DE2345096A1 - METHOD OF MANUFACTURING A FIELD EMISSION CATODE - Google Patents

Description

- 519

Linfield Research Institute, Mc Minnville, Oregon, U.S.A.

Method of making a field emission code

The invention relates to "a method for producing a Field emission cathode made of a metal wire and for reshaping the same after operation.

The need for electron sources that are capable of a finely focused electron beam with a stable and generating high current density is constantly growing as such electron beam applications become more widespread and refined such as scanning electron microscopes, electron scanning devices, Devices for the production of masks for integrated circuits, computer storage devices with Electron beam access or the like. Accordingly, have field emission cathodes due to their high current densities, which are much greater than those of thermionic cathodes or Hot cathodes achievable current densities, found an increased interest. Furthermore, in the case of a field emission cathode, the apparent

A098U / 0867

bare electron source can be much smaller, so that it is possible to use the electron beam accordingly focus smaller point without consuming additional Devices for focusing or zooming are required. So far, however, the field emission cathodes found no practical application, since they are quite unstable and the service life of the cathodes under practical vacuum- conditions is very limited. Furthermore, under practical vacuum conditions, an undesirably high level of flicker occurs. noise when field emission cathodes are at room temperature operate.

One therefore has the applicability of heated field emission cathodes to increase the cathode life with less drastic ones Vacuura values were considered, theoretically increasing the current density and other characteristics of the electron beam should be improved. The expected improvements are based on the thermal field emission cathode on the possibility of the surface tension with the mechanical tension caused by the electric field ddrart to equilibrate so that the cathode tip is deformed into a better shape for emission, i.e. to a very sharp point. However, it was found that in the manufacture and use of such a thermal Field emission cathode certain undesirable effects occur. A significant emission can only be achieved if

4098 U / 0867

when the electric field is above the equilibrium value addition is increased.

However, if this value is exceeded, it changes the geometry of the emitter tip in a disordered manner the time, which results in current fluctuations.

Changes in the emitter geometry also lead to a drastic one Change in the spatial emission distribution, which is intolerable from the point of view of fine focusing. Thus, the thermal field emission cathodes have found no practical application, although theoretically they are numerous Would offer advantages. It was generally believed that a final shape of the cathode tip exhibiting a high current density and allows a narrow electron beam angle not to be in an equilibrium state preventing further geometric changes can be brought. There was also a need for extremely low pressures to work, making it cumbersome to use and requiring expensive peripheral equipment.

Nevertheless, occur during an operation of a thermal Field emission cathode has two additional major advantages in conjunction with the aforementioned advantages, and these Knowledge has given rise to the thermal field emission further investigate cathodes as electron beam sources. First of all, it is possible to remove residual gases from the emitter surface cause undesirable current reductions, as well as undesirable

40981 4/0867

Just out au eliminate. This is done through thermal desorption. Second, it is possible to prevent microscopic surface roughness due to ion bombardment, which causes instability of the surface density and spatial distribution, since the micro-roughness due to the surface tension of the heated cathode is immediately smoothed again.

It is therefore an object of the invention to provide a method of production a field emission cathode and to reshape the same after operation to create which in stable operation due to a small tip radius allows a high current density, a stable geometric configuration and a has an extremely long service life in a moderate vacuum and generates an extremely finely focusable electron beam.

This object is achieved according to the invention in that one under vacuum

(a) a si oo) -oriented metal wire to a temperature heated, during which impurities are removed from the wire, and that one

(b) applying an electrical potential to the wire so that a minimum total current is used, and that

(c) The current building the £ 100 / - levels to a fine one Peak is maintained until current stabilization occurs,

wherein steps (a) and (b) can be carried out simultaneously or in reverse order.

4098U / 0867

According to the invention, a successful / ioo7 panel structure is used for the first time reached the cathode tip. Tips with an extremely small radius and an extremely stable configuration are achieved. The lifespan is extremely long, even if you use less drastic vacuum values compared to conventional ones Procedure, works. The electron beam formed by the cathode is finely focused and has an extremely high current density, and in particular, this electron beam can be controlled by simple electrostatic or magnetic lens adjustments be focused to a fine point. The cathode is made of a material with a body-centered cubic crystal structure, such as tungsten or molybdenum or the like in the axial direction a <"1oo)> - orientation (Miller indices) is present. The mentioned properties will be reliable and reproducibly achieved at elevated temperature. Either extreme temperatures or extreme vacuum values are required, see above that the process can be carried out economically. You can work with different direct currents or pulsed currents.

