Techniques for producing thin films with improved uniformity and printing speed

Abstract

To provide a method for processing a substrate by using a printer which deposits liquid so as to process a layer having prescribed thickness.SOLUTION: An improved manufacturing method includes automated adjustment or print parameters based on ink or substrate characteristics which have been specifically measured or estimated. In one embodiment, ink spreading characteristics are used to select droplet size used to produce a particular layer, and / or to select a specific baseline volume / area or droplet density that is then scaled and / or adjusted to provide for layer uniformity. In a second embodiment, details per expected droplet are used to carefully control mixing of the deposited droplets and to interleave the droplets so as to assist in layer uniformity. The liquid layer is then cured or baked so as to provide a permanent structure.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] (CROSS-REFERENCE TO RELATED APPLICATIONS) This disclosure is incorporated herein by reference in its entirety. No. 62 / 264,076, filed December 7, 2015, entitled "Techniques for Manufacturing Thin Films with Improved Homogeneity and Print Speed"; U.S. Patent No. 9,352,561 (USSN 14 / 340403), issued May 31, 2016, entitled "Techniques for Print Ink Droplet Measurement and Control to Deposit Liquids within Precise Tolerances"; U.S. Patent No. 9,010,899, issued April 21, 2015, entitled "Techniques for Print Ink Volume Control to Deposit Liquids Within Precise Tolerances"; and U.S. Patent No. 8,995,022, issued March 31, 2015, entitled "Ink-Based Layer Fabrication Using Halftoning To Control No. 20150373305 (USSN 14 / 840343), filed Aug. 31, 2015, entitled “Fast Measurement of Droplet Parameters in Industrial Printing System” (USSN 14 / 840343, filed Aug. 31, 2015); and U.S. Patent Application Publication No. 20150298153 (USSN 14 / 788,609), filed Jun. 30, 2015, entitled “Techniques for Arrayed Printing of a Permanent Layer with Improved Speed ​​and Accuracy.” Priority is claimed to U.S. Patent Application Serial No. 62 / 264,076. [Background technology]

[0002] Printers are used in a wide variety of industrial processes for fabrication. In one application, the printer can be used to jet a liquid onto a substrate, which is then cured, dried, or otherwise treated to form a permanent layer of an electronic device. The technique is particularly useful for the deposition of materials that cannot be easily deposited using other techniques, such as organic materials (such as plastics). One exemplary application involves the fabrication of display panels (such as televisions or other display screens), solar panels, and light-emitting devices. For example, the printer can be used to deposit organic LED materials that help generate light under the influence of an electrical stimulus, as well as organic encapsulation layers that help protect sensitive internal layers from exposure to oxygen, moisture, or other contaminants. In conventional processes used to fabricate large panel electronic devices, these printers are "room-sized" and can be used to print on very large substrates within very precise parameters as part of an assembly-line type process. Large panel devices can be formed from these substrates (e.g., large TVs or solar panels), or alternatively, many small devices can be formed on these substrates and then separated from each other. Of course, these applications are merely illustrative and the printer can be used in many processes to print many different materials (organic or otherwise) for many types of devices other than electronic devices.

[0003] As technology improves, manufacturing specifications typically require deposited layers to be thinner and less variable, all while providing the same or better performance characteristics. However, as dimensions shrink, it also becomes more difficult to produce very thin layers using the techniques described above. For example, while slight unevenness (e.g., micron-scale roughness) can typically be tolerated within the surface of thick layers (e.g., 30 microns or more for the applications listed above), the same variations can result in layer failure or other unacceptability in thinner layers (e.g., layers that are less than about 10 microns thick for the applications listed above). Unexpected thinness (i.e., thickness variations) across the layer geometry can potentially cause moisture or oxygen failure, loss of proper seal at the layer perimeter, or otherwise create significant performance defects.

[0004] To cite one example of these difficulties, it is expected that for organic light-emitting diode displays (OLEDs), manufacturing specifications will require layers that are less than 10 microns thick, with controlled uniformity of layer thickness and very precise boundary control. Such precision is difficult to achieve, especially when it is considered that conventional printheads can deposit 20-30 micron diameter droplets at pitches of 20-30 microns or more (e.g., if the printed droplets do not spread properly, they may produce layers with voids / gaps or that are either locally too thick or too thin). In this regard, droplet spreading and mixing of adjacent droplets can be a complex and non-precise process, especially outside of a laboratory setting. Factors such as droplet viscosity, surface tension, atmospheric temperature, substrate material, surface chemistry and geometry, static electricity, substrate temperature, nozzle errors, particulate contamination, and other factors can make it difficult to provide a manufacturing-grade reproducible process. Quite the opposite, it is typically desirable to have a printing process that is as fast as possible, e.g., to print a room-sized substrate in less than about 60-120 seconds, and the commercial viability of the process may be threatened if manufacturing takes too long. That is, for example, if the deposited liquid requires excessive settling time to form an even liquid coat, or if the deposited liquid spreads poorly or irregularly, the production process may be excessively expensive, or the resulting layer may have defects such as undesirable variations, irregularities, or gaps. To reiterate the points asserted above, it is difficult to print precision layers of liquid in a manufacturing process where the liquid thickness translates to final material thickness for very thin layers.

[0005] What is needed is a liquid that can be processed (cured, dried, or otherwise made a permanent part of the product) to form a layer having a desired thickness. A set of techniques that facilitates a reproducible manufacturing process using a printer to deposit layers. Additionally, what remains needed is a set of techniques for processing layers with greater control, so that the deposited layers have greater reliability and consistency, especially for thin organic layers. The present invention addresses these needs and provides other related advantages. Summary of the Invention [Problem to be solved by the invention]

[0006] The present disclosure provides techniques for improved manufacturing using printers that deposit liquids to fabricate layers having defined thicknesses and other desired layer properties. [Means for solving the problem]

[0007] The first set of principles involves manufacturing methods in which ink and / or substrate properties are specifically measured (or estimated or inferred) and then used to select or adjust the droplet deposition environment via printhead selection or control, or other control over specific ink droplet parameters. In this regard, taking into account the expected ink spread in the deposition process (e.g., depending on the substrate, process, ink, or other properties), the process can be specifically adjusted or selected to produce an acceptable deposition within specified tolerances of layer thickness and deposition time. In one application, the volume of ink per unit area (volume / area), droplet deposition pattern, and / or the linkage between droplet size and layer thickness is set or specifically adjusted depending on the expected spread properties of the deposition liquid (i.e., "ink") for a given manufacturing equipment, deposition environment, and printing process. For example, U.S. Patent No. 8,995,022 for "Ink Based Layer Fabrication Using Halftoning To Control Thickness" relates to a process in which layer thickness values ​​are converted to "ink packing" density (e.g., amount of ink per unit area of ​​substrate) to build a layer to have a desired thickness; the ink as printed has limited spreading properties, and once the relationship between ink volume / area and resulting layer thickness is understood, the scaled volume / area can then be used to "build" a layer to have a desired thickness. Among other things, this disclosure provides specific techniques that can be used to select suitable values ​​for that relationship. To convey an example, if a nominal amount of ink per unit area produces a blanket coverage of a substrate at 1 micron thickness, and a layer 3 microns thick is desired, the nominal amount of ink per unit area can be tripled to produce a layer of the desired thickness of 3 microns. According to the first set of techniques discussed herein, when droplet spreading is understood or quantified (e.g., through measurement, estimation, or inference), a reference volume per unit area can be selected that will result in blanket coverage and an increased understanding of the resulting thickness that would be produced using that reference amount of ink per unit area.This disclosure provides techniques for selecting deposition parameters that can be used to further this understanding, and the scaling techniques referenced above are then relied upon to produce appropriate layer thicknesses within specified tolerances at reduced deposition times.

[0008] It should be noted that several different parameters can be adjusted to achieve these objectives. For example, if a given ink and deposition process produces thick drops of limited spread (i.e., the ink drops cannot produce a layer of the desired thickness because the drops are too thick and the resulting layer would be too thick), a different print head can be selected or the nozzle electronic controls can be adjusted to produce smaller drops. In one design, instead of (or in addition to) changing the print head or nozzle drive signals, the nozzle printing grid (e.g., drop pitch) can be varied, e.g., by changing the firing frequency of the nozzles to vary the in-scan nozzle pitch, or by adjusting the print head to vary the effective cross-scan nozzle pitch. In yet another design, a print head is used that can feature multiple discrete drop sizes (e.g., 1x volume, 2x volume, 3x volume, 4x volume, etc.) and drop size selection based on measured, estimated, or estimated diffusion properties. Note that while one embodiment features drop sizes that are integer multiples of each other (or close to such values ​​in terms of volume), other embodiments feature much more subtle differences, with tuned or "customized" waveforms used to precisely customize the deposition volume. To the extent that the drop volume can be varied between multiple sizes, in one implementation, the largest drop size of a set of candidate acceptable drop sizes that can produce the desired layer is selected for use. For example, a method premised on these specific principles can determine whether each available drop size can produce the desired layer (in the volume per unit area required to obtain a blanket coverage) (i.e., the layer is not too thick after curing) based on expected diffusion. The maximum drop volume (and associated process characteristics) that meets this test is then selected, possibly in combination with other smaller drop sizes. If desired, an optional "buffer" can also be injected into the process to allow for margin of error associated with process variations. For example, if a given drop size produces a blanket coating at "exactly" the desired thickness, such a drop size can be rejected, with the selection of the maximum drop instead subject to the ability to produce the desired layer thickness minus some artificially injected buffer, e.g., minus 10 percent.

[0009] Numerous implementations and variations will occur to those skilled in the art. In the following description, various systems, devices, apparatus, processes, and combinations that accomplish these principles will become apparent.

[0010] A second set of principles involves understanding nozzle errors (e.g., volume aberrations, nozzle position or "nozzle bow" errors) or droplet landing location errors (e.g., a function of nozzle bow and / or droplet trajectory and / or tendency to "wick") and correcting nozzle selection and / or timing and / or droplet size to deposit ink such that the droplets have appropriate mixing characteristics. Note that any of these techniques can be optionally mixed and matched with the first set of techniques discussed above (and other techniques discussed below). In this regard, the surface tension characteristics of the droplets can be such that the droplets exhibit wicking characteristics that distort the effective droplet positioning and essentially pull sibling droplets out of position. Such problems can be greatly exacerbated by the factors and / or errors described above. Techniques are presented herein to mitigate this tendency through the use of filters that "stack" droplets over multiple scan passes such that the intermixing of adjacent droplets is carefully controlled (e.g., by adding a "linked" droplet at the end, or in an otherwise planned manner). Note that in one embodiment, nozzle and / or droplet position errors are intentionally taken into account in the print planning process, such that measured expected variations in droplet size and / or landing location are relied upon in planning the interlacing discussed above. For example, in a specifically contemplated implementation, droplets can be repeatedly measured per nozzle (or per available waveform per nozzle for all nozzles) to develop a statistical ensemble of measurements per nozzle (or per "nozzle-waveform combination"). This is in accordance with the well-known n 1 / 2This helps to reduce the measurement error according to the relationship (i.e., with respect to the number of measurements n) and helps to define accurate expectations and deviation / variance measures of the drop volume and / or landing position per nozzle and / or per nozzle-waveform combination. The statistical parameters can then be taken into account in the layer-by-layer and / or printing process to generate a uniform drop density using the layer-by-layer multi-scan process discussed above. As above, other techniques discussed herein may also be specifically addressed, even if not specifically addressed in connection with each detailed embodiment. Although not listed above, these techniques can be mixed and matched at will. For example, different droplet sizes can be optionally combined with interleaving, for example, using larger volume droplets in one pass and "interleaved" smaller volume droplets in another pass. To cite another example, these techniques can be optionally combined with in situ dynamically measured per nozzle data, as disclosed by U.S. Pat. Nos. 9,010,899 and 9,352,561, and other patents and publications incorporated by reference. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is an illustrative diagram showing several techniques for producing thin films with improved uniformity and printing speed. [Figure 2A] FIG. 2A is an illustrative diagram of a droplet of liquid (209) from a printer immediately after the droplet has established contact with a substrate (205), where the desired thickness of the film is represented by horizontal line 207 and the effective diameter of the droplet's contact area is represented by "d1." [Figure 2B] FIG. 2B is an illustrative diagram similar to FIG. 2A, but in which the droplets (213, 215, and 217) after spreading are both "too thick" and "too far apart," resulting in height error ("he") and gap error ("ge"), where the effective droplet diameter is represented by "d2" while the maximum droplet height is represented by "h2." [Figure 2C]FIG. 2C is an illustrative view similar to FIG. 2B, in which the droplets have spread and mixed to provide a continuous film having a height difference ("hd") for the desired thin film thickness (207), where the effective droplet diameter is represented by "d3" while the maximum droplet height is represented by "h3A." [Figure 2D] FIG. 2D is an illustrative view similar to FIG. 2C, where the droplets have spread and mixed to a further extent such that the layer (223) is uniform and a reasonable approximation of the desired thin film thickness (207), and the effective droplet contact area diameter after spreading is effectively represented by "d4." [Figure 2E] FIG. 2E is an illustrative diagram similar to FIG. 2D in which the droplets are spreading or otherwise deposited to form a uniform layer (223), but are taking a very long time to "settle" (i.e., to completely spread), represented conceptually by the two illustrated droplets and the varying diameter arrow "d5," and it may be difficult to form the desired film thickness in a reasonable amount of time. [Figure 3A] FIG. 3A is an illustrative diagram showing a hypothetical printing grid 305 and possible landing locations for droplets during printing, where two idealized droplet landing locations 308 and 311 are designated with distinct spreads represented by circles (numbered 309 and 312). [Figure 3B] FIG. 3B is an illustrative diagram similar to FIG. 3A, but instead showing undesirable wicking of the center of the droplet from FIG. 3A to have an effective center at 316 along with potentially unpredictable dot gain 317 and coverage area. [Figure 3C] FIG. 3C is an illustrative diagram showing how droplet stacks can be layered to prevent undesirable wicking. [Figure 3D] FIG. 3D is an illustrative diagram showing how a filter (335) can be applied to the droplet deposition pattern (334) to achieve layer-by-layer stacking (see FIG. 3C). [Figure 3E]FIG. 3E is an illustrative diagram showing unintended droplet volume variations (e.g., depicted dot gain differences for individual droplets, represented by the “circles” numbered 343, 345, 347, and 349) and trajectory offsets and variations (represented by the arrows shown for each deposited droplet). [Figure 4A] FIG. 4A is an illustrative diagram similar to FIG. 3E, but showing pattern / filter adjustments (509) to pre-distort the interleaving pattern to accommodate droplet volume variations and / or offsets and trajectory variations (i.e., such that the desired interleaving results in proper ordering and positioning of droplets despite the aforementioned variations). [Figure 4B] FIG. 4B is a hypothetical droplet deposition pattern showing the landing positions (and volumes / densities) of deposited droplets interleaved using the adjustment technique of FIG. 4A, where the droplets are effectively selected to compensate for unintended droplet variations across the nozzles (e.g., droplet size variations, or nozzles that do not fire at all, such as "nozzle number 5"). [Figure 4C] FIG. 4C is an illustrative diagram similar to FIG. 4B, but in which a selected nozzle (e.g., “Nozzle No. 4”) has been selected to produce an extra-large droplet size (473), i.e., to effectively compensate for an unintended droplet variation (e.g., the inability of “Nozzle No. 5” to fire a droplet). [Figure 4D] FIG. 4D is an illustrative diagram similar to FIG. 4B but showing the use of a print head step or offset between scans so that the same nozzle can be used in multiple locations and so that multiple nozzles can be used to deposit different sized / emitted drops at the same print grid location. [Figure 5A] FIG. 5A is a flow diagram illustrating an embodiment of a technique for producing thin films with improved uniformity and printing speed. [Figure 5B] FIG. 5B is a plot of layer thickness as a function of the volume of ink per unit area of ​​substrate (eg, "ink packing density"). [Figure 6A]FIG. 6A illustrates several alternative techniques for printing process planning / adjustment to produce thin films with improved uniformity and printing speed. [Figure 6B] FIG. 6B also shows several alternative techniques for printing process planning / adjustment to produce thin films with improved uniformity and printing speed; optionally, per numeral 665, the deposition per nozzle can be adjusted by varying (programming) the waveform used to drive any individual nozzle. [Figure 6C] FIG. 6C is a flow diagram for an embodiment in which multiple alternative nozzle driving waveforms are programmed into the circuitry associated with a given nozzle. [Figure 6D] FIG. 6D shows a hypothetical nozzle driving waveform in an embodiment in which the nozzle driving waveform (and associated droplet size, velocity, trajectory, etc.) can be customized. [Figure 7A] FIG. 7A illustrates one embodiment of a processing apparatus for flat panel electronic devices that includes multiple modules (eg, 703, 705, and 707), one of which includes a printer. [Figure 7B] FIG. 7B is an illustrative diagram of the control of a printer used in the apparatus of FIG. 7A. [Figure 7C] FIG. 7C is a block diagram of the electronic control system for the printer module of FIG. 7A. [Figure 7D] FIG. 7D is an illustrative diagram showing the integration of a drop measurement device with a printer. [Figure 8A] FIG. 8A is a plan view showing printing onto a substrate (801) via multiple scans (eg, 807 and 808), where multiple large panel devices are represented (eg, 802). [Figure 8B] FIG. 8B is an enlarged view of the substrate. [Figure 8C] FIG. 8C is a cross-sectional view of the substrate of FIG. 8B taken along arrow CC from FIG. 8B. [Figure 8D] FIG. 8D is a top view of a pixel well. [Figure 9A] FIG. 9A is similar to FIG. 8C, but with an encapsulation layer 902 added to it. [Figure 9B] FIG. 9B is a plot of layer thicknesses of different layers produced using certain edge treatment techniques discussed herein. [Figure 9C] FIG. 9C is a plan view of a corner of a flat panel electronic device having an edge treatment technique applied to it. [Figure 9D] FIG. 9D is another plan view of a corner of a flat panel electronic device with an edge treatment technique applied to it. [Figure 9E] FIG. 9E is another plan view of a corner of a flat panel electronic device with an edge treatment technique applied to it. [Figure 9F] FIG. 9F is an illustration of a printing grid droplet deposition pattern specially configured for use in the corners of a layer. [Figure 9G] FIG. 9G is an illustration of another printed grid droplet deposition pattern specifically configured for use in the corners of a layer (eg, using multiple droplet sizes). DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] The subject matter defined by the recited claims may be better understood by reference to the following detailed description, which should be read in conjunction with the accompanying drawings. This description of one or more specific embodiments, which are designed below to enable construction and use of various implementations of the technology described by the claims, is not intended to limit the recited claims, but to illustrate their application to certain methods, systems, apparatus, and devices. In particular, specific examples are presented in the context of fabricating solar panels, displays such as organic light emitting diode ("OLED") display devices, and other light emitting devices, and more particularly, in the context of fabricating very specific thin film layers, but the principles described herein may also be applied to other methods, systems, apparatus, and devices. (Detailed Description)

[0013] With the general principles thus introduced, the present disclosure will now describe several detailed implementations.

