Local fading accelerator and the origin of TeV cosmic ray electrons

The cosmic ray electron spectrum exhibits a break at a particle energy of similar to 1 TeV and extends without any attenuation up to similar to 20 TeV. Synchrotron and inverse Compton energy losses strongly constrain the time of emission of similar to 20 TeV electrons to approximate to 2 x 10(4) yr and the distance of the potential source(s) to approximate to 100-500 pc, depending on the cosmic ray diffusion coefficient. This suggests that maybe one nearby discrete source may explain the observed spectrum of high energy electrons. Given the strong energy dependence (proportional to 1/E) of the cooling time of TeV electrons, the spectral shape of the electron spectrum above the similar to 1 TeV break strongly depends on the history of injection of these electrons from the source. In this paper we show that a local, continuous (on timescales of similar to 10(5) yr) but fading electron accelerator, with a characteristic decay time of similar to 10(4) yr, can naturally account for the entire spectrum of cosmic ray electrons in the TeV domain. Although the standard "nearby pulsar" scenario naturally meets this time condition, it is (almost) excluded by recent measurements of the positron fraction, which above similar to 100 GeV saturates at a level well below 0.5 and drops above similar to 400-500 GeV. The second potential source population, the supernova remnants, accelerate mostly electrons, rather than positrons. However, they hardly can provide an effective production of multi-TeV electrons via the standard diffusive shock acceleration scenario for similar to 10(5) yr. A third possibility are stellar wind shocks, which however are likely to be continuous with nearly constant luminosity on timescales >> 10 kyr and probably cannot match the time requirement of our potential source. Therefore, we face a real challenge in the identification of the origin of the source of multiTeV electrons. Thus, the link of this source with known particle accelerators would require a dramatic revision of the standard paradigms of acceleration and escape in such objects