JP2013021353A - Systems and methods for implementing interaction between laser shaped as line beam and film deposited on substrate - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide systems and methods for positioning a film for interaction with a laser shaped as a line beam and for controlling parameters of the shaped line beam, for example, to melt an amorphous silicon film, for example, to crystallize the film for the purpose of manufacturing thin film transistors (TFTs).SOLUTION: A laser crystallization apparatus and method are provided for selectively melting a film such as amorphous silicon that is deposited on a substrate. The apparatus may comprise an optical system for producing stretched laser pulses for use in melting the film. In another aspect of an embodiment of the present invention, a system and a method are provided for stretching a laser pulse. In further another aspect, a system is provided for maintaining a divergence of a pulsed laser beam (stretched or non-stretched) at a location along a beam path within a predetermined range.

Description

CROSS REFERENCE The present invention to related application is claims priority to May 26, U.S. Patent Application No. 11 / 138,100, filed 2005, now US Patent No. 6,693,939 A long time application filed on Nov. 13, 2003, which is a continuation-in-part of US patent application Ser. No. 10 / 141,216, entitled “Laser Lithographic Light Source with Beam Delivery”, filed May 7, 2002. This is a continuation-in-part application of US patent application Ser. The entire disclosures of these documents are hereby incorporated by reference herein.
The present invention is also a continuation-in-part of US patent application Ser. No. 10 / 781,251 entitled “Ultra High Energy High Stability Gas Discharge Laser Surface Treatment System” filed on Feb. 18, 2004.
The present invention is also a continuation-in-part of US patent application Ser. No. 10 / 425,361, filed Apr. 29, 2003, entitled “Lithography laser with beam delivery and beam pointing control”.
The present invention positions the film against interaction with a laser shaped as a line beam and shapes the film to crystallize, for example, to melt an amorphous silicon film to produce a thin film transistor (TFT), for example. The present invention relates to a system and method for controlling line beam parameters.

  Crystallization by laser of an amorphous silicon film deposited on a substrate, for example glass, represents a promising technique aimed at producing a material film having a relatively high electron mobility. Once crystallized, this material can then be used to make TFTs suitable for use in thin film transistors (TFTs) and relatively large liquid crystal displays (LCDs) in one particular application. . Other applications of crystallized silicon films can include “organic LEDs (OLED)” and “system on panel (SOP)”. More quantitatively, it is believed that a mass production system capable of rapidly crystallizing a film having a thickness of about 90 nm and a width of about 70 mm or longer will be commercially available in the near future. This step can be carried out using a pulsed laser optically formed into a line beam, for example a laser focused on a first axis, for example a short axis and extended to a second axis, for example a long axis. it can. In general, the first and second axes are orthogonal to each other, and both axes are substantially orthogonal to the central ray traveling toward the membrane. An exemplary line beam for laser crystallization can have a beam width of less than about 20 microns and a beam length of about 700 mm. Using this configuration, the film is scanned or stepped in a direction parallel to the beam width and continuously melted to crystallize a film having a substantial length, for example 700 mm or longer. be able to.

In some cases it may be desirable to ensure that portions of the silicon film are exposed to a laser energy density that is controlled to a preselected energy density range during melting. In particular, for positions along the shaped line beam, energy density control to a preselected range is generally desired, and a somewhat constant energy density is desirable when the line beam is scanned over the silicon film. High energy density levels can cause the film to flow, leading to undesirable “fading”, ie, a non-planar surface profile and poor particle quality. This non-uniform distribution of film material is often referred to as “aggregation” and can make the crystallized film unsuitable for certain applications. On the other hand, low energy density levels can result in incomplete melting and poor particle quality. By controlling the energy density, a film having substantially homogeneous properties can be achieved.
One factor that can affect the energy density in the exposed film is the spatial relationship of the thin film to the depth of focus (DOF) of the pulsed laser. This DOF depends on the focusing lens, but for a typical lens system configured to produce a line beam having a beam width of 20 microns, a good DOF estimate is considered to be about 20 microns.

  With the above in mind, it will be appreciated that the portion of the silicon film that is completely within the laser DOF will experience a different energy density than the portion of the silicon film that is only partially within the laser DOF. To do. That is, surface fluctuations on the silicon film, glass substrate, and vacuum chuck surface holding the glass substrate, can result in unwanted energy density fluctuations between film positions, even with only a few micro fluctuations that are unknown. . Furthermore, even under controlled manufacturing conditions, the overall surface variation (ie, vacuum chuck + glass substrate + film) can be about 35 microns. It will be appreciated that these surface variations can be particularly problematic for focused narrow beams having a DOF of only about 20 microns.

In addition to surface variations, unwanted film movement relative to the shaped line beam can also cause energy density variations. For example, a small movement may occur during stage vibration. Also, improper alignment of the stage to the shaped line beam and / or improper alignment of the stage to the scan plane can result in unwanted energy density fluctuations.
Other factors that can lead to energy density variations between film positions include changes in laser power characteristics during scanning (eg, pulse energy, beam pointing, beam divergence, wavelength, bandwidth, pulse duration, etc. Change). In addition, the position and stability of the shaped line beam and the quality of the beam focus (ie, shape) during the scan can also affect the energy density uniformity.
With the above in mind, Applicants disclose several systems and methods for achieving interaction between a shaped line beam and a film deposited on a substrate.

  Disclosed are systems and methods for generating pulses having pulse characteristics suitable for interaction with a film deposited substrate such that a shaped beam forms a short axis and a long axis. In one aspect of embodiments of the present invention, systems and methods are provided for stretching laser pulses. The system includes a beam splitter for directing a first portion of the pulse along a first beam path and a second portion of the pulse along a second delayed beam path; and the delayed beam Arranged along a path, arranged to invert the second beam part, and in cooperation with the beam splitter, at least a part of the inverted second beam part is transferred to the first beam path. And a plurality of reflective elements disposed on the surface.

  Another aspect of embodiments of the present invention provides a system and method for maintaining the divergence of a pulsed laser beam within a predetermined range at a location along the beam path. The system includes an adjustable beam expander, means for measuring a divergence and generating a signal, adjusting the beam expander in response to the signal, and the pulsed laser beam And a controller for maintaining divergence within the predetermined range.

