How to measure gases

[chrisxxx]

Hi All,

I'm a nubie at optics, but I would like some advice if possible on how to analyse the composition of an airflow. i.e. O2,CO2,NH4,CH4 etc. I am interested in this for horticultural purposes. Is it possible to measure these using some kind of optical device and a broadband frequency source (without spending a fortune)?
I don't mind spending time on research, but I would appreciate some guide as to the right direction.
I am a software/hardware engineer and I would have no problem with the analysis of information from a CCD or similar. I was just interested to see if there were any DIY designs out there!

Thanks Chris

[VDX]

... look here - http://messkonzept.de/e_index.html

The company sells gas sensors, which measures concentrations of two (or more) different gases in a mixture by super precise measuring the thermal conductivity of the mixture and calculating the composition with this values ...

Viktor

[mixedgas]

This is done with differntial path Mid band IR. The cost of parts, ie halogen source, mid-iR bandpass filters, etc takes it out of the range of the hobbyist.
Steve

[VDX]

... the thermal conductivity sensors I've linked are based on thin silizium membranes, coated with Nickel or Platinum, heated and measured to get the gas mixture numbers.

Some years ago I've developed another type of TC-sensor with 10 nanometer thick platinum wires, assembled to ressemble another type of gas-sensor - similar to the silizium type, but more in the DIY-range ;-)

So there are different methodes to get the values ...

Viktor

[chrisxxx]

Thanks VDX,

However, the airflow contains ammonia which is not very good for sensitive components.
Also there are dust and spores which would probably cause a problem.

I suppose the only way would be to use an optical path to measure the absorption of radiation.
I cant think of any other method.

Chris

[mixedgas]

Most machines for high percentages of trace gasses use a reference beam path with pure air, a second path with the sampled air, a mid-IR source, two detectors and a filter wheel. co2 is easy, you need no filter wheel for that, just a bandpass filter and a lock in amplifier.

I get asked this question about once every three months. I usually do not like the reason I get asked this. Do you work on commercial plant monitoring systems?


Steve

[VDX]

... dust-filtering is common for this sensors - made mostly with sintered ceramic filter-plates.

And for the ammonia - the most aggressive gases were hot NOx mixtures, which eroded the polyimide-silver conducting glue, but didn't touch the platinum.

I've 'toughened' the connections by replacing the thermoset conducting glues with a DIY mixture of eutectic gold-tin microspheres and gold dust in dexpanthenol, which I've 'brazed' with my first IR-laserdiode setup ... this was the main reason, I've started to develop diode-laser setups ;-)

Viktor

[chrisxxx]

Thanks again both of you.
I have worked as a software engineer in the horticultural industry. I have used many different types of sensor systems e.g. Temp/RH/CO2/O2/CH4/NH4 etc. but it is always difficult
to find a sensor that will measure gases that contain lots of contaminants. A friend has asked me to design an O2 analyser which can be left in the airstream for long periods of time without complicated
filtering systems.
For instance, to measure O2 levels in compost, the best sensor available currently is the Zirconium based O2 sensor with a water bubble filter to remove the ammonia (a by-product of the composting process) These sensors and the associated electronics are quite expensive and I keep thinking 'there must be another way'!!!
Also the airflow (even after the water filter) contains all sorts of fine dust and other contaminants which slowly cause the sensor to fail, especially if the water filter stops working!
I'm working on this partially as a hobby, because I find the horticultural aspect quite fascinating.
Its probably quite boring to most people, especially my wife!!!!
Chris

[planters]

If you limit your analysis to several gasses such as ammonia, methane and carbon dioxide (4.3um) What about obtaining several filters centered on the strongest line for each. You then set up a broadband IR (thermal) source and measure the intensity of each pass-band as the test mixture flows in front of each filter. You can then deconvolve only the real combination of gases that would produce the measured values. The maths are not trivial because each gas will probably absorb light at bands that are at the peak of another gas. The greater the number of filters the more burdensome the analysis, but the greater the range of gasses and the smaller the error factor. You could do this as well by building a scanning grating that scanned the source across a slit and then analyzed the absorbance of the light as it passed through the gas and on to the detector. You could calibrate with pure gasses of each type and confirm your system with a mixed (known blend) of gasses.

[mixedgas]

Usually four filters and a reference filter (no major gas adsorption for this band, serves as a intensity reference) plus a blank space are placed on a spinning wheel. The resulting waveform is sort of self decoding after that.

Steve