OPERATION MODES OF HOLLOW CATHODE

ARCS WITH REDUCED GAS FLOW

K. Groh, S.E. Walther

To cite this version:

K. Groh, S.E. Walther. OPERATION MODES OF HOLLOW CATHODE ARCS WITHREDUCED GAS FLOW. Journal de Physique Colloques, 1979, 40 (C7), pp.C7-277-C7-278.<10.1051/jphyscol:19797136>. <jpa-00219109>

HAL Id: jpa-00219109

https://hal.archives-ouvertes.fr/jpa-00219109

Submitted on 1 Jan 1979

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, estdestinee au depot et a la diffusion de documentsscientifiques de niveau recherche, publies ou non,emanant des etablissements d’enseignement et derecherche francais ou etrangers, des laboratoirespublics ou prives.

JOURNAL DE PHYSIQUE CoZloqw. C7, suppZdment au n07, Tome 40, JuiZZet 1979, page C7- 277

OPERATION MODES OF H O L L O W CATHODE ARCS W I T H REDUCED GAS F L O W

K.H. Groh and S.E. ~alther ' .

I s t Insti tute of Physics Giessen University, Heirrich-Buff-Ring 16, 0-6300 Giessen, F.R.G.

Abstract

In 1971 the first research work on hollow cathode arcs (abbreviated HCA) started at Giessen University. Emphasis has been put on the operation of this hollow cathode arc for the ion beam neutralization of the elec- tric propulsion unit RIT 10. In this ap- plication the HCA works as an electron source delivering electrons for main dis- charge starting and beam neutralization. But beside this, diagnostics have been per- formed dealing with the physical phenomena of HCA's. Different measurements including Langmuir probe analysis led to a simplified discharge model. The determination of the discharge current parts carried by the ca- thode and the electron emitter inside the cathode gave us hints to explain the ob- served operation modes, the plum mode, and the spot mode by assuming an electrical double sheath.

Introduction

Hollow cathodes have been developed and in- vestigated in the scope of the research and development program of the radio frequency ion thrusters (RIT) for the use as electron source (plasma bridge neutralizer) (1). The main aspect of these activities was me- chanical stability, reliable function, gas economy, and high efficiencies. The dis- charge mechanism has never been understood entirely and different theories have been published by various authors trying to ex- plain the observed phenomena more or less.

For our experiments we used a hollow cathode reducing gas flow by a small orifice in the cathode disc. One can observe two operation modes, the plume mode showing a large lu- minous plasma region outside the cathode orifice like a plume and the spot mode where the discharge is concentrated at the cathode orifice. Between plume and spot mode there exists an unstable operation, the so-calleti transition mode where the discharge oscil- lates between plume and spot mode. Different theories try to explain this dis' charge modes. Csiky and Groh (2,3) assume that the cathode space charge sheath pene- trates the cathode boring which results in the change from the plume to the spot mode. According to Philip's opinion (4) the change of the electron emission from the cathode disc to the cathode interior is responsible for the change of the operation modes. How- ever, our most recent results do not agree * Now with ERNO-Raumfahrttechnik, Bremen

with one of the mentioned theories but point to the presence of an electrical double sheath which can explain the observed plume and spot mode operation. In the following a short description of the experiments and their interpretation will be given.

Experimental Arrangement

For the measurements we used our standard- ized 3 mm hollow cathode being sketched in fig. 1. The hollow cathode consists of a. 3 mm molybdenum tube being enclosed for gas flow reduction purposes at the downstream end by an electron beam welded tboriated tungsten disc with a 0.3 mm diameter boring. In 2 mm distance the discharge anode is located with a centric hole of 1.5 mm dim- meter. At the upstream end of the hollow cathode tube the gas supply is connected, a mercury vaporizer or a gas inlet for inert gas operation. In the cathode interior just behind the ca- thode disc the so-called insert is located a low work function material supply acting as electron emitter. This insert is in- sulated electrically against the cathode body and is connected via a feed through to an outer power Supply as sketched in fig. 1, too. The discharge is stroke by heating up the cathode to about 1 1 0 0 ~ ~ until a sufficient thermionic electron emission is available at an applied voltage of 150 - 200 volts between cathode and anode. If a gas flow is added in the order of magnitude of 0.02 mg/sec mercury vapour discharge ignit- ion occurs and the discharge voltage dxops to about 20 volts. Now, the cathode heater can be switched off since the discharge sustains itsself heating the cathode by ion impingement. Due to the insulated insert the discharge can be operated either between cathode~and anode with floating insert or between anode' and insert with floating cathode showing the below described characteristics.