[0014] First, it should be noted that some terms used herein should be introduced. First, references to "ink" should be understood to refer to a liquid that has been or will be printed onto a substrate, and then the inherent liquid coloration or specific shade is generally processed (e.g., cured, dried, polymerized, solidified, or otherwise converted into a permanent part of the product) to form a material having a thickness that is not immediately significant, which is different from, for example, graphics applications where the ink has a specific color or shade property that is transferred to a medium to directly impart an image pattern, i.e., such "ink", once applied, is typically deposited to produce an immediate visual effect without a substantial (or critical) thickness. In the applications discussed herein, the deposited liquid may ultimately be used for some graphics or display application (e.g., the deposited liquid may be used to form a red, green, or blue pixel emissive layer, or otherwise used to produce light, color, or other optical effects when ultimately electrically activated, according to the principles herein), but the point is that the material is deposited for some type of mechanical, structural, optical, or electrical property that will interact with other layers. Second, references to "halftone" or "halftoning" as used herein similarly do not refer to the traditional graphics use of these terms, but rather should be understood to suggest that inks may be printed to provide a specific volume per unit area and impart thickness, flow rate, diffusion, mixing, or other mechanical (or material) properties of the deposited layer. That is, while "halftoning" in graphics applications traditionally refers to the use of droplets of one or more colors in relative amounts to directly convey a visual characteristic, here "halftone" as used herein simply refers to a droplet pattern that will create a liquid layer of a specified thickness after layer processing."Density," as used with respect to ink, refers to such volume per unit area, and can be adjusted, for example, by (1) varying individual drop size / volume, (2) changing the printing grid pitch or otherwise adjusting the density of drops per unit area of ​​the substrate (e.g., more drops are printed into a given unit area), or (3) other techniques. A "printing grid" is the area in which ink drops will theoretically fit into a given mark. It refers to an imaginary mesh representing all possible "dots" or concrete locations that can be deposited given the electromechanical properties used to print in a print scan. Typically, the separation in this grid along a first axis is a function of the effective nozzle pitch (nozzle separation) in the direction of the first axis, while the separation along the second axis is a function of the digital electronics (e.g., representing the maximum frequency at which the nozzles can be fired in a given print scan). It should be noted that some techniques discussed herein can be described in terms of varying the print grid spacing, e.g., adjusting the frequency of the digital signal used to time (i.e., trigger) the nozzle firing of droplets, changing the mechanical scan speed, rotating the print head (to effectively change the nozzle pitch according to a sine function of the rotation angle), partially / incrementally advancing the print head (between or between scans) along the in-scan / cross-scan or other axis, and the like, and these various mechanisms are non-limiting examples. A "circuit" can refer to analog or digital electronic elements (e.g., special purpose logic gates) arranged either as a special purpose circuit that, when electrically stimulated, necessarily performs a certain function, or as a general purpose circuit (e.g., a processor, FPGA, or other configurable circuit) that is controlled or otherwise configured by instructions (software) to adapt the circuit to perform a specific function and cause the circuit to behave as if it were a special purpose circuit. In the case of software or other instruction logic, the instructions are typically written or designed in a manner that has a certain structure (architectural features) such that, when these instructions are ultimately executed, they necessarily cause one or more general purpose circuits or hardware devices to perform a certain described task. "Logic" can refer to software logic (i.e., instruction logic) or hardware logic (e.g., digital chip or board design), or a combination of these."Non-transitory machine-readable medium" means any tangible (i.e., physical) storage medium, including, but not limited to, random access memory, hard disk memory, optical memory, floppy disk or CD, server storage, volatile memory, memory card, and / or other tangible mechanism from which instructions may be subsequently read by a machine, regardless of the manner in which the data on the medium is stored. The machine-readable medium may be in a stand-alone form (e.g., in a program disk, a solid-state memory card, whether bootable or executable or otherwise, or other memory), or may be embodied as part of a larger mechanism, such as a laptop computer, a portable or mobile device, a server, a data center, a "blade" device, a subsystem, an electronic "card", a storage device, a network, or other set of one or more other forms of devices. The instructions may be implemented in different forms, such as, for example, as metadata that is effective to invoke a certain action when invoked, as Java code or script, as code written in a specific programming language (e.g., as C++ code), as a processor-specific instruction set, or in some other form. The instructions can also be executed by the same processor or common circuitry, or by different processors or circuits, depending on the embodiment. For example, "instructions stored on a non-transitory machine-readable medium" typically refers to software stored on a disk or other physical memory or storage device, which is structured such that when later (eventually) installed or executed by an operator or end user, it configures a machine (e.g., one or more processors) to operate in a specified manner. In one implementation, the instructions on the non-transitory machine-readable medium can be executed by a single computer, and in other cases as described, can be stored and / or executed on a distributed basis, for example, using one or more servers, web clients, or application-specific devices, whether co-located or remote from each other.Each function described in this disclosure or in the figures may be implemented as part of a composite program, or on a single media representation (e.g., a single floppy disk), or on multiple separate storage devices, or on dedicated circuitry or the like. The software may be implemented as a stand-alone software module (i.e., a program or subroutine that can be run or called) stored together, either in the form of a circuit combined with such software. "Blanket" coverage or "complete" coverage refers to the notion that, at least locally speaking, for example, the deposited liquid (after any allowable spreading or settling time) per unit area of ​​the substrate has no voids or gaps that are not covered by the liquid. Note that in most processes discussed herein, droplets are deposited at discrete locations on the substrate, and these droplets spread and then mix together according to some decay function to spread and wet the substrate, and "blanket" coverage typically assumes that the droplets require some non-zero spreading time after printing for mixing in this manner. "Module" as used herein refers to a structure dedicated to a specific function, e.g., a "first module" performing a first specific function and a "second module" performing a second specific function refer to mutually exclusive sets of code when used in the context of instructions (e.g., computer code). When used in the context of mechanical or electromechanical structures (e.g., a "printer module," which refers to a dedicated set of components that may also include software, depending on the embodiment), for example, a "printing module" and a "curing module" will refer to mutually exclusive structural elements dedicated to performing those functions. In all cases, the term "module" is used to refer to a specific structure for performing a function or operation that would be understood by those skilled in the art to which the subject matter pertains as a conventional structure used in the particular technical field (e.g., a software module or a hardware module) and not as a general placeholder or "means for any and all structures" to perform the described function (e.g., a "pack of bulls")."Processing," when used to refer to a post-deposition step applied to a liquid ink on a substrate, refers to a step applied to the liquid ink to make it a permanent part of the device, for example, by curing, solidifying, drying, polymerizing, or otherwise converting the liquid into a form (flexible or otherwise) suitable as a permanent structure of the device.

[0015] FIG. 1 is an illustrative diagram showing several techniques for producing thin films with improved uniformity and printing speed. More specifically, FIG. 101 provides a series of functions, generally identified by reference numeral 101. These techniques include the use of at least one processor (represented by computer icon 103) to plan and / or control printing by an industrial processing printer 105. The at least one processor 103 operates under the aid of software, represented by floppy disk icon 107. When they are executed, the software (i.e., instructions) cause the at least one processor to perform specific functions (109) that will plan the printing to produce a desired layer thickness. As previously discussed, this layer can be a thin layer of a specific material that will be used as a solar panel, a display (e.g., an electronic flat panel device), or another form of light emitting device. For example, these processes can be used to plan and then effect printing on a substrate (130) that will ultimately form a mobile device display (139), another type of display (e.g., a TV screen 141), or a solar panel (143), or an array of such devices or other electronic devices (represented by the arrayed substrate 137).

[0016] As represented by dashed starting block 113, the processing equipment and / or its control system receives layer thicknesses for layers to be processed on the substrate. The layers can be deposited only locally (e.g., specifically within the confines of a pixel well to form a light emitting element layer or other type of layer in an electronic device stack) or over a large area of ​​the substrate (e.g., to span many electronic structural components or to provide other effects, such as optical effects produced using a high refractive index material or specific combination of materials, smooth surfaces, or other layers. The ink may be deposited in a variety of applications (as a barrier, electrode, insulator, or encapsulation layer). Regardless of the application, in this embodiment, the layer is assumed to have a flat top surface and a specifically desired (uniform) thickness in a defined target area, and the instructions cause the control system to select a nominal volume per unit area of ​​ink, as represented by numeral 111. Note that, as previously mentioned, in other examples, this thickness may be varied by target area if desired (e.g., to compensate for variations in the underlying surface geometry) or adjusted in a manner consistent with an edge taper or other specific contour. The control system optionally utilizes the (measured, inferred, or estimated) diffusion properties (115) of the ink that will be used for printing to select this nominal volume per unit area. Note that for the applications discussed herein, it is generally desired to have a uniform film free of defects within the intended deposition area, i.e., without voids / gaps and with minimal unintended thickness variations. To this effect, the reference volume per unit area is typically achieved through the density of droplets that, once deposited (i.e., printed), will mix with each other and spread to provide blanket coverage of the intended target area of ​​the substrate. In the case where the specific layer is an element of an electronic layer stack (e.g., a functional layer in a light-emitting element), the target area is the intended geometry of the layer (e.g., with a pre-established layer height and width). In the case where the specific layer is a continuous layer spanning many electronic components (e.g., an encapsulation, barrier, smoothing, or similar layer), the target area has an area that is at least as large as or smaller than the structural feature size. In simple terms, the reference volume per unit area is what produces a blanket ink coating that covers the entire area of ​​interest (i.e., the "target area"), where the layer has a uniform height with as few defects or aberrations as possible.Generally speaking, the droplets typically spread over a period of time depending on the ink viscosity, surface energy, chemistry and topography, and other properties, and after the spreading period the printed liquid then substantially stabilizes and is then treated to cure or otherwise solidify the ink, harden or otherwise convert it into a permanent structure in the electronic device.

[0017] To improve layer uniformity, smoothness, printing time, and other layer properties, the system plans the printing according to the principles discussed herein (117). The planning can include adjusting the reference volume per unit area or other printing process parameters according to the techniques discussed herein. In one optional embodiment as introduced above, a scan filter (118) can be utilized to layer droplets over multiple scans (so that droplets coalesce in a more predictable manner), and the filter and / or pattern (or density) can optionally be adjusted to account for nozzle errors and / or other process variations (121). As an example, if individual nozzles and / or combinations of specific nozzles with specific drive waveforms ("nozzle-waveform combinations") produce aberrations (123) relative to droplets from other nozzles, this can be specifically measured in situ (125) and applied to adjust the scan filter or otherwise adjust the droplet deposition pattern (121). In another optional embodiment, details of the nozzle-waveform combinations are dynamically measured and re-measured to develop statistical populations of measurements and to so derive measures for drop volume and / or drop landing location and spread measures (e.g., variance, standard deviation, 3 sigma measures, etc.) to account for changing conditions (e.g., ink viscosity, temperature, nozzle clogging, and many other possible process parameters). These values ​​can be stored in system accessible memory (127) for use in the print planning process (117), scan filter selection (118), or adjustment process (121) described above. In yet another optional embodiment, the drop firing signal (or nozzle pitch) is adjusted, for example, by adjusting trigger signal frequency or delay values, or by rotating the printhead (thereby effectively changing the printing grid spacing), so as to vary the reference volume per unit area or other printing process details, as applied. Other configurations and / or forms of adjustment are possible, and some of these techniques can be mixed and matched with one another and are considered optional features relative to one another. When planning is completed, the result is a set of control data (129) that can be stored in the system memory (127) and / or immediately applied to printing, for example, as referenced above, printing can be performed according to the control data (i.e., on a new substrate 132, according to the adjusted filters / data, as represented by numeral 131). The substrate can then be advanced for processing (curing or other processing) 133 as appropriate, or otherwise finished (135). Depending on the application, each substrate can ultimately be formed into a dedicated product (e.g., a single large HDTV screen, etc.), or can be processed as an array of panels, which are then cut into a number of electronic panel devices, as per numeral 137.

[0018] It should be noted that, as indicated above, the diffusion characteristics may be measured in advance in one implementation. In one embodiment, this measurement can be performed using one or more test deposition processes on a test substrate, ideally using the same printer, ink, printhead, deposition environment, temperature, and other process factors that would be used in an assembly line setting. That is, because process corners may vary from application to application and from equipment to equipment, in one embodiment, the processing device arrangement is used in situ in one or more trial runs, using inks that would be used during assembly line type processing, which are then subjected to automated (e.g., image-based) or manual analysis to measure diffusion parameters, which can then be programmed into the system or otherwise applied in the manner indicated herein. If desired, even for fixed processing applications, measurements can be re-performed over time to account for changing process conditions (e.g., equipment wear, ink fluctuations, temperature changes, or other factors). As previously noted, the use of measured parameters (and / or measured statistical parameters based on multiple trial runs) is not required for all embodiments.

[0019] Although some basic processes have been thus introduced, it may be useful at this point to explore certain characteristics of ink printing for industrial processing, where the ink will become a permanent structural component of an electronic device. An exemplary application that will be used in these discussions is the processing of organic LED ("OLED") display panels, however, as previously mentioned, the techniques discussed herein are not so limited. In contrast to printing of liquids that will be absorbed into a medium (e.g., paper to impart specific color properties, thus becoming "dimensionless" for most applications), these applications deposit liquids that for the most part are not absorbed into a support medium, but rather that retain structural characteristics such as height that will be processed to form a desired structural layer (e.g., by curing a deposited monomer liquid in situ to form a polymer layer with a specifically planned thickness, etc.). Figure 2A is used to introduce such a deposition process, generally through the depiction designated by the numeral 201.

[0020] More specifically, a droplet 203 of liquid ink will be ejected from an inkjet printer and "directed" to a specific desired landing location 204 on a substrate 205 (an "x" marks the spot). The substrate can be any desired support medium, e.g., glass, metal, or another material, with or without other layers already deposited to create the underlying surface geometry. For example, the surface onto which the droplet is deposited can be any type of structure or layer stack, with potentially complex chemistry and / or geometry. The droplet may flatten slightly on impact, as represented by numeral 209, and land slightly off-center relative to the specific desired landing location 204, depending (of note) on the process corners for the particular fabrication process, nozzle, printhead, drive electronics, etc. This type of aberration will be discussed in more detail below. The desired final The actual bed height is represented by line 207 and, as the droplets land, the value "d 1", and it should be noted that this droplet coverage area is modeled as a circle, however, droplets in practice can land in many different shapes (e.g., elliptical, teardrop-shaped, etc.) and the concept of a circular impact area will be used in conjunction with Figures 2A-E simply for illustration and explanation.

[0021] After the droplets collide, they "settle" and spread out slightly according to a time decay function, i.e., they land with a shape modeled by the depiction in FIG. 2A and then flatten out (or "spread out") according to properties such as ink viscosity, surface energy of the substrate and / or any underlying surface geometry, temperature, ink surface tension, and other parameters.

