A method for preparing an inorganic molecular crystal heterojunction

By preparing inorganic molecular crystal heterojunctions, the systematic deficiencies in the study of the interface characteristics between IMC and low-dimensional semiconductors have been addressed, enabling precise control and performance improvement of heterostructures and promoting the development of low-dimensional electronic devices.

CN122266515APending Publication Date: 2026-06-23UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610390318.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-03-27
Publication Date
2026-06-23

AI Technical Summary

Technical Problem

In the existing technology, there is a lack of systematic research on the interface characteristics between cage-like inorganic molecular crystals (IMC) and low-dimensional semiconductors. This results in unclear interface coupling mechanisms and a lack of systematic interface performance evaluation criteria, making it difficult to meet the needs of high-performance heterogeneous devices in various scenarios for precise control of interface performance and diversified material selection.

Method used

An inorganic molecular crystal heterostructure fabrication method was adopted. By constructing the unit cell of IMC and low-dimensional semiconductor, the lattice and atomic positions were optimized, static calculations and surface model construction were performed, the degree of lattice matching was tested, the heterostructure interface layer spacing was optimized, and the leakage current density under the action of electric field was calculated. The performance of the gate dielectric heterostructure was evaluated using the FN equation.

Benefits of technology

It enables precise control of heterostructures, significantly improves the performance of low-dimensional electronic devices, provides a new research approach for the development of low-dimensional electronic devices, and can efficiently evaluate the performance of gate dielectric heterostructures and precisely control the channel carrier state.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122266515A_ABST
    Figure CN122266515A_ABST
Patent Text Reader

Abstract

The application discloses a preparation method of an inorganic molecular crystal heterojunction and belongs to the field of microelectronic and optoelectronic devices, and the specific steps are as follows: I: constructing a crystal cell of an IMC and a low-dimensional semiconductor, and optimizing the crystal lattices and atom positions of the two types of crystal cells; II: performing static calculation on the two types of optimized crystal cells, and constructing a surface model according to the static calculation result; the inorganic molecular crystal heterojunction can be obtained, the performance of the gate dielectric heterojunction can be efficiently evaluated by using the F-N equation, the channel carrier state can be accurately controlled, and the performance of the low-dimensional electronic device is significantly improved, so that a brand-new research thought is provided for the development of the low-dimensional electronic device.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of microelectronics and optoelectronic devices, and more particularly to a method for preparing an inorganic molecular crystal heterostructure. Background Technology

[0002] Cage-like inorganic molecular crystals (IMCs) possess excellent bandgap-dielectric constant balance characteristics and naturally occurring van der Waals surfaces without dangling bonds, enabling them to construct high-quality heterointerfaces with low-dimensional semiconductors. Furthermore, they exhibit extremely high physical stability during dimensionality reduction. Existing research has confirmed that typical IMCs, represented by Sb₂O₃, can form ideal I-type band alignments with low-dimensional channels. This not only provides a high electron barrier, significantly improving the performance of N-type devices, but also serves as an effective protective layer, enhancing the device's environmental stability and photoelectric response speed.

[0003] Despite the enormous application potential of typical IMC systems, theoretical research on the interface properties of IMCs / low-dimensional semiconductors remains scarce. Research is overly limited to single interfaces formed by a very few known materials such as Sb₂O₃ and traditional semiconductors, and issues such as unclear interface coupling mechanisms and a lack of systematic interface performance evaluation criteria persist. These limitations and technical deficiencies have resulted in a lack of systematic understanding within the industry of the interface coupling, band alignment, and charge transfer mechanisms between different types of cage-like IMC materials and different low-dimensional semiconductors. This hinders the ability to meet the practical needs of high-performance heterostructures in various scenarios for precise control of interface performance and diverse material selection, representing a crucial step in promoting the application of IMC materials. Therefore, we propose a method for preparing inorganic molecular crystal heterostructures. Summary of the Invention

[0004] The purpose of this invention is to address the deficiencies in the existing technology by proposing a method for preparing inorganic molecular crystal heterojunctions.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A method for preparing an inorganic molecular crystal heterostructure, the specific steps of which are as follows: Ⅰ: Construct unit cells for IMC and low-dimensional semiconductors, and optimize the lattice and atomic positions of the two types of unit cells; II: Perform static calculations on the two types of optimized unit cells, and construct a surface model based on the static calculation results; III: Obtain the corresponding parameters from the static calculation results, and test the degree of lattice matching between the materials based on the surface model; IV: Construct a heterojunction interface, test and determine the stable configuration of the heterojunction interface, and then optimize the interlayer spacing of the heterojunction interface. V: Perform static self-consistent calculations on the optimized heterojunction and analyze the electronic characteristics based on the calculation results. At the same time, calculate the leakage current density of the heterojunction under the action of an electric field.

