Electric field sensor gain increasing method based on nitride quantum dots

Abstract

The invention relates to the technical field of electric field sensors, in particular to an electric field sensor gain increasing method based on nitride quantum dots. The method comprises the following steps: firstly, constructing a nitride and graphene heterojunction model by utilizing a first principle, calculating quantum capacitance, and screening out optimal quality structure configuration; preparing a sensor according to the configuration and the initial process parameters, and obtaining the actual measurement sensitivity in the simulated electric field; further constructing an optimization model taking the quantum dot size and the dielectric layer thickness as input and the sensitivity as a target, calculating a parameter gradient by utilizing a gradient ascending method, and outputting a correction value; and finally, carrying out closed-loop iteration of preparation, testing, calculation and adjustment until the objective function is converged, and outputting final process parameters. According to the invention, collaborative optimization of microcosmic electronic structure screening and macroscopic process parameters is realized, and the detection sensitivity and the signal-to-noise ratio of the electric field sensor are remarkably improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of electric field sensors, and particularly relates to a gain improvement method for an electric field sensor based on nitride quantum dots. BACKGROUND

[0002] With the rapid development of smart grid, ubiquitous power Internet of Things and advanced industrial automation technology, the demand for non-contact voltage detection and weak electric field sensing is increasingly urgent. As the core sensing element for obtaining the operation state of a power system, monitoring environmental electromagnetic safety and realizing high-precision medical detection, the performance of an electric field sensor directly determines the reliability and sensitivity of the system. Among numerous sensing materials, wide-bandgap semiconductor materials (such as gallium nitride, aluminum nitride and other nitrides) have become a research hotspot due to their high breakdown field strength, excellent thermal stability and piezoelectric properties. In particular, when these nitrides exist in the form of quantum dots, their energy level structure changes significantly under the influence of quantum confinement effect, and they can exhibit stronger photoelectric response capability. At the same time, graphene, as a two-dimensional material with ultra-high carrier mobility and single-atom layer thickness, is often used as a high-sensitivity charge transport channel. Combining nitride quantum dots and graphene to construct a heterojunction composite material can theoretically combine the advantages of both and realize efficient capture and amplification of weak electric field signals, which is an important development direction for the next generation of high-performance electric field sensors.

[0003] However, although the nitride quantum dot and graphene heterojunction has great application potential in theory, it still faces many severe challenges in actual device design and preparation process. The current research and development mode mostly relies on the traditional "empirical trial and error method", that is, by adjusting the material ratio, deposition temperature or quantum dot size through a large number of experimental repetitions. This method lacks in-depth understanding and prediction of the micro-electronic behavior of the heterojunction interface (such as energy band alignment, interface charge transfer and state density distribution), resulting in great blindness in the experimental process. At the micro level, due to the lack of accurate theoretical guidance, the prepared heterojunction often has serious lattice mismatch and interface defect states, which will cause strong carrier scattering, significantly reduce the interface coupling efficiency, and then introduce large low-frequency noise (1 / f noise), greatly limiting the signal-to-noise ratio of the sensor. In addition, the size distribution control of quantum dots and the matching of dielectric layer thickness usually use rough process parameter setting, which is difficult to achieve precise regulation at the nanometer level. In complex actual application environments (such as temperature fluctuations and humidity changes), this non-optimized structure design is easy to cause sensor sensitivity drift, linearity decline and long-term stability deficiency, which cannot meet the stringent requirements of high-voltage transmission lines or precision instruments for high-fidelity detection of weak electric fields.

[0004] Therefore, the technical problem to be solved in the prior art is how to overcome the blindness of the traditional empirical development mode, establish a systematic method capable of deeply accessing the micro-electronic structure level, accurately matching the band characteristics of the material interface, and effectively solving the complex nonlinear coupling relationship between the key process parameters such as the quantum dot size and the thickness of the dielectric layer and the macroscopic performance (such as sensitivity and gain) of the sensor, so as to significantly improve the gain performance and response consistency of the electric field sensor for weak signals under the premise of ensuring long-term stability. SUMMARY

[0005] The purpose of the present application is to solve the above problems and provide a gain improvement method for an electric field sensor based on nitride quantum dots.

[0006] In a first aspect, a gain improvement method for an electric field sensor based on nitride quantum dots adopts the following technical solution: A gain improvement method for an electric field sensor based on nitride quantum dots, the method comprising the following steps: Step (1), obtain the crystal structure data of graphene and nitride, construct a heterojunction model of nitride and graphene using first principles, and set a vacuum layer in the Z-axis direction; perform geometric structure optimization and single-point energy calculation on the heterojunction model to obtain band structure and state density parameters; based on the converged electronic structure calculation results, further calculate the quantum capacitance value of the system, and select the heterojunction model with the highest quantum capacitance value as the optimal heterostructure configuration; Step (2), according to the optimal heterostructure configuration selected in step (1), set the initial quantum dot size and dielectric layer thickness as process parameters, prepare a nitride-graphene heterojunction material, and load the material on the surface of an electrode to obtain a to-be-tested electric field sensor; place the to-be-tested electric field sensor in a simulated electric field environment, obtain the output current and voltage response data of the sensor under the preset electric field intensity and temperature conditions; and calculate the sensitivity and signal-to-noise ratio of the sensor; Step (3), construct an optimization model with quantum dot size and dielectric layer thickness as input variables and sensitivity as the objective function; use the gradient ascent method to calculate the gradient direction of the objective function with respect to the input variables according to the sensitivity index obtained in step (2), and output the quantum dot size correction value and dielectric layer thickness correction value for the next iteration; Step (4), adjust the quantum dot size and dielectric layer thickness in step (2) according to the correction values output in step (3) to re-prepare the electric field sensor, and repeat steps (2) to (3) until the change rate of the objective function is lower than a preset threshold, and output the final sensor preparation process parameters.

