Discrimination of background events in the PolarLight X-ray polarimeter

TL;DR

This paper presents a method for effectively discriminating background events in the PolarLight X-ray polarimeter, significantly improving its sensitivity by removing over 70% of background noise through image property analysis.

Contribution

It introduces a novel background discrimination technique based on image property comparison, approaching the theoretical removal limit and enhancing polarimetric sensitivity.

Findings

Over 70% of background events can be removed.

Background contamination reduced from 25% to 8%.

Achieves a polarimetric sensitivity of around 0.2 Crab.

Abstract

PolarLight is a space-borne X-ray polarimeter that measures the X-ray polarization via electron tracking in an ionization chamber. It is a collimated instrument and thus suffers from the background on the whole detector plane. The majority of background events are induced by high energy charged particles and show ionization morphologies distinct from those produced by X-rays of interest. Comparing on-source and off-source observations, we find that the two datasets display different distributions on image properties. The boundaries between the source and background distributions are obtained and can be used for background discrimination. Such a means can remove over 70% of the background events measured with PolarLight. This approaches the theoretical upper limit of the background fraction that is removable and justifies its effectiveness. For observations with the Crab nebula, the background contamination decreases from 25% to 8% after discrimination, indicative of a polarimetric sensitivity of around 0.2 Crab for PolarLight. This work also provides insights into future X-ray polarimetric telescopes.