Inspiration
With the electron-like properties of Qubits, we now have fascinating new ways to model atoms. This groundbreaking new research leads to many meaningful and impactful applications.
What it does
Our model calculates the ground-state energy of HeH+ over bond lengths from 0.5 to 3.0 angstroms using the Variational Quantum Eigensolver (VQE).
How we built it
Using Qiskit Nature, we mapped the HeH⁺ molecular Hamiltonian onto qubits using Parity mapping. From this, we were able to use the UCCSD ansatz and SLSQP optimizer in a VQE workflow, then compared the results with exact and ideal simulations. Finally, we evaluated the optimized circuits on IBM’s ibm_fez quantum hardware to measure real device performance.
Challenges we ran into
Common challenges were learning how to construct HeH+ using PySCFDriver, how electrons have half-integer spins and are fermionic, how to construct a fermionic Hamiltonian, how to acquire an ansatz using UCCSD, and so on. Conducting cross-method validation and using the two-qubit system instead of four was of special difficulty.
Accomplishments that we're proud of
From our model, we are able to find the least costly combination using Parity+UCCSD+SLSQP, which achieves the best runtime and acceptable error.
What we learned
We successfully learned how molecular Hamiltonians can be mapped to quantum circuits and solved using VQE. Comparing exact, ideal, and IBM hardware results showed us how noise and limited sampling affect accuracy. We also realized that choosing an efficient mapping, ansatz, and optimizer is essential for practical quantum chemistry experiments.
What's next for Quantum Chemistry
To make this pipeline more common and potentially useful in microchemistry modeling.
Built With
- chat-gpt
- python
- qiskit