[0007] Further, in step (1), the geometric structure optimization includes the following process: The exchange-correlation potential is described using a PBE functional under the generalized gradient approximation (GGA). An ultrasoft pseudopotential is employed to handle the interaction between the ionic core and valence electrons. A cutoff energy of 380 eV is set, and a k-point grid with a density of 4×4×1 is generated using the Monkhorst-Pack method. The convergence criterion is an energy convergence threshold of 1.0×10⁻⁶. -5 eV / atom, interatomic force convergence threshold of 0.03 eV / Å, and maximum internal stress below 0.05 GPa.

[0008] Furthermore, in step (1), the single-point energy calculation includes the following process: The band structure is a high-symmetry point path calculation electronic band, which is used to directly obtain the electronic transport parameters of the material's band gap width and effective carrier mass; The density of states includes the total density of states and the partial density of states, which are used to qualitatively and quantitatively assess the contribution of different atomic orbitals to electronic states near the Fermi level.

[0009] Furthermore, in step (1), the formula for the quantum capacitance value is as follows: ; in, For quantum capacitors, For elementary charge, The density of states at the Fermi level; The heterojunction model of the nitride and graphene is selected from one of aluminum nitride / graphene, boron nitride / graphene, or silicon nitride / graphene.

[0010] Furthermore, in step (2), the method for preparing nitride-graphene heterojunction materials includes chemical synthesis or physical deposition. The chemical synthesis method includes the following process: under an inert atmosphere, a nitride precursor is mixed with a graphene dispersion, the reaction temperature is 150℃~300℃, the reaction time is 1h~5h, quantum dots are induced to grow in situ on the graphene surface to form a uniform heterojunction, and the interface is annealed at a temperature of 300℃~500℃ to obtain the nitride-graphene heterojunction material. The physical deposition method includes the following process: using chemical vapor deposition or magnetron sputtering technology, nitride materials are deposited onto a pretreated graphene substrate in a vacuum environment, and by adjusting the deposition rate and substrate temperature, quantum dots are ensured to be distributed in a single layer or a few layers, and the interface is annealed at a temperature of 300℃~500℃ to obtain the nitride-graphene heterojunction material.

[0011] Furthermore, step (2), specifically the step of loading the nitride-graphene heterojunction material onto the electrode surface, includes: The prepared nitride-graphene heterojunction material is coated on the surface of the interdigital electrode by using a spin coating or drop coating technology; and low-temperature annealing is carried out under an inert atmosphere, wherein the temperature of the low-temperature annealing is 100 DEG C to 200 DEG C, and the time of the low-temperature annealing is 30 min to 60 min.

[0012] Further, in step (2), the simulated electric field environment is constructed by a programmable electric field generator and an environment control cabin, the electric field strength is 0.1 kV / m to 10 kV / m, the temperature is -20 DEG C to 80 DEG C, and the frequency is 50 Hz to 1000 Hz. The sensitivity is the ratio of the output current change amount to the electric field strength change amount.

[0013] Further, in step (3), the gradient ascent method is provided with a parameter constraint condition, wherein the constraint range of the quantum dot size is 10 nm to 100 nm, and the constraint range of the dielectric layer thickness is 50 nm to 200 nm; and the preset threshold value is that the change rate of the sensitivity is less than 1%.

[0014] In a second aspect, an electric field sensor adopts the following technical solution: An electric field sensor is prepared by the above method. The electric field sensor comprises an interdigital electrode and a nitride-graphene heterojunction material loaded on the surface of the interdigital electrode, wherein the average size of the nitride quantum dots in the nitride-graphene heterojunction material is 5 nm to 7 nm, and the dielectric thickness is 20 nm to 30 nm.

[0015] In a third aspect, an electric power system monitoring device adopts the following technical solution: The electric power system monitoring device comprises the electric field sensor, a data acquisition unit and a signal processing unit, and is configured to monitor the weak electric field change around a high-voltage transmission line, and the detection limit is lower than 0.01 kV / m.

[0016] The present application has the following advantages: This invention provides a method for enhancing the gain of an electric field sensor based on nitride quantum dots. It employs a closed-loop approach combining first-principles calculations and gradient-ascent iterative optimization. First, based on microscopic electronic structure simulation, nitride-graphene heterojunction configurations with optimal band alignment and interface coupling characteristics are selected using quantum capacitance as a key indicator. This computationally guided design strategy directly optimizes carrier transport paths at the atomic scale, effectively suppressing nonradiative recombination and noise caused by interface defect states, and significantly enhancing the heterojunction's initial capture and amplification capabilities for weak electric field signals. Furthermore, by establishing a mathematical model between sensitivity and process parameters and utilizing the gradient-ascent algorithm for closed-loop feedback of experimental data, this method overcomes the challenges of multi-parameter nonlinear coupling that traditional single-factor experiments struggle with, achieving precise nanoscale coordinated control of quantum dot size and dielectric layer thickness. This dynamic optimization mechanism ensures optimal matching between the quantum confinement effect of the quantum dots and the polarization characteristics of the dielectric layer, thereby maximizing sensor gain performance while minimizing energy loss. Ultimately, this method not only significantly shortens the development cycle of high-performance sensors and reduces trial-and-error costs, but also significantly improves the detection sensitivity, signal-to-noise ratio, and batch consistency of finished electric field sensors, enabling them to maintain excellent detection limits and long-term reliability even in complex environments. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a method for enhancing the gain of an electric field sensor based on nitride quantum dots, provided in an embodiment of the present invention.

