JP2006205261A - Drilling device of printed circuit board - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a working device appropriate for a drilling method of a printed circuit board which improves the reliability of working and the quality of a hole and eliminates the need for a desmear process by performing working so as not to allow films to remain in the bottom of the hole. <P>SOLUTION: The substrate working device arranged in such a manner that a laser beam radiated from a UV laser source is distributed by a light deflector to two directions and the each energy density, energy space distribution and beam diameter can be set by discrete setting means is used. The energy density of the first UV laser beam is made higher than the resolution energy threshold value of a conductive layer and a hole is drilled down to an insulating layer immediate before the desired conductor layer. Next, the laser beam is switched to a second UV laser beam made lower than the resolution energy threshold value of the conductive layer and made higher than the resolution energy threshold value of the insulation layer and the remaining insulating layer is removed. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a printed circuit board processing apparatus, and more particularly to a printed circuit board processing apparatus suitable for processing a bottomed hole (blind hole) for connecting an upper conductor layer and a lower conductor layer using a laser beam.

FIG. 17 is a configuration diagram of a conventional laser processing apparatus. In this laser processing apparatus, the laser beam 2 output from the laser oscillator 1 is enlarged or reduced in diameter by M times by the collimator 3 and shaped by the aperture 4 to a diameter suitable for processing. Mirror 14 of the laser light shaping the corner mirror 5 and the machining head Z, via the first and second two galvanometer mirrors 15 _a, 15 _b is incident on the fθ lens 16, the galvano mirror 15 _a and the galvanometer mirror is positioned by 15 _b, normally incident on a predetermined position of the working surface from the fθ lens 16. Processing is performed for each processing region 18 corresponding to the fθ lens 16, and the regions are moved in the order of 18 ₁ ... 18 _{N in} the drawing by an XY work table (not shown).

  FIG. 18A is a diagram illustrating the operation of the collimator 3 and the aperture 4. In each distribution below, the vertical axis represents the laser beam energy, and the horizontal axis represents the beam diameter. Since the energy space distribution at the exit from the laser oscillator is generally a Gaussian distribution, the energy space distribution of the beam that has passed through the collimator 3 is also a Gaussian distribution. The laser light can change the energy space distribution by changing the magnification M of the collimator 3. That is, for example, when the enlargement ratio M is reduced, as shown in “a ′ distribution (dotted line)”, the energy space distribution of a small beam diameter and high energy density (high output density) is increased, and when the enlargement ratio M is increased. As shown in “b ′ distribution (dotted line)”, the beam diameter is increased, and an energy space distribution having a low energy density (low output density) is obtained.

  When the diameter of the aperture 4 is increased with respect to the beam diameter, energy is concentrated in the center portion, and the processed hole bottom (that is, the inner conductor layer) may be damaged. Therefore, by cutting out a relatively uniform energy distribution at the center of the beam by the aperture 4, an “A ′ distribution (dotted line)” or “B ′ distribution (solid line)” in which the energy intensity in the processed portion is substantially uniform. The bottom of the blind hole is not damaged. Here, the full output energy space distribution when the aperture 4 is removed from the optical axis and only the collimator 3 is used is hereinafter referred to as “C ′ distribution”.

  On the other hand, FIG. 18B shows the energy space distribution when the beam shaping unit 30 is inserted in the optical path of the laser light 2. By placing the beam shaping unit 30 in the optical path of the laser light 2, the energy space distribution is rectangularized, and is expanded or reduced by the collimator 3 (a distribution and b distribution in the figure), and further cut by the aperture 4. Thus, the uniformity of the energy magnitude in the processed part can be remarkably improved (solid lines in the figure: “A distribution”, “B distribution”). Hereinafter, this rectangular distribution is referred to as a “top hat shape”. As the beam shaping unit 30, a commercially available product such as a combination of an aspheric lens or a combination of diffractive optical elements can be used. Here, the full output energy space distribution when the aperture 4 is removed from the optical axis is hereinafter referred to as “distribution C”.

  The configuration of the printed circuit board is as follows: a substrate with FR-4 glass fibers (hereinafter referred to as “glass-containing substrate”) in which the surface is a conductor layer and the conductor layers and insulating layers containing glass fibers are alternately arranged; Copper foil with resin (hereinafter referred to as “RCC substrate”) on which a laminated plate of an insulating layer and a conductor layer is attached, and a substrate coated with a resin film or resin on the surface of the conductor layer (hereinafter referred to as “resin direct substrate”) and so on. For the insulating layer, an organic material such as epoxy or polyimide resin is mainly used. An inorganic material such as ceramics may be used as a filler instead of glass fiber as a reinforcing material for the insulating layer.

Conventionally, a processing method using a CO ₂ laser having a wavelength of 10.6 μm is widely known. For example, a method of processing a blind hole in an insulating layer of a resin direct substrate was introduced by an IPC review in the United States in 1982 and put into practical use as a CO ₂ resin direct method. Further, as a method of blind holes drilled in the glass-containing substrate, Patent Document 1 and Patent Document 2, after forming a window in advance remove the conductive layer corresponding to the hole diameter in the chemical etching or drilling, CO ₂ A method of processing an insulating layer with a laser is disclosed.

Further, in Patent Document 3, as a method of processing a plurality of conductor layers and insulating layers to be laminated, a conductor by a circumferential operation using ultraviolet (hereinafter referred to as UV) laser light that can efficiently process a metal. UV + CO ₂ laser method is disclosed for processing a blind hole or through hole by repeating the removal of the insulating layer by window processing and CO ₂ lasers layers.

However, in the case of drilling a resin layer by this CO ₂ laser, it is known that a film called “smear” having a thickness t _C (0.2 to 3 μm) remains on the bottom of the hole (immediately above the conductor layer). . Moreover, it has been found through experiments by the inventors that the thickness of the remaining film thickness t _C hardly changes even when the energy density of the CO ₂ laser pulse and the number of shots are increased.

