Improved performance of ridge waveguide device structure and fabrication method

Abstract

This application discloses a ridge waveguide device structure and fabrication method with improved performance, comprising: forming a waveguide layer on a substrate; partially etching the waveguide layer through a first hard mask to form a first waveguide pattern; performing a first treatment on the first sidewall of the first waveguide pattern to reduce the surface roughness of the first sidewall; after removing the first hard mask, fully etching the waveguide layers on both sides of the first waveguide pattern through a second hard mask covering the first waveguide pattern; forming a second waveguide pattern below the first waveguide pattern to form a ridge waveguide; performing a second treatment on the second sidewall of the second waveguide pattern to reduce the surface roughness of the second sidewall; and after removing the second hard mask, performing a third treatment on the first and second sidewalls to further reduce the surface roughness of the first and second sidewalls. This application can significantly improve the sidewall smoothness of the ridge waveguide, thereby improving the key performance of the device.

Description

Technical Field

[0001] This application relates to the field of semiconductor processing technology, and in particular to a ridge waveguide device structure and fabrication method with improved performance. Background Technology

[0002] Currently, in optical waveguide device chips, there is a problem with the poor sidewall roughness of patterned ridge waveguides, which mainly affects the following key performance characteristics: (1) Propagation Loss: The rougher the sidewalls, the more scattering loss occurs when light propagates in the waveguide. When the light field propagates in the silicon waveguide, some of the energy will change direction due to the roughness of the sidewalls and leak out of the waveguide, resulting in a decrease in signal strength.

[0003] (2) Backscattering: Rough sidewalls cause light to reflect backward, forming backscattering. This can interfere with the forward-propagating optical signal, introduce noise, and may affect the stability of the laser (especially when the laser is coupled to a waveguide).

[0004] (3) Polarization Performance: Sidewall roughness can induce mode coupling, especially crosstalk between TE and TM modes, leading to a decrease in polarization extinction ratio. For polarization-sensitive photonic integrated circuits (PICs), this can severely affect system performance.

[0005] (4) Phase noise and coherence: In coherent optical communication or sensing systems, random phase disturbances caused by sidewall roughness will introduce phase noise and reduce the sensitivity of coherent detection.

[0006] The main cause of poor sidewall roughness is the insufficient etching capability of existing patterning processes for forming ridge waveguides. Therefore, it is necessary to study a process method that can significantly improve the sidewall roughness problem of ridge waveguides. Summary of the Invention

[0007] The purpose of this application is to overcome the aforementioned problems in the prior art and to provide a ridge waveguide device structure and fabrication method with improved performance.

[0008] To achieve the above objectives, the technical solution of this application is as follows: According to a first aspect of this application, embodiments of this application provide a method for fabricating a performance-improved ridge waveguide device structure, comprising: Provide substrate; A waveguide layer and a first hard mask are sequentially formed on the substrate; The waveguide layer is partially etched using the first hard mask to form a first waveguide pattern on the waveguide layer; The first sidewall of the first waveguide pattern is subjected to a first processing, which includes multiple alternating processing steps and passivation steps. The processing steps use a first free radical to remove protrusions on the first sidewall to reduce the surface roughness of the first sidewall. The passivation steps use a second free radical to passivate the surface of the first sidewall for protection. The first hard mask is removed, and a second hard mask covering the first waveguide pattern is formed on the waveguide layer. The waveguide layer on both sides of the first waveguide pattern is fully etched through the second hard mask, and a second waveguide pattern is formed below the first waveguide pattern, thereby forming a ridge waveguide including the first waveguide pattern and the second waveguide pattern. Using a third free radical, a second treatment is performed on the second sidewall of the second waveguide pattern. A polymer layer is deposited on the second sidewall using the byproducts generated during the second treatment process to protect the second sidewall while removing the protrusions present on the second sidewall, thereby reducing the surface roughness of the second sidewall. The second hard mask is removed, and the first and second sidewalls are subjected to a third treatment using metastable excited fourth free radicals to remove the remaining protrusions on the first and second sidewalls, thereby further reducing the surface roughness of the first and second sidewalls.

[0009] In some embodiments, the waveguide layer is made of silicon.

[0010] In some embodiments, the material of the second hard mask includes spin-coated carbon.

[0011] In some embodiments, the first free radical includes a fluorine-containing free radical, a first nitrogen-containing free radical, and a first oxygen-containing free radical, and the second free radical includes a second nitrogen-containing free radical and a second oxygen-containing free radical.

[0012] In some embodiments, the third free radical includes a first hydrogen-containing free radical.

[0013] In some embodiments, the fourth free radical includes a second hydrogen-containing free radical.

[0014] In some embodiments, the fluorine-containing free radical, the first nitrogen-containing free radical, and the first oxygen-containing free radical are obtained by exciting a first gas, the first gas including CF4, N2, and O2. The fluorine-containing free radical is used to remove protrusions on the first sidewall, the first nitrogen-containing free radical and the first oxygen-containing free radical are used to promote the dissociation of CF4, and the first nitrogen-containing free radical is also used to passivate the surface of the first sidewall for protection. The second nitrogen-containing free radical and the second oxygen-containing free radical are obtained by exciting a second gas, the second gas including N2 and O2. The second nitrogen-containing free radical is used to promote the dissociation of O2 to obtain the second oxygen-containing free radical, and the second oxygen-containing free radical is used to oxidize the surface of the first sidewall to achieve surface passivation.

[0015] In some embodiments, the first hydrogen-containing free radical is obtained by exciting a third gas, the third gas including H2, the first hydrogen-containing free radical is used to remove protrusions present on the second sidewall, and the polymer layer is formed by reacting the first hydrogen-containing free radical with free active carbon atoms generated after the surface of the second hard mask is bombarded by plasma, and the resulting byproducts are deposited on the second sidewall.

[0016] In some embodiments, the second hydrogen-containing free radical is obtained by exciting a fourth gas, which includes H2. The H2 is excited to form a plasma by using helium metastable particles, and the charged particles therein are filtered out to obtain the second hydrogen-containing free radical. The second hydrogen-containing free radical is used to remove the protrusions remaining on the first sidewall and the second sidewall.

[0017] In some embodiments, when performing the processing steps, the flow rate of O2 is 100 sccm to 2000 sccm, the flow rate of CF4 is 0.5 to 2 times the flow rate of O2, the flow rate of N2 is 0.5 to 1 times the sum of the flow rates of CF4 and O2, the temperature is 50°C to 350°C, the pressure is 500 mTorr to 2000 mTorr, the source power is 500 W to 2000 W, the bias power is 1 W to 30 W, and the time is 3 s to 20 s.

[0018] In some embodiments, during the passivation step, the flow rate of O2 is 100 sccm to 2000 sccm, the flow rate of N2 is 0.1 to 1 times that of O2, the temperature is 50°C to 350°C, the pressure is 500 mTorr to 2000 mTorr, the source power is 500 W to 2000 W, the bias power is 1 W to 30 W, and the time is 3 s to 20 s.

[0019] In some embodiments, when performing the second process, the flow rate of H2 is 50 sccm to 1000 sccm, the temperature is 100°C to 350°C, the pressure is 500 mTorr to 2000 mTorr, the source power is 500 W to 2000 W, the bias power is 1 W to 30 W, and the time is 30 s to 120 s.

[0020] In some embodiments, when performing the third process, the flow rate of H2 is 50 sccm to 1000 sccm, the temperature is 50°C to 250°C, the pressure is 100 mTorr to 2000 mTorr, the source power is 200 W to 2000 W, the bias power is 0 W, and the time is 30 s to 180 s.

[0021] In some embodiments, when performing the first processing, the first hard mask is used to protect the apex corner of the first waveguide pattern.

[0022] In some embodiments, when performing the second processing, the first waveguide pattern and the apex corner of the second waveguide pattern are protected using the second hard mask.

[0023] In some embodiments, during the first processing, a first processing temperature of 200°C to 350°C is also used to repair lattice damage to the first sidewall.

[0024] In some embodiments, when performing the second treatment, the first hydrogen-containing free radical is also used to remove impurities on the second sidewall and repair lattice damage. Specifically, the second treatment is performed at a temperature of 100°C to 200°C to reduce the activity of the first hydrogen-containing free radical and decrease the reaction rate, or the second treatment is performed at a temperature of 200°C to 350°C to activate the defective lattice on the second sidewall and improve the reaction efficiency of the first hydrogen-containing free radical.

