AKCIXG OF ELECTIUCAL CONTACTS IN TKLErilONlO SWITCHING CIKCUITS 557 



increases with surface cleaning.* For clean Pd contacts in atmospheric 

 air an average limiting current density of 30 amps/cm , or about G times 

 [ho normal glow current density, was obtauied. This sudden transition 

 fiom the low current density glow to the very high current density arc 

 represents a high rate of change in the emission process. With con- 

 taminated contacts, this is probably due to the presence of low work 

 function high emission spots on the cathode. These spots may be elimi- 

 nated by proper cleaning thus allowing glow discharge to be maintained 

 at higher current densities. The observed glow-arc transitions for clean 

 contacts, consistently occurring at about 30 amps/cm for Pd, may still 

 be attributed to the formation of a surface film on the cathode through a 

 cathode-atmosphere reaction. f 



JMeasurcments have also indicated that under certain conditions, 

 glow discharge cannot be obtained even at currents much below the 

 limiting currents discussed above. It appears that there is a limiting rate 

 of rise of ciu'rcnt with time above which glow discharge cannot be main- 

 tained. In Table IV, column 6, the initial rates of current rise are given. 

 In all cases where the rate of current rise was greater than about 3 X 10 

 amps/sec, lines 1, 2, 3 and 7, no glow was obtained. The experiment 

 was repeated with two other cathode diameters of 0.2 and 0.05 cm. The 

 limiting rates of rise obtained were approximately the same as given 

 above, indicating that the limiting rate of current rise is independent of 

 the cathode area. This seems reasonable since at the beginning of the 

 transient the currents are very small and the emission area is only a 

 ver}' small fraction of the cathode area. No detailed explanation, how- 

 e\-er, can be furnished at this time as to why such a limit of the rate of 

 current rise does exist. It is obvious, nevertheless, that while the rate of 

 current rise can be increased without limit by manipulating the ciivuit 

 parameters, the conduction mechanism in the contact gap, will, in gen- 

 eral, have its own limitations as determined by the emission processes 

 involved. 



ACKNOWLEDGMENT 



I am indebted to ]\Iiss R. E. Cox for assistance with many of the ex- 

 periments and calculations reported here. 



* With a contaminatod cathode surfaco a transition into an arc may (jccur (hir- 

 ing the noiDial gh)\v ]j(Mi()(l woU Ix'forc the curi'cnt is high enough to allow normal 

 glow to cover the entire cathode surface. This is particularly true with larger 

 cathod(; areas which are usually hai'd Xa clean satisfactorily 1)\' the above pro- 

 cedure. 



t A recent unpublished study !)>■ V . K. Haworth has shown that in the absence 

 of the usual surface contaminants, glow discharge is capable of activating pal- 

 ladium and silver contacts through the formation of surface films. These surface 

 reactions appear to be strongly clej)endent on the atmosphere. 



