Rapid Evolution of the Solar Atmosphere During the Impulsive Phase of a Microflare Observed with Hinode's EIS: Hints of Chromospheric Magnetic Reconnection

We obtained rapid cadence (11.2 s) EUV stare spectra of a solar microflare with the Extreme-ultraviolet Imaging Spectrometer (EIS) aboard Hinode. The intensities of lines formed at temperatures too cool to be found in the corona brightened by factors around 16 early during this event, indicating that energy must be deposited in the chromosphere. The flare started earlier in its transition region emission than it did in its hottest emission, which rules out thermal conduction from a directly heated coronal source as the means of energy transport to the chromosphere. All of the lines showed an increase in nonthermal turbulent velocity by factors of about 3 during the flare. None of the lines exhibited systematic, significant upflows starting at flare onset. We derive the density evolution of the flare plasma at temperature around 2 MK from the intensity ratio of Fe XIV lines at 264.789 and 274.204 A. From both lines we removed the bright pre-flare quiescent emission, and from 274.204 we removed the blended emission of Si VII 274.180 based on the Si VII 274.180/275.361 intensity ratio, which varies only slightly with density. In this way the flare electron density is derived with emission from only the flare plasma. The density increased by an order of magnitude from its pre-flare quiescent average of 3.43+/-0.19 x 10^9 /cm^3 to its maximum impulsive phase value of 3.04+/-0.57 x 10^10 /cm^3 in 2 minutes. The fact that this rapid increase in density is not accompanied by systematic, large upward velocities indicates that the density increase is not due to the filling of loops with evaporated chromospheric material, but rather due to material being directly heated in the chromosphere, likely by magnetic reconnection. The density increase may be due to a progression of reconnection sites to greater depths in the chromosphere, where it has access to larger densities, or it may be due to compression of 2 MK plasma by the 10 MK plasma as it attempts to expand against the high density chromospheric plasma. The further absence of flows during the cooling phase indicates that flare plasma was not falling down (draining from heated, filled loops), but cooling locally in the chromosphere. This work was supported by NASA grant NNX10AC08G.