[0018] Figure 2 The diagram shows the microscopic atomic model of the nitride-graphene heterojunction constructed using first principles and optimized for geometry in this embodiment of the invention; wherein, (a) is the aluminum nitride / graphene heterojunction model, (b) is the boron nitride / graphene heterojunction model, and (c) is the silicon nitride / graphene heterojunction model.

[0019] Figure 3 The above are the band structure diagrams of three different heterojunction systems calculated based on the optimized model in this embodiment of the invention; wherein, (a) corresponds to the aluminum nitride / graphene system, (b) corresponds to the boron nitride / graphene system, and (c) corresponds to the silicon nitride / graphene system.

[0020] Figure 4 The diagram shows the density of states (DOS) distribution of three different heterojunction systems calculated in the embodiments of the present invention; where (a) corresponds to the aluminum nitride / graphene system, (b) corresponds to the boron nitride / graphene system, and (c) corresponds to the silicon nitride / graphene system. Detailed Implementation

[0021] In order to make the objects, technical solutions and advantages of the present application clearer, the following further describes the present application with reference to the accompanying drawings, the described embodiments should not be regarded as limitations to the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0022] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. Unless otherwise defined, all technical and scientific terms used in the embodiments of the present application have the same meaning as understood by those skilled in the art to which the embodiments of the present application belong. The terms used in the embodiments of the present application are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0023] Those skilled in the art should understand that in the following description of the embodiments of the present application, the order of the serial numbers does not mean the order of execution, and some or all steps can be executed in parallel or in sequence, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0024] The terms used in the embodiments of the present application are only for the purpose of describing the specific embodiments and are not intended to limit the present application. The singular forms "a" and "the" used in the embodiments of the present application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise.

[0025] Those skilled in the art should understand that in the embodiments of the present application, the numerical range should be understood as also specifically disclosing each intermediate value between the upper limit and the lower limit of the range. Each smaller range between any stated value and stated range and any stated value or intermediate value within the stated range is also included in the present application. The upper limit and lower limit of these smaller ranges can be independently included or excluded from the range.

[0026] Unless otherwise specified, the technical / scientific terms used herein have the same meaning as generally understood by those skilled in the art of the present application. Although only preferred methods and materials are described in the present application, any methods and materials similar or equivalent to those described herein can also be used in the embodiments or test examples of the present application. All documents mentioned in the specification are generally incorporated by reference to disclose and describe the methods and / or materials related to the documents. In the event of conflict between any incorporated document and the content of the present application, the content of the present application shall prevail.

[0027] It should be noted that all raw materials and / or reagents in the embodiments of the present application are purchased on the market or prepared according to conventional methods well known to those skilled in the art.

[0028] The embodiment provides a method for improving the gain of a nitride quantum dot-based electric field sensor, which improves the gain of the electric field sensor by Figure 1 The method solves the technical problems of high trial and error cost, low interface coupling efficiency and difficulty in accurately matching macroscopic performance and microscopic structure parameters in traditional electric field sensor research and development. The method not only covers material screening at the atomic scale to macroscopic device preparation and testing, but also introduces a mathematical optimization algorithm to dynamically adjust key process parameters, thereby comprehensively improving the sensitivity, signal-to-noise ratio and stability of the electric field sensor.

[0029] I. First-principles calculation simulation and heterostructure screening.

[0030] As shown in Figure 1 The embodiment first performs step S1, i.e., calculation simulation and heterostructure screening. This step aims to theoretically screen the most sensitive material combination to electric field response, avoiding blind experimental exploration.

[0031] 1. Model construction and parameter setting The crystallographic data of graphene and the nitride to be screened (such as aluminum nitride AlN, boron nitride BN, silicon nitride SiN, etc.) are obtained, which are usually extracted from inorganic crystal structure databases or related crystallographic literature, including lattice constants, atomic coordinates and space group information. Use the Visualizer module in the Materials Studio software or similar atomic-scale modeling tools to construct a periodic supercell model of graphene. In order to simulate the characteristics of two-dimensional materials and eliminate the periodic boundary conditions, a non-physical interlayer interaction must be introduced in the normal direction of the graphene plane (Z axis). A vacuum layer with a thickness of 15 Å~20 Å must be set in the normal direction of the graphene plane (Z axis) to ensure that the electronic wave functions between adjacent mirror layers do not overlap.

[0032] Construct a cut surface model of the nitride material and place it on the graphene substrate to form a heterojunction. In this process, the key is to handle the lattice mismatch between the two materials. The embodiment uses the coherent approximation method to control the lattice constant mismatch between the two materials within 5% by appropriately rotating or expanding the cell (constructing a supercell, such as a 4x4 graphene matching a 3x3 nitride), thereby minimizing the stress introduced at the interface. As shown in Figure 2 Aluminum nitride / graphene (AlN / G) Figure 2 a), boron nitride / graphene (BN / G) Figure 2 b), and silicon nitride / graphene (SiN / G) Figure 2c) the microscopic heterojunction initial model, in which the ball-and-stick model clearly shows the arrangement of different atoms (blue for nitrogen atoms, gray for carbon atoms, pink for aluminum atoms, and yellow for silicon atoms) at the interface.

