Method for controlling configuration of laser induced breakdown and ablation

The present invention relates mainly to a method of changing both internal and external aspects such as ablation (cutting) of a material or modification of physical properties in a material structure by using a laser beam. The present invention can be applied to various materials.

[0002] Laser-induced fracture of materials results in chemical changes and physical, chemical and physical fractures, decomposition, cutting and evaporation. Lasers are an excellent means of controlling procedures that require precision, such as engraving micro-patterns. Pulsed rays are more effective than continuous rays for many procedures, including medical treatment. Pulsed laser beams have extremely short bursts or pulses of light, for example, with a duration of about 10 nanoseconds. There is usually a pause between these pulses. The peak value of each pulse is often relatively large around gigawatts.<u style="single">10</u>^{<u style="single">13</u>}<u style="single">w / cm</u>^{<u style="single">2</u>}It can be as strong as it is. The focal point of the laser beam extends over an area of a certain radius, but the influence of the beam extends not only to the focal area or spot but also to the peripheral area adjacent to the spot, which adversely affects them. The surrounding area affected by this may even be several times wider than the spot itself. This causes problems such as affecting tissues in medical care. In the field of laser machining, current lasers, which utilize nanosecond pulse, cannot achieve high accuracy and control, especially when using non-absorbable wavelengths.

[0003] An object of the present invention is to provide a method for localizing laser-induced breakdown in order to solve such a problem. Another object is to provide a method of inducing a certain pattern of fracture inside or outside the material.

[0004] [Means for Solving the Problems] In one aspect of the present invention, a method for destroying a material by laser induction is provided. In this case, the material is characterized by a sudden, rapid and distinct change in its gradient at a constant pulse width value in relation to the fluence fracture threshold (Fth) and the laser beam pulse width (T). This method includes generating a laser pulse beam having a width equal to or less than a constant laser pulse width value for each pulse. The light beam focuses on a point on or below the surface of the material for which laser-induced breakdown is desired.

[0005] Further, in one embodiment, the present invention can be understood by defining a constant laser pulse width value as follows. That is, the relationship between the fluence destruction threshold and the laser pulse width causes the fluence destruction threshold (Fth) to be the square root of the pulse width (<u style="single">T</u>^{<u style="single">1/2</u>}) Is determined for a curve having a first portion over a range of relatively long (high) pulse widths that vary in proportion to). The curve has a second part over a shorter (lower) pulse width than the first part. The proportional relationship between the fluence failure threshold and the pulse width differs between the first and second parts of the curve, and the constant pulse width value follows the curve and with the first and second parts. Come to one point between. In other words, it goes without saying that the constant pulse width value is such that the Fth vs. τp relationship does not apply, and this relationship does not apply to pulse width values shorter than the constant pulse width.

[0006] Scaling the fluence failure threshold (Fth) as a function of pulse width (T) is the square root of pulse width (T).<u style="single">T</u>^{<u style="single">1/2</u>}) Is expressed as Fth and is demonstrated by the pulse width over the nanosecond range. In the present invention, the fracture threshold (Fth) is the square root of the pulse width (<u style="single">T</u>^{<u style="single">1/2</u>}) Shall provide a plurality of effective methods with pulse widths up to the picosecond and femtosecond ranges found not to fluctuate.

[0007] The pulse width duration in the range from nanosecond to femtosecond is achieved by generating short optical pulses with a constant duration from the optical oscillator. Next, the time of the short optical pulse is extended by about 500 to 10,000 times to generate a time-extended optical pulse to be amplified. The time-extended optical pulse is then amplified with a solid-state amplification medium. This includes combining time-extended optical pulses with optical pulses generated by a second laser used to optical pump solid-state amplification media. After this, the amplified pulse is compressed and returns to the original pulse duration.

[0008] In one embodiment, the laser oscillator produces very short pulses of about 10-100 femtoseconds at relatively low energy levels of about 0.001 to 10 nanojoules. Next, it is stretched to an energy of about 0.001 to 10 nanojoules with a pulse width of about 100 picoseconds to 1 nanosecond. Then, it is usually amplified to about 0.001 to 1,000 millijoules, 100 picoseconds to 1 nanosecond, and then compressed again. Eventually, it will be 10-200 femtoseconds and 0.001-1,000 millijoules. Although the pulse generation system is not uniform, it is desirable that the laser medium be sapphire containing titanium impurities that control the function of the laser.

