« AnteriorContinuar »
METHOD FOR FORMING EMITTERS FOR
FIELD EMISSION DISPLAYS
STATEMENT OF GOVERNMENT INTEREST
This invention was made with government support under Contract No. DABT 63-97-C-0001 awarded by Defense Advanced Research Projects Agency (DARPA). The government has certain rights in this invention.
TECHNICAL FIELD 10
This invention relates to the production of field emission displays and more particularly to a method for forming emitters for field emission displays using image reversal lithography. 15
BACKGROUND OF THE INVENTION
Flat panel displays are widely used in a variety of applications, including computer displays. In addition to 2Q liquid crystal and plasma displays, one type of device well suited for such applications is a field emission display. Field emission displays typically include a generally planar substrate having an array of electron emitters. In many cases, the emitters are conical projections integral to the substrate. 2J
FIG. 1 is a simplified side cross-sectional view of a portion of a field emission display 110 including a faceplate 120 and a baseplate 121 in accordance with the prior art. FIG. 1 is not drawn to scale. The faceplate 120 includes a transparent viewing screen 122, a transparent conductive 30 layer 124 and a cathodoluminescent layer 126. The transparent viewing screen 122 supports the layers 124 and 126, acts as viewing surface and as a wall for a hermetically sealed package formed between the viewing screen 122 and the baseplate 121. The viewing screen 122 may be formed 35 from glass or other transparent material. The transparent conductive layer 124 may be formed, for example, from indium tin oxide. The cathodoluminescent layer 126 may be segmented into localized portions. In a conventional monochrome display 110, each localized portion of the cathod- 40 oluminescent layer 126 forms one pixel of the monochrome display 110. Also, in a conventional color display 110, each localized portion of the cathodoluminescent layer 126 forms a green, red or blue sub-pixel of the color display 110. Materials useful as cathodoluminescent materials in the 45 cathodoluminescent layer 126 include Y203:Eu (red, phosphor P-56), Y3(Al,Ga)5012:Tb (green, phosphor P-53) and Y2(Si05):Ce (blue, phosphor P-47) available from Osram Sylvania of Towanda, Pa. or from Nichia of Japan.
The baseplate 121 includes emitters 130 formed on a 50 planar surface of a substrate 132 that is preferably a semiconductor material such as silicon. The substrate 132 is coated with a dielectric layer 134. In one embodiment, this is effected by deposition of silicon dioxide via a conventional TEOS process. The dielectric layer 134 is formed to 55 have a thickness that is approximately equal to or just less than a height of the emitters 130. This thickness is on the order of 0.4 microns, although greater or lesser thicknesses may be employed. A conductive extraction grid 138 is formed on the dielectric layer 134. The extraction grid 138 60 may be formed, for example, as a thin layer of polysilicon. An opening 140 is created in the extraction grid 138 having a radius that is also approximately the separation of the extraction grid 138 from the tip of the emitter 130. The radius of the opening 140 may be about 0.4 microns, 65 although larger or smaller openings 140 may also be employed.
In operation, the extraction grid 138 is biased to a voltage on the order of 100 volts, although higher or lower voltages may be used, while the substrate 132 is maintained at a voltage of about zero volts. Signals coupled to the emitters 130 allow electrons to flow to the emitter 130. Intense electrical fields between the emitter 130 and the extraction grid 138 cause emission of electrons from the emitter 130.
A larger positive voltage, ranging up to as much as 5,000 volts or more but usually 2,500 volts or less, is applied to the faceplate 120 via the transparent conductive layer 124. The electrons emitted from the emitter 130 are accelerated to the faceplate 120 by this voltage and strike the cathodoluminescent layer 126. This causes light emission in selected areas, i.e., those areas opposite the emitters 130, and forms luminous images such as text, pictures, and the like.
The brightness of the light produced in response to the emitted electrons depends, in part, upon the number of electrons striking the cathodoluminescent layer 126 in a given interval. Field emission microscopy of the emitters 130 reveal that electrons are emitted from only a few atomic sites at the tip of the emitters. The emitting area is very small, generally from 1-5 nm in diameter. Uniformity in the shape, height, and placement of the emitters 130 is an important factor in the quality of the field emission display 110. These parameters affect differences in the number of electrons striking areas of the cathodoluminescent layer 126 that may be perceived by the viewer as bright and dark areas, or as other defects.
For instance, if an emitter 130 is shorter than other emitters, electrons emitted from the tip of the taller emitter may have a tendency to spread out more as they are directed to the cathodoluminescent layer 126. This could cause electrons to bleedover to areas of the cathodoluminescent layer 126 other than those intended, creating a picture defect. Similarly, emitters 130 that are longer than the others may have a tendency to not spread out as much as desired. Mis-located emitters 130 may tend to create a surplus of electrons in one area and a deficiency of electrons in others, also making a deficient picture.