[0033] After the model is constructed, the Dmol3 module or VASP software based on density functional theory is used for geometric structure optimization. In order to ensure the balance between the accuracy of the calculation results and the calculation efficiency, the exchange correlation potential is selected as the PBE functional under the generalized gradient approximation (GGA), which can well describe the electronic structure of solid materials and the surface adsorption energy. For the interaction between the ionic and valence electrons, the ultra-soft pseudo-potential is used, which helps to reduce the number of plane wave basis and improve the calculation speed. The key calculation parameters are set as follows: the plane wave cutoff energy is set to 380 eV to ensure the convergence accuracy of the wave function expansion; the Monkhorst-Pack method is used to generate the k-point grid for the Brillouin zone integration, and the grid density is set to 4x4x1 to accurately capture the electronic properties of two-dimensional materials in reciprocal space.

[0034] 2. Geometric optimization and convergence criterion The purpose of geometric structure optimization is to find the lowest energy point on the potential energy surface of the system, i.e., the most stable atomic configuration. During the optimization process, the atomic positions and cell parameters are constantly adjusted until the following strict convergence criteria are met: the total energy change of the system is less than 1.0x10 -5 eV / atom, the maximum force acting on each atom is less than 0.03 eV / Å, and the maximum internal stress in the cell is less than 0.05 GPa. Only when the above conditions are met simultaneously, the structure is considered to have reached a stable state, as shown in the optimized structure Figure 2 , it can be seen that a stable van der Waals contact or weak chemical bonding is formed between the nitride layer and the graphene layer, and the interface distance is within a physically reasonable range (usually about 3.0 Å).

[0035] 3. Electronic structure calculation and quantum capacit screening Based on the stable structure of the optimized convergence, further single-point energy calculation is carried out to obtain detailed electronic properties. Key parameters include the band structure and density of states.

[0036] As shown in Figure 3 , the band structures of aluminum nitride / graphene Figure 3 a), boron nitride / graphene Figure 3 b), and silicon nitride / graphene Figure 3 c) along the high-symmetry point path The band structure of a material is shown in the diagram. Band diagrams visually reflect the material's electrical conductivity and bandgap characteristics. It is clearly visible in the diagram that different heterojunctions exhibit significant differences in band dispersion near the Fermi level (shown by the dashed line, typically set to 0 eV). For example, Figure 3 The a-value shows that a tiny band gap of about 0.01 eV is opened near the Fermi level, while Figure 3 b shows a larger band gap (0.101 eV). Figure 3 The band gap of c is 0.009 eV. The size of the band gap directly affects the thermal noise level of the sensor at room temperature and the excitation efficiency of charge carriers.

[0037] like Figure 4 As shown, the corresponding density of states (DOS) was calculated, including the total density of states (TDOS) and the partial density of states (PDOS). The density of states plot reveals the distribution density of electronic states in different energy ranges. By analyzing the DOS peaks near the Fermi level, the charge accumulation capability of the heterojunction interface can be qualitatively assessed. For example, Figure 4 The aluminum nitride / graphene system exhibits a sharp DOS peak near the Fermi level, suggesting that it may have a high density of states.

[0038] To quantitatively screen for the optimal heterojunction configuration, this embodiment introduces "quantum capacitance (C)". Q "Quantum capacitance" is used as the core screening criterion. In low-dimensional nanomaterials, due to the finite density of states, the amount of charge required to fill the energy level when the potential changes is finite; this effect manifests as quantum capacitance. For electric field sensors, higher quantum capacitance means that under the same external electric field, the sensing layer can accumulate more induced charge, thereby generating a larger output signal. The formula for calculating quantum capacitance is as follows:

[0039] in, elementary charge ( C), This represents the density of states at the Fermi level. In practice, the program automatically extracts the calculated value. Numerical values ​​are substituted into formulas to calculate the values ​​of each heterojunction system. The system calculates and sorts the values. Finally, it selects the heterojunction model with the highest quantum capacitance value (such as the aluminum nitride / graphene or silicon nitride / graphene combination with a specific crystal facet selected in this embodiment) as the optimal heterostructure configuration and outputs it to the subsequent fabrication steps.

[0040] II. Material preparation and sensor integration.

[0041] like Figure 1As shown in the flowchart, after the theoretical screening is completed, the initial process parameters are set and the material preparation phase is entered. The purpose of this phase is to convert the optimal model calculated in theory into an actual physical device and ensure accurate control of the microstructure.

[0042] 1. Set initial process parameters According to the optimal heterojunction type output in step S1, the initial process parameters are set. The key control variables are the quantum dot size and the dielectric layer thickness. For example, the initial quantum dot size is set to 5 nm and the dielectric layer thickness is set to 25 nm. These parameters will serve as control targets for subsequent chemical or physical preparation processes.

[0043] 2. Preparation of nitride-graphene heterojunction material This embodiment provides two alternative high-precision preparation paths: chemical synthesis method and physical deposition method, which can be selected according to the equipment conditions and yield requirements.

[0044] Path one: chemical synthesis method (solvent thermal method / colloidal chemical method) This method is suitable for large-scale and low-cost preparation. The specific operation is carried out in an inert gas atmosphere glove box filled with argon or nitrogen to prevent the precursor from being oxidized. The selected nitride precursor (such as anhydrous aluminum chloride for preparing AlN or borane ammonia for preparing BN) is dissolved in an organic solvent, and a previously prepared single-layer or few-layer graphene dispersion solution is added. Ultrasonic dispersion is used to ensure uniform suspension of graphene nanosheets. Then, the mixed solution is transferred to a high-pressure reaction kettle for solvent thermal reaction at a temperature range of 150°C to 300°C, and the reaction time is controlled within 1h to 5h. Under this high-temperature and high-pressure environment, the nitride precursor decomposes and heterogeneously nucleates on the defect sites or lattice matching sites of the graphene surface, growing in situ nitride quantum dots. In order to accurately control the size of the quantum dots (so that they meet the initial set process parameters, such as 2nm to 10nm), an appropriate amount of surfactant (such as oleic acid, oleylamine or polyvinylpyrrolidone PVP) is added to the reaction solution to limit the excessive growth and agglomeration of the crystal nucleus by adjusting the ratio of the precursor concentration to the surfactant. After the reaction is completed, the product is centrifuged, washed and dried to obtain a powder material. Finally, in order to optimize the interface quality, the powder is annealed at 300°C to 500°C to eliminate residual organic matter in the synthesis process, repair lattice defects and enhance interface chemical bonding.

