Process debugging method for preparing thin films by reactive magnetron sputtering

Abstract

The present invention provides a process debugging method for preparing a thin film by using reactive magnetron sputtering, and the method comprises: step one, determining a target working power of reactive magnetron sputtering and a target flow rate of a working gas according to an arc discharge condition on a target material surface; step two, based on the target working power and the target flow rate of the working gas, measuring the surface roughness of a thin film prepared by reactive magnetron sputtering at different powers of an auxiliary source, and determining the auxiliary source power corresponding to a surface roughness of less than 0.2 nm as the target auxiliary source power; step three, based on the target working power, the target flow rate of the working gas and the target auxiliary source power, using reactive gases with different flow rates, testing the absorption loss of a thin film on a substrate surface by a weak absorption analysis method, the absorption loss being used to characterize the reaction degree of the thin film prepared by reactive magnetron sputtering, and determining the reaction gas flow rate corresponding to an absorption loss of less than 10 ppm as the target reaction gas flow rate.

Description

Technical Field

[0001] The invention relates to the technical field of optical coating and precision measurement, and in particular to a process debugging method for preparing a thin film by using reactive magnetron sputtering. Background Art

[0002] Physical vapor deposition is currently the main means of producing dielectric optical films, including electron beam evaporation, magnetron sputtering, ion beam sputtering, ion-assisted deposition and other methods.

[0003] Electron beam evaporation has the characteristics of low cost and fast plating rate, but the plated film has a loose columnar structure, which easily absorbs water to cause refractive index changes and optical absorption in the infrared band. In addition, the film surface roughness is poor, which will produce significant scattering losses, which seriously limits the application of electron beam evaporation in the field of high-quality optical films. Optical films prepared by ion beam sputtering usually have the lowest absorption loss and scattering loss, but the equipment is expensive and maintenance is complicated, making the cost of high-quality optical films (such as mirrors with a reflectivity greater than 99.999%) high.

[0004] Among them, reactive magnetron sputtering equipment has a moderate cost, but problems such as arcing, target poisoning, and hysteresis have caused great difficulties in the process parameter debugging of reactive magnetron sputtering equipment, severely limiting its application in the production of high-quality optical films. Summary of the invention

[0005] In view of this, in order to at least partially solve at least one of the above-mentioned technical problems, the present invention provides a process debugging method for preparing a thin film by reactive magnetron sputtering.

[0006] According to an embodiment of one aspect of the present invention, a process debugging method for preparing a thin film by using reactive magnetron sputtering is provided, comprising: step one, determining a target working power of the reactive magnetron sputtering and a target flow rate of the working gas according to the arc discharge condition on the surface of the target material; step two, based on the target working power and the target flow rate of the working gas, measuring the surface roughness of the thin film prepared by the reactive magnetron sputtering at different powers of the auxiliary source, and determining the auxiliary source power corresponding to the surface roughness of less than 0.2 nm as the target auxiliary source power; step three, based on the target working power, the target flow rate of the working gas and the target auxiliary source power, using reactive gases with different flow rates, testing the absorption loss of the thin film on the surface of the substrate by a weak absorption analysis method, the absorption loss is used to characterize the reaction degree of the thin film prepared by reactive magnetron sputtering, and determining the reaction gas flow rate corresponding to the absorption loss below 10 ppm as the target reaction gas flow rate.

[0007] According to an embodiment of the present invention, the reaction gas includes oxygen or nitrogen, and the working gas is argon; the film thickness in step 2 is ≥100 nm; and the film thickness in step 3 is 450-550 nm.

[0008] According to an embodiment of the present invention, the reaction degree includes the content of suboxide in an oxide film formed by preparing a thin film by reactive magnetron sputtering or the content of nitride in a nitride film.

[0009] According to an embodiment of the present invention, the auxiliary source power and / or the flow rate of the reaction gas is increased; the auxiliary source includes a radio frequency (RF) bias or a plasma source.

[0010] According to an embodiment of the present invention, the flow rate of the reaction gas is increased in increments of 1 to 4 sccm.

[0011] According to an embodiment of the present invention, when the absorption loss is lower than 1000 ppm, the flow rate is increased by 1 sccm.

[0012] According to an embodiment of the present invention, when the absorption loss just drops below 10 ppm, the reaction gas flow rate at this time is determined as the target reaction gas flow rate.

[0013] According to an embodiment of the present invention, the target material for preparing the thin film by reactive magnetron sputtering is titanium, hafnium, niobium, tantalum or silicon; the substrate for preparing the thin film is a silicon substrate or a quartz substrate.

[0014] According to an embodiment of the present invention, the process debugging method further includes: performing post-annealing treatment on the thin film prepared by reactive magnetron sputtering.

[0015] According to an embodiment of the present invention, the post-annealing treatment lasts for 10 hours, the heating and cooling rate is 5°C / min, and the annealing temperature of the post-annealing treatment is ≤ the crystallization temperature of the oxide - 100°C or the crystallization temperature of the nitride - 100°C.

