HAL Id: jpa-00229912
https://hal.archives-ouvertes.fr/jpa-00229912
Submitted on 1 Jan 1989
HAL
is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire
HAL, estdestinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
FACTORS INFLUENCING THE STABILITY OF COLD-CATHODES FORMED BY COATING A
PLANAR ELECTRODE WITH A METAL-INSULATOR COMPOSITE
S. Bajic, M. Mousa, R. Latham
To cite this version:
S. Bajic, M. Mousa, R. Latham. FACTORS INFLUENCING THE STABILITY OF COLD-CATHODES FORMED BY COATING A PLANAR ELECTRODE WITH A METAL- INSULATOR COMPOSITE. Journal de Physique Colloques, 1989, 50 (C8), pp.C8-79-C8-84.
�10.1051/jphyscol:1989814�. �jpa-00229912�
COLLOQUE DE PHYSIQUE
Colloque C8, suppl6ment au no 11, Tome 50, novembre 1989
FACTORS INFLUENCING THE STABILITY OF COLD-CATHODES FORMED BY COATING A PLANAR ELECTRODE WITH A METAL-INSULATOR COMPOSITE
S. BAJIC, M.S. MOUSA'') and R.V. LATHAM
Department of Electronic Engineering and Applied Physics, University of Aston, GB-Birmingham, B4 7 E T , Great-Britain
Abstract
An investigation has been undertaken into the stability of populations of electron emission sites on extended-area, composite-coated cathodes for applied fields of S 20 MVIm. These cathodes have been shown to give rise to total current instabilities (S) typically in the range 10 to 40 % for total emission currents of 2 100 PA. An optical imaging technique has revealed that the current instability is associated with temporal changes in the emission site distributions which appear to be influenced by the coating composition. Operation under increased pressure H2, N2, O2 and CO environments (P 2 lo4 mbar) has been shown to only slightly degrade the current stability. Furthermore, extended tests have revealed that emission sites on the composite cathodes are resistant to He-conditioning.
It has been previously demonstrated 111 that if a planar metal electrode is coated with a suspension of graphite particles encapsulated in a resin dielectric medium, it is possible to draw field-induced electron emission currents in excess of 5 mA per cm2 at macroscopic field levels of 20 MVIm. By using a simple optical imaging technique, it has been shown that the measured currents are derived from a random distribution of a few hundred point sources that tend to vary individually in intensity, or even switch on and off randomly with time;
for this reason, the total emitted current tends to be relatively unstable. More recently, it has been established that the emission characteristics of such cathodes, including their stability, is significantly influenced by the material nature of the suspended particles 121.
The aim of the present study was to evaluate the potential technological performance of this type of composite cathode under "commercial" vacuum conditions; here particular significance was given to the stability of the emitted current, since an earlier study 131 had indicated that a dielectric-coated cathode is electronically
"protected from the effects of the residual gas environment, and gives more stable emission characteristics. In this paper, a range of cathodes, composed of various composite coatings, have been studied dynamically under various residual gas environments ( He, N2, H2, O2 and CO ) , and within the pressure range to lo-*
mbar. In addition, an in siru specimen hot-stage has been used to investigate the effects of an initial low- temperature ( 200 - 250°C ) de-gassing bake-out procedure.
2. Experimental Svstem and Procedure
The current-voltage ( I-V ) characteristics, site distribution and stability data for the cathodes used in this study were obtained using the plane-parallel electrode module and its associated testing circuitry 141 shown in Figure 1.
Here, the electrode regime is seen to consist of a 14 mm- diameter, extended area test cathode, which is electrically isolated from the vacuum system via a 10 mm-thick pyrophillite insulating block, and is pre-set with a known separation (typically 0.5 mm) from a transparent tin- oxide coated anode. Under normal testing conditions, electron emission is obtained by applying a high positive potential (< 5 kV) to the anode via a current limiting resistor. Emission current measurements are made using a picoammeter connected between the test cathode and earth. In order to monitor the current stability of a particular cathode, chart recorder current-time (I-t) traces can be obtained from the output of the picoammeter
.
