JP2009244616A - Laser direct drawing method and laser direct drawing device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a laser direct drawing method and device which has an LD as a light source and is capable of drawing an inexpensive DFR requiring a light amount of 20 to 100mJ/cm<SP>2</SP>. <P>SOLUTION: In the laser direct drawing device wherein an object to be drawn is moved in a sub-scanning direction while deflecting laser beams arrayed in a line direction and a column direction, in a main scanning direction and the laser beams modulated on the basis of raster data are condensed in the sub-scanning direction using a cylindrical lens to draw a desired pattern on the object, a setting means which sets the number of optical axes of laser beams in the line direction to p which is a positive integer and is a product of q and s and which sets a light amount in a drawing position and a rotating means which rotates the cylindrical lens in a matrix plane are provided, and the position of the cylindrical lens is determined in accordance with the light amount set by the setting means, and laser beams are radiated to the object while shifting s line at a time in the line direction, so that q-exposure of a light quantity in the drawing position which is q times single exposure is carried out. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention moves the drawing object in the sub-scanning direction while deflecting the optical axis of the laser beam in the main scanning direction with respect to the drawing object, and modulates (ON / OFF) the laser beam based on the raster data. The present invention relates to a laser direct drawing method and a laser direct drawing apparatus (LDI: Laser Direct Imaging) for drawing a desired pattern.

  The laser direct drawing device converts the design data (CAD data) at the time of circuit pattern design into vector data, finds the ON / OFF pixel of the laser beam from each intersection of the contour line and the pixel unit, and converts it into raster data for drawing. Then, a desired pattern is drawn on the drawing object by irradiating a laser beam based on the obtained raster data.

  Hereinafter, a conventional laser direct drawing apparatus for drawing using a laser beam having a plurality of laser diodes (hereinafter referred to as “LD”) as a light source will be described.

  8 is a perspective view of a conventional laser direct drawing apparatus, FIG. 9 is a schematic view of a light source system, (a) is a side view, (b) is a front view, and (c) is an enlarged view of a reduced laser beam. It is.

  As shown in FIG. 9B, the plurality of LDs 1 serving as light sources are arranged in a parallelogram array. In the case shown in the figure, 16 rows are arranged at intervals Py in the vertical direction (row direction) in the figure, and 8 columns are arranged at intervals Px in the left-right direction (column direction) in the figure. Further, the LDs 1 in the vertical direction are arranged so as to be shifted from each other by Px / 16.

  As shown in FIG. 8, the laser beam 4 emitted from each LD 1 is transmitted through the collimator lens 2 and then reduced by a lens (not shown). The laser beam is reduced by the prism 3 in the row direction to Py / 16. It becomes bundle 4. The laser beam bundle 4 is reflected by the corner mirrors 5a and 5b, then reduced by the imaging lens 6, enters the polygon mirror 7 through the corner mirror 5c, is deflected and scanned by the rotation of the polygon mirror 7, and is fθ lens. 8 is incident. The scanning direction by the polygon mirror 7 is the main scanning direction (X direction in the figure). Then, the laser beam bundle 4 transmitted through the fθ lens 8 is directed downward by the folding mirror 9, transmitted through the cylindrical lens 10, and DPI determined on the photosensitive member such as DFR or photoresist on the drawing object 11. (Dot Per Inch. 6.35 μm in 4000 DPI) The incident light is incident at a pitch (see FIG. 9C), and the photosensitive member is drawn (exposed). At this time, the table 12 on which the drawing object 11 is mounted is moved at a constant speed in the sub-scanning direction (Y direction in the figure) by the linear motor 13 and the linear guide 14. (Patent Document 1).

  Here, the front focal point of the fθ lens 8 is positioned on the reflection surface of the polygon mirror 7, and the component parallel to the XY plane of the laser beam 4 reflected by the polygon mirror 7 is parallel, and the component perpendicular to the XY plane is The diffused light starts from the reflection point of the polygon mirror 7. Therefore, the component parallel to the XY plane of the laser beam 4 is collected by the fθ lens 8, but passes through the cylindrical lens 10 as it is. On the other hand, the component perpendicular to the XY plane of the laser beam 4 is converted into parallel by the fθ lens 8 and condensed by the cylindrical lens 10.

