Figure 2. The g-factor experiment for highly charged ions.

The triple Penning-trap set-up (gold) comprises: (i) The PT with a homogeneous magnetic field to determine the frequency ratios Γ* by measuring the three motional eigenfrequencies and probing the Larmor frequency. (ii) The AT to detect the spin-state of the valence electron. (iii) The Creation trap (CT) for ion creation within a miniature electron beam ion source (mEBIS). To enhance the production rate of ⁴⁸Ca ions, an enriched calcium target is used with the following isotope composition: ⁴⁰Ca: 78.77%, ⁴²Ca: 3.02%, ⁴³Ca: 0.62%, ⁴⁴Ca: 9.55%, ⁴⁶Ca: 0.02% and ⁴⁸Ca: 8.02%. The set-up is placed in a cryogenic (T=4.2 K) ultra-high vacuum chamber (P<1 × 10⁻¹⁶ mbar). In a the axial resonator noise spectrum is shown including the dip-signal of a thermalized single ⁴⁸Ca¹⁷⁺ ion. In b the spin-state of the ⁴⁸Ca¹⁷⁺ ion is detected as an axial frequency jump at an absolute axial frequency of ν_z,off=412.4 kHz. In c the spin-flip probability is shown in dependence of the measured Γ*-values, scaled by the final central Γ value Γ_mean=5138.837 974 37 (58). The black points represent binned data to guide the eye. This data binning is not relevant for the Gaussian maximum-likelihood (ML) fit, shown in red. The dark grey-shaded area illustrates the uncertainty of Γ_mean and the bright grey area represents the binomial errors considering the amount of cycles of binned data and the probability of the ML fit. Error bars represent the uncertainties of each single axial frequency measurement point is related to the 1 sigma standard deviation.