[0025] In some embodiments, during the third processing, the second hydrogen-containing free radical is also used to remove impurities on the surface of the ridge waveguide and to repair lattice damage.

[0026] In some embodiments, another implementation of the first free radical is to replace the fluorine-containing free radical, the first nitrogen-containing free radical, and the first oxygen-containing free radical with a third hydrogen-containing free radical; another implementation of the second free radical is to replace the second nitrogen-containing free radical and the second oxygen-containing free radical with hydroxyl free radicals; the third hydrogen-containing free radical is obtained by exciting a fifth gas, the fifth gas including H2, by exciting H2 with helium metastable particles to form a plasma, and after filtering out charged particles therein, the third hydrogen-containing free radical is obtained; the hydroxyl free radical is obtained by exciting a sixth gas, the sixth gas including H2O, by exciting H2O with helium metastable particles to form a plasma, and after filtering out charged particles therein, the hydroxyl free radical is obtained; the third hydrogen-containing free radical is used to remove the protrusions on the first sidewall, and the hydroxyl free radical is used to adsorb and form Si-OH bonds on the first sidewall to passivate the surface of the first sidewall, so as to avoid the third hydrogen-containing free radical from reacting with silicon during the processing steps.

[0027] In some embodiments, when performing the processing steps, the flow rate of H2 is 80 sccm to 150 sccm, the temperature is 110°C to 350°C, the pressure is 10 mTorr to 50 mTorr, the source power is 400 W to 800 W, the bias power is 1 W to 30 W, and the time is 3 s to 20 s.

[0028] In some embodiments, during the passivation step, the H2O flow rate is 3 sccm to 10 sccm, the temperature is 110°C to 350°C, the pressure is 10 mTorr to 50 mTorr, the source power is 400 W to 800 W, the bias power is 1 W to 30 W, and the time is 3 s to 20 s.

[0029] In some embodiments, during the first treatment, the third hydrogen-containing free radical is also used to remove impurities on the first sidewall and repair lattice damage.

[0030] According to a second aspect of this application, embodiments of this application also provide a ridge waveguide device structure, which is obtained using the performance-improved ridge waveguide device structure fabrication method provided in any of the embodiments of the first aspect above.

[0031] The embodiments of this application may have, or at least have, the following advantages: (1) By performing a first treatment on the first sidewall of the first waveguide pattern, the protrusions on the first sidewall can be preferentially removed by the first free radical (fluorine-containing free radical, first nitrogen-containing free radical, and first oxygen-containing free radical / third hydrogen-containing free radical) used in the treatment step, thereby reducing the roughness of the first sidewall. The surface of the first sidewall is then passivated by the second free radical (second nitrogen-containing free radical and second oxygen-containing free radical / hydroxyl free radical) used in the passivation step, so as to protect the surface of the first sidewall during the treatment step. Thus, the first sidewall can be finely roughened by multiple alternating cycles of the treatment step and the passivation step, so that the surface roughness of the first sidewall can be effectively reduced. The second sidewall of the second waveguide pattern with the second hard mask is then treated by using the third free radical (first hydrogen-containing free radical). The second treatment utilizes the polymer layer deposited on the second sidewall by the byproducts generated during the second treatment process to protect the second sidewall while simultaneously removing protrusions on the second sidewall using the first hydrogen-containing free radical, thus reducing the surface roughness of the second sidewall. After removing the second hard mask, a metastable excited fourth free radical (the second hydrogen-containing free radical) is used to perform a essentially non-destructive third treatment on the first and second sidewalls at a lower energy level, removing any remaining protrusions on the first and second sidewalls and further reducing their surface roughness. Thus, the overall sidewall roughness of the ridge waveguide is reduced through stepwise processing, improving the surface smoothness of the ridge waveguide sidewalls. Therefore, it can significantly improve key performance characteristics of the device, such as propagation loss, backscattering, polarization performance, phase noise, and coherence.

[0032] (2) By using a first gas including CF4, N2 and O2 for processing and exciting the first gas, the obtained first nitrogen-containing free radicals and first oxygen-containing free radicals can be used to promote the dissociation of CF4 and generate volatile NO groups / CO groups, which can generate more fluorine-containing free radicals to efficiently remove the protrusions on the first sidewall, thus improving efficiency. The first nitrogen-containing free radicals can be used to passivate the surface of the first sidewall to protect the surface of the first sidewall while removing the protrusions. By using a second gas including N2 and O2 for passivation and exciting the second gas, the obtained second nitrogen-containing free radicals can be used to promote the dissociation of O2, thereby improving the efficiency of the obtained second oxygen-containing free radicals in the surface oxidation of the first sidewall, thus providing timely and effective protection for the apex of the first waveguide pattern and maximizing the preservation of the sharp turning shape at the apex.

[0033] (3) By using a fifth gas including H2 for the processing step and exciting the fifth gas, the obtained third hydrogen-containing free radicals can be used to remove the protrusions on the first sidewall; by using a sixth gas including H2O for the passivation step and exciting the sixth gas, the obtained hydroxyl free radicals can be used to adsorb and form Si-OH bonds as dangling bonds on the first sidewall, thereby passingivating and protecting the surface of the first sidewall, avoiding the undesirable etching reaction between the third hydrogen-containing free radicals and the silicon material on the first sidewall during the processing step, and effectively maintaining the lateral critical dimension; and by using helium metastable particles to excite the fifth and sixth gases, a basically damage-free surface smoothing process can be performed at a lower energy, thereby enabling a fine roughening process for the first sidewall.

[0034] (4) By using spin-coated carbon (SOC) material to form a second hard mask, the formed first waveguide pattern is completely covered. This isolates the physical bombardment and chemical corrosion of the etching gas during the formation of the second waveguide pattern, avoiding process damage such as loss, roughness, and defects in the first waveguide pattern, and ensuring the integrity and dimensional accuracy of the overall ridge waveguide structure. Furthermore, by performing a second processing while retaining the second hard mask, the high activity and low etching selectivity of the first hydrogen-containing free radical (third free radical) can be utilized to selectively react with the microscopic high-energy defects (protrusions) on the second sidewall. Through the microscopic modification effect, the surface of the second sidewall can be "peak-shaving and valley-filling" without over-etching the silicon material on the second sidewall, repairing defects such as burrs, unevenness, and microscopic holes on the second sidewall after etching, and reducing the roughness of the second sidewall. At the same time, relying on the plasma emitted from the SOC under the bombardment of plasma, The free active carbon atoms have a reaction buffering capacity, which can react with the excess first hydrogen-containing free radicals in the system, consume the redundant active hydrogen, and inhibit the excessive etching of the silicon material on the second sidewall by the first hydrogen-containing free radicals. Only an appropriate amount of the first hydrogen-containing free radicals are retained for defect modification, and a temporary and selective polymer layer is formed on the surface of the second sidewall using the reaction byproducts as a protective film, which plays a secondary protection role. This helps to ensure that the etching process of the silicon atom layer on the surface of the second sidewall only occurs on the "peaks" that need to be removed, without destroying the overall structure of the surface of the second sidewall. Furthermore, depending on the degree of roughness of the second sidewall surface, the second treatment can be performed in different temperature ranges. When the roughness is relatively mild, the second treatment can be performed at a lower temperature of 100℃ to 200℃. At this temperature, the activity of the first hydrogen-containing free radical is relatively low, which can make the reaction rate slow down. This mainly focuses on impurity removal and slight repair of micro-defects, which is suitable for fine repair of slightly rough second sidewall structures. When the roughness is severe, the second treatment can be performed at a higher temperature of 200℃ to 350℃. By appropriately increasing the temperature, the defect lattice of silicon on the second sidewall surface can be activated, the reaction efficiency of the first hydrogen-containing free radical and high-energy defect silicon atoms can be improved, the etching of the protrusion and the rearrangement of surface atoms can be accelerated, and the surface smoothness of the second sidewall can be significantly improved. This approach not only solves the problems of structural damage, sidewall roughness, and impurity residue that easily occur in dry etching of silicon optical waveguides, but also precisely balances the structural protection and surface modification effects. It can effectively reduce the light scattering loss of silicon optical waveguides and significantly improve the forming accuracy and optical transmission performance of optical waveguide devices, thereby enabling the fabrication of high-performance silicon-based optical waveguide devices.

