About 22 years ago, the physics world was briefly rocked by claims of evidence for a new “5th force”, based on reanalysis of data from an early 20th century experiment. Baron Roland von Eötvös, a Hungarian nobleman, had performed extensive measurements of the correlation between inertial mass and gravitational mass and published them in 1922. The lead article in the January 6, 1986 issue of Physical Review Letters had the unassuming title: "A Reanalysis of the Eötvös Experiment" by Ephriam Fischbach, et al. Two days later the New York Times ran an article with the headline: "Hints of Fifth Force in Universe Challenge Galileo's Findings" describing the possible implications of Fischbach's work.
Peculiar experimental results from Eötvös’ gravity measurements and from the behavior of "strange" K-mesons (kaons) were explained in the Fischbach paper by introducing a new theory that proposed a "hypercharge" force, a new fifth force of nature that was gravity-like, but that repelled rather than attracting nearby masses. (See my 15th AV column “Antigravity II: A Fifth Force?” published in the September-1986 issue of Analog.) Fischbach’s work prompted a number of precision experimental “5th Force” tests. These results, when they became available a few years later, provided compelling experimental evidence that there was NO 5th force of the type and strength that Fischbach’s group had predicted. The 5th force idea had been falsified.
Well, Ephriam Fischbach is back with a new re-analysis of old data. In a paper entitled “Evidence for Correlations Between Nuclear Decay Rates and Earth-Sun Distance”, his group at Purdue University has re-analyzed data from long-duration radioactive decay experiments performed at Brookhaven National Laboratory (BNL) in 1986 and at Germany’s Physikalisch-Technische Bundesanstalt (PTB) in 1998.
Both of these experiments were precision determinations of the half-lives of long lived radio-isotopes. The BNL experiment was a study of the beta decay of the isotope silicon-32 (32Si), which has a half life of about 172 years. Data was collected over a period spanning more than 4 years. As a control, the equipment also monitored the beta decay of the isotope chlorine-36 (36Cl), which has a half-life of 301,000 years. Analysis of the data computed the ratio of 32Si to 36Cl decay rates in order to suppress apparatus-dependent systematic variations.
The PTB experiment was a study of the decay of the alpha-decay radioactive isotope europium-152 (152Eu), which has a half life of about 13.5 years. The measurement spanned over 15 years, and the equipment also monitored the alpha particle decay of the isotope radium-226 (226Ra), which has a half-life of 1,600 years. The PTB group similarly used the data to compute the ratio of 152Eu to 226Ra decay rates, in order to suppress apparatus-dependent systematic variations.
The re- analysis of these data by the Fischbach group indicated that in both experiments there were similar time-dependent variations in the measured counting rates at the level of about 0.1 %. Moreover, during a period of about 3 years in which both experiments were collecting data at the same time, the observed time variations showed a remarkable correspondence.
The observed time variations had a period of one year and appeared to be roughly seasonal. The Fischbach group compared the variations with a number of variables that had an annual cycle. They settled on one particular variable, the distance R between the Earth and the Sun, which varies annually by about ±1.7% because the Earth’s orbit is slightly elliptical, so that the Earth is closest to the Sun (R = 147,098,074 km) on about January 3 and farthest from the Sun (R = 152,097,701 km) around July 4. (Curiously, because the seasons are dominated by the tilt of the Earth’s rotation axis, the coldest weather in the Northern Hemisphere occurs when the Earth is closest to the Sun and the sunlight is most intense, and the warmest weather occurs when the total sunlight is weakest.)
When the Fischbach group plotted the variations in 1/R2 (i.e., an inverse square law form) with those of the two radioactive decay rate measurements, they found remarkably good correlations. The formal probabilities that the observed correlations could have occurred by uncorrelated random statistical variation were extremely small, 6 x 10-18 for the BNL data and 2 x 10-246 for the PTB data. The variations appear to be real and to be correlated with the Earth-Sun distance.
Radioactive decays are supposed to be fundamental processes that are well insulated from the influence of environmental effects. The Fischbach work raises the question of what possible physical phenomenon could produce this kind variation in the decay rate of two distinctly different kinds of radioactive decay, one dominated by the weak interaction and the other by the strong interaction? Or alternatively, what kind of unsuspected systematic error could produce almost identical artifact variations in two distinctly different physical measurements performed thousands of miles apart?