The reason why the film remains in this way can be estimated as follows. CO ₂ laser processing is a method in which an insulating layer absorbs laser light to raise the temperature of the resin and cause thermal decomposition. Therefore, the thermal conductivity of copper, which is the inner layer conductor, is about three orders of magnitude higher than that of the resin. As a result, when the insulating layer becomes thinner, the heat flows to the inner layer conductor, so that the resin temperature cannot reach the decomposition temperature. A film having a thickness of 0.2 to 3 μm remains at the bottom of the hole.

  When the film remains in this manner, a chemical desmear process (conditioning, water washing, boiling, cooling, water washing, swelling, water washing, oxidation desmear, water washing, neutralization, water washing, drying, etc.) is performed to remove the film remaining at the bottom of the hole. Consisting of steps). Further, when the hole diameter is 100 μm or less, the wettability of the treatment liquid is reduced (the treatment liquid is difficult to enter the processed hole), and thus the desmear process reliability is lowered. The desmear treatment is originally intended to remove the resin film remaining on the bottom of the hole. However, when removing the film remaining on the bottom of the hole, the hole wall is removed by 3 to 5 μm, and the hole diameter is usually reduced. There was also a problem that the maximum diameter was increased by 10 μm.

  On the other hand, a method of blind hole processing on a resin direct substrate using a UV laser was introduced by the US IPC review in 1987 and put into practical use as a UV resin direct method. Further, Patent Document 4 discloses a UV direct method in which a conductor layer and an insulating layer are processed only by a UV laser as a method for processing a plurality of conductor layers and insulating layers to be laminated.

When a UV laser is used, an insulating layer does not remain at the bottom of the hole as in the case of the CO ₂ laser. However, if sufficient pulse energy is used for processing in order to obtain a practical processing speed, the energy density is excessive, so that the conductor layer at the bottom of the hole is scraped, and is provided to ensure plating strength. The irregularities on the surface of the conductor layer are not melted and decomposed. In particular, in the case of a wavelength conversion type UV laser that converts a wavelength using a non-linear optical element or the like, the pulse energy cannot be changed during the processing by the pulse width or the pulse frequency like a CO ₂ laser. Since the thickness variation is as large as about 20 μm with respect to the thickness of 65 μm, the hole bottom conductor layer is damaged as a result. Further, when the energy absorption rate of the insulating layer is low, the amount of energy reaching the hole bottom increases, thereby increasing the energy accumulation of the hole bottom conductor layer. For this reason, the resin on the upper surface of the conductor layer may decompose and vaporize. At this time, the insulating layer may be peeled off by the vaporization energy, and the insulating layer around the hole bottom corner may be peeled off in a ring shape. However, if the pulse energy is reduced, damage to the hole bottom can be reduced, but the number of pulse shots increases, so the processing speed decreases.

  In addition, when a glass-containing substrate is processed, not only the conductor layer is scraped due to excessive energy, but also glass protrudes from the side wall of the hole, and the side surface of the hole is punched out in a barrel shape.

The reason why the insulating layer becomes excessive in energy can be estimated as follows. For example, when the wavelength is 355 nm, the energy absorption rate of the UV laser light is about 30 to 80% epoxy material, about 70 to 75% copper, about 20% glass (70% transmission, 10% reflection), and the difference is large. Moreover, it is a difference in thermal conductivity, that is, epoxy-based material 0.8 to 0.85 Wm ⁻¹ K ⁻¹ , copper 386 Wm ⁻¹ K ⁻¹ , glass 1.04 to 1.09 Wm ⁻¹ K ⁻¹ , and the difference is large. Therefore, in the case of a glass-filled substrate, about 80% of energy is reflected and diffused inside the hole by the glass fiber. Therefore, if the pulse period is set to 0.0033 ms or less (pulse frequency of 3 kHz or more), the resin on the hole side wall It is made into a barrel shape, and the protrusion of the glass fiber becomes large and the hole quality is lowered.

Further, in this processing method, processing is performed at an energy density of 3 J / cm ² or more in order to remove the conductor layer even when glass fibers are not contained in the insulating layer. For this reason, the control of heat becomes difficult due to the difference in the material property values with respect to the UV light, and the conductor layer at the bottom of the hole is damaged. Therefore, it is difficult to obtain a practical hole quality.

  Further, when this UV light is used, as shown in FIG. 18 (a), processing is performed with A ′ or C ′ distribution up to the middle of the hole, so that unevenness occurs in the bottom of the hole, and the remaining film after In some cases, the time required for the removal process becomes long, or the conductor layer at the bottom of the recess is partially damaged.

  As a means for solving such a problem of damaging the bottom conductor, for example, as disclosed in Non-Patent Document 1, the energy density of the laser beam is higher than the decomposition energy threshold of the resin layer and the decomposition energy of the conductor layer. There has been proposed a method of selectively etching only the resin layer portion by setting it lower than the threshold value.

  Here, the decomposition energy threshold is an energy density (referred to as fluence) that is the product of the irradiation power density of laser light and the pulse width (referred to as a fluence). It is known that processing does not start), but that value.

Further, in Patent Document 5, in order to remove residues and smears at the bottom of the hole and in the peripheral part, the UV light of the excimer laser is converted into a uniform distribution of a line shape or a rectangular shape with a beam shaping optical element. After that, a method of irradiating and cleaning a plurality of holes at the same time is disclosed. However, if this is applied to the outermost resin layer, the surface of the resin layer will be damaged.
JP 58-64097 A US Pat. No. 5,010,232 JP-A-1-266683 US Pat. No. 5,593,606 Japanese Patent No. 2983481 "Laser Ablation and its Applications", Corona, (1999), page 146, lines 6 to 13.

