HK1078050A - Laser machining method and laser machining apparatus - Google Patents

Description

Laser processing method and laser processing apparatus

Technical Field

The present invention relates to a laser processing method and a laser processing apparatus suitable for processing a printed circuit board or a semiconductor.

Background

The pulse energy of the UV laser depends on the output release time relative to the excitation time, and in general, the lower the frequency, the greater the resulting pulse energy, and the higher the frequency, the lower the pulse energy.

The peak output power and the pulse width depend on the pulse frequency, the kind of crystal generating the fundamental wave, the length, and the LD (laser diode) output power exciting the crystal, and therefore the output release time, i.e., the pulse width, becomes larger as the frequency increases. Since the crystal may be damaged when the pulse frequency is low, the crystal is not oscillated in the low frequency region to protect the crystal, and the maximum position of the output power is set at a position slightly higher than the lower limit frequency region. The peak output power decreases slowly with increasing frequency.

Fig. 10 is a characteristic diagram showing UV laser output characteristics of high peak and narrow pulse width excited by YVO4, which is typical and suitable for printed circuit board processing.

The YVO4UV laser has an External Cavity structure of its fundamental crystal and a 3 rd harmonic (THG) wavelength conversion crystal, and when a Q-switch (Q-SW) is turned on, fundamental waves having a wavelength of 1064nm oscillate and are simultaneously wavelength-converted. Since an electro-optical device of a polarization modulation system having a high response speed to Q-SW is used, there is almost no fundamental wave remaining and the pulse width is narrow, but the crystal is easily damaged because of high peak intensity. Therefore, the position of the wavelength conversion crystal is periodically moved to ensure the device lifetime.

As shown in this figure, the maximum average output power at a pulse frequency of 40KHz is about 11W, and decreases with increasing frequency, about 5.5W at 100 KHz. As the frequency increases, the change in the rising edge of the pulse is small, but the slope of the falling edge of the pulse increases slowly, increasing the pulse width from 25ns to 35 ns. The pulse energy is reduced to 25% at 100KHz versus 50KHz, and the peak output power is reduced to 19%.

Fig. 11 is a characteristic diagram showing the output characteristics of a typical YAG laser with a low peak and a large pulse width, which is suitable for printed circuit board processing, like the YVO4UV laser, and the frequency of the YVO4UV laser shown in fig. 10 is shown by a broken line for convenience of comparison.

The YAGUV laser has an Internal Cavity structure of its fundamental crystal and THG wavelength conversion crystal, and when Q-SW is turned on, the fundamental wave oscillates and simultaneously performs wavelength conversion. Since an audio optical device having a slow response speed to the Q-SW is used, the fundamental wave continues to pass through the conversion crystal after the Q-SW is turned off before the end of the beam passage. Therefore, the output time, i.e., the pulse width is large but the peak intensity is low, and thus the crystal damage is small.

As shown in this figure, the maximum average output power at a pulse frequency of 40KHz is about 11W, and decreases with increasing frequency, about 5.5W at 100 KHz. The change in the rising edge of the pulse is small as the frequency increases, but the slope of the falling edge of the pulse increases more, increasing the pulse width from 130ns to 180 ns. The pulse energy is reduced to 25% at 100KHz versus 50KHz, and the peak output power is reduced to 15%.

Comparing fig. 10 and fig. 11, both average output powers of the pulse frequency in the range of 40 to 100KHz are almost the same. However, if the YAGUV laser has a peak output power of 20% at a frequency of 50KHz and about 4% at a frequency of 100KHz, based on a frequency of 50KHz (100%) of the YVO4UV laser.

As can be seen from FIGS. 10 and 11, the pulse width of the conventional laser is determined by the frequency in the pulse frequency range of 40-100 KHz, the YVO4UV laser is limited to 20-35 ns, and the YAGUV laser is limited to 130-180 ns. Therefore, as shown by oblique lines in fig. 11, the insulating layer cannot be processed with a pulse width suitable for the processing.

