Patent Grant
7172789
The present invention relates generally to the field of creating assemblies through fluidic self assembly and to the assemblies so created.
There are many examples of large arrays of functional components which can provide, produce or detect electromagnetic signals or chemicals or other characteristics. An example of such a large array is that of a display where many pixels or sub-pixels are formed on an array of electronic elements. For example, an active matrix liquid crystal display includes an array of many pixels or sub-pixels which are fabricated on amorphous silicon or polysilicon substrates which are large. As is well known in the art, it is difficult to produce a completely flawless active matrix liquid crystal display (LCD), when the display area is large, such as the LCD's on modem laptop computers. As the display area gets larger and larger, the yield of good displays decreases. This is due to the manner in which these display devices are fabricated.
Silicon VLSI can be used to produce such an array over a silicon wafer's surface, but silicon wafers are limited in size, limited in conductivity, and not transparent. Further, processing of large areas on silicon wafers can be expensive. Displays which valve the light coming through them need to be transparent. Single crystal silicon can be bonded to a glass substrate and then etched to remove most of the area to achieve transparency, but this is intrinsically wasteful in that, for the sake of maximizing light transmission, the majority of the processed material is discarded and becomes chemical waste. The under-utilization of the precious die area wastes resources, causes greater amounts of chemical waste to be generated in the process, and is generally inefficient and expensive. Another example is photodiode arrays which may be used to collect solar energy. Large arrays of silicon photodiodes with concentrating lenses have been made by sawing wafers and using a pick and place assembly, but thermal dissipation is poor for large elements, and the small elements require too much assembly time.
Alternative approaches to fabricating arrays such as displays include fabricating the desired circuitry in an amorphous or polycrystalline semiconductor layer which has been deposited on a substrate, such as glass or quartz. These approaches avoid the limitations of the size of the available single crystal silicon wafers, and avoid the cost of the single crystal wafers, but require expensive deposition of the semiconductor layer, and they still require processing of the entire large substrate to form the active elements in an array, still resulting in the production of much chemical waste and wasted resources. These processes also limit the choice of the substrate; for example, plastic substrates cannot be used due to the nature of the processes which deposit the semiconductor layers. Furthermore, amorphous or polycrystalline silicon semiconductor elements do not perform as well as those made from single crystal semiconductor material. For displays, as an example, it is often difficult or impossible to form some of the desired circuitry out of the amorphous or polycrystalline semiconductor materials. Thus, high frequency edge drivers may be impossible to form out of these materials. This results in the difficulty and expense of attaching an electrical lead for each and every row and column of an array, such as an active matrix liquid crystal display array.
As noted above, another difficulty with the existing techniques is that the large number of elements in a large array results in a low probability that all of them will work properly and thus the yield of acceptably good arrays from the manufacturing process is low. Furthermore, there is no possibility of testing any of the elements until the assembly is complete, and then any imperfection in the array must be tolerated or the entire array could be discarded or special and expensive techniques must be used to repair it. These problems result from the fact that the various elements in the array are fabricated on the array rather than separately.
It is possible to separately produce elements, such as pixel drivers and then place them where desired on a different and perhaps larger substrate. Prior techniques can be generally divided into two types: deterministic methods or random methods. Deterministic methods, such as pick and place, use a human or robot arm to pick each element and place it into its corresponding location in a different substrate. Pick and place methods place devices generally one at a time, and are generally not applicable to very small or numerous elements such as those needed for large arrays, such as an active matrix liquid crystal display. Random placement techniques are more effective and result in high yields if the elements to be placed have the right shape. U.S. Pat. Nos. 5,545,291 and 5,904,545 describe methods which use random placement. In this method, microstructures are assembled onto a different substrate through fluid transport. This is sometimes referred to as fluidic self assembly (FSA). Using this technique, various blocks, each containing a functional component, may be fabricated on one substrate and then separated from that substrate and assembled onto a separate substrate through the fluidic self assembly process. The process involves combining the blocks with a fluid and dispensing the fluid and blocks over the surface of a receiving substrate which has receptor regions (e.g. openings). The blocks flow in the fluid over the surface and randomly align onto receptor regions.
