A method and apparatus for preparing a patterned graphene silicon wafer

Abstract

The application relates to a preparation method of a patterned graphene silicon wafer, which comprises the following steps: performing activation treatment on the surface of a silicon wafer; coating a specific pattern on the silicon wafer by using a graphene dispersion liquid to form a graphene dispersion liquid film, wherein the dispersion liquid is graphene oxide or graphene; airing the graphene dispersion liquid film after coating to form a graphene oxide film; performing heat curing treatment on the graphene oxide functionalized graphene oxide film; placing the heated silicon wafer in a reducing atmosphere; irradiating the graphene oxide film in the reducing atmosphere by using a laser to obtain a silicon wafer with grown patterned graphene; and realizing industrial application by forming the patterned graphene silicon wafer by the method.

Description

Technical Field

[0001] This invention relates to the field of graphene technology, and more specifically, to a method and apparatus for preparing patterned graphene silicon wafers. Background Technology

[0002] Since its discovery in 2004, graphene has been extensively studied and proven to be a high-performance material, exhibiting excellent electrical and thermal conductivity, as well as a large specific surface area. It has attracted increasing attention from researchers. Modern microelectronic devices are extremely small in size and require high precision, which places stringent demands on the quality of graphene used in them. Therefore, how to prepare high-quality graphene and transfer it intact onto silicon substrates has become a significant challenge for researchers in this field. To date, many methods for preparing graphene have been reported, mainly including micromechanical exfoliation, SiC epitaxy, chemical exfoliation, and CVD. Among these, CVD is simple to operate and yields high-quality graphene with large surface areas, making it the primary method for obtaining high-quality graphene.

[0003] Transferring graphene films grown on metal substrates to silicon substrates inevitably introduces impurities, voids, and other non-ideal factors, thus reducing film quality. To address this, researchers have employed a method of depositing a thin metal layer on a silicon substrate to directly grow graphene, achieving promising results. However, the quality of graphene grown using this method still falls short of the requirements for practical applications.

[0004] Therefore, the market demands a technology that can directly grow patterned graphene films on silicon wafers. The aim is to reduce the steps involved in transferring graphene films onto silicon substrates, thereby avoiding damage and the introduction of impurities. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method and equipment for preparing patterned graphene silicon wafers, addressing the above-mentioned deficiencies of the prior art.

[0006] The technical solution adopted by this invention to solve its technical problem is:

[0007] A method for fabricating patterned graphene silicon wafers includes the following steps:

[0008] Step S1: Activate the surface of the silicon wafer;

[0009] Step S2: A specific pattern is coated on the silicon wafer using a graphene dispersion to form a graphene oxide dispersion film, wherein the dispersion is graphene oxide or functionalized graphene oxide.

[0010] Step S3: Dry the coated graphene dispersion film to form a graphene oxide film.

[0011] Step S4: Heat curing treatment of the graphene oxide film;

[0012] Step S5: Place the heated silicon wafer in a reducing atmosphere;

[0013] Step S6: Irradiate the graphene oxide film placed in a reducing atmosphere with a laser to obtain a silicon wafer with patterned graphene grown on it.

[0014] Preferably, in step S1, the activation treatment of the silicon wafer surface is performed by one or more of the following methods: high-temperature heating treatment, plasma bombardment treatment, and piranha solution treatment. The activation treatment can be selected by heating the silicon wafer at 600 to 1000 degrees Celsius for 30 to 120 minutes, plasma bombardment for 30 to 120 minutes, or piranha solution treatment for 5 to 60 minutes.

[0015] Preferably, in step S2, the functionalized graphene oxide contains imide groups, benzene ring structures, or sulfonyl groups. The preparation method of the graphene oxide or functionalized graphene oxide is not limited.

[0016] Preferably, in step S4, the silicon wafer is heated and cured at a temperature of 80-200°C for 5-60 minutes.

[0017] Preferably, in step S5, the reducing atmosphere comprises one or more gases such as hydrogen, hydrogen sulfide, ammonia, and carbon monoxide; the silicon wafer is placed in the reducing atmosphere and then heated to 150-400 degrees Celsius. This atmosphere can reduce the power required for laser irradiation and enable functional doping of the graphene film.

[0018] Preferably, in step S6, the laser is a pulsed ruby ​​laser or a carbon dioxide laser, the power of the laser irradiation is 1-50W, and the total laser irradiation time is 20S-70S.

[0019] Preferably, in step S2, the concentration of the graphene oxide dispersion is 0.1%-2%.

[0020] Preferably, in step S3, the graphene dispersion film is dried at room temperature for 0.5-6 hours, and the thickness of the graphene after drying is 2 μm.

[0021] Preferably, in step S1, X-ray photoelectron spectroscopy is used to detect whether the silicon wafer has oxygen-containing sites; if oxygen-containing sites are detected, the activation process is stopped. In step S2, laser reflection measurement is used to monitor the coating of the graphene dispersion in real time, and the coating process is adjusted in real time according to the monitoring results to maintain the uniformity of graphene distribution. In step S3, the graphene dispersion film is dried more quickly by infrared drying or microwave drying, and the drying is completed by measuring the weight of the graphene dispersion film in real time. In step S6, the heated silicon wafer is placed in a reducing atmosphere, and the graphene oxide film is irradiated by laser. The thickness and uniformity data of the graphene oxide film are obtained by micro Raman spectroscopy, and the laser power is controlled according to the data.