The following relationships exist for average output power, peak output power, pulse energy, pulse frequency, pulse width, material removal, energy density, and output power density:

pulse energy (J) mean output power (W)/pulse frequency (Hz)

Peak output power (W) pulse energy (J)/pulse width(s)

Removal amount ≈ pulse energy × number of pulses

Energy density (J/cm2) ═ pulse energy (J)/beam area (cm2)

Output power density (W/cm2) ═ peak output power (W)/beam area (cm2)

That is, pulse energy is inversely proportional to pulse frequency and peak output power is inversely proportional to pulse width. In general, since the output power density (energy density per unit time) is sufficiently higher than the decomposition threshold of a material except for a metal material such as copper having high thermal conductivity, the removal amount is substantially proportional to the total energy (pulse energy × number of pulses). Therefore, the efficiency and the processing speed can be increased when processing is performed near the maximum output.

On the other hand, for FR-4 material, which is a typical printed circuit board material, a copper conductor layer and an insulating layer (a layer in which resin is impregnated into glass fiber) as conductor layers have a difference of about 10 times between an output power density and an energy density threshold value (copper: glass fiber: resin ≈ 10: 3: 1) depending on physical parameters of the material, and appropriate conditions are required for each material. The following is a detailed description.

(1) When the copper conductor layer on the surface is processed, if the energy density and the output power density are too high, the insulating layer immediately below the conductor layer is processed at a high output power at the moment of removing the copper conductor layer, and therefore, the bottom cut of the insulating layer becomes large after removing the resin. On the other hand, if the energy density and the output density are too low, the supplied heat spreads around the processing portion, and the removal amount per pulse decreases. Therefore, the number of pulses increases and the processing speed decreases. Therefore, the energy density and output power density suitable for processing the copper conductor layer are about 5-10J/cm 2 and about 150-300 MW/cm2 under the condition of high peak value and narrow pulse width, and about 10-20J/cm 2 and about 100-200 MW/cm2 under the condition of low peak value and large pulse width.

(2) When the energy density and the output power density are too high in processing the glass fiber, the resin surface of the glass fiber is largely protruded after the resin around the glass fiber is removed by the surface reflection light, and the remaining amount of the insulating layer is reduced as the processing proceeds, thereby damaging the copper conductor layer at the bottom of the hole. On the other hand, if the energy density and the output power density are too low, the removal amount per pulse decreases, and therefore the number of pulses increases and the processing speed decreases.

Therefore, the energy density and output power density suitable for glass fiber processing are about 2-6J/cm 2 and about 100-200 MW/cm2 at high peak, narrow pulse width. About 3-8J/cm 2 and about 60-120 MW/cm2 at low peak, large pulse width.

(3) When the resin is processed, if the energy density and the output power density are too high, the copper conductor layer at the bottom of the hole is damaged. On the other hand, if the energy density and the output power density are too low, resin residue at the bottom of the hole increases, and the processing speed decreases due to the increase in the number of pulses.

Therefore, the energy density and output power density suitable for resin processing are about 0.5-1.5J/cm 2 and about 15-30 MW/cm2 at high peak, narrow pulse width. About 0.7-1.5J/cm 2 and about 10-20 MW/cm2 at low peak, large pulse width.

As described above, when the hole quality, the hole shape, and the machining speed are used as references, the threshold values of the energy density and the output density suitable for each material of the workpiece have the upper limit value and the lower limit value, and the hole quality and the hole shape are deteriorated when machining is performed near the maximum output. For example, when a pore size of 50 μm is processed on a copper conductor layer, a glass fiber and a resin using the UV laser shown in fig. 10 and 11, practical pulse frequencies are limited to ranges of about 40KHz or less, about 60KHz or less and about 100KHz or less, respectively, from the viewpoints of suitable energy density and output power density.

Therefore, in the prior art, as the processing is performed (i.e., each time the material to be processed is changed), the processing is performed by changing the pulse frequency in the frequency range of 40KHz to 100KHz with reference to the energy (or peak output power) necessary for the processing of the copper conductor layer, the processing of the glass fiber, and the processing of the resin, or by determining the frequency with reference to the energy (or peak output power) necessary for the processing of the copper conductor layer, and adjusting the energy (or peak output power) by changing the LD output power (i.e., the output power of the laser oscillator) without changing the pulse frequency with respect to the glass fiber and the resin.

