EP3314633A1 - Fiber laser-based system for uniform crystallization of amorphous silicon substrate - Google Patents

Fiber laser-based system for uniform crystallization of amorphous silicon substrate

Info

Publication number
EP3314633A1
EP3314633A1 EP16818736.7A EP16818736A EP3314633A1 EP 3314633 A1 EP3314633 A1 EP 3314633A1 EP 16818736 A EP16818736 A EP 16818736A EP 3314633 A1 EP3314633 A1 EP 3314633A1
Authority
EP
European Patent Office
Prior art keywords
panel
harmonic
line
beams
assembly
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
EP16818736.7A
Other languages
German (de)
French (fr)
Other versions
EP3314633A4 (en
Inventor
Manuel LEONARDO
Michael Von Dadelszen
Yuri Erokhin
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
IPG Photonics Corp
Original Assignee
IPG Photonics Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by IPG Photonics Corp filed Critical IPG Photonics Corp
Publication of EP3314633A1 publication Critical patent/EP3314633A1/en
Publication of EP3314633A4 publication Critical patent/EP3314633A4/en
Pending legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/02—Manufacture or treatment of semiconductor devices or of parts thereof
    • H01L21/02104—Forming layers
    • H01L21/02365—Forming inorganic semiconducting materials on a substrate
    • H01L21/02656—Special treatments
    • H01L21/02664—Aftertreatments
    • H01L21/02667—Crystallisation or recrystallisation of non-monocrystalline semiconductor materials, e.g. regrowth
    • H01L21/02691—Scanning of a beam
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/02—Manufacture or treatment of semiconductor devices or of parts thereof
    • H01L21/04—Manufacture or treatment of semiconductor devices or of parts thereof the devices having at least one potential-jump barrier or surface barrier, e.g. PN junction, depletion layer or carrier concentration layer
    • H01L21/18—Manufacture or treatment of semiconductor devices or of parts thereof the devices having at least one potential-jump barrier or surface barrier, e.g. PN junction, depletion layer or carrier concentration layer the devices having semiconductor bodies comprising elements of Group IV of the Periodic System or AIIIBV compounds with or without impurities, e.g. doping materials
    • H01L21/26—Bombardment with radiation
    • H01L21/263—Bombardment with radiation with high-energy radiation
    • H01L21/268—Bombardment with radiation with high-energy radiation using electromagnetic radiation, e.g. laser radiation
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/02—Manufacture or treatment of semiconductor devices or of parts thereof
    • H01L21/02041—Cleaning
    • H01L21/02057—Cleaning during device manufacture
    • H01L21/02068—Cleaning during device manufacture during, before or after processing of conductive layers, e.g. polysilicon or amorphous silicon layers
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/02—Manufacture or treatment of semiconductor devices or of parts thereof
    • H01L21/02104—Forming layers
    • H01L21/02365—Forming inorganic semiconducting materials on a substrate
    • H01L21/02518—Deposited layers
    • H01L21/02521—Materials
    • H01L21/02524—Group 14 semiconducting materials
    • H01L21/02532—Silicon, silicon germanium, germanium
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/02—Manufacture or treatment of semiconductor devices or of parts thereof
    • H01L21/02104—Forming layers
    • H01L21/02365—Forming inorganic semiconducting materials on a substrate
    • H01L21/02518—Deposited layers
    • H01L21/02587—Structure
    • H01L21/0259—Microstructure
    • H01L21/02592—Microstructure amorphous
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/02—Manufacture or treatment of semiconductor devices or of parts thereof
    • H01L21/02104—Forming layers
    • H01L21/02365—Forming inorganic semiconducting materials on a substrate
    • H01L21/02656—Special treatments
    • H01L21/02664—Aftertreatments
    • H01L21/02667—Crystallisation or recrystallisation of non-monocrystalline semiconductor materials, e.g. regrowth
    • H01L21/02675—Crystallisation or recrystallisation of non-monocrystalline semiconductor materials, e.g. regrowth using laser beams
    • H01L21/02678—Beam shaping, e.g. using a mask
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/02—Manufacture or treatment of semiconductor devices or of parts thereof
    • H01L21/02104—Forming layers
    • H01L21/02365—Forming inorganic semiconducting materials on a substrate
    • H01L21/02656—Special treatments
    • H01L21/02664—Aftertreatments
    • H01L21/02667—Crystallisation or recrystallisation of non-monocrystalline semiconductor materials, e.g. regrowth
    • H01L21/02675—Crystallisation or recrystallisation of non-monocrystalline semiconductor materials, e.g. regrowth using laser beams
    • H01L21/02683—Continuous wave laser beam
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/02—Manufacture or treatment of semiconductor devices or of parts thereof
    • H01L21/02104—Forming layers
    • H01L21/02365—Forming inorganic semiconducting materials on a substrate
    • H01L21/02656—Special treatments
    • H01L21/02664—Aftertreatments
    • H01L21/02667—Crystallisation or recrystallisation of non-monocrystalline semiconductor materials, e.g. regrowth
    • H01L21/02675—Crystallisation or recrystallisation of non-monocrystalline semiconductor materials, e.g. regrowth using laser beams
    • H01L21/02686—Pulsed laser beam
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/02—Manufacture or treatment of semiconductor devices or of parts thereof
    • H01L21/04—Manufacture or treatment of semiconductor devices or of parts thereof the devices having at least one potential-jump barrier or surface barrier, e.g. PN junction, depletion layer or carrier concentration layer
    • H01L21/18—Manufacture or treatment of semiconductor devices or of parts thereof the devices having at least one potential-jump barrier or surface barrier, e.g. PN junction, depletion layer or carrier concentration layer the devices having semiconductor bodies comprising elements of Group IV of the Periodic System or AIIIBV compounds with or without impurities, e.g. doping materials
    • H01L21/30—Treatment of semiconductor bodies using processes or apparatus not provided for in groups H01L21/20 - H01L21/26
    • H01L21/324—Thermal treatment for modifying the properties of semiconductor bodies, e.g. annealing, sintering