[0016] According to an embodiment of the present invention, by first determining the target working power of the reactive magnetron sputtering and the target flow rate of the working gas, and then determining the target auxiliary source power, and then based on the above-determined parameters, using a weak absorption analysis method to test the absorption loss, a systematic and efficient adjustment of the process parameters of the reactive magnetron sputtering equipment is achieved, and arcing is avoided, and target poisoning and hysteresis phenomena are reduced, which helps to continuously and stably prepare high-quality optical films with low optical losses.

[0017] The present invention proposes a systematic process for process debugging of reactive magnetron sputtering equipment, which is conducive to reducing the dependence on the experience of the debugging personnel and improving the efficiency of the debugging process. On the auxiliary source assisted reactive magnetron sputtering equipment, the thin film prepared by this method can achieve atomic-level flatness, and the film absorption loss can reach the sub-ppm (0.4~0.9ppm) level, which is basically the same level as the ion beam sputtering equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The above and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:

[0019] Figure 1 A schematic structural diagram of a reactive magnetron sputtering device according to an embodiment of the present invention is shown;

[0020] Figure 2 A flow chart showing a process debugging method for preparing a thin film by reactive magnetron sputtering according to an embodiment of the present invention is shown;

[0021] Figure 3 A schematic diagram showing the variation of absorption loss with oxygen flow rate during oxygen flow rate adjustment using weak absorption analysis during the preparation of an oxide film according to an embodiment of the present invention;

[0022] Figure 4 The scanning electron microscope image of the cross section of the alternating film layer prepared by weak absorption analysis and debugging in Example 1 of the present invention is shown;

[0023] Figure 5 A scanning electron microscope image of a cross section of an alternating film layer prepared in Comparative Example 1 without weak absorption analysis and debugging is shown.

[0024] [Description of Reference Numerals]

[0025] 1- Power supply;

[0026] 2- Target material;

[0027] 3- Working gas vent;

[0028] 4- Target gun;

[0029] 5- auxiliary source;

[0030] 6-reaction gas vent;

[0031] 7-RF power supply;

[0032] 8-Metal grid;

[0033] 9-workpiece tray;

[0034] 10-RF bias power supply;

[0035] 11- Chamber vent. DETAILED DESCRIPTION

[0036] Below, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present invention. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of embodiments of the present invention. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of concepts of the present invention.

[0037] The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. The term "comprising" used herein indicates the existence of features, steps, operations, but does not exclude the existence or addition of one or more other features.

[0038] In the case of using expressions such as "at least one of A, B, and C, etc.", it should generally be interpreted in accordance with the meaning of the expression generally understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.). In the case of using expressions such as "at least one of A, B, or C, etc.", it should generally be interpreted in accordance with the meaning of the expression generally understood by those skilled in the art (for example, "a system having at least one of A, B, or C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0039] In the process of coating using reactive magnetron sputtering equipment, problems such as arcing, target poisoning and hysteresis are prone to occur.

[0040] Arcing refers to the arc discharge phenomenon between the target and the anode, which will damage the target and affect the cleanliness of the film surface. Target poisoning refers to the reaction gas forming a thin layer of dielectric material on the surface of the target, which will produce a capacitor-like effect, accumulating positive charge on the surface of the dielectric material, thereby reducing the energy of the sputtered ions when they reach the target surface, causing the sputtering rate to decrease, and may even reduce the sputtering rate to zero, causing the target to fail to work. The first case of the hysteresis effect can be understood as when there is too much reaction gas, the coating rate will be reduced through target poisoning. At this time, the combined reaction gas is reduced, the reaction gas accumulates too much, and the coating rate further decreases. The two processes promote each other, resulting in a continuous drift in the coating rate. The second case of the hysteresis phenomenon can be understood as when there is too little reaction gas, there will be a similar mechanism that makes the reaction gas more insufficient, and the film stoichiometric mismatch continues to deteriorate. However, both processes are relatively long in time, and the coating rate will not produce significant differences within a few hours, making the process debugging of reactive magnetron sputtering require a long time period.

[0041] In the process of realizing the concept of the present invention, it is found that by observing the arc discharge on the surface of the target material, the target working power and the target flow rate of the working gas are first determined, and then the target auxiliary source power is determined based on the above parameters. Finally, based on the various parameters that have been determined, different flow rates of reactive gases are used to test the absorption loss of the reactive magnetron sputtering film prepared on the surface of the target material using weak absorption analysis, thereby determining the target reactive gas flow rate. Based on the above steps, the present invention can systematically and efficiently debug the process parameters of the reactive magnetron sputtering equipment and avoid arcing, target poisoning and hysteresis.