The versatility of this facility has been further improved by including an in situ cathode heating stage,which enables the temperature of the cathode to be elevated during operation to >250 O C . This temperature is monitored by an inserted thermocouple probe which is used in conjunction with an automatic cold-junction compensated digital thermometer."'Visiting Scientist from Mu'tah University, Al-Karak, JORDAN.
Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1989814
resiatlve neatkg
cuwent I l m l w camera
mmitw
u- ---
image I
processing
?
V.CR.
A u
DYrODhl O.lmm mlca spacer
Fig. 1- A schematic representation of the emission imaging module and experimental systems 141.
The electrode module, shown schematically in Figure 1, provides a means of directly imaging the spatial distribution of the constituent cathode emission sites. Under the influence of an applied field, electrons are emitted from sites on the test cathode, and are accelerated in the electrode gap to subsequently impinge on the Sn02 anode surface with typical energies of a few KeV. As a consequence, these high energy incident electrons induce excitations within the anode surface, and hence give rise to pin-points of bluish light that have been termed " anode spots " 151. Thus, the anode spot formation effectively " marks " the corresponding emission site distribution on the test cathode, and can be recorded using either conventional photographic or vedeo techniques.
In order to study the effects of the residual gas invironment on cathode operation, an auxiliary UHV gas handling facility is interfaced to the main chamber for the introduction of any commercially-available gas species to a chosen pressure ( typically in the range to 10'~ mbar ).
The composite cathodes used in this study, were produced by spinning and subsequently curing a single-layer, resin/conducting particle suspension onto planar 14 mm diameter Cu-substrates. This technique, which has been described in detail elsewhere /1,6/ , has been used to produce seven cathode types composed of geometrically irregular C, MoS2, Au, Si, Sic, S and Fe particles of typical densities 200
-
2000 / cm2 and particle dimensions of 50-
250 pm.GAP FIELD ( M V d )
-
IOS/&"Fig.2- A comparison of the reversible I-V characteristics pig.3- example of the of a typical of (a) a typical C-com~osite coated cathode, and (b) an C-composite cathode operating at a total current level uncoated Cu-cathode. of (a) i0 PA, (b) 100
i~
and@) 1 rnA.(a) t=O (b) t=3 min (c) t=6 min (d) t=9 min
(a) t=O (b) t=3 min (c) t=6 min (d) t=9 min
Fig.4- A time sequence of constant-field site images illustrating the spatial stability of (i) a typical C-composite cathode (ii) a typical Si-composite cathode. E=10 MV/m and Ie=0.5 m4.
Recent studies /1,2,6/ have shown that these cathodes give rise to a localised "non-metallic" emission phenomenon at anomolously low-fields (<lOMV/m), and exhibit an irreversible " switch-on " behaviour followed by a low field reversible I-V characteristic. In order to illustrate the extremely high emissivity of these cathodes, Figure 2 compares the reversible current-voltage (I-V) characteristics typically obtained from (a) a C- composite emitter and (b) an uncoated Cu-electrode. By monitoring the total emission current under constant field conditions, at various points on the I-V characteristic, it is found that the composite cathodes exhibit a significant current instability, i.e. such as illustrated in Figure 3. This shows the temporal fluctuations in the emission current for a typical C-composite cathode operating at a chamber pressure of 10
-*
mbar and mean current levels (Ie) of (a) A, lo4 A, and (c) A.Refemng to Figure 3(a), it is conventional to define the current instability, S, as
Thus it will be seen in Figure 3 that the stability of a C-composite emitter typically improves from
-
50 to -2 %for increasing current levels in the range to 10-3 A. Table 1 compares the typical values of S obtained at I,
= A for the seven composite cathode types used in this study, and indicates that the C, Si and S i c composites generally give rise to the most stable constant-field emission characteristics, i.e. S
-
10-
15 %.o ! . l . l . l ~ l . ~ m l
0.0 0.2 0.4 0.6 0.8 1 .O 1.2
Emission Current I, (mA)
o2
Q CO
E H,or N,
0 UHV
Fig.5- The influence of 02, CO, N2 and Hz- environments on the stability of a typical C-composite cathode operating at various current levels. ~=10-' mbar.