[0006] As a further aspect of the present invention, the specific steps for optimizing the lattice and atomic positions of the two types of unit cells in step I are as follows: S1.1: Based on the known inherent crystal structure characteristics of cage-like inorganic molecular crystals (IMCs) and low-dimensional semiconductors, corresponding initial unit cells were constructed. The Projected Added Wave (PAW) method was used to mathematically describe the pseudopotential interaction between valence electrons and ion cores in the unit cell system, clarifying their interaction relationship. At the same time, the exchange correlation between electrons within the unit cell was defined using the GGA-PBE functional, determining the theoretical model for this structural optimization calculation. Among them, cage-like inorganic molecular crystals (IMCs) include As2O3, Sb2O3, and Bi2O3; low-dimensional semiconductors include transition metal chalcogenides (TMDs), black phosphorus and its analogues, and one-dimensional carbon nanotubes (CNT-(16,0)). S1.2: The cutoff energy of the plane wave function basis vector is set to 500 eV, and the energy convergence accuracy threshold in the iterative calculation of the generalized gradient approximation correlation function (GGA) is set to 1.0 × 10⁻⁶ eV. -6 If, during iterative calculations, the energy change between two consecutive iterations is lower than this threshold, the energy calculation is considered to have reached convergence, and the upper limit of the force exerted on each atom is set to 0.01 eV / m². If the force values ​​of all atoms in any unit cell are lower than this threshold, then the convergence condition of atomic forces is satisfied, and then 2×0.03 / The K-point sampling density is used to perform gridding and sampling of the Brillouin zone of the unit cell; S1.3: Based on all the set theoretical methods, calculation parameters and convergence criteria, iterative optimization calculations are performed on the constructed IMC initial unit cell and the low-dimensional semiconductor initial unit cell respectively. During the calculation process, the lattice constant and the spatial position of the atoms of the two types of initial unit cells are adjusted simultaneously. Iterative optimization calculations are performed continuously until the two types of initial unit cells simultaneously meet the dual preset judgment criteria of energy convergence accuracy and atomic force convergence. Then the iteration is stopped and the unit cell structure at this time is extracted to obtain the corresponding stable unit cell structure after optimization.

[0007] As a further aspect of the present invention, the specific steps for constructing the surface model in step II are as follows: S2.1: Retrieve the set calculation parameters, set the ion step value to 1 and the broadening value to 0.05, and then perform static self-consistent calculations on the optimized IMC cell and low-dimensional semiconductor cell based on all the set calculation parameters to obtain the corresponding electronic structure and lattice parameters. S2.2: Based on the IMC unit cell data obtained by static calculation, the IMC crystal is subjected to different crystal orientations by crystal section simulation method, and multiple sets of IMC crystal surface samples with different orientations are prepared. Then, the structural stability and physical properties of each crystal surface sample are tested one by one, and stable crystal surfaces with physical properties that meet the preset standards are selected. S2.3: Based on the test results of the IMC crystal planes of each orientation, the (111) crystal orientation is determined to be the high-performance stable crystal plane of IMC. Based on the lattice parameters and atomic arrangement characteristics of the IMC (111) crystal plane, the corresponding IMC surface model is constructed. Then, the lattice constants and core lattice parameters of the IMC surface model and the low-dimensional semiconductor cell are extracted. Then, the lattice mismatch degree between IMC and different low-dimensional semiconductors is calculated by using the lattice mismatch rate calculation formula, so as to clarify the basic situation of lattice matching of various material combinations. S2.4: Based on the lattice mismatch calculation results, differentiated supercell expansion designs are performed for IMC and low-dimensional semiconductors. When targeting black phosphorus BP, a cell expansion method of IMC 2×1 and BP 5×3 is adopted; when targeting CNT-(16,0), a cell expansion method of IMC 1×2 is adopted, and CNT-(16,0) is simultaneously expanded axially by 2 times. =2c, and then, based on the set cell expansion parameters, the optimized low-dimensional semiconductor initial cell is subjected to supercell expansion processing to construct the corresponding supercell model; S2.5: Based on the cell expansion design scheme, the expanded IMC model is spatially stacked with the low-dimensional semiconductor to construct a heterogeneous interface supercell model of IMC and low-dimensional semiconductor. The actual lattice mismatch rate of the heterogeneous interface supercell model is calculated, and the cell expansion parameters are adjusted according to the calculation results to ensure that the lattice mismatch rate of the final IMC and low-dimensional semiconductor heterogeneous interface supercell model is strictly controlled within 5%.

[0008] As a further aspect of the present invention, the specific steps for testing the lattice matching degree between the materials in step III are as follows: S3.1: Obtain the stable structure of the IMC surface model and the low-dimensional semiconductor cell, while keeping all the set calculation parameters, and additionally enable the dipole correction terms in the x, y and z directions in the configured calculation parameters. Then, based on the adjusted calculation parameters, perform electronic structure calculations for each material in the stable structure of the IMC surface model and the low-dimensional semiconductor cell, and extract the vacuum energy level, work function and bandgap parameters of each material from the results after the calculation is completed. S3.2: Using the vacuum energy level as a reference, the energy level parameters of all extracted IMCs and low-dimensional semiconductors are calibrated, and the vacuum energy levels of different materials are uniformly aligned to 0 eV. Then, the lattice constants and lattice orientations of the IMC surface model and the stable structure of the low-dimensional semiconductor unit cell are extracted. Subsequently, the degree of lattice matching between different IMCs and different low-dimensional semiconductors is tested one by one through the lattice matching degree calculation method. At the same time, combined with the previous requirement of 5% lattice mismatch rate, the material combinations that meet the preset standard of lattice matching are preliminarily screened. S3.3: Based on the energy level parameters after unified alignment of vacuum energy levels, calculate the specific values ​​of conduction band offset CBO and valence band offset VBO of the candidate IMC and low-dimensional semiconductor combinations that have been initially screened by lattice matching. Then, using CBO and VBO greater than 1eV as the screening criteria, judge the CBO and VBO values ​​of all candidate heterocombinations, eliminate combinations that do not meet the criteria, and retain the IMC and low-dimensional semiconductor heterocombinations that meet the requirements.

