CN1180447A - Method and apparatus for material deposition using primer - Google Patents

Abstract

A liquid primer is misted, flowed into a deposition chamber (2) and deposited on a substrate (5). A liquid precursor (64) is misted, flowed into a deposition chamber (2) and deposited on the substrate (5). The primer and precursor are dried to form a solid thin film, which is then annealed to form a part of an electronic component (1112) in an integrated circuit (1110), such as the dielectric (1130) in a memory cell. The primer is a solvent, and the precursor includes a metal carboxylate, a metal alkoxide, or a metal alkoxycarboxylate in a precursor solvent. Preferably, the primer and the precursor solvent are the same solvent, such as 2-methoxyethanol, xylenes, n-butyl acetate or hexamethyl-disilazane.

Description

Method and apparatus for material deposition using primer

Background

1. Field of the invention

The present invention relates to a method for depositing high quality films of composite (compound) materials on a substrate, and to an apparatus for carrying out such a method. More particularly, the invention relates to the fabrication of integrated circuits by applying a liquid precursor to a wafer and then drying the applied liquid to form an integrated circuit element.

2. Description of the related Art

Prior art methods for depositing thin films of complex compounds such as ferroelectrics and metal oxides for high dielectric constant applications in integrated circuits include: vacuum evaporation (i.e., electron beam, laser ablation, etc.); vacuum sputtering (i.e., electron beam, d.c., r.f., ion beam, etc.); treating gold with powder; reactive chemical vapor deposition, including metalorganic chemical vapor deposition (MOCVD); and liquid application methods using sol-gels (alkoxides) or carboxylates. However, none of these known methods is capable of producing metal oxides with sufficiently good characteristics for use on integrated circuits. For example, metal oxides produced for ferroelectric applications can fatigue very quickly and have excessive leakage currents for high dielectric constant applications. Furthermore, none of the prior art methods, other than sputtering, can produce films that are sufficiently thin for use in integrated circuits, and the resulting films have significant physical defects, such as cracking, peeling, and the like. With prior art methods, particularly sputtering, it is not possible to reliably and repeatedly produce metal oxides having a particular stoichiometric composition within the tolerances required for integrated circuits. Certain processes, such as CVD, may be hazardous or toxic. All methods require high temperatures, which can damage the integrated circuit and provide poor "step coverage" to the substrate to be covered; that is, the prior art results in a significant deposit build-up on the boundary at any break on the substrate. In prior art liquid deposition methods, it is not possible to control the thickness with the accuracy required for integrated circuit fabrication. As a result, until now, metal oxides have not been used in integrated circuits except for one or two special relatively expensive applications, such as the use of sputtered PZT in ferroelectric integrated circuits, but which are expected to have short operating lifetimes.

Recently, some us inventors have described an aerosol deposition method and apparatus for producing thin films for integrated circuit applications, see us patent NO,5456,945, published as 10/1995. While the process described therein provides a large improvement over prior art processes in thin films of about 1000 angstroms or less, it creates serious problems in step coverage and film quality. Because thinner films allow for more compact integrated circuits, it would be highly desirable to have a method and apparatus for producing high quality thin films of complex compounds, such as metal oxides, having thicknesses less than 1000 angstroms for integrated circuits and other applications.

Summary of The Invention

The present invention overcomes many of the problems and disadvantages associated with known methods by incorporating a primer deposition step into the method described in US patent No.5,456,945. The primer is applied prior to or simultaneously with the deposition of the precursor. A primer mist generator is incorporated into the apparatus.

The present invention provides a method of manufacturing an integrated circuit, the method comprising the steps of: (a) providing a liquid primer; (b) providing a liquid precursor; (c) placing a substrate into an enclosed deposition chamber; (d) generating a primer mist of the liquid primer; (e) flowing a primer mist through the deposition chamber to form a primer liquid layer on the substrate; (f) generating a precursor mist of the liquid precursor; (g) flowing the precursor mist through a deposition chamber to form a precursor liquid layer on the substrate; (h) treating each liquid layer deposited on the substrate to form a film of solid material; and (i) completing the fabrication of the integrated circuit such that elements of the integrated circuit comprise at least a portion of a solid material film. Preferably, the liquid primer includes a coating solvent selected from the group consisting of: 2-methoxy ethanol, dimethylbenzene and n-butyl acetate. Hexamethyl-disiloxane (HMDS) primers have also proven useful. Preferably, the precursor comprises a metal compound in a precursor solvent, the metal compound being selected from the group consisting of: metal alkoxides and metal carboxylates, and metal alkoxycarboxylates. Preferably the precursor solvent is the same as the coating solvent. Preferably, the step of flowing each mist into the deposition chamber is performed while maintaining the substrate at ambient temperature and maintaining a vacuum in the deposition chamber. Preferably, the step of flowing the primer mist into the deposition chamber and the step of flowing the precursor mist into the deposition chamber are performed simultaneously. Preferably, the vacuum is between about 100 torr and 800 torr. Preferably, the method further comprises the step of filtering the primer mist prior to the flowing step. Preferably, the flowing step includes injecting the primer mist into the deposition chamber proximate to and around one side of the substrate and discharging the primer mist from the deposition chamber in a region proximate to and around a periphery of an opposite side of the substrate to form a substantially uniformly distributed stream of primer mist on the substrate. Preferably, the method includes the additional step of mixing a plurality of different primer mists outside the deposition chamber to form a primer mist mixture and flowing into the deposition chamber. Preferably, the method further comprises the additional step of applying ultraviolet radiation to one of the primer mist and the precursor mist as the mist flows through the deposition chamber. Preferably, the treating step comprises applying ultraviolet radiation to one of the primer layer and the precursor layer deposited on the substrate. Preferably, the step of generating the primer mist includes ultrasonically vibrating a quantity of the liquid primer to form the primer mist. Preferably, the step of ultrasonically vibrating includes adjusting the particle size of the primer mist by controlling one of the frequency and the amplitude of the ultrasonic vibration. Preferably, the step of generating the precursor mist comprises ultrasonically vibrating a quantity of the liquid precursor to form the precursor mist, the step of ultrasonically vibrating comprising adjusting the particle size of the precursor mist by controlling one of the frequency and the amplitude of the ultrasonic vibration. Preferably, the step of treating comprises one or more steps of a set of steps of drying, heating and annealing the layer deposited on the substrate. Preferably, the treating step comprises drying the liquid primer and liquid precursor layers deposited on the substrate. Preferably,the step of drying includes maintaining a sub-atmospheric pressure in the deposition chamber.

