Fourier Transform Infrared (FTIR) Spectroscopy

Group 1 (Monday 1-5pm)

Author
Reshma Reji

Outline
Introduction
Objective of the Experiment
Theory
* FTIR Spectroscopy * FTIR Spectrophotometer * Samples

Experimental
Physical properties of reagents used
Solution Preparation
Procedure
Instrument settings

Data 1. IR spectrum of Chloroform and D-chloroform 2. Rotational Spectrum of CO2 (Standard Resolution) 3. Rotational Spectrum of CO2 (High Resolution) 4. Carbonyl stretch in 2-butanone solutions (wavenumber vs. % T) 5. Carbonyl stretch in 2-butanone solutions (wavenumber Vs Absorbance) 6. Calibration curve of 2-butanone solutions (concentration vs. absorbance) *
Calculations
a. Preparation of solutions b. Concentration of the unknown c. Percent error of observed and theoretical ratios of CH, CD stretch frequencies

Results and Conclusion

References

Objective The goal of the first part of the experiment was to study the effects of isotopes on bond stretching. In the second part of the experiment, the influence of instrument resolution on the rotational spectrum of carbon dioxide was studied. The purpose of the third part of the experiment was to create a calibration curve to find the unknown concentration of 2-butanone.
Introduction
Fourier Transform Infrared Spectroscopy (FTIR) is a very useful analytical technique used for qualitative and quantitative analysis of organic and inorganic compounds. It can be used for a variety of purposes such as; finding the concentrations of unknown solutions, determining the constituents of unknown solutions, forensic purposes or finding individual components in paints, drugs, fibers residues, contaminated chemicals etc. Solids, liquids and gases can be analyzed using this method. The most important benefit of FTIR spectroscopy is its ability to identify the different types of functional groups (CH, CD, C=O, CN, CF etc) in molecules. The chemical bonds in these functional groups correspond to the wavelengths (wavenumbers) of absorbed light and represent specific areas(peaks)of the obtained spectrum. The identification of different functional groups and their chemical bonds facilitate the characterization of molecules. The spectra obtained from pure compounds are highly distinctive and therefore helps to recognize molecular structures. FTIR spectra therefore serve as “molecular finger prints” for several compounds [6].
Theory
FTIR spectroscopy is used to record the infrared intensity at different wave numbers of light. Infrared light is classified into three regions; far infrared (FR), mid infrared (MI) and near infrared (NI). The wavenumbers range from 4 ~ 400cm-1 for FR, 400 ~ 4,000cm-1 for MI and 4,000 ~ 14,000cm-1 for NI. The infrared region in the electromagnetic spectrum can be seen in Figure 1. The Infrared spectrum is formed due to the transitions within the vibrational energy states that are quantized. FTIR spectroscopy detects the vibrations of the chemical bonds of the functional groups in the sample [8].

Figure 1: Electromagnetic Spectrum (IR region to the left of visible light region)

The number of degrees of freedom for a non linear molecule which consists of N atoms is 3N, of which three motions represent translational motions and another three correspond to rotational motions. The degrees of freedom remaining in the molecule is 3N-6, which represents the number of vibrational modes of the molecule. The number of ways that atoms can vibrate in the molecule is the number of its vibrational modes [1]. The frequencies at which chemical bonds vibrate depend on the elements and the nature of bonds. Vibrations of individual bonds can occur at different frequencies depending on the ground states and excited states of bonds. Molecular vibrations occur at lower frequencies for ground states and higher frequencies for excited states [6]. A ground state chemical bond can be brought to its vibrational excited state through the absorption of IR radiation. Chemical bonds can stretch, contract or bend when exposed to IR light. Absorptions due to stretching normally generate stronger peaks than bending absorptions. Each functional group in a molecule absorbs infrared light at a unique wavenumber range (or frequency). This does not depend on the structure of the remaining parts of the molecule. Different molecules may contain equivalent functional groups. However, the stretch of a particular functional group will always appear in the same wavenumber range (or frequency) in all molecules [1,4]. Some of the common functional groups and the frequencies at which they absorb can be seen in Table 1 and Figure 2.
Table 1: Common functional groups and their frequencies of absorption Functional Groups | Frequency (cm-1) | Wavelength (nm) | Alkanes, C-H | 2850 - 3000 | 3509 – 3333 | Alkynes, C≡C | 2100 - 2500 | 4762 – 4000 | Amines, NH, RNH | 3300 - 3500 | 3030 – 2857 | Ketones, C=O | 1690 – 1780 | 5917 – 5618 | Nitriles, C≡N | 2100 - 2270 | 4762 – 4405 |