[0009] As a further aspect of the present invention, the specific steps for optimizing the interlayer spacing of the heterogeneous interface in step IV are as follows: S4.1: Summarize the actual lattice mismatch rate, CBO, and VBO values ​​of the heterojunction supercell model. Using lattice mismatch rate ≤ 5% and CBO and VBO > 1eV as dual screening criteria, select heteromaterial combinations that simultaneously meet the dual screening criteria from the candidate combinations. Then, based on the selected heteromaterial combinations, spatially stack the IMC surface model with the low-dimensional semiconductor cell stable structure, build the initial model of the heterojunction interface according to the conventional interface contact method, and retain the set calculation parameters. S4.2: Fix the IMC surface model in the initial model of the heterojunction interface, and gradually translate the low-dimensional semiconductor model on one side along the x or y direction of the crystal coordinate system. Then set multiple sets of different translation sites to form a series of heterojunction interface test models with different site configurations. Subsequently, perform system energy calculation on all heterojunction interface test models, obtain the total system energy of the heterojunction interface under each site configuration, and organize and record all energy data. S4.3: Compare the total system energy data of all site configurations, select the site configuration with the minimum total system energy, which is the stable site configuration of the heterojunction interface. Then, take the heterojunction interface model of the stable site configuration as the calculation object, and remove the restriction of ion step = 1 in the set calculation parameters. Then, with the goal of minimizing the system energy, the interlayer spacing of the heterojunction interface under the stable site configuration is gradually adjusted and iterated. During the calculation process, the spatial position of atoms at the interface is optimized simultaneously. S4.4: When the interlayer spacing is adjusted to the point where the total energy of the system reaches the minimum value again, and the atomic forces at the interface and the system energy both meet the preset convergence criteria, stop the iterative calculation and determine the optimized heterogeneous interface interlayer spacing.

[0010] As a further aspect of the present invention, the specific steps for calculating the leakage current density of the heterojunction under the action of the electric field in step V are as follows: S5.1: Extract the optimized heterojunction interface stability configuration model and perform static self-consistent calculation on it using the removed calculation parameters to obtain the corresponding full electronic structure data. Then, perform element projection analysis on the calculated heterojunction interface electronic structure data to extract the total density of states and the partial density of states of each component element from the heterojunction interface electronic structure data, forming a density of states distribution map. S5.2: Extract the original electrostatic potential distribution data within the heterointerface lattice from the calculation results, and perform planar averaging on the data along the normal direction of the heterointerface to obtain the planar average electrostatic potential distribution curve. At the same time, keep the calculation parameters before the calculation and perform self-consistent calculations three times. The first calculation is to calculate the charge density of the overall heterointerface system, the second calculation is to calculate the charge density of the IMC component in the heterointerface separately, and the third calculation is to calculate the charge density of the low-dimensional semiconductor component in the heterointerface separately. After the calculation is completed, accurately extract the original charge density data obtained from the three calculations and record them. S5.3: According to the preset calculation formula, the three sets of extracted charge density data are subtracted to obtain the differential charge density data of the heterogeneous interface. Combining the differential charge density data of the heterogeneous interface with the planar average electrostatic potential distribution curve, the spatial distribution and intensity of charge reconstruction at the heterogeneous interface are analyzed to determine the direction and peak magnitude of charge transfer. Simultaneously, the interface electric field intensity is extracted from the electrostatic potential distribution. To clarify the distribution characteristics, strength variation patterns, and influencing factors of the built-in electric field; S5.4: Retrieve the heterojunction interface stability configuration model after electronic characteristic analysis, extract the heterojunction barrier heights CBO and VBO values ​​and dielectric constant obtained from previous analysis, and then calculate the leakage current density of the heterojunction under the action of the electric field using the FN equation. The specific form of the FN equation is as follows: In the formula, q, E, ,φ,Vg,m*,T, kB represents the elementary charge, electric field strength within the dielectric layer, Planck's constant, barrier height, applied voltage, effective carrier mass, system temperature, dielectric constant, and Boltzmann constant, respectively. The temperature T is 300K room temperature, and the internal electric field strength is E=Vg / t, where Vg is selected at 355mV with reference to the IRDS route specification.

