Letters to Nature

Nature 400, 849-852 (26 August 1999) | doi:10.1038/23655; Received 4 May 1999; Accepted 30 June 1999

Measurement of gravitational acceleration by dropping atoms

Achim Peters1, Keng Yeow Chung1 & Steven Chu1

  1. Physics Department, Stanford University, Stanford, California 94305-4060, USA

Correspondence to: Steven Chu1 Correspondence and requests for materials should be addressed to S.C. (e-mail: Email: schu@leland.Stanford.EDU).

Laser-cooling of atoms and atom-trapping are finding increasing application in many areas of science1. One important use of laser-cooled atoms is in atom interferometers2. In these devices, an atom is placed into a superposition of two or more spatially separated atomic states; these states are each described by a quantum-mechanical phase term, which will interfere with one another if they are brought back together at a later time. Atom interferometers have been shown to be very precise inertial sensors for acceleration3,4, rotation5 and for the measurement of the fine structure constant6. Here we use an atom interferometer based on a fountain of laser-cooled atoms to measure g, the acceleration of gravity. Through detailed investigation and elimination of systematic effects that may affect the accuracy ofthe measurement, we achieve an absolute uncertainty of Deltag/g approximately 3 times 10-9, representing a million-fold increase in absoluteaccuracy compared with previous atom-interferometer experiments7. We also compare our measurement with the value of g obtained at the same laboratory site using a Michelson interferometer gravimeter (a modern equivalent of Galileo's 'leaning tower' experiment in Pisa). We show that the macroscopic glass object used in this instrument falls with the same acceleration, to within 7 parts in 109, as a quantum-mechanical caesium atom.