Carbon Dioxide Detection 1

[ventolin21]

Hi All

I'm doing a project on carbon dioxide detection in the human breath. I'm considereing using infrared source and detector(thermopile) for the detection proces.

Is any any out there who can provide me with some help?

 

[fred29653]

IR is the way to do it.
The wavelength is 4.25u and you will need to use a dual wavelenth or similar technology to eliminate interference from water vapour in the exhaled breath.
A small lamp bulb will make a good IR source.
Murata make a pyroelectric detector with a built in CO2 band filter, but there are others.
The Murata device is an AC only detector so you will need to use a flashing IR source.
If you need fast response you might have to consider using a mechanical shutter to interrupt the IR source, if speed is not critical then just flash the bulb at about 1 Hz.
CO2 absorbtion of IR is not linear so be prepared to do some curve straightening.
Good luck

 

[ventolin21]

Hi Fred

1. Will a Dual Channel Thermopile Detector do?

2. I realise that i need some kind of chopper(mechanical) wheel to modulate the source but i really do not know what are the components required to build the mechanical shutter.

3. Are there any specific light bulb? as in the model number i can get?

 

[IRstuff]

Only detection, no quantification?

Without some sort of control system, aren't you going to wind up measuring the atmosphere?

TTFN

 

[ventolin21]

hi IR

So what do i do about quantification. Currently i've having lots of difficulties even getting the correct components for the detection cicuitry.

1. The emitter Source(Incandescent Source).
2. The appropriate Detector. currently i'm using a TPS2534 provided by perkin elemer and i'm not really sure if it is the right way to go about.

So care to advice what components i require? before i hit into the quantification bit...

 

[IRstuff]

Something like mass spectroscopy on samples carefully collected without contamination from the atmosphere would seem to be the most reasonable approach.

Other possibilities:
gas chromatography
infrared spectrometry
chemical precipitation

TTFN

 

[ventolin21]

yes ir

I am using infrared spectroscopy .....but what components are required for the detection which is what i'm not sure about...

 

[CyberPrestidigitator]

I was hoping that you could help me with a little mystery we're having with an IR Camera. We are using IR to measure respiration and, in hopes of increasing contrast, have installed a bandpass filter to block all but the wavelengths charicteristically blocked by CO2 absorption. When the breath is contrasted against the person, we see the breath as a region of decreased temperature - as would be expected. However, when the subject turns to the side, and the respiration is measured against a wall. The breath is seen as a region of increased temperature. Which is rather unexpected. Theories here abound, from water-vapor interference to background noise to camera sensitivity limits. Any thoughts that you might add to the pool are greatly appreciated.

 

[IRstuff]

"decreased temperature - as would be expected"

I'm not convinced that this is "as would be expected." The net signal of an IR detection is a combination of transmission and emission.

The CO2 absorption band does not imply zero transmission, particularly at short ranges, as might be configured in a lab. MODTRAN shows a 26% transmission at 4.255 microns at a distance of 4 meters.

Therefore, your subject's warm breath against a cold wall will appear somewhat warmer. The same subject's warm breath against a warm body may or may not appear colder.



TTFN

 

[CyberPrestidigitator]

we are indeed seeing extrema at both locations. The breath is measurable as variations against the backgrounds. The problem comes in the explanation for the polarity of the extrema. I expect that the filter would be non-ideal, But I question whether or not we are measuring CO2 at all giving the low CO2 content in breath (2-3% ?). From what I have come across, water-vapor is much more likely to be causing the variations in temperature. But I still don't understand why I would be getting two different results.

 

[IRstuff]

Why would you expect otherwise?

How tight is your bandpass filter? Don't forget that if your band pass filter is external to the dewar, it's emitting and reflecting as well and that stuff reaches the detector. Did you move the camera to do the different measurements?

The CO2 spike at short ranges is VERY narrow. The same MODTRAN run shows transmissions higher than 60% at both 4.167 and 4.348 microns at the same range. That means that your measured signal is a weighted average:
~50%*background + ~50%*breath.

But, the radiance of breath or air, for that matter, is considerably lower than that of a solid material at the same temperature. The net effect is a net radiance that's warmer than room temperature, but colder than external body temperature.

TTFN

 

[CyberPrestidigitator]

the subject moves, but the camera does not. I don't see that the emissivity of the filter could change the results in this way. These observations are based on a single film clip in which the subject moves their head while breathing. While I can see that the breath would be cooler (avg) than the body, and warmer (avg) than the wall; the two temperatures are not the same. The breath in front of body is cooler than the breath against the wall.

 

[BrianR]

As fred29653 said, the CO2 band is centered on 4.25u, while the 4.0u point has virtually no absorbtion. You could take a ratiometric measurement between those two bands but whai is also needed to measure the amount of CO2 is to measure the width of the 4.25 band and that needs a spectrometer. You can determine water vapour content by reading 5u and longer.