Methods, systems, and manufacturing methods for silicon carbide wafers

Abstract

This application discloses a method, system, and manufacturing method for cleaving silicon carbide wafers. The cleaving method includes the following sequential steps: a pre-cleaving step: guiding and extending microcracks within the modified layer of a silicon carbide ingot that has undergone internal laser-modified layer scanning using a first ultrasonic field; a faceting step: connecting and connecting the extended cracks within the modified layer plane using a second ultrasonic field to form a continuous separation interface; and a separation step: applying a separation force to separate the wafer from the ingot body along the continuous separation interface. This application's technical solution, through a three-step method of guided extension, faceting, and gentle separation, achieves precise stepwise application of ultrasonic energy, effectively reducing the risk of wafer edge chipping and microcracks, and improving wafer separation yield.

Description

Technical Field

[0001] This application relates to the field of semiconductor material processing technology, and in particular to a method, system, and manufacturing method for silicon carbide wafers. Background Technology

[0002] Silicon carbide crystals are hard and brittle. Using lasers to form a modified layer inside the crystal before wafer stripping is an effective way to improve material utilization. The dicing process is crucial to the success of wafer stripping and the quality of the wafer.

[0003] Currently, the mainstream wafer cleaving process uses a combination of suction cup pulling and ultrasonic vibration. Ultrasonic energy is used to assist in the propagation of laser-induced microcracks in a liquid environment, helping to separate the wafer from the ingot.

[0004] However, this method has significant drawbacks: the microcracks formed by the laser are unevenly distributed within the modified layer and fail to connect completely. If the ultrasonic energy used is low, the resulting cavitation effect is insufficient to completely connect the cracks, leading to cleaving failure or the inability to completely separate the wafer. If the ultrasonic energy is increased to improve the cleaving success rate, it can easily cause the cracks to propagate uncontrollably, intruding into the wafer or ingot body, and triggering the risk of edge chipping, microcracks, or even overall fragmentation.

[0005] Therefore, existing technologies face the contradiction of "insufficient energy is ineffective, while excessive energy is easily damaged" in the process of ultrasonic-assisted wafer cleaving. It is difficult to ensure the integrity and yield of the wafer while guaranteeing the success rate of cleaving, which restricts the full realization of the advantages of laser wafer stripping technology. Summary of the Invention

[0006] This application provides a method, system, and manufacturing method for cleaving silicon carbide wafers, aiming to solve the problem in the prior art where the success rate of cleaving and the integrity of the wafer are difficult to balance due to the contradiction in ultrasonic energy control.

[0007] To achieve the above objectives, this application proposes a method for cleaving silicon carbide wafers. This method includes the following steps performed in sequence: Pre-cracking step: For silicon carbide ingots that have completed internal laser modification layer scanning, microcracks within the modification layer are guided and extended by a first ultrasonic field; Surface formation step: The extended cracks are connected and penetrated in the plane of the modified layer using a second ultrasonic field to form a continuous separation interface; Separation step: Apply a separation force to separate the wafer from the ingot body along the continuous separation interface.

[0008] In some embodiments, the first ultrasonic field is a focused ultrasonic field, and the trajectory of the ultrasonic focal point corresponds to the plane of the modified layer.

[0009] In some embodiments, the frequency of the focused ultrasound field is 0.5 to 2 MHz and the power is 250 to 500 W.

[0010] In some embodiments, the second ultrasonic field is generated by a flat-headed ultrasonic transducer with a frequency of 80–150 kHz and a power of 80–150 W.

[0011] In some embodiments, the separation force is a vibration shear force generated by a third ultrasonic field, which is used to vibrate out the residue at the separation interface to achieve separation of the wafer and the ingot body. The energy density of the third ultrasonic field is lower than that of the first and second ultrasonic fields.

[0012] In some embodiments, the third ultrasonic field is provided by an ultrasonic cleaning device with a power of 400-600W and an action time of 3-5 minutes.

[0013] In some embodiments, the separation force is a mechanical pull-out force applied to the wafer surface.

[0014] This application also provides a silicon carbide wafer cleaving system for performing the cleaving method described above. The silicon carbide wafer cleaving system includes: a pre-cleaving module for generating the first ultrasonic field; a faceting module for generating the second ultrasonic field; and a separation module for providing the separation force.

