Skip to Content

Hydrogen BEC at MIT

[Historical report created from GSU BEC Homepage:]

"The MIT group has provided this report of progress in the long search for BEC in spin polarized hydrogen:

Dear Colleague:

From the inquiries we have received we know that there is interest in the BEC community about our recent observation of BEC in atomic hydrogen. In response, we have prepared this memorandum. The discovery is quite recent (June 12, 1 A.M., is when we broke out the Champagne). We are still interpreting the data and would like to take some more time before publication. Meanwhile, we thought that it would be helpful to tell you what we have seen. However, these findings are tentative and subject to revision. If you wish to refer to these findings, this should be described as a private communication. And if you should wish to refer to any of the numbers, please emphasize that they are preliminary and subject to revision. What took us over the threshold of BEC is implementation of radio frequency evaporation. Previously, we evaporated by lowering the magnetic field in our Ioffe-Pritchard trap that confined the atoms axially. At low temperature this process becomes inefficient. With rf evaporation, the atoms can emerge in all directions, leaving wherever the magnetic field satisfies the resonance condition for the applied rf field. Important to the advance was our observation and measurement of a pressure shift for the 1S-2S transition. This had been calculated by Dalgarno, Jamieson and Doyle, but we had not previously seen it clearly. Our results are in reasonable agreement with theory. Using the pressure shift, we directly determine the density from the shift in the spectral line. The shift for a density of 1013 atoms/cm3 is approximately 2 kHz (@243nm).

Our spectroscopic technique was the same as used previously: excitation of the 1S-2S two photon transition by a pulse of laser light at 243 nm, followed by application of an electric field to quench the 2S state, causing Lyman alpha photons to be emitted, which we detect.

As we decreased the temperature the density increased as expected. However, when we crossed the BEC transition line, the density decreased rapidly with decreasing temperature. This is expected with hydrogen. Two-body dipole relaxation in the condensate causes a rapid loss of atoms and limits the condensate fraction to a few percent.

The smoking gun was detection of a separate spectral feature, shifted far to the red. (Far being up to 700 kHz.) Explanation: the pressure shift in the condensate is extremely high due to the high density. So, the condensate displays a spectral feature distinct from the normal gas. The size of the signal, its pressure shift, and its behavior in time, were in good agreement with what one expects for a condensate. This was all done by Doppler free spectroscopy- i.e. absorption of one photon from each of two counter-propagating beams.

There should be a recoil-shifted signal from the condensate arising from absorbing two photons from a single laser beam. We have not yet seen this, but our search has not been thorough. We have, however, seen the Doppler broadened signal of the normal gas. The width of this signal tells us the gas temperature. The result agrees with other estimates.

Some TENTATIVE numbers:

  • Size of condensate: the number of atoms exceeds 108.
  • Condensate fraction: a few percent.
  • Condensate density: a few times 1015 atoms/cm3. (Note - the three body loss rate in hydrogen is very small.)
  • Transition temperature: about 40 µK.
  • Normal density at transition: about 2 X 1014 atoms/cm3.
  • Lifetime of condensate: up to five seconds.

We will prepare numbers for publication as soon as possible, but meanwhile this lets you know more or less what we are up to.

At the moment we are letting the dilution refrigerator warm up a bit to overcome the system's emotional trauma from having worked, which resulted in one of the lines becoming plugged.

The members of our research group who carried out this last phase in our hunt for BEC in hydrogen were Dale Fried, Tom Killian, Lorenz Willmann, David Landhuis and Stephen Moss.

Tom Greytak email: tom@noleak.mit.edu
Dan Kleppner email: dk@amo.mit.edu

June 14, 1998"