[0022] FIG. 2B shows that three droplets (213, 215, and 217) are deposited and allowed to settle on substrate 205, with each droplet having a maximum height after spreading ("h") that represents the amount of substrate covered by the droplet. 2 " ) and diameter (" d 2 2B provides a depiction (211) of the droplets that are considered to have a droplet height ("h") of 0.1 mm. As depicted in this figure, the deposited droplets do not contact and mix with one another, resulting in a gap error ("ge") that represents voids and other defects in the resulting structure layer. FIG. 2B also provides a depiction (211) of the droplet height ("h") of 0.1 mm. 2" ) exceeds the desired layer height in some places (represented by line 207), thus producing a layer that is too thick in some places and too thin in others. This type of deposition may be unacceptable for many applications, for example, for electronic applications, current through the deposited layer may be impeded or shorted by the depicted voids, or otherwise produce irregularities. For color filter applications (e.g., backlit displays) and emissive layers, these aberrations may result in color variations, thus producing line effects and other undesirable defects in the resulting display device. For encapsulation applications, these aberrations may expose sensitive electronic components to moisture, oxygen, and other harmful contaminants, thus degrading device life. Other problems may also be produced depending on the application, for example, if the layer is to be a smoothing layer or barrier, the result can be a smoothing layer that is not smooth or does not otherwise provide an effective barrier.

[0023] FIG. 2C is similar to that described above, except that the droplets have spread to a greater extent ("d 3 "), where the height ("h 3 ”) and are intermixed to the extent that there are no more gaps or void defects, and thus these droplets and process details are now provided as potentially suitable for use in producing the desired layer thickness 207. 3 Note that the maximum thickness (maximum height "h") is "close" to the desired layer thickness 207, and thus, by appropriate "halftone" selection (including, for example, droplet density adjustment or other volume per unit area adjustment), the layer parameters can be adjusted to "enlarge" the maximum thickness to approximately equal the desired layer height 207. However, the depicted droplets 213, 215, and 217, while they do diffuse and mix, still do so only to a limited extent, and thus the apex (maximum height "h") is "close" to the desired layer thickness 207. 3A "), valley (having a minimum height "h 3B "), and the height difference corresponding to the variation between these two limits ("h d"). The layer depicted is suboptimal because, due to the surface roughness, portions of the deposited layer would be too thick and / or too thin (potentially representing weak points where oxygen or moisture may penetrate the layer or other defects may occur as discussed above). Note that manufacturers' specifications typically set maximum allowable roughness or height variation permitted in the deposited layer, and this standard may be used in a particular application to determine the extent to which roughness (e.g., as depicted in FIG. 2C) may be considered acceptable or excessive. However, even within a manufacturer's stated tolerances, non-uniformity can adversely affect device performance, potentially causing optical (if This can lead to optical or electrical variations that produce distortions (or other) in the image.

[0024] FIG. 2D depicts (via depiction 231) a layer 233 that corresponds to a printing process detail that is highly suitable for producing the desired layer 207. That is, as can be seen, the deposited droplets are very large (i.e., “d 4 ) so that the deposited layer 233 is perfectly flat and has a height ("h 4 That is, assuming that the volume per unit area is less than a maximum value (e.g., a drop density that does not use all drop firing points), the volume per unit area can be adjusted to a height ("h") to exactly match the desired layer thickness. 4 For example, the plotted value ("h 4 ") is 60% of the desired bed height (207) and the height ("h 4") is well below the maximum drop density, then increasing the drop density by two-thirds should produce exactly the desired layer. The depiction of FIG. 2D represents an optimal layer that is much less susceptible to defects than the previous example, since the top surface of the layer is flat and lacks the surface roughness depicted in FIG. 2C. Also, note that if the volume per unit area represented by the figure is close to the maximum density supported by the printing grid, then additional scans can be added to the printing process to produce the desired thickness after curing (e.g., using additional scans to deposit ink at a volume per unit area above the maximum obtainable with a single scan), if such is permitted within any manufacturing timing budget.

[0025] Finally, FIG. 2E shows that layer 243 has an excessive settling time (i.e., it eventually grows to height “h 5 "), providing a depiction 241. In this figure, layer thickness is not an issue, but rather the solid and dashed droplets and the variable diameter "d 5" is illustrated to imply that a very long deposition process will ultimately be required. In this regard, it is noted that many consumer products have very sensitive price points, and if, for example, the deposition (e.g., printing) process takes too long, this increases the manufacturing cost per product, and the resulting (increased) price point can make a substantial difference in consumer demand, for example, with respect to whether the buying public is one to purchase a flat panel OLED TV or a competing technology. Generally speaking, in the context of current manufacturing applications envisioned by this disclosure, the illustrated system and process must be able to print a liquid layer over a "room-sized" substrate in less than about 120 seconds to be commercially viable versus other technologies, and it is considered that for future product generations, this time is expected to be significantly less (e.g., 45 seconds or less). Thus, the deposition details represented by FIG. 2E and its associated description are less desirable than those represented by FIG. 2D and its associated description, since FIG. 2D is generally associated with a much faster overall printing process.

[0026] These considerations are not the only ones that affect the ability of a printing process to produce a desired layer. Figures 3A-3E are used to introduce other considerations regarding some of the techniques described herein. Generally speaking, these figures each depict a print head 303 and a print grid 305, the latter having a vertical axis (e.g., 306) and a horizontal axis (e.g., 307) that each define a nodal point at their intersection that may be a location where an ink droplet can potentially be printed during the scanning motion (304) of the print head 303 relative to the substrate (not seen in Figure 3A). The spacing between the vertical axes represents the nozzle pitch, which may appear in several ways. For example, in one embodiment, a single print head may be used, and the spacing between adjacent nozzles of the print head determines this "cross scan pitch" (e.g., "Δx" between nozzles 1, 2, 3, 4, 5, etc.). In yet another embodiment, the depicted print head may be moved an incremental distance, for example, between successive scans, such that the nozzles are spaced apart by 10 degrees. 0 microns apart, but if the printheads can be moved "horizontally" in 25 micron increments, then the "cross-scan pitch" would be 25 microns and multiple scans would be used to obtain 25 micron spacing between drops. In yet a third variation, multiple printheads could be used (e.g., each offset "horizontally" from one another by an amount (e.g., 25 microns) to obtain this same effect during a single scan). In a fourth variation, the printheads could be rotated to effectively adjust the "cross-scan pitch". Each of these options is collectively represented in these figures by a single printhead depiction for ease of illustration purposes, but it should be understood that each of these options is represented by the depicted printhead and that whatever design is used, the effective nozzle spacing for purposes of scanning is modeled as the cross-scan pitch. During scanning, the printhead also moves a distance (304) per unit time ("Δt"), thereby defining the scan speed. As the print head moves, nozzle firing signals can be triggered by a digital clock at a maximum firing frequency that defines the "in-scan pitch" (represented as "Δy" in the figures and the associated separation between the horizontal axes). Again, it should be understood that there are other designs for placing droplets at "vertical intervals" along the depicted print grid, and Figures 3A-E should be considered to encompass these various options.

[0027] 3A shows a depiction 301 in which droplets are fired at two nodes 308 and 311, and the droplets are assumed to land exactly as planned at the depicted locations (308 and 311), and are assumed to have dot gains corresponding to circles 309 and 312. The depicted droplets will impact the substrate and spread as shown above, ideally mixing together in the overlap area of ​​the depicted dot gains (i.e., where circle 309 meets circle 312).

[0028] However, in practice, surface energy, ink liquid surface tension, and other parameters mentioned above can affect how the droplets mix together, leading to unpredictable results. This is generally illustrated in FIG. 3B in the depiction designated by reference number 315. That is, rather than having ink covering the substrate, as represented by the merging of circles 309 and 312 from FIG. 3A, the droplets can pull (e.g., "wick") away from one another, producing, in this example, an effective droplet center at 316 (which does not correspond to a printing grid nodal location, but instead represents an unpredictable location and coverage represented by ellipse 317). If it were erroneously assumed that the droplets would provide coverage as shown by the merging of circles 309 and 312 in FIG. 3A, the actual merging represented by FIG. 3B could also result in defects such as voids / gaps, excessive thickness, and surface roughness.

[0029] 3C provides a depiction 321 used to illustrate a potential solution to this problem, specifically interleaving (i.e., distributing) droplet landing locations into different scans in a manner depicted by circles (droplets) labeled "A" and circles (droplets) labeled "B." The droplets labeled "A" are deposited on a first pass of the print head 303 over the substrate, while the droplets between (labeled "B") are deposited on a second pass of the print head 303 over the substrate. In this way, the droplets labeled "A" do not mix (or only slightly mix) on their own, and the droplets labeled "B" are then added later such that they are positioned between the "A" droplets (and much closer to the "A" droplets than the other "A" droplets) and simultaneously induce mixing of the surrounding droplets in a manner that reduces the likelihood of undesired wicking. The properties of the liquid ink (and any other relevant factors) are such that this inter-layering and simultaneous intermixing with the lower density droplets deposited in pass "A" does not substantially alter the intended centers of the various droplets. That is, for the example of FIG. 3B, the "B" droplets simultaneously intermix with multiple neighbors and the deposited droplets. It is not intended to lead to unpredictable or undesirable "wicking" of the printhead. Note that in this example, it is assumed that the print grids for scan "A" and scan "B" are the same, that only two scans are used (e.g., it is possible to use three or more scans to interleave the droplets), and that the droplet sizes in passes "A" and "B" are the same, but none of these are required by any embodiment. That is, for example, contemplated variations feature droplet sizes in passes "A" and "B" that are purposefully selected to be different (e.g., droplets in pass "A" have three times the volume of those in pass "B"), the use of multiple droplets in some locations (e.g., at nodal locations labeled "A"), and / or the use of three or more scans, and in fact many such variations are possible.

[0030] FIG. 3D illustrates one possible method for performing layer-by-layer deposition, generally designated by numeral 331. Again, the described technique can be embodied as an apparatus or system, and can be embodied in the form of instructions (e.g., software) stored on a non-transitory machine-readable medium. The method first identifies or reads a reference volume per unit area 332, e.g., a density of specific volumetric droplets or a combination of droplets of different volumes, calculated according to the principles discussed above and stored in a machine-accessible digital memory. This volume per unit area or corresponding droplet volume / firing pattern can be selected to produce a minimum thickness "blanket" layer (e.g., the thinnest layer that can be printed using the available droplet size while providing blanket coverage, i.e., there are no voids in the resulting liquid coat after settling and mixing), within constraints such as those discussed above. This reference would then be scaled to produce the correct amount of thickness corresponding to the intended layer. In one embodiment, the intended layer formed from the liquid ink is 5 microns or less, in other embodiments, it is 4 microns or less, 3 microns or less, 2 microns or less, or 1 micron or less, etc. For example, according to the principles discussed above, if a given reference volume per unit area of ​​"two 10.0 picoliter (pL) drops per square millimeter" produces a 1 micron thick layer with complete coverage, and a 1.60 micron thick layer is desired, the reference volume per unit area may be expanded to "3.2 to 10.0 pL drops per square millimeter." A tentative deposition pattern is then generated (334) for the printable area of ​​the substrate (e.g., with the printing pattern selected to print droplets only in those regions of the substrate where the product is being formed).In one preferred embodiment, the pattern is generated by invoking (i.e., as a software module) a "halftoning" subroutine that receives a definition of the printable area (e.g., height and width) and printhead details (e.g., nozzle pitch, print grid details, etc.) as well as an upscaled density (e.g., "3.2 to 10.0 pL drops per square millimeter") and returns drop pattern information equal to the nozzle firing decisions for the print grid that is superimposed on the printable area of ​​the substrate. In contemplated variations, as above, (a) the drop size can be upscaled instead of or in addition to the number of drops (e.g., the upscaled value could be "two 32.0 pL drops per square millimeter" or "ten 32.0 pL drops per square millimeter") and / or (b) the print grid can be effectively adjusted (e.g., "0.3125 mm" instead of "0.3125 mm"), effectively keeping the existing drop pattern but by changing the print grid density. 2 "two 10.0 pL droplets per square millimeter"); and / or (c) multiple droplet sizes can be used together in a single scan; and / or (d) other techniques can be used. The use of subroutines to calculate firing commands (e.g., select nozzles for firing to spatially distribute droplets) is known, as introduced by U.S. Pat. No. 8,995,022 (already incorporated by reference). In effect, the desired parameters are specified (e.g., "1 square millimeter"). The software then requests a firing map ("ten 32.0 pL droplets per Torr"), with the software returning a firing map that spatially distributes the requested droplets. The pattern generated by such software is effectively represented by numerals 334 in FIG. 3D, although it should be noted that the software may also be integrated with other depicted functions (e.g., droplet size selection and / or alternating stacking, the latter represented by the use of filter 335).

[0031] Continuing with this description, FIG. 3D illustrates a hypothetical pattern of "1's" and "0's" for a hypothetical print grid having five rows and five columns (i.e., as a tentative deposition pattern) for purposes of illustration and explanation. Notwithstanding this simplified example, it should be understood that in practice, the print grid will have thousands of rows and thousands of columns. In the depicted matrix, a "1" represents a decision that the associated nozzle will be "fired" at the corresponding print grid location (node), while a "0" represents a decision that the associated nozzle will not be "fired" at the corresponding print grid location. As indicated by numeral 335, an alternating layer filter (represented as A, B / B, A) is applied to the tentative deposition pattern to separate the nozzle firings into separate scans. This filter, illustrated in this example as a 2×2 matrix, is applied to the digital firing decisions (i.e., the 1's and 0's of the print grid firing pattern) to simply distribute adjacent droplets in each column of the print grid into separate scans. It is noted that many other filters may be applied, and much more complex filters may be used (e.g., referring to adjusted print grid firing decisions 337, note that feature "A" in the lower left corner may pass droplets close to each other, providing suboptimal droplet distribution, and a more robust filter may be used so that sparse droplets are interleaved with even greater precision). Whatever type of interleaving filter or process is used, one or more processors under software control may operate on the interim deposition pattern (334) as stored in memory 336 and apply a digital filter (i.e., filter 335) related to the design to the stored data in order to automatically transform the stored data using matrix mathematics. The results (337) are again stored in memory.Obviously, there are many alternatives for performing this conversion; for example, in another embodiment, the halftoning subroutine could be called a first time, passed an expanded density (e.g., "1.60 drops per square millimeter") to identify the first pass drop deposition pattern, and then called a second time with a position or similar offset (e.g., passed the same expanded density for the second pass). Again, note that the use of digital filters is not required for all embodiments. As shown by the right side of FIG. 3D, the result is then a set of firing instructions for the first scan ("A"), a set of firing instructions for the second scan ("B"), and possibly a set of firing decisions for further scans (i.e., as represented by the presence of ellipses 339). The stored / modified data can then be applied to the printer to control printing and to create physical devices having layers with physical structural properties imparted / dictated by the described techniques, where layer-by-layer deposition as described is applied to help improve droplet mixing and thus promote the formation of more uniform layers. Again, this can greatly aid in improving the fabrication of electronic devices given thin layers.

[0032] Depending on the embodiment, additional measures can also be taken to further enhance layer uniformity. For example, Figure 3E shows an example 341 of a hypothetical per-nozzle error in terms of deposited droplets. That is, unintended variations in droplet size (represented as circles of different sizes, e.g., 345, 347, and 349) and landing position errors (compare the trajectories and landing positions depicted by the arrows for droplet 343 versus droplet 345) can potentially disrupt layer uniformity.

[0033] More specifically, FIG. 3E illustrates hypothetical, repeatable droplet or nozzle firing errors for each of nozzles 1-5 of a print head. Expected drop volumes are represented by circles (e.g., circle 343) centered on the print grid node locations, while expected drop landing locations and / or firing trajectories are represented by arrows emanating from each print grid node location. Nozzle 1 is considered in this illustration to produce drops having a median expected volume per pass (e.g., of 10.04 picoliters or "pL"), with a trajectory that is slightly off-center to the right. Note that this illustration illustrates the use of the drop-by-drop technique introduced with respect to FIG. 3C above, for example, one drop (343) from nozzle 1 is considered to have a downward trajectory to the right as part of "scan A," while the next drop (345) from the same nozzle is considered to have an upward trajectory to the right as part of "scan B" (i.e., the print head / substrate is moved relatively in the opposite direction to the direction of "scan A"). Droplets from nozzle 2 (347) are considered to have a larger expected volume (e.g., 10.47 pL) than nozzle 1, but a larger expected velocity error and a relatively small lateral error (e.g., slightly "left" relative to the droplets from nozzle 1). Droplets from nozzle 3 (349) produce droplets with a smaller expected volume (e.g., 9.03 pL), lower expected velocity, and no expected lateral tilt relative to the droplets from nozzles 1 or 2. FIG. 3E does not show droplets depicted from nozzle 5, representing the fact that our hypothetical nozzle 5 is inoperative and therefore does not produce droplets as expected. Depending on the degree of deviation from the desired droplet parameters, deposition using these droplets may result in extremely thin layer problems.