[0011] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention constructs an initial unit cell based on the inherent structure of cage-like IMCs and low-dimensional semiconductors, uses the PAW method to describe the interaction between valence electrons and ion cores, and defines electron exchange correlation through GGA-PBE functionals. Simultaneously, it sets the cutoff energy to 500 eV and the energy convergence accuracy to 1.0 × 10⁻⁶ eV. -6 eV, the upper limit of atomic force is 0.01 eV / and 2×0.03 / The K-point sampling density was used to iteratively optimize the lattice constant and atomic spatial position of the two initial unit cells until the energy and atomic force convergence criteria were met. After obtaining the stable unit cell, the parameters were retrieved and the ion step was increased to 1 and the broadening was increased to 0.05. Then, the optimized unit cell was subjected to static self-consistent calculation to obtain the electronic structure and lattice parameters. Based on the electronic structure and lattice parameters, the IMC was subjected to different crystal orientations and performance tests. The (111) crystal orientation was determined to be a high-performance stable crystal plane and a surface model was constructed. Then, the core lattice parameters of the two were extracted and the corresponding values ​​were calculated. To determine the lattice mismatch, differentiated supercell extensions were performed for BP and CNT-(16,0) materials. A heterointerface supercell model was constructed, and parameters were adjusted to keep the lattice mismatch rate within 5%. Subsequently, the calculation parameters were maintained, and dipole correction terms in the x, y, and z directions were enabled. Electronic structure calculations were performed on each material, and parameters such as the vacuum energy level were extracted and aligned to 0 eV. The lattice matching rate was then tested at 5% mismatch rate for initial screening of combinations. Candidate combinations were calculated for CBO and VBO values, and further screening was conducted if both values ​​were greater than 1 eV. Finally, the lattice mismatch rate was used as the determining factor. For heterojunctions with a concentration ≤5% and CBO and VBO >1eV, suitable heterojunction combinations are screened. Then, an initial heterojunction model is built. The IMC surface model is fixed, and multiple sites are established by translating the low-dimensional semiconductor along the x or y direction. The total energy of each configuration system is calculated, and the stable site configuration with the lowest energy is selected. The ion step=1 constraint is removed, and the heterojunction interface interlayer spacing is iteratively optimized until convergence. The optimized model is then extracted for static self-consistent calculation to obtain full electronic structure data. The density of states distribution spectrum is obtained through elemental projection. Electrostatic potential data is extracted and averaged along the interface normal plane to obtain the distribution curve. Three self-consistent calculations are performed, and the differential charge density is obtained by subtraction. The interface charge and built-in electric field characteristics are analyzed in conjunction with the electrostatic potential curve. Finally, relevant models are retrieved, and the barrier height and dielectric constant are extracted. The leakage current density of the heterojunction under the electric field is calculated using the FN equation. This allows for the acquisition of lattice-matched heterostructures. Furthermore, the FN equation can be used to efficiently evaluate the performance of gate dielectric heterostructures and precisely control the channel carrier states, thereby significantly improving the performance of low-dimensional electronic devices and providing a new research approach for the development of low-dimensional electronic devices. Attached Figure Description

[0012] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0013] Figure 1 This is a flowchart of a method for preparing an inorganic molecular crystal heterostructure proposed in this invention. Figure 2 This is a flowchart illustrating the interface construction and electronic property analysis of a method for preparing an inorganic molecular crystal heterojunction proposed in this invention. Figure 3The diagram shows the IMC and cell structure of a low-dimensional semiconductor in the method for preparing an inorganic molecular crystal heterojunction proposed in this invention. Figure 4 This is a band edge alignment diagram of the heterojunction interface in the method for preparing an inorganic molecular crystal heterojunction proposed in this invention. Figure 5 The BP test site configuration diagram on the IMC is for the preparation method of the inorganic molecular crystal heterojunction proposed in this invention. Figure 6 The differential charge density, surface average differential charge density, and interface electrostatic potential distribution of the heterojunction are shown in the diagram of the preparation method of inorganic molecular crystal heterojunction proposed in this invention. Figure 7 This is a trend diagram of gate leakage current in the IMC and BP stacked structures of the inorganic molecular crystal heterojunction preparation method proposed in this invention. Detailed Implementation

[0014] Example 1, referring to Figures 1-7 A method for preparing an inorganic molecular crystal heterostructure, the specific steps of which are as follows: We construct unit cells for IMC and low-dimensional semiconductors, and optimize the lattice and atomic positions of the two types of unit cells.

[0015] For details, please refer to Figure 3 , Figure 4 Based on the known inherent crystal structure characteristics of cage-like inorganic molecular crystals (IMC) and low-dimensional semiconductors, corresponding initial unit cells were constructed. The Projected Added Wave (PAW) method was used to mathematically describe the pseudopotential interaction between valence electrons and ion cores within the unit cell system, clarifying their relationship. Simultaneously, the exchange correlation between electrons within the unit cell was defined using the GGA-PBE functional, determining the theoretical model for this structural optimization calculation. The cutoff energy of the plane wave function basis vector was set to 500 eV, and the energy convergence accuracy threshold during the GGA iterative calculation of the generalized gradient approximation correlation function was set to 1.0 × 10⁻⁶ eV. -6 If, during iterative calculations, the energy change between two consecutive iterations is lower than this threshold, the energy calculation is considered to have reached convergence, and the upper limit of the force exerted on each atom is set to 0.01 eV / m². If the force values ​​of all atoms in any unit cell are lower than this threshold, then the convergence condition of atomic forces is satisfied, and then 2×0.03 / The K-point sampling density is used to mesh and sample the Brillouin zone of the unit cell. Based on all the set theoretical methods, calculation parameters and convergence criteria, iterative optimization calculations are performed on the constructed IMC initial unit cell and the low-dimensional semiconductor initial unit cell respectively. During the calculation, the lattice constant and the spatial position of the atoms of the two types of initial unit cells are adjusted simultaneously. Iterative optimization calculations are performed until the two types of initial unit cells simultaneously meet the dual preset judgment criteria of energy convergence accuracy and atomic force convergence. Then the iteration is stopped and the unit cell structure at this time is extracted to obtain the corresponding stable unit cell structure after optimization.

[0016] It should be further noted that cage-like inorganic molecular crystals (IMCs) include As2O3, Sb2O3, and Bi2O3; low-dimensional semiconductors include transition metal chalcogenides (TMDs), black phosphorus and their analogues, and one-dimensional carbon nanotubes (CNT-(16,0)).

