Test of the principle of equivalence by a null gravitational red-shift experiment

A test of the Einstein equivalence principle (EEP) was performed by carrying out a "null" gravitational red-shift experiment. The experiment compared the rates of a pair of hydrogen maser clocks with those of a set of three superconducting-cavity stabilized oscillator clocks as a function of the solar gravitational potential. If EEP were not valid, the relative rates could vary with potential. During the experiment, the solar potential in the laboratory varied approximately linearly at 3 parts in ${10}^{12}$ per day because of the Earth's orbital motion, and diurnally with an amplitude of 3 parts in ${10}^{13}$ because of the Earth's rotation. An upper limit on the relative frequency variation of 1.7 parts in ${10}^{2}$ of the external potential was set. The accuracy was limited by the frequency stability of the clocks and by un-modeled environmental effects. The result is consistent with the EEP at the two percent level. The experiment can also be viewed as setting a limit on a possible spatial variation of the fine-structure constant.