[0034] To address these unintended variations, in one embodiment, these errors are actually exploited in generating the print drop firing pattern, i.e., expected per nozzle drop volume and / or landing position errors are identified and actually relied upon (e.g., by halftoning subroutines or other software) to generate a distributed drop pattern and / or drop size, and the per nozzle or per waveform errors in drop size and landing position are taken into account in planning the nozzles that will fire, when they will fire, and the nozzle driving waveforms or instructions that will be used, which promotes smooth layer formation and minimizes "wicking" and other problems. These various errors can be used to generate or adjust the print grid firing decisions (interim or final) and / or scan / scan positions, or to pre-warp the firing map and adjust filters to exploit these expected variations.

[0035] 4A-D provide further examples illustrating how such unintended errors / variations may be accommodated and / or reduced.

[0036] FIG. 4A illustrates another possible method for performing droplet-by-drop deposition, generally designated by numeral 401. The method is similar to that described above in connection with FIG. 3E, with similar steps described using similar reference numbers. These techniques can again be embodied as an apparatus or system, and can be embodied in the form of instructions (e.g., software) stored on a non-transitory machine-readable medium. The method first identifies or retrieves a reference volume per unit area (332), e.g., calculated according to the principles discussed above and stored in a machine-accessible digital memory. This volume per unit area can be expressed as a minimum number of droplets of a given size per unit area, or a pattern of droplets, or raw volume, or in some other manner, selected to generate a "blanket layer" of a reference thickness within a minimum scan time or a minimum number of printhead scans (relative to the substrate). This quantity is then scaled (333) or adjusted in light of the desired thickness (e.g., by the ratio of the desired thickness to the nominal thickness of the processed permanent layer). In one hypothetical embodiment, the nominal thickness may be a 1 micron thick layer and the final thickness of the desired layer may be, for example, 3.2 microns. For example, according to the principles discussed above, if a given reference density of "two 10.0 pL drops per square millimeter" produces a 1 micron thick layer with complete coverage, and a 1.60 micron thick layer is desired, the reference density may be scaled to "3.2 drops per square millimeter" or some mathematical equivalent. A tentative deposition pattern is then generated for the entire printable area of ​​the substrate (334). In one preferred embodiment, this pattern is generated by calling a "halftoning" subroutine (i.e., as a software module) that receives a definition of the printable area (e.g., height and width) and printhead details (e.g., nozzle pitch, print grid details, etc.) as well as the scaled volume / area or drop density (e.g., "3.2 drops per square millimeter") and returns drop pattern information equivalent to the nozzle firing decisions for the print grid that is superimposed on the printable area of ​​the substrate.

[0037] As indicated by numeral 335, an interleaved filter can then be applied to the interim deposition pattern to again separate the nozzle firings into separate scans. This filter can be stored in machine-accessible memory 403. The compiled (e.g., measured in situ) nozzle-to-nozzle or nozzle-waveform combination aberrations can also be stored in machine-accessible memory (407) and retrieved and used to adjust and / or create the stored filter (and / or interim deposition pattern), as indicated by numeral 409. The results of this process can then be stored in digital memory 411 (e.g., this can be the same memory or a different memory, as indicated by numerals 403 and 407, and represents a transformation of at least a portion of the data previously stored in the memory). Again, nearly any type of filter can be applied to separate the droplet deposition pattern into multiple scans to "interleave" the droplet depositions. In this embodiment, one or more processors under software control can operate on the interim deposition pattern (334) as stored in memory 411 and apply a digital filter (i.e., filter 335) related to the design to automatically transform the stored data using matrix mathematics, with the result (413) being stored back into memory or used to overwrite the original firing decision. As was true previously, here too there are many alternatives for performing the referenced transformation of the droplet data and the associated sorting of droplet firing responsibilities into multiple scans. This is illustrated by the right side of FIG. 4A by the depiction of a set of firing instructions for a first scan ("A"), a set of firing instructions for a second scan ("B"), and a set of possible firing decisions for further scans (i.e., represented by the presence of ellipses 339).Because the referenced aberrations do not necessarily align "nicely" with the printing grid pitch, in one embodiment, any scan can be offset by a small amount by repositioning the print head in the cross-scan dimension; for example, the print head can be incrementally advanced by a "step offset" for every "A" and "B" (or other scan), or for both, as indicated by optional process blocks 415 and 416. In this regard, one optional design for the printer mechanism includes a stepper motor that moves the print head in the cross-scan dimension by very small amounts, e.g., micron-scale increments. Considering that for many equipment implementations, the minimum printing grid cross-scan pitch will be 20 microns or greater, this "fine-tuning" capability provides greatly increased precision in positioning droplets to provide droplet coalescence with fewer aberrations and the production of more uniform layers.

[0038] Again, the techniques discussed above include (a) receiving droplet data per nozzle or per nozzle-waveform combination, (b) planning droplet positioning and / or droplet selection in a manner that is dependent on the variations represented by the received data to position droplets in a manner that will result in a more uniform layer, and / or (c) determining whether the droplets are to be positioned in a uniform layer. ) It should be noted that this can be implemented in many different ways, including as software controlling one or more processors to adjust printing parameters as necessary to obtain deposited layers (and solidified layers) of liquid ink with improved uniformity using faster printing times.

[0039] FIG. 4B illustrates a hypothetical print pattern (451) similar to that discussed above in connection with FIG. 3E, but planned or replanned in a manner that depends on the variation in drop detail. Note that unlike FIG. 3E, the circles representing deposited drop volumes are now depicted centered at the landing locations (rather than on print grid nodes) and are shown as representing a sparse print drop pattern to better illustrate uniformity. More specifically, FIG. 4B again shows print head 303, print grid 305, and arrows representing drop trajectories (for those drops fired as part of the sparse print pattern). To obtain a relatively uniform ink density that depends on the expected volume of the drops and the expected landing location variation, two drops are used from nozzle 1, two drops from nozzle 2, and five drops from each of nozzles 3 and 4. The use of only two droplets from nozzles 1 and 2 correlates with the fact that these nozzles are measured to produce larger droplets than nozzles 3 and 4, which, conversely, use droplets at a greater firing frequency, respectively. The two droplets from nozzle 1 are interleaved in different scans in this example (e.g., according to their similar trajectories), while the two droplets from nozzle 2 are sorted in this example into the same scan (scan "B") to better distribute these droplets. To account for the inability to produce droplets from nozzle 5, either additional scans (with printhead offset or step in the cross-scan dimension) or relatively frequent use of droplets from nozzles 4 and 6 (the latter not featured in the figure) can be relied upon to impart uniform ink density. As mentioned above, interleaving is used to deposit droplets for improved coalescence properties, but nozzle variations are also considered in embodiments, leading to better layer uniformity. Again, this diagrammatic depiction is provided as an example for purposes of explanation, and the design of a suitable filter (and conditioning process) for layer-by-layer stacking that takes into account the aforementioned variations will be within the capabilities of one of ordinary skill in the art and will be highly implementation dependent.

[0040] Figure 4C provides an example 471 similar to that presented in Figure 4B, but illustrating intentional variation in droplet size to promote layer uniformity and accurate deposition of a uniform liquid coat; i.e., in comparing the depicted droplet spread (and associated droplet volume) from nozzle 4 (designated by numeral 473) to the similar depiction in Figure 4B, it should be noted that the droplets in Figure 4C are larger. In one embodiment, print planning or halftoning software, coded to take into account nozzle-to-nozzle variation, detects the need for deposition of a larger volume from nozzle 4 and, in response, varies the droplet size or the firing waveform used to create the droplets. In various embodiments, this additional volume can be achieved through (a) multiple scan passes, (b) use of a “customized” nozzle drive waveform (e.g., programming or varying the drive waveform for nozzle 4 to enhance volume generation or vary the drop trajectory), (c) use of supported “enlarged” drop sizes (e.g., using a printhead that enables the generation of “dual” drop volumes via application of dual drive pulses), or (d) other means.

[0041] FIG. 4D illustrates a technique in which multiple scan passes are applied with the printhead incrementally offset between passes. Thus, in a first pass "A," during which the printhead advances downward relative to the drawing page, nozzles 1-5 are controlled to produce droplets as represented by the downward arrows. The print planning software then offsets the printhead position by moving the printhead two positions to the right of the drawing page such that nozzle 1 matches the position that nozzle 3 had in the first pass. The printhead is then moved in the opposite direction in a second pass, Droplets are deposited as represented by the upward arrows (i.e., with associated volume and / or trajectory per nozzle). Firing decisions and print head offsets are each planned to interleave the droplets as appropriate to provide a uniform liquid coat (and processed layer). For example, nozzle 2 as depicted in this figure can be used in a bidirectional scan to deposit droplets at two different locations (483), and conversely, nozzle 3 (and / or other nozzles) can be used to deposit overlapping droplets as desired (e.g., per numeral 485). In one embodiment, print head offsets, droplet size and trajectory variations occurring through the techniques discussed herein, and interleave can be applied in three or more scans to provide a high degree of deposition uniformity.

[0042] FIG. 5A depicts yet another embodiment of a technique (501) that can be used to improve layer uniformity and printing speed. More specifically, the method begins with a specification of a layer thickness, as represented by numeral 503. A method, apparatus, or system embodying these techniques then selects fabrication process parameters for producing the layer of interest (505). The process parameters are then applied to print a liquid ink onto a substrate (506), which will then be processed to cure or otherwise convert the liquid ink into a permanent structure (507). In one embodiment, the liquid is deposited as discrete droplets that settle and mix together as previously described to form a uniform liquid coat carrying the organic material. The liquid coat is then cured or otherwise baked or dried (e.g., via UV exposure) to convert the liquid into a structural element of a light-emitting device. It should be noted that in an assembly line type fabrication process, where the same process would be used for many substrates, production testing may be performed first on test substrates to experimentally measure (509) how the ink droplets spread, thus providing feedback (511) that may be used to define a suitable volume per unit area (513) or equivalent measure (e.g., density of droplets of a given size per unit area). As previously mentioned, in one embodiment, the method may identify a minimum volume per unit area (e.g., pattern density of "10.0 pL" droplets) that will produce a "blanket" liquid coverage of the area of ​​interest, taking into account the expected droplet spread, as well as other process parameters that still ensure that the given process parameters are capable of producing the desired layer (e.g., within relevant specification parameters). This "criteria" may then be scaled (515) to produce the desired layer thickness. In one embodiment, these techniques are used to fabricate layers that are 5.0 microns or less thick, 4.0 microns or less thick, 3.0 microns or less thick, 2.0 microns or less thick, or 1.0 micron or less thick.

[0043] In one optional embodiment, collectively represented by the numeral 517, the printing infrastructure provides the ability to produce multiple discrete drop sizes. In one version, multiple different print heads can be used, each producing a separate discrete drop size. For example, one print head may produce drops of "about" 10.0 pL per drop, another may produce 7.0 pL per drop, another may produce 4.5 pL per drop, etc. In such an embodiment, each one of these is separately calibrated for drop spreading characteristics and ability to produce a layer of the desired thickness, and one of these is selected for use (assuming that at least one of the drop sizes 1-n is capable of producing the desired thickness). For example, as depicted, a first drop size ("drop size 1") 518 may be configured to have (a) a blanket coating (523) using a drop density available for a particular printing process, (b) a layer thickness less than or equal to a thickness (T) that is regulated by specifications (e.g., fn{MV / A(minimum volume / unit area)}≦T, as shown by numeral 525 in the figure), and (c) a layer thickness that is regulated by specifications using one or more scans, but not including the thickness (T) of the blanket coating (523), as shown by numeral 527 in the figure. As shown by (529), the thickness (T) of the droplets can be evaluated for its ability to produce a thickness (T) within a maximum specified deposition time t (e.g., fn{j*MV / A}≦t). Note that for number 525, buffers can be artificially injected, i.e., ideally, a droplet size is not selected that can only produce the desired thickness “just exactly” in the minimum volume per unit area (MV / A) if there are alternatives available, and ideally, the process details will provide some buffer above the referenced minimum (required to obtain blanket coverage) to accommodate nozzle variations, unexpected errors, etc. (hence the reference of “buffer” by process box 525 (i.e., fn{MV / A}≦T(buffer)). “Droplet size 2” (519) is then evaluated according to these criteria, etc., through “Droplet size n” (529). After evaluation of multiple droplet sizes, multiple alternative “acceptable” droplet sizes are selected. If each of these alternatives can be used according to these tests, in one optional embodiment, the one of these alternatives with the largest drop size (529) is selected for use (e.g., such that would minimize deposition time or otherwise enhance deposition rate). Other selection criteria can also or instead be used. As a non-limiting example, the drop size that results in the smoothest layer can be selected instead, and such criteria can result in the smallest drop size of the available alternatives being selected. Obviously, other alternatives are possible (e.g., the fastest deposition process). In another variation, a single print head can be used to generate multiple drop sizes (e.g., using different alternative waveforms applied to a given nozzle).

[0044] Whatever methodology is employed, firing decisions are then assigned to the print grid according to reference numeral 531 based on the use of the preferred selected droplet size. If desired, as previously referenced, nozzle-waveform combination aberrations (533) of expected droplet volume or landing location (e.g., trajectory, velocity, nozzle bow, and / or other factors) can be taken into account in planning the firing decisions (e.g., based on the mean or standard deviation and other statistical parameters such as statistical variance) as discussed above. As implied by this description, in one embodiment, one nozzle-waveform combination may be used in place of another combination, while the associated statistical variance (or statistical parameter) is small, since the mean (i.e., expected volume or expected landing location) may not be optimal. In an embodiment in which a statistical ensemble is constructed based on many measurements per nozzle (or nozzle-waveform combination), i.e., per droplet parameter, to reduce the effect of measurement errors, many different statistical parameters can be calculated and applied in this manner. As indicated by reference numeral 531, the printing pattern and / or filter and / or pattern selection can then be adjusted based on the indicated variation in nozzle-to-nozzle (or nozzle-waveform combination) parameters.

[0045] It should be noted that FIG. 5A shows two further options that the designer may choose to implement, depending on the application. First, the system may be designed to provide the local (i.e., within-substrate) variation of the layer parameters (535) and associated topography-specific minimum volume / unit area required to obtain blanket coverage (537). In this regard, it will be noted that the droplet spreading characteristics may vary depending on many factors, including substrate surface chemistry, substrate surface energy, and various factors that contribute to these. This dependency leads to the consequence that for the same fabrication process, the same ink may spread in a first portion of the same substrate differently than in a second portion of the substrate, thus generating undesirable thickness variations (e.g., in situations where the process is being used to deposit large-scale films such as encapsulation layers). In relation to the optional technique represented by numerals 535 and 537, this variation is specifically measured on a test panel, during calibration or in advance, and then programmed into the system in a manner that the parameters vary depending on the portion of the substrate being printed. For example, the volume / area reference associates an ink with a specific "minimum" layer thickness (i.e., associated with a blanket coating), which , in some embodiments, may vary depending on the area of ​​the substrate. Perhaps stated differently, in areas where the ink spreads further, a different (e.g., larger) volume / area or drop density may be used, etc. Note that other embodiments may simply use one set of parameters per panel or substrate. For example, if a layer is to be fabricated in individual "pixel wells" on a substrate, it may be desirable to treat each pixel well that will receive the layer in exactly the same way. Even for embodiments where relatively large layers are deposited (e.g., encapsulation, smoothing, or barrier layers), many applications may be desirable to model deposition parameters as being consistent throughout the substrate, with parameters pre-calibrated and stored for application in an assembly line process. However, if the parameters are specific to the substrate area, these considerations may optionally be taken into account when determining which of multiple alternative drop sizes should be selected (e.g., as additional tests or requirements beyond criteria 523, 525, and 527). According to numeral 539, whichever of the described tests is used to determine the printing process details, the printing is scheduled based on these parameters and the result is a completed electronic device with improved layer uniformity that is fabricated using faster print times than would otherwise be possible.

[0046] FIG. 5B provides a graph 551 that is used to explain various issues associated with scaling depth of a layer during the process of depositing liquid ink. FIG. 5B shows a thickness profile of a processed film obtained using a stylus profilometer, which is useful in connection with the calibration process and scaling a reference ink volume or drop density per unit area as discussed above. After generating actual test layers of material or simulating these layers including ink drop spreading properties, the ink volume can be correlated to different stages of layer thickness. For example, a first curve 553 representing a layer of 1.0 micron thickness is associated with an ink volume per unit area (for the print grid maximum drop population) that corresponds to 8% of the maximum ink volume per scan per unit area, assuming drop firing at all print grid nodes. Note that there is no gap in the center of the layer represented by curve 853, which is considered to be continuous, i.e., having a substantially uniform thickness. For a subsequent fabrication process, if a layer thickness of 1.0 micron is assigned by the received layout data for the deposited layer, this quantity of 1.0 micron would be used along with the nominal volume / area associated with the ink (i.e., dependent on its spreading properties) to select a drop pattern (e.g., a "halftone" pattern) that will distribute the drops to the various print grid points to achieve a uniform deposited layer (after drop spreading). Similarly, a second curve 855 is considered to represent a uniform 2.0 micron thick layer, corresponding to a 16% print grid fill. For example, based on a nominal volume / area of ​​8%, a 16% ink volume for a particular substrate area may be generated to result in a 2.0 micron thick layer. The mapping between layer thickness values ​​and drop pattern ("halftone") selections can also be estimated using this process. As an example, if the layout data called for a 1.5 micron thick encapsulation layer, a scaling selected to correspond to a point approximately between these two values ​​(12%) may be applied (e.g., halfway between 8% and 16%).Other illustrated curves 857, 859, 861, and 863, corresponding to layers of 3.0, 4.0, 5.0, and 6.0 microns thickness, respectively, are associated with 24%, 32%, 40%, and 50% fill, respectively. By specifically matching different "enlarged" volumes / areas or droplet densities to individual layer thicknesses and associating enlarged volumes to deliver ink quantities to target areas, designers can customize ink deposition to any desired thickness in a manner that will lead to predictable results. This provides a high degree of control over the thickness of material deposited via liquid ink. As these examples imply, multiple ink volumes / areas (that provide complete coverage) can be measured and used to interpolate fill for a desired thickness, such as the 1.5 micron thick example provided above.