[0017] Static calculations were performed on the two types of optimized unit cells, and surface models were constructed based on the results of the static calculations.

[0018] For details, please refer to Figure 3 , Figure 4 It can be seen that by retrieving the pre-set calculation parameters, setting the ion step value to 1 and the broadening value to 0.05, and then performing static self-consistent calculations on the optimized IMC cell and low-dimensional semiconductor cell based on all the set calculation parameters, the corresponding electronic structure and lattice parameters are obtained. Based on the IMC cell data obtained by static calculation, the IMC crystal is subjected to different crystal orientations through the crystal section simulation method, and multiple sets of IMC crystal surface samples with different orientations are prepared. The structural stability and physical properties of each crystal surface sample are then tested one by one, and stable crystal surfaces that meet the preset standards are selected. Based on the test results of each orientation of the IMC crystal surface, the (111) direction is determined to be the high-performance stable crystal surface of the IMC. Based on the lattice parameters and atomic arrangement characteristics of the IMC(111) crystal plane, a corresponding IMC surface model is constructed. Then, the lattice constants and core lattice parameters of the IMC surface model and the low-dimensional semiconductor cell are extracted. Using the lattice mismatch rate calculation formula, the lattice mismatch degree between the IMC and different low-dimensional semiconductors is calculated to clarify the basic lattice matching conditions of various material combinations. Based on the lattice mismatch degree calculation results, differentiated supercell expansion designs are implemented for the IMC and low-dimensional semiconductors. For black phosphorus BP, an IMC 2×1 and BP 5×3 expansion method is used; for CNT-(16,0), an IMC 1×2 expansion method is used, and CNT-(16,0) undergoes axial 2x expansion. =2c), and then, based on the set cell expansion parameters, the optimized low-dimensional semiconductor initial cell is subjected to supercell expansion processing to construct the corresponding supercell model. Based on the cell expansion design scheme, the expanded IMC model and the low-dimensional semiconductor are spatially stacked to construct a heterogeneous interface supercell model of IMC and low-dimensional semiconductor. The actual lattice mismatch rate of the heterogeneous interface supercell model is calculated, and the cell expansion parameters are adjusted according to the calculation results to ensure that the lattice mismatch rate of the final IMC and low-dimensional semiconductor heterogeneous interface supercell model is strictly controlled within the range of 5%.

[0019] The relevant parameters are obtained from the static calculation results, and the degree of lattice matching between the materials is tested based on the surface model.

[0020] For details, please refer to Figure 5 It is known that to obtain the stable structure of the IMC surface model and the low-dimensional semiconductor cell, while maintaining all the set calculation parameters, and additionally enabling dipole correction terms in the x, y, and z directions among the configured calculation parameters, electronic structure calculations are performed on each material in the stable structure of the IMC surface model and the low-dimensional semiconductor cell based on the adjusted calculation parameters. After the calculation is completed, the vacuum energy level, work function, and bandgap parameters of each material are extracted from the results. Using the vacuum energy level as a reference, the energy level parameters of all extracted IMC and low-dimensional semiconductors are calibrated, and the vacuum energy levels of different materials are uniformly aligned to 0 eV. Finally, the stable structure of the IMC surface model and the low-dimensional semiconductor cell is extracted. The lattice constant and lattice orientation were determined, and then the degree of lattice matching between different IMCs and different low-dimensional semiconductors was tested one by one using the lattice matching degree calculation method. At the same time, combined with the previous requirement of 5% lattice mismatch rate, material combinations that meet the preset standard of lattice matching were initially screened. Based on the energy level parameters after the vacuum energy level was uniformly aligned, the specific values ​​of conduction band offset CBO and valence band offset VBO of the candidate IMC and low-dimensional semiconductor combinations that were initially screened by lattice matching were calculated. Then, with CBO and VBO greater than 1eV as the screening standard, the CBO and VBO values ​​of all candidate heterocombinations were judged, and combinations that did not meet the standard were eliminated, while the IMC and low-dimensional semiconductor heterocombinations that met the requirements were retained.

[0021] It should be further explained that different values ​​of the effective tunneling mass of charge carriers were assumed, and the leakage current at the IMC / MoSe2 interface was calculated under these conditions. To simultaneously consider EOT performance and leakage current isolation performance, the EOT was preset to 1nm in this calculation. The results show that although As2O3 can form the highest barrier with the channel semiconductor, its dielectric constant is low, and the leakage current is generally high under the assumption of 1nm EOT. Only when the effective tunneling mass of charge carriers is greater than 1.5m0 can the leakage current standard of the logic cell be met. Sb2O3 and Bi2O3 have higher dielectric constants and larger physical thicknesses under the same EOT, which can effectively isolate charge carrier tunneling. Among them, Bi2O3 has better EOT performance, making it almost unaffected by the size of the effective tunneling mass of charge carriers. Even under relatively low barrier conditions, the leakage current can be controlled to <10. -7 A / cm 2 This level of performance confirms that inorganic molecular crystals (IMCs) can construct high-quality van der Waals heterointerfaces with good physical compatibility and extremely low interface state density with low-dimensional semiconductors. Among them, Sb2O3 and Bi2O3 achieve excellent leakage current blocking capabilities, while Bi2O3 also shows the application prospects of a new type of gate dielectric with ultra-low power consumption and high gate control capability due to its highly advantageous electrostatic control potential. The IMC / MoSe2 stacked structure identified in this invention has good performance and is expected to become a design solution for high-performance electronic device structures.

