Optical frequency standard with a single 171Yb+ ion

With the development of atomic frequency standards, frequencies and time intervals became the physical quantities that can be measured with the highest precision. One of the most promising developments in the ongoing research for even more accurate frequency standards are frequency standards based on narrow optical transitions in laser-cooled ions stored in radio-frequency traps. These single-ion optical frequency standards have the potential to reach instabilities and systematic uncertainties beyond the capabilities of the best microwave standards available at present. In this work, the 2S1/2(F=0, mF=0)→2D3/2(F=2, mF=0) electric quadrupole transition in a single 171Yb+ laser cooled ion stored in a radio frequency Paul trap is investigated as a candidate for such an optical frequency standard. Frequency comparisons between two such frequency standards were used as a method to investigate the instability properties and systematic uncertainties of the frequency standard. For this purpose a new ion trap system was set up in addition to an existing one. An instability measured by an Allan deviation of σy(t)≅1.1 x 10-14t-1/2 of the frequency difference between the two standards was demonstrated. This is close to the stability limit derived from numerical simulations that include quantum projection noise as the only source of noise. A mean relative frequency difference between the two frequency standards of 3.8(6.1) x 10-16 was found, which is comparable to the best result achieved with cesium fountain clocks. Two systematic effects, the quadratic Stark effect and the quadrupole shift, were studied in detail and the relevant atomic parameters were measured for the first time. For the quadratic Stark effect, the relevant static atomic scalar- and tensor electric polarizabilities were measured to be ΔαS=-6.9(1.4) x 10-40 Jm2/V2 and αT(D3/2)=-13.6(2.2) x 10-40 Jm2/V2 respectively. The quadrupole moment of the 2D3/2 state was determined to be Θ(D3/2)=9.32(48) x 10-40 Cm2, which is in good agreement with theoretical calculations. Estimates for all known systematic uncertainties of the frequency standard were derived leading to a total systematic relative uncertainty of 1.0 x 10-15 for the present state of the experiment. Experimental strategies are discussed that will allow to improve the relative uncertainty to a few parts in 1017 over the next years. (orig.)