Experimental Results

Operating the HCA in this manner we found that both the cathode and the insert carry the discharge current as shown in fig. 2. The main part of the discharge current is drawn from the insert at low total currents. This behaviour changes at high discharge currents when the main part is takeh over by the cathode. This current distribution

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:19797136

depends on the pressure inside the hollow cathode which,has been investigated, too. The result is graphed in fig. 3 where the discharge current parts are shown versus the hollow cathode pressure. Obviously, the current from the cathode disc predominates at high gas pressures and vice versa the insert current at low pressures. That means looking at fig. 2 that the cross over pqint of both current graphs is shift- ed to lower discharge currents with in- creasing gas pressure.

Discussion

This behaviour of the HCA can be explained by the moving of the basic point of the dis charge. The basic point or active zone of the discharge penetrates the cathode's in- terior more or less dependend on the pres- sure or the mean free path, repectively. Thus, we interpretthe current sharing to ca thode and insert (5) . But as we see from fig. 3 plume and spot mode operation is possible between cathode and anode and between insert and anode. This disproves former theoretical models assuming cathode fall sheaths penetrating the cathode boring or the shift of the emission from an outer to an inner surface. Rather, we explain the observed discharge modes by an electric double sheath which appears at discontinous cross-section re- ductions, as is well known (6). As a double sheath we understand a small limited area of high electrical field strength within a plasma. Such sheaths are created if the diff.usion current A.j is smaller than the reauired current J,. fn order to deliver , U the needed current the charge carriers mist be accelerated which happens in a double sheath as sketched in fig_,4.-- In our case, the plume mode operation can be observed at low discharge pressures and/or at low discharge currents. Conse- quently, it follows that the drift current from the interior plasma into the cathode boring is too small and the electrons must be accelerated in a double sheath. So we identify the plume mode operation with the existence of a double sheath. The double sheath-voltage depends on the pressure and the discharge current and is in the order of magnitude of some volts.

With increasing gas pressure or increasing discharge current the diffusion current den sity j grows resulting in a higher drift curren? from the internal plasma. No longer the accelerating sheath is necessary which we identify with spot mode operation. From our Lansmuir probe measurements in the

References ( 1 ) S. Walther, K. Groh, H. Loeb, AIAA-

Paper No. 78-706, San Diego, 1978 (2) G. Csiky, NASA TN D-4966, 1969 (3) K. Groh, Doctorate Thesis, Giessen 1973 (4) D. Fearn, C. Philip, AIAA-Paper No.

72-416, Bethesda, 1973 (5) K. Groh, S. Walther, Proc. 13. ICPIG,197; (6) Th. Wasserrab, "Gaselektronik II", 1972 (7) K. Groh, S. Walther, H. Loeb, Proc.

C, T W W T i x r n r n n n ? 1 0 7 Q

ANODE HOLLOW CATHODE INSULATED INSERT

CL-

l-2 LJ LI D LI 1%

Fig. 1: Cross section of the hollow cathode -- - .

Fig.

g plume mode

- n I I I I I

0 0.2 01 0.6 0.8 1 DISCHARGE CURRENT, a

Discharge current distribution cathode JCa and insert JIns

external discharge-plasma ( 7) we have ob- tained plasma data which allow us a rough estimation of the above mentioned diffusion curren As uming a plasma density of n = 10 ".cm-' and an electron temperature of 10,000 K we achieve for plume mode ope- ration at 0.3 a discharge current a drift current in the order of magnitude 0.05-0.1 amp. In accordance with our model a double sheath is required to enable the discharge current. Fig

HOLLOW CATHODE PRESSURE, mb

Fig. 3: Discharge current distribution to cathode JCa and insert J and mass flow rate in versusl€Ee gas pressure

ril b

GAS FLOW

HOLLOW CATHODE

4: Sketch of the discountinous cross- section reduction with electrical double sheath