JP2007152420A - Method of removing film on substrate - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method of removing selectively only a film on a substrate without damaging the substrate. <P>SOLUTION: Including a condensing means 3 for converging a ultrashort light pulse laser L, the method of removing a film on a substrate includes: a condensing point setting step in which the condensing point p of the condensing means 3 is set at a position away from the boundary 2b between the substrate 1 and the film 2 thereon to the film 2 side by a first prescribed distance D; and a film removing step in which the film 2 is removed through laser ablation by convergently irradiating the film 2 with the ultrashort light pulse laser L. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a method for removing a film formed on a substrate by laser ablation.

  Up to now, a laser-induced breakdown (LIB: Laser Induced Breakdown) has been performed on a material having a two-layer structure made of one or more metals, a dielectric material, and a semiconductor material. There has been known a method that does not affect (see, for example, Patent Document 1).

  When this method is applied to the removal of a film formed on a substrate, the focal point is trepanned to increase the width of removal (in this specification, moving the focal point to a circle is “trepanned”). ")" Is desirable.

As a method of trepanning the condensing point, for example, a method as shown in FIG. 10 is known (for example, see Patent Document 2). A laser beam 71 is incident on an inclined incident surface of a dove prism (Dove-Prism) 74 that is inclined with respect to the main axis by two parallel blocks 72 having two parallel surfaces and two wedge plates 73 and rotated about the main axis. To do. The tilted laser beam 71 emitted from the dove prism 74 is incident on the condensing lens 75 and is condensed on the edge of the circle 76 centering on the main axis 100. This condensing point draws a circle 76 (repanning) by the rotation of the Dove prism 74, and the circle 76 is removed.
JP 2002-205179 A JP 2001-516648 A

  The above-described conventional LIB affects approximately one layer and does not affect other layers, but does not disclose how it can affect approximately one layer and not affect other layers. Therefore, for example, when the copper of 4.5 μm thickness on the polyimide substrate is focused and irradiated with an ultrashort optical pulse laser to remove the copper by ablation, the surface of the polyimide substrate is also removed. That is, conventionally, the film on the substrate could not be removed without damaging the substrate.

  On the other hand, in the conventional trepanning method described above, a bulk glass optical component having a long length of a medium through which a laser beam passes, such as a parallel block, two wedge plates, and a dove prism, is frequently used. There is a problem that the time width (pulse width) is widened or the diameter of the focused spot is increased due to chromatic aberration.

  The present invention has been made in view of the above problems of the conventional ablation processing method, and an object thereof is to provide a method for selectively removing only a film on a substrate without damaging the substrate.

  In addition, the present invention has been made in view of the above-mentioned problem of trepanning, and the width to be removed is reduced by trepanning using a method that does not cause a pulse width broadening due to wavelength dispersion or a condensing spot diameter due to chromatic aberration. Another challenge is to provide a removal method that can be expanded.

  The invention according to claim 1, which has been made to solve the problem, has condensing means for condensing an ultrashort optical pulse laser, and the condensing point of the condensing means is set to the substrate of the film on the substrate and the substrate. A condensing point setting step for setting a position separated from the interface of the film by a first predetermined distance toward the film side, and removal for removing the film by laser ablation by condensing and irradiating the ultrashort light pulse laser on the film And a step of removing the film on the substrate.

  As shown in FIG. 1A, when the condensing point (beam waist) of the condensing lens 3 which is a condensing means is located on the surface 2a of the film 2, the increase in the beam diameter before and after the condensing point is gentle. Ablation processing proceeds in the thickness direction 2 (Z direction), and the substrate 1 is also processed. In the first aspect of the invention, the focal point is a position defocused to the film 2 side by a first predetermined distance (D) from the interface 2b between the substrate 1 and the film 2 (see FIG. 1B), and therefore the surface 2a of the film 2 The beam diameter increases rapidly as the condensing diameter increases at a distance from the surface 2a in the thickness direction (z direction). As a result, even if ablation occurs on the surface 2a, the fluence decreases abruptly only by being slightly separated from the surface 2a in the thickness direction, and ablation does not easily occur, and the substrate 1 is not processed. Moreover, since the condensed diameter on the surface 2a of the film 2 is large, the film can be widely removed without trepanning. Here, the first predetermined distance is a distance at which the fluence at the interface is equal to or less than the ablation threshold value of the substrate, in other words, a distance larger than the Rayleigh range (see equations (7) and (7) ′ described later). . The optimum value for the first predetermined distance varies depending on the material and film thickness of the substrate and the film, so it must be determined experimentally (see Example 1 described later).