[0022] Example 2, refer to Figures 1-7 A method for preparing an inorganic molecular crystal heterostructure, the specific steps of which are as follows: A heterojunction interface is constructed, and the stable configuration of the heterojunction interface is tested and determined. Then, the interlayer spacing of the heterojunction interface is optimized.

[0023] For details, please refer to Figure 6 , Figure 7It can be seen that by summarizing the actual lattice mismatch rate, CBO, and VBO values ​​of the heterojunction interface supercell model, and using a lattice mismatch rate ≤5% and CBO and VBO >1eV as dual screening criteria, heteromaterial combinations that simultaneously meet the dual screening criteria are selected from the candidate combinations. Then, based on the selected heteromaterial combinations, the IMC surface model and the stable low-dimensional semiconductor cell structure are spatially stacked. An initial model of the heterojunction interface is built according to conventional interface contact methods, while retaining the pre-set calculation parameters. The IMC surface model in the initial heterojunction interface model is fixed, and the low-dimensional semiconductor model on one side is gradually translated along the x or y direction of the crystal coordinate system. Multiple sets of different translation sites are then set to form a series of heterojunction interface test models with different site configurations. Subsequently, all heterojunction interface test models are tested. The system energy is calculated to obtain the total system energy of the heterojunction interface under each site configuration. All energy data are organized and recorded. By comparing the total system energy data of all site configurations, the site configuration with the minimum total system energy is selected. This configuration is the stable site configuration of the heterojunction interface. Then, the heterojunction interface model of the stable site configuration is used as the calculation object. At the same time, the restriction of ion step=1 in the pre-set calculation parameters is removed. Then, with the goal of minimizing the system energy, the interlayer spacing of the heterojunction interface under the stable site configuration is gradually adjusted and iteratively calculated. During the calculation process, the spatial position of atoms at the interface is optimized simultaneously. When the interlayer spacing is adjusted to the point that the total system energy reaches the minimum value again, and the forces on the atoms at the interface and the system energy both meet the preset convergence criteria, the iterative calculation is stopped, and the optimized heterojunction interface interlayer spacing is determined.

[0024] Static self-consistent calculations were performed on the optimized heterojunction, and the electronic characteristics were analyzed based on the calculation results. At the same time, the leakage current density of the heterojunction under the action of an electric field was calculated.

[0025] For details, please refer to Figure 6 , Figure 7It can be seen that the optimized heterojunction interface stability configuration model is extracted, and static self-consistent calculation is performed on it using the unresolved calculation parameters to obtain the corresponding full electronic structure data. Then, elemental projection analysis is performed on the calculated heterojunction interface electronic structure data to extract the total density of states and the partial density of states of each component element, forming a density of states distribution spectrum. The original distribution data of electrostatic potential in the heterojunction interface lattice is extracted from the calculation results, and the data is averaged along the normal direction of the heterojunction interface to obtain the plane-averaged electrostatic potential distribution curve. At the same time, the calculation parameters before unresolved calculation are maintained, and self-consistent calculation is performed three times. The charge density of the overall heterojunction system was calculated. The charge density of the IMC component within the heterojunction was calculated separately in the second calculation, and the charge density of the low-dimensional semiconductor component within the heterojunction was calculated separately in the third calculation. After the calculations, the raw charge density data from the three calculations were accurately extracted and recorded. Following a pre-defined calculation formula, the three sets of extracted charge density data were subtracted to obtain the differential charge density data of the heterojunction. Combining the differential charge density data of the heterojunction with the planar average electrostatic potential distribution curve, the spatial distribution and intensity of charge reconstruction at the heterojunction were analyzed to determine the direction and peak magnitude of charge transfer. Simultaneously, the interfacial electric field intensity was extracted from the electrostatic potential distribution. The distribution characteristics, strength variation patterns, and influencing factors of the built-in electric field were clarified. The heterojunction interface stability configuration model obtained after electronic characteristic analysis was retrieved, and the heterojunction barrier height (CBO and VBO values) and dielectric constant obtained from previous analysis were extracted. Then, the leakage current density of the heterojunction under the action of the electric field was calculated using the FN equation. It should be further explained that the specific form of the FN equation is as follows: In the formula, q, E, ,φ,Vg,m*,T, kB represents the elementary charge, electric field strength within the dielectric layer, Planck's constant, barrier height, applied voltage, effective carrier mass, system temperature, dielectric constant, and Boltzmann constant, respectively. The temperature T is 300K room temperature, and the internal electric field strength is E=Vg / t, where Vg is selected at 355mV with reference to the IRDS route specification.

Claims

1. A method for preparing an inorganic molecular crystal heterostructure, characterized in that, The specific steps of this preparation method are as follows: Ⅰ: Construct unit cells for IMC and low-dimensional semiconductors, and optimize the lattice and atomic positions of the two types of unit cells; II: Perform static calculations on the two types of optimized unit cells, and construct a surface model based on the static calculation results; III: Obtain the corresponding parameters from the static calculation results, and test the degree of lattice matching between the materials based on the surface model; IV: Construct a heterojunction interface, test and determine the stable configuration of the heterojunction interface, and then optimize the interlayer spacing of the heterojunction interface. V: Perform static self-consistent calculations on the optimized heterojunction and analyze the electronic characteristics based on the calculation results. At the same time, calculate the leakage current density of the heterojunction under the action of an electric field.

