Method for separating workpieces

Abstract

The present invention relates to a method for separating a workpiece (1) by means of laser pulses (20) of a laser beam (2) along a separation line (10), wherein the laser beam (2) is split into a plurality of sub-laser beams (26) by means of a beam splitter optical unit (62), and the sub-laser beams (26) are focused onto the surface (12) and / or volume of the workpiece (1) by means of a focusing optical unit (64), such that the sub-laser beams (26) are arranged side by side and spaced apart from each other along the separation line (10), wherein the laser pulses (20) of the sub-laser beams (26) are introduced into the workpiece along the separation line (10) to cause material ablation, and the laser power of each sub-laser beam (26) is adjusted according to the ablation depth (AT) achieved in the workpiece (1).

Description

Technical Field

[0001] This invention relates to a method for separating workpieces along a separation line using laser pulses from a laser beam. Background Technology

[0002] The practice of separating workpieces by applying laser pulses is known, for example, for separating sections of a workpiece from the workpiece, wherein the ablation of material in the workpiece can be achieved by applying a laser beam to the workpiece, the ablation being achieved, for example, by sublimation of the workpiece or by melting along with the subsequent expulsion of the melt.

[0003] However, due to shielding and changes in hole geometry, the effectiveness of the cutting process decreases with increasing processing depth, making the laser energy insufficient for rapid material ablation. In other words, the energy required for material ablation increases with processing depth, therefore each point along the desired separation line (along which the workpiece should separate) must be swept by the laser beam multiple times, thus processing time is directly proportional to the number of points swept.

[0004] Here, it is known to use so-called multi-point optics to form multiple sub-beams of a laser beam, and to use these sub-laser beams for workpiece separation by means of a scanner and / or a transfer stage. The ablation of the material can be configured more effectively by introducing multiple sub-laser beams simultaneously.

[0005] While this method has resulted in an overall reduction in processing time and an increase in efficiency, it still suffers from the following drawback: the efficiency of material ablation with increasing processing depth is not compensated. In other words, material ablation slows down with increasing processing depth, thus requiring multiple, time-consuming scans of the sub-laser beam along the separation line even with increased ablation depth.

[0006] Furthermore, a known device (WO 2020 / 016362 A1) enables the arrangement of sub-laser beams arranged side-by-side along a line using a microlens array, wherein the number of sub-laser beams is determined by varying the distance between the individual microlens arrays. Here, the energy of the laser beam provided by the laser can be divided into multiple sub-laser beams, thus the energy of each sub-laser beam varies with the number of sub-laser beams. However, this method does not include a function for workpiece separation. Summary of the Invention

[0007] Based on known prior art, the objective of this invention is to suggest an improved method for separating workpieces along a separation line by means of laser pulses from a laser beam.

[0008] This task is accomplished by separating the workpiece along a separation line using laser pulses from a laser beam. Advantageous extensions of this method are derived from this specification and the accompanying drawings.

[0009] Accordingly, a method is proposed for separating a workpiece along a separation line using laser pulses from a laser beam, wherein the laser beam is split into multiple sub-laser beams by means of a beam splitter optical unit, and each of the sub-laser beams is focused onto the surface and / or volume of the workpiece by means of a focusing optical unit, such that the sub-laser beams are arranged side-by-side and spaced apart from each other along the separation line. According to the invention, material ablation occurs by introducing laser pulses into the workpiece along the separation line, and the laser power of each sub-laser beam is adjusted according to the ablation depth achieved in the workpiece.

[0010] Here, the laser provides laser pulses, wherein the laser pulses move along a beam axis defined by the optical unit of the laser, thereby providing a laser beam of the laser along the beam axis accordingly. The laser is preferably a short-pulse laser providing pulses in the nanosecond range or an ultrashort-pulse laser providing laser pulses in the picosecond or femtosecond range.

[0011] Ultrashort pulse lasers emit laser pulses with exceptionally high energy densities within a very short timescale, thereby producing particularly sharp edges along the separation line. Here, the pulse length of the ultrashort laser pulse can be, for example, shorter than 100 ps, ​​especially shorter than 10 ps, ​​and particularly preferably shorter than 1.5 ps. However, the pulse length can also be, for example, longer than 300 fs. Typically, the pulse length is 1 ps. The peak flux (i.e., the intensity of the incident laser beam or sub-beam) can be 0.3 J / cm². 2 Up to 30J / cm 2 The focal diameter of the incident laser beam or sub-laser beam can be between 1 μm and 100 μm, and is particularly preferably 9 μm.

[0012] The focal diameter of a non-radially symmetric beam profile can typically be defined by the second moment of the power density distribution of the beam profile, as described in ISO 11146-3: "Lasers and laser-related equipment—Test methods for laser beam widths, divergence angles and beam propagation ratios—Part 3: Intrinsic and geometrical laser beam classification, propagation and details of test methods," where the shorter of the two principal axes of the beam profile is assumed to be the focal diameter in this case.

[0013] The laser beam is guided through a beamsplitter optical unit, which, for example, includes a beamsplitter cube that splits the incident laser beam into two sub-beams of equal intensity. However, the beamsplitter optical unit can also include optical elements that split the beam into multiple sub-beams and deflect them such that all sub-beams extend parallel to each other and at a distance from one another. The beamsplitter optical unit can also include multiple optical elements, wherein one element splits one or more incident laser beams into multiple sub-beams, while another element deflects the sub-beams in a parallel direction.

[0014] For example, a beam splitter optics unit can ensure that an incident laser beam is split into five, ten or more sub-laser beams.

[0015] Furthermore, by using a beamsplitter optical unit, the high pulse energy of the pulsed laser can be efficiently distributed to different sub-laser beams, thereby improving the efficiency of the processing. Specifically, all sub-laser beams can have the same laser energy, or they can have different laser energies. For example, the energy fluctuation between sub-laser beams can be as high as 50% or more. For example, the sub-laser beams can have an increase or decrease in laser energy along one or more directions, resulting in an energy gradient in the beam aggregation. In particular, this energy gradient can be generated in a targeted manner by means of coherent beam combining, or individual beams can be shut off.

[0016] A separation line should be understood as the line along which the workpiece should be separated. For example, a separation line can be set as a straight line. However, a separation line can also be a curved line. In particular, a separation line can also be a more complex separation line or part of a separation geometry.

[0017] For example, if a rectangular shape is to be cut from a workpiece, the separation line can be, for example, just the edge of the rectangle, but it can also be pre-defined by the entire outline to be cut.

[0018] The sub-laser beams are arranged side-by-side and spaced apart from each other along the separation line. Here, the distance between adjacent sub-laser beams is the distance between their geometric centers along the separation line. However, this distance can also be determined along the separation line between the maximum intensity values ​​of adjacent sub-laser beams. The start and end points of the distance measurement can also be those points where the laser beam or beam axis intersects the workpiece surface.

[0019] For all adjacent sub-laser beams, the distance between adjacent sub-laser beams can be the same. For example, each sub-laser beam can be 5 μm, 10 μm, or 100 μm away from its adjacent sub-laser beams. However, it is also possible that the distances between adjacent sub-laser beams are different. Thus, for example, one sub-laser beam can be 5 μm or more away from one adjacent sub-laser beam, but only 3 μm, 10 μm, or less away from another adjacent sub-laser beam.

[0020] In particular, flexible beam splitting can be achieved through coherent beam combining, in which the sub-laser beams have adjustable distances, number of spots, and spot intensities.

[0021] In particular, each sub-laser beam has only two directly adjacent sub-laser beams because the sub-laser beams are arranged along the separation line.

[0022] The number and spacing of the sub-laser beams can be selected so that the sub-laser beams are spaced apart from each other along the entire desired separation line. In other words, the entire region to be separated, defined by the separation line, is covered by the sub-laser beams.

[0023] Each sub-laser beam is focused using a focusing optics unit. For example, the focusing optics unit can include a lens or other diffractive element that convergently deflects the sub-laser beams into a focal region. This specifically means that the intensity of the sub-laser beams is concentrated into a minimal spatial area within the focal region. For example, a separate focusing optics unit can be provided for each sub-laser beam. However, it is preferable to focus all sub-laser beams using a single focusing optics unit.

[0024] Alternatively, the focusing optics can deflect sub-laser beams incident at different angles or lacking a parallel beam axis onto a parallel beam axis. This specifically means that all sub-laser beams are oriented parallel to each other after passing through the focusing optics. However, it is also possible for the beam axis to be deflected by the focusing optics onto a diverging or converging beam axis. For example, this could be meaningful: during material ablation, the shading effect of material edges in the process region should be avoided. This diverging beam axis can be achieved, for example, using non-telecentric objectives or non-telecentric optical structures.

[0025] For example, the focusing optics unit can be adapted to the beam splitter optics unit such that sub-laser beams emitted from the beam splitter optics unit are incident on the focusing optics unit at different angles, and then the sub-laser beams are oriented parallel to each other by the focusing optics unit. This, for example, allows the beam splitter optics unit to be located at the focal point of the focusing optics unit. For example, the focusing optics unit and the beam splitter optics unit can also be located in each other's conjugate planes.

[0026] Sub-laser beams can be focused onto the surface and / or volume of a workpiece. This means that the focal zone is, for example, located exactly on the surface of the material or below the material surface in the direction of the beam. All sub-laser beams are preferably focused into the same plane.