[0022] Preferably, in step S2, the method for adjusting the coating process in real time based on the detection results is as follows:

[0023] The thickness information of the graphene dispersion on the silicon wafer surface is characterized by laser reflection data, and a thickness distribution model is generated by the thickness information of the graphene dispersion film.

[0024] Obtain a preset thickness reference model of the graphene dispersion film, establish a position coordinate system in the thickness distribution model, establish coordinate axes with the center of the silicon wafer as the endpoint, uniformly divide the surface of the silicon wafer into several points, and determine the coordinate value of each point.

[0025] The thickness distribution model is superimposed on the thickness reference model for comparison. The thickness value of each point in the thickness distribution model is compared with the thickness value of the thickness reference model to obtain the thickness comparison value, and this value is recorded. If the absolute value of the thickness comparison value is less than the correction value, it is marked as a normal point and its thickness comparison value is recorded as 0. If the absolute value of the thickness comparison value is greater than the correction value, it is marked as an abnormal point. The abnormal points that are connected to each other are integrated to obtain multiple abnormal regions.

[0026] Obtain the outer contour of each abnormal region on the graphene dispersion film; and determine the coordinate region of the abnormal region based on the coordinate values ​​of the abnormal point, thereby calculating the actual location of the abnormality; the method for determining the coordinate region is as follows: take the outermost point of the four sides of the abnormal region as the tangent point, draw the tangent line, and determine the range occupied by the abnormal region by the coordinate of the tangent line.

[0027] Determine whether the abnormal area is a depression or a convexity; radiate outwards in a circle from the center of the abnormal area; mark the points where the thickness comparison value is greater than 0 and less than 0 respectively; then, the abnormal area with the thickness comparison value greater than 0 is determined to be a depression; the abnormal area with the thickness comparison value greater than 0 is determined to be a convexity.

[0028] The abnormal areas identified as depressions are coated a second time until the thickness of all the abnormal areas identified as depressions is greater than the thickness of the thickness reference model. The uniformity of coating thickness is then checked in the preset area around the coated abnormal areas after the second coating. It is determined whether the thickness comparison value of the area is greater than the correction value. If the thickness comparison value is greater than the correction value, it is marked as an abnormal point. The interconnected abnormal points are integrated to obtain multiple abnormal areas. Then, the multiple abnormal areas on the graphene dispersion film are removed.

[0029] In real-time monitoring, the collected data is input into a trained prediction model for classification. The thickness of the graphene dispersion film is predicted based on the coating data, and the coating amount is adjusted according to the prediction results. The training method for the model is as follows:

[0030] Collect data related to the preparation process of graphene oxide films, including coating amount and corresponding changes in the thickness of graphene dispersion films;

[0031] The collected data is preprocessed, including noise removal, handling missing values, and data standardization. Features are extracted from the preprocessed data, and feature selection is performed to determine the feature set used to train the SVM model.

[0032] The extracted feature sets are labeled to distinguish between normal and abnormal conditions, and integrated into a dataset. The dataset is then divided into training and test sets using cross-validation.

[0033] The SVM model is trained using a training set. By inputting labeled data, the SVM model learns and recognizes feature patterns of normal and abnormal situations.

[0034] Use the test set to validate the trained model, evaluate the model's performance and generalization ability through the test set, and adjust the model's hyperparameters and retrain it if the target is not met; if the target is met, the training is complete.

[0035] The trained SVM model is applied to real-time prediction.

[0036] Preferably, during the preparation of graphene, the surface morphology, thickness change, deposition time, deposition temperature, surface chemical composition, and functional group changes of the graphene oxide film are obtained through monitoring equipment;

[0037] Preferably, in step S2, the method of coating the graphene oxide dispersion onto the silicon wafer surface is any one of the following: drop coating, casting, impregnation, screen printing, inkjet printing, roll coating, coating, scraping coating, and spin coating.

[0038] The present invention also provides a patterned graphene silicon wafer device, which is manufactured based on the aforementioned method for preparing patterned graphene silicon wafers, comprising a silicon wafer substrate and a patterned graphene film covering the surface of the silicon wafer substrate.

