Fig. 3

High-resolution electronic paramagnetic resonance imaging and spectroscopy. a T ₁ decay rate as a function of applied magnetic field B ₀ for 1 mM of nitric acid, (intrinsic-green diamonds) and 100 mM of Cu²⁺ ions, (intrinsic + target-blue circles). The uncertainty in the relaxation is given by the rms error on the T ₁ stretched exponential fit. b Relaxation rate spectra, \({\rm{\Gamma }}_1^{\left( {{\rm{Cu}}} \right)}\), identified by subtracting \(\Gamma _1^{\left( {{\rm{I}} + {\rm{T}}} \right)} - {\rm{\Gamma }}_1^{\left( {\rm{I}} \right)}\) from the field of view (FOV) (35 × 35 µm²) and region of interest (ROI) (1.6 × 1.6 µm²). The absorption peak at 486 ± 1 G corresponds to a g-factor of 2.21 ± 0.02, consistent with the effective g-factor for aqueous Cu²⁺ ions from conventional CW EPR, see dashed grey line. c Histogram of the measured g-factor from each imaging voxel for Cu²⁺ ions and free electrons. The distribution of g-factors for the free electrons is centred at g = 2.003 with a standard deviation of 0.007, while the Cu²⁺ distribution exhibits a distribution centred at g = 2.199 with a standard deviation of 0.03. d Measured relaxation rate, \({\rm{\Gamma }}_1^{\left( {{\rm{Cu}}} \right)}\), on resonance (B ₀ = 486 G) vs. Cu²⁺ concentration. The error bars are given by the rms error on the T ₁ stretched exponential fit. The minimum detectable number of spins in a sensing volume defined by 0.025 µm³ or (25 aL) is ~ 75,000 spins per voxel or ~ 125 zmol per voxel