However, in order to stabilize the UV laser output, it is necessary to control the temperature of the wavelength conversion crystal to be within 0.1 ℃. That is, if the thermal equilibrium state of the wavelength conversion crystal (LBO, CLBO, BBO) for SHG, THG is broken when the pulse frequency is changed or when the peak output power is controlled by the LD output power, the beam exit angle changes due to the refractive index change accompanying the crystal temperature change, and as a result, the hole position accuracy and the hole shape deteriorate.

However, even if the wavelength conversion crystal temperature is controlled within 0.1 ℃, due to time delay or the like, for example, if the beam exit angle is changed by 40 μ rad at 60KHz and 60 μ rad at 80KHz with reference to a frequency of 40KHz, the hole position accuracy is deteriorated by about 5 μm at maximum. Moreover, output power varies, beam mode (spatial energy distribution) is destroyed, and hole quality and hole shape deteriorate.

Further, since conditions suitable for each material cannot be secured, desired hole position accuracy, hole quality, and hole shape cannot be obtained.

For example, patent document 1 discloses the following processing method: the pulse frequency was set constant, the peak output power was switched outside the laser oscillator, the copper conductor layer was processed at a high energy density and a high output power density, then the medium energy density and the medium output power density were switched to process a glass fiber, and finally the low energy density and the low output power density were switched to process a resin.

[ patent document 1] Japanese patent application laid-open No. 2002-335063

Disclosure of Invention

The technique described in patent document 1 can set the pulse frequency and the peak output power for the copper conductor layer, the glass fiber, and the resin, but when the peak output power is reduced and the pulse width is increased, the resin residue at the hole bottom is increased. Therefore, in order to reduce the resin residue, it is necessary to reduce the pulse width in a state where the peak output power is increased.

Accordingly, an object of the present invention is to provide a laser processing method and a laser processing apparatus which are excellent in hole position accuracy and hole quality, in view of the above-described problems of the prior art.

In order to solve the above problem, a first aspect of the present invention is a laser processing method for shaping a waveform of laser light output after a narrow pulse processing and supplying the shaped laser light to a processing unit.

A second aspect of the present invention is a laser processing apparatus including a laser oscillator that outputs a pulsed laser beam, and processing an object to be processed with the pulsed laser beam, the laser processing apparatus including a pulse shaper that is disposed on an optical path of the laser beam and processes the object with the first aspect of the present invention.

The present invention shapes the waveform of laser light output after narrow pulse processing and then supplies the shaped laser light to a processing unit, so that laser processing with excellent hole position accuracy and hole quality can be performed.

Drawings

Fig. 1 is a schematic diagram showing a laser processing apparatus according to embodiment 1 of the present invention.

Fig. 2 is a timing chart showing the operation of the present invention.

Fig. 3 shows a diagram of the relationship between the exit beam 6 and the #1 branch beam 7 according to the invention.

Fig. 4 shows a relationship diagram of the outgoing beam 6, the #1 branched beam 7 and the #0 branched beam 8 according to the present invention.

Fig. 5 is a graph showing a relationship between a workpiece and a peak output power according to the present invention.

Fig. 6 is a schematic diagram showing a laser processing apparatus according to embodiment 2 of the present invention.

FIG. 7 is a timing chart showing the operation of embodiment 2 of the present invention.

FIG. 8 is a schematic view of a process according to the present invention.

FIG. 9 is a view illustrating a process of the present invention.

Fig. 10 is a graph showing the output characteristics of UV laser light excited by YVO 4.

Fig. 11 is a graph showing the output characteristics of YAG-excited UV laser.

Detailed Description

Embodiments of the present invention are described below with reference to the drawings.

[ example 1]

Fig. 1 is a schematic diagram showing a laser processing apparatus according to embodiment 1 of the present invention.

