Method for characterizing materials used in electronic devices and semiconductor devices

By calculating the quality factor of semiconductor device materials using the forward conduction mode of unit cells and combining it with machine learning, the problems of complex and inefficient calculations in the prior art are solved, and efficient screening and evaluation of materials suitable for shrinking nodes is achieved.

CN110021385BActive Publication Date: 2024-11-26SAMSUNG ELECTRONICS CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN201811256113.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2018-05-21
Filing Date
2018-10-26
Publication Date
2024-11-26
Estimated Expiration
2038-10-26

AI Technical Summary

Technical Problem

The prior art is difficult to effectively evaluate and select appropriate materials for semiconductor device interconnect materials, especially when reducing the size nodes, where conductivity resistivity increases, traditional methods are computed and are not suitable for machine learning or big data analysis.

Method used

The forward conduction mode using unit cells only is used to calculate the material's factor of quality (FOM) and combine it with machine learning to simplify calculations and increase efficiency to screen materials suitable for use in scaled-down nodes.

Benefits of technology

Provides simpler and more efficient methods to evaluate material suitability, accurately predict conductivity differences between nanowires and bulk materials, improves computational efficiency, and is better integrated with existing machine learning algorithms for screening Suitable interconnection materials.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN110021385B_ABST
    Figure CN110021385B_ABST
Patent Text Reader

Abstract

The present disclosure relates to a method for characterizing a material used in an electronic device and a semiconductor device. A method for characterizing a material used in a semiconductor device and a semiconductor device using the material are described. The material has a unit cell and a crystal structure. The method includes determining a figure of merit (FOM) of the material using only the forward conduction mode of the unit cell. The FOM is the resistivity multiplied by the mean free path. The FOM can be used to determine the suitability of a material for use in a semiconductor device.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to methods of characterizing materials used in electronic devices and semiconductor devices. Background Art

[0002] Back-end-of-line (BEOL) interconnects connect logic devices to each other and to power and clock signals in semiconductor devices. The conductive portion of the BEOL interconnect nominally consists of highly conductive metals. For example, conventional interconnects may use materials with low bulk resistivity such as Cu, Al, W. High conductivity / low resistivity materials are utilized to meet the desired performance goals of the interconnect.

[0003] The trend in semiconductor technology is to scale to smaller sizes. Both the pitch and width of the interconnect are scaled down. Although metals have low bulk resistivity, the resistivity of the metal increases as it is scaled down to small sizes. This is called the size effect of metal resistivity. At sufficiently small sizes, the resistivity of the interconnect material increases to a resistivity that exceeds the bulk resistivity. Materials such as Cu experience a significant increase in resistance as the pitch is scaled. For example, a ten nanometer wide copper interconnect can have a resistivity that is an order of magnitude greater than the bulk resistivity. Therefore, copper interconnects cannot continue to scale up at future nodes without exceeding performance targets. Although Co and Ru have been studied for lower nodes, it is still desirable to study other materials for interconnects.

[0004] Recently, it has been shown that the product of resistivity (ρ) and mean free path (λ) can be used as a figure of merit (FOM) for resistivity scaling. The lower the figure of merit, the better the material performs when scaled to smaller sizes. The conventional method for determining the FOM of a material relies on a semiclassical description of the resistivity of bulk metals. The method involves determining the energy-momentum / band structure relationship for a dense grid of momentum points and calculating the Fermi surface of the solid from the band structure. With some simplifying assumptions, the method involves calculating:

[0005]

[0006] Where λ is the mean free path, ρ 0 is the volume resistivity, e is the electron charge, is the reduced Planck constant, v is the velocity as a function of the wave vector k, n is the index of the energy band, dS is the differential surface area, is the Fermi surface. Therefore, for larger unit cells, this method is computationally very cumbersome and may not be suitable for machine learning or data mining in large datasets. In addition, this method obtains the FOM in bulk metals and assumes that this FOM is scalable to thin films and nanowires. However, as mentioned above, the conductance degrades with confinement. Therefore, this method cannot provide usable results for narrow pitches. Therefore, an improved mechanism for determining appropriate materials to be used in future scaled nodes is desired. Summary of the invention

[0007] A method of characterizing a material for use in a semiconductor device and a semiconductor device using the material are described. The material has a unit cell and a crystal structure. The method includes determining a figure of merit (FOM) of the material using only a forward conducting mode of the unit cell (e.g., an energy band that spans the Fermi surface / moves in a direction of a canonical transport orientation described below). The FOM is the resistivity multiplied by the mean free path. The FOM can be used to determine the suitability of a material for use in a semiconductor device.

