US 3394066 A
Description (OCR text may contain errors)
July 23, 1968 j M|LES 3,394,066
METHOD OF ANODIZING BY APPLYING A POSITIVE POTENTIAL TO A BODY IMMERSED IN A PLASMA Original Filed April 12, 1965 4 Sheets-Sheet 1 30 I30. 29 3| 2%: j A
v-MlI v 32 33 l l 21 l [I n1. 1 II Fig.l
INVENTOR. John L. Miles July 23, 1968 J. L. MILES 3,394,066
METHOD OF ANODIZING BY APPLYING A POSITIVE POTENTIAL To A BODY IMMERSED IN A PLASMA Original Filed April l2, 1963 4 Sheets-Sheet 2 CONDUCTING RESISTIVE BODY FORMED BARRIER OF MATERIAL FORMED OF M $1 NEGATIVE N IONS QQQ QQ: 0 SUPPLIED FROM 0 PLASMA \\\\\\\Q \\Q QR 51 O \:Q
ELECTRIC POTENTIAL F I9. 2
2503 4 PERFORMANCE OF M203 nuns FORMED ON AI SURFACE BY THERMAL OXIDATION AND BY METHOD DESCRIBED 3 (SEE TABLE 1) E I 7 O 5 3 9| A 3 E ii 2 5 CURRENT IN MILLIAMPERES Fig.3
INVENTOR. John L. Miles Amzrney July 23, 1968 J. L MILES 3,394,066
METHOD OF ANODIZING BY APPLYING A POSITIVE POTENTIAL TO A BODY IMMERSED IN A PLASMA Original Filed April l2, 1963 4 Sheets-Sheet 3 1 HIIHHIIIUIW INVENTOR.
John L. Miles Fig. 5 BY J. L. MILES 3,394,066 METHOD OF ANODIZING BY APPLYING A POSITIVE POTENTIAL July 23, 1968 TO A BODY IMMERSED IN A PLASMA Original Filed April 12, 1963 4 Sheets-Sheet 4.
RESISTIVE BARRIER ALUMINUM OR (OTHER METAL 46 ALUMINUM F ig. 7
l N203 THICKNESS Fig. 8
INVENTOR- John L. Miles Fig. IO
United States Patent 3,394,066 METHOD OF ANODIZING BY APPLYING A POSI- TIVE POTENTIAL TO A BODY IMMERSED IN A PLASMA John L. Miles, Belmont, Mass., assignor to Arthur D. Little, Inc., Cambridge, Mass., a corporation of Massachusetts Continuation of application Ser. No. 282,187, Apr. 12, 1963, which is a continuation-in-part of application Ser. No. 225,100, Sept. 20, 1962. This application July 17, 1967, Ser. No. 654,009
18 Claims. (Cl. 204-164) ABSTRACT OF THE DISCLOSURE A method of forming a resistive barrier film on an electrically conducting body of an anodizable material through reaction of negative ions with the positive ions of the anodizable material. After the formation of a thin priming layer, the resistive barrier is caused to increase to a desired thickness by impressing an electrical potential across the priming layer.
This application is a continuation of application Serial No. 282,187, filed April 12, 1963, now abandoned which in turn is a continuation-in-part of my copending application Ser. No 225,100 filed Sept. 20, 1962, now abandoned.
This invention relates to a method for forming a resistive barrier film and to the resulting film produced thereby. In particular, this invention relates to forming films capable of controlling the passage of electricity in circuits.
It has long been known that certain metals can be chemically treated to form a thin layer of dielectric on their surface, the dielectric being a compound which is the reaction product of the metal and the chemical treating agent. Generally, these compounds deposited on the metal surface form a barrier against further reaction and thus additional thickening of the barrier is in some cases difiicult, in others impossible. Probably the best example of such a barrier is the formation of aluminum oxide on aluminum. An aluminum surface will readily oxidize in normal atmospheres to form a layer of aluminum oxide on this surface, the thickness of which is limited to about 20 A. If heated in an oxidizing atmosphere to about 500 C., an aluminum surface will oxidize readily to form a layer of aluminum oxide up to about 40 A. Since aluminum oxide is known to be highly resistive to further oxidation as well as to many other corrosive reactions it is purposely formed on many aluminum articles.
In order to form, in a controllable manner, a layer of aluminum oxide having a consistent thickness throughout, it is customary to anodize it. The now-well-known process of anodizing comprises immersing the aluminum article in a liquid electrolyte and applying a potential across it such that negatively charged oxygen ions, derived from the electrolyte, are directed to the surface of the aluminum article serving as the anode in the circuit. In recent years this type of oxidation, which may be termed wet anodization to contrast with the method described herein, has been applied successfully to many other metals including, but not limited to, tantalum, titanium, zirconium, niobium, uranium, berryllium, manganese, magnesium and the like. However, as will be shown below, there are certain inherent disadvantages in forming oxide films on many of these metals by the wet anodization process.