[0009] In another aspect, a method according to the invention provides a laser beam that determines a spot with a lateral Gaussian profile. A feature of the profile is that laser-induced fracture results in cleavage of the spot's internal region because the fluence at or near the center of the beam spot is greater than the threshold fluence. Maximum intensity is obtained at the center of none other than the beam waist. The beam waist is the point of the beam that makes the wave surface perfectly flat. That is, the radius of curvature at this point is infinite. At this center point, the radius R = 0 on the XY axes and Z = 0 on the Z axis. This allows the material to be damaged with a very small volume of Z = 0 and R = 0. Therefore, the feature can be smaller than the spot size on the XY focal plane and smaller than the Rayleigh range (depth of focus) on the Z axis. The pulse width duration may be higher if it is less than the pulse width determined by the sudden or discernible change in gradient in relation to the fluence destruction threshold and the pulse width of the laser beam, but is in the femtosecond range. Is desirable.

[0010] In another embodiment, the optical path is provided with a diaphragm, a disk or a mask to block at least a portion of the beam so that the beam has the desired geometry. In yet another embodiment, the desired beam configuration can be achieved by changing the beam spot size or by forming a Fourier transform (FT) pulse to achieve a geometric shape with a special frequency distribution. ..

[0011] The energy of the beam should be in the range of 10 nJ (nanojoules) to 1 millijoule, and the fluence of the beam should be in the range of 0.1 J / cm2 to 100 J / cm2 (joules per square centimeter). The wavelength is preferably 200 nm (nanometer) to 1 μm (micron).

[0012] In addition, the present invention utilizes new methods and duration ratings to determine the optimum pulse width duration for a particular material to create precisely shaped cut surfaces or voids inside or on the surface of the material. Has the advantage of being obtained. Such ratings can be reproduced for a given material using the method according to the invention. Further, there is an advantage that extremely high intensity can be obtained from this method with a modest amount of energy, and the spot size may be very small. Damage to adjacent areas is minimal and can provide important convenience for human and animal tissues.

[0013] Other features and conveniences according to the present invention will become apparent from the following description of preferred embodiments, claims and accompanying drawings.

[Embodiments of the Invention] Figure 1 shows a laser-induced breakdown threshold determined as a function of laser pulse width in the nanosecond to femtosecond range using a chirped pulse amplification (CPA) laser system. It is a performance test device. The basic configuration of such a CPA system is disclosed in United States Patent No. 5,235,606 granted to the assignee of the present invention. The inventor of the patent is the same as the inventor of the present application. In addition, the United States Patent No. 5,235,606 is a part of this application by quoting the whole.

The Chirped Pulse Amplification System is a book entitled "Laser Focus World" published by Pennwell in June 1992 by co-inventors Jeffrey Squireer and Gerard Morrow of the present application. It is explained in. According to the book, CPA systems are broadly divided into four categories. The first category belongs to high-energy systems with few repetitions, such as ND glass lasers with an output of several joules, but the number of shots is less than one shot per minute. The second category has an output of about 1 joule and a repetition rate of 1 to 20 hertz. The third category is a group of millijoule levels that operate at 1-10 kHz repeats. Lasers in the fourth category operate at 250-350 kHz and emit 1-2 microjoules per pulse. In US Pat. No. 5,235,606, multiple solid amplification materials were identified and US Pat. No. 5,235,606 was identified. The invention of No. is described with alexandrite. In the examples below, titanium / sapphire is used for explanation, and although there are some variations, the basic process of US Pat. No. 5,235,606 is explained.

[0016] The following examples show the case where the pulse duration is in the range of several hundred picoseconds or less, the pulse energy is less than microjoules at a frequency of around 1 kilohertz, and most of them are in the nanojoule range. However, these examples are for illustration purposes only and the present invention is not limited thereto.

[0017] In the basic method of CPA, a short pulse is first generated. The oscillator pulse is ideal and short enough that there is no need to compress the pulse again. The post-occurrence pulse is stretched by a grating pair that allows for positive group velocity variations. How long the pulse is stretched depends on the amount of amplification. If it is less than 1 millijoule, a few tens of picoseconds are sufficient. Usually, the first amplification step is performed by a reproduction amplifier or a multipath amplifier. In one configuration, it consists of a gain medium, a Pockels cell and an optical resonator with a thin film wave changer. After the regenerative amplification step, the pulse can be recompressed or continuously amplified. The compressor consists of a grating or a grating pair, and has a negative group velocity variation. The grating should be used in a compressor to accommodate the elongation stage grating. Details of the representative system are disclosed in US Pat. No. 5,235,606, which was previously part of this specification.