  In yet another aspect of embodiments of the present invention, a laser crystallization apparatus and method for selectively melting a film deposited on a substrate is provided. The apparatus includes a laser source that generates a pulsed laser output beam, an optical system that generates a pulse stretcher output by stretching pulses in the laser output beam, and an optical apparatus that generates a line beam from the pulse stretcher output. Can be included.

  In yet another aspect of embodiments of the present invention, systems and methods are provided for maintaining energy density in a predetermined range at the film during the interaction between the film and the shaped beam. The system uses an autofocus sensor to measure the distance between the membrane and the focusing lens, and adjusts light source parameters using the measurement to maintain the energy density in the preselected range at the membrane. And a controller.

1 is a schematic diagram of the main components of an exemplary manufacturing system for crystallizing an amorphous silicon film. FIG. It is a figure which shows the apparatus for determining whether a line beam is focused on the film | membrane deposited on the board | substrate. Three exemplary beams: a first beam having the best focus on the table plane, a second beam having the best focus at 10 microns from the table plane, and 15 microns from the table plane FIG. 6 is a graph showing intensity variation as a function of a short-axis beam for a third beam having the best focus. FIG. 6 is a graph showing an energy density as a function of lateral growth length and a region where partial melting and agglomeration may occur. 1 is a perspective view of a vacuum chuck assembly for holding a workpiece during interaction with a line beam. FIG. 1 is a schematic plan view showing a system for positioning a silicon film for interaction with a line beam and its usage. FIG. 1 is a schematic plan view showing a system for positioning a silicon film for interaction with a line beam and its usage. FIG. 1 is a schematic plan view showing a system for positioning a silicon film for interaction with a line beam and its usage. FIG. 1 is a schematic plan view showing a system for positioning a silicon film for interaction with a line beam and its usage. FIG. 1 is a schematic plan view showing a system for positioning a silicon film for interaction with a line beam and its usage. FIG. 1 is a schematic plan view showing a system for positioning a silicon film for interaction with a line beam and its usage. FIG. 1 is a schematic plan view showing a system for positioning a silicon film for interaction with a line beam and its usage. FIG. 1 is a schematic plan view showing a system for positioning a silicon film for interaction with a line beam and its usage. FIG. 1 is a schematic plan view showing a system for positioning a silicon film for interaction with a line beam and its usage. FIG. 1 is a schematic plan view showing a system for positioning a silicon film for interaction with a line beam and its usage. FIG. 1 is a schematic plan view showing a system for positioning a silicon film for interaction with a line beam and its usage. FIG. 1 is a schematic plan view showing a system for positioning a silicon film for interaction with a line beam and its usage. FIG. 1 is a schematic plan view showing a system for positioning a silicon film for interaction with a line beam and its usage. FIG. 1 is a schematic plan view showing a system for positioning a silicon film for interaction with a line beam and its usage. FIG. 1 is a schematic plan view showing a system for positioning a silicon film for interaction with a line beam and its usage. FIG. 1 is a schematic plan view showing a system for positioning a silicon film for interaction with a line beam and its usage. FIG. 1 is a schematic plan view showing a system for positioning a silicon film for interaction with a line beam and its usage. FIG. FIG. 2 is a schematic diagram of a portion of the system shown in FIG. 1 illustrating aspects of an embodiment of the present invention. It is a detailed view of a 6 mirror pulse stretcher. FIG. 6 is a plot of intensity and time for extended and non-extended pulses. FIG. 5 is a plot of intensity and vertical width showing an increase in vertical uniformity of an extended pulse when compared to a non-extended pulse. FIG. 4 shows an actively controllable beam expander that can be independently adjusted in two axes.

  Referring initially to FIG. 1, a schematic diagram that is not to scale of the major components of an exemplary manufacturing system, generally designated as system 10, for crystallizing amorphous silicon film 12 is shown. As shown, the system 10 includes a laser source 20 that generates a pulsed laser beam, a pulse stretcher 22 for increasing the pulse duration, a mechanism for actively guiding the beam, and / or an active beam expander. And a beam delivery unit 24 that can. The system 10 is stable for measuring one or more beam characteristics, eg, wavefront and / or beam pointing, and generating control signals used by the active guidance unit and / or active beam expander. The processing method module 26 may be further included. The system 10 also holds and positions an optical instrument module 28 for beam homogenization, beam shaping, and / or beam focusing, and a silicon film 12 deposited on a substrate 32, which may be, for example, glass. And a movable stage system 30 for.

  In summary, the system 10 shown in FIG. 1 and described in more detail below, includes a width (short axis) at a membrane 12 of about 20 microns or less (short axis), 700 mm or longer (major axis), And can be configured to produce a focused thin beam 34, such as a line beam, having a depth of focus (DOF) of about 10 microns to 20 microns. Each pulse of the focused thin beam can be used to melt the strip-shaped amorphous silicon. The strip is then crystallized. In particular, the strip crystallizes during the lateral growth process in which the grains grow in a direction parallel to the minor axis. The grains grow inward from the edge (parallel to the minor axis) and satisfy the creation of a ridge (so-called grain boundary protrusion) along the center of the strip extending from the plane of the silicon film. Next, the stage is moved piecewise or continuously to expose a second strip that is parallel to the first strip and that overlaps a portion thereof. During exposure, the second strip melts and then recrystallizes. Sufficient overlap can be utilized to melt the ridge again. By melting the ridge portion again, a flat film surface (for example, a peak-to-peak value of ˜15 nm) can be maintained. This process, hereinafter referred to as fine beam continuous lateral solidification (tb-SLS), is generally repeated until the entire film is crystallized.

  FIG. 2 shows an apparatus for determining whether the fine beam pulse laser 34 is properly focused on the silicon film 12 deposited on the substrate 32. Each part of the optical instrument module 28 (see FIG. 1) may include a short axis field stop 36 and a short axis focusing optical instrument 37. In general, the beam is initially focused on the field stop 36 and then imaged to produce an intensity profile as shown in FIG. 3A (plot 62) at the membrane 12. FIG. 3A shows the profile (plot 62) of field stop 36 configured as a slit having a small dimension in the minor axis. Using this configuration, a beam width (FWHM) of about 13 μm, intensity uniformity better than 5% along the flat top of the profile, and less than about 3 um between 10% and 90% of the total intensity The profile shown in FIG. 3A can be generated with steep edge ramps that can be made. Beams having a width between about 5 μm and 10 μm can also be utilized. A single edge (ie knife blade) is used instead of a slit at the field stop to melt again during a steep trailing edge ramp (ie tb-SLS) without affecting the leading edge A beam profile having an edge corresponding to the material that is not to be produced. Although shown as a single lens, it should be appreciated that the focusing optics 37 can include several optical components including, but not limited to, various types of lenses.