[0015] In some embodiments, the pre-splitting module includes a focused ultrasonic generator with a frequency of 0.5 to 2 MHz and a power of 250 to 500 W; the surface-forming module includes a flat-headed ultrasonic transducer with a frequency of 80 to 150 kHz and a power of 80 to 150 W; and the separation module is an ultrasonic cleaning device or a mechanical pulling device.

[0016] This application also provides a method for manufacturing a silicon carbide wafer, the method comprising: A ring-shaped modified layer is formed inside a silicon carbide ingot by laser scanning; The wafer is separated from the ingot body using the dicing method described above.

[0017] This application proposes a method for cleaving silicon carbide wafers. The cleaving method includes the following sequential steps: a pre-cleaving step: for a silicon carbide ingot that has undergone internal laser-modified layer scanning, microcracks within the modified layer are guided and extended using a first ultrasonic field; a faceting step: a second ultrasonic field is used to connect and penetrate the extended cracks within the plane of the modified layer, forming a continuous separation interface; and a separation step: a separation force is applied, causing the wafer to separate from the ingot body along the continuous separation interface. This application's technical solution achieves precise, stepwise application of ultrasonic energy through a three-step method of guided extension, faceting, and gentle separation. The pre-cleaving step allows for the application of higher energy without separation risk even when cracks are not fully penetrated, while the faceting step ensures complete formation of the separation interface. This allows the final separation step to be completed with minimal energy, fundamentally avoiding the dilemma of energy levels in traditional methods, effectively reducing the risk of wafer edge chipping and microcracks, and improving wafer separation yield. This application also provides a silicon carbide wafer cleaving system and manufacturing method. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, wherein: Figure 1 This is a schematic flowchart of a silicon carbide wafer dicing method according to an embodiment of this application; Figure 2 This is a schematic flowchart of a method for preparing a silicon carbide wafer according to an embodiment of this application; Figure 3 This is a schematic diagram of the growth of microcracks within a silicon carbide ingot after the internal laser-modified layer scanning has been completed. Figure 4 A schematic diagram of the growth of internal cracks after microcracks in the modified layer are guided and extended using a first ultrasonic field. Figure 5 This is a schematic diagram of the internal cracks after being treated with the second ultrasonic field. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0020] It should be noted that, unless otherwise stated or limited, all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of this application are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0021] It should also be noted that, unless otherwise stated or limited, when an element is referred to as "fixed to" or "set on" another element, it may be directly on the other element or there may be an intervening element present. When an element is referred to as "connected to" another element, it may be directly connected to the other element or there may be an intervening element present.

[0022] Furthermore, unless otherwise stated or limited, the descriptions involving "first," "second," etc., in this application are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. Additionally, the technical solutions of the various embodiments can be combined with each other, but only on the basis of being achievable by those skilled in the art. When the combination of technical solutions is contradictory or impossible to implement, such a combination of technical solutions should be considered non-existent and not within the scope of protection claimed in this application.

[0023] This application proposes a method for cleaving silicon carbide wafers, see reference. Figure 1 The diagram shows a flow chart of a silicon carbide wafer dicing method. This silicon carbide wafer dicing method includes the following steps performed in sequence: Step S10, Pre-cracking step: For the silicon carbide ingot that has completed the scanning of the internal laser modified layer, microcracks in the modified layer are guided and extended by the first ultrasonic field.

[0024] Understandably, after laser scanning, a large number of random and discontinuous microcracks and defects already exist in the modified layer. Traditional methods attempt to use one-time, global energy (whether ultrasonic or mechanical force) to brute force through these cracks. If the force is too small, it cannot be pushed through; if the force is too large, it will cause random fragmentation.

[0025] This step aims to refine crack treatment while maintaining the wafer and ingot as a single, mutually supporting and constrained mechanical unit. Even with high energy, the ingot's inherent structure prevents uncontrolled crack propagation into the wafer or ingot, thus preventing damage. Specifically, a first ultrasonic field with a specific frequency and amplitude is applied to the ingot to selectively excite stress concentration areas at the microcrack tips within the modified layer. This promotes the orderly extension and interconnection of microcracks along the laser scanning path, forming pre-cracked channels. This process avoids the risk of uncontrollable fracture caused by direct external mechanical forces, while improving the uniformity and predictability of crack propagation, laying the foundation for subsequent separation processes.

