15, 528, 1904. 1 ) A. WBHNBLT, Ann. d. Phys. (4) 14, 425--468, 1904.
A piece of platinum sheet 0.01 mm thick was folded in the middle parallel to the long edges along a line AB and then a cylindrical bulge CD of the sheet was produced on the folded sheet in the middle by pressing it over a small metal tube of about 2 mm thickness, in such a way that the folded sheet and ( cf. Fig. 1 b ) hard - soldered at the ends A'A ' ' and B ' B " to thick copper leads carried a small electrically heatable tube C' D' of 2 mm width in the middle , which was open at the top at C' and closed at the bottom at D'.
The sheet was also provided, as can be seen in Fig. 1 b , with folds parallel to C' D' ; the area A' B' BA" was approximately 2 x 0.4 cm. - This electrode ,
which could be filled with salts , then served as the anode; the heating current for the anode was supplied by an accumulator battery , which was insulated from the battery used to heat the WEHNELT cathode.
a..
Fig. 1.
If such an anode, whose cylindrical b.
tube C' D' was filled with sodium carbonate , Ä
was heated to a dark red glow , an intensely luminous yellow torch , sometimes spherical , sometimes elongated , emanated from the opening
(J,) , extending to the walls of the (spherical) discharge vessel. The cathode remained
,-th..e·
II 1 1 s· D'
enveloped in blue light , and each electrode thus appeared to be the starting point and center of a radiant , brilliant light phenomenon. The spectrum of the anodic light phenomenon contained intense D lines.
After a few minutes, the anode phenomenon became paler and finally disappeared completely; one then had the usual view of the tube and it was mainly the blue rays emanating from the cathode that were conspicuous to the observer , while at the anode only a bluish-white colored glow , mostly bouncing back and forth, was visible.
Almost the same phenomenon as with sodium carbonate as anode material also occurred with NaCl , but in the latter case, the sodium lines were also present in the spectrum of the blue rays coming from the cathode. Thallium chloride produced a splendid green anode flare ; in the latter spectrum , only the very bright green thallium line was observed.
- Also investigated so far have been LiCl , Li, CO8 , KCl , K,CO8, RbCl, CsCl, CuCl,, BaC11, SrC12, InC13 • All these salts gave intense, characteristically colored anode flares with the spectra of the corresponding metals; the lines of the spectra are very sharp, but otherwise seem to agree with those obtained in Bunsen flames.
The alkaline earth oxides , which according to Wehnelt (1. c. ) , mediate the emission of negative electrons from the cathode , were ineffective. Other oxides, such as Al2O8 and CuO, used as anodes , also did not exhibit the above phenomena , at least not at the anode temperatures we employed. It therefore appears that strongly dissociated or evaporating salts are primarily capable of this.
By applying sufficiently strong currents , the auxiliary current heating the anode could be switched off. The anode was then heated by the heat from the current flowing through the tube itself.
§ 5.
A so-called Faraday cylinder was placed at a distance of 2 cm from an anode constructed according to Fig. 1 , which contained sodium carbonate. The outer sheath of the cylinder was connected directly to earth , and the inner sheath was connected to earth via a galvanometer (sensitivity 10-8 Amp. ). The heating currents for the cathode and anode were supplied by two batteries that were isolated from each other. When current passed through the tube, a strong positive deflection of 100 S. -T. and more was obtained in the galvanometer, as long as a yellow flare emanated from the anode. As the flare faded , the galvanometer deflection also decreased, reaching zero and even becoming negative; this latter effect can probably be explained by cathode rays passing from the cathode into the interior of the Faraday cylinder.
The current flowing through the tube fell at the last Try from about 0.85 to 0.2 amps. The same result as above was also obtained with TlCl1 and CdCl1 as anode filling ; the galvanometer a1188Ch strokes were smaller , but also longer lasting. The Faraday cylinder, which was attached to a spring-loaded rod , could be placed opposite either the anode or the cathode. In the first case , as mentioned, positive deflections were obtained. However, as the cylinder was increasingly turned toward the cathode , the deflections decreased, passed through zero, and finally became strongly negative when the cylinder was opposite the cathode.
The experiments were repeated with different tubes and varying the distance of the Faraday cylinder from the electrodes with qualitatively the same success.
§ 6.
From the observations described we conclude that rays carrying positive charges emanate from a hot salt anode, which we shall briefly call anode rays.