Definitions

  • This disclosure relates to fiber laser-based method and system for manufacturing flat panel displays. Particularly, the disclosure relates to a fiber laser system for annealing amorphous silicon panel by a harmonic laser beam so as to fabricate substantially uniform polycrystalline silicon displays, and a method of operating the inventive system.
  • the Flat Panel Display (FPD) fabrication environment is among the world's most competitive and complex technologies. Generally, this technology is in constant need for compact products equipped with the FPD and featuring high resolution, bright large display, low power consumption and fast video capability and, of course, low cost.
  • FPD Flat Panel Display
  • the thin film transistor (TFT) technology is the basis for the FPD fabrication that can be either high-resolution, high-performance liquid crystal display (LCD), or organic light emitting diode (OLED).
  • TFT thin film transistor
  • LCD liquid crystal display
  • OLED organic light emitting diode
  • display circuits were made on a thin opaque layer of amorphous silicon ("a-silicon or a-Si") and arranged in a backplane across the layer to correspond to respective pixels.
  • a-silicon or a-Si amorphous silicon
  • Each pixel includes a great number of crystalline grains which are each formed with a specific grain area and orientation.
  • a-Si was at least partially replaced by poly-silicon (p-Si) which has the carrier mobility approximately two orders of magnitude greater than that of a-Si, substantially improves the aperture ratio, pixel resolution, and reduces the pixel size.
  • p-Si poly-silicon
  • a-Si is initially thermally treated to convert into liquid amorphous Si, and then it is maintained in the molten state for a certain period of time.
  • the temperature range sufficient to maintain the molten state is selected to allow the initially formed poly-crystallites to grow or crystallize.
  • the surface is cooled to induce a solidification phase of the processed material.
  • the LTPS approach is based on two generic methods - Excimer Laser Annealing (ELA) or partial melt LA and sequential lateral
  • the SLS method includes melting the entire thickness of the a-Si film by a beam from an excimer laser that operates at a 3xx nm wavelength.
  • crystallization fronts grow from opposite sides of the molten film.
  • the growth is lateral.
  • the laterally developed crystalline grains can be elongated to large horizontal dimensions. The latter is advantageous since electron mobility increases as grains grow larger.
  • both SLA and partial annealing (EL) methods are viable options.
  • excimer lasers used in both LA and SLS processes, dominate annealing of TFT flat panel displays.
  • Excimer lasers provide a wide range of processing power, with average range of processing powers up to 300 W and higher, energies higher than 1 J and pulse width from 10 to 250 ns.
  • excimers deliver UV light at the wavelength 3xx nm, which is directly absorbed in a-Si without additional frequency conversion.
  • the pulse frequency of the excimer laser is relatively low. To the best of applicants' knowledge, it does not exceed 6 kHz in SLS processes and considerably lower in standard ELA . Such a relatively low frequency in the standard ELA may explain the necessity of multiple irradiations of each location since the duration between subsequent pulses is greater than the time constant (of the process) beyond which excited atoms lose their mobility. As to the SLS, with KHz frequencies leading to high energies, the excimer requires multiple gas changes over a daylong period of operation which makes it unsuitable for mass production. Many of the above disclosed and other disadvantages of the excimer lasers are not characteristic to fiber lasers. [0009] A need therefore exists for using a fiber laser-based annealing system operable to provide a substantially uniform p-crystalline structure (p-Si) by using both SLS and partial anneal (LA) methods with a narrow width laser beam.
  • p-Si substantially uniform p-crystalline structure
  • LA partial anneal
  • the SLS processes require a line beam on target, wherein the long axis is orders of magnitude larger than the short axis (e.g. 2mm short ( ⁇ 400: 1)).
  • Primary scanning is typically conducted in the direction of the short axis, but long axis scanning is not precluded.
  • the dimensions of the line beam are effectively driven by 3 parameters: desired width of the short axis (function of desired poly-Si grain size and short axis intensity profile), desired fluence (J/cm 2 ) on target, and pulse energy reaching the target.
  • desired width of the short axis function of desired poly-Si grain size and short axis intensity profile
  • desired fluence J/cm 2
  • pulse energy reaching the target For example, a 5 ⁇ wide line with ⁇ pulse energy and 1 J/cm 2 required fluence would allow for a line beam length of «2mm.
  • the specific relationships are dependent on the intensity profile of the short axis.
  • a top hat short axis profile for example, would allow for a longer line beam than a Gaussian short axis profile with all other parameters being the same. Nevertheless, the above values are indicative of the relative relationships between the various parameters under consideration.
  • the panels to be annealed with said line beams are orders of magnitude larger than the length of the line beams that can be achieved in SLS with individual burst mode fiber lasers. It will be necessary to stitch together beams from individual or multiple lasers to achieve effectively continuous poly-Si grain structures.
  • the partial LA process requires considerably stronger requirements for fluence and higher average powers. However, partially because of beam width above 5 micron which is greater than a 1 to 2 micron linewidth in SLS, the stitching problem does not exist.
  • aspects of the disclosed system include scanning/stepping in each axis, multiple laser compatibility, requisite speeds and accuracies, auto-focus
  • the disclosed system for crystallizing amorphous Si (a-Si) panels by partial melt laser annealing (LA) or sequential lateral solidification (SLS) annealing process includes at least one single transverse mode (SM) quasi- continuous wave (QCW) fiber laser source, emitting a pulsed harmonic beam at a pulse repetition rate of at least 80 MHz, along a path,
  • the system further has a delivery system including a beam conditioning assembly located downstream from the fiber laser source and configured to transform the harmonic beam such that the harmonic beam has desired divergence and spatial distribution characteristics.
  • the delivery system further has a beam velocity and profile assembly operative to provide the conditioned harmonic beam with a desired intensity profile at a desired scanning velocity in an object plane.
  • the delivery system a beam imaging assembly for imaging the conditioned harmonic beam in the object plane onto an image plane in at least one beam axis at a desired demagnification such that a width of the conditioned harmonic beam is reduced to a narrow linewidth of at least 1 ⁇ at the image plane.
  • the system is provided with a panel handling assembly operative to provide relative position and velocity between the imaged narrow linewidth beam and the panel such as to irradiate each location of the a-Si panel at least two times with an exposure time during each time of at least 100 ns in order to provide transformation of a-Si to polysilicon (p-Si) structure with a uniform grain size of at most 1 ⁇ .
  • system described in the 1 aspect further has multiple SM QCW fiber laser sources emitting respective beam that are combinable together.
  • the system of any of the above aspects is provided with the beam velocity and profile system which is configured to convert a high-ratio Gaussian harmonic beam into a flat-top harmonic and selected from a beam segmentation and recombination system, beam re-apodization system, beam combining system, and beam cropping system.
  • Another aspect of the inventive system of any of the above aspects concerns the beam segmentation and recombining system which is selected from a fly's eye or bi-prism optical arrangements.
  • the fly's eye is imaging or non-imaging homogenizer.
  • Another aspect of the inventive system of any of the above aspects relates to the beam combining system configured to overlap and intersperse multiple harmonic beams to generate a central homogeneous central top portion of an intensity profile which is cropped at the object plane.
  • the beam combining system is operative to overlap multiple harmonic beams. One of the overlapped beams is everted so that a resulting harmonic beam is homogeneous in a direction of long axis.
  • the beam re- apodization system is configured with at least one or multiple acylindrical optical elements converting a high-ratio Gaussian harmonic beam into a flat-top intensity profile in at least of beam axes.
  • the beam velocity and profile system is configured with a scanner operative to provide the conditioned harmonic beam with a desired velocity so that the imaged narrow linewidth beam homogeneously and continuously generate a line of crystallization without stitching.
  • the beam imaging assembly is configured with a focusing lens focusing the conditioned harmonic beam with the desired intensity profile at the desired scanning velocity in a short beam axis direction onto a first mask,
  • the latter defines the object plane and has cutting knives for sharpening edges of the beam in a long axis direction, and an objective lens located downstream from the first mask and adjacent to the panel.
  • the beam imaging assembly further includes a second mask located between the first mask and objective lens and configured to vignette residual inhomogeneity of the line of crystallization.
  • the beam imaging assembly is configured with an anamorphic lens arrangement providing the different desired demagnification along orthogonal beam axes of the conditioned harmonic beam with the desired intensity profile.
  • the current aspect of the system of any of the above and following aspects relates to the beam imaging assembly which is anamorphic and includes two spaced masks providing different demagnification in respective orthogonal beam axes and having different object planes.
  • Yet another aspect of the inventive of any of the above and following aspects concerns the beam imaging assembly which has a proximity mask configured to define a desired length of a beamline on the panel, the proximity mask is spaced from the panel at a distance limiting edge diffraction.
  • the system of any of the above and following aspects also includes the panel handling assembly with a support supporting the panel to be annealed such that the panel is displaceable in orthogonal XY planes relative to the fixed beam imaging assembly.
  • Still another aspect of the inventive system of any the above and following aspects relates to the panel handling assembly including a support supporting the panel to be annealed and configured such that the panel is stationary relative to the beam imaging assembly displaceable in orthogonal XY planes.
  • the panel handling assembly is configured with a support supporting the panel to be annealed such that the panel is displaceable on one of XY planes and the beam imaging system is displaceable on another of XY planes.
  • the system of this aspect related to any of the above and following aspect is configured with the SM QCW fiber laser source which is mounted in a fixed position relative to the displaceable beam image system or displaceable therewith.
  • the system of any of the above aspect and following aspects further includes an auto focus system, beam profiler and MURA measuring system.
  • system of any of the above and following aspect Is configured with the fly's eye homogenizer having delay step glass elements to eliminate coherence effects.
  • Still another aspect of the inventive system of any of the above and following aspects includes a dithering system operative to dither the narrow linewidth beam onto the panel such that residua] iiihomogeneous regions of the p-Si structure are at different locations in sequential lines effectively smoothing out the residual inhomogeneities and reducing the Mura to a predetemiine reference range.
  • Yet another aspect of the disclosed system defined in any of the above or following aspects 24 relates to the dithering system which is operative to oscillate the panel or any suitable component downstream form the object plane, or an optical component of the beam delivery system, such as a lens or mirror, or an optical component along the beam path in the direction of the conditioned harmonic narrow width line beam during the SLS annealing process, or the mask defining the object plane.
  • system is configured to generate a continuous line in both SLS and AL processes with one or multiple passes.
  • the inventive system is operative provide interdigitation of multiple beams after two or more passes.
  • the inventive system of any of the above and following aspects is configured to provide pulse picking utilizing a mechanical scanning, acousto-optic or electro- optic method.
  • FIG. 1 is a general highly schematic view of the inventive system
  • FIG. 2A is one modification of the inventive system of FIG. 1;
  • FIG. 2A is another embodiment of the inventive system of FIG. 1;
  • FIG. 3 is a view of beam conditioning/homogenizing and beam profiling subassemblies of the inventive systems of FIG. 2 A and 2B;
  • FIG. 4 is a view of beam imaging subassembly of FIGs. 2A and 2B;
  • FIG. 5 is a view of panel handling sub-assembly of FIG. 1 ;
  • FIG. 6 is an optical schematic of the beam homogenizing subassembly with a beam polarizing combiner
  • FIG. 7 is a graphical representation of principle of operation of the homogenizing subassembly of FIG. 6;
  • FIGs. 8A through 9B are respective further graphical representation of principle of operation of the homogenizing subassembly of FIG. 6;
  • FIGs. 1 OA- IOC are respective arrangement operating for combining multiple beams;
  • FIGs. 11A-11C are respective orthogonal, side and top views of another configuration of the beam homogenizing subassembly
  • FIGs. 12A - 12C illustrate a thermal transformation of a single line to be crystallized by a laser beam
  • FIGs. 12A illustrates a general principle of operation of scanning subassembly
  • FIGs. 13B-13D illustrate a harmonic laser beam with respective different intensity profiles.
  • FIGs. 14A -14C is respective views of a polygon scanner and system utilizing the polygon.
  • FIGs. 15A - 15B are identical to FIGs. 15A - 15B;
  • FIGs. 16A-16B are identical to FIGs. 16A-16B;
  • FIGs 18A-18B are identical to FIGs 18A-18B;
  • FIGs. 19A-19B are identical to FIGs. 19A-19B;
  • FIGs20A-20D are identical to FIGs20A-20D.
  • FIG. 22
  • FIG. 23 FIGs. 24-27.
  • the inventive fiber laser-based system is configured to increase the productivity of silicon annealing processes including both standard ELA and SLS and greatly reduce the cost of manufacturing and operation of currently available annealing systems.
  • the inventive system is configured to output at least a 1 ⁇ wide harmonic pulsed beam at 3xx and 5xx mn pulsed beam, which is incident on a-Si panel of all known generations of Si panels. As a result a p-Si crystalline structure having uniform sub- ⁇ grains over the entire area of the panel to be treated is produced.
  • FIG. 1 illustrates a general layout of the inventive modular system 10 that can be utilized both in SLS and LA processes for annealing small and large a-Si panels.
  • inventive system 10 is based on a fiber laser source 12 which generates a substantially diffraction-limited laser pulsed beam propagating along a beam path through an optical schematic of beam delivery system which includes several sequentially located subassemblies.
  • the upstream sub-assembly 14 is configured to control divergence and beam size for further homogenization and scanning sub-assemblies 16.
  • the homogenization and scanning subassembly 16 is configured to control an intensity and velocity of the conditioned beam at a mask plane.
  • the following sub-assembly 18 is operative to image the conditioned beam on the a-Si surface with desired demagnification via a panel handling subassembly 20 which operates to provide different patterns of displacement between the panel to be annealed and the beam delivery system.