[0035] (5) By performing the first processing while retaining the first hard mask, the first hard mask can be used as a capping layer on top of the first waveguide pattern to protect the apex of the first waveguide pattern from damage and morphological changes; by performing the second processing while retaining the second hard mask, the second hard mask can be used as a capping layer on top of the second waveguide pattern to protect the apex of the second waveguide pattern from damage and morphological changes while covering and protecting the first waveguide pattern; by performing the third processing on the sidewalls (first sidewall and second sidewall) of the ridge waveguide using the second hydrogen-containing free radical obtained by excitation of helium metastable particles, a surface roughening smoothing process with virtually no damage can be performed at a lower energy, completely eliminating longitudinal damage, and refining the morphological defects that may exist at the apex of the first and second waveguide patterns after removing the second hard mask, further improving the overall pattern quality of the ridge waveguide.

[0036] (6) By reducing the sidewall roughness and repairing the lattice damage during the first, second and third treatments, a comprehensive treatment effect was achieved.

[0037] In summary, by performing the above-described processing, the embodiments of this application can significantly improve key performance characteristics of the device, such as propagation loss, backscattering, polarization performance, phase noise, and coherence.

[0038] Other advantages of this application will be described in the following detailed description. Attached Figure Description

[0039] Figure 1 This is a flowchart illustrating a method for fabricating a performance-improved ridge waveguide device structure, provided as a preferred embodiment of this application.

[0040] Figure 2 This is a schematic diagram of a structure formed on a substrate after a first waveguide cladding, a waveguide layer, a first hard mask layer, and a first photoresist pattern are formed, according to a preferred embodiment of this application.

[0041] Figure 3 This is a schematic diagram of the structure after the first hard mask is formed by etching the first hard mask layer through the first photoresist pattern, which is a preferred embodiment of this application.

[0042] Figure 4 This is a schematic diagram of the structure after removing the first photoresist pattern, provided as a preferred embodiment of this application.

[0043] Figure 5 This is a schematic diagram of the structure after partially etching the waveguide layer to form a first waveguide pattern using a first hard mask, according to a preferred embodiment of this application.

[0044] Figure 6This is a schematic diagram of the structure after removing the first hard mask, provided as a preferred embodiment of this application.

[0045] Figure 7 This is a schematic diagram of a structure after forming a second hard mask layer, an anti-reflection layer, and a second photoresist pattern on a waveguide layer, according to a preferred embodiment of this application.

[0046] Figure 8 This is a schematic diagram of the structure after forming a second waveguide pattern by fully etching the waveguide layer through a second hard mask, which is a preferred embodiment of this application.

[0047] Figure 9 This is a schematic diagram of the structure after removing the second hard mask, provided as a preferred embodiment of this application.

[0048] Figure 10 This is a schematic diagram of the structure after forming a second waveguide cladding on a ridge waveguide, according to a preferred embodiment of this application.

[0049] In the figure: 10. Substrate; 11. Waveguide layer; 111. First waveguide pattern; 112. Second waveguide pattern; 113. Ridge waveguide; 12. First photoresist pattern; 13. First hard mask layer; 131. First hard mask; 14. First waveguide cladding; 15. Second hard mask layer; 151. Second hard mask; 16. Second photoresist pattern; 17. Anti-reflection layer; 18. Second waveguide cladding. Detailed Implementation

[0050] To address the issue of sidewall roughness in existing patterned ridge waveguides, which negatively impacts key device performance, this application provides a method for fabricating a performance-improved ridge waveguide device structure, including: Provide substrate; A waveguide layer and a first hard mask are sequentially formed on the substrate; The waveguide layer is partially etched using the first hard mask to form a first waveguide pattern on the waveguide layer; The first sidewall of the first waveguide pattern is subjected to a first processing, which includes multiple alternating processing steps and passivation steps. The processing steps use a first free radical to remove protrusions on the first sidewall to reduce the surface roughness of the first sidewall. The passivation steps use a second free radical to passivate the surface of the first sidewall for protection. The first hard mask is removed, and a second hard mask covering the first waveguide pattern is formed on the waveguide layer. The waveguide layer on both sides of the first waveguide pattern is fully etched through the second hard mask, and a second waveguide pattern is formed below the first waveguide pattern, thereby forming a ridge waveguide including the first waveguide pattern and the second waveguide pattern. Using a third free radical, a second treatment is performed on the second sidewall of the second waveguide pattern. A polymer layer is deposited on the second sidewall using the byproducts generated during the second treatment process to protect the second sidewall while removing the protrusions present on the second sidewall, thereby reducing the surface roughness of the second sidewall. The second hard mask is removed, and the first and second sidewalls are subjected to a third treatment using metastable excited fourth free radicals to remove the remaining protrusions on the first and second sidewalls, thereby further reducing the surface roughness of the first and second sidewalls.

[0051] This application embodiment first processes the first sidewall of the first waveguide pattern and the second sidewall of the second waveguide pattern formed in steps independently (first processing and second processing), which can effectively reduce the surface roughness of the first sidewall and the second sidewall respectively. Then, a third processing with minimal damage is performed on the first sidewall and the second sidewall at low energy, which further reduces the surface roughness of the first sidewall and the second sidewall. Thus, the sidewall roughness of the ridge waveguide as a whole is reduced through step processing, and the surface smoothness of the ridge waveguide sidewall is improved. Therefore, it can significantly improve the key performance of the device such as propagation loss, backscattering, polarization performance, phase noise and coherence.

[0052] This application also provides a ridge waveguide device structure, which is obtained using the above-described improved ridge waveguide device structure fabrication method.

[0053] The specific embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0054] refer to Figure 1 This application provides a method for fabricating a ridge waveguide device structure with improved performance, which may include the following steps in sequence: Step S11: Provide a substrate.

[0055] refer to Figure 2 In some embodiments, a substrate 10 is used for further forming a ridge waveguide on the substrate 10 to form the ridge waveguide device structure provided in the embodiments of this application. The substrate 10 may include any suitable type of semiconductor substrate and material. For example, the substrate 10 may include a silicon (Si) substrate, a germanium (Ge) substrate, or a germanium-silicon (SiGe) substrate, or a III / V compound semiconductor substrate, such as a gallium arsenide (GaAs) substrate, an indium gallium arsenide (InGaAs) substrate, or similar materials.

[0056] In some embodiments, substrate 10 may include a wafer.

[0057] Step S12: Form a waveguide layer and a first hard mask sequentially on the substrate.

[0058] refer to Figure 2 In some embodiments, a deposition process may be used to sequentially form a first waveguide cladding layer 14, a waveguide layer 11, and a first hard mask layer 13 on the surface of the substrate 10.

[0059] In some embodiments, the material of the first waveguide cladding 14 may include silicon dioxide, etc.

[0060] In some embodiments, the waveguide layer 11 may be made of silicon or the like.

[0061] In some embodiments, the material of the first hard mask layer 13 (the first hard mask material) may include silicon dioxide, etc.

[0062] In some embodiments, a spin coating process can be used to form a first photoresist layer on the surface of the first hard mask layer 13. Then, a photolithography process is used to etch the first photoresist layer, forming a first photoresist pattern 12 on the surface of the first hard mask layer 13. Next, a plasma dry etching process can be used, using the first photoresist pattern 12 as a mask, to etch the first hard mask layer 13, stopping at the surface of the waveguide layer 11, thereby forming a first hard mask 131 on the surface of the waveguide layer 11, as shown below. Figure 3 As shown. The first hard mask 131 serves as a mask for subsequent etching of the waveguide layer 11 to form the first waveguide pattern.

[0063] In some embodiments, an etching gas comprising a fluorine-containing gas (e.g., at least one of CF4, C4F8, NF3, etc.), an oxygen-containing gas (e.g., O2), and a dilution gas (e.g., at least one of N2, Ar, etc.) can be used to etch the first hard mask layer 13 to form the first hard mask 131.