[0018] One of the important aspects of the present invention is found in the characteristic curve of the fluence fracture threshold Fth as a function of the laser pulse width peculiar to the material. In such a property curve, identify points where the gradient properties of the material change rapidly, clearly and rapidly, or change identifiablely. In general, it is more desirable to operate beyond this point because it allows more precise control of the laser-induced breakdown (LIB) or laser-induced cutting threshold. (Example 1-Opaque material) FIG. 1 shows an experimental device for determining the threshold fluence by determining the incident fluence for the scattered energy or the pulse width for the threshold fluence. The system includes the above-mentioned means for generating a pulsed laser and means for collecting light emitted from a target to a photomultiplier tube, usually a lens. The change in the light beam that has passed through the transparent sample is measured by an energy meter.

[0019] FIG. 2 shows a graph of data obtained from gold, which is an absorption medium using 150 fs pulses, and FIG. 3 shows the relationship between the pulse width and the threshold fluence. The arrows in FIG. 3 indicate that the relationship between the threshold fluence and the pulse width changes significantly.

[0020] Under the experimental conditions of a wavelength of 800 nm and a pulse of 200 fs for gold (Fig. 3), the absorption depth is 275 Å and the diffusion distance is 50 Å. When the pulse is in nanosecond units, the diffusion distance is 10 μm (micron) in diameter, much longer than the absorption depth, and as a result thermal diffusion is a limiting factor in feature size resolution. Empirical evidence that these two ratings exist is shown in Figure 3. Figure 3 is a graph of the experimental and theoretical cut thresholds as a function of pulse width. The arrow at a pulse width of about 7 picoseconds (indicated here as T and τp) is the point where the thermal diffusivity (1th) is equal to the absorption depth (1 / a) (or the boundary region closest to that point). Obviously, if the spot size is small, the pulse must be short (small). For spot sizes of about 1000 Å or less, set the pulse width to 100. It will be necessary to keep it below the femtosecond level. As is clear from the figure, this is the point where the cut threshold gradually fluctuates as a function of the pulse width, or changes from an almost constant value to a value that depends on the pulse time. This result is noteworthy. It has been demonstrated that the electron thermalization time of the energy stored inside gold by the laser beam is around 500 fs or less, and the electron-lattice interaction time is 1 ps. As a result, energy is included in the beam spot in the case of ultrafast laser pulses. In fact, when the energy is at or near the cutting threshold, the cross section of the space of the laser beam will determine the size and shape of the area to be cut (FIGS. 4 and 5).

[0021] Further experiments were continued to measure the amount of recombination light produced as a function of fluence colliding with the gold thin film. In this case, the technique based on the experimental device described above was used. The basic idea is that the intensity of light is proportional to the amount of material to be cut. In Figure 4, the removed material is graphed as a function of fluence. There is a clear threshold fluence at the beginning of material removal. By taking only a small portion of the Gaussian beam whose fluence is greater than the threshold, the area to be cut can be limited to this small area. In Figure 4, Ra is the radial position on the beam where the fluence is at the threshold. At this time, cutting occurs only inside the radius Ra. With the correct selection of incident fluence, it is clear that in principle the spot or hole to be cut is smaller than the spot size Rs. Figure 5 illustrates this concept graphically. Figure 4 shows the data for a 150 fs pulse, but the movement of this threshold is shown over a wide range of pulse widths. However, in the case of a pulse longer than this, it is impossible to cut the sub-spot size because heat diffusion is dominant, as will be described later.

Further experiments with opaque materials were continued, with 800 nm titanium / sapphire oscillator pulses extended with grating pairs, amplified with a 1 kHz regenerative amplifier, and recompressed with yet another grating pair. As a result, a pulse width of 7ns ~ 100fs was obtained. This beam was focused with an objective lens with a magnification of 10. That is, the theoretical spot size was 3.0 μm in diameter. Two holes with a diameter of about 0.3 μm were photographed by scanning electron micrographs of the holes to be cut of a silver thin film applied to glass using a pulse with a pulse width of 200 fs and a pulse energy of 30 nJ (fluence is 0.4 J / cm2). I found it. Similar results were obtained with aluminum.