Figure 2: Various functional groups and their frequencies and wavelengths of absorption.
Instrument
The Infrared instruments used in the earlier periods were dispersive types. In these types of instruments, specific frequencies of energy from the radiation were split using a prism or a grating. The detector measures the energy at each wavelength and yields a spectrum of intensity versus frequency. The main intricacy of the dispersive instruments was their slow paced scanning. This limitation was resolved by the development of an optical device called the interferometer. In a FTIR spectrophotometer, an interferometer is used instead of the regular monochromator [10]. Albert Michelson invented the Michelson Interferometer in 1880. However, sensitive detector arrays and computers for FTS were not available at that time and therefore the technique was not commonly used. Today, with the use of Fourier Transforming computers and highly sensitive detector arrays, FTS has now become a widely used technique for interferometric measurements [3].
Contrary to dispersive instruments, FTIR spectrophotometers can detect all IR frequencies at the same time. FTIR spectroscopy is preferred over the dispersive methods because of its high-speed, precision (does not need external calibration), high signal to noise performance, sensitivity and mechanical simplicity [10].
A schematic design of an interferometer is displayed in Figure 3. The main components of an FTIR spectrophotometer include an Infrared light source, interferometer, sample compartment, detector and the computer. Infrared light is obtained from a radiant blackbody source [10]. A small aperture regulates the amount of light that passes through the sample. The interferometer is an optical device that has the ability to detect all of the IR frequencies at the same time. The distinctive signal that the interferometer generates has all the frequencies of IR radiation “encoded” into it [9]. The interferometer splits IR radiation using a beam splitter into two beams. The radiation is reflected by the mirrors present inside the interferometer in such a way that the signal which passes out of the interferometer is a result of the interference of the two beams. The signal obtained is referred to as the “interferogram”. Every data point in the interferogram contains information about every beam that originated from the source. Therefore, when the interferogram signal is obtained, all the infrared frequencies are measured simultaneously [10]. However, the desired spectrum is not an interferogram, but a frequency spectrum. In a frequency spectrum, frequency is plotted against intensity. The interferogram is complex and cannot be analyzed the way it is. Therefore, the individual frequencies are “decoded” to obtain the desired spectrum. This process is accomplished by the computer using a famous technique called “Fourier Transformation”.

Figure 3: Schematic design of an FTIR spectrophotometer
Samples
Solid samples and liquid samples are prepared differently for FTIR analysis. For liquids, drops of liquid samples are injected in a dismountable liquid cell (Figure 4). The structure of the liquid cell prevents leakage and therefore the sample gets trapped between the salt plates present inside. The salt plates for liquid samples are generally made of sodium chloride (NaCl). The liquid cell is then placed in the sample holder of the spectrophotometer for analysis. For solids, the samples are initially crushed with potassium bromide (KBr). The powdery mixture is then made into a thin pellet and a great pressure is applied to dry it. This is placed in the sample holder and analyzed. An alternative method for solid analysis is to dissolve a small amount of solid in a solvent and place it in a salt plate. One example of a solvent that is normally used is methylene chloride (CH2Cl2). Evaporating the solvent leaves a thin film (also called as cast film) of the solid sample and is analyzed. This method is generally used to identify polymers [6,7].

Figure 5: A dismountable liquid cell

Salts are typically used in IR analysis because they are transparent to infra red rays and therefore do not absorb in the infra red region. In addition to NaCl, plates made of caesium iodide(CsI), silver chloride (AgCl) or Germanium(Ge) can also be used. However, the salts plates that are chosen depend on the nature of the sample that is analyzed. Salt plats made of NaCl, Kbr or CsI are commonly used for analyzing organic compounds. Plates made of non hydroscopic substances such as AgCl and Ge are preferred for samples containing water. A downside of using salt plates is the fact that they lose their transparency and appear to be fogged as they get exposed to moisture. The values obtained for transmittance in the IR spectrum rely highly on the condition of the salt plates. As the windows get exceedingly fogged, the signal strength decreases. Therefore, higher concentrations becomes necessary to augment the signal strength and offset the scattering [6, 7].