In the following the invention is explained in more detail with reference to drawings. Show it

Figure 1 is a general flow diagram of the inventive Procedure;

Figure 2 with a space-centered cubic crystal Miller, indices;

A098U / 0867

FIG. 3 shows a greatly enlarged view of an emitter tip before carrying out the method according to the invention;

Figure k shows an entire hairpin cathode structure with the
Field emission tip;

FIG. 5 shows a cathode tip according to FIG. 3 after heating;

FIG. 6 shows the cathode tip according to FIG. 5 according to the invention outside field;

FIG. 7 shows part of the cathode tip according to FIG. 6,
which is above level 7-7, in stylized form;

FIG. 0 shows a schematic representation of a scanning electron microscope for carrying out the method according to the invention;

FIG. 9 shows a detailed flow diagram for illustration of the process according to the invention.

Thermal field emission cathodes are made from a single crystal wire with <1 oo) orientation. This means that the crystallographic alignment of the cube surface of the single crystal unit cell is perpendicular to the axis of the wire. The ¹ Ci oo ^ -oriented wires have spoken more precisely according to.
the Miller index designation the levels (loo), (o1o) or (ooi) in the direction of the wire length. Such - ^ 1 oo) -oriented wires made of tungsten or similar metal can be produced by zone melting processes.

409814/0867

2345098

For a full understanding of the invention, a certain knowledge of the Miller index designation is required, since this method of drawing is used to characterize body-centered cubic crystal structures such as tungsten crystals or molybdenum crystals. For this purpose, the body-centered cubic structure of FIG. 2 is to be explained. The unit cell 1 comprises nine atoms, eight atoms (3, 4, 5 ȟ, 9> lo, 11, 12) are located at the corners of the cube and the ninth Atom (2) is in the center of the cube. To derive the Miller indices, a line is drawn between two vertically aligned atoms and designated as the Y-axis. Furthermore, a line is drawn between two horizontally oriented atoms and is designated as the X-axis and finally a line is drawn between two horizontally oriented atoms, which runs at right angles to the X-axis and the Y-axis and is designated as the Z-axis. The three axes intersect, for example, in your cube according to Figure 2 in atom k.

Planes that run through the cubic structure, can be identified by three-digit numbers. The first number 1 denotes the position of the cube where the Plane cuts through the X-axis. The second number indicates the point where the Y-axis is intersected and the third number indicates the point at which the Z-axis is intersected. Thus, according to FIG. 2, the (ioo) plane cuts through the x-axis at position 1 and includes four atoms.

40981 4/0867

2345098

") lche form a ..- exte of ii ^. 'jua, while both the Y-axis as well as the Z-axis are not intersected. Similarly, the plane (oio) only crosses the if axis and the plane (ooi) only the Z-axis. The name of this Levels is just a matter of nomenclature, so one can use the levels (loo), (o1o) and (ooi) as the family which can denote Jj oo] -planes.

Any plane that passes through unit cell one can be designated using the Miller indices by dividing the distance between two adjacent atoms along the three reference axes into increments. For example, the atom 3 is in the "(1)" position on the X axis. The X-axis also runs through the atom k. Between these two atoms intermediate positions "(2)" and "(3)" are provided at a distance from each other, namely at the point of half the distance to the "(1)" position and at the point which is a third of the distance to the " (1) "position. Similar designations are provided for the positions between atoms h and 5 on the Y-axis and between atoms k and 6 on the X-axis. Thus an (i12) -plane runs through the atom on the X-axis, through the atom 5 on the Y-axis and through the position "(2)" between atom k and atom 6 on the Z-axis. This plane (112) running through unit cell one is shown generally in FIG. 2 by the lines 7 & t 7 ^ ^u ° d 7 °.

4098H / 0867

BATH ORIGINAL

Another example is given below. A (31o) -plane runs through the position "(3)" between the atoms 3 and h and through, but the atom 5 »the Z axis is not intersected. This plane, which runs parallel to the Z axis, is denoted in FIG. 2 by the lines 8a, 8b, oc and Gd. Other planes through the slement cell of the body-centered crystal cube are denoted in an analogous manner by Miller indices.