[0047] In many applications, it is also desirable to provide sharp linear edges at the boundary regions. For example, as represented in FIG. 5B by numeral 565, if the ink density is maintained on a consistent basis right up to the layer periphery, the surface tension of the ink may cause the liquid ink to form "horns" or "bat ears", which may subsequently solidify to form undesirable thickness variations in the finished layer. Discussed below are techniques for varying the ink density at the boundary regions to mitigate this and related effects, for example by adjusting the volume / area or drop pattern (or related parameters such as drop size) in portions of the layer adjacent to the layer edge. These techniques will be reviewed later in this discussion.

[0048] FIG. 6A illustrates still further optional techniques that can be employed to provide better layer uniformity and printing speed. These further techniques are generally referred to by the numeral 601. As previously described, a layer thickness value for at least one target region is first read (603). Such readout can be, for example, from a machine-accessible digital memory in response to operator input or preprogramming associated with a digital (e.g., computer-controlled) assembly-line type fabrication process. For example, a system in the form of one or more processors defines layer parameters as previously described, again utilizing techniques that improve layer uniformity and printing speed (i.e., via process 405). For example, as previously described, the system calculates a minimum volume / area (413) for a given ink / substrate process configuration (i.e., a minimum areal density that provides a "blanket" ink coverage), and then scales this quantity (415) to produce the desired layer (i.e., according to the ratio of the desired layer thickness to the thickness produced through the use of the minimum). The right side of FIG. 6A illustrates several optional techniques for improving layer uniformity and printing speed that can be used in addition to or instead of those techniques already discussed (e.g., in any desired combination or permutation as relevant to a particular implementation). As indicated by numeral 605, the first such technique involves varying the in-scan pitch or separation between the horizontal axes of the print grid (see reference numeral 307 provided above in connection with FIG. 3A). See also the circuitry and associated discussion provided in FIG. 14A-B of U.S. Pat. No. 9,010,899 and FIG. 5D of the aforementioned U.S. Patent Publication No. 20150373305 (USSN14 / 840343), which are incorporated by reference, particularly in connection with the discussion of clocks and triggers.By varying the frequency of the clock (or trigger signal) used to time the firing of a given waveform, one can effectively vary the drop density and mixing properties of a given drop size and / or print head, thereby varying the degree to which a given drop size and / or print head is suited to producing a uniform film within a prescribed print time. In one embodiment, these techniques are applied in conjunction with the filter and filter / pattern adjustment techniques discussed above (see discussion of FIG. 4A ) via trigger rate definition on a per-scan basis to layer drops to provide a mix of drops that mitigate undesired wicking and better position drops for uniform layer processing. That is, in one embodiment, a first print grid spacing can be used for a first scan, and a second, different spacing (along the vertical or horizontal print grid axis or both) can be used for a second scan. Each scan can deposit similar drop sizes (and landing locations) from a common set of nozzles, or distinct drop sizes (or locations) from distinct sets of nozzles. To cite another example, by further combining these techniques with variable step printhead positioning (see, e.g., numerals 416 and 417 of FIG. 4A and supporting discussion), the combination of these various techniques can precisely position droplets of specific expected volumes and trajectories to provide a precisely calculated volumetric density distribution covering a given unit area by carefully distributing both the size and relative position of droplets from different nozzles or nozzle-waveform combinations (e.g., numeral 607 by rotating the printhead). , shows another optional variant in which the cross-scan print grid pitch is further adjusted. That is, as mentioned above, in one embodiment, the printer can advantageously include a rotation mechanism (i.e., some type of power drive) that rotates the print head, thereby effectively varying the cross-scan pitch according to the sine or cosine of the relative angle between the print head and the cross-scan axis (e.g., numeral 307 in FIG. 3A). Such a technique varies the separation between the vertical axes (306) from FIG. 3A, and can optionally be used in conjunction with or in place of the timing (Δy) variation discussed immediately above. Numerals 609 and 611 relate to an option in which the effective drop size is varied, for example, by using a print head adapted to use multiple drive waveforms to generate drop sizes with specifically desired expected volumetric properties (and variance between expected volumetric properties), or by changing one print head (e.g., one that produces drops having an average volume of 10.0 pL per nozzle or some other value) to another print head (e.g., one that produces drops having an average volume of 7.0 pL per nozzle). In one embodiment, the print head can be selected to effectively deposit multiple similarly sized droplets at the same time (e.g., two droplets using a dual drive pulse), or to be one with a fixed number of droplet sizes (e.g., 1x, 2x, 3x, 4x) rather than being programmable and customizable. According to numeral 613, it is also possible to vary the ink used to achieve a particular deposition. For example, if it is determined that the prescribed layer thickness (T) is too large for a given droplet size (e.g., the droplets for a given printer and printing process do not spread sufficiently upon contact, as illustrated for Figs. 2B and 2C), it is possible to select a different ink or vary the composition of the ink (including but not limited to introducing a spreading agent), or modify the substrate to possess a different surface energy.In yet another embodiment, a printhead and / or printing process is used that produces ink droplets having an expected volume average (615) of less than or equal to 7.0 pL on average, or that produces a film thickness (617) that is less than 4.00 microns. In yet another embodiment, ink density, halftone patterns, or other ink or process parameters are applied as a form of edge or border treatment 619 to produce sharp edges. These techniques, also discussed in the aforementioned U.S. Pat. No. 8,995,022 (incorporated by reference), provide powerful techniques for producing layers of consistent thickness, with customized or tailored tapering, all the way to the intended periphery of the desired layer. Finally, as represented by numeral 621, other techniques can also be used to vary or adjust the drop density and / or ink volume applied to a given unit area, and / or the way the deposited droplets mix together, to prevent droplet wicking, provide smooth layer formation with improved printing times, and otherwise produce very thin film layers, even though the droplets themselves may have an inherent diameter larger than the intended film thickness.

[0049] FIG. 6B is used to discuss an embodiment (631) in which the droplet size is effectively varied between multiple discrete alternatives. As discussed above, this type of variation, if supported by the characteristics of a particular machine (system) implementation, in one embodiment, the ink volume / area and droplet mixing can be adjusted to achieve layer uniformity and / or optimize print times within specifications and other improvements referenced above. Referring to the first example, if it is determined that the specifications for a particular assembly line type process call for an organic encapsulation layer of 1.0 micron ±10%, and if a particular machine produces droplets having a diameter of 30 microns (which do not spread sufficiently and thus produce layers that are too thick, see, e.g., the discussion provided above in connection with FIGS. 2B and 2C), in one embodiment, the print head is modified to one that produces a smaller droplet diameter (e.g., 12 microns), or the nozzle drive waveforms per nozzle are modified and / or programmed to support firing of the smaller droplet size. These techniques may also be used in other applications. Note that the above types of nozzle errors (e.g., errors in trajectory, volume, or velocity per nozzle as introduced in connection with FIGS. 3E and 4B and supporting discussion) can be used to reduce and / or vary. If desired, a printhead can be selected that provides discrete drop sizes that are predefined or otherwise programmed into the printhead driver circuitry in order to provide drop properties that can be selected or varied in situ. Thus, if a given drop size does not work to meet the required layer characteristics (e.g., does not produce a layer as illustrated by the discussion for FIG. 2D above), in one embodiment, the drop properties are varied until such a layer can be obtained. If multiple "all work" drop sizes are available, the techniques identified above can be used to arbitrate between these acceptable alternatives according to any desired criteria, for example, selecting the largest of the discrete drop sizes or the process with the smallest print grid (e.g., to minimize print time) or producing the smoothest layer (e.g., potentially the smallest of the discrete drop sizes). In variations, multiple sizes can be used together (e.g., a first scan in which small droplets are deposited in a dispersed pattern followed by larger droplets for interleaving, or vice versa). Note that such techniques do not imply that multiple drop sizes cannot be used in one or more scans (e.g., drop sizes can be combined in any desired permutation, e.g., as part of a variable drop volume halftone, or edge contour management technique).

[0050] As shown in FIG. 6B, the depicted process assumes a reference printing grid definition (633), a specification-defined layer thickness (635), ink spreading characteristics (637), and associated ink definition information (which in this embodiment includes "contact angle", drop thickness or height, radius, or other properties, per numeral 639). The method then evaluates each alternative drop size in succession, per numeral 641. [Note that "contact angle" refers to the angle made between the drop and the substrate at its perimeter, e.g., the average contact angle of a spherical drop may be modeled as this section at the perimeter of a flattened section of a sphere. See numerals 298 and 299 in FIGS. 2A and 2B, respectively. ]Of course, in accordance with the above discussion, depending on the embodiment, each of these ink / droplet and / or printing process parameters may be varied, but for purposes of this discussion, it should be assumed that the multiple drop sizes (however generated) are evaluated individually for purposes of selecting one of the multiple drop sizes for use in the printing process to promote layer uniformity and / or print time, all within specification constraints for layer processing. In order to evaluate a given drop size, the parameters represented by numerals 633, 635, 637, and 639 are assumed to be fixed for the time being in order to evaluate whether a given printing process size is suitable for use in a given printing process, and whether a different one of the alternative drop size(s) should be selected.

[0051] Thus, as represented by block 641, these parameters are assumed to be fixed and each possible discrete drop size is evaluated. The method iteratively takes each discrete drop size, starting from the smallest, and determines whether that drop size is suitable for producing a layer of the desired thickness (635). For each drop size evaluated, the method calculates the minimum volume / area (643) for the drop size in question, i.e., for a continuous film with no gaps / voids, taking into account the printing grid definition (633) and diffusion properties (637) that would produce a "blanket" coating of ink. The first consideration is whether the drop will produce a layer that is too thick, taking into account diffusion (see, e.g., FIG. 2B and supporting discussion), i.e., the thickness after layer processing / curing will be less than the desired (μ 0 -Δ 0 ) is exceeded. For example, μ 0 represents the specified maximum value of the layer thickness, and Δ 0 represents the desired buffer as introduced above. If the droplet size is not capable of producing a "thin enough" layer according to numeral 645, then the droplet size is adjusted to the desired buffer size, as shown in FIG. 6B in this embodiment. As represented by the "Next" reference in , it is deemed unsuitable for use. In other embodiments, instead of discarding a drop size, other techniques are evaluated or used to determine whether the given drop size can be used at all, e.g., whether print grid adjustments, ink property modifications, etc. can be varied so that the given drop size can still be used to produce a "thin enough" layer. If the given drop size is suitable, the method saves the drop size as potentially usable and continues to evaluate the next drop size in the same manner.

[0052] Once all alternative discrete drop sizes have been evaluated in this manner and a set of alternatives established, the method then proceeds to select one of these alternatives for use in the fabrication process (647), applying other of the considerations discussed above. For example, according to function block 649, the method can then determine whether a scan time constraint would be violated if a particular alternative were used, taking into account the maximum drop density (e.g., assuming a fully populated print grid for the target area, i.e., all possible print grid nodes represented a drop firing), and if such constraints would be violated, the particular drop may be eliminated from use (assuming other alternatives were still available). Again, in other embodiments, instead of discarding a drop size (or if other alternatives were not available), other techniques can be evaluated to determine whether the drop size could be used at all, e.g., if print grid adjustments, ink property modifications, etc. were varied, it would produce an acceptable fabrication process that met the specifications without violating the constraints. In one embodiment, these criteria can be expressed as a maximum number of scans and / or a maximum time, according to numeral 651. Also, as represented by numeral 649, the method can use smoothness criteria to evaluate droplets of different sizes, for example, the system obtains a measure of layer smoothness (e.g., RMS, as measured and entered by an operator or automatically measured on a test substrate) and chooses among the remaining droplet candidates according to the one that produces the smoothest layer in the fewest number of scans. In one contemplated implementation, for any given layer thickness (e.g., 0-30 microns), a range of nozzle droplet sizes (and associated drive signals) is provided in advance sufficiently that multiple droplet sizes (653) will be suitable for use according to the evaluation procedure introduced above, and smoothness / timing constraints will be used to arbitrate between these "acceptable candidates."In such an embodiment, the system can access a single file from memory (e.g., a file indexing all measured properties, such as thickness, surface roughness, contact angle, etc., according to droplet size, if desired, for the printing process in question, on a basis that varies with the substrate Cartesian coordinates) and can run automated processes according to the software to select the droplet size that will be used. In systems in which the printhead is changed to vary the droplet size, the software can engage a different printhead (e.g., on the same or a different traveler), or can alert the operator that the printhead needs to be changed manually. Other alternatives are also possible. Many other criteria can also be used to select between droplet alternatives, either in addition to, or instead of, those criteria sets discussed above.

[0053] As shown by the remaining figures, the depicted method selects one of the alternative drop sizes for use in the printing process and then calculates the nozzle firing data as discussed above. For example, for the selected drop size, the method selects the reference volume / area or drop density (655) to be the minimum (for the particular drop size) to obtain "complete" ink coverage of the intended target area. The drop distribution or density can then be determined according to a formula that assumes the ink spreading characteristics and the printing grid in-scan and cross-scan pitch dimensions. Again, in the figures, the value Δ 0 represents optional buffering, and the value μ 0 is the nominal volume / area or droplet density (and corresponding nominal thickness) Note that, per numeral 659, the criterion is then scaled as necessary to obtain the proper layer thickness through a drop pattern that distributes the drops in a manner that results in a uniform layer of ink of the desired thickness after drop mixing and conversion of the layer into a permanent structure, i.e., without voids / gaps and with minimized height variation. As indicated by numeral 661, in one embodiment, this function is established by calling a subroutine and passing the drop size, printing grid pitch, and drop spread parameters to that subroutine, which then returns a large area firing map (663) in which the drop firings have been distributed within both each individual scan (e.g., to control wicking) as well as the cross scan in a manner that promotes proper mixing, per the techniques discussed above. A second subroutine can then be called (665) to test the firing map to ensure ink density uniformity to detect aberrations on a unit area basis, depending on the digital memory contents representing the volume and / or landing position average and spread per nozzle (666). If local ink volume varies (e.g., within any target area) above a predefined threshold error amount, or if a processor-applied modeling algorithm predicts that undesired wicking or thickness defects will occur beyond a threshold error amount (i.e., dependent on expected average and spread values ​​for the various nozzles used to print), an exception process (667) can be triggered and used to make corrective firing decisions. For example, if it is determined that a particular row of drops will be off-center, as determined from expected per-nozzle stored data, a different drive waveform can be tested, or a different nozzle can be substituted for use of the first nozzle (e.g., in a subsequent scan), or the same nozzle can be used (albeit with a printhead position offset or "step"), etc. Many variations using the various capabilities presented in this disclosure can be applied by those skilled in the art without departing from the spirit of the techniques discussed herein.Once the launch data described by the launch map has been adjusted / corrected as appropriate, the launch map is stored and / or sent for printing, as indicated by process box 668.

[0054] 6C is a block diagram of a method 671 for defining multiple waveforms for any given print nozzle. Note that in one embodiment (as described in the aforementioned patents and patent publications, which are incorporated by reference), each nozzle has a specifically shaped waveform that is predefined or selected under associated assumptions regarding drop volume and landing location. See, for example, but not limited to, the discussion in U.S. Patent Publication No. 20150099059 (e.g., Figures 15A-D), the similar discussion in U.S. Patent No. 9,010,899 regarding Figures 14A-C, and the discussion in U.S. Patent Publication No. 20150298153 regarding Figure 5D, which are incorporated by reference. Each of these references describes mechanisms for providing waveforms to printhead nozzles on an individualized basis, for example, for printheads with thousands of nozzles. Per numerals 673 and 675, the method selects a specific waveform for each nozzle, then measures the volume, landing location (e.g., trajectory), and other parameters per nozzle-waveform combination, and compiles a statistical population of many independent measurements to develop a statistical confidence in the average and spread parameters. Based on this information, the method can then select a set of "n" waveforms per nozzle that can be expected to produce distinct droplet volumes and distinct landing locations (677). While any criteria desired can be used to identify this set, in one embodiment, the waveforms are selected to provide evenly distributed values ​​across the range of droplet volumes and distinct landing locations (e.g., within ±10% of the expected value). For example, if two waveforms produce nearly ideal expected volumes and landing locations (albeit nearly identical), the method can discard one of them and instead choose the second waveform that produced 110% of the volume and / or droplets, or that represents an intentional expected location offset from the desired landing location, i.e., provides variability. This then allows the droplets (from the same or different nozzles) to mix and match for any given intended deposition area. This provides a more efficient set of options for combining, for example, a 9.90 pL expected drop volume from a first nozzle / waveform with a 10.10 pL expected drop volume from a second nozzle / waveform to obtain an expected total volume of 20.00 pL. See, for example, U.S. Pat. Nos. 9,010,899 and 9,352,561, referenced above. Finally, once a set of values ​​has been selected, the method then programs (679) the drive circuitry for each nozzle with the corresponding waveform such that the programmed waveform will be used to drive the corresponding nozzle at the appropriate "trigger time." As described in the patents and patent publications incorporated by reference, in one embodiment, up to 16 different waveforms can be selected and pre-programmed for each nozzle, and the selection of a particular one of these alternatives is accomplished by a waveform selection value that is "written" into the digital circuitry for the particular nozzle. This is not required for all embodiments.

