Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

The Journal of Chemical Physics

Complexity and simplicity of optimal control theory pulses shaped for controlling vibrational qubits

Dmytro Shyshlov, Dmitri Babikov

Published Nov 21, 2012• 7 citations
Publisher

Abstract

In the context of molecular quantum computation the optimal control theory (OCT) is used to obtain shaped laser pulses for high-fidelity control of vibrational qubits. Optimization is done in time domain and the OCT algorithm varies values of electric field in each time step independently, tuning hundreds of thousands of parameters to find one optimal solution. Such flexibility is not available in experiments, where pulse shaping is done in frequency domain and the number of "tuning knobs" is much smaller. The question of possible experimental interpretations of theoretically found OCT solutions arises. In this work we analyze very accurate optimal pulse that we obtained for implementing quantum gate CNOT for the two-qubit system encoded into the exited vibrational states of thiophosgene molecule. Next, we try to alter this pulse by reducing the number of available frequency channels and intentionally introducing systematic and random errors (in frequency domain, by modifying the values of amplitudes and phases of different frequency components). We conclude that a very limited number of frequency components (only 32 in the model of thiophosgene) are really necessary for accurate control of the vibrational two-qubit system, and such pulses can be readily constructed using OCT. If the amplitude and phase errors of different frequency components do not exceed ±3% of the optimal values, one can still achieve accurate transformations of the vibrational two-qubit system, with gate fidelity of CNOT exceeding 0.99.

DOI
10.1063/1.4765344

Related papers

Towards Molecular Quantum Computing: Laser Pulse Shaping of Quantum Logic Gates on Diatomic Molecules

Ryan R. Zaari

The intent of this study is to determine the feasibility of diatomics as molecular quantum computing candidates and shed insight into the use of such experimental laser pulse shaping methods to represent quantum logic gates. Four appropriate rovibrational states of model diatomic molecules are encoded as the qubit states. A set of 2-qubit quantum logic gates (ACNOT, CNOT, NOT, Hadamard) are represented by amplitude and phase shaped laser pulses. The combinations of amplitudes and phases that produce the optimal laser pulse representation, for each quantum logic gate, are determined by a Genetic Algorithm optimization routine. The theoretical laser pulse shaping is analogous to current experimental frequency-domain pulse shaping apparatus with amplitude and phase control at individual frequencies. A model set of diatomics is sampled in order to determine a relationship between optimal laser pulse shaping and the choice of diatomic molecule. We show that the choice of diatomic molecule greatly influences the ability to produce optimal laser pulse shapes to represent quantum logic gates. Tuneable parameters specific to laser pulse shaping instruments are varied to determine their effect on optimal pulse production. They include varying the number of amplitude and phase components, adjusting the number of frequency components, and altering the frequency resolution which is synonymous with altering the laser pulse duration. A time domain analytic form of the original frequency domain laser pulse function is derived, providing a useful means to infer the laser pulse dependencies on these parameters. Initially, we show that the appropriate choice of rovibrational state qubits of carbon monoxide (12C16O) and the use of simple shaped binary pulses, 2 amplitude and 2 phase components, can provide significant control for specific quantum gates. Further amplitude variation at each frequency component is shown to be a crucial requirement for optimal laser pulse shaping, whereas phase variation provides minimal contribution. We show that the generation of optimal laser pulse shapes is highly dependent upon the frequency resolution and increasing the number of frequency components provides incremental improvements to optimal laser pulses.

•Jan 1, 2012•0 citations

High Fidelity Quantum Gates with Vibrational Qubits

Eduardo Berríos, Martin Gruebele, Dmytro Shyshlov +2 more

Physical implementation of quantum gates acting on qubits does not achieve a perfect fidelity of 1. The actual output qubit may not match the targeted output of the desired gate. According to theoretical estimates, intrinsic gate fidelities >99.99% are necessary so that error correction codes can be used to achieve perfect fidelity. Here we test what fidelity can be accomplished for a CNOT gate executed by a shaped ultrafast laser pulse interacting with vibrational states of the molecule SCCl(2). This molecule has been used as a test system for low-fidelity calculations before. To make our test more stringent, we include vibrational levels that do not encode the desired qubits but are close enough in energy to interfere with population transfer by the laser pulse. We use two complementary approaches: optimal control theory determines what the best possible pulse can do; a more constrained physical model calculates what an experiment likely can do. Optimal control theory finds pulses with fidelity >0.9999, in excess of the quantum error correction threshold with 8 × 10(4) iterations. On the other hand, the physical model achieves only 0.9992 after 8 × 10(4) iterations. Both calculations converge as an inverse power law toward unit fidelity after >10(2) iterations/generations. In principle, the fidelities necessary for quantum error correction are reachable with qubits encoded by molecular vibrations. In practice, it will be challenging with current laboratory instrumentation because of slow convergence past fidelities of 0.99.