Thus the process which uses fluidic self assembly typically requires forming openings in a substrate in order to receive the elements or blocks. Methods are known in the prior art for forming such openings and are described in U.S. Pat. No. 5,545,291. One issue in forming an opening is to create its sidewalls so that blocks will self-align into the opening and drop into the opening. The substrate having openings in the glass layer 10 may be used as a receiving substrate to receive a plurality of elements by using a fluidic self assembly method.
While the foregoing methods allow an assembly to be created through a fluidic self-assembly process, these methods do not always provide perfect yields. Moreover, achieving a high yield may require longer processing times than is desired. For example, the fluidic self-assembly process can still produce poor results if the elements do not flow smoothly over the service containing the openings. For example, if there is too much friction between the elements and the surface, the elements tend to stick to the surface and do not flow into the receptor regions. Furthermore, the elements may become dislodged from the openings in subsequent processing steps.
From the foregoing description it can be seen that it is desirable to provide methods for allowing improved fluidic self-assembly.
The present invention provides various methods for assembling a structure onto a substrate and also provides apparatuses resulting from these methods. A method according to one aspect of the present invention includes dispensing the slurry onto a substrate wherein the slurry includes a fluid and a first plurality of elements, each of which is designed to mate with a receptor region on the substrate and each of which comprises a functional element, and wherein the slurry also includes a second plurality of elements which are not designed to mate with receptor regions on the substrate. In one particular embodiment of a method according to this aspect of the invention, the second plurality of elements facilitate movement of the first plurality of elements over the substrate and do not include any functional elements.
A method according to another aspect of the invention includes dispensing in a flow having a first direction a slurry onto a substrate, wherein the slurry comprises a fluid and a first plurality of elements, each of which is designed to mate with a receptor region on the substrate and each of which includes a functional element. The method further includes vibrating the substrate in a second direction which is substantially perpendicular to the first direction.
A method according to another aspect of the present invention includes the use of nozzles. In one exemplary method according to this aspect, a slurry is created which includes a fluid and a plurality of elements each of which is designed to mate with a receptor region on the substrate and each of which comprises a functional element. The slurry is dispensed onto the substrate from a nozzle or a series of nozzle jets. A second fluid projected may be from another nozzle (or series of nozzles) may be used. In other embodiments, a nozzle providing a suction may also be used.
A method according to another aspect of the invention includes dissolving a bonding agent into a solvent to create a fluid (and a surfactant may also be dissolved into the solvent), and dispensing the slurry onto a substrate, wherein the slurry comprises the fluid and a plurality of elements each of which is designed to mate with a receptor region on the substrate and each of which includes a functional element, and evaporating the solvent after each of the plurality of elements has mated with a corresponding receptor region, wherein the bonding agent bonds each of the plurality of elements to the corresponding receptor region.
These aspects as well as other features of the present invention will be described further below.
The present invention is illustrated by way of example and not limitation in the figures of the accompanying drawings in which like references indicate similar elements.
The present invention relates to methods and apparatuses for forming openings in a receiving substrate. The following description and drawings are illustrative of the invention and are not to be construed as limiting the invention. Numerous specific details are described to provide a thorough understanding of the present invention. However, in certain instances, well known or conventional details are not described in order to not unnecessarily obscure the present invention in detail.
The present invention relates generally to the field of creating openings in a receiving substrate and to apparatuses having these openings. The present invention may be used to fabricate openings for various different types of arrays. Typically, each element in the array includes a functional component which may be an electrical component, a chemical component, or an electromechanical structural element or a micro electromechanical structural element or a micro-mechanical structural element. The various methods of the present invention are illustrated in certain detailed examples with regard to the manufacture of an active matrix liquid crystal display, but it will be recognized that the invention will have wider applicability. Merely by way of example, the invention may be applied to the fabrication of an electronic array which can be used to deliver precise voltages for the control of liquid crystal cells to create a liquid crystal display or may be used for other types of displays such as electro-luminescent displays or light emitting diode displays or displays using electrophoretic ink display pixels, such as microencapsulated electrophoretic ink display pixels, and also for other applications requiring sampling or producing electrical signals over a large array of electrodes, such as memories and imaging arrays and photo diode arrays. Further, the present invention may be used with electromagnetic signal detectors (e.g. antennas), or solar cells, or chemical sensors.