Theoretical and empirical studies on the application of field emission cathodes aeigen that for a cathode point the following relationship applies:

E = βV (1).

In this equation, the factor ß means the reciprocal value of five times the effective tip radius r of the cathode. The symbol V denotes the anode voltage and E denotes the electric field strength at the emission surface. The current density at room temperature for a given The radius of curvature that can originate from the cathode is calculated using the following equation:

J = aE ² exp QB φ ^{3 // 2} / eJ, (2).

In this equation, J is the current density in amperes and E is the intensity of the electric field at the top, expressed in volts per centimeter, Jj is the work function of the tip material in electron volts, 13 one

40981 A / 0867 '

Constant equal to 6, o3 x 10 in the above units and a a function of (P (a = 3.5 χ 1 O ~ 'A / V, where for tungsten = k, 5 eV).

The greatest possible current density axis is aimed for. This current density maximum is due to the relative dullness limited to the top. The smaller the effective tip radius r, the greater the current density, as long as all the others Factors stay the same.

The best known method of making a field emission cathode with a faceted shape is to that first devi UoIfram wire is micropolished and shaped according to Figure 3 at :: t. The effective tip radius in Range from 3oo_l - 3ooo ^ - "is achieved by the tip heated, whereupon this under the influence of the Surface tension is rounded off and smoothed. The tip after the rounding process is shown in Figure 3, wherein the letter "r" denotes the effective tip radius.

•; ^r hen the ßraissionsspitvie subjected to an electric field, the emission starts in areas a low work function (or high-Vert i3) of the emitter. The emission usually stays within a gas angle range of about 60 uui around the apex of the emitter surface. To initiate the build-up of the emitter surface the ßniitter-Le.pjerature must be increased sufficiently that the surface becomes mobile. The electrostatic force of the electric field is

409814/0867

acts a migration of the surface atooe zuia! - / niitterscheitel, whereby the enitter is built up.

Z ~ the field strength at which a balance ordered: see i- of the surface tension which results in the rounding of the tip, and the electrostatic -u: if baukraf t, which leads to the formation of a peak, is given by the following relationship:

1/2 (3),

where V is the surface tension. For i-olfram at 2ooo K this is about 29oo dynes / cm. A state of equilibrium is reached at E ", but there is a minimum field strength

'7
of 4 χ 10 V / cui is required in order to achieve a just usable emission during the control. It follows from equation (3) »that r must not exceed a value of about υοο ^. Sotnit has this method (namely to work in the equilibrium point, in which the jitter shape geu. 2 "igur 5 is retained) a definite limitation, which consists in the fact that the current density, which before the fundamental geometrical deformation of the tip (with a corresponding drastic change of the local radius r) can be achieved, is strictly limited to a value which is several orders of magnitude below the desired or in most cases usable \ 7ert

409814/0867

BATH ORIGINAL

- i'd-

niello. "is equal to ürmi-j; so that i'iio lledingunfi of the equation (* 3) only in eitw.,: local area of the,> i., issionsi'i;'. clie is fulfilled a generally unstable geometri— «before Zuptauü, with the exception of the <ioo> construction method.

It i ^ t beka-.ηΐ that heating a · Feldetnissionskatode which has been previously placed according to the shape of Figure 3, in the presence of an electric field of about 2 k-'10 V / cm (that is, under hall;.. The Gleiclycrichtswertes ) hu a structure of the lii'iiss ioiisf läclie leads over surface migrations on the opitze. Thereafter, the emitter can be operated in a stable manner at room temperature, provided that a sufficiently large vacuum is maintained.

The method of producing a built-up field emitter and operating it at room temperature (including the associated restrictions) is described in an article by AV Grewe, DN Eggenberger, J.Wall and L, M. Welter in "The Ileview of Scientific Instruments", Vol. 39, 57'0-5o3, April 19G.J, under the title "Electron Gun Using a. Field Emission. Source" dealt with.