The Journal of Physical Chemistry A•Jul 18, 2012•15 citations

Anharmonic Properties of the Vibrational Two-Qubit System

Yingying Gu

My thesis focuses on theoretical studies in the molecular quantum computing with vibrational qubits. We studied how the molecular vibrational properties affect the fidelity of the logic gates. Our approach is to encode states of the quantum information register into the vibrational eigenstates of a molecule and use the shaped laser pulse (femtosecond, infrared) to control the vibrational state-to-state transition. By using the Optimal Control Theory to shape a femtosecond laser and numerical propagation of laser-driven vibrational wave packets, we analyzed how the vibrational properties of a two-qubit system affect the accuracy of the quantum gates and how to effectively control the vibrational state-to-state transitions. From these studies, we found the anharmonicities are very important for the control of the vibrational state-to-state transitions and an intricate interplay between the frequencies and anharmonicities in the two-qubit system leads to the occurrence of resonances between different transitions in the vibrational manifold. Such resonances cause "leakage" of population into other vibrational states and hinder the control. From theoretical analysis of resonances, we formulate several criteria for selection of molecule for implementation of the vibrational two-qubit system. These criteria can help experimentalists to choose the best molecules to achieve accurate qubit transformations in the experiment.

•Feb 11, 2015•0 citations

Quantum computing based on vibrational eigenstates: Pulse area theorem analysis

Taiwang Cheng, Alex Brown

In a recent paper [D. Babikov, J. Chem. Phys. 121, 7577 (2004)], quantum optimal control theory was applied to analyze the accuracy of quantum gates in a quantum computer based on molecular vibrational eigenstates. The effects of the anharmonicity parameter of the molecule, the target time of the pulse, and the penalty function on the accuracy of the qubit transformations were investigated. We demonstrate that the effects of all the molecular and laser-pulse parameters can be explained utilizing the analytical pulse area theorem, which originates from the standard two-level model. Moreover, by analyzing the difference between the optimal control theory results and those obtained using the pulse area theorem, it is shown that extremely high quantum gate fidelity can be achieved for a qubit system based on vibrational eigenstates.

The Journal of Chemical Physics•Jan 17, 2006•74 citations

With Vibrationally Excited Thiophosgene Molecule and Double-Well Ion Traps

Dmytro Shyshlov

For practical realization of quantum information processing we need a quantum system that provides reliable preparation of the initial state, high-fidelity quantum gate operations, error tolerance, readout of the result of quantum computation and scalability of the system to increase the number of qubits. In this dissertation we show how these requirements can be addressed for molecular quantum computer. For computational study of quantum information processing with molecules we employ thiophosgene (SCCl2) molecule that has been used as a test system for quantum control experiments [Mol. Phys. 105, 1999 (2007)]. We investigate the gateway scheme of control in which transitions between the vibrational states that encode qubits are only allowed through the intermediate “gateway” state in the B electronic state. This scheme of control provides reliable preparation of the initial qubit state and allows using UV/vis laser pulses. We demonstrate that high-fidelity quantum gates are possible to achieve in molecular quantum computer. The optimal control theory is employed to obtain a shape of laser pulse that performs CNOT gate with ~0.9999 fidelity. Analysis of frequency profile of the optimal pulse shows that preparation of the high-fidelity computational pulse requires only 64 frequency channels. Error tolerance of the computational pulse is studied by modifying amplitudes and phases of frequency components. It is shown that gate fidelity remains high after small modifications are introduced to the optimal pulse. The scheme of readout of quantum information using quantum beat spectroscopy is proposed. The quantum beat signal is obtained from excitation of the final superposition of the qubit states to a readout state in the B electronic state with a time-delayed pulse. We find that fitting the quantum beat signal with a standard fitting expression produces a phase error and propose a new accurate expression that includes phase correction term. The system of two atomic ions trapped in a double-well potential is also studied as a first step towards scalable quantum computation with trapped molecular ions. The rigorous computational treatment of the system provides explanation of the vibrational energy transfer between the ions in terms of wave packet dynamics in the accurate asymmetric potential.