In another aspect, the present invention provides a method of manufacturing an integrated circuit, the method comprising the steps of: (a) providing a liquid primer; (b) providing a liquid precursor; (c) placing a substrate into the closed deposition chamber; (d) generating a primer mist of the liquid primer; (f) generating a precursor mist of the liquid precursor; (e) flowing each mist into a deposition chamber to form a liquid mixture of primer and precursor on the substrate; (h) treating the liquid mixture deposited on the substrate to form a film of solid material; and (i) completing the fabrication of the integrated circuit such that elements of the integrated circuit comprise at least a portion of the solid material film.

In yet another aspect, the present invention provides an apparatus for manufacturing an integrated circuit, the apparatus comprising: (a) a deposition chamber; (b) a substrate holder located in the deposition chamber; (c) means for generating a mist of the liquid primer coating; (d) means for generating a mist of liquid precursor; (e) means for flowing the primer mist and the precursor mist into a deposition chamber to form liquid layers on the substrate, the liquid layers comprising a primer liquid and a precursor liquid; and (h) means for treating the liquid layer deposited on the substrate to form a film of solid material on the substrate.

The use of a primer prior to the deposition of the metal oxide can produce a metal oxide film exhibiting better surface morphology and a dielectric with low leakage current. An excellent quality composite compound film can be produced with a thickness that is half or even one-third the thickness of the thinnest high quality film that can be produced with the prior art methods and apparatus. Further objects, advantages and features of thepresent invention will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings.

Brief description of the drawings

FIG. 1 is a cut-away side view of a deposition chamber portion of an aerosol deposition system according to the present invention;

FIG. 2 is a plan view of the inlet and outlet nozzle assemblies of the system of FIG. 1;

FIG. 3 is an enlarged plan view of an input nozzle of the system of FIGS. 1 and 2;

FIG. 4 is a schematic side view of a mist generator of the aerosol deposition system according to the present invention;

FIG. 5 is a schematic plan view of a buffer chamber and associated inlet and outlet portions according to the present invention;

FIG. 6 is a flow chart illustrating a method of manufacturing an integrated circuit according to the present invention;

FIG. 7 is a top view of a preferred embodiment of an atomized deposition system according to the invention;

FIGS. 8 and 9 show the shutter assembly and substrate in two different positions to illustrate the adjustable relationship between the shutter and the substrate;

FIG. 10 is a perspective view showing placement of an ultraviolet radiation source within a deposition chamber according to the invention;

FIG. 11 shows a cross-sectional side view of a portion of an integrated circuit wafer made using the method and apparatus of the present invention;

FIG. 12 is a plot comparing the film thickness, dielectric constant, and leakage current of BST thin films for eachBST film made with and without a primer;

fig. 13 is an electron micrograph of a cross section of a BST capacitor fabricated using a primer; and

fig. 14 is an electron micrograph of a cross section of a BST capacitor fabricated without the use of a primer.

Description of the preferred embodiments 1, overview

A flow chart of a preferred embodiment of a method according to the invention is shown in fig. 6 and manufactured by the methodThe set ofA portion of the integrated circuit is shown in fig. 11. At step P1, a substrate 5 is provided. In the art, the term "substrate" is used in a generic sense to mean any one or more layers of material 5 upon which is deposited a interest layer 1130, and in a specific sense to mean a silicon wafer 1122 upon which is formed an integrated circuit 1110. Unless otherwise indicated herein, the term "substrate" will be used herein to refer to any object on which a layer of material is deposited using the methods and apparatus of the present invention. The substrate provided in step P1 preferably comprises a P-type silicon wafer 1122. At step P2, initial integrated circuit layers 1124, 1126 and 1128 are fabricated to form a substrate 5 on which a metal oxide layer 1130 is deposited. First, an approximately 5000 angstroms of silicon dioxide insulating layer 1124 is wet grown. Typically, the SiO₂The layers are etched to form the desired shape, which, once the appropriate titanium layer 1126, platinum layer 1128, dielectric layer 1130, and platinum layer 1132 are deposited, produces a given integrated circuit device 1112. Bottom electrode 1127 comprises a thin layer 1126 of titanium metal deposited on silicon dioxide 1124, preferably by in situ sputtering, and a 2000 angstrom thick platinum electrode deposited on titanium 1126, preferably by in situ sputtering. By"in-situ" is meant that both titanium and platinum are sputtered without breaking the vacuum. Titanium layer 1126 is optional. In use, it diffuses into the silicon dioxide and platinum and helps the platinum 1128 adhere to the silicon dioxide 1124. A layer of material 1130, such as PZT or BST, is then deposited using the method and apparatus of the present invention as will be discussed below. Another 2000 a layer of platinum is deposited over layer 1130. The wafer 1110 was then annealed and patterned using a photomask process and etched to the electrode layer 1128 to produce capacitive integrated circuit devices 1112, one of which is shown in cross-section in fig. 11, which were tested by connecting one lead of the test apparatus to the platinum electrode layer 1128 and contacting a thin probe connected to the other lead of the test apparatus to the other electrode layer 1132.

A primer was prepared at step P6. In the preferred embodiment, this step includes providing an amount of a single solvent, such as 2 methoxyethanol, xylenes, n-butyl acetate, or Hexamethyldisiloxane (HMDS), but it may also include the step of combining several solvents, such as combining two or more of the four solvents described above. Such preferred solvents, whether a single solvent or a combination of solvents, are the final solvents of the precursor, that is, the solvent of the precursor applied in step P22 will be described below. Certain solvents that may be used as primers along with their boiling points include: alcohols, for example 1-butanol (117 ℃), 1-pentanol (117 ℃), 2-pentanol (119 ℃), 1-hexanol (157 ℃), 2-hexanol (136 ℃), 3-hexanol (135 ℃), 2-ethyl-1-butanol (146 ℃), 2-methoxyethanol (124 ℃), 2-ethoxyethanol (135 ℃) and 2-methyl-1-pentanol (148 ℃); ketones, such as 2-hexanone (methyl, butyl (methyl) ketone) (127 ℃), 4-methyl-2-pentanone (methyl isobutyl (methyl) ketone) (118 ℃), 3-heptanone (butyl ethyl (ethyl) enone) (123 ℃), and cyclohexanone (156 ℃); esters, such as butyl acetate (127 ℃), 2-methoxyethyl acetate (145 ℃) and 2-ethoxyethyl acetate (156 ℃); ethers such as 2-methoxyethyl ether (162 ℃ C.) and 2-ethoxyethyl ether (190 ℃ C.); and aromatic hydrocarbons such as xylene (138 ℃ to 143 ℃), toluene (111 ℃), ethylbenzene (136 ℃) and Hexamethyldisiloxane (HMDS) (125 ℃).