[0027] In an extended scheme where a separate focusing optics unit is set for each sub-laser beam, it is also possible to precisely focus one sub-laser beam onto the surface while focusing another sub-laser beam into the volume of the material.

[0028] The sub-laser beams are focused by a focusing optics unit, such that each sub-laser beam converges within a focal region. This means that the intensity of the laser beam is highest within the focal region. For example, in the case of a Gaussian beam profile, the geometric focal region can be defined by the focal point where the maximum intensity of the beam occurs. However, the geometric focal region can also be a non-Gaussian beam, such as a beam with a flat-top focal region or a non-diffractive beam, such as a Bessel beam or a Bessel-like beam, especially the maximum intensity of a Bessel-Gaussian beam. The shape of the focal region can vary depending on the beam profile used.

[0029] Introducing a laser pulse into a workpiece along a separation line can be equivalent to the laser pulse being fully or partially absorbed by the workpiece. For example, this can cause the workpiece to be intensely heated by the laser pulse. This can cause the heated material in the workpiece to detach from the overall workpiece composite due to melting or sublimation. In particular, this can mean laser ablation, laser drilling, or laser cutting.

[0030] Especially when removing material from a workpiece, the effectiveness of the cutting process decreases with increasing machining depth. Here, machining depth is the distance between the original workpiece surface and the new workpiece surface in the trench, the new workpiece surface being generated by the laser pulses of the sub-laser beam along the separation line due to the ablation process. In particular, the machining depth can be different along the separation line, for example because the workpiece is non-planar or unevenly placed on the workpiece holder, or because the workpiece has different densities or different layers that respond differently to the introduced sub-laser beam, or because the sub-laser beam used to deepen the trench has different energies from the start of the method; correspondingly, the different ablation rates of the various regions of the separation line machined by the respective sub-laser beams result in different machining depths.

[0031] Here, the effectiveness of the cutting process describes how much laser energy is used to achieve what ablation or ablation depth. For example, the effectiveness of the cutting process decreases with increasing processing depth, but it can also increase depending on the characteristics of the workpiece.

[0032] If the laser power of each sub-laser beam is adjusted, it could mean that more laser energy is introduced through each sub-laser beam compared to the case of unadjusted sub-laser beams. However, it could also mean that less laser energy is introduced through each sub-laser beam compared to the case of unadjusted sub-laser beams.

[0033] The adjustment of the laser energy of the sub-laser beam depends on the ablation depth, which means that the effectiveness of material ablation at the corresponding ablation depth should be considered when adjusting the laser energy of the sub-laser beam. If the sub-laser beam efficiently ablates the material from the workpiece at, for example, a first ablation depth, and ablates the material less efficiently at a second ablation depth due to the low laser power of the sub-laser beam, then the laser power of the sub-laser beam can be increased.

[0034] However, the efficiency of material ablation can also be related to the ablation rate per unit time, meaning that material ablation must occur along a certain length of the separation line after a given time. For example, the laser power of each sub-laser beam can be adjusted to achieve the maximum ablation rate along the separation line. Furthermore, optimization of the ablation depth and length along the separation line is also possible. In particular, the ablation volume, i.e., the total amount of material ablated from the workpiece, can be optimized by adjusting the power of each sub-laser beam.

[0035] Therefore, the proposed method has the advantage of improving the efficiency and edge quality of the workpiece (e.g., silicon) cutting process by continuously utilizing the provided laser power. Here, the ablation power per unit time at the corresponding position processed by the corresponding sub-laser beam can be kept substantially constant, for example.

[0036] The laser power of each sub-laser beam can be adjusted by adjusting the number of sub-laser beams given a given input power of the laser.

[0037] This means that, in the first ablation depth, it is particularly efficient to ablate the workpiece material using a first number of sub-laser beams. Here, the laser power is divided into the first number of sub-laser beams by the beam splitter optical unit.

[0038] For example, a laser beam can be split into five sub-beams, so that each sub-beam introduces one-fifth of the laser energy into the workpiece.

[0039] For example, the ablation efficiency using five sub-laser beams at the first ablation depth can be very high, but the efficiency at the second ablation depth can be very low. In this case, for the second ablation depth, the number of sub-laser beams can be reduced to, for example, four, so that each sub-laser beam introduces one-quarter of the laser power into the workpiece material. By increasing the laser power of the sub-laser beams at the second ablation depth, the efficiency of the ablation process can also be improved.

[0040] However, for example, at the second ablation depth, five sub-laser beams can introduce more energy into the workpiece than is sufficient for the ablation process, thus improving the efficiency of the method by increasing the number of sub-laser beams from five to six. Although each sub-laser beam therefore introduces less energy into the workpiece, this energy is introduced into the workpiece simultaneously by a larger number of sub-laser beams, allowing for simultaneous processing at more locations. This increases the number of sections where processing is performed simultaneously along the separation line using sub-laser beams, further enhancing the efficiency of the ablation process.

[0041] The laser power of each sub-laser beam can be increased with the increase of ablation depth.

[0042] This could mean, for example, that the laser power of the input laser increases with increasing ablation depth, but it could also mean that the number of sub-laser beams decreases with increasing ablation depth, resulting in the laser energy being distributed across fewer sub-laser beams while maintaining the same laser power. Thus, the laser energy introduced into the workpiece can be increased by utilizing each sub-laser beam.

[0043] The beam splitter optical unit can include and be composed of multiple microlens arrays, for example, four microlens arrays.

[0044] A microlens array comprises an arrangement of multiple microlenses. Here, a microlens is a small lens, particularly a lens with a typical center-to-center distance (“pitch”) of 0.1 to 10 mm, preferably 1 mm, wherein each individual lens in the arrangement is capable of having the effect of a normal macro lens.

[0045] An angular spectrum is generated from a (at least substantially) collimated input laser beam using a multi-microlens array. This spectrum is imaged onto a focal plane, typically located on or in the workpiece, by means of Fourier lenses. Multiple sub-laser beams are generated here through interference and diffraction effects, depending on the spacing of the microlens array. The number of sub-laser beams varies due to variable modifications to the interference pattern. The flux and intensity of the sub-laser beam spots are inversely proportional to the number of sub-laser beams, and the sum of the laser powers of the sub-laser beams is substantially equivalent to the laser power of the input laser beam.

[0046] The focal plane is synonymously used for the focusing surface of the sub-laser beams. A microlens array arranges microlenses in at least one direction perpendicular to the beam propagation direction. A uniform beam profile with sub-laser beams of uniform spacing is generated in the focal plane, wherein the sub-laser beams are arranged in a direction perpendicular to the beam propagation direction (e.g., the X-direction). Therefore, the sub-laser beams can be arranged in a direction in which the microlenses (e.g., constructed as cylindrical lenses) are also arranged.

[0047] Preferably, four microlens arrays are used, through which the laser beam passes. On the one hand, a simple and compact structure is still achieved using four microlens arrays. On the other hand, this allows for flexible adjustment of the number of sub-laser beams.

[0048] The beam splitter optical unit may also include at least two pairs of microlens arrays, wherein the microlens arrays of the pair of microlens arrays are spaced at a fixed distance from each other.

[0049] For example, this means that one microlens array pair can be shifted relative to another microlens array pair. For example, the distance between the first microlens array and the third microlens array can be fixed. For example, the distance between the second microlens array and the fourth microlens array can be fixed simultaneously. At the same time, the relative distance between the first and second microlens array pairs can be changed. However, it is also possible for other combinations of microlens arrays to be fixedly connected to each other to maintain a constant distance.

[0050] By arranging two microlens arrays at a fixed distance, the degrees of freedom of the optical arrangement can be restricted or predefined in a targeted manner, making it easier to adjust the number of sub-laser beams.

[0051] The laser power of the sub-laser beam and / or the number of sub-laser beams can be adjusted by shifting the two microlens arrays relative to each other.

[0052] The microlens array produces a change in the interference pattern relative to each other, that is, a change in the number of sub-laser beams.

[0053] Therefore, an adjustment mechanism can be used to adjust the optical characteristics of at least some of the microlens arrays in the beam path. This allows for the adjustment of the effective focal length of the entire microlens array and thus the number of sub-laser beams. By appropriately selecting the adjustment position of the adjustment mechanism, a uniform intensity distribution across the different sub-laser beams can be achieved.

[0054] Here, the adjustment mechanism can be motor-driven and equipped with electronic control devices, wherein different adjustment positions associated with various numbers of sub-laser beams are programmed in the electronic control devices and can be automatically activated by means of the electronic control devices. This enables convenient and rapid switching between different numbers of sub-laser beams when ready for the next processing task.

[0055] The first and third microlens arrays can be arranged in a fixed position within the beam path itself, while the second and fourth microlens arrays, located within the beam path, can be arranged on a movable carriage, or vice versa. Then, within a certain adjustment range of the carriage, there is a linear relationship between the number of sub-laser beams and the carriage position, thereby allowing for particularly convenient adjustment of the number of sub-laser beams.

[0056] Microlens arrays or pairs of microlens arrays can include convex microlenses.

[0057] Here, the convex microlens has a so-called virtual focal point located in front of the microlens in the direction of beam propagation. A microlens array composed of convex lenses is also called a convex microlens array.

[0058] If the incident laser beam is incident on the convex microlens array, the beam is broadened. In particular, this reduces the intensity of the laser beam behind the convex microlens array in the beam propagation direction. This prevents damage to other optical elements in the beam path.