[0042] The invention provides a reactive magnetron sputtering device. Figure 1 FIG. 2 shows a schematic diagram of the structure of a reactive magnetron sputtering device according to an embodiment of the present invention. Figure 1 As shown, it includes: power supply 1, target material 2, working gas vent 3, target gun 4, auxiliary source 5, reaction gas vent 6, RF power supply 7, metal grid 8, workpiece disk 9, RF bias power supply 10, chamber vent 11. It should be noted that the vacuum chamber, vacuum pump group and coating monitoring device are not shown in FIG. Figure 1 The vacuum chamber and vacuum pump group are used to provide a vacuum environment for the coating process. The coating monitoring device is used to observe and control the relevant parameters in the coating process.

[0043] The target gun 4 is connected to the target material 2 and is electrically connected to the power supply 1. The power supply 1 may include a DC power supply or a DC pulse power supply. The power supply 1 applies an electric field to perform magnetron sputtering. The target gun 4 contains a magnet that can constrain the electrons above the target material 2 to move in a circular motion through a magnetic field, continuously ionize the working gas, and the generated positive ions are accelerated toward the target material 2 under the action of an external electric field, thereby sputtering out atoms in the target material 2, and these atoms are deposited on the substrate to form the desired film. There can be one or more target guns 4, and the present invention does not specifically limit the number of target guns 4. A baffle can also be set on the target gun 4 as needed. The working gas is introduced into the chamber through the working gas vent 3.

[0044] The auxiliary source 5 can be at least one of a radio frequency bias, a plasma source, and an ion source, preferably an inductive radio frequency plasma source. Radio frequency bias (RF bias) generates plasma using radio frequency coupling capacitors. Due to the asymmetry of the workpiece disk 9 and the chamber, a negative bias is generated on the surface of the workpiece disk 9, thereby accelerating the bombardment of the substrate by positive ions, promoting the reaction of the reaction gas with the sputtered target atoms and improving the surface flatness of the film, while the plasma source generates plasma using radio frequency coupling capacitors, which has a similar effect to RF bias. The radio frequency power supply 7 introduces radio frequency current into the coil wrapped around the quartz cup to excite and maintain the plasma. The reaction gas on which the plasma is generated is introduced through the reaction gas vent 6, and auxiliary gases such as argon can also be introduced into the reaction gas. The metal grid 8 has openings, and a bias is generated by an external electric field or because of the presence of radio frequency plasma, which provides an acceleration effect for the positive ions coming out of the auxiliary source 5, and provides a certain directionality, so that the ion beam is focused on the film area. The radio frequency bias power supply 10 applies radio frequency bias and is connected to the workpiece disk 9, which can generate capacitively coupled plasma, which helps to improve the atomic mobility of the generated film surface, thereby reducing the roughness of the film. In addition, gas can be introduced into the chamber through the chamber vent 11 as needed.

[0045] Preferably, the frequency of the radio frequency power source 7 of the present invention is 13.56 MHz.

[0046] Preferably, the energy utilization efficiency of the ion beam is improved through the superposition effect of the metal grid 8 and the auxiliary source 5 .

[0047] It should be noted that with the rapid development of optical atomic clocks, cavity-enhanced trace gas detection, ring laser gyroscopes and other fields, the demand for high-quality optical films such as high-reflectivity mirrors is also increasing. The performance of these devices is positively correlated to the finesse of the internal optical cavity, and the finesse of the cavity is positively correlated to the finesse of the internal optical cavity. Reflectivity of cavity mirror pass In correlation, the reflectivity of the cavity mirror is mainly limited by the absorption loss and scattering loss of the film. Therefore, the absorption loss and scattering loss levels of the film are key parameters for measuring the quality of optical films, and these two losses are closely related to the coating process parameters.

[0048] The following will describe the process of adjusting the process parameters for preparing the thin film using the reactive magnetron sputtering device.

[0049] Specifically, according to an embodiment of one aspect of the present invention, a process debugging method for preparing a thin film by reactive magnetron sputtering is provided. Figure 2 FIG. 1 is a flow chart showing a process debugging method for preparing a thin film by reactive magnetron sputtering according to an embodiment of the present invention. Figure 2 As shown, the process debugging method includes steps one to three.

[0050] In step one, the target working power and the target flow rate of the working gas for reactive magnetron sputtering are determined according to the arc discharge condition on the target surface.

[0051] An arc phenomenon will occur between the target material 2 and the anode cover. This is because the voltage or gas pressure is too high, which may cause a breakdown discharge phenomenon. The arc phenomenon can be seen directly from the observation window, or it can be judged based on whether there is a sudden increase in current in the power supply. To avoid the occurrence of arc phenomenon, the working gas flow rate should not be too high, so as to keep the gas pressure in the chamber at 2~3mTorr. The working power or working voltage needs to be adjusted. If an arc phenomenon is observed, the working power or working voltage needs to be reduced. In the process of adjusting the working power or working voltage, the power of the auxiliary source 5 can be selected as half of the rated power of the auxiliary source 5, so that the reaction gas flow rate is sufficient to make the coating color transparent.