•Jan 1, 2015•0 citations

Rovibrational wave-packet manipulation using shaped midinfrared femtosecond pulses toward quantum computation: Optimization of pulse shape by a genetic algorithm

Masaaki Tsubouchi, Takamasa Momose

We have numerically examined optimization of the experimentally achievable shapes of mid-infrared femtosecond laser pulses to construct elementary quantum gates using molecular rovibrational states of a single molecule. Instead of optimizing the electric field by optimal control theory, the method generally used in theoretical studies of coherent control, the transmittance and the phase shift of conventional pulse shapers were optimized in the frequency domain by a genetic algorithm. The target molecular system we examined was the rotation-vibration states of CO in the $X({^{1}\ensuremath{\Sigma}}^{+})$ ground electronic state. Although the existence of the rotational degrees of freedom makes the quantum gate operation complicated, high fidelity of over 0.95 was achieved for one-rovibrational-qubit systems, which indicates that a molecular quantum computer may be feasible for at least one-qubit calculations on a diatomic molecule. On the other hand, two-qubit calculations are more difficult to achieve in a single CO molecule using rotational and vibrational degrees of freedom as different qubits. The importance of the control of the rotational in addition to the vibrational wave packet in order to realize a molecular quantum computer is discussed.

Physical Review A•May 20, 2008•57 citations
In the context of molecular quantum computation the optimal control theory (OCT) is used to obtain shaped laser pulses for high-fidelity control of vibrational qubits. Optimization is done in time domain and the OCT algorithm varies values of electric field in each time step independently, tuning hundreds of thousands of parameters to find one optimal solution. Such flexibility is not available in experiments, where pulse shaping is done in frequency domain and the number of "tuning knobs" is much smaller. The question of possible experimental interpretations of theoretically found OCT solutions arises. In this work we analyze very accurate optimal pulse that we obtained for implementing quantum gate CNOT for the two-qubit system encoded into the exited vibrational states of thiophosgene molecule. Next, we try to alter this pulse by reducing the number of available frequency channels and intentionally introducing systematic and random errors (in frequency domain, by modifying the values of amplitudes and phases of different frequency components). We conclude that a very limited number of frequency components (only 32 in the model of thiophosgene) are really necessary for accurate control of the vibrational two-qubit system, and such pulses can be readily constructed using OCT. If the amplitude and phase errors of different frequency components do not exceed ±3% of the optimal values, one can still achieve accurate transformations of the vibrational two-qubit system, with gate fidelity of CNOT exceeding 0.99.
The intent of this study is to determine the feasibility of diatomics as molecular quantum computing candidates and shed insight into the use of such experimental laser pulse shaping methods to represent quantum logic gates. Four appropriate rovibrational states of model diatomic molecules are encoded as the qubit states. A set of 2-qubit quantum logic gates (ACNOT, CNOT, NOT, Hadamard) are represented by amplitude and phase shaped laser pulses. The combinations of amplitudes and phases that produce the optimal laser pulse representation, for each quantum logic gate, are determined by a Genetic Algorithm optimization routine. The theoretical laser pulse shaping is analogous to current experimental frequency-domain pulse shaping apparatus with amplitude and phase control at individual frequencies. A model set of diatomics is sampled in order to determine a relationship between optimal laser pulse shaping and the choice of diatomic molecule. We show that the choice of diatomic molecule greatly influences the ability to produce optimal laser pulse shapes to represent quantum logic gates. Tuneable parameters specific to laser pulse shaping instruments are varied to determine their effect on optimal pulse production. They include varying the number of amplitude and phase components, adjusting the number of frequency components, and altering the frequency resolution which is synonymous with altering the laser pulse duration. A time domain analytic form of the original frequency domain laser pulse function is derived, providing a useful means to infer the laser pulse dependencies on these parameters. Initially, we show that the appropriate choice of rovibrational state qubits of carbon monoxide (12C16O) and the use of simple shaped binary pulses, 2 amplitude and 2 phase components, can provide significant control for specific quantum gates. Further amplitude variation at each frequency component is shown to be a crucial requirement for optimal laser pulse shaping, whereas phase variation provides minimal contribution. We show that the generation of optimal laser pulse shapes is highly dependent upon the frequency resolution and increasing the number of frequency components provides incremental improvements to optimal laser pulses.
Physical implementation of quantum gates acting on qubits does not achieve a perfect fidelity of 1. The actual output qubit may not match the targeted output of the desired gate. According to theoretical estimates, intrinsic gate fidelities >99.99% are necessary so that error correction codes can be used to achieve perfect fidelity. Here we test what fidelity can be accomplished for a CNOT gate executed by a shaped ultrafast laser pulse interacting with vibrational states of the molecule SCCl(2). This molecule has been used as a test system for low-fidelity calculations before. To make our test more stringent, we include vibrational levels that do not encode the desired qubits but are close enough in energy to interfere with population transfer by the laser pulse. We use two complementary approaches: optimal control theory determines what the best possible pulse can do; a more constrained physical model calculates what an experiment likely can do. Optimal control theory finds pulses with fidelity >0.9999, in excess of the quantum error correction threshold with 8 × 10(4) iterations. On the other hand, the physical model achieves only 0.9992 after 8 × 10(4) iterations. Both calculations converge as an inverse power law toward unit fidelity after >10(2) iterations/generations. In principle, the fidelities necessary for quantum error correction are reachable with qubits encoded by molecular vibrations. In practice, it will be challenging with current laboratory instrumentation because of slow convergence past fidelities of 0.99.