  FIG. 2 shows that the beam 36 is not collimated in the long axis 38 and branches from the field stop 36 to the membrane 12. As described above, the length of the beam 36 at the major axis 38 may be about 700 mm or longer. On the other hand, as shown in FIG. 1, the beam 36 is focused in the minor axis 40 by an optical instrument module 28 that can include a focusing optics 38. With this structural arrangement, the light 42 reflected from the film 12 continues to diverge from the optical axis 44 so that it is analyzed by the detection system and the beam 36 is properly focused on the short axis 40 (shown in FIG. 1). It can be determined whether or not.

  With continued reference to FIG. 2, the detection system may include a total reflection mirror 46 that directs the reflected light 42 to the image plane 48. The magnifying mirror 50 is positioned so that an enlarged image of the image plane 48 can be obtained by the camera 52. For a detection system, the image plane 48 is approximately equal to the distance traveled by the same reflected light 42 from the membrane 12 to the image plane 48 from the short axis field stop 36 to the membrane (eg, in the art). Relative to each other within an acceptable tolerance). The camera output can then be used to adjust one or more system variables to improve focus on the membrane 12 or energy on the membrane 12 as described in more detail below. The density can be changed. For example, the stage 30 can be moved relative to the focusing optics 37 to adjust the focus.

  In some cases, it may be desirable to include a second detection system similar to that described above having a mirror 54, a magnifying mirror 56, and a camera 58, as shown in FIG. In combination, two detection systems can be used to simultaneously determine whether the light is focused on the long axis 44 at both ends of the beam (at the short axis). One functional aspect of note of the detection system shown in FIG. 2 is that the detection system, particularly the mirrors 46, 54, are positioned so as not to interfere with the light traveling from the beam 36 to the substrate 12. Further, this configuration allows the fine beam focus to be analyzed and adjusted during exposure of the membrane 12.

  FIG. 3A shows intensity variation as a function of short axis beam width for a focused beam (plot 62), a beam that is 10 microns out of focus (plot 64), and a beam that is 15 microns out of focus (plot 66). The graph is shown. The various plots shown relate to a focusing lens having a numerical aperture (NA) of approximately 0.15. An interesting feature of these plots is that all of the plots 62, 64, 66 have relatively steep sidewalls. These steep sidewalls are the result of the optical configuration shown in FIG. 2 where a short axis field stop 36 is used. Thus, FIG. 3A shows that the variation in beam intensity when the beam is out of focus is more pronounced than the corresponding change in beam width, as described above in the Background section. For membrane 12, it may be desirable to maintain the energy density in a preselected range. More specifically, energy density control in the film 12 can be achieved over a certain range of focus conditions by changing the characteristics of the laser pulse, eg, pulse energy, with small changes in beam width. .

With the above in mind, Applicants disclose a system and method for maintaining the energy density in a preselected range in the membrane 12, for example, by changing the pulse characteristics to compensate for changes in focus conditions. . This change in focus condition may occur, for example, during the scanning operation of the stage 30 with respect to the line beam. More specifically, the energy density obtained with a slightly out of focus beam (eg, plot 66) can be selected as the target energy density. With this target, focus conditions are measured using, for example, the detection system shown in FIG. It will be appreciated that alternative methods of determining focus conditions can be used to include an autofocus sensor (active or passive) or other suitable technique that can measure the distance between the membrane 12 and the focusing optics 37. And When the focus condition is measured, pulse characteristics such as pulse energy can be changed to reach the target energy density. Therefore, when it is found from the measurement result that an out-of-focus condition exists, the first pulse energy E ₁ corresponding to the target energy density of the out-of-focus condition is used. On the other hand, if the measurement results show that the film is in the DOF, the second pulse energy E ₂ corresponding to the target energy density of the focus condition is used, and E ₁ <E ₂ .

FIG. 3B shows a graph showing the energy density as a function of the lateral growth length of a 50 nm thick Si film and the region where partial melting and agglomeration may occur. FIG. 3 also shows that the energy range of lateral growth may be quite wide (about 450 mJ / cm ² and 820 mJ / cm ² ) with the lateral growth length increasing in proportion to the energy density. ing. Nevertheless, larger lateral growth lengths are considered useful by allowing larger scan pitches (and higher throughput) while melting the central ridge again.

  Several methods can be used to adjust the pulse energy between pulses in some cases as needed. For example, in the case of an excimer laser source, a preselected pulse energy can be achieved by changing the discharge voltage. Alternatively, an adjustable attenuator can be positioned along the beam path of the line beam to selectively change the pulse energy. For this purpose, any apparatus known in the art for reducing pulse energy, including but not limited to filters and pulse trimmers, can be used. Other pulse characteristics that can be varied to correct the focus condition and maintain the energy density at a preselected range at different locations on the film 12 are not limited to the following, but are adjustable, for example: A pulse spectrum (wavelength) using a line narrowing module or a line selection module can be included. Alternatively, an adaptive optical instrument capable of fast focus control can be used as the focusing optical instrument 37 responsive to the measurement focus condition in the control feedback loop.

  4 and 5A-5Q illustrate the system and corresponding method used to position the membrane 12 for interaction with a line beam focused from a laser source. As shown in FIG. 4, an exemplary positioning configuration is positioned on or formed as an integral part of a so-called ZPR table 102 (see FIG. 5A) that can include a movable wedge assembly, for example. For those skilled in the art, a vacuum chuck 100 having a substantially planar surface 101 (also shown in FIG. 5A) that is planar within manufacturing tolerances may be included. 4 and 5A, the ZPR table 102 has a Z direction, a pitch direction in which the chuck 100 is rotated around the X axis, and a rolling direction in which the chuck 100 is rotated around the Y axis. It can be assumed that the vacuum chuck 100 can be moved back and forth independently. 5A is a schematic view, the system of the present invention includes an X stage 104 that moves the vacuum chuck 100 back and forth in the X direction, and a Y stage 106 that moves the vacuum chuck 100 back and forth in the Y direction. It can be included. In a typical and exemplary setting, X, Y, and Z are three mutually orthogonal axes. As shown, both stages 104, 106 are relative to a stable reference block 108, eg, a granite block, that forms a substantially planar reference surface (eg, flat within manufacturing tolerances for those skilled in the art). And can be moved. In general, air bearings can be employed between the stages 104, 106 and the granite block 108.