The laser oscillator 1 that outputs YAGUV laser light is connected to the system controller 3 via the laser power controller 4. The pulse shaper 2 disposed on the optical axis of the outgoing beam 6 output from the laser oscillator 1 is connected to the system controller 3 via a pulse shaper controller 5. The #1 branched light beam 7 (machining light beam) is guided to a machining unit (not shown), and the #0 branched light beam 8 is guided to a damper (not shown) and converted into heat. The laser power controller 4 outputs a Q-SW on/off signal 11 arranged inside the laser oscillator 1 in accordance with a pulse output command signal 9 specifying the pulse mode (pulse width, i.e., frequency, and peak output power) of the outgoing beam 6 output from the system controller 3.

The pulse shaper controller 5 outputs a switching signal 12 for controlling the pulse shaper 2(AOM or eom, here AOM.) in accordance with an output command signal 10 of a prescribed pulse pattern (peak output power and pulse width) output from the system controller 3, and forms a #1 branched beam 7 for processing from the outgoing beam 6. The pulse shaper 2 can control the penetration rate and can control the peak output power of emergent light relative to incident light within the range of 100-20%.

The operation of this embodiment is explained below.

Fig. 2 is a timing chart showing operations, (a) shows a pulse output command signal 9 and an output command signal 10, (b) shows a current value supplied to the Laser Diode (LD) inside the laser oscillator 1, (c) shows the outgoing beam 6, (d) shows the #1 branched beam 7, (e) shows the on/off signal 11 of Q-SW, and (f) shows the on/off signal 12 of the pulse shaper 2.

The current is constantly supplied to the LD, Q-SW is turned on according to the pulse output instruction signal 9, and the outgoing beam 6 is output. Then, the pulse shaper 2 is turned on in accordance with the output command signal 10, and the #1 branched beam 7 is supplied to the machining unit.

In the figure, the on time T1 of the Q-SW is the laser output time, the off time T2 of the Q-SW is the laser gain storage time, and T1+ T2 is the pulse period T. The longer the off-time T2 (i.e., the longer the pulse period T), the larger the laser gain storage amount, and the higher the peak output power WP of the outgoing beam 6. Therefore, by changing the time (TPA, TPB, TPC in the figure) for turning on the pulse shaper 2 within the on time T1 of Q-SW, the pulse width of the #1 branch beam 7 can be adjusted.

TWA, TWB, TWC are shown as the actual pulse width (half-value width on a general scale) of the #1 branch beam 7 of the on-times TPA, TPB, TPC of the pulse shaper 2, and the peak output powers WPA, WPB, WPC, respectively.

The pulse shape of the #1 branch beam 7 is explained below.

Fig. 3 shows a diagram of the relationship between the outgoing beam 6 and the #1 branch beam 7. The penetration of the pulse shaper 2 is 100%. In the figure, a waveform a (corresponding to a waveform 14 in fig. 2) is various waveforms of #1 beamlet 7 when the on-time TPA of the pulse shaper 2 is set to have a pulse width TWA of 125ns, a waveform B (corresponding to a waveform 15 in fig. 2) is set to have an on-time TPB of the pulse shaper controller 5 to have a pulse width TWB of 80ns, and a waveform C (corresponding to a waveform 16 in fig. 2) is set to have an on-time TPC of the pulse shaper 2 to have a pulse width TWC of 60 ns.

For example, comparing waveform A with waveform C, as shown in the figure, the pulse rising edge is unchanged, but the energy at the pulse tail (the second half of the pulse) is removed, and the energy of waveform C is about 1/4 of waveform A. Therefore, by changing the on-time and the transmittance of the pulse shaper 2, the energy of the #1 branch beam 7 can be adjusted with high accuracy within a range of 100 to 50% of the energy of the outgoing beam 6.