At step P8, a primer is applied to the substrate 5. In the preferred embodiment, the primer is atomized, screened with a mesh screen filter 310 and applied to the substrate 5 in a deposition chamber 2, as described in detail below. The term "mist" as used herein is defined as liquid particles carried by a gas. The term "mist" includes an aerosol, which is generally defined as a colloidal suspension of solid or liquid particles in a gas. The term "mist" also includes vapors, fog, and additional spray suspensions of precursor solvents in gases. Since each of the above words is taken from the usage of common words. The definitions are not necessarily precise, overlapping and may be used differently by different authors. The term "aerosol" is intended herein to include all suspensions encompassed by the "aerosol science and technology" written by Parker C.Reist, Mc Graw-Hill, New York, 1983, which is incorporated herein by reference. The term "mist" as used herein is intended to be broad in meaning to the term "aerosol" and includes suspensions not included under the terms aerosol, vapor, fog. Ultraviolet (UV) radiation is applied to the primer mist as it flows into and through the deposition chamber 2 or after it is applied to the substrate 5, as shown by dashed lines P11 and P12, respectively. However, in the preferred embodiment, these two steps P11 and P12 are skipped.

It has been found that the use of a primer prior to deposition of the precursor produces a film with better surface morphology and lower leakage current than the film produced without the primerstep P8.

A precursor liquid is prepared at step P20. The precursor is preferably a metal alkoxycarboxylate as described in U.S. patent No.5,514,822 issued 5/7 1996, a detailed example of which is given below. The precursor prepared in step P20 is typically prepared in bulk and stored until needed. Immediately prior to the application of the precursor, a solvent exchange step, a concentration adjustment step, or both, will be performed to provide the optimal precursor for application. The solvent exchange step is described in detail in U.S. patent application serial No.08/165,082. The final precursor solution is preferably used as the sole source for the entire deposition process following the application of the primer. However, the present invention also contemplates the use of multiple precursor sources in parallel or in series. In particular, it is also possible to use further sources in parallel for doping or modifying the finally desired compound.

The precursor liquid used in the present invention is a stable solution. By "stable" is meant herein that the critical oxygen-metal bond of the desired final chemical compound is formed during the precursor formation process and is stable after such formation. This is two-fold, the first being that the solution does not react or deteriorate when stored over a reasonably long period; in a second aspect, the bonds formed in forming the precursor remain stable throughout the deposition process and at least form part of the bonds in the final desired chemical compound. That is, the metal-oxygen bond in the precursor remains stable and passes through the deposition process to form the metal-oxygen bond of the final desired metal-oxygen compound.

According to the method of the invention, the mist filtered by the precursor liquid flows uniformly into and onto the substrate 5 at ambient temperature. The ambient temperature here refers to the temperature of the surroundings. That is, no additional heat is applied to the substrate other than from the environment. When UV rays are applied, the ambient temperature will be higher than the chamber temperature, and when the substrate is processed without applying UV rays but with vacuum applied, the ambient temperature will be lower than the room temperature. Based on the above, typically the ambient temperature may be between about-50 ℃ and 100 ℃. Preferably, the ambient temperature is between about 15 ℃ and 40 ℃.

As will be described below, a key aspect of the inflow process is that the mist flows through the substrate 5 via a plurality of inlet ports and is exhausted from the area above the substrate 5 via a plurality of exhaust ports, and the ports are distributed proximate or around the periphery of the substrate 5 so as to produce a substantially uniformly distributed mist flow through the substrate 5.

During, after, or both during and after deposition, the precursor liquid is processed to form a thin film of solid material on the substrate 5. In this context, "treating" means any one or combination of the following: exposure to vacuum, ultraviolet irradiation, electrode formation, drying, heating and annealing. In the preferred embodiment, UV radiation is applied to the precursor solution during deposition at step P24. It is also preferred to apply ultraviolet radiation after the deposition at step P28. After deposition; the material deposited on the substrate 5, which in this preferred embodiment is liquid, is also preferably exposed to vacuum for a certain time, then heated and then annealed. The chemistry of the UV treatment process is not fully understood. It is believed that UV assists in the dissociation of the metal oxide particles or other elements including the desired final chemical compound from the solvent and the organic or other fragments of the precursor compound.

An important parameterof many synthetic thin films, such as ferroelectric films, is that they are required to be very thin (e.g., in the range of 200 to 5000 angstroms). Such a film thickness can easily be obtained by the method and the device according to the invention. The invention can also be used to produce much thicker films if desired.

The invention is particularly suitable for depositing high quality thin films of compounds such as those of ferroelectrics, superconductors, materials with high dielectric constants, and gemstones. For exampleThe invention can be used for depositing a material having ABO₃Comprises PbT1O₃，Pb_xZr_yTiO₃，Pb_xLa_yZr_zTiO₃And YMnO₃Wherein Y represents any rare earth element. In addition, the invention can also be used for depositing barium strontium titanate [ (Ba, Sr) TiO₃]Strontium titanate (SrTiO)₃) And additionally thin films of multielement compounds, such as described in U.S. patent No.5,519,234 issued 5/21 1996.2. Deposition apparatus

As shown in fig. 1, a thin film deposition apparatus according to an exemplary embodiment of the present invention is shown and generally indicated by reference numeral 1. The apparatus 1 includes a deposition chamber 2 comprising a substrate holder 4, a baffle 6, an input nozzle assembly 8, an exhaust nozzle assembly 10, and an ultraviolet radiation source 16. The deposition chamber 2 includes a body 12 and a lid 14 securable to the body 12 to define an enclosed space within the deposition chamber 2. The chamber is connected to a plurality of external vacuum sources to be described below. The lid 14 is pivotally connected to the body 12 using a hinge indicated at 18. In operation, the mist and inert carrier gas are fed through tube 45 in direction 43 and through feed nozzle assembly 8, from where the mist will be deposited onto substrate 5. Excessmist and carrier gas are drawn out of the deposition chamber 2 through the exhaust nozzle 10.