These results suggest that even smaller holes can be machined by reducing the spot size, which is a function of aperture and wavelength. We have demonstrated that the fourth harmonic (200 nm) can be generated by using a non-linear crystal. Thus, by using a stronger objective lens with a 200 nm ray, it is possible to form a hole with a diameter of 200 angstroms in principle.

[0024] These examples show that using femtosecond unit pulses, the spatial resolution of the cutting / machining process is significantly lower than the wavelength of laser radiation used to generate it. can do. The area or diameter of the hole to be cut is smaller than the area or diameter of the spot size. In special cases such as diffraction-limited spot size, the size (diameter) of the hole to be cut is smaller than the basic wavelength. We formed a laser cutting hole with a diameter smaller than the spot diameter and a little less than 10% smaller than the spot size of the laser beam. When an ultrafast pulse is applied to metal, the thermal diffusion distance 1th =<u style="single">(Dt)</u>^{<u style="single">1/2</u>}(Where D is the thermal diffusivity and t is the pulse time) is much smaller than the absorption depth (1 / a). Here, a is the absorption coefficient in the case of radiation.

Those skilled in the art will appreciate that by utilizing the basic methods of the present invention, another embodiment can be realized according to the desired configuration of induced fracture. Examples include using a mask in the optical path, changing the spot size, moving the lens to adjust the focal position, adjusting the cavity design of the laser, Fourier transform (FT) shaping, lasers other than TEMoo. There are adjustments such as the use of operation modes, Rayleigh range depth of focus, and beam waist. However, it is not limited to these.

[0026] Fig. 6 (A) and 6 (B) show how to use the mask. The basic method is to place the mask in the optical path or in the target itself. If you want to block part of the beam, the mask must be made of opaque material (Fig. 6 (A)) and suspended in the optical path. Alternatively, an absorbent mask may be placed on the target to form the contour of the target following the shape of the mask (Fig. 6 (B)).

The change in spot size can be achieved by changing the f / # of the laser, i.e. changing the focal length of the range, or adjusting the aperture to match the size of the incident beam to the lens. ..

[0028] Operating in a mode other than the TEMoo mode means that the upper horizontal mode may be used. This will have the following effects on the beam and materials: That is, the beam intensity does not have to be circular and does not have to be Gaussian. The material is cut according to the beam shape.

The Rayleigh range (Z-axis) may be adjusted by varying the diameter of the beam. In this case, the focal plane is in the XY axes. (Example 2: Transparent material) A series of tests were performed on SiO2 (glass) samples, and laser-induced breakdown (LIB) as an action of a laser pulse width of 150 fs to 7 ns was measured using a CPA laser system. The short pulsed laser used was a 10 Hz titanium / sapphire oscillator amplifier system based on the CPA technique. A laser pulse was focused on the inside of the SiO2 sample using an f = 25 cm lens. The Rayleigh length of the focused beam is approximately 2 mm. The size of the focal point was measured in its original position by the objective lens of the microscope. The spot size FWHM (maximum half width) measured was 26 μm in diameter in Gauss mode. A fused silica sample having a thickness of 0.15 mm was prepared from Corning 7940.

[0030] These samples were optically polished on both sides at a scratch / dig ratio of 20:10. Each sample was washed with methanol before being subjected to the experiment. The purpose of using this sample was to avoid complicating the problem by self-focusing the laser pulses individually. The SiO2 sample was mounted on a computer-controlled electric XY moving stage. Laser irradiation to each position of the sample was performed only once.