In recent years with the development of new active electronic elements (e.g., cryotrons, tunneling devices, field emissions amplifiers, thin magnetic film elements and the like) it has come to be recognized that there must be 3,394,066 Patented July 23, 1968 developed methods for forming film barriers capable of controlling the passage of electricity, i.e. behaving as true insulators, or as dielectric layers which are capable of permitting the passage of substantial tunneling currents. Moreover, the rather recent developments in microminiaturization have pointed out the need for extremely small inactive or passive electronic devices among which may be listed capacitors and resistors. The size of such devices, the requirement for extremely exacting dimensions, the need to be able to make them in quantity, and the absolute necessity of good quality control has, of course, indicated the need for the development of techniques other than soldering, laminating and the like which are limited to the extent to which they can be applied to very small devices. This, in turn, has meant that techniques such as anodizing and vapor deposition in patterns have been used to overcome the drawbacks inherent in the heretofore standard techniques.
As an example, aluminum oxide appears very attractive as an electron barrier material (either as an insulation or as a dielectric layer for tunneling). However, in order to put a coating of aluminum oxide on an aluminum surface it is necessary to either oxidize it in air or in some other oxidizing atmosphere, or to anodize it. The former technique, it carried out at room temperature, limits the ultimate thickness of the aluminum oxide film; and if carried out at elevated temperature requires intense heating of the device being formed.
Although wet anodizing offers the possibility of forming thicker films of aluminum oxide, of controlling the thickness of the oxide, and of making a film of uniform thickness there are associated with this process a number of inherent disadvantages. Among these disadvantages may be listed the necessity for immersing the article in an electrolyte, the possibility that the film formed is soluble or partially soluble in the electrolyte, the strong probability that the film will have included within it constituents derived from the electrolyte that leave weak spots, and finally the limitation on the type of negatively charged ions which can be furnished by the electrolyte and hence upon the chemical characteristics of the barrier film itself. Thus both the metal and the barrier film surface are limited in quality and kind.
It would therefore be desirable to have available a method for forming a barrier film capable of controlling the passage of electron current, the thickness of the film being one which could be accurately controlled and could be formed to any desired magnitude. It would also be desirable to have such a method which could be performed at room temperature without exposing the metal surface to a liquid and which would no longer be limited to certain metals and to specific barrier films.
It is therefore the primary object of this invention to provide an improved method for depositing a barrier film on a surface of an electrically conducting body through chemical reaction with the surface. It is another object of this invention to provide a method of the character described which can be carried out at room temperature and without the use of liquid electrolytes. It is another object to provide such a method which is extendable to metals and other electrically conducting materials heretofore not usable, and to the formation of barrier films heretofore not formable on metal surfaces. It is another primary object of this invention to provide a method for making circuit elements of the character described which are particularly suited for micro-miniaturized circuits.
It is yet another object of this invention to provide films which do not contain constituents which would give them weak spots. It is another object to provide such films which are capable of good quality control as well as thickness control. It is yet another object to provide such films which are incorporated in active and passive circuit elements and form a part thereof. Other objects of the invention will in part be obvious and will in part be apparent hereinafter.
The invention accordingly comprises the several steps and the relation of one or more of such steps with respect to each of the others, and the article possesses the features, properties and relation of elements which are exemplified in the following detailed disclosure, and the scope of the invention will be indicated in the claims.
For a fuller understanding of the nature and objects of the invention reference should be had to the following detailed description taken in connection with the accompanying drawings in which FIG. 1 is one modification of an apparatus suitable for carrying out the method of this invention;
FIG. 2 is a diagrammatic representation showing the formation of a barrier film;
FIG. 3 is a series of plots of current versus voltage for four aluminum oxide films of varying thickness formed by thermal oxidation and by the method of this invention using an external voltage source;
FIG. 4 is another modification of the apparatus showing the use of liquid nitrogen for condensation of moisture;
FIG. 5 illustrates an evaporated test pattern;
FIG. 6 is a much enlarged view of one of the crossovers in the test pattern of FIG. 5;
FIG. 7 is a much enlarged cross-sectional view of the actual cross-over;
FIG. 8 is a plot of barrier thickness buildup vs. discharge exposure time;
FIGS. 9a and 9b are contour maps of resistances with and without the use of an extremely thin layer of aluminum under the test pattern;
FIG. 10 is a top plan view of a cryotron formed using the method of this invention; and
FIG. 11 is a cross-sectional view showing the construction of the cryotron of FIG. 10.