[0039] The beneficial effects of this invention are as follows: This invention directly coats a silicon substrate with a patterned graphene oxide or a functionalized graphene oxide solution, and then directly obtains a silicon wafer with a patterned graphene film grown by laser reduction. There is no step of transferring the graphene film onto the silicon substrate, avoiding damage and the introduction of impurities, while simultaneously obtaining a patterned and high-quality graphene film, which is of great significance for the application of graphene on silicon substrates. Through the method of this invention, patterned graphene silicon wafers are formed, thereby realizing industrial applications. Laser reduction directly grows patterned graphene films on the wafer, significantly improving production efficiency. Furthermore, using a graphene oxide solution containing sulfonic acid groups allows for good dispersion in water, preventing graphene agglomeration, and the sulfonic acid groups can react with the hydroxyl groups on the surface of the surface-activated silicon wafer, enhancing the bonding ability between graphene and the silicon wafer and reducing graphene breakage and shedding. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be further described below in conjunction with the accompanying drawings and embodiments. The drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort:

[0041] Figure 1 This is a step diagram of the patterned graphene silicon wafer fabrication method according to an embodiment of the present invention. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, a clear and complete description will be provided below in conjunction with the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the protection scope of the present invention.

[0043] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Similarly, the terms "an" or "a" and similar terms do not indicate a quantity limitation, but rather indicate the presence of at least one. The terms "connected" or "linked" and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up," "down," "left," "right," etc., are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship also changes accordingly.

[0044] The steps of the patterned graphene silicon wafer and its preparation method of the present invention are as follows: Figure 1 As shown, it includes the following steps:

[0045] Step S1: Activate the silicon wafer surface to create oxygen-containing sites. This activation process alters the surface's chemical properties, exposing reactive functional groups and increasing its reactivity. This allows subsequent chemical reactions to occur on the wafer surface. Oxygen-containing sites on silicon wafers are typically located at lattice defects, such as surface pits, dangling bonds, and interstitial silicon atoms. These sites readily combine with reactive gas molecules, thus increasing the surface reactivity and improving atomic adhesion during deposition, ultimately enhancing the quality and performance of the silicon wafer. Therefore, silicon wafer surface activation treatment fills the surface with functional groups, thereby increasing reactivity.

[0046] Methods for activating the surface of silicon wafers include high-temperature heating, plasma bombardment, and / or piranha solution treatment. Specifically, in this invention, high-temperature heating involves heating the silicon wafer to 600 to 1000 degrees Celsius for 30 to 120 minutes. High-temperature heating is the simplest activation method, involving short-term treatment of the silicon wafer at a high temperature. High-temperature heating can introduce oxygen atoms or generate oxides on the silicon wafer surface, thereby forming oxygen-containing sites. This method is simple to operate and has low cost. In this invention, plasma bombardment involves bombarding the silicon wafer surface with oxygen plasma for 30 to 120 minutes. Plasma bombardment, by bombarding the silicon wafer surface with oxygen plasma generated by gas discharge, can remove surface defects and stains, and can attach oxygen-containing groups to the surface, thereby increasing surface reactivity and improving the surface quality and chemical properties of the silicon wafer. The plasma bombardment treatment in this invention is a piranha solution treatment in which the silicon wafer is placed in a piranha solution for 5-60 minutes. Treating the silicon wafer with the piranha solution can remove defects and stains on the surface of the silicon wafer, and at the same time activate the chemical reactions therein, thereby increasing the surface reactivity.

[0047] Step S2: A specific pattern is coated onto the silicon wafer using a graphene dispersion to form a graphene oxide dispersion film. The dispersion is graphene oxide or functionalized graphene oxide; the functionalized graphene oxide contains imide groups, benzene ring structures, or sulfonyl groups. The function of the graphene dispersion is to uniformly disperse graphene in a solvent, making it a soluble graphene solution and preventing it from agglomerating into clumps. The prepared graphene solution has good flowability and stability, facilitating subsequent processing and application. The preparation of the graphene oxide dispersion of this invention is not particularly limited; it can be commercially available or prepared using conventional methods known to those skilled in the art. Exemplarily, the graphene oxide dispersion is prepared using the Hummers method or a modified oxidative exfoliation method. The source of the graphite is not particularly limited; any conventional graphite known to those skilled in the art is acceptable. Exemplarily, it can be nanographite, natural flake graphite, etc. The mesh size of the graphite is not particularly limited; for example, it can be 100-8000 mesh. The concentration of the graphene oxide dispersion is 0.1%-2%. The graphene oxide dispersion can be coated onto the surface of a silicon wafer by any of the following methods: drop coating, casting, impregnation, screen printing, inkjet printing, roll coating, coating, blade coating, or spin coating.

[0048] Step S3: Dry the coated graphene dispersion film to form a graphene oxide film. The graphene dispersion film is dried at room temperature for 0.5-6 hours, and the thickness of the graphene after drying is 2 μm.

[0049] Step S4: Heat-curing the graphene oxide film; heat-curing the silicon wafer at 80-200℃ for 5-60 minutes. This heat treatment alters the band structure of graphene oxide, improving its light absorption capacity (i.e., increasing laser energy utilization) and thus enhancing its laser reduction efficiency. The heating mode for the graphene oxide dispersion can be a single temperature mode, a staged heating mode, or a continuous linear heating mode, depending on the actual effect.

[0050] Step S5: Place the heated silicon wafer in a reducing atmosphere; the reducing atmosphere includes one or more gases such as hydrogen, hydrogen sulfide, ammonia, and carbon monoxide; after placing the silicon wafer in the reducing atmosphere, heat it to 150-400 degrees Celsius. The reducing atmosphere can reduce the power required for laser irradiation and functionalize the graphene film.