  The object of the present invention is to clarify the specific value of the decomposition energy threshold for the above-mentioned resin and conductor layer, and to use the difference to selectively remove the residual film on the bottom of the laser beam. It is an object of the present invention to provide an apparatus capable of switching from the energy density to consistently performing the process from drilling to removing the residual film on the bottom of the hole and eliminating the desmear process.

According to the experiments by the inventors, the decomposition energy threshold in the case of a UV laser with a wavelength of 355 nm is an epoxy-based material of 0.3 to 0.5 J / cm ² , copper of 0.8 to 1.0 J / cm ² , it has been found that a glass 5~6J / cm ^2. The present invention finds that this decomposition energy threshold is slightly different between the epoxy-based material and copper, and using this, the energy density of the second UV laser beam is made to be higher than the decomposition energy threshold of the insulating layer. The target conductor layer was successfully exposed by processing the remaining insulating layer at an energy density lower than the decomposition energy threshold of the conductor layer.

  Furthermore, in the present invention, when processing is performed using the first UV laser light in the step before performing the finishing process using the second UV laser light, the energy space distribution of the first UV laser light is topped. It turned out that it was necessary to make a hat shape and to flatten the hole bottom.

  Therefore, in order to solve the above problems, at least two UV laser light sources are prepared, the energy space distribution is made into a top hat shape by the beam shaping unit, and the beam diameter and energy density of the two systems can be set independently. What should I do? At this time, when the beam emitted from one UV laser light source is switched to two systems by an acousto-optic device or the like, it is further preferable because the size of the apparatus is reduced. Further, if the two beams are irradiated on the workpiece with the same optical axis, it is not necessary to move the XY stage when switching the laser beam, so that the processing time is shortened.

  According to the present invention, in the method for processing a blind hole in a conductor layer provided inside a printed circuit board in which conductor layers and insulating layers are alternately laminated, the energy density of the first UV laser light is changed to the conductor layer. And an energy density higher than the decomposition energy threshold of the insulating layer is processed halfway through the insulating layer immediately before the inner conductor layer, and the energy density of the second UV laser beam is set to the decomposition energy threshold of the inner conductor layer. Since the remaining film thickness of the insulating layer is removed with an energy density lower than the value and higher than the decomposition energy threshold value of the insulating layer, the inner conductor layer can be exposed with almost no damage.

  That is, since the remaining film of the insulating layer at the bottom of the hole can be removed, a desmear treatment process is unnecessary or the processing time can be shortened.

  Moreover, since peeling of the hole bottom corner portion does not occur, the hole quality is improved.

  Furthermore, since the laser beam is configured to be distributed using a deflector, the laser beam for processing the conductor layer and the UV laser beam for removing the remaining film of the insulating layer with one UV laser source. Therefore, it is possible to reduce the size and cost of the apparatus.

Further, when the UV laser light and the UV laser light are coaxially combined with the laser light distribution means, the UV processing by the UV laser light 2 _i and the hole bottom finishing by the UV laser light 2 _k are coaxial in the two-head configuration. Thus, the number of table movements can be reduced, and the processing time can be shortened.

Hereinafter, embodiments of the present invention will be specifically described using examples.
(Device Example 1)
FIG. 1 is a configuration diagram of a laser processing apparatus according to the first embodiment of the present invention. In the figure, the UV laser oscillator 1 uses polarized Q-switched YVO ₄ laser light (repetition frequency: 10 kHz to 100 kHz) as the third harmonic (355 nm) using a nonlinear optical element LBO (LiB ₃ O ₅ ). . Here, the laser oscillator 1 was installed so that the electric field vector of the laser light 2 was in the direction parallel to the paper surface. The acoustooptic deflector 6 generally has a structure in which a piezoelectric element is bonded to an acoustooptic element. When an operating voltage (RF voltage) is not applied to the piezoelectric element, the incident light travels straight (transmits). When an operating voltage is applied, the incident light is Bragg diffracted by the ultrasonic wave generated in the acousto-optic element via the piezoelectric element, and the emission angle is changed and deflected. Using this acousto-optic deflector 6, the laser beam 2 is caused to travel straight (2 _k ) or deflect (2 _i ). The deflected laser light 2 _i is made into a top hat-shaped energy space distribution by the beam shaping unit 30 _i , narrowed by the collimator 3 _i, and shaped to a diameter suitable for processing by the aperture 4 _i . Then, the light is reflected by the corner mirror 5 _i , and the polarization direction is converted by the half-wave plate 11 so that the electric field vector is in the direction perpendicular to the paper surface S _i (hereinafter referred to as S-polarized light). It is reflected by the polarization beam combiner (polarization beam splitter) 10 for reflecting incident on machining head Z through the corner mirror 5 _k, composed of a head portion Z (2 pieces of galvano-mirror and the fθ lens. FIG. 17) and is incident perpendicularly to the processing surface.

On the other hand, the UV laser light 2 _k when traveling straight passes through the beam shaping unit 30 _k, is enlarged or reduced by the collimator 3 _k , and is shaped to a diameter suitable for processing by the aperture 4 _k . Since 2 _k is P-polarized light (P _k ), it passes through the polarizing beam combiner (polarizing beam splitter) 10 and enters the machining head Z via the corner mirror 5 _k , and the head portion Z (two galvanometer mirrors). And the fθ lens) and is incident perpendicularly to the processing surface. Here, the optical paths of the laser beams 2 _i and 2 _k were adjusted so as to travel on the same optical axis.