The method of this invention can be briefly described as comprising the steps of forming a priming barrier on the surface of an electrically conducting body by reaction with a gaseous reactant, exposing the barrier to a plasma comprising negatively charged ions of the gaseous reactant, and impressing across the priming barrier an electrical potential to increase the thickness of the priming barrier and form the resistive barrier film which is characterized as being the reaction product of the negatively charged ions and the conducting material. The
negatively charged ions are of a species which form with the conducting body material a compound which in film form is capable of controlling the passage of electricity.
The step of impressing an electrical potential across the priming barrier may be accomplished either by providing an external source of electrical current such as by contact with an electrode which is part of a suitable circuit for supplying an electron current, or by providing a sufficient supply of negatively charged ions on the priming barrier surface to establish a potential between these negative ions and the positive ions within the conducting body. The first of these modifications makes it possible to deposit a somewhat thicker resistive barrier than when the econd is used. Both modifications will be illustrated.
The resulting resistive barrier is characterized as being of consistent thickness throughout, essentiall free of occluded materials, and completely controllable with respect to its thickness. Moreover, the film need not be an oxide and the surface on which it is formed need not be a metal or be restricted as heretofore was the case.
FIG. 1 illustrates a typical apparatus suitable for carrying out the steps of this invention when the impressing of the electrical potential is achieved through the use of an external voltage source. The modification of this invention in which the potential is established by the supplying of sufficient negatively charged ions on the priming bar- Ill 4 rier will be illustrated with references to FIGS. 49. The effect of reducing moisture vapor content on the uniformity of resistance obtained over a given area will also be illustrated in conjunction with the discussion of these figures.
Since it will normally be convenient to carry out the deposition of the barrier film in a vacuum system, the apparatus of FIG. 1 shows a bell jar 10 which defines within its interior an atmosphere of reduced pressure or plasma 11. Typically, to begin the process the bell jar is evacuated to from about 10* to 10 mm. mercury. A vacuum line 12 with a suitable check valve 12a connects the bell jar to a vacuum pump (not shown). Conduit 13, through which the gas making up the plasma in the bell jar is introduced, is likewise provided with a suitable valve 13a.
In the formation of circuit components, or of circuit elements such as may be used in microminiaturized circuits, it will be convenient to affix the conducting body to a suitable substrate 15 which is seen in FIG. 1 to be supported in the plasma 11 of the bell jar 10. Such a substrate is normally a nonconducting body such as a glass microscope slide on which a film of the electrically conducting material is deposited from the vapor phase. This, of course, may be done in the bell jar as a first step. In FIG. 1 thin film 16 is shown to be deposited and adhered to the substrate 15 which is maintained within the bell jar on support 17. For convenience this support extends through the bell jar so that it may be oriented within the jar from without.
In the arrangement shown in FIG. 1 the bell jar itself rests upon a metal electrically grounded base 20 which in turn sits upon a base support 21 and is afiixed to a foundation 22. Any other suitable ground other than base 20 may be supplied if it is not convenient to use the base as the ground. The required negative ions of the gaseous reactant are provided in the necessary ionized state through the establishment of an electrical field which in FIG. 1 is formed between the high voltage electrode 24 and the electrically grounded base 20. The high voltage electrode 24 is in turn connected through the wall of the bell jar by means of a vacuum-tight seal through a high voltage line 25 to a suitable power source 26. A switch 27 is provided in high voltage line 25.
The electrode 28 (which for convenience is referred to as the anodizing electrode) enters the interior of the bell jar through a vacuum'tight seal, and through power line 29 is connected to a power source 30. An attachment to the electrically grounded base 20 through an ammeter 31, voltmeter 32 and switch 33 completes the circuit.
FIG. 1 illustrates the use of DC power. An AC source of electricity may be used equally well for the anodizing electrode 28 and high voltage electrode 24; but the DC source is preferred inasmuch as deposition of the film barrier will of course take place only during half of the AC cycle.
The formation of an aluminum oxide film on an aluminum surface may be used as an example to describe a typical operation of the apparatus in FIG. 1 in the deposition of a film barrier in accordance with this invention. A clean microscope slide serving as a substrate 15 is attached to the support 17 and the bell jar placed in position. After the atmosphere within the bell jar has been pumped down to a vacuum of about l() mm. mercury or less the substrate 15 is brought into position over a mask or stencil and aluminum is vaporized on the substrate in a pattern which corresponds to the openings in the mask. Inasmuch as the mechanism for depositing the printed pattern of aluminum (the conducting body) on the slide and the apparatus required for doing this are not part of this invention, neither is illustrated in FIG. 1. This represents a standard technique for forming printed circuits.
After the aluminum 16 has been deposited upon the slide it is raised in position so that that film which is to be coated is brought into contact with the anodizing electrode 28. A small amount of dry purified oxygen gas is then introduced through line into the bell jar. It has been found that the oxygen pressure in the bell jar has two optimum values namely about 1 mm. mercury and about 50 microns mercury. Either of these pressure levels may be used in the process. Other pressures may also of course be used but these have been found to be optimum for oxygen and aluminum. If desired, an inert gas such as argon may be mixed with the ionized reactant gas in the bell jar. The oxygen present is then ionized by connecting the high voltage electrode 24 to the power source 26 through switch 27 and the anodizing circuit is completed by closing switch 33 making the necessary connection between the power source 30 and the anodizing electrode 28.