[0051] Step S6: Irradiate a graphene oxide film placed in a reducing atmosphere with a laser to obtain a silicon wafer with patterned graphene grown on it. The laser is a pulsed ruby ​​laser or a carbon dioxide laser, and the laser irradiation power is 1-50W; the total laser irradiation time is 20S-70S. Laser treatment can significantly improve the reduction degree of graphene, thereby improving the overall electrochemical performance and achieving deep reduction of the graphene structure. Furthermore, the pulsed ruby ​​laser or carbon dioxide laser used in this invention has a longer wavelength, stronger photothermal effect, and better reduction effect on graphene oxide.

[0052] The present invention also includes a patterned graphene silicon wafer device, manufactured based on the aforementioned method for preparing a patterned graphene silicon wafer, comprising a silicon wafer substrate and a patterned graphene film covering the surface of the silicon wafer substrate.

[0053] This invention involves directly coating a silicon substrate with a patterned graphene oxide or functionalized graphene oxide solution, followed by laser reduction to directly obtain a silicon wafer with a patterned graphene film. This eliminates the step of transferring the graphene film onto the silicon substrate, avoiding damage and the introduction of impurities, while simultaneously yielding a patterned and high-quality graphene film, which is of great significance for the application of graphene on silicon substrates. The method of this invention forms patterned graphene silicon wafers, enabling industrial applications. The laser reduction method for preparing graphene films directly grows patterned graphene films on silicon wafers, significantly reducing laser power. It allows for the use of low-power lasers to reduce graphene in a single step, lowering costs and significantly improving production efficiency.

[0054] The difference between this invention and the prior art is that the laser, with its photonic and thermal effects, destroys the CO bonds in the graphene oxide matrix and completes the deoxygenation reaction with the help of high-frequency pulses and reducing gases to generate graphene. Furthermore, the method of preparing graphene films by laser reduction using reducing gases directly grows patterned graphene films on silicon wafers, which significantly reduces the laser power.

[0055] The present invention will be explained below with reference to embodiments. Those skilled in the art will understand that the following embodiments are for illustrative purposes only and should not be considered as limiting the scope of the invention. Where specific techniques or conditions are not specified in the embodiments, they are performed according to the techniques or conditions described in the literature in the field or according to the product instructions. Reagents or instruments whose manufacturers are not specified are all conventional products that can be purchased commercially.

[0056] Example 1

[0057] (1) The silicon wafer is activated by bombarding the surface of the silicon wafer with ions for 60 minutes to form oxygen-containing sites on the surface of the silicon wafer.

[0058] (2) A dispersion of functionalized graphene oxide containing sulfonic acid groups is coated onto a silicon crystal to form a specific pattern, thereby forming a graphene oxide dispersion film with a specific pattern.

[0059] (3) After the silicon wafer is naturally dried, heat it at 150 degrees Celsius for 30 minutes to cure it.

[0060] (4) Place the silicon wafer from the previous step in a hydrogen atmosphere and heat it to 300 degrees.

[0061] (5) Irradiate the silicon wafer surface for 20 seconds in a hydrogen atmosphere using a pulsed carbon dioxide laser set to 20w.

[0062] (6) Remove the silicon wafer to obtain a silicon wafer on which patterned graphene has been directly grown.

[0063] Example 2

[0064] (1) The silicon wafer is activated by bombarding the surface of the silicon wafer with ions for 60 minutes to form oxygen-containing sites on the surface of the silicon wafer.

[0065] (2) The dispersion of graphene oxide is coated onto a silicon crystal to form a specific pattern, thus forming a graphene oxide dispersion film with a specific pattern.

[0066] (3) After the silicon wafer is naturally dried, heat it at 150 degrees Celsius for 30 minutes to cure it.

[0067] (4) Place the silicon wafer from the previous step in an ammonia atmosphere and heat it to 300 degrees.

[0068] (5) Irradiate the silicon wafer surface for 20 seconds using a pulsed ruby ​​laser set to 20w in an ammonia atmosphere;

[0069] (6) Remove the silicon wafer to obtain a silicon wafer on which patterned graphene has been directly grown.

[0070] Example 3

[0071] (1) Activate the silicon wafer by heating it at 800 degrees Celsius for 60 minutes to form oxygen-containing sites on the surface of the silicon wafer.

[0072] (2) A dispersion of functionalized graphene oxide containing sulfonic acid groups is coated onto a silicon crystal to form a specific pattern, thereby forming a graphene oxide dispersion film with a specific pattern.

[0073] (3) After the silicon wafer is naturally dried, heat it at 150 degrees Celsius for 30 minutes to cure it.

[0074] (4) Place the silicon wafer from the previous step in a hydrogen atmosphere and heat it to 300 degrees.

[0075] (5) Irradiate the silicon wafer surface for 50 seconds in a hydrogen atmosphere using a pulsed ruby ​​laser set to 20w;

[0076] (6) Remove the silicon wafer to obtain a silicon wafer on which patterned graphene has been directly grown.