My thesis focuses on theoretical studies in the molecular quantum computing with vibrational qubits. We studied how the molecular vibrational properties affect the fidelity of the logic gates. Our approach is to encode states of the quantum information register into the vibrational eigenstates of a molecule and use the shaped laser pulse (femtosecond, infrared) to control the vibrational state-to-state transition. By using the Optimal Control Theory to shape a femtosecond laser and numerical propagation of laser-driven vibrational wave packets, we analyzed how the vibrational properties of a two-qubit system affect the accuracy of the quantum gates and how to effectively control the vibrational state-to-state transitions. From these studies, we found the anharmonicities are very important for the control of the vibrational state-to-state transitions and an intricate interplay between the frequencies and anharmonicities in the two-qubit system leads to the occurrence of resonances between different transitions in the vibrational manifold. Such resonances cause "leakage" of population into other vibrational states and hinder the control. From theoretical analysis of resonances, we formulate several criteria for selection of molecule for implementation of the vibrational two-qubit system. These criteria can help experimentalists to choose the best molecules to achieve accurate qubit transformations in the experiment.
In a recent paper [D. Babikov, J. Chem. Phys. 121, 7577 (2004)], quantum optimal control theory was applied to analyze the accuracy of quantum gates in a quantum computer based on molecular vibrational eigenstates. The effects of the anharmonicity parameter of the molecule, the target time of the pulse, and the penalty function on the accuracy of the qubit transformations were investigated. We demonstrate that the effects of all the molecular and laser-pulse parameters can be explained utilizing the analytical pulse area theorem, which originates from the standard two-level model. Moreover, by analyzing the difference between the optimal control theory results and those obtained using the pulse area theorem, it is shown that extremely high quantum gate fidelity can be achieved for a qubit system based on vibrational eigenstates.
For practical realization of quantum information processing we need a quantum system that provides reliable preparation of the initial state, high-fidelity quantum gate operations, error tolerance, readout of the result of quantum computation and scalability of the system to increase the number of qubits. In this dissertation we show how these requirements can be addressed for molecular quantum computer. For computational study of quantum information processing with molecules we employ thiophosgene (SCCl2) molecule that has been used as a test system for quantum control experiments [Mol. Phys. 105, 1999 (2007)]. We investigate the gateway scheme of control in which transitions between the vibrational states that encode qubits are only allowed through the intermediate “gateway” state in the B electronic state. This scheme of control provides reliable preparation of the initial qubit state and allows using UV/vis laser pulses. We demonstrate that high-fidelity quantum gates are possible to achieve in molecular quantum computer. The optimal control theory is employed to obtain a shape of laser pulse that performs CNOT gate with ~0.9999 fidelity. Analysis of frequency profile of the optimal pulse shows that preparation of the high-fidelity computational pulse requires only 64 frequency channels. Error tolerance of the computational pulse is studied by modifying amplitudes and phases of frequency components. It is shown that gate fidelity remains high after small modifications are introduced to the optimal pulse. The scheme of readout of quantum information using quantum beat spectroscopy is proposed. The quantum beat signal is obtained from excitation of the final superposition of the qubit states to a readout state in the B electronic state with a time-delayed pulse. We find that fitting the quantum beat signal with a standard fitting expression produces a phase error and propose a new accurate expression that includes phase correction term. The system of two atomic ions trapped in a double-well potential is also studied as a first step towards scalable quantum computation with trapped molecular ions. The rigorous computational treatment of the system provides explanation of the vibrational energy transfer between the ions in terms of wave packet dynamics in the accurate asymmetric potential.
We have numerically examined optimization of the experimentally achievable shapes of mid-infrared femtosecond laser pulses to construct elementary quantum gates using molecular rovibrational states of a single molecule. Instead of optimizing the electric field by optimal control theory, the method generally used in theoretical studies of coherent control, the transmittance and the phase shift of conventional pulse shapers were optimized in the frequency domain by a genetic algorithm. The target molecular system we examined was the rotation-vibration states of CO in the $X({^{1}\ensuremath{\Sigma}}^{+})$ ground electronic state. Although the existence of the rotational degrees of freedom makes the quantum gate operation complicated, high fidelity of over 0.95 was achieved for one-rovibrational-qubit systems, which indicates that a molecular quantum computer may be feasible for at least one-qubit calculations on a diatomic molecule. On the other hand, two-qubit calculations are more difficult to achieve in a single CO molecule using rotational and vibrational degrees of freedom as different qubits. The importance of the control of the rotational in addition to the vibrational wave packet in order to realize a molecular quantum computer is discussed.