In a preferred embodiment of the present invention, an opening has a predetermined size which is designed to receive an element also of the same predetermined size so that the element fits into the opening. By controlling the fit between the element and the opening, it is possible to have the elements self assemble into the openings. Accordingly, in a preferred embodiment of the present invention, each element has a trapezoidal cross-sectional shape which fits into a trapezoidal opening in which the top of the opening has a larger area than the bottom of the opening. Furthermore, it is desirable to form a deep opening without making the opening too wide.
The method shown by
In one example of the present invention, the etch to create the opening 105 is a wet etch. The preferred etch mix depends on the type of glass used. The etch mix contains dilute hydrofluoric acid and, preferably, an additional acid such as nitric acid, sulfuric acid, perchloric acid, or more preferably a halide acid such as hydrobromic acid or hydrochloric acid or hydriodic acid. The ratio of water to the second acid to hydrofluoric acid ranges from 2:1:1 to 50:20:1, depending on the type of glass used for the substrate. The substrate is etched for roughly one quarter of the total etch time. In one particular embodiment, the ratios of an etchant which includes water, hydrochloric acid and hydrofluoric acid is 4:1:1 (water:hydrochloric:hydrofluoric) and the total etch time is such that the total depth of the etchant produces an opening having a depth of 14.5 microns.
The glass can be etched with any hydrofluoric acid-based etch solution, with hydrofluoric acid concentration ranging from concentrated hydrofluoric acid to 1% concentrated hydrofluoric acid in water. In a preferred embodiment, the glass is soda lime glass, and is etched with a mixture of hydrofluoric acid:counter acid:water in which the water content ranges from 50% to 95%, the hydrofluoric acid content ranges from 25% to 1%, and the counter acid content ranges from 40% to 5%. The counter acid could be nitric acid or sulfuric acid, or more preferably, hydriodic acid, or hydrobromic acid, or most preferably, hydrochloric acid. The temperature of the etching solution is maintained at a temperature in the range of from −10° C. to 40° C. In a more preferred embodiment, the glass is borosilicate or aluminosilicate glass, such as Corning 7059 or Corning 1737, respectively, and is etched in an etch mix as above, in which the water content ranges from 60% to 95%, the hydrofluoric acid content ranges from 10% to 1%, and the counter acid, as described above, ranges from 30% to 5%. In the most preferred embodiment, the temperature of the borosilicate or aluminosilicate glass etch described above is controlled to within 0.5 degrees C., at a specified temperature between −5 and 5 degrees C. In the most preferred embodiment, the etch solution is 1 part HF to 10 parts HCl to 100 parts H2O.
After the opening 105 has been formed in the substrate 101 as shown in
After the beveled edge is created at the top of the opening, another amorphous silicon layer 108 is deposited onto the entire top of the substrate. The amorphous silicon may be deposited by a PECVD or CVD or a sputtering process to a total thickness of about a thousand Angstroms. An optional tungsten layer (which is preferably about 100 to 1000 Angstroms thick) may then be deposited (e.g. by sputtering) on the layer 108. Then a photoresist layer 109 is applied to the top of the amorphous silicon layer 108 (or on top of the optional tungsten layer if the tungsten layer is applied), and this photoresist layer 109 is patterned to create the opening 110 in the photoresist layer 109 (and in the optional tungsten layer if present) as shown in
By means of an example, given a block 353 with the cross-section shape shown in
The following parameters create a block and opening combination which produces good self-assembly.