The fracture threshold Fth was measured using two diagnostic techniques. First, the lens focused the plasma radiation from the focal region into a photomultiplier tube with a suitable filter. Second, the fluctuation of the transmittance penetrating the sample was measured with an energy meter. (See Fig. 1.) Visual inspection confirmed fracture at pulse duration in nanosecond units. Figure 7 is a graph of the relationship between typical plasma radiation and transmitted light signals for incident laser energy at a laser pulse width of τp = 300fs. What should be noted here is that the transmittance gradually changed near Fth. This can be explained by the spatiotemporal behavior of fracture at ultrashort pulses. Fracture remains in a localized state where the plasma generated due to the threshold is reached at the focal center due to the variation in intensity in space and the pulse duration is short. The transmitted light is attenuated due to reflection, scattering and absorption by the plasma. Assuming that the laser intensity has a Gaussian profile in both time and space, and that an avalanche occurs and reaches a threshold at all pulse durations, the input energy U The transmitted laser energy Ut as a function of is given by the following formula. That is, Ut = kU U UthUt = kUth [1 + 1n (U / Ut)<b>h</b>)] U&gt; Uth where k is the linear transmission coefficient. The solid line in Fig. 7 is a graph of the above equation (1) with Uth as a parameter. In contrast, nanosecond laser pulse rupture results in cutting the transmitted beam near the pulse peak value, showing different behavior in space-time.

[0032] Figure 8 shows the fluence destruction threshold Fth as a function of the laser pulse width. Between 10ps and 7ns, the fracture threshold follows scaling at relatively long pulse width ratings (triangular and square). This is also shown for comparison. It can be seen that this data agrees with the research results shown above only in the part of the curve where the pulse width is large. As the pulse width becomes shorter than a few picoseconds, the threshold begins to increase. As mentioned in the section on opaque materials (metals), it is noteworthy that such short pulse widths increase accuracy.

It can be seen that the damage threshold accuracy is significantly improved, which is consistent with the multiphoton avalanche fracture theory. (See Figures 8 and 9.) Features can be smaller than the spot size in the XY focal plane and smaller than the Rayleigh range (depth of focus) in the longitudinal or Z-axis. These elements are essential for features smaller than the spot size or Rayleigh range. (Example 3: Tissue) Using a CPA laser system, a series of experiments were performed to determine the corneal fracture threshold as a function of the laser pulse width in the range of 150 fs to 7 ns. As mentioned earlier, in this CPA laser system, the laser pulse width can be changed without changing parameters other than the laser pulse width (spot size, wavelength, energy, etc.). The laser focused on a spot size (FWHM) with a diameter of 26 μm. Plasma radiation was recorded as the action of pulse energy and the tissue damage threshold was measured. The damage was also evaluated histologically.

[0034] The fracture threshold calculated from the plasma radiation data is the scaling law, Fth, as in the case of metal and glass.<u style="single"> T</u>^{<u style="single">1/2</u>} It shows that it deviates from. As shown in FIG. 9, the scaling rule of the fluence threshold holds up to a pulse width of about 10 ps, but does not apply when the pulse width is shortened to less than a few picoseconds. As shown in FIGS. 10 and 11, the cut or LIB threshold shows dramatic fluctuations at high (long) pulse widths.

The threshold is extremely accurate at short pulse widths. When the pulse was short, the standard deviation from the fracture threshold measurement was significantly reduced. In addition, as a result of the analysis, it was clarified that the pulse of less than 10 ps causes less damage to the adjacent tissue.

[0036] The fracture threshold at ultrashort pulses (&lt;10ps) is smaller and the standard deviation is smaller than at long pulses. Less damage to adjacent tissue is the result of ultrashort laser pulses.

[0037] In summary, sub-wavelength holes can be machined in a metal surface using femtosecond unit laser pulses. Such an effect can be physically understood by the term thermal diffusion distance, which exceeds the pulse fixation time limit and is smaller than the absorption depth of the incident light beam. Such an idea is based on the fact that the hole diameter is determined by the lateral Gaussian distribution of the pulse with respect to the evaporation and cutting thresholds.

[0038] Laser-induced optical breakdown dielectric can be said to consist of the following three stages. That is, free electron generation and proliferation, plasma heating, and material deformation or destruction. Avalanche ionization and polyphoton ionization are two processes that cause destruction. The threshold for laser-induced breakdown in dielectric materials depends on the pulse width of the laser pulse. The empirical scaling law of the fluence fracture threshold as a function of pulse width is given by Fth τp ,. Alternatively, the intensity fracture threshold Ith = Fth / τp. Is given. This scaling law can be applied to pulse widths from nanoseconds to tens of picoseconds, but from previously unknown ratings in the present invention: 7 picoseconds for gold and 10 picoseconds for SiO2. We take advantage of the fact that the decay threshold does not follow the scaling law when a reasonably short laser pulse is used.