2. The method for preparing an inorganic molecular crystal heterostructure according to claim 1, characterized in that, The specific steps for optimizing the lattice and atomic positions of the two types of unit cells described in Step I are as follows: S1.1: Based on the known inherent crystal structure characteristics of cage-like inorganic molecular crystals (IMCs) and low-dimensional semiconductors, corresponding initial unit cells were constructed. The Projected Added Wave (PAW) method was used to mathematically describe the pseudopotential interaction between valence electrons and ion cores in the unit cell system, clarifying their interaction relationship. At the same time, the exchange correlation between electrons within the unit cell was defined using the GGA-PBE functional, determining the theoretical model for this structural optimization calculation. Among them, cage-like inorganic molecular crystals (IMCs) include As2O3, Sb2O3, and Bi2O3; low-dimensional semiconductors include transition metal chalcogenides (TMDs), black phosphorus and its analogues, and one-dimensional carbon nanotubes (CNT-(16,0)). S1.2: The cutoff energy of the plane wave function basis vector is set to 500 eV, and the energy convergence accuracy threshold in the iterative calculation of the generalized gradient approximation correlation function (GGA) is set to 1.0 × 10⁻⁶ eV. -6 If, during iterative calculations, the energy change between two consecutive iterations is lower than this threshold, the energy calculation is considered to have reached convergence, and the upper limit of the force exerted on each atom is set to 0.01 eV / m². If the force values ​​of all atoms in any unit cell are lower than this threshold, then the convergence condition of atomic forces is satisfied, and then 2×0.03 / The K-point sampling density is used to perform gridding and sampling of the Brillouin zone of the unit cell; S1.3: Based on all the set theoretical methods, calculation parameters and convergence criteria, iterative optimization calculations are performed on the constructed IMC initial unit cell and the low-dimensional semiconductor initial unit cell respectively. During the calculation process, the lattice constant and the spatial position of the atoms of the two types of initial unit cells are adjusted simultaneously. Iterative optimization calculations are performed continuously until the two types of initial unit cells simultaneously meet the dual preset judgment criteria of energy convergence accuracy and atomic force convergence. Then the iteration is stopped and the unit cell structure at this time is extracted to obtain the corresponding stable unit cell structure after optimization.

3. The method for preparing an inorganic molecular crystal heterostructure according to claim 2, characterized in that, The specific steps for constructing the surface model described in step II are as follows: S2.1: Retrieve the set calculation parameters, set the ion step value to 1 and the broadening value to 0.05, and then perform static self-consistent calculations on the optimized IMC cell and low-dimensional semiconductor cell based on all the set calculation parameters to obtain the corresponding electronic structure and lattice parameters. S2.2: Based on the IMC unit cell data obtained by static calculation, the IMC crystal is subjected to different crystal orientations by crystal section simulation method, and multiple sets of IMC crystal surface samples with different orientations are prepared. Then, the structural stability and physical properties of each crystal surface sample are tested one by one, and stable crystal surfaces with physical properties that meet the preset standards are selected. S2.3: Based on the test results of the IMC crystal planes of each orientation, the (111) crystal orientation is determined to be the high-performance stable crystal plane of IMC. Based on the lattice parameters and atomic arrangement characteristics of the IMC (111) crystal plane, the corresponding IMC surface model is constructed. Then, the lattice constants and core lattice parameters of the IMC surface model and the low-dimensional semiconductor cell are extracted. Then, the lattice mismatch degree between IMC and different low-dimensional semiconductors is calculated by using the lattice mismatch rate calculation formula, so as to clarify the basic situation of lattice matching of various material combinations. S2.4: Based on the lattice mismatch calculation results, differentiated supercell expansion designs are performed for IMC and low-dimensional semiconductors. When targeting black phosphorus BP, a cell expansion method of IMC 2×1 and BP 5×3 is adopted; when targeting CNT-(16,0), a cell expansion method of IMC 1×2 is adopted, and CNT-(16,0) is simultaneously expanded axially by 2 times. =2c, and then, based on the set cell expansion parameters, the optimized low-dimensional semiconductor initial cell is subjected to supercell expansion processing to construct the corresponding supercell model; S2.5: Based on the cell expansion design scheme, the expanded IMC model is spatially stacked with the low-dimensional semiconductor to construct a heterogeneous interface supercell model of IMC and low-dimensional semiconductor. The actual lattice mismatch rate of the heterogeneous interface supercell model is calculated, and the cell expansion parameters are adjusted according to the calculation results to ensure that the lattice mismatch rate of the final IMC and low-dimensional semiconductor heterogeneous interface supercell model is strictly controlled within 5%.

