The hydroxyperoxy (HO2) and organic peroxy radicals (RO2) have a strong influence on the ozone production in the troposphere under polluted conditions. They have very low concentrations because of their high reactivity.
The chemical amplification technique (CA) offers a relatively light weight, low cost approach to the measurement of the sum of HO2 and RO2.
In this study tests were performed in order to characterize and improve
a photolytic calibration source, which is based on the photolysis of H2O
at 185 nm in the presence of air, producing HO2 and OH radicals.
Critical aspects of the source and its optimisation will be discussed.
Once sampled the peroxy radicals are converted into NO2 in
the inlet via a chain reaction involving CO and NO which are added continuously
to the mixture [1]:
The chain reaction is terminated by radical radical reactions and radical
wall losses:
The air sample is pumped through a home made NO2 luminol detector at a constant flow rate of 2 sLmin-1, where a photomultiplier tube is used to detect the chemiluminiscence produced from the reaction of NO2 with a luminol solution on a glass fibre filter paper. In order to assure that the sampled concentrations are in the linear response regime of the NO2 detector, the output of a NO2 permeation tube is continuously added prior to the detector.
A 14 points NO2 calibration having concentrations in the range 0 to 150 ppb, obtained by diluting the output of NO2 permeation tubes kept at 40°C, is used to determine the NO2 sensitivity.
The ambient NO2 and that produced from the reaction of ambient O3 and NO at the inlet can be distinguished from that produced in the chain reaction by alternating the CO addition point in the inlet. The measurement cycle has typically a 60 second period. To establish stable flow conditions N2 is added at the other addition point (figure 1).
When the CO is added at point 1 the chain reaction takes place while
in the background mode (CO at point 2) only ambient Ox (NO2
+ O3) and PAN is measured. PAN indicates the thermal decomposition
of PAN in the reactor.
The chain length (CL) of the ROx detector is defined by the
expression:
The source for peroxy radicals is based on the photolysis of water at
185 nm wavelength [2]:
The OH produced at the source is converted to HO2 by adding
small amounts of CO to the source. The photolysis of molecular oxygen at
185 nm wavelength between lamp and glass tube decreases the light flux.
By changing the position of the penray lamp different light intensities in the source and therefore different amounts of radicals result.
The O3 also produced inside the glass tube is measured simultaneously
with RO2 and is used as an actinometer in the determination
of the amount of radicals.
The detector samples in the center of the source in order to minimize radical losses due to wall reactions.
A glass frit is built in the source to reduce the length of the tube
required for laminar flow conditions (figure 2). Gas flows are controlled
by electronic flow controllers and mixed in a teflon T-piece. Distilled
water is added to a mixing chamber by a peristaltic pump.
The CL of the CA detector was measured using the calibration source.
The amount of NO added to the CA was varied. Figure 3 shows the behaviour
of the determined CL as a funktion of the NO concentration.
By adding CH4 instead of CO to the source, CH3O2
radicals are produced in a mixed source:
Similar CLs are obtained for both HO2 and HO2+CH3O2 mixture for all NO concentrations used.
The differences in the CLs are expected to come from radical losses
in the inlet and differences in sensitivity of the detector for CH3O2
and HO2. Considering that wall losses are expected to be more
important for HO2 than for CH3O2 radicals,
a similar CL for the pure and mixed source indicates an adequate sampling
point with minima losses of radicals.
The profile of O3 in the tube has to be determined in order to select a sampling point having constant O3 concentrations. At the highest lamp intensity the O3 profile was measured through a glass tube (4 mm id, 8cm length) by a conventional photometric UV ozone monitor (figure 4). The radical and O3 profiles are assumed to be similar because both are produced by photolysis.
Due to the laminar flow profile the concentration increases near to
the walls of the tube. On the left side of the tube nearer to the lamp
high ozone gradients are observed. In figure 4 the nearly homogeneous region
used for radical calibration is shown.
Thus the calculated radical amounts may be somewhat incorrect. First tests indicated that an error is likely to be small.
Part of this work was supported by the Senator für Bildung und
Wissenschaft of Bremen and the EC.
[1] Hastie, D.R. et al., Calibrated Chemical Amplifier for Atmospheric ROx Measurements, Anal. Chem., 63, 1991
[2] Schultz, M. et al., Calibration source for peroxy radicals with built-in actinometry using H2O and O2 photolysis at 185 nm, J. Geophys. Res., 100, 1995
[3] Hofzumahaus, A., private communication, 1997
All text on this server is © 1997 Institute
of Environmental Physics, University of Bremen, Germany, and may not
be quoted, copied or used without permission.
Page created by Dirk
Stöbener, last change: 1997/12/20.