  The invention according to claim 2 is the method for removing a film on the substrate according to claim 1, wherein the condensing point setting step first sets the condensing point to a second predetermined distance (D ′) from the interface. In this case, the focusing point is set to a position closer to the film than the second predetermined distance, and the removing step is repeated until the substrate appears. .

  Since the condensing point setting step first sets the condensing point at a position separated from the interface by a predetermined distance toward the film side, and then sequentially repeats the removal step by setting the condensing point at a position closer to the interface than the predetermined distance. Even if the film thickness is large, only the film can be reliably removed without damaging the substrate. Here, the second predetermined distance is a distance at which the fluence at the interface is not more than the ablation threshold of the substrate, and is a distance larger than the Rayleigh range. The second predetermined distance is greater than the first predetermined distance.

  The invention according to claim 3 for solving the problem has a condensing means for condensing the ultrashort optical pulse laser, and the condensing point of the condensing means is changed from the surface of the film on the substrate to the Rayleigh range. A condensing point setting step for setting within the range of the position where the Rayleigh range has entered the film from a position that is far away, and a fluence control step for making the fluence of the condensing point approximately equal to the ablation threshold of the film, A method for removing a film on a substrate, comprising:

As shown in FIG. 1C, if the Rayleigh range is z _r and the focal length of the condensing lens 3 as the condensing means is f, in the invention of claim 3, the condensing point is z upward from the surface 2 a of the film 2. _r Set within the range from the up position to the down z _r position. Further, the fluence of the condensing point is made substantially equal to the ablation threshold value of the film 2. For example, when the thickness of the film 2 is small, the condensing point is set so that the surface 2a has a dotted line in FIG. 1C, that is, the distance from the condensing lens 3 to the surface 2a of the film 2 is (f + z _r ). As a result, the film is ablated by the first laser pulse and the film is removed. However, even if the next laser pulse attacks the substrate 3, the beam diameter increases, the fluence decreases, and ablation cannot occur. . Here, the fluence of the condensing point that is substantially equal to the ablation threshold value of the film changes depending on the film thickness, the setting position of the condensing point, etc., but is not uniquely determined, but 10% or more of the ablation threshold value of the film, A film ablation threshold value or less is preferred. This is because the fluence is an average value of the beam condensing region, and if this is larger than the ablation threshold of the film, the fluence near the center becomes further larger and it becomes difficult to prevent damage to the substrate.

  The invention according to claim 4 is the method for removing a film on a substrate according to any one of claims 1 to 3, further comprising: the ultrashort optical pulse laser incident on the condensing means. It has a deflection rotation step for deflecting and rotating the differential motion beam.

  By deflecting and rotating an ultrashort pulse laser with a rotating wedge plate, a precessing beam can be formed, so that it is possible to prevent widening of the pulse width due to wavelength dispersion and increase of the focused spot diameter due to chromatic aberration, so that trepanning processing can be performed. can do.

  The invention according to claim 5 is the method for removing a film on the substrate according to claim 4, wherein the condensing means is a condensing lens, and the nodes of the precessing beam are collected in the concentrating lens. It has a transmission step for transmitting to the entrance pupil of the optical lens.

  A telecentric lens (a lens in which the entrance pupil of the lens is in a telecentric position) can be used as the condensing lens, and the precessing beam can be condensed to the diffraction limit of the lens.

  The invention according to claim 6 is the method for removing a film on the substrate according to any one of claims 1 to 5, wherein the NA of the condenser lens is 0.4 or more. Yes.

  As shown in FIG. 1A, NA (numerical aperture) is the optical axis (z) of the light beam having the maximum opening angle among the light beams going to the condensing point p (point away from the lens 3 by the focal length f) in geometric optics. Assuming that the angle formed by (axis) is (γ / 2), NA = sin (γ / 2). Therefore, the larger the NA, the more rapidly the beam diameter changes with the z-axis direction distance. Therefore, when NA becomes 0.4 or more, the range in the z-axis direction of the beam diameter having a fluence equal to or greater than the ablation threshold becomes narrow, and only the film can be selectively removed further.

  The invention according to claim 7 is the method for removing a film on the substrate according to any one of claims 1 to 6, wherein the film is made of a material having a threshold value smaller than an ablation threshold value of the substrate. It is a feature.

  By making the fluence of the condensing point substantially equal to the ablation threshold value of the film, damage to the substrate can be more effectively prevented.

  Since the condensing point is a position defocused from the interface between the substrate and the film to the film side by a first predetermined distance, the condensing diameter on the surface of the film is large, and the beam diameter suddenly increases as the distance from the surface in the thickness direction increases. To increase. As a result, even if ablation occurs on the surface, the fluence decreases abruptly only by being slightly away from the surface in the thickness direction, making it difficult for ablation to occur, and the substrate is not processed.