[0045] Path two: physical deposition method (CVD / PVD) This method is suitable for scenarios with extremely high requirements for film quality and thickness. Using a chemical vapor deposition (CVD) or magnetron sputtering (PVD) system, the nitride precursor is deposited on the graphene surface in a high vacuum (background vacuum better than 10 The deposition process is carried out in an environment of (Pa). First, a high-quality graphene film is transferred onto a substrate as a deposition template. For magnetron sputtering, a high-purity nitride target (or a metal target with nitrogen gas for reactive sputtering) is used, and the deposition rate is precisely controlled (typically between 0.1 Å / s and 1 Å / s) by adjusting the sputtering power (RF / DC power supply), working gas pressure, and target-substrate distance. To obtain discrete quantum dots rather than continuous films, the deposition time needs to be strictly controlled, and the surface diffusion length of adsorbed atoms is adjusted using the substrate temperature (adjustable from room temperature to 400 °C), thereby forming island-like quantum dots using the Volmer-Weber growth mode. Similarly, after deposition, in-situ or ex-situ annealing is performed at 300 °C to 500 °C to promote the rearrangement of interface atoms and form a heterojunction interface with low defect density.

[0046] 3. Sensor Integration and Electrode Loading After the heterojunction material is prepared, it needs to be integrated onto the sensor electrode. The electrode is usually fabricated as an interdigitated electrode on a flexible PI substrate or a rigid ceramic substrate using photolithography. The electrode material is selected from chemically stable materials such as gold (Au) or platinum (Pt) to form a good ohmic contact.

[0047] If the powder material is prepared using chemical methods, it needs to be dispersed in a suitable solvent to form an electronic paste, which is then coated onto the surface of the interdigitated electrodes using spin coating or drop coating techniques. The spin coating speed (e.g., 1000 rpm to 3000 rpm) and time determine the uniformity and thickness of the coating. If a physical method is used, the material is deposited directly onto the electrode.

[0048] After coating, the sensor is placed in a tube furnace and subjected to low-temperature annealing under an inert atmosphere. The annealing temperature is strictly controlled between 100℃ and 200℃, and the time is 30 min to 60 min. The purpose of this low-temperature annealing step is to remove solvent residue, enhance the physical adhesion between the heterojunction material and the metal electrode, and at the same time avoid the oxidation or agglomeration of quantum dots caused by high temperature, thus ensuring the mechanical stability of the sensor structure.

[0049] III. Performance Testing and Data Acquisition.

[0050] like Figure 1 As shown in the process flow, the fabricated sensor under test is sent to a customized testing platform for performance evaluation. This testing step is crucial in connecting fabrication and optimization, providing realistic and quantitative feedback data.

[0051] 1. Construction of simulated electric field environment The test platform consists of an environmental control chamber, a programmable electric field generator, and a high-precision data acquisition unit. To simulate operating conditions in real power systems or complex environments, the environmental control chamber is equipped with a Peltier temperature control module and a humidity generator, capable of generating temperature fluctuations from -20℃ to 80℃ and varying humidity levels to test the sensor's environmental robustness. The electric field generator uses a high-voltage amplifier to drive a parallel-plate capacitor structure, generating a uniform and controllable electric field between the two plates. The preset electric field strength range covers 0.1 kV / m to 10 kV / m, and the frequency range covers power frequency (50 / 60 Hz) to intermediate frequency (1 kHz) to comprehensively evaluate the sensor's dynamic response characteristics.

[0052] 2. Data Collection and Indicator Calculation Place the sensor in the simulated electric field described above, and connect its output to a lock-in amplifier or a high-precision digital oscilloscope via a shielded cable. The lock-in amplifier utilizes phase-sensitive detection technology to extract the weak sensor response signal from a strong noise background. At a preset electric field strength (… Under the conditions of temperature and temperature, the output current of the sensor is recorded in real time. ) or voltage ( ) Response data.

[0053] After acquiring the raw data, it is processed by the built-in processing unit. First, the sensitivity (S) is calculated. Sensitivity is defined as the ratio of the change in the output signal to the change in the input electric field intensity, i.e.:

[0054] or

[0055] Sensitivity is typically determined by linearly fitting the electric field strength-output signal curve, and the slope of the curve is the sensitivity.

[0056] Secondly, the signal-to-noise ratio (SNR) is calculated. A Fast Fourier Transform (FFT) is performed on the baseline signal of the sensor when there is no electric field input to obtain the noise power spectral density. The total noise power is then obtained by integration. The SNR is the ratio of signal power to noise power under a specific electric field. The sensitivity and SNR data obtained from these calculations will serve as the input values ​​for the objective function of subsequent optimization algorithms.

[0057] IV. Iterative optimization using the gradient ascent method.

[0058] Unlike the traditional single-variable controlled variable method, this embodiment introduces a gradient-based automatic optimization algorithm to solve the complex nonlinear coupling problem between quantum dot size and dielectric layer thickness.