In order to investigate the origin of the
UHV
current instability, the transparent-anode imaging technique has been used to monitor the populations of emission sites on the composite cathodes under constant-field conditions. Thus, figure 4(i) shows a time sequence of coristant-field site distribution images for a typical C- composite cathode, and clearly demonstrates a spatial site instability for 1, = 0.5 mA and E = 9 MVIm. It will be seen that, of the sixty or so sites identified in Figure 4(i) (a)-
(d), only 60-
70 % of these sites are in an emitting"on" state at any moment in time. In addition to the site switching instabilities, it is also evident that many sites are subject to random variations in intensity ( i.e. site current ) with time. In contrast, Figure 4(ii) (a)
-
(d)demonstrates how the Si-composites t'ypically give rise to the most stable populations of emission sites under identical experimental conditions. Here, it is found that
-
85 % of the total site population remains in an "on"state at any frame of reference over the 9 minute testing period. At this point, it should be noted that the comparable values of S for the C and Si composites may arise due to an " averaging effect " associated with the greater number of sites on the C cathode.
Table 1
-
Collated data of typical instabilities, S, atI,
= lo4 A for each cathode type.The cathode stability measurements described above have been extended to investigate the influence of H2, N2, CO and O2 environments at " poor-vacuum " pressures of
-
mbar. Thus, Figure 5 shows how the.underlying influence of each particular gas environment varies significantly, where it is observed that O2 and CO give rise to the largest increases in S, i.e. a degredation in performance. Whilst this represents only a small deviation from the normal UHV performance, it is significant to note that the composite cathodes appear to be' unaffected during operation under poor-vacuum H2 and N2 environments. In addition to these tests, a series of
Composite Cathode
s
at I, = lo4 A (%)He-conditioning experiments were conducted at pressures of -lo4 mbar and applied voltages of 5
-
10 KV( obtained at increased gap spacings). Under con~tant~field conditions, one would typically observe a steady decrease in the total emission current of an uncoated cathode /4,71 by approximately one order of magnitude, over a period
-
20 rnins. However, tests with the composite-coated cathodes have revealed that no such conditioning occurs; i.e. the composite emitters appear to be "protected", or resistant to operation under these conditions.MoS2 49 C
15
Finally, in an attempt to improve the current stability of the composite-coated cathodes, an in-situ cathode hot- stage has been used to subject the cathodes to a 4 hour thermal pre-treatment at 2 0 0 ' ~ as described in detail elsewhere /3,6/. However, in contrast to the improvements in stability typically observed with pre-heated dielectric-coated microtips /3,8-101, it was found that this process invariably led to a de-stabilisation of the emission current and spatial distribution of sites. In quantitative terms, heat treatment was generally found to degrade the cathode stability by a factor of 2 , e.g. from S = 32 % to S = 60 % for I, = A.
4.