Further, by adjusting the magnifications M _i and M _k of the collimators 3 _i and 3 _k while keeping the aperture diameter constant, energy space distributions A _i and B _k suitable for processing can be obtained. (See FIG. 18B.) Further, when the aperture 4 _i is removed from the optical path, the energy space distribution can be made to be a full output C _i distribution.

  This apparatus example is very simplified as an apparatus having two laser optical systems for processing and capable of setting energy distribution almost independently, and the apparatus can be miniaturized. Further, in the case of a deflector using a conventional acousto-optic device or the like, the use efficiency of energy is improved because transmitted light that has been input to a beam damper or the like and discarded is used. Further, since the two laser beams are on the same optical axis at the time of irradiating the workpiece, the machining time can be shortened without moving the stage in the case of consistent machining.

  In this example of the apparatus, the acoustooptic deflector 6 is known in which the deflection direction of the input light and the output light changes by 90 degrees. However, when this is used, it is not necessary to convert the polarization direction. The wave plate 11 becomes unnecessary.

(Processing example 1)
FIG. 2A is a diagram showing the energy distribution and the hole shape for the glass substrate processing method when the apparatus example 1 is used, and FIG. 2B is a timing chart showing the processing timing at this time. is there.

Here, the specifications of the laser beams 2 _i and 2 _k are defined in the present specification as follows. Note that the subscript a is a division of the laser beams i and k.
E _Pa : Pulse energy (= E _{P0a × (} d _Aa / d _0a / M _a ) ² )
E _P0a : Pulse energy at the laser oscillator exit port d _Aa : Aperture diameter d _0a : Beam diameter of laser beam 2 _a before entering 30 _a M _a : Collimator magnification T _Pa : Pulse width W _Pa : Peak Output (= E _Pa / T _Pa )
E _dSa : Energy density (= E _Pa / [π (d _Sa / 2) ² ])
d _Sa : Processing spot diameter (= d _Aa × [(L _a / f _a ) -1])
L _a : Distance from fθ lens to workpiece f _a : Focal length of fθ lens T _PPa : Pulse period T _GC : Galvano mirror positioning period E _Sa : Decomposition energy threshold N _a : Number of pulse shots V _a : Material removal amount

In FIG. 2A, 21 is an outer conductor layer made of copper (thickness 9 μm), 22 is an insulating layer made of epoxy (thickness 50 μm), 24 is an inner conductor layer made of copper, and t is the first laser. This is the film thickness left after processing with light 2 _i . First, a first laser beam 2 _i deflected by applying a high frequency voltage to the acoustooptic deflector 6 was obtained. Here, the decomposition energy threshold E _Si for the first laser light 2 _i is the resin (0.3 to 0.5 J) of the outer conductor layer 21 (0.8 to 1.0 J / cm ² ) and the insulating layer 22. The higher value of the decomposition energy threshold value with respect to / cm ² ) is 0.8 to 1.0 J / cm ² .

Since the energy density required for the efficient removal of the conductor layer of the laser beam 2 _i is about 3.0 J / cm ² or more from the experiment, it is necessary to focus the beam, so that the laser beam can be obtained as large as possible. 2 _i applied the _{C i} distribution. Pulse width T _{Pi is set} to 25 ns, pulse period T _PPi is set to 0.03 ms (frequency 30 kHz), _peak output W _{Pi is set to} 2.4 _{to 4.0} kW, and pulse energy E _Pi is set to 0.06 to 0.10 mJ. When the spot diameter d _Si was 40 μm, the pulse energy density E _dSi could be _4.8 to 8.0 J / cm ² , and most of the conductor layer 21 and the insulating layer could be removed. Further, in order to process a window having a hole diameter of 100 μm on a conductor layer having a thickness of 9 μm, it is necessary to cause the beam spot to make several concentric circular movements (arrows in the figure). In this case, the total number of necessary pulses _Ni was 100 shots. Here, the thickness t leaving the insulating layer is preferably 5 to 10 μm. Here, since the energy space distribution of the first laser beam 2 _i has a top hat shape, the remaining film thickness can be made substantially uniform.

The remaining film t of the insulating layer 22 was removed by the second laser light 2 _k obtained by stopping the application of the high frequency voltage to the acoustooptic device 6. Energy density of the UV laser required for removing the insulating layer of the hole bottom of the laser beam _{2 k} is 0.3~0.5J / cm ² or more (practically the 0.5 J / cm ² or higher). For this reason, it is possible to remove the insulating layer uniformly and to remove the oxide layer on the surface of the inner layer conductor to improve the peeling strength of the resin, and the top hat shape hardly damages the conductor layer material A B _k distribution was applied. If the pulse energy E _Pk of the second laser beam 2 _k is higher than the decomposition energy threshold value E _{Sk and} lower than the decomposition energy threshold value of the inner conductor layer 24, the inner conductor layer 24 is not damaged. . Therefore, the pulse width T _Pk is 25 ns, the pulse period T _PPk is 0.03 ms (wave number 30 kHz), the peak output W _Pk is 2.3 to 3.6 kW, and the pulse energy E _Pk is 0.06 to 0.09 mJ. As a result, the processing spot diameter d _Sk was set to 120 μm larger than the window diameter of 100 μm, and the pulse energy density E _dSk was set to _0.5 to 0.8 J / cm ² . This condition is higher than the decomposition energy density of 0.3 to 0.5 J / cm ² of the epoxy insulating layer which is practically necessary, so that the remaining film of the insulating layer can be removed. Since the decomposition energy density was lower than 0.8 to 1.0 J / cm ² , the inner conductor layer was not damaged. At this time, the insulating layer removal rate was about 0.5 μm / pulse, and the required number of pulses N _k was 30 shots.