[0052] In step 2, based on the target working power and the target flow rate of the working gas, the surface roughness of the film prepared by reactive magnetron sputtering is measured at different powers of the auxiliary source 5, and the auxiliary source power corresponding to the surface roughness less than 0.2 nm is determined as the target auxiliary source power.

[0053] After step 1, the target working power and the target flow rate of the working gas are fixed, and the roughness of the coating is preferably tested using an atomic force microscope to adjust the auxiliary source power. The roughness of the optical film is related to the scattering loss (scatter), and the two meet .in, is the root mean square of the film surface height, which is used to measure the roughness level of the film. Indicates the wavelength of incident light. The film roughness is less than 0.2nm as the standard for debugging. By optimizing the surface roughness of the film, the optical scattering loss of the film is suppressed.

[0054] It should be noted that in the process of film coating adjustment based on roughness, it is preferred to use a silicon wafer with a surface roughness of less than 0.1nm as the substrate, which can reduce the error introduced in the film roughness test process. When the roughness of the coating is tested using an atomic force microscope, it can more accurately reflect the roughness level of the coating. The scanning area of ​​the atomic force microscope is preferably ≤10×10μm 2 .

[0055] In step three, based on the target working power, the target flow rate of the working gas and the target auxiliary source power, reaction gases with different flow rates are used to test the absorption loss of the thin film on the substrate surface by a weak absorption analysis method. The absorption loss is used to characterize the degree of reaction of the thin film prepared by reactive magnetron sputtering, and the reaction gas flow rate corresponding to the absorption loss below 10 ppm is determined as the target reaction gas flow rate.

[0056] In the present invention, the prepared thin film can be understood as a thin film having an atomically flat surface, low absorption loss and uniform thickness.

[0057] After the target working power, the target flow rate of the working gas, and the target auxiliary source power are fixed through steps one and two, the reaction gas flow rate is adjusted according to the thin film absorption loss level measured by the weak absorption analyzer. When the flow rate of the reaction gas introduced is insufficient, there are a large number of unpaired electrons in the target element of the film, which will produce significant absorption loss. At this time, the thickness of the oxide layer on the surface of the target may also gradually become thinner, causing the coating rate to gradually increase over time (a case of hysteresis); the absorption loss will be further aggravated due to insufficient reaction. Therefore, the absorption loss is used to characterize the reaction degree of the film prepared by reactive magnetron sputtering, which can effectively determine whether there is a shortage of reaction gas. When the absorption loss is below 10ppm, it can be understood that the flow rate of the reaction gas is just right.

[0058] The present invention screened the equipment used in the characterization process of optical film absorption loss, and tried ellipsometer, spectrophotometer, and weak absorption analyzer respectively. Among them, the measurement accuracy of ellipsometer for material extinction coefficient is only 10 -2 The measurement accuracy of the absorption loss is 10 -1 The spectrophotometer can measure the extinction coefficient of materials with an accuracy of 10 -4 The measurement accuracy of the absorption loss is 10 -3 In the screening process, it was found that the method of the weak absorption analyzer of the present invention can measure the absorption loss with an accuracy of 10 -7 The magnitude of this factor can be used to adjust the process parameters for preparing thin films by reactive magnetron sputtering.

[0059] According to the embodiment of the present invention, suitable process parameters are determined in sequence through step 1, step 2 and step 3. The arc phenomenon is avoided through step 1, so that the target material can work smoothly; the appropriate auxiliary source power is determined by step 2, which is conducive to reducing the surface roughness of the film prepared by reactive magnetron sputtering and improving the surface flatness of the film; the reaction gas flow rate is set in a suitable range through step 3, which reduces the target poisoning phenomenon and hysteresis phenomenon, and is conducive to the continuous and stable preparation of high-quality optical films with low optical loss.

[0060] The present invention proposes a systematic process for process debugging of reactive magnetron sputtering equipment, which is conducive to reducing the dependence on the experience of the debugging personnel and improving the efficiency of the debugging process. On the auxiliary source assisted reactive magnetron sputtering equipment, the thin film prepared by this method can achieve atomic-level flatness, and the film absorption loss can reach the sub-ppm (0.4~0.9ppm) level, which is basically the same level as the ion beam sputtering equipment.

[0061] According to an embodiment of the present invention, the specific operation process of step one may include the following operations: after the magnetron sputtering starts (the step from power supply being turned on to a stable working state), gradually increase the power (or voltage) applied to the target surface, and observe through the observation window whether there is an arc on the target surface. The power (or voltage) allowed when there is no arc is used as the power (or voltage) range for normal operation of the reactive magnetron sputtering thin film preparation equipment, and determine the highest possible working power within the range, that is, the target working power.