Arrays of emitters 130 can be formed by chemical mechanical polishing steps such as those taught in U.S. Pat. No. 5,372,973, assigned to Micron Technology, Inc. and incorporated herein by reference. These arrays of emitters 130 can also be formed by typical semiconductor fabrication processes such as wet or dry etching of the silicon substrate 132. One example of forming emitters 130 by semiconductor fabrication steps is seen in U.S. Pat, No. 5,766,829 assigned to Micron Technology, Inc. and incorporated herein by reference. In the '829 patent, printed features for defining the size and location of emitter sites are made using phase shift lithography. As seen in FIG. 2 of the '829 patent, by using this method, the phase of exposure energy such as visible light or x-rays is controlled through a reticle in two orientations so that exposed and non-exposed regions or "islands" are produced on a photoresist by destructive or constructive interference. The islands are hardened and then used as etching masks. Isotropic or anisotropic etching is performed on the exposed substrate, while leaving the areas under the islands intact. Etching continues until the areas of the substrate under the islands form points; then the islands are removed. These points become the emitters of the flat panel display.
A problem in using phase shift lithography is that it is difficult to control the photoresist onto which the exposure energy is directed, causing the islands formed on the baseplate to be non-uniform. Later processing with nonuniform
islands cause nonuniform emitters to be formed, and ultimately creates a substandard field emission display.
Other semiconductor fabrication technologies have been used to make arrays of emitters 130. For instance, a negative photoresistive material layered on the substrate has been 5 used. Using a negative photoresist to make an array of emitters 130 requires exposing the photoresist only where the islands are to be formed. The exposing energy directed to the negative photoresist hardens the exposed areas and later developing removes the nonexposed areas. This creates 1° an array of islands of exposed photoresist for later processing into an array of emitters 130. However, using a negative photoresist is disfavored for many reasons. It is extremely temperature sensitive, so that normal variations in processing temperatures create nonuniform islands. Some negative :5 photoresist has a tendency to swell during developing, thus preventing its use in very small pattern making. It also has a limited depth of focus. Additionally, developing some negative photoresist requires organic solvents that are flammable as well as difficult and expensive to safely dispose. 20
A positive photoresistive material can be layered on the substrate, patterned then exposed, but this process also has difficulties. When using a positive photoresist, areas that receive the exposing energy are removed and areas that are shielded from the exposing energy remain after developing. 25 In order to form an array of small islands, most of the positive photoresist is exposed, e.g., over 95%. Trying to create uniform islands of non-exposed positive photoresist is difficult with so much exposing energy applied to the positive photoresist. For instance, if the exposing energy is 30 visible light, excess light tends to undercut the mask, thereby exposing the positive photoresist meant to be covered. In addition, the exposing light is reflected, refracted, and scattered around the photoresist. The same effects occur with
x-ray or other exposing energies during exposing times. Unfortunately, these effects are nonuniform which causes the islands of positive photoresist to be nonuniform as well. As described above, it is impossible to create uniform emitters 130 from nonuniform islands.
Thus, it would be desirable to devise a method for creating uniform emitters using fabrication steps that are currently existing.
SUMMARY OF THE INVENTION
According to one aspect of the invention, a method for forming an emitter for a field emission device is provided. A photoresist layer is deposited on a semiconductor material used to form emitters. Islands of the photoresist are exposed by an exposing energy. The temperature of the substrate is 50 increased while in the presence of a basic gas, making the exposed islands non-photoreactive. The photoresist is then flood exposed and developed, leaving behind the hardened islands. Finally, the semiconductor is selectively removed until the semiconductor material below each island is 55 emitter-shaped.
Other aspects of the invention include a field emission device emitter and an array of emitters in a field emission device formed by the above steps.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a simplified side cross-sectional view of a portion of a field emission display according to the prior art.
FIG. 2 is a cross-sectional view of a substrate used to form 65 emitters for field emission displays according to an embodiment of the present invention.
FIG. 3 is a cross-sectional view of the substrate shown in FIG. 2 after layers have been deposited on the substrate.
FIG. 4 shows a cross-sectional view of the substrate shown in FIG. 3 after adding a masking layer.
FIG. 5 shows a cross-sectional view of the substrate of FIG. 4 after it has been exposed by an exposing energy and the masking layer removed.
FIG. 6 shows the substrate of FIG. 5 after the exposed areas had been neutralized.