  • the inventive modular system 10 features several configurations of each of the listed above subassemblies as disclosed below in detail.
  • FIG. 2A illustrates modular system 10 mounted on a movable console 22 enclosing an IR pump laser, cooling system, control circuits and other peripheral components allowing the operation of fiber laser source 12 .
  • the system 10 is configured as Gantry machine having an inverted U-shaped bracket 24 and a base 26 which is mounted on console 22.
  • stage 20 supporting the panel to be laser treated guides the panel in X-Y planes, while beam imaging sub-assembly 18 remains stationary such that the surface of the workpiece is laser treated in accordance with a variety of topological patterns discussed below.
  • both the panel and beam-imaging sub-assembly 18 are each movable in respective planes. Still another configuration, which is particularly
  • the fiber laser source 12 is disclosed in co-pending and co-owned US '790 application. Briefly, laser source 12 is configured in addition to the laser pump which can be located in console 22. The fiber laser pump operates in a quasi-continuous regime (QCW) outputting a substantially diffraction limited, pump beam in a 1 ⁇ fundamental wavelength range at a kHz - MHz repetition rate range.
  • QCW quasi-continuous regime
  • the pump beam is coupled into a laser head 28 configured with a beam guiding optic and harmonic generator.
  • the harmonic beam at either about a 532 nm wavelength or about 355 nm wavelength is output for propagation along a beam path through sub-assemblies of system 10.
  • the single- or low-mode (SM) pump source 12 is operative to output packets of pulses at pocket repetition rate (PaR ) of up to 2 MHz and packet duration between 50 and 500 ns.
  • the pulses within each packet are output at a frequency varying from about 80 MHz and up to 200 MHz.
  • the source 12 is configured to output a pulsed beam with a substantially Gaussian intensity profile in bursts of pulses or continuous beam of pulses.
  • the latter in combination with at least 1 micron-wide line beam on target, controllable fluence, incremental stage velocity and other beam and system parameters create a p-Si structure having a uniform grain length at most equal to but preferably smaller than 1 micron and as small as 2 microns with a low coherence time of about 20-30 ps.
  • the output harmonic beam with M equal to 1 or close to 1 further propagates through beam conditioning sub-assembly 14.
  • the configuration of the latter is specifically tailored to the needs of either the homogenizing subassembly and/or scanning sub-assembly located downstream from sub-assembly 14.
  • system 10 shown in FIGs. 2-5 includes the homogenizing sub-assembly operative to convert a Gaussian intensity profile into a flat top or top hat profile
  • beam conditioning subassembly 14 is configured with upstream and downstream collimators.
  • the harmonic fiber laser beam at the above disclosed wavelengths is typically elliptical with a high aspect ratio and a gaussian intensity distribution in each of the long and short axes.
  • the upstream collimator component includes cylindrical lenses configured to provide the harmonic beam with the desired size in the short axis. It is nominally configured as Gallilean telescope with negative and positive cylinder lenses 30.
  • the downstream collimator including negative and positive cylindrical lenses 32, provides the harmonic beam with the desired size in the long axis direction. At this point the collimated harmonic beam is conditioned for further homogenization.
  • the elliptical (but a circular beam is not excluded) harmonic beam with the gaussian intensity profile is prone to interference when overlapped.
  • the coherence properties also eliminate diffractive homogenization solutions due to speckle related phenomena. Accordingly, the coherence should be somehow mitigated.
  • both, the SLS and LA processes require a line beam with sufficient homogeneity along the length of the line. Even though the SLS process is up to an order of magnitude less sensitive to homogeneity than the ELA annealing process, there must still be adequate homogeneity. The more homogenous the intensity profile, the more the process window is opened up for pulse to pulse energy variation and depth of focus.
  • the homogenizing subassembly 16 addresses the coherence problem and provides the line beam on target with a substantially uniform intensity distribution along the length of the beam line.
  • the short axis may be less than 5 wavelengths in size, and there are limits as to how homogenous this axis can be made with physically reasonable optical components, due to diffraction resolution (point spread function) limits. Except for very large beam widths (e.g. ⁇ or larger), there is not much advantage to be gained from the homogenization line width.
  • the homogenization is based on beam segmentation and recombining technique utilized with two standard techniques: fly's eye homogenizer, as used in system 10 of FIGs. 2-5, and bi-prisms.
  • fly's eye homogenizer as used in system 10 of FIGs. 2-5, and bi-prisms.
  • the principle of the segmentation and recombining technique is well known to one of ordinary skill. If the Gaussian-like laser beam, like the one generated in this inventive system, is split into multiple beams and then recombined after modifying the profile of one or more of the split beams, it is possible to generate a homogenous beam in at least one axis. Fly's eye homogenizers overlap multiple segments of a beam, and require beams with very low spatial coherence. Otherwise they suffer from severe speckle and other interference related phenomena.
  • Fiber lasers are usually not considered candidates for fly's eye homogenization due to their normally high coherence. Experiments have indicated, however, that the laser sources disclosed here and in US '790 have sufficiently short coherence times, such that sections of delay glass may be added to the individual segments to ensure that coherence time is exceeded at the overlap location. Fly's eye homogenizers are an option for these lasers. Accordingly, the disclosed fly's eye and bi-prism homogenizers successfully utilize the disclosed QCW lasers. [0055] The homogenizing sub-assembly 16 of FIGs. 2-5 provides the homogenized line beam at the plane of mask 54 (FIG. 4).
  • sub-assembly 16 may further have a one-dimensional mask 40 provided with such an aperture that only the desired, most homogeneous part of the
  • homogenized beam propagates further down the path.
  • the homogenized beam is then incident on a condenser lens 48.
  • the imaging and nonimaging fly's eye homogenizer both of which are incorporated in the inventive system.
  • the term "imaging” makes much sense, because the basic mechanism is the imaging of first array of lenslets 34 to the mask or object plane.
  • the imaging homogenizer also includes a second array of lenslets 44 in FIG. 3 located between the phase shifter 36, 38 and condenser lens 48.
  • the lenses of second array 44 and condenser lens 48 image the individual field diaphragms to the mask/object plane.
  • This type of the homogenizer requires the precise alignment of the two lens arrays, the condenser lens and the optical axis of the incident beam.
  • a non-imaging homogenizer consists of one, first lens array 34 followed by delay glasses 36, 38 and 42 and condenser lens 48. Similarly to the imaging homogenizer, lens array 34 splits the harmonic incident coherent beam into beamlets propagating through the phase shifter before they pass through condenser lens 48 and overlap at the homogenization plane located in the back focal plane of the condenser lens. The intensity pattern in the
  • homogenization plane is related to the spatial frequency spectrum generated by the lens array.
  • the lens array should distribute the light at equal intensities in the desired angular spectrum images.
  • the condensed homogeneous beam passes through beam imaging subassembly 18, which is configured to deliver the homogenized beam to the image plane on the surface of the panel with a desired demagnification factor.
  • the homogenized beam is first projected and focused onto mask 54 by a focusing lens 50 in a direction of the short axis of the beam.
  • the mask 54 is configured with blades cropping edges of the projected homogenized beam in the long axis direction.
  • This mask plane is then re-imaged onto objective lens 60 at some optimized demagnification with edge resolution limited by the numerical aperture of the imaging system.
  • the demagnification for example lOx or 3 Ox, reduces sensitivity to pointing stability.
  • An optional mask 56 receiving the line beam may be viewed as a vignetting aperture removing residual inhomogeneity of the previously sharpened edges.
  • the de-magnified line beam with the desired degree of uniformed intensity and width further impinges upon a turning mirror 58 before coupling into objective lens 60 and illuminating the a-Si glass panel which lies upon a stage 62.
  • the panels, particularly large panels, are hardly ideally flat.
  • the use of auto-focusing assembly 64 allows maintaining the desired focal distance between spherical or anamorphic objective lens 60 and thus control the desired uniformity range of the the beam intensity of the line beam on target.
  • the properties may include the inhomogeneity of diffraction efficiency, diffraction angle and polarization state of the diffracted light. If the measured parameter or parameters of MURA is/are beyond a predetermined range, then providing one or multiple feedback controlling loops allows readjustment of any of the above-disclosed subassemblies in real time.
  • Still another measuring sub-assembly is configured as a beam profilometer 70 position anywhere along the beam path downstream from homogenizing subassembly.
  • the beam profilometer 70 may be a camera-based beam profiling system configured with a camera and analysis software. Often times, this system needs to be used with beam attenuation or beam sizing accessories, depending on the task at hand.
  • the advantage to camera-based beam profiling is the real-time viewing and measuring of a laser's structure with high accuracy measurements. If the measured beam profile parameter or parameters are outside the predetermined range, the annealing process terminates and troubleshooting of the identified malfunctioning subassemblies begins.
  • homogenizing sub-assembly 16 of system 10 based on the segmentation and recombining technique includes bi-prisms.
  • the inventive bi-prism-based subassembly 16 is configured to impart an optical path difference between two beamlets.
  • the path difference should be such that delay is longer than the coherence length/time.
  • a preconditioned harmonic circular or elliptical Gaussian beam 80 propagates through a ⁇ /2 waveplate 82, configured to control splitting energies, and is further fragmented in an upstream polarizing beam splitter 72 directing two orthogonally polarized beamlets 84 and 86 along respective short and long paths.
  • the beamlet 86 is guided along a longer, delay path of sub-assembly 16 including multiple turning mirrors 88 a polarizing beam combiner 74.
  • the other beamlet 84 propagates through a single axis eversion prism 76 where it is everted and further coupled into polarizing beam combiner 74.
  • path 86 is lm longer than path 84, then there will be «3ns time difference between the two paths. For 1.5ns pulses at 150MHz repetition rate, this delay is sufficient to ensure that pulses within a burst from the two paths will arrive at separate times, and there will be no interference.
  • blue 86 and red 84 beamlets pass through respective long and short paths with a 3.3. ns delay without overlapping one another in time and avoiding interference.
  • the beamlets of the same polarization follow one another at 150 MHz
  • the interleaved beamlets are output at 300 MHz effective rep rate in a ⁇ 303 ns burst. It should be taken into consideration that if the time delay between two paths is shorter than the pulse duration but longer than the laser coherence tine, there will also be no interference.
  • FIGs. 8A-8B and 9A-9B illustrate another modification of homogenizing sub-assembly 16.
  • This type of homogenizer relates to the bi-prism segmentation and recombining of FIG. 6 and generally can be referred to as beam combining.
  • the time slice bi-prism homogenizer can be extended to beam combining homogenization techniques via the geometric overlapping of multiple beams and/or beam segments, as long as no two pulses within a pulse burst arrive at a specific location with temporal overlap.
  • FIGs. 8A and 8B illustrate four (4) beams 88, 90, 92 and 94 respectively which are overlapped and temporally interspersed to generate a central homogenous section 96.
  • FIG. 8B illustrates a 4 beam overlap in which beams 88 and 94 are simultaneous, whereas beams 90 and 92 are interspersed. As can be seen, the beams are not temporally or spatially overlapped. In the context of system 10, this homogenous section is cropped by a mask and then imaged onto the panel surface.
  • FIGs. 9A and 9B illustrate a 3 beam overlap with eversion.
  • the outside beam is everted such that the majority of the line beam is homogenous in the long axis.
  • Such a method allows for extended length of the line beams, and reduces the quantity of line stitching required for fiber laser beam used for annealing large panels as will be explained below. While more efficient than the cropping technique of FIGs. 8A-8B, this technique still has discarded laser power at the ends of the homogenized line beam.
  • the simplest method allowing a Gaussian beam profile to be converted to a top hat beam profile is to crop the center portion of the beam that is within the homogeneity requirements.
  • the cropping must be done at the mask (object) plane 54 of FIG. 4, which is then imaged to the combined object/process (image) plane via objective lens 60 of FIG. 5.
  • the objective lens 60 in this method may use anamoiphic cylindrical elements, with different demagnification in each axis, or spherically symmetric elements that provide constant demagnification. This method discards the majority of the beam energy.
  • the beam combining method is more efficient than simply cropping the central, homogeneous portion of a single beam, but still has the same problem.
  • FIGs. 10A - IOC illustrate respective beam combining configurations.
  • FIG.10A illustrates a system including a polarizing beam combiner receiving two linearly p-polarized and s-polarized input laser beams. The input beams propagate along respective legs of the shown system each of which may be provided with a 1 ⁇ 2 waveplate.
  • FIG. 10B illustrates a beam combining structure capable of angularly combining multiple beams that pass through a field lens capable of focusing these beams into a single beam waist.
  • FIG. IOC illustrates a beam combining arrangement in which multiple converging beams of different orders impinge upon a diffractive beam combiner configured to combine the passing beams into a single output beam.
  • beam homogenizing subassembly 16 is configured to remap the apodization of a Gaussian beam to a top hat apodization utilizing aspherical optical elements.
  • the best known example of this type is the piShaper.
  • the latter is a telescope where intensity profile is transformed in a controlled manner one of the basic principles is zero wave aberration of the entire system which distinguishes this type of homogenizer from, for example, the segmentation and recombining-based.
  • the SLS process only absolutely requires homogenization in the long axis.
  • the optimum top hat re-apodization will therefore be either one axis only, or anamorp c.
  • cylinder lenses are required.
  • the shown system includes, for example, an anamorphic crossed acylinder lens 98 designed to transform a circular beam to a 1mm long line beam, homogenized in each axis to the diffraction limited point spread. This lens directly transforms the beam intensity profile at focus.
  • This method can be used to illuminate a mask plane, which is then imaged onto the process plane via an anamorphic or spherically symmetric objective lens.
  • the use of a single anamorphic lens converting the beam profile in the long axis is required.