In the presence of even the slightest amount of oxygen, air or water vapor within the bell jar there is deposited upon the aluminum film 16 a coat of aluminum oxide which in this case forms the required priming coat. This means that when anodizing electrode 28 is brought into physical contact with the aluminum film 16 there is already present the priming layer of the resistive barrier of aluminum oxide which is required as will be explained in greater detail with reference to FIG. 2.
The thickness of the barrier film is controlled by the amount of voltage supplied by power source 30. In the operation of the apparatus of FIG. 1, it has been found convenient to increase the anodizing voltage gradually, or by increments. Normally it is preferred to start with a low voltage, for example of the order of 1.5 volts, and observe the current flow. When this has decreased to a small value the voltage is increased by another increment and the process repeated until the desired thickness of barrier film has been built up.
Several theories have been offered to explain the ability of selected conducting bodies to build up resistive films on their surfaces. Although I do not wish to be bound by any one theory, that set forth in the Transactions of the Faraday Society, volume XLIII, pages 429 134 (1947), appears to offer a reasonable explanation of the process by which a resistive barrier is formed by con- 'ventional wet anodization. It appears reasonable to apply essentially the same theory to the method of this invention.
Turning now to FIG. 2, it will be seen that there is provided a conducting body of a material M having a surface S on which the protective barrier is to be built. The barrier film itself has a surface S By the theory offered it is necessary at surface S to provide negative ions from the surrounding plasma and positively charged ions from the conducting body. It will be seen that as the resistive barrier, which is the reaction product of the negative and positive ions, grows thicker there is presented the problem of bringing the required positively charged ions to the surface S This is done by impressing an electrical field across the resistive barrier. It is believed that the presence of this electrical field is responsible for the growth of the barrier. This can be explained by postulating that the positive ions of the conducting body go into solid solution in the resistive barrier across which the electrical potential exists. This means that the positive ions can readily diffuse in the resistive barrier and, under the influence of the field or concentration gradient, are capable of drifting across the barrier to combine with the negative ions on the surface. When the field across the resistive barrier becomes insufiicient to effect any further solubility of the positively charged ions in the barrier it is necessary to increase the field strength to force more of the positive ions to diffuse through the barrier. This is readily shown (as indicated above in connection with the description of the operation of the apparatus of FIG. 1) by the necessity for periodically increasing the anodizing voltages. Thus when the barrier has reached a thickness where the potential across it no longer causes solubility and diffusion of the positive ions in it, no more positive ions are available at surface S for reaction with negative ions. However, application of additional potential will enable the barrier formation to begin again.
In the process there appears to be no theoretical limitation upon the thickness of the barrier which can be built or deposited in this manner. However, there are practical limitations dictated by the equipment itself. Aluminum oxide films 400 A. thick have been built up by this method with no difficulty.
In the prior art process of anodizing, in which the surface to be anodized is immersed in a liquid electrolyte, experience has shown that a fixed and essentially constant film thickness build-up is associated with a unit of applied voltage. For example the value in terms of angstroms/ volt for an aluminum oxide film on aluminum is 14. In the method of this invention the comparable figure for forming an aluminum oxide film on aluminum is 22 angstroms/ volt.
Once the required resistive barrier film has been formed to the desired thickness, the entire assembly thus formed, or at least the barrier film, may be encapsulated to prevent moisture or other contaminants from affecting the performance of the film. This is conveniently done by vapor-depositing a film of a few microns (e.g. two) thickness of arsenic trisulfide, silicon monoxide, or other protective film material, over the barrier film or the entire assembly.
The physical properties of the aluminum oxide films formed in accordance with this invention were evaluated to compare with films formed by other techniques. To do this, aluminum oxide films were formed on pure aluminurn surfaces which in turn had been deposited by vacuum deposition techniques on a glass microscope slide. All samples were formed by depositing a 1 mm. wide strip of aluminum on the slide, then forming the oxide layer as indicated and finally depositing another 1 mm. wide strip of aluminum at right angles to the first. Voltagecurrent characteristics were measured by a four-probe method. Table 1 summarizes the data obtained and FIG. 3 is a plot of selected data obtained on samples prepared in the same manner. In FIG. 3 curves 2, 3 and 4 repre sent performances for Examples 12, 3 and 8 respectively; while curve 1 represents the performance of a typical thermally produced oxide film.
TABLE 1.FORMATION OF ALUMINUM OXIDE FILMS Calculated Measured Measured Example No. A1203 Film Capacitance, tan 6 loss 1 Thickness, A. L/cm.