Discussion
Au 46
The composite cathodes described in this paper have been shown to exhibit I-V characteristics which are noteable for their extremely low initial switch-on field (-1.5 MV/m) and a subseq~lent reversible behaviour that shows a rapid increase in current,with applied field, eventually delivering current densities of -1 rnNcm2 at fields of
-
10 MVIm.It is believed that the electron emission from these cathodes is associated with a metal-insulator-metal-insulator- vacuum (M-I-M-I-V) regime involving the type of structure shown schematically in figure 6. Initially, it is assumed that the resin forms a "blocking" contact between the conducting particle and the metal substrate in region 1. However, as the applied field is increased across the electrode gap, the antenna effect of the particle 11 11 results in an enhanced voltage drop, and hence field, across the dielectric layer in region 1. The magnitude of this field enhancement will be approximately = Ns, where h is the maximum height of the particle above the substrate, and s is the thickness of the dielectric layer between the particle and the substrate Ill. When the applied field reaches a certain critical value, a "primary" M-I-M switch-on event occurs in this region. This will lead to a current avalanche between the substrate and the conducting particle, i.e dielectric breakdown, and, as a
Si 15
Sic 12
S 22
Fe 23
M&aI Substrate
Fig.6
-
A shematic representation of field-induced electron emission via a two-stage switch-on mechanism associated with a M-I-M-I-V microregime.result, the potential of the particle falls rapidly (<I ns) to the substrate potential
.
In consequence, there will be a sudden enhancement in the surface field above the conducting particle, particularly at any protruding sub-feature such as in region 2. The magnitude of such a surface field enhancement factor,P,
can be estimated by considering a corresponding "conducting spike" of height h and tip radius r shown dotted in figure 6, and will be approximated by P=Wr 1121. Hence, a "secondary" M-I-V switch-on process occurs due to the enhanced field appearing across the dielectric fim surface in region 2, as described by Latham 11 31 andBayliss and Latharn /14/. This produces electron emission by a field-induced hot-electron emission (FIHEE) mechanism from an electmfomed conducting channel within the dielectric layer at the M-I-V region.To explain the inherent instability of this two-stage emission mechanism, it is necessary to consider the factors likely to influence its operation. In the above two-stage "antenna" model, it is assumed that the switch-on process occurs following the charging of an isolated conducting particle so that its potential becomes equal to the substrate potential. However, although evidence exists in support of a dielectric breakdown mechanism between some particles and the substrate 161, it is not yet clear whether the charging of the particle is a "transient"
or "permanent" state of the emission process. In fact, an alternative mechanism may exist in which the potential of the particle is maintained at an equilibrium value, i.e. the net flow of charge into the particle is equal to the rate of electrons emitted into vacuum. Such a mechanism would imply that a finite resistance region exists between the substrate and the particle (region 1 in figure 6). which could support a steady-state M-I-M conduction mechanism; a process that could result from the formation of either "electroformed" conducting filaments /15,16/, or metallic filaments due to an electromigration process 117-201 following the intial M-I-M switch-on event.
In fact, it is likely that both the M-I-M and M-I-V emission processes could exist in a dynamic state, such that the potential fluctuates about a mean value as a result of a dynamic inbalance in the M-I-M conduction current,I,, and the emission current, I,. In order to estimate a likely frequency for this behaviour, it can be assumed that the emitted current per site, I,, effectively "drains" the supply of electrons from a charged particle of radius, r, i.e in a repetitive pulsed manner. By comparing the capacitance of an isolated sphere with I,, it follows that that the repetition rate, or "frequency", of this process can be approximated 161 by
f =
re
14 RTE,%V~ (2)where E, is the relative permittivity of the resin and Vp is the potential of the particle. Thus, assuming that I, = 5 pA, Er=4, r=100 pm and Vp= 5 V, equation 2 yields a frequency, f, of the order 10 MHz; a high frequency emission mechanism has previously been predicted by Halbritter 1211. It is also of interest to note that the inherent capacitance, and transient charging phenomenon, would be likely to influence the pulsed-field emission characteristics of a composite-coated cathode; in particular, by introducing a time delay between the 1,- response and pulsed-field waveform.
