The new sediment trap has a 0.5 m2aperture, a funnel slope of 34° and is capable of collecting 21 samples at programmed intervals (1 min–1 year) during deployment in the deep ocean. The trap has been designed to allow reliable data on trace inorganic and organic components (such as trace elements,n‐alkanes, PCBs, PAHs, amino/fatty acids) in addition to the standard biogeochemical variables in the collected particles. Due to the exclusive use of synthetic (such as fiberglass, PVC, PTFE, or POM‐Delrin®) and highly resistant metallic materials contamination problems have been eliminated for these species. Blank values determined in several tests in the open ocean were as low as 1% or even less of the amounts present in trap material, even at low particulate loadings. Another major aim was the elimination of loss of dissolved components from the sample cups into seawater. Microbial and chemical processes modify the collected particles, thereby mobilizing originally particulate species into solution. It is thus essential to avoid greater losses of dissolved species through diffusion into the surrounding seawater after collection. This was achieved by means of an especially designed sealing mechanism. Tests with tracer substances during field studies proved that losses of dissolved components from supernatants during one year of deployment are as low as 10%. Additionally, the relationship between flow characteristics around the traps and their excursions from the vertical position in a bottom‐tethered array was studied during a one‐year deployment in the North Atlantic. ©1996 American Institute of Physics.