[0039] Although not bound by a specific theory, the ionization process of a solid dielectric irradiated with a strong laser pulse can be expressed by the following general formula. That is, dne (t) / dt = η (E) ne (t) + (dne (t) / dt) PI- (dne (t) / dt) loss where ne (t) is a free electron (plasma) Density, η (E) is the avalanche coefficient, and E is the electric field strength. The second term on the right side represents the amount of photoionization, and the third term is the loss due to electron diffusion and recombination. If the pulse width is within the picosecond range, electron loss can be ignored for the duration of the short pulse.

The photoelectric separation or ionization component can be evaluated by the tunnel ratio. When the pulse is as short as E ~ 108V / cm, the tunnel ratio is estimated to be w ~ 4 x109 / sec-1, which is small compared to the Avalanche tunnel ratio calculated as follows. However, photoionization provides the initial electrons required for the avalanche process at short pulse widths. For example, according to the data at 1ps, the rms field threshold is about 5 x 107 V / cm. The field will reach 3.5 x 107 V / cm (rms) at 0.5 ps by the time the pulse peaks and w ~ 100 sec-1. The electron density during the period of t ~ 100fs can reach ne ~ nt [1 --exp (-w t)] ~ 1011cm-3. Here, nt to 1022 are electron densities in the valence band.

Ignoring the last two terms, this is an electron avalanche process, where collision ionization by major electrons is driven by a laser electric field. Then the electron density is given by ne (t) = no x exp (n (E) t), where no is the density of the initial free electrons. These initial electrons can be generated by surface trap or photoionization. Fracture becomes more statistically significant with the help of photoionization at short pulse widths. The fracture condition is given by ητp ~ 18, subject to the condition that fracture occurs when the electron density exceeds the initial densities of nth ~ 1018 cm-3 and no ~ 1010 cm-3. For experimental purposes, it is more convenient to use the critical density of the plasma, nth ~ 1.6 x 1021 cm-3, so the threshold will be reached at ητp ~ 30. The definition of plasma density related to the fracture threshold is somewhat arbitrary. However, even if a specific density is selected as the plasma density, the threshold dependence as a function of pulse duration does not change (scaling law).

[0042] In the experiment, the applied electric field is several tens of MV / cm or more. In such a strong electric field, the average energy of electrons is ~ 5eV, and the electron collision time τ is 0.4fs for electrons with energy of U 5-6eV. Electrons repeat two or more collisions during an electrical collision. Therefore, in the case of these high energy electrons, the electric field is essentially a DC electric field. Destruction at optimum frequency is Ermsth (W) = Edcth<u style="single">(1 + w</u>^{<u style="single">2</u>}<u style="single">τ</u>^{<u style="single">2</u>}<u style="single">)</u>^{<u style="single">1/2</u>} It has been shown to correspond to t.dc destruction by. Here, w is the optical frequency and τ is the collision time.

[0043] In DC fracture, the avalanche process η = α (E) Vdrift is described using the ionization rate α per unit length. Here, Vdrift is the drift velocity of electrons. When the electric field is several MV / cm strong, the drift velocity of free electrons is saturated and does not depend on the laser electric field Vdrift ~ 2 x 107 cm / s.

[0044] The ionization rate per unit length of an electron is exactly the probability of eE / Ui P (E) times, so the electron energy is Ui or more or α (E) = (eE / Ui) P ( E). Defining EKT.ZP and Ei as the electron threshold electric fields prevents the thermal, quantum and ionization scattering effects from slowing down. Then, since the electric field becomes E &lt;EKT and can be ignored, the distribution is almost exclusively the heat distribution, and P (E) can be simplified to exp (-Ui / kT). It is estimated that when EKT &lt;E &lt;EP, it becomes P (E) ~ exp (-const / E), and when the electric field (E&gt; EP) is stronger than this, it becomes P (E) ~ exp (-const / E2). Summing up the above three cases, the formula that satisfies both the upper and lower limits of the electric field is as follows. That is, α (E) = (eE / Ui) exp (-Ei / E (1 + E / Ep) + EKT). Therefore, FthαE2τp ~ 1 / τp, that is, in the case of ultrashort pulse, the fluence threshold is E&gt;. If EpEi is satisfied, it will increase.