The substrate holder 4 consists of two circular plates 3, 3 'of an electrically conductive material, for example stainless steel, the top plate 3 being insulated from the bottom plate (field plate) 3' by an electrically insulating material 7, for example "delrin". In one exemplary embodiment, a 5 inch diameter substrate 5 is used, and the substrate holder 4 is typically 6 inches in diameter and is supported on a rotating shaft 20, which shaft 20 is in turn connected to a motor 18 to enable rotation of the holder 4 and substrate 5 during the deposition process. The insulating shaft 22 electrically insulates the substrate holder 4 and substrate 5 supported thereon from a DC voltage applied to the deposition chamber body 12, thereby creating a DC bias between the substrate holder 4 and baffle plate 6 (through the chamber body 12). The DC bias can be used, for example, to polarize the electric field of the thin film as it is being deposited onto the substrate 5. The insulated shaft 22 is connected to the shafts 20 and 20' by couplings 21. The power supply 102 is operatively connected by a wire 106 to a connection point 108 on the body 12 of the deposition chamber 2 and by a wire 104 through the feed-through 23 to the brass collar 25 to generate a DC bias between the field plate 3' and the baffle 6.

Baffle plate 6 is formed of an electrically conductive material, such as stainless copper, and is of sufficient size to extend substantially parallel over substrate 5 so that the vapor source or mist emitted through inlet tube 26 and nozzle assembly 8 is forced to flow between baffle plate 6 and substrate holder 4 onto substrate 5. The baffle 6 preferably has the same diameter as the substrate 5. It has been found that the best results are obtained if the area of the baffle 6 in the plane parallel to the substrate varies by 10% or less withrespect to the area of the substrate 5. That is, the area of the baffle 6 is at most 10% larger than the area of the substrate 5 or at most 10% smaller than the area of the substrate 5. The shutter 6 is preferably connected to the lid 14 by a plurality of rods 24 as shown in fig. 1, so that the shutter 6 will move away from the substrate 5 when the lid is opened.

Fig. 8 and 9 show the spacers 6 located at various distances from the substrate holder 4. Typically, each rod 24 is a stainless steel rod attached to the deposition chamber lid 14. Each rod 24 is bored so as to receive a screw 35 (fig. 1) with which the rod 24 is connected to the baffle 6. Each rod 24 is apertured to receive an adjustment screw 36 which secures the threaded rod 35 to the rod 24. By loosening the adjustment screws 36, repositioning the rods 24 relative to the threaded rods 35, and then tightening the adjustment screws 36, the effective length of each rod can be adjusted to up to 1/2 inches without requiring the rods 24 to be removed from the chamber lid 14. Each of the rods 24 is removable, which allows multiple sets of rods of different lengths L, L ', etc. to be replaced to coarsely adjust the respective distances S, S', etc. between the shutter plate 6 and the substrate holder 4 (and substrate 5) depending on the source material, flow rate, etc. For example, the rod length L may be adjusted to provide a distance S in the range of 0.10-2.00 inches. Once in position, the rod 24 may also be adjusted as described above. The rod 24, the threaded rod 35 and the adjusting screw 36 thus constitute adjusting means for adjusting the shutter 6. When a barium strontium titanate precursor liquid prepared as described below is deposited, the distance between substrate 5 and baffle 6 is preferably between about 0.35 inches and 0.4 inches. It is preferable that the shutter 6 has a smoothness tolerance of at most 5% of the distance between the shutter 6 and the substrate 5. That is, the difference between the distancebetween the substrate 5 and the baffle 6 at any given point and the distance between the substrate 5 and the baffle 6 at any other point is 5% or less of the average distance between the substrate 5 and the baffle 6. For example, if the average distance between the substrate 5 and the baffle 6 is 0.38 inches, there will be no points on the substrate that are more than 0.40 inches from the baffle or less than 0.36 inches from the baffle.

It has been found that baffles within the above tolerance range, i.e., baffles having an area close to the same as the substrate area and a smoothness of 5% or less, provide better thickness uniformity and higher deposition rates than with baffles outside the above tolerance range.

Fig. 7 is a top view of an apparatus according to an exemplary embodiment of the present invention. As shown in FIG. 7, a 0-1000 Torr temperature compensated capacitance manometer 710 monitors the pressure in the deposition chamber 2, and its signal is used to control a downstream control value (not shown) to maintain an accurate pressure in the deposition chamber 2. The deposition chamber 2 is pumped to below 5.0 x 10^-6The high vacuum below torr is accomplished when valve 713 is opened. The high vacuum pumping of the deposition chamber 2 is used to facilitate the extraction of moisture from the chamber walls and from the substrates 5 located within the chamber prior to the deposition operation.

The deposition chamber is evacuated to a pressure of between about 100 and 800 torr during the deposition operation. The deposition chamber exhaust system includes a liquid nitrogen cooling elbow 709 connected to the process chamber 2 through a valve 726. Access from the deposition chamber 2 to an external chamber (not shown) is provided by an air operated slot valve 703. The deposition chamber 2 can be viewed through viewing port 718 during a deposition operation.

A mass flow controller 708 and VCR valve 725-3 are provided for the precursor liquid to control the deposition rate of the precursor through the buffer chamber 42 into the deposition chamber 2 by adjusting the flow of an inert gas, such as argon from the source 736 to the mist generator 46-1. An additional mass flow controller 748 and valve 725-4 are connected to mist generator 46-2 which is connected to buffer chamber 42 through VCR valve 725-5 to control the deposition rate of primer into deposition chamber 2 through buffer chamber 42 by regulating the flow of inert gas, such as argon, from source 736 to mist generator 46-2. A separate mass flow controller 758 is used to introduce oxygen and/or another inert or process gas from source 738 into buffer chamber 42 through VCR valve 725-7.

The inlet nozzle assembly 8 and the outlet nozzle assembly 10 will be described in more detail with reference to fig. 2. The input nozzle assembly 8 includes an input tube 26 that receives atomized solution from a buffer chamber 42 as discussed below with reference to fig. 5. The inlet tube 26 is connected to an arcuate tube 28 having a plurality of apertures or inlet ports 31 for receiving removable screws 30 and removable inlet nozzles 33, each of which is positioned 1/4 inches center to center along the inner circumference of the tube 28.

Fig. 3 shows a plan view of one of the inlet nozzles 33. It includes a screw 33 having an enlarged hollow screw head 301, the screw head 301 having a rim 303 and a hollow screw stem 39 (fig. 2), and a mesh filter 310. Mesh filter 310 is preferably frictionally secured to the inside of screw head 301 prior to attachment of head 301 to stem 39, but may also be brazed to the outer surface of rim 303. Preferably, all of the nozzle portion 33, including the mesh filter 310, is made of stainless steel. Mesh filter 310 is preferably a stainless steel mesh filter having approximately one square micron gap 315 between the wires. It has been found that, while otherwise identical, the use of the mesh filter results in a somewhat reduced deposition rate, it is easily overcome by increasing the number of ports 31 and/or the size of the ports. It is believed that the filter aligns the mist so that the flow of the mist onto the substrate is more uniform and less turbulent so that anomalies in the mist flow that cause non-uniformities are less likely to occur.