  10A and 10B are diagrams showing the arrangement of the start sensor, where FIG. 10A is a diagram viewed from the X direction in FIG. 8, and FIG. 10B is a diagram viewed from the Y direction in FIG.

  As shown in FIG. 8, a reflection mirror 15 and a start sensor 16 are disposed below the left side of the cylindrical lens 10 in FIG. Then, in order to align the scanning start position for each row, drawing of one scan in the main scanning direction is started after a predetermined time after the start sensor 16 detects the laser beam 4 reflected by the mirror 15 (in the illustrated case, The distance between the detection position and the drawing start position is 10 mm).

  In FIG. 11, the laser beam emitted from the LD 1 arranged in a 4 × 4 parallelogram array is deflected by one surface of the polygon mirror 7 and a vertical line k consisting of points K1 to K4 is formed on the drawing object. It is a figure explaining the procedure to draw. Here, the interval between the LDs 1 in each column is 2e, and the LDs in the adjacent upper and lower columns are shifted by the interval e. That is, LDb1 to LDb4 in the second column (in the figure, the LD in the first row is LDa, the LD in the second row is LDb, the LD in the third row is LDc, and the LD in the fourth row is LDd. 1 and 4 are distinguished from each other in the column direction from the left side to the right side.) Are arranged in the middle of the first column LDa1 to LDa4.

  Now, assuming that the laser beam moves to the right in the figure at a speed e / t, the moving speed of the drawing object is ignored.

  If the distance between LDd4 and the vertical line k is e at time T = 0 (FIG. A), the optical axis of LDd4 overlaps with the point K4 at time T = e, so LDd4 indicated by hatching in FIG. Is turned on (FIG. B). Further, at time T = 2e, the optical axis of LDc4 overlaps with point K3, so that LDc4 is turned on (FIG. 3c). At time T = 3e, the optical axis of LDb4 overlaps with point K2, and the optical axis of LDd3 overlaps with point K4, so LDb4 and LDd3 are turned on (d in the figure). Thereafter, similarly, at time T = 10e, the optical axis of LDa1 overlaps with the point K1, so that LDa1 is turned on (k in the figure). That is, each of the points K1 to K4 is irradiated with the laser beam four times. Since the diameter of the laser beam 4 incident on the drawing object is, for example, 0.025 mm, the straight line k drawn with 4000 DPI becomes a straight line with a width of 0.025 mm.

  FIG. 12 shows a case where the object to be drawn is moved in the sub-scanning direction together with the scan (No. 1 scan) in FIG. In addition, the optical axis of each LD1 in No2-No4 scan is the center of the circle which changes and shows the diameter (it represents small sequentially). As shown in the figure, a straight line k composed of points K1 to K16 can be drawn by No1 to No4 scans. In the case of the light source system shown in FIG. 9, each of the points K1 to K16 is irradiated with the laser beam eight times.

Here, the output w of the LD 1 in FIG. 8 is 0.06 watts, the optical system efficiency η is 50%, the polygon mirror has 10 surfaces, the rotation speed R is 5000 rpm, and the focal length f of the fθ lens is. When the thickness is 350 mm, the total amount of light E irradiated to the 1 cm ² area in the case of 4000 DPI is obtained by Equation 1.

E = W × t = 10.31 mJ / cm ² (Formula 1) Here, W is the input energy per 1 cm ² , and t is the irradiation time, and are obtained by Formulas 2 and 3, respectively. W = w × η × number of spots = w × η × (1 cm / 0.00635 mm) ² × 8 (row) = 595, 201.2 (w / cm ² ) (Expression 2) t = tan ⁻¹ (6.35 × 10 ⁻³ /2/350)/30,000°=17.325×10 ⁻⁹ (seconds) (Equation 3)
That is, DFR equivalent to DFR sensitivity of 10 mJ / cm ² can be exposed.
JP2007-94122A

By the way, most commercially available DFRs require ²⁰ to 100 mJ / cm ² as the light quantity, and DFRs with a required light quantity of 10 mJ / cm ² are expensive.