  The best mode for carrying out the present invention will be described below with reference to the drawings.

(Embodiment 1)
FIG. 2 is a schematic configuration diagram of the substrate film removal apparatus according to the first embodiment using an ultrashort optical pulse laser. Reference numeral 10 denotes a laser generator that generates an ultrashort optical pulse laser, 3 denotes a condensing lens that is a condensing means, 2 denotes a film, 1 denotes a substrate, and 20 denotes a triaxial (xyz) moving stage.

For example, a mode-locked Ti: sapphire laser can be used as the laser generator 10 that generates an ultrashort optical pulse laser, but a mode-locked fiber laser-based FCPA (Fiber Charpe Pulse Amplification) laser is preferably used. This is because when the film material is a metal material, the ablation threshold of the metal material is 0.5 J / cm ² or less, and therefore, in a mode-locked Ti: sapphire laser with 1 pulse energy of 1 mJ, it is necessary to significantly attenuate with an attenuator However, one pulse energy of the FCPA laser is about 10 μJ, so it is not necessary to attenuate it. Here, the pulse width of the ultrashort optical pulse laser is 10 ps or less in which laser-induced breakdown occurs. The lower limit is not particularly limited, but is about 10 fs. The pulse width is preferably smaller than 3 ps, more preferably 1 ps or less. This is because when the substrate is a polymer material and the film is a metal material, if the pulse width is large, the ablation threshold of the film becomes higher than that of the substrate, and it becomes difficult to suppress damage to the substrate.

  As the condenser lens 3, a microscope objective lens whose aberration is corrected may be used. A larger NA is desirable for the reasons described above.

The substrate 1 is not particularly limited, such as a polymer material such as polyimide, polyethylene terephthalate, polycarbonate, polytetrafluoroethylene, and polymethyl methacrylate. When the substrate is made of a polymer material, the ablation threshold is 0.5 to 1 J / cm ² .

The film 2 is not particularly limited, such as a metal material such as copper, aluminum, titanium, tin, and stainless steel, but aluminum is desirable when the substrate 1 is made of a polymer material. Since the ablation threshold value of aluminum is 0.07 to 0.09 J / cm ^2, which is smaller than the ablation threshold value of the substrate made of a polymer material, it becomes easy to selectively remove only the film without damaging the substrate. .

  Hereinafter, each step will be described.

<Condensing point setting step>
First, the condensing point p of the condensing lens 3 is adjusted to a position Dt above the surface 2a of the film 2 (D: first predetermined distance). This adjustment can be performed by lowering the z-axis stage 21 of the moving stage 20 by Dt after aligning the focal point p with the surface 2a of the film 2. Here, t is the thickness of the film 2. In addition, the method of making the condensing point p correspond with the surface 2a of the film | membrane 2 inserts a mirror between the laser generator 10 and the condensing lens 3, for example, and makes the condensing lens 3 an objective lens through the mirror. The microscope is configured to move the z-axis stage up and down so that the surface 2a of the film 2 can be clearly seen.

  When the condensing point p is set at a position Dt above the surface 2a of the film 2, D may be determined experimentally. However, if it is as follows, the amount of experiment for determining the condition can be reduced.

The laser beam can be approximated with a normal Gaussian beam. When the Gaussian beam is collected by a condenser lens having a focal length f, the beam diameter of the beam waist is 2w ₀ , the diameter of the laser beam incident on the condenser lens is d, and the diameter is 1 / e ^2. If the full spread angle is γ,
2w ₀ = (4λ / π) (f / d) = (4λ / πγ) (1)
It is expressed. Here, λ is a wavelength. In addition, the change in the beam diameter in the vicinity of the beam waist as shown in FIG.
(2w) ² = (2w ₀ ) ² + γ ² z ² (2)
It becomes. Here, 2w is a diameter at a distance ± z along the beam axis (z axis) from the beam waist.

  When the condensing point p is set at the position Dt above the surface 2a of the film 2, the surface when the condensing point p is at the Dt position using the above equations (1) and (2). The amount of the experiment can be reduced by making the fluence of the above the ablation threshold value of the film 2 or more.

<Removal step>
Next, by generating an ultrashort optical pulse laser having a predetermined one-pulse energy from the laser generator 10, the region irradiated with the laser is ablated, and has a diameter 2a and a depth δ as shown in FIG. A hole is dug. In this state, for example, by moving the x-axis stage in the plus direction, a groove having a width 2a and a depth δ parallel to the x-axis can be dug. Further, in this state, the reciprocating process is performed by moving the x-axis stage in the minus direction, and a groove deeper than δ can be dug. If the thickness t of the film 2 is δ or less, a hole reaching the substrate 1 is formed. If ask the x-axis stage in this state is moved by x _0, film 2 width 2a, over the length x _0, is removed.