In one example of the method shown in
It will be appreciated for the process shown in
After the opening 154 is created, a second mask material 155 is applied over the substrate 151 and into the hole 154 after the patterned mask layer 153 has been removed. This results in the structure shown in
Another example of a method according to the present invention will now be described in conjunction with
The various different methods of the present invention will typically create an array of openings on a receiving substrate. These openings are then filled with a plurality of elements, each of which typically include at least one function component, such as a pixel driving circuit for driving a liquid crystal cell in an active matrix liquid crystal display or other display driving elements for other types of displays. Co-pending U.S. patent applications Ser. Nos. 09/251,220 and 09/251,268 filed Feb. 16, 1999 by John Stephen Smith and assigned to the same Assignee of the present invention describe an example of the electrical circuitry disposed on each element which is to be assembled into an opening. These co-pending applications are hereby incorporated herein by reference. Generally, these elements resemble tapered blocks having a trapezoidal cross-section where the top of the block is wider than the bottom of the block. An example of such a block is shown as block 16 in
In processing operation 307, the assembly is planarized. In one example, the substrate with the blocks is spin-coated with partially polymerized benzocyclobutene to a uniform thickness of greater than 2 microns. This planarization layer is then cured.
Vias are then etched in the planarization layer and electrical connections are made to bonding pads on the integrated circuit in the active circuit layer 327. Processing operation 309 as shown in
The Fluidic Self Assembly (FSA) process can be performed in a variety of conditions, with a variety of fluids. Proper selection of the block and substrate surface chemistry as well as the fluid to be used results in faster, higher yielding assemblies with dramatically fewer excess blocks left on the surface prior to rinsing, inspection, and repair. The surface chemistry interaction with the fluid effectively reduces the amount of friction between the blocks and the substrate. The choice of block surface chemistry, substrate surface chemistry, and fluid are all usually interrelated. For example, hydrophobic surface chemistries work much better with organic fluids than with aqueous solutions. The treatment of the surfaces usually, but not necessarily, alters both sets of surfaces so that they are of the same type. For example, both the elements and substrate may be treated to have a hydrophilic surface and the elements, so treated, may be dispersed in a polar solvent (e.g. H2O) and then dispensed over the so treated substrate in an FSA process. In an alternative example, both the elements and the substrate may be treated to have a hydrophobic surface and the elements and substrate, so treated, may be combined in a solvent (e.g. hexane) which is non-polar in an FSA process.
In the case of the block surface chemistry, any surface treatment that cleans the surface of the block results in improved FSA's. A surface treatment fluid may be in a liquid form, a gas form or a plasma form. The improvement is more significant if the surface treatment leaves a desired, uniform surface chemistry on all of the blocks for a given assembly. Hydrophilic surface chemistries can be achieved by rinsing the blocks in any oxidizing solution, such as aqueous KMnO4, H2SO4 with H2O2, NH4OH with H2O2, or ozonated deionized water or by exposing the blocks to an oxygen plasma. The surface chemistry is more uniform if the blocks can be agitated while being immersed in one of the preceding chemistries. The agitation could be caused by a variety of systems, such as from a recirculating fluid flow, from a megasonic or ultrasonic bath, or by stirring. Hydrophilic surfaces with good uniformity can be achieved by oxidizing the surfaces and reacting a functional, self-assembled monolayer onto the surface, and then oxidizing the functional group in the self-assembled monolayer. For example, octenyltrimethoxysilane could be reacted onto block surfaces cleaned by oxidation with H2SO4 and H2O2. The alkene group can then be oxidized into a hydrophilic functional group by treatment with a mild, aqueous KMnO4 solution.
Hydrophobic surfaces on blocks can be created by cleaning and oxidizing the blocks as described above, and then reacting a self-assembled monolayer onto the surface, such that the hydrocarbon chains of the self-assembled monolayer are topmost at the surface. Typically the contact angle of water on these hydrophobic surfaces is at least 90 degrees. TEFLON, which can be deposited or formed on the surfaces of the blocks, will also act to create hydrophobic surfaces on the blocks. Other types of coatings of a hydrophobic nature may be used or a fluorine plasma may be used to create hydrophobic surfaces on the blocks.