An object of the present invention is to provide a laser direct drawing method and a laser direct drawing apparatus capable of drawing an inexpensive DFR having a required light quantity of 20 to 100 mJ / cm ² even in a laser direct drawing apparatus using an LD as a light source. There is.

  In order to solve the above-mentioned problem, the first means of the present invention is a method in which the object to be drawn is deflected in the main scanning direction which is the same as the column direction while deflecting the optical axis of the laser beam aligned in the column direction and the row direction. In the laser direct drawing method for drawing a desired pattern on the drawing object by irradiating the drawing object with the laser beam modulated based on raster data, moving in the sub-scanning direction that is the same as the row direction, The number of optical axes of the laser beam in the row direction is set as p which is a product of q and s which are positive integers, and the object to be drawn is irradiated with the laser beam while shifting by s rows in the row direction, It is characterized in that q-fold exposure is performed, which is q times that in the case of single exposure.

  Further, the second means of the present invention is configured to deflect the object to be drawn in the same direction as the row direction while deflecting the optical axis of the laser beam aligned in the column direction and the row direction in the main scanning direction which is the same direction as the column direction. The laser beam modulated in accordance with the raster data is condensed in the sub-scanning direction using a cylindrical lens and irradiated onto the object to be drawn, thereby forming a desired pattern on the object to be drawn. In the laser direct drawing apparatus for drawing, the number of optical axes of the laser beam in the row direction is set to p, which is a product of q and s, which are positive integers, and the setting means for setting the amount of light at the drawing position; A rotation unit that rotates the cylindrical lens in the matrix plane, determines the position of the cylindrical lens according to the light amount set in the setting unit, and shifts the column by s rows in the row direction. By irradiating to the object to be drawn body Zabimu, the amount of the drawing position is characterized by q multiple exposure is q times in the case of 1 double exposure.

Even in a laser direct drawing apparatus using an LD with a small output as a light source, it is possible to draw a drawing object having a required light amount of 20 to 100 mJ / cm ² , so that the types of drawing objects that can be drawn can be increased. In addition, since an inexpensive DFR can be used, the total cost of the product can be reduced.

  The present invention will be described below.

  FIG. 1 is a diagram for explaining the gist of the present invention. FIG. 1A shows a first example, and FIG. 1B shows a second example. As in the case of FIG. 11, the interval between the LDs in each column is 2e, and the LDs in the adjacent upper and lower columns are shifted by the interval e. Similarly to the case of FIG. 12, the optical axis of each LD in the No1 to No4 scans is the center of a circle representing the diameters sequentially reduced.

  Assume that the moving speed of the drawing object is halved with respect to the LDs 1 arranged in a parallelogram array of 4 rows × 4 columns. In this case, as shown in FIG. 5A, the points K3 and K4 are irradiated with the laser beam eight times by the first scan and the second scan. That is, twice the amount of light is supplied to the points K3 and K4 as compared to the case of FIG. 11 (single exposure) (double exposure). Similarly to the points K3 and K4, the laser beam is irradiated 8 times for the points after the point K5. That is, twice the amount of light is supplied compared to the case of FIG.

  Further, as shown in FIG. 5B, assuming that the moving speed of the drawing object is ¼, the laser beam is irradiated 16 times for the points after the point K4. That is, four times the amount of light is supplied compared to the case of FIG. 11 (quadruple exposure).

  When the arrangement of LD1 is 8 columns × 16 rows as shown in FIG. 9, it is possible to perform 8-fold exposure and 16-fold exposure by setting the sub-scanning direction feed speed to 1/8, 1/16.

  In actual exposure, the amount of light to be supplied to the object to be drawn (that is, the number of overlapping exposures) is determined first, and the moving speed of the object to be drawn, that is, the table feed speed must be determined according to the result. Hereinafter, the calculation procedure of the table feed speed v will be described. It is assumed that the polygon mirror has 10 surfaces, M has a rotational speed R of 5000 rpm, and the fθ lens has a focal length f of 350 mm, and is drawn at 4000 DPI.

  FIG. 2 is an explanatory diagram of the scanning angle α, and FIG. 3 is a plan view of the cylindrical lens 10.

