Molecular energy calculation method

Abstract

Provided is a molecular energy calculation method for calculating, with high accuracy, the energy of the electronic state of a molecule of practical size.　The present invention comprises: an initial calculation step in which a natural orbital of a molecule is calculated using a perturbation method that takes into account, as a perturbation term, the deviation of the electronic state of the molecule approximated using a quantum chemical calculation method from the true electronic state; a basis transformation step in which a subsystem that governs the electronic state of the molecule is extracted on the basis of the calculation results of the initial calculation step; a circuit construction step in which a quantum circuit that is a calculation circuit for a quantum computer is constructed on the basis of the selection results of the basis transformation step; and a quantum computer calculation step in which the energy value of the electronic state of the molecule is calculated by operating the quantum circuit constructed in the circuit construction step.

Description

How to calculate molecular energy

[0001] The present invention relates to a method for calculating molecular energy.

[0002] In recent years, quantum chemical calculations have been widely used to analyze and predict molecular structures, electronic states, and reaction mechanisms, and the analytical and predictive results are being applied to a variety of fields. For example, quantum chemical calculations are being used as a theoretical basis for the development of new molecules, and applications such as calculating the properties of new molecules are being recognized.

[0003] When calculating material properties, reliable calculations including electronic correlation are possible by using the Full-Configuration Interaction method, which includes all possible mixing between electronic configurations. However, since all mixing coefficients between different configurations must be recorded, the required computational resources increase exponentially with the size of the molecule. Quantum chemical calculations become difficult when the target molecule is large, and methods for achieving accurate quantum chemical calculations for large molecules are being researched.

[0004] Specifically, Patent Document 1 proposes a calculation method for quantum chemistry calculations that can be performed even on a quantum computer, in which the number of times the overlap integral value is calculated is reduced by selecting molecular orbitals to be calculated using an active space.

[0005] International Publication No. 2022-97298

[0006] In the quantum chemical calculations of Patent Document 1, molecular orbitals using an active space are selected to reduce the number of calculations. However, if molecular orbitals are selected as they are, there is a possibility that calculations will not be performed with sufficient accuracy. Even if an electron configuration that contributes greatly to the effect of electron correlation is selected based on the overlap integral value of a molecular orbital pair, if the accuracy of the molecular orbitals before selection is insufficient, the accuracy of the calculation results obtained by reducing the number of calculations will also decrease.

[0007] An object of the present invention is to provide a molecular energy calculation method for calculating the energy of electronic states in molecules of practical size with high accuracy in a quantum computer.

[0008] The molecular energy calculation method of the present invention includes an initial calculation step of calculating the natural orbitals of a molecule using a perturbation method in which the deviation from the true electronic state is taken into account as a perturbation term for the electronic state of the molecule approximated using a quantum chemical calculation method that approximates the electron-electron interactions of the target molecule with a mean field potential; a basis conversion step of extracting a subsystem that governs the electronic state of the molecule based on the calculation result of the initial calculation step; a circuit construction step of constructing a quantum circuit, which is an arithmetic circuit for a quantum computer, based on the selection result of the basis conversion step; and a quantum computer calculation step of operating the quantum circuit constructed in the circuit construction step to calculate the energy value of the electronic state of the molecule.

[0009] In the basis transformation step, it is preferable to extract a subsystem by utilizing an active space using natural orbitals.

[0010] In the basis transformation step, it is preferable to use the active space to extract subsystems to a number less than a predetermined specific number.

[0011] The quantum computer calculation step preferably involves operating the quantum circuit using a quantum simulator to calculate the energy value.

[0012] The quantum simulator is preferably implemented on a general-purpose graphics computing device.

[0013] The quantum chemical calculation method for approximating the electron interactions of the target molecule with a mean field potential is preferably the Hartree-Fock method.

[0014] The perturbation method in which the deviation from the true electronic state is taken into consideration as a perturbation term is preferably the Moller-Plesset method.