The term "build-up" refers to a process in which atoms migrating in the surface are replaced by one of externally applied electrostatic force can be directed into areas of lower energy, resulting in a lateral

A098U / 0867

BATH ORIGINAL

Growth of levels with low Ililler indices leads to the development of large facets. For a comprehensive discussion of the phenomena associated with your "structure and faceting", refer to an article by PC Bettler and FM Charbonuier, "The Physical Ueviev /", ISd. I ι j ¹ , No. 1, υ.5-93 »I July 19 '~ o »u'.tec referred to the title" ictivation ßoergy for the Surface Migration of Tungsten in the Presence of a High-Electric Field ".

Different types of bpit build-up and faceting have been observed in V / olfarn cathodes. Construction takes place as successively smaller overlapping planes extend outward from the imitter slot. Such a construction is necessarily connected with faceting of other planes which represent the sides of the constructed hermaphrodite. In an observed build strong way Ewissionsstellen arise at l 112j surfaces due to faceting of jj1q} -and | j O oj planes. Another Body leads -to strong iSmission of £ 5lol planes at the same time a faceting of the Π whether] | ~ 11 2j and [^ l 1 takes place oj levels.

Another and very important type of structure causes strong emission disturbances at the [Ίοοΐ levels by faceting levels 1.11ο], j_1 Ί2] and jlo \. By far the greatest reduction in the effective local tip radius r and thus the greatest improvement in the beam angle control

^409814/0867 EADOBGINAL

occurs in a ^ loo ^ -iu f construction with {'(oo ^ -oriented wire, since the primary build-up takes place along the emitter axis with the formation of a pyramid, on the ΓΪ od] level. Only the extremely small spit of the The pyramid acts as the primary eu-issionotelle, since other <1 oo7 structural parts are angularly separated from the vertex ^ 1oo ^ - on the construction site by 9o. The decrease in the electric field with a vertex angle of 9o is sufficient to generate a substantial emission of the 9o (oo!) -.. .iufliaus share au prevent the Gegensat ;; result to other _αιΓ huuendforiüen in a variety of Siiiissionsstellen within a iiinkelverte ilung, ■■ / oak sufficiently close to the longitudinal axis of the emitter is situated, ii- : i allow a substantial e ission from all directors xu. This leads to u an iD "! ission in a broad range.

The highly ilaiie er ^ 'i Ui see <loo> structure has so far only been observed occasionally; md as a transition state, as from an article by L. ... Swanson and LC Crouser, "Journal of Applied Physics", p . ho, mr. 12, h7k \ -h ~ / k9, November lyop, under the title ".iiigular Oonfineuient of Field Electron and Ion Emission". So far, however, it has generally been assumed that under the conditions of thermal field emission, no end farms built up in the field, including the>(loo> structure s of a / loo ^ -oriented wire, can reach a state of equilibrium, because of two factors which have hitherto been considered inevitable Once the geometry of the emitter tip changes apparently inappropriately

A098U / 0867

BM ORIGINAL

casual with the zoit, creating strong current density Fluctuations due to a change in the local opjtzenradius r take place. Second, these changes cause the Z-nitter geometry a drastic jinderi'ng of the ränylic-ien : 3r.iissiDi] distribution. The latter is not tolerable at ., Applications that require fine focusing is so that the ray must be delimited by Oestimmte Iliutel

It has now been established that, contrary to the previous results and contrary to the previous ideas of the specialist community, jittersipit'ienendforyen with ^ Ioo7 -orientierteu: structure uiit] urit extretü small effective radius in; more reliable: nd reproducible ,, eise ar.s einci λΊο · ο7 "^ vI.? · .-. tierte ~. r ·. *: ■ ": everything could be divided j ι. κι iu tue r ^r nisehe. Field can be operated in etaoiler lieine.

The reasons why an unpredictable structure under the conditions which will be explained below are not yet completely clear. It is assumed, however, that three factors are of great importance, Uta one xu achieve loo / -Aufbau reliably acc. to the invention. First, the slightest leads carbon contamination, which is almost inevitable at operating temperatures of the tip of more than i.2oo K and below the temperature of the thermal desorption of the carbon, seemingly

40981 A / 0867

ORIGINAL BATHROOM

'bar from a reduction in the surface free energy of the £ "looj levels, whereby a non- ^ loo) * structure takes place. Second, leads to the presence of an oxide surface, which is observed at peak temperatures below about 195ο Κ apparently leads to an increase in the surface free energy of the £ 100] planes over that of the £ i12j and planes ^ 3IcQ, which promotes a <1oo> build-up. Thirdly, the high degree of field localization to the narrow area of the tip where ß is large makes it possible that the equilibrium conditions (equation 3) on a Place of the Euiitterflache are met, with other areas the surface is not subject to a sufficient build-up area to be able to be further deformed.