The rotation time tm of one surface of the polygon mirror is 1.2 × 10 ⁻³ (seconds). If 16 lines are drawn (single exposure) while one surface of the polygon mirror rotates, the table moving distance at this time is 0.1016 mm (0.00635 × 16 lines). v / N = 84.67 / N (mm / second).

At this time, when the exposure width (scan length in the main scanning direction) of the drawing object is Lmm, the effective exposure angle of the polygon mirror is φ, and the rotation angle of the polygon mirror 1 surface is Pm, the scanning angle α is φ = tan ^{− Since 1} (L / 2 / f) (°) and Pm = 360 / M (°), the scanning angle α can be obtained by Equation 4. Therefore, as shown in FIG. 3, the cylindrical lens 10 is disposed by being rotated by an angle α with respect to the X axis (in the illustrated case, it is disposed so as to rise to the right). α = tan ⁻¹ [(tm × φ / Pm × v) / L] (°) (Formula 4)
The cylindrical lens 10 is held as follows. That is, the cylindrical lens 10 is supported by the lens holder 35. A circular portion 38 provided in the lens holder 35 is fitted to the outer periphery of the boss 34a so that the axis of the cylindrical lens 10 passes through the center of the boss 34a. The boss 34a has a hollow horseshoe shape in which a part of the annular shape is cut out, and its center is positioned at the drawing start point. The reason why the boss 34 a is notched is to prevent the boss 34 a from interfering with the laser beam 4 transmitted through the cylindrical lens 10. The inner radius of the boss 34a is a radius (15 mm here) that does not prevent the laser beam 4 from entering the start sensor 15. Accordingly, the lens holder 35, that is, the cylindrical lens 10, can be positioned around the drawing start point around the center of the boss 34a, that is, the drawing start point. A hole 37 is formed in the lens holder 35 so that the laser beam 4 that has passed through the cylindrical lens 10 within a range in which the lens holder 35 can rotate does not interfere with the lens holder 35. The scanning angle α can be set with high accuracy by a driving device (not shown).

  FIG. 4 is a sectional view in a direction perpendicular to the axis of the cylindrical lens, and P is the axis of the cylindrical lens 10.

  As described above, the components in the width direction of the cylindrical lens 10 of the laser beam 4 incident on the cylindrical lens 10 are parallel. Therefore, when the central axis of the laser beam 4 perpendicularly incident on the cylindrical lens 10 passes through the central axis P, the laser beam 4 is collected at a position away from the central axis P by F (where F is the focal length of the cylindrical lens 10). Shine. Here, when the cylindrical lens 10 is moved to the right of the figure by δ as shown by a two-dot chain line in the figure with the central axis of the laser beam 4 fixed (that is, the central axis P is moved to the right of the figure only). ), The laser beam 4 is focused on F0 to the right by δ from F in the figure. That is, even if the central axis of the laser beam 4 is the same, if the position of the central axis P of the cylindrical lens 10 is shifted, the condensing position is moved by the shifted distance. That is, when the central axis P of the cylindrical lens 10 is inclined by an angle α with respect to the traveling direction of the workpiece, the condensing position of the laser beam 4 is also inclined by the angle α with respect to the traveling direction. Therefore, the scanning angle α can be set as an angle of the central axis P of the cylindrical lens 10 with respect to the scanning direction.

  In the case of this embodiment, since the laser beams are arranged in 8 columns × 16 rows, not only single exposure but also double, quadruple, 8-fold, and 16-fold exposures are possible. That is, the amount of light per unit area can be 10 mJ, 20 mJ, 40 mJ, 80 mJ, and 160 mJ.

  In general, when the number p of optical axes of laser beams in the row direction is set to a product of positive integers q and s (p = qs), the object to be drawn is irradiated with the laser beam while shifting by s rows in the row direction. This makes it possible to perform q-fold exposure in which the amount of light at the drawing portion is q times that in the case of single exposure.

  FIGS. 5A and 5B are explanatory diagrams of the exposure start position, where FIG. 5A shows the case of single exposure, FIG. 5B shows the case of double exposure, and FIG. 5C shows the case of quadruple exposure. FIG. 6 is an explanatory diagram of the exposure start position.