[0008] The method described here can be used to calculate the FOM of bulk materials and / or nanowires using only the forward conduction mode. FOM can be used to screen the material used as the interconnection in the node scaled down. In addition, the method providing FOM is simpler than conventional methods and can have improved runtime efficiency that is superior to conventional methods. When calculating the FOM of various materials, the structure and the obtained FOM can be used in combination with machine learning to extract a model that captures the relationship between FOM and atomic structure. Then, the model can be used to significantly speed up the prediction of the FOM of other materials. Therefore, the method can be simpler, more efficient and easier to integrate with the machine learning algorithm of the prior art in concept and practice. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 is a flow chart depicting an exemplary embodiment of a method for characterizing a material used in an electronic device.

[0010] Figure 2 is a flow chart depicting another exemplary embodiment of a method of determining a quality factor of a material used in an electronic device.

[0011] Figure 3-5 are exemplary embodiments showing materials characterized using the methods described herein.

[0012] Figure 6 Depicted is an exemplary embodiment of an electronic structure calculated using the methods described herein.

[0013] Figure 7A-7B is a graph that plots the number of forward conduction modes versus cross-sectional area.

[0014] Figure 8 An exemplary embodiment of a structure in an electronic device formed using materials selected using a FOM is depicted. DETAILED DESCRIPTION

[0015] The exemplary embodiments relate to the selection of materials for electronic devices, and in particular the selection of materials for interconnects used in semiconductor devices. The following description is given to enable one of ordinary skill in the art to implement and use the invention, and is provided in the context of a patent application and its requirements. Various modifications to the exemplary embodiments and the general principles and features described herein will be readily apparent. The exemplary embodiments are described primarily in terms of specific methods and systems provided in specific implementations. However, the methods and systems will operate effectively in other implementations.

[0016] Phrases such as "exemplary embodiment," "one embodiment," and "another embodiment" may refer to the same or different embodiments, and may also refer to multiple embodiments. The exemplary embodiments will be described in the context of a particular method having certain steps. However, the methods and systems operate effectively for other methods having different and / or additional steps and steps in a different order that do not contradict the exemplary embodiments. The embodiments will also be described with respect to systems and / or devices having certain components. However, the systems and / or devices may include more or fewer components than shown, and changes in the arrangement and type of components may be made without departing from the scope of the invention. Therefore, the present invention is not intended to be limited to the embodiments shown, but is intended to conform to the widest scope consistent with the principles and features described herein.

[0017] The use of the terms "a", "an", "the" and similar referents in the context of describing the invention (especially in the context of the appended claims) are to be construed to cover both the singular and the plural, unless otherwise indicated herein or clearly contradicted by context. The terms "comprising", "having", "including", and "containing" are to be construed as open-ended terms (i.e., meaning "including, but not limited to") unless otherwise noted.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the art to which the present invention belongs. Note that the use of any and all examples or exemplary terms provided herein is intended only to better illustrate the present invention and is not intended to limit the scope of the present invention unless otherwise indicated. In addition, unless otherwise defined, all terms defined in general dictionaries are not to be over-interpreted.

[0019] A method of characterizing a material used in a semiconductor device and a semiconductor device using the material are described. The material has a unit cell and a crystal structure. The method includes determining a figure of merit (FOM) of the material using only a forward conduction mode of the unit cell. The FOM is the resistivity multiplied by the mean free path. The FOM can be used to determine the suitability of a material for use in a semiconductor device.

[0020] Figure 1 is a flow chart depicting an exemplary embodiment of a method 100 for characterizing materials used in electronic devices. For simplicity, some steps may be omitted, may be performed in another order, and / or may be combined. The method 100 may be used for bulk computing or nanowires, where the material is confined in a direction perpendicular to the conduction direction. Thus, the method 100 may be used to screen materials for use in interconnects or other structures. The material has a crystal structure that can be represented by a unit cell including atomic types and positions.