[0059] 1. Optimize model construction In step S3, a mathematical optimization model is constructed. Set the input variable vector:

[0060] wherein is the quantum dot size, is the dielectric layer thickness. Set the objective function is the sensitivity of the sensor . The goal of optimization is to find the optimal such that is maximized.

[0061] To ensure the feasibility of the process, set the parameter constraint condition: the constraint range of the quantum dot size is 10 nm~100 nm, and the constraint range of the dielectric layer thickness is 50 nm~200 nm.

[0062] 2. Gradient calculation and parameter update The gradient ascent method is used for iterative optimization. The core of the algorithm is to calculate the gradient direction of the objective function with respect to the input variable . In experiments, the expression of the analytical gradient is usually unknown, so the numerical differentiation method (such as finite difference method) is used to estimate the gradient. Specifically, a small perturbation is applied near the current parameter point (for example, a group of samples slightly deviating from the current size and thickness is prepared), and the partial derivative is calculated according to the test results:

[0063]

[0064] After obtaining the gradient vector, the parameter correction value of the next iteration is calculated according to the following formula:

[0065] wherein, is the learning rate (step size), which is used to control the amplitude of parameter adjustment to avoid oscillation. The system outputs the new quantum dot size correction value and dielectric layer thickness correction value calculated.

[0066] 3. Closed-loop iteration and convergence As shown in the judgment block in Figure 1 , according to the correction value output in step S3, adjust the process parameters in step S2 (for example, change the precursor concentration or deposition time to adjust the size, change the spin coating speed or deposition time to adjust the thickness), re-prepared a batch of electric field sensors, and repeat the test and calculation process of steps S2 to S3.

[0067] After each iteration, a convergence condition is checked. A preset threshold is usually set to be a relative change rate of the objective function (sensitivity) less than 1%, or a preset maximum number of iterations is reached, or the parameter update tends to zero. When the convergence condition is met, the algorithm terminates.

[0068] V. Product and system application 1. Final process parameters and sensor structure After the above-mentioned multiple rounds of closed-loop iterative optimization of "preparation-test-calculation-adjustment", the system finally outputs a set of optimal sensor preparation process parameters. In a typical experiment of the present embodiment, the parameters finally converge to: quantum dot size of 5-7 nm, dielectric layer thickness of 20-30 nm.

[0069] The high-gain electric field sensor prepared based on the optimal parameters has a macroscopic structure including a substrate, a gold or platinum interdigital electrode on the substrate, and a nitride-graphene heterojunction sensing layer loaded on the surface of the electrode. Microscopically, the sensing layer is composed of a single-layer or few-layer graphene sheet and nitride quantum dots with uniform size (5-7 nm) grown or deposited in situ on the surface thereof, and the overall dielectric layer thickness is precisely controlled to be 20-30 nm. This structure design maximizes the quantum confinement effect of quantum dots and the high conductivity of graphene, and the optimized dielectric layer thickness ensures insulation while allowing effective field effect modulation, thereby achieving extremely high detection sensitivity and a detection limit as low as 0.01 kV / m.

[0070] 2. Power system monitoring device The high-gain electric field sensor can be further integrated into a power system monitoring device. The device includes the above-mentioned sensor probe, a front-end signal conditioning circuit (low-noise preamplifier), a data acquisition card, and a back-end signal processing unit (DSP or host computer). The device is configured to non-contact monitor the weak electric field changes around high-voltage transmission lines and substation equipment. In actual operation, the signal processing unit processes the collected time-domain waveforms, and using the optimized signal-to-noise ratio characteristics, it can sensitively capture the tiny electric field distortions caused by insulator degradation, partial discharge, or wire icing, thereby achieving early fault warning. In addition, due to the excellent biocompatibility (nitride and carbon materials) and high sensitivity of the sensor, the device can also be modified for non-contact detection of bioelectric fields (such as electrocardiogram and electroencephalogram), and has broad application prospects.

[0071] In summary, the present embodiment successfully overcomes the defects of low research and development efficiency and poor performance consistency in the prior art by scientifically rigorous theoretical calculation to guide material design and advanced closed-loop optimization algorithm to guide process parameter adjustment, and provides a reproducible, high-performance electric field sensor gain enhancement complete solution.

[0072] Based on the foregoing embodiments, in order to further demonstrate the effectiveness of the "computational simulation screening + iterative optimization" method described in the present application, two groups of preferred examples (Example 1, Example 2) and three groups of comparative examples (Comparative Example 1, Comparative Example 2, Comparative Example 3) are provided below.

[0073] The tests of these examples and comparative examples are all carried out under the same standard test environment: ambient temperature 25°C, relative humidity 40%, input electric field frequency 50Hz, electric field intensity range 0.1~10 kV / m.

[0074] Example Example 1 This Example 1 uses the method of the above-mentioned embodiments to prepare an electric field sensor.

[0075] 1. Screening stage (step S1): Through first-principle calculation, three kinds of heterojunctions of aluminum nitride (AlN) / graphene, gallium nitride (GaN) / graphene and zinc oxide (ZnO) / graphene are compared. The calculation results show that the state density of AlN / graphene heterojunction at the Fermi level is the highest, and the calculated quantum capacitance value is 32.5 (as a comparison, ZnO / graphene is only 18.2 ), so AlN / graphene is selected as the optimal configuration.

[0076] 2. Preparation and iteration: Initial parameter setting: the quantum dot size is set to 15 nm, and the dielectric layer thickness is set to 50 nm.