[0026] Optionally, the first ultrasonic field is a focused ultrasonic field, which focuses ultrasonic energy onto a very small point or linear area (focal point) and scans along the plane of the modified layer; that is, the trajectory of the ultrasonic focal point corresponds to the plane of the modified layer. The focused ultrasonic field has extremely high energy density per unit area at the focal point, which can efficiently ablate or vibrate the crack tip, promoting its directional extension. The frequency of the focused ultrasonic field is between 0.5 and 2 MHz, and the power is between 250 and 500 W.

[0027] Alternatively, in some other embodiments, the first ultrasonic field can also employ a plane wave of a specific mode. By designing a special transducer, a waveform that can resonate or strongly couple with the crack tip can be generated to achieve the same directional propagation of the crack. However, its control precision and energy efficiency are usually lower than those of focused ultrasound, and it is prone to causing vibration interference in non-target areas. Therefore, focused ultrasound is the preferred solution.

[0028] Step S20, surface formation step: The extended cracks are connected and connected within the plane of the modified layer using a second ultrasonic field to form a continuous separation interface.

[0029] Through the pre-splitting step, the crack has developed from a point into a longer line segment, but there are still unconnected breaks between the segments. This step aims to robustly and gently connect these segments into a complete, closed two-dimensional separation surface. At this point, since the pre-splitting step has consumed a large amount of fracture energy, the material strength in the remaining fractured areas is very low, so the high-energy-density ultrasound used in the pre-splitting step is not necessary.

[0030] In this step, a second ultrasonic field is applied to the entire modified layer plane with low power and wide coverage, achieving physical connection and gentle separation of the entire separation interface with relatively low energy. The frequency is set between 0.2 and 1 MHz, and the power is controlled between 80 and 150 W, selected based on the size, mass, and thickness of the spade, forming a uniform and non-focused vibration field. Then, by adjusting the phase and scanning path of the ultrasonic waves, the energy is continuously applied to the weak connection areas between cracks, inducing the slow fusion of microcrack ends and gradually closing the fracture points.

[0031] This step avoids the formation of new, uncontrollable secondary cracks due to excessive local energy during the bonding process, thus ensuring the flatness of the separation interface and the quality of the back side of the wafer.

[0032] Optionally, the second ultrasonic field uses an ultrasonic transducer with a flat end face, such as a flat-headed ultrasonic transducer. The flat-headed ultrasonic transducer operates at a low to medium frequency of 80–150 kHz and a low to medium power (e.g., 80–150 W). The ultrasonic field it generates has a relatively uniform energy distribution on the action surface, and can act on multiple breakpoints simultaneously, gently breaking them or causing fatigue fracture, thus achieving connection.

[0033] In some other implementations, the second ultrasonic field is achieved by high-power wide-area ultrasound, such as a high-power ultrasonic cleaning tank. Although the energy density is lower and the effect is gentler, the crack surface can still be connected by extending the action time or optimizing the frequency matching.

[0034] Step S30, Separation Step: Apply a separation force to separate the wafer from the ingot body along a continuous separation interface.

[0035] Through the pre-crack and faceting steps, the wafer and ingot are physically connected only by a very small amount of residual powder or tiny unbroken bridges, and their mechanical strength has been reduced to a very low level. At this point, any slight disturbance is enough to separate them. Therefore, this step aims to break this last weak connection in the gentlest and least impactful way possible, in order to finally achieve a non-destructive separation between the wafer and the ingot body.

[0036] Optionally, the separation force is the vibrational shear force generated by the third ultrasonic field, used to vibrate away residues at the separation interface to achieve separation. Specifically, low-energy wide-area ultrasound, such as cleaning-grade ultrasound, can be used to achieve physical separation. The crystal ingot is placed in a conventional ultrasonic cleaner, utilizing its cavitation effect and liquid vibration to easily vibrate away residual powder at the interface, achieving contactless automatic separation. The ultrasonic cleaner has a power of 400–600W, a processing time of 3–5 minutes, and requires no precision clamps.

[0037] Alternatively, in some embodiments, the separation force is a mechanical pulling or warping force applied to the wafer surface. Separation can be achieved with only a very small, controllable pulling or warping force, and since the interface is completely broken, even when using mechanical methods, there is no risk of tearing due to incomplete crack penetration, as is seen in traditional drawing processes.