First, the basic concept of the invention will be explained with the aid of a simplified flow chart according to FIG. The Ilaar needle cathode arrangement according to Figure h is produced according to conventional methods. This hairpin arrangement 2o according to FIG. H can be made of tungsten and comprises a 1oo> -oriented tungsten emitter wire 21 which ends at a point which is shown in FIG. 3 in a greatly enlarged manner. This hairpin assembly is placed in a partially evacuated chamber and heated sufficiently to desorb contaminants, including carbon in the form of volatile oxide. However, care is taken to ensure that the temperature does not reach or exceed a value at which a thin residual tungsten oxide layer is thermally removed.

A098U / 0867

While the cathode arrangement is now kept sufficiently hot so that a surface change at the tip is possible, a relatively strong electric field is now applied to initiate the build-up. This is the preferred operation at this stage, since the effective surface free energy of the £ iooj planes has been increased while the effective surface free energy of the £ 11ol, JJ123 and J_31o] planes has been decreased. The jTiooJ planes are therefore built up preferentially and the O3, £ i 12J and Q) 10] planes preferably form facets.

In Figure 1, the first block relates to the manufacture of the cathode assembly using a metal wire with a <looj> -autnentered cubic crystal structure, the end of which is shaped into an emitter tip with an effective radius between 5oo Λ and 4ooo A. This wire is made by spot welding with an Ilaarnadeldralit, which serves as a bracket, connected. In the second block according to Figure 1, the tip is heated to such a temperature that although the impurities (carbon as CO) are desorbed, but the remaining oxide layer on the metal wire not removed in the area of the emitter tip will. In the third block, the cathode is heated further in order to maintain surface mobility at the single-emitter tip to obtain. At the same time, an electric field is applied to begin the <^ loo7 build-up, see below.

40981 A / 0867

-13-

Figure ό shows a Eraitter tip, which was formed according to the present invention. The arrow, which is aligned with the axis of the wire, denotes the <1 oo *) direction, while the arrow perpendicular to the wire axis denotes the. At any given time, the vertex 33 of the Euiitter point does not converge to a perfect point, but to a very slightly rounded shape. The faceted £ "lloj planes ^x jo gradually merge into four £ 0lo ~] structural planes which are spaced $ 0 from the apex 53 of the wire and are arranged in the circumferential direction at an angular distance of 90 from one another at the ends of the wire. These (olo) build-up levels are only small thickenings on the sides of the wire, as they are twisted against the axial electric field.

Figure 7 shows an ideal picture of the area of the Erüitterspit ^ e above the level 7-7 in Figure o. This shows that the Zinitterspitae has a generally tetrahetrical shape with its apex 53 i «^ loo) -lichtung and • old four inclined Lateral surfaces (^ o, 5l) due to the (11o) faceting and with a base surface $ 2, which is defined by the (loo) plane 7-7.

As the build-up progresses, the top end becomes finer and the effective local tip radius r decreases accordingly. The current density and the total current become clear according to equation (2) increased. It is necessary

40981 A / 0867

to reduce the electric field strength as the build-up progresses , in order to keep the total current below a limit value at which a catastrophic arc would be generated in the vacuum. When the current is stabilized, a ^ loo / * build-up to an extremely fine point is achieved, and the cathode is ready for operation.

Above only a very rough spelling of the jJrfiu- rl "ng given. For a detailed presentation of the invention FIG. 9 is explained in the following, as well as there: aatiscbe Diagram of Figure 0 showing a defrosting electron microscope represents. As already mentioned, the first stage is to have a - < 1oo> -built thermal field cathode from a (loo) -oriented Manufacture wire by first the tip 22 of the emitter 2 1 according to the configuration of the Figure 3 represents hearing according to conventional methods. This stage can be effectively done by the following electrochemical procedure.