[0055] Figure 6D illustrates how a "customized" waveform generated by a digital-to-analog converter (DAC) for a given nozzle may be predefined as a series of discrete signal levels, for example defined by digital data. Number 681 in Figure 6D identifies a graph of waveform 683, having discrete signal levels 685, 687, 689, 691, 693, 695, and 697. In one embodiment, each nozzle driver may include circuitry that receives multiple waveforms (e.g., up to 16 or another number), each waveform defined as a series of signal levels of variable voltage and duration. Each signal level in this series may be represented individually as a multi-bit voltage and multi-bit duration. That is, in such an embodiment, the pulse width can be effectively varied by defining different durations for one or more signal levels, and the drive voltage can be waveform-shaped in a manner selected to provide subtle drop size, velocity, or trajectory (landing position) variations, with the drop volume measured to provide specific volumetric gradations, such as in units of 0.01 pL. Wave shaping thus provides the ability to adjust the drop volume and flight parameters to more closely approximate ideal values. These waveform shaping techniques also facilitate strategies to reduce or eliminate visible defects, for example, in one optional embodiment, a single assigned nozzle drive waveform is pre-tuned for each nozzle so that all nozzles provide a uniform drop volume (e.g., as close to 10.00 pL as possible). As described, alternative predetermined waveforms are optionally made available to each nozzle, with a dynamic calibration (or another process) used to select (e.g., program) "one" of the alternative predetermined waveforms to be applied for a short period of time. Other possibilities exist.

[0056] The effect of different drive waveforms and the resulting droplet volume can be measured in advance. In one embodiment, up to 16 different drive waveforms per nozzle can be stored in a dedicated nozzle-specific 1k static random access memory (SRAM) for efficient later use in providing discrete volume variations as selected by software (see previously referenced and incorporated by reference applications, publications, and patents). With the different drive waveforms at hand, each nozzle is then instructed on a drop-by-drop basis as to which waveform to apply via programming of data that achieves the specific drive waveform.

[0057] It should be noted that in alternative embodiments, the drive waveform can be fixed, with one of several discrete waveforms being applied to each nozzle as described above. For example, as previously discussed, one embodiment employs a printhead in which fixed drop volumes can be multiples of one another, e.g., a printhead is used that can feature multiple drop sizes (e.g., 1x, 2x, 3x, 4x, etc.), with different waveforms or programmed values ​​being used to distinguish each selection. Thus, in such a system, a range of drop sizes, For example, 1.0 pL, 2.0 pL, 4.0 pL, etc. can be selected for each nozzle as required. Note that there is no inherent requirement that the droplet volumes relate as integer multiples, and in other embodiments, the volumes can be arbitrary. In yet another contemplated embodiment, a single waveform is applied to all nozzles for a given pass (scan), but this waveform is scaled between scans, and such action can effectively vary (e.g., scale) the droplet details for groups of nozzles on the fly (e.g., the waveform can be reset for groups of nozzles per scan). Again, whatever technique is used, the control software advantageously reads per-drop (per nozzle, per waveform) statistics and uses these to plan droplet combinations and distributions and place individual droplets relative to one another in a manner that produces better layer uniformity with a high degree of predictability.

[0058] FIG. 7A shows one contemplated multi-chamber processing apparatus 701 that can be used to apply the techniques disclosed herein. Generally speaking, the depicted apparatus 701 includes several general modules or subsystems, including a transfer module 703, a printing module 705, and a processing module 705. Each module maintains an atmospheric environment that is controlled to be different from the ambient air and to exclude unwanted items or oxygen or moisture, and printing can be performed, for example, by the printing module 705 in a first controlled atmosphere, and other processing, for example, another deposition process such as an inorganic encapsulation layer deposition or a curing process (for the printed material, for example), can be performed in a second controlled atmosphere (note that these atmospheres can optionally be the same, optionally nitrogen or other inert gas atmosphere). The apparatus 701 uses one or more mechanical handlers to move substrates between modules without exposing the substrates to uncontrolled atmospheres. Within any given module, other substrate handling systems and / or specific devices and control systems can be used that are adapted to the processing performed in that module.

[0059] Various embodiments of the transfer module 703 can include an input load lock 709 (i.e., a chamber that provides buffering between different environments while maintaining a controlled atmosphere), a transfer chamber 711 (also having a handler for transporting the substrate), and an atmospheric buffer chamber 713. Other substrate handling mechanisms can be used within the print module 705, such as a floating table for stable support of the substrate during the printing process. In addition, an xyz motion system, such as a split axis or gantry motion system, can be used for precise positioning of at least one print head relative to the substrate, as well as providing a y-axis transport system for transport of the substrate through the print module 705. It is also possible to use multiple inks for printing within the print chamber, for example, using separate print head assemblies, so that, for example, two different types of deposition processes can be performed within the print module in a controlled atmosphere. The print module 705 can include a gas enclosure 715 that houses the inkjet printing system, along with means for introducing an inert atmosphere (e.g., nitrogen) and otherwise controlling the atmosphere (e.g., temperature and pressure), gas components, and the presence of particulate matter for environmental conditioning.

[0060] Various embodiments of the processing module 707 can include, for example, a transfer chamber 716, which also has a handler for transporting substrates. In addition, the processing module can also include an output load lock 717, a nitrogen stack buffer 719, and a curing chamber 721. In some applications, the curing chamber can be used to cure, bake, or dry the monomer film into a uniform polymer film. For example, two specifically contemplated processes include a heating / baking process and an ultraviolet radiation curing process.

[0061] In one application, the apparatus 701 is adapted for mass production of liquid crystal display screens or OLED display screens, for example, fabricating an array of (for example) eight screens at a time on a single large substrate. These screens can be used in televisions and as display screens for other forms of electronic devices. In a second application, the apparatus can also be used in the mass production of solar panels in the same manner. In yet another embodiment, the apparatus can be used to produce other types of light emitting devices or other electronic products.

[0062] The printing module 705 can be advantageously used in such applications to deposit organic light-emitting layers or encapsulation layers that help protect sensitive elements of an OLED display device. For example, the depicted apparatus 701 can be loaded with a substrate and controlled to move the substrate back and forth between various chambers in a manner that is not interrupted by exposure to uncontrolled atmosphere during the encapsulation process. The substrate can be loaded via an input load lock 709. A handler positioned in the transfer module 703 can move the substrate from the input load lock 709 to the printing module 705, and following completion of the printing process, the substrate can be moved to the processing module 707 for curing. By repeated deposition of subsequent layers, each of controlled thickness, a total encapsulation can be built up to suit any desired application. Again, it is noted that the techniques described above are not limited to encapsulation processes, and many different types of tools can be used. For example, the configuration of apparatus 701 can be varied to place the various modules 703, 705, and 707 in a different juxtaposition, and additional, fewer, or different modules can also be used. Such an embodiment is well suited for assembly line type manufacturing, where, because of the large number of nozzles present, the aforementioned nozzle / series of nozzle waveform calibrations can be re-measured at intervals (or otherwise intermittently during run time) while the previous substrate is removed from the print chamber and a new substrate is introduced.

[0063] While Figure 7A provides one example of a set of linked chambers or processing components, clearly many other possibilities exist. The techniques introduced above can be used to control processing processes performed with the device depicted in Figure 7A, or indeed with any other type of deposition equipment.

[0064] FIG. 7B is an illustrative diagram depicting some functional components within a printer (e.g., within a print module) and depicting a top view of a substrate as it may appear during a deposition process. The print chamber is generally designated by reference numeral 725, the substrate to be printed is generally designated by numeral 729, and the support table used to transport the substrate is generally designated by numeral 727. Generally speaking, any xy coordinate of the substrate is reached by a combination of movements including y-dimension movement (optionally x-dimension movement) of the substrate by the support table (e.g., using a floating support), and generally using a "slow axis" x-dimension movement of one or more printheads 735 along a traveler 733, as represented by arrow 737. The floating table and substrate handling infrastructure are used to move the substrate and advantageously provide tilt rejection control along one or more "fast axes" (731) as needed. The printhead is considered to have multiple nozzles 739, each separately controlled by a firing pattern derived from a firing map (e.g., to achieve printing of columns corresponding to printer grid points). When relative motion between one or more printheads and the substrate is provided in the direction of the fast axis (i.e., the y-axis), the printing typically represents swaths that follow the printer grid points of the individual rows. The printhead can also be advantageously adjusted to vary the effective nozzle spacing (e.g., by rotation of one or more printheads, per numeral 741). Multiple such nozzles can be controlled separately, as desired. Note that multiple print heads may be used together, oriented with x-, y-, and / or z-dimensional offsets relative to one another (see axis legend in the upper left corner of FIG. 7B). The printing operation continues until the entire target area (and any border areas) has been printed with ink, as desired. Following deposition of the required amount of ink, the substrate is completed, either by evaporating the solvent to dry the ink (e.g., using a thermal process) or by use of a curing process, such as an ultraviolet curing process.

[0065] The right side of FIG. 7B depicts the interaction of a control system (e.g., comprising one or more processors 743 driven under the assistance of control software 745) in defining suitable printing process parameters and applying these parameters to control printing. More specifically, the control software causes the system to receive a base layer design that defines the topography (on the substrate) over which a desired layer is to be printed (e.g., in terms of a Cartesian map that defines pixel wells or substrate areas that are to receive liquid ink, as well as the desired thickness of the layer in question, according to numeral 747). The software stores or reads this information from memory, and then causes the system to receive process parameters (749), such as nozzle cross-scan pitch, nozzle positions, number of nozzles, and other information related to the printhead and the expected spreading characteristics of the ink. This various data is then applied in a print planning and / or print optimization step (751), where the planning utilizes the techniques referenced above to improve layer uniformity and / or print speed. For example, as described above, the software causes the system to calculate a reference volume / area (753), equate this to a thickness per unit (755), and then scale this thickness as a function of Cartesian location on the substrate (757) to create a firing map. As previously described, the firing map can be generated according to the average and spread measures per nozzle or per nozzle-waveform combination measured, and can be optionally filtered to provide multiple maps corresponding to scans depositing interleaved drops. Once such processing is performed, information representative of the nozzle firing details is then stored in a digital memory and subsequently passed to the printer for use in the fabrication process. As indicated by numeral 761, in one embodiment, the software causes the control system to select or vary one or more process parameters (droplet size, print grid definition, etc.) using the techniques discussed above, with any variations instantiated in the stored print data (759). During run time, the nozzle drivers are then driven in response to the stored print data as optimized.These drivers are implemented in the form of application specific integrated circuits, for example, each driving a number of nozzles, such as 256 nozzles, to control printing in response to an associated firing command, as shown by Figure 7B. Printing thus results in a layer of liquid ink on the substrate, which is subsequently advanced to a processing module to cure the liquid ink or otherwise convert it into a permanent form.

[0066] FIG. 7C provides additional details about the structure of the electronic control system (765) for the printer. Coordination across the various subsystems is provided by a set of processors 767 (not shown in FIG. 7C), acting under instructions provided by software. In one embodiment, these processors include a supervisory processor or general-purpose CPU and a set of additional processors used for parallel processing. In one specifically contemplated implementation, these additional processors take the form of a multi-core processor or a graphics processing unit (GPU, e.g., with hundreds or more cores). Each core is assigned a portion of the nozzle firing instructions to process. For example, each core can be assigned a set of nozzles and, under firmware control, can test these instructions to model whether the previously stored nozzle firing instructions meet threshold tests for layer uniformity and minimized wicking, or for other processing purposes, with exception handling applied if they do not. Parallel processing and / or This delegation is not required for all embodiments. During the fabrication process, the processor feeds data to the print head 768 to cause the print head to eject various volumes of ink in response to the firing instructions provided. The print head 768 typically has a number of inkjet nozzles arranged in rows or arrays and associated reservoirs that allow the ejection of ink in response to activation of a piezoelectric or other transducer. Such transducers cause individual nozzles to eject controlled amounts of ink in amounts governed by electronic firing waveform signals applied to corresponding piezoelectric transducers. Other firing mechanisms can also be used. The print head applies ink to the substrate 769 at various xy locations that correspond to grid coordinates. The variation in position is accomplished by both the print head motion system 770 and the substrate handling system 771 (e.g., causing the print to appear as one or more swaths across the substrate). In one embodiment, the print head motion system 771 moves the print head back and forth along the traveler while the substrate handling system provides both stable substrate support and "x" and "y" dimensional transport (as well as rotation) of the substrate, e.g., for alignment and tilt elimination. During printing, the substrate handling system provides relatively fast transport in one dimension (e.g., the "y" dimension for FIG. 7C), while the print head motion system 770 provides relatively slow transport in another dimension (e.g., the "x" dimension for FIG. 7C), e.g., for print head offset. In another embodiment, multiple print heads can be used with the primary transport handled by the substrate handling system 771. The image capture device 772 can be used to identify the location of any references and assist in alignment and / or error detection functions applied to the substrate position and orientation.

[0067] The apparatus also includes an ink delivery system 773 and a printhead maintenance system 774 to support printing operations. The printheads may be periodically calibrated or undergo a maintenance process. To this end, during a maintenance sequence, the printhead maintenance system 774 is used to perform appropriate priming, ink or gas purging, testing and calibration, and other operations as appropriate for the particular process. Such processes may also include, for example, individual measurements of parameters such as drop volume, velocity, and trajectory, as referenced by numerals 777 and 778.

[0068] As introduced above, the printing process can be performed in a controlled environment, i.e., in a manner that presents a reduced risk of contaminants that may degrade the effectiveness of the deposition layer. To this effect, the apparatus includes a chamber control subsystem 775 that controls the atmosphere in the chamber, as represented by functional block 776. Optional process variations can include performing the ejection of deposition material in the presence of an ambient nitrogen gas atmosphere (or another inert environment having specifically selected gases and / or controlled to exclude unwanted particulate matter), as described. As represented by numeral 779, the apparatus also includes a memory subsystem that can be used to store halftone pattern information or halftone pattern generation software, template print image data, and other data, as needed. For example, the memory subsystem can be used as an operating memory for the conversion of a previously generated print image according to the techniques introduced above, to internally generate printer control instructions that govern the firing (and timing) of each droplet. Where some or all of such rendering is performed elsewhere and the task of the apparatus is to fabricate device layers in accordance with received printer instructions, the received instructions may be stored in memory subsystem 779 for use during the printing process and / or maintenance, as appropriate. As represented by numeral 780, in one optional embodiment, individual drop details may be varied (e.g., to correct nozzle aberrations) through variation of the firing waveform of any given nozzle. In one embodiment, a set of alternative firing waveforms may be pre-selected and made available to each nozzle, on a shared or dedicated basis, and optionally (e.g., as described in U.S. Patent Publication No. 20150298 ). 153)。 While some embodiments use a predetermined scan path (i.e., despite the errors), with compensation for errors achieved using different nozzles and / or drive waveforms for certain nozzles, in other embodiments, print optimization (782) is performed to reassess scan path details and potentially improve deposition time.

[0069] FIG. 7D provides an illustrative diagram of a system 785 premised on optical measurements used in situ (i.e., represented by nozzle plates of individual printheads 787A / 787B, each with a large number of individual nozzles, e.g., 788) to measure droplet parameters of a relatively large printhead assembly 786, where in a typical implementation there will be hundreds to thousands of nozzles. An ink supply (not shown) is coupled to each nozzle (e.g., at nozzle 788), and a piezoelectric transducer (also not shown) is used to eject droplets of ink under control of an electronic control signal per nozzle. The nozzle design maintains a slight negative pressure of ink at each nozzle (e.g., nozzle 788) to avoid flooding the nozzle plate, and an electrical signal for a given nozzle is used to activate a corresponding piezoelectric transducer and pressurize the ink for the given nozzle, thereby expelling one or more droplets from the given nozzle. In one embodiment, the control signal for each nozzle is normally 0 volts, and a positive pulse or signal level at a given voltage is used for a specific nozzle to eject droplets for that nozzle (one per pulse). In another embodiment, a different tuned pulse (or other more complex waveform) can be used for each nozzle. However, in conjunction with the example provided by FIG. 7D, it will be assumed that it is desired to measure the droplet volume produced by a specific nozzle (e.g., nozzle 788) that is ejected downward from the printhead (i.e., in the direction of "h", which represents the z-axis height relative to a three-dimensional coordinate system 789) as the droplet is collected by the ejection well 790. Note that in a typical application, the dimension of "h" is typically on the order of one millimeter or less, and that there are many thousands of nozzles (e.g., 10,000 nozzles) that will have their individual droplets individually measured in this manner within an operational printer.Thus, in order to optically measure each droplet precisely (i.e., within a measurement window of about one millimeter as described, a droplet originating from a specific one of thousands of nozzles in a large print head assembly environment), certain techniques are used in the disclosed embodiments to precisely position the optical measurement components, the print head assembly 786, or both, relative to each other for optical measurement.