[0064] refer to Figure 4 In some embodiments, a photoresist stripping process can be used to remove the first photoresist pattern 12, so that the top surface of the first hard mask 131 is fully exposed.

[0065] Step S13: Partially etch the waveguide layer using the first hard mask to form the first waveguide pattern on the waveguide layer.

[0066] refer to Figure 5 In some embodiments, a plasma dry etching process can be used, with the first hard mask 131 as the etching mask, to etch the waveguide layer 11. A protruding first waveguide pattern 111 can be formed on the etched surface of the waveguide layer 11 by partially etching the waveguide layer 11 (i.e., etching depth is less than the thickness of the waveguide layer 11).

[0067] Step S14: Perform a first treatment on the first sidewall of the first waveguide pattern with the first hard mask on top to reduce the surface roughness of the first sidewall.

[0068] refer to Figure 5 In some embodiments, since the surface roughness of the sidewalls (first sidewalls) of the first waveguide pattern 111 formed by etching is generally poor, the surface roughness of the first sidewalls can be reduced by performing a first treatment on the first sidewalls of the first waveguide pattern 111, which still has the first hard mask 131 on top. The first treatment of the first sidewalls of the first waveguide pattern 111 is performed concurrently with the first hard mask 131. Thus, the first hard mask 131 can be used as a capping layer on top of the first waveguide pattern 111 to protect the apex corners of the first waveguide pattern 111, preventing damage and morphological changes to the apex corners during the first treatment.

[0069] In some embodiments, the first process includes multiple alternating processing steps and a passivation step. The processing steps use a first free radical to remove protrusions (rough peaks) present on the first sidewall, thereby reducing the surface roughness of the first sidewall. The passivation step uses a second free radical to passivate the surface of the first sidewall, protecting it during the processing steps. Thus, through the alternating cycle of processing and passivation steps, a refined roughness reduction process is performed on the first sidewall, effectively reducing its surface roughness.

[0070] In some embodiments, the processing steps in the first process are performed using first free radicals contained in a plasma obtained by exciting a first gas. The first free radicals may include fluorine-containing free radicals, first nitrogen-containing free radicals, and first oxygen-containing free radicals. The fluorine-containing free radicals, first nitrogen-containing free radicals, and first oxygen-containing free radicals are obtained by exciting the first gas to form a plasma and then filtering out charged particles from the plasma.

[0071] In some embodiments, the first gas may include CF4, N2, and O2. Fluorine-containing free radicals are used to remove protrusions on the first sidewall (the main cause of poor roughness), while first nitrogen-containing free radicals and first oxygen-containing free radicals promote the dissociation of CF4 into fluorine-containing free radicals, generating volatile NO / CO groups. This generates more fluorine-containing free radicals, thus efficiently removing protrusions from the first sidewall and improving processing efficiency. Furthermore, the first nitrogen-containing free radicals also have a certain nitriding effect on the first sidewall, which can be used for surface passivation of the first sidewall to protect its surface while removing protrusions.

[0072] In some embodiments, the passivation step in the first treatment is performed using second free radicals contained in a plasma obtained by exciting a second gas. The second free radicals include a second nitrogen-containing free radical and a second oxygen-containing free radical. The second nitrogen-containing free radical and the second oxygen-containing free radical are obtained by exciting a second gas to form a plasma and then filtering out charged particles from the plasma.

[0073] In some embodiments, the second gas may include N2 and O2. The second oxygen-containing free radical is used to perform surface oxidation on the first sidewall, thereby achieving surface passivation and protection of the first sidewall. The second nitrogen-containing free radical promotes the dissociation of O2 into the second oxygen-containing free radical, improving the efficiency of the resulting second oxygen-containing free radical in oxidizing the first sidewall. Thus, the surface oxidation effect generated by the second oxygen-containing free radical can effectively and promptly protect the apex corner of the first waveguide pattern 111, maximizing the preservation of the sharp corner morphology and improving device performance.

[0074] In some embodiments, during the processing steps, the flow rate of O2 is 100 sccm to 2000 sccm, the flow rate of CF4 is 0.5 to 2 times the flow rate of O2, and the flow rate of N2 is 0.5 to 1 times the sum of the flow rates of CF4 and O2. For example, the flow rate of O2 can be 100 sccm, 200 sccm, 400 sccm, 1000 sccm, 1550 sccm, 1840 sccm, or 2000 sccm, or any value between any two of the aforementioned flow rate values. The flow rate of CF4 can be 0.5, 0.6, 0.8, 1, 1.3, 1.6, 1.9, or 2 times the flow rate of O2, or any value between any two of the aforementioned multiple values. The flow rate of N2 can be 0.5 times, 0.65 times, 0.7 times, 0.8 times, 0.9 times, or 1 times the sum of the flow rates of CF4 and O2, or any value between any two of the aforementioned multiples.

[0075] In some embodiments, during the passivation step, the flow rate of O2 is 100 sccm to 2000 sccm, and the flow rate of N2 is 0.1 to 1 times the flow rate of O2. For example, the flow rate of O2 can be 100 sccm, 300 sccm, 500 sccm, 700 sccm, 1001 sccm, 1500 sccm, 1800 sccm, or 2000 sccm, or any value between any two of the aforementioned flow rate values. The flow rate of N2 can be 0.1, 0.3, 0.5, 0.7, 0.9, or 1 times the flow rate of O2, or any value between any two of the aforementioned multiple values.

[0076] In some embodiments, the temperature is 50°C to 350°C when performing the processing step and / or passivation step. For example, the temperature may be 50°C, 70°C, 90°C, 100°C, 120°C, 150°C, 200°C, 250°C, 300°C, or 350°C, or any value between any two of the aforementioned temperature values.

[0077] In some embodiments, the pressure is 500 mTorr to 2000 mTorr during the processing and / or passivation steps. For example, the pressure may be 500 mTorr, 700 mTorr, 900 mTorr, 1000 mTorr, 1200 mTorr, 1400 mTorr, 1700 mTorr, or 2000 mTorr, or any value between any two of the aforementioned pressure values.

[0078] In some embodiments, the source power is 500W to 2000W when performing the processing step and / or passivation step. For example, the source power may be 500W, 750W, 950W, 1005W, 1350W, 1700W, 1850W or 2000W, or any value between any two of the aforementioned power values.

[0079] In some embodiments, the bias power is 1W to 30W when performing the processing step and / or passivation step. For example, the bias power may be 1W, 3W, 6W, 9W, 10W, 15W, 20W, 25W or 30W, or any value between any two of the aforementioned power values.

[0080] In some embodiments, the processing step and / or passivation step takes place over a period of 3 to 20 seconds. For example, the time may be 3 seconds, 6 seconds, 9 seconds, 10 seconds, 12 seconds, 15 seconds, 18 seconds, or 20 seconds, or any value between any two of the aforementioned time values.

[0081] Thus, through the coordinated control of the aforementioned flow rate, flow ratio, temperature, pressure, source power, bias power, and time, the first treatment can be performed at low energy, avoiding direct bombardment by plasma and preventing physical damage to the crystal from high-energy ions. This results in a smooth first sidewall of the first waveguide pattern 111 after the first treatment. Furthermore, the combination of the treatment and passivation steps not only improves the smoothness of the first sidewall but also effectively protects the sidewall morphology, preventing sidewall bending. The number of cycles for the treatment and passivation steps can be determined based on the treatment effect.

[0082] In some embodiments, the first treatment can be performed at a first processing temperature of 200°C to 350°C to repair internal lattice damage (defects such as dislocations and lattice distortions in the shallow layer) present on the first sidewall, thereby achieving a comprehensive treatment effect.

[0083] In some other embodiments, a first free radical contained in the plasma obtained by exciting a fifth gas can be used for the processing steps in the first treatment, and a second free radical contained in the plasma obtained by exciting a sixth gas can be used for the passivation steps in the first treatment. The first free radical may include a third hydrogen-containing free radical, which is obtained by exciting the fifth gas to form plasma and then filtering out charged particles from the plasma. The second free radical may include a hydroxyl free radical, which is obtained by exciting the sixth gas to form plasma and then filtering out charged particles from the plasma.

[0084] In some embodiments, the fifth gas may include H2. Furthermore, a third hydrogen-containing radical can be obtained by exciting H2 with helium metastable particles to form a plasma and filtering out charged particles therein.