The substrate surface treatment is analogous to the block surface treatment. It is possible to perform fluidic self-assemblies in which the block receives one surface treatment, and the substrate receives a different surface treatment. For example, the substrate may be exposed to a surface treatment fluid which creates a metal coating (film) on the surface of the substrate (a hydrophilic surface normally) or a surface treatment fluid which is an oxygen plasma (to create a hydrophilic surface) or a surface treatment fluid which oxidizes a coating (or otherwise creates an oxidized coating) on the substrate while the blocks are exposed to a surface treatment fluid which is dissolved ozone in water. Alternatively, the same surface treatment fluid (e.g. an oxygen plasma) may be used on both the blocks and the substrate. It is also possible to pattern the surface treatment on the substrate, such that the surfaces in the bottom of the receptor sites are hydrophilic, and the rest of the substrate is hydrophobic, for example.
As mentioned above, the choice of fluid for any given FSA depends on the surface chemistry of the blocks and the substrate. Assemblies processed with hydrophobic surface chemistries (e.g. the blocks and substrate have each been exposed to surface treatment fluid(s) which have created hydrophobic surfaces on the blocks and the substrates) proceed more rapidly if an organic solvent, such as toluene or hexane, is used as the slurry fluid in the FSA process. Conversely, if the blocks and substrate are hydrophilic (because the blocks and substrate have been exposed to surface treatment fluid(s) which have created hydrophilic surfaces on the blocks and the substrate), water or hydrophilic solvents such as acetone (or other polar or water soluble solvents) are more appropriate as the slurry fluid in the FSA process.
A preferred method of surface treatment for the block devices is to oxidize the surface with a dilute aqueous solution containing 0.008% potassium permanganate, 0.025% sodium periodate, and 0.415% potassium carbonate. The block devices are stirred in the oxidizing solution at 75° C. for 2 hours. The preferred method for stirring the block devices in the solution is to separate the block devices from the stirrer by a stainless steel sieve material. After 2 hours of exposure to the oxidizing solution, the block devices are collected on the sieve, and rinsed in a 0.3M aqueous sodium bisulfite solution. Then the devices are rinsed in a 0.1M solution of acetic acid. Finally, they are rinsed in DI (deionized) water.
The most preferred method is to oxidize the block devices while they recirculate in a solution of 5 to 125 ppm dissolved ozone in DI water. The most preferred ozone concentration is in the range of 50–125 ppm dissolved ozone. The block devices are recirculated for 1 hour, rinsed in DI water, and then transferred to the slurry fluid prior to fluidic self assembly.
The most preferred method for treating a glass or silicon substrate is to place the substrate in a mix of 95 parts concentrated sulfuric acid and 5 parts hydrogen peroxide for 10 minutes. The substrate is then rinsed in DI water prior to fluidic self assembly.
A preferred treatment for treating plastic substrates (e.g. a flexible plastic substrate) is to expose them to gaseous ozone for 5 minutes. The most preferred method for treating a plastic substrate is to expose the substrate to an oxygen plasma for 1 minute prior to fluidic self assembly.
Besides surface interaction, the optimal fluids for FSA have low viscosity and can be used safely. It is possible to add reagents to the fluid to improve the FSA process. For example, surfactants or certain water soluble polymers can be added to water to reduce the block-substrate interaction, thereby reducing the number of excess blocks left on the substrate at the end of an assembly. If a polymer is used, it is desirable that the polymer serve as a lubricant while dissolved. Preferred polymers would consist of water soluble polymers such as Union Carbide Polyox (polyethylene glycol) or DuPont Elvanol (polyvinyl alcohol). If a surfactant is used, it is preferable to use a non-ionic surfactant so that the circuitry is not damaged by counterions in the surfactant. The surfactants normally have a molecular form which includes a hydrophobic portion and a hydrophilic portion; often, the non-ionic surfactant will include a hydrophilic portion which is an ethylene oxide oligomer. In a preferred method, a non-ionic surfactant such as Union Carbide MinFoam 1X or Triton 190 or DuPont Zonyl FS-300 is used in water to perform fluidic self assembly of blocks which have been surface treated with aqueous KMnO4 (or ozonated water) on a glass substrate that was oxidized with H2SO4 and H2O2. In this case, the water and surfactant and the blocks form the slurry for the FSA process, and the slurry is dispensed onto the pretreated substrate. Typically, the substrate is immersed in the FSA fluid (e.g. water in the immediately preceding example) which is the same as the fluid used to create the slurry having the blocks, and the slurry is added to the substrate while it is immersed in the FSA fluid. The substrate may be immersed in exactly the same fluid and a surfactant as the fluid and surfactant which make up the slurry.
