Patent Application
20080090022
Referring now to the drawings and particularly to
The apparatus 10 as depicted in
The apparatus 10 further includes at least a first elongated conduit 24 of diameter d, wherein d is preferably between about 0.5 to 3.0 cm, extending through a side wall 12a into the interior of said evacuated enclosure 12. The first, hollow conduit 24 includes at least one portion, the distal end portion 24a thereof, having an aperture 26 formed therein. The first conduit means 24 and the aperture portion 24a thereof are adapted to, respectively, transmit and introduce an energy transferring gas from a source (not shown) into the interior of said evacuated enclosure 12, preferably to a point immediately adjacent apparatus adapted to provide activated species from said energy transferring gas. In the preferred embodiment depicted in
Regardless of the composition of the energy transferring gas employed, the aperture 26 formed at the distal end of the first conduit 24 must be capable of delivering a selected flow rate (as measured in terms of SCCM, i.e., standard cubic centimeters per minute) of said energy transferring gas. The flow rate is selected to provide a sufficient pressure of the energy transferring gas adjacent said aperture 26 for initiating the activation of said energy transferring gas at a power-pressure aperture size regime which is at the substantial minimum of the modified Paschen curve.
The first conduit 24 may further include means reducing the flow path for the energy transferring gas to create a “choke-condition” in the first conduit 24 adjacent the aperture 26 so as to provide a localized high pressure of the energy transferring gas. As used herein, the term “choke condition” refers to the condition which occurs when the speed of the energy transferring gas passing through the aperture 26 in the first conduit 24 reaches transonic speed. The choke condition generally is that condition which occurs in compressible gas or fluid flow when, for a conduit of a uniform size, the speed of the gas passing through said conduit reaches transonic velocity. It is at this condition that any rise in the flow rate of the energy transferring gas results in an increase in pressure rather than velocity. It is also this condition which defines the choke mode and it is precisely this effect which allows the pressure to be increased for operation at the substantial minimum of the Paschen curve. This localized high pressure creates a sufficient pressure zone for the energy transferring gas flowing through the first conduit adjacent the aperture portion 24a so that a plasma may be initiated and maintained at a power level which can be independently controlled. In an alternative embodiment, the pressure within the aperture portion 24a of the first conduit 24 may be easily controlled by employing a solenoid therewithin, which solenoid may be selectively constricted or relaxed so as to regulate the flow rate of energy transferring gas passing therethrough and hence the pressure which exists adjacent said aperture portion 24a.
Note that, and as will be specified in detail hereinafter, the activated species of the energy transferring gas forms a plume of pressure isobars adjacent the aperture portion of the first conduit 24, which plume defines an activation region of activated species. As will also be detailed hereinafter, the boundaries of the plume of activated species are specified by the pressure differential which exists between the gas flowing through the interior of the first conduit and the background pressure existing in the interior of the enclosure. As should be apparent, material which is sputtered from the surface of the first conduit 24 would degrade the quality of the activated species in the plume 34; and more importantly, the quality of the subsequently deposited thin film material would correspondingly deteriorate. Thus, a protective overcoat is preferably fabricated from a material which is resistant to a high temperature sputtering environment; or alternatively, a material which is relatively benign when incorporated into the ultimately deposited film. In a preferred embodiment, graphite is employed as the material from which the protective overcoat is fabricated. Graphite is not only highly resistant to high temperature and high temperature sputtering processes, but when the apparatus 10 is employed for the deposition of semiconductor alloy materials, graphite is substantially electrically benign to the desired characteristics of that deposited semiconductor film.
Returning now to
As clearly depicted in
In order to achieve the function of coupling the introduced microwave energy to the introduced energy transferring gas, the first conduit 24 extends through an aperture 30 formed in the side face 32 of the waveguide 28 for receiving the energy transferring gas. Accordingly, aperture 30 is adapted to facilitate the passage of the first conduit 24 and the energy transferring gas carried therewithin into an activation region 34 formed adjacent the aperture portion 24a of the first conduit 24 so that the plume of activated species extends from the interior of said applicator 28.
The radiant microwave applicator 28 further includes cut-away section 36 formed in the face 35 thereof opposite the face 32 in which the aperture 30 is formed. The cut-away section 36 has a diameter larger than the diameter of the aperture 30 and preferably at least about 2 inches so as to provide for the movement of expanding pressure isobars of the plume of activated species without having those activated species degrade too much of the microwave applicator material. It should therefore be understood that the applicator cut-away section 36 is adapted to provide a means of directed escape for the activated species of the energy transferring gas from within said applicator 28. The microwave applicator 28 further includes a closed end plate 40 to prevent the escape of unused microwave energy into the interior of the evacuated enclosure 12. It is to be noted that the maximum size of the cut-away section 36 of face 35 of the applicator 28 will be determined by balancing the facts that (1) the smaller the opening is made, the more material therefrom will be etched away, but the more the microwave energy is confined; while (2) the larger the opening is made, the less material is etched therefrom, but the more the microwave energy leaks into the enclosure. The cutaway section 36 may further include a microwave absorptive or reflective screen or other means adapted to prevent the microwave energy from entering the enclosure. This becomes particularly significant as the pressure differential between the background pressure and the pressure of the energy transferring gas in the first conduit is reduced to approach the aforementioned factor of at least 5.