The discharge nozzle assembly 10 includes an arcuate discharge tube 29 having a plurality of small holes or discharge ports 31' with removable screws 30. The construction of the discharge nozzle assembly 10 is substantially the same as that of the input nozzle 8 assembly, except that it does not include the input nozzle 33 and a tube 34 having an induction vacuum/discharge source (not shown). The end caps 32 of the tubes 28 and 29 can be removed for cleaning. The arced tube 28 of the input nozzle assembly 8 and the corresponding arced tube 29 of the discharge nozzle assembly 10 each surround two oppositely disposed circumferential portions 4-1, 4-2 of the substrate holder 4.

In one exemplary embodiment in which the BST film is to be deposited, the centers of the holes 31, 31' in the tubes 28 and 29 are nominally located 0.375 inches above the substrate holder 4. However, as shown in fig. 8 and 9, the distance may be adjusted to suit a particular deposition process.

Each tube 28, 29 is typically made of 1/4 "O, D stainless steel and has an inside diameter of about 3/16". The inner wall of each tube 28, 29 is preferably electropolished. The holes 31, 31' in the tubes 28, 29 are each centered a distance 1/4 "and are tapped to receive 4-40 (1/8") male screws.

With this structure, and by adjusting the position of the nozzle 33 by selectively inserting the nozzle 33 into the position of the screw 30 of the arc tube 28 and adjusting the position of the opened discharge hole 31' by selectively removing the screw 30 in the arc tube 29, the solution or mist flow of the vapor above the substrate 5 can be well controlled in various solutions and flow rates thereof, so as to obtain uniform deposition of a thin film on the substrate 5.

Referring to fig. 1 and 2, the substrate holder 4, baffle plate 6, input nozzle assembly 8 and discharge nozzle assembly 10 cooperate to define a relatively small semi-enclosed deposition region 17 surrounding the upper/exposed surface of the substrate 5 within which the solution of vapor is substantially contained throughout the deposition process.

While exemplary embodiments of the substrate holder 4, baffle 6, input nozzle assembly 8, and output nozzle assembly 10 have been described and illustrated, it is understood that various modifications of these structures may be used within the scope of the invention. For example, the arcuate inlet and outlet pipes 28 and 29 may be replaced by pipes of another configuration, such as V-shaped and U-shaped pipes, or may simply be replaced by a plurality of individual nozzles and individual discharge ports.

Figure 5 shows a cross-sectional view of a manifold assembly 40 according to the present invention. The manifold assembly 40 is used to supply a solution (mist or aerosol) of vapor to the input nozzle assembly 8 and generally comprises: a buffer chamber 42; a plurality of inlets 44 connected to respective mist generators via respective valves 725-2, 725-5, 725-7; a deposition valve 725-1 for regulating the flow of mist from the buffer chamber 42 to the nozzle assembly 8; and an exhaust valve 725-6. It is a feature of the present invention that the inlet 44 from the valves 725-2, 725-5 and 725-7 is at 90 degrees to the outlet 49 to the deposition valve 725-1. The buffer chamber 42 is sufficiently large thatthe mist takes on average about 1 to 5 minutes, preferably about 2.5 minutes, in the chamber. This time frame and the 90 degree between inlet 44 and outlet 49 allow for any large droplets that may cause film topography problems. I.e. droplets larger than 2 microns precipitate out. When more than one mist is used at the same time, for example when a primer and a precursor are co-introduced (see below), it enables the mists to mix until they form a single, uniform mist. In the preferred embodiment, the buffer chamber 42 is preferably a cylinder having an inner diameter of 3 inches (vertical in FIG. 5) and a length of about 4 inches (horizontal in FIG. 5) and is made of stainless steel.

In use, one or more mist generators 46 "are used to generate one or more different mists which then flow into the buffer chamber 42 through valve 725" and inlet 44.

The mists flowing into the buffer chamber 42 are mixed into a single uniform atomized solution which then flows into the deposition chamber 2 through the valve 725-1 and the input tube 26 at an appropriate flow rate. Valve 725-1 may optionally be closed to evacuate the deposition chamber 2 when needed or to purge and purge the manifold system when necessary. Similarly, the outlet of the exhaust valve 725-6 is connected to a vacuum source (not shown) so that the valve 725-1, the valve 725-6 and the one or more valves 725-can be closed and opened when necessary to vent/purge the one or more mist generators 46 and the buffer chamber 42 can be purged by providing a vacuum through a pump (not shown) or using a standard negative pressure suction type exhaust port to suck the buffer chamber 42 to purge and purge the mist generators 46 and buffer chamber 42.

The stabilized precursor solutions are ultrasonically agitated to atomize or form a spray of the solutions to produce a mist of the stabilized precursor solutions prior to their introduction into the deposition chamber 2. Fig. 4 is a schematic side view of an exemplary embodiment of the mist generating device used in the present invention. The mist generator 46 includes a closed container 54, a TDKTU-26B or equivalent ultrasonic transmitter 56 liquid and vacuum sealed in the bottom of the container 54, and a power source 72 whose frequency and amplitude are adjustable. The vessel 54 is a modified microporous sheet-mantle T-Line (Millipore Waferguard T-Line) gas filter unit (catalog No. yy5000500) without an internal filter support, with the direction of gas flow indicated by arrow 420 being in the opposite direction to that used during normal operation of the filter. The transmitter 56 is mounted in a slot in the bottom of the mist generator 46. The mist generator 46 also includes an inlet 60 and an outlet 62 for the carrier gas to pass through the container 54. The power supply 72 includes a frequency control device, i.e., a frequency control dial 73 that can be rotated to adjust the frequency of the transmitter 56 and an amplitude control device 75, i.e., an amplitude control dial 75 that can be rotated to adjust the amplitude of the output of the transmitter 56. By adjusting the frequency and amplitude of the transmitter, the particle size of the mist can be controlled. Adjusting the particle size allows one to adjust the surface morphology, step coverage, and deposition rate of the deposition process.

