1 Introduction

Gamma-ray experiments in low-Earth orbit, such as the Compton telescope COMPTEL onboard the Compton Gamma-Ray Observatory (CGRO), operate in an intense and variable radiation environment. The main constituents of the ambient radiation fields are primary cosmic-ray particles, geomagnetically trapped radiation-belt particles, as well as albedo neutrons and $\gamma$-ray photons. The different particle species interact with the spacecraft and detector materials, resulting in the emission of instrumental background photons (for a review, see e.g. Dean et al. 1991). COMPTEL data, dominated by instrumental background, have a typical signal-to-noise ratio of a few percent. Hence, a qualitative and quantitative understanding of the instrumental background is crucial for conducting astrophysical measurements, in particular of the cosmic diffuse gamma-ray background (hereafter CDG), and of the $\gamma$-ray line emission in the interstellar medium or from supernovae and their remnants.

The instrumental background experienced by COMPTEL is subdivided into two major components according to their signature in energy space: first, a continuum background discussed by Ryan et al. (1997); second, the instrumental line background, the focus of this paper. The latter arises from a number of different radioactive isotopes generated in the instrument material. This primarily occurs from activation by trapped protons during passages through the South-Atlantic Anomaly (SAA), from neutron absorption, and from primordial radioactivity. An earlier report on activation in the COMPTEL telescope was given by Morris et al. (1997a).

The discussion is structured as follows. After a brief description of the COMPTEL instrument in Sect. 2, general characteristics of the instrumental (line) background are summarized in Sect. 3. In Sect. 4, identifications of specific isotopes are discussed. In Sect. 5, the variations of the activity of individual isotopes are described. In Sect. 6, a comparison of instrumental line backgrounds in different low-energy $\gamma$-ray experiments is given. Also, average values for the activity of spacecraft materials are presented. The results of this work are summarized and discussed in Sect. 7. Finally, appendices give the event selections used in these line studies, and provide detailed descriptions of the procedures employed for determining the background contributions of individual isotopes in the CDG analysis, and also - with slight modifications - in the analysis of the galactic 1.8 MeV line emission from ²⁶Al.