[0062] Furthermore, the flow rate of the working gas during magnetron sputtering is increased to observe whether an arc occurs. In the same manner as above, the flow rate range of the working gas allowed when the reactive magnetron sputtering thin film preparation equipment is working normally is determined, and the working gas flow rate that is as high as possible within the range is determined, that is, the target flow rate of the working gas.

[0063] According to an embodiment of the present invention, the reaction gas includes oxygen or nitrogen, and the working gas is argon; the film thickness in step 2 is ≥100nm; the film thickness in step 3 is 450~550nm. The reaction gas can select oxygen or nitrogen as needed, oxygen is used as the reaction gas when preparing the oxide film, and nitrogen is used as the reaction gas when preparing the nitride film. The working gas can use argon, neon, krypton, etc., for example. In terms of work efficiency and cost-effectiveness, the working gas is preferably argon. In step 2, an atomic force microscope is preferably used to measure the surface roughness of the film prepared by reactive magnetron sputtering. When the film thickness is less than 100nm, the roughness detected by atomic force microscopy may be affected by the roughness of the substrate. When the film thickness is above 100nm, the surface roughness of the coating can be reflected. Preferably, the thickness of the film is less than 1μm, and more preferably, the thickness of the coating is less than or equal to 500nm.

[0064] The coating thickness in step three is affected by the weak absorption analyzer. When the thickness is within the above range, the weak absorption analyzer can more accurately test the absorption loss of the film.

[0065] According to an embodiment of the present invention, the reaction degree includes the content of suboxide in an oxide film formed by preparing a thin film by reactive magnetron sputtering or the content of nitride in a nitride film.

[0066] It can be understood that absorption loss refers to the attenuation of light intensity due to the absorption of light by the material when light passes through the material. In other words, the absorbed light intensity is stronger than the incident light. When the film's absorption loss of light is 1ppm, it means that one millionth of the incident light is absorbed.

[0067] The reason why oxide or nitride dielectric films are often used in visible or near-infrared optical thin film materials, such as silicon oxide and tantalum oxide, is that they have large band gaps and there is no energy level absorption of electrons transitioning from the valence band to the conduction band in the visible or near-infrared range, so they can be considered to be basically transparent to visible light or near-infrared light. However, due to defects inside the material, such as the presence of suboxides or nitrides, the film will still have a small amount of absorption at this time. The presence of suboxides or nitrides means that some electrons in the atoms are not bonded, and these electrons will introduce band gap states into the energy band structure, resulting in absorption loss. Therefore, the magnitude of the absorption loss is also positively correlated with the content of suboxides or nitrides (that is, positively correlated with the degree of oxygen or nitrogen deficiency in the material). In the case of a good reaction degree, it means that the content of suboxides in the oxide film or the content of nitrides in the nitride film is approximately 0, that is, the absorption loss of the film prepared by reactive magnetron sputtering is low, so that the film is fully oxidized or nitrided.

[0068] Specifically, taking the preparation of tannin oxide film as an example, Figure 3 The figure shows the schematic diagram of the absorption loss changing with the oxygen flow rate during the preparation of the tantalite film according to the embodiment of the present invention using weak absorption analysis to debug the oxygen flow rate. Figure 3 As shown, by using oxygen flow rates of 16 sccm, 19 sccm, 22 sccm, and 25 sccm respectively to coat films on a fused quartz substrate and measuring the absorption loss of a single-layer tantalum oxide film, when the film loss drops to a certain level and no longer changes, the oxygen flow rate at this time can be used as the working oxygen flow rate.

[0069] According to an embodiment of the present invention, the auxiliary source power and / or the flow rate of the reaction gas are increasing. Testing the auxiliary source power and the flow rate of the reaction gas in an increasing manner makes it easier to find a suitable target auxiliary source power and a target reaction gas flow rate. The auxiliary source includes a radio frequency bias or a plasma source. The radio frequency bias can generate an AC electric field on the substrate, enhance the acceleration of ions in the plasma toward the substrate, and thus increase the energy of ions bombarding the substrate. The plasma source can generate high-density free electrons and ions, which can better activate the reaction gas and make the oxidation or nitridation reaction more sufficient.

[0070] Preferably, during the debugging process, an increasing reaction gas flow rate may be used to plate a single-layer dielectric film with a thickness of about 500 nm on the quartz substrate, and a weak absorption analyzer may be used to measure the absorption loss of the dielectric film.

[0071] More preferably, the incremental increase of the reaction gas flow rate is 1-4 sccm, for example, 1 sccm, 2 sccm, 3 sccm or 4 sccm. When the incremental increase is adjusted within the above range, it helps to find a more suitable reaction gas flow rate more accurately.