[0077] Example 4

[0078] (1) The silicon wafer is activated by treating the surface of the silicon wafer with piranha solution for 60 minutes to form oxygen-containing sites on the surface of the silicon wafer.

[0079] (2) A dispersion of functionalized graphene oxide containing sulfonic acid groups is coated onto a silicon crystal to form a specific pattern, thereby forming a graphene oxide dispersion film with a specific pattern.

[0080] (3) After the silicon wafer is naturally dried, heat it at 150 degrees Celsius for 30 minutes to cure it.

[0081] (4) Place the silicon wafer from the previous step in a hydrogen atmosphere and heat it to 300 degrees.

[0082] (5) Irradiate the silicon wafer surface for 20 seconds in a hydrogen atmosphere using a pulsed carbon dioxide laser set to 20w.

[0083] (6) Remove the silicon wafer to obtain a silicon wafer on which patterned graphene has been directly grown.

[0084] Sheet resistance tests were performed on the graphene silicon wafers obtained after implementing Examples 1-4, and the results are shown in the table below:

[0085] Example Example 1 Example 2 Example 3 Example 4 sheet resistor 560 958 863 714

[0086] The difference between Example 1 and Example 2 lies in the different dispersions used. Example 1 uses a dispersion of functionalized graphene oxide containing sulfonic acid groups for coating, while Example 2 uses a common graphene oxide dispersion for coating. In Example 1, the presence of sulfonic acid groups in the graphene oxide allows the graphene oxide solution to disperse well in water, preventing graphene agglomeration. Furthermore, the sulfonic acid groups can react with the hydroxyl groups on the surface of the activated silicon wafer, enhancing the bonding ability between graphene and the silicon wafer and reducing graphene breakage and shedding. Another difference lies in the irradiation process. Example 1 used a 20W carbon dioxide laser to irradiate the silicon wafer in a hydrogen atmosphere, while Example 2 used a ruby ​​laser in an ammonia atmosphere. 20W is considered low-power irradiation, while carbon dioxide laser irradiation typically ranges from tens to hundreds of watts, whereas ruby ​​laser irradiation generally ranges from thousands to tens of thousands of watts. Therefore, carbon dioxide laser irradiation is more suitable for low-power applications, resulting in better performance, stronger graphene reduction, and lower energy consumption. Example 2 used ordinary graphene oxide and reduced it with a ruby ​​laser in an ammonia atmosphere. Compared to Example 1, which used sulfonic acid-modified graphene oxide reduced with a carbon dioxide laser in a hydrogen atmosphere, the sulfonic acid-modified graphene oxide bonded better to the silicon wafer, and the carbon dioxide laser reduction was more effective. Therefore, the sheet resistance of Example 1 was significantly lower than that of Example 2.

[0087] The differences between Example 1 and Example 3 include the silicon wafer surface treatment method and the laser used for irradiation. In Example 3, the silicon wafer was heated to 800 degrees Celsius for 60 minutes, while in Example 1, the silicon wafer surface was bombarded with ions for 60 minutes. Because heating to 800 degrees Celsius for 60 minutes resulted in fewer oxygen-containing sites on the silicon wafer surface compared to ion bombardment, the sulfonic acid-based graphene oxide did not bond better to the silicon wafer. Another difference is that Example 3 used a ruby ​​laser for reduction, which had a worse reduction effect than the carbon dioxide laser used in Example 1. Therefore, the sheet resistance of Example 1 was significantly lower than that of Example 3. However, because the dispersion of sulfonic acid-based functionalized graphene oxide used in Example 3 could react with the hydroxyl groups on the silicon wafer surface, the sheet resistance was slightly lower than that of Example 2.

[0088] The difference between Example 1 and Example 4 lies in the silicon wafer surface treatment method. In Example 1, the silicon wafer surface treatment method was ion bombardment for 60 minutes, while in Example 4, the silicon wafer surface treatment method was piranha solution treatment for 60 minutes. From the final measured sheet resistance, it can be seen that the result of Example 4 is slightly worse than that of Example 1. The reason is that the silicon wafer surface formed by piranha solution treatment for 60 minutes is less than that formed by ion bombardment, resulting in the sulfonic acid group graphene oxide not being better bonded to the silicon wafer. The resulting graphene film is easily damaged and detached, resulting in defects in the film. Therefore, the sheet resistance of Example 1 is significantly lower than that of Example 4. However, since Example 4 uses carbon dioxide laser irradiation, its sheet resistance is slightly lower than that of Example 3.

[0089] As demonstrated by the above embodiments, this invention directly coats a silicon substrate with a patterned graphene oxide or functionalized graphene oxide solution, and then directly obtains a silicon wafer with a patterned graphene film grown by laser reduction. The step of transferring the graphene film onto the silicon substrate is eliminated, avoiding damage and the introduction of impurities, while simultaneously obtaining a patterned and high-quality graphene film, which is of great significance for the application of graphene on silicon substrates. The method of this invention forms a patterned graphene silicon wafer, thereby realizing industrial applications. The method of preparing graphene films by laser reduction directly grows patterned graphene films on silicon wafers, significantly reducing laser power. It allows the use of low-power lasers to reduce graphene in a single step, reducing costs and significantly improving production efficiency.