Experimental Table 2: Chemicals used Reagent | Formula | Structure | Molecular Weight(g/mol) | Density(g/ml) | Melting Point(°C) | Boiling point(°C) | CAS No: | Chloroform | CHCl3 | | 119.38 | 1.474 | -63.5 | 61.2 | 67-66-3 | Chloroform-D | CDCl3 | | 120.38 | 1.500 | -64.0 | 60.9 | 8665-49-6 | 2-butanone | C4H8O | | 72.11 | 0.804 | -86 | 79.64 | 78-93-3 |

Table 3: Preparation of Solutions Solutions | Weight % of 2-Butanone | 2-Butanone (ml) | 2-Butanone (l) | Chloroform (ml) | 1 | 0.05 | 0.0230ml | 23.0l | 25ml | 2 | 0.075 | 0.0343ml | 34.3l | 25ml | 3 | 0.10 | 0.0460ml | 46.0l | 25ml | 4 | 0.125 | 0.5720ml | 57.2l | 25ml | 5 | 0.150 | 0.0680ml | 68.0l | 25ml | The volume of 2-Butanone required to prepare each solutions were calculated using Equation 2. To derive Equation 2, the following ratio and Equation 1 were used (see below)
Mass of ButanoneMass of Chloroform=Weight% of Butanone100

Mass of 2-Butanone = (Density of 2 Butanone*Volume of 2-Butanone) Mass of Chloroform = (Density of Chloroform*Volume of Chloroform) Wt% of each 2-Butanone solution = 0.10%, 0.050%, 0.125%, 0.075%,0.30%

Equation 1:
Volume of 2-Butanone= mass of 2-butanone(g)Density of 2-butanone(gml)

Equation 2:
Using Equation 1 and the ratio,
Volume of 2-Butanone(ml)= Density of CHCl3*Volume of CHCl3*(wt% 2butanone/100)Density of 2-butanone

Wt% of each 2-Butanone solution = 0.10%, 0.050%, 0.125%, 0.075% and 0.30% Volume of CHCl3 used to prepare each solution = 25ml Density of CHCl3=1.474g/ml Density of 2-Butanone=0.804g/ml

Procedure
Part 1
In the first part of the experiment, two isotopes; Chloroform (CHCl3) and Deuterated Chloroform (CDCl3), were each scanned using a Fourier Transform Infrared (FTIR) spectrophotometer under standard resolution. To analyze CHCl3, several drops of CHCl3 were injected into salt plates made of sodium chloride. After obtaining the spectrum of CHCl3, the salt plate was rinsed with CDCl3. The spectrum of CDCl3 under standard resolution was also taken. Using the IR spectra of CHCl3 and CHCl3, the frequencies of the CH stretch and the CD stretch was determined. In the spectra, the CH stretch was observed at 3007.46 cm-1and the CD stretch was seen in 2250.53cm-1.The experimental ratio of these stretch frequencies was calculated.

Part 2
Background scans were performed to obtain the spectrum of carbon dioxide in air. The scans were taken under standard resolution and high resolution. The two spectra were compared.
Part 3
In the third part of the experiment, five solutions of 2-Butanone with varying concentrations (by weight %) were prepared in chloroform using the amounts listed in Table 3. The volumes of 2-butanone needed to prepare these solutions were calculated using Equation 2. All solutions were prepared in 25ml of chloroform. Initially a back ground spectrum was obtained. Using a syringe, the sample solutions were injected into a dismountable liquid cell (Figure 5) and placed into the sample holder of the spectrophotometer. Infrared spectrum of each solution was obtained under standard resolution.

Instrument
Fourier transform Infrared Spectrometer
Manufacturer: Mattson Instruments
Model: 4020 Galaxy Series
Settings
Part 1: Standard resolution (4.0cm-1)
Part 2: Standard resolution(4.0nm) and High resolution(0.4cm-1)
Part 3: Standard resolution(4.0cm-1) Data
Figure 6: Infrared Spectrum of Chloroform and D-chloroform (Absorbance vs. Wave numbers)

Figure 7: FT-IR Spectrum of Carbon dioxide under standard resolution

Figure 8: Rotational spectrum of Carbon dioxide under high resolution

Figure 9: Wavenumber vs. Percent Transmittance spectrum displaying the carbonyl stretch ( at 1710.561cm-1) in varied concentrations(wt%) of 2-butanone solutions.