FIG. 12 is a graph of α as a function of the electric field E. α was calculated from the experimental data as ητp = 30 and η = aVdrift. The curved part shown by the solid line is calculated from the above formula using Ei = 30MV / cm, Ep = 3.2MV / cm and Ekt = 0.01MV / cm. These parameters were calculated from U = eE1. Where U is the appropriate value taken for heat, quantum and ionization energies, where 1 is the corresponding energy-related length (1kt = 1p ~ 5Å atomic space, and 1i ~ 30Å).

[0046] This is the same saturation as the experimental data. Correcting the dotted line with a correction factor of 1.7 is in good agreement with the experimental data. This correction factor of 1.7 is not very important. That is, systematic modifications can be used, and it is possible that the fracture first occurs on the surface and therefore the threshold may be smaller. Furthermore, the uncertainty of the saturation value Vdrift can also be a correction factor. The most important thing here is that the shape (gradient) of the curve given by the above equation is in perfect agreement with the experimental data. Therefore, the laser-induced breakdown mechanism of molten silica using a short pulse of 150 fs and a wavelength of 780 nm (Example 2) also seems to be dominated by the avalanche process.

The curves in FIGS. 3, 8 and 9 show common properties between opaque and transparent materials. That is, each is Fth vs.<u style="single">T</u>^{<u style="single">1/2</u>}It starts with the behavior of, but there is no doubt that a clear change will appear from this behavior next. From the point of view of deviation, each curve is not always the same. This is because the materials themselves are already different. Since the physical properties of each material are different, material-specific analysis is required.<u style="single">SiO</u>_{<u style="single">2</u>}In the case of (Fig. 8), the energy fixing mechanism is due to dielectric fracture. Synchrotron radiation releases densely coupled electrons by polyphoton ionization (MPI), which in turn accelerates them into high-energy electrons by the strong electric fields of the two lasers. Before the laser works, there are only a few relatively high-energy electrons. These electrons collide with other constrained electrons and emit these electrons during the avalanche process. In the case of metallic materials, free electrons are present, and energy is instantly absorbed and redistributed. Regardless of the material, as the pulse becomes shorter, laser-induced breakdown (LIB) or cutting occurs only in the region where the laser intensity exceeds LIB or the cutting threshold. This is not enough time for the surrounding area to react thermally. As the pulse becomes shorter, steam is generated from the cutting material after the pulse is fixed rather than during the pulse fixing. This is because the duration of the pulse is extremely short. It was also found that the laser intensity may fluctuate with the propagation axis (Fig. 13). The beam intensity can be expressed as a function of R or Z as follows. That is, I (Z, R) = Io<u style="single">/ (1 + Z / Z</u>_{<u style="single">R</u>}<u style="single">)</u>^{<u style="single">2</u>} Exp<u style="single">(</u><u style="single">-</u><u style="single">2R</u>^{<u style="single">2</u>}<u style="single">/ W</u>^{<u style="single">2</u>}_{<u style="single">Z</u>}<u style="single">)</u>here,<u style="single">Z</u>_{<u style="single">R</u>}Is Rayleigh Range<u style="single">Z</u>_{<u style="single">R</u>}<u style="single">= </u><u style="single">π</u><u style="single">Wo</u>^{<u style="single">2</u>}<u style="single">/</u><u style="single">λ</u>Is. Wo is the beam size at the waist (Z = 0).

[0048] It can be seen that the maximum electric field value occurs at the center of the waist when Z = R = 0. If the threshold is set accurately, the material can be damaged accurately at the waist, and the damaged volume can represent only a part of the waist in the R or Z direction. Accurate control of damage thresholds or laser intensity variations is extremely important.

[0049] For example, if it is known that the damage threshold or laser intensity variation is within 10%, I (0, Z) / I0 = 1 / (1+ (1+) on the axis (R = 0). With Z / ZR) 2) = 0.9, the damaged volume can be generated at a distance of ZR / 3. Again, the ZR is in the Rayleigh range. Therefore, the beam waist of WO = λ is ZR = πWo2 / λ = πλ, and the distance d to the hole can be ZRπλ / 3 as shown in FIG.