Once the blocks and the substrate are prepared in the desired manner (e.g. the desired surface treatments have been completed), and the appropriate carrier fluid is selected, the blocks can be deposited onto the substrate in a number of ways. They can flow down an inclined tube containing the fluid so that they fall through the FSA fluid and onto the substrate. One or a plurality of these tubes can be used to deposit the blocks onto the substrate. Also, the tube or tubes can be moved across the substrate such that the entire substrate can be covered with block slurry prior to or during assembly. In another embodiment, the blocks could be carried through a tube or a number of tubes by a fluid flow that is either laminar or turbulent. The flow(s) could effectively spray the block slurry over the substrate to cover a selected portion of or the entire substrate. The aforementioned tubes could be circular, fan-shaped, or have a multitude of ports.
The fluidic self assembly process can be accomplished in a variety of methods. The blocks can be moved across the substrate surface in a number of ways, including forced fluid flow, suction, gravity, magnetic fields if the blocks have magnetic characteristics, or any combination of these driving forces. Forced fluid flow can drive the blocks across a tilted or horizontal substrate either by providing a uniform flow over the surface, or by using a very localized flow that can be directed in any desired manner. A magnetic field applied to magnetized blocks could draw the blocks into the receptor sites. Suction could be applied to holes located in the bottom of the receptor sites that go through the substrate to a vacuum source on the other side of the substrate. Gravity is used by tilting the substrate from horizontal to an angle not greater than 55 degrees. The blocks then slide down the substrate under the force of gravity. In a preferred embodiment, the tilted substrate is vibrated to drive the blocks down the substrate. The vibration frequency ranges from 50 Hz to 10,000 Hz, and can have a square, sine, sawtooth, or any other waveform. The direction of the vibration is in one embodiment transverse to the direction of the block slurry flow down the tilted substrate. In some instances it may be desirable to have components of the vibration in the direction of the block flow down the substrate, and in the direction normal to the substrate.
Once the blocks are moving across the substrate, the fluidic self assembly can proceed by either allowing all of the excess blocks (blocks that do not fill a receptor site) to move completely off of the substrate, or by forcing excess blocks to move off of the substrate. The excess blocks can be driven off of the substrate with fluid flow over the substrate, by altering the conditions of the vibration applied to the substrate during the assembly, by increasing the angle of inclination of the substrate, by suctioning off blocks, or by any combination of these methods.
The fluidic self-assembly process can be accelerated in a number of ways in addition to proper selection of the process parameters listed above. One method for speeding up the FSA process is to deposit blocks regularly or uniformly over the entire surface. This deposition process could result in a uniformly dense layer of blocks across the surface. This could be accomplished by spraying the block slurry across the substrate with one or more deposition nozzles. If desirable, the nozzle(s) could be designed to sweep across the surface to achieve more uniform coverage. Alternatively, the blocks could be deposited in selected areas of the substrate. These blocks could be placed by one or more fixed or movable deposition nozzles that deliver the blocks with the assistance of some combination of fluid flow, gravity, and vibrational impulse. Placing the blocks uniformly or regularly over the substrate surface accelerates the overall FSA process because less time is required for the blocks to move over the entire substrate. For example, if 10 rows of blocks are deposited regularly over the substrate surface, the blocks will only have to move one-tenth the distance across the substrate as compared to the case in which just one row of blocks is placed at the top of the substrate. Once a sufficient number of blocks has covered the areas of the substrate containing the receptor sites, such that the desired number of the receptor sites are filled, the excess blocks can be forced off of the substrate.
