University of Applied Sciences Jena SciTec Scientific Instrumentation
FTIR Gas Sensing and Aerosol Measurement Lab Course
Tina Bischof (631450) Wazeem Basheer Karunnapallil (636784) Amal Raj (636785) August August 5, 2013
Grou Group p numb number er:: 10 Experimental Experimental Proced Procedure: ure: April 9, 2013 (Part 1) 1) and May May 4, 2013 2013 (Part 2) Hando Ha ndover ver Date: Date: ??? ??? Valuation: Date of Valuation: Signature:
FTIR
Part I.
Qualitative and quantitative calibration 1. Procedure At first we use the program “MOLSPEC” to generate automatically spactral lines of CO, CO2 and NO with a data base / calculation software. These spectral lines show the dependence of the absorbance on the concentration of the gas. The absorbed intensity was set on the y- axis while x- axis displays the wavenumber. After this we measure the spectral lines / absorbance of a test gas with the help of the FTIR by using the program “Essential FTIR”: 1. measurement of background (air in cuvette) 2. evacuation of cuvette (pressure
≈
0 psi)
3. measurement of sample (pressure
≈
15psi =ˆ 1034.2 mbar)
The absorbance was measured for di ff erent concentrations of the test gas / sample. The concentration was changed by using a gas divider (see table 1). The measurement of the sample takes place in the following way: The percentage of the test gas was set to one of the shown values, the spectral lines were measured and after this the cuvette was evacuated. Now the procedure begins again by setting a new value. The maximum of the concentration of the test gas is 90 ppm. The gas flow was set l to 3.5 min . Table 1: Gas divider percentage and concentration measurement number 1 2 3 4 5 6
percentage at gas divider (%) 100 80 60 40 20 0
T. Bischof, W. B. Karunnapallil and A. Raj
concentration (ppm) 90 72 54 36 18 0
2
FTIR
2. Expectation 2.1. Spectral lines of CO , NO and CO 2 The absorbance depends on the number of particles which absorb the incoming radiation. By increasing the concentration the absorbance increases too. Therefore we expect that a higher concentration of the gas leads to higher absorbance.
2.2. Qualitative and quantitative calibration For these low concentrations of the test gas parts we expect a linear behaviour of the calibration function.
3. Discussion 3.1. Spectral lines of CO , CO 2 , NO and H 2 O 3.1.1. Carbon Monoxide (CO) spectrum
Figures 1 and 2 show the absorption spectra of CO with di ff erent concentrations and the axis of the graph plotted wave number verses absorption. From the graph, we can quintessentially say that absorption increases with increase in concentration. In this experiment we used di ff erent concentration of CO (0.1 and 0.000001) and absorbance was observed at the same frequency range from 2090 cm 1 to 2200 cm 1 −
−
Figure 1: Carbon Monoxide: c = 0.000001
T. Bischof, W. B. Karunnapallil and A. Raj
3
FTIR
Figure 2: Carbon Monoxide: c = 0.1 3.1.2. Carbon Dioxide (CO2 ) spectrum
Wave numbers 2300 cm 1 to 2380 cm 1 represents the absorbance of CO 2 (see figures 3 and 4). At frequency 2370 cm 1 , it shows maximum absorbance. Like carbon monoxide the absorbance property of carbon dioxide also increases with increase in concentration. −
−
−
Figure 3: Carbon Dioxide: c = 0.000001
T. Bischof, W. B. Karunnapallil and A. Raj
4
FTIR
Figure 4: Carbon Dioxide: c = 0.1 3.1.3. Nitrogen Oxide (NO) spectrum
Absorbance frequency range of nitrogen monoxide is from 1800 cm 1 to 1950 cm 1 (see figures 5 and 6). For 0.01 concentration of (NO), we got maximum absorbance peak value at 902.4789 at frequency 1870 cm 1 . −
−
−
Figure 5: Nitrogen Oxide: c = 0.000001
T. Bischof, W. B. Karunnapallil and A. Raj
5
FTIR
Figure 6: Nitrogen Oxide: c = 0.1 3.1.4. Water (H2 O) spectrum
For H2 O, maximum absorbance takes place in the frequency around 1700 cm 1 . Absorbance properties of H 2 O increases with increase in concentration and vice versa (see figures 7 and 8). −
Figure 7: Water: c = 0.000001
T. Bischof, W. B. Karunnapallil and A. Raj
6
FTIR
Figure 8: Water: c = 0.1 3.1.5. Summary of Spectral Behaviour
When the concentration was increased by a factor of 100 the corresponding absorbance also increased by a factor of 100. From this we infer that concentration of the element is directly related to absorbance. For instance, concentration of CO was initially set to 0.000001 and maximum absorbance was observed at 0.0602 and when the concentration was increased by a factor of 100, maximum absorbance was observed at 6.023. Table 2: Frequency Range of Absorption in cm Element CO CO2 H2 O NO
Lower Range 2090 2300 1350 1800
T. Bischof, W. B. Karunnapallil and A. Raj
1
−
Uppeer Range 2200 2380 1900 1950
7
FTIR
3.2. Qualitative and quantitative calibration 3.2.1. Background Spectrum
This graph shows the background spectrum and this is subtracted from the sample spectrum to get the final result.