A predetermined amount of precursor liquid 64 is introduced into the vessel 54 prior to operation. During operation, the transmitter 56 is electrically activated to generate a mist 66 of precursor liquid, and an inert carrier gas is passed through the port 60 into the mist 60 where the carrier gas becomes moist or saturated with mist, and the moist carrier gas then flows from the outlet 62 into the manifold assembly 40. The carrier gas is typically an inert gas such as argon, helium or nitrogen, but may also include reactive gases under appropriate conditions.

The mist generator 46 shown in fig. 4 is particularly advantageous because it produces a solution of vapor that can be efficiently flowed or injected into the deposition chamber 2 without difficulties such as icing.

Fig. 10 is a perspective view showing the placement of the ultraviolet irradiation source 16 in the deposition chamber 2. Photo-enhancement of the method of the invention is achieved by providing UV (ultraviolet) light during and after the deposition process, which is believed to promote dissociation of the solvent and organic from the precursor, thereby accelerating the drying process. In addition, the use of UV radiation prior to the deposition process may facilitate the removal (absorption) of moisture from the deposition chamber 2 and from the substrate 5. The position of the uv light source 16 in the deposition chamber is not critical due to the fact that uv radiation is reflected by the stainless steel walls of the deposition chamber 2 into the space between the input nozzle 8 and the exhaust nozzle 10 and also onto the substrate 5 where its illumination provides the light enhancement effect described above.

The UV source 16 comprises at least one UV lamp located in the deposition chamber 2 for applying an ultraviolet bath therein. Sources of spectrum that can be used include ultraviolet lamps and excimer lasers. In each case, the radiation bath applied by the UV source 16 may be adjusted to optimize the dissociation of the desired chemical compounds from the solvent and the organic or further fragments. When in the first case the radiation emitted by the laser is spectrally modified to correspond to the energy required to dissociate or cleave solvent bonds, precursor chemical compound bonds and/or any intermediate organic complex bonds formed during the deposition process which maintains the desired compound in a given precursor liquid. Alternatively, if the UV source 16 is a UV lamp (or lamps), then "tuning" is accomplished by replacing the UV lamp (or group) with another UV lamp (or group) having a more desirable spectrum.

If a ferroelectric algae film is deposited from a vaporous alkoxycarboxylate source, such as when depositing a precursor to form Barium Strontium Titanate (BST) as described below, it is preferred to use a Danielson matrix phototube PSM-275UV radiation source 16 that emits UV radiation having a wavelength of about 180-260 nanometers. UV radiation in this wavelength range is particularly effective for resonating and dissociating the bonds that maintain BST in vapor alkoxycarboxylates, sol-gels, MOD, or another liquid chemical source.

The apparatus 1 shown in fig. 1 comprises an electrical device 102 for applying a DC bias in the deposition chamber 2 during a deposition operation. The electrical device 102 includes a DC input 104 and an output 106. The Direct Current (DC) potential applied between the input collar 25 and the deposition chamber body 12 is typically 350V. The DC bias achieves polarization on the ferroelectric film in situ that increases the film mass. Dipole alignment along the C-axis (main polarization axis) of the crystal is often required and the resulting alignment reduces dislocation density responsible for fatigue and stiffness problems. DC bias voltages greater or less than 350V may also be used to produce the above results. Further, the combination of ultraviolet irradiation and DC bias may be applied in chamber 2 either together or sequentially, and repeatedly, as deposition occurs.

An additional heating device such as a heating plate (not shown) may be used to bake and/or anneal the precursor liquid film previously deposited on the substrate. The baking and annealing can be performed in the deposition chamber 12, but is preferably performed in an auxiliary chamber, as discussed in connection with steps P11 and P12 of fig. 6. The annealing is preferably carried out in an oxygen combustion furnace. High energy density ultraviolet radiation, such as radiation from a diffusion excimer laser source, is also a preferred mode of annealing. 3. Example of the method

Detailed examples of methods for preparing Barium Strontium Titanate (BST) precursors and for fabricating capacitors using BST as the capacitive medium are given below. In table 1, "FW" represents the formula weight "grams" in units of weight: "grams", "mmoles" means milligram molecules, and

compound (I)	FW	grams	mmole	Equiv.
					Barium salt	137.327	9.4255	68.635	0.69986
2-ethyl hexanoic acid	144.21	19.831	137.51	1.4022
					Strontium salt	87.62	2.5790	29.434	0.30014
2-ethyl hexanoic acid	144.21	8.5005	58.945	0.60107
					Titanium isopropoxide	284.26	27.878	98.072	1.0000

"Equiv" represents the equivalent number of moles in solution. The amount of material shown in table 1 was measured to begin step P20 (fig. 6). Barium was placed in 100ml of 2-methoxyethanol and allowed to react. A first metered amount of 2-ethylhexanoic acid was applied to the mixture and stirred. Strontium is then added to the mixture. Once the reaction is complete, a second metered amount of 2-ethylhexanoic acid is applied to the mixture. The mixture was heated to a maximum temperature of 115 ℃ and stirred and all water was distilled off, allowing the mixture to cool. Titanium isopropoxide was added to the mixture, which was then diluted to 220ml with additional 2-methoxyethanol. The mixture was heated to a maximum temperature of 116 ℃ and stirred. All isopropanol and water were then distilled off and step P20 was completed. At the point of the step P21,the mixture was then diluted to exactly 200ml with additional 2-methoxyethanol. The resulting mixture had a concentration of 0.490M, with a Ba to Sr ratio of 0.69986: 0.30014.

The chemical reaction involved the formation of a precursor solution consisting of barium 2-ethylhexanoate, strontium 2-ethylhexanoate, titanium 2-methoxyethanolate as described below

Example I, 2-Ethyl barium hexanoate

(barium Metal) + (2-ethylhexanoic acid) → (2-ethylhexanoic acid barium) + (Hydrogen)

Example II strontium 2-ethylhexanoate

(strontium metal) + (2-ethylhexanoic acid) → (strontium 2-ethylhexanoate) + (hydrogen):

EXAMPLE III (2-Methoxyethanolate titanium)

(titanium isopropoxide) + (2-methoxyethanol) → (2-methoxyethanol titanium) + (isopropanol)

.

The use of 2-methoxyethanol as solvent enables the water arising from distillation to be removed, since the higher boiling point of 2-methoxyethanol enables it to be distilled in H₂The O can be still remained after distillation. The precursor thus formed is substantially anhydrous. Barium and strontium 2-ethylhexanoate were used because: films formed using medium chain length carboxylates such as those in the precursor do not suffer from cracking, bubbling or flaking during baking, as do films formed using long chain carboxylates. The barium and strontium 2-methoxyethanol salts were tested, but proved to be sensitive to air and water in excess. Titanium 2-methoxyethanolate gives a better film than titanium 2-ethylhexanoate, which is not sensitive to air, but although titanium 2-methoxyethanolate is sensitive to air, it is not as sensitive to air as titanium isopropoxide.

