We have previously introduced a new type of rotatable flow-through directional passive air sampler (DPAS) for monitoring trace air pollutants in ambient air. In wind tunnel tests, the sampler turns into the prevailing wind direction and retains NO(2) (used as a test pollutant) on an internal sampling medium ring of triethanolamine (TEA)-coated meshes to indicate the source of pollution. However, these meshes can become saturated, after exposure times of tens of hours or a few days, due to the relatively small masses of TEA which can be coated onto them. This paper outlines the saturation problem and presents a possible redesign of the DPAS sampling approach, to allow longer-term (weeks-months) deployments, where air passes over larger volumes of TEA retained in a compartmentalised carousel. Investigations varying the volume, depth and mixing of TEA in sampling compartments suggested that the limiting step of NO(2) uptake was its rate of supply from the atmosphere. In wind tunnel trials, NO(2) uptake into TEA continued linearly in response to a stable air concentration over periods of tens of days, showing no signs of saturation. Uptake was wind velocity dependent across the range of 0.50, 2.00, 5.00 and 8.00 m s(-1). Results indicate that the total sampling time and storage capacity of the TEA for NO(2) can be varied to meet deployment time requirements, indicating that long-term, cheap, directional passive air sampling is achievable.