Fig. 4 shows a relationship diagram of the outgoing beam 6, the #1 branched beam 7, and the #0 branched beam 8. The energy of the #0 branch beam 8 is the difference between the energies of the outgoing beam 6 and the #1 branch beam 7, and therefore is the waveforms 17, 18, and 19 shown in fig. (d) of the figure. The diagram (a) shows the on/off signal 12 of the pulse shaper 2 ((f) in fig. 2), (b) shows the outgoing beam 6 ((c) in fig. 2), and (c) shows the #1 branched beam 7 ((d) in fig. 2).

Specific processing examples are described below.

Fig. 5 is a graph showing the relationship between the workpiece and the peak output, (a) shows the #1 branched beam 7, (b) shows the beam pattern and the threshold value specific to the material, and (c) shows the machining result. For glass-doped materials, an insulating layer 25 of glass fibers 26 integrally formed with a resin 27 is disposed between an outer copper conductor layer 24 and an inner or inner copper conductor layer 28. The ratio of the energy threshold THCU of copper, the energy threshold THG of glass, and the energy threshold THR of resin is about 10: 3: 1, but a hole having an excellent shape accuracy and no resin 27 remaining at the bottom of the hole can be processed by processing the copper conductor layer 24 by irradiating the #1 branched beam 7 having a pulse width of 100ns or more (for example, the waveform a shown in fig. 3) as the beam pattern 29 a predetermined number of times, processing the glass fiber 26 by irradiating the #1 branched beam 7 having a pulse width of less than about 100ns (for example, the waveform B shown in fig. 3) as the beam pattern 30 a predetermined number of times, and processing the resin 27 at the bottom of the hole by irradiating the #1 branched beam 7 having a pulse width of less than about 100ns (for example, the waveform C shown in fig. 3) as the beam pattern 31 a predetermined number of times.

The following more specifically describes the present invention.

1) About quality of processing

For example, at a pulse frequency of 40KHz, a pulse width of over 100ns, and an energy density of about 12J/cm²The output power density is about 160MW/cm²Under the conditions of (1) processing a copper conductor layer, and then performing pulse width less than 100ns and energy density of about 4J/cm²The output power density is about 70MW/cm²Under the conditions of (1), and then performing glass processing under the conditions of pulse width of less than 80ns and energy density of about 1J/cm²The output power density is about 12MW/cm²By performing the resin processing under the conditions of (1), it is possible to perform the processing with excellent hole quality.

When processing a resin, the amount of removal can be made the same regardless of the pulse width and the peak output power, with a constant processing energy. However, since the damage of the hole bottom and the resin residue at the hole bottom have a trade-off relationship, when the peak output power is low and the pulse width is large, the resin residue at the hole bottom is large. In contrast, when the pulse width is reduced in a state where the frequency is reduced and the peak output power is secured, resin residue can be reduced, and the quality of the hole can be improved.

When a blind via (bottomed via) is processed by a UV laser, the via bottom is damaged when the energy distribution is uneven because the energy absorption rate of the material is high. Therefore, the beam shape of the #1 branched beam 7 is preferably made a so-called flat-top beam by a beam shaping means not shown.

2) With respect to machining position accuracy

For example, when processing an FR-4 material having a copper conductor layer of 12 μm and an insulating layer of 80 μm in thickness, the number of pulses necessary for processing is 10 to 15 pulses for the copper conductor layer, 50 to 70 pulses for the glass portion of the insulating layer, and 5 to 10 pulses for the bottom of the hole. Therefore, in the conventional method of adjusting the pulse frequency and the LD current, the change in the beam exit angle due to the thermal change of the crystal is about 60 μ rad, and the hole position accuracy error of the optical system is about 5 μm.

In contrast, the present invention minimizes thermal changes in the crystal (i.e., the pulse frequency and pulse output power are constant, and the actual pulse width and actual peak output power are adjusted outside the oscillator), thereby reducing the change in the exit angle of the beam to a maximum of 20 μ rad or less, and reducing the error in the aperture position accuracy of the optical system to 2 μm or less.

In the present invention, both the peak output and the pulse width of the same beam can be adjusted to expand the output adjustment range by 20 times, and an FR-4 material (composed of copper and a glass fiber-doped resin) having an energy threshold of about 10 times difference can be processed with high quality.