The BST precursors formed as described above are used in the inventive method shown in fig. 6 and the inventive devices shown in fig. 1-5 and 7-10 are used to fabricate the capacitor shown in fig. 11.

The BST precursor as described above is placed into the container 54 of the mist generator 46-1 (fig. 7), and the 2-methoxyethanol solvent is placed into the container 54 of the mist generator 46-2. Initially, a substrate including a silicon wafer having a silicon dioxide layer and a platinum layer deposited thereon was pre-baked in an oven at atmospheric pressure (@ Colorado Springs, Colorado) and 180 ℃ for 10 minutes. The substrate is placed on a substrate holder 4 in a deposition chamber. The deposition chamber is pumped to 0.4 torr by a backing pump (not shown) connected to valve 726. The substrate rotation motor 18 is then activated to rotate the substrate holder 4. The UV source 16 is then activated to absorb the moisture in the deposition chamber and any moisture on the substrate the deposition chamber is slowly backfilled with an inert gas source 704, such as argon or nitrogen, through valves 727 and 707 to a pressure of about 595 torr. Next, the vacuum line 702 of the process is opened to stabilize the chamber pressure at about 595 Torr. Valve 725-6 is closed and injection valve 725-1 and deposition valves 725-4 and 725-5 are then opened to begin the flow of argon from source 736 through ultrasonic mist generator 46-2, which is then turned on for 1 minute to deposit an about 100 angstroms film of primer on the substrate at ambient temperature. Then, the deposition valve 725-1 is closed, then the valve 725-6 is opened and the transmitter 56 is turned off along with the mist generator 46-2, and the buffer chamber 42 is exhausted through the exhaust port 705 until the mist generator 46-2 reaches the room temperature. The buffer chamber 42 is purged by applying argon from the source 736 and through the exhaust 705. Valves 725-4 and 725-5 are then closed. Deposition valve 725-1 is reopened and valves 725-3 and 725-2 are also opened to allow argon gas to flow from source 736 through ultrasonic mist generator 46-1, which is then turned on for 10 minutes to allow about 600 angstroms of film to be deposited on the substrate at ambient temperature. The deposition process uses an argon carrier gas to flow both the primer mist and the BST precursor mist over the substrate 5. After a sufficient amount of BST precursor is deposited on the substrate and a thin film is generated, the mist generator 46-1 and the substrate rotation motor are turned off. Deposition valve 725-1 is closed and valve 725-6 is opened to exhaust buffer chamber 42 through exhaust 705 until mist generator 46-1 reaches room temperature. The buffer chamber 42 is purged by applying argon from the source 736 and through the exhaust 705. While the wafer remained in the deposition chamber, the chamber was slowly pumped down to 0.4 torr. The UV source 16 is then turned off. Then, the valve 713 is shut and the deposition chamber is vented to atmospheric pressure. The wafer was then removed from the deposition chamber and "post baked" at 400 ℃ for two minutes. The sheet was then annealed at 800 ℃ for 80 minutes in an oxygen atmosphere. The wafer is then etched using well-known photolithographic techniques to produce a plurality of electronic devices 1112. One sample prepared by this method is hereinafter referred to as sample a.

The above procedure was repeated for another sample (referred to as sample B) except that step P8 was skipped. That is, for sample B, no primer was applied, and the BST precursor was applied directly to the substrate 5.

Both devices a and B are made from one half of a silicon wafer comprising the plate capacitor shown in fig. 11, and the other half of the wafer comprising a series of upper and lower steps between two different layers, as shown in fig. 13 and 14.

For capacitor samples a and B formed on the panel capacitor in the half-chip fabricated in each of the two processes, the thickness of the BST film 1130, the dielectric constant of the BST film 1130, and the leakage current density through the BST film 1130 were measured. In each case, the electric field at which the leakage current was measured was 500 kilovolts per centimeter. The results are shown in FIG. 12 as a function of the sample. The leakage current density is scaled to the right of the graph and is expressed in amperes per square centimeter. The scale of the dielectric constant is given on the left side of thefigure. The dielectric constant for sample a, with the primer applied, was significantly increased by over 100, 1/3. The leakage current density was 2.2X 10-8 amps/cm 2 for the sample with the primer applied, compared to 6X 10-8 amps/cm 2 for sample B without the primer applied. These results indicate that the use of primers can significantly improve the electrical properties that are critical to integrated circuit performance.

Observations were made regarding the surface morphology and step coverage characteristics of each capacitor a and B. Micrographs were taken for each flat surface. The surface of sample a with primer was smooth and had only a single small pinhole defect. The surface is sufficiently smooth to allow the fabrication of integrated circuit devices. The surface of sample B without primer had a number of pinhole defects and the film was broken in many areas. This surface is far from suitable for manufacturing reliable integrated circuits.

Fig. 13 and 14 are images of electron micrographs of a portion of an actual device of samples a and B, respectively, taken on the stepped region of the capacitor structure. That is, FIG. 13 shows an electron micrograph of a cross section of a capacitor fabricated by using a primer prior to depositing a BST dielectric, and FIG. 14 shows a cross section of a capacitor fabricated by using a primer prior to depositing a BST dielectricElectron micrographs of one section of a capacitor made without the primer. In both cases, the device includes a silicon wafer 1122, a SiO step structure as described with reference to FIG. 11, except that the step structure is removed₂ Layer 1124, a bottom electrode 1127, a BST layer 1130, and a platinum top electrode 1132. The bottom electrode is shown as a single layer, rather than separate layers of titanium and platinum, because after annealing they diffuse into each other at their interface and they are substantially indistinguishable on an electron micrograph.

In each of the figures, steps 1310, 1410 are formed in layer 1124 with bottom electrode 1127 deposited over it and BST layer 1130 on top. Comparing fig. 13 and 14, the thickness of the BST layer 1130 is very uniform. In the "high" portion 1314 of step 1310, the thickness is measured to be 45 nanometers (nm), while in the "low" portion 1314 of step 1310, the thickness is measured to be 52nm, with a difference of 7 nm. For comparison, the thickness was measured to be 51nm in the high portion 1414 and 78nm in the low portion 1416 in FIG. 14, with a difference of 27 nm. In FIG. 13, the BST follows the contour of the step better, with the difference between its thinnest point 1312 and thickest point 1318 being significantly less than the difference between points 1412 and 1418 of FIG. 14. Finally, layer 1130 is generally thinner in FIG. 13. This feature indicates that for very thin films, the dielectric formed with the primer is much superior. The relative difference is less pronounced for thicker films, although it is still present. The quality and reliability of the process is also significantly increased for thicker films when primers are used. It is understood that the difference in surface tension between the processes with and without the paint can generally explain the difference in their results.

