Pressure-dependent calibration of the OH and HO2 channels of a FAGE HOx instrument using the Highly Instrumented Reactor for Atmospheric Chemistry (HIRAC)

The calibration of field instruments used to mea- sure concentrations of OH and HO2 worldwide has tradi- tionally relied on a single method utilising the photolysis of water vapour in air in a flow tube at atmospheric pres- sure. Here the calibration of two FAGE (fluorescence as- say by gaseous expansion) apparatuses designed for HOx (OH and HO2) measurements have been investigated as a function of external pressure using two different laser sys- tems. The conventional method of generating known con- centrations of HOx from H2O vapour photolysis in a turbu- lent flow tube impinging just outside the FAGE sample in- let has been used to study instrument sensitivity as a func- tion of internal fluorescence cell pressure (1.8-3.8 mbar). An increase in the calibration constants COH and CHO2 with pressure was observed, and an empirical linear re- gression of the data was used to describe the trends, with 1COHD (17 11) % and 1CHO2 D (31.6 4.4) % increase per millibar air (uncertainties quoted to 2 ). Presented here are the first direct measurements of the FAGE cali- bration constants as a function of external pressure (440- 1000 mbar) in a controlled environment using the Univer- sity of Leeds HIRAC chamber (Highly Instrumented Reac- tor for Atmospheric Chemistry). Two methods were used: the temporal decay of hydrocarbons for calibration of OH, and the kinetics of the second-order recombination of HO2 for HO2 calibrations. Over comparable conditions for the FAGE cell, the two alternative methods are in good agree- ment with the conventional method, with the average ra- tio of calibration factors (conventional : alternative) across the entire pressure range, COH(conv)=COH(alt)D 1.19 0.26 and CHO2.conv/=CHO2.alt/D 0.96 0.18 (2 ). These alterna- tive calibration methods currently have comparable system- atic uncertainties to the conventional method: 28 % and 41 % for the alternative OH and HO2 calibration methods respectively compared to 35 % for the H2O vapour photoly- sis method; ways in which these can be reduced in the future are discussed. The good agreement between the very differ- ent methods of calibration leads to increased confidence in HOx field measurements and particularly in aircraft-based HOx measurements, where there are substantial variations in external pressure, and assumptions are made regarding loss rates on inlets as a function of pressure.