  • This method is alignment sensitive to input beam profile and divergence, as well as centration and angle of incidence. Any beam delivery that utilizes such a method must be capable of ensuring beam centration of the order of 10's of microns, and orthogonality of 10's of micro radians.
  • FIGs. 12A ?? 12B illustrate the general operating principle configurations implementing the scanning technique.
  • FIG. 11 A a constant beam of pulses irradiating a long line in the direction of arrow A. Every location along the line is exposed to, for example, the same top flat beam (FIG. 11C) such that at a certain point of time, initial portion 101 is fully crystallized.
  • the recently irradiated line stretch 103 is still in the process of crystallization and will be fully crystallized as the panel moves further along arrow A with line stretch 105 being currently irradiated and stretchl07 is yet to be irradiated.
  • the thermal profile of the line to be crystalized shown in FIG. 11B graphically illustrates the status of the above disclosed line stretches 101- 107.
  • the required line beam length and scanning velocity to achieve desired fluence and exposure time can be determined as follows:
  • the scanning fluence at any point is:
  • FIGs. 13A - 13D conceptually illustrate the scanning configuration.
  • the laser source outputs the harmonic beam passing through the beam conditioning subassembly which generates the desired beam profile in the mask plane or directly at the image plane after the scanning assembly.
  • the desired profile may be Gaussian in both axes 104, Gaussian/Quasi Gaussian in the direction of one axis and top hat in the direction of the other axis 106 and the top hat profile in the direction of both axes as indicated by 108. If the top hat profile is desired, then the homogenizing subassembly is required, and its configuration depends on laser beam characteristics and desired profile.
  • the beam is incident on the scanner and scanning optics before it passes through the mask plane or directly at the image plane panel obtaining the desired beam profile.
  • FIGs. 14A -14E illustrate one of possible configurations of beam delivery and scanning subassembly 18 provided with a rotating mirror, such as a monogon or a multifaceted polygon 100.
  • scanner 100 may be mounted to the Gantry machine instead of turning mirror 46 and used with or without the homogenizer to illuminate the mask.
  • the facets can be oriented at any angle, such as 45 0 or 90°, and deflect the harmonic beam along the light path towards a telecentric, cylindrical Galilean expander shown in FIGs. 14B-14E.
  • the expanded beam then passes through short axis field lens 50, mask 54 and turning mirror 8 before coupling into objective lens 60.
  • AOD optical solid state deflectors relying on the acousto-optic effect and referred further to as AOD.
  • the AODs do not contain moving parts and therefore exhibit high deflection velocities and more reliable than mechanical scanners.
  • the AOD is based on a periodically changing refractive index in an optically transparent material, induced by propagating sound waves in the material.
  • the changing refractive index is the result of refraction and compression of the material, inducing a changing density of the material.
  • This periodically changing refractive index acts like an optical grating, moving at the speed of sound in the crystal that will diffract a laser beam traveling through the material.
  • the panels to be annealed with line beams are orders of magnitude larger than the length of the line beams that can be achieved with individual burst mode fiber lasers. Accordingly, it is necessary to stitch together beams from individual or multiple lasers to achieve effectively continuous poly-Si grain structures with grain size smaller than 1 micron and as small as 0.2 micron.
  • Excimer lasers used for ELA processes have high energy per pulse with low repetition rate. This makes them suitable for single line annealing of large panels, wherein the line beam encompasses the entire panel width.
  • Burst mode fiber lasers with equivalent total power output have orders of magnitude lower pulse energy and requisite orders of magnitude higher repetition rate.
  • the pulse energy is too low to allow for a line beam from a single laser to encompass the entire panel width.
  • the region of interest may be the entire panel area, or portions thereof.
  • Another possibility is to tailor the intensity profiles of the ends of the beams such that when adjacent beams overlap the effective intensity remains homogenous. Even though there are coherence/speckle issues with overlapping coherent beams, it is possible to intersperse the adjacent beams such that the individual pulses in each burst do not arrive at the same time. [0085]
  • the following disclosure covers various methods of stitching beam lines for continuous poly-Si coverage. Controlling the beam delivery and panel handling subassemblies, is possible to provide a variety of line patterns and ways by which individual line beams can be stitched to encompass the regions of interests.
  • continuous line with 1-D scanning is conceptually the simplest method used to align individual beams such that they constitute an effectively continuous line, somewhat analogous to the LA continuous line beam. This effective line beam would then be scanned in one (1) dimension along the entire panel as shown in FIG. 16A.
  • FIGs. 15A and 15B show respective configurations that mitigate the tight packing issues.
  • fold mirrors 110 By placing fold mirrors 110 near the waists of the individual beams, i.e., close to the image plane, where there is sufficient free space between beams. The the individual
  • objective/projection lens or lenses 60 in case of multiple laser beams, can be sequentially placed in different orientations to free up space for fixturing of these lenses.
  • the two illustrated arrangements on respective FIGs. 15A and 15B show alternate direction fold, and three direction fold stacking of individual line beams. This basic concept can be extended to other configurations to achieve the same goal. Challenges with generating a continuous line are primarily related to alignment, coherent interference, autofocus and mechanical/thermal stability.
  • the individual line beams must be carefully aligned, in both position and angle, such that the resultant continuous line beam is sufficiently homogenous in both intensity and profile to support the annealing process.
  • the individual beams must be aligned relative to each other to «0.1 ⁇ tolerances along the full continuous crystallization line length.
  • burse repetition rate is too high for crystallization to occur with a single pass, placing every second burst on a first pass and then filling in the remaining bursts on a second pass may be required.
  • a further modification of continuous line includes multiple passes as shown in FIG. 16B. If the length of the continuous line is limited to a portion of the panel width, then the entire panel could be covered by using several passes, with the line beam offset between passes. If the line is 1 ⁇ 2 the panel width, then 2 passes would be required. 1/3 the panel width would require 3 passes, and so on. While this approach eases some of the challenges associated with a full panel width line beam, there are challenges associated with maintaining continuity of poly-Si grain structure between passes. The burst in consecutive passes must be aligned in both axes to «0. ⁇ accuracy. Again, if the repetition rate is too high, physical separation of sequential bursts will be required.
  • 16B and 17A-B may be combined such that the separate line beams of of FIG. 17 are each composed of a number of continuously stitched individual line beams of FIG. 16.
  • the requirement to physically separate sequential busts may arise due to repetition rate and/or interdigitation constraints. There are two basic methods to physically separate sequential pulses.
  • One technique includes multiple passes with increased scan speed.
  • a single line beam is scanned in the short axis direction with respect to the panel at sufficient speed such that the individual pulses are physically separated to generate every second line beam.
  • An ensuing pass is offset in the short axis direction such the missing alternate line beams are placed.
  • This above technique is the most conceptually simple compare to the other technique requiring line beam spacing, but it also requires doubling the scan speed as compared to continuously placing the line beams in a single pass. For example, if the laser is operating at 1MHz, and the line beams must be spaced every 2 ⁇ , then a single pass would require a scanning speed of 2m/s, while a double pass would require 4m/s. These are extremely high speeds at which to scan beams over large panels with 0.1 ⁇ level accuracy.
  • the pulse bursts are sequentially separated into different beam lines, repeating at a fixed modulus value, such that the effective repetition rate for an individual beam line is reduced by the modulus.
  • the previous example requiring a single beam line at 2m/s would, for example, be modified to 4 beam lines at 500mm/s if the modulus is 4.
  • Interdigitation also has the added advantage of separating the placement in time of beams that are physically adjacent on the panel, and thereby reducing or eliminating the implications of inadequate crystallization time between pulses due to high repetition rate.
  • interdigitation can be achieved by placing each offset with individual passes. For example, if two-beam interdigitation is desired as shown in FIG. 19B, then the scan speed would be such that the pulse spacing is exactly 2 line widths. The beams would then be stepped with respect to the panel by one half the beam length. The second pass would place pulses exactly between the pulses of the first pass. The beam would then be stepped again by one half the beam length, and the third pass would place the beams exactly aligned with the beams of the first pass, as shown in FIG. 18C.
  • FIG. 20 A - 20D and 21 A - 21 C rather than placing all the pulses individually, one pass at a time, it is possible to place multiple beams simultaneously, such that all required Interdigitation takes place within a single pass.
  • This figure illustrates single pass 2- beam and 3-beam interdigitation. The beams are placed with respect to each other such that, during a single scan, multiple beams are interdigitated. As with the mode for individual pulse placement, the number of interdigitated beams is only limited by the practicality of
  • this method reduces the effective repetition rate in terms of crystallization time linearly with the number of interdigitated pulses. Not only does this method reduce the number of passes required to fill in the
  • the beam stitching methods can be implemented in combinations to achieve the required p-Si grain characteristics within the constraints of laser and system specifications. Furthermore, each of the disclosed techniques, and combinations thereof, can be implemented with multiple, coordinated laser sources, and also in conjunction with the pulse picking methods described later.
  • the pulse picking technique can be implemented in different ways.
  • One of the ways includes directing the pulse picked beams into individual beam lines which are then scanned similarly to a single line. This can be realized using two approaches - continuous line and interdigitated line beams - and has the advantage of reducing the effective repetition rate of the individual beam lines, but requires multiple beam deliveries per laser.
  • the first approach allows the pulse picked beams to be delayed with respect to each other and combined, to form a single long homogenous continuous line that can be passed through a single beam delivery. This results in a line beam analogous to one that can be achieved with a lower repetition rate, higher pulse energy laser.
  • the alternative approach relates to the interdigitation and includes using the pulse picked beams to generate multiple pass
  • Another way of pulse picking includes long axis scanning in conjunction with a rotating scanner, such as polygon 100 shown in FIG. 23 or AOD.
  • the redirected beam is incident on a proximity mask where knife edges define the length of the beam irradiating the surface.
  • the beam's linewidth is formed using the above disclosed short axis mask with the degree of demagnification different from that of the proximity mask.
  • consecutive bursts of pulses produce adjacent, closely positioned lines separated along the long axis and stitched in a continuous longer line at effectively reduced burst rep rate with respect to the short axis, of the This is different from the line dragging technique since the latter is characterized by the continuity and obviously does not require stitching.
  • poiycrystaliine grain structure may cause visible Mura in the finished display, especially for stitched lines wherein the patterning will be periodic.
  • One mitigation technique is to dither the line beam onto the panel, such that inhomogeneous regions (including stitching) are at different locations in sequential lines, to effectively smooth out the inhomogeneities and reduce the resultant Mura to acceptable levels.
  • a dithering system may include a rotating wedge or diffusing element placed into the beam path. This can be very effective for removing speckle in coherent beams, but may be difficult for use in the SLS annealing process because rotating elements smear the beam in two dimensions, and SLS annealing requires a very narrow line beam. Any
  • dithering/smearing must be in the direction of the long axis of the line beam only.
  • the dithering may be periodic or stochastic in nature. Periodic dithering will follow a periodic profile, such as a saw tooth or sinusoid. Stochastic dithering will result in randomized (or pseudo-randomized) distributions of the inhomogeneities. The preferred type of dithering will depend on the dithering method and acceptable levels of Mura. Stochastic dithering is expected to be more effective in Mura reduction, whereas periodic dithering may be less complicated/costly to implement while achieving acceptable Mura.
  • the magnitude and periodicity (if not stochastic) of the dithering will depend on the types of inhomogeneities, whether or not there is stitching, and what levels of Mura are acceptable.
  • the magnitudes and periodicities could range from the order of ten microns to, potentially, greater than one millimeter.
  • One dimensional dithering can be achieved by several alternative subassemblies shown in FIGs. 24-28
  • FIG. 24 illustrates a dithering system 120 operative oscillate the panel or any suitable component after the mask.
  • the panel is oscillated/vibrated in the direction of the line beam during the SLS annealing process.
  • this can be either a periodic or stochastic oscillation, as illustrated in FIG. 27.
  • periodic oscillation wherein consecutive passes follow the same periodic path. This will both ensure constant overlap of the stitched lines, and smooth the overlap region along the scan path, if stochastic dithering is used with stitched lines, then the overlap between consecutive scans will be highly variable. This may not be acceptable for many types of device, particularly OLED devices.
  • FIG. 25 shows another configuration of dithering system 120 oscillate an optical component of the beam delivery system, such as a lens or mirror, which is oscillated/vibrated in the direction of the line beam during the SLS annealing process, to produce the requisite inhomogeneity smoothing, while maintaining sharp definition of the line beam.
  • the dithering is introduced into the beam ahead of a mask plane, then the mask may be used to maintain a straight path of the line beam edge on the panel. In this case, either periodic or stochastic dithering will be equally applicable. If the dithering is introduced after the mask, then the resultant line distributions will be equivalent to oscillating the panel and the same arguments for periodic vs. stochastic dithering hold as illustrated in FIG. 27.
  • FIG. 26 illustrates still another configuration of dithering system 120 oscillating the mask.
  • the mask is oscillated/vibrated in the direction of the line beam during the SLS annealing process. This will smooth out the stitching zone between consecutive passes, but will not smooth out inhomogeneities within the line beam.
  • periodic vs. stochastic dithering hold as for panel oscillation.

