Quantum information processing (QIP) offers the potential to create new frontiers in fields ranging from quantum biology to cryptography. Two key figures of merit for electronic spin qubits, the smallest units of QIP, are the coherence time (T-2), the lifetime of the qubit, and the spin lattice relaxation time (T-1), the thermally defined upper limit of T-2. To achieve QIP, processable qubits with long coherence times are required. Recent studies on (Ph4P-d(20))(2)[V(C8S8)(3)], a vanadium-based qubit, demonstrate that millisecond T-2 times are achievable in transition metal complexes with nuclear spin free environments. Applying these principles to vanadyl complexes offers a route to combine the previously established surface compatibility of the flatter vanadyl structures with a long T-2. Toward those ends, we investigated a series of four qubits, (Ph4P)(2)[VO(C8S8)(2)] (1), (Ph4P)(2)[VO(beta-C3S5)(2)] (2), (Ph4P)(2)[VO(alpha-C3S5)(2)] (3), and (Ph4P)(2)[VO(C3S4O)(2)] (4), by pulsed electron paramagnetic resonance (EPR) spectroscopy and compared the performance of these species with our recently reported set of vanadium tris(dithiolene) complexes. Crucially we demonstrate that solutions of 1-4 in SO2, a uniquely polar nuclear spin free solvent, reveal T-2 values of up to 152(6) mu s, comparable to the best molecular qubit candidates. Upon transitioning to vanadyl species from the tris(dithiolene) analogues, we observe a remarkable order of magnitude increase in T-1 attributed to stronger solute solvent interactions with the polar vanadium-oxo moiety. Simultaneously, we detect a small decrease in T-2 for the vanadyl analogues relative to the tris(dithiolene) complexes. We attribute this decrease to the absence of one nuclear spin free ligand, which served to shield the vanadium centers against solvent nuclear spins. Our results highlight new design principles for long T-1 and T-2 times by demonstrating the efficacy of ligand-based tuning of solute-solvent interactions.