[0077] Iteration process: after the first round of preparation and testing, the sensitivity is 2.1 mV / (kV / m). Substituting into the gradient ascent method model calculation, it is suggested to reduce the size and thickness.

[0078] Final parameters: after 4 rounds of closed-loop iterative optimization, the algorithm converges. The final process parameters are determined as: the average size of aluminum nitride quantum dots is 6.2 nm, and the dielectric layer thickness is 24 nm.

[0079] 3. Final product: the heterojunction with the above-mentioned parameters is prepared by solvothermal method, and integrated on a gold interdigital electrode, and annealed at 150°C.

[0080] Example 2 This Example 2 also uses the method of the above-mentioned embodiments, but selects a second preferred material in the screening stage.

[0081] 1. Screening stage (step S1): the calculation shows that the quantum capacitance value of boron nitride (BN) / graphene is 28.4 It is slightly lower than AlN, but better than traditional materials, and was selected as an alternative.

[0082] 2. Preparation and iteration (steps S2-S4): Prepared by physical magnetron sputtering.

[0083] Iterative process: Initially, the dimensions were set at 8 nm and the thickness at 40 nm. After algorithm optimization, it was found that BN has better insulation properties, allowing for thinner dielectric layers without breakdown.

[0084] Final parameters: After three rounds of iteration, the final process parameters were determined as follows: average size of boron nitride quantum dots 5.5 nm, and dielectric layer thickness 21 nm.

[0085] Final product: The material is magnetron sputtered onto a platinum interdigitated electrode and then annealed at 400°C.

[0086] Comparative Example Comparative Example 1 Comparative Example 1 omits the calculation and screening in step S1 and directly selects zinc oxide (ZnO) quantum dots and graphene composites, which are commonly used in the prior art.

[0087] 1. Material selection: ZnO / graphene.

[0088] 2. Process parameters: Referring to the general parameters in existing literature, the quantum dot size was set to 6 nm and the dielectric layer thickness to 25 nm.

[0089] 3. Preparation method: The same solvothermal method as in Example 1 was used.

[0090] Comparative Example 2 Comparative Example 2 used the selected AlN / graphene material, but omitted the iterative optimization steps S3-S4 and adopted empirical parameters.

[0091] 1. Material selection: AlN / graphene.

[0092] 2. Process parameters: The quantum dot size distribution is 15~20 nm, and the dielectric layer thickness is 100 nm.

[0093] 3. Preparation method: Prolonged preparation time leads to quantum dot aggregation and film thickness.

[0094] Comparative Example 3 This comparative example 3 illustrates the case where miniaturization was pursued but no algorithmic constraints were applied.

[0095] 1. Material selection: AlN / graphene.

[0096] 2. Process parameters: quantum dot size <2 nm, dielectric layer thickness <10 nm.

[0097] Preparation method: very short time of deposition.

[0098] Experimental results and analysis The above 5 groups of sensors were tested under the same conditions, and the results are shown in Table 1 below.

[0099] Table 1 Performance test data of Examples 1-2 and Comparative Examples 1-3

[0100] As can be seen from the data in Table 1, under the condition that the physical size is similar (both 6 nm / 25 nm), the sensitivity of Example 1 (AlN / G) (5.82) is much higher than that of Comparative Example 1 (ZnO / G, 2.30). This verifies the effectiveness of step S1. First-principle calculation predicts that the AlN system has a higher quantum capacitance (32.5 vs 18.2 ), which means that under the same external electric field, the heterojunction interface of Example 1 can accumulate more charges, thereby generating a larger induced signal. If no calculation screening is performed and only empirical selection of materials is used, the best gain substrate cannot be obtained.

[0101] Using the same material, the sensitivity of Example 1 is more than 5 times that of Comparative Example 2 (5.82 vs 1.15). In Comparative Example 2, the quantum dot size is too large (15-20 nm), which weakens the quantum confinement effect, resulting in insufficient energy level discretization and poor photoelectric response; at the same time, the thick dielectric layer (100 nm) increases the electric field shielding effect and tunneling resistance, making it difficult to effectively regulate the channel current with weak electric field signals. The gradient rising method of the present application successfully finds the best balance point of quantum confinement effect and field effect control (5-7 nm / 20-30 nm).

[0102] Although the sensitivity of Comparative Example 3 is acceptable (3.50), the signal-to-noise ratio is extremely low (42 dB) and the long-term stability is extremely poor (drift > 15%). This is because there is a "constraint condition" in the algorithm. When the dielectric layer is too thin (< 10 nm), serious quantum tunneling leakage occurs, resulting in a dramatic increase in leakage current noise (1 / f noise), which overwhelms the effective signal. At the same time, a small quantum dot is prone to oxidation and thermal instability. The method of the present application not only pursues sensitivity, but also ensures high signal-to-noise ratio and stability within the constraint range through iterative optimization.

[0103] Example 2 shows that even if the material is changed to BN, through the same set of processes (calculation + iteration) of the present application, excellent performance (sensitivity 4.95, better than all comparative examples) can still be obtained. This proves that the method of the present application has good universality and scalability.

[0104] The experimental data fully prove that the method of "first-principle screening and gradient ascent iterative optimization" can significantly improve the gain performance (sensitivity is improved by more than 2 times) of the nitride quantum dot electric field sensor, reduce noise, and consider the stability of the device, and solve the technical problems of "blind trial and error" and "parameter mismatch" in the prior art.

[0105] Other variations and modifications can be made to the above-described embodiments without departing from the scope of the application. The above-described embodiments are merely meant to be illustrative and not limiting.