If the anode rays present a phenomenon parallel to the cathode rays , they must also be magnetically deflectable. Indeed , we observed that the anodic light flare was deflected by an electromagnet in the sense of positive particles emanating from the anode. However, we do not wish to attach any significance to this experiment , since the entire light phenomenon in the tube was so strongly altered by the magnetic field that the observed strong magnetic deflectability was presumably due to secondary effects.
The velocity of the luminous particles, provided that it is sufficiently large, can be determined more reliably by Doppler's principle , since , analogous to the line shifts in the canal radiation spectrum discovered by 8TARK 1) , such shifts should also occur in the spectrum of the anode rays. However, given the small velocities to be expected here, one can probably only determine this with the aid of strongly
1 ) J. STARK, Pbyÿ. zs. 6, 892-897, 1905.
resolving spectral devices can perceive the DOPPLER effect.
us.
§ 7.
The anode rays described have been observed several times before
First and foremost , we would like to point out that E. Warburg 1 ) had
already carried out an experiment in 1890 that bore a great deal of resemblance to those described in § 3 and 4 and was essentially identical. We quote the relevant passage from Warburg 's treatise below :
=="§ 5.== Associated with the evolution of sodium at a glass anode is a phenomenon that is not without interest for glow discharge and is most reliably observed in hydrogen gas of somewhat higher pressure ( from 6 mm upwards). One notices a broad, yellow halo of sodium vapor around the negative glow , illuminated by the cathode rays. This halo disappears immediately when the platinum wire A is used as the anode , thus no new sodium is evolved , and immediately reappears when the current is allowed to enter the glass ; it therefore undoubtedly originates from sodium particles that , released at the anode , are carried along in the direction of the positive current. "
There are also numerous other references scattered throughout the literature that are closely related to the phenomena we have described. We would like to point to Warburg 's 2 observations on the tip discharge, further to Riecke and Stark's 3 experiments on the arc in air, and finally to a report by Wiedemann and Wehnkelt 4 on the vacuum arc at metal anodes. Wiedemann and Wehnkelt have also already pointed out the spectroscopic significance of the spectra produced by the vapors of their anodes in a vacuum :
Of interest to us are the effects of salt anodes
1 ) E. WARBURG, Wied. Ann. 40, 1-17, 1890. ') E. WARBURG, Ann. d. Phys. (4) 2, 299-302, 1900. ") E. RIECKE and J. STABK, Phys. ZS. ö, 537-538, 1904-. 4 ) E. W1EDBMANN und A. WBHNBLT, Phys. ZS. 6, 690, 1905.
Anode rays also for the interpretation of a number of sub Richardson's investigations 1). Richardson found , among other things, that a freshly annealed platinum wire emits positive ions in a vacuum , but loses this property over time and does not regain it on its own. With suitable treatment, such as irradiation with a secondary, luminous discharge, the platinum wire regains its lost emissivity. Richardson expresses the opinion that a foreign substance on the platinum may be at work here. We would like to endorse this opinion and express the suspicion that in many of Richardson's experiments , slight contamination of the platinum anodes with salts, primarily NaCl, had occurred. Of course, we do not wish to deny that pure platinum itself does not have the ability to emit positive ions. Other metals may also possess this property at sufficiently high temperatures. - Similar to salts , occluded gases such as hydrogen and hydrocarbons also appear to produce anode rays. We occasionally observed intense, positively charged anode rays that displayed the spectrum of the very sharp carbon bands intensely and otherwise exhibited only weak D lines. The nature of the gas in the discharge tube therefore appears to have some influence on the anode rays ; this is also the case with canal rays. Incidentally, Richardson has also already provided evidence for this.
§ 8.
Finally , we note that even a Wehnelt cathode containing some chloride salt , especially at very high temperatures , occasionally emits positive particles. For we occasionally obtained positive charges from a Faraday cylinder placed opposite the cathode at the first moment of current passage. In this case, the negative glow also exhibited metallic lines in the spectrum.
Furthermore, this is probably related to the fact that we occasionally observe the extinction of the WEHNELT tube when increasing
') OW Ri:cHARDBON , Phys. ZS. 5, 6--1 1, 1904; 6, 914-915 Cambr. Phil. Soc. 13, ,1905 192, 1906..
the temperature of the cathode; we would like to explain this by the fact that the positive ions emanating from the cathode at the high temperature create a gas envelope of high resistance, similar to CR00KES's dark space. - We have not observed a phenomenon that would indicate an emission of negative particles from the anode. ,
Charlottenburg, Physical-Technical Reich Institute, October 1906.