Abstract

A system for crystallizing amorphous Si (a-Si) panels by partial melt laser annealing (LA) or sequential lateral sol idi ficat ions (SLS) annealing process is provided. The system comprises at least one single transverse mode (SM) quasi -cont inuous wave (QCW) fiber laser source, emitting a pulsed harmonic beam; a beam conditioning assembly located downstream from the fiber laser source and configured to transform the harmonic beam such that the harmonic beam has desired divergence and spatial distribution characteristics; a beam velocity and profile assembly operative to provide the conditioned harmonic beam with a desired intensity profile at a desired scanning velocity in an object plane; a beam imaging assembly for imaging the conditioned harmonic beam in the object plane onto an image plane in at least one beam axis at a desired demagni ficat ion such that a width of the conditioned harmonic beam is reduced to a narrow linewidth; and a panel handling assembly operative to provide relative position and velocity between the imaged narrow linewidth beam and the panel such as to irradiate each location of the a-Si panel at least two times with an exposure time.

Description

FIBER LASER-BASED SYSTEM FOR UNIFORM CRYSTALLIZATION
OF AMORPHOUS SILICON SUBSTRATE
BACKGROUND OF THE DISCLOSURE
Field of the Invention
[0001] This disclosure relates to fiber laser-based method and system for manufacturing flat panel displays. Particularly, the disclosure relates to a fiber laser system for annealing amorphous silicon panel by a harmonic laser beam so as to fabricate substantially uniform polycrystalline silicon displays, and a method of operating the inventive system.
Prior Art Discussion
[0002] The Flat Panel Display (FPD) fabrication environment is among the world's most competitive and complex technologies. Generally, this technology is in constant need for compact products equipped with the FPD and featuring high resolution, bright large display, low power consumption and fast video capability and, of course, low cost.
[0003] The thin film transistor (TFT) technology is the basis for the FPD fabrication that can be either high-resolution, high-performance liquid crystal display (LCD), or organic light emitting diode (OLED). At the dawn of the TFT technology display circuits were made on a thin opaque layer of amorphous silicon ("a-silicon or a-Si") and arranged in a backplane across the layer to correspond to respective pixels. Each pixel includes a great number of crystalline grains which are each formed with a specific grain area and orientation.
[0004] Later a-Si was at least partially replaced by poly-silicon (p-Si) which has the carrier mobility approximately two orders of magnitude greater than that of a-Si, substantially improves the aperture ratio, pixel resolution, and reduces the pixel size. As a result of these properties of poly-Si, portable/mobile electronic devices now feature high resolution flat panel small displays.
[0005] There are two fundamentally different approaches for converting the a-Si into poly-Si through crystallization (annealing). One is a thermal annealing (TA) approach, and the other is a low-temperature poly-silicon annealing (LTPS) approach, which is of particular interest here. In the latter, a-Si is initially thermally treated to convert into liquid amorphous Si, and then it is maintained in the molten state for a certain period of time. The temperature range sufficient to maintain the molten state is selected to allow the initially formed poly-crystallites to grow or crystallize. After the crystallization step is complete, the surface is cooled to induce a solidification phase of the processed material. The LTPS approach is based on two generic methods - Excimer Laser Annealing (ELA) or partial melt LA and sequential lateral
solidification (SLS), with the latter being of particular interest here and also the subject matter of U.S. Pat. Appl. 14/790,170 (US '790) co-owned with and fully incorporated in this application by reference.
[0006] In contrast to the traditional ELA, the SLS method includes melting the entire thickness of the a-Si film by a beam from an excimer laser that operates at a 3xx nm wavelength. As a result, crystallization fronts grow from opposite sides of the molten film. In other words, the growth is lateral. The laterally developed crystalline grains can be elongated to large horizontal dimensions. The latter is advantageous since electron mobility increases as grains grow larger. However, despite advantages and disadvantages of each method relative to the other method, both SLA and partial annealing (EL) methods are viable options.
[0007] Historically, excimer lasers, used in both LA and SLS processes, dominate annealing of TFT flat panel displays. Excimer lasers provide a wide range of processing power, with average range of processing powers up to 300 W and higher, energies higher than 1 J and pulse width from 10 to 250 ns. Also, excimers deliver UV light at the wavelength 3xx nm, which is directly absorbed in a-Si without additional frequency conversion.
[0008] The pulse frequency of the excimer laser is relatively low. To the best of applicants' knowledge, it does not exceed 6 kHz in SLS processes and considerably lower in standard ELA . Such a relatively low frequency in the standard ELA may explain the necessity of multiple irradiations of each location since the duration between subsequent pulses is greater than the time constant (of the process) beyond which excited atoms lose their mobility. As to the SLS, with KHz frequencies leading to high energies, the excimer requires multiple gas changes over a daylong period of operation which makes it unsuitable for mass production. Many of the above disclosed and other disadvantages of the excimer lasers are not characteristic to fiber lasers. [0009] A need therefore exists for using a fiber laser-based annealing system operable to provide a substantially uniform p-crystalline structure (p-Si) by using both SLS and partial anneal (LA) methods with a narrow width laser beam.
SUMMARY OF THE DISCLOSURE
[0010] The SLS processes require a line beam on target, wherein the long axis is orders of magnitude larger than the short axis (e.g. 2mm short (∞ 400: 1)). Primary scanning is typically conducted in the direction of the short axis, but long axis scanning is not precluded.
[0011] The dimensions of the line beam are effectively driven by 3 parameters: desired width of the short axis (function of desired poly-Si grain size and short axis intensity profile), desired fluence (J/cm2) on target, and pulse energy reaching the target. For example, a 5μηι wide line with ΙΟΟμΙ pulse energy and 1 J/cm2 required fluence would allow for a line beam length of «2mm. The specific relationships are dependent on the intensity profile of the short axis. A top hat short axis profile, for example, would allow for a longer line beam than a Gaussian short axis profile with all other parameters being the same. Nevertheless, the above values are indicative of the relative relationships between the various parameters under consideration.
[0012] The panels to be annealed with said line beams are orders of magnitude larger than the length of the line beams that can be achieved in SLS with individual burst mode fiber lasers. It will be necessary to stitch together beams from individual or multiple lasers to achieve effectively continuous poly-Si grain structures.
[0013] The partial LA process requires considerably stronger requirements for fluence and higher average powers. However, partially because of beam width above 5 micron which is greater than a 1 to 2 micron linewidth in SLS, the stitching problem does not exist.
[001 ] The inventive system addressing all of the above and other issues, which are
characteristic to both SLS and LA processes of both SLS, allows implementation of the selected methods for annealing a large panel. Aspects of the disclosed system include scanning/stepping in each axis, multiple laser compatibility, requisite speeds and accuracies, auto-focus
compatibility, thermal management, active alignment and thermal management which are summarized immediately below.
[0015] In accordance with one aspect of the disclosure, the disclosed system for crystallizing amorphous Si (a-Si) panels by partial melt laser annealing (LA) or sequential lateral solidification (SLS) annealing process includes at least one single transverse mode (SM) quasi- continuous wave (QCW) fiber laser source, emitting a pulsed harmonic beam at a pulse repetition rate of at least 80 MHz, along a path, The system further has a delivery system including a beam conditioning assembly located downstream from the fiber laser source and configured to transform the harmonic beam such that the harmonic beam has desired divergence and spatial distribution characteristics. The delivery system further has a beam velocity and profile assembly operative to provide the conditioned harmonic beam with a desired intensity profile at a desired scanning velocity in an object plane. Also, the delivery system a beam imaging assembly for imaging the conditioned harmonic beam in the object plane onto an image plane in at least one beam axis at a desired demagnification such that a width of the conditioned harmonic beam is reduced to a narrow linewidth of at least 1 μιη at the image plane. The system is provided with a panel handling assembly operative to provide relative position and velocity between the imaged narrow linewidth beam and the panel such as to irradiate each location of the a-Si panel at least two times with an exposure time during each time of at least 100 ns in order to provide transformation of a-Si to polysilicon (p-Si) structure with a uniform grain size of at most 1 μΐΆ.
[0016] In accordance with another aspect, the system described in the 1 aspect further has multiple SM QCW fiber laser sources emitting respective beam that are combinable together.
[0017] In another aspect, the system of any of the above aspects is provided with the beam velocity and profile system which is configured to convert a high-ratio Gaussian harmonic beam into a flat-top harmonic and selected from a beam segmentation and recombination system, beam re-apodization system, beam combining system, and beam cropping system.
[0018] Another aspect of the inventive system of any of the above aspects concerns the beam segmentation and recombining system which is selected from a fly's eye or bi-prism optical arrangements.
[0019] In another aspect of the inventive system disclosed in any of the above-discussed aspects, the fly's eye is imaging or non-imaging homogenizer.
[0020] Another aspect of the inventive system of any of the above aspects relates to the beam combining system configured to overlap and intersperse multiple harmonic beams to generate a central homogeneous central top portion of an intensity profile which is cropped at the object plane. [0021] In still another aspect of the inventive system of any of the above-disclosed aspects the beam combining system is operative to overlap multiple harmonic beams. One of the overlapped beams is everted so that a resulting harmonic beam is homogeneous in a direction of long axis.
[0022] The system of any of the above aspects in accordance with another aspect in which the beam combining system is configured with a polarizing beam combiner or field lens combiner or diffractive beam combiner or fly's eye.
[0023] The system of any of the above aspects and following aspects, wherein the beam re- apodization system is configured with at least one or multiple acylindrical optical elements converting a high-ratio Gaussian harmonic beam into a flat-top intensity profile in at least of beam axes.
[0024] The system of any of the above and following aspects in which the beam velocity and profile system is configured with a scanner operative to provide the conditioned harmonic beam with a desired velocity so that the imaged narrow linewidth beam homogeneously and continuously generate a line of crystallization without stitching.
[0025] The system of each of the above and follow aspects in which the scanner is selected from a rotating mirror or acousto optical deflector or galvanometer.
[0026] In a further aspect of the system disclosed of any of the above and following aspects the beam imaging assembly is configured with a focusing lens focusing the conditioned harmonic beam with the desired intensity profile at the desired scanning velocity in a short beam axis direction onto a first mask, The latter defines the object plane and has cutting knives for sharpening edges of the beam in a long axis direction, and an objective lens located downstream from the first mask and adjacent to the panel.
[0027] In accordance with the system of any of the above and following aspects, the beam imaging assembly further includes a second mask located between the first mask and objective lens and configured to vignette residual inhomogeneity of the line of crystallization.
[0028] According to another aspect of the system described in any the above and following aspects, the beam imaging assembly is configured with an anamorphic lens arrangement providing the different desired demagnification along orthogonal beam axes of the conditioned harmonic beam with the desired intensity profile. [0029] The current aspect of the system of any of the above and following aspects relates to the beam imaging assembly which is anamorphic and includes two spaced masks providing different demagnification in respective orthogonal beam axes and having different object planes.
[0030] Yet another aspect of the inventive of any of the above and following aspects concerns the beam imaging assembly which has a proximity mask configured to define a desired length of a beamline on the panel, the proximity mask is spaced from the panel at a distance limiting edge diffraction.
[0031] The system of any of the above and following aspects also includes the panel handling assembly with a support supporting the panel to be annealed such that the panel is displaceable in orthogonal XY planes relative to the fixed beam imaging assembly.
[0032] Still another aspect of the inventive system of any the above and following aspects relates to the panel handling assembly including a support supporting the panel to be annealed and configured such that the panel is stationary relative to the beam imaging assembly displaceable in orthogonal XY planes.
[0033] Still another aspect of the inventive system disclosed in any of the above and below aspects, the panel handling assembly is configured with a support supporting the panel to be annealed such that the panel is displaceable on one of XY planes and the beam imaging system is displaceable on another of XY planes.
[0034] The system of this aspect related to any of the above and following aspect is configured with the SM QCW fiber laser source which is mounted in a fixed position relative to the displaceable beam image system or displaceable therewith.
[0035] The system of any of the above aspect and following aspects further includes an auto focus system, beam profiler and MURA measuring system.
[0036] In this aspect the system of any of the above and following aspect Is configured with the fly's eye homogenizer having delay step glass elements to eliminate coherence effects.
[0037] Still another aspect of the inventive system of any of the above and following aspects includes a dithering system operative to dither the narrow linewidth beam onto the panel such that residua] iiihomogeneous regions of the p-Si structure are at different locations in sequential lines effectively smoothing out the residual inhomogeneities and reducing the Mura to a predetemiine reference range. [0038] Yet another aspect of the disclosed system defined in any of the above or following aspects 24 relates to the dithering system which is operative to oscillate the panel or any suitable component downstream form the object plane, or an optical component of the beam delivery system, such as a lens or mirror, or an optical component along the beam path in the direction of the conditioned harmonic narrow width line beam during the SLS annealing process, or the mask defining the object plane.
[0039] I accordance with still another aspect of the invention disclosed in any of the above discussed aspect, the system is configured to generate a continuous line in both SLS and AL processes with one or multiple passes.
[0040] Yet in another aspect of the disclosure as discussed in any of the above aspects, the inventive system is operative provide interdigitation of multiple beams after two or more passes.
[0041] In yet another aspect, the inventive system of any of the above and following aspects is configured to provide pulse picking utilizing a mechanical scanning, acousto-optic or electro- optic method.
BRIEF DESCRIPTION OF THE DRAWINGS
[0042] The above and other aspects of the disclosed system will become more readily apparent from the following specific description in conjunction with drawings, in which:
FIG. 1 is a general highly schematic view of the inventive system;
FIG. 2A is one modification of the inventive system of FIG. 1;
FIG. 2A is another embodiment of the inventive system of FIG. 1;
FIG. 3 is a view of beam conditioning/homogenizing and beam profiling subassemblies of the inventive systems of FIG. 2 A and 2B;
FIG. 4 is a view of beam imaging subassembly of FIGs. 2A and 2B;
FIG. 5 is a view of panel handling sub-assembly of FIG. 1 ;
FIG. 6 is an optical schematic of the beam homogenizing subassembly with a beam polarizing combiner;
FIG. 7 is a graphical representation of principle of operation of the homogenizing subassembly of FIG. 6;
FIGs. 8A through 9B are respective further graphical representation of principle of operation of the homogenizing subassembly of FIG. 6; FIGs. 1 OA- IOC are respective arrangement operating for combining multiple beams;
FIGs. 11A-11C are respective orthogonal, side and top views of another configuration of the beam homogenizing subassembly;
FIGs. 12A - 12C illustrate a thermal transformation of a single line to be crystallized by a laser beam;
FIGs. 12A illustrates a general principle of operation of scanning subassembly;
FIGs. 13B-13D illustrate a harmonic laser beam with respective different intensity profiles.
FIGs. 14A -14C is respective views of a polygon scanner and system utilizing the polygon.
FIGs. 15A - 15B;
FIGs. 16A-16B;
FIGs, 17A-17B;
FIGs 18A-18B;
FIGs. 19A-19B;
FIGs20A-20D;
FIGs. 21A-21C
FIG. 22; and
FIG. 23 FIGs. 24-27.
SPECIFIC DESCRIPTION
[0043] Reference will now be made in detail to embodiments of the invention. Wherever possible, same or similar numerals are used in the drawings and the description to refer to the same or like parts or steps. Some of the drawings are in simplified form and not to precise scale. Unless specifically noted, it is intended that the words and phrases in the specification and claims be given the ordinary and accustomed meaning to those of ordinary skill in the diode and fiber laser arts. The word "couple" and similar terms do not necessarily denote direct and immediate connections, but also include mechanical and optical connections through free space or intermediate elements.
[0044] The inventive fiber laser-based system is configured to increase the productivity of silicon annealing processes including both standard ELA and SLS and greatly reduce the cost of manufacturing and operation of currently available annealing systems. In the SLS context, the inventive system is configured to output at least a 1 μηι wide harmonic pulsed beam at 3xx and 5xx mn pulsed beam, which is incident on a-Si panel of all known generations of Si panels. As a result a p-Si crystalline structure having uniform sub-μηι grains over the entire area of the panel to be treated is produced.
[0045] FIG. 1 illustrates a general layout of the inventive modular system 10 that can be utilized both in SLS and LA processes for annealing small and large a-Si panels. Regardless of a particular annealing method, inventive system 10 is based on a fiber laser source 12 which generates a substantially diffraction-limited laser pulsed beam propagating along a beam path through an optical schematic of beam delivery system which includes several sequentially located subassemblies. The upstream sub-assembly 14 is configured to control divergence and beam size for further homogenization and scanning sub-assemblies 16. The homogenization and scanning subassembly 16 is configured to control an intensity and velocity of the conditioned beam at a mask plane. The following sub-assembly 18 is operative to image the conditioned beam on the a-Si surface with desired demagnification via a panel handling subassembly 20 which operates to provide different patterns of displacement between the panel to be annealed and the beam delivery system. The inventive modular system 10 features several configurations of each of the listed above subassemblies as disclosed below in detail.
[0046] FIG. 2A illustrates modular system 10 mounted on a movable console 22 enclosing an IR pump laser, cooling system, control circuits and other peripheral components allowing the operation of fiber laser source 12 . The system 10 is configured as Gantry machine having an inverted U-shaped bracket 24 and a base 26 which is mounted on console 22. In the illustrated configuration, stage 20 supporting the panel to be laser treated guides the panel in X-Y planes, while beam imaging sub-assembly 18 remains stationary such that the surface of the workpiece is laser treated in accordance with a variety of topological patterns discussed below.
Alternatively, as illustrated in FIGs. 2B both the panel and beam-imaging sub-assembly 18 are each movable in respective planes. Still another configuration, which is particularly
