Transcript The nitrogen dioxide sonde

The nitrogen dioxide sonde
W.W. Sluis, M.A.F. Allaart, A.J.M. Piters, L.F.L. Gast
See: Sluis et al., AMT, 2010
Content list
•
•
•
•
•
•
•
•
Why an NO2 sonde?
Characteristics
Design of the instrument
The detector
The luminol solution
Calibration
Profiles
Future development
Why an NO2 sonde?
• Air quality
monitoring with
high vertical
resolution.
• Satellite and AQmodel validation.
• Alternative to
scattered light
measurements.
• Can be applied in
an operational
network.
http://www.temis.nl/airpollution/no2.html
Characteristics
• Measurement range
1 – 100 ppbv.
• Fast (1 second)
• Light weight (0.7 kg)
• Little power
• Cheap (disposable)
• Not dangerous to
finder / environment
Design of the instrument
The detector
The circuit has been designed to convert femtoamperes from the
photodiodes to millivolts at the output.
From left to right:
8 photodiodes  Current to voltage converter  a low-pass RC filter, with time
constant = 1 sec.  A times 1 amplifier, to reduce the output impedance.
The luminol solution (1/2)
• Luminol: The active compound  reacts with NO2.
• Potassium hydroxide: To solve luminol in water. First step of the reaction.
• Sodium sulfite: Removes ozone, sulfur dioxide, and PAN interference.
• Ethanol: Increases the light signal.
• Sodium EDTA: In combination with Triton X-100, it increases the function of
sodium sulfite.
•Triton X-100: Lowers the surface tension of water, thereby enhancing the contact
surface between air and liquid.
• Oxygen
The luminol solution (2/2)
The chemicals are optimized with respect to the light signal.
This is done with an iterative process by varying the concentration of one component
while holding the concentrations of all the other compounds constant at their optimum
values, as found in a previous iteration step.
Calibration
Applied corrections sonde data:
• Acidification of the luminol solution. The sonde can
measure for 2.5 – 3 hours (pH course 12.5 – 11.6).
Signal variation approximately 10%.
• The temperature dependence of the luminol solution is
3% per degree Celsius.
• Temperature dependence of the dark current of the
detector.
• Sensitivity of the luminol solution appears to increase
with age (still under investigation). Need for absolute
calibration before each launch.
Comparison of the NO2 sonde
with an in-situ monitor
•
Special thanks to Lou Gast, RIVM
Winter 2010/2011 profiles
Future development
• Investigating the effect of different treatment of the
solution.
• Finding the best procedure for absolute calibration.
• Making the signal reproducible within about 10%.
• Better characterization of the sensor at low pressure,
e.g. pump efficiency.
• Investigating possible impact of other atmospheric
compounds.
• Starting cooperation with GRAW radiosondes
(Germany) for possible future production.
The end.
Questions?