[0050] The maximum intensity is just centered on the beam waist (Z = 0, R = 0). If the threshold is clear, the transparent insulating material can be damaged in a small volume centered near the origin (Z = 0, R = 0). The damage in this case would be much smaller than the beam waist measured in the R direction. Small cavities, holes or damage can be smaller in size than the Rayleigh Range (ZR) within the volume of clear insulating material. In another variation, the lens can be moved to increase the size of the hole or cavity in the Z dimension. In this case, moving the focal point along the Z axis increases the length of the hole or cavity. These features are important for the above-mentioned applications as well as applications such as micro-dimension machining, integrated circuit manufacturing, and data coding to data storage media.

Although the present invention has been described so far in relation to its embodiments, the present invention is not limited by the above description, but is limited only by the claims shown in the claims.

[0052] Embodiments of the invention for which exclusive ownership or privilege is claimed are defined in the claims.

[Effect of the Invention] One of the advantages of the present invention is to identify a rating at which the fracture threshold fluence does not obey the scaling law, and to improve the accuracy of laser-induced breakdown by using such a rating. And also to induce fracture in a certain pattern inside or on the surface of the material. That is, according to the present invention, it is possible to operate the laser at a place where the breaking or cutting threshold becomes accurate. The accuracy can be clearly shown by the I-shaped bars along the curves of FIGS. 8 and 9. These I-shaped bars consistently show low deviations and high accuracy corresponding to ratings below a predetermined pulse width.

BRIEF DESCRIPTION OF THE DRAWINGS [Fig. 1] Fig. 1 is a schematic diagram of a laser-induced breakdown experimental system including a chirped pulse amplification laser system and means for detecting scattered energy and transmitted energy. In the case of a transparent sample, the transmitted energy can also be measured.

FIG. 2 is a graph of the relationship of scattered energies to incident fluence obtained for an opaque sample (gold) when scanned with a pulse duration of 150 femtoseconds (fs) in the system of FIG.

[Fig. 3] The relationship between the fluence threshold and the pulse width is graphed using the calculated value and the experimental value in the case of gold, and the experimental value of gold is obtained by operating the system of Fig. 1 at a wavelength of 800 nm. is there. The arrow indicates Fth on the graph<u style="single">T</u>^{<u style="single">1/2</u>}Indicates a point that is not proportional to. Fth<u style="single">T</u>^{<u style="single">1/2</u>}The relationship of being proportional to is applicable only to pulse widths above a certain level as shown by the solid line.

FIG. 4 is a graphical representation of gold subspot size ablation and machining based on arbitrary units, showing the fluence threshold required to initiate material removal, Fth. Where Rs is the spot size of the incident ray and Ra is the radius of the hole cut in the XY plane.

FIG. 5 is a schematic showing a light intensity profile showing that in the case of laser microdimension machining with ultrafast pulses according to the present invention, only the peak value of the light intensity profile exceeds the cutting / machining intensity. ..

FIG. 6 is a schematic diagram of light rays showing the installation of a disc-shaped mask in an optical path.

FIG. 7 is a plot of scattered plasma radiation and transmitted laser pulses as a function of incident laser pulse energy in a transparent glass sample SiO2.

FIG. 8 is a graph showing the relationship between the fluence threshold (Fth) and the pulse width (T) for the transparent glass sample of FIG.<u style="single">T</u>^{<u style="single">1/2</u>}It is shown that Fth, which fluctuates with, holds only for pulses within the range shown by the solid line. There are previously published reports showing cases of large pulse widths (Squares, Smith Optical Eng 17, 1978 and Triangles, Stokowsky, NBS Spec 541, 1978).

FIG. 9 is a graph of the fluence threshold of corneal tissue, that is, the relationship between the corneal damage threshold and the pulse width. In this case as well, the proportional relationship between Fth and the pulse width is only for a relatively long pulse width.<u style="single">T</u>^{<u style="single">1/2</u>}It shows that it follows the relational expression of.

FIG. 10 is a graph of the relationship between plasma emission and laser fluence, with Fth determined very clearly at a pulse width of 170 fs, showing human corneal rupture data at 170 fs and 7 ns. ..

FIG. 11 is a graph of the relationship between plasma emission and laser fluence, showing that the 7ns pulse width contrasts with the extremely obscure Fth. It shows human corneal destruction data at 170fs and 7ns.

FIG. 12 is a graph of the impact ionization rate per unit distance determined by experiments and theoretical calculations.

FIG. 13 is a schematic representation of the beam shape along the longitudinal Z-axis direction and the exact control position of damage with dimensions in the Z-axis direction.