When the predetermined one pulse energy is E and the ablation threshold of the film 2 is Ath, when there is no loss,
E / (πa ² ) ≧ Ath (3)
There is a relationship. If the loss in the optical element between the laser generator 10 and the film 2 is ρ,
E (1-ρ) / (πa ² ) ≧ Ath (3) ′
There is a relationship.

  If the thickness t of the film 2 is large, the above steps are repeated. That is, in the condensing point setting step, the distances between the condensing point p and the surface 2a of the film 2 are sequentially set as D′−t−δ, D′−t−2δ,... D′−t−nδ. The removal step is repeated each time (D ′: second predetermined distance). The number of repetitions depends on the thickness t of the film 2. Repeat at least until the substrate 1 appears. Since δ varies depending on the number of reciprocations, instead of decreasing the distance between the condensing point p and the surface 2a in a certain step as described above, the value of δ to be decreased may be changed.

  Normally, the intensity distribution of a laser beam has a Gaussian distribution. When such a Gaussian beam is focused by, for example, a condensing lens having an NA of 0.8, the cross-sectional shape of the ablated portion is as shown in FIG. V-shaped or U-shaped following the lens NA. In the case where the substrate 1 is an insulating material and the film 2 is a conductive material, with this shape, the removal width of the film near the substrate is narrowed, and the insulation of the film across the removal region may be uncertain. There is. By inserting a flat top beam shaper between the laser generator 10 and the condensing lens 3, the Gaussian beam can be corrected to a flat top type, and the cross-sectional shape can be made a shape close to a rectangle. As a result, more reliable insulation is possible. The flat top beam shaper can be a Newport beam shaper or an aspheric lens designed to correct a Gaussian beam to a flat top shape.

(Embodiment 2)
FIG. 4 is a schematic configuration diagram of the substrate film removal apparatus according to the second embodiment. The removal apparatus of this embodiment is provided with a deflection rotation means 4 for deflecting and rotating an ultrashort optical pulse laser incident on the condenser lens 3 into the precession beam in the removal apparatus of Embodiment 1 of FIG. 2 is basically the same as FIG. 2 except that the moving stage 20 is omitted. The same components as those in the first embodiment are denoted by the same reference numerals and description thereof is omitted.

  The deflection rotation means 4 has an optical axis O.D. A rotating body 42 that rotates around A and a wedge plate 41 that is attached to the rotating body 42 and rotates together with the rotating body 42.

Next, a mechanism for trepanning by the deflection rotating means 4 will be described. An optical axis O.D. is formed on the wedge plate 41 indicated by a solid line whose angle between the incident surface 411 and the output surface 412 is α. When the pulse laser L is incident from the A direction, the optical axis O.D. The laser beam is deflected in the direction of θ with A and emits a laser La indicated by a solid line. There is a relationship of the following equation between θ and α.
θ = sin ⁻¹ [nsin α] −α (4)
Here, n is the refractive index of the wedge plate 41.

Optical axis O.D. When a laser La that forms A and θ enters the condenser lens 3, the optical axis O.D. on the surface of the film 2 located at a focal distance f away from the lens 3 as indicated by a solid line. The light is condensed at a position from A to r and becomes a condensing point p. r and θ have the following relationship.
r = ftanθ (5)
When the laser L enters the wedge plate 41 rotated by 180 ° indicated by a dotted line, the optical axis O.D. The laser beam Lb is deflected in a direction that forms A with -θ and emits a laser Lb indicated by a dotted line.

  Optical axis O.D. When the laser Lb that forms A with -θ is incident on the condenser lens 3, the optical axis O.D. on the surface of the film 2 is again shown by the dotted line. The light is condensed at a position from A to r and becomes a condensing point p.

  4B is a view of the film 2 of FIG. 4A viewed from the z-axis (optical axis OA) direction. When the wedge plate 41 rotates in the arrow 11 direction, the laser beam L incident on the wedge plate 41 is: It deflects and rotates to become precession beams La and Lb, and the condensing point p also moves sequentially in the direction of the arrow 12 to draw a trepanning circle 8 of radius r.

  It can be seen from the equations (4) and (5) that the size (radius r) of the trepanning circle 8 can be controlled by the focal length f of the condenser lens 3 and the wedge angle α of the wedge plate 41.

  Hereinafter, each step will be described.