[0090] The present invention also provides another embodiment, which differs from the previous embodiment in that, in step S1, X-ray photoelectron spectroscopy is used to detect whether the silicon wafer has oxygen-containing sites, and if oxygen-containing sites are detected, the activation process is stopped; in step S2, laser reflection measurement is used to monitor the coating of the graphene dispersion in real time, and the coating process is adjusted in real time according to the monitoring results to maintain the uniformity of graphene distribution; in step S3, the graphene dispersion film is dried more quickly by infrared drying or microwave drying, and the drying is completed by measuring the weight of the graphene dispersion film in real time; in step S6, the heated silicon wafer is placed in a reducing atmosphere, and the graphene oxide film is irradiated by laser. The thickness and uniformity data of the graphene oxide film are obtained by micro Raman spectroscopy, and the laser power is controlled according to the data.

[0091] Detecting oxygen-containing sites using X-ray photoelectron spectroscopy helps determine the chemical state of the silicon wafer surface. This serves as a key indicator to determine whether to continue activation processing, ensuring the smooth progress of subsequent steps. Real-time monitoring of the graphene dispersion coating using laser reflectance measurements helps control the coating process and maintain a uniform distribution of graphene. This helps ensure the uniformity and quality of the film. Infrared or microwave drying can accelerate the drying process of the graphene dispersion film. Real-time measurement of the film weight to determine whether drying is complete is an effective method to control the drying process and avoid over-drying. Micro-Raman spectroscopy can obtain data on the thickness and uniformity of the graphene oxide film. Using this data to control the laser power allows for adjustment of the laser irradiation intensity, ensuring that the treatment of the film is effective and does not affect its properties and uniformity.

[0092] In step S2, the method for adjusting the coating process in real time based on the detection results is as follows:

[0093] The thickness of the graphene dispersion on the silicon wafer surface is characterized by laser reflection data. A thickness distribution model is generated based on the thickness information of the graphene dispersion film. By using laser reflection data, the thickness information of the graphene dispersion film can be monitored in real time and accurately, ensuring that the coating process can be accurately adjusted to achieve the expected coating thickness.

[0094] A pre-defined thickness reference model of the graphene dispersion film is obtained. A positional coordinate system is established within this model, with the center of the silicon wafer as the endpoint. The silicon wafer surface is uniformly divided into several points, and the coordinate value of each point is determined. This coordinate system and point division allow for more accurate analysis of the coating thickness uniformity. Comparing the thickness distribution model with the reference model, and comparing the thickness values ​​at each point, helps to detect and record abnormal points, thereby enabling real-time monitoring and control of the coating process uniformity.

[0095] The thickness distribution model is superimposed on the thickness reference model for comparison. The thickness value of each point in the thickness distribution model is compared with the thickness value of the thickness reference model to obtain the thickness comparison value, and this value is recorded. If the absolute value of the thickness comparison value is less than the correction value, it is marked as a normal point and its thickness comparison value is recorded as 0. If the absolute value of the thickness comparison value is greater than the correction value, it is marked as an abnormal point. The abnormal points that are connected to each other are integrated to obtain multiple abnormal regions.

[0096] The thickness distribution model is compared with a reference model, and the thickness difference at each point is calculated. This effectively identifies points of thickness anomalies—areas where the thickness deviation exceeds a set correction value. Discrete anomaly points are grouped into regions, thus more clearly locating anomalies during the coating process. After marking the anomaly points and regions, targeted corrective measures can be taken, such as adjusting the coating amount or performing local repairs, to ensure the coating thickness is within the expected range, thereby improving coating quality and uniformity. Recording information on thickness comparison values, anomaly points, and anomaly regions facilitates subsequent data analysis and quality improvement, allowing for the tracing and analysis of anomalies during the coating process and providing data support for optimizing the coating workflow.

[0097] The process involves obtaining the outer contour of each anomalous region on the graphene dispersion film and determining the coordinate region of the anomalous region based on the coordinate values ​​of the anomalous points. This allows for the calculation of the actual location of the anomalous region. The method for determining the coordinate region is as follows: tangent lines are drawn using the outermost points of the four sides of the anomalous region as tangents, and the coordinates of these tangent lines determine the extent of the anomalous region. Obtaining the outer contour of the anomalous region involves identifying anomalous points and integrating connected anomalous points into multiple anomalous regions. The outer contour of each anomalous region is then determined, and algorithms such as connection of points within the anomalous region or edge detection are used to obtain the region boundary. Based on the outermost points of the four sides of the anomalous region's outer contour, four outermost points can be found. Connecting these points forms four boundary lines, i.e., tangent lines. The coordinates of these four tangent lines determine the extent of the anomalous region. These tangent lines form a bounding box or polygon, the interior of which represents the actual extent of the anomalous region. By processing the coordinates of the tangent lines, the actual location of the anomalous region or its position on the entire silicon wafer can be obtained.