[0015] Preferably, the molecules include benzene, chlorobenzene, and nitrobenzene.

[0016] According to the present invention, a molecular energy calculation method can be realized in a quantum computer, which calculates the energy of the electronic state of a molecule of practical size with high accuracy.

[0017] Fig. 1 is a block diagram of an arithmetic device for performing quantum chemical calculations. Fig. 2 is an explanatory diagram of a quantum circuit used for quantum chemical calculations. Fig. 3 is an explanatory diagram of transition absorption wavelengths obtained as calculation results in Examples 1-3. Fig. 4 is a flowchart showing a series of steps for calculating the energy of the electronic state of a molecule in the present invention.

[0018] The energy calculation method of the present invention calculates the energy of an electronic state of a molecule using an algorithm for performing quantum chemical calculations. The energy calculation method includes an initial calculation step, a basis transformation step, a circuit construction step, and a quantum computer calculation step, and each step is performed by a computing device. The computing device is, for example, a general-purpose graphics computing device equipped with a GPGPU (General-Purpose Computing on Graphics Processing Units), which is a GPU (Graphical Processing Unit) that can also perform general-purpose data processing other than image processing.

[0019] The arithmetic unit has programs for each process stored in a program memory (not shown). A central control unit, which is made up of a processor, executes the programs in the program memory to realize the function of performing calculations in each step.

[0020] As shown in FIG. 1 , the computing device 10 includes a quantum chemistry calculation unit 11 that performs quantum chemistry calculations. The quantum chemistry calculation unit 11 includes an initial calculation unit 12 that performs an initial calculation step, a basis conversion unit 13 that performs a basis conversion step, a circuit construction unit 14 that performs a circuit construction step, a quantum computer calculation unit 15 that performs a quantum computer calculation step, an output control unit 16 that controls the execution content of each step and the output of the calculation results, and a memory unit 17 that stores information such as molecular structure information. The quantum chemistry calculation unit 11 can perform each of the above steps by executing a quantum chemistry calculation program. The computing device 10 also includes an input receiving unit (not shown) that is electrically connected to a user interface and receives user operations. The user interface is any device that can be operated by a user, such as a touch panel, switch buttons, a mouse, or a keyboard.

[0021] In the initial calculation step executed by the initial calculation unit 12, atomic orbitals are acquired based on the specification of the type and coordinates of atoms in the molecule to be calculated and the specification of the calculation level. The coordinates are expressed as x-axis coordinates, y-axis coordinates, and z-axis coordinates. The calculation level uses the STO-nG basis set or split valence basis set. Quantum calculations are performed on the acquired atomic orbitals using Hartree-Fock calculations that perform approximation, and the molecular orbitals (MOs) of the ground state molecule are acquired by solving the Schrödinger equation.

[0022] Furthermore, a natural orbital (NO), which is an orbital including the influence of an excited state, is obtained by calculation using the second-order Moller-Plesset method for the molecular orbital, which is the electronic state of the molecule approximated using the Hartree-Fock calculation, and is then substituted into the molecular orbital.

[0023] Both molecular orbitals and natural orbitals are orbitals formed by linear combinations of the orbitals of each atom that make up a molecule, but molecular orbitals are optimized for energy, while natural orbitals are optimized for occupancy. Natural orbitals contain higher energy states than molecular orbitals, so when performing high-precision quantum chemistry calculations using the same number of orbitals, using natural orbitals allows for more accurate energy evaluation than molecular orbitals.

[0024] In the basis conversion step executed by the basis conversion unit 13, orbitals that are a subsystem governing the electronic state of the molecule are extracted from the molecular orbitals obtained in the initial calculation step using a complete active space (CAS) using natural orbitals, and a Hamiltonian to be used for constructing a quantum circuit is derived. In quantum chemical calculations, the larger the molecule, the greater the required computational resources. Therefore, if the number of orbitals obtained in the initial calculation step is equal to or greater than a predetermined specific number, orbitals to be used in subsequent calculations are extracted and a Hamiltonian is constructed. On the other hand, if the number of orbitals is less than a specific number, such as in hydrogen molecules, a Hamiltonian may be constructed from all orbitals obtained in the initial calculation step. In other words, the number of orbitals obtained in the initial calculation step is extracted to a number less than a predetermined specific number.