  As best seen in FIG. 5B, the system of the present invention is fixedly attached to the granite block 108 through three autofocus sensors 112a-112c, eg, the overhead housing 114, for the illustrated embodiment. A plurality of optical sensors can be included that are active or passive autofocus sensors. As shown, the three autofocus sensors 112A-112C are spaced along the X axis. In general, the three autofocus sensors 112A-112C can be positioned along or on a line parallel to the X axis. Further, as shown, each autofocus sensor 112a-112c is oriented to measure a distance, such as a distance 116, parallel to the Y axis between the respective autofocus sensor 112a-112c and the surface 101. . Thereby, a distance is obtained between the surface 101 and the reference plane 110 that is parallel to the Y axis. Although three light sensors are shown, it will be appreciated that a system having more than three and only one light sensor can be employed to implement some or all of the functional aspects detailed below. Shall.

  5B-5E illustrate how the system of the present invention can determine the roll angle α between the surface 101 and the reference surface 110. Specifically, starting from FIG. 5B, the first measurement (distance 116) can be performed between the autofocus sensor 112A and the surface 101 with the table 102 in the first position along the X axis. I understand. Next, as shown in FIG. 5C, the table can be positioned at the second position along the X axis by translation of the table 102 along the X axis by the operation of the X stage. At this second position, a second distance measurement can be performed between the autofocus sensor 112A and the surface 101. Although two measurements are sufficient, FIG. 5D shows that the system of the present invention can make a third measurement with the table in a third position along the X axis. These measurement results can then be processed in the algorithm to determine the roll angle α between the surface 101 and the reference surface 110 as shown in FIG. 5E. Note that the table 102 is moved along the Y axis and the same procedure (not shown) is performed to determine the inclination angle of the surface 101 with respect to the reference plane.

  With the roll angle α (and the tilt angle if necessary) determined, the ZPR table 102 is selectively actuated to move the surface 101 until it is substantially parallel to the reference plane 110 as shown in FIG. 5F. Can be moved. At this point, the stage coordinate system can be established. Further, as shown in FIG. 5G, the three autofocus sensors 112a-112c can be calibrated for linearity over the distance to the surface 101 and the measurement range. With this calibration, the surface 101 can be used as a reference (eg, autofocus reference plane) for future measurements.

  In one implementation of the system of the present invention, the spatial position and orientation of the focused line beam of the laser can be determined. An exemplary focused beam that can be characterized by a substantially linear line beam axis 118 is shown as a dotted line in FIG. 5H. For the system shown, the pulsed laser light reaches the beam axis above and before the overhanging housing 114. Further, a portion of the ZPR table 102 extends outwardly from the overhanging housing 114 along the Y axis so that at least a portion of the surface 101 of the table 102 can be exposed to the pulsed laser.

  As further shown in FIG. 5H, the system of the present invention may be a line beam camera 120 that measures positions for a plurality of fine beam focus positions (eg, best focus positions) for the illustrated embodiment. Possible detectors can be included. More specifically, as shown, the line beam camera 120 can be mounted on the table 102 and thus can be moved with the table 102. It will be appreciated that a configuration having multiple line beam cameras (not shown) can be used to measure multiple line beam focal positions without movement of the X stage.

  FIGS. 5I to 5L illustrate how the system of the present invention having one camera 120 can determine the spatial position of the beam axis 118 and the relative angle φ between the surface 101 and the beam axis 118. . More specifically, the table 102 is in a first position along the X axis between the beam axis 118 and the reference plane 110 by the camera 120 starting from FIG. 5I and representing the parallel (distance 122a) to the Y axis. It can be seen that the first measurement can be performed in the state. Next, as shown in FIG. 5J, the table 102 can be translated by the operation of the X stage 104, and the table can be positioned at the second position along the X axis. At this second position, the second distance measurement 122a can be performed by the camera 120 in parallel with the Y axis. While perhaps two measurements are sufficient, FIG. 5K shows that the system of the present invention can make a third measurement (distance 122c), for example, with the table in a third position along the X axis. Is shown. These measurement results can then be processed in an algorithm to determine the relative angle φ between the surface 101 and the beam axis 118 as shown in FIG. 5L.

  With the relative angle φ between the surface 101 and the beam axis 118 determined, the ZPR table 102 is selectively actuated to move the table 102 so that the surface 101 is substantially aligned with the beam axis 118 as shown in FIG. 5M. To an alignment that is parallel to (eg, within tolerances acceptable to those skilled in the art). In alignment, FIG. 5N uses autofocus sensors 112a-112c to measure the position of surface 101 (ie, autofocus reference plane) and calibrate autofocus sensors 112a-112c on the autofocus reference plane. It shows that you can. It then establishes the laser and stage coordinate system.

  FIG. 5O shows that the glass substrate 32 and the deposited film 12 can now be positioned on the vacuum chuck (ie, surface 101). As shown in FIG. 5O, the X stage 104 can be actuated to move the table 102 to a position that is advantageous to assist in positioning the film 12 on the surface 101. With the membrane positioned on the table 102, the table 102 can be moved relative to the interaction with the autofocus sensors 112a to 112c, as shown in FIG. 5P. Therefore, the height of the film 12 can be determined using the autofocus sensors 112a to 112c. With the film 12 at a known height, the table 102 can be actuated to move the film 12 within the depth of focus (DOF) of the focused line beam, as shown in FIG. 5Q. With the film 12 in the DOF of the laser, the laser can be operated to expose and melt the strip-shaped film 12 as part of the fine beam continuous lateral solidification (tb-SLS) described above, for example.