The K 1oo> -oriented tungsten wire emitter 21 can have a diameter in the order of magnitude of o, o7'6 inta to o.23 ram. This wire is connected by spot welding or electron beam welding to a hairpin-shaped holder 20 made of tungsten, which can have a diameter of the order of magnitude of 0.25 mm. Willingly. Figure ü is the cathode structure 32 within a tube,

4098 1-A / 0867

BAD ORiGfNAL

-2ο-

which one encloses a chamber 3o, and which one then to a pressure in the range of 10 Torr with a vacuum pump 31 can be evacuated. Then the temperature of the cathode 32 is increased enough so that the impurities evaporate and the emitter surface getn. Figure 5 smoothed will. This is achieved in that a current pulse from a direct current source or an alternating current source 35 is sent through the tungsten hairpin structure 2o, wherein the cathode 32 is heated to a temperature between 17oo ° K and 19-OO K during several short bursts of temperature, The temperature to which the cathode 32 is heated can can be easily regulated by setting a current limiting resistor 3 ^. The aim of this stage is to remove all impurities to remove from the single point 22 (Figure 3), except for a thin remaining oxide layer, which is on the emitter tip due to previous contact with free oxygen is in the surrounding atmosphere.

Any carbon inclusions within the surface area diffuse to the surface and combine with the oxygen in the original oxygen layer or with the oxygen within the chamber 3o to carbon oxides and mainly to carbon monoxide. The carbon oxides are thermally desorbed, namely »N at a speed which depends on the temperature of the cathode structure 32, so that a faster desorption occurs when the cathode is within the stated range

4098U / 0867

heated to a higher temperature. The upper temperature limit, on which cathode 32 is allowed to be heated during this and subsequent steps will be discussed below explained.

As FIG. 6 shows, the tip 22 becomes hemispherical with an effective radius r due to the surface tension which acts on the increasingly mobile heated cathode 32. It should be noted that if there are enough carbon impurities in the \ loo7-oriented wolf wire, it may be necessary to introduce a certain amount of additional oxygen into the chamber 3 »while the emitter is heated to temperatures between 130K and 150K Uta to bring about a complete combination of the carbon on the surface and in the bulk with oxygen. If a high-purity tungsten wire is used, the oxygen in the starting material is usually sufficient to carry out the cleaning. In most cases there is always a sufficient amount of oxygen present, and it was either in the absorbed elemental form or in the form of iJolf ram oxide, so that the carbon can mix with it! Can combine hydrogen to form volatile oxides. Furthermore, the carbon can be thermally desorbed directly from the surface by heating to temperatures above 250K. ^Here-: by, however, a significant blunting of the emitter is effected, which very often is undesirable. The latter procedure may also have additional steps to introduce

A098U / 0867

a small amount of oxygen necessary so that this reduces costs.

In order to cause the impurities to be completely gasified in the tip area, the cathode 32 is opened several times by shock heating to higher temperatures, namely in the previously mentioned Dreich. The temperature is allowed but not a value in the order of magnitude of 195ο Κ exceed. In addition, the temperature may have a specific temperature limit value in a respective specific atmosphere not exceed, in which a very thin residue - layer of "wolf rough" oxide, which normally forms the cathode 32 is completely removed, leaving a bare tungsten tip. The desired top condition is then achieved when all carbon impurities are removed and a residual thin layer of wolf rani oxide remains.

The field-aiifbauverfahren can now begin the tip 22 by applying a voltage between Euiitterspitze 22 and a nearby anode 3 ^>. This creates a mechanical tension. The B-jxtter must be kept sufficiently high (1 ^ 00 IC to Κίοο K), while sufficient surface mobility is guaranteed. The temperature, however, must not exceed the upper limit mentioned above, at which the oxide layer would be removed.

0981 4/0867

The electrical field is increased until an equilibrium value is exceeded, whereupon a manageable, stable structure of the tip with the formation of a very fine tip begins. The initial current at the beginning of the peak build-up is typically around 2o «./? . One observes first that the total current initially remains constant for a certain period of time and then increases rapidly over time xvird, which indicates a (l ooi) structure. After entering of the tip construction, it is desirable to use the maximum of the total flow to a range from üo - 1 oo ^ au,

/ by appropriately adjusting the cathode-anode dimensions and thus the electric field lowers «The total current stabilization indicates that the structure of the cathode is extremely fine axially aligned tip 23 according to FIG.