The deposition apparatus 10 further includes at least one remotely located, generally planar substrate 50 operatively disposed within the enclosure 12 and at least spaced a distance from the activation region 34 sufficient to prevent said thin film material depositing thereupon from direct exposure to the electrons present in that region. The apparatus further includes at least one electrically biased screen or mesh 70. The screen(s) or mesh(s) are disposed between the energy transferring gas activation region 34 and the collision region 65. The screen 70 is electrically biased. The bias may be any of 1) a positive bias to repel the ionic species with the plume of activated species as it passes therethrough, 2) a negative bias to attract and neutralize the charged ionic species, or 3) a plurality of screens with opposite biases. The electrically biased screen 70 also acts (along with the positive ions) to attract the electrons within the plume and insure that they do not reach collision region 65. The screen 70 is spaced far enough from the plasma activation region 34 so as to insure that the screen is not destroyed by the plasma. The screen is also made of a material that is resistant to the effects of the plasma. Preferred materials include graphite, tungsten, nickel and nickel plated materials. The screen 70 is also spacedly disposed from the collision region 65 such that any electrons which pass through the screen before they are captured thereby do not impinge upon the collision region, as such could create undesirable species in the depositing materials. The Apparatus may further include a plurality of meshes or screens, each one providing an additional degree of separation (fractionation) of the charged species from the neutral species within the plume of activated species.
The apparatus 10 may further preferably include means 52 adapted to heat and or apply an electrical or magnetic bias to the substrate 50. It is to be understood, however, that the use of heat or a bias is not required to practice the invention disclosed herein. In a preferred embodiment, the substrate 50 is operatively disposed so as to be substantially aligned with the first conduit 24 so that a flux of the activated species generated in the activation region 34 can be directed thereat for deposition thereupon.
In a preferred embodiment, the deposition apparatus 10 may also be equipped with a second elongated, hollow conduit 60 having at least one aperture 62 formed at the distal end 60a thereof. The apertured end 60a of the second conduit 60 extends through the top wall 12b of the enclosure 12 and into the interior thereof so that the aperture 62 terminates in close proximity to said substrate 50. The second conduit 60 is adapted to deliver a flow of a precursor deposition gas from a source (not shown) into a collision region 65 which is created adjacent said substrate 50. The precursor deposition gas is typically selected from the group consisting essentially of a silicon-containing gas, a germanium-containing gas, a carbon-containing gas, and combinations thereof. Specific examples of preferred precursor deposition gases include, but are not limited to, SiH4, SiF4, Si2H6, GeH4, Ge2H6, GeF4, CH4, C2H6, and combinations thereof.
As previously mentioned, the precursor deposition gas is introduced by the second conduit 60 into the collision region 65. The collision region 65 is disposed in the path of travel of the neutral free radicals of the activated species of the energy transferring gas as those activated species are directed from said activation region 34 toward the substrate 50. In this manner neutral free radical activated species from the activation region 34 are directed towards the collision region 64 where said species collide and interact with said precursor deposition gas so as to create a desired energized deposition species. It is to be noted that the collision region 64 should be disposed a distance from said substrate 50 selected so that the desired deposition species created in the collision region 64 will uniformly deposit over the entire surface of the substrate 50 without encountering multiple collisions with either other activated species or other deposition species formed in the collision region. It should also be noted that as the pressure changes from the activation region to the collision region, so does the mean-free-path length of the activated species. The path length increases as the pressure decreases such that a plasma can be formed in the activation region and cannot be formed in the collision region. In a preferred embodiment, the background pressure to which the enclosure 12 is evacuated provides for a mean free path for the free radical deposition species of approximately 1-15 cm. Therefore, by spacing the substrate a distance of 1-15 cm from the collision region, the entire surface thereof will be covered with a uniform thin film of material.
Finally, with reference to
While the apparatus described hereinabove has been designed so as to be particularly adapted to carry out the principles of the instant invention, it is to be understood that other modified embodiments of this apparatus may be used with equal advantage and it is the practice of the method rather than the specific apparatus which defines the true scope of the instant invention. It will, however, be necessary to make periodic references to the aforedescribed apparatus in order to more clearly explain the practice of the operative concepts of the method disclosed herein.