  In another aspect of embodiments of the present invention, the system of the present invention illustrated in FIGS. 5A-5Q can be used to correct a film 12 having an imperfect non-planar surface. This surface profile variation may result from dimensional imperfections in the film 12, the glass substrate 32, and / or the vacuum chuck surface 101. By correcting for surface profile variations, a substantially constant energy density can be maintained at different locations on the membrane 12. For this purpose, as shown in FIG. 5P, the method uses three autofocus sensors 112a to 112c to determine the three respective distances between the sensor and the membrane 12 parallel to the Y axis. A first stage of doing can be included. A line beam camera 120 can be used to manually adjust the ZPR table 102 (by changing Z, pitch, roll) to position the surface 101 along the best focus line (ie, beam axis 118). it can. The respective distances between each autofocus sensor 112a-112c and the membrane 12 can then be stored as a reference distance, thus obtaining three coordinate points on the membrane 12. A linear fit through these three coordinate points can be used to determine the calculated best focus line (axis 118). During exposure, when the film 12 is scanned along the Y axis, the distance to the film 12 can be measured, for example, by three autofocus sensors 112a to 112c, thus obtaining three new coordinate points. The best fit line through these new coordinate points can then be calculated and the ZPR table 102 is adjusted through computer control so that the best fit line substantially coincides with the calculated best focus line (axis 118). The table 102 can be aligned to match (eg, match within acceptable tolerances to those skilled in the art).

  FIG. 6 shows a portion of the system 10 shown in FIG. 1 in more detail. Specifically, FIG. 6 illustrates an exemplary embodiment having a two-chamber excimer laser 20. Other types of laser sources are used in the system 10, semiconductor lasers, excimer lasers having one chamber, more than two, eg, an oscillation chamber and two amplification chambers (amplification chambers in parallel or in series). It should be appreciated that a semiconductor laser can be included that provides seed light to one or more excimer amplification chambers. Other designs are possible. Further details of the two-chamber laser source 20 shown in FIG. 6 can be found in US patent application Ser. No. 10 / 631,349, 2003 entitled “Control System for Two-Chamber Gas Discharge Laser” filed July 30, 2003 US patent application Ser. No. 10 / 356,168, entitled “Automatic Gas Control System for Gas Discharge Laser”, filed on Jan. 31, 2003, “Output of Gas Discharge Laser MOPA System” filed on Dec. 18, 2003 US patent application Ser. No. 10 / 740,659, entitled “Method and Apparatus for Controlling”, US Patent Application Ser. No. 10/740, entitled “Gas Discharge MOPA Laser Spectrum Analysis Module”, filed Sep. 30, 2003. No. 676,907, filed Sep. 30, 2003 "Optics for Gas Discharge MOPA Laser Spectrum Analysis Module" US patent application Ser. No. 10 / 676,224, filed Sep. 30, 2003, US patent application Ser. No. 10 / 676,175 filed Sep. 30, 2003 No. 10 / 631,349, filed July 30, 2003, entitled "Control System for Two-chamber Gas Discharge Laser", "Ultra-Narrow Band Two Chamber" filed July 24, 2003 US Patent Application No. 10 / 627,215 entitled “High Repetition Number Gas Discharge Laser”, US Patent Application No. “Method and Apparatus for Cooling Magnetic Circuit Elements”, filed Jun. 25, 2003 No. 10 / 607,407, filed August 20, 2004 entitled "Timing Control for Two Chamber Gas Discharge Laser System" US patent application Ser. No. 10 / 922,692, US Pat. No. 6,625,191 entitled “High Repetition Number MOPA Laser System”, and US Pat. No. 6,625,191 entitled “Basic Modular MOPA Laser System”. No. 6,567,450, the entire disclosure of these patents is hereby incorporated by reference herein.

  As an overview, FIG. 6 shows that the two-chamber laser source 20 can include a master oscillator 208 and a power amplifier 210 and is therefore often referred to as a so-called MOPA laser source. In one implementation of the tb-SLS method described above, a 6 Khz (6000 pulses / second) MOPA laser can be used with a pulse energy of about 150 mJ. With this configuration, a 730 mm × 920 mm membrane can be processed for about 75 seconds (with 60% overlap).

  Master oscillator 208 and power amplifier 210 each have two elongated electrodes, a laser gas, eg, XeCl, XeF, ArF, or KF, a cross-flow fan that circulates a gas discharge laser between the electrodes, and one or more. It includes a discharge chamber that can accommodate a number of water-cooled finned heaters (not shown). For example, the master oscillator 208, which can be amplified by two passes through the power amplifier 210 to produce the laser beam 214B, produces the first laser beam 214A. The master oscillator 208 can include a resonant cavity formed by the output coupler 208A and a line narrowing module 208B, both of which are described in detail in previously referenced applications and patents. The gain medium for the master oscillator 208 can be generated between two electrodes, each 30 to 50 centimeters long and housed in the master oscillator discharge chamber.

  The power amplifier 210 can include a discharge chamber similar to the discharge chamber of the master oscillator 208 that provides a gain medium between the two elongated electrodes. However, unlike the master oscillator 208, the power amplifier 210 generally does not have a resonant cavity, and the gas pressure can generally be maintained higher than the gas pressure of the master oscillator 208. With the MOPA configuration shown in FIG. 6, the master oscillator 208 can be designed and operated to maximize beam quality parameters such as wavelength stability and obtain ultra-narrow bandwidth, while power amplifier 210. Can be designed and operated to maximize power output.

  The output beam 214A of the main oscillator 8 can be amplified by, for example, two passes through the power amplifier 210 to generate the output beam 214B. The optical components that accomplish this are the three that the Applicant named the main oscillator wavefront engineering box “MO WEB” 224, the power amplifier wavefront engineering box “PA WEB” 226 ”, and the beam inverter“ BR ”228. Can be housed in a module. Together with the line narrowing module 208B and the output coupler 208A, these three modules can all be mounted on a single vertical optical bench that is independent of the discharge chamber 208C and the discharge chamber of the power amplifier 210. With this configuration, chamber vibrations caused by acoustic shock and fan rotation can be substantially isolated from the optical components.