[0085] In some embodiments, the sixth gas may include H2O. Furthermore, hydroxyl radicals can be obtained by exciting H2O with helium metastable particles to form a plasma, and then filtering out charged particles therein.

[0086] By exciting H2 and H2O with helium metastable particles, highly reactive third hydrogen-containing free radicals and hydroxyl free radicals are obtained. These can be used to perform low-energy treatment on the first sidewall, avoiding direct plasma bombardment and achieving a gentle, non-destructive surface treatment. The highly reactive third hydrogen-containing free radicals and hydroxyl free radicals arrive at the wafer (substrate) surface as low-energy neutral particles, undergoing only a chemical reaction. This completely avoids the physical bombardment damage to the crystal caused by high-energy ions, thus enabling refined roughness reduction treatment of the first sidewall.

[0087] In the first process, a third hydrogen-containing free radical is used to remove the protrusions on the first sidewall during the process, and a hydroxyl free radical is used to adsorb and form Si-OH bonds as dangling bonds on the first sidewall to passivate the surface of the first sidewall. This avoids the third hydrogen-containing free radical from undergoing an undesirable etching reaction (H·+Si→SiH4↑) with the silicon material on the first sidewall during the process, thus effectively maintaining the critical lateral dimension.

[0088] In some embodiments, the flow rate of H2 during the processing steps is 80 sccm to 150 sccm. For example, the flow rate of H2 can be 80 sccm, 100 sccm, 125 sccm, 140 sccm, or 150 sccm, or any value between any two of the aforementioned flow rate values.

[0089] In some embodiments, the H2O flow rate during the passivation step is 3 sccm to 10 sccm. For example, the H2O flow rate can be 3 sccm, 4 sccm, 8 sccm, 9 sccm, or 10 sccm, or any value between any two of the aforementioned flow rate values.

[0090] In some embodiments, when H2 is used for the processing step and / or H2O is used for the passivation step, the temperature is 110°C to 350°C. For example, the temperature may be 110°C, 150°C, 185°C, 200°C, 250°C, 300°C or 350°C, or any value between any two of the aforementioned temperature values.

[0091] In some embodiments, when H2 is used for the processing step and / or H2O is used for the passivation step, the pressure is 10 mTorr to 50 mTorr. For example, the pressure can be 10 mTorr, 20 mTorr, 30 mTorr, 40 mTorr or 50 mTorr, or any value between any two of the aforementioned pressure values.

[0092] In some embodiments, when H2 is used for the processing step and / or H2O is used for the passivation step, the source power is 400W to 800W. For example, the source power may be 400W, 500W, 600W, 700W or 800W, or any value between any two of the aforementioned power values.

[0093] In some embodiments, when H2 is used for the processing step and / or H2O is used for the passivation step, the bias power is 1W to 30W. For example, the bias power may be 1W, 4W, 7W, 10W, 15W, 20W, 25W or 30W, or any value between any two of the aforementioned power values.

[0094] In some embodiments, when H2 is used for the processing step and / or H2O is used for the passivation step, the time is 3s to 20s. For example, the time can be 3s, 6s, 9s, 10s, 15s or 20s, or any value between any two of the aforementioned time values.

[0095] By coordinating and controlling the flow rate, temperature, pressure, source power, bias power, time, etc., and performing the first treatment at low bias power and lower energy, the direct bombardment of plasma can be avoided, and the physical bombardment damage of high-energy ions to the crystal can be avoided, making the first sidewall smooth after the first treatment.

[0096] In some embodiments, when using H2 and H2O for the first treatment, a third hydrogen-containing free radical can also be used to remove impurities (such as carbon) on the first sidewall and repair lattice damage on the surface of the first sidewall, thereby achieving a comprehensive treatment effect.

[0097] Step S15: After removing the first hard mask, a second hard mask covering the first waveguide pattern is formed on the waveguide layer. The waveguide layers on both sides of the first waveguide pattern are fully etched through the second hard mask to form a second waveguide pattern below the first waveguide pattern.

[0098] refer to Figure 6 In some embodiments, a wet process can be used to remove the first hard mask 131, so that the top surface of the first waveguide pattern 111 is fully exposed. For example, a wet solution comprising hydrofluoric acid and ammonium fluoride can be used to etch and remove the first hard mask 131.

[0099] refer to Figure 7 In some embodiments, a second hard mask layer 15, an anti-reflection layer 17, and a second photoresist pattern 16 may be sequentially formed on the surface of the waveguide layer 11 after the first hard mask 131 has been removed.

[0100] In some embodiments, the material of the second hard mask layer 15 (second hard mask) may include spin-coated carbon (SOC).

[0101] In some embodiments, the antireflective layer 17 may include a silicon-containing antireflective layer (SiARC).

[0102] In some embodiments, a spin-coating process can be used to form a spin-coated carbon layer (second hard mask layer 15) on the surface of the waveguide layer 11 after the first hard mask 131 has been removed, completely covering the top surface of the first waveguide pattern 111. Then, a deposition process can be used to form a silicon-containing anti-reflective layer on the surface of the spin-coated carbon layer. Subsequently, a spin-coating process can be used again to form a second photoresist layer on the surface of the silicon-containing anti-reflective layer, and a photolithography process can be used to lithographically form the second photoresist layer, forming a second photoresist pattern 16 on the surface of the silicon-containing anti-reflective layer. The second photoresist pattern 16 is located above the first waveguide pattern 111, the width of the second photoresist pattern 16 is greater than the width of the first waveguide pattern 111, and the orthographic projection of the second photoresist pattern 16 onto the surface of the waveguide layer 11 needs to completely cover the first waveguide pattern 111.

[0103] refer to Figure 7 and Figure 8In some embodiments, the antireflective layer 17 (containing a silicon antireflective layer) and the second hard mask layer 15 (spin-coated carbon layer) can be sequentially etched using the second photoresist pattern 16 to form a second hard mask 151 covering the first waveguide pattern 111 on the surface of the waveguide layer 11 outside the sides of the first waveguide pattern 111 through pattern transfer. The second hard mask 151 is then used to perform full etching on the surface of the waveguide layer 11 outside the sides of the first waveguide pattern 111 (i.e., etching through the waveguide layer 11 outside the sides of the first waveguide pattern 111, stopping at the surface of the underlying first waveguide cladding 14). A second waveguide pattern 112 is then formed below the first waveguide pattern 111, with a width greater than the width of the first waveguide pattern 111. This results in the formation of a ridge waveguide 113 including the first waveguide pattern 111 and the second waveguide pattern 112 on the surface of the first waveguide cladding 14. By utilizing different selectivity ratios between the spin-coated carbon layer and the silicon-containing anti-reflective layer, after etching, only the second hard mask 151 of the SOC material remains above the ridge waveguide 113, while the silicon-containing anti-reflective layer and the second photoresist pattern 16 are completely etched away. Figure 8 As shown.

[0104] By using spin-coated carbon material to form a second hard mask 151, the high fluidity of the spin-coated carbon material can be utilized to completely cover the formed first waveguide pattern 111, creating a smooth surface and improving pattern transmission quality. Simultaneously, during the etching process to form the second waveguide pattern 112, the second hard mask 151 acts as a hard protective layer, isolating the first waveguide pattern 111 from physical bombardment and chemical corrosion by etching gases. This prevents process damage such as loss, roughness, and defects in the first waveguide pattern 111. Etching is performed only on the silicon material outside the two sides of the first waveguide pattern 111, enabling precise formation of the second waveguide pattern 112 structure and ensuring the integrity and dimensional accuracy of the final ridge waveguide 113 structure.

[0105] After etching to form the second waveguide pattern 112, the second hard mask 151 is retained to provide basic protection and reaction support conditions for the subsequent roughening treatment (second treatment) of the sidewall of the second waveguide pattern 112.

[0106] Step S16: Perform a second treatment on the second sidewall of the second waveguide pattern with the second hard mask on top to reduce the surface roughness of the second sidewall.