[0021] Via step 102, the quality factor (FOM) of the material is determined using only the forward conduction mode of the unit cell. FOM is the resistivity of the material multiplied by the mean free path. As described above, resistivity increases as size decreases. The mean free path may also be different from the mean free path obtained in the bulk material. Therefore, the FOM determined in step 102 may be different from the bulk resistivity multiplied by the bulk mean free path, particularly for nanowires. Step 102 may be performed using boundary conditions for bulk or nanowires. Therefore, a FOM suitable for use may be calculated. Step 102 includes determining the electronic structure of the reference transport orientation for the unit cell, determining the Fermi level of each electronic structure, and determining the number of energy bands across the Fermi level of each electronic structure. The reference transport orientation corresponds to the direction of the forward motion of a given nanowire. Each nanowire has a different cross-sectional area (i.e., the number of unit cells in the direction transverse to the reference transport direction is different for different nanowires). Here, the cross-sectional area refers to the number of unit cells in the direction transverse to the reference transport direction. The number of energy bands across the Fermi level is the number of forward conduction modes. It has been shown previously that for lower nanowire cross-sectional areas, the band number versus cross-sectional area is linear. See, e.g., "A tight-binding model for transport in bulk metals, metal nanowires, alloys, and heterostructures I and II," Hegde et al. J. Appl. Phys. 2014. The slope k of the band number versus cross-sectional area can be used to calculate the base transport orientation FOM for each base transport orientation using the following formula:

[0022]

[0023] where λρ is the FOM for the base transport orientation. Note again that λρ can be different from λρ 0A statistical average of the FOMs of the reference transport orientations of the unit cell may then be determined. The statistical average may be considered the FOM of the material. Step 102 may be repeated for a plurality of materials.

[0024] Then, the FOM can be optionally used to determine the suitability of the material for use in a semiconductor device via step 104. Depending on the application, a FOM for a specific orientation or a FOM for a material can be used for determination. The accurate FOM of the nanowire can be determined and used to determine the material for the application of the node that is scaled down. For example, method 100 indicates that Rh, Ir, or Ni can be selected in step 104 for interconnection of smaller-sized nodes. For the FOM determined using the bulk boundary conditions in step 102, the FOM of different materials can be used to predict the conductivity difference of the bulk material. Although the FOM may not be numerically accurate for the bulk, the trend is accurate and can be used to select between materials for bulk applications. However, the FOM determined for the bulk material in step 102 should not be compared with the FOM determined for another material in the nanowire / interconnection system in step 102. In other words, the comparison between the FOMs in step 104 should be performed between materials or orientations with similar boundary conditions (e.g., both are nanowires or both are bulks).

[0025] In addition to selecting materials directly based on FOM, step 104 can also include integrating FOM with machine learning to predict the FOM of other materials. These predictions can also be used to select materials used in semiconductor devices. For example, a database of FOM (for bulk and / or nanowires) and the atomic structure of the material determined by FOM can be created. Features used for machine learning can include descriptors of crystal structure, valence electron configuration, bispectrum or radial distribution function, which are derived from the position of atoms in the unit cell and their translation and / or other structural characteristics. Then, the machine learning algorithm can be applied to map the structure to FOM. Once completed, the user can simply input the structural characteristics of the new material (they) and obtain the predicted FOM for the new material (they). The predicted FOM can be compared with other predicted and / or calculated FOMs to determine the suitability of the material for use in semiconductor devices.

[0026] Method 100 provides multiple benefits. The method provides accurate FOM for nanowires, thin films and bulk materials within a single unified framework. For bulk materials, the FOM trends of different materials can be used to predict the conductivity differences of materials. For long nanowires, FOM can be accurate. This is an advantage over conventional methods that calculate the FOM of bulk and then extend the results to nanowires. As described above, due to size effects, extending the bulk results to nanowires / interconnects with cross-sectional dimensions in the nanometer or tens of nanometer ranges may not provide accurate FOM. In addition, nanowires usually have transport orientations. Generally, materials are not truly isotropic. Therefore, restrictions in the lateral direction (perpendicular to the transport orientation) can lead to different FOMs for different transport orientations of the same material. Because the method determines the FOM based on the reference transport orientation, the FOM of the desired orientation can be determined. Method 100 is also relatively simple. Instead of using large grid points to interpolate on the Fermi surface, method 100 uses the forward conduction mode for each reference transport orientation and counts the energy bands. This is computationally easier. Furthermore, determining the FOM from the number of forward conduction modes involves linear interpolation, which is relatively simple. Due to the relative simplicity of the calculations, method 100 can provide improved run-time efficiency. The method can also be relatively easily extended to use with machine learning to improve predictions of results for additional materials.

