Direct detection experiments turn to lose sensitivity of searching for a sub-MeV light dark matter candidate due to the threshold of recoil energy. However, such light dark matter particles can be accelerated by energetic cosmic-rays such that they can be detected with existing detectors. We derive the constraints on the scattering of a boosted light dark matter and electron from the XENON100/1T experiment. We illustrate that the energy dependence of the cross section plays a crucial role in improving both the detection sensitivity and also the complementarity of direct detection and other experiments.

Recently, DAMPE has released its first results on the high-energy cosmic-ray electrons and positrons (CREs) from about $25$ GeV to $4.6$ TeV, which directly detect a break at $\sim 1$ TeV. This result gives us an excellent opportunity to study the source of the CREs excess. In this work, we used the data fo proton and helium flux (from AMS-02 and CREAM), $\bar{\mathrm{p}}/\mathrm{p}$ ratio (from AMS-02), positron flux (from AMS-02) and CREs flux (from DAMPE without the peak signal point at $\sim 1.4$ TeV) to do global fitting simultaneously, which can account the influence from the propagation model, the nuclei and electron primary source injection and the secondary lepton production precisely. For extra source to interpret the excess in lepton spectrum, we consider two separate scenarios (pulsar and dark matter annihilation via leptonic channels) to construct the bump ($\gtrsim 100$ GeV) and the break at $\sim 1$ TeV. The result shows: (i) in pulsar scenario, the spectral index of the injection should be $\nu_{\mathrm{psr}} \sim 0.65$ and the cut-off should be $R_{c} \sim 650$ GV; (ii) in dark matter scenario, the dark matter particle's mass is $m_{\chi} \sim 1208$ GeV and the cross section is $\langle \sigma v \rangle \sim 1.48 \times 10^{-23} \mathrm{cm}^{3} \mathrm{s}^{-1}$. Moreover, in the dark matter scenario, the $\tau \bar{\tau}$ annihilation channel is highly suppressed, and a DM model is built to satisfy the fitting results.