Astron. Astrophys. 331, 1147-1156 (1998)

5. Conclusions

We have considered the propagation of a monoenergetic neutral beam travelling in a dense background plasma, comparing our results with those of Simnett and Haines (1990). In the first instance, we treated the problem without including the response of the background plasma, as was done in Simnett and Haines (1990) (they only considered it a posteriori, concluding that it only reduced the size and steepness of the potential drop). Although electrons do suffer a greater collisional deceleration, careful consideration of the generated electric field shows that the protons will always "drag" the electrons along with them. In this case therefore, no significant separation of the beam electrons and protons can occur, the charge separation being oscillatory in nature and on a scale extremely small compared to the beam non-thermal Debye length. A particular consequence of this result is that the formation of a spatially extended double layer, as assumed by Simnett and Haines (1990), is not possible under solar chromospheric conditions, invalidating their subsequently proposed mechanism for electron runaway. In terms of the unresponsive plasma description presented in this paper, runaway is still possible in principle, with some electrons gaining as much as 75% of the beam proton energy (the 100% quoted by Simnett and Haines (1990) arose from an erroneous factor of appearing in their proton equation of motion). However, in order to avoid "destroying" the self-consistent electric field that accelerates them, only a very small number of such electrons can actually runaway. We conclude, therefore, that electron runaway from neutral beams is negligible in this scenario.

When the response of the background plasma is considered, the end conclusion is the same, but for different reasons. As the beam electrons are decelerated, the generated electric field serves to accelerate the background plasma electrons, which in turn serves to neutralise the current. During this initial stage, both beam and plasma electrons undergo oscillations of the same amplitude and frequency. Ultimately, the beam electrons will be decelerated until they have the same mean speed as the plasma electrons. The two populations of electrons will then become indistinguishable, with the beam protons proceeding alone until stopped by collisions. In this more realistic scenario, the response of the background plasma masks the charge separation, and in doing so makes runaway again negligible.

All of the results described above were obtained using analytic mean particle methods. To add weight to our conclusions, we compared our results to those from an electrostatic particle code, confirming all the above results. In addition to this confirmation, further numerical simulations were carried out to investigate the collective effects. It was shown that a monoenergetic neutral beam generates Langmuir waves in two steps, firstly, like an electron beam and then like a proton beam. This means that at the very beginning the instability growth rate is high, corresponding to an electron beam instability. Then, after a mixing phase, a proton beam instability starts with a much lower growth rate. The final high saturation level of plasma waves corresponds to that of a proton beam.

These effects of neutral beam propagation have interesting consequences for interpretation of radio emission from chromospheric layers during the flare impulsive phase as we discuss in a subsequent paper. According to Simnett and Haines (1990) they also provide a new mechanism for production of flare hard X-ray bursts, though they do not make it clear how. Emission of deka-keV HXRs needs deka-keV electrons with, in a nonthermal beam model, a very large particle flux (Brown 1971). If these electrons were part of a neutral beam the accompanying protons would carry an energy flux times larger and far in excess of the total flare power available. If, as Simnett and Haines suggest, the HXR electrons are runaways accelerated by the E fields created by the neutral beam, then even if the E field were capable of such acceleration, the runaway would have to be extremely energy efficient, transferring most of the flare power from the protons to the runaways and a large electron beam current. Such a mechanism does not eliminate the need for intense electron beams, though it produces them via neutral and proton beam processes rather than directly as a first generation acceleration product.

© European Southern Observatory (ESO) 1998

Online publication: March 3, 1998
helpdesk.link@springer.de