But there may be a case where the on-off signal 11 of Q-SW is delayed by the time TD with respect to the pulse output command signal 9.

[ example 2]

Fig. 6 is a schematic diagram showing a laser processing apparatus according to embodiment 2 of the present invention, and the same components or components having the same functions as those in fig. 1 of embodiment 1 are given the same reference numerals, and redundant description thereof is omitted.

The optical sensor 22 is connected to the pulse shaper controller 5, and outputs a detection signal 10a to the pulse shaper controller 5 after detecting the reflected light 21 reflected by the surface of the pulse shaper 2. The operation speed of the optical sensor 22 is about several ns.

In this embodiment, the system controller 3 indicates the pulse mode signal 23, i.e. the on-time T2 of the pulse shaper 2, to the pulse shaper controller 5.

Fig. 7 is a timing chart showing the operation of this embodiment, wherein (a) shows a pulse output command signal 9 and an output command signal 10, (b) shows a current value supplied to the LD inside the laser oscillator 1, (c) shows the outgoing beam 6, (d) shows a #1 branched beam 7, (e) shows an on/off signal 11 of Q-SW, (f) shows a detection signal 10a of the optical sensor 22, and (g) shows an on/off signal 12 of the pulse shaper 2.

As shown in the figure, in this embodiment, even if the on/off signal 11 of Q-SW is delayed by the time TD with respect to the pulse output command signal 9, the output command signal 12 can be substantially synchronized with the on/off signal 11 of Q-SW because the pulse shaper 2 is operated after the output of the outgoing beam 6 is confirmed. As a result, the waveform of the #1 branch beam 7 can be made highly accurate and uniform.

Instead of providing the optical sensor 22 outside the oscillator 1, the optical sensor 22 may be provided inside the oscillator 1 to capture reflected light, branched light, and scattered light.

In the above-described embodiments 1 and 2, the pulse width and the peak output power of the #1 branched beam 7 are changed in steps according to the material of the workpiece, but the pulse width and the peak output power of the #1 branched beam 7 may be continuously changed as the processing proceeds, or may be adjusted in steps and continuously. Such control, continuously or stepwise and continuously, for adjusting the pulse width and peak output power of the #1 branched beam 7, can improve the quality of, for example, the interface between copper and an insulating layer or the interface between an insulating layer and a conductor layer at the bottom of a hole.

When a copper conductor layer is processed with a large pulse width (i.e., the #1 branched beam 7 having a large pulse width), the formation of scattered matter at the entrance of the hole, the increase in the shape of the entrance of the hole, and the like are affected, but the transmittance and the on-time T2 of the pulse shaper 2 can be improved by selecting them depending on the processing material.

However, when an audio optical device (AOM or EOM) is used for the pulse shaper 2, the pulse width is affected by the in-crystal beam diameter passage time τ (τ ═ D/V, where D is the beam diameter and V is the crystal passage speed of the ultrasonic wave). That is, the larger the incident beam diameter D entering the pulse shaper 2, the longer the transit time (response speed) τ, and the smoother the output rise.

The crystal is SiO₂Since V is 5.96Km/s, the response speed τ is 50ns when D is 0.3mm, for example. Further, the delay time of the ultrasonic wave reaching the beam was 50ns when the distance from the transducer (oscillation source of the ultrasonic wave) to the beam was 0.3 mm. Therefore, in practical applications, the range for controlling the pulse width is 50ns or more.

On the other hand, when the frequency of the ultrasonic wave is 40MHz, the electric dissociation energy is 25ns (full-wave rectification is 12.5 ns). Therefore, in order to minimize the decomposition energy, the electrical delay can be shortened to several ns by synchronizing the on signal of the AOM (pulse shaper 2) with respect to the detection signal of the optical sensor 22, or setting a plurality of clocks and selecting the closest clock to turn on the AOM.

In addition, when the pulse shaper 2 is formed by combining the photoelectric device of the polarization modulation mode, the lambda/4 plate and the Brewster plate, the response speed can be improved to several ns, and the pulse width can be controlled to be 20-100 ns. Moreover, since the pulse rising edge can be controlled, the shape of the #1 branch beam 7 can be more finely controlled.