[0025] The specific number is determined based on the number of molecules or the computing capacity of the computing device 10 relative to the number of types of atoms constituting the molecules, for example, to prevent computational resources such as computation time from becoming excessively large.

[0026] The complete active space is a method used to identify a subsystem (electronic configuration) that governs the electronic state of a molecule. In this embodiment, extraction is performed based on the highest occupied natural orbital (HONO), which is the orbital with the highest energy among the natural orbitals occupied by electrons, and the lowest unoccupied natural orbital (LUNO), which is the orbital with the lowest energy among the natural orbitals not occupied by electrons. HONO is the highest occupied molecular orbital (HOMO), and LUNO is the lowest unoccupied molecular orbital (LUMO). Orbitals are extracted based on the energy of each natural orbital compared to HONO and LUNO, symmetry with respect to the molecule, and the like. The number of orbitals to be extracted is set to less than a specific number.

[0027] The electrons occupying each orbital extracted from the complete active space are used in subsequent calculations, while the interactions between other electrons are averaged and the degrees of freedom are frozen. Since the static interaction energy of electrons that are frozen is the only constant to be considered, subsequent calculations are omitted and they are not converted into quantum bits during the circuit construction process. This significantly reduces the amount of calculations.

[0028] In the circuit construction step executed by the circuit construction unit 14, a quantum circuit for performing calculations in the quantum computer calculation step is constructed. A quantum circuit is an arithmetic circuit for a quantum computer. A Jordan-Wigner transformation is performed on the integral of one electron or two electrons obtained from orbitals less than a specific number obtained in the basis conversion step, and a quantum gate corresponding to the integral of the electrons is obtained. The transformation is performed using a Hamiltonian using a Pauli gate, which corresponds to the quantum gate. One spin orbital is associated with one quantum bit by the Jordan-Wigner transformation. If the orbital is occupied, a different quantum bit is assigned, and if not, an auxiliary quantum bit is assigned. The overall state of the system is obtained as the product of these quantum bits.

[0029] In the Jordan-Wigner transformation, for example, in the Hartree-Fock state of six electrons with six atomic orbitals (12 spin orbitals), such as benzene, the six lowest orbitals are occupied and the other six orbitals are unoccupied, and this state is converted into a quantum bit representation of |111111000000>, which is the product of six |1> and six |0>. In addition, in the transformation of quantum computer operations, a unitary gate is constructed.

[0030] The quantum phase estimation algorithm (QPEA) is a computational technique for quantum phase estimation. Of the multiple derivatives of the quantum phase estimation algorithm, the iterative quantum phase estimation algorithm (IQPEA), which can be performed with only one auxiliary quantum bit, can be selected to perform the quantum phase estimation computation with the smallest number of quantum bits required.

[0031] As shown in FIG. 2 , this embodiment uses IQPEA in the quantum phase estimation calculation, and constructs a quantum circuit 20 to be used in the calculation based on the obtained quantum gate. The quantum circuit 20 performs calculations using an iterative quantum phase estimation algorithm. The quantum circuit 20 includes an auxiliary quantum bit 21, a first quantum bit 22, a second quantum bit 23, a Hadamard gate 31, a first rotation gate 32, a second rotation gate 33, a Hadamard gate 34, a measurement 35, and a unitary gate 36. The auxiliary quantum bit 21 is the first quantum bit represented by |0>, and the quantum bits other than the auxiliary quantum bit 21 are provided in order, with the first quantum bit being the first quantum bit 22 and the second quantum bit being the second quantum bit 23. Depending on the result of the Jordan-Wigner transformation, quantum bits are provided up to the nth quantum bit (n is a natural number), which is the nth quantum bit. The first rotation gate 32 and the second rotation gate 33 are rotation gates in the z direction.