[0059] Sub-laser beams can be focused onto the same focal plane.

[0060] This means that all geometric focal points along the direction of the sub-laser beams are at the same distance from the surface. For example, all sub-laser beams can be precisely focused onto the surface. However, for example, all sub-laser beams can also be focused 10 μm below the surface of the workpiece. Furthermore, all focal points of the sub-laser beams can also lie in a plane perpendicular to the beam propagation direction, independent of the surface topology of the workpiece.

[0061] This has the following advantages: the separation process and ablation depth are well controlled and produce uniform separation edges.

[0062] The sub-laser beam can move along the separation line from its original position by a deflection value.

[0063] To achieve precise separation of the workpieces, the sub-laser beam preferably moves along the separation line. In particular, in the case of a straight separation line, the movement of the sub-laser beam is preferably a linear deflection following the straight shape of the separation line, and in the case of a curved separation line, the deflection of the sub-laser beam is preferably a deflection following the curved shape of the separation line.

[0064] Here, the initial position is the initial location of the sub-laser beam on the workpiece surface. Specifically, the initial position is the position of the sub-laser beam before it undergoes deflection or leaves the workpiece. The deflection value is the maximum distance between the moving sub-laser beam and its initial position. The deflection value is measured in a manner similar to the distances of different sub-laser beams along the separation line.

[0065] The laser emits laser pulses during the movement of the sub-laser beam. This movement enables the laser pulses to be introduced along the separation line. In particular, it allows the laser pulses to be introduced into the workpiece along the entire separation line.

[0066] For example, the sub-laser beam can be moved along the separation line by means of a scanner system, so that the laser beam is introduced into the workpiece along the separation line and the workpiece is separated along the separation line. In particular, this can be achieved by a scanner system that is preferably arranged after the beam splitter optical unit in the beam propagation direction and guides the sub-laser beam along the separation line.

[0067] The motion of the sub-laser beam is preferably related to the repetition rate of the laser providing the laser beam to ensure uniform material ablation.

[0068] With the sub-laser beam moving at a constant speed, the repetition rate or pulse rate of the laser can be selected accordingly, ensuring that the separation line is swept without gaps. For example, this could mean that with a motion speed of 10 μm / μs and a focal diameter of 1 μm, the repetition rate must be 10 MHz. This ensures that the sub-laser beam sweeps through every point on the separation line during its motion and introduces laser pulse energy. However, this could also specifically mean selecting a significantly faster repetition rate, causing the focal areas of successively introduced laser pulses from the same sub-laser beam to overlap, i.e., the distance between successively introduced laser pulses is less than the focal diameter or the diameter of the focal area.

[0069] Conversely, if the motion speed is not constant, the repetition rate or pulse rate can be matched to the motion speed. For example, in the case of non-uniform motion along a separation line, fewer laser pulses can be introduced in slow motion segments (where the motion speed is particularly low), while more laser pulses must be introduced in fast motion segments (where the motion speed is very high). For example, the pulse rate can therefore be inversely proportional to the motion speed. In particular, this also applies to periodic motion discussed below.

[0070] Therefore, the laser energy of the sub-laser beam can be uniformly distributed along the separation line, thereby ensuring uniform material ablation. In particular, uniform material ablation means that the ablation depth is substantially the same along the separation line (provided the material surface is flat).

[0071] Sub-laser beams can first move along the separation line from their original positions with a deflection value, wherein the deflection value is less than or equal to the distance between two adjacent sub-laser beams, preferably equal to the distance between the sub-laser beams, and can then move along the separation line back to their original positions, wherein the movement can be particularly periodic.

[0072] The deflection value is less than or equal to the distance between two adjacent sub-laser beams, preferably equal to the distance between the sub-laser beams. This specifically means that the first sub-laser beam moves to the original position of the adjacent second sub-laser beam. Simultaneously, the second sub-laser beam can, for example, move to the original position of the adjacent third sub-laser beam, wherein the third sub-laser beam is different from the first sub-laser beam, and so on. This ensures that the laser pulse is introduced into the workpiece along the entire separation line.

[0073] However, the first sub-laser beam can also move only half the distance to the adjacent sub-laser beam. The sub-laser beam can also move two-thirds of the distance in the direction of the separation line, and then move one-third of the distance against the separation line after returning to its original position. Movement in both directions along the separation line is particularly feasible.

[0074] After deflection, the sub-laser beam moves back to its original position along the separation line. The return motion can occur at the same velocity as the previous motion, or it can have a different velocity profile.

[0075] All sub-laser beams preferably perform the same motion simultaneously.

[0076] The laser emits laser pulses during the movement of the sub-laser beam. This movement introduces the laser pulses along the separation line. The introduction of the laser pulses is equivalent to the laser pulses being absorbed entirely or partially by the material. For example, this can cause the material to be intensely heated by the laser pulses. This can cause the heated material to detach from the overall material composite.

[0077] This has the advantage that the movement line (i.e., deflection value) of each sub-laser beam can be significantly shorter than the arrangement of the sub-laser beams (i.e., the entirety of the sub-laser beams generated by the beam splitter optical unit) must be guided over the entire workpiece. Therefore, the complex axial or scanner positioning and orientation required for the actual separation process can be eliminated. For example, it also eliminates the need to pass beyond the workpiece edge, where the workpiece receiving part would be damaged. Overall, this method allows for the processing of individual components on a workpiece (e.g., on a silicon wafer) instead of processing the entire wafer at once. Therefore, overall, the flexibility of the geometry of different components on the workpiece is increased.

[0078] Periodic motion means that sub-laser beams deflect from their original positions and return to their original positions after fixed time intervals. This specifically means that the deflection occurs along a separation line, and the sub-laser beam returns to its original position after reaching a deflection value, then deflects against the separation line, and returns to its original position again after reaching a deflection value. However, it is also possible that periodic motion occurs between the original positions of one sub-laser beam and the original positions of adjacent sub-laser beams.

[0079] In other words, the sub-laser beam starts from the initial position and performs an oscillating motion along the separation line, preferably a periodic oscillating motion.

[0080] Preferably, after at least one movement in which the sub-laser beam deflects from its original position by a deflection value and returns to its original position, the original position is shifted along the separation line by a displacement value, and the movement of the sub-laser beam is repeated. This displacement value can also be achieved, particularly by means of a scanner system. This configuration is preferably chosen if the beam focusing of the sub-laser beam cannot cover the entire separation line, or if the feed motion should be superimposed with scanning motion (e.g., in the case of high speed and / or complex structures), and separation along the separation line must be performed accordingly gradually. Here, the displacement value preferably corresponds to the range of the sub-laser beam focusing in the direction of the separation line.

[0081] In other words, the simultaneous processing width (which is pre-given by the focusing of the sub-laser beams applied to the exit and return motions) is used as the shift value, such that the simultaneous processing widths are placed side by side successively. Thus, the processing or separation of the workpiece along the separation line first occurs with the first processing width, and after the separation of this segment of the separation line corresponding to this processing width is completed, the focusing of the sub-laser beams is shifted by the shift value, so that separation in the processing width can occur again.

[0082] However, it is preferable that each structure to be separated from the workpiece can be separated without displacement, using a processing width of the focused sub-laser beams. For this purpose, the number of sub-laser beams and their distance from each other are preferably adapted to enable simultaneous processing of the entire length important to the respective structure. In other words, it is preferable to attempt, for example, to simultaneously cut one side of the chip.

[0083] The separation line can be a straight line, and the periodic motion of the laser beam can be a linear deflection.

[0084] This means that the separation line is not curved, that is, it has no geometric curvature. The sub-laser beam can only move along the separation line, so if the sub-laser beam is guided on a straight separation line, linear deflection occurs.

[0085] In other words, the sub-laser beams are preferably always located on the separation line, even when performing their oscillating motion.

[0086] Preferably, after at least one movement in which the sub-laser beam leaves the initial position with a deflection value and returns to the initial position, the initial position is shifted along the separation line with a shift value, and the movement of the sub-laser beam is repeated. This configuration is preferably chosen if the beam focusing of the sub-laser beam cannot cover the entire separation line and separation along the separation line must be carried out gradually accordingly. Here, the shift value preferably corresponds to the range of the sub-laser beam focusing in the direction of the separation line.

[0087] In other words, the pre-defined simultaneous processing width of the sub-laser beams (which are subjected to leave and return motion) is used as a shift value, such that the simultaneous processing widths are placed side by side. Thus, the processing or separation of the workpiece along the separation line first occurs with a first processing width, and after the separation of this segment of the separation line corresponding to this processing width is completed, the sub-laser beams are shifted by the shift value, so that separation in the processing width can occur again.

[0088] The periodic motion of the sub-laser beam can be generated by the periodic motion of the focusing optical unit.

[0089] Accordingly, the focusing optics unit can move mechanically, for example, laterally, that is, perpendicular to the beam direction. However, this could also mean that the focusing optics unit is tilted at an angle relative to the beam direction.

[0090] Because the sub-laser beams have different incident positions on the focusing optics, they extend through the focusing optics along different beam paths, thus being imaged onto different locations within or on the workpiece after leaving the focusing optics. Therefore, the amplitude of motion of the focusing optics largely determines the deflection value along the separation line. If the focusing optics are not deflected, the imaging of the sub-laser beam onto the workpiece determines its original position on the workpiece.