Claims

1. A method for gain enhancement of an electric field sensor based on nitride quantum dots, characterized by, The method comprises the following steps: Step (1), obtaining the crystal structure data of graphene and nitride, constructing a heterojunction model of nitride and graphene by using first principle, and setting a vacuum layer in the Z-axis direction; performing geometric structure optimization and single-point energy calculation on the heterojunction model to obtain the band structure and state density parameters; based on the converged electronic structure calculation result, further operation is performed to obtain the quantum capacitance value of the system, and the heterojunction model with the highest quantum capacitance value is selected as the optimal heterostructure configuration; Step (2), according to the optimal heterostructure configuration selected in step (1), setting the initial quantum dot size and dielectric layer thickness as process parameters, preparing a nitride-graphene heterojunction material, and loading the material on the surface of an electrode to obtain a to-be-tested electric field sensor; the to-be-tested electric field sensor is placed in a simulated electric field environment, and the output current and voltage response data of the sensor are obtained under the preset electric field intensity and temperature conditions; the sensitivity and signal-to-noise ratio of the sensor are calculated; Step (3), constructing an optimization model with quantum dot size and dielectric layer thickness as input variables and sensitivity as target function; using the gradient ascent method, the gradient direction of the target function with respect to the input variables is calculated according to the sensitivity index obtained in step (2), and the quantum dot size correction value and dielectric layer thickness correction value of the next iteration are output; Step (4), according to the correction value output in step (3), adjusting the quantum dot size and dielectric layer thickness in step (2) to reprepare the electric field sensor, and repeating steps (2) to (3) until the change rate of the target function is less than a preset threshold, and outputting the final sensor preparation process parameters.

2. The method of claim 1, wherein, In step (1), the geometric structure optimization comprises the following process: The generalized gradient approximation (GGA) under the PBE functional was used to describe the exchange correlation potential, the ultra-soft pseudo-potential was used to treat the interaction between the core and valence electrons, the cutoff energy was set to 380 eV, and the Monkhorst-Pack method was used to generate a k-point grid with a density of 4×4×1; the convergence judgment parameters: energy convergence threshold was 1.0×10 -5 eV / atom, the interatomic force convergence threshold was 0.03 eV / Å, and the maximum internal stress was less than 0.05 GPa.

3. The method of claim 1, wherein, In step (1), the single-point energy calculation comprises the following process: The band structure is the electronic band calculated by the high-symmetry point path, which is used to directly obtain the electronic transmission parameters of the band gap width and carrier effective mass of the material; The state density includes the total state density and the partial wave state density, which is used to qualitatively and quantitatively evaluate the contribution of different atomic orbits to the electron state near the Fermi level.

4. The method of claim 1, wherein, In step (1), the formula of the quantum capacitance value is as follows: ； wherein, is the quantum capacitance, is the elementary charge, is the density of states at the Fermi level; The heterojunction model of the nitride and graphene is selected from one of aluminum nitride / graphene, boron nitride / graphene or silicon nitride / graphene.

5. The method of claim 1, wherein, In step (2), the method for preparing the nitride-graphene heterojunction material comprises a chemical synthesis method or a physical deposition method; The chemical synthesis method comprises the following process: under an inert atmosphere, the nitride precursor is mixed with the graphene dispersion liquid, the reaction temperature is 150-300 DEG C, the reaction time is 1-5 hours, the quantum dots are induced to grow in situ on the surface of the graphene, a uniform heterojunction is formed, and interface annealing treatment is performed at a temperature of 300-500 DEG C, and the nitride-graphene heterojunction material is obtained. The physical deposition method comprises the following steps: depositing the nitride material on the pretreated graphene substrate in a vacuum environment by using chemical vapor deposition or magnetron sputtering technology, and adjusting the deposition rate and the substrate temperature to ensure that the quantum dots are distributed in a single layer or a few layers, and performing interface annealing treatment at a temperature of 300-500 DEG C, so as to obtain the nitride-graphene heterojunction material.

6. The method of claim 1, wherein, In step (2), the step of loading the nitride-graphene heterojunction material on the surface of the electrode specifically comprises the following steps: The prepared nitride-graphene heterojunction material is coated on the surface of the interdigital electrode by using a spin coating or drop coating technique; and low-temperature annealing is performed in an inert atmosphere, wherein the temperature of the low-temperature annealing is 100-200 DEG C, and the time of the low-temperature annealing is 30-60 min.

7. The method of claim 1, wherein, In step (2), the simulated electric field environment is constructed by a programmable electric field generator and an environment control cabin, wherein the electric field strength is 0.1-10 kV / m, the temperature is-20-80 DEG C, and the frequency is 50-1000 Hz. The sensitivity is the ratio of the output current change amount to the electric field strength change amount.

8. The method of claim 1, wherein, In step (3), the gradient ascent method is provided with a parameter constraint condition, wherein the constraint range of the quantum dot size is 10-100 nm, and the constraint range of the dielectric layer thickness is 50-200 nm; and the preset threshold value is that the change rate of the sensitivity is less than 1%.

9. An electric field sensor, characterized by The electric field sensor is prepared by the method according to any one of claims 1-8. The electric field sensor comprises an interdigital electrode and a nitride-graphene heterojunction material loaded on the surface of the interdigital electrode, wherein the average size of the nitride quantum dots in the nitride-graphene heterojunction material is 5-7 nm, and the dielectric thickness is 20-30 nm.

10. A power system monitoring device, characterized by The device comprises the electric field sensor according to claim 9, a data acquisition unit and a signal processing unit; the device is configured to monitor the weak electric field change around the high-voltage transmission line, and the detection limit is lower than 0.01 kV / m.

Images