[0027] Figure 2 1 is a flow chart depicting an exemplary embodiment of a method 120 for determining the FOM of a material that can be used in an electronic device. For simplicity, some steps may be omitted and may be performed in another order and / or combination. A material has a crystal structure that can be represented by a unit cell including the type and position of atoms. For example, a unit cell may be used for face-centered cubic (FCC), hexagonal close-packed (HCP), or other crystal structures. The unit cell may be determined from information about the material available in any one of many known databases or similar sources. Such a source may also be used to select a material whose FOM is desired to be calculated.

[0028] Via step 122, the reference transport orientation (direction) of the unit cell is determined and one of the reference transport orientations is selected. For example, FCC metals such as Cu, Au, Ag, Al, Ni, Rh, and Ir have reference transport orientations of

[100] ,

[110] ,

[111] ,

[112] . HCP materials such as Ru, Co have reference transport orientations of [00-11] and [11-20]. For A x B y For composite materials of this type, the selected reference transport orientations may be

[100] ,

[010] ,

[001] ,

[110] , and

[111] . Other orientations are possible, particularly for other materials. Figure 3An exemplary embodiment of a unit cell of a material having an FCC crystal structure is depicted.

[0029] Via step 124 , a plurality of unit cells are created along the selected reference transport orientation. Figure 4 A plurality 210 of unit cells 200 are depicted that have been connected along a reference transport orientation (the

[100] -z direction in the illustrated embodiment). Although only two unit cells 200 are shown, a significantly greater number of unit cells are typically repeated.

[0030] Via step 126, the plurality of unit cells 210 are extended into a transverse orientation. The transverse orientation is perpendicular to the reference transport orientation. Thus, along the transverse orientation(s) (e.g. Figure 3-4 The unit cell 210 is repeated to form a nanowire (x and y) of the unit cell 200 shown. The transport direction of the nanowire is along the reference transport orientation. The cross section of the nanowire is in the transverse orientation. As part of step 126, this repetition of the unit cell 210 is repeated to form nanowires with different cross-sectional areas. Thus, step 126 forms a plurality of nanowires for the reference transport orientation. Figure 5 An exemplary embodiment of a portion of one such nanowire 220 for a unit cell 200 formed using step 126 is depicted. In the illustrated embodiment, the aspect ratio of the cross section is 1:1. Thus, the nanowire 220 illustrated has a cross section of five unit cells by five unit cells. A plurality of nanowires similar to nanowire 220 are formed in step 126, with different numbers of unit cells (different cross-sectional areas) in the cross section. Although only two unit cells 200 are shown in the reference transport orientation, nanowire 220 typically has a greater number of unit cells along this direction. However, the total number of unit cells 200 for a single nanowire may be relatively small. For example, in some embodiments, the total number of unit cells 200 does not exceed one hundred. In addition, in other embodiments, nanowires having other aspect ratios and / or shapes may be used. For example, a cylindrical cross section, an elliptical cross section, or other cross section may be used instead of a rectangular cross section. The aspect ratio may also be different. For example, the aspect ratio may be 1:2, 1:3, 1:4, 3:1, and / or another value.

[0031] Via step 128, the electronic structure or band structure of each nanowire 220 is calculated for the unit cell 200 in the nanowire. This can be done using any electronic structure technique that accurately represents the electronic structure / band structure of the material under consideration. Step 128 should provide the band structure for a reference transport orientation. In other words, the energy relative to forward transverse momentum (E relative to k) of each nanowire is calculated for the reference transport orientation. z , where k z For volume calculations, for each transverse mode (k x,k y ) to obtain different band structures E(k z )relation. Figure 6 An exemplary embodiment of an electronic structure 250 calculated in step 128 is depicted. Electronic structure 250 is for illustration purposes only and is not intended to limit the electronic structures that may be determined using method 120. Furthermore, in some embodiments, steps 122, 124, and 126 may be considered part of step 128, as these steps are performed to obtain the electronic structure.