Figure 10: Wavenumber vs. Absorbance Spectrum displaying the carbonyl stretch (at 1710.561cm-1) in varied concentrations (wt%) of 2-butanone solutions

Figure 11: Standard calibration curve of varying concentrations (wt %) of 2-butanone solutions, displaying the absorbencies of each solution at 1710.561cm-1

Lineast of concentration vs absorbance for Figure 11. m | 3.255041 | 0.153407 | b | Sm | 0.255667 | 0.027118 | Sb | R2 | 0.981828 | 0.020212 | Sy |

Calculations

1. Finding the volume needed to prepare 2-Butanone solutions:

Equation 2 was derived using the calculations shown in experimental section:

Equation 2:

Volume of 2-Butanone(ml)= Density of CHCl3*Volume of CHCl3*(wt% 2butanone/100)Density of 2-butanone

Wt% of each 2-Butanone solution = 0.10%, 0.050%, 0.125%, 0.075% and 0.30% Volume of CHCl3 used to prepare each solution = 25ml Density of CHCl3=1.474g/ml Density of 2-Butanone=0.804g/ml

Weight % | Volume of 2-Butanone(ml) | 0.050 | 0.0230ml | 0.075 | 0.0343ml | 0.10 | 0.0460ml | 0.125 | 0.5720ml | 0.150 | 0.0680ml | The calculated volumes of 2-butanone used to prepare solutions is shown in the table to the right.

2. a. Calculation of unknown concentration: y=mx+b y= absorbance of the unknown obtained from Figure 10 = 0.598637841 x= unknown concentration

Equation obtained from calibration curve(Figure 11),

y=3.255x+0.1534 0.598637841=3.255x+0.1534 x= 0.137 wt%
Concentration of unknown 2 butanone solution =x=0.137 ±0.0068 wt%

b. Uncertainty:
Sx = Sy|m| 1k + 1n + (y-mean y)²m²Σ(xi-mean x)² n=5 n=5 k=1, y= 0.598637841, mean y=0.478911 y-mean y2=0.598637841-0.478911²=0.0143345 m= 3.255
Σxi-mean x2=0
Sy= 0.020212
Sx = 0.020212|3.255| 11 + 15 + 0.01433453.255²*0
Sx = ± 0.0068 (%wt)

3. Percent error of observed ratio and theoretical ratio of CH, CD stretch frequencies

a. Observed Ratio of frequencies: νCHνCD = 3007.46 cm-12250.53cm-1= 1.336

b. Theoretical Ratio:
According to Hooke’s Law, frequency; ν=12πkμ Where, k= Hooke’s constant (equal for both CH and CD) µ = the reduced mass of the two molecules

Therefore theoretical ratio of frequencies is,
Equation A νCHνCD=12πkμCH12πkμCD Reduced mass of C-H and C-D, μCH=massCmassHmassC+massH=12.01(1.008)12.01+(1.008)=0.929949=0.9299 μCH=massCmassDmassC+massD=12.01(2*1.008)12.01+(2*1.008)=1.726234=1.726234
From Equation A, the ratio of Theoretical frequencies, νCHνCD=μCDμCH=1.7220.9299=1.36 Therefore Observed Ratio or Actual Ratio = 1.336 Theoretical Ratio= 1.36

Percent error = [ |observed - theoretical | / theoretical value ] x 100 = = [(1.336-1.36)/1.36]*100
=1.76%