advantageous for generation 6 to generation 8 panels, allows the panel to remain stationary, while the beam imaging sub-assembly moves in X-Y planes. The inventive system has an optical schematic, which in turn may have a variety of configurations operating with all possible modifications of the displacement between the panel and beam delivery subassembly practically without any structural alterations. [0047] The fiber laser source 12 is disclosed in co-pending and co-owned US '790 application. Briefly, laser source 12 is configured in addition to the laser pump which can be located in console 22. The fiber laser pump operates in a quasi-continuous regime (QCW) outputting a substantially diffraction limited, pump beam in a 1 μη fundamental wavelength range at a kHz - MHz repetition rate range. The pump beam is coupled into a laser head 28 configured with a beam guiding optic and harmonic generator. Depending on the number of frequency/wavelength converting stages, the harmonic beam at either about a 532 nm wavelength or about 355 nm wavelength is output for propagation along a beam path through sub-assemblies of system 10. The single- or low-mode (SM) pump source 12 is operative to output packets of pulses at pocket repetition rate (PaR ) of up to 2 MHz and packet duration between 50 and 500 ns. The pulses within each packet are output at a frequency varying from about 80 MHz and up to 200 MHz. The source 12 is configured to output a pulsed beam with a substantially Gaussian intensity profile in bursts of pulses or continuous beam of pulses. The latter in combination with at least 1 micron-wide line beam on target, controllable fluence, incremental stage velocity and other beam and system parameters create a p-Si structure having a uniform grain length at most equal to but preferably smaller than 1 micron and as small as 2 microns with a low coherence time of about 20-30 ps.
[0048] Referring to FIG. 3 in combination with FIG. 2, the output harmonic beam with M equal to 1 or close to 1 further propagates through beam conditioning sub-assembly 14. The configuration of the latter is specifically tailored to the needs of either the homogenizing subassembly and/or scanning sub-assembly located downstream from sub-assembly 14.
[0049] Since system 10 shown in FIGs. 2-5 includes the homogenizing sub-assembly operative to convert a Gaussian intensity profile into a flat top or top hat profile, beam conditioning subassembly 14 is configured with upstream and downstream collimators. The harmonic fiber laser beam at the above disclosed wavelengths is typically elliptical with a high aspect ratio and a gaussian intensity distribution in each of the long and short axes. Hence the upstream collimator component includes cylindrical lenses configured to provide the harmonic beam with the desired size in the short axis. It is nominally configured as Gallilean telescope with negative and positive cylinder lenses 30. Similarly, the downstream collimator, including negative and positive cylindrical lenses 32, provides the harmonic beam with the desired size in the long axis direction. At this point the collimated harmonic beam is conditioned for further homogenization. [0050] The elliptical (but a circular beam is not excluded) harmonic beam with the gaussian intensity profile is prone to interference when overlapped. The coherence properties also eliminate diffractive homogenization solutions due to speckle related phenomena. Accordingly, the coherence should be somehow mitigated.
[0051] Both, the SLS and LA processes require a line beam with sufficient homogeneity along the length of the line. Even though the SLS process is up to an order of magnitude less sensitive to homogeneity than the ELA annealing process, there must still be adequate homogeneity. The more homogenous the intensity profile, the more the process window is opened up for pulse to pulse energy variation and depth of focus.
[0052] The homogenizing subassembly 16 addresses the coherence problem and provides the line beam on target with a substantially uniform intensity distribution along the length of the beam line. The short axis may be less than 5 wavelengths in size, and there are limits as to how homogenous this axis can be made with physically reasonable optical components, due to diffraction resolution (point spread function) limits. Except for very large beam widths (e.g. ΙΟμιη or larger), there is not much advantage to be gained from the homogenization line width.
[0053] Conceptually, the homogenization is based on beam segmentation and recombining technique utilized with two standard techniques: fly's eye homogenizer, as used in system 10 of FIGs. 2-5, and bi-prisms. The principle of the segmentation and recombining technique is well known to one of ordinary skill. If the Gaussian-like laser beam, like the one generated in this inventive system, is split into multiple beams and then recombined after modifying the profile of one or more of the split beams, it is possible to generate a homogenous beam in at least one axis. Fly's eye homogenizers overlap multiple segments of a beam, and require beams with very low spatial coherence. Otherwise they suffer from severe speckle and other interference related phenomena.
[0054] Fiber lasers are usually not considered candidates for fly's eye homogenization due to their normally high coherence. Experiments have indicated, however, that the laser sources disclosed here and in US '790 have sufficiently short coherence times, such that sections of delay glass may be added to the individual segments to ensure that coherence time is exceeded at the overlap location. Fly's eye homogenizers are an option for these lasers. Accordingly, the disclosed fly's eye and bi-prism homogenizers successfully utilize the disclosed QCW lasers. [0055] The homogenizing sub-assembly 16 of FIGs. 2-5 provides the homogenized line beam at the plane of mask 54 (FIG. 4). It includes a first array of cylinder lenslets 34 splitting the incident Gaussian beam into respective beamlets. The beamlets then impinge upon respective delay glasses 36, 38, 42 which as a unit provide an optical delay among these beamlets so as to minimize the interference and thus mitigate the effects of laser beam coherence which leads to the formation of a flat-top intensity profile. The flat-top beam may not have an ideally uniform intensity. Therefore, optionally, sub-assembly 16 may further have a one-dimensional mask 40 provided with such an aperture that only the desired, most homogeneous part of the
homogenized beam propagates further down the path. The homogenized beam is then incident on a condenser lens 48.
[0056] There are two main types of flu's eye homogenizer: the imaging and nonimaging fly's eye homogenizer both of which are incorporated in the inventive system. The term "imaging" makes much sense, because the basic mechanism is the imaging of first array of lenslets 34 to the mask or object plane. Structurally, the imaging homogenizer also includes a second array of lenslets 44 in FIG. 3 located between the phase shifter 36, 38 and condenser lens 48. The lenses of second array 44 and condenser lens 48 image the individual field diaphragms to the mask/object plane. This type of the homogenizer requires the precise alignment of the two lens arrays, the condenser lens and the optical axis of the incident beam.
[0057] A less complicated version of an optical integrator is referred as non-imaging
homogenizer. A non-imaging homogenizer consists of one, first lens array 34 followed by delay glasses 36, 38 and 42 and condenser lens 48. Similarly to the imaging homogenizer, lens array 34 splits the harmonic incident coherent beam into beamlets propagating through the phase shifter before they pass through condenser lens 48 and overlap at the homogenization plane located in the back focal plane of the condenser lens. The intensity pattern in the
homogenization plane is related to the spatial frequency spectrum generated by the lens array. To achieve a good flat-top uniformity using a non-imaging homogenizer, the lens array should distribute the light at equal intensities in the desired angular spectrum images.
[0058] Referring to FIG. 4, the condensed homogeneous beam passes through beam imaging subassembly 18, which is configured to deliver the homogenized beam to the image plane on the surface of the panel with a desired demagnification factor. The homogenized beam is first projected and focused onto mask 54 by a focusing lens 50 in a direction of the short axis of the beam. The mask 54 is configured with blades cropping edges of the projected homogenized beam in the long axis direction. This mask plane is then re-imaged onto objective lens 60 at some optimized demagnification with edge resolution limited by the numerical aperture of the imaging system. The demagnification, for example lOx or 3 Ox, reduces sensitivity to pointing stability. If the imaging system gets out of focus, however, diffraction patterns will be seen at the ends of the line beam due to diffraction from the sharp edges of the mask. As with any imaging system, depth of focus and resolution are inversely proportional to each other, as long as the imaging has diffraction limited resolution. An optional mask 56 receiving the line beam may be viewed as a vignetting aperture removing residual inhomogeneity of the previously sharpened edges.
[0059] Referring to FIG. 5, the de-magnified line beam with the desired degree of uniformed intensity and width further impinges upon a turning mirror 58 before coupling into objective lens 60 and illuminating the a-Si glass panel which lies upon a stage 62. The panels, particularly large panels, are hardly ideally flat. The use of auto-focusing assembly 64 allows maintaining the desired focal distance between spherical or anamorphic objective lens 60 and thus control the desired uniformity range of the the beam intensity of the line beam on target. Depending both on the system implementation and on the uniformity of the panel thickness, it may be necessary to implement dynamic autofocus, wherein the individual beams have their focal planes
continuously adjusted to maintain homogenous poly-Si grain structure throughout the panel area. Whether or not auto-focus will be required, and the accuracy that is needed, will depend on the depth of focus of the line beam optical delivery system.
[0060] Sometimes, observing the annealed panel, it is possible to notice iridescence which is caused by periodic micro structure created during the crystallization process. Non-uniformity of the microstructure likely causes variation in intensity known as "mura" believed to originate when two annealed panel areas are stitched together in a beam long axis direction. The subassembly 68 of measuring morphological characteristics of the annealed panel is disclosed in U.S. provisional patent application 62334881 which is fully incorporated herein by reference. The presence of Mura may require an automatic readjustment of system 10. More particularly, standalone Mura measuring system 68 is operative to measure properties of the light which is generated by an independent laser source and diffracted from the panel in real time. The properties may include the inhomogeneity of diffraction efficiency, diffraction angle and polarization state of the diffracted light. If the measured parameter or parameters of MURA is/are beyond a predetermined range, then providing one or multiple feedback controlling loops allows readjustment of any of the above-disclosed subassemblies in real time.
[0061] Still another measuring sub-assembly is configured as a beam profilometer 70 position anywhere along the beam path downstream from homogenizing subassembly. The beam profilometer 70 may be a camera-based beam profiling system configured with a camera and analysis software. Often times, this system needs to be used with beam attenuation or beam sizing accessories, depending on the task at hand. The advantage to camera-based beam profiling is the real-time viewing and measuring of a laser's structure with high accuracy measurements. If the measured beam profile parameter or parameters are outside the predetermined range, the annealing process terminates and troubleshooting of the identified malfunctioning subassemblies begins.
[0062] Referring to FIGs. 6 and 7, homogenizing sub-assembly 16 of system 10 based on the segmentation and recombining technique includes bi-prisms. The inventive bi-prism-based subassembly 16 is configured to impart an optical path difference between two beamlets. The path difference should be such that delay is longer than the coherence length/time.
[0063] In particular, a preconditioned harmonic circular or elliptical Gaussian beam 80 propagates through a λ/2 waveplate 82, configured to control splitting energies, and is further fragmented in an upstream polarizing beam splitter 72 directing two orthogonally polarized beamlets 84 and 86 along respective short and long paths. The beamlet 86 is guided along a longer, delay path of sub-assembly 16 including multiple turning mirrors 88 a polarizing beam combiner 74. The other beamlet 84 propagates through a single axis eversion prism 76 where it is everted and further coupled into polarizing beam combiner 74.
[0064] For example, if path 86 is lm longer than path 84, then there will be «3ns time difference between the two paths. For 1.5ns pulses at 150MHz repetition rate, this delay is sufficient to ensure that pulses within a burst from the two paths will arrive at separate times, and there will be no interference. As illustrated in FIG. 7, blue 86 and red 84 beamlets pass through respective long and short paths with a 3.3. ns delay without overlapping one another in time and avoiding interference. The beamlets of the same polarization follow one another at 150 MHz
corresponding to 6.7 ns rep rate. Accordingly, the interleaved beamlets are output at 300 MHz effective rep rate in a ~ 303 ns burst. It should be taken into consideration that if the time delay between two paths is shorter than the pulse duration but longer than the laser coherence tine, there will also be no interference.
[0065] While this method is certainly simpler than the fly's eye, the better homogeneity of the beam can be obtained with the fly's eye method because of multiple lenslets and thus multiple beamlets compared to only two rays in the bi-prism method.
[0066] FIGs. 8A-8B and 9A-9B illustrate another modification of homogenizing sub-assembly 16. This type of homogenizer relates to the bi-prism segmentation and recombining of FIG. 6 and generally can be referred to as beam combining. In particular, the time slice bi-prism homogenizer can be extended to beam combining homogenization techniques via the geometric overlapping of multiple beams and/or beam segments, as long as no two pulses within a pulse burst arrive at a specific location with temporal overlap.
[0067] FIGs. 8A and 8B illustrate four (4) beams 88, 90, 92 and 94 respectively which are overlapped and temporally interspersed to generate a central homogenous section 96. FIG. 8B illustrates a 4 beam overlap in which beams 88 and 94 are simultaneous, whereas beams 90 and 92 are interspersed. As can be seen, the beams are not temporally or spatially overlapped. In the context of system 10, this homogenous section is cropped by a mask and then imaged onto the panel surface.
[0068] FIGs. 9A and 9B illustrate a 3 beam overlap with eversion. Here the outside beam is everted such that the majority of the line beam is homogenous in the long axis. Such a method allows for extended length of the line beams, and reduces the quantity of line stitching required for fiber laser beam used for annealing large panels as will be explained below. While more efficient than the cropping technique of FIGs. 8A-8B, this technique still has discarded laser power at the ends of the homogenized line beam.
[0069] The simplest method allowing a Gaussian beam profile to be converted to a top hat beam profile is to crop the center portion of the beam that is within the homogeneity requirements. The cropping must be done at the mask (object) plane 54 of FIG. 4, which is then imaged to the combined object/process (image) plane via objective lens 60 of FIG. 5. The objective lens 60 in this method may use anamoiphic cylindrical elements, with different demagnification in each axis, or spherically symmetric elements that provide constant demagnification. This method discards the majority of the beam energy. In this regard, the beam combining method is more efficient than simply cropping the central, homogeneous portion of a single beam, but still has the same problem.
[0070] FIGs. 10A - IOC illustrate respective beam combining configurations. FIG.10A illustrates a system including a polarizing beam combiner receiving two linearly p-polarized and s-polarized input laser beams. The input beams propagate along respective legs of the shown system each of which may be provided with a ½ waveplate. FIG. 10B illustrates a beam combining structure capable of angularly combining multiple beams that pass through a field lens capable of focusing these beams into a single beam waist. FIG. IOC illustrates a beam combining arrangement in which multiple converging beams of different orders impinge upon a diffractive beam combiner configured to combine the passing beams into a single output beam.
[0071] Referring to FIG. 11 A - 11C, beam homogenizing subassembly 16 is configured to remap the apodization of a Gaussian beam to a top hat apodization utilizing aspherical optical elements. The best known example of this type is the piShaper. The latter is a telescope where intensity profile is transformed in a controlled manner one of the basic principles is zero wave aberration of the entire system which distinguishes this type of homogenizer from, for example, the segmentation and recombining-based.
[0072] As mentioned above, the SLS process only absolutely requires homogenization in the long axis. The optimum top hat re-apodization will therefore be either one axis only, or anamorp c. In either case cylinder lenses are required. The shown system includes, for example, an anamorphic crossed acylinder lens 98 designed to transform a circular beam to a 1mm long line beam, homogenized in each axis to the diffraction limited point spread. This lens directly transforms the beam intensity profile at focus. This method can be used to illuminate a mask plane, which is then imaged onto the process plane via an anamorphic or spherically symmetric objective lens. In practical application of this invention, the use of a single anamorphic lens converting the beam profile in the long axis is required. This method is alignment sensitive to input beam profile and divergence, as well as centration and angle of incidence. Any beam delivery that utilizes such a method must be capable of ensuring beam centration of the order of 10's of microns, and orthogonality of 10's of micro radians.
[0073] Turning now to the scanning system of profile and scanning subassembly 16, if a line beam with a certain length and intensity is scanned along the full length of the desired crystallization line with a certain velocity, it is possible to irradiate the full line homogenously with a desired exposure time and fluence. This method allows a shorter beam to be dragged in the long axis direction to generate a continuously long line of arbitrary length making thus the line stitching unnecessary. The profile of the beam in the short axis is less important, but it must remain constant along the length of the beam. A top hat or even supergaussian short axis profile will allow for more use of the laser power, but is not essential for the process to be effective. The scanning technique is realized by either a rotating mirror of FIGs. 16A-16D or acousto- optical deflector (AOD), as disclosed below.
[0074] FIGs. 12A ~ 12B illustrate the general operating principle configurations implementing the scanning technique. FIG. 11 A a constant beam of pulses irradiating a long line in the direction of arrow A. Every location along the line is exposed to, for example, the same top flat beam (FIG. 11C) such that at a certain point of time, initial portion 101 is fully crystallized. The recently irradiated line stretch 103 is still in the process of crystallization and will be fully crystallized as the panel moves further along arrow A with line stretch 105 being currently irradiated and stretchl07 is yet to be irradiated. The thermal profile of the line to be crystalized shown in FIG. 11B graphically illustrates the status of the above disclosed line stretches 101- 107.
[0075] For a given laser power and line beam width, the required line beam length and scanning velocity to achieve desired fluence and exposure time can be determined as follows:
Assume top hat line beam of FIG. 11 C with the line beam of length, Lb and traveling velocity v. At this velocity, an exposure time
T = Lb/v.
Hence for the exposure time T, and with line beam length, Lb, required velocity:
v = Lb/T
For on target laser power, P, and line beam width, Wb, the intensity is:
I = P/Lb Wb
The scanning fluence at any point is:
Summarizing the above it is easy to see that for desired fluence, H, and exposure time T, with laser power P and line beam width, Wb, the required line length and scanning velocity are:
Lb = P T/H Wb v = P/H Wb
Example:
Laser power = 150 W
Line beam width = 5 μηι
Required exposure time = 300 ns
Required rluence = 0.7 J/cm2 (7,000 J/m2)
Based on the above, the line beam length Lb = 1.3 mm
Scanning velocity = 4,300 m/s.