FIG. 2B shows a timing chart of this method. First in time that a laser beam 2 _i is the second laser beam 2 _k seen to be superimposed weakly. This occurs because the diffraction efficiency of the acousto-optic element is not 100%, but it is lower than the decomposition energy threshold value of the resin. Therefore, there is an adverse effect on the processing of the first laser beam 2 _i. There wasn't.

  According to this method, no peeling occurred between the conductor layer and the insulating layer in the periphery of the hole bottom corner. In addition, after the drilling, there was a case where a slight dispersal and scattered matter remained on the conductor layer at the bottom of the hole. This is a soft etching process that is the first step of the plating process after the laser processing. There was no problem because it can be removed together with the oxide layer.

(Processing example 2)
FIG. 3A is a diagram showing the energy distribution and the hole shape for the processing method of the resin direct substrate when the apparatus example 1 is used, and FIG. 3B is a timing chart showing the processing timing at this time. is there. In FIG. 3A, 22 is an epoxy insulating layer (thickness 50 μm), 24 is an inner conductor layer made of copper, and t is a film thickness left after processing with the first laser beam 2 _i . Here, the decomposition energy threshold value E _Si for the first laser light 2 _i is determined in the decomposition energy threshold value for the resin of the insulating layer 22 and is 0.3 to 0.5 J / cm ² .

Since the processing method is almost the same as the processing example 1, only different points will be described. Since the energy density required for the resin layer of the first laser beam 2 _i is lower than that for the conductor layer, the A _i distribution was applied. Pulse width T _Pi is 25 ns, pulse period T _PPi is 0.03 ms (frequency 30 kHz), peak output W _{Pi is} 1.0 _to 1.6 kW, and pulse energy E _Pi is 0.025 _to 0.040 mJ. When the spot diameter d _Si was 50 μm, the pulse energy density _EdSi was 1.3 to 2.0 J / cm ² , and most of the insulating layer could be removed. The required number of pulses _Ni was 25 shots. The thickness t leaving the insulating layer was set to 5 to 10 μm in consideration of the variation in the insulating layer thickness. Here, since the energy space distribution of the first laser beam 2 _i is a top hat shape, the remaining film thickness can be made substantially uniform.

The remaining film t of the insulating layer 22 is removed by the second laser light 2 _k having a top hat-shaped B _k distribution. 25ns pulse width _{T Pk,} the pulse period _{T PPk 0.03ms} (wavenumber 30 kHz), peak - a click output _{W Pk} was applied 0.4～0.6KW. Here, unlike machining example 1, the machining spot diameter d _Sk cannot be made larger than the window diameter of 50 μm in order to avoid damage other than the hole bottom, so that the pulse energy E _Pk is set to 0.010 to 50 μm, which is the same as the window diameter. The pulse energy density E _dSk was set to _0.5 to 0.8 J / cm ² by _decreasing to 0.016 mJ. At this time, the insulating layer removal rate was about 0.5 μm / pulse, and the required number of pulses N _k was 15 shots.

(Device example 2)
FIG. 4 is a block diagram of a laser processing apparatus according to the second embodiment of the present invention, in which the insulating layer can be removed with a CO ₂ laser after the window is opened with a UV laser. In the figure, a UV laser beam 2 _i output from a laser oscillator 1 _i passes through a beam shaping unit 30 _i , is enlarged or reduced in diameter by a collimator 3 _i, and is shaped to a diameter suitable for processing by an aperture 4 _i. The Then, the light enters the machining head Z _i through the corner mirror 5 _i, is condensed by the head portion Z _i , and enters the machining surface perpendicularly.

The CO ₂ laser light 2 _j output from the laser oscillator 1 _j passes through the beam shaping unit 30 _j, and the diameter is enlarged or reduced by the collimator 3 _j, and is shaped to a diameter suitable for processing by the aperture 4 _j . Then, the light enters the machining head Z _j via the corner mirror 5 _j, is condensed by the head portion Z _j , and enters the machining surface perpendicularly.

The UV laser light 2 _k output from the laser oscillator 1 _k passes through the beam shaping unit 30 _k , is expanded or reduced in diameter by M _k times by the collimator 3 _k, and is shaped to a diameter suitable for processing by the aperture 4 _k. The Then, entering the machining head Z _k through the corner mirror 5 _k, perpendicularly incident on the processed surface is condensed by the head portion Z _k.

By controlling the beam shaping units 30 _i , 30 _j , and 30 _k for any of the laser beams 2 _i , 2 _j , and 2 _k , the energy space distribution can be shaped from a Gaussian distribution to a top hat shape.

Further, by changing the aperture diameters d _Ai , d _Aj , and d _Ak , it is possible to adjust the machining beam diameters d _Si , d _Sj , and d _{Sk on} the workpiece surface while keeping the energy density constant.

Further, different energy space distributions A _i can be _obtained by changing the enlargement factors M _i , M _j , M _k of the collimators 3 _i , 3 _j , 3 _k while keeping the aperture diameters d _Ai , d _Aj , d _Ak constant. , B _i , A _j , B _j , A _k , B _k . Further, when the aperture is removed from the optical path, the energy space distribution can be set to full output C _i , C _j , C _k .

Further, the head Z _i , the head Z _j , and the head Z _k can each be processed continuously, and each has a moving amount capable of processing the entire printed board. Laser processing is performed so that the inter-axis distances L _ij and L _jk are minimized so that the movement distance of the XY work table is minimized, and the same printed circuit board can be processed on the work table. It is arranged linearly on the device.

(Processing example 3)
FIG. 5 is a diagram showing the energy distribution and the hole shape when processing a glass-containing substrate using the apparatus example 2, and FIG. 6 is a timing chart showing the processing timing at this time. In these drawings, (a) to (c) in FIG. 5 correspond to (a) to (c) in FIG.

