The integrated OH exposure (typical units: molec cm-3 sec), that is, the product of OH
concentration and residence time, is calibrated as a function of UV intensity and/or ozone concentration input to the PAM reactor, usually at fixed residence time and relative humidity.
Shown below are example time series from "OFR185" and "OFR254" OH exposure calibration experiments that used SO2 as a reactive tracer
species.
Here, the total flow through the PAM chamber is 10 L/min and the relative humidity ranges from 28-30% and 31-35% during the
OFR185 and OFR254 calibrations, respectively.
In the OFR185 calibration, the dimming voltage applied to the ozone-producing UV lamp ballast is stepped from 0-10 VDC, which has the effect of varying the
UV irradiance and the ozone mixing ratio in the PAM chamber. Irradiance and ozone mixing ratio are positively correlated in the "OFR185" mode of operation.
In the OFR254 calibration, the dimming voltage applied to the ozone-free UV lamp ballast is stepped from 0-10 VDC and the dimming voltage applied to the
ozone chamber lamp ballast is maintained at 10 VDC. Thus, in this mode, irradiance and ozone mixing ratio measured downstream of the PAM chamber are negatively correlated (the input ozone
mixing ratio is constant).
The lamps are turned off about halfway through and at the end of the experiment to get multiple measurements of unreacted SO2, which can then be
averaged or interpolated across each condition.
Note that because SO2 is sticky, it takes a relatively long time to reach steady-state levels when significant changes are made to the
conditions in the chamber - especially when the lamps are turned off. In general, it may be necessary to allow more time than the 5-min step changes shown here for illustrative
purposes.
The integrated OH exposure is calculated by measuring the decay of a reactive tracer
species X with known OH rate constant kOH, then applying the integrated rate law
of X, assuming pseudo-first order conditions with respect to OH,
i.e. depletion of OH is negligible from reaction with X:
Below is a plot of OH exposure as a function of irradiance. This calibration curve is obtained from averaging the last
~30-60 seconds of data at each condition in the time series above, then applying the OH exposure equation shown. The fit parameters can be loaded into the PAMControls software to apply the
calibration data in real time.
Tracer species that have been most commonly used to calibrate the OH exposure in the PAM reactor include SO2 and CO.
At 298K and 1 atm, kOH+SO2 ~
9.4*10-13 cm3 molec-1 sec-1 and kOH+CO ~
2.3*10-13 cm3 molec-1 sec-1 (JPL Evaluation
15)
SO2 is most accurate for calibrating OH exposures ranging from approximately 1*1011 to 2*1012 molec cm-3
sec.
CO is most accurate in calibrating exposures ranging from approximately 5*1011to 1*1013molec cm-3sec.
Li et al. (2015) (attached) provides a detailed discussion of choosing appropriate tracer species over a range of
calibrated OH exposure conditions. In general, measured OH exposure values
are most accurate in the range of