[0091] When using microlens arrays, it is particularly possible to generate the motion of sub-laser beams by shifting the elements within the microlens array device.

[0092] Pulsed lasers can be ultrashort pulse lasers.

[0093] Ultrashort pulse lasers emit laser pulses with exceptionally high power densities on a small timescale, thereby producing particularly sharp edges along the separation line. Here, the pulse length of the ultrashort laser pulse can be, for example, shorter than 100 ps, ​​especially shorter than 10 ps, ​​and particularly preferably shorter than 1.5 ps. However, the pulse length can also be, for example, longer than 300 femtoseconds. Typically, the pulse length is 1 ps. The peak flux (i.e., the intensity of the incident laser beam or sub-beam) can be between 0.3 J / cm² and 30 J / cm², typically between 1 J / cm² and 5 J / cm². The focal diameter of the incident laser beam or sub-beam can be between 1 μm and 100 μm, and the focal diameter is particularly preferably 9 μm.

[0094] The sub-laser beam and the workpiece can be displaced relative to each other; in particular, the sub-laser beam and the workpiece can be displaced by a feed along the feed direction.

[0095] The displacement of the sub-laser beam and the workpiece relative to each other means that the sub-laser beam and / or the workpiece can move, wherein the offset of the sub-laser beam relative to the original position of the two objects is achieved by the movement of one or both objects.

[0096] In particular, displacement can be achieved using a scanner or axis system (e.g., an XYZ stage), which can, for example, move the workpiece in all spatial directions. This displacement can be superimposed with oscillating motion.

[0097] The feed can have a feed rate along the feed direction. This means that the relative motion has a certain velocity, thereby guiding the sub-laser beam relative to the workpiece at a certain velocity. In particular, the feed velocity vector is oriented parallel to the separation line, that is, the feed direction is given by the separation line. During the motion, due to the introduction of pulsed laser into the workpiece, material ablation occurs along the separation line.

[0098] The input laser beam and / or sub-laser beam can be decomposed into two mutually orthogonal polarization components during its passage through the birefringent polarizer element.

[0099] Birefringence is understood as the ability of an optical material to separate an incident laser beam into two sub-beams that are perpendicular to each other. This occurs because the optical material has different refractive indices depending on the polarization and the angle of incidence of the light relative to the optical axis of the material.

[0100] In the context of this application, sub-laser beams polarized perpendicularly to each other are understood as linearly polarized sub-laser beams whose polarization directions are oriented at 90° angles to each other. However, sub-laser beams polarized perpendicularly to each other are also understood as circularly polarized sub-laser beams with opposite rotation directions, that is, two circularly polarized sub-laser beams, one left-handed and the other right-handed. The conversion from linearly polarized sub-laser beams with mutually perpendicular polarization directions to circularly polarized sub-laser beams with opposite rotation directions can be achieved, for example, by means of appropriately oriented delay plates (λ / 4 plates).

[0101] The emitted laser beam, decomposed by a birefringent polarizing element according to its polarization components, can have angular and / or positional offsets relative to each other.

[0102] This can also be explained by the anisotropy of the refractive index of the optical material of the polarizing element in different polarization directions.

[0103] For example, the sub-laser beam can have an angular offset after passing through the birefringent polarizing element. This means that the sub-laser beam with the first polarization behind the birefringent polarizing element does not extend parallel to the sub-laser beam with the second polarization.

[0104] To produce an angular offset (without a positional offset), a birefringent polarizer element can have a beam exit surface tilted at an angle relative to the beam incident surface. In this case, the optical axis of the birefringent crystal is typically oriented parallel to the beam incident surface. In this configuration, at the beam exit surface, two sub-laser beams exit from the birefringent crystal at the same position and with a defined angular offset.

[0105] For example, the sub-laser beams can have a positional shift after passing through the birefringent polarizing element. This means that the sub-laser beam with the first polarization extends parallel to the sub-laser beam with the second polarization after passing through the birefringent polarizing element. However, the parallel shift of these two sub-laser beams makes the distance between them finite.

[0106] To produce a positional shift (without an angular shift), a birefringent polarizer element can have, for example, a parallel-oriented, typically flat beam incident surface and beam exit surface. In this case, the optical axis of the birefringent crystal is typically oriented at an angle relative to the beam incident surface. If the input laser beam strikes the beam incident surface perpendicularly, a pure positional shift is produced at the beam exit surface.

[0107] The distance between individual sub-laser beams with different polarizations can be determined by polarization elements (e.g. during manufacturing) or by the orientation of the optical axis of the crystal relative to the incident laser beam.

[0108] Sub-laser beams arranged side-by-side and spaced apart from each other along a separation line can have alternating polarization.

[0109] If adjacent sub-laser beams have mutually orthogonal polarizations, then the polarizations are alternating. For example, if the first sub-laser beam has right-hand circular polarization, the second sub-laser beam has left-hand circular polarization, and the third sub-laser beam has right-hand circular polarization, then the polarizations are alternating. Alternatively, if the first sub-laser beam is polarized along the separation line, the second sub-laser beam is orthogonal to the separation line, and the third sub-laser beam is polarized along the separation line, then the polarizations are alternating.

[0110] If a laser beam, for example, generated by a single-mode laser and having a Gaussian beam profile, is divided into two or more sub-beams and these sub-beams are at least partially superimposed, this can lead to undesirable interference effects if the sub-beams have the same or similar polarization. Therefore, when focusing the sub-beams, the focal regions or focusing sections cannot be arbitrarily close together, and thus the sub-beams are typically focused at focal regions spaced apart from each other on the workpiece.

[0111] When using sub-laser beams with mutually perpendicular polarization states, (partial) superposition will not cause interference effects from laser beams from different positions or angular ranges, provided that the polarization states of the corresponding sub-laser beams are consistent across the entire relevant beam cross section or corresponding focal region. Therefore, the polarization of the corresponding sub-laser beams should vary as little as possible across the beam cross section or focal region depending on their position. In this case, focal regions can be arbitrarily close to each other, partially or possibly completely overlapping, and even forming uniform focal regions, specifically both laterally (i.e., perpendicular to the propagation direction of the sub-laser beams) and longitudinally (i.e., along the propagation direction of the sub-laser beams).

[0112] The polarization of each sub-laser beam can be oriented along or orthogonal to the separation line.

[0113] The polarization energy along the separation line means that, in the case of linear polarization, the polarization axis is parallel to the separation line. In the case of a curved separation line, this energy means that the polarization axis is tangent to the separation line, where the focal point lies on the separation line. Especially in the case of a straight separation line, this energy means that the polarization axis lies entirely on the separation line.

[0114] Orienting energy orthogonal to the separation line means that, in the case of linear polarization, the polarization axis is orthogonal to the separation line. In the case of a curved separation line, this energy means that the polarization axis is orthogonal to the tangent of the separation line, where the focal point lies on the separation line and the tangent forms at the focal point.

[0115] For example, when the polarization is perpendicular to the feed direction (i.e., perpendicular to the separation line), higher quality cutting results can be achieved in the case of silicon wafers, such as improved fracture resistance. For example, the workpiece can be ablated efficiently by means of polarization parallel to the feed direction.

[0116] However, depending on the material, the effect of polarization can be exactly the opposite or different, thus achieving higher quality cutting results when polarization is parallel to the feed direction than when polarization is perpendicular to the feed direction. In any case, polarization can influence the cutting quality.

[0117] By using delay plates and / or delay elements and filter elements, all sub-laser beams can have the same polarization.

[0118] For example, the delay plate can be a λ / 2 plate, by means of which the linear polarization of the incident laser line can be rotated. For example, p-polarization can be converted to s-polarization by means of the λ / 2 plate. However, the delay plate can also be a λ / 4 plate, by means of which circular polarization is converted to linear polarization. For example, the incident laser beam can have circular polarization and p-polarization after the laser line has passed through the λ / 4 plate.

[0119] For example, the filter element can be a polarizer, such as a thin-film polarizer, or any other polarizer element. In particular, the filter element can be mounted after the polarizer element in the fabricated optical unit. Furthermore, the filter element can also be switched into the beam path or turned off.

[0120] This allows a beam path with one polarization to be filtered out from the polarizer element, so that a sub-laser beam with only one polarization direction is still introduced into the workpiece.

[0121] The entire separation process can be divided into different sub-separation processes, each of which is performed using a p-polarized or s-polarized sub-laser beam or a sub-laser beam with different polarizations.

[0122] Using sub-laser beams with different polarizations is advantageous because different polarization directions relative to the separation line affect the ablation depth and the quality of the ablation edge for each pulse. For example, polarization parallel to the feed direction is suitable for achieving particularly efficient material ablation per laser pulse. Conversely, polarization perpendicular to the feed direction can produce particularly smooth edges with high fracture resistance.

[0123] For example, a certain ablation depth can be achieved in the first step using a sub-laser beam polarized parallel to the feed direction. For example, the cut edge can be post-processed in the second step using a sub-laser beam polarized perpendicular to the separation line. Furthermore, different sequences can be determined, according to which the separation line is traversed by sub-laser beams of different polarizations. In particular, sub-laser beams with different polarizations can be introduced simultaneously.

[0124] For example, sub-laser beams with different polarizations can be placed side by side. Because of their different polarizations, there is no interference effect between adjacent sub-laser beams, allowing them to be placed close together.

[0125] Therefore, efficiency and quality can be improved by processing workpieces along or perpendicular to the separation line using different polarizations within different processing strategies and under different processing conditions.

[0126] For example, the polarization of the sub-laser beam can be determined by a polarizer. This could be a thin-film polarizer that allows only one polarized sub-beam to pass through. Such a polarizer can be introduced into the beam path before the sub-laser beam is deflected onto the workpiece. In particular, the polarization of the sub-beam can also be determined directly by the polarizer element, where the orientation of the optical axis relative to the incident beam can be adjusted. Therefore, the polarization of the sub-beam can be determined and selected.

[0127] By alternating between parallel and perpendicular polarization, effects related to both quality and efficiency can be simultaneously utilized. The process can also be divided into two, three, or more process steps. In this case, high efficiency and improved economy can be achieved, for example, at the beginning of the process using sub-laser beams with alternating polarization or only parallel polarization. At the end of the process, the polarization can be rotated by 90°, for example, using a λ / 2 plate or a switching liquid crystal, liquid crystal display, lithium niobate crystal, or Pockels cell, thereby improving workpiece quality through polarization oriented perpendicular to the feed direction.

[0128] The workpiece can be a wafer, especially a silicon wafer, from which individual chips should be separated.

[0129] In particular, in chip manufacturing, what is possible is that the wafer only needs to be accurately positioned, without needing to be oriented. This means that the orientation of the wafer along the feed axis no longer allows for deviations of a few micrometers across the entire wafer width. Attached Figure Description

[0130] Preferred further embodiments of the invention will be described in more detail below with reference to the accompanying drawings. Herein lies:

[0131] Figure 1A , Figure 1B This is a schematic diagram without a method for adjusting the laser intensity;

[0132] Figure 2A , Figure 2B , Figure 2C It is a sketch used to determine the distance of the sub-laser beams;

[0133] Figure 3 This is a schematic diagram illustrating the correlation between laser intensity and the number of sub-laser beams;

[0134] Figure 4A , Figure 4B This is a schematic diagram of a microlens array optical unit;

[0135] Figure 5A , Figure 5B This is a schematic diagram of the time correlation when a sub-laser beam is introduced with periodic deflection;

[0136] Figure 6A , Figure 6B This is a schematic diagram of a polarizer element;

[0137] Figure 7 This is a schematic diagram of the processing of optical units and beam paths;

[0138] Figure 8 This is a schematic diagram of the method;

[0139] Figure 9A , Figure 9B , Figure 9C This is another schematic diagram of the method;

[0140] Figure 10 It is a measurement curve used to determine the erosion efficiency;

[0141] Figure 11 It is a measurement curve used to determine the erosion efficiency;

[0142] Figure 12 It is a measurement curve used to determine the erosion efficiency; and

[0143] Figure 13 It is a comparison of the ablation of materials with different parameters. Detailed Implementation

[0144] Preferred embodiments are described below with reference to the accompanying drawings. Here, in different drawings, the same reference numerals are used for elements that are the same, similar, or have the same function, and repeated descriptions of these elements are omitted to avoid redundancy.

[0145] Figure 1A and Figure 1B A method for separating workpiece 1 (e.g., for separating a silicon wafer) at two different times is illustrated schematically.

[0146] exist Figure 1A In this process, a sub-laser beam 26 is introduced into the workpiece 1, whereby, for example, the workpiece 1 is heated in the focal zone by laser pulses extending along the sub-laser beam 26, causing the heated material of the workpiece 1 to be released from the surrounding bulk composite. This process is known as laser ablation, laser drilling, or laser cutting.

[0147] exist Figure 1A At the start of the process (when surface 12 is still flat, i.e., especially untreated), sub-laser beam 26 strikes surface 12 perpendicularly, for example, everywhere. Sub-laser beam 26 has a beam cross-section, for example, a symmetrical beam cross-section (e.g., generated by a Gaussian beam profile), such that the laser intensity at the beam focus is maximized in the lateral direction and decreases toward the edge of the beam.

[0148] Correspondingly, the workpiece is heated more intensely at the center of the beam than at the edges, making it easier for the material heated at the beam center to detach from the workpiece than the material at the beam edges. As a result, the ablation depth of each pulse is unevenly distributed across the beam cross-section, thus making the bottom of the borehole uneven.

[0149] Furthermore, at the start of the processing, most of the focal zone is located below surface 12 or within the volume of workpiece 1, so that for each laser pulse, a relatively large amount of laser energy is absorbed by the material of workpiece 1, thereby making the processing very efficient.

[0150] Figure 1B This unevenness is amplified with increasing ablation depth because the sub-laser beam 26, with its radially decreasing beam profile compared to the edges of the beam, causes more and more material to be released from the center of the borehole. In particular, the intensity distribution in the borehole corresponds to the intensity distribution within the beam profile, resulting in a significant increase in the surface area processed by the laser beam and a decrease in intensity during the progressive processing. This, in particular, reduces the efficiency of the material ablation process. To improve the efficiency of the process, it is therefore necessary to increase the intensity of the sub-laser beam 26 with increasing ablation depth.

[0151] Figure 2AA sketch is shown to illustrate the determination of the distance between the sub-laser beams 26. Here, it is assumed that the surface 12 of the workpiece 1 has curvature or is uneven. However, the distance determination can be similarly performed for a flat surface 12. In this case, the beam splitter optics unit splits the laser beam 2 into a first sub-laser beam 260 and a second sub-laser beam 262. The first sub-laser beam 260 extends substantially parallel to the second sub-laser beam 262, with the focal area 24 of the first sub-laser beam 260 located within the volume of the workpiece 1, while the focal area 24 of the second sub-laser beam 262 is located on the surface 12 of the workpiece 1. The only factor relevant to determining the distance between the two sub-laser beams is the direct distance L between the sub-laser beams 260 and 262, which is measured between the beam axes of the first and second sub-laser beams.

[0152] Figure 2B Shown in top view Figure 2A In this case, two sub-laser beams 260 and 262 are introduced into the workpiece 1 at intervals and side by side along the separation line 10. The curvature of the surface 12 of the workpiece 1 is not visible from the bird's-eye view, therefore a direct geometric connection of the two beam axes of the sub-laser beams 260 and 262 is suitable for determining the distance L between the two sub-laser beams 260 and 262.

[0153] Figure 2C Another side view of the method is shown; in particular, it shows the ability to introduce multiple sub-laser beams 26 into the workpiece 1. All sub-laser beams 26 are focused into the same focal plane. This means that all focal areas lie in a single plane. In this case, the focal plane is given by the surface 12 of the workpiece 1.

[0154] Figure 3 The different sub-beams and their intensities in focal region 24, depending on the number of sub-laser beams 26 used, are shown. In the top row, the incident laser beam (not shown) is split into three sub-laser beams 260, 261, and 262. The intensities of these three sub-laser beams in focal region 24 are very high, as indicated by dark shading. For example, the three sub-laser beams can be introduced into the workpiece on their first pass along the separation line 10 (e.g., in the positive x-direction).

[0155] In the second row, the same incident laser beam (not shown) is split into four sub-laser beams. These four sub-laser beams have reduced intensity in the focal region, as indicated by mid-tone shading. Specifically, the incident laser beam used to generate the sub-laser beams remains unchanged; the intensity is adjusted by varying the number of sub-laser beams. For example, these four sub-laser beams can be used along the separation line 10 in the negative x-direction on the second pass.

[0156] The third row shows five sub-laser beams, all generated by the same incident laser beam. These five sub-laser beams have significantly lower intensities than the four sub-laser beams and also significantly lower intensities than the three sub-laser beams. For example, the five sub-laser beams can be introduced into the workpiece again on the third pass along the separation line 10 in the positive x-direction.

[0157] Therefore, the intensity of each sub-laser beam can be adjusted simply by changing the number of sub-laser beams, without having to change the incident laser beam. Consequently, the intensity introduced into workpiece 1 by each laser pulse is also independent of the number of sub-laser beams.

[0158] In particular, it is possible to use a large number of sub-beams during the first pass along the separation line 10 and reduce the number of sub-beams during further passes along the separation line 10, thereby increasing the laser energy of each sub-beam. In this sense, the first pass is performed according to the third step, while the last pass is performed according to the first step.

[0159] In particular, as the number of sub-laser beams increases, the sub-laser beams extend over increasingly larger distances D, thus requiring smaller feed motions to sweep across the entire separation line 10. Therefore, if the introduced laser energy is sufficient for material ablation, the process speed can also be increased.

[0160] To achieve material separation of workpiece 1, a total of 500 to 2500 single passes are required on the separation line 10. Here, the laser power of each sub-laser beam 26 can be adjusted at each pass, either after scanning the material of workpiece 1 to a predetermined ablation depth AT, or after a predetermined number of single passes; for example, the laser power of each sub-laser beam 26 is adjusted after 200 passes.

[0161] Figure 4A The microlens array optical unit is shown, namely the beam splitter optical unit 62 mounted for beam splitting. The input laser beam 2 of the laser 21 is guided through an optional telescopic optical unit 61 (shown in dashed lines) to adjust the diameter of the laser beam 2, and then guided through the microlens array optical unit to generate multiple sub-laser beams 26. Here, the input laser beam 2 is collimated at least in the x-direction perpendicular to the beam propagation direction (z-direction).

[0162] The input laser beam 2 passes through four microlens arrays MLA1-MLA4. Each microlens array MLA1-MLA4 has microlenses 622 arranged along the x-direction, each with an aperture a. The aperture a measured along the x-direction is identical in all microlens arrays MLA1-MLA4. Microlens arrays MLA1-MLA4 or their microlenses 622 can have all different focal lengths f1-f4. The focal lengths f1-f4 of the microlenses 622 are consistent within the respective microlens arrays MLA1-MLA4.

[0163] exist Figure 4A The 622 microlens is entirely composed of convex lenses. However, in Figure 4B In the alternative embodiment shown, the microlens 622 of MLA1 is a concave lens, while the microlenses of MLA2-MLA4 are convex lenses. This allows for the expansion of the laser beam 2, thereby attenuating the peak intensity of the laser beam 2 between the microlens arrays 622. This prevents damage to subsequent optical components. However, in the structure with four microlenses, the number of convex microlens arrays is limited to two.

[0164] Due to the microlens arrays MLA1-MLA4, each point of the corresponding microlens in the plane of the first microlens array MLA1 is imaged at an angle, i.e., an angular spectrum is generated. This angular spectrum is imaged onto the focal plane by means of the focusing optics unit 6, for example, onto the workpiece 1 or onto the workpiece. Here, the sub-laser beam 26 appears due to interference and diffraction effects in the beam path.

[0165] The microlens array optical unit 62 has an adjustment mechanism 63 by means of which at least some of the mutual spacing of the microlens arrays MLA1-MLA4 can be adjusted. For example, each microlens array MLA1-MLA4 can be provided with a dedicated carriage 632 on a guide 631 (e.g., a rack), which can be moved by a motor (e.g., by means of a driven gear engaged in the rack), wherein the carriage can be controlled by an electronic control device 633. Multiple adjustment positions of the entire microlens arrays MLA1-MLA4 are stored in the control device 633, wherein the corresponding adjustment positions include the target position to be moved relative to the z-direction for different carriages 632.

[0166] It should be observed that the positions of the several microlens arrays MLA1-MLA4 can be coupled and / or fixed. For example, the first microlens array MLA1 and the third microlens array MLA3 can be interconnected. Similarly, the second microlens array MLA2 and the fourth microlens array MLA4 can also be interconnected. Here, the number of sub-laser beams in the focal plane can be adjusted by the relative offset of the microlens arrays.

[0167] Generally, the adjustment mechanism 63 can move to multiple adjustment positions, wherein the number of sub-laser beams 26 can be adjusted by adjusting the position.

[0168] The resulting sub-laser beam 26 is focused by the focusing optical unit 64 onto or into the workpiece 1. For example, in Figure 4A and Figure 4B Five sub-laser beams 26 are generated, which are positioned side-by-side along a line in the x-direction, for example, on the separation line 10. In this case, the microlens array has microlenses arranged only in the x-direction, and none arranged in the y-direction. The laser beam has been focused in the focal plane in the y-direction, for example, by means of cylindrical lenses between the laser source and the first microlens array in the beam path.

[0169] The movable carriages 632 of the microlens array optical units can be coupled to each other. For example, the movable carriage of the first microlens array MLA1 can be coupled to the carriage of the third microlens array MLA3. This, for example, means that the distance between microlens arrays MLA1 and MLA3 is always the same. However, it is also possible that the movable carriage of the first microlens array MLA1 is coupled to the carriage of the second microlens array MLA2. It is also possible that two or more carriages are coupled to each other.

[0170] In particular, the coupled microlens arrays can be displaced relative to the uncoupled microlens arrays. For example, microlens arrays MLA1 and MLA2 can be displaced relative to microlens arrays MLA3 and MLA4. This means that the distance between microlens arrays MLA1 and MLA2, or between microlens arrays MLA3 and MLA4, remains constant, but the distance between microlens arrays MLA2 and MLA3 changes. By displacing the microlens arrays, the interference pattern of different beams in the beamsplitter optical unit changes, thereby allowing adjustment of the number of sub-laser beams 26 introduced into the workpiece. Here, the intensity in each focal region of the sub-laser beam 26 is inversely proportional to the number of sub-laser beams.

[0171] Finally, multiple sub-laser beams can be guided onto the workpiece using a scanner system or axis system, so that the sub-laser beams 26 scan across the workpiece along the separation line 10 and cut the workpiece 1 by introducing laser pulses along the separation line 10. The number of sub-laser beams 26 can be adjusted accordingly based on the ablation depth and process progress to improve the efficiency of the ablation process.

[0172] Figure 5A Another time-varying curve shows the possible motion and deflection of sub-laser beams 260, 262, and 264. For example, these sub-laser beams are observed as laser beam 2 passes through according to... Figure 4A and Figure 4BThe beam splitter optical unit generates the beam and focuses it onto or into the workpiece 1 or onto the surface 12 of the workpiece 1 by means of a focusing optical unit. In this embodiment, the sub-laser beams are evenly spaced apart from each other, that is, the distance L between two adjacent sub-laser beams 26 is always the same.

[0173] Furthermore, the x-axis of the figure extends along the separation line 10, while the y-axis represents the time axis. The first sub-laser beam 260 is focused into the workpiece 1, such that a laser pulse from, for example, an ultrashort pulse laser, causes material ablation in the focal region 24 of the first sub-laser beam 260. This can be achieved, for example, by mechanisms known as laser ablation, laser drilling, or laser cutting. Definite material ablation is achieved by each pulse or each pulse train of the pulsed laser. However, this material ablation must be carried out uniformly along the separation line 10, for example, to avoid material stress.

[0174] The following description specifically refers to the first sub-laser beam 260, but it is similarly applicable to the other sub-laser beams. The first sub-laser beam 260, starting from its original position X01, first deflects along the separation line 10, specifically by a deflection value of 300. During the deflection of the sub-laser beam 260, a pulsed laser emits pulses, for example, continuously. Thus, a total of three laser pulses are emitted along the path between the original point X01 and the point where the deflection value 300 is reached. Therefore, the first sub-laser beam 260 moves back to its original position X01, thus emitting a total of five laser pulses upon reaching the original position X01. Sub-laser beam 262 then moves against the separation line 10 until it reaches a deflection value 302. In the figure, the deflection value 302 differs from the deflection value 300, thus the movement is particularly asymmetrical. However, the deflection always occurs along the separation line 10. A total of nine laser pulses are emitted before reaching the reversal point of the deflection value 302. Next, the sub-laser beam 260 returns to its original position X01 along the separation line 10, during which four additional laser pulses are introduced into the workpiece.

[0175] The motion in the first part of the movement is performed at a first speed, in which a laser pulse is introduced to the right of the original position. The motion in the second part of the movement is performed at a second speed, in which a laser pulse is introduced to the left of the original position. In particular, these movements can also be performed using velocity curves; for example, sawtooth curves, sine curves, cosine curves, or trigonometric curves can be used. The pulse rate of the laser pulse can be adjusted here so that each sub-laser beam 26 on the separation line 10 uniformly introduces the pulse into the workpiece 1, thereby causing uniform material ablation. In particular, the movement speed can also be adapted to the deflection value, so that the same number of laser pulses are always introduced into the workpiece 1 for each deflection value.

[0176] Sub-laser beam 260, along with second sub-laser beam 262 and third sub-laser beam 264, is introduced into workpiece 1. All sub-laser beams are located at their original points X01, X02, and X03 at the initial time T0. Subsequently, all sub-laser beams shift by a deflection value 300 along the separation line 10, then return to their original positions, then shift by an offset value 302 against the separation line 10, and finally return to their original positions. Throughout this process, the distance L of the sub-laser beams remains constant. Here, the deflection values ​​300 and 302 along and against the separation line can be selected such that their sum exactly corresponds to the distance of the sub-laser beams. This ensures that adjacent sub-laser beams sweep across the same point on the separation line 10. For example, when the second sub-laser beam 262 deflects by a maximum deflection value 302 against the separation line 10, it sweeps across the same point on the separation line 10 reached by the first sub-laser beam 260 when it deflects by a maximum deflection value 300 along the separation line 10. This similarly applies to the two sub-laser beams 262 and 264. Since adjacent sub-laser beams sweep across the same point on the separation line 10, uniform material ablation is ensured. In particular, this ensures gapless material ablation, so that no workpiece with the original thickness of workpiece 1 is retained along the separation line 10.

[0177] Figure 5B The image shows laser pulses from different sub-laser beams 260, 262, and 264 introduced onto workpiece 1 as seen from a bird's-eye view. It is clearly visible that there are no points along the separation line 10 that have not been swept by the sub-laser beams, thus ensuring uniform material ablation.

[0178] If adjacent sub-laser beams have the same polarization, then adjacent sub-laser beams 260, 262, and 264 cannot be introduced into workpiece 1 at arbitrarily small distances from each other. This is due to an interference effect, through which adjacent sub-laser beams partially cancel each other out (destructive interference) or enhance each other (constructive interference). Conversely, if adjacent sub-laser beams have different, especially orthogonal, polarizations, then the sub-laser beams cannot interfere with each other. For this reason, laser beam 2 can be split into different polarization components, and the sub-laser beams introduced into workpiece 1 can be formed from these polarization components.

[0179] Figure 6A , Figure 6B A birefringent polarizer element 4 is shown, which can decompose the laser beam 2 into different polarization components. Figure 6A , Figure 6B The birefringent polarizer element 4 in the form of a birefringent crystal is schematically shown. Different birefringent materials can be used as the crystal components of the polarizer element 4, such as α-BBO (barium α-borate), YVO4 (yttrium vanadate), crystal quartz, etc.

[0180] Figure 6AThe birefringent polarizer element 4 is constructed in a wedge shape, meaning that the incident plane beam incident surface 40 for the input laser beam 2 and the plane beam exit surface 42 of the polarizer element 4 are oriented at a (wedge-shaped) angle to each other. The optical axis 44 of the crystal workpiece is oriented parallel to the beam incident surface 40. The wedge-shaped birefringent polarizer element is the preferred embodiment here.

[0181] A non-polarized or circularly polarized input laser beam 2, entering the birefringent polarizer element 4 perpendicular to the beam incident surface 40, is split into two mutually perpendicular sub-laser beams 26 (s-polarized and p-polarized, respectively) at the beam exit surface 42, which is tilted at an angle relative to the beam incident surface 40. Figure 5A Typically, the s-polarized laser beam 26 is identified by a dot, while the second p-polarized laser beam 26 is identified by a double-headed arrow. The first p-polarized laser beam 26 is refracted to a lesser degree than the second s-polarized laser beam 26 when it exits from the birefringent polarizer element 4, resulting in an angular offset α between the first and second sub-laser beams 26. Here, the first and second sub-laser beams 26 exit from the birefringent polarizer element 4 at the same position on the beam exit surface 42; that is, an angular offset α occurs between the two sub-laser beams 26, but there is no positional offset.

[0182] exist Figure 6B In the case of polarizer element 4 shown, the beam incident surface 40 and the beam exit surface 42 are oriented parallel to each other, and the optical axis 44 of the crystal workpiece is oriented at a 45° angle relative to the beam incident surface 40. In this case, the input beam 2 incident perpendicular to the beam incident surface 40 is split at the beam incident surface 2a into a first sub-laser beam 26O of ordinary form and a second sub-laser beam 26AO of non-ordinary form. The two sub-laser beams 26 are emitted parallel to each other at the beam exit surface 42, that is, without angular offset, but with positional offset Δx.

[0183] therefore, Figure 6A , Figure 6B The fundamental difference between the two birefringent polarizer elements 4 shown in the figure is that... Figure 5A The polarizer element 4 shown produces an angular offset α (no positional offset), while Figure 5B The polarizer element 4 shown produces a positional offset Δx (no angular offset). Two polarizer elements 4 can form components of a processing optical unit 6, which can, for example, […]. Figure 7 The structure is shown. It should be understood that the processing optical unit 6 can also have polarizer elements that produce both positional offset Δx and angular offset α, as is the case in conventional prism polarizers, which typically have two birefringent optical elements.

[0184] Figure 7The corresponding processing optical unit 6 is shown, which includes: a birefringent polarizer element 4 for decomposing the laser beam 2 into sub-laser beams that are perpendicularly polarized to each other; according to Figure 4A and Figure 4B The system includes a beam splitter optical unit 62 for generating multiple pairs of sub-laser beams 26 with different polarizations; and a focusing optical unit configured to focus the sub-laser beams 26 into a focal region 24 that extends along the separation line 10 in the focusing plane. Additionally, an optional λ / 2 plate 70 and a switchable polarization filter element 72 are mounted prior to the processing optical unit 6.

[0185] Figure 7 The processing optical unit 6 shown has a focusing optical unit 64, which is used to focus the two sub-laser beams 26 onto a focusing surface, the focusing surface being... Figure 6A and Figure 6B It is located on surface 12 of the workpiece 1 to be processed. (And...) Figure 6A and Figure 6B As shown, the focusing surface can be located above surface 12 or within workpiece 1. For example, an input laser beam 2 from a pulsed laser or an ultrashort pulse laser is incident on a birefringent polarizer element 4, where two sub-laser beams 26 are generated.

[0186] exist Figure 7 In the case of the processing optical unit 6 shown, the beam splitter optical unit 62 is arranged after the polarizer element 4 in the beam path. However, the beam splitter optical unit 62 can also be arranged before the polarizer element 4.

[0187] Figure 7 The processing optical unit 6 shown is at least partially movable relative to the workpiece 1, enabling the sub-laser beam 26 to move along the separation line 10. In particular, the number of sub-laser beams can also be changed by means of the beam splitter optical unit, enabling the power adjustment of the sub-laser beams when passing through the separation line 10 multiple times.

[0188] exist Figure 7 In the case of the processing optical unit 6 shown, the birefringent polarizer element 4 is arranged in a plane optically conjugate with respect to the focusing plane. The optically conjugate plane is associated with the focusing plane by means of an angle-to-position transformation (Fourier transform) generated by the focusing optical unit 7. Angles in the optically conjugate plane correspond to positions in the focusing plane, and vice versa. The two sub-laser beams 26 emitted from the polarizer element 4 with an angular offset α are thus focused in the focusing plane with a position offset Δx between the two centers of the focal region 24. Here, the position offset is less than the diameter of the focal region 24, so that the two focal regions 24 overlap each other. Since there is no position offset between the two sub-laser beams 26 emitted from the polarizer element 4, the two sub-laser beams 26 are oriented parallel to and perpendicular to the focusing plane 8 after passing through the focusing optical unit 64.

[0189] The focal regions 24 of a pair of adjacent, differently polarized sub-laser beams are formed by corresponding sub-laser beams 26 that are polarized perpendicularly to each other, and these sub-laser beams partially overlap. Furthermore, the focal regions 24 of two corresponding directly adjacent pairs of perpendicularly polarized sub-laser beams 26 also overlap. Figure 7 Of all the focal zones 24 shown, only the sub-laser beams 26 that are perpendicularly polarized to each other overlap, so that no interference effect occurs between the sub-laser beams 26 with the same polarization. The substantially linear intensity distribution produced in this way in the focal plane extends along the separation line 10.

[0190] If only one polarization direction is desired in the arrangement of the sub-laser beams in the workpiece, this can be achieved, for example, by varying the λ / 2 plate 70 and the polarizing filter 72. For instance, the λ / 2 plate 70 can be used to determine the polarization direction of the laser beam 2 incident on the polarizer element 4. For example, a p-polarized or s-polarized beam can thus be incident on the polarizer element 4 without splitting into multiple polarizations after the polarizer element 4. However, it is also possible that the incident laser beam is rotated by the λ / 2 plate 70 so that the two polarization components have the same strength after the polarizer element 4.

[0191] To filter out potentially unwanted polarization directions after the processing optical unit 6, a polarizer 72 can be introduced after the processing optical unit 6. For example, the polarizer can be switchable. For example, this could mean that the polarization of the sub-laser beam 26 filters out the first polarization direction but allows the second polarization to pass through. However, it could also be exactly the opposite. In particular, the polarizer 72 could be removed from the beam path so that both polarization components can be guided to the workpiece unimpeded. However, it could also mean that the polarizer has no effect on the sub-beam 26.

[0192] With similar Figure 7 In this way, the intensity profile of alternating polarization can also be determined by... Figure 6B The sub-laser beams are generated by a birefringent polarizing element 4, in which the sub-laser beams are positioned offset away from the birefringent polarizing element 4. However, since an angular offset is required in the conjugate plane for focusing the sub-laser beams 26 parallel to each other onto the workpiece 1, at least one additional lens assembly is required to correct the beam angle (not shown) of the sub-laser beams 26 onto the workpiece 1.

[0193] Figure 8 The diagram schematically illustrates a method in which adjacent sub-laser beams have different, and particularly orthogonal, polarization directions. The laser beam 2 of the pulsed laser is split into multiple sub-laser beams 26 by means of a processing optical unit, for example, as... Figure 6A and Figure 6BAs shown, adjacent sub-laser beams have orthogonal polarizations, and the distance between adjacent sub-laser beams can be, for example, smaller than the diameter of the focal region 24. This causes adjacent focal regions 24 of sub-laser beams 26 with different polarizations to overlap. However, due to the different polarizations, there is no interference effect. However, it is also possible for different sub-laser beams 26 to have a spacing significantly larger than the diameter of the focal region 24. In principle, the distance L between sub-laser beams with different polarizations can be set by using polarization elements. Conversely, the number of sub-laser beams with the same polarization and the distance between sub-laser beams 26 can be adjusted by means of the beam splitter optical unit 62.

[0194] Using sub-laser beams 26 with different polarizations is advantageous because different polarization directions relative to the separation line 10 affect the ablation depth and the quality of the ablation edge for each pulse. For example, polarization parallel to the feed direction is suitable for producing particularly smooth edges along the separation line. Conversely, polarization perpendicular to the feed direction enables particularly efficient material ablation for each laser pulse. By filling the space between two identically polarized sub-laser beams with a sub-laser beam of a different polarization, the efficiency and quality improvements of different polarizations can thus be achieved through different processing strategies.

[0195] Figure 9A , Figure 9B and Figure 9C Different processing strategies and variations of the method are illustrated. Figure 9A In the first step, multiple sub-laser beams 26 are introduced into the workpiece 1 along the separation line 10. Adjacent sub-laser beams 26 have different polarizations. In particular, adjacent sub-laser beams 26 are introduced into the workpiece 1 at a small distance from each other. The sub-laser beams 26 are deflected or periodically deflected along the separation line 10 such that adjacent sub-laser beams 26 sweep across the same position on the separation line 10. However, it is also possible that the deflection is smaller than the distance between adjacent sub-laser beams.

[0196] By such a small distance, the intensity distribution of the sub-laser beam along the separation line 10 is very uniform, thereby achieving uniform material ablation along the separation line 10. In the second step, the polarization component parallel to the separation line 10 can then be turned off, for example, by changing or filtering one of the optical elements in the processing optical unit 6, so that only the sub-laser beam 26 with polarization parallel to the separation line 10 is introduced into the workpiece 1. By introducing only the sub-laser beam with polarization parallel to the separation line 10 into the workpiece 1 in the second step, the ablation depth can be increased efficiently.

[0197] Figure 9BA variation of the method is shown, in which, in the first step, only a sub-laser beam 26 with polarization orthogonal to the separation line is used. This orthogonal orientation generates high fracture resistance at the cutting edge in the first step. In the second step, subsequently, only a sub-laser beam with polarization parallel to the separation line 10 is used, thereby enabling efficient separation of the workpiece.

[0198] Figure 9C Another variation of the method is shown, in which a similar approach is used in the first step. Figure 9A The sub-laser beams are introduced into workpiece 1 in a manner in which adjacent sub-laser beams are orthogonally polarized. In the second step, high fracture resistance at the separation edge can then be generated by means of sub-laser beams polarized parallel to the separation line 10, wherein, in the third step, only sub-laser beams polarized perpendicular to the separation line 10 are used to rapidly and efficiently increase the ablation depth. Overall, through Figure 9A , Figure 9B , Figure 9C Different methods ensure high quality of the cut edges.

[0199] Figure 10 The erosion efficiency of the method for different pulse lengths and fluxes is shown. For all the pulse lengths shown, it is evident that the erosion efficiency, measured in kilojoules per cubic millimeter, increases with increasing flux, measured in joules per square centimeter. This means that the greater the intensity of the sub-laser beam, the more material is released from the entire workpiece composite per kilojoule. However, this value stagnates above a flux of approximately 1.5 J / cm², making further increases in the flux or intensity of the sub-laser beam meaningless. However, to further improve the erosion efficiency, it is therefore meaningful to split the available energy of the input laser beam into multiple sub-beams. The erosion efficiency then increases with the number of sub-laser beams.

[0200] In particular, Figure 10 The results show that shorter pulse durations have significantly higher erosion efficiency, which is why pulse durations of less than 1.2 ps are preferred.

[0201] Figure 10 The data is displayed in different ways. Figure 11 In this diagram, the maximum erosion efficiency is plotted relative to the pulse length. It is also evident that for short pulse lengths less than 1.2 ps, the maximum erosion efficiency is very high, specifically in the range of 2.5-3 mm³ / kJ, while for longer pulse lengths, the erosion efficiency decreases.

[0202] Figure 12The maximum ablation rate based on the repetition rate of the pulsed laser is shown. For this purpose, the number and repetition rate of the sub-laser beams are varied uniformly. For example, using only one sub-laser beam results in a low ablation efficiency of 1000 kHz. Using two sub-laser beams at only 500 kHz improves the efficiency. Further efficiency improvements are achieved using three sub-laser beams (each at only 330 kHz). Efficiency is further improved when four sub-laser beams are introduced into the workpiece at a repetition rate of 250 kHz. Efficiency is further improved when eight sub-laser beams are introduced into the workpiece at a repetition rate of 125 kHz. In particular, with short pulse lengths of less than 1 ps, even when depositing the same total energy in the workpiece, the efficiency of this method with eight sub-laser beams is four times that of the method with only one sub-laser beam.

[0203] Figure 13 Multiple samples processed with different pulse lengths and feed rates are shown. It is clearly visible that significantly stronger material ablation occurs with shorter pulse lengths. It is also evident that the sub-laser beam produces continuous ablation geometries at low feed rates, while only a single "drill hole" is observed at high feed rates.

[0204] Within the scope of application, all individual features presented in the embodiments can be combined and / or interchanged with each other without departing from the scope of the invention.

[0205] List of reference numerals

[0206] 1. Workpiece

[0207] 10 Separation Line

[0208] 12 Surface

[0209] 2. Laser beam

[0210] 20 laser pulses

[0211] 21 Lasers

[0212] 24 Coke Area

[0213] 26 sub-laser beams

[0214] 260 First Sub-laser Beam

[0215] 262 Second Sub-laser Beam

[0216] 264 Third Sub-Laser Beam

[0217] 280 First polarization

[0218] 282 Second Polarization

[0219] 3. Shift

[0220] 30 deflection

[0221] 300 deflection along the separation line

[0222] 302 Deflection against the separation line

[0223] 4. Birefringent polarizer element

[0224] 40 beam incident surface

[0225] 42 Beam exit surface

[0226] 44 optical axes

[0227] 6. Processing optical units

[0228] 61 Telescopic device

[0229] 62 beam splitter optical units

[0230] 622 microlenses

[0231] 63 Adjustment mechanism

[0232] 64 focusing optical units

[0233] 70 Delay board

[0234] 72 Switchable polarizing filters

[0235] L distance

[0236] MLA microlens array

[0237] Span between D sub-laser beams

Claims

1. A method for separating a workpiece (1) along a separation line (10) by means of a laser pulse (20) of a laser beam (2), in, The laser beam (2) is split into multiple sub-laser beams (26) by means of a beam splitter optical unit (62), and the sub-laser beams (26) are focused onto the surface (12) and / or volume of the workpiece (1) by means of a focusing optical unit (64), such that the sub-laser beams (26) are arranged side by side and spaced apart from each other along the separation line (10). Its features are, The laser pulses (20) of the sub-laser beams (26) are introduced into the workpiece along the separation line (10) to cause material ablation, and the laser power of each sub-laser beam (26) is adjusted according to the ablation depth achieved in the workpiece (1).

2. The method according to claim 1, characterized in that, The laser power of each sub-laser beam (26) is adjusted by adjusting the number of sub-laser beams (26) given the input power of the laser beam (2).

3. The method according to claim 1 or 2, characterized in that, The laser power of each sub-laser beam (26) increases with the increased ablation depth.

4. The method according to claim 1, characterized in that, The beam splitter optical unit (62) consists of multiple microlens arrays.

5. The method according to claim 4, characterized in that, The beam splitter optical unit (62) consists of at least two microlens array pairs, each consisting of two microlens arrays, wherein the microlens arrays of the microlens array pair have a fixed distance from each other.

6. The method according to claim 5, characterized in that, The laser power of the sub-laser beam (26) is adjusted by shifting the two microlens arrays relative to each other.

7. The method according to claim 4, characterized in that, Microlens arrays or pairs of microlens arrays include convex microlenses.

8. The method according to claim 1 or 2, characterized in that, The sub-laser beams (26) are focused into the same focal plane.

9. The method according to claim 1 or 2, characterized in that, The sub-laser beam (26) moves away from its original position by a deflection value along the separation line (10).

10. The method according to claim 9, characterized in that, The sub-laser beam (26) moves away from its original position along the separation line (10) by a deflection value less than or equal to the distance between two adjacent sub-laser beams (26), and the sub-laser beam then moves back to its original position along the separation line (10).

11. The method according to claim 9, characterized in that, The separation line (10) is a straight line, and the motion of the sub-laser beam (26) is a linear deflection.

12. The method according to claim 10, characterized in that, The motion is a periodic motion, which is generated by the periodic motion of the focusing optical unit (64).

13. The method according to claim 1 or 2, characterized in that, The laser beam (2) is an ultrashort pulse laser.

14. The method according to claim 1 or 2, characterized in that, The sub-laser beam (26) and the workpiece (1) are capable of shifting relative to each other.

15. The method according to claim 1 or 2, characterized in that, The laser beam (2) and / or the sub-laser beam (26) are decomposed into two mutually orthogonal polarization components as they pass through the birefringent polarizer element (4).

16. The method according to claim 15, characterized in that, The emitted sub-laser beams (26), decomposed by the birefringent polarizer element (4) according to the polarization components, have angular and / or positional offsets from each other.

17. The method according to claim 16, characterized in that, The focused sub-laser beams (26) have alternating polarizations along the separation line (10), or the polarization of each sub-laser beam (26) is oriented along or orthogonal to the separation line (10).

18. The method according to claim 16 or 17, characterized in that, The entire separation process is divided into different sub-separation processes, each of which is performed using a p-polarized or s-polarized sub-laser beam (26).

19. The method according to claim 1 or 2, wherein, The workpiece (1) is a wafer.

20. The method according to claim 2, characterized in that, The number of sub-laser beams (26) decreases with increasing ablation depth.

21. The method according to claim 4, characterized in that, The beam splitter optical unit (62) consists of four microlens arrays.

22. The method according to claim 5, characterized in that, The laser power of each sub-laser beam (26) is adjusted by adjusting the number of sub-laser beams (26) given the input power of the laser beam (2), and the number of sub-laser beams (26) is adjusted by shifting the two microlens array pairs relative to each other.

23. The method according to claim 9, characterized in that, The motion is related to the repetition rate of the laser to ensure uniform material ablation.

24. The method according to claim 10, characterized in that, The deflection value is equal to the distance between two adjacent sub-laser beams (26).

25. The method according to claim 14, characterized in that, The sub-laser beam (26) and the workpiece (1) can be moved relative to each other along the feed direction by a feed.

26. The method according to claim 19, wherein, The workpiece (1) is a silicon wafer.

Images