  The optical components within the line narrowing module 208B and output coupler 208A are described in more detail in previously referenced applications and patents. In summary, the line narrowing module (LNM) 208B can include a three-point or four-point prism beam expander, an ultrafast response tuning mirror, and a grating arranged in a Littrow configuration. The output coupler 208A may be a partially reflective mirror that typically reflects about 20% of the output beam for KrF systems and about 30% for ArF systems. The remaining non-reflected light passes through the main oscillator 208 and enters the line center analysis module (LAM) 207. From LAM 207, light can enter into “MO WEB” 24. The “MO WEB” may include a total internal reflection (TIR) prism (or a first surface mirror with a high reflectivity coating) and an alignment component that accurately directs the output beam 214A into the “PA WEB” 26. .

  The “PA WEB” 226 may include a TIR prism (or a first surface mirror with a high reflectivity coating) and an alignment component that directs the laser beam 214A to a first pass through the power amplifier gain medium. The beam reversal module 228 can include a two-reflection beam reversing prism that relies on total reflection and therefore does not require an optical coating. Alternatively, the beam inverter 228 can be a total reflection mirror. In either case, beam inverter 228 is adjustable in response to a control signal from a measurement device, eg, SMM 26, to direct the partially amplified beam on a preselected beam path through the power amplifier gain medium. Can do. In particular, the beam inverter can be adjusted to correct for beam pointing errors and reduce the beam divergence of the beam exiting the pulse stretcher 22 as described above.

After inversion in beam inverter module 228, partially amplified beam 214A can pass through the gain medium in power amplifier 210 again and exit through spectrum analysis module 209 and “PA WEB” 226 as power amplifier output beam 214B. . From “PA WEB” 226, the beam, for example, increases pulse duration, reduces intensity variation across the beam section (ie, averages or smoothes the intensity profile), and reduces beam interference, as detailed below. A 6-mirror pulse stretcher 22 can be entered. By increasing the pulse duration, the peak intensity of each laser pulse is reduced while maintaining the pulse energy. For the system 10 shown in FIG. 1, the optical components in the optical instrument module 28 may include relatively large lenses that are difficult to manufacture and expensive to manufacture. These expensive optical components are often prone to degradation resulting from billions of high-intensity ultraviolet pulses. Furthermore, optical damage is known to increase with increasing laser pulse intensity (ie, optical power per cm ² (energy / time), or mJ / ns-cm ² ). Thus, reducing the pulse intensity by increasing the pulse duration can extend the lifetime of these optical components. Furthermore, extending the pulse duration is believed to be useful in the crystallization process. Instead of or in addition to the 6-mirror pulse stretcher 22, a current patent-pending U.S. Patent Application No. 10/10 entitled “Long Delayed High TIS Pulse Stretcher” filed on November 13, 2003. One or more of the pulse stretchers disclosed in US Pat. Nos. 712, 545 are used to stretch pulses for use in the thin beam continuous lateral solidification (tb-SLS) method described herein. In particular, a pulse stretcher with 200 ns time integral square (TIS) output pulses can be used. US patent application Ser. No. 10 / 712,545 is hereby incorporated by reference herein.

  FIG. 7 shows a more detailed view of the six mirror pulse stretcher 22 showing the beam path through the pulse stretcher 22. The beam splitter 216 can be selected to reflect a predetermined percentage of the partially amplified beam 214A to the delay path created by the six focusing mirrors 320A, 320B, 320C, 320D, 320E, and 320F. The remaining light is transmitted through the beam splitter 216. It will be appreciated that the reflection / pass characteristics of the beam splitter may affect the output pulse duration and / or the output pulse shape. For stretcher 22, each mirror 320A-320F may be a focusing mirror, such as a concave spherical mirror. In general, one or more of the six mirrors can be adjustable, eg, tip / tilt adjustable, to assist in alignment of the pulse stretcher 22.

  As shown in FIG. 7, the reflected light from the beam splitter 216 can travel in an unfocused state (ie, substantially collimated) along the optical path 301A to the mirror 320A, and then the mirror 320A The reflection portion is focused on a point 302 located in the middle of the mirror 320A and the mirror 320B along the optical path 301A. The beam can then be expanded and reflected from mirror 320B, which converts the expanding beam into a collimated (ie, substantially collimated) beam to mirror 320C along optical path 301C. Lead. Next, the mirror 320C can focus the reflection portion along the optical path 301D at a point 304 that can be positioned in the middle of the mirror 301C and the mirror 301D. The beam can then be expanded and reflected from mirror 320D, which converts the expanding beam into a collimated (ie, substantially collimated) beam to mirror 320E along optical path 301E. Lead. Next, the mirror 320E can focus the reflection portion along the optical path 301F to the mirror 301E and a point 306 located in the middle of the mirror 301F. The beam can then be magnified and reflected from mirror 320F, which converts the beam being magnified into a parallel (ie, substantially collimated) beam that is beam splitter along optical path 301G. Lead to 216. Beam splitter 216 can reflect the beam from mirror 320F onto optical path 301H, where the beam combines with the portion of the pulse transmitted through beam splitter 216. The pulse portion transmitter and delay are combined to establish a pulse stretcher beam output 214C as shown. The stretch pulse 400 is plotted in intensity versus time in FIG. 8, and again can be compared to the shape of the power amplifier output pulse 402 (non-stretch pulse) plotted in FIG. For the stretched pulse shown, the pulse can be shaped so that it has two large, approximately equal peaks, with a smaller tapering peak over time after the first two peaks. It will be appreciated that the shape of the stretched pulse can be modified using beam splitters having different reflectivities.

  FIG. 7 shows that the delayed beam can be focused and expanded separately three times. Because of this odd number of (ie, not even) focusing steps, the delayed beam is inverted (in both horizontal and vertical directions) with respect to the portion of the pulse transmitted through beam splitter 216. Thus, the output beam 214C from the six mirror pulse stretcher 22 can include a combined beam or a mixed beam. By this mixing, intensity fluctuation can be reduced. The pulse stretcher 22 can also mix different interfering cells from different parts of the beam, thus reducing beam interference. The effect on the vertical uniformity of an exemplary beam is shown in FIG. Specifically, the stretched pulse 404 is plotted in FIG. 9 as a width in the direction perpendicular to the intensity, and again, the shape of the power amplifier output pulse 406 (non-stretched pulse) plotted in FIG. Can be compared. If the beam has a near Gaussian distribution in the horizontal axis, which is often common when using an excimer laser source, the effect of the pulse stretcher 22 on the horizontal intensity variation may be negligible.