  As shown in FIG. 5, at the time of double exposure, a single point is generated in the row direction, and at the time of quadruple exposure, a single point and a double point are generated. In the 16-fold exposure when the LDs 1 are arranged in 8 columns × 16 rows, the distance in the Y direction until the 16-fold exposure is 95.25 μm (6.35 μm × 15), which is a length that cannot be ignored. Further, the start position at which the desired number of overlaps differs depending on the number of exposures N in the sub-scanning direction. Therefore, as shown in FIG. 6, when the exposure start position in the sub-scanning direction is determined and scanning is performed, the LD above the position is turned off, and the LD below the position is used as the drawing data. If the drawing data is assigned so as to be turned ON / OFF correspondingly, the exposure result in the sub-scanning direction can be made uniform even in the overlap exposure.

  Therefore, for example, a setting unit for setting the number N of overlaps, a cylindrical lens rotation positioning unit for changing the tilt angle α of the cylindrical lens, and a calculation unit for rewriting drawing data for N double exposure are provided. When the number N is set, the table feed speed v, the tilt angle α of the cylindrical lens is set based on the set overlap number N, and the writing data is rewritten for N double exposure automatically. If the control device is configured, the exposure operation is further facilitated.

  FIG. 7 is a diagram showing an example of the wavelength-absorbance characteristics of DFR. For example, when a mercury lamp is used as a light source, the light from the mercury lamp includes i-line having a wavelength of 365 nm and h-line having a wavelength of 405 nm, and both i-line and h-line are involved in exposure. To do. Therefore, when the required light quantity is shown based on a mercury lamp, it cannot be said that exposure with the numerical value of irradiation energy is possible with an LD of one wavelength. However, even if the efficiency is 60%, a DFR of 96 mJ≈100 mJ can be exposed if 16-fold exposure is performed.

It is a figure explaining the summary of this invention. It is explanatory drawing of the scanning angle (alpha). It is a top view of a cylindrical lens. It is sectional drawing of the direction orthogonal to the axis line of a cylindrical lens. It is explanatory drawing of an exposure start position. It is explanatory drawing of an exposure start position. It is a figure which shows the example of the wavelength-absorbance characteristic of DFR. It is a perspective view of the conventional laser direct drawing apparatus. It is the schematic of a light source system. It is a figure which shows arrangement | positioning of a start sensor. It is a figure explaining the drawing procedure at the time of arrange | positioning at 4 rows x 4 columns. It is explanatory drawing of a drawing result.

Explanation of symbols

4 Laser beam 10 Cylindrical lens 11 Drawing object

Claims (3)

While the optical axis of the laser beam aligned in the column direction and the row direction is deflected in the main scanning direction that is the same as the column direction, the drawing object is moved in the sub-scanning direction that is the same as the row direction, and the raster data is converted. In a laser direct drawing method for drawing a desired pattern on the drawing object by irradiating the drawing object with the laser beam modulated based on the method,
The number of optical axes of the laser beam in the row direction is set as p which is a product of q and s which are positive integers, and the object to be drawn is irradiated with the laser beam while shifting by s rows in the row direction, A laser direct drawing method characterized by performing q-fold exposure, which is q times the amount of light at a drawing position in the case of single exposure. While the optical axis of the laser beam aligned in the column direction and the row direction is deflected in the main scanning direction that is the same as the column direction, the drawing object is moved in the sub-scanning direction that is the same as the row direction, and the raster data is converted. In a laser direct drawing apparatus for drawing a desired pattern on the drawing object by condensing the laser beam modulated based on the cylindrical lens in the sub-scanning direction and irradiating the drawing object,
The number of optical axes of the laser beam in the row direction is set to p, which is a product of q and s, which are positive integers, and setting means for setting the amount of light at the drawing position and the cylindrical lens are rotated in the matrix plane A rotating means for causing
The position of the cylindrical lens is determined according to the amount of light set in the setting means,
A laser direct drawing apparatus characterized in that the laser beam is irradiated to the object to be drawn while being shifted by s rows in a row direction, thereby performing q-fold exposure in which the amount of light at the drawing position is q times that of single exposure. .   3. The laser direct drawing apparatus according to claim 2, wherein a center of rotation of the cylindrical lens is made coincident with a scanning start point of the laser beam.

Images