[0038] For specific effects, please refer to [link / reference]. Figures 3-5 As shown, where, Figure 3This is a schematic diagram of the growth of microcracks in a silicon carbide ingot after the internal laser-modified layer scanning has been completed. The red lines in the diagram represent the laser cutting path, and the yellow lines represent the cracks generated around the laser path. After the laser action, the cracks grow near the laser cutting path, do not connect together, and show a tendency to grow in a disordered manner. Figure 4 This is a schematic diagram of the growth of internal cracks after the microcracks in the modified layer are guided and extended using the first ultrasonic field. Specifically, after treatment with focused ultrasound, the cracks around the laser path extend along the ultrasonic direction, and the cracks intersect and connect together, forming a surface from lines, in preparation for subsequent processing. Figure 5 This is a schematic diagram of the internal cracks after the second ultrasonic field treatment; it can be seen that the yellow cracks have covered the entire modified layer. Based on the fact that the focused ultrasound has connected the cracks together, the entire modified layer is covered with cracks. Under these circumstances, only an ordinary ultrasonic cleaner or a micro-mechanical separation device is needed to shatter the modified layer and separate the wafer from the ingot body.

[0039] In summary, the technical solution of this application decomposes the complex wafer dicing task into three sub-functions: extension, connection, and separation, with each step having a single and clear objective. Furthermore, it matches the most suitable energy form and magnitude (focused high energy, planar medium energy, and wide-area low energy / micro-mechanical energy) to each functional step, achieving precise, efficient, and safe energy utilization. The highest-risk operation (high-energy processing) is placed in the initial stage when the structure is most stable, and the risk gradually decreases as the process progresses until the final step is almost risk-free. Ultimately, this transforms the traditional high-risk, high-loss wafer dicing process into a low-stress, low-damage, controllable separation process, significantly improving wafer yield and surface quality.

[0040] This application also provides a silicon carbide wafer cleaving system for performing the cleaving method described above. The silicon carbide wafer cleaving system includes a pre-cleaving module for generating a first ultrasonic field. Its core task is to safely apply high-energy-density ultrasound under structural constraints, achieving directional guidance and initial propagation of the crack. It may include the following components: High-frequency focused ultrasound generator: Typically employs a piezoelectric ceramic transducer with a frequency of 1 MHz or higher (e.g., 0.5–2 MHz), combined with an acoustic lens or curved self-focusing design, to focus ultrasonic energy onto a very small focal point. High-precision motion platform: Carries and drives the focused ultrasound head, enabling its focal point to scan along a preset three-dimensional path perfectly aligned with the plane of the laser-modified layer. High-frequency power source and matching network: Provides 250–500W (adjustable) of high-frequency electrical power and ensures efficient conversion of electrical energy into acoustic energy via the matching network, which is then transmitted to the transducer. Coupling medium circulation system: Provides and maintains a coupling fluid such as deionized water to ensure efficient transmission of ultrasonic waves from the transducer to the ingot.

[0041] The surface-forming module is used to generate a second ultrasonic field; its core task is to robustly connect the pre-cracked crack segments into a complete and continuous two-dimensional separation interface using a uniform and gentle energy field. It may include the following components: Low-to-medium frequency flat-head ultrasonic transducer: Typically employs a sandwich-type piezoelectric transducer (i.e., flat-head transducer) with a frequency range of 80–150 kHz. Its flat emitting surface results in a relatively uniform energy distribution of the generated ultrasonic waves in the near-field region. Adjustable power drive source: Provides 80–150 W (adjustable) drive power and can adaptively adjust according to ingot size, thickness, and other information. Mechanical contact or near-field coupling mechanism: Designed to maintain a constant, uniform, minute distance or gentle contact between the transducer end face and the ingot surface, ensuring effective transmission of ultrasonic energy to the target plane inside the ingot.

[0042] The separation module provides the separation force, and its core task is to ultimately separate the wafer, which has already formed a separation interface, from the ingot body with minimal mechanical or energy disturbance. It may include the following components: Configuration A (All-ultrasound Solution): Wide-area ultrasonic cleaning tank: A standard or custom ultrasonic cleaning tank equipped with common cleaning-grade transducers with a frequency of approximately 20–40 kHz and a power of 400–600 W. Its cavitation effect and macroscopic liquid vibration are sufficient to dislodge interface residues. Loading and conveying mechanism: Smoothly moves the ingots processed in the first two steps into and out of the cleaning tank.

[0043] Configuration B (Hybrid Solution): Micro-force mechanical separation devices: such as vacuum suction cup lifting mechanisms or precision warping mechanisms with highly sensitive pressure feedback. Since the interface is completely broken, micro-force mechanical separation devices can achieve separation by providing a tiny, almost impact-free force, far less than that required by traditional pulling.