[0072] Particularly preferably, when the absorption loss is less than 1000 ppm, the flow rate increment is 1 sccm. When the absorption loss is less than the above range, continue to refer to Figure 3 It can be seen that when the oxygen flow rate is 22 sccm, the absorption loss is about 10000 ppm; when it is 25 sccm, the absorption loss is close to 0. Therefore, it is necessary to reduce the flow rate increment when it is lower than 1000 ppm, so that it is easier to find a suitable target reaction gas flow rate.

[0073] According to an embodiment of the present invention, when the absorption loss just drops below 10ppm, the reaction gas flow rate at this time is determined as the target reaction gas flow rate. It should be noted that with the increase of the reaction gas flow rate, the absorption loss of the film gradually decreases. When the absorption loss of the film just drops below 10ppm, it can be considered that the reaction gas flow rate is debugged. The reason is that if the reaction gas flow rate is further increased, the reaction gas will gradually tend to be excessive, the oxide layer or nitride layer on the surface of the target material 2 will gradually become thicker, and the coating rate will also tend to be unstable, resulting in the first case of hysteresis caused by excessive reaction gas flow.

[0074] It should be noted that before the weak absorption analysis method is used to debug the reaction gas flow rate, the reaction gas flow rate is also roughly adjusted. The increment of the rough adjustment flow rate is 3~6sccm, such as 3sccm, 4sccm, 5sccm or 6sccm. For example, the rough adjustment can be made by observing the color method or measuring the refractive index and extinction coefficient of the film using an ellipsometer. Specifically, for example, the debugging of the oxygen flow rate during the tantalum oxide coating process can be performed by observing the color of the tantalum oxide film, or using an ellipsometer to measure the refractive index and extinction coefficient of the film to preliminarily obtain the approximate range of the oxygen flow rate. When the oxygen flow rate is seriously insufficient, the second situation of the hysteresis phenomenon will occur. There is unoxidized tantalum in the film, and the color shows a metallic luster. As the oxygen flow rate increases, the film gradually changes to a transparent color. It can be understood that, for example, the cutoff point for rough adjustment can be set according to the color change of the film, and then the weak absorption analysis method can be further used for testing and adjustment.

[0075] After debugging, the target oxygen flow rate is 10-50 sccm, for example, 10 sccm, 20 sccm, 30 sccm, 40 sccm or 50 sccm. The target nitrogen flow rate is 10-50 sccm, for example, 10 sccm, 20 sccm, 30 sccm, 40 sccm or 50 sccm.

[0076] According to an embodiment of the present invention, the target material for preparing the thin film by reactive magnetron sputtering is titanium, hafnium, niobium, tantalum or silicon, preferably tantalum or silicon. Titanium, hafnium, niobium, tantalum and silicon are relatively common target materials and can be applied to the process debugging method of the present invention. Tantalum has a high melting point and corrosion resistance, and when used as a target material for reactive magnetron sputtering, it can react chemically with reactive gases, such as oxygen or nitrogen, to form tantalum oxide or tantalum nitride. The substrate for preparing the thin film is a silicon substrate or a quartz substrate. The silicon substrate has a good lattice matching and thermal expansion coefficient matching with silicon-based materials such as silicon oxide and silicon nitride, which helps to maintain a good interface and reduce stress. The quartz substrate, that is, silicon dioxide, has the characteristics of optical transparency. In the process of process debugging, it is not easy to introduce errors, and it is easier to find and determine the reaction gas flow range for coarse adjustment and weak absorption analysis test.

[0077] It can be understood that the films prepared in step 2 and step 3 of the present invention are both single-layer films, and the films prepared in step 2 and step 3 are prepared on substrates respectively. And after determining the target auxiliary source power in step 2, it is necessary to re-prepare the single-layer film in step 3 on a clean substrate to avoid the error caused by the film prepared in step 2 during the detection process. Preferably, the substrate used in step 2 is a silicon substrate, and the substrate used in step 3 is a quartz substrate. The required optical film samples can be prepared based on the process parameters determined in steps 1 to 3 as needed.

[0078] The optical film samples prepared above can also be post-annealed as needed. After the debugging of the above three steps, the optical film samples have already had extremely low optical loss. Due to the presence of auxiliary sources in the coating process, continuous ion bombardment may cause a large number of defects in the film, which will affect the absorption loss and mechanical properties (such as loss angle) of the film. The performance of the optical film samples can be further improved through the post-annealing process. The atmosphere of the post-annealing process should be kept consistent with the reaction gas in the coating process.

[0079] According to an embodiment of the present invention, the post-annealing treatment time is 10 hours, the heating and cooling rate is 5°C / min, and the annealing temperature of the post-annealing treatment is ≤ the crystallization temperature of the oxide -100°C or the crystallization temperature of the nitride -100°C. It can be understood that the post-annealing temperature is determined by the crystallization temperature of the oxide or nitride and the temperature that the film can withstand, so as to avoid the increase of scattering loss caused by crystallization or damage of the film. If bubble defects are found in the film after annealing, the annealing temperature should be further reduced at intervals of 50°C for testing. For example, the appropriate annealing temperature for tantalum oxide film is 300°C~550°C.