In practical applications such as the scanning electron microscope According to Figure ο, a complete tip build-up is achieved indicated by a high quality image transmission. Generally, a beam passage is through a small opening the best sign of a (loo) buildup. In a similar way Thus, peak degradation is indicated by degraded transmission.

The aim of the tip construction method described is to create a pyramid-shaped structure as shown in FIGS reach. With a ^ loo ^ structure of the tungsten, the

4098 U / 0867 b * d original

-2k-

JJooj, Q 12 | and (J31oJ planes large facets, such as the facets 5o and 51 generally shown in FIG. 7 · Since the partially built up ^ olo) planes according to FIG Electron emission at. The entire electron emission directed in the longitudinal direction emanates essentially from the apex 53 and is limited to a narrow angular range.

It was found that after reaching the built-up tip end shape according to FIGS. 6 and 7, the emitter for a long period of time in the temperature range between 1200 K and 17oo K within a total current range of 1-3oo Ajjfy and under an ambient pressure in the range of 10 to 10 ^υ Torr can be operated. If the pressure is in the range -9

of 10 Torr, a lower temperature range can also be selected. In-course operation is also possible at even lower temperatures, including room temperature, whereby the useful life is inversely related to the ambient pressure of the emitter. The limit for the smallest possible spot size in the focused beam depends in part on the chromatic aberration, which in turn is related to the width of the energy distribution. Since the width of the energy distribution increases with the temperature, it may be desirable, in fact, for a short period to work at such low temperatures -to.

4098U / 0867

If the emitter in high vacuum to temperatures of more than about 2,000 K is heated, the oxide layer is removed. With such a clean emitter, which one is free is made up of traces of carbon impurities, it comes from according to the laws of probability and function the hierarchy of the surface free energies of the different families of planes to a poo ^ -teak structure, if the above conditions are observed. However, if carbon is present, there will be an (I oo) build-up among them Conditions not predicted and of course will not occur.

During operation at low temperatures or at unwanted operation at a high temperature the emission properties of the emitter can be deteriorated. It has now been found that the original tips— Conditions during operation can be restored by keeping the electric field for 15 to 3o see. relaxed at a temperature in the range from 1ooo K to 1 / Oo K or reduced and then applied a sufficient voltage to realize a total current of Io to 2oi * ft. The surroundings of the emitter can of course contain a certain amount of oxygen sufficient to restore the oxide layer and any carbon contamination of the surface, which originate from inclusions or from adsorbed carbon sources to remove. A Strou-Itestabilization i; tends to restore the top.

409814/0867

In the following, Figure ο Be.jug should again be used. In this figure, a scanning electron beam device is shown in which a narrow beam 40 is emitted from the cathode 32 . This beam is focused on a grounded sample by means of a single electrostatic or electromagnetic lens k2. The cathode 32, the anode 'JL, the lens 42 and the grounded sample 4i are held at an increasingly higher potential) (-10,000 volts, -0,000 volts, -000 volts and 0 volts), which is indicated by the' sources 33 »3 <-> and 44 is displayed. A tilt generator 4 for a display tube 46 is connected to a deflection system 47 so that the beam 40 scans the sample 41. A detector 4-J is provided behind the sample 4l, and the display tube 46, which is used to display the sample, is connected to the detector 4o. If the L.pit ^ e is faulty or if the (loo) - άτι buildclf orifl deteriorates, the tip can be restored according to the oiiigeu procedure without having to remove the cathode from the apparatus, if the -i / ütastelek fcroneuuikoskop ge,;. Figure '-' ■ is switched off, the Suitterte: lyeratnr : \\ -iMoIi ^ t verri '^ - jert i / earth, Lii the Ji.aufbaukoiiiiguration "freeze", after which the inode / cathode ^ paunuug is switched off . When the electron microscope is now to be put back into place, the voltage is first switched on and then the cathode is heated to the operating temperature. In the event that the described shutdown procedure is not followed, the tip can be used. lost. In this case, however, this structure can be restored very early by repeating the heating and voltage application steps.