[0098] The process involves determining whether an abnormal area is a depression or a convexity. A circular pattern radiates outwards from the center of the abnormal area. Points with thickness comparison values ​​greater than 0 and less than 0 are marked. An abnormal area with a thickness comparison value greater than 0 is classified as a depression, and an abnormal area with a thickness comparison value less than 0 is classified as a convexity. Based on the thickness differences observed during the coating process, the nature of the abnormal area is inferred.

[0099] For abnormal areas identified as depressions, a second coating is applied until the thickness of all such areas exceeds the thickness reference model. The uniformity of the coating thickness is then checked in a pre-defined area surrounding the second-coated abnormal area. If the thickness comparison value of this area exceeds a correction value, it is marked as an abnormal point. Connected abnormal points are then integrated to obtain multiple abnormal areas. These multiple abnormal areas on the graphene dispersion film are then removed. For the depressions identified as abnormal areas, a second coating operation is performed until the thickness of all such areas exceeds the thickness reference model. This aims to correct insufficient coating in the abnormal areas to achieve the desired coating thickness. The uniformity of the coated area is then confirmed. For areas that still exhibit abnormalities after coating (thickness comparison values ​​exceeding the correction value), connected abnormal points are integrated into multiple abnormal areas. At this point, all abnormal areas are raised, and these abnormal areas are then removed.

[0100] In real-time monitoring, the collected data is input into a trained prediction model for classification. The thickness of the graphene dispersion film is predicted based on the coating data, and the coating amount is adjusted according to the prediction results. The training method for the model is as follows:

[0101] Collect data related to the preparation process of graphene oxide films, including coating amount and corresponding changes in the thickness of graphene dispersion films;

[0102] The collected data is preprocessed, including noise removal, handling missing values, and data standardization. Features are extracted from the preprocessed data, and feature selection is performed to determine the feature set used to train the SVM model.

[0103] The extracted feature sets are labeled to distinguish between normal and abnormal conditions, and integrated into a dataset. The dataset is then divided into training and test sets using cross-validation.

[0104] The SVM model is trained using a training set. By inputting labeled data, the SVM model learns and recognizes feature patterns of normal and abnormal situations.

[0105] Use the test set to validate the trained model, evaluate the model's performance and generalization ability through the test set, and adjust the model's hyperparameters and retrain it if the target is not met; if the target is met, the training is complete.

[0106] The trained SVM model is applied to real-time prediction.

[0107] By using a trained model, the thickness of the graphene dispersion film can be predicted after real-time data collection. This enables real-time monitoring and adjustment of the coating amount during the coating process, helping to control the coating thickness. The trained model can identify normal and abnormal situations, helping to promptly detect and handle anomalies. This helps reduce potential problems during the coating process, improving production efficiency and coating quality. The ability to adjust the coating amount based on the prediction results facilitates a more consistent graphene dispersion film thickness, improving coating uniformity and quality stability. Real-time monitoring and adjustment allow for more effective control of the coating process, reducing unnecessary coating waste and improving production efficiency. Automated prediction and adjustment processes help reduce the impact of human operation, lower the possibility of human error, and improve the stability and consistency of the production process.

[0108] In the description of this specification, references to terms such as "one embodiment," "some embodiments," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. It should be understood that those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A method for fabricating a patterned graphene-silicon wafer, the method comprising: The method comprises the following steps: ​ Step S1, activating the surface of the silicon wafer to make the silicon wafer have oxygen-containing sites; the method for activating the surface of the silicon wafer is one or more of high-temperature heating treatment, plasma bombardment treatment and piranha solution treatment; Step S2, coating a specific pattern on the silicon wafer using a graphene dispersion liquid to form an oxidized graphene dispersion liquid film, wherein the dispersion liquid is oxidized graphene or functionalized oxidized graphene; Step S3, air-drying the graphene dispersion liquid film after coating to form an oxidized graphene film; Step S4, heating and curing the oxidized graphene film; Step S5, placing the heated silicon wafer in a reducing atmosphere; after the silicon wafer is placed in the reducing atmosphere, heating is performed at 150-400 degrees; Step S6, irradiating the oxidized graphene film placed in the reducing atmosphere by laser to obtain a silicon wafer on which patterned graphene is grown.

2. The method of claim 1, wherein the graphene is patterned by a photolithography process. In step S2, the functionalized oxidized graphene contains functional groups such as sulfonic acid groups, imide groups and benzene ring structures.

3. The method of claim 1, wherein the graphene is patterned by a photolithography process. In step S4, the silicon wafer is heated and cured at a temperature of 80-200°C for 5-60 minutes.

4. The method of claim 1, wherein the graphene is patterned by a photolithography process. In step S5, the reducing atmosphere includes an atmosphere composed of one or more of hydrogen, hydrogen sulfide, ammonia and carbon monoxide.

