Electron Acceleration And Relevant Radiation At Coronal Shocks Of Solar Eruptions

Solar eruptions are intense energy release activities in the solar atmosphere. Coronal mass ejections (CMEs) and flares are the most important ones. CMEs are large amount of magnetized plasma ejected from the Sun, representing the largest-scale eruptive phenomenon in the solar system. Flares are sudden and intense brightening in the chromosphere and the corona. When a CME or flare occurs, energy up to1031-1032erg is released rapidly and transformed to the acceleration of particles, heating of the plasma, and radiation across the entire electromagnetic spectrum. It can also have serious impacts on the Earth’s s-pace environment. It is widely believed that the released energy is magnetic energy accumulated in the corona. However, the physical mechanism of particle acceleration remains unclear. Magnetohydrodynamic shocks are driven by solar eruptions both in the coronal and interplanetary space, known to be efficient accelerators in the space. Solar eruptions-coronal shocks-particle acceleration-electromagnetic radiation are the most important physical processes in solar-terrestrial studies. In this thesis, we focus on electron acceleration at coronal shocks, and type Ⅱ radio bursts and hard X-ray and γ-ray emissions produced by energetic electrons. In chapters2and3, we explore the roles of streamers in electron acceleration by shocks and excitation of type Ⅱ bursts; in chapters4and5, we study the spectral hardening at high energies observed in flare continuum emissions and electron acceleration at flare termination shock.We report an intriguing type Ⅱ radio burst that occurred on2011March27. Its dynamic spectrum was featured by a sudden spectral break on the well-observed harmonic branch. Based on plasma emission mechanism, we suggest that the slow-drift period before the break was generated inside the streamer by a coronal eruption driven shock, and the spectral break is a consequence of the radio source region at the shock front crossing the streamer boundary where density drops abruptly. Using the measurements of the shock speed, the radio frequency drift and the temporal duration of the radio emission after the break, one can estimate the thickness and the electron density gradient at the streamer boundary. We find that the density drops from～5×106cm-3to～2×106cm-3at a distance of～0.1RQ(?)。Considering the shock may not propagate perpendicularly across the streamer boundary, this distance should be considered as an upper limit of the above density drop. In spite of this, it is comparable to the electron density measurements in previous observations. We suggest that it is a promising approach to diagnose the properties of the radio emission region by combining specific morphological features of type Ⅱs with imaging observations of solar eruptions.We also examine the other coronal type Ⅱ event occurring on2011March25. Some common observational features of the two events are summarized as follows:(1) Both CMEs erupted from the same active region beneath a well-observed helmet streamer, and the sweeping process of the CME front through the streamer structure can be observed clearly;(2) the heights of CME front obtained from coronagraph imaging observations are consistent with that deduced from the type Ⅱ spectral fitting using a reasonable density model;(3) type Ⅱ radio emission ended when CME/shock fronts passed the streamer cusp region, subject to observational uncertainty. These observational results indicate that both type Ⅱ bursts were possibly related to the shock-streamer interaction. Based on this observational deduction, we suggest that an effective particle acceleration system is established when the shock propagating outward and sweeping through the closed field of the streamer, which may contribute to the acceleration of energetic electrons accounting for type Ⅱ bursts.We further develop a simplified shock-streamer model and carry out a test-particle simulation to study the energization of electrons. Simulation results show that only those electrons that are injected within the closed field region-s can be accelerated efficiently. The shock-streamer trapping effect allows the trapped electrons to return to the shock front multiple times and be repetitive-ly accelerated. Therefore, the shock-streamer configuration plays an important role in our study. Our simulation also shows that electrons which are energetic enough to excite radio bursts, mainly concentrate in the shock upstream within its immediate neighborhood, and around the tip of relevant closed field lines. The locations of energetic electrons can serve as a proxy of the type II radio bursts. This prediction needs to be further verified by future high-spatial resolution ra-dioheliographs at the appropriate metric wavelength. Considering the fact that most solar eruptions originate from closed field regions, electron acceleration by a shock propagating in a closed field structure may be important to the generation of metric type Ⅱs in a general manner. This scenario also provides an alternative explanation to the long-standing issue of the disconnection between metric and interplanetary type Ⅱ bursts.The observed hard X-ray and γ-ray continuum in solar flares are interpreted as bremsstrahlung emissions of accelerated non-thermal electrons. It has been noted for a long time that in many flares the energy spectra show a hardening at energies around or above300keV. We perform a systematic examination of185flares from the SMM/GRS γ-ray detector and identify23electron-dominated events whose energy spectra show clear double power-laws. We then conduct a statistical study of these events and15spectral hardening events that were studied in previous literature. For38events that have γ1, all but one are in the range of2.5to4.5. For34events that have γ2,31of them are in the range of1.5to2.5. In7events we have γ1—γ2>2. It is also found that the spectral index below the break (γ1) anti-correlates with the break energy (εb). Furthermore, γ1also anti-correlates with Fr, the fraction of photons above the break to the total non-thermal photons. The statistical results provide a stringent constraint on the underlying electron acceleration mechanism. Spectral breaks as large as2are hard to explain by merely including electron-electron bremsstrahlung or relativistic correction of the electron-nucleon cross section.We suggest that the hardening in photon spectrum reflects an intrinsic hard-ening in the source electron spectrum. Then we propose an electron acceleration model based on diffusive shock acceleration mechanism at a finite-width flare ter-mination shock. The intensity of the turbulence I(k) determines the magnitude of the diffusion coefficient k, which in turn controls whether a spectral hard-ening will occur. At low energies electrons resonate with the dissipation range turbulence, while at high energies electrons resonate with the inertial range tur-bulence. As a result of the difference of the turbulence spectral indices between the dissipation range and inertial range, we obtain a broken electron spectrum with hardening at high energies by numerically solving the1-D time-dependent Parker’s particle transport equation. We assume the turbulence spectral index in the inertial range εd=2.7, and consider three cases for the dissipation range∈i.5/3,1.5and1.0. We find that in these three cases, the obtained electron spectra can be well fit by a double-power-law, and the fitted parameters are consistent with those observed in flare emission spectra. We also find that the proposed scenario can explain some of the observational results. Therefore we offer a direct and promising explanation for the observed spectral hardening in solar flares.