In the most general terms of one embodiment of the instant invention, there is disclosed a novel method of generating activated species from an energy transferring gas in an activation region 34 which is located interiorly of the evacuated enclosure. The ions of the activated species of the energy transferring gas may then be separated from the neutral free radicals thereof. The neutral free radicals are then directed to collide with a precursor deposition gas in a collision region 65 so as to yield a high density of only desired energized deposition species, which energized species react with the exposed surface of a substrate spacedly located relative to the activation region.
In operation, the method of the subject invention is carried out in a evacuated enclosure or chamber of the type generally described with reference to
As specifically described, there is introduced into the interior of the evacuated enclosure, inter alia, the energy transferring gas, which introduction is accomplished by a first conduit preferably having at least one aperture formed at the distal end thereof. The aperture is typically dimensioned to have a diameter of between about 0.25 to 3.0 cm and may be equipped with a solenoid operated aperture reduction structure and/or a protective overcoat, both of which elements have been fully described hereinabove.
The energy transferring gas is typically selected from the group consisting of the noble gases, hydrogen, methane, etchant gases, and combinations thereof. In a preferred embodiment, the energy transferring gas is helium, which when appropriately excited, yields long-lived neutral activated species. Additionally, the energy transferring gas (or alternatively the precursor deposition gas) may include a number of diluent or dopant gases including, but not limited to, O2, NH3, N2, NH4, H2, CH4, PH3, PH5, BF3, BF5, B2H6, BH4, and combinations thereof. The function of the diluent or dopant gas is to provide a source of an element to be incorporated into the deposited film. For example, if a thin film of silicon/germanium alloy material is being deposited upon the spacedly disposed substrate, the film may be rendered slightly p-type by the addition of small amounts of a p-type dopant, such as BF3, into the flow of the energy transferring gas (or deposition precursor gas). Alternatively, in the deposition of a layer of insulating SiOx material, it will be necessary to include small amounts oxygen in the stream of the energy transferring gas. Regardless of composition, it is important that the energy transferring gas be delivered through said first conduit to a point immediately adjacent the activation means, such as directly into the interior of the radiant microwave applicator discussed hereinabove.
It should be noted that the energized species formed from the activation gas and/or the deposition precursor gas may be all of or some portion of the total gas molecule used. For example if SiH4 is used as one of the deposition precursor gases, then the activated species thereof may include activated forms of SiH4, SiH3, SiH2, SiH, Si, and H, not all of which will form or be desirable depositing species.
While the activation energy may be selected from the group consisting of a.c. energy, d.c. energy, r.f. energy, microwave energy, photoactivation energy, and combinations thereof. In fact, any electromagnetic energy from 0 Hz to 5 Ghz could be used to activate the energy transferring gas. In a preferred embodiment, the activation energy is microwave energy and the activation means is a radiant microwave applicator, such as the aforementioned radiant microwave waveguide which extends into the interior of the evacuated enclosure. The radiant microwave applicator is adapted to provide about 1-10 kilowatts of microwave power at a frequency of 2.45 GHz. The energy transferring gas is delivered from the aperture in the first conduit through an opening hole formed in the side wall of the microwave applicator into an activation region located at least partially within the hollow interior of said applicator. The activation region is clearly defined as that region in which an ionized plasma of activated species of the energy transferring gas is formed. The high flow rate of the energy transferring gas exiting the first conduit relative to the background pressure within the enclosure also defines a series of pressure isobars which serve to limit the volume occupied by the ionized plasma of the activated species of that energy transferring gas.
The energy transferring gas is preferably delivered by the first conduit at a flow rate of at least about 100 SCCM, and more preferably approximately between 100-2000 SCCM. In this way, it is possible to maintain a preferred pressure differential of at least about a factor of five times difference between the background pressure that exists within the interior of the enclosure (less than about 50 torr, and preferably 0.1-10 mtorr) and the pressure of the energy proximate the aperture of the first conduit (which pressure may be as high as about 10-30 torr). It should be apparent that the pressure within any given isobar decreases with distance away from the aperture in the firs conduit. Therefore, at any given power, the slope of the Paschen curve will provide a pressure-determined boundary of the activation region.