In FIG. 5A, the outer conductor layer 21 (thickness 9 μm) is removed by the energy in the range where the pulse energy E _Pi of the laser light 2 _i is higher than the decomposition energy threshold E _Si . Therefore, 30 ns pulse width _{T Pi} of the laser beam _{2 i,} the pulse period _{T PPi} 0.04 ms (frequency 25 kHz), peak - 4000 W _a click output _{W Pi,} a _{C i} distribution of 0.12mJ pulse energy _{E Pi} applied Then, the machining spot diameter d _Si was set to 60 μm, and the pulse energy density _EdSi was _set to 4 J / cm ² . This condition satisfies a practically required energy density of 3 J / cm ² or more (which is an experimental value), so that the conductor layer can be removed. Further, in order to process a window having a hole diameter of 100 μm in the conductor layer that is larger than the beam diameter, the beam spot was moved in a circumferential direction (arrow in the figure). In this case, the required number of pulses _Ni was 80 shots.

In FIG. 5B, the insulating layer 22 (thickness 50 μm) is almost removed by the laser beam 2 _j . Here, 10 [mu] s pulse width _{T Pj,} Pi - 800 W a click output _{W Pj,} pulse energy _{E Pj} a _{C j} distribution of 8 mJ, and the working spot diameter _{d Sj} in greater 150μm than window size 100 [mu] m, the pulse energy density E _dSj was 45 J / cm ² . The number of pulses _Nj was 3 shots. Under this condition, most of the insulating layer can be removed, but a smear having a thickness t _c of 0.1 to 3 μm remained at the bottom of the hole.

In FIG. 5 (c), the smear of the insulating layer 22 is removed by the laser beam _{2 k.} Thus, 30 ns pulse width _{T Pk,} the pulse period _{T PPK} 0.04 ms (wavenumber 25 kHz), peak - applying 4000W a click output _{W Pk,} the _{B k} distribution of 0.12mJ pulse energy _{E Pk,} working spot diameter d _Sk was set to 150 μm larger than the window diameter of 100 μm. In this case, the pulse energy density E _dSk is 0.7 J / cm ² . At this time, the insulating layer removal rate was about 0.5 μm / pulse, and the required number of pulses N _k was 10 to 15 shots.

And according to the above method, the process of a conductor layer and the process of the insulating layer containing glass fiber can be performed most efficiently. The total energy of the UV laser beam reaching the hole bottom is about 0.55 mJ (≈0.12 mJ × 10 shots × (φ100 / φ150) ² ), and the total energy in the case of directly processing the insulating layer is 6 mJ (0 .10 mJ × 50 shots) or less. Therefore, even if the material has a low absorption rate of UV laser light, the bottom of the hole is not damaged, and no peeling occurs between the conductor layer and the insulating layer in the periphery of the hole bottom corner. Further, in the case of this processing example, even if the thickness of the insulating layer varies, the remaining film thickness (smear thickness) after processing the CO ₂ laser beam 2 _j does not change, so that the processing reliability is improved.

(Processing example 4)
FIG. 7 is a diagram showing the energy distribution and hole shape when processing the RCC substrate using the apparatus of apparatus example 2, and FIG. 8 is a timing chart showing the processing timing at this time. (A) to (c) in FIG. 7 correspond to (a) to (c) in FIG.

In the RCC substrate, the conductor layer 21 was processed in the same manner as in FIG. 5A as shown in FIG. The removal processing of the insulating layer 22 by the CO ₂ laser beam 2 _{j in} FIG. 7B has a lower decomposition energy density threshold value than that of the glass-containing substrate, so that the pulse width _TPj of the B _j distribution is 10 ms, The conditions of an output W _Pj of 500 W and a pulse energy E _Pj of 5 mJ were applied. When the machining spot diameter d _{Sj is set} to 150 μm, a pulse energy density E _dSj of 30 J / cm ² is obtained, which satisfies a practical pulse energy density of 10 J / cm ² or more. According to this condition, most of the insulating layer can be removed by 1 to 2 shots, but a residual film (smear) having a thickness t _c of 0.1 to 3 μm remains at the bottom of the hole. In the same manner, the remaining film was removed.

(Processing example 5)
FIGS. 9A and 9C are diagrams showing energy distributions and hole shapes when processing an FR-4 substrate without a conductor layer on the surface using the apparatus of apparatus example 2, and FIG. , (D) is a timing chart showing the timing at that time. In the FR-4 substrate processing without a conductor layer on the surface, as shown in FIG. 5A, the hole entrance diameter is such that the pulse energy E _Pj of the laser beam 2 _j for removing the insulating layer and the decomposition energy of the insulating layer Except for being determined by the threshold value E _{Sj, the same} process as the removal process of the insulating layer in FIG. Further, FIG. (B) the removal of the insulating layer remaining film of the hole bottom by the laser beam 2 _k of the insulating layer 22 were performed as in the residual film removal hole bottom in FIG. 5 (c).

(Processing example 6)
FIGS. 10A and 10C are views showing energy distributions and hole shapes when processing a resin substrate having no conductor layer on the surface using the apparatus of apparatus example 2, and FIGS. d) is a timing chart showing the timing at that time. In the case of a resin substrate having no conductor layer on the surface, as shown in FIG. 4A, the hole entrance diameter is the pulse energy E _Pj of the laser beam 2 _j for removing the insulating layer and the decomposition energy threshold value E _{Sj of the} insulating layer. Except as determined by the above, it was performed in the same manner as the processing of the insulating layer in FIG. Further, residual film removal of the insulating layer of the hole bottom by the laser beam 2 _k of the insulating layer 22 in FIG. (C) was performed in the same manner as the residual film removal hole bottom in FIG. 7 (c).

