Increased detector response in optical overfilled measurements due to gas lens formation by nitrogen flow through the entrance aperture

According to our experimental results, a nitrogen flow used to prevent dust and moisture entering a detector may influence measurements performed with trap detectors in overfilled conditions. A stable light source was measured with a wedged trap detector with 4 mm aperture, and the nitrogen flow rate was varied. The nitrogen flow was found to have the largest effect of up to 0.8% on the responsivity of the detector at around 1.0 l min-1 flow rate. The effect of nitrogen flow can be removed down to 0.02% by an added crossflow which removes the nitrogen out of the optical axis. In another experiment, the effect was removed almost completely by changing the flowing gas from nitrogen to synthetic dry air. We also present measurement results that indicate the responsivity changes with nitrogen to be smaller than 0.05% with underfilled beam geometry, even without the added crossflow. Based on simulations, the nitrogen flow through the detector forms a gradient-index type gas lens in front of the detector increasing the effective aperture area and thus the responsivity. In the underfilled measurement geometry there is no light close to the aperture edge which could be refracted inside the detector. Finally, we consider methods to ensure that the responsivity changes due to the gas flow remain below 0.05% in overfilled measurement geometry, without compromising the cleanliness of the detector with too small gas flow rate.