Alternate View Column AV-49
Keywords: 17 keV neutrinos beta decay solid state detectors weak interactions
Published in the December-1991 issue of Analog Science Fiction & Fact Magazine;
This column was written and submitted 5/10/91 and is copyrighted ©1991 by John G. Cramer.
All rights reserved. No part may be reproduced in any form without
the explicit permission of the author.
This page now has an access count of:
Neutrinos are very peculiar particles. About 610 trillion neutrinos produced by the sun are passing through your body in the second it takes to read this line. If it is night where you are, the neutrinos from the sun are passing through the earth in order to reach you. Because neutrinos have no electric charge and little or no mass, they interact with matter only through the two weakest forces, gravity and the weak interaction. They can pass through solid matter almost as freely as through vacuum, and light years of lead would be needed to absorb them.
There are three distinct "flavors" of neutrinos (e-, µ-, and -neutrinos) and they all seem to have zero charge and negligible mass. Careful measurements indicate that if e-neutrinos have any mass at all, it could be no more than 2 electron-volts or 4 millionths of an electron's mass. Why do neutrinos, unlike their electron brothers and their quark cousins, have zero or near-zero mass? The present standard model of particle physics provides no way of answering this question. Unlike the photon, which must have zero mass because it is the mediating particle of the infinite-range electromagnetic force, there is no fundamental reason why neutrinos should be massless. They just are, as nearly as we can tell.
Or perhaps not. There is at present a growing body of experimental evidence that some neutrinos may have a mass of 17,000 electron-volts (17 keV) or about 1/30 of the mass of an electron. This Alternate View column is about these 17 keV neutrinos and their implications.
The radioactive weak-interaction process through which the neutrino first revealed itself is called beta decay and comes in two vaieties, electron beta decay and positron beta decay. In electron beta decay a radioactive nucleus reduces its mass-energy slightly by converting one of its neutrons into a proton, emitting an electron and an anti-neutrino in the process. The excess energy generated in the decay is distributed randomly between the electron and the anti-neutrino.
Before the neutrino was hypothesized, it was assumed that only the electron was emitted during beta decay. If this was the case, the electron should carry away all of the available energy. Early measurements of electron energy, however, showed that the energy was not a sharp value. Instead it was smeared out in a lumpy distribution that extended from zero to the maximum energy available in the decay. Some of the energy was somehow disappearing, so that the books didn't balance in the energy column. Theoretical considerations also indicated another problem: neutrons, protons, and electrons all have an intrinsic spin angular momentum of 1/2 a unit. Therefore, in converting a neutron to an electron and a proton (1/2 -> 1/2 +/- 1/2), the books didn't balance in the spin column either.
Wolfgang Pauli solved both of these problems at a stroke by hypothesizing the neutrino, another spin 1/2 particle which carried away the missing energy, thereby preserving the laws of conservation of energy and angular momentum. Enrico Fermi was later able to use this idea to accurately predict the observed partition of energy between the electron and the neutrino in beta decay. This energy partition determines the spectrum of electron energies collected from a large number of nuclear decays.
The region near the high energy end of this electron energy spectrum is sensitive to the mass of the e-neutrino (actually the e-anti-neutrino, which should have identical mass). If the neutrino has zero mass, the distribution smoothly tails into the baseline. But if the neutrino has a small mass, the distribution is chopped off early, producing a "nose" with an abrupt edge at the end of the distribution. Careful measurements of electron energy spectra in this region, searching for such a nose, have produced the best upper limit (about 2 eV) on the mass of the e-neutrino. Similar considerations at higher energies have lead to more generous upper limits on the masses of the µ-neutrino and -neutrino.
John Simpson of the University of Guelph in Canada was attempting to pull down the upper limit on the e-neutrino mass when, about six years ago, he found the first evidence for neutrinos with a 17 keV mass. To avoid scattering and absorption in the material that contains the decaying radioisotope source, Simpson implanted the radioisotope tritium, with a half life of 12 years and a maximum beta decay energy of 18.6 keV, directly into the silicon diode detector with which he was measuring the energies of its emitted electrons. The technique, as it turned out, could not be used to improve the limit on the mass of the e-neutrino, but it showed an unexpected effect in the low energy end of the spectrum. There was a break in slope, a "kink", in the otherwise smooth electron energy distribution at an energy about 17 keV below the maximum electron energy.