[0032] Via step 130, a Fermi level for the electronic structure and the reference transport orientation is determined for each nanowire 220. Thus, the Fermi level is indicated in the electronic structure 250. Known suitable techniques including, but not limited to, the dichotomy method may be used to determine the Fermi level.

[0033] Via step 132, the number of energy bands that cross the Fermi level of the electronic structure 250 of each nanowire 220 is calculated. Step 132 essentially involves counting the number of energy bands that cross the Fermi level. For bulk calculations, this process is performed for each different transverse mode, and the total number of modes is summed over all allowed transverse modes. The number of energy bands that cross the Fermi level is equal to the number of forward conduction modes for the baseline transport orientation of the nanowire. For each nanowire 220, this number of forward conduction modes can be stored relative to the cross-sectional area of ​​the nanowire 220. Therefore, after step 132, the number of forward conduction energy bands is obtained for each cross-sectional area (each nanowire) for the baseline transport orientation. If method 120 is used in conjunction with a thin film, a forward conduction band count can be obtained for each independent transverse mode (e.g., k x or k y )Execute steps 128, 130 and 132.

[0034] Via step 134, a linear fit is made to the number of forward conduction modes versus the cross-sectional area of ​​the nanowire 220. The slope k of the resulting line provides the ratio of the number of forward conduction modes versus the cross-sectional area for a reference transport orientation. Fig. 7A and 7B 260 and 270 are graphs that plot mode number versus cross-sectional area for a bulk and nanowire of a material, respectively. Graphs 260 and 270 are for illustrative purposes only and are not intended to limit method 120. Although the slope is different for different reference transport orientations in a nanowire, the data for each reference transport orientation can be fitted to a line. Furthermore, the slope for the bulk can be different from the slope for the nanowire.

[0035] Via step 136, the FOM of the material for the reference transport orientation is determined based on the slope. Calculating the FOM, λρ, using the slope k is relatively simple and is indicated in the equation provided above. The FOM can be used to determine the suitability of the material for an interconnect or other structure having the reference transport orientation as the direction of the drive current. However, for polycrystalline interconnects, the FOM of the material is calculated as described below.

[0036] The following steps are repeated for the remaining reference transport orientations of the unit cell, via step 138: selecting a reference transport orientation for the unit cell (122), creating the unit cell along the selected reference transport orientation (124), providing the unit cell along a transverse orientation (126), determining the electronic structure (128), determining the Fermi level (130), determining the number of energy bands that span the Fermi level (132), linear fitting (134), and determining the FOM for the reference transport orientation (136). Although in Figure 2 , but these calculations for different reference transport orientations can be performed in parallel. Thus, the FOMs for all reference transport orientations of the unit cell (also referred to herein as reference transport orientation FOMs) can be determined.

[0037] Via step 140, the FOM of the material is calculated using the statistical average of the baseline transport orientation FOM of the baseline transport orientation. The statistical average can be determined in various ways. For example, when the FOM is not greatly different for each baseline transport orientation, a simple average of the baseline transport orientation FOM for each boundary condition can be used. If the distribution between the FOMs of the baseline transport orientations is large, the median of the baseline transport orientation FOM can be used. A weighted average can also be determined in step 140. For example, an average weighted by the number of baseline transport orientations in the bulk crystal structure can be calculated in step 140. For example, for the FCC crystal structure, the

[100] orientation has six equivalent orientations and the

[110] orientation has twelve equivalent orientations. In this case, the transport orientation weighted average can be calculated as (6×

[100] FOM+12×

[110] FOM) / 18. In another embodiment, the statistical average can be an average weighted by the expected statistical distribution of the baseline transport orientation in the experimental sample of the nanowire, thin film or bulk metal. Other statistical averages can be used in other embodiments or situations. Thus, the FOM of a material may be calculated. The FOM of other material(s) may be determined using method 120. Furthermore, the FOM may be used to characterize a material for a variety of purposes including, but not limited to, those discussed herein.