[0107] refer to Figure 8In some embodiments, since the surface roughness of the sidewalls (second sidewalls) of the etched second waveguide pattern 112 is generally poor, the surface roughness of the second sidewalls can be reduced by performing a second processing on the second sidewalls of the second waveguide pattern 112, which still has the second hard mask 151 on top. The second processing of the second sidewalls of the second waveguide pattern 112 is performed concurrently with the second hard mask 151. Thus, the second hard mask 151 can be used as a capping layer on top of the second waveguide pattern 112 to protect the apex corners of the second waveguide pattern 112, preventing damage and morphological changes to the apex corners during the second processing.

[0108] In some embodiments, a third free radical is used to perform a second treatment on the second sidewall of the second waveguide pattern 112 with the second hard mask 151 on top. Furthermore, a polymer layer formed on the second sidewall by utilizing byproducts generated during the second treatment can be deposited to protect the second sidewall while removing protrusions present on the second sidewall to reduce its surface roughness.

[0109] In some embodiments, a second treatment is performed using a third free radical contained in a plasma obtained by exciting a third gas. The third free radical may include a first hydrogen-containing free radical. The first hydrogen-containing free radical is obtained by exciting a third gas to form a plasma and then filtering out charged particles from the plasma.

[0110] In some embodiments, the third gas may include H2. The first hydrogen-containing free radical is used to remove protrusions present on the second sidewall. The polymer layer is formed by reacting free active carbon atoms generated after the surface of the second hard mask 151 of the SOC material is bombarded by plasma with the first hydrogen-containing free radical, and the resulting byproducts are deposited on the second sidewall.

[0111] The second sidewall surface of the second waveguide pattern 112 of the etched silicon material often has a large number of microscopic protrusions, peaks, lattice defects, and rough grooves. These are inherent process defects caused by the anisotropic etching of dry etching gas. By performing a second treatment while retaining the second hard mask 151, the high activity and low etching selectivity of the first hydrogen-containing free radical (compared to high-energy etching gas, the first hydrogen-containing free radical has an extremely low etching rate on silicon) can be utilized to selectively react with the microscopic high-energy defects (protrusions) on the second sidewall. (At a certain controllable temperature (e.g., 100℃~350℃), the first hydrogen-containing free radical will not etch a large amount of the flat silicon sidewall substrate plane, but will only preferentially act on the microscopic high-energy defect region of the silicon sidewall.) Through the microscopic modification effect, the surface of the second sidewall can be "shaving peaks and filling valleys" without over-etching the silicon material on the second sidewall, repairing defects such as burrs, unevenness, and microscopic holes on the second sidewall after etching, and reducing the roughness of the second sidewall.

[0112] Meanwhile, the second hard mask 151 of the retained SOC material not only continuously protects the structure of the first waveguide pattern 111, but its high-density carbon content is also a key auxiliary medium for smoothing the second sidewall. On the one hand, by utilizing the second hard mask 151 of the SOC material, excessive first hydrogen-containing free radicals can be prevented from diffusing to the surface of the first waveguide pattern 111, completely eliminating the hydrogenation modification and etching damage to the intact silicon structure of the first waveguide pattern 111, thus ensuring the dimensional accuracy and optical performance of the upper pattern structure of the ridge waveguide 113. On the other hand, relying on the reaction buffering capacity of the free active carbon atoms escaping from the second hard mask 151 of the SOC material under plasma bombardment, they can react with the excess first hydrogen-containing free radicals in the system during the second processing, consume the redundant active hydrogen, suppress the excessive etching of the silicon material on the second sidewall by the first hydrogen-containing free radicals, retain only an appropriate amount of the first hydrogen-containing free radicals for defect modification, and use the reaction byproducts to form a temporary, selective polymer layer on the surface of the second sidewall as a protective film, playing a secondary protection role. This can help ensure that the etching process of the silicon atom layer on the surface of the second sidewall only occurs on the "peaks" that need to be removed, without destroying the overall structure of the surface of the second sidewall.

[0113] In some embodiments, when performing the second process, the flow rate of H2 is 50 sccm to 1000 sccm. For example, the flow rate of H2 can be 50 sccm, 60 sccm, 90 sccm, 100 sccm, 300 sccm, 500 sccm, 800 sccm, or 1000 sccm, or any value between any two of the aforementioned flow rate values.

[0114] In some embodiments, the temperature during the second processing is 100°C to 350°C. For example, the temperature may be 100°C, 110°C, 130°C, 150°C, 180°C, 200°C, 210°C, 250°C, 300°C, 325°C, or 350°C, or any value between any two of the aforementioned temperature values.

[0115] In some embodiments, the pressure during the second processing is 500 mTorr to 2000 mTorr. For example, the pressure may be 500 mTorr, 600 mTorr, 900 mTorr, 1000 mTorr, 1100 mTorr, 1500 mTorr, 1700 mTorr, or 2000 mTorr, or any value between any two of the aforementioned pressure values.

[0116] In some embodiments, when performing the second processing, the source power is 500W to 2000W. For example, the source power may be 500W, 700W, 900W, 1000W, 1250W, 1500W, 1850W or 2000W, or any value between any two of the aforementioned power values.

[0117] In some embodiments, the bias power is 1W to 30W when performing the second processing. For example, the bias power can be 1W, 3W, 5W, 8W, 10W, 15W, 20W, 25W or 30W, or any value between any two of the aforementioned power values.

[0118] In some embodiments, the time for performing the second processing is 30s to 120s. For example, the time can be 30s, 50s, 70s, 90s, 100s, 110s, or 120s, or any value between any two of the aforementioned time values.

[0119] Thus, by coordinating the control of the above-mentioned flow rate, temperature, pressure, source power, bias power, time, etc., and performing the second processing under low bias power and low energy, the second sidewall of the second waveguide pattern 112 after the second processing can be smoothed.

[0120] In some embodiments, during the second treatment, a first hydrogen-containing free radical is also used to remove impurities (such as carbon) on the second sidewall and to repair lattice damage, thereby achieving a comprehensive treatment effect.

[0121] Temperature is a core process parameter for the second treatment, and the entire temperature range offers precise process adaptability. For example, by performing the second treatment at a temperature of 100℃ to 200℃ (the low-temperature range), the activity of the first hydrogen-containing free radicals can be reduced, thus decreasing the reaction rate. In this low-temperature range, the activity of the first hydrogen-containing free radicals is relatively low, resulting in a slower reaction rate. This primarily manifests as impurity removal and minor repair of microscopic defects, with almost no etching of the silicon substrate, making it suitable for finely repairing slightly rough second sidewall structures.

[0122] For example, by performing a second treatment at a temperature of 200℃ to 350℃ (a medium-high temperature range), the defect lattice on the second sidewall is activated, improving the reaction efficiency of the first hydrogen-containing free radical. Within this medium-high temperature range, by appropriately increasing the temperature, the defect lattice of silicon on the second sidewall surface can be activated, increasing the reaction efficiency between the first hydrogen-containing free radical and high-energy defect silicon atoms. This accelerates the etching of the protrusion and the rearrangement of surface atoms, significantly improving the surface smoothness of the second sidewall, making it suitable for repairing relatively rough second sidewall structures. This solves the problems of structural damage, sidewall roughness, and impurity residue that easily occur in dry etching of silicon optical waveguides, while precisely balancing structural protection and surface modification effects. It can effectively reduce the light scattering loss of silicon optical waveguides, significantly improving the forming accuracy and optical transmission performance of optical waveguide devices. It is a key fine modification process for fabricating high-performance silicon-based optical waveguide devices, and the process is simple and highly compatible.

[0123] Step S17: After removing the second hard mask, perform a third treatment on the first and second sidewalls to further reduce the surface roughness of the first and second sidewalls.

[0124] refer to Figure 9 In some embodiments, after the second processing, plasma obtained by exciting a seventh gas can be used to remove the second hard mask 151, completely exposing the top surface of the ridge waveguide 113. The seventh gas may include, for example, N2 and H2, for removing the second hard mask 151 of the SOC material.

[0125] After etching to form the ridge waveguide 113, the sidewalls (first sidewall and second sidewall) of the ridge waveguide 113 may still have a certain roughness. Furthermore, during the first processing, the shading effect of the first hard mask 131 on the apex of the first waveguide pattern 111 may weaken the processing effect at the apex. Similarly, during the second processing, the shading effect of the second hard mask 151 on the apex of the second waveguide pattern 112 may also weaken the processing effect at the apex. Therefore, after removing the second hard mask 151, a third processing can be performed on the first and second sidewalls to refine the apex region of the first waveguide pattern 111 and the apex region of the second waveguide pattern 112, eliminate the surface roughness that may exist in the apex region, and further smooth the entire exposed surface of the ridge waveguide 113 after the third processing.