<Condensing point setting step>
The condensing point p of the condensing lens 3 is adjusted near the surface of the film 2. Here, the vicinity of the surface means that when the Rayleigh range is z _r , the condensing point p in geometric optics (the beam waist position z = 0 in wave optics) goes down from the position where it rises z _r above the surface of the film 2. This is the range up to the position where _{zr is} lowered. That is, as shown in FIG. 1C, the surface 2a of the film 2 is in the range of ± z _r (the distance from the condenser lens 3 to the surface 2a of the film 2 is in the range of f ± z _r ). The Rayleigh range z _r is a distance in which the beam diameter at the beam waist z = 0 is within √2 times, and outside this, the beam with the distance is shown as an asymptotic line (straight line indicated by a dotted line) of the laser beam spread angle γ. The diameter increases.

When the film 2 is thin, it is preferable to set the condensing point so that the surface 2a has a dotted line in FIG. 1C, that is, the distance from the condensing lens 3 to the surface 2a of the film 2 is (f + z _r ). Although the film is ablated by the first laser pulse and the film is removed, even if the next laser pulse attacks the substrate 3, the beam diameter increases, the fluence decreases, and ablation cannot occur.

When the thickness of the film 2 is large and the thickness is around 2z _r , the surface 2a is the solid line in FIG. 1C, that is, the distance from the condenser lens 3 to the surface of the film 2 is (f−z _r ). In addition, a condensing point may be set. Even down from the surface 2z _r does not increase much the beam diameter, it is possible to sequentially remove the film 2. If the beam diameter further falls below 2z _r , the beam diameter increases as an asymptotic line with the distance, so that the fluence decreases and ablation cannot be caused. As a result, the substrate 1 is not processed.

If the beam diameter of the beam waist when the Gaussian beam is collected by the condenser lens is 2w ₀ ,
z _r = πw ₀ ² / λ (6)
It is expressed. From the expressions (1) and (6), z _r = (4λ / π) (f / d) ² = (4λ / π) (1 / γ ² ) (6) ′
It is. For example, when a 50 × microscope objective lens is used as the condenser lens 3, f = 3 mm. Therefore, when λ = 1.5 μm and d = 2 mm, z _r = 4.3 μm.

<Fluence control step>
The one-pulse energy E generated from the laser generator 10 is adjusted so that the fluence of the condensing point p is substantially equal to the ablation threshold value of the film 2. That is, if the loss in the optical element between the laser generator 10 and the film 2 is ρ,
0.1 × Ath ≦ E (1-ρ) / (πw ₀ ² ) ≦ Ath (7)
To satisfy.

<Deflection rotation step>
The laser pulse L controlled so as to satisfy the expression (7) in the fluence control step is incident on the rotating wedge plate 41, and the precession beams La and Lb are emitted. Then, the condensing beam cross-sectional area on the surface 2a of the film 2 by the condensing lens 3 is ablated, and a donut-shaped groove having a radius r is formed.

In this state, a three-axis moving stage (not shown) for example, when to x ₀ moves in the x-axis direction, it is possible to process the groove width 2r length x _0.

  When so-called trepanning processing is performed as in the present embodiment, the processing can be performed with a wide (2r) cross section close to a rectangle, so that it is not necessary to use a flat top beam generator or the like as in the first embodiment.

Example 1
The copper film 2 having a thickness t = 4.5 μm formed on the polyimide substrate 1 having a thickness of 25 μm was removed by the substrate film removing apparatus of the first embodiment shown in FIG.

As the laser generator 1, an FCPA laser generator having the following characteristics was used.
Center wavelength: 1.56 μm
Pulse energy: 1μJ / pulse Pulse width: 0.9ps
Repetition frequency: 194 kHz
Beam intensity distribution: Gaussian Beam diameter: 2 mm

  When the metal material film on the polyimide substrate, which is a polymer material, is removed as in this embodiment, the center wavelength of the laser is preferably 500 to 2000 nm. When shorter than 500 nm, the photon energy increases and the ablation threshold of the polymer material decreases. As a result, it becomes difficult to eliminate damage to the substrate.

The condenser lens 3 was an infrared microscope objective lens having a magnification of 100 times. The condensing diameter at the condensing point is about 3.6 μm (calculated value). Therefore, the fluence at the condensing point is estimated to be 4.9 J / cm ² when the transmittance of the condensing lens 3 is 50%. Since the copper ablation threshold is 0.47 J / cm ² , this fluence is about an order of magnitude larger than that and naturally satisfies the equation (3) ′.