[0032] In the quantum computer calculation step executed by the quantum computer calculation unit 15, the quantum circuit 20 constructed in the circuit construction step is operated to perform calculations of the quantum phase estimation method and calculate the energy value of the molecule's electronic state. The energy value is the energy difference ΔE between the ground state energy value E0 and the excited state energy value E1. In other words, it is calculated using the formula E1-E0=ΔE.

[0033] The transition absorption wavelength λ can be calculated by substituting the energy difference ΔE into the formula ΔE = hc / λ. h is Planck's constant, c is the speed of light in a vacuum, and λ = 1239.8 / ΔE is used. In the quantum computer calculation process, the energy value is calculated by operating the quantum circuit 20 using a quantum simulator implemented on a computing device such as a general-purpose graphics processing unit (GPGPU). The quantum circuit 20 is also operated on a quantum computer. The quantum circuit 20 may be in the form of a quantum circuit simulated by a classical computer, such as a tensor network, neural network, graph neural network, matrix calculation, etc.

[0034] We will now explain an example of calculating the energy of a compound's electronic state. In each case, the energy of the ground state and the first spin singlet excited state is calculated, and the π-π* transition wavelength is estimated. In the first example, the energy values ​​of benzene are calculated using quantum chemical calculations. In the second example, chlorobenzene is calculated, and in the third example, nitrobenzene is calculated.

[0035] In the first example, the optical absorption wavelength of benzene is calculated. Specifically, the energy difference between the ground state and the first spin singlet excited state is calculated to estimate the π-π* transition wavelength. In the initial calculation step, the type and coordinates of the benzene atoms are specified, and 6-311G(d) is used as the basis function for the calculation level, and the electronic state indicating the initial configuration of each atom is obtained by Hartree-Fock calculation. Based on the electronic state, the natural orbitals of benzene are obtained by second-order Moller-Plesset perturbation calculation, and then converted into molecular orbitals.

[0036] In the basis conversion step, a basis is selected that is a subsystem that governs the electronic state of benzene using the complete active space. Since both the HOMO and LUMO of benzene are antisymmetric with respect to the plane containing the benzene, orbitals with energies close to these orbitals and exhibiting the same symmetry as these orbitals are selected.

[0037] The orbitals selected from the complete active space are six: HONO-2, HONO-1, HONO, LUNO, LUNO+1, and LUNO+2, which exhibit the same symmetry as the 2pzπ orbital of an aromatic ring. Benzene contains 42 electrons, but only the six electrons that occupy these orbitals in the ground state are used, and the degrees of freedom of the other 36 electrons are frozen (fixed). The integrals for one and two electrons can be obtained from these orbitals.

[0038] In the circuit construction process, the electron integral obtained from the basis conversion process is converted into a quantum gate operation using the Jordan-Wigner transformation. In the quantum computer calculation process, the quantum computer operation is executed on a quantum computer simulator configured on a GPGPU unit to obtain an energy value. For example, a Tesla-V-100 (manufactured by NVIDIA) is used as the GPGPU unit.

[0039] The transition absorption wavelength in benzene calculated by IQPEA is approximately 204 nm. The calculation time is approximately 19 hours.

[0040] In the second example, the optical absorption wavelength of chlorobenzene is calculated. Note that explanations of other calculation conditions similar to those in the first example will be omitted. For the natural orbitals calculated in the initial calculation step, in the basis conversion step, a basis that is a subsystem that governs the electronic state of chlorobenzene is selected using the complete active space. In chlorobenzene, since both the HOMO and LUMO are antisymmetric with respect to the plane containing chlorobenzene, orbitals with energies close to these orbitals and exhibiting the same symmetry as these orbitals are selected.