[0038] In one embodiment, the following steps may be repeated for at least one additional material: selecting a reference transport orientation (122), creating a unit cell along the selected reference transport orientation (124), providing a unit cell along a transverse orientation (126), determining an electronic structure (128), determining a Fermi level (130), determining the number of energy bands that span the Fermi level (132), performing a linear fit (134), determining a FOM for the reference transport orientation (i.e., a reference transport orientation FOM) (136), repeating steps 122-136 for the remainder of the plurality of reference transport orientations (138), and determining a FOM for the material using a statistical average of the reference transport orientation FOM (140). In this embodiment, the material and the at least one additional material may be arranged in descending order of FOM, with the material having a lower FOM indicating greater conductivity.

[0039] Via step 142, the results can be optionally applied to machine learning. In one embodiment, the FOM determined in step 140 can be stored in a database, for example as a reference output. The boundary conditions (bulk / nanowire), atomic structure and other descriptors of the material are also stored. Therefore, the FOM of the candidate material can be stored as a training sample in the database relative to the boundary conditions (bulk, nanowire) and structure. The features of machine learning can be composed of structural and electronic descriptors, such as the bispectrum or radial distribution function or the valence electron number of the material. Therefore, a database of calculated FOMs, materials of FOMs and structural properties of materials can be created.

[0040] Step 142 may also include using the database to train a machine learning algorithm that nonlinearly maps inputs (structural properties) to outputs (FOM). The machine learning algorithm may be parametric or nonparametric. Examples of machine learning algorithms suitable for this purpose may include deep learning (e.g., deep convolutional neural networks), kernel ridge regression, Bayesian regression (e.g., Gaussian process regression), random forests, boosted nonlinear regression, and / or K-nearest neighbor regression. Standard practices such as test-train database splitting, cross validation, and hyperparameter estimation are performed during the training process.

[0041] As part of step 142, the FOMs of other material(s) can then be calculated from the structural and electronic configuration information using the model trained on the database. This bypasses the need to perform steps 122 to 140 for each FOM for each new material to be characterized. This can make method 120 extremely computationally efficient for new materials. This also allows unsupervised learning of the database to perform functions such as clustering, thereby grouping materials with similar target FOMs in order to find the desired structural and electronic features in candidate replacement materials. Therefore, it can be convenient to identify potential materials for use in various applications including interconnections at smaller size nodes.

[0042] The FOM determined via steps 122 to 140 or predicted in step 142 may optionally be used to determine the suitability of a material for use in a semiconductor device, via step 144. Depending on the application, a FOM for a specific orientation or a FOM for a material may be used for the determination. An accurate FOM may be determined for nanowire 220 and may be used to determine materials for use in applications at scaled-down nodes.

[0043] The FOM can be for both nanowires and bulk. The method 120 can directly calculate the FOM and / or predict the FOM using machine learning. The method can also be relatively easily extended to be used with machine learning to improve the prediction of results for additional materials.

[0044] Figure 8 A semiconductor device 300 fabricated using method 100 or 120 is depicted. Semiconductor device 300 may include a substrate / various layers 302 that have been fabricated. Device 304 to which electrical connections are desired is also shown. Metal layer 310 includes material(s) selected using method 100 and / or 120. Metal layer 310 may form interconnects at scaled-down nodes. Without selection using method 100 and / or 120, the material selected for metal layer 310 may be inappropriate, for example due to size effects. Thus, interconnect / metal layer 310 may have desired properties, and semiconductor device 300 functions as desired.

[0045] A method and system for screening materials used in electronic / semiconductor devices has been described. The method and system have been described based on the exemplary embodiments shown, and one of ordinary skill in the art will readily recognize that variations can be made to the embodiments, and any variations will be within the spirit and scope of the method and system. Therefore, many modifications may be made by one of ordinary skill in the art without departing from the spirit and scope of the appended claims.

[0046] This application claims the benefit of patent application No. 62 / 587,826 filed on November 17, 2017 and patent application No. 15 / 985,543 filed on May 21, 2018, and is incorporated herein by reference.