5. The method of claim 1, wherein the graphene is patterned by a photolithography process. In step S6, the laser is a pulsed ruby laser or a carbon dioxide laser, the power of the laser irradiation is 1-50 W, the total time of the laser irradiation is 20-70 seconds, in step S2, the concentration of the oxidized graphene dispersion liquid is 0.1%-2%, and in step S3, the air-drying temperature of the graphene dispersion liquid film is room temperature, and the air-drying time is 0.5-6 hours.

6. The method of claim 1, wherein the graphene is patterned by a photolithography process. In step S1, whether the silicon wafer has oxygen-containing sites is detected by X-ray photoelectron spectroscopy, and if the formation of the oxygen-containing sites is detected, the activation treatment is stopped; in step S2, the surface properties of the graphene dispersion liquid are detected in real time by laser reflection measurement to judge the uniformity of the coating, and the coating process is adjusted in real time according to the monitoring result; in step S3, the graphene dispersion liquid film is accelerated to dry by infrared drying or microwave drying, and whether the drying is completed is judged by measuring the weight of the graphene dispersion liquid film in real time; in step S6, the heated silicon wafer is placed in a reducing atmosphere, and the oxidized graphene film is irradiated by laser, the thickness and uniformity data of the oxidized graphene film are obtained by micro-Raman spectroscopy, and the laser power is controlled according to the data content.

7. The method of claim 1, wherein the graphene is patterned by a photolithography process. In step S2, the method for adjusting the coating process in real time according to the detection result is as follows: According to the laser reflection data, the thickness information of the graphene dispersion liquid on the surface of the silicon wafer is characterized, and a thickness distribution model is generated based on the thickness information of the graphene dispersion liquid film; A position coordinate system is established in the thickness distribution model, a coordinate axis is established with the center of the silicon wafer as an endpoint, the surface of the silicon wafer is uniformly divided into a plurality of points, and the coordinate values of each point are determined; The thickness distribution model is compared with the thickness reference model, the thickness value of each point of the thickness distribution model is compared with the thickness value of the thickness reference model to obtain a thickness comparison value, and the thickness comparison value is recorded; if the absolute value of the thickness comparison value is less than the correction value, the point is marked as a normal point, and the thickness comparison value is recorded as 0; if the absolute value of the thickness comparison value is greater than the correction value, the point is marked as an abnormal point, and the abnormal points connected to each other are integrated to obtain a plurality of abnormal regions; An outer contour of each abnormal region on the graphene dispersion liquid film is obtained; The coordinate region of the abnormal region is determined according to the coordinate value of the abnormal point, and the actual position of the abnormality is calculated; wherein the method for determining the coordinate region is: taking the outermost points of the upper, lower, left and right four sides of the abnormal region as tangent points, drawing tangent lines, and determining the range occupied by the abnormal region according to the coordinates of the tangent lines; The abnormal region is determined to be a depression or a protrusion, and the specific method is: taking the center point of the abnormal region as the center of a circle to radiate outward in a circular manner; the points with a thickness comparison value greater than 0 and less than 0 are marked respectively; the abnormal region with a thickness comparison value greater than 0 is determined to be a depression; and the abnormal region with a thickness comparison value greater than 0 is determined to be a protrusion; The abnormal region determined to be a depression is subjected to secondary coating until the thickness values of all the abnormal regions determined to be depressions are greater than the thickness value of the thickness reference model; the surrounding predetermined region of the coated abnormal region after the secondary coating is subjected to coating thickness uniformity detection; it is determined whether the thickness comparison value of the region is greater than the correction value; if the thickness comparison value is greater than the correction value, the point is marked as an abnormal point, and the abnormal points connected to each other are integrated to obtain a plurality of abnormal regions; and the plurality of abnormal regions on the graphene dispersion liquid film are removed; In real-time monitoring, the collected data is input into the trained prediction model for classification, the thickness of the graphene dispersion liquid film is predicted according to the coating data, and the coating amount is adjusted according to the prediction result; wherein the method for training the model is: Collecting data related to the preparation process of the graphene oxide film, including the coating amount and the corresponding thickness change of the graphene dispersion liquid film; Pretreating the collected data, including removing noise, processing missing values and standardizing data, extracting features from the pretreated data, and selecting features to determine a feature set for training the SVM model; Labeling the extracted feature set to distinguish between normal preparation and abnormal conditions, and integrating the feature set into a data set; the data set is divided into a training set and a test set by cross-validation; The training set is used to train the SVM model, which learns and identifies the feature patterns of normal and abnormal conditions by inputting the labeled data; The trained model is verified using the test set to evaluate the performance and generalization ability of the model; if the indicators are not met, the hyperparameters of the model are adjusted and retrained; if the indicators are met, the training is completed; The trained SVM model is applied to real-time prediction.

8. The method of claim 1, wherein the graphene is patterned by a photolithography process. In step S2, the method for coating the graphene oxide dispersion liquid on the surface of the silicon wafer is any one of drop coating, casting, immersion, screen printing, inkjet printing, roller coating, doctor blade coating and spin coating.

9. A patterned graphene-on-silicon wafer device manufactured based on the method of claim 1-8, comprising a silicon wafer substrate, and a patterned graphene film covering the surface of the silicon wafer substrate.

Images