[0126] In some embodiments, after removing the second hard mask 151, a metastable excited fourth free radical can be used to perform a third treatment on the first sidewall and the second sidewall to remove the remaining protrusions on the first sidewall and the second sidewall, so as to further reduce the surface roughness of the first sidewall and the second sidewall.

[0127] In some embodiments, a third treatment may be performed using a fourth free radical contained in a plasma obtained by exciting a fourth gas. The fourth free radical may include a second hydrogen-containing free radical. The second hydrogen-containing free radical is obtained by exciting a fourth gas to form a plasma and then filtering out charged particles from the plasma.

[0128] In some embodiments, the fourth gas may include H2. Furthermore, a second hydrogen-containing free radical can be obtained by exciting H2 with helium metastable particles to form a plasma and filtering out charged particles therein, which is used to gently remove residual protrusions on the first and second sidewalls.

[0129] In some embodiments, when performing the third processing, the flow rate of H2 is 50 sccm to 1000 sccm. For example, the flow rate of H2 can be 50 sccm, 70 sccm, 85 sccm, 100 sccm, 200 sccm, 550 sccm, 900 sccm or 1000 sccm, or any value between any two of the aforementioned flow rate values.

[0130] In some embodiments, the temperature during the third processing is 50°C to 250°C. For example, the temperature may be 50°C, 70°C, 90°C, 100°C, 130°C, 150°C, 200°C, or 250°C, or any value between any two of the aforementioned temperature values.

[0131] In some embodiments, the pressure during the third processing is 100 mTorr to 2000 mTorr. For example, the pressure may be 100 mTorr, 200 mTorr, 500 mTorr, 800 mTorr, 1000 mTorr, 1200 mTorr, 1400 mTorr, 1800 mTorr, or 2000 mTorr, or any value between any two of the aforementioned pressure values.

[0132] In some embodiments, when performing the third processing, the source power is 200W to 2000W. For example, the source power may be 200W, 500W, 800W, 1000W, 1200W, 1500W, 1800W or 2000W, or any value between any two of the aforementioned power values.

[0133] In some embodiments, when performing the third processing, the bias power is 0W (i.e., the bias power is turned off).

[0134] In some embodiments, the time for performing the third processing is 30s to 180s. For example, the time can be 30s, 50s, 80s, 96s, 100s, 120s, 150s or 180s, or any value between any two of the aforementioned time values.

[0135] By coordinating the control of the above-mentioned flow rate, temperature, pressure, source power, time, etc., and performing the third processing at a lower energy level without bias voltage (bias power off), a gentler and non-damaging surface smoothing process can be achieved. Furthermore, it can refine the morphological defects that may exist at the apex of the first waveguide pattern 111 and the apex of the second waveguide pattern 112 after removing the second hard mask 151, thereby further improving the overall pattern quality of the ridge waveguide 113.

[0136] The embodiments of this application reduce the overall sidewall roughness of the ridge waveguide 113 by implementing step-by-step processing (first processing, second processing, and third processing), and improve the surface smoothness of the sidewall of the ridge waveguide 113. Therefore, it can significantly improve the key performance of the device, such as propagation loss, backscattering, polarization performance, phase noise, and coherence.

[0137] In some embodiments, during the third processing, a second hydrogen-containing free radical can also be used to remove impurities (such as carbon) on the surface of the ridge waveguide 113 and repair lattice damage on the surface of the ridge waveguide 113, thereby achieving a comprehensive processing effect.

[0138] In some other embodiments, N2 can be used as a protective gas, and the ridge waveguide 113 can be annealed at a temperature of 200°C to 400°C to smooth and repair the entire exposed surface of the ridge waveguide 113 and repair lattice damage. Annealing not only comprehensively repairs the inherent defects and lattice damage of the ridge waveguide 113 material, but also utilizes high temperature to improve the surface smoothness of the ridge waveguide 113, further enhancing the key performance of the device and improving the deposition quality of the subsequent film layer (second waveguide cladding).

[0139] Step S18: Form a second waveguide cladding that covers the ridge waveguide on the first waveguide cladding.

[0140] refer to Figure 10 In some embodiments, after annealing, an atomic layer deposition process can be used to form a second waveguide cladding 18 on the surface of the first waveguide cladding 14, and the formed second waveguide cladding 18 covers the exposed surface of the ridge waveguide 113, completely covering the surface of the ridge waveguide 113, including the first waveguide pattern 111 and the second waveguide pattern 112. This forms a smooth, well-shaped ridge waveguide device structure on the substrate 10, significantly improving key performance characteristics such as propagation loss, backscattering, polarization performance, phase noise, and coherence.

[0141] In some embodiments, the material of the second waveguide cladding 18 may include silicon dioxide, etc.

[0142] In a second aspect, embodiments of this application also provide a ridge waveguide device structure, which is obtained using the performance-improved ridge waveguide device structure fabrication method provided in any of the embodiments of the first aspect above.

[0143] refer to Figure 10 In some embodiments, the ridge waveguide device structure is built on a substrate 10, including a first waveguide cladding 14 disposed on the surface of the substrate 10, a ridge waveguide 113 disposed on the surface of the first waveguide cladding 14, and a second waveguide cladding 18 disposed on the surface of the first waveguide cladding 14 and covering the entire surface of the ridge waveguide 113. The first waveguide cladding 14 and the second waveguide cladding 18 form an enclosure of the ridge waveguide 113, thereby forming a ridge waveguide device structure on the substrate 10. The ridge waveguide 113 includes a second waveguide pattern 112 located on the surface of the first waveguide cladding 14, and a first waveguide pattern 111 connected above the second waveguide pattern 112. The lateral width of the second waveguide pattern 112 is greater than the lateral width of the first waveguide pattern 111, forming a ridge-shaped waveguide morphology.

[0144] Specifically, by performing a first processing on the first sidewall of the first waveguide pattern 111, the roughness of the first sidewall is reduced; by performing a second processing on the second sidewall of the second waveguide pattern 112, the surface roughness of the second sidewall is reduced; and by performing a third, essentially non-destructive processing on the first and second sidewalls again at a lower energy level, the surface roughness of the first and second sidewalls is further reduced. Thus, by performing stepwise processing, the overall sidewall roughness of the ridge waveguide 113 is reduced, and the surface smoothness of the ridge waveguide 113 sidewalls is improved. Therefore, it can significantly improve key performance characteristics of the device, such as propagation loss, backscattering, polarization performance, phase noise, and coherence.

[0145] In a third aspect, embodiments of this application also provide a plasma processing apparatus for performing the performance-improved ridge waveguide device structure fabrication method corresponding to the above embodiments to fabricate the ridge waveguide device structure corresponding to the above embodiments. The plasma processing apparatus includes inductively coupled plasma (ICP) processing equipment or capacitively coupled plasma (CCP) processing equipment, etc.

[0146] In other aspects, embodiments of this application also provide an electronic device, including a ridge waveguide device structure obtained using the performance-improved ridge waveguide device structure fabrication method of the above embodiments. The electronic device can be a storage device, mobile phone, computer, tablet computer, electronic instrument, television, artificial intelligence device, etc.

[0147] In summary, the embodiments of this application first process the first sidewall of the first waveguide pattern 111 and the second sidewall of the second waveguide pattern 112 formed in steps independently (first processing and second processing), which can effectively reduce the surface roughness of the first sidewall and the second sidewall respectively. Then, a third processing with minimal damage is performed on the first sidewall and the second sidewall at a lower energy, which further reduces the surface roughness of the first sidewall and the second sidewall. Thus, the overall sidewall roughness of the ridge waveguide 113 is reduced through step processing, and the surface smoothness of the sidewall of the ridge waveguide 113 is improved. Therefore, it can significantly improve the key performance of the device, such as propagation loss, backscattering, polarization performance, phase noise and coherence.