Conclusion As mentioned in the objective, the goal of the first part of the experiment was to study the effects of isotopes on bond stretching. The two isotopes used in this experiment were chloroform (CHCl3) and D-chloroform (CDCl3). An FTIR spectrum of these two chemicals was obtained. In the spectra, the CH stretch was observed at 3007.46 cm-1and the CD stretch was seen in 2250.53cm-1.The observed ratio was calculated to be 1.336 and theoretical ratio obtained was 1. 36. The theoretical ratio is slightly higher than the observed ratio. However, the percent error between these rations is 1.76%. The very low percent error shows the relative accuracy of the FTIR analysis. An important aspect that affects the stretching absorption frequency of bonds is the identity of the atoms that are involved. The two atoms involved in CHCl3 and CDCl3 are the hydrogen atom (H) and the deuterium atom (D). Deuterium is a heavy isotope of hydrogen. Theoretically, the higher the masses of the atoms involved, the lower will be the stretching frequencies of their absorption [9]. This can be understood from the Hooks Law which states that frequency is inversely proportional to the reduced mass: ν=12πkμ
This analogy can be seen in IR spectra of CHCl3 and CDCl3 obtained from this experiment (Figure 6). Since deuterium has a higher mass than hydrogen, the CD bond absorbs IR radiation at a lower stretch frequency of 2250.53cm-1 compared to the CH stretch frequency of 3007.46 cm-1. Therefore, heavy atoms make their attached bonds to absorb IR radiation at lower frequencies [9]. Theoretically, the CH stretch in alkanes should occur at 2850-3000cm-1. The absence of the C-H signal around the 2850-3000cm-1 range in D-chloroform (Figure 6) makes the spectrum unique and illustrates the fact that the IR spectra can be used as a finger print for the identification of molecules.
In the second part of the experiment, advantages and disadvantages of using high and low resolutions is understood. The spectra of carbon dioxide in both standard (4cm-1) and high resolutions (0.4cm-1) were obtained. Spectral resolution is defined as the capacity of a sensor to define and identify signals at very small wavelength intervals [11]. The resolving power is also described as the capability of the spectrograph to “resolve” features in the spectrum [11]. The resolution plays a major role in the IR spectrum that is obtained. This is obvious from the CO2 spectra displayed in Figure 7 and 8. The low resolution (4cm-1) spectrum in Figure 7 is broad and shows a CO2 stretch at 2356.6cm-1. However, the high resolution (0.4cm-1 ) rotational spectrum (Figure 8) displays a regularly spaced spectrum and is a more detailed version of the maxima seen in the low resolution spectrum (Figure 7). The spacing of the data in high resolution is at every 0.4cm-1 compared to 4.0cm-1 in low resolution spectrum. Therefore high resolution provides with a very detailed rotational spectrum at very fine wavelengths. High resolution will always provide with quality data which will resolve all bands. However, for analytical analyses, a lowest acceptable resolution is picked to increase the signal to noise ratio. Low resolutions will give maximum precision and sensitivity in the quantification of gas species. For analytical purposes, a high resolution spectrum would be too noisy[12].
The peaks in the high resolution spectrum represent the rotational transitions of the carbon dioxide stretch. Using the rotational spectrum, the change in energy of rotation can be found. The change in energy of the low resolution spectrum is higher than the change in energy of the high resolution spectrum. Therefore the spectrograph cannot resolve the rotational bands at low resolutions.
In an FTIR spectrophotometer, a background scan should be performed before obtaining the spectra of the sample. To obtain the rotational spectrum of CO2 (Figure 7, 8), no back ground scans were performed. This was intentionally done to understand how the presence of carbon dioxide affects IR spectra of any analyte. Atmospheric components (CO2, H2O) can interfere with the analyte and their signals might appear in the IR spectrum. This effect can be understood from the spectrum of CO2 in Figure 10 and 11. Running a background scan avoids this problem. Background spectrum contains the characteristics of the instrument. When the background spectrum is removed, the data obtained wholly comes from the sample. The background spectrum is automatically subtracted from the spectrum of the analyte and helps generate accurate data. Before recording the IR spectra of the 2-butanone solutions, a background scan was performed [9, 10].
In the third part of the experiment, the IR spectra of five butanone solutions were recorded in standard resolution. High resolution was not used because the carbonyl peak obtained from the spectra of 2-butanone is very broad. Running the solutions at high resolutions will give a much detailed spectrum. However, because the carbonyl peak is broad, a detailed spectrum will not be useful and will lead to higher acquisition times.
In an FTIR spectrum, wavenumber (cm-1) of absorption is plotted against transmittance (%T). Transmittance (T) is a fraction of the amount of light at a unique wavelength that passes through a sample. It is a ratio of the intensity of incident light and the light that passes out of the sample. Absorbance is the negative log of transmittance; A= -log (T). Therefore the %T values obtained from the spectra of the 2-butanone solutions were converted to absorbance values to graph the calibration curve. As seen in Figures 9 and 10, he carbonyl stretch of the 2-butanone solutions were seen in 1710cm-1. This was the wavenumber of maximum absorption. Using the data from the IR spectra, a calibration curve for the 2-butanone solutions (Figure 11) with concentration versus absorbance was obtained. The linearity of the graