METHOD FOR SINGLETING SEMOCONDUCT DIES FROM SILICON CARBIDE SUBSTRATES

Abstract

Implementations of a method for singulating silicon carbide may include providing a silicon carbide substrate (2, 30) of a thickness; irradiating the silicon carbide substrate (2, 30) with a laser beam (4) in a plurality of X-direction die roads to a focal point at a depth in the thickness in a predetermined number of X passes; and irradiating the silicon carbide substrate (2, 30) in a Y-direction (90) with the laser beam (4) in a Y-direction to a focal point at a depth in the thickness in a predetermined number of Y passes. The method may also include fracturing (50) the silicon carbide substrate (2, 30) first in the Y-direction (90) and then in the X-direction (88) using an anvil (54). and include extending a band (32) connected to the silicon carbide substrate (2, 30) to separate a plurality of dies from the silicon carbide substrate (2, 30).

Description

BACKGROUND 1. Technical field

[0001] Aspects of this document generally concern systems and methods for singulating semiconductor dies from semiconductor substrates. More specific implementations relate to the singulation of semiconductor dies from silicon carbide substrates. 2. State of the art

[0002] Semiconductor substrates are used for the purpose of fabricating various semiconductor devices. Many different types of semiconductor devices have been developed, including transistors, diodes, rectifiers, and the like. SUMMARY

[0003] Implementations of a silicon carbide singulation process may include providing a silicon carbide substrate of a certain thickness; and in a plurality of X-direction die paths: irradiating the silicon carbide substrate in one X-direction with a laser beam focused on a first focal point at a first distance to the specified thickness in a first X-pass; irradiating the silicon carbide substrate in the X-direction with the laser beam focused on a second focal point at a second distance to the specified thickness in a second X-pass; irradiating the silicon carbide substrate in the X-direction with the laser beam focused on a third focal point at a third distance to the specified thickness in a third X-pass;and irradiating the silicon carbide substrate in the X-direction with the laser beam, focused on a fourth focal point at a fourth distance, to the thickness in a fourth X-pass. The process can also include, in a variety of Y-direction die passes: irradiating the silicon carbide substrate in a Y-direction with the laser beam, focused on a first focal point at a first distance, to the thickness in a first Y-pass; irradiating the silicon carbide substrate in a Y-direction with the laser beam, focused on a second focal point at a second distance, to the thickness in a second Y-pass; irradiating the silicon carbide substrate in a Y-direction with the laser beam, focused on a third focal point at a third distance, to the thickness in a third Y-pass;Irradiation of the silicon carbide substrate in a Y direction with the laser beam, focused on a fourth focal point at a fourth distance, to the thickness in a fourth Y pass; irradiation of the silicon carbide substrate in a Y direction with the laser beam, focused on a fifth focal point at a fifth distance, to the thickness in a fifth Y pass. The method may include fracturing the silicon carbide substrate in the X direction and in the Y direction along the plurality of X-direction die roads and the plurality of Y-direction die roads, respectively, using an anvil; and stretching a band associated with the silicon carbide substrate to separate a plurality of dies from the silicon carbide substrate.

[0004] Implementations of a method for singulating silicon carbide may include one, all, or any of the following: The first distance in the first X-pass can extend further into the thickness than the second distance in the second X-pass, the second distance in the second X-pass can extend further into the thickness than the third distance in the third X-pass, and the fourth distance in the fourth X-pass can extend further into the thickness than the third distance in the third X-pass. The first distance in the first X-pass can be -26 micrometers, the second distance in the second X-pass can be -19 micrometers, the third distance in the third X-pass can be -13 micrometers, and the fourth distance in the fourth X-pass can be -14 micrometers. The first distance in the first Y-pass can extend further into the thickness than the second distance in the second Y-pass; the second distance in the second Y-pass can extend further into the thickness than the third distance in the third Y-pass; the fourth distance in the fourth Y-pass can extend further into the thickness than the third distance in the third Y-pass; and the fourth distance in the fourth Y-pass can extend further into the thickness than the fifth distance in the fifth Y-pass. The first distance in the first Y-pass can be -26 micrometers, the second distance in the second Y-pass can be -21 micrometers, the third distance in the third Y-pass can be -13 micrometers, the fourth distance in the fourth Y-pass can be -17 micrometers, and the fifth distance in the fifth Y-pass can be -14 micrometers.

[0005] The sampling rate used in the first Y pass, second Y pass, fourth Y pass and fifth Y pass can be 510 mm / second, and a sampling rate used in the third Y pass can be 150 mm / second.

[0006] The sampling rate used in the first X-pass, second X-pass and fourth X-pass can be 525 mm / second, and the sampling rate used in the third X-pass can be 150 mm / second.

[0007] The laser power used in the first and fourth X-passes can be 0.18 W; the laser power used in the second X-pass can be 0.12 W; the laser power used in the third X-pass can be 0.04 W; the laser power used in the first, second, fourth, and fifth Y-passes can be 0.23 W; and the laser power used in the third Y-pass can be 0.04 W.

[0008] Implementations of a silicon carbide singulation process can include providing a silicon carbide substrate of a certain thickness; and, in a plurality of X-direction die paths, irradiating the silicon carbide substrate in an X-direction with a laser beam focused on a focal point at a distance to the thickness in four X-passes; and, in a plurality of Y-direction die paths, irradiating the silicon carbide substrate in a Y-direction with the laser beam focused on a focal point at a distance to the thickness in five Y-passes. The process can also include fracturing the silicon carbide substrate first in the Y-direction and then in the X-direction along the plurality of X-direction die paths, respectively.the multiple Y-direction die roads using an anvil with a predetermined overrun height, an anvil spacing of 0.39 mm and a chopper lowering speed of 20 mm / second; and extending a band connected to the silicon carbide substrate to separate multiple dies from the silicon carbide substrate at a temperature of 60 °C.

[0009] Implementations of a method for singulating silicon carbide may include one, all, or any of the following: If the thickness of the silicon carbide substrate is approximately 100 micrometers, the predetermined clearance height can be 1.23 mm for the X-direction die roads and 1.21 mm for the Y-direction die roads. If the thickness of the silicon carbide substrate is approximately 200 micrometers, the predetermined clearance height can be 1.14 mm for the X-direction die roads and 1.12 mm for the Y-direction die roads.

[0010] The expansion of the belt can further include expansion with an expansion height of 8 mm, an expansion rate of 10 mm / second and a holding time of 30 seconds.

[0011] Implementations of a method for singulating silicon carbide can include providing a silicon carbide substrate of a thickness; and in a plurality of X-direction die roads, irradiating the silicon carbide substrate in an X-direction with a laser beam focused on a focal point at a depth in the thickness in a predetermined number of X-passes, each X-pass of the predetermined number of X-passes having a different laser spot diameter; and in a plurality of Y-direction die roads, irradiating the silicon carbide substrate in a Y-direction with the laser beam focused on a focal point at a depth in the thickness in a predetermined number of Y-passes, each Y-pass of the predetermined number of Y-passes having a different laser spot diameter.The process may include fracturing the silicon carbide substrate first in the Y direction and then in the X direction along the plurality of X-direction die roads or the plurality of Y-direction die roads using an anvil; and stretching a band associated with the silicon carbide substrate to separate a plurality of dies from the silicon carbide substrate.

[0012] Implementations of a method for singulating silicon carbide may include one, all, or any of the following: In the X direction, a first laser spot diameter of a first X pass can be larger than a second laser spot diameter of a second X pass, and a third laser spot diameter of a third X pass can be smaller than a fourth laser spot diameter of a fourth X pass. In the Y direction, a first laser spot diameter of a first Y pass can be larger than a second laser spot diameter of a second Y pass, a third laser spot diameter of a third Y pass can be smaller than a fourth laser spot diameter of a fourth Y pass, and a fifth laser spot diameter of a fifth Y pass can be smaller than the fourth laser spot diameter of the fourth Y pass.

[0013] The first depth of a first X-pass can be -26 micrometers, a second depth of a second X-pass can be -19 micrometers, a third depth of a third X-pass can be -13 micrometers, and a fourth depth of a fourth X-pass can be 14 micrometers.

[0014] The first depth of a first Y-pass can be -26 micrometers, a second depth of a second Y-pass can be -21 micrometers, a third depth of a third Y-pass can be -13 micrometers, a fourth depth of a fourth Y-pass can be -17 micrometers, and a fifth depth of a fifth Y-pass can be -14 micrometers.

[0015] The fourth laser spot diameter of the fourth X-pass can generate a modified region in sections of the multitude of X-direction die roads that are not covered by a structure.

[0016] The fourth laser spot diameter of the fourth X-pass can burn or melt at least a section of a structure present in sections of the multitude of X-direction die roads.

[0017] The first laser spot diameter, second laser spot diameter, and third spot diameter can generate a modified region in sections of the multitude of X-direction die roads covered by a structure.

[0018] The foregoing and other aspects, features and advantages will be evident to professionals from the DESCRIPTION and DRAWINGS as well as from the CLAIMS. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The following describes implementations in conjunction with the attached drawings, where identical reference symbols denote similar elements and: Fig. Figure 1 is a cross-sectional diagram of an implementation of a silicon carbide substrate during irradiation with a laser beam in a die road; Fig.Figure 2 is another cross-sectional diagram of a silicon carbide substrate during pulsed laser irradiation in a die mill during the feeding of the silicon carbide substrate; Fig. Figure 3 is a cross-sectional view of an implementation of a breaking system; Fig. Figure 4 shows a side view and a top view of a silicon carbide substrate before and during expansion using an expansion system; Fig. Figure 5 is a flow chart with corresponding diagrams of a silicon carbide substrate during processing with an implementation of a laser system, a fracturing system and an expansion system; Fig. Figure 6 is a cross-sectional view of a silicon carbide substrate, showing the position in the silicon carbide substrate material for laser irradiation in different X-direction passes (X passes) and different Y-direction passes (Y passes); Fig. 7 is a flowchart for the implementation of a process for singulating silicon carbide; Fig. Figure 8 is a diagram of die specimens used in three-point bending tests and a representation of a three-point bending test system; Fig. Figure 9 is a cross-sectional view of an implementation of a crushing system and a detailed view of the tip of an implementation of an anvil during operation; Fig. Figure 10 illustrates a sequence of processing steps for a silicon carbide substrate during an expansion process; Fig. 11 is a top-view micrograph of an X-axis die road after laser cutting; Fig. Figure 12 is a diagram of a die road during lasering through a structure in the die road; Fig.Figure 13 is a diagram showing the change in laser spot diameter corresponding to a change in the depth of a focal point in a thickness of a silicon carbide substrate; and Fig. 14 a top-view micrograph of a structure in a die street before lasering; Fig. 15 is a top-view micrograph of the structure of Fig. 14 after a first round; Fig. 16 is a top-view micrograph of the structure of Fig. 15 after a second round; Fig. 17 is a top-view micrograph of the structure of Fig. 16 after a third round; Fig. Figure 18 is a micrograph of a side wall of a silicon carbide substrate, showing modified regions where no structure is present and under a structure in the die street; Fig.Figure 19 is a diagram of the effect of the laser spot diameter on a structure in a die street during three passes; and Fig. Figure 20 is a diagram of the effect of the laser spot diameter on a structure in a die street during four passes. DESCRIPTION

[0020] This disclosure, its aspects and implementations are not limited to the specific components, assembly procedures or process elements disclosed herein. Many additional components, assembly procedures and / or process elements known in the field and compatible with the intended methods for singulating semiconductor substrates are apparent from this disclosure for use in certain implementations.Accordingly, for example, although certain implementations are disclosed, such implementations and implementing components may include any shapes, sizes, designs, types, models, versions, dimensions, concentrations, materials, quantities, process elements, steps and / or the like as are known in the field for such a process for processing semiconductor substrates and implementing components and processes that are compatible with the intended operation and procedures.

[0021] The various methods for singulating semiconductor substrates disclosed herein employ focused laser irradiation to create a damaged / modified region within the semiconductor substrate, followed by fracturing the semiconductor substrate along the modified region and separating a multitude of dies from the semiconductor substrate via a ribbon expansion process. This overall process is referred to as "stealth dicing." The stealth dicing method utilizes a laser system, a fracturing system, and an expansion system in combination with a substrate assembly system. Although stealth dicing works in theory, the ability to apply the method to accurately and repeatably singulate dies from semiconductor substrates for integration into semiconductor packages capable of passing reliability tests requires significant experimentation, which is dependent on the semiconductor substrate material.The dependence on the semiconductor substrate material is also a function of the specifications of the respective semiconductor substrate material, which, as non-limiting examples, may include semiconductor material type, crystallographic orientation, crystal plane orientation to the surface, doping concentration, doping type, number of crystal imperfections / defects, type of crystal imperfections / defects, orientation of crystal imperfections / defects, thickness of the semiconductor substrate, size of the semiconductor substrate, die-street orientation (X or Y), and many other attributes / parameters of a semiconductor substrate material.

[0022] For this reason, attempting to apply stealth dicing parameters used for one semiconductor substrate type to a stealth dicing process for a different semiconductor substrate type, or even for a different thickness of the same semiconductor substrate type, does not yield predictable results. Therefore, the significant experiments detailed in this document focused on developing a stealth dicing process specific to a particular semiconductor substrate material—in this case, silicon carbide. The experimental results presented in this document were unpredictable and unexpected. Due to the extreme hardness of silicon carbide, dicing the semiconductor substrate by sawing using diamond-coated / impregnated saw blade technology is slow and difficult.The ability to utilize stealth dicing to produce dies from a silicon carbide substrate that can be incorporated into packages and pass reliability tests can be very valuable. Such a process can increase the number of wafers and units per hour that can be processed in a packaging / assembly process. This method can also enable a reduction in die size, as the die lane width can be reduced because the kerf width of a given saw blade no longer needs to be considered when determining the die lane width.

[0023] The silicon carbide substrates described in the examples disclosed herein are of the N-type, 4H-polytype, with a crystal orientation of 4 degrees off-axis. The dislocation density of the silicon carbide substrates is approximately 5 × 10⁻⁶ 3 cm 2 with a micropipe density of less than 0.1 cm 2The principles revealed herein could also be applied to silicon carbide substrates with different dislocation densities and micropipe densities.

[0024] It will be on Fig.Figure 1 refers to a cross-sectional diagram of an implementation of a silicon carbide substrate 2 during irradiation with a laser beam 4 in a die using a laser system. As shown, a lens 6 (or lens group) is used to focus the laser irradiation 4 onto / into the silicon carbide substrate 2. As shown, the focused laser beam 4 is refracted at an angle 8 upon entering the silicon carbide substrate 4, which is determined by the material and the specific wavelength / energy of the laser beam 4 irradiation. The combination of the focusing applied by the lens 6 and the angle of refraction 8 determines the depth / position in the silicon carbide substrate 2 where the focal point 10 of maximum laser beam 4 energy is located. Fig.1. The laser beam 4 travels in a direction perpendicular to the paper (into the paper and out of the paper). At the focal point 10, the energy of the focused irradiation of the laser beam 4 modifies the material of the silicon carbide substrate to create a modified region.

[0025] Although in Fig. Since the focal point 10 is represented as a point, the absorption of the irradiation energy from the laser beam 4 actually occurs in a more linear direction (in the form of a column / cylinder) in the direction of the laser beam into the material. Because the laser beam operates in a pulsed mode rather than a continuous wave mode, when the silicon carbide substrate 2 is fed / scanned under the laser beam 4 at a fixed velocity, a structure of modified regions 12 can be observed in cross-section, corresponding to each focused pulse of the laser beam 4 irradiation, as shown in the diagram of Fig. 2 shown. Depending on the overlap of each pulse with each other pulse, which is set by the pulse repetition rate and / or feed rate of the silicon carbide substrate 2, the modified regions 12 in the silicon carbide substrate 2 material can be separated by unmodified material (as shown in Fig. 2 shown) or can merge into one another to form a continuous / essentially continuous modified region. For this reason, the terms “modified region” and “modified layer” are used synonymously in this document.

[0026] The depth of the focal point 10 into the silicon carbide substrate 2 can be adjusted using the lens 6 and / or by changing the physical distance between the lens 6 and the upper surface 13 of the silicon carbide substrate 2. If multiple passes of the laser beam over the silicon carbide substrate 2 are used, the depth of each pass can be independently adjusted to be the same as, deeper into, or closer to the upper surface 13 of the silicon carbide substrate 2 than the previous pass. Here, the term "upper surface" 13 refers to the surface of the silicon carbide substrate facing the laser beam 4.In various process implementations, the upper surface could be either the side of the silicon carbide substrate containing electrically active devices of the silicon carbide substrate (active side), or the opposite surface of the silicon carbide substrate (back side).

[0027] In the various process implementations disclosed herein, two further main processes are also employed to achieve the separation of the various dies from the silicon carbide substrates, fracturing, and expansion. Referring to Fig.Figure 3 illustrates an implementation of a crushing system 14. Here, the crushing system 14 includes a chopper 16, which is shown above the substrate 18, on which a cover strip 20 and a mounting strip 22 have been attached on their respective sides. The chopper 16 is positioned equidistantly between sections of the anvil 24, which are separated from each other on each side by the anvil spacing 28, in order to generate a bending moment in the substrate 18 when the chopper is pressed downwards against the mounting strip 22. Determining the distance the chopper 16 should travel downwards against the assembly belt 22 during operation to produce a clean and repeatable fracturing of the silicon carbide substrate at the modified region(s) / layer(s) in the die road is the result of several calibrations and calculations that take into account the thickness of the cover belt 20, the thickness of the assembly belt 22 and the thickness of the silicon carbide substrate 18.

[0028] In one particular process implementation, an absolute chopper height is calibrated by simply placing masking tape over the anvil and lowering the chopper until the masking tape reaches a point where it can no longer be pulled out from under the chopper. In one implementation, where the chopper is 91.34 mm long / high, the absolute chopper height is 91.378 mm, with a masking tape thickness of 0.038 mm. In various process implementations, a chopper travel height is used to describe the distance the chopper travels from a zero point of the drive motor to the surface of the silicon carbide substrate (which would pass through the thickness of the assembly belt, if present).To account for the thickness of the mounting tape, the masking tape, and the substrate thickness for a given absolute chopper clearance height, a parameter is calculated that is called the wafer-related relative height and has been varied in the experiments disclosed in this document.

[0029] In a specific implementation with the previously specified chopper height, the previously specified masking tape thickness, and for a 200-micrometer-thick silicon carbide substrate, the calculation for the wafer-related relative height is performed by adding the silicon carbide substrate thickness, the chopper override height, the mounting tape thickness, and the masking tape thickness, and then subtracting 1378 micrometers. The result, for a mounting tape thickness of 90 micrometers, a masking tape thickness of 50 micrometers, an absolute chopper override height of 1.14 mm, and an absolute chopper height of 91.14 mm, is a wafer-related relative chopper height of 102 micrometers. Referring to the larger view of the crushing system of the Fig. 3, which are shown in the cross-sectional view of the Fig.As shown in Figure 9, this wafer-related relative chopper height 29 reflects the distance the chopper at contact position 26 pushes below the original level of the silicon carbide substrate 18 below the assembly belt 22 during downward deflection over the anvil 24. In other words, the wafer-related relative chopper height 29 reflects the force that must be exerted on the assembly belt / silicon carbide substrate / cover belt stack to achieve fracture of the silicon carbide substrate at a given anvil distance. In the studies in this document, the wafer-related relative height 29 and the actual anvil distance 28 showed statistical significance when investigating the occurrence of undivided dies and the presence of lateral cracks.The study identified a process window with a wafer-related relative height of approximately 134 to 144 micrometers and an X-axis anvil distance between the chopper and the anvil side of approximately 3108 to 3260 micrometers as producing acceptable results, both with regard to undivided dies and the presence of lateral cracks. In the Y-axis direction, a wafer-related relative height of approximately 102 to 120 micrometers with a Y-axis anvil distance of approximately 2628 micrometers was identified as optimal for undivided dies (the study identified no effect on lateral cracking in the Y-axis).

[0030] After the dies are broken, since they are only separated by the actual width of the crack between them in a stealth-dicing process, the ability to remove the dies from the assembly belt without causing chipping is limited. To increase the ability to successfully pick up the dies after breaking, the assembly belt is stretched / expanded using an expansion system. (Referring to...) Fig. Figure 4 shows a mounted silicon carbide substrate 30, connected to mounting tape 32 and frame 34, in a top view and in a side cross-sectional view on the left. Here, the fracture lines 40 in the silicon carbide substrate 30 are shown as dashed lines, as they are difficult to see visually due to the very narrow width of the fractures. On the right side, the Fig.4. The silicon carbide substrate 30 is stretched using a chuck that is raised a predetermined height / distance beneath the assembly belt 32 and includes rollers 38 that assist the belt in stretching uniformly across the width of the silicon carbide substrate 30. The aim of the belt stretching process is to create sufficiently wide gaps 42 between the dies to allow the individual dies to be removed from the belt during a die pick-up process without die chipping, while avoiding excessive stretching of the belt to prevent die sagging during pick-up, which would compromise pick-up accuracy.The process variables that support the expansion process include the height to which the chuck is raised (expansion height), the temperature at which the expansion is performed, the time the belt is held in the expansion position by the chuck (holding time), and the speed at which the chuck is raised (expansion rate). After the expansion process, the assembled silicon carbide substrate is then moved to a die-picking operation, where the individualized dies can be picked up and placed either directly onto a package substrate or into a pick-up belt for subsequent use in a package assembly process.

[0031] Referring to Fig. Figure 5 illustrates a flowchart for the implementation of a stealth-dicing method for a silicon carbide substrate 44. As in Fig.As shown in Figure 5, a tape mounting process (step 46) is used to apply a mounting tape and / or a masking tape prior to stealth dicing. In some process implementations, no tape is required on the top surface of the silicon carbide substrate; in others, a (masking or mounting) tape may be applied to the top surface. Following the tape mounting process, the stealth dicing process (step 48) is performed. As disclosed later in this document, the stealth dicing process may involve multiple passes in the X-axis scribe lines / die roads (X passes) and multiple passes in the Y-axis scribe lines / die roads (Y passes) using a laser and lens. The laser and lens operate at a given wavelength and laser power.Although the term “laser power” is used in this document, since the laser can be operated in pulsed mode, “laser power” is a time-averaged calculation of the average power of the set of pulses generated by the laser, as opposed to a constant output power from a laser operating in continuous wave mode.

[0032] Following the stealth dicing process, the silicon carbide substrate is then processed using the crushing system (step 50), which comprises the chopper 52 and the anvil 54, which can be any of those disclosed in this document. As in Fig.As shown in Figure 5, in the fracturing process the downward force 56 exerted by the chopper 52 is used in combination with the corresponding moment 58 provided by the two sections of the anvil, which are separated by the anvil distance, which act in such a way as to cause the substrate to fracture along the entire line of the chopper in the die-street area of ​​the silicon carbide substrate.

[0033] After the crushing process, the assembled silicon carbide substrate is then processed by an expansion system, which serves to expand the substrate from the center outwards, as indicated by the four arrows 60 in Fig.5 is displayed. The result of the expansion system is to create sufficient spacing between the dies to allow a die-picking device to remove the various dies without die chipping or problems caused by sagging from overstretched tape. The ability to perform the expansion at a higher temperature than during the picking process allows the tape to plastically deform during the stretching process, but then regain its tensile strength for the picking process as it cools.

[0034] Referring to Fig.Figure 7 illustrates a flowchart for another implementation of a method for singulating a silicon carbide substrate 62. In this implementation, a manual belt assembly process is used (step 64), although in other implementations an automatic belt assembly device may be used. Then, any laser stealth dicing process disclosed herein is applied (step 66), followed by the application of a cover strip to the back side of the silicon carbide substrate (the side that was not facing the laser during laser cutting) (step 68). The application of the cover strip can be performed either manually or automatically using an automatic belt assembly device in various method implementations. The silicon carbide substrate is then ready for processing using a crushing system such as any one disclosed herein (three-point crushing process, step 70).

[0035] As in Fig. As shown in Figure 7, the procedure includes a verification of the silicon carbide substrate orientation (wafer orientation check, step 72) to ensure that the wafer flattening (or other orientation structures) is / are in the correct orientation when the silicon carbide substrate is loaded onto the expansion system. This procedure ensures that the die roads are correctly aligned relative to the expansion forces exerted when the chuck is lifted beneath them during the expansion process. This helps to ensure a uniform / desired spacing between the dies after the expansion process. The silicon carbide substrate then undergoes the expansion process, which may be any one disclosed in this document (step 73). In a particular implementation of the procedure, with reference to Figure 7, the process may be carried out as follows: Fig.10, a reassembly process (step 82) is applied. In various implementations, for example, but not limited to, the initial size of the silicon carbide substrate 74 is six inches in diameter, mounted using the mounting band 76 on a frame / ring 78 of ten inches in diameter. In this embodiment, after the expansion process, a second eight-inch ring 80 is applied to the mounting band 76 while the silicon carbide substrate 74 is still mounted, and the first ring 78 is then removed in a second assembly process. During this process, the now-expanded mounting band is tensioned as it is applied to the second ring 80, which can further prevent sagging / loosening during subsequent die-holding processing. The silicon carbide substrate 74 is then processed during subsequent steps while attached to the second ring 80. As in Fig.Figure 7 shows that the process steps with solid line outlines are those carried out with the first ring, and those in dashed lines are those carried out with the placed second ring.

[0036] These additional process operations can, as in Fig. Figure 7 shows an automated optical inspection (AOI, step 84) followed by a die pickup operation, which is performed either concurrently with or prior to a die attachment process in which each die is attached to a substrate during a die packaging operation (DA, step 86).

[0037] The tensile strength of the die at the die roads after stealth dicing was also measured using a three-point bending test. This three-point bending technique was applied to collect data that differs from ordinary die strength data obtained by three-point bending. In ordinary die strength data collection, a single die is subjected to three-point bending to evaluate the strength of the die after thinning and / or singulation. In the tests performed here, with reference to Fig.8. Two dies 114 are separated from a silicon carbide substrate after stealth dicing is performed between the two dies 114, but without breaking in the die road 116 between the two dies. The two dies are then placed on two supports 118 positioned remotely on either side of the die road, with the front (active side) 120 of the two dies 114 in contact with the two supports 118. Subsequently, a chopper 122 is placed against the rear (124) of the two dies 114 on the die road 116 and then pressed against the rear 124 until the two dies break on the die road 116. As in Fig.As illustrated in Figure 8, the tensile strength of die roads 126 in the Y direction was tested using two dies 130, and the tensile strength of die roads 128 in the X direction was tested using two dies 132. In one particular implementation, it was observed that the tensile strength for 100 micrometer thick silicon carbide dies was on average about 1 Newton higher on the Y-direction die roads than on the X-direction die roads.

[0038] This document discloses various process parameters for different implementations of the stealth dicing method. These are exemplary and reflect the results of sets of statistically designed experiments, including reliability tests of assembled dies, to validate that the singulation processes offer long-term stability for a desired design lifetime.

[0039] Referring to Fig.Figure 6 shows a diagram of the laser passes in an implementation of a multi-pass laser process with X-direction 88 and Y-direction 90. Here, four passes are performed in each X-die lane and five passes are performed in each Y-die lane. Fig.Figure 6 shows that in this particular process implementation, the four passes are performed in the X-die paths, with the first path reaching a first deepest distance into the silicon carbide substrate material (first depth), the second path reaching a second distance into the silicon carbide substrate material (second depth), which is not as deep as the first, the third path reaching a third distance into the silicon carbide substrate material (third depth), which is less deep than the second, and the fourth path reaching a fourth deepest distance into the silicon carbide substrate material (fourth depth), which is deeper than the third distance / depth. In other words, the first path reaches further into the thickness of the silicon carbide substrate in the X direction than the second path, the second path reaches further into the thickness than the third path, and the fourth path reaches further into the thickness than the third path.In this implementation, the fourth path is less deep than the second path. The reasons for this difference in the depth of the third path and the presence of the fourth path are discussed below.

[0040] In the Y direction, the five paths are executed as shown, with the first path reaching a first, shallowest distance into the silicon carbide substrate, the second path reaching a second, shallower distance into the silicon carbide substrate, the third path reaching a third, shallowest distance into the silicon carbide substrate, the fourth path reaching a fourth distance, shallower than the second path, and the fifth path reaching a fifth distance, shallower than the fourth path but deeper than the third path.In other words, the first distance of the first Y-pass extends further into the thickness of the silicon carbide substrate than the second distance of the second Y-pass, the second distance extends further into the thickness than the third distance of the third Y-pass, the fourth distance of the fourth Y-pass extends further into the thickness than the third distance, and the fourth distance extends further into the thickness than the fifth distance of the fifth Y-pass. These same paths, with these relative distances and sequences, can be used for both 100-micrometer-thick and 200-micrometer-thick silicon carbide substrates.

[0041] The effect of multiple passes is to create modified regions / layers within the thickness of the silicon carbide substrate. These modified regions / layers help to support crack propagation through the thickness and along the length of each X-road and Y-road of the silicon carbide substrate. Provided sufficient modified regions / layers are present, during the fracturing operation, the crack that isolates the die propagates in a controlled manner along the modified region / layers along each die road. In contrast, uncontrolled crack propagation can cause the crack to leave the die road area and penetrate into the active areas of the die surrounding the die road as lateral cracks.Uncontrolled tearing is mainly detected through chip failures after singulation, as the uncontrolled torn sections of the chip damage it to such an extent that it subsequently no longer functions properly.

[0042] The presence of modified regions / layers in the thickness of the silicon carbide substrate is difficult to detect during a visual examination under a microscope from above, and therefore the location of these regions in Fig. Figure 11, indicated by dashed lines, is drawn over the micrograph of an implementation of four dies of a silicon carbide substrate. This silicon carbide substrate is one like any disclosed in this document. Here, the dashed lines show the position of the laser-machined X-die road 92 and the laser-machined Y-die road 94 at the die junction 96. What in Fig.The structure 98 formed in the die road is visually perceptible in die road 92. Structure 98 is fabricated in various implementations from a wide variety of die stack materials, distinct from the silicon carbide / oxide / passivation materials present elsewhere in the die road, including, as non-limiting examples, metals, metal alloys, or underbump metallization materials. These materials absorb the laser light more completely than the other silicon carbide / oxide / passivation materials in the die road, thus preventing more laser light from reaching the focal point of the laser beam in the thickness. They also absorb visually perceptible radiation and therefore appear differently colored in the micrograph.

[0043] Fig.Figure 12 is a diagram illustrating the effect of structure 98 on the uniformity of the laser beam 100 as the laser beam attempts to reach the focal point 102 at the desired depth 104. As illustrated, the additional absorption of light in the laser beam 100 reduces the amount of light that can then reach the focal point 102. If the absorption of light is sufficient (as in a situation where the laser beam is emitted with sufficiently high power), the diagram shows Fig.12, how a burning / melting 106 of the material of structure 98 takes place. One of the challenges posed by the burned / melted material 106 is that it is generally now more absorbent or more reflective of the laser light than the material of structure 98 originally was. In other words, a second pass of laser light over the burned / melted material 106 will now be attenuated even more at the focal point 102 than the pass in which the burned / melted material 106 was originally formed. If the power of the laser beam in the second pass is sufficiently high to produce more burned / melted material in the second pass, then in a third pass, structure 98 will be even more absorbent of laser light than it was in the second pass.This further reduces the laser beam's ability to create the desired modified region / layer in the silicon carbide substrate beneath the structure. Fig. Figure 11 shows a burn / melt structure 109 on the material of the underlying structure. The dotted / alternating pattern in the burn / melt structure 109 reflects that the laser used for stealth dicing is a pulsed laser, as disclosed in this document.

[0044] The degree and / or speed of burning / melting is a function of the laser spot diameter. The smaller the laser spot diameter, the higher the laser power density and the faster the burning / melting occurs at the same laser power. (Referring to...) Fig.Figure 13 shows a diagram illustrating the relationship between the laser spot diameters α and β and the corresponding laser focal heights 110 and 108, respectively. Here, the laser spot diameter α is the widest, which occurs when the focal height 110 extends deepest into the silicon carbide substrate 112. The laser spot diameter β is narrower, which occurs when the focal height 108 extends less deeply into the silicon carbide substrate. For the same laser beam power, the laser power density is highest for the laser spot diameter β. Since the laser focal height changes when the laser lens(es) is moved up or down, the focal length of the lens does not change during this movement, thus maintaining the shape of the laser beams 134, 137.

[0045] Equation 1 is a relationship which expresses the laser power density (I0) in gigawatts / cm². 2as a function of the laser energy (E) in joules, the pulse duration (τ) in nanoseconds, the laser absorption rate (γ, %), the laser power (P) in gigawatts and the laser spot diameter (d) in centimeters. I0=4γEπτd2=4γPπd2

[0046] Inspection reveals that when the laser absorption rate γ increases due to changes in the material exposed to the laser (through burning / melting), the laser power density increases accordingly. Furthermore, the laser power density increases with the inverse of the square of the laser spot diameter, meaning that the strongest effect on the laser power density is the upward shift of the focal point height, which reduces the laser spot diameter. Thus, at the same laser power, the reduction in laser spot diameter causes the burning / melting process to deteriorate due to the significant increase in laser power density.

[0047] An experiment was conducted to investigate the effect of the laser spot diameter on a silicon carbide substrate 136, which contained a structural region that was visible in the micrograph of the Fig. Figure 14 illustrates this. Here, structure 138 appears as a lighter area due to its higher reflectivity for optical light. Fig. Figure 14 shows the silicon carbide substrate 136 before any laser processing. Subsequently, a first laser pass was performed with a laser power of 0.18 W at a focal height of 26 micrometers, and the resulting effect on the left edge of the structure is shown in the micrograph of the Fig. Figure 15 illustrates this. At this laser spot diameter, little or no visually perceptible effect can be observed. Then a second laser pass was performed with a laser power of 0.18 W and a focal point height of 19 micrometers. The micrograph of the Fig.Figure 16 illustrates that at this laser spot diameter, a significant visually detectable burned / melted region 140 is now present in the structure 138 of the silicon carbide substrate 136. A third laser pass was then performed with a laser power of 0.18 W and a focal spot height of 13 micrometers, and the results are shown in the micrograph of the Fig.Figure 17 illustrates this. Here, the burned / melted region 142 is now even darker, reflecting the fact that, in addition to experiencing burning / melting due to the smaller laser spot diameter, the burned / melted region 142 has increased the laser absorption rate for the wavelength of the laser beam's energy due to the change in the material's properties within the burned / melted region 142. Thus, the amount of laser light that would have been expected to reach the focal point at the focal height is reduced even more than would have been expected in the second laser pass. Therefore, the additive effect of melting / burning the structure 138 can cause one or more laser passes to fail, essentially creating a modified region / layer within the silicon carbide, thereby increasing the likelihood of uncontrolled cracking during the fracturing operation.The effect of this, viewed using micrographs or scanning electron microscopy of the sidewalls of the isolated die, is that one or more of the passes may be “missing” or appear to have failed to modify the silicon carbide substrate material, resulting in a generated modified layer.

[0048] In the system and process implementations disclosed herein, the laser parameters were optimized to understand the presence of a structure in the X-direction die roads. No corresponding structure was found in the Y-direction die roads in these experiments. However, the principles disclosed herein regarding the modification of the laser parameters to handle the structure in the X-direction die roads could be applied to help modify the laser parameters in the Y-direction die roads accordingly, although suitable experiments would need to be performed to validate the laser and refraction parameters disclosed herein.Furthermore, while it was found that using four laser passes in the X-direction minimizes yield losses due to the presence of the structure in localized areas of the X-direction die roads, a complete elimination of yield losses due to uncontrolled breakage was not achieved. This indicates the difficulty the structure causes in promoting the generation of lateral cracks and die cracks. Removing the structure would eliminate such failures but would render other essential manufacturing functions, such as electrical testing, die sorting, and other metrology operations, typically performed in the die road region, unusable.Furthermore, the need / use of protective tapes / structures, where the laser treatment stops partially or completely on the protective tape materials, would create the need to develop laser parameters that capture the structure of the protective tapes / structures.

[0049] Fig. Figure 18 is a micrograph of the side wall of a single silicon carbide die with a total thickness of about 100 micrometers after lasering using the four-pass X process disclosed in this document (see Figure 18). Fig.(See Figure 6 for a diagram). Here, over section 144, which corresponds to a segment of the X-direction die road that exhibits no structure on it, the presence of modified layers corresponding to the first, second, third, and fourth paths (arranged approximately 1 micrometer apart in depth) can be observed. However, over section 146, which lies beneath a structure not visible in this cross-sectional view, a modified layer corresponding to the fourth path is not significantly evident, and the degree of material modification for each of the three paths is reduced. Interestingly, the position of the modification layer induced by the three paths also appears to extend somewhat deeper into the thickness of the silicon carbide die 148, as the depth 150 over section 144 was approximately 27.26 micrometers, while the depth 152 over section 146 was approximately 34.63 micrometers.The observed lower degree of modification and the absence of modification from the fourth pathway indicate that crack propagation into the material beneath the structure along pathway 146 differs slightly from propagation beneath pathway 144. However, if the degree of modification beneath the structure is sufficient, propagation along pathway 146 can be reliably achieved to minimize yield losses due to uncontrolled cracking and lateral cracking. Furthermore, the use of the fourth pathway helps ensure sufficient modification occurs in those portions of the die road not located beneath the structure, thereby further stimulating the material beneath the structure to propagate the crack originating from the other silicon carbide material.The various experiments disclosed in this document were designed to help determine the otherwise unpredictable laser parameters and other parameters that will help maximize the yield despite the presence of the intermittently occurring structure in the X-direction die roads.

[0050] The impact of the in Fig. 14, Fig. 15, Fig. 16 to Fig. The observed change in the laser spot diameter of 17 is shown in the diagram of the Fig.Figure 19 illustrates a process with three passes at constant laser power from pass to pass. As shown, the first pass / path has the focal point 154, which extends to the deepest depth 156 in the silicon carbide substrate 158, thereby generating a laser spot diameter α on the structure 160. During the processing of this first path, no burning / melting of the structure 160 occurs. However, as the focal point 162 moves to depth 164, the laser spot diameter narrows to β. The resulting increase in laser power density leads to the burning / melting of the structure 160 represented by cloud 166. Since the structure 160 is now at least partially burned / melted, increased burning / melting of the structure 160 occurs when the focal point 168 shifts upwards to depth 170, thereby reducing the laser spot diameter to γ.This increase in absorption due to burning and the increase in laser power density can cause the complete loss of the modification of the crystal structure of the silicon carbide substrate 158 from a single pass / path. This diagram illustrates how using the same laser power in each pass / path while sequentially decreasing the laser spot diameter enhances the burning / melting of the material of structure 160 and reduces the formation of the modified region / layer beneath the structure.

[0051] The diagram of Fig.Figure 20 illustrates the effect of a four-path / pass process, where the first path / pass is performed at a deepest focal point 172 at a depth 174 with a laser spot diameter α. During this pass, which occurs at a first laser power level, no damage to the material of the structure 160 is observed. During a second pass / path, the laser power is reduced, and the depth 176 of the focal point 178 is also reduced, resulting in a laser spot diameter β. Due to the reduction in laser power, the laser power density resulting from the smaller laser spot diameter β does not lead to damage to the material of the structure 160. In the third path / pass, the focal point 180 is located at a further reduced depth 182, reducing the laser spot diameter to γ, and the laser power is further reduced to a lowest power level.Again, the resulting laser power density does not damage structure 160. During the fourth path / pass, the focal point 184 is moved to a depth 186, which is slightly less than the depth 182, but with the same power as in the first path with a laser spot diameter of δ. As shown, this higher laser power density than in all other passes causes the burning / melting of structure 160 represented by cloud 188, but ensures that the silicon carbide substrate 158 material not located beneath the structure receives significant exposure to modifying laser energy to support the uniform formation of a crack that can propagate into the area beneath structure 160, as previously discussed.Thus, although the burning / melting in the final pass of structure 160 affects the modification of the silicon carbide substrate material under the structure, it does not weaken the laser energy as completely as the third pass through already burned / melted material that is in . Fig. 19 is shown.

[0052] Several statistically designed experiments were conducted using 100-micrometer-thick silicon carbide substrates, such as those disclosed herein, to determine the factors influencing stealth dicing and crushing quality / capability. The results of several of these experiments are summarized in this document to reveal the ranges of operating parameters where the best results were achieved. These results are also applicable to 200-micrometer-thick silicon carbide substrates, although these were not used in the tests.

[0053] The experiments used an initial set of laser, refractive, and expansion parameters as a starting point, which was the result of significant factorial experimental design work on the various parameters. These parameters are presented in Tables 1, 2, 3, and 4 below: Table 1 contains the set of initial laser parameters used for both 100 micrometer thick and 200 micrometer thick silicon carbide substrates: Table 1 path wavelength Power (W) Focal height (µm) Scanning speed (mm / s) path wavelength Laser power (W) Focal height (µm) Scanning speed (mm / s) X0 1064 nm 0,18 -26 525 Y0 1064 nm 0,23 -26 510 X1 1064 nm 0,18 -19 525 Y1 1064 nm 0,23 -21 510 X2 1064 nm 0,18 -13 525 Y2 1064 nm 0,04 -13 150 Y3 1064 nm 0,23 -17 510 - Y4 1064 nm 0,23 -14 510 Table 2 contains the set of initial fracturing parameters for use with the fracturing system for silicon carbide substrates approximately 100 micrometers thick. The fracturing sequence includes fracturing the Y-direction roads first, followed by fracturing the X-direction roads second. Table 2 Direction Clearance height (mm) Anvil distance (ratio multiplier) Chopper lowering speed (mm / s) X 1,2 0,39 20 mm / s J 1,2 0,39 20 mm / s Table 3 contains the set of initial fracturing parameters for use with the fracturing system for silicon carbide substrates approximately 200 micrometers thick. The fracturing sequence includes fracturing the Y-direction roads first, followed by fracturing the X-direction roads second. Table 3 direction (see Fig. 6) Clearance height (mm) Anvil distance (ratio multiplier) Chopper lowering speed (mm / s) X 1,12 0,39 20 mm / s J 1,12 0,39 20 mm / s Table 4 is the set of initial expansion parameters for use with the expansion system for both approximately 100 micrometers and approximately 200 micrometers thick silicon carbide substrates. Table 4 Expansion height temperature Holding time Expansion rate 8 mm 60 °C 30 seconds 10 mm / s

[0054] A planned experiment was conducted to investigate the effect of varying the laser power in three X-direction passes on the distance between a structure-area crack line and the die polyimide ring (SL remaining) and a good structure-area die (die in the region of a structure in the die road for which no observed offset fracture was present). Ten passes were performed, and the results of the experiment showed that changing the laser power in the X-direction passes alone would not provide a statistically significant advantage for road-area crack formation.

[0055] Subsequently, twelve additional tests were performed, including increasing the laser power, decreasing the laser power of the second pass / path, adding an additional pass in the X direction, decreasing the sampling rate of the added pass, increasing the sampling rate of the added pass, adding five micrometers of laser compensation, using a refraction sequence of X first then Y, decreasing the anvil spacing, using a bottom-to-top refraction sequence, and matching the laser parameters in the X direction to those in the Y direction. The best results for 1.5-inch wafers were achieved when the laser power on the second pass was reduced, an additional pass at 0.04 W was added, the sampling rate of the additional pass was reduced, and laser compensation at 5 micrometers was applied.

[0056] Then, using the identified parameters, three additional silicon carbide wafers were processed, resulting in a reduction of die failures due to cracking caused by the structure in the X-direction die streets from 1428 PPM to 159 PPM, a statistically significant result. The combination of changes to the parameters affected by laser modifications and changes to the fracturing process was unexpected and surprising, especially since the initial tests showed that changing the laser power had no statistically significant effect on reducing yield losses due to uncontrolled lateral die cracking in the structural regions on the wafer.

[0057] The resulting X-die road structure parameters are shown in Tables 5 to 8 below: Table 5 includes the set of specific laser parameters that were used for both 100 micrometer thick and 200 micrometer thick silicon carbide substrates and that resulted from the experiments above: Table 5 path wavelength Power (W) Focal height (µm) Scanning speed (mm / s) path wavelength Laser power (W) Focal height (µm) Scanning speed (mm / s) X0 1064 nm 0,18 -26 525 Y0 1064 nm 0,23 -26 510 Xl 1064 nm 0,12 -19 525 Y1 1064 nm 0,23 -21 510 X2 1064 nm 0,04 -13 150 Y2 1064 nm 0,04 -13 150 X3 1064 nm 0,18 -14 525 Y3 1064 nm 0,23 -17 510 Y4 1064 nm 0,23 -14 510 Table 6 contains the set of specific fracturing parameters for use with the fracturing system for silicon carbide substrates approximately 100 micrometers thick. The fracturing sequence includes fracturing the Y-direction roads first, followed by fracturing the X-direction roads second. Table 6 Direction Clearance height (mm) Anvil distance (ratio multiplier) Chopper lowering speed (mm / s) X 1,23 0,39 20 mm / s J 1,21 0,39 20 mm / s Table 7 contains the set of specific fracturing parameters for use with the fracturing system for silicon carbide substrates approximately 200 micrometers thick. The fracturing sequence includes fracturing the Y-direction roads first, followed by fracturing the X-direction roads second. Table 7 Direction Clearance height (mm) Anvil distance (ratio multiplier) Chopper lowering speed (mm / s) X 1,14 0,39 20 mm / s J 1,12 0,39 20 mm / s Table 8 is the set of specific expansion parameters for use with the expansion system for both approximately 100 micrometers and approximately 200 micrometers thick silicon carbide substrates. Table 8 Expansion height temperature Holding time Expansion rate 8 mm 60 °C 30 seconds 10 mm / s

[0058] The ability to singulate silicon carbide substrates using stealth dicing, while simultaneously significantly reducing lateral and uncontrolled cracking due to structured areas in the die roads, can yield additional benefits by eliminating processing steps applied during sawing. For example, eliminating high-pressure water jets and compressed air on the wafer's top surface during singulation can prevent observable delamination defects of the solderable top metal after stealth dicing. Eliminating chipping from a saw blade can allow for a reduction in die road size and a corresponding increase in wafer density.Other process improvements can be observed in the fact that the substrates per hour or wafers per hour that can be processed using stealth dicing can be measurably higher compared to other processes such as dual-blade saw cutting (2.4 wafers per hour), Sakasa blade cutting (9 wafers per hour), or laser full-cutting (8 wafers per hour). Since the stealth dicing process does not involve the use of water, surfactant chemicals, or blade wear parts, a significant reduction in operating costs could also be achieved compared to a dual-blade sawing process.

[0059] It is understood without further ado that where the foregoing description refers to particular implementations of singulation processes for semiconductor substrates and implementing components, subcomponents, processes and subprocesses, a number of modifications may be made without deviating from its idea, and that these implementations, implementing components, subcomponents, processes and subprocesses may also be applied to other singulation processes for semiconductor substrates.

Claims

Method for singulating silicon carbide, comprising: providing a silicon carbide substrate (2, 30) comprising a thickness; and in a plurality of X-direction die roads: irradiating the silicon carbide substrate (2, 30) in an X-direction (88) with a laser beam (4) focused on a first focal point at a first distance, to the thickness in a first X-pass; irradiating the silicon carbide substrate (2, 30) in the X-direction (88) with the laser beam (4) focused on a second focal point at a second distance, to the thickness in a second X-pass; irradiating the silicon carbide substrate (2, 30) in the X-direction with the laser beam (4) focused on a third focal point at a third distance, to the thickness in a third X-pass;Irradiation of the silicon carbide substrate (2, 30) in the X direction with the laser beam (4), which is focused on a fourth focal point at a fourth distance, into the thickness in a fourth X pass; and in a multitude of Y-direction die paths: Irradiation of the silicon carbide substrate (2, 30) in a Y direction (90) with the laser beam (4) focused on a first focal point at a first distance, into the thickness in a first Y pass; Irradiation of the silicon carbide substrate (2, 30) in a Y direction (90) with the laser beam (4) focused on a second focal point at a second distance, into the thickness in a second Y pass; Irradiation of the silicon carbide substrate (2, 30) in a Y direction (90) with the laser beam (4) focused on a third focal point at a third distance, into the thickness in a third Y pass;Irradiating the silicon carbide substrate (2, 30) in a Y direction (90) with the laser beam (4), which is focused on a fourth focal point at a fourth distance, to the thickness in a fourth Y pass; Irradiating the silicon carbide substrate (2, 30) in a Y direction (90) with the laser beam (4), which is focused on a fifth focal point at a fifth distance, to the thickness in a fifth Y pass; and fracturing (50) the silicon carbide substrate (2, 30) in the X direction and in the Y direction along the plurality of X-direction die roads and the plurality of Y-direction die roads, respectively, using an anvil (54); and extending a band (32) connected to the silicon carbide substrate (2, 30) to separate a plurality of dies from the silicon carbide substrate (2, 30). The method of claim 1, wherein the first distance in the first X-pass extends further into the thickness than the second distance in the second X-pass, the second distance in the second X-pass extends further into the thickness than the third distance in the third X-pass, and the fourth distance in the fourth X-pass extends further into the thickness than the third distance in the third X-pass. The method of claim 1, wherein the first distance in the first X-pass is -26 micrometers, the second distance in the second X-pass is -19 micrometers, the third distance in the third X-pass is -13 micrometers, and the fourth distance in the fourth X-pass is -14 micrometers. The method of claim 1, wherein: the first distance in the first Y-pass extends further into the thickness than the second distance in the second Y-pass; the second distance in the second Y-pass extends further into the thickness than the third distance in the third Y-pass; the fourth distance in the fourth Y-pass extends further into the thickness than the third distance in the third Y-pass; and the fourth distance in the fourth Y-pass extends further into the thickness than the fifth distance in the fifth Y-pass. The method of claim 1, wherein the first distance in the first Y-pass is -26 micrometers, the second distance in the second Y-pass is -21 micrometers, the third distance in the third Y-pass is -13 micrometers, the fourth distance in the fourth Y-pass is -17 micrometers, and the fifth distance in the fifth Y-pass is -14 micrometers. Method according to claim 1, wherein a sampling rate used in the first Y pass, second Y pass, fourth Y pass and fifth Y pass is 510 mm / second and a sampling rate used in the third Y pass is 150 mm / second. The method according to claim 1, wherein the sampling rate used in the first X-pass, second X-pass and fourth X-pass is 525 mm / second and the sampling rate used in the third X-pass is 150 mm / second. The method of claim 1, wherein: the laser power used in the first X-pass and fourth X-pass is 0.18 W; the laser power used in the second X-pass is 0.12 W; the laser power used in the third X-pass is 0.04 W; the laser power used in the first Y-pass, second Y-pass, fourth Y-pass and fifth Y-pass is 0.23 W; and the laser power used in the third Y-pass is 0.04 W. Method for singulating silicon carbide, comprising: providing a silicon carbide substrate (2, 30) comprising a thickness; and in a plurality of X-direction die roads, irradiating the silicon carbide substrate (2, 30) in an X-direction (88) with a laser beam (4) focused on a focal point at a distance, to the thickness in four X-passes; in a plurality of Y-direction die roads, irradiating the silicon carbide substrate (2, 30) in a Y-direction (90) with the laser beam (4) focused on a focal point at a distance, to the thickness in five Y-passes; fracturing (50) the silicon carbide substrate (2, 30) first in the Y-direction and then in the X-direction along the plurality of X-direction die roads, respectively.the plurality of Y-direction die roads using an anvil (54) with a predetermined overrun height, an anvil spacing of 0.39 mm and a chopper lowering speed of 20 mm / second; and extending a belt (32) connected to the silicon carbide substrate (2, 30) to separate a plurality of dies from the silicon carbide substrate (2, 30) at a temperature of 60 °C. Method according to claim 9, wherein, when the thickness of the silicon carbide substrate (2, 30) is 100 micrometers, the predetermined overrun height is 1.23 mm for the X-direction die roads and 1.21 mm for the Y-direction die roads. Method according to claim 9, wherein, when the thickness of the silicon carbide substrate (2, 30) is 200 micrometers, the predetermined overrun height is 1.14 mm for the X-direction die roads and 1.12 mm for the Y-direction die roads. Method according to claim 9, wherein the stretching of the band (32) further comprises stretching with a stretching height of 8 mm, a stretching speed of 10 mm / second and a holding time of 30 seconds. Method for singulating silicon carbide, comprising: providing a silicon carbide substrate (2, 30) comprising a thickness; and in a plurality of X-direction die paths, irradiating the silicon carbide substrate (2, 30) in an X-direction (88) with a laser beam (4) focused on a focal point at a depth in the thickness, in a predetermined number of X-passes, each X-pass of the predetermined number of X-passes having a different laser spot diameter; in a plurality of Y-direction die paths, irradiating the silicon carbide substrate (2, 30) in a Y-direction (90) with the laser beam focused on a focal point at a depth in the thickness, in a predetermined number of Y-passes, each Y-pass of the predetermined number of Y-passes having a different laser spot diameter;Breaking (50) the silicon carbide substrate (2, 30) first in the Y direction (90) and then in the X direction (88) along the plurality of X-direction die roads and the plurality of Y-direction die roads, respectively, using an anvil (54); and extending a band (32) connected to the silicon carbide substrate (2, 30) to separate a plurality of dies from the silicon carbide substrate (2, 30). Method according to claim 13, wherein in the X direction (88) a first laser spot diameter of a first X pass is larger than a second laser spot diameter of a second X pass and a third laser spot diameter of a third X pass is smaller than a fourth laser spot diameter of a fourth X pass. Method according to claim 13, wherein in the Y direction (90) a first laser spot diameter of a first Y pass is larger than a second laser spot diameter of a second Y pass, a third laser spot diameter of a third Y pass is smaller than a fourth laser spot diameter of a fourth Y pass, and a fifth laser spot diameter of a fifth Y pass is smaller than the fourth laser spot diameter of the fourth Y pass. The method of claim 13, wherein a first depth of a first X-pass is -26 micrometers, a second depth of a second X-pass is -19 micrometers, a third depth of a third X-pass is -13 micrometers, and a fourth depth of a fourth X-pass is 14 micrometers. The method of claim 13, wherein a first depth of a first Y-pass is -26 micrometers, a second depth of a second Y-pass is -21 micrometers, a third depth of a third Y-pass is -13 micrometers, a fourth depth of a fourth Y-pass is -17 micrometers, and a fifth depth of a fifth Y-pass is -14 micrometers. Method according to claim 14, wherein the fourth laser spot diameter of the fourth X-pass generates a modified region in sections of the plurality of X-direction die roads that are not covered by a structure. Method according to claim 14, wherein the fourth laser spot diameter of the fourth X-pass burns a structure that is present in sections of the plurality of X-direction die roads. Method according to claim 14, wherein the first laser spot diameter, the second laser spot diameter and the third spot diameter generate a modified region in sections of the plurality of X-direction die roads covered by a structure.

Images