An anomaly is observed when comparing fig. 12 with fig. 13 and 14. In the flat areas of the sheet, the dielectric 1130 formed using the primer is thicker, while in the step areas of the sheet, the dielectric 1130 formed without the primer is thicker. This is not completely understood, but it can be explained as follows: in the case of flat areas, the thickness is measured using an isotope detector, while in the step areas, the thickness is detected by measurements made from electron micrographs. The two measurement formats are not exactly identical and it may be that the thickness in the flat and stepped regions is substantially the same, or at least much closer, for the process using the primer than that shown in figures 12 and 13. If at the same time the precursor tends to collect more in the lower regions without the primer process, a distinction will be drawn between fig. 12 and fig. 13 and 14.

An important feature to note in the above examples is the ability to fabricate high quality BST films of about 50 nm suitable for use in integrated circuits by the method of the present invention. It is not known in the art to successfully produce such high quality and such thin BST films. In prior art aerosol deposition methods, BST films typically need to be made much thicker if dielectrics suitable for integrated circuits must be successfully fabricated. If the quality of the integrated circuit has to be achieved, even thicker films are to be made by alternative methods such as sputtering.

An alternative deposition procedure was also performed which was similar to the procedure described in the previous example, except that the primer step P8 was performed simultaneously with the precursor deposition step P22. That is, the mist generators 46-1 and 46-2 are turned on and the valves 725-1, 725-2, 725-3, 725-4 and 725-5 are all opened simultaneously, and the precursor and primer mist are mixed in the buffer chamber 42 before entering the deposition chamber 12. Valve 725-1 is then closed, mist generators 46-1 and 46-2 are turned off, valve 725-6 is opened and both mist generators 46-1 and 46-2 are vented to atmosphere until they cool to room temperature. This procedure also produced better film surface morphology and less leakage current than the no primer procedure, although the results were not as good as those obtained by performing steps P8 and P22 separately. It is believed that it will become the preferred process when more experience is gained in the deposition process parameters through this process.

The present invention is advantageous in depositing synthetic thin films of materials such as ferroelectrics, superconductors, materials with high dielectric constants, gemstones, and the like, but is not limited to depositing such synthetic thin films.

While there has been described what are presently considered to be the preferred embodiments of the invention, it is to be understood that the invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive. The scope of the invention is defined by the appended claims rather than by the foregoing description.

Claims (14)

1. A method of fabricating an integrated circuit (1110), the method comprising: (a) providing a liquid precursor (64); (b) placing a substrate (5) into a closed deposition chamber (2); (c) generating a precursor mist of the liquid precursor; (d) flowing the precursor mist through the deposition chamber to form a layer of the precursor liquid on the substrate; (e) processing the liquid layer deposited on the substrate to form a film (1130) of solid material; and (f) completing fabrication of the integrated circuit (1110) such that at least a portion of the solid material film is included in elements (1112) of the integrated circuit; the method is characterized by further comprising the steps of: (g) providing a liquid primer; (h) generating a mist of the liquid primer; and (i) flowing the primer mist through the deposition chamber (2) to form a layer of the primer liquid on the substrate (5) prior to the step of flowing the precursor mist.

2. A method of manufacturing an integrated circuit (1110), the method comprising: (a) providing a liquidprecursor (64); (b) placing a substrate (5) into a closed deposition chamber (2); (c) generating a precursor mist (66) of the liquid precursor; the method is characterized by further comprising the steps of: (d) providing a liquid primer; (e) generating a primer mist of the liquid primer; (f) flowing the precursor and primer mist through the deposition chamber to form a liquid mixture of the primer and precursor on the substrate; (g) processing the liquid mixture deposited on the substrate to form a film (1130) of solid material; and (h) completing fabrication of the integrated circuit (1110) such that elements (1112) of the integrated circuit include at least a portion of the solid material film.

3. An apparatus (1) for manufacturing an integrated circuit (1110), the apparatus comprising: (a) a deposition chamber (2); (b) a substrate holder (4) located in the deposition chamber; (c) device (46-1) for generating a mist of liquid precursor, the device being characterized by further comprising the following means: (d) means (46-2) for generating a mist of the liquid primer coating; (e) means (8, 10, 736, 725, etc.) for flowing the primer mist and the precursor mist through the deposition chamber (2) so as to form liquid layers on the substrate (5), the liquid layers comprising the primer liquid and the precursor liquid; and (f) means (16) for treating the liquid layer deposited on the substrate to form a film (1130) of solid material on the substrate (5).

4. A method according to claim 1 or 2, or apparatus according to claim 3, wherein said liquid primer comprises a primer solvent.

5. The method or apparatus of claim 4, wherein the primer solvent comprises a solvent selected from the group consisting of: 2-methoxyethanol, xylenes, n-butyl acetate and hexamethyl-disiloxane.

6. The method or apparatus of claim 5, wherein the precursor comprises a metal compound in a precursor solvent, the metal compound being selected from the group consisting of: metal alkoxides and metal carboxylates, and metal alkoxycarboxylates.

7. The method or apparatus of claim 5, wherein the precursor solvent is the same as the primer solvent.

8. A method according to claim 1 or 2, wherein the mists flow into the deposition chamber while the substrate is maintained at ambient temperature.

9. The method of claim 1 wherein said mists are flowed simultaneously into said deposition chamber.

10. A method according to claim 1 or 2, wherein each mist flows into the deposition chamber while maintaining a vacuum in the deposition chamber.

11. A method according to claim 1 or 2, comprising the additional step of mixing a plurality of different primer mists outside said deposition chamber to form a primer mist mixture and flowing into the deposition chamber.

12. The method according to claim 1 or 2, comprising the additional step of applying ultraviolet radiation to one of said primer mist and said precursor mist as said mist flows through the deposition chamber.

13. A method according to claim 1 or 2, wherein the processing step comprises applying ultraviolet radiation to one of the primer layer and the precursor layer deposited on the substrate.

14. A method according to claim 1 or 2, comprising the step of applying a DC bias between said deposition chamber (2) and said substrate (5).

Images