[0041] The seven orbitals selected are HONO-10, HONO-3, HONO-1, HONO, LUNO, LUNO+1, and LUNO+3, which exhibit the same symmetry as the 2pzπ orbital of an aromatic ring. Chlorobenzene contains 58 electrons, but only the eight electrons that occupy these orbitals in the ground state are used, and the degrees of freedom of the other 50 electrons are frozen (fixed). One-electron and two-electron integrals can be obtained from these orbitals.

[0042] The transition absorption wavelength in chlorobenzene calculated by IQPEA is approximately 206 nm. The calculation time is approximately 95 hours.

[0043] In the third example, the optical absorption wavelength of nitrobenzene is calculated. Note that explanations of other calculation conditions similar to those in the first example will be omitted. For the natural orbitals calculated in the initial calculation step, in the basis conversion step, a basis that is a subsystem that governs the electronic state of nitrobenzene is selected using the complete active space. In nitrobenzene, since both the HOMO and LUMO are antisymmetric with respect to the plane containing nitrobenzene, orbitals with energies close to these orbitals and exhibiting the same symmetry as these orbitals are selected.

[0044] The eight orbitals selected are HONO-6, HONO-3, HONO-2, HONO-1, HONO, LUNO, LUNO+1, and LUNO+6, which exhibit the same symmetry as the 2pzπ orbital of an aromatic ring. Nitrobenzene contains 64 electrons, but only the 10 electrons that occupy these orbitals in the ground state are used, and the degrees of freedom of the other 54 electrons are frozen (fixed). One-electron and two-electron integrals can be obtained from these orbitals.

[0045] The transition absorption wavelength in nitrobenzene calculated by IQPEA is approximately 209 nm. The calculation time is approximately 206 hours.

[0046] As shown in FIG. 3 , transition absorption wavelengths are obtained as calculation results by the quantum chemical calculations of Examples 1 to 3. The quantum phase estimation calculation result 41 is the transition wavelength and calculation time obtained by the first to third examples using IQPEA in this embodiment, and the comparative example calculation result 42 is the transition wavelength obtained by a calculation method using the configuration interaction (CI) method (CAS-CI method) utilizing the complete active space (CAS). The quantum phase estimation calculation result 41 shows a similar trend to the comparative example calculation result 42. In the quantum phase estimation calculation result 41, benzene is the calculation result obtained in Example 1, chlorobenzene is the calculation result obtained in Example 2, and nitrobenzene is the calculation result obtained in Example 3.

[0047] In the basis transformation process in the first to third embodiments, trajectories are extracted using the complete active space, but if the number of trajectories is less than a specific number, the circuit construction process is carried out without extraction using the complete active space.

[0048] The quantum chemical calculation of this embodiment can be used with conventional computers (classical computers) and quantum computers. By limiting the number of orbitals for which the Jordan-Wigner transformation is performed to a specific number or less, quantum chemical calculations can be performed with fewer computational resources, even in quantum computer calculations. Computational resources include time, the number of CPU cores, or the number of quantum bits.

[0049] Furthermore, in this embodiment, quantum chemical calculations are performed with fewer quantum bits, which reduces the burden of error correction and makes it possible to realize a quantum computer with fewer errors that can withstand long-term operation. In other words, quantum chemical calculations can be performed as an error-tolerant quantum computer.

[0050] In the energy calculation method of this embodiment, the number of quantum bits used in the calculation can be reduced by applying an active space to extract the orbitals to be calculated, so that even a quantum computer with a limited number of available quantum bits can perform quantum chemical calculations on molecules of practical size. Molecules of practical size are, for example, molecules with molecular weights and numbers of atoms larger than those of hydrogen molecules and lithium hydride, and include compounds such as benzene.