FIG. 7 shows a cross-sectional view of the substrate of FIG. 6 after it has been developed.
FIG. 8 is a flowchart of a process for manufacturing a field emission display according to an embodiment of the present invention.
FIG. 9 is a simplified block diagram of a computer including a field emission display formed by embodiments of the present invention.
DETAILED DESCRIPTION OF THE
FIGS. 2-7 are cross-sectional views of various stages of manufacturing an emitter baseplate, according to embodiments of the present invention. A procedure 300 for making emitters using image reversal lithography is shown in FIG. 8. This procedure 300 parallels, to some extent, the images shown in FIGS. 2-7 and those shown in the previously incorporated '829 patent.
FIG. 2 shows a substrate 200 of emitter material. The substrate 200 may be, for example, silicon, molybdenum, palladium oxide, or diamond-like carbon. Additionally, the substrate 200 may be a layer of amorphous silicon disposed over an insulative substrate such as glass. The substrate 200 may be small, having a diagonal measurement of a few centimeters, or less, or the substrate may be large, such as one that can be used for a computer or television monitor. Embodiments of this invention will be described as having the substrate 200 made of silicon and not disposed on an insulative layer, but the invention is in no way limited to this description or limited to the examples cited above.
FIG. 3 shows the substrate 200 of FIG. 2 after a hardmask layer 202 and a positive photoresist layer 204 have been deposited. Using the hardmask layer 202 is optional, but preferred. The hardmask layer 202 can be any material suitable for its purpose and preferably is an oxide layer such as silicon dioxide (Si02). Additionally, the hardmask layer 202 could comprise a nitride layer, such as silicon nitride (Si3N4). Further, nickel or chrome or some other suitable metal can be added to the hardmask layer 202. The thickness of the hardmask layer 202 can range between 0.05 and 0.3 microns (/an), and in the preferred embodiment is 0.2 fim thick. The positive photoresist layer 204 is one of many available positive photoresist materials, such as those available from Olin. A positive photoresist layer 204 could be, for example, Olin part number HiPR6509 or HPR504. Most preferably, the positive photoresist layer 204 is about 1.1 fim thick although it preferably ranges from 0.6 to 2 fim. The step of depositing a hardmask layer 202 and a photoresist layer 204 on a substrate 200 of an emitter material is shown in FIG. 8 as a step 302.
With reference to FIG. 4, following deposition of the hardmask layer 202 and the positive photoresist layer 204 in the step 302, a mask 206 is created and positioned above the positive photoresist layer 204 as is known in the art. This is shown as a step 304 in FIG. 8. Those skilled in the art will appreciate that the mask 206 is typically comprised of a
substrate of transparent material, such as quartz, attached to an opaque film. The material attached to the opaque film need only be transparent to the exposing energy, and not necessarily transparent to visible light, unless visible light is used to expose the positive photoresist layer 204. In FIG. 4, 5 only the opaque film portion of the mask 206 is depicted. When referring to the mask 206, it is understood that it is the opaque film portion of the mask 206 that is being referenced.
An exposing energy 210 is directed to the mask 206 and positive photoresist layer 204. The mask 206 has holes or 10 openings through which the exposing energy 210 can pass unperturbed. Typically, when using a positive photoresist layer 204, openings 212 are made in the mask 206 where the underlying photoresist material is to be developed and removed. However, when using image reversal lithography, 15 the pattern of the mask 206 is reversed. That is, when using image reversal lithography, such as is used in the method of FIGS. 2-8, openings 212 are made in the mask 206 where hardened material is eventually desired. The exposing step corresponds to a step 306 in FIG. 8. Areas exposed by the 20 exposing energy 210 through the negative mask 206 will later form islands that are then used to make the emitters 130, as is described in greater detail below.
In a step 308, (FIG. 8) the mask 206 used in FIG. 4 is removed, leaving the substrate 200 substantially as shown in 25 FIG. 5. (This Figure also corresponds to FIG. 4 of the '829 patent, although is shown from a different point of reference.) As illustrated in FIG. 5, the exposing energy 210 chemically alters the positive photoresist layer 204. This creates locations within the positive photoresist layer 204 of 30 exposed areas 214 and nonexposed areas 216. The exposed areas 214 substantially align with where the openings 212 within the mask 206 were located. Exposing the positive photoresist layer 204 to the exposing energy 210 causes a release of photo-generated acid within the exposed areas 35 214. In typical processing of a positive photoresist layer 204, the presence of the acid within the exposed areas 214 is desired, because when the positive photoresist layer 204 is developed, the acid first etches the material within the exposed areas 214. This etched material is later removed. 40 However, when using image reversal lithography, this acid present in the exposed areas 214 is detrimental and must be neutralized.