Claims

1. A method of characterizing a material used in an electronic device, the material having a unit cell and a crystal structure, the method include: determining a figure of merit (FOM) of the material using only the forward conduction mode of the unit cell, the figure of merit being the resistivity multiplied by the mean free path; as well as using the quality factor to determine the suitability of the material for use in the electronic device, The step of determining the quality factor comprises: selecting a plurality of base transport orientations of the unit cell; determining, for each of a plurality of nanowires, an electronic structure of the material using the forward conduction mode, each of the plurality of nanowires being based on the unit cell of the material and a reference transport orientation of the unit cell corresponding to the forward conduction mode, each of the plurality of nanowires having a cross-sectional area; for each of the plurality of nanowires, determining a Fermi level for the electronic structure and the baseline transport orientation; determining an energy band number spanning the Fermi level of the electronic structure of each of the plurality of nanowires, the energy band number being a forward conduction mode number for the reference transport orientation of each of the plurality of nanowires; linearly fitting the forward conduction mode number to a cross-sectional area of ​​a nanowire in the plurality of nanowires to provide a slope of the forward conduction mode number with respect to the cross-sectional area; and A reference transport orientation figure of merit is determined for each of the plurality of reference transport orientations of the material based on the slope.

2. The method of claim 1 , wherein the step of determining the electronic structure of each of the plurality of nanowires further comprises: include: creating a first plurality of unit cells along the reference transport orientation; providing a second plurality of unit cells along at least one lateral orientation of each nanowire, the at least one lateral orientation being perpendicular to the reference transport orientation, the second plurality of unit cells corresponding to the cross-sectional area of ​​each of the plurality of nanowires; as well as The electronic structure of each of the plurality of nanowires is determined using the first plurality of unit cells and the second plurality of unit cells.

3. The method according to claim 1, further comprising: include: repeating the steps of determining electronic structure, determining a Fermi level, determining the number of energy bands spanning the Fermi level, linear fitting, and determining a reference transport orientation figure of merit for each of the remaining portions of the plurality of reference transport orientations for the material; as well as A figure of merit for the material is determined using a statistical average of the reference transport orientation figures of merit for the plurality of reference transport orientations.

4. The method according to claim 3, further comprising: include: Repeat the following steps: Step A, selecting a reference transport orientation; Step B, determining the electronic structure; Step C, determining the Fermi level; Step D, determining the number of energy bands spanning the Fermi level; Step E, linear fitting; Step F, determining a reference transport orientation quality factor; and repeating steps AF for the rest of the plurality of reference transport orientations; and determining said figure of merit using a statistical average of said baseline transport orientation figure of merit for at least one additional material; as well as The materials and the at least one additional material are ranked in descending order of the quality factor, with materials having lower quality factors indicating greater conductivity.

5. The method according to claim 3, further comprising: include: Repeat the following steps: Step A, selecting a reference transport orientation; Step B, determining the electronic structure; Step C, determining the Fermi level; Step D, determining the number of energy bands spanning the Fermi level; Step E, linear fitting; Step F, determining a reference transport orientation quality factor; and repeating steps AF for the rest of the plurality of reference transport orientations; and determining said figure of merit using a statistical average of said baseline transport orientation figure of merit for at least one additional material; as well as The figures of merit of the material and the at least one additional material are stored in a database relative to a structure as a reference output having an input including an electronic descriptor.

6. The method according to claim 5, further comprising: include: A machine learning algorithm is trained using the database to non-linearly map the input to the reference output.

7. The method according to claim 6, further comprising: include: The machine learning algorithm is used to calculate a figure of merit for the new candidate material based on its electronic descriptor.

8. The method of claim 3, wherein the quality factor is a bulk quality factor of the material.

9. The method of claim 3, wherein the quality factor is a nanowire quality factor of the material.

10. The method of claim 1, wherein a cross-section of each of the plurality of nanowires corresponds to an aspect ratio, the plurality of nanowires sharing the aspect ratio. The method of claim 10 , wherein the aspect ratio is different from 1:

1.

12. The method of claim 1, wherein the cross section is selected from a circular cross section, an elliptical cross section, and a rectangular cross section.