[0080] According to the embodiment of the present invention, the above-mentioned process debugging method makes the coating rate of the reactive magnetron sputtering equipment stable, and the prepared optical film has both low absorption loss and low surface roughness. The film prepared by the above-mentioned method can obtain a low optical loss film with stable performance, which is helpful for application in various optical films such as high reflection film and anti-reflection film, and the absorption loss of the prepared film can reach the sub-ppm level. The prepared film can basically achieve consistency with the expensive ion beam sputtering equipment, which is helpful for promoting the application of reactive magnetron sputtering.

[0081] For ease of understanding, the operation process of step one and step two is described in detail with examples.

[0082] Select the initial power (or voltage) according to the power range (or voltage range) of the sputtering power supply for safe operation. For example, for a power supply with a maximum power of 600W, you can start debugging from a sputtering power of 300W. Adjust the sputtering gas flow rate. Ar gas is usually used as the working gas to maintain the chamber pressure at 3E-3Torr when the molecular pump group is working normally, and gradually increase the power to 350W, 400W, 450W, 500W, and 550W. However, it is not limited to this. Observe whether there is arc discharge on the surface of the target material, and set the working power allowed when there is no arc discharge. The corresponding working power and gas flow rate can be selected according to the demand for the coating rate. The coating rate can be measured using one of the methods such as a film thickness meter, an ellipsometry spectrometer, an electron scanning microscope, and a spectrophotometer to determine the target working power and target flow rate.

[0083] By adjusting the process parameters of magnetron sputtering for preparing thin films of various oxide films and nitride films, the suitable range of target operating power is between 100 and 600 W, for example, it can be 100 W, 200 W, 300 W, 400 W, 500 W or 600 W, and the flow rate of working gas, such as argon, is between 10 and 100 sccm, for example, it can be 10 sccm, 20 sccm, 30 sccm, 40 sccm, 50 sccm, 60 sccm, 70 sccm, 80 sccm, 90 sccm or 100 sccm.

[0084] Taking a single-layer grid RF plasma source as an example, a step sequence of powers of 200W, 300W, 400W, and 500W (not limited to these) is used to coat a film on a silicon substrate. Before coating, the roughness of the silicon substrate measured by an atomic force microscope on an area of ​​5x5 square microns is less than or equal to 0.15nm, and the film thickness is greater than or equal to 100nm; after coating, the surface roughness on an area of ​​5x5 square microns is measured using an atomic force microscope, and the roughness is less than or equal to 0.2nm, and the corresponding auxiliary source power is selected as the target auxiliary source power.

[0085] After debugging, the appropriate auxiliary ion source power is between 100 and 600 W, for example, it may be 100 W, 200 W, 300 W, 400 W, 500 W or 600 W, and the RF substrate bias power is between 25 and 150 W, for example, it may be 25 W, 40 W, 55 W, 70 W, 85 W, 100 W, 115 W, 130 W, 145 W or 150 W.

[0086] The present invention is further described below by examples, drawings and related test experiments and results thereof. In the detailed description below, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present invention. However, it is obvious that one or more embodiments may also be implemented without these specific details. And, in the absence of conflict, the details in the following embodiments may be combined arbitrarily into other feasible embodiments.

[0087] It should be noted that the following specific embodiments are only for illustration and the protection scope of the present invention is not limited thereto.

[0088] Embodiment 1:

[0089] Taking the use of silicon target to plate silicon oxide and the use of tantalum target to plate tantalum oxide as an example, the process debugging method for preparing the thin film includes the following steps:

[0090] Sputtering power and sputtering gas flow test:

[0091] Use a power supply with a maximum power of 600W. Start debugging from a sputtering power of 300W, adjust the sputtering gas flow, use argon as the working gas, and maintain the chamber pressure at 3E-3Torr when the molecular pump group works normally. Gradually increase the power to 350W, 400W, 450W, 500W, and 550W, observe whether there is arc discharge on the surface of the target material, and set the allowable working power when there is no arc discharge.

[0092] After determination, the appropriate range of sputtering power is between 100~600W, and the flow rate of argon gas is between 10~100sccm.

[0093] Debugging of auxiliary source power:

[0094] Use a step sequence of 200W, 300W, 400W, and 500W to plate tandem oxide or silicon oxide on a silicon substrate. Before coating, the roughness of the silicon substrate measured by atomic force microscopy on an area of ​​5x5 square microns should not be higher than 0.15nm, and the coating thickness should not be less than 100nm. Use atomic force microscopy to measure the surface roughness on an area of ​​5x5 square microns after coating. The roughness should not be higher than 0.2nm, and the corresponding auxiliary source power is selected.

[0095] After debugging, the appropriate auxiliary ion source power is between 100 and 600 W, and the RF substrate bias power is between 25 and 150 W.