Wet anodized 270 0.310 0. 050
l Capacitance and tan 8 loss (dissipation factor) measured at a frequency of 1 kc.
2 Reported in Proceedings of the IRE, 48: 1482 (August 1960).
Examples 1-9 were formed using DC current while Examples 1012 were formed using AC current. Examples 5, 6, 8 and 9 were formed using an oxygen plasma pressure of 1 mm. mercury and Examples 14, 7 and 10-12 using a plasma pressure of 50 microns mercury.
It will be seen from Table 1 that aluminum oxide films formed by my method exhibited capacitance values which compared favorably with those of films formed by recent techniques of wet anodizing, and were superior to the prior art films in their lower dissipation factors under the test conditions used.
In describing the process of this invention it has been pointed out that in the formation of the resistive barrier layer it is necessary to impress an electrical potential across the barrier layer in order to make it possible for the positively charged aluminum ions to migrate to the surface where barrier layer growth takes place. FIG. 1 illustrates an apparatus wherein this potential is applied from an external source. FIG. 4 illustrates apparatus wherein such an external source of current is not employed. In providing an oxygen (or other gaseous) plasma, negatively charged ions are furnished at the surface S of the resistive barrier as shown in FIG. 2. If a sufiicient number of these negatively charged ions are present at surface S then an external current is not necessary since the required electrical potential will be established between these negative ions and the positive ions in the conducting body. It is, therefore, not necessary in the apparatus of FIG. 4 to supply the external circuitry or the anodizing electrode 28.
The apparatus of FIG. 4 illustrates an additional feature, that of the use of a well 40 extending into the evacuated bell jar 10 and containing a cryogenic fluid such as liquid nitrogen 41. The purpose of the well is to afford a refrigerated surface within the bell jar on which moisture vapor will condense. This has been found to be one effective way of reducing the moisture content within the bell jar. Of course other ways of reducing or minimizing water vapor content may be used, such as baking the system at elevated temperatures.
In forming a resistive barrier layer in apparatus of FIG. 4 the same steps are performed as in the use of the apparatus of FIG. 1, beginning with the step of building up a priming layer on the conducting body surface, and then impressing an electrical potential across the priming layer to effect the growth of the resistive barrier. It will be appreciated that in many cases the priming layer forms spontaneously with the introduction of the plasma or spontaneously with the residual gases present in the vacuum system. The following description will again be given in terms of an aluminum oxide barrier layer on an aluminum surface. However, it is not meant to restrict the process of this invention to the formation of aluminum oxide on aluminum.
In order to determine the effect of discharge exposure time on the resistances of a number of aluminum oxide barriers, and hence on their thicknesses, 100 individual crossings were evaporated onto a single one-inch by oneinch glass substrate in the pattern illustrated in part in FIG. 5. This was accomplished by first depositing on the substrate 400 electrical connectors 45 (deposited as 20 rows of 20 each). These connectors were deposited from the vapor phase through an appropriate stencil. The bell jar 10 was evacuated to a pressure of 10 mm. Hg. prior to the formation of the test pattern. The next step was the vacuum deposition by use of an appropriate stencil of the 100 aluminum lower electrodes 46 (FIG. 6). At this point sufficient dry pure oxygen to give a pressure of 50, was introduced through inlet 13 by opening valve 1311. This was sufiicient oxygen to form a thin priming barrier film of A1 0 on the lower electrodes 46.
In order to form resistive barrier films of A1 0 of increasing thicknesses on the cross-over areas to be evaluated, the oxygen plasma discharge was turned on by closing switch 27 for two minutes. When the discharge was turned oif two rows comprising 20 electrodes were then completed by evaporating conducting metal strips 48 to intersect lower electrodes 46 and form cross-over test areas 49 having A1 0 barrier layers. Then the plasma discharge was turned on again for a brief period and an additional 20 cross-over areas completed by forming upper electrodes 48 on them. In similar manner five different data points were obtained each represented by 20 individual crossings having A1 0 layers of increasing thicknesses. The resistances of each of the cross-over areas (A1 0 barrier films) were measured by applying current and measuring the voltage as indicated in FIG. 6. Thicknesses of the A1 0 films formed were then calculated from these resistances and plotted against time of exposure to the oxygen plasma in FIG. 8. These data illustrates that it is possible to build up resistive barrier layers in this manner, and to control thicknesses very accurately and conveniently.
In constructing a number of test patterns such as illusstrated is FIGS. 57 it was found that in general the resistances of the cross-over areas near the edges and in the corners of the patterns were higher for a given discharge exposure time. It is believed that the presence of water vapor exerts some influence on the growth of the barrier films, particularly on those specimens close to the edges where it would have a lesser distance to travel to the films. Thus it is preferable in some applications to remove or counteract the effect of the water vapor to minimize the variation in aluminum oxide thickness and hence resistivity of the barrier.