[0044] This application also provides a method for manufacturing silicon carbide wafers, see reference. Figure 2 The diagram shows a process flow chart for fabricating a silicon carbide wafer; the method for manufacturing this silicon carbide wafer includes: Step S01: A ring-shaped modified layer is formed inside the silicon carbide ingot by laser scanning.

[0045] This step sets a precise physical separation surface for subsequent dicing. Utilizing the nonlinear absorption effect of ultrashort pulse lasers (such as picosecond and femtosecond lasers), the laser focus is precisely positioned at a predetermined cutting depth inside the ingot, and a weakening layer composed of dense microcracks, phase transformations, or defects is formed through scanning.

[0046] Understandably, traditional cutting methods (such as diamond wire saws) produce continuous cutting tracks and debris. In contrast, the laser-modified layer is a discrete network of defect points. This layer provides the optimal starting point and energy coupling interface for subsequent pre-splitting and surface-forming ultrasonic processes. The cavitation and vibration effects of ultrasonic energy can efficiently interact with these microscopic defects, driving their expansion and connection.

[0047] Step S02: Using the dicing method described above, the wafer is separated from the ingot body.

[0048] The cleaving method in this step has been described in detail above. It adopts all the technical solutions of all the above-mentioned cleaving scheme embodiments, and therefore has at least all the beneficial effects brought about by the technical solutions of the above embodiments. It will not be repeated here.

[0049] In summary, because the dicing process is controllable and gentle, the initial surface damage layer of the separated wafer is extremely shallow, with no macroscopic scratches or warping. This can significantly reduce the burden on subsequent thinning and surface treatment processes such as grinding and polishing, shorten the processing cycle, and ultimately improve the crystal quality and mechanical strength of the substrate.

[0050] The above description is only a part or preferred embodiment of this application. Neither the text nor the drawings should limit the scope of protection of this application. All equivalent structural transformations made using the content of this application's specification and drawings under the overall concept of this application, or direct / indirect applications in other related technical fields, are included within the scope of protection of this application.

Claims

1. A method for dicing silicon carbide wafers, characterized in that, The steps are executed in the following order: Pre-cracking step: For silicon carbide ingots that have completed internal laser modification layer scanning, microcracks within the modification layer are guided and extended by a first ultrasonic field; Surface formation step: The extended cracks are connected and penetrated in the plane of the modified layer using a second ultrasonic field to form a continuous separation interface; Separation step: Apply a separation force to separate the wafer from the ingot body along the continuous separation interface.

2. The dicing method according to claim 1, characterized in that, The first ultrasonic field is a focused ultrasonic field, and the trajectory of the ultrasonic focal point corresponds to the plane of the modified layer.

3. The dicing method according to claim 2, characterized in that, The frequency of the focused ultrasound field is 0.5 to 2 MHz, and the power is 250 to 500 W.

4. The dicing method according to claim 1, characterized in that, The second ultrasonic field is generated by a flat-headed ultrasonic transducer with a frequency of 80–150 kHz and a power of 80–150 W.

5. The dicing method according to claim 1, characterized in that, The separation force is the vibration shear force generated by the third ultrasonic field, which is used to shake out the residue at the separation interface to achieve the separation of the wafer and the ingot body. The energy density of the third ultrasonic field is lower than that of the first and second ultrasonic fields.

6. The dicing method according to claim 5, characterized in that, The third ultrasonic field is provided by an ultrasonic cleaning device with a power of 400-600W and an action time of 3-5 minutes.

7. The dicing method according to claim 1, characterized in that, The separation force is a mechanical pull-out force applied to the wafer surface.

8. A dicing system for silicon carbide wafers, used to perform the dicing method as described in any one of claims 1 to 7, characterized in that, include: A pre-splitting module is used to generate the first ultrasonic field; A surface-forming module is used to generate the second ultrasonic field; A separation module is used to provide the separation force.

9. The dicing system according to claim 8, characterized in that, The pre-splitting module includes a focused ultrasonic generator with a frequency of 0.5–2 MHz and a power of 250–500 W; the surface-forming module includes a flat-headed ultrasonic transducer with a frequency of 80–150 kHz and a power of 80–150 W; the separation module is an ultrasonic cleaning device or a mechanical pulling device.

10. A method for manufacturing a silicon carbide wafer, characterized in that, include: A ring-shaped modified layer is formed inside a silicon carbide ingot by laser scanning; The wafer is separated from the ingot body using the dicing method as described in any one of claims 1 to 7.

Images