409814/0867

BATH ORIGINAL

One of the most essential properties of the tip constructed in accordance with the invention, which is observed during operation of the tip, is the ability to operate at extremely high current rates for a long period of time at pressures which are orders of magnitude higher than previously thought possible with field emission cathodes. With respect to the vacuum system and other peripheral devices, as well as in terms of basic equipment significant savings can be made so that a large U ^r irtschaf tlichkeit, also with regard to the operation and particularly with respect to the initial investment can be received.

As mentioned earlier, molybdenum also has a space-centered cubic crystalline structure, and corresponding results are ti it with a <(l oo) -oriented molybdenum wire similar procedure, but at slightly lower temperatures. The surface tension (V ') for molybdenum is in the range of 22oo dynes / cm at 1700 K. This changes the intensity of the electric field at which the equilibrium point is achieved according to equation (3). The behavior of a molybdenum emitter manufactured according to the above process is similar to that of a tungsten emitter. It is in turn a basic requirement that all carbon impurities are removed and that a thin residual oxide layer remains on the surface, so the work function of the i / 100] levels and that of the

409814/0867

_1 1 ο) - , Lj ί 2J - and I3I0J -planes is lowered, whereby the (1 oo> - iupbau is preferred. Certain other Iletalle such as tantalum are also suitable for the inventive method, but tungsten is generally preferred as starting material , because it has a high melting point and a low vapor pressure and a relatively high electrical and thermal conductivity and high mechanical strength.

4098U / 0867

Claims (1)

1Λ - 519 Claims Method for manufacturing a field emission cathode from a metal wire and for reshaping the same after operation « characterized in that one under vacuum a) a ^/ Λ oo / -oriented metal wire heated to a temperature at which impurities are removed from the wire, and that one b) applies an electrical potential to the wire so that a minimum sanitary current begins, and that c) the current under building the Γΐοοΐ levels au a fine tip is maintained until the current stabilizes, wherein steps (a) and (b) can be carried out simultaneously or in reverse order. 409814/0867 2. The method according to claim 1, characterized in that that the metal wire is made of a metal with a rough, cubic-crystalline structure and preferably made of tungsten or molybdenum. 3. Method according to one of the claims! or ?., characterized in that one is a metal wire and a hemispherical one with a conical tip / end with an ul liadius of less than 2'joo 1. k. Method according to one of the claims I "jxe 3", characterized in that the total current is kept below a predetermined maximum value at which the arc discharge takes place. i>. Method according to one of Claims 1 to 4, characterized characterized in that the temperature of step (a) is kept below a limit value at which essentially all of the surface oxide layer is removed or where a geometric deformation the ^ pit ^ e enters. ό. A method according to any one of claims 1 to j, characterized in that in stage (a) a temperature is selected at which the carbon of the metal wire diffuses from the surface and reacts with oxygen to form carbon oxides which are thermally desorbed . 4098U / 0867 BAD ORIGINAL 7. The method according to any one of claims 1 to 6, characterized marked that uian before, after or during the Step (b) adjusts the temperature to a value at which there is a surface mobility of the wire tip is guaranteed and preferably to a value between 1200 K and 19 <> o K and in particular between 1300 ° K to 1500 ° K and 1800 ° K. 3. The method according to any one of claims 1 to 7> characterized in that the wire consists of tungsten and its temperature in step (a) has a value between 1200 K and 19oo K to 2'ioo H and preferably a value between 14oo iC to I700 Ii and 1900 ° K. 9. The method according to one of claims 1 to G, characterized -7 marked that at a pressure of 10 to -9
10 Torr is worked. Io. Method according to one of Claims 1 to 9> thereby characterized in that after stage (a) and optionally after stage (b) or before stage (c) oxygen is supplied and / or that a metal wire with an oxide layer is used in step (a). 409 814/08 67 BATH ORIGINAL 11. The method according to any one of claims 1 to 1o, characterized in that a metal wire with a Diameter from 0.12 to 0.25 mm is used. 12. The method according to any one of claims 1 to 11, characterized in that in step (b) an initial flow of 5 to 25M-B and preferably 1o to 2ojt π is selected. 13 · The method according to any one of claims 1 to 12, characterized in that during stage (c) the Measures current and after a £ ioo_) structure indicating Current rise sets the applied potential to achieve a predetermined current strength. 4098U / 0867 Le θ rs θ 11e