The excess blocks can be removed from the substrate by the methods listed above. The excess block clearing process can be accelerated with the addition of components designed to push blocks off of the substrate. Any combination of a number of different components could accomplish this task. One component that can increase clearing is a wiper blade that moves across the substrate. Blocks that are in receptor sties will be unaffected, but blocks on the surface will be swept away by the wiper blade. Alternatively, a brush could be used in place of or in addition to the wiper blade. The components that clear excess blocks from the surface could be one or more large items, relative to the size of the blocks, that move down the substrate with the blocks, pushing blocks down and off the substrate as they move. These items may be referred to as helper blocks. Because these items are normally significantly larger than the blocks, gravity will drive them down the substrate faster than the blocks, and they will be able to push a large number of the blocks off of the substrate as they move down. These relatively large items could be glass, plastic, or metal balls, cylinders, or rectangular solids. More preferably, these items could have the same approximate shape as the blocks (but larger in size), or they could be shaped pieces of material that are designed to contact the substrate only where there are no receptor sites. Most preferably, these helper blocks could be rectangular pieces made of magnetic stainless steel. After removing the excess blocks from the surface, these items could be recovered and readily separated from the excess blocks by sieving or some other particle separation method or separated with a magnetic field in the case where the helper blocks are magnetic (e.g. the helper blocks are comprised of magnetic material).
After the excess blocks have been removed, the substrate can be inspected for faults such as empty holes or blocks that have not seated properly in their receptor sites. Empty sites can be filled by placing a small number of blocks in a slurry form and depositing this slurry on or above the empty site, and forcing or allowing the blocks to fill the empty site(s) by means analogous to those used to perform the assembly. Empty sites can also be repaired by a pick-and-place process, in which an individual block is placed in an individual receptor site. A receptor site with an improperly seated block can be repaired by removing the block with either suction, fluid flow, or mechanical means, and then filling the now empty receptor site as described above. Alternatively, blocks that are tilted or rotated in their receptor site could be repaired after the assembly is complete with a hot press process.
Once all desired repairs have been made, the blocks may be bonded into their receptor sites with an organic or polymeric agent that serves as an adhesive in this application. The bonding agent could be dissolved into the original FSA fluid prior to the start of the assembly, or it could be added any time later during the assembly, assuming that the bonding agent is compatible with the fluid used for the assembly. If the bonding agent is incompatible with the fluid used for the assembly, a fluid displacement process can be used at any time after the start of the assembly to replace the first fluid with a second fluid which is compatible with the bonding agent. In one example, acetone is used as the FSA fluid, and a water soluble polymer is used as the bonding agent. After completion of the inspection and repair process, water is added to the system and acetone is removed at approximately the same rate, until there is a sufficient percentage of water in the resulting water/acetone mix such that the bonding agent is soluble in the mix. In a different example, the FSA is performed in toluene containing benzocyclobutene oligomers which serves as a bonding agent. In another example, the FSA is performed in basic water that contains surfactant and a water soluble polymer that serves as the bonding agent. Alternatively, the FSA can be performed in water containing surfactant, and then a water soluble bonding agent dissolved in water can be added to the water/surfactant mixture. Typically the completed assembly is left immersed in the bonding solution for 5–30 minutes to allow the bonding agent to diffuse around the blocks in the receptor sites. At this time, if the substrate were tilted it is preferable to reduce or eliminate the angle of inclination, and, in some cases, to apply a low amplitude vibration to the substrate while the angle of inclination of the substrate is being decreased.
After the filled substrate has remained immersed in the bonding solution for the desired length of time, the substrate is either lifted out of the solution, or more preferably, the solution is drained away or evaporated from the substrate. Then the substrate is allowed to dry, either by free or forced convection. During the drying process, it is beneficial to tilt the substrate slightly, so that excess fluid does not pool and dry in the array of blocks.
Once the substrate is dry, the fluidic self assembly process is complete. At this time, if necessary, repair of blocks that are tilted in their receptor sites can be accomplished with the use of heat and mechanical pressure in a number of different ways. The substrate can be heated to a temperature that softens the bonding agent, and then pressure can be applied either locally with a point source, more globally with a roller apparatus, or globally with weight or other form of applied pressure. In this step it is important that the surface pressed against the filled substrate will not stick to the blocks in the receptor sites or to the substrate. In one example, Solutia Scripset 550 was used as the bonding agent, the substrate was heated to 120° C., and then pressed between perfluoroalkoxy polymer films with approximately 500 psi for 1 minute. The substrate is then cooled while under pressure. The pressure is released and the perfluoroalkoxy films were removed. In another example with the same bonding agent, a TEFLON roller, made out of TEFLON tubing placed over a steel rod, was rolled across the substrate after it was heated to 120° C. Both of these processes reduced the number of tilted blocks that remained after the completion of the FSA.