[0070] In one embodiment, these techniques utilize a combination of (a) xy motion control (791A) of at least a portion of the measurement system (e.g., in a dimensional plane 792) to precisely position a measurement area 793 immediately adjacent to any nozzle that is to generate droplets for optical calibration / measurement, and (b) sub-surface optical recovery (791B) (e.g., thereby allowing easy placement of a measurement area next to any nozzle despite the large printhead surface area). Thus, in an exemplary embodiment having about 10,000 or more printing nozzles, the motion system is capable of positioning at least a portion of the optical system at as many as 10,000 (e.g.) discrete locations proximate to the ejection path of each individual nozzle of the printhead assembly. In one embodiment, a continuous motion system, or a system with even finer positioning capabilities, can be used. Optical measurement techniques contemplated include shadowgraphy, interferometry, and the use of transparent tape to optically measure droplets from many individual nozzles at once (i.e., see U.S. Patent Publication No. 20150373305, for “Fast Measurement of Droplet Parameters in Industrial Printing System,” USSN 14 / 840343, filed August 31, 2015, which is incorporated by reference). The first two of these techniques allow for the use of a microscope to capture droplets in flight (e.g., in the case of shadowgraphy, to effectively image the shadow of the droplet) or to measure the droplets from multiple nozzles simultaneously. As the droplet lands on the tape or measurement surface (e.g., according to USSN 14 / 840343), the optics are typically adjusted in place so that a precise focus is maintained on the measurement area. Note that because a typical droplet can be on the order of a few microns in diameter, the optical arrangement is typically extremely precise, presenting challenges in terms of the relative positioning of the print head assembly and the measurement optics / measurement area. In some embodiments, to aid in this positioning, optics (mirrors, prisms, etc.) are used to orient the light capture path for sensing below the multi-dimensional surface 792 so that the measurement optics can be placed close to the measurement area without interfering with the relative positioning of the optical system and print head. This allows for effective position control in a manner that is not limited by the millimeter-order pile height h within which the droplet is imaged, or the large-scale x and y widths occupied by the print head being monitored. With interferometry-based drop measurement techniques, separate light beams incident on the droplet from different angles create interference patterns detectable from a viewpoint nearly orthogonal to the light path, and thus the optics in such systems capture light from angles approximately 90 degrees off the path of the source beam, but in a manner that utilizes sub-surface optical recovery to measure the drop parameters. Other optical measurement techniques can also be used as described. In yet another variation of these systems, the motion system 791A is optionally and advantageously made to be an xyz motion system that allows selective engagement and disengagement of the drop measurement system without moving the printhead assembly during drop measurement. Briefly, in an industrial processing device having one or more large printhead assemblies, it is considered that in order to maximize manufacturing uptime, each printhead assembly will be "parked" at a service station from time to time to perform one or more maintenance functions, and given the sheer size of the printhead and the number of nozzles, it may be desirable to perform multiple maintenance functions on different parts of the printhead at once.To this effect, in such an embodiment, it may be advantageous to move the measurement / calibration device around the printhead rather than vice versa. [This then allows for the engagement of other non-optical maintenance processes, e.g., on different nozzles if desired.] To facilitate these operations, the printhead assembly may optionally be "parked" with the system identifying a specific nozzle or set of nozzles that are subject to optical calibration. Once the printhead assembly or a given printhead is stationary, the motion system 791A is engaged to move at least a portion of the optical system relative to the "parked" printhead assembly to precisely position the measurement area 793 at a suitable location for detecting droplets ejected from a specific nozzle, the use of the z-axis of movement allowing for selective engagement of the light recovery optics from well below the face of the printhead, facilitating other maintenance operations instead of or in addition to optical calibration. Perhaps described differently, the use of the xyz motion system allows for selective engagement of the drop measurement system that is independent of other tests or test devices used in the service station environment. It should be noted that this structure is not required for all embodiments and other alternatives are possible where only the print head assembly moves and the measurement assembly is stationary, or where parking of the print head assembly is not required.

[0071] Generally speaking, the optics used for droplet measurement will include a light source 794, an optional set of light delivery optics 795 (which optionally direct light from light source 794 to measurement area 793), one or more light sensors 796, and a set of recovery optics 797 that direct light used to measure the droplet from measurement area 793 to one or more light sensors 796. Motion system 791A moves any one or more of these elements along with spittoon 790 in a manner that allows for the directing of light after droplet measurement from measurement area 793 to a subsurface location around spittoon 790, while optionally also providing a container (e.g., spittoon 790) to collect the ejected ink. In one embodiment, light delivery optics 795 and / or light recovery optics 797 move in a manner that is perpendicular to droplet travel, parallel to droplet travel. A motion system moves each of elements 794, 795, 796, 797, and the drainage basin 790 as an integral unit during droplet measurement using mirrors that direct light to / from the measurement area 793 along the dimension. This setup offers the advantage that the focus does not need to be recalibrated to the measurement area 793. As described by numeral 791C, light delivery optics are also optionally used to provide source light from a location below the measurement area dimensional surface 792, for example, as generally shown, with both light source 794 and light sensor 796 directing light on either side of drainage basin 790 for measurement purposes. As described by numerals 798 and 799, the optical system can optionally include lenses for focusing purposes, as well as photodetectors (e.g., for non-imaging techniques that do not rely on the processing of multi-pixelated "photos"). Again, note that the optional use of z-motion control for the optical assembly and spittoon allows for optional engagement and disengagement of the optical system, and precise positioning of the measurement area 793 adjacent to any nozzle at any time while the printhead assembly is "parked." Such parking of the printhead assembly 786 and xyz motion of the measurement system are not required for all embodiments. For example, in one embodiment, laser interferometry is used to measure droplet characteristics, and the printhead assembly (and / or optical system) is moved either in or parallel to the deposition plane (e.g., in or parallel to plane 792) to image droplets from the various nozzles. Other combinations and permutations are possible.

[0072] 8A-8D are used to illustrate the problems for the fabrication of one or more flat panels on a substrate. Examples are presented in these various figures illustrating the scanning motion of a print head relative to the substrate, techniques for encapsulating various electronic components (e.g., of an electronic display), fabrication of layers within pixel wells or banks, and techniques for layer edge contour management (e.g., when no such banks are present).

[0073] FIG. 8A depicts a substrate 801 with several dashed boxes representing individual panel products. One such product, seen at the bottom left of the figure, is designated using reference numeral 802. Each substrate (in a series of substrates) has, in one embodiment, several registration marks as represented by numeral 803. In one embodiment, two such marks 803 are used for the substrate as a whole to allow adjustment for substrate position offset, rotation error, and scale error, while in another embodiment, three or more such marks 803 are used to facilitate adjustment for tilt and additive type errors. In yet another embodiment, each panel (such as any of the four depicted panels) is accompanied by a per-panel registration mark, such as mark 805. These registration marks can be used for independent error handling per panel or per product. Again, a sufficient number or density of such marks (e.g., three or more per panel or other product) allows compensation for non-linear errors. These marks may be in addition to or instead of the substrate fiducials 803. In yet another embodiment, the registration marks are replicated at regular intervals regardless of panel position (e.g., as represented by oval 809). Whatever approach is used, one or more cameras 806 are used to image the registration marks to detect the errors referenced above. In one contemplated embodiment, a single stationary camera is used, and the printer's transport mechanism (e.g., a handler and / or air floating mechanism) moves the substrate to position each registration mark in turn within the field of view of the single camera. In a different embodiment, the camera is mounted on a motion system for transport relative to the substrate. In yet another embodiment, low and high magnification images are taken, the low magnification image is used to coarsely locate a fiducial for high resolution magnification, and the magnified image is used to identify a precise fiducial position according to the printer coordinate system, as discussed below. In consideration of the foregoing discussion, in one embodiment, the printer's transport mechanism controls the motion to within about 1 micron of the intended position.

[0074] In a typical implementation, printing will be performed to deposit a given layer of material over the entire substrate at once (i.e., with a single printing process that provides layers for multiple products). To illustrate this, FIG. 8A shows two illustrative scans 807 and 808 of a print head along the long axis of the substrate, where in a split-axis printer, the substrate is typically moved back and forth (e.g., in the direction of the depicted arrows) and the printer advances the print head positionally (i.e., perpendicular to the drawing page) between scans. Note that while the scan paths are depicted as linear, this is not required in any embodiment. Also, while the scan paths (e.g., 807 and 808) are depicted as adjacent and mutually exclusive in terms of the area covered, this is also not required in any embodiment (e.g., the print head can be moved on a portion-by-portion basis relative to the print swath, if desired). Finally, also note that any given scan path typically traverses the entire printable length of the substrate to print layers for multiple products in one pass. Each pass uses nozzle firing decisions according to a print grid firing map (as distorted or corrected), and each firing decision applies a waveform selected and programmed to produce the desired drop volume, trajectory, and velocity. Advantageously, a processor built into the printer (as introduced above) both qualifies and updates nozzle / droplet measurements and parameters, detects per-substrate or per-panel details, corrects the template print plan, and programs the nozzle firing data. Once printing is complete, the substrate and wet ink (i.e., the deposited liquid) can be allowed to settle (e.g., to promote mixing and fixing) and then transported for curing or processing of the deposited liquid into a permanent layer.For example, returning briefly to the discussion of FIG. 7A, a substrate can have "ink" applied in a printing module 705 and then subsequently transported to a curing chamber 721, all without breaking the controlled atmosphere (i.e., nitrogen or other inert atmosphere, advantageously used to prevent moisture, oxygen, ozone, or particulate contamination).

[0075] FIG. 8B is used to help illustrate one aspect of a process for the fabrication of OLED panels in such an environment. Specifically, FIG. 8B shows the substrate at a later stage of the fabrication process after structures have been added to the substrate and an organic encapsulation layer is formed over these elements. The assembly is generally represented by the numeral 811 and is considered to feature an array of panels on a common substrate. Features specific to one panel will be designated using a numeral followed by a separate letter, e.g., "A" for the first panel, "B" for the second panel, etc. Each panel 812A / B has several regions that serve different purposes. For example, numerals 813A / 813B refer to the active display area that contains the light-emitting layer (e.g., many pixels). Generally speaking, each active area 813A / B will include the electrodes and luminescent layers necessary to provide pixilation and associated routing of electrical signals, such as for control and powering. For active matrix OLED panel designs, the active areas also contain electrical circuitry (e.g., capacitors, transistors, conductive lines) that direct current to selected light-emitting luminescent pixels. The circuits in the active areas of the individual panels are powered and controlled through electrical contact with the individual terminals (e.g., 815A / B, 816A / B) associated with the terminal blocks 817A / 817B of the individual panels. Typically, an encapsulation layer will provide a protective "blanket" over only the active areas (i.e., to enclose the electroluminescent material) while allowing unimpeded external access to the terminal blocks 817A / B. Thus, the encapsulation printing process must deposit liquid ink in a manner that reliably and uniformly covers the active areas (813A / 813B) without gaps, holes, or other defects, while not simultaneously reliably and uniformly covering the terminal blocks 817A / 817B. The terminal blocks form part of the "exposed areas" that will not receive ink, while the exposed areas will receive ink deposited to form the desired layer. In Figure 8B, note the use of numeral 818 to represent the xyz coordinate system, and see the use of numeral 819 to refer to a separate set of ellipses to indicate the presence of any number of panels replicated in the x as well as y dimensions of the array.

[0076] FIG. 8C is used to illustrate the surface topography of the underlying electronic components that would encapsulate, for example, the emissive or other active layers of an OLED display pixel. FIG. 8C depicts a cross-sectional view of the assembly 811 taken along arrow CC from FIG. 8B. Specifically, the view shows the area of ​​the substrate associated with panel A, the active area 813A of panel A, and the conductive terminal (815A) of panel A that is used to achieve electronic connection to the active area. A small oval region 821 of the view is seen enlarged on the right side of the view to illustrate the layers for one pixel well in the active area above the substrate 812A. These layers include, respectively, an anode layer 829, a hole injection layer ("HIL") 831, a hole transport layer ("HTL") 833, an emissive or emissive layer ("EML") 835, an electron transport layer ("ETL") 837, an electron injection layer ("EIL") 838, and a cathode layer 839. Additional layers such as hole blocking layers, electron blocking layers, polarizers, barrier layers, primers, and other materials may also be included. Once completed, the OLED device will generally operate as follows: Application of a forward bias voltage (anode positive with respect to the cathode) will result in hole injection from the anode 829 and electron injection from the cathode layer 839. Recombination of these electrons and holes will then result in the formation of excited states of the light-emitting layer 835 material, which relax to the ground state with the emission of a photon of light. In the case of a "bottom-emitting" structure, light is emitted through a transparent anode layer 829 formed below the hole injection layer 831. A typical anode material can be formed, for example, from indium tin oxide (ITO). In a bottom-emitting structure, the cathode layer 839 is reflective and opaque. Typical bottom-emitting cathode materials include Al and Ag, typically with thicknesses greater than 100 nm. In a top-emitting structure, emitted light exits the device through the cathode layer, and for optimal performance, the anode layer is highly reflective and the cathode is highly transparent. A commonly used reflective anode structure includes a layered structure with a transparent conducting layer (e.g., ITO) formed over a highly reflective metal (e.g., Ag or Al) to provide efficient hole injection.Commonly used transparent top light-emitting cathode layer materials that provide good electron injection include Mg:Ag (10-15 nm with an atomic ratio of about 10:1), ITO, and Ag (10-15 nm). HIL831 is typically a transparent high work function material that readily accepts holes from the anode layer 829 and injects holes into the HTL833 layer. HTL833 is another transparent layer that passes the holes accepted from the HIL831 layer to the EML layer. Electrons are provided from the cathode layer 839 to the electron injection layer (EIL838). Electron injection into the electron transport layer is followed by injection from the electron transport layer into the EML where recombination with holes occurs after subsequent emission of light. The emission color depends on the EML layer material and is typically red, green, or blue for a full color display. The emission intensity is controlled by the electron-hole recombination rate, which depends on the driving voltage applied to the device.

[0077] As described, layers in the active area can be degraded through exposure to oxygen and / or moisture. It is therefore desirable to enhance OLED lifetime by encapsulating these layers on both the faces (822) and sides (823) of these layers. The purpose of the encapsulation is to provide an oxygen-resistant and / or moisture barrier, as described. The techniques disclosed herein, in addition to being used to deposit the organic light-emitting layers, can also be used for processing the encapsulation layers. FIG. 9A will be used to discuss encapsulation layer processing, while FIG. 8D, discussed immediately below, discusses layer processing within the pixel banks or "wells."

[0078] FIG. 8D shows an example of a pixel well, generally represented by numeral 841. The dashed box 843 indicates that this pixel well is only one example of many such wells (e.g., millions of wells for an HDTV screen, with separate pixel wells for red, green, and blue pixels in this example) defined on the substrate. The numeral 847 identifies a structural ridge that serves to "dam" the liquid printed in the central region 845; that is, to form the light-emitting element of the pixel, a conductor can first be deposited throughout the entire region 843 (or in a manner that provides electrical contact to a point in the region 845 and then a ridge is built over the structure), and a light-emitting element, such as various of the layers identified above in connection with FIG. 8C, is then deposited as a liquid in the well and processed to form a permanent layer. Within such a well, it is desirable to have good layer uniformity, and the techniques presented above are advantageously used for this purpose as well.

[0079] It should be noted that all layers are deposited within a containment structure, e.g., the encapsulation layer may in some applications be deposited so as to span many electronic components, such as the pixels of a display device. In some designs, a containment bank structure (or photolithographic structure) can optionally be used, as in the case of the pixels discussed above, but this need not be the case in all designs. For example, in many electronic display designs, it is possible to deposit layers so that they form natural boundaries or transitions as part of the deposition process.

[0080] As mentioned above, it is frequently desirable to provide sharp linear edges in boundary regions. For example, briefly returning to FIG. 5B (i.e., see numeral 565), if the ink density is maintained on a consistent basis right up to the layer periphery, the surface tension of the ink may cause the liquid ink to form "horns" or "bat ears". FIGS. 9A-9G are used to discuss techniques for varying the ink density in boundary regions to mitigate this and related effects, for example, by adjusting the volume / area "scaling" in the portions of the layer adjacent to the layer edges. This scaling can be done to customize the thickness in these regions and increase or decrease the edge height as relevant to the design. In one embodiment, this technique involves reducing the scaling of the drop density near the edges or corners of the intended print area. In another embodiment, the drop size can be varied (e.g., scaled) or locally deposited ink volumes controlled / customized differently to achieve these benefits.