  As described above, the performance of the laser crystallization process may depend on the uniformity of the energy density. Unlike lithographic lasers, which are multiple firing processes that provide averaging between exposures during exposure, laser crystallization is primarily a one-shot process, so averaging is an intensity average within a single pulse. May be limited to Some of the factors that determine energy density uniformity are line beam uniformity and beam space interference. In general, the optical instrument can be included in the optical instrument module 28 (FIG. 1) to homogenize the beam. In these optical instruments, it is believed that an array of microlenses is used to split the beam into beamlets. A large aperture lens can be used to redirect the beamlets so that the beamlets overlap each other exactly in the focal plane of the spherical lens. When these beamlets are integrated, any intensity variation can be substantially eliminated and a flat top beam profile is obtained. The averaging can be improved as the beam is split into beamlets. However, if the microlens aperture is too small, for example smaller than one interference area of the line beam, there may be a repetitive pattern of microlenses in the interference area, which will produce undesirable results. . In short, the amount of homogenization achieved using an array of microlenses can be limited. With this in mind, if the spatial interfering cells in the pulse stretcher 22 are averaged, the beam delivered to the microlens array will result in inferior interference, which in turn is It will minimize intensity fluctuations due to interference and / or allow miniaturization of the microlens array aperture.

  One feature to note of the pulse stretcher 22 shown in FIG. 7 is that the beam divergence of the output beam (ie, beam 214C) may increase when the beam pointing error of the input beam (ie, beam 214B) increases. That is the point. This increase in beam divergence is often undesirable for laser crystallization, and therefore it is desirable to minimize the beam pointing error of the beam entering the pulse stretcher (ie, beam 214B). FIG. 6 shows that the active beam guidance unit 500 can be positioned upstream of the pulse stretcher 22 to minimize the beam pointing error of the beam 214B entering the pulse stretcher. The active beam guidance unit may be responsive to beam directing measurements performed upstream of the pulse stretcher 22 and / or divergence measurements performed downstream of the pulse stretcher 22, for example, , SMM 26 and can be used to control the active beam guidance unit 500. Structurally, the active beam guidance unit 500 is described in more detail below and in several applications previously incorporated herein by reference to actively control beam guidance in the beam delivery unit 238. One or more adjustable mirrors similar to the described mirrors 240A, 240B can be included. Alternatively or in addition to the active beam guidance unit 500, the orientation of the beam inverter 228 can be actively adjusted to control the beam orientation upstream of the pulse stretcher 22. Specifically, the adjustable beam inverter 228 may be responsive to beam pointing measurements performed upstream of the pulse stretcher 22 and / or divergence measurements performed downstream of the pulse stretcher 22. it can.

  FIG. 6 shows that the system 10 can include a beam delivery unit 24 and a stabilization method module (SMM 26). Functionally, these elements can cooperate with laser source 20 and pulse stretcher 22 to provide a pulsed beam at the output of SMM 26 that meets the set of beam specifications of the present application. In practice, the beam specifications at the input of the optical instrument module 28 (see FIG. 1) may depend on the design of the optical instrument module 28 (illuminator). Specific beam parameters include, but are not limited to, intensity, wavelength, bandwidth, wavefront (eg, also called wavefront curvature, beam divergence), polarization, intensity profile, beam size, beam pointing, dose Stability and wavelength stability can be included. For optical instrument modules capable of generating, for example, the 20 micron x 700 mm line beam described above for laser crystallization, directional stability to within 20 μrad, wavefront curvature change to less than 10%, and ± 2% It may be necessary to maintain energy stability to within. Furthermore, it would be desirable to obtain these characteristics without having to operate continuously over a relatively long period of time until the laser is “stable” to avoid wasted firing.

  The SMM 26 monitors the incident light and provides a feedback signal to the control system to ensure that the light is delivered to the optics module 28 with desirable parameters including beam orientation, beam size, wavefront, and pulse energy. It can be positioned upstream of the input port of the optical instrument module 28 so that it can. For example, pulse energy, beam orientation, and beam position may be achieved using techniques described in US patent application Ser. No. 10 / 425,361 (the '361 application) previously incorporated herein by reference. Thus, it can be monitored between the pulses by the measuring instrument in the SMM 26. Specifically, FIG. 10B of the '361 application shows a structural arrangement for measuring pulse energy, beam pointing, and beam position between pulses. As will be described in more detail below, the SMM 26 can also be configured to monitor wavefront curvature and beam size. Using a DSP-based processor in combination with a high-speed CMOS linear photodiode array can allow beam characteristics to stabilize with rapid calculation of beam characteristics at up to 8 kHz as well as rapid feedback. .

  Vertical and horizontal beam pointing and position errors can be evaluated with the SMM 26 for all pulses of light generated by the laser. Overall, there are four types of independent sensor measurements: vertical pointing error, horizontal pointing error, vertical position error, and horizontal position error. In one exemplary implementation, vertical and horizontal orientation measurements are made with linear photodiode arrays (such as the “S903 NMOS linear image sensor” provided by “Hamamatsu Corporation”, Bridgewater, NJ, USA). This can be done by placing a far field image on a PDA) device. In general, the pointing error can be defined from the target position defined at the exit of the SMM 26. Measurements of vertical and horizontal positions can be made by placing a reduced image of the beam near the BDU exit on the PDA element. Measurement of the pulse energy of the beam can be performed with the SMM 26 in a calibrated photocell circuit. Signals from sensors in the SMM 26 can be sent through electrical connectors to a “stabilization controller” that can form part of the SMM 26.

  Beam pointing control is achieved by using an active beam guidance unit 500 upstream of the pulse stretcher 22 (described above) and / or within the BDU 24 and selectively adjusting the orientation of the beam inverter 228 (also described above). Can do. Specifically, the BDU 24 can include two beam directing mirrors 240A and 240B, which can control one or both and perform tip and tilt correction to change the beam pointing. The beam pointing can be monitored within the SMM 26 by feedback controlling one or both of the beam directing mirrors 240A, 240B. For example, the error signal can be sent to a stabilization controller in the SMM 26 that processes the raw sensor data and generates commands to drive the high speed induction rotating mirrors 40A and 40B. These two high speed induction rotating mirrors, each having two control axes, can be placed upstream of the SMM 26 as shown. Each of the rotating mirrors can be attached to a high speed induction motor. In a specific embodiment, a piezoelectric mirror drive can be installed to allow rapid (200 Hz) beam pointing and position correction.