After the completion of the FSA, and the press repair if desired, the residual bonding agent that remains on the surface of the substrate can be removed with an oxygen plasma descum without damaging the bonding of the blocks in their receptor sites. At this point, the substrate is ready for planarization and the rest of the down stream processes required to electrically interconnect the blocks and construct an electrical apparatus.
An exemplary method according to another aspect of the invention will now be described in conjunction with
An exemplary method according to another aspect of the invention will now be described in conjunction with
There are several methods that may be used to make block receptor sites. The choice of method depends in part on the material to be used as the organic layer on the glass. If the organic material is an amorphous or semicrystalline polymer, the receptor sites may be embossed into the polymer material with a mold that matches the block size and pitch for the device being produced (e.g. an active matrix LCD). In this case, it may be preferable to adjust the mold to take into account the differences in thermal coefficient of expansion between the polymer and the mold material.
Embossing is just one method of forming the receptor sites in a polymeric coating on the glass. The coating may alternatively be injection molded onto the glass substrate during the coating process. Alternatively, the receptor sites could be formed in the polymer by solvent casting a polymer solution onto a receptor site mold. The solvent cast sheet could then be laminated or transferred to the glass substrate.
It is also possible to form the receptor sites in a thin film of either thermoset plastic or crosslinkable organic material on the glass substrate. The receptor sites are formed by placing the liquid organic starting material on the mold. It is preferable to treat the mold with a release coating, such as an oil, a fluorinated coating, or a low-surface energy self-assembled monolayer coating, or one of these materials combined with a metal layer that separates readily from these coatings, such as silver. The glass substrate is then pressed on top of the liquid organic material, such that the liquid flows around all of the features of the mold, and any bubbles in the liquid are removed. The glass/organic/mold stack is then exposed to sufficient heat or UV light, if necessary, for a sufficient time to cure the liquid organic material into a crosslinked solid material. The mold is then separated from the organic film on the glass.
It is also possible to pattern the material without a mechanical mold by using light energy. There are two techniques to accomplish this. First, standard photolithography techniques could be used on a photopatternable material to expose and develop away the volume of material in the receptor site. In this case, the photopatternable material serves as the substrate material. Alternatively, a mask layer could be used on top of the organic material, such that the organic material could be etched with a plasma etch system. This method would apply to a larger class of organic substrate materials because the organic material does not need to be photopatternable. It is also possible to form the receptor sites in the organic layer by using a laser drilling or laser ablation technique as described above in conjunction with
While the foregoing description has provided examples of the present invention, it will be apparent that various modifications may be made within the spirit and scope of the invention which is limited only by the following claims. For example, the order of the processing operations may be modified and the same or similar result achieved in the resulting structure.
This application is a divisional application of U.S. patent application Ser. No. 09/432,291, filed Nov. 2, 1999 now U.S. Pat. No. 6,623,579.
| Number | Name | Date | Kind |
|---|---|---|---|
| 3967004 | Oda et al. | Jun 1976 | A |
| 4194668 | Akyurek | Mar 1980 | A |
| 4397325 | Van Roeyen | Aug 1983 | A |
| 4542397 | Biegelsen et al. | Sep 1985 | A |
| 5167989 | Dudek et al. | Dec 1992 | A |
| 5403624 | DiMaio et al. | Apr 1995 | A |
| 5429676 | Uchida et al. | Jul 1995 | A |
| 5545291 | Smith et al. | Aug 1996 | A |
| 5904545 | Smith et al. | May 1999 | A |
| Number | Date | Country | |
|---|---|---|---|
| 20040026031 A1 | Feb 2004 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 09432291 | Nov 1999 | US |
| Child | 10634350 | US |