[0051] The flow of operations for calculating the electronic state energy of a molecule in the present invention will be described with reference to the flowchart shown in Figure 4. A target molecule, which is a molecule for which the electronic state energy is to be calculated, is determined (step ST100). The types and coordinates of atoms are specified based on the target molecule, and atomic orbitals of each atom are determined based on the calculation level specified by a user operation (step ST110). Quantum calculations are performed on each atomic orbital using the Hartree-Fock method to calculate the molecular orbitals of the target molecule (step ST120). Quantum calculations are performed on the calculated molecular orbitals using the second-order Moller-Plesset method to calculate the natural orbitals of the target molecule (step ST130).

[0052] The calculated natural orbitals are replaced with molecular orbitals to construct a Hamiltonian for calculating the energy of the electronic state (step ST140). If the number of orbitals in the constructed Hamiltonian is equal to or greater than a specific number (Y in step ST150), the complete active space is used to extract molecular orbitals to be transformed by the Jordan-Wigner transformation to a number less than a specific number (step ST160). If the number of orbitals in the constructed Hamiltonian is less than a specific number (N in step ST150), the extraction is omitted.

[0053] The Jordan-Wigner transformation is performed on the integral of one or two electrons in molecular orbitals less than a specific number to obtain a quantum gate corresponding to the integral of the electrons (step ST170). A quantum circuit 20 represented by a quantum computer gate according to the quantum phase estimation algorithm is constructed (step ST180).

[0054] A quantum phase estimation calculation is performed using a quantum computer simulator or a quantum circuit 20 constructed on a quantum computer (step ST190). As the calculation results of the quantum phase estimation calculation, the energies of the ground state and excited state of the target molecule are calculated (step ST200). The transition absorption wavelength can be obtained from the energy difference between the ground state and the excited state. After the transition absorption wavelength is calculated, the series of steps ends.

[0055] In this embodiment, the processes in the quantum chemistry calculation unit 11, the output control unit 16, the storage unit 17, and the input reception unit are performed by an arithmetic device equipped with a processor. The hardware structure of the processing unit that executes the processing of each step is the following various processors. The various processors are GPGPUs (General-Purpose Computing on Graphics Processing Units), which execute software (programs) and function as various processing units to perform general-purpose data processing other than image processing.

[0056] General-purpose processors that function as various processing units other than GPGPUs include programmable logic devices (PLDs), which are processors whose circuit configuration can be changed after manufacturing, such as CPUs (Central Processing Units), FPGAs (Field Programmable Gate Arrays), and GPUs (Graphical Processing Units), and dedicated electrical circuits, which are processors with circuit configurations designed specifically for executing various processes.

[0057] A single processing unit may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (e.g., multiple FPGAs, a combination of a CPU and an FPGA, or a combination of a CPU and a GPU). Multiple processing units may also be configured with a single processor. Examples of multiple processing units configured with a single processor include: a first configuration, as typified by client or server computers, in which a single processor is configured with a combination of one or more CPUs and software, and this processor functions as multiple processing units; and a second configuration, as typified by system-on-chip (SoC), in which a processor is used to realize the functions of an entire system including multiple processing units on a single IC (Integrated Circuit) chip. In this way, the various processing units are configured with one or more of the above-mentioned various processors as a hardware structure.

[0058] Furthermore, the hardware structure of these various processors is, more specifically, an electric circuit formed by combining circuit elements such as semiconductor elements, and the hardware structure of the storage unit is a storage device such as a hard disk drive (HDD) or a solid state drive (SSD).

[0059] The quantum chemical calculations of this embodiment can be performed using a quantum computer instead of a conventional computer. From the above description, the molecular energy calculation methods described in Supplementary Notes 1 to 8 below can be understood.