In a step 310 of FIG. 8, the substrate 200 is placed in an oven in the presence of a basic gas (not shown). This baking 45 step 310 causes the exposed areas 214 of the positive photoresist layer 204 to neutralize the photo-generated acid by reacting with the basic gas. Any known method for neutralizing the photo-generated acid could be used. Once these exposed areas 214 are baked in the presence of a basic 50 gas, they are no longer sensitive to light. This is indicated in FIG. 6 by showing the exposed areas 214 of FIG. 5 as exposed and baked areas 224. The nonexposed areas 216 of the positive photoresist layer 204 remain sensitive to light. Neither the increased temperature nor the presence of the 55 basic gas has an effect on the nonexposed areas 216 of the positive photoresist layer 204. These areas 216 remain substantially unchanged.
In a preferred embodiment, the substrate 200 is baked at a temperature of 95° C. in the presence of 100% anhydrous 60 ammonia for between 5 minutes and 2 hours at a pressure slightly below atmospheric pressure, for instance 600 torr. As is known in the art, the procedure to bake the substrate 200 in the presence of 100% anhydrous ammonia is to first place the substrate in an oven that can be evacuated. Then, 65 the oven is evacuated of as much air as practical, and pure nitrogen is pumped into the evacuated space. These steps of
evacuating the oven and introducing nitrogen are repeated several times. Each cycle of evacuation and introducing additional nitrogen removes a further quantity of the air originally in the oven. By repeating these steps, nearly all (greater than 99%) of the air can be evacuated. After the final evacuation, anhydrous ammonia is allowed into the oven until a small vacuum remains. This creates a pressure within the oven that is slightly below atmospheric pressure. As stated above, the substrate 200 is then soft-baked at 95° C. for a time between 5 minutes and 2 hours.
In a step 312 of FIG. 8, the entire substrate 200 is flood exposed with an exposing energy 210 which may or may not be the same exposing energy described in step 306. As is known in the art, flood exposure is directing exposing energy 210 at a substrate 200 where no mask 206 is used. This is represented in FIG. 6. Recall that since the areas 224 of the positive photoresist layer 204 have been baked in the presence of a basic gas, they are no longer photosensitive, and are thus not changed by the exposing energy 210. The nonexposed areas 216 of the positive photoresist layer 204 are, however, still photoreactive and release acid when exposed to the exposing energy 210.
In a step 314 of FIG. 8, the positive photoresist layer 204 of the substrate 200 is then developed as normal. It can be developed using tetra-methyl-ammonium-hydroxide (TMAH), which is commercially available from a variety of sources. Other developers such as sodium hydroxide (NaOH) or potassium hydroxide (KOH) may be used for developing. This developing step 314 uses the photogenerated acid from the flood exposure to dissolve the previously non-exposed areas 216 of the positive photoresist layer 204. Full development leaves only the exposed and baked areas 224 of the positive photoresist layer 204 remaining on the hardmask layer 202.
Once the step 314 is complete, the substrate 200 will look substantially as shown in FIG. 7, with the exposed and baked areas 224 resting on top of the hardmask layer 202. FIG. 7 directly corresponds to FIG. 5A of the '829 patent, with the substrate 200 corresponding to a baseplate, the hardmask layer 202 corresponding to a mask layer, and the exposed and baked areas 224 corresponding to solid areas of photoresist.
From the state of the substrate 200 as shown in FIG. 7, arrays of emitters 130 can be made by a known method. Examples of making emitters 130 from this stage are described in the '829 patent, column 6, lines 17-64, which has been previously incorporated by reference. Highlights of this description are shown in steps 316-322 of FIG. 8 and as described herein.
A step 316 of FIG. 8 calls for the hardmask layer 202 to be etched, using the exposed and baked areas 224 as an etching mask. If the hardmask layer 202 was not used, then steps 316 and 318 of FIG. 8 are omitted. As described in the '829 patent, etching the hardmask layer 202 can be performed with either a wet etch or a dry etch depending on materials used. For example, if the hardmask layer 202 is silicon nitride, it can be etched with an SF6 based plasma etch. Following the etching of the hardmask layer 204 in step 316, a step 318 shown in FIG. 8 directs that the exposed and baked areas of photoresist 224 be removed. For removal of a positive photoresist, a solution of concentrated H2S04 and H202 at about 150° C. can be used. Following this step, the substrate 200 would look substantially similar to the baseplate as shown in FIG. 5B of the '829 patent, with islands of the etched hardmask layer 202 sitting over the substrate 200.