Measurement of the ground-state hyperfine splitting of antihydrogen

Full text: The hydrogen atom is one of the most extensively studied atomic systems, and its ground state hyperfine splitting (GS-HFS) of νHFS = 1.42 GHz has been measured with an extremely high precision of δνHFS/νHFS ∼ 10-12. Therefore, the antimatter counterpart of hydrogen, the antihydrogen atom, consisting of an antiproton and a positron, is an ideal laboratory for studying the CPT symmetry. As a test of the CPT invariance, measuring νHFS of antihydrogen can surpass in accuracy a measurement of the 1S-2S transition frequency proposed by other groups. In fact, it has several advantages over a 1S-2S measurement. Firstly, it does not require the (neutral) antihydrogen atoms to be trapped. Secondly, the only existing consistent extension of the standard model, which is based on a microscopic theory of CPT and Lorentz violation, predicts that νHFS should be more sensitive to CPT violations. In addition, the parameters introduced by Kostelecky et al. have the dimension of energy (or frequency). Therefore, by measuring a relatively small quantity on an energy scale (like the 1.42 GHz GS-HFS splitting), a smaller relative accuracy is needed to reach the same absolute precision for a CPT test. This makes a determination of νHFS with a relative accuracy of 10-4 competitive to the measured relative mass difference of K0 and --K0 of 10-18, which is often quoted as the most precise CPT test so far. The ASACUSA collaboration at CERN's Antiproton Decelerator (AD) has recently submitted a proposal to measure νHFS of antihydrogen in an atomic beam apparatus similar to the ones which were used in the early days of hydrogen HFS spectroscopy. The apparatus consists of two sextupole magnets for the selection and analysis of the spin of the antihydrogen atoms, and a microwave cavity to flip the spin. This method has the advantage that antihydrogen atoms of temperatures up to 150 K, 'evaporating' from a formation region, can be used. Numerical simulations show that such an experiment is feasible if ∼ 100 antihydrogen atoms per second can be produced in the ground state, and that an accuracy of better than 10-6 can be reached within reasonable measuring times. (author)