The QSPICE app implements well-known methods in the field to derive the circuit Hamiltonian from a lumped element circuit model. However, we include one key improvement to these existing methods. The QSPICE app incorporates an estimate of the parasitic capacitances between different elements in the circuit. This rough estimate is made from the circuit drawing.
There are two important implications of this addition of parasitic capacitances to quantum circuit models. Firstly, previous models limited the type of lumped element circuits that could be quantized. Including parasitic capacitances removes this limitation, allowing any lumped element circuit to be quantized and thereby making it simpler to automate this quantization process.
More meaningfully, noise is an important factor that is limiting the power of quantum computers. This noise can come from outside sources, such as the electronics that control and measure the qubits, or imperfections in the circuit fabrication. But each qubit is also a source of noise to its neighbors through leakage and parasitic elements. This is the same type of noise that is limiting the classical semiconductor industry. This noise source depends intimately on the circuit layout, and many classical software tools exist to describe it.
We aim to make QSPICE the first quantum software tool to accurately describe noise in quantum circuits. Our white paper describes how parasitic elements can be included in a lumped element model by discarding the assumptions contained within Kirchhoff’s laws.
The first design we have created with QSPICE is the null flux qubit. Its sensitivity to low frequency flux noise has been greatly reduced through use of a null flux geometry. See our calculation of its properties.