(Processing example 7)
11 (a) and 11 (c) show the energy distribution and hole shape when processing a resin substrate without a conductor layer (or FR-4 substrate without a conductor layer) on the surface using the apparatus of apparatus example 2. FIG. (B), (d) is a timing chart showing the timing at that time. In the resin substrate having no conductor layer on the surface, as shown in FIG. 5A, the _Bi distribution of the UV laser beam 2 _i is applied instead of the laser beam 2 _j , and the pulse energy E _Pi and the material of the insulating layer are used. the insulating layer in consideration of the variation amount of the thickness of the insulating layer, t is processed to leave 5 to 10 [mu] m, as shown in FIG. (c), the hole bottom by the laser beam 2 _k remaining film thickness t Was removed.

(Device Example 3)
FIG. 12 is a configuration diagram of a laser processing apparatus according to the third embodiment of the present invention. Components having the same or the same functions as those in FIG. 4 are denoted by the same reference numerals and description thereof is omitted.

In the figure, 10 is a polarization beam combiner (polarization beam splitter) installed on the optical axis of laser light 2 _i , 11 is a half-wave plate installed on the optical axis of laser light 2 _k , and 12 is all It is a reflective corner mirror. The total reflection corner mirror 12 is arranged so as to be coaxial with the laser beam 2 _i that reflects the laser beam 2 _k and passes through the polarization beam combiner 10. The laser oscillator 1 _i and the laser oscillator 1 _k are arranged so that the polarization direction and the optical axis are parallel to each other.

As a result, the P-wave component P _i of the laser beam 2 _i of B _i distribution, the laser light coaxially of the S-wave component S _k of the laser beam 2 _k of B _k distribution is simultaneously or serially added on the time axis The That is, it is possible to supply beams having different energy densities, output densities, and spot diameters to the processing portion in a state where the output of the head Z _i is constant. Further, by removing the half-wave plate 11 and the total reflection corner mirror 12 from the optical axis, simultaneous and individual processing can be performed with the laser light 2 _i and the laser light 2 _k as in the case of FIG.

When the laser oscillator 1 _i and the laser oscillator 1 _k are installed so that the electric field vector of the laser light 2 _i is parallel to the paper surface and the electric field vector of the laser light 2 _k is vertical to the paper surface, the polarization direction of the laser light 2 _k is set. Becomes the S wave component _Sk , so that the half-wave plate 11 in FIG. 12 can be omitted.

(Processing example 8)
FIG. 13 is a diagram illustrating a processing example of the RCC substrate by the apparatus of the apparatus example 3.
In the same figure, the residual resin thickness at the bottom of the hole of the C _i distribution laser beam 2 _i and B _k distribution in which the conductor layer processing and the insulating layer processing are set to the conductor layer removal energy density can be removed, but the conductor layer is damaged. the laser beam 2 _k without energy density, leaving 10μm for a given insulating layer thickness, was simultaneously processing halfway. Subsequently, the remaining resin thickness was continuously shot with a laser beam 2 _k with a head Z _k to make a hole. As a result, the conductor layer and the insulating layer can be removed without being affected by the thickness error of the insulating layer in the manufacturing process, and the hole quality is improved because the blind hole can be processed without damaging the hole bottom. The total pulse shot time of 0.012 seconds (25 kHz, 300 shots) after positioning of the galvanometer mirror was not changed by simultaneous processing.

(Processing example 9)
FIG. 14 is a diagram illustrating a processing example of a resin substrate (layer thickness: 40 μm) by the apparatus of the third embodiment.
In the figure, the laser light 2 _i with A _i distribution set to the insulating layer removal energy density and the laser light 2 _k with energy density that can remove the insulating layer with B _k distribution but do not damage the conductor layer, Processing was simultaneously performed halfway leaving 10 μm with respect to the insulating layer thickness, and then continuously shot with a laser beam 2 _k with a head Z _k to make a hole. As a result, the conductor layer and the insulating layer can be removed without being affected by the thickness error of the insulating layer in the manufacturing process, and the hole quality is improved because the blind hole can be processed without damaging the hole bottom. In addition, the total pulse shot time of 0.001 second (40 kHz, 40 shots) after positioning of the galvanometer mirror can be reduced by half to 0.002 second (40 kHz, 80 shots) when used alone.

(Example 4)
FIG. 15 is a diagram showing a configuration of a laser processing apparatus according to the fourth embodiment of the present invention. Components having the same or the same functions as those in FIGS. 4 and 12 are denoted by the same reference numerals and description thereof is omitted. To do. The laser source 1 _i was installed so that the electric field vector of the laser beam 2 that is linearly polarized light becomes a P-wave component in the direction parallel to the paper surface. In the figure, reference numerals 6 _a and 6 _b denote acousto-optic deflectors, which are installed on the optical axis of the laser beam 2 whose electric field vector is parallel polarized light. Reference numeral 8 denotes a beam damper for 0th-order diffracted light (transmitted light).

Since the laser beam 2 _i and the laser beam 2 _k cause energy loss due to the acoustooptic deflectors 6 _a and 6 _b , the energy density and the output density are the same as the laser beam 2 _i and the laser beam 2 _k in FIG. , Each decrease by about 15%. However, since the characteristics such as the energy space distribution do not change, by adjusting the output, it was possible to perform substantially the same processing as the apparatus example 1 shown in FIG.

(Processing example 10)
In the case of processing a glass-containing substrate, the procedure was as follows. First, the conductor layer is removed by the laser beam 2 _i having the C _i distribution with the energy density set to the conductor layer removal energy density, the insulating layer is then removed by the laser beam 2 _{j having the} A _j distribution, and the B _k distribution The remaining film at the bottom of the hole was removed by laser light 2 _k to form a blind hole. In this case, the longest shot time after positioning the galvanometer mirror was 0.0012 seconds when the conductor layer was removed. In addition, the removal of the insulating layer was 0.003 seconds, and the removal of the remaining film at the bottom of the holes was 0.0004 seconds. Therefore, since the remaining film at the bottom of the hole can be removed while the galvanometer mirror is moved to remove the conductor layer, the substantial processing speed is the same as that of the apparatus example 1 shown in FIG. As a result, it is possible to alternate the operation of the laser source 1 _i and the laser source 1 _k in FIG. 4 in one laser source 1 _i, it was possible to reduce the cost of the apparatus.