<Condensing point setting step>
First, the condensing point p of the condensing lens 3 is adjusted to a position Dt above the surface 2a of the copper film 2. This adjustment can be performed by lowering the z-axis stage 21 of the moving stage 20 by Dt after aligning the focal point p with the surface 2a of the copper film 2. The value of D−t was determined as follows.

From equation (1), γ = 0.55 rad. Since it is necessary to make the fluence at the position of Dt approximately equal to 1/10 of the fluence at the condensing point, that is, approximately equal to the ablation threshold value of copper, using equation (3), a = 5.6 μm Estimated. Therefore, by setting 2w = 11.2 μm, 2w ₀ = 3.6 μm, and γ = 0.55 rad in equation (2), z = D−t = 20 μm.

  Since the Rayleigh range zr of this embodiment is zr = 1.6 μm from the equation (6) ′, the predetermined distance D (= 20 + 4.5 = 24.5 μm) from the interface between the substrate and the film to the condensing point p is , About 15 times larger than the Rayleigh range zr.

<Removal step>
D = 24.5 μm, and irradiating the copper film 2 with the ultrashort light pulse from the laser generator 10, the x-axis moving stage 23 is set to 30 mm / over the entire width of the copper film 2 in the x-axis direction of 370 μm. One reciprocating scan was performed at a speed of sec (L1 in FIG. 5). In order to monitor the degree to which the copper film 2 is removed from the resin substrate 1, processing was performed while measuring the resistance value with the removal groove interposed therebetween.

  Next, returning to <Condensing Point Setting Step>, the y-axis stage is moved by 50 μm in the y-axis direction, and the z-axis stage 21 is used to set the distance between the condensing point p and the surface 2a of the copper film 2 to D−t−δ. = (20-1) = 19 μm <Removal step> was performed (L2 in FIG. 5).

  Next, returning to <Condensing Point Setting Step>, the y-axis stage is moved by 50 μm in the y-axis direction, and the distance between the condensing point p and the surface 2a of the copper film 2 is set to D−t−2δ = (20−2). ) = 18 μm and the <removal step> was performed (L3 in FIG. 5).

  Thereafter, the same operation was sequentially repeated until the resistance value increased rapidly.

  FIG. 5 is a surface photograph of the copper film 2, and black lines L 1, L 2,... L 15 parallel to the x-axis are processing marks of the copper film 2. FIG. 6 shows the result of monitoring the change in resistance value, and the horizontal axis is Dt-nδ. Although the resistance is low until L14, the resistance value rapidly increases at L15, and it can be seen that the copper film 2 on the polyimide substrate 1 is completely removed.

  FIG. 7 is a drawing of a cross-sectional observation result of line AA of L15 in FIG. 5, which is a relatively wide tapered groove having a width of 16.9 μm on the surface of the copper film 2 and a width of 8 μm at the bottom. I understand that. Moreover, the damage of the surface (interface with the copper film 2) of the polyimide substrate 1 was hardly seen.

  Dt = 6 μm corresponding to L15, that is, D = 10.5 μm (= 6 + 4.5) is the first predetermined distance in claim 1. By setting the focal point p to the 10.5 μm film side from the interface between the substrate and the film, the film could be selectively removed without damaging the substrate.

(Example 2)
In this example, the copper film 2 of Example 1 is made of an aluminum film, and the position of Dt, that is, the fluence at the film surface 2a is 0.08 J / s slightly larger than the aluminum ablation threshold (0.07 J / cm ² ). Except for adjusting to cm ² , the same as Example 1.

  As a result of selective removal of the 4.5 μm thick aluminum film on the 25 μm thick polyimide substrate, the aluminum film could be selectively removed without damaging the polyimide substrate at all.

This aluminum film is removed, the laser at fluence of 0.08 J / cm ² is also irradiated to the polyimide substrate, due to the ablation threshold of the polyimide 0.5 J / cm ² or more is considered to not be damaged.

(Example 3)
This example was implemented using the substrate film removal apparatus shown in FIG. The apparatus for removing a film on a substrate used in the present embodiment includes a laser generator 10 that generates a laser beam L, an attenuator 45 that adjusts the intensity of the laser beam L, a shutter 46 that controls ON / OFF of the laser beam L, and A bending mirror 15, a wedge plate 41 attached to a motor 42 for converting the laser beam L to precession beams La and Lb, and a first relay lens 7a for transmitting a node N of the precession beams La and Lb. A second relay lens 7b; a dichroic mirror 21 that reflects the precessing laser beams La and Lb; a condensing lens 3 that condenses the precessing laser beams La and Lb that reflect the dichroic mirror 21; The film 2 formed on the substrate 1 on which the precessing laser beams La and Lb collected by the optical lens 3 are incident from the z-axis direction is moved in the x, y, and z-axis directions. It comprises a movable stage 20 for, and the control computer 40.

