QASoft Application Examples

Infrared Study of Breath and Breathing


Trace gases in the breath can be measured by their infrared absorption spectra. Water and carbon dioxide are the main molecular species added to air in the lungs. This is a bit of a problem, because the spectra of these two molecules interfere with the measurement of the trace gases that are indicators of illness.

In the present study, Infrared Analysis, Inc. has addressed the following problems:


(1) Removing the absorption features of water and CO2 from the infrared spectra of breath samples and air samples.

(2) Pushing the minimum detection limits of the trace gases down to the parts-per-billion level.


(3) Measuring gases that the lungs add to the air.

(4) Measuring gases that the lungs remove from the air.

This is not a study that attempts to link breath components to illness. It is only a study of breath analysis technique.


Figure 1. is a picture of the apparatus used. A multiple pass absorption cell was used at a total optical path of 28 meters. This cell is mounted in the sample compartment of the FT-IR spectrometer. On the table in front of the spectrometer are gas handling components: (1) a tube to introduce 10 percent nitrogen into the air and breath samples, (2) a tube to humidify nitrogen for preparation of water subtraction spectra, and (3) gas syringes for introduction of CO2 into nitrogen. The breath sample bag is hanging down in front ot the table. The vacuum pump is on the floor in front of the nitrogen tank.

The operating procedure of this study was as follows.

1. The spectrometer was set for one wavenumber operation with 256 scans per spectrum.

2. The 28 meter cell was filled with nitrogen.

3. A background spectrum was recorded through the nitrogen. Vacuum was not used for the background because the vacuum may shift small interference fringes in the spectrum and then they may not cancel out against the interference fringes of the air and breath spectra.

4. The cell was evacuated, and nitrogen from the tank was then passed through the wetting tube and into the cell. To prevent condensation of water on the mirror surfaces and windows, a small portion of dry nitrogen (10% of the cell volume) was carried into the cell in front of the wet nitrogen.

5. The spectrum of the wetted nitrogen was recorded.

6. The cell was pumped out again and nitrogen was used to carry CO2 gas into the cell to about six percent by volume.

7. The spectrum of the CO2 in nitrogen was recorded.

8. These three spectra—dry nitrogen, wetted nitrogen, and CO2 in nitrogen—were set aside for future use. Presumably, they could be used in the processing of breath samples from many individuals.

9. Room air was placed in the cell and its spectrum was recorded.

10. The sample bag was filled with breath, and this sample was placed in the cell, carrying the 10% portion of dry nitrogen ahead of it. The breath spectrum was recorded.

11. The spectra were processed to provide an absorbance spectrum of the breath with the room air spectrum as background and also to provide absorbance spectra of the water and CO2 samples with the dry nitrogen spectrum as background. The QASoft spectrum processing program includes a feature that removes the lines of CO impurity that may be in the nitrogen.

12. The concentrations of trace gases were measured by the method of QASoft.

Here are some questions and answers pertaining to the measurements.


Q.       How do you get the noise level in the spectrum low enough to see parts-per-billion concentrations of trace gases in the breath?


A.       (1) Have the subject fill a good-sized bag with breath so that a fair-sized long path cell with good throughput may be used.


(2) Use the best possible infrared detector, such as narrow-band mercury-cadmium-telluride. See discussions in the book PROCEDURES IN INFRARED ANALYSIS OF GASES, available in the website:


(3) Use as many interferometer scans as is practical for all spectra–background, room air, breath, CO2 reference and H2O reference.


Q.        How do you ensure a clean subtraction of the water and CO2 lines?


A.       Make the water and CO2 subtraction spectra under exactly the same instrumental conditions as exist during the recording of the breath spectra. See discussions in the book PROCEDURES IN INFRARED ANALYSIS OF GASES available in the website:


Q.      What method of quantitative analysis do you use?


A.        We use RIAS—Region Integration and Subtraction. This works within the QASoft program of Infrared Analysis, Inc. and it is an automation of the traditional band area method of measurement. After each component is measured, its spectrum is subtracted from the sample spectrum. The sample spectrum is thus reduced to a blank, one compound at a time. A sequence of measurements may be run automatically. For details, see discussions in the book PROCEDURES IN INFRARED ANALYSIS OF GASES, which is readily available in the website:

The Spectra and the Results of their Manipulation.

Figure 2 shows a portion of the breath spectrum plotted in absorbance form, using the room air spectrum as background. The upper spectrum is what appears before removal of the water and CO2 lines, while the lower spectrum is the residue after the lines are subtracted away.

breathfig5.jpgFigure 2. Upper - Absorbance spectrum of breath with room air as background. Lower - the same spectrum with water and CO2 lines subtracted away. The absorbance scale is the same for both cases.


Figure 3 shows the residual spectra of breath (upper) and air (lower) after water and CO2 lines have been removed. Methanol and isoprene are identified as having been introduced into the breath sample by the lungs, and ammonia is identified as having been removed from the air sample.



breathfig4.jpg Figure 3. Spectra with water and CO2 lines removed. Upper - Breath; Lower - Room air.











Figure 4 shows the methane in the breath and the methane in the room air. Note an approximate six-fold increase in concentration. Note also in the upper spectrum the absence of the C-H band of non-methane hydrocarbons.



breathfig2.jpgFigure 4. Methane region. Upper - Methane in breath, with air methane subtracted away. Lower - Methane in room air. Same absorbance scale for both. The lower spectrum is displaced for the presentation.




Figure 5 shows the bands of N2O and CO in the room air and in the breath. It also shows the difference of the two spectra. Note that the difference spectrum shows that both N2O and CO have been added to the air in the lungs.


breathfig3.jpgFigure 5. N2O and CO. Lower - Room air; Middle - breath; Upper - Breath minus room air.

            Same absorbance scale for all three spectra.





Five groups of compounds have been encountered.


(1) Compounds illustrated here that have been seen in the spectra of breath samples at concentrations exceeding those seen in room air. These are water, carbon dioxide, methane, carbon monoxide, nitrous oxide, methanol and isoprene.


(2) Compounds that have been seen in the breath spectra, but are not illustrated here: ethanol and acetone.


(3) Compounds frequently seen in room air, but which have been seen to be reduced in the breath, apparently because of absorption in the lungs: acetone, methanol, ethanol, and ammonia.


(4) Compounds seen at low concentrations in room air and at about the same concentration in the breath: hydrocarbon vapors.


(5) Compounds of interest that should be detectable at the parts-per-billion level, but that have not yet been seen: ethane, nitric oxide.


Here is a tabulation of the concentrations measured for the gases illustrated here.


Water: 100 percent humidity at body temperature.


Carbon dioxide: about six percent by volume.


Methane: 10.2 parts-per-million added in the lungs.


Carbon monoxide: 1.02 parts-per-million added in the lungs.


Methanol: 0.26 parts-per-million added in the lungs.


Isoprene: 0.09 parts-per-million added in the lungs.


Nitrous oxide: 0.14 parts-per-million added in the lungs.


Ammonia: 0.016 parts-per-million removed from the air in the lungs.



This study has shown that when samples and reference gases are handled appropriately and their spectra are properly manipulated, information on breath composition is available in the infrared spectrum. Possibly, qualified medical researchers might pursue the study further with breath measurements on patients that have physical disorders.






For additional information on the measurement technique used in this work and on the RIAS analytical procedure of QASoft, see the book PROCEDURES IN INFRARED ANALYSIS OF GASES, which appears in the website


For other literature references, search the web under the topic BREATH ANALYSIS.

Return to Table of Contents