Finally, estimates of the rate of rise of the particle temperature ( dTIdt ) are only typically
-
1 Ws for a metallic particle of radius r =I00 p n 161. However$ may be possible that larger local temperature fluctuations occur in the vicinity of an emission site and give rise to a varying thermal-stimulation of the emission mechanism 1221, which would in turn give rise to current instabilties.At this stage, it is not clear how the above mechanisms, ( or others ) are influenced by the material composition of the emission regime. Nor indeed is the role of the particle size and geometry fully understood. However, from practical point of view, it appears that the Si composite has a performance which exhibits a significantly.
better stability than most of other structures studied. Further investigation is therefore warranted into their possible use as a stable cold cathode electron source. In common with all composite cathodes, this regime also has the important practical advantage of being relatively insensitive to the residual gas pressure for P I mbar; i.e. sensitive M-I-M region is protected. They do however appear to have the practical disadvantage of being degraded by long-term thermal cycling.
Field-induced electron emission (FIEE) has been observed from seven composite-coated cathode types under various experimental conditions. These emission tests have revealed that :-
( 9 (ii) (iii)
for total emission currents of 5 lo4 A, the current instability is limited to 2 10 %,
the inherent instability is associated with random "on / off switching" of the individual sites of a population the current stability is only slightly degraded during operation under"poor-vacuum" H2, N2, CO and O2 environments
(iv) a thermal pre-treatment has the undesirable effect of degrading the current stability, and
(v) the composite-coated cathodes appear to be "resistant" to the wen- known He-conditioning process.
These observations have been discussed in terms of a "non-metallic", two-stage FIEE mechanism associated with metal-insulator-metal-insulator composite microsrmctures.
Acknowledgement
The work reported in this paper forms part of a programme sponsored by SDIO I IST and managed by the Space Power Institute, Auburn University, Alabama, USA.
Relerences
I1
I
Bajic, S. and Latham, R.V., J. Phys. D : Appl. Phys.2
(1988) 200.I2
1
Bajic, S. and Latham, R.V., Proc. 2nd Int. Conf. on Vac. Microelectronics Bath UK (1989).I3 1 Latham, R.V. and Mousa M.S., J. Phys. D : Appl. Phys.
19
(1986) 699.I4
I
Bajic, S. and Latham, R.V., J. Phys. D : Appl. Phys.2
(1988) 943.I5 / Young, R. W., Vacuum 24 (1973) 167.
I6 / Bajic, S., PhD Thesis Univ. of Aston Birmingham (1989).
/7 / Bajic, S., Abbot, A. and Latham, R.V., Proc. Int. Symp. on Discharges and Elec. Insul. in Vac, (1988) 8.
181 Mousa, M.S. andLatham,R.V., J. dePhys.
42
C7 (1986) 139.191 Mousa, MS., J. dePhys.
48
C6 (1987) 109./lo/ Latham, R.V. and Salim, M.A., J. Phys. E : Sci. Instrum. 20 (1987) 181.
11 11 Athwal, C.S., Bayliss, K., Calder, R. and Latham R.V., IEEE Trans. Plasma Sci. PS-13 (1983, 225.
1121 Latham, R.V., High Voltage Vacuum Insulation : The Physical Basis (New York : Academic) (1981).
1131 Latham, R.V., Vacuum 22 (1982) 137.
1141 Bayliss, K.H. and Latham, R.V., Proc. R. Soc. A 403 (1986) 285.
1151 DearnaIey, G., Stoneham, A.M. and Morgan, D.V., Rep. Prog. Phys.
3
(1970) 1129.1161 Adler, D,, Henish, H.K. and Mott, N.F., Rev. Mod. Phys. (1978) 209.
1171 Richards, B.P. and Footner, P.K., GEC J. Res. 2 (1984) 157.
1181 Couch, N.R., Movaghar, B. and Girling, I.R., Solid State Commun.
a
(1986) 7.1191 Couch, N.R., Montgomery, C.M. and Jones, R., Thin Solid Films (1986) 173.
1201 Cade, N.A., Cross, G, Lee, R.A., Bajic, S. and Latham, R.V.,J. Phys. D : Appl. Phys
.a
(1988) 148.1211 Halbritter, J., Appl. Phys.
A39
(1986) 49.1221 Xu, N.S. and Latham, R.V., J. de Phys. C2 (1986) 73.