The presence of the refrigerated surface within the evacuated area such as is created by the liquid nitrogen in well 40 of FIG. 4 contributes to the removal of the moisture vapor from the atmosphere. However, it appears that the glass substrate (or any other substrate) also has water vapor associated with it and it has been found that a major portion of this water vapor can be removed or at least partially counteracted by baking or by depositing an extremely small amount of aluminum on the substrate surface before depositing the conducting body on which the resistive barrier is to be built. Ireferably this initial alumin um film is sufiiciently thin so that it is nonconducting. Apparently, it need not be a continuous film.
In order to illustrate the effect of the use of the initial aluminum film to trap unwanted moisture vapor a test pattern, as described in connection with FIGS. 57, was laid down upon a glass substrate, the upper half of which contained the thin nonconducting layer of aluminum, the lower half of which was untreated prior to the deposition of the aluminum strips. Resistances were measured by the technique illustrated in FIG. 6 and plotted as a contour map shown in FIG. 9. FIG. 9a represents the resistances measured over that portion of the slide which contained the initial nonconducting layer of aluminum. It will be seen that the maximum variation in resistance lay between 8 and 12 ohms. In FIG. 9b resistances were plotted as a contour map for the lower half which did not receive the initial nonconducting aluminum film. Here it will be seen that resistance varied between 15 and 40 ohms, or by a factor of three.
The materials on which the barrier films may be built may be defined as those which are capable of conducting an electrical current. This means that they will normally be metals, although they are not limited to metals since non-metallic materials such as silicon and germanium are conductors. The electrically conducting materials must also be capable of forming with the negatively charged ions a compound which itself, in film form, is capable of controlling the passage of an electrical current. Normally this means that the film barriers will be insulators, although when sufiiciently high voltages are applied su stantially tunneling or field emission currents may flow.
The materials from which the electrically conducting body may be formed include, but are not limited to, aluminum, magnesium, antimony, bismuth, tantalum, chromium, beryllium, niobium, titanium, zirconium, tungsten, boron, lead, silicon and germanium.
The negative ions may be any ions which are capable of reacting with the conducting metal to form a compound which in film form is capable of controlling the passage of an electron current. The most common of these are, of course, oxygen ions. However, some nitrides and sulfides are known to be resistive and thus nitrogen ions and sulfide ions are within the scope of this inven- TABLE 2.-FORMATION OF MAGNESIUM OXIDE FILMS Calculated MgO Measured Measured Example N0. Film Thickness, Capacitance, tan loss 1 A. rt/cm. l
Capacitance and tan 6 loss (dissipation factor) measured at a frequency of 1 kc.
In a similar manner an antimony oxide barrier film was formed on an antimony surface and a tantalum oxide barrier film on tantalum. The antimony oxide film had a capacitance value of 0.5 af/cm. and a tan 6 loss of 0.08. The tantalum oxide film exhibited the normal interference colors associated with tantalum oxide formed by anodization using a liquid electrolyte.
Using the apparatus FIG. 4, i.e. without employing an external source of electrical current, an electrical resistive barrier film of lead oxide was formed on a lead film and of silver oxide on a silver film. In each case, the lead or silver film was deposited on a glass slide from the vapor phase. After the priming layer of the oxide had been formed, dry oxygen was continuously circulated through the vacuum bell jar at a pressure of 50' microns. This continual pumping of oxygen through the system serves to flush out at least :a portion of any moisture present in the system. While oxygen was thus being supplied the switch 27 was closed to ionize the oxygen and provide the necessary oxygen ion plasma.
The resistive barrier films were allowed to build up for about 30 minutes under the conditions described. The lead oxide film thus formed had a breakdown voltage of about 0.5 volt. Inasmuch as lead is widely used in thin film circuitry (cryotrons, tunneling devices and the like) the ability to form an insulation layer of lead oxide directly on the lead film takes on great significance and offers a materially improved technique in the formation of these thin film circuits.
The resistive barrier layer of silver oxide on silver film grew very thick and very rapidly in these experiments. The barrier of silver oxide thus formed had a breakdown voltage of about 2 volts and proved to be an electrical resistive film.
An electrically resistive barrier film was also formed on silicon by the process of this invention. In forming this barrier film a single silicon crystal was used and the silicon oxide was formed on one of its faces. The apparatus of FIG. 1 was used and an anodizing electrode was brought into contact with the surface of the silicon crystal. The oxygen pressure within the bell jar was maintained at 50 microns and after the priming layer of silicon oxide was formed the required electrical potential was applied to generate an oxygen ion-containing plasma, and the external electrical current to the anodizing electrode was started at 2 volts. As the current dropped off across the growing silicon oxide film, this externally applied voltage was increased by increments of between 2 and 3 volts. Such stepwise voltage increases were continued until the voltage across the resistive barrier film of silicon oxide reached 12 volts. This required about 20 minutes. By this process an electrically resistive barrier film was formed on a semiconductor material.