  The motor operates the mirror angle in two axes and can thus redirect the optical path of the laser beam. Two motors with two control axes can allow the BDU stabilization controller to independently adjust vertical and horizontal beam pointing and position errors. The control system can correct the beam error between pulses. That is, the beam error from each laser pulse can be supplied to the feedback control system to generate a command for the induction motor. The electronics used to operate the feedback control system can be located in a “stabilizing controller”. By arranging a mirror as shown in FIG. 6, drift due to lasers, attenuators (if provided), and other optical instruments can be corrected. Thus, in some cases, a stable beam (in position and orientation) can be projected to the entrance of the optical instrument module 28 having a stability of up to 10 μrad.

The pulse energy monitored by the SMM 26 can be used as a feedback signal and input to the laser energy control algorithm. For gas discharge lasers, the pulse energy can be varied by adjusting the discharge voltage of the laser. Because the energy control algorithm can stabilize the energy at the SMM 26 (at the input of the optical instrument module 28), it can correct any short-term or long-term drift in pulse energy due to light absorption or other causes. it can.
As described above, the SMM 26 can also measure beam size and beam divergence (wavefront curvature). In general, the beam size from the laser can be fixed using the opening of the laser exit. However, the beam divergence from the laser may vary due to optical instrument heating, laser energy, laser voltage, and F2 concentration in the discharge gas when using a fluorine excimer laser.

  As shown in FIGS. 6 and 7, beam divergence and beam size can be actively controlled using an adjustable beam expander 502 that can be positioned along the BDU 24. As shown in FIG. 2, the beam expander 502 can include four lenses: two horizontal lenses 504A, 504B and two vertical lenses 504C, 504D. In one configuration, the beam expander 502 has a length L of about 0.30 m and has a nominal input of 12 mm on the horizontal axis, 9 mm on the vertical axis, 5 mm on the horizontal axis, and 18 mm on the vertical axis. The size can be determined to have an output section. In an exemplary configuration, the lens 504A can be a plano-convex cylindrical lens with f = 507.0 mm, the lens 504B can be a plano-convex cylindrical lens with f = 202.8 mm, and the lens 504C has F = 202.8 mm and a plano-convex cylindrical lens, and the lens 504D can be a plano-convex cylindrical lens with f = 405.6 mm. In an alternative configuration, lenses 504A and 504C can instead use a single lens. Beam divergence and beam size changes can be achieved by adjusting the spacing of the beam expander lenses. Specifically, the spacing between lenses 504A and 504B can be changed to change the beam on the horizontal axis, and the spacing between lenses 504C and 504D can be changed to change the beam on the vertical axis. . In one embodiment, the movable lens can be mounted on a linear electric drive. Thus, the expander 504 can allow independent control of the horizontal and vertical beam wavefronts.

  The aspects of the embodiments of the invention disclosed above are intended to be preferred embodiments only, and in any respect are intended to limit the disclosure of the invention to particular preferred embodiments only. It will be appreciated by those skilled in the art that this is not the case. Many changes and modifications may be made to the disclosed aspects of embodiments of the disclosed invention that will be understood and appreciated by those skilled in the art. The claims are intended to encompass, within their scope and meaning, not only the disclosed aspects of the embodiments of the present invention, but also equivalents and other modifications and changes that would be apparent to one skilled in the art. . "Interaction between a laser shaped as a line beam and a film deposited on a substrate, as described and shown in this patent application in the details required to meet" 35 USC § 112 " Specific aspects of the embodiments of the "systems and methods for realizing" are any of the above-mentioned objectives of the above-described aspects of the embodiment or aspects of the above-described embodiments to the problem solved by the above-described embodiments. Although any other reason or purpose may be adequately achieved, the described aspects of the described embodiments of the invention are merely exemplary and broadly contemplated by the invention It will be understood by those skilled in the art that The scope of the presently described and claimed aspects of the embodiments is intended to encompass completely other embodiments that are presently apparent or will be apparent to those skilled in the art based on the teachings herein. . The scope of the present invention “systems and methods for realizing interaction between a laser shaped as a line beam and a film deposited on a substrate” is only and completely limited by the claims, Nothing exceeds the claims of the claims. Reference to an element in such a claim in the singular means "one and only one" in the interpretation of such claim element, unless explicitly stated otherwise. It is not intended and should not be meant, but rather shall mean "one or more". All structural and functional equivalents to any of the above-described aspects of embodiments known to those of ordinary skill in the art or later become known are incorporated herein by reference and are expressly claimed. It is intended to be covered by a range. Any term used in the specification and / or claims and expressly given meaning in the specification and / or claims in this application shall have the dictionary or It shall have such meanings regardless of other commonly used meanings. The apparatus or method described herein as any aspect of an embodiment is intended to solve each and every problem sought to be solved by the aspects of the embodiment disclosed herein, as it is encompassed by the claims. It is not intended or necessary to deal with. Any element, component, or method step of the disclosure of the present invention is generally given regardless of whether the element, component, or method step is explicitly recited in the claims. Not intended. Any element of a claim in the claims, unless that element is described as a "step" instead of "action" in the case of a method claim, or in the case of a method claim “35 USC § 112” shall not be construed under the provisions of paragraph 6.

Claims (5)

A system for maintaining energy density in a preselected range in the membrane during the interaction of the membrane with a line beam shaped from a light source by a shaping optic;
The membrane has an imperfect non-planar surface;
The system
An autofocus sensor for measuring the distance between the membrane and the focusing lens;
A controller that adjusts light source parameters utilizing the measurement to maintain the energy density in the preselected range at the film;
A system characterized by including.   The system of claim 1, wherein the light source is a pulsed laser source and the light source parameter is pulse energy.   The system of claim 2, wherein the laser source is a gas discharge laser source and the pulse energy is adjusted by changing a discharge voltage of the laser source.   The system of claim 1, wherein the controller adjusts a variable attenuator to maintain the energy density in the membrane at the predetermined range.   The system of claim 1, wherein the light source generates a light source spectrum, and the controller changes the light source spectrum to maintain the energy density in the predetermined range at the film.

Images