The other parts not specifically described are the same in structure and function as those in embodiment 1 described above.

The present invention is not limited to the glass-doped material, and is also effective for bottomed hole processing using a build-up layer to which a resin copper conductor layer RCC material (composed of a conductor layer of a copper conductor layer and a resin insulation layer) is attached, and bottomed hole processing of a build-up layer only of a resin insulation layer, as shown in fig. 8.

In embodiment 1, when the delay time TD of the on/off signal 11 of Q-SW with respect to the pulse output command signal 9 is known, if a timing element is provided and the pulse shaper 2 operates after the delay time TD after the pulse output command signal 9 is output, the optical sensor 22 described in embodiment 2 does not have to be provided.

As shown in fig. 9, the pulse shaper 2 of the present invention is suitable for the through hole (through hole) processing of a multilayer circuit board (composed of an outer copper conductor layer, an inner copper conductor layer, a glass-doped composite layer, and a glass-doped core layer 32) by applying the most appropriate beam pattern depending on the material.

The narrow pulse processing of the present invention can be applied to long pulse laser beams of Internal Cavity structure and external Cavity structure, for example, UV laser beams, visible laser beams, and near-infrared laser beams, which have fundamental waves of YAG laser beams and YLF laser beams (YVO4 laser beams).

The embodiment as described above can produce the following effects.

That is, since the pulse frequency and the pulse output power of the laser oscillator are constant and the actual pulse width and the actual peak output power are adjusted outside the laser oscillator, the thermal change inside the laser oscillator can be minimized and the positional accuracy of the processing can be improved.

Further, since the pulse energy can be adjusted with high accuracy within the range of 100% to 5%, even an FR-4 material (made of copper and a glass fiber-doped resin) having an energy threshold value of about 10 times difference can be processed with high quality.

Furthermore, since the processing can be performed with a pulse width (at 40 KHz) of, for example, 25 to 110ns, which cannot be employed in the prior art, it is easy to control the hole quality and the hole shape.

Claims (8)

1. A laser processing method for processing a workpiece by irradiating the workpiece with a laser beam, characterized in that a laser beam waveform outputted after a narrow pulse processing is shaped and then the shaped laser beam is supplied to a processing unit.

2. The laser processing method according to claim 1, wherein the shaping of the laser light is started in synchronization with a start of the output of the laser light output after the narrow pulse processing.

3. The laser processing method according to claim 1 or 2, wherein the object is processed by a laser beam shaped to have a pulse width of 100ns or more when at least one of a metal material layer, an organic material layer, and an inorganic material layer constituting the object is stacked in a thickness direction, and at least one of the organic material layer and the inorganic material layer is processed by a laser beam shaped to have a pulse width of less than 100 ns.

4. A laser processing method according to any one of claims 1 to 3, wherein the pulse width and the peak output power of the shaped laser light are changed as the processing proceeds.

5. A laser processing apparatus having a laser oscillator for outputting a pulse laser beam and processing an object to be processed by the pulse laser beam, wherein a pulse shaper is provided on an optical path of the laser beam, a waveform of the laser beam outputted after a narrow pulse processing is shaped by the pulse shaper, and the shaped laser beam is supplied to a processing unit.

6. The laser processing apparatus according to claim 5, wherein the pulse shaper starts shaping the laser light in synchronization with a start of output of the laser light output after the narrow pulse processing.

7. The laser processing apparatus according to claim 5 or 6, wherein the pulse shaper changes a pulse width when processing at least one of the metal material layer, the organic material layer, and the inorganic material layer, which constitute the object to be processed, in a state where the at least one of the metal material layer, the organic material layer, and the inorganic material layer is stacked in a thickness direction.

8. The laser processing apparatus according to claim 7, wherein the pulse width is changed to 100ns or more when the metallic material layer is processed and changed to less than 100ns when at least one of the organic material layer and the inorganic material layer is processed.