4. The method for preparing an inorganic molecular crystal heterojunction according to claim 3, characterized in that, The specific steps for testing the lattice matching degree between the materials in step III are as follows: S3.1: Obtain the stable structure of the IMC surface model and the low-dimensional semiconductor cell, while keeping all the set calculation parameters, and additionally enable the dipole correction terms in the x, y and z directions in the configured calculation parameters. Then, based on the adjusted calculation parameters, perform electronic structure calculations for each material in the stable structure of the IMC surface model and the low-dimensional semiconductor cell, and extract the vacuum energy level, work function and bandgap parameters of each material from the results after the calculation is completed. S3.2: Using the vacuum energy level as a reference, the energy level parameters of all extracted IMCs and low-dimensional semiconductors are calibrated, and the vacuum energy levels of different materials are uniformly aligned to 0 eV. Then, the lattice constants and lattice orientations of the IMC surface model and the stable structure of the low-dimensional semiconductor unit cell are extracted. Subsequently, the degree of lattice matching between different IMCs and different low-dimensional semiconductors is tested one by one through the lattice matching degree calculation method. At the same time, combined with the previous requirement of 5% lattice mismatch rate, the material combinations that meet the preset standard of lattice matching are preliminarily screened. S3.3: Based on the energy level parameters after unified alignment of vacuum energy levels, calculate the specific values ​​of conduction band offset CBO and valence band offset VBO of the candidate IMC and low-dimensional semiconductor combinations that have been initially screened by lattice matching. Then, using CBO and VBO greater than 1eV as the screening criteria, judge the CBO and VBO values ​​of all candidate heterocombinations, eliminate combinations that do not meet the criteria, and retain the IMC and low-dimensional semiconductor heterocombinations that meet the requirements.

5. The method for preparing an inorganic molecular crystal heterostructure according to claim 4, characterized in that, The specific steps for optimizing the interlayer spacing at the heterogeneous interface described in step IV are as follows: S4.1: Summarize the actual lattice mismatch rate, CBO, and VBO values ​​of the heterojunction supercell model. Using lattice mismatch rate ≤ 5% and CBO and VBO > 1eV as dual screening criteria, select heteromaterial combinations that simultaneously meet the dual screening criteria from the candidate combinations. Then, based on the selected heteromaterial combinations, spatially stack the IMC surface model with the low-dimensional semiconductor cell stable structure, build the initial model of the heterojunction interface according to the conventional interface contact method, and retain the set calculation parameters. S4.2: Fix the IMC surface model in the initial model of the heterojunction interface, and gradually translate the low-dimensional semiconductor model on one side along the x or y direction of the crystal coordinate system. Then set multiple sets of different translation sites to form a series of heterojunction interface test models with different site configurations. Subsequently, perform system energy calculation on all heterojunction interface test models, obtain the total system energy of the heterojunction interface under each site configuration, and organize and record all energy data. S4.3: Compare the total system energy data of all site configurations, select the site configuration with the minimum total system energy, which is the stable site configuration of the heterojunction interface. Then, take the heterojunction interface model of the stable site configuration as the calculation object, and remove the restriction of ion step = 1 in the set calculation parameters. Then, with the goal of minimizing the system energy, the interlayer spacing of the heterojunction interface under the stable site configuration is gradually adjusted and iterated. During the calculation process, the spatial position of atoms at the interface is optimized simultaneously. S4.4: When the interlayer spacing is adjusted to the point where the total energy of the system reaches the minimum value again, and the atomic forces at the interface and the system energy both meet the preset convergence criteria, stop the iterative calculation and determine the optimized heterogeneous interface interlayer spacing.

6. The method for preparing an inorganic molecular crystal heterojunction according to claim 5, characterized in that, The specific steps for calculating the leakage current density of the heterojunction under the influence of the electric field, as described in step V, are as follows: S5.1: Extract the optimized heterojunction interface stability configuration model and perform static self-consistent calculation on it using the removed calculation parameters to obtain the corresponding full electronic structure data. Then, perform element projection analysis on the calculated heterojunction interface electronic structure data to extract the total density of states and the partial density of states of each component element from the heterojunction interface electronic structure data, forming a density of states distribution map. S5.2: Extract the original electrostatic potential distribution data within the heterointerface lattice from the calculation results, and perform planar averaging on the data along the normal direction of the heterointerface to obtain the planar average electrostatic potential distribution curve. At the same time, keep the calculation parameters before the calculation and perform self-consistent calculations three times. The first calculation is to calculate the charge density of the overall heterointerface system, the second calculation is to calculate the charge density of the IMC component in the heterointerface separately, and the third calculation is to calculate the charge density of the low-dimensional semiconductor component in the heterointerface separately. After the calculation is completed, accurately extract the original charge density data obtained from the three calculations and record them. S5.3: According to the preset calculation formula, the three sets of extracted charge density data are subtracted to obtain the differential charge density data of the heterogeneous interface. Combining the differential charge density data of the heterogeneous interface with the planar average electrostatic potential distribution curve, the spatial distribution and intensity of charge reconstruction at the heterogeneous interface are analyzed to determine the direction and peak magnitude of charge transfer. Simultaneously, the interface electric field intensity is extracted from the electrostatic potential distribution. To clarify the distribution characteristics, strength variation patterns, and influencing factors of the built-in electric field; S5.4: Retrieve the heterojunction interface stability configuration model after electronic characteristic analysis, extract the heterojunction barrier heights CBO and VBO values ​​and dielectric constant obtained from previous analysis, and then calculate the leakage current density of the heterojunction under the action of the electric field using the FN equation. The specific form of the FN equation is as follows: In the formula, q, E, ,φ,Vg,m*,T, kB represents the elementary charge, electric field strength within the dielectric layer, Planck's constant, barrier height, applied voltage, effective carrier mass, system temperature, dielectric constant, and Boltzmann constant, respectively. The temperature T is 300K room temperature, and the internal electric field strength is E=Vg / t, where Vg is selected at 355mV with reference to the IRDS route specification.