PEBS - Positron Electron Balloon Spectrometer > PEBS physics > Positron fraction

Positron fraction

Because of their low mass, electrons and positrons are strongly affected by radiative energy losses when propagating from their sources to the observer. The energy loss rate through inverse Compton scattering in the cosmic microwave background, and from synchrotron radiation in the galactic magnetic field, increases with the square of the electron energy. Consequently, a power-law energy spectrum E at the source would be steepened to E-(α+1) at the observation site for electrons at high energy (E>10 GeV). For shock-acceleration in supernova remnants, values of α slightly larger than 2.0 are expected and seem to be consistent with many of the observations. For secondary electrons/positrons, the source spectrum would follow the energy spectrum of ambient protons, approximately ≈E-2.7 (which is softer than the spectrum at the source of protons because of energy dependent diffusion of cosmic rays through the galaxy), and then this power law would be further steepened by radiative losses to approach an observed spectrum ≈E-(2.7+1.0) . Hence, if all positrons in the cosmic rays are exclusively of interstellar origin but a substantial intensity of electrons comes from primary sources, the positron fraction e+/(e++e-) should continuously and smoothly decrease with energy. Experimentally, the separation of positrons and electrons must be performed via deflection of the particles in a magnetic field, and the identification of positrons is severely complicated by the background of protons which may exceed the positron intensity by 4 or 5 orders of magnitude. The first positron measurements, performed with a magnet spectrometer on balloons up to about 10 GeV seemed to confirm the decline with energy of the fraction of positrons. However, the situation became more complicated when more measurements, extending to higher energy, became available. An experiment which used the earth’s magnetic field as an analyzer reported an unexpectedly large positron fraction around 20 GeV which, however, could not be confirmed with subsequent measurements with the superconducting magnet spectrometers HEAT and CAPRICE. Nevertheless, the HEAT results indicated a small excess of positrons above about 6 GeV which is not expected in conventional secondary production models. Measurements in space with AMS-1 gave results consistent with HEAT. However, none of these measurements could reach sufficiently high energies to answer the crucial question of how the positron feature extended to higher energy.

Recent developments

Very recently, results between 1 to 100 GeV from the PAMELA space mission were released. These data, shown below together with previous measurements, have generated significant excitement as they seem to exhibit two new features: at energies below 5 GeV, the reported positron fraction is well below most previous measurements, but at higher energies, a monotonic increase appears up to 100 GeV.

Compilation of measurements of the positron fraction, figure taken from arXiv:0810.4995

This increase is not inconsistent with the HEAT and AMS data in the region where overlap exists, but is well above any reasonable prediction for an exclusively interstellar secondary origin of the positrons. The low-energy behavior of the PAMELA results could be a solar modulation effect, due to a positive/negative charge asymmetry of solar modulation which alternates from one solar cycle to the next. This will be discussed further below. However, the stunning high-energy increase requires new physics or astrophysics for explanation. Is it a long-sought sign of the contribution of dark matter particle decays, or does it point towards pair production processes in a nearby pulsar or an exploding Wolf-Rayet star? Or is it an experimental artifact due to unresolved background in the data? We cannot here review the numerous attempts at interpretation that have appeared in the literature (an example is shown in the figure below),

A possible origin of the positron excess from the Geminga pulsar, figure taken from arxiv:0810.2784

but we emphasize the obvious conclusion: The present data must be confirmed by an independent measurement with superb positron-proton discrimination power, and the energy coverage should be extended beyond the TeV range: will the positron-fraction continue to increase, or saturate and perhaps fall again? These, exactly, are objectives of the PEBS project. For energies beyond a few hundred GeV, the separation of electrons and positrons is not currently possible, but a few measurements exist that attempt to determine the all electron (e++e-) intensity into the TeV region. This is an extremely interesting region, because radiative energy losses become so strong at TeV energies that only electrons from relatively close (of the order of one kpc or less) and young (≤105y) sources are expected to reach the solar system. Thus, unless there are fairly local pulsars or supernova-remnants acting as accelerators, the electron intensity is expected to drop off rapidly at those energies. In fact, evidence for such a cutoff could already be hinted by indirect observations from the ground with the HESS air-Cherenkov telescope.

There are considerable disagreements between individual electron data sets, even when all data sets are normalized to agree at a given energy. It seems that the overall high-energy slope of the spectrum is quite steep, with a power-law exponent of about 3.3, but one must realize that all data above about 300 GeV so far have come from a single experiment, the series of observations with the Japanese emulsion chambers. The situation changed recently when results from long-duration balloon flights of the ATIC instrument and from PPB-BETS were reported. These new data are interpreted as providing evidence for a spectral feature, with a peak in the electron energy spectrum (when multiplied with E3) between 300 and 600 GeV. Thus, is there indeed an additional component to the electron intensity in this region, and what is its origin? Or could there be a systematic problem with the measurement? There are numerous phenomenological studies that attempt to explain these findings and often try to identify a connection between this feature and the apparent increase of the positron fraction at lower energy. It is obviously very important to clarify this issue with independent new observations. Measuring this region with excellent hadronic background rejection is a central goal of the PEBS-1 project and will be extended by PEBS-2.

Intended measurement

We plan to build and fly a permanent magnet spectrometer combined with a TRD and Calorimeter using state-of-the-art technology from European science groups working at CERN and also with AMS-2. This balloon mission we call PEBS-1. PEBS-1 will be flown first in the northern hemisphere from Kiruna to Canada/Alaska in 2012. A second flight is envisaged in Antarctica in 2013/14. As well as these high-impact science measurements we plan to demonstrate on these flights that the newly-developed high-resolution scintillating fiber tracker is a viable option for future missions. The expected measurements for PEBS-1 on a single northern balloon flight are shown in the following figure.


The expected statistics from a single flight of PEBS-1 shown as blue circles
assumed to conform to GALPROP model. PAMELA is shown in black, and a weighted
mean of previous measurements is shown in red. The different curves show expectations from
different charge dependent solar modulation models.

Here the statistics and energy range (up to ~20GeV) of the positron fraction are shown as blue symbols. The lines are various versions of the expectations from the GALPROP simulation package developed by Strong and Moskalenko with no additional sources of positrons/electrons. The PEBS-1 data are clearly of high statistical significance and will have no hadronic background contamination. The expected number of positrons+electrons at high energies >1TeV for PEBS-1 in a single northern 5-day flight is ~10.