Design and evaluation of a gas chromatographic detector based on the measurement of the power reflected from a microwave plasma

The operation of the microwave induced plasma, gas chromatographic detector is based on the measurement of the change in reflected power arising from the interaction of the analyte with an atmospheric pressure argon plasma sustained in the highly efficient TM{dollar}sb{lcub}010{rcub}{dollar} resonant cavity. Microwave forward power and tangential gas flow are optimized for n-pentane. Using the same sample, the analytical performance of the detector is evaluated by obtaining a calibration curve. The results showed that this detection method yields a quadratic calibration curve. The effect of plasma torch design on the analytical performance (sensitivity and detection limits) of the detector is evaluated using a tangential flow torch and a water-cooled capillary plasma torch. The sensitivity with the water-cooled capillary torch was found be to higher by a factor of 7500 and the limit of detection superior by 4 orders of magnitude over the tangential flow torch. Noise analysis of the reflected power signal revealed that the major noise types in the frequency window considered (0-200 Hz) are white noise, low frequency 1/f noise and high frequency discrete noise. The overall magnitude of the noise spectra exhibited a marked dependence on the degree of coupling of power to the microwave cavity and, to a lesser extent, on the flow rate of the plasma gas. The noise did not change consistently as plasma tube cooling water flow rate was changed. The detector's responses to several alkanes, alcohols, and the permanent gases, hydrogen, oxygen, and carbon dioxide are compared with those exhibited by the flame ionization detector. To describe the reflected power signal production mechanism, a simple mathematical model is proposed and evaluated. The model assumes that the major process generating the signal is atomic ionization.