[0060] [Supplementary Note 1] A method for calculating molecular energy, comprising: an initial calculation step of calculating natural orbitals of a molecule using a perturbation method in which deviations from the true electronic state of the molecule approximated using a quantum chemical calculation method that approximates the electron interactions of the target molecule with a mean-field potential are considered as perturbation terms; a basis conversion step of extracting a subsystem that governs the electronic state of the molecule based on the calculation results of the initial calculation step; a circuit construction step of constructing a quantum circuit, which is an arithmetic circuit for a quantum computer, based on the selection result of the basis conversion step; and a quantum computer calculation step of operating the quantum circuit constructed in the circuit construction step to calculate an energy value of the electronic state of the molecule. [Supplementary Note 2] The method for calculating molecular energy according to Supplementary Note 1, in which the basis conversion step extracts the subsystem using an active space using the natural orbitals. [Supplementary Note 3] The method for calculating molecular energy according to Supplementary Note 2, in which the basis conversion step extracts the subsystem using the active space to a number less than a predetermined specific number. [Appendix 4] The method for calculating molecular energy according to any one of Appendices 1 to 3, wherein the quantum computer calculation step calculates the energy value by operating the quantum circuit using a quantum simulator. [Appendix 5] The method for calculating molecular energy according to Appendices 4, wherein the quantum simulator is implemented on a general-purpose graphics processing device. [Appendix 6] The method for calculating molecular energy according to any one of Appendices 1 to 5, wherein the quantum chemical calculation method that approximates the electron interactions of the target molecule with a mean field potential is the Hartree-Fock method. [Appendix 7] The method for calculating molecular energy according to any one of Appendices 1 to 6, wherein the perturbation method that takes into account the deviation from the true electronic state as a perturbation term is the Moller-Plesset method. [Appendix 8] The method for calculating molecular energy according to any one of Appendices 1 to 7, wherein the molecules include benzene, chlorobenzene, and nitrobenzene.

[0061] 10 Calculation device 11 Quantum chemistry calculation unit 12 Initial calculation unit 13 Basis conversion unit 14 Circuit construction unit 15 Quantum computer calculation unit 16 Output control 17 Memory unit 20 Quantum circuit 21 Ancillary quantum bit 22 First quantum bit 23 Second quantum bit 31 Hadamard gate 32 Rotation gate 33 Rotation gate 34 Hadamard gate 35 Measurement 36 Unitary gate 41 Quantum phase estimation calculation result 42 Comparative example calculation result ST100 to ST200 Steps

Claims

1. A method for calculating molecular energy comprising: an initial calculation step of calculating the natural orbitals of a molecule using a perturbation method in which the electronic state of the molecule is approximated using a quantum chemical calculation method that approximates the electronic interactions of the target molecule with a mean field potential, and the natural orbitals of the molecule are calculated using a perturbation method in which the deviation from the true electronic state is taken into account as a perturbation term; a basis conversion step of extracting a subsystem that governs the electronic state of the molecule based on the calculation results of the initial calculation step; a circuit construction step of constructing a quantum circuit, which is an arithmetic circuit for a quantum computer, based on the selection result of the basis conversion step; and a quantum computer calculation step of operating the quantum circuit constructed in the circuit construction step to calculate the energy value of the electronic state of the molecule.

2. The method for calculating molecular energy according to claim 1, wherein the basis transformation step extracts the subsystem by utilizing an active space using the natural orbitals.

3. The molecular energy calculation method according to claim 2, wherein the basis transformation step utilizes the active space to extract the subsystems to a number less than a predetermined specific number.

4. The method for calculating molecular energy according to claim 1, wherein the quantum computer calculation step calculates the energy value by operating the quantum circuit using a quantum simulator.

5. The molecular energy calculation method according to claim 4, wherein the quantum simulator is implemented on a general-purpose graphics computing device.

6. A method for calculating molecular energy according to claim 1, wherein the quantum chemical calculation method for approximating the electron interactions of the target molecule with a mean field potential is the Hartree-Fock method.

7. A method for calculating molecular energy according to claim 1, wherein the perturbation method in which the deviation from the true electronic state is taken into account as a perturbation term is the Moller-Plesset method.

8. A method for calculating molecular energy according to any one of claims 1 to 7, wherein the molecules include benzene, chlorobenzene, and nitrobenzene.

Images