(Device Example 5)
FIG. 16 is a diagram showing a configuration of a laser processing apparatus according to the fifth embodiment of the present invention, which is the same as the above-described FIGS. 4, 12, and 15 related to apparatus example 2, apparatus example 3 or apparatus example 4. Or the thing of the same function attaches | subjects the same code | symbol and abbreviate | omits description.

Which in apparatus example 4, arranged acousto-optic device 6 _a, 6 _b, by arranging the 1/2-wavelength plate 11 and the total reflecting corner mirror 12 and a polarization beam combiner 10, from a single laser source 1 _i The laser beam 2 _i and the laser beam 2 _k are separated or made coaxial.

Further, by removing the half-wave plate 11 and the total reflection corner mirror 12 from the optical axis, the laser beam 2 _i and the laser beam 2 _k (one-dot chain line) are individually processed as in the case of FIG. Can also be done.

(Processing example 11)
The resin direct substrate (the thickness of the insulating layer is 40 μm) was processed by the laser processing apparatus according to apparatus example 5 in the same manner as in apparatus example 3. As a result, the hole quality was improved as in the apparatus example 3. Further, although the laser light 2 _i and the laser light 2 _k after the positioning of the galvano mirror were shot with different times, the total pulse shot time 0.001 second (40 kHz) by increasing the output of the laser light 2 _i. , 40 shots).

  Further, by removing the half-wave plate 11 for coaxializing the laser light and the total reflection corner mirror 12 from the optical axis, the function of the laser processing apparatus shown in the apparatus example 3 can be made equivalent. . That is, the same function as that of the apparatus example 3 can be provided by one laser source. As a result, the apparatus cost could be reduced.

1 is a configuration diagram of a laser processing apparatus according to a first embodiment of the present invention. It is a figure which shows the energy distribution and the shape of a hole of the process part using the laser processing apparatus which concerns on this invention in order of a process. It is a figure which shows the energy distribution and the shape of a hole of the process part using the laser processing apparatus which concerns on this invention in order of a process. It is a block diagram of the laser processing apparatus which concerns on the 2nd Embodiment of this invention. It is a figure which shows the energy distribution and the shape of a hole of the process part using the laser processing apparatus which concerns on this invention in order of a process. It is a figure which shows the timing chart of FIG. It is a figure which shows the energy distribution and the shape of a hole of the process part using the laser processing apparatus which concerns on this invention in order of a process. It is a figure which shows the timing chart of FIG. It is the energy distribution of the process part using the laser processing apparatus which concerns on this invention, the shape of a hole, and a timing chart. It is the energy distribution of the process part using the laser processing apparatus which concerns on this invention, the shape of a hole, and a timing chart. It is the energy distribution of the process part using the laser processing apparatus which concerns on this invention, the shape of a hole, and a timing chart. It is a block diagram of the laser processing apparatus which concerns on the 3rd Embodiment of this invention. It is the energy distribution of the process part using the laser processing apparatus which concerns on this invention, the shape of a hole, and a timing chart. It is the energy distribution of the process part using the laser processing apparatus which concerns on this invention, the shape of a hole, and a timing chart. It is a block diagram of the laser processing apparatus which concerns on the 4th Embodiment of this invention. It is a block diagram of the laser processing apparatus which concerns on the 5th Embodiment of this invention. It is a block diagram of the conventional laser processing apparatus. It is a figure which shows the effect | action of a beam shaping unit, a collimator, and an aperture.

Explanation of symbols

1, 1 _i , 1 _j , 1 _k laser oscillator 2 laser light 3, 3 _i , 3 _j , 3 _k collimator 4, 4 _i , 4 _j , 4 _k aperture 5, 5 _i , 5 _j , 5 _k corner mirror 14 mirror 15 _a, 15 _b galvanometer mirror 16 f [theta] lens 30,30 _{i, 30} j, _{30 k} beam shaping unit 2,2 _{i, 2} j, _{2 k} laser beam _{Z, Z i, Z j,} Z k head

Claims (4)

  A beam emitted from one ultraviolet laser source is distributed in two directions by an optical deflector, and each energy density, energy space distribution, and beam diameter are set by individual setting means, and then both beams follow the same optical path. A printed circuit board processing apparatus characterized by being advanced.   A beam emitted from one ultraviolet laser source is distributed in two directions by an optical deflector, and the side of the deflected light is squeezed by a beam diameter adjusting means to form a high energy first ultraviolet laser beam, and its transmitted light 2. The printed circuit board processing apparatus according to claim 1, wherein the second side is expanded by a beam diameter adjusting means to form a low energy second ultraviolet laser beam, and then both beams travel on the same optical path.   A beam emitted from one laser source is distributed in two directions by an optical deflector, and each energy density, energy space distribution, and beam diameter are set by individual setting means, and then both beams travel on the same optical path. A printed circuit board processing apparatus characterized by being configured as described above.   A beam emitted from one laser source is distributed in two directions by an optical deflector, and the side of the deflected light is squeezed by a beam diameter adjusting means to form a high energy first laser beam, and the side of the transmitted light 4. The printed circuit board processing apparatus according to claim 3, wherein the laser beam is expanded by a beam diameter adjusting means to form a low energy second laser beam, and then both beams travel on the same optical path.

Images