[0081] FIG. 9A shows an aggregate structure 901 in which an encapsulation 902 has been added to the substrate, where a boundary or taper is used (e.g., instead of deposition in a containment structure). Note that the encapsulation 902 now encapsulates the face 822 and lateral edges 823 relative to the substrate 812A (from FIG. 8C), and the encapsulation extends laterally to occupy a larger deposition area than the underlying active layer. At the ends of this area, the encapsulation forms a taper or boundary region to help encapsulate / seal the lateral edges and leave the ends 815A exposed. The boundary region in this case provides a transition between the target area for the layer of interest (e.g., the area where the electronic components are encapsulated) and the area where the layer of interest is not to be formed (e.g., the area where the ends are exposed, or the product or substrate edge). This is depicted on the left side of FIG. 9A in the enlarged ellipse region 907. As seen in this enlarged view, the encapsulation comprises several thin layers, e.g., alternating organic and inorganic layers that provide a barrier against moisture and oxygen. The organic encapsulation layer can be advantageously deposited using the techniques introduced above, with the thickness of each individual layer being adjusted using the techniques described. For a particular organic encapsulation layer 909, a first region 911 overlies the underlying structures, such as the electrodes and other OLED layers discussed above. A second region 913 is formed by laminating the first region 911 over the underlying structures, such as the electrodes and other OLED layers discussed above. The first region 911 acts as an impingement region, i.e., to maintain a substantially uniform surface 915 that is planar with the first region 911. Optionally, the deposited thickness can be the same in both regions 911 and 913, but this need not be the case for all deposition processes. Regardless of the region, an inkjet printing process using techniques described herein can be used to control the thickness and promote uniformity of the particular encapsulation layer 909 where at least the structures are encapsulated, and optionally, within the second buffer region 913. Finally, a third gradient or boundary region 917 represents a transition to the exposed area of ​​the underlying substrate (e.g., to provide electrical terminals for the active region). Numeral 919 indicates an associated taper in the encapsulation surface as it transitions to the exposed substrate.

[0082] FIG. 9B represents a graph 921 illustrating how the techniques described herein can be applied to shape layer edges. Such techniques are particularly useful when no containment structure is used, although of course it is optional to use such techniques within a containment structure. More specifically, three curves 923, 925, and 927 obtained using stylus profilometer measurements of a processed 6.0 micron encapsulation layer are represented in FIG. 9B. The difference between these curves was generated by the density applied to a unit area of ​​the substrate ("printed cell") in the area abutting the intended layer's lateral edge. In contrast to the reference represented by curve 923, curve 925 represents a process in which the ink volume / density for a printed cell is reduced approaching a boundary within the encapsulation layer (e.g., in front of the encapsulation layer perimeter). In contrast, curve 927 represents a process in which the ink volume / density is increased for printed cells abutting the same boundary. Note that the resulting layer thickness actually increases slightly just prior to the boundary, e.g., at x-locations of 2,000μ and 17,000μ. By adjusting the ink volume / density or other details for the boundary region, the designer can adjust the edge buildup at the layer boundary in a desired manner, including for purposes of providing a uniform layer thickness or surface, or smoothing or enhancing a transition. Note that the amount of ink buildup adjacent to a layer edge depends in large part on ink properties such as surface tension (and its dependence on temperature), and many of the adjustment principles described herein can be used to adjust this buildup. For example, some inks may naturally form a lip or so-called capillary ridge (e.g., as represented at point 929 of curve 927), and in such cases, the volume / drop density adjustment process described herein can be applied to remove this lip to help adjust the final layer thickness, for example, by reducing a scaling value or otherwise adding an adjustment or correction factor for the printing cells abutting the layer edge so that the permanent layer contour more closely matches curve 923.Other processes can also be used to the same effect, for example using "larger" or "smaller" drop sizes in boundary regions, or drop sizes that are "tipped" close to the boundary, or pattern modification techniques (e.g., where drops are selectively removed from the pattern to adjust the edge profile). Many other techniques will occur to those skilled in the art.

[0083] The spreading properties of the ink can potentially affect the sharpness of layer edges and corners on a broad scale, for example, as dot gain increases (i.e., the ink spreads too much), such sharpness becomes more difficult to control. Thus, as introduced above, ink volume, drop density, and / or drop size or density can be varied at or adjacent to an edge or corner.

[0084] Figure 9C provides a depiction 931, no grid points shown, used to illustrate the effect in corners 939 of a deposited layer 933. To produce thin films, the ink volume / droplet density can be made relatively sparse within areas 935 and 937 for portions of the intended layer (933) that are not near the layer edges, which helps to mitigate the "corners" or "bat ears" discussed above (see, e.g., Figures 5B and 9B). However, depending on the diffusion characteristics, if sparse ink deposition is used in regions 935, 937, and 939, the associated "low" ink volume / density may produce sharp edges. Thus, the ink volume (e.g., drop size, drop density, etc.) in areas 935, 937, and 939 may also be intentionally increased, or the deposition details may be otherwise modified to improve edge linearity. Alternatively, dedicated scan passes of the effective grid spacing using alternative nozzle driving waveforms (e.g., selected to vary drop volume), or techniques described herein may also be applied in the same spirit.

[0085] FIG. 9D provides a depiction 941 that is used to help illustrate one processing technique that adjusts material thickness at layer edges in the context of an OLED panel. For example, in an encapsulation process such as the one discussed, it may be desirable to ensure consistent thickness in the solidified layer up to the planned encapsulation perimeter to provide reliable edge sealing of any underlying sensitive material layers. The substrate is viewed in plan view, but the depiction of electrical terminals is omitted. Note the use of fiducials 951 to match the process so that the organic encapsulation layer is printed correctly over the underlying substrate. The substrate topography onto which the encapsulation layer is deposited is considered to comprise areas 943 and 945 (e.g., roughly corresponding to the use of numbers 911 and 913 as discussed above in connection with FIG. 9A). Considering the effects of spreading of the deposited ink and surface energy / tension of that ink, rather than having undesired edge effects, the scaling of the ink volume / area or drop density can be adjusted in this embodiment along the intended layer edges, and in doing so, modify the edge profile at the substrate / layer perimeter. For example, the scaling in region 945 can be optionally increased as depicted by the grayscale value seen at the left of FIG. 9D to increase the ink volume in areas approaching the boundary, or a correction factor can be added to the volumetric plan. In this regard, it is noted that each unit area of ​​the substrate (represented by a value between 0 and 255, respectively) can initially be associated with a particular thickness, represented, for example, in this example by a virtual grayscale value of "220". If it is experimentally determined that a transition (e.g., at the boundary between regions 945 and 947) provides insufficient coverage due to ink spreading, the grayscale value can be selectively increased to provide relief, for example, by increasing the value of one or more rows or columns of a target region representing, for example, a layer periphery, corresponding to region 945 and associated scaling value 954 (e.g., from "220" to "232" in FIG. 9D).The corrections can be pre-stored as correction images or firing maps (e.g., because the corrections can vary as a function of process, temperature, ink, and other factors, depending on the application), or can be otherwise incorporated into the original nozzle firing instructions or other stored data. In conjunction with the techniques described in the above-referenced U.S. Pat. No. 8,995,022, these grayscale values ​​can be used to select the drop density (e.g., halftone density). Alternatively, these values ​​can be used in the second embodiment to adjust (or select) the waveforms applied to the nozzle groups. Note that when multiple boundary conditions exist, e.g., the intersection of two boundaries, it may be desirable to provide further adjustments, such as a depicted grayscale value of "240" for the corner target area 955. Clearly many possibilities exist, and the use of grayscale values ​​as illustrated by FIG. 9D is not required for all embodiments. By adjusting the volume / area, drop volume, and / or drop density in these boundary regions (or otherwise adjusting the effective ink volume according to pre-measured properties), the techniques introduced above allow for customized control over the layer edges in any manner suitable for the particular deposition process at issue, for example to facilitate edge sealing of flat panel devices. Note that it is also possible for the software to automatically provide adjusted drop size / density scaling any time the software detects a printed cell within a defined distance from the layer edge, depending on the desired embodiment or effect, as will be shown below. Either in addition to, or in addition to, such volume / density adjustments, the optional technique of "fencing" can also be employed.

[0086] The choice of a particular technique for adjusting edge buildup (or other transitions in layer thickness) can be made by the designer depending on several factors. For example, adjusting the drop density may make sense when the typical layer thickness assumes (for a particular layer deposition process) a printhead scan with intermediate density relative to the printing grid, i.e., so that there is headroom to modify the ink volume / area (e.g., without compromising "blanket" substrate coverage in the edge area). Conversely, varying the print drop size (without varying the drop density) may be effective when one has the ability to change the drop size (to larger or smaller sizes) without also affecting the "blanket" substrate coverage in the area of ​​interest. Employed together, these techniques can be extremely powerful for generating edge profiles that meet almost any desired goal.

[0087] 9E helps illustrate the "fencing" that improves edge integrity and shows a portion of a substrate 961 having a central region 963 of uniform layer thickness, a border region 965 of "controlled" ink volume / area, drop density, or drop size (i.e., selected to avoid edge buildup), and a set of fencing clusters 967 selected to provide edge uniformity. Perhaps described differently, the central region 963 represents a substantial area of ​​ink, the border region 965 represents an area of ​​controlled ink volume or density relative to the central region, and the fencing clusters 967 represent a relatively dense ink pattern selected to provide a crisp, well-defined layer edge in the final layer. It is believed that the mix of fixed structure within the random, sparser structure of the tapered edge will lead to less irregularities and reduce the possibility of larger voids when the ink volume is over-extended close to the edge. In alternative designs, the fencing clusters can be placed directly adjacent to the central region (e.g., with no border region between the cluster and the central region) or can be made part of the central region itself (i.e., the edge). As should be appreciated by these examples, many variations are possible that rely on additional density variation to adjust edge buildup and / or provide desired edge characteristics.

[0088] It should be noted that in addition to adjusting the grayscale values ​​of the boundary regions, it is also possible to adjust the location or pattern of the printing drop firing decisions applied to such regions, i.e., modify the local firing decisions, thus providing a "tilted" drop pattern near the edge. For example, FIG. 9F shows an exemplary printing grid pattern 981 that may be used when a region represents a corner of a deposited film. FIG. 9F is similar to FIGS. 3A-3C, 3E, and 5B in its reference of the printing grid. The particular drop pattern represented in FIG. 9F represents an ink volume similar to the pattern seen in FIG. 5B (i.e., 13 of 25 possible drops are fired). However, while the pattern in FIG. 9F is characterized by a relatively high density use of drops along the top edge 983 of the substrate and the left edge 985 of the film, the interior region 987 is left relatively sparse, i.e., to produce relatively sharp left and top edges. Such techniques can be combined to facilitate this variation of drop size and / or grid spacing and / or other deposition details referenced above to customize the edge profile of the finished layer. In the variant represented by Figure 9G, variable drop sizes can also be used for this purpose, e.g., a printed grid pattern 991 can feature large drop size selections (995) at the periphery of the pattern, e.g., to provide fencing or for other effects, while smaller drop sizes (993) or sparse drop firing (e.g., 987) can be used elsewhere within the pattern.

[0089] The use of various edge customization, framing, or "fencing" techniques is It should be noted that this is not required of the embodiment, and that this is within the ability of one of ordinary skill in the art to determine the best strategy for a particular application, ink, and process technology. Also, while the focus of the example in Figures 9A-G is on an encapsulation layer, these same principles can be applied to the formation of any desired layer. For example, it is expressly contemplated that the described printing principles can be used to fabricate any of the HIL, HTL, EML, ETL, or other layers of an OLED device, for example, by way of example, with individual pixel wells, or on another patterned or non-patterned basis.

[0090] Considering the techniques presented above, what is described is a set of processes for improving the uniformity of processed solidified layers originally deposited as discrete droplets representing layers of ink. Through careful control of the printing process and parameter adjustments in the manner described, it is possible to plan printing details in advance in a manner that deposits droplets to provide locally uniform ink volumes and even controls the mixing of adjacent droplets in a manner that mitigates against wicking and associated layer defects. By automatically adjusting the process based on expected droplet characteristics, including ink spreading, it is possible to improve printing speed and provide more uniform layers than would otherwise be possible, thereby improving layer quality and product quality and ultimately reducing manufacturing costs.

[0091] It should be appreciated that the present disclosure provides specific examples of systems, methods, services, and software that significantly enhance electronic device processing and electronic flat panel processing. These techniques may be extended into systems, articles, or software, or other forms of implementation.

[0092] Other applications will readily occur to those familiar with printing processes for electronic device fabrication in light of the teachings provided above, or in light of the present invention as defined by the claims set forth below. Thus, the foregoing discussion is intended to be illustrative only, and nevertheless, other designs, uses, substitutions, modifications, and improvements are possible that are within the scope of this disclosure, which is defined only as limited by the following claims and their equivalents.

Claims

1. 1. A method of processing a substrate using an inkjet printer, the method comprising the steps of: depositing droplets of a liquid onto the substrate and processing the liquid to form a finished layer; receiving a desired thickness of the finished layer to be formed on the substrate; receiving data representative of diffusion properties of the liquid on the substrate; receiving data representative of a plurality of drop sizes to be produced by the inkjet printer; for each droplet size of the plurality of droplet sizes, determining a minimum volume per unit area for depositing the droplets on the substrate at a droplet size that results in a continuous liquid layer without voids on the substrate based on the spreading characteristics; identifying a droplet size among the plurality of droplet sizes that can be deposited in a smallest volume per unit area to form a continuous liquid layer that, after processing, results in the finished layer having a thickness less than or equal to the desired thickness; selecting from said identified droplet sizes droplet sizes that meet one or more additional criteria; determining a droplet density for depositing the droplets of the liquid at the selected droplet size to form a continuous liquid layer that, after processing, results in the finished layer having a thickness less than or equal to the desired thickness; assigning nozzle firing decisions to each printing nozzle of the inkjet printer based on the drop density; printing a successive liquid layer using the selected drop size and in accordance with the assigned nozzle firing decisions; and treating the printed continuous liquid layer to form the finished layer.

2. The method of claim 1 , wherein the one or more additional criteria include scan time, layer uniformity, or both.

3. 2. The method of claim 1, wherein identifying a droplet size from the plurality of droplet sizes that can be deposited in a smallest volume per unit area to form a continuous liquid layer that, after processing, results in the finished layer having a thickness less than or equal to the desired thickness comprises starting with a smallest droplet size from the plurality of droplet sizes and progressing to the next droplet size.

4. The method of claim 1 , further comprising determining a droplet density for each of the plurality of droplet sizes to form a void-free liquid layer within a unit area.

5. The method of claim 1 , wherein the step of assigning nozzle firing decisions comprises using halftoning.

6. The method of claim 1 , wherein allocating the nozzle firing decisions comprises sorting the nozzle firings into a number of scans.

7. The method of claim 1 , wherein the diffusion characteristics are determined using a test deposition process that includes image-based analysis of a test substrate.

8. The method of claim 1 , wherein the step of assigning nozzle firing decisions includes applying an alternating stack filter.

9. The method of claim 1 , wherein the droplets are deposited using the printing nozzles of a printhead, and the plurality of droplet sizes comprises a range of droplet sizes for each of the printing nozzles.

10. 2. The method of claim 1, wherein determining a droplet density at which to deposit the droplets at the selected droplet size to form a continuous liquid layer that, after processing, results in the finished layer having a thickness less than or equal to the desired thickness includes determining a reference droplet density based on a minimum volume per unit area for the selected droplet size, and scaling the reference droplet density based on the desired layer thickness.

11. 1. A method of processing a substrate using an inkjet printer, the method comprising the steps of: depositing droplets of a liquid onto the substrate and processing the liquid to form a finished layer; receiving a desired thickness of the finished layer to be formed on the substrate; receiving data representative of diffusion properties of the liquid on the substrate; receiving data representative of a plurality of drop sizes to be produced by the inkjet printer; determining, for each of the plurality of droplet sizes, a thickness of the resulting finished layer formed by depositing droplets of liquid on the substrate with a minimum volume per unit area such that the liquid layer is free of voids based on the diffusion characteristics; identifying a droplet size from among the plurality of droplet sizes that results in the determined finished layer thickness being equal to or less than the desired thickness; selecting, from the identified droplet sizes, droplet sizes that satisfy one or more additional criteria; determining a droplet density for depositing the droplets of the liquid at a selected droplet size to form a continuous liquid layer that, after processing, results in the finished layer having a thickness less than or equal to the desired thickness; assigning nozzle firing decisions to each printing nozzle of the inkjet printer based on the drop density; assigning a waveform for firing each of the printing nozzles to vary the drop size, drop position, or both; printing a successive liquid layer using the selected drop size in accordance with the assigned nozzle firing decisions and drive waveforms; and processing the printed successive liquid layer to form a finished layer.

Images