[0076] FIGs. 13A - 13D conceptually illustrate the scanning configuration. The laser source outputs the harmonic beam passing through the beam conditioning subassembly which generates the desired beam profile in the mask plane or directly at the image plane after the scanning assembly. Turning briefly to FIG. 13B-13D, the desired profile may be Gaussian in both axes 104, Gaussian/Quasi Gaussian in the direction of one axis and top hat in the direction of the other axis 106 and the top hat profile in the direction of both axes as indicated by 108. If the top hat profile is desired, then the homogenizing subassembly is required, and its configuration depends on laser beam characteristics and desired profile. The beam is incident on the scanner and scanning optics before it passes through the mask plane or directly at the image plane panel obtaining the desired beam profile.
[0077] FIGs. 14A -14E illustrate one of possible configurations of beam delivery and scanning subassembly 18 provided with a rotating mirror, such as a monogon or a multifaceted polygon 100. Turning briefly back to FIG. 2, scanner 100 may be mounted to the Gantry machine instead of turning mirror 46 and used with or without the homogenizer to illuminate the mask. The facets can be oriented at any angle, such as 450 or 90°, and deflect the harmonic beam along the light path towards a telecentric, cylindrical Galilean expander shown in FIGs. 14B-14E. The expanded beam then passes through short axis field lens 50, mask 54 and turning mirror 8 before coupling into objective lens 60.
[0078] Another type of the scanning-based beam delivery system includes the use of optical solid state deflectors relying on the acousto-optic effect and referred further to as AOD. The AODs, not shown but well-known to one of ordinary skill, do not contain moving parts and therefore exhibit high deflection velocities and more reliable than mechanical scanners. The AOD is based on a periodically changing refractive index in an optically transparent material, induced by propagating sound waves in the material. The changing refractive index is the result of refraction and compression of the material, inducing a changing density of the material. This periodically changing refractive index acts like an optical grating, moving at the speed of sound in the crystal that will diffract a laser beam traveling through the material.
[0079] In the context of SLS, the panels to be annealed with line beams are orders of magnitude larger than the length of the line beams that can be achieved with individual burst mode fiber lasers. Accordingly, it is necessary to stitch together beams from individual or multiple lasers to achieve effectively continuous poly-Si grain structures with grain size smaller than 1 micron and as small as 0.2 micron.
[0080] Excimer lasers used for ELA processes have high energy per pulse with low repetition rate. This makes them suitable for single line annealing of large panels, wherein the line beam encompasses the entire panel width.
[0081] Burst mode fiber lasers with equivalent total power output, however, have orders of magnitude lower pulse energy and requisite orders of magnitude higher repetition rate. The pulse energy is too low to allow for a line beam from a single laser to encompass the entire panel width.
[0082] For this reason, it is necessary to stitch together or scan individual line beams, such that the resultant poly-Si grain structure is continuous over the region of interest. The region of interest may be the entire panel area, or portions thereof.
[0083] Furthermore, at very high repetition rates, there may be insufficient time between individual bursts to allow sufficient cooling for SLS crystallization to occur. In this case it may be necessary to place every second line beam in a first pass, and then place the intervening line beams in a second pass, such that there is sufficient cooling at each beam location for crystallization to occur. Alternatively, a pulse picking method, or a method for simultaneously scanning in the long axis direction, may be implemented. Pulse picking may used in lieu of, or in conjunction with, placing every second line per pass. Stitching together 2 adjacent lines in the long axis provides significant challenges. Ideally each line beam will have a perfectly sharp edge and the edges of adjacent beams will exactly abut.
[0084] Another possibility is to tailor the intensity profiles of the ends of the beams such that when adjacent beams overlap the effective intensity remains homogenous. Even though there are coherence/speckle issues with overlapping coherent beams, it is possible to intersperse the adjacent beams such that the individual pulses in each burst do not arrive at the same time. [0085] The following disclosure covers various methods of stitching beam lines for continuous poly-Si coverage. Controlling the beam delivery and panel handling subassemblies, is possible to provide a variety of line patterns and ways by which individual line beams can be stitched to encompass the regions of interests.
[0086] Referring to FIG. 15 A -15B, continuous line with 1-D scanning is conceptually the simplest method used to align individual beams such that they constitute an effectively continuous line, somewhat analogous to the LA continuous line beam. This effective line beam would then be scanned in one (1) dimension along the entire panel as shown in FIG. 16A.
Depending on the dimensions of the individual line beams, this can result in very tight packing of objective/projection lenses and other optical elements, such that mounting and alignment fixturing would be extremely challenging.
[0087] FIGs. 15A and 15B show respective configurations that mitigate the tight packing issues. By placing fold mirrors 110 near the waists of the individual beams, i.e., close to the image plane, where there is sufficient free space between beams. The the individual
objective/projection lens or lenses 60 (FIG. 5), in case of multiple laser beams, can be sequentially placed in different orientations to free up space for fixturing of these lenses. The two illustrated arrangements on respective FIGs. 15A and 15B show alternate direction fold, and three direction fold stacking of individual line beams. This basic concept can be extended to other configurations to achieve the same goal. Challenges with generating a continuous line are primarily related to alignment, coherent interference, autofocus and mechanical/thermal stability.
[0088] The individual line beams must be carefully aligned, in both position and angle, such that the resultant continuous line beam is sufficiently homogenous in both intensity and profile to support the annealing process. For the SLS process, the individual beams must be aligned relative to each other to «0.1 μιη tolerances along the full continuous crystallization line length.
[0089] Some overlap is required where the individual line beams meet. The extent of overlap will depend on the specific beam profiles. The fiber lasers used for the SLS process have high spatial and temporal coherence, and are prone to interference effects when overlapped. Either neighboring individual beams must have random phase with respect to each other, or the pulse bursts must be timed such that the individual pulses within each burst amve at separate times for neighboring line beams as discussed above in reference t profile and velocity subassembly. [0090] Even if the individual line beams can be configured to deliver the required continuous line beam parameters, they must maintain alignment while continuously processing large area panels. Given the large dimensions of generation 8 panels for example, even a small temperature differential of 1°C can result in ΙΟμηι, or more, of misalignment over the panel dimensions due to differential thermal expansion.
[0091] Furthermore, if the burse repetition rate is too high for crystallization to occur with a single pass, placing every second burst on a first pass and then filling in the remaining bursts on a second pass may be required.
[0092] Finally, the depth of focus of the laser process will be limited, and it is probable that the panels will have insufficient flatness to satisfy the required depth of focus. In this case, it will be necessary to implement dynamic autofocus for each of the individual line beams.
[0093] A further modification of continuous line includes multiple passes as shown in FIG. 16B. If the length of the continuous line is limited to a portion of the panel width, then the entire panel could be covered by using several passes, with the line beam offset between passes. If the line is ½ the panel width, then 2 passes would be required. 1/3 the panel width would require 3 passes, and so on. While this approach eases some of the challenges associated with a full panel width line beam, there are challenges associated with maintaining continuity of poly-Si grain structure between passes. The burst in consecutive passes must be aligned in both axes to «0. Ιμηι accuracy. Again, if the repetition rate is too high, physical separation of sequential bursts will be required.
[0094] Referring to FIGs. 17A and 17B, rather than attempt to create a continuous line beam from multiple shorter line beams, it is possible to distribute individual line beams evenly spaced across the panel, such that the beams can be scanned in one direction and stepped in the other to provide continuous poly-Si grain structure. For simplicity, they are scanned in the same direction step after step. Alternatively, it is possible to scan in alternate directions after stepping for each pass. The actual number of beams could vary from a few to tens, depending on the number of lasers, the specific characteristics of the lasers, individual line beam dimensions, and how the individual line beams are generated. Again, the burst repetition rate may require physical separation of sequential bursts. The embodiments shown in FIG. 16B and 17A-B, may be combined such that the separate line beams of of FIG. 17 are each composed of a number of continuously stitched individual line beams of FIG. 16. [0095] The requirement to physically separate sequential busts may arise due to repetition rate and/or interdigitation constraints. There are two basic methods to physically separate sequential pulses.
[0096] One technique includes multiple passes with increased scan speed. Here a single line beam is scanned in the short axis direction with respect to the panel at sufficient speed such that the individual pulses are physically separated to generate every second line beam. An ensuing pass is offset in the short axis direction such the missing alternate line beams are placed.
[0097] This above technique is the most conceptually simple compare to the other technique requiring line beam spacing, but it also requires doubling the scan speed as compared to continuously placing the line beams in a single pass. For example, if the laser is operating at 1MHz, and the line beams must be spaced every 2μηι, then a single pass would require a scanning speed of 2m/s, while a double pass would require 4m/s. These are extremely high speeds at which to scan beams over large panels with 0.1 μπι level accuracy.
[0098] If multi-pass interdigitation, is implemented, wherein the interdigitation ratio is greater than 2, then the scanning speed issue is compounded linearly with the interdigitation ratio.
[0099] If the pulse bursts are sequentially separated into different beam lines, repeating at a fixed modulus value, such that the effective repetition rate for an individual beam line is reduced by the modulus. The previous example requiring a single beam line at 2m/s would, for example, be modified to 4 beam lines at 500mm/s if the modulus is 4.
[00100] If multiple scans are used, whether multiple pass continuous line beams, separate line beam locations, or a combination thereof, there will always be challenges at the seams between the individual passes. Maintaining continuity of the poly-Si grain structure across the seams will be a challenge. The specific requirements for the final devices to be manufactured from the SLS annealed panels will define the allowable seam discontinuities.
[00101] If the requirements are too tight to be met with continuous seams, then it is possible to implement a beam interdigitation scheme, wherein the boundaries between individual beams are staggered across the panel, as illustrated in FIGs. 18, 19, 20 and 21. Depending on the degree of interdigitation, it can be guaranteed that no device transistor will have more than a specific fractional area of beam stitching boundaries. [00102] Interdigitation also has the added advantage of separating the placement in time of beams that are physically adjacent on the panel, and thereby reducing or eliminating the implications of inadequate crystallization time between pulses due to high repetition rate.
[00103] Referring to FIGs. 18A - 18C, interdigitation can be achieved by placing each offset with individual passes. For example, if two-beam interdigitation is desired as shown in FIG. 19B, then the scan speed would be such that the pulse spacing is exactly 2 line widths. The beams would then be stepped with respect to the panel by one half the beam length. The second pass would place pulses exactly between the pulses of the first pass. The beam would then be stepped again by one half the beam length, and the third pass would place the beams exactly aligned with the beams of the first pass, as shown in FIG. 18C.
[00104] If three-beam Interdigitation is required, as shown in FIGs. 19A -19E, the same procedure would be followed, except that the pulses would be spaced by three line widths, and the step size would be one third of the beam length between passes. The number of mterdigitated beams can be chosen to be arbitrarily large. This method will be limited by the maximum achievable system scan speed. For a given pulse repetition rate and line width, the required scan speed will increase linearly with the number of interdigitated beams. Interdigitation
automatically eliminates any issues related to repetition rate and crystallization time.
[00105] Referring to FIG. 20 A - 20D and 21 A - 21 C, rather than placing all the pulses individually, one pass at a time, it is possible to place multiple beams simultaneously, such that all required Interdigitation takes place within a single pass. This figure illustrates single pass 2- beam and 3-beam interdigitation. The beams are placed with respect to each other such that, during a single scan, multiple beams are interdigitated. As with the mode for individual pulse placement, the number of interdigitated beams is only limited by the practicality of
implementation.
[00106] By separating the individual beams in the scan direction, this method reduces the effective repetition rate in terms of crystallization time linearly with the number of interdigitated pulses. Not only does this method reduce the number of passes required to fill in the
interdigitated pattern, it also works well in combination with the pulse picking method of FIG. 22.
[00107] The beam stitching methods can be implemented in combinations to achieve the required p-Si grain characteristics within the constraints of laser and system specifications. Furthermore, each of the disclosed techniques, and combinations thereof, can be implemented with multiple, coordinated laser sources, and also in conjunction with the pulse picking methods described later.
[00108] As previously discussed, high laser repetition rates can result in insufficient time for crystallization between pulses in SLS. Rather than use multiple passes to fill in each scan line with physically separated pulses, it is possible to pick pulses into separate beam lines with a fixed modulus. The pulse picking is associated with the scanning subassembly 100 using a mechanical (scanning polygon, galvanometer, etc.), acousto-optic methods as disclosed above or electro-optic method known to one of ordinary skill worker.
[00109] The pulse picking technique can be implemented in different ways. One of the ways includes directing the pulse picked beams into individual beam lines which are then scanned similarly to a single line. This can be realized using two approaches - continuous line and interdigitated line beams - and has the advantage of reducing the effective repetition rate of the individual beam lines, but requires multiple beam deliveries per laser.
[00110] The first approach allows the pulse picked beams to be delayed with respect to each other and combined, to form a single long homogenous continuous line that can be passed through a single beam delivery. This results in a line beam analogous to one that can be achieved with a lower repetition rate, higher pulse energy laser. The alternative approach relates to the interdigitation and includes using the pulse picked beams to generate multiple pass
interdigitations or arranged to produce a single pass interdigitated pattern, as disclosed above.
[001 11 ] Another way of pulse picking includes long axis scanning in conjunction with a rotating scanner, such as polygon 100 shown in FIG. 23 or AOD. The redirected beam is incident on a proximity mask where knife edges define the length of the beam irradiating the surface. The beam's linewidth is formed using the above disclosed short axis mask with the degree of demagnification different from that of the proximity mask. As a result, consecutive bursts of pulses produce adjacent, closely positioned lines separated along the long axis and stitched in a continuous longer line at effectively reduced burst rep rate with respect to the short axis, of the This is different from the line dragging technique since the latter is characterized by the continuity and obviously does not require stitching. Scan along the long axis of the beam such as consecutive bursts are stitched continuously in line on the panel. [00112] Despite all the configurations and techniques disclosed above, localized, persistent regions of inhomogeneity in the line beam may result in patterning in the
poiycrystaliine grain structure. These patterns may cause visible Mura in the finished display, especially for stitched lines wherein the patterning will be periodic. One mitigation technique is to dither the line beam onto the panel, such that inhomogeneous regions (including stitching) are at different locations in sequential lines, to effectively smooth out the inhomogeneities and reduce the resultant Mura to acceptable levels.
[00113] Structurally, a dithering system may include a rotating wedge or diffusing element placed into the beam path. This can be very effective for removing speckle in coherent beams, but may be difficult for use in the SLS annealing process because rotating elements smear the beam in two dimensions, and SLS annealing requires a very narrow line beam. Any
dithering/smearing must be in the direction of the long axis of the line beam only.
[00114] The dithering may be periodic or stochastic in nature. Periodic dithering will follow a periodic profile, such as a saw tooth or sinusoid. Stochastic dithering will result in randomized (or pseudo-randomized) distributions of the inhomogeneities. The preferred type of dithering will depend on the dithering method and acceptable levels of Mura. Stochastic dithering is expected to be more effective in Mura reduction, whereas periodic dithering may be less complicated/costly to implement while achieving acceptable Mura.
[00115] The magnitude and periodicity (if not stochastic) of the dithering will depend on the types of inhomogeneities, whether or not there is stitching, and what levels of Mura are acceptable. The magnitudes and periodicities could range from the order of ten microns to, potentially, greater than one millimeter. One dimensional dithering can be achieved by several alternative subassemblies shown in FIGs. 24-28
[00116] FIG. 24 illustrates a dithering system 120 operative oscillate the panel or any suitable component after the mask. The panel is oscillated/vibrated in the direction of the line beam during the SLS annealing process. For device length lines, this can be either a periodic or stochastic oscillation, as illustrated in FIG. 27. For stitched lines, it may be necessary to use periodic oscillation, wherein consecutive passes follow the same periodic path. This will both ensure constant overlap of the stitched lines, and smooth the overlap region along the scan path, if stochastic dithering is used with stitched lines, then the overlap between consecutive scans will be highly variable. This may not be acceptable for many types of device, particularly OLED devices.
[00117] FIG. 25 shows another configuration of dithering system 120 oscillate an optical component of the beam delivery system, such as a lens or mirror, which is oscillated/vibrated in the direction of the line beam during the SLS annealing process, to produce the requisite inhomogeneity smoothing, while maintaining sharp definition of the line beam. If the dithering is introduced into the beam ahead of a mask plane, then the mask may be used to maintain a straight path of the line beam edge on the panel. In this case, either periodic or stochastic dithering will be equally applicable. If the dithering is introduced after the mask, then the resultant line distributions will be equivalent to oscillating the panel and the same arguments for periodic vs. stochastic dithering hold as illustrated in FIG. 27.
[00118] FIG. 26 illustrates still another configuration of dithering system 120 oscillating the mask. The mask is oscillated/vibrated in the direction of the line beam during the SLS annealing process. This will smooth out the stitching zone between consecutive passes, but will not smooth out inhomogeneities within the line beam. Again, the same arguments for periodic vs. stochastic dithering hold as for panel oscillation.
[00119] While there is shown and described herein certain specific structure embodying the invention, it will be manifest to those skilled in the art that various modifications and rearrangements of the parts may be made without departing from the spirit and scope of the underlying inventive concept and that the same is not limited to the particular forms herein shown and described except insofar as indicated by the scope of the appended claims.