After evacuating the enclosure, applying the electromagnetic field of microwave energy by means of the radiant microwave applicator and introducing a sufficient flow of the energy transferring gas (which, in conjunction with the background pressure in the enclosure and the power of the electromagnetic field), ignite a plasma of activated species from the high density of the energy transferring gas residing within the activation region; activated species of the energy transferring gas travel towards the collision region 65 illustrated in
It is to be recalled that the location of the collision region is defined by a second conduit having at least one aperture disposed at the distal end thereof, which distal end of the conduit extends into the interior of the evacuated enclosure and terminates in close proximity to the substrate. The function of the second conduit is to deliver the precursor deposition gas into the collision region 65 so that the neutral activated species of the energy transferring gas may interact with the precursor gas to yield an energized deposition species. Preferred precursor deposition gases delivered to the collision region include, but are not limited to, silicon containing gases, germanium containing gases, carbon containing gases, tin containing gases and combinations thereof. The precursor deposition gas is typically delivered into the collision region at a flow rate of at least about 10 SCCM and preferably between about 10 and 200 SCCM, with a preferred flow rate of between about 25 and 100 SCCM, and a most preferred flow rate of approximately 40 SCCM.
In the collision region, activated species of the energy transferring gas (i.e., helium) interact with the precursor deposition species (e.g., silane) in the following manner:
He*+SiH4->SiH3*+H+He
The resulting SiH3*, H and He then migrate from the collision region and the neutral free radicals are deposited upon the exposed surface of the substrate, which substrate is located a distance from the collision region that is within the length of the mean free path of the energized deposition species. The He, because it is a non-excited neutral species, is now a harmless by product, does not cause ionic bombardment of the depositing material or in any other way deleteriously effect the growing film. The mean free path of the energized deposition species is the distance that the energized deposition species may travel without colliding with either other deposition species (such as free radicals) or encountering a second collision with activated species of the energy transferring gas. The mean free path of the energized deposition species should be approximately the length of the longest dimension of the substrate so that the uniform deposition of the energized deposition species over the entire exposed surface of the substrate is assured.
As is well known to those skilled in the art, the length of the mean free path of the energized deposition species is directly dependent upon the background pressure existing within the evacuated enclosure through which those species will diffuse. For example, if the background pressure within the enclosure were relatively high, i.e., approximately one torr, the mean free path of the energized deposition species would be quite short (on the order of 1 mm or less). If on the other hand, the background pressure of the evacuated enclosure is maintained at substantially sub-atmospheric levels, as approximately 1 millitorr, the mean free path of the energized deposition species will be considerably longer, on the order of 5-10 cm. It can thus be appreciated that the size of the substrate upon which the energized deposition species is to be deposited, will be one of the critical factors used in determining the length of the mean free path, and correspondingly, the background pressure at which the evacuated enclosure must be maintained.
It should be equally apparent that once the background pressure of the evacuated enclosure is determined, and the deposition rate is selected, each of the other parameters critical in initiating a plume of a given volume of activated species of the energy transferring gas are likewise determined. As was discussed hereinabove, as a rule of thumb, the flow rate of the energy transferring gas through the first conduit must be sufficient to create a pressure differential of at least about a factor of five between the energy transferring gas and the background pressure of the evacuated enclosure for the uniform deposition of energized deposition species onto a substrate having a surface area of about 100 square cm. Thus, knowing the required length of the mean free path, the background pressure may be selected, and this background pressure dictates the possible range of flow rates at which the energy transferring gas must be introduced into the enclosure in order to maintain a significant pressure differential between the background pressure and the flow within the first conduit. This also determines the size aperture which will provide transonic flow at that flow rate of energy transferring gas.
Taking the usable range of flow rates of the energy transferring gas for a given background pressure, it then becomes possible to use the Paschen curve to determine an optimized power/pressure regime in which to operate for a given volume of activated species. At a given power/pressure/aperture size regime, it is possible to determine the approximate percentage of the energy transferring gas which has been excited to form the activated species, (the typical range is 1-5% of, for example, He being promoted to the He* activated species). Knowing this percentage allows the operator to regulate the flow rate of the precursor deposition gas into the collision region, thereby maximizing the ratio of He* to precursor gas molecules to avoid the possibility of multiple collisions between the energized deposition species and the precursor deposition gas.
It should noted that the collision region may consist of a unique form of plasma. The unique plasma is a filtered, neutralized plasma which is formed of a volume of gaseous species including energized and non-energized neutral gaseous species but has a substantially reduced fraction of ionized gaseous species. Preferably the fraction of ionized gaseous species may be reduced by at least 50%, more preferably by at least 75% and most preferably by at least 90%.
The method and apparatus of the instant invention will be capable of creating a semiconductor material having reduced density of subgap defects. The material may be silicon or a silicon alloy. The material may further optionally include one or more band gap adjusting elements. The material preferably has a non-single crystal microstructure and has fewer than 5.0×1015/cm3 subgap defects. More preferably, the semiconductor material will have fewer than 1.0×1015/cm3 subgap defects, most preferably fewer than 5.0×1014/cm3 subgap defects.
It should be recognized that the foregoing description and discussion are merely meant to illustrate the principles of the instant invention and not meant to be a limitation upon the practice thereof. It is the following claims, including all equivalents, which are meant to define the true scope of the instant invention.