  The on-substrate film removing apparatus further includes an observation light source 18 that generates visible light for illuminating and observing the film 2 with visible light, and the visible light from the observation light source 18 is bent by 90 ° and incident on the dichroic mirror 21. A coupler 19 including a half mirror, a condenser lens 3, a dichroic mirror 21, and a CCD camera 22 that images the film 2 through the coupler 19 are provided.

  The substrate film removal apparatus further includes an optical bench 30 on which the bending mirror 15, the rotating wedge plate 41, the first relay lens 7 a, the second relay lens 7 b, the dichroic mirror 21, and the condenser lens 3 are disposed, and the optical bench 30. and a drive unit (not shown) for driving in the z-axis direction.

  The controller 20 that controls the laser generator 10, the attenuator 45, the shutter 46, the triaxial moving stage 20, the CCD camera 22, and the optical bench 30 are connected to the control personal computer 40. Control of the generated laser beam L and its intensity, ON / OFF control of the shutter 46, imaging data processing of the CCD camera 22, drive control of the drive unit and the triaxial moving stage 20 are performed.

  The laser generator 10 is an FCPA laser generator, which has a center wavelength of 1.045 μm, a maximum average output of 400 mW, a pulse width of 0.45 ps, a repetition frequency f of 162.3 kHz, and a beam diameter of 2 mm. A beam L is emitted.

  The motor 42 is a hollow shaft air turbine and can rotate up to 200,000 rpm.

  The wedge plate 41 is made of BK7 and has a wedge angle of 0.5 °.

  Each of the first relay lens 7a and the second relay lens 7b is a combination of infrared achromat lenses having a focal length of 100 mm, and can transmit a 1: 1 real image. That is, the node N on the wedge plate 41 by the first relay lens 7a and the second relay lens 7b is transmitted to the entrance pupil 31 of the condenser lens 3 and becomes N '.

  The condensing lens 3 is a microscope infrared objective lens with a magnification of 50 times (f = 3.6 mm), the entrance pupil 31 is located at a telecentric position in the lens barrel, and the NA (numerical aperture) is 0.55. .

  The substrate 1 is a polyimide having a thickness of 25 μm, and the film 2 is a copper film having a thickness of 4.5 μm.

The spot diameter at the condensing point p is 2w ₀ = 2.4 μm from equation (1), the Rayleigh range is z _r = 4.3 μm from equation (6) ′, and the radius of the trepanning circle is (4), ( 5) From the formula, r = 15.7 μm.

  Hereinafter, each step will be described.

<Condensing point setting step>
First, the attenuator 45 is controlled to set the average power of the laser beam L to 0.5 mW, and the shutter 46 is turned on. Next, while observing the image picked up by the CCD camera 22 on the monitor of the personal computer 40, the optical bench 30 is moved in the z-axis direction by the drive unit of the optical bench 30 so that the focused spot p is located on the surface of the copper film 2 ( Finely move in the direction of arrow 32).

<Fluence control step>
Next, the shutter 46 is turned off and the attenuator 45 is controlled so that the fluence of the condensing point p becomes substantially equal to the ablation threshold of the copper film 2.

In this embodiment, the average power of the laser beam L is set to 2.5 mW, that is, the pulse energy is set to 0.0156 μJ / pulse (= 2.5 mW / 162 kHz) from the result of the preliminary experiment. The fluence at the condensing point p at this time is 0.2 J / cm ² because the transmittance of the condensing lens 3 is 58% and the condensing diameter is about 2.4 μm. This value is about half of the copper ablation threshold (about 50%) and satisfies the equation (7). The reason for this will be described later.

<Deflection rotation step>
Next, the shutter 46 is turned on while the hollow shaft air turbine 6 is rotated at 153,500 rpm, and the precession beams La and Lb are emitted from the rotating wedge plate 42.

<Transmission step>
The nodes N of the precession beam La and Lb are transmitted to the entrance pupil of the condenser lens 3 by the relay lenses 7a and 7b.

<Removal step>
The copper film 2 is reciprocated by 165 μm 1 in the x-axis direction at a speed of 30 mm / sec.

  Next, the polyimide substrate 1 / copper film 2 is brought closer to the condenser lens 3 by 2 μm on the z-axis stage 21 and the <removal step> is repeated. Furthermore, the polyimide substrate 1 / copper film 2 is brought closer to the condenser lens 3 by 2 μm on the z-axis stage 21 and the <removal step> is repeated.