Claims

1. A system for crystallizing amorphous Si (a-Si) panels by partial melt laser annealing (LA) or sequential lateral solidification (SLS) annealing process, comprising:
at least one single transverse mode (SM) quasi-continuous wave (QCW) fiber laser source, emitting a pulsed harmonic beam at a pulse repetition rate of at least 80 MHz, along a path; a beam conditioning assembly located downstream from the fiber laser source and configured to transform the harmonic beam such that the harmonic beam has desired divergence and spatial distribution characteristics;
a beam velocity and profile assembly operative to provide the conditioned harmonic beam with a desired intensity profile at a desired scanning velocity in an object plane;
a beam imaging assembly for imaging the conditioned harmonic beam in the object plane onto an image plane in at least one beam axis at a desired demagnification such that a width of the conditioned harmonic beam is reduced to a narrow linewidth of at least 1 μηι at the image plane; and
a panel handling assembly operative to provide relative position and velocity between the imaged narrow linewidth beam and the panel such as to irradiate each location of the a-Si panel at least two times with an exposure time during each time of at least 100 ns in order to provide transformation of a-Si to polysilicon (p-Si) structure with a uniform grain size of at most 1 μιη.
2. The system of claim further comprising multiple SM QCW fiber laser sources.
3. The system of claims 1 or 2, wherein the beam velocity and profile system is configured to convert a high-ratio Gaussian harmonic beam into a flat-top harmonic and selected from a beam segmentation and recombination system, beam re-apodization system, beam combining system, and beam cropping system.
4. The system of any of above claims, wherein the beam segmentation and recombining system is selected from a fly's eye or bi-prism optical arrangements.
5. The system of any of the above claims, wherein the fly's eye is imaging or non-imaging homogenizer.
6. The system of claim 1 or 2 or 3, wherein the beam combining system is configured to overlap and intersperse multiple harmonic beams to generate a central homogeneous central top portion of an intensity profile which is cropped at the object plane.
7. The system of claim 1 or 2 or 3, wherein the beam combining system is operative to overlap multiple harmonic beams wherein one of the overlapped beams being everted so that a resulting harmonic beam is homogeneous in a direction of long axis.
8. The system of claim 1 or 2 or 3, wherein the beam combining system is configured with a polarizing beam combiner or field lens combiner or diffractive beam combiner or fly's eye.
9. The system of claims lor 2, wherein the beam re-apodization system is configured with at least one or multiple acylindrical optical elements converting a high-ratio Gaussian harmonic beam into a flat-top intensity profile in at least of beam axes.
10. The system of any of the above claims, wherein the beam velocity and profile system is configured with a scanner operative to provide the conditioned harmonic beam with a desired velocity so that the imaged narrow linewidth beam homogeneously and continuously generate a line of crystallization without stitching.
11. The system of claim 10, wherein the scanner is selected from a rotating mirror or acousto optical deflector or galvanometer.
12. The system of any of the above claims, wherein the beam imaging assembly is configured with a focusing lens focusing the conditioned harmonic beam with the desired intensity profile at the desired scanning velocity in a short beam axis direction onto a first mask defining the object plane and having cutting knives for sharpening edges of the beam in a long axis direction, and an objective lens located downstream from the first mask and adjacent to the panel.
13. The system of claim 12, wherein the beam imaging assembly further includes a second mask located between the first mask and objective lens and configured to vignette residual
inhomogeneity of the line of crystallization.
14. The system of any of 1 through 11 claims, wherein the beam imaging assembly is configured with an anamorphic lens arrangement providing the different desired demagnification along orthogonal beam axes of the conditioned harmonic beam with the desired intensity profile.
15. The system of any of 1 through 11 claims, wherein the beam imaging assembly is anamorphic and includes two spaced masks providing different demagnification in respective orthogonal beam axes and having different object planes.
16. The system of any of 1 through 12 claims, wherein the beam imaging assembly has a proximity mask configured to define a desired length of a beamline on the panel, the proximity mask being spaced from the panel at a distance limiting edge diffraction.
17. The system of any of the above claims, wherein the panel handling assembly includes a support supporting the panel to be annealed such that the panel is displaceable in orthogonal XY planes relative to the fixed beam imaging assembly.
18. The system of any of 1 through 16 claims, wherein the panel handling assembly includes a support supporting the panel to be annealed and configured such that the panel is stationary relative to the beam imaging assembly displaceable in orthogonal XY planes.
19. The system of any of the above claims, wherein the panel handling assembly is configured with a support supporting the panel to be annealed such that the panel is displaceable on one of XY planes and the beam imaging system is displaceable on another of XY planes.
20. The system of claim 18 or 19, wherein the SM QCW fiber laser source is mounted in a fixed position relative to the displaceable beam image system or displaceable therewith.
21. The system of any of the above claims further comprising an auto focus system, beam profiler and MURA measuring system.
22. The system of one of claims 1-5, wherein the fly's eye homogenizer configured with delay step glass elements to eliminate coherence effects.
23. The system of one of the above claims further comprising a dithering system operative to dither the narrow linewidth beam onto the panel such that residual inhomogeneous regions of the p-Si structure are at different locations in sequential lines effectively smoothing out the residual inhomogeneities and reducing the Mura to a predetermine reference range.
24. The system of one of the above claims, wherein the dithering system is operative to oscillate:
the panel or any suitable component downstream form the object plane, or
an optical component of the beam delivery system, such as a lens or mirror, or an optical component along the beam path in the direction of the conditioned harmonic narrow width line beam during the SLS annealing process, or
the mask defining the object plane.
EP16818736.7A 2015-06-29 2016-06-29 Fiber laser-based system for uniform crystallization of amorphous silicon substrate Pending EP3314633A4 (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US201562186057P 2015-06-29 2015-06-29
US201662315310P 2016-03-30 2016-03-30
PCT/US2016/040222 WO2017004280A1 (en) 2015-06-29 2016-06-29 Fiber laser-based system for uniform crystallization of amorphous silicon substrate

Publications (2)

Publication Number Publication Date
EP3314633A1 true EP3314633A1 (en) 2018-05-02
EP3314633A4 EP3314633A4 (en) 2019-04-10

Family

ID=57609576

Family Applications (1)

Application Number Title Priority Date Filing Date
EP16818736.7A Pending EP3314633A4 (en) 2015-06-29 2016-06-29 Fiber laser-based system for uniform crystallization of amorphous silicon substrate

Country Status (4)

Country Link
EP (1) EP3314633A4 (en)
KR (1) KR102509883B1 (en)
CN (1) CN107924827B (en)
WO (1) WO2017004280A1 (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11673208B2 (en) * 2017-07-31 2023-06-13 Ipg Photonics Corporation Fiber laser apparatus and method for processing workpiece
WO2019028082A1 (en) * 2017-07-31 2019-02-07 Ipg Photonics Corporation Laser apparatus and method of processing thin films
EP3642384A4 (en) * 2017-07-31 2021-03-31 IPG Photonics Corporation Laser apparatus and method of processing thin films
US20200238441A1 (en) * 2017-10-13 2020-07-30 The Trustees Of Columbia University In The City Of New York Systems and methods for spot beam and line beam crystallization

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050189329A1 (en) * 2003-09-02 2005-09-01 Somit Talwar Laser thermal processing with laser diode radiation
JP4567984B2 (en) * 2004-01-30 2010-10-27 株式会社 日立ディスプレイズ Flat panel display manufacturing equipment
US7700463B2 (en) * 2005-09-02 2010-04-20 Semiconductor Energy Laboratory Co., Ltd. Method for manufacturing semiconductor device
WO2007046290A1 (en) * 2005-10-18 2007-04-26 Semiconductor Energy Laboratory Co., Ltd. Semiconductor device and manufacturing method thereof
TWI479660B (en) * 2006-08-31 2015-04-01 Semiconductor Energy Lab Thin film transistor, manufacturing method thereof, and semiconductor device
CN101933201A (en) * 2007-08-01 2010-12-29 深度光子公司 Method and apparatus for pulsed harmonic ultraviolet lasers
JP5688203B2 (en) * 2007-11-01 2015-03-25 株式会社半導体エネルギー研究所 Method for manufacturing semiconductor substrate
JP2009135448A (en) * 2007-11-01 2009-06-18 Semiconductor Energy Lab Co Ltd Method for manufacturing semiconductor substrate, and method for manufacturing semiconductor device
KR101094294B1 (en) * 2009-11-17 2011-12-19 삼성모바일디스플레이주식회사 Method for manufacturing organic light emitting diode display
TWI545627B (en) * 2012-06-13 2016-08-11 Sumitomo Heavy Industries Semiconductor device manufacturing method and laser annealing device
JP5918118B2 (en) * 2012-12-18 2016-05-18 株式会社日本製鋼所 Method for manufacturing crystalline semiconductor film
US9343307B2 (en) * 2013-12-24 2016-05-17 Ultratech, Inc. Laser spike annealing using fiber lasers

Also Published As

Publication number Publication date
KR102509883B1 (en) 2023-03-13
KR20180014439A (en) 2018-02-08
EP3314633A4 (en) 2019-04-10
WO2017004280A1 (en) 2017-01-05
CN107924827B (en) 2022-07-01
CN107924827A (en) 2018-04-17

Similar Documents

Publication Publication Date Title
US6943086B2 (en) Laser annealing apparatus, TFT device and annealing method of the same
KR101167324B1 (en) Laser thin film poly-silicon annealing optical system
KR101115077B1 (en) Laser thin film poly-silicon annealing system
JP4322359B2 (en) Laser processing equipment
TWI260701B (en) Laser annealing apparatus and annealing method of semiconductor thin film using the same
JP2017135390A (en) Novel thermal treatment apparatus
KR102509883B1 (en) Fiber laser-based system for uniform crystallization of amorphous silicon substrates
US11465235B2 (en) Laser light irradiating device
JP2004103628A (en) Laser annealing device and method of laser-annealing tft substrate
JP2010062559A (en) Process and device for changing structure of semiconductor layer
JP2005217267A (en) Laser irradiation device
US20230405713A1 (en) Fiber laser apparatus and method for processing workpiece
CN111065759B (en) Laser device and method for processing thin film
JP4628879B2 (en) Manufacturing method of display device
JP4442537B2 (en) Laser process equipment
TW202231394A (en) Apparatus for generating a laser line on a working plane
US11600491B2 (en) Laser apparatus and method of processing thin films
JP2006080205A (en) Device and method for laser machining
JP2004241561A (en) Laser processing system

Legal Events

Date Code Title Description
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE

PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20180123

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

AX Request for extension of the european patent

Extension state: BA ME

DAV Request for validation of the european patent (deleted)
DAX Request for extension of the european patent (deleted)
A4 Supplementary search report drawn up and despatched

Effective date: 20190308

RIC1 Information provided on ipc code assigned before grant

Ipc: H01L 21/20 20060101AFI20190304BHEP

Ipc: H01L 21/268 20060101ALI20190304BHEP

Ipc: H01L 21/324 20060101ALI20190304BHEP

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: EXAMINATION IS IN PROGRESS

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: EXAMINATION IS IN PROGRESS

17Q First examination report despatched

Effective date: 20200923

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: EXAMINATION IS IN PROGRESS