One possibility raised in the paper is that there might be a “scalar field” produced by the Sun that modulates the value of the terrestrial fine structure constant a and thereby changes the radioactive decay rates. They cite a theory that predicts such an effect, but note that the coupling strength that would be needed to produce the observed variations is about 14 orders of magnitude larger than the theory would predict.
Another possibility raised in the paper is that the radioactive decays are reacting in some novel way with the flux of neutrinos coming from the Sun. The intensity of solar neutrino flux would follow the inverse square law, and might also be modulated by changes in solar activity. In a follow-up paper, Jenkins and Fischbach noted that a statistically significant drop in the decay rate of the radioactive isotope manganese-54 (54Mn) during the solar flare of December 13, 2006, with several drops in decay rate correlated with spikes in x-rays from the solar flare as detected by GOES satellites.
However, there seems to be a problem with this link to solar flares. The first paper suggests that a decrease in neutrino flux with increasing Earth-Sun distance reduces the radioactivity decay rate, while the solar flare paper suggests that the increase in neutrino flux during a solar flare reduces the radioactivity decay rate. You cannot have it both ways.
Is it plausible that solar neutrinos could modulate the rate of radioactive decays present on the Earth? I would have to say, no. There is no known link between solar neutrino flux and radioactive decay processes (particularly those like the 152Eu alpha-decay that proceeds through the strong interaction,) and our present understanding of fundamental interactions suggests that such a hypothetical link is extremely unlikely.
This, however, leaves us with two questions: (1) what is producing the variations that the Fischbach group reports? and (2) if such variations do, for whatever reason, depend on the distance from the Sun, are there other ways of looking for them?
Let’s consider question (1) first. Decay rates, which are measured as particle detections per second, require a time standard that precisely divides time into well defined intervals. This is usually done with a crystal oscillator that, because of its physical dimensions, oscillates at a frequency that corresponds to an electrical standing wave in the crystal. The dimensions of such a crystal change with temperature. Temperatures, even in a controlled laboratory environment, can have seasonal variations that reflect the external outdoor temperature of environment and the peculiarities of the heating or cooling system.
Therefore, I would be very suspicious that the reported variations on radioactive decay rate might actually be reflecting some tiny seasonal variations in the time standard used in the measurements. I have no reason to believe that this is the problem, but I think that it should be carefully examined before invoking new physics to explain the observations.
As for question (2), the Earth’s orbit has a very small eccentricity, so the annual variations in R are small. A better way of testing whether radioactive decay rates depend directly on 1/R2 would be to monitor a radioactive decay process within a space vehicle in a long elliptic orbit with a large eccentricity, so that R has a very large variation. As it happens, NASA has a number of space probes that match this description, because many space probes, particularly those that venture into the outer reaches of the Solar System, are powered by radioisotope-driven thermoelectric power sources containing a strong radioactive decay source that produces enough energy as heat to power the vehicle. The power levels of such thermoelectric generators are carefully monitored because they constitute the principal power source of the vehicle.
Therefore, as a test of Fischbach’s hypothesis, one should ask NASA scientists whether there is any evidence for variations in the output level on radioisotope power sources in vehicles in long elliptic orbits that would correspond to a 1/R2 variation in the radioactive decay rate. However, I suspect that I already know the answer to this question. If such power-level variations were present, they would be large and would long ago have been reported by the scientists and engineers of the space program.
In any case, it’s an interesting situation. We have suggestive but unexplained observations that may reflect a new and unexpected physical phenomenon. It’s a situation that is familiar to readers of hard science fiction and usually leads to spectacular discoveries that drive the plot line. That could happen in this case, but I doubt it.
AV Columns Online:
Electronic reprints of about 149 "The Alternate View" columns by John G. Cramer,
previously published in
“Evidence for Correlations between Nuclear Decay Rates and Earth-Sun Distance”, Jere H. Jenkins, Ephriam Fischbach, Hohn B. Buncher, John T. Gruenwald, Dennis Krause, and Joshua J. Mattes, arXiv preprint 0808.3283v1 [astro-ph], August 25, 2008.
“Perturbation of Nuclear Decay Rates During the Solar Flare of 13 December 2006”, Jere H. Jenkins and Ephriam Fischbach, arXiv preprint 0808.3156 [astro-ph], August 22, 2008.