A new method that improves gas-chromatography-electroantennographic detection through lock-in amplification is demonstrated. Here, measurements of antennal responses to major pheromone component, <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex Notation="TeX">$Z11$</tex></formula> -16:Ald, are performed under more optimal conditions, using saline electrical connections to excise antennae from male H. subflexa. Matched filtering in colored noise is applied to traditional enhanced graphics adapter recordings, allowing the signal to noise ratio of to be increased by about 6.1 dB (uncertainty is dependent on dosage). A dose-response model, including model parameter uncertainties is then used to estimate and compare performance in terms of naïve error rates involved in the detection of insect responses to GC peaks. Results indicate that relative performance is dosage dependent. Without a visible flame ionization detector (FID) reference peak for determining elution time, the model predicts the detection limit (placed at 5.0% expected naïve error rate) to be approximately 12 times lower using chopper modulation than when using traditional methods. At the highest traditional dosage tested, 10 pg, the equivalent chopper modulated dosage is estimated to be about 92 times lower. When a reference FID peak is clearly visible, the predicted detection limit (at 5.0% expected naïve error rate) is expected to be approximately 7 times lower using chopper modulation than when using traditional methods. At the traditional dosage of 10 pg, the predicted equivalent chopped dosage is estimated to be about 66 times lower.
Read full abstract