13. A method of determining a figure of merit (FOM) of a material used in a semiconductor device, the material having a unit cell and a crystal structure, the method include: selecting a base transport orientation of the unit cell; creating a first plurality of unit cells along the reference transport orientation; providing a second plurality of unit cells for each of the plurality of nanowires along at least one lateral orientation, the at least one lateral orientation being perpendicular to the reference transport orientation, the second plurality of unit cells corresponding to a cross-sectional area of ​​each of the plurality of nanowires, one of the plurality of nanowires having a cross-sectional area different from a cross-sectional area of ​​another of the plurality of nanowires; determining an electronic structure of each of the plurality of nanowires using the first plurality of unit cells and the second plurality of unit cells; for each of the plurality of nanowires, determining a Fermi level for the electronic structure and the baseline transport orientation; determining an energy band number across the Fermi level of the electronic structure of each of the plurality of nanowires, the energy band number being a forward conduction mode number for the reference transport orientation of the nanowire; linearly fitting the forward conduction mode number to a cross-sectional area of ​​a nanowire in the plurality of nanowires to provide a slope of the forward conduction mode number with respect to the cross-sectional area; determining a reference transport orientation figure of merit for the material based on the slope, the reference transport orientation figure of merit being the resistivity multiplied by a mean free path for the reference transport orientation; Repeating the steps of selecting a reference transport orientation, creating the first plurality of unit cells, providing the second plurality of unit cells, determining an electronic structure, determining a Fermi level, determining a number of energy bands spanning the Fermi level, linear fitting, and determining a reference transport orientation quality factor for each of a plurality of remaining reference transport orientations of the material; as well as A figure of merit of the material is determined using the reference transport orientation and the reference transport orientation figures of merit of the plurality of remaining reference transport orientations.

14. A semiconductor device, include: substrate; as well as A plurality of interconnects having conductive portions on the substrate, the conductive portions being a material selected by: Using only the forward conduction mode of the unit cell, the material's figure of merit (FOM) is determined, which is the resistivity multiplied by the mean free path; as well as using the quality factor to determine suitability of the material for use in the semiconductor device, The step of determining the quality factor comprises: selecting a plurality of base transport orientations of the unit cell; determining, for each of a plurality of nanowires, an electronic structure of the material using the forward conduction mode, each of the plurality of nanowires being based on the unit cell of the material and a reference transport orientation of the unit cell corresponding to the forward conduction mode, each of the plurality of nanowires having a cross-sectional area; for each of the plurality of nanowires, determining a Fermi level for the electronic structure and the reference transport orientation; determining an energy band number spanning the Fermi level of the electronic structure of each of the plurality of nanowires, the energy band number being a forward conduction mode number for the reference transport orientation of each of the plurality of nanowires; linearly fitting the forward conduction mode number to a cross-sectional area of ​​a nanowire in the plurality of nanowires to provide a slope of the forward conduction mode number with respect to the cross-sectional area; A baseline transport orientation figure of merit (FOM) for the material is determined based on the slope.

15. The semiconductor device according to claim 14, wherein the step of determining the quality factor further comprises: include: repeating the steps of determining electronic structure, determining a Fermi level, determining the number of energy bands spanning the Fermi level, linear fitting, and determining a reference transport orientation figure of merit for each of the remaining portions of the plurality of reference transport orientations for the material; as well as A figure of merit for the material is determined using a statistical average of the reference transport orientation figures of merit for the plurality of reference transport orientations.

16. The semiconductor device according to claim 15, wherein determining the electronic structure of each of the plurality of nanowires further comprises: include: creating a first plurality of unit cells along the reference transport orientation; providing a second plurality of unit cells along at least one lateral orientation of each nanowire, the at least one lateral orientation being perpendicular to the reference transport orientation, the second plurality of unit cells corresponding to the cross-sectional area of ​​each of the plurality of nanowires; as well as The electronic structure of each of the plurality of nanowires is determined using the first plurality of unit cells and the second plurality of unit cells.

17. The semiconductor device according to claim 15, wherein the material is further selected by the following steps: Repeat the following steps: step A, select a reference transport orientation; step B, determine the electronic structure; step C, determine the Fermi level; step D, determine the number of energy bands across the Fermi level; step E, linear fit; step F, determine the quality factor of the reference transport orientation; repeat steps AF for the rest of the plurality of reference transport orientations ; and determining said figure of merit using a statistical average of said baseline transport orientation figure of merit for at least one additional material; as well as The materials and the at least one additional material are ranked in descending order of the quality factor, with materials having lower quality factors indicating greater conductivity.