Can an injection model replenish filaments in a weak magnetic environment?

We observed an H surge that occurred in NOAA Active Region 12401 on 2015 August 17, and we discuss its trigger mechanism, and kinematic and thermal properties. It is suggested that this surge was caused by a chromospheric reconnection which ejected cool and dense material with transverse velocity of about 21-28 km s^-1 and initial Doppler velocity of 12 km s^-1. This surge is similar to the injection of newly formed filament materials from their footpoints, except that the surge here occurred in a relatively weak magnetic environment of ∼100 G. Thus, we discuss the possibility of filament material replenishment via the erupting mass in such a weak magnetic field, which is often associated with quiescent filaments. It is found that the local plasma can be heated up to about 1.3 times the original temperature, which results in an acceleration of about-0.017 km s^-2. It can lift the dense material up to 10 Mm and higher with an inclination angle smaller than 50°, namely the typical height of active region filaments, but it can hardly inject the material up to those filaments higher than 25 Mm, like some quiescent filaments. Thus, we think that the injection model does not work well in describing the formation of quiescent filaments.