Quantum computing is an emerging area within the information sciences revolving around the concept of quantum bits (qubits). A major obstacle is the extreme fragility of these qubits due to interactions with their environment that destroy their quantumness. This phenomenon, known as decoherence, is of fundamental interest. There are many competing candidates for qubits, including superconducting circuits, quantum optical cavities, ultracold atoms and spin qubits, and each has its strengths and weaknesses. When dealing with spin qubits, the strongest source of decoherence is the magnetic dipolar interaction. To minimize it, spins are typically diluted in a diamagnetic matrix. For example, this dilution can be taken to the extreme of a single phosphorus atom in silicon, whereas in molecular matrices a typical ratio is one magnetic molecule per 10,000 matrix molecules. However, there is a fundamental contradiction between reducing decoherence by dilution and allowing quantum operations via the interaction between spin qubits. To resolve this contradiction, the design and engineering of quantum hardware can benefit from a 'bottom-up' approach whereby the electronic structure of magnetic molecules is chemically tailored to give the desired physical behaviour. Here we present a way of enhancing coherence in solid-state molecular spin qubits without resorting to extreme dilution. It is based on the design of molecular structures with crystal field ground states possessing large tunnelling gaps that give rise to optimal operating points, or atomic clock transitions, at which the quantum spin dynamics become protected against dipolar decoherence. This approach is illustrated with a holmium molecular nanomagnet in which long coherence times (up to 8.4 microseconds at 5 kelvin) are obtained at unusually high concentrations. This finding opens new avenues for quantum computing based on molecular spin qubits.

Enabling the rational synthesis of molecular candidates for quantum information processing requires design principles that minimize electron spin decoherence. Here we report a systematic investigation of decoherence via the synthesis of two series of paramagnetic coordination complexes. These complexes, [M(C2O4)3](3-) (M = Ru, Cr, Fe) and [M(CN)6](3-) (M = Fe, Ru, Os), were prepared and interrogated by pulsed electron paramagnetic resonance (EPR) spectroscopy to assess quantitatively the influence of the magnitude of spin (S = (1)/2, (3)/2, (5)/2) and spin-orbit coupling (ζ = 464, 880, 3100 cm(-1)) on quantum decoherence. Coherence times (T2) were collected via Hahn echo experiments and revealed a small dependence on the two variables studied, demonstrating that the magnitudes of spin and spin-orbit coupling are not the primary drivers of electron spin decoherence. On the basis of these conclusions, a proof-of-concept molecule, [Ru(C2O4)3](3-), was selected for further study. The two parameters establishing the viability of a qubit are a long coherence time, T2, and the presence of Rabi oscillations. The complex [Ru(C2O4)3](3-) exhibits both a coherence time of T2 = 3.4 μs and the rarely observed Rabi oscillations. These two features establish [Ru(C2O4)3](3-) as a molecular qubit candidate and mark the viability of coordination complexes as qubit platforms. Our results illustrate that the design of qubit candidates can be achieved with a wide range of paramagnetic ions and spin states while preserving a long-lived coherence.

The electronic and magnetic properties of the complexes [Co(terpy)Cl2 ] (1), [Co(terpy)(NCS)2 ] (2), and [Co(terpy)2 ](NCS)2 (3) were investigated. The coordination environment around Co(II) in 1 and 2 leads to a high-spin complex at low temperature and single-molecule magnet properties with multiple relaxation pathways. Changing the ligand field and geometry with an additional terpy ligand leads to spin-crossover behavior in 3 with a gradual transition from high spin to low spin.