  The result of selectively removing the copper film on the polyimide substrate by the above method is shown in FIG. The copper film 2 could be completely removed over a width of about 30 μm and a length of 165 μm. When the removal region was observed with a laser microscope, the surface in contact with the copper film 2 of the polyimide substrate 1 was not processed at all.

Next, the reason why the above good results can be obtained by setting the fluence at the condensing point p to 0.2 J / cm ² and about ½ of the copper ablation threshold will be discussed.

First, the reason why ablation processing can be performed even when the fluence is about 1/2 of the ablation threshold value of copper is that the beam intensity is a Gaussian distribution and the beam diameter at which the intensity becomes 1 / e ² is 2w ₀ . The fluence 0.2 J / cm ² at the condensing point p is an average fluence within 2w ₀ = 2.4 μm, and naturally, the center of the condensing point p has the highest fluence, and the fluence becomes lower toward the periphery. As a result, it is considered that the fluence near the center exceeds the ablation threshold value of 0.47 J / cm ² for copper and is ablated.

Further, since z _r = 4.3 μm, when the fluence at the condensing point p (the surface of the copper film 2) was set to a copper ablation threshold of 0.47 J / cm ² , it decreased by 4.5 μm from the condensing point p. Even on the surface of the polyimide substrate 1, the fluence is close to 0.4 J / cm ^2, and the polyimide substrate 1 is also ablated.

  Based on the above consideration, the polyimide substrate 1 is completely damaged by reciprocating the condensing point p while approaching the polyimide substrate 1 by 2 μm with the fluence of the condensing point p being about ½ of the copper ablation threshold. Therefore, it is concluded that only the copper film 2 could be selectively removed.

  The present invention is a method for selectively removing only a film on a substrate, which can be used for dry etching of a semiconductor integrated circuit, a display panel, etc., and has very high industrial applicability.

It is a figure explaining the positional relationship of a condensing point and a film | membrane. It is a schematic block diagram of the board | substrate top film removal apparatus of Embodiment 1. FIG. It is a figure which shows an ablation process area | region typically. It is a schematic block diagram of the board | substrate top film removal apparatus of Embodiment 2. FIG. In the surface photograph of the copper film on the polyimide substrate of Example 1, black lines L1, L2,... Parallel to the x axis are the processing marks of the copper film. It is a graph which shows the change of resistance value. FIG. 6 is a drawing of a cross-sectional observation result of A15 line L15 in FIG. 5. It is a schematic block diagram of the board | substrate top film removal apparatus used in Example 3. FIG. 4 is a surface photograph of a copper film on a polyimide substrate of Example 3. It is a schematic block diagram of the conventional trepanning processing apparatus.

Explanation of symbols

1 ... substrate 2 ... membrane 2b ... interface 3 ... condensing lens (collection) Light means)
p: Focusing point L: Ultrashort pulse laser D: First predetermined distance D '... ······ Second predetermined distance Z _r ··· Rayleigh range La, Lb ··· Precession beam N, N '······ Precession Shaped beam section

Claims (7)

A position having condensing means for condensing the ultra-short optical pulse laser, and the condensing point of the condensing means is separated from the interface between the substrate and the film on the substrate by a first predetermined distance toward the film side A focusing point setting step to be set to
A removal step of condensing and irradiating the ultrashort optical pulse laser on the film and removing the film by laser ablation;
A method for removing a film on a substrate, comprising:   The condensing point setting step initially sets the condensing point at a position separated from the interface by a second predetermined distance toward the film side, and thereafter sequentially brings the condensing point closer to the interface than the second predetermined distance. 2. The method of removing a film on a substrate according to claim 1, wherein the removal step is repeated until the substrate appears after setting the position. A condensing means for condensing the ultrashort optical pulse laser, and a position where the condensing point of the condensing means is within the Rayleigh range from a position away from the surface of the film on the substrate by the Rayleigh range. A focusing point setting step to be set within the range;
A fluence control step for making the fluence of the condensing point approximately equal to the ablation threshold of the film;
A method for removing a film on a substrate, comprising:   The substrate according to claim 1, further comprising a deflection rotation step for deflecting and rotating the ultrashort optical pulse laser incident on the condensing unit into a precessing beam. How to remove the film.   5. The film on a substrate according to claim 4, wherein the condensing means is a condensing lens, and further includes a transmission step of transmitting a node of the precessing beam to an entrance pupil of the condensing lens. Removal method.   The method for removing a film on a substrate according to claim 1, wherein NA of the light collecting means is 0.4 or more.   7. The method of removing a film on a substrate according to claim 1, wherein the film is made of a material having a threshold value smaller than an ablation threshold value of the substrate.

Images