Figure 9: Background spectrum
3.2.2. Measured spectra
Calibration is an essential step in all the analytical experiments; here in FTIR experiment also we do the calibration. For the purpose of doing the calibration we analyzed the area of the entire spectrum of the gas mixture containing CO, CO 2 , NO and H2 O by varying the concentration from 100 % to 0 % (of 90 ppm mixture). 100% corresponds to air + test gas and 0% is just the air. The figure 10 show the spectral ranges of CO2 , CO and NO. As the concentration of test gas is reduced the absorbance decreases.
T. Bischof, W. B. Karunnapallil and A. Raj
8
FTIR
(a) c = 40%
(b) c = 80%
(c) c = 100%
Figure 10: Measured spectra
T. Bischof, W. B. Karunnapallil and A. Raj
9
FTIR Following are the results obtained after the area analysis. Area1 shows the measured the area of the curves for the element specific wavenumber range. Area2 was calculated by measuring the area of the 2 highest peaks of the curve of NO. The graph 11(a) and table 3 represent the level of absorbance for corresponding increase in the concentration of the gas mixture. By closely studying the graph and the table it is quite evident that CO shows a linear response when compared to NO. So, we came to the conclusion that CO can be used for the calibration process in the experiment. The linear response can be represented by: c( A) = m · A + n
(1)
m = 8.799
n = 67.6
The graph 11(b) and table 4 are the result of analyzing two peaks corresponding to NO, while studying the graph it’s again clear that NO cannot be used for calibration as it doesn’t give a linear response. Theoretically, NO should also give a linear response, but due to some factors like no constant temperature, change in humidity, dust present at mirror and beam divider it doesn’t show this behaviour. Table 3: Area1 concentration (ppm) 90 72 54 36 18 0
CO 3.3618 -0.5203 -1.9039 -3.9924 -4.9876 -7.4033
NO 1.6311 -5.4167 -9.6252 -9.3359 -11.6612 -11.6408
Table 4: Area2 concentration (ppm) 90 72 54 36 18 0
T. Bischof, W. B. Karunnapallil and A. Raj
NO Peak 1 0.6084 -1.2372 -1.3064 -1.0465 -1.0612 -2.1617
NO Peak 2 0.338 -0.1934 -0.0106 0.3757 0.3272 0.719
10
FTIR
(a) Area1
(b) Area2
Figure 11: Qualitative and quantitative calibration
T. Bischof, W. B. Karunnapallil and A. Raj
11
FTIR
Part II.
Calculation of concentrations 4. Procedure The aims of this part are the emission- measurement and the calculation of the concentration of an exhaust gas from a duct of a combustion engine. At first it is necessary to measure the background spektrum which is subtracted from the sample spectrum to get the final result. Therefore the cuvette was evacuated with a pump up to 0 psi. Then pure air was filled into it (14 psi) and the measurement was started (software Essential FTIR). The next step was to measure the exhaust gas. After the engine was switched on we wait for a constant concentration of CO at the measuring device Siemens Ultramat. Then the flow of air and gas into this device was adjusted at the dilution machine to get a defined concentration of CO. This mixture of the gas and air was inserted into the cuvette. After some minutes the measurement of the spectrum was started by using the software Essential FTIR. This procedure was repeated with another defined concentration of CO.
5. Discussion 5.1. Background Spectrum This graph shows the background spectrum and this is subtracted from the sample spectrum to get the final result. While comparing the background 1 (see figure 9) and background 2 (see figure 12) some diff erences could be found this is due to errors while measuring and error during calculation of area. The errors are due to humidity, in homogeneous emission of IR source, change in laser modes, and no constant compound of test gas.
Figure 12: Background spectrum Due to the insufficient concentration of NO and NO2 in the gas mixture is the wavenum ber range of 1800 m 1 - 1950 m 1 not considered. Therefore the measured wavenumber range was set to 2000 m 1 - 2500 m 1 . −
−
−
−
T. Bischof, W. B. Karunnapallil and A. Raj
12
FTIR
5.2. Comparison
(a) c = 55 ppm
(b) c = 127 ppm
Figure 13: Measured Spectra Table 5: Measured values of the concentration of CO concentration (ppm) 127 55
Baseline 1.98 1.14
Height of Peak 0.139 0.083
The wavenumber of the measured peak equals 2110.41 m 1 . The wavenumber range for the area is 2042 m 1 to 2225 m 1 . Substituting the baseline values at 127 ppm and 55 ppm in the equation1 we found that the calculated and the theoretical values are di ff erent. Calculated Value: −
−
−
For 127 ppm: Baseline value: 1.985 Calculated concentration: 85 ppm For 55 ppm: Baseline value: 1.7103 Calculated concentration: 82 ppm We infer that this di ff erence in the calculated value is due to errors while measuring and error during calculation of area. The errors are due to humidity, in homogeneous emission of IR source, change in laser modes spectrum, no constant compound of test gas and gas flow of CO2 from air into measurement system.
T. Bischof, W. B. Karunnapallil and A. Raj
13