It was a small effect which might have been produced if about 1% of the counts in the spectrum had been shifted down in energy by 17 keV. Simpson went to great lengths to eliminate the possibility that this was some artifact produced by electronics, cosmic rays, background radioactivity, or some solid-state physics effect such as incomplete energy absorption in the crystal lattice of the detector. Finally he convinced himself that he was dealing with a real effect. He found that the measured data could be explained if one applied conventional weak-interaction theory but assumed that 99% of the time the decay proceeded with a conventional massless neutrino but that 1% of the time a neutrino with a mass of 17 keV was emitted.
Simpson published his results and proposed this explanation. The paper was met with great skepticism by most of the physics community. Many papers were published by others that suggested flaws in Simpson's measurement technique or proposed more conventional explanations of the data.
Simpson is a stubborn man and a careful experimentalist. Over the past five years he has carefully exvaluated the criticisms of his original measurements and has carried out two new experiments, one again using tritium as the beta-decay source and the outher using the isotope sulfur-35, a beta decay source that emits electrons with a maximum energy of 167 keV. Both new experiments showed the same effect, a kink in the spectrum 17 keV below the maximum energy end-point of the spectrum.
Other experimental groups have recently began to take an interest in Simpson's results. Groups at Oxford University, at the Institute Ruder Boskovicin Zagreb, Yugoslavia, and at the Lawrence Berkeley Laboratory have all made measurements that provide supporting evidence for the emission in beta decay about 1% of the time of a neutrino with a mass of 17 keV.
The Berkeley results are particularly interesting because the radioisotope used, carbon-14, was grown into the crystal lattice along with the germanium of which the detector is made. Since carbon and germanium lie in the same column of the periodic table and have similar valence properties, they can exist in the same crystal lattice structure with out dislocations or imperfections. Carbon-14 has an electron end-point energy of 156 keV, and the energy spectrum shows the familiar kink 17 keV below the end point. The leader of the LBL group, Eric Norman, estimates that there is a less than 1% chance that their result is a statistical fluke.
All of these results are preliminary, but the evidence is growing. The one worry that many physicists have is that all of the measurements showing evidence for a 17 keV neutrino have been obtained with radioisoptopes (3H, 14C, 35S, and 51Fe) implanted in either silicon or germanium crystals. No evidence of similar effects has yet been found using magnetic analysis of the electrons, a more standard technique for measuring electron energy spectra. Thus, there remains a lurking doubt that the measured effect may have more to do with properties of crystals than with neutrinos.
Nevertheless, let's accept the result for the moment and ask if a 17 keV neutrino can be fitted into our persent understanding of fundamental particles. As I said above, there is no particular reason, within the standard model, why a neutrino could not have a non-zero mass, even a mass as large as 17 keV. But if any neutrino has a non-zero mass, all three flavors should. Since the decay into the 17 keV neutrino happens only about 1% of the time, it cannot be the e-neutrino and is probably the -neutrino.
Under the scenario that all three neutrino flavors have mass, it is expected that a beta decay will produce a "mixed state" with some probability of producing each of the three neutrino flavors. It is not unreasonable, then, that a -neutrino might be produced with a 1% probability.
The difficulties arise when we attempt to reconcile a 17 keV -neutrino with two other neutrino-related puzzles of contemporary astrophysics, the solar neutrino problem and the dark matter problem.
Evidence for 17 keV Neutrinos:
J. J. Simpson, Physical Review Letters 54, 1891 (1985);
J. J. Simpson and A. Hime, Phys. Rev. D39, 1825 (1989);
A. Hime and J. J. Simpson, Phys. Rev. D39, 1837 (1989);
E. B. Norman, Bull. Am. Phys. Soc. 36, 1260 (1991).
This page was created by John G. Cramer on 7/12/96.