The ability to form an insulation layer of silicon oxide on a silicon body by the method disclosed herein is an important improvement over the present method of heating the silicon body at high temperatures for an extended period of time. By the metthod of this invention the sili con oxide is formed at room temperature in a relatively short time and within a vacuum system in which other steps in the construction of an integrated circuit may be accomplished.
In the use of the method of this invention in the construction of entire circuits or parts of circuits which are formed by vapor deposition techniques in apparatus similar or equivalent to that shown in FIGS. 1 and 4 it will be necessary to form the resistive barrier layer in only certain predetermined areas. As an example, if a cryotron is to be formed on a substrate 55 as illustrated in FIG. 10, the resistive barrier layer which is to serve as insulation between gate 50 and control 51 should preferably extend only over the crossover area 52 or just beyond it, leaving sufiicient surface areas on the gate 50 to electrically connect leads 53 which in turn may lead to soldering points 54. By the use of a stencil 56 (FIG. 11) having suitable patterns 58 cut in it and maintained adjacent to the surface of the electrically conducting body (cryotron gate 50 in FIGS. 10 and 11) it is possible to build up the resistive barrier layer 52 over the desired areas only. Although stencil 56 may come in direct. contact with the surface of gate 50, it may be desirable to maintain a small gap 5% between them to prevent scratching or other forms of marring of the surface of the electrically conducting body 50. The Width of this gap is not critical so long as it is less than the mean free path of the discharge gases in the evacuated atmosphere.
The effectiveness of the use of a stencil may be illustrated by the following example. A film of aluminum was deposited on a glass slide by vacuum deposition. A stencil in which a pattern (including very fine lines) had been cut out was placed adjacent the aluminum film surface and after a priming layer had been formed on the film, the stencil-covered film Was exposed to an oxygen plasma (50 microns pressure) for 30 minutes. Samples were made with and without the use of an anodizing electrode. Examination of all of the samples showed that the aluminum oxide electrical resistive barrier layer grew only in those areas of the aluminum film where the pattern was cut through the stencil exposing the film. Sharp lines off-demarcation defining the patterns were observed by means of an ellipsometer, an instrument described in the literature. (See, for example, An Ellipsometer for Film Measurement, Rev. Sci. Instr. 16:26 (1945).) The size and shape of the aluminum oxide pattern grown on the aluminum film corresponded exactly to the pattern cut in the stencil. Such a stencil is, of course, equally adaptable to the formation of patterns on surfaces other than aluminum.
A stencil may also be used in a manner to build up low resistance contact areas on the electrically conducting film prior to the formation of the electrically resistive barrier films in areas where they are required. Then, in the subsequent formation of the barrier films the stencil or mask need not be used. As an example, assume that it is desired to construct a cryotron in which aluminum is to be used as the gate element and lead as the control. A film of aluminum is deposited on a substrate in a pattern which consists of a strip having circular contact areas at each end. A stencil which covers the strip and exposes the circular areas is then placed adjacent the substrate and a low resistance metal film which is not subject to any appreciable anodization, e.g., gold or tin, is deposited on the exposed circular areas. These gold or tin film-covered areas are then available for making electrical contact. The stencil is removed; a layer of aluminum oxide is built up on the now exposed aluminum strip in accordance with the method of this invention; and then a lead gate film is deposited across the aluminum strip which has an insulating layer of aluminum oxide. This completes the construction of an electronic circuit area by a method which is entirely carried out in a vacuum system.
It will be evident from the above description and examples that the method of this invention provides the possibility of forming improved barrier films in greater variety. It also provides for forming improved circuit elements.
It will thus be seen that the objects set forth above, among those made apparent from the preceding description, are efficiently attained, and since certain changes may be made in carrying out the above method and in the constructions set forth without departing from the scope of the invention, it is intended that all matter contained in the above description or shown in the accom panying drawings shall be interpreted as illustrative and not in a limiting sense.
1. A method of forming an electrically resistive barrier film of controlled thickness, comprising the steps of (a) forming on the surface of an electrically conducting body of anodizable material a thin priming layer of the resistive barrier film to be formed, said priming layer being the product resulting from a chemical reaction between said body and a gaseous reactant;
(b) exposing said priming layer to a plasma of said gaseous reactant; and
(c) during exposure to said plasma, impressing an electrical potential across said priming layer by contacting said conducting body with an electrode and supplying to said electrode an electrical current to make said conducting body electrically positive with respect to said plasma, said potential being at least of a sufiicient magnitude to cause said priming layer to increase in thickness by further reaction between said conducting body and said gaseous reactant to form said resistive barrier film; said conducting body serving as the sole source of positive ions and initially providing no substantial support for said plasma, with said gaseous reactant serving as the sole source of negative ions for said further reaction; said negatively charged ions being of species which react with said conducting body to for ma dielectric film.
