Photonic qubits simulate lithium hydride, cutting qubit needs for chemistry

Researchers at Sapienza Università di Roma and the Istituto di Fotonica e Nanotecnologie, Consiglio Nazionale delle Ricerche (IFN-CNR) have successfully simulated the electronic structure of lithium hydride (LiH) using a quantum density functional theory approach. This work demonstrates a quantum algorithm requiring only O( log_2(N)) qubits to simulate a system of N orbitals, a significant reduction in the computational resources typically needed for molecular modeling.

By encoding the problem on a programmable integrated photonic circuit, the team establishes a pathway for simulating larger molecular systems with advanced hardware for near-term quantum chemistry applications. The results show the viability of integrating qubit-efficient algorithms with reconfigurable photonic hardware.

Photonic Q-DFT Simulates Lithium Hydride Electronic Structure

The lithium hydride (LiH) molecule served as the initial test case for a new quantum simulation approach. This qubit efficiency is achieved by reformulating the complex many-body problem into a more manageable Kohn-Sham reference system, minimizing resource overhead for larger molecular simulations. The simulation used a quantum photonic platform, integrating a deterministic quantum dot single-photon source with a programmable integrated photonic circuit to encode the problem with qubits.

This architecture represents a departure from conventional quantum computing approaches reliant on large numbers of qubits, offering a potential pathway for near-term quantum chemistry applications. The team employed the ensemble extension of the variational quantum eigensolver (VQE) paradigm, minimizing the energy of the LiH molecule within the subspace defined by its ground and first excited states.

Through state-vector simulations using software named qlass, the ensemble VQE circuit mapped onto the photonic encoding was emulated, paired with a classical COBYLA optimizer to refine the simulation. Evaluations extended beyond noiseless simulations to include a noisy model incorporating partial photon distinguishability and photon-counting statistics. The researchers state in their work that the successful demonstration of this Q-DFT implementation on a photonic circuit establishes a pathway for simulating increasingly complex molecular systems.

Q-DFT Algorithm Reduces Qubit Scaling to O(log₂N)

Researchers successfully mapped the Kohn-Sham problem for LiH onto a quantum photonic platform, achieving a qubit scaling of O(log₂N) for a system of N orbitals. This approach differs from conventional quantum computing architectures, which typically demand a linear increase in qubit count alongside system size. This optimization process relies on iterative measurements and classical computations to refine the system’s parameters, ultimately converging on a low-energy solution.

Nathan Shammah of Unitary Fund France and Bruno Senjean of ICGM, Université de Montpellier, CNRS, ENSCM, also contributed to the research. The results demonstrate the viability of this integrated platform for tackling increasingly complex molecular simulations.

Encoding Molecular Systems on Integrated Photonic Circuits

This demonstration bypasses the typical linear scaling of qubit demands with system size, a major hurdle in simulating complex molecules with conventional quantum computers. The team’s implementation of Q-DFT reduces the qubit requirement to logarithmic scaling, specifically O(log₂N) qubits for a system of N orbitals, allowing for simulations previously inaccessible due to hardware limitations. The researchers evaluated both the ground and first excited-state properties of LiH to validate the algorithm and hardware.

The programmable integrated photonic circuit allows for dynamic reconfiguration of the quantum algorithm, adapting to the specific requirements of different molecular systems. The researchers note that the accuracy of the simulation is primarily limited by the approximate density functional employed, suggesting that further improvements in theoretical methods could lead to even more precise and reliable results.

LiH Molecule as a Two-Qubit Quantum Simulation

The lithium hydride (LiH) molecule was successfully simulated using only two qubits, a feat previously requiring twelve qubits with conventional wavefunction theory approaches. This reduction in qubit demand stems from the implementation of quantum density functional theory (Q-DFT) on a programmable integrated photonic circuit, allowing for a more efficient encoding of the molecular system. The team used a Q-DFT algorithm that scales with log base 2 of the number of orbitals, dramatically lowering resource requirements for larger molecular simulations.

The researchers employed an ensemble variational quantum eigensolver, a method that simultaneously estimates the energy and density of the Kohn-Sham system to determine the ground-state energy of the physical system. The self-consistent field loop required only six iterations in the noiseless quantum simulation, mirroring the seven iterations needed by the reference PySCF results.

Convergence was defined by an energy change below ${10}^{–5}$ Hartree and a root mean square deviation change in density below ${10}^{–4}$ between iterations. The use of a photonic platform allowed for the simulation of LiH with only two qubits, as the molecule, in a minimal basis set, requires only log base 2 of four qubits at the Kohn-Sham density functional theory level. Removing two of the lithium’s orbitals, the 2p x and 2p y, further reduced the qubit requirement.

Ten simulations, each initialized with random parameter values and using ${10}^9$ shots, showed consistent convergence, with the solid purple curves representing the results and the dashed line indicating their mean. “Hence, only 2 qubits are needed to compute the energy of the LiH molecule in the minimal basis at the KS-DFT level of theory,” the researchers state, highlighting the efficiency of their method. This advancement suggests a viable path toward simulating larger, more complex molecular systems with reduced quantum resource demands.

Photonic Platform Advances for Quantum Chemistry

This demonstration bypasses the typical qubit scaling challenges associated with quantum chemistry simulations, offering a new pathway for modeling molecular properties with reduced computational resources. The approach uses quantum density functional theory (Q-DFT), a reformulation of the many-body problem that reduces qubit demands to logarithmic scaling with the number of orbitals, O(log₂N) qubits for N orbitals. This efficiency stems from representing the molecular system as a non-interacting Kohn-Sham system, simplifying the quantum computation.

Coincidence events, indicative of successful quantum operations, were recorded using a time-to-digital converter and processed by a classical computer to determine expectation values. The single-photon source, based on a semiconductor quantum dot excited by a pulsed laser, played a major role in generating the necessary quantum states.

DFT Foundations and Limitations in Quantum Chemistry

This photonic platform uses a deterministic quantum dot single-photon source, critical for generating the necessary quantum states, coupled with a reconfigurable circuit that allows for precise control and manipulation of these photons. The team acknowledges that sources of error beyond those currently accounted for remain significant, but the results suggest a viable path toward simulating larger molecular systems.

More like this

Quantum PhysicsHidden entanglement survives decoherence, new simulations from Ulm University suggest

Quantum AlgorithmsQoro Quantum expands cloud simulation for quantum algorithms

Quantum Research NewsTensor networks model mixed states in topological quantum matter

Quantum AlgorithmsPsiQuantum’s quantum circuits aid porous material simulation