[0148] The above are merely preferred embodiments of this application. These embodiments are not intended to limit the scope of protection of this application. Therefore, any equivalent changes made based on the description and drawings of this application should also be included within the scope of protection of this application.

Claims

1. A method for fabricating a ridge waveguide device structure with improved performance, characterized in that, include: Provide substrate; A waveguide layer and a first hard mask are sequentially formed on the substrate; The waveguide layer is partially etched using the first hard mask to form a first waveguide pattern on the waveguide layer; The first sidewall of the first waveguide pattern is subjected to a first processing, which includes multiple alternating processing steps and passivation steps. The processing steps use a first free radical to remove protrusions on the first sidewall to reduce the surface roughness of the first sidewall. The passivation steps use a second free radical to passivate the surface of the first sidewall for protection. The first hard mask is removed, and a second hard mask covering the first waveguide pattern is formed on the waveguide layer. The waveguide layer on both sides of the first waveguide pattern is fully etched through the second hard mask, and a second waveguide pattern is formed below the first waveguide pattern, thereby forming a ridge waveguide including the first waveguide pattern and the second waveguide pattern. Using a third free radical, a second treatment is performed on the second sidewall of the second waveguide pattern. A polymer layer is deposited on the second sidewall using the byproducts generated during the second treatment process. This protects the second sidewall while removing protrusions to reduce its surface roughness. The material of the second hard mask includes spin-coated carbon, and the third free radical includes a first hydrogen-containing free radical. The polymer layer is formed by reacting free active carbon atoms generated after the surface of the second hard mask is bombarded by plasma with the first hydrogen-containing free radical, and the resulting byproducts are deposited on the second sidewall. The second hard mask is removed, and the first and second sidewalls are subjected to a third treatment using metastable excited fourth free radicals to remove the remaining protrusions on the first and second sidewalls, thereby further reducing the surface roughness of the first and second sidewalls.

2. The method for fabricating the performance-improved ridge waveguide device structure according to claim 1, characterized in that, The waveguide layer is made of silicon, the first free radical includes a fluorine free radical, a nitrogen free radical and a oxygen free radical, the second free radical includes a second nitrogen free radical and a second oxygen free radical, and the fourth free radical includes a second hydrogen free radical.

3. The method for fabricating the performance-improved ridge waveguide device structure according to claim 2, characterized in that, The fluorine-containing free radical, the first nitrogen-containing free radical, and the first oxygen-containing free radical are obtained by exciting a first gas, which includes CF4, N2, and O2. The fluorine-containing free radical is used to remove the protrusions on the first sidewall. The first nitrogen-containing free radical and the first oxygen-containing free radical are used to promote the dissociation of CF4. The first nitrogen-containing free radical is also used to passivate the surface of the first sidewall for protection. The second nitrogen-containing free radical and the second oxygen-containing free radical are obtained by exciting a second gas, which includes N2 and O2. The second nitrogen-containing free radical is used to promote the dissociation of O2 to obtain the second oxygen-containing free radical. The second oxygen-containing free radical is used to oxidize the surface of the first sidewall to achieve surface passivation. The first hydrogen-containing free radical is obtained by exciting a third gas, the third gas including H2, and the first hydrogen-containing free radical is used to remove the protrusions present on the second sidewall; The second hydrogen-containing free radical is obtained by exciting a fourth gas, which includes H2. The H2 is excited to form a plasma by using helium metastable particles, and the charged particles therein are filtered out to obtain the second hydrogen-containing free radical. The second hydrogen-containing free radical is used to remove the protrusions remaining on the first sidewall and the second sidewall.

4. The method for fabricating the performance-improved ridge waveguide device structure according to claim 3, characterized in that, During the aforementioned processing steps, the flow rate of O2 is 100 sccm to 2000 sccm, the flow rate of CF4 is 0.5 to 2 times the flow rate of O2, the flow rate of N2 is 0.5 to 1 times the sum of the flow rates of CF4 and O2, the temperature is 50℃ to 350℃, the pressure is 500 mTorr to 2000 mTorr, the source power is 500 W to 2000 W, the bias power is 1 W to 30 W, and the time is 3 s to 20 s; and / or, during the aforementioned passivation steps, the flow rate of O2 is 100 sccm to 2000 sccm, the flow rate of N2 is 0.1 to 1 times the flow rate of O2, the temperature is 50℃ to 350℃, the pressure is 500 mTorr to 2000 mTorr, the source power is 500 W to 2000 W, the bias power is 1 W to 30 W, and the time is 3 s to 20 s; And / or, when performing the second treatment, the flow rate of H2 is 50 sccm to 1000 sccm, the temperature is 100℃ to 350℃, the pressure is 500 mTorr to 2000 mTorr, the source power is 500 W to 2000 W, the bias power is 1 W to 30 W, and the time is 30 s to 120 s. And / or, when performing the third process, the flow rate of H2 is 50 sccm to 1000 sccm, the temperature is 50℃ to 250℃, the pressure is 100 mTorr to 2000 mTorr, the source power is 200 W to 2000 W, the bias power is 0 W, and the time is 30 s to 180 s.

5. The method for fabricating the performance-improved ridge waveguide device structure according to claim 1, characterized in that, When performing the first processing, the first hard mask is used to protect the apex corner of the first waveguide pattern; and / or, when performing the second processing, the second hard mask is used to protect the apex corner of both the first waveguide pattern and the second waveguide pattern.

6. The method for fabricating the performance-improved ridge waveguide device structure according to claim 2, characterized in that, During the first treatment, a first treatment temperature of 200℃ to 350℃ is used to repair lattice damage on the first sidewall; and / or, during the second treatment, the first hydrogen-containing free radical is used to remove impurities on the second sidewall and repair lattice damage, wherein the second treatment is performed at a temperature of 100℃ to 200℃ to reduce the activity of the first hydrogen-containing free radical and decrease the reaction rate, or the second treatment is performed at a temperature of 200℃ to 350℃ to activate the defective lattice on the second sidewall and improve the reaction efficiency of the first hydrogen-containing free radical; and / or, during the third treatment, the second hydrogen-containing free radical is used to remove impurities on the surface of the ridge waveguide and repair lattice damage.

7. The method for fabricating the performance-improved ridge waveguide device structure according to claim 2, characterized in that, Another implementation of the first free radical is to replace the fluorine-containing free radical, the first nitrogen-containing free radical, and the first oxygen-containing free radical with a third hydrogen-containing free radical. Another implementation of the second free radical is to replace the second nitrogen-containing free radical and the second oxygen-containing free radical with hydroxyl free radicals. The third hydrogen-containing free radical is obtained by exciting a fifth gas, which includes H2. The H2 is excited using helium metastable particles to form a plasma, and the charged particles are filtered out to obtain the third hydrogen-containing free radical. The hydroxyl free radical is obtained by exciting a sixth gas, which includes H2O. The H2O is excited using helium metastable particles to form a plasma, and the charged particles are filtered out to obtain the hydroxyl free radical. The third hydrogen-containing free radical is used to remove the protrusions on the first sidewall, and the hydroxyl free radical is used to adsorb and form Si-OH bonds on the first sidewall to passivate the surface of the first sidewall, thereby preventing the third hydrogen-containing free radical from reacting with silicon during the processing steps.

8. The method for fabricating the performance-improved ridge waveguide device structure according to claim 7, characterized in that, During the aforementioned processing steps, the flow rate of H2 is 80 sccm to 150 sccm, the temperature is 110°C to 350°C, the pressure is 10 mTorr to 50 mTorr, the source power is 400 W to 800 W, the bias power is 1 W to 30 W, and the time is 3 s to 20 s; and / or, during the aforementioned passivation steps, the flow rate of H2O is 3 sccm to 10 sccm, the temperature is 110°C to 350°C, the pressure is 10 mTorr to 50 mTorr, the source power is 400 W to 800 W, the bias power is 1 W to 30 W, and the time is 3 s to 20 s.

9. The method for fabricating the performance-improved ridge waveguide device structure according to claim 7, characterized in that, During the first treatment, the third hydrogen-containing free radical is also used to remove impurities on the first sidewall and repair lattice damage.

10. A ridge waveguide device structure, characterized in that, The improved ridge waveguide device structure was obtained using the fabrication method described in any one of claims 1-9.

Images