2. A method in accordance with claim 1 wherein said electrically conducting body is a metal.
3. A method in accordance with claim 2 wherein said metal is aluminum.
4. A method in accordance with claim 2 wherein said metal is beryllium.
5. A method in accordance with claim 2 wherein said metal is magnesium.
6. A method in accordance with claim 1 wherein said anodizable material is a semiconductor.
7. A method in accordance with claim 6 wherein said semiconductor is silicon.
8. A method in accordance with claim 1 including the step of placing a stencil adjacent to the surface of said electrically conducting body whereby said electrically resistive barrier is formed on said body in a pattern.
9. A method in accordance with claim 1 further characterized by the additional steps of (d) increasing said electrical potential to a higher level; and
(e) repeating step (d) until an electrical resistive barrier film of a desired thickness is formed.
10. A method in accordance with claim 1 wherein said gaseous reactant is oxygen.
11. A method in accordance with claim 1 wherein said gaseous reactant is diluted with an inert gas.
12. A method in accordance With claim 1 including the step of encapsulating at least said resistive barrier film thereby to prevent any appreciable change in the properties of said film.
13. A method in accordance with claim 1 including as a first step depositing said electrically conducting body in the form of a film on an electrically nonconducting substrate.
14. A method in accordance with claim 1 including the step of controlling the water vapor content of said atmosphere.
15. A method of forming an electronic circuit element incorporating an aluminum oxide film on an aluminum surface, comprising the steps of (a) depositing an aluminum film on an electrically nonconducting substrate;
(b) exposing the free surface of said aluminum film to an oxygen-containing atmosphere thereby to form a priming coat of aluminum oxide on said surface;
(0) exposing said priming coat of aluminum oxide to an oxygen plasma; and
(d) during exposure to said oxygen plasma, impressing across said priming coat an electrical potential by contacting said aluminum film with an electrode and supplying to said electrode an external electrical current to make said aluminum film electrically positive with respect to said plasma and initially providing no substantial support for said plasma, thereby to increase the thickness of said aluminum oxide to a predetermined value; said aluminum film serving as the sole source of aluminum ions and said plasma serving as the sole source of oxygen ions.
16. A method of forming a silicon oxide film on a silicon surface, comprising the steps of (a) exposing the surface of a silicon body to a gaseous atmosphere containing oxygen thereby to form on said silicon surface a priming coat of silicon oxide;
(b) exposing said priming coat of silicon oxide to an oxygen plasma; and
(0) during exposure to said oxygen plasma, impressing across said priming coat an electrical potential by contacting said silicon body with an electrode and supplying to said electrode an external electrical current to make said silicon body electrically positive with respect to said plasma and initially providing no substantial support for said plasma, thereby to increase the thickness of said silicon oxide to a predetermined value; said silicon body serving as the sole source of silicon ions and said plasma serving as the sole source of oxygen ions.
[7. A method of forming a circuit element incorporating an electrically conducting film having a dielectric film layer over a predetermined p0rtion thereof, comprising the steps of (a) forming a first film of an electrically anodizable conducting material in a pattern on a substrate;
(b) depositing a second film of a low electrical resistance material on that portion of the pattern of said first film over which said dielectric film layer is not desired;
(c) exposing said first film to a plasma of a gaseous reactant which reacts with said first film of electrically conducting material to form a dielectric film, but which is essentially unreactive with said second film of low resistance material, whereby a priming layer of said dielectric film is formed on the exposed area of said first film; said priming layer being the product resulting from a chemical reaction between the exposed surface of said first film and said gaseous reactant; and
(d) continuing said exposing and during exposure of said exposed surface of said first film to said plasma, impressing an electrical potential across said priming layer by contacting said first film with an electrode and supplying to said electrode an electrical current to make said first film electrically positive with respect to said plasma and initially providing no substantial support for said plasma, said potential being at least of a sufiicient magnitude to cause said priming layer to increase in thickness by further reaction between said exposed surface of said first film and said gaseous reactant to form said dielectric film; said first film serving aS the sole source of positive ions and said gaseous reactant serving as the sole source of negative ions for said further reaction.
13 '18. A method in accordance with claim 17 wherein said low electrical resistance material is gold.
References Cited UNITED STATES PATENTS 2,346,483 5/ 1944 Goss 204-192 2,955,998 10/1960 Berghaus et a1 204-177 14 3,035,205 5/ 1962 Berghaus et al. 204-192 3,077,444 2/ 1963 Hub 204-l92 3,108,900 10/1963 Papp 204192 OTHER REFERENCES APC application of Berghaus et al., Serial No. 283,312, published May v1943.
ROBERT K. MIHALEK, Primary Examiner.