Logo PTB
symbolic picture: "magazines"

Are fundamental constants really constant?

The discussion about possible variations of fundamental constants was excited in recent years because astrophysical observations seem to indicate a significant change in the value of the fine structure constant, a, around 5 - 10 billion years ago. PTB engaged in a search to detect a possible variation of a and carried out a precise comparison of different atomic frequencies over a period of three years. The result excludes a present day variation of a larger than a relative change of 2 · 10–15 per year.

The diagram shows the results of the measurements of transition frequencies in Yb and Hg ions and in the H atom (with their respective 1 s uncertainties). The relative change in the Rydberg frequency (Ry) is plotted as a function of the relative change of the fine structure constant, a. Since the point of origin of the diagramme, where Ry and a are strictly constant, still remains within the area of “combined uncertainty” (central ellipse), the measurements do not indicate a change in Ry and a.

The postulate of the constancy of fundamental quantities – like, for instance, the speed of light – is basic to how physics describes the universe and, therefore, is decisive for metrology. It must be experimentally verifiable. New theoretical models for a unified description of the fundamental interactions of physics indeed allow such variations or even describe them as indispensable.

The most important test case is the so-called fine structure constant, a, a dimensionless number composed of the speed of light, the elementary charge and Planck’s constant. It appears in the description of several atomic and electric phenomena. By comparing atomic clocks based on different atomic transition frequencies a very precise test of the constancy of a can be carried out: if a changes in time, the clocks will increasingly drift out of synchronism as a effects the different atomic transition frequencies each in a specific manner.

At PTB an optical frequency standard based on a trapped ytterbium ion was compared to the primary caesium atom clock CSF1 (“fountainclock”) twice over an interval of three years. Within the combined uncertainty, the two mea-surements agreed very well. The frequency of the optical standard had not changed in comparison to the caesium clock. Similar results were obtained during 1999 to 2003 by NIST (Boulder, USA) for the optical frequency of a trapped mercury ion and – in a collaboration between MPQ (Garching) and BNM-SYRTE (Paris) – in an investigation on atomic hydrogen.

The combination of these results allows a clear conclusion on the constancy of a. Should there have been dynamic changes of the fundamental constants in the early stages of the universe they have obviously decreased in our time to the point that they are below the detection limits of the best available precision measurements.

Contact at PTB:

Phone: +49-531-592-0