[0096] Debugging of reaction gas flow:

[0097] Rough adjustment is done by observing the color of the tantalum oxide film, or measuring the refractive index and extinction coefficient of the film with an ellipsometer to obtain an approximate oxygen flow rate. When the oxygen flow rate is seriously insufficient, there is unoxidized tantalum in the film, and the color shows a metallic luster. As the oxygen flow rate increases, the film gradually becomes transparent, and this is the end of the rough adjustment stage. However, this does not mean that the tantalum oxide film is fully oxidized, and it may still contain sub-oxides. At this time, a weak absorption analyzer can be used for fine adjustment. For example, the oxygen flow rates of 17, 19, 21, and 23 sccm are used to coat the film on a fused quartz substrate, and the absorption loss of the single-layer film is measured. When the film loss drops to a certain level and no longer changes, the oxygen flow rate at this time is used as the target reaction gas flow rate.

[0098] After determination, the target oxygen flow rate is 10~50sccm; the target nitrogen flow rate is 10~50sccm.

[0099] Post annealing process:

[0100] The parameters of the above three steps are fixed respectively to prepare alternating film layers of tantalum oxide and silicon oxide. The post-annealing time is 10 hours, the cooling rate is 5°C per minute, and the appropriate annealing temperature for tantalum oxide film is 300 to 550 degrees.

[0101] Comparative Example 1:

[0102] The preparation process of this comparative example 1 is substantially the same as that of Example 1, except that this comparative example 1 does not use a weak absorption analyzer to perform a step of fine-tuning the reaction gas flow rate.

[0103] Figure 4 The scanning electron microscope image of the cross section of the alternating film layer prepared by weak absorption analysis and debugging in Example 1 of the present invention is shown; Figure 5 The scanning electron microscope image of the cross section of the alternating film layer prepared in Comparative Example 1 without weak absorption analysis and debugging is shown. Figure 4~Figure 5 As shown, it can be seen Figure 4 The thickness of the alternating film layers is relatively uniform, and Figure 5 The thickness of the film layer shows a trend of change, indicating that the process parameters adjusted through weak absorption analysis can produce a relatively uniform film with atomic-level flatness, can optimize the optical loss performance of the film, and the coating process is relatively stable. Figure 5 Due to the occurrence of target poisoning and hysteresis effects, the prepared film is not uniform, the thickness is difficult to unify, and it is difficult to apply and promote it in the optical field.

[0104] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A process debugging method for preparing a thin film by reactive magnetron sputtering, characterized in that: include: Step 1, determining the target working power and target flow rate of the working gas for reactive magnetron sputtering according to the arc discharge condition on the target surface; Step 2, based on the target working power and the target flow rate of the working gas, the surface roughness of the thin film prepared by reactive magnetron sputtering is measured at different powers of the auxiliary source, and the auxiliary source power corresponding to the surface roughness less than 0.2 nm is determined as the target auxiliary source power; Step 3: Based on the target working power, the target flow rate of the working gas and the target auxiliary source power, the reaction gas with different flow rates is used to test the absorption loss of the thin film on the surface of the substrate by a weak absorption analysis method. The absorption loss is used to characterize the reaction degree of the thin film prepared by reactive magnetron sputtering. The reaction gas flow rate corresponding to the absorption loss of less than 10 ppm is determined as the target reaction gas flow rate; The auxiliary source power and / or the flow rate of the reaction gas is increased; when the absorption loss is less than 1000 ppm, the flow rate is increased by 1 sccm; When the absorption loss just drops below 10 ppm, the reaction gas flow rate at this time is determined as the target reaction gas flow rate.

2. The process debugging method according to claim 1, characterized in that: The reactive gas includes oxygen or nitrogen, and the working gas is argon; The film thickness in step 2 is ≥ 100 nm; The film thickness in step 3 is 450~550nm.

3. The process debugging method according to claim 1 or 2, characterized in that: The reaction degree includes the content of suboxide in the oxide film formed by reactive magnetron sputtering to prepare the thin film or the content of nitride in the nitride film.

4. The process debugging method according to claim 1 or 2, characterized in that: Assist sources include RF bias or plasma sources.

5. The process debugging method according to claim 1 or 2, characterized in that: The target materials for preparing thin films by reactive magnetron sputtering are titanium, hafnium, niobium, tantalum or silicon; The substrate for preparing the thin film is a silicon substrate or a quartz substrate.

6. The process debugging method according to claim 1 or 2, characterized in that: The process debugging method also includes: The thin films prepared by reactive magnetron sputtering were post-annealed.

7. The process debugging method according to claim 6, characterized in that: The post-annealing treatment lasts for 10 hours, the heating and cooling rate is 5°C / min, and the annealing temperature of the post-annealing treatment is ≤ the crystallization temperature of the oxide - 100°C or the crystallization temperature of the nitride - 100°C.

Images