The control of the RP re-encounter probability finds a direct application to improve the performance of chemical devices. Here, we show how a simple-to-implement control scheme highly enhances the sensitivity of a model chemical magnetometer by up to two orders of magnitude. The basic idea behind a chemical magnetometer is that, since a change in the magnetic field modifies the amount of singlet products, one can reverse the reasoning and measure the chemical yield to estimate B. Intuitively, the magnetic sensitivity is high when a small change in the magnetic field intensity produces large effects on the singlet yield. Formally, it is defined as:(2)Λs(B)≡∂Φs(B)∂B=∫0∞pre(t)gs(B,t)dt,with gs(B,t)≡∂fs(B,t)∂B being the instantaneous magnetic sensitivity. The functional form of fs(B,t)=Sρel(t)S strongly depends on the specific realization of the radical pair, in particular on the number of the surrounding nuclear spins. Here, we consider a radical pair in which the first electron spin is devoid of hyperfine interactions, while the second electron spin interacts isotropically with one spin-1 nucleus, e.g. nitrogen. In the context of the chemical compass (i.e. when the task is determining the magnetic field direction through anisotropic hyperfine interactions), an analogous configuration (with only one spin-1/2 nucleus) has been proposed [3], and numerically characterized [8], as being optimal: Additional nuclear spins would perturb the intuitive ‘reference and probe’ picture. The Hamiltonian then simplifies to H=-γeB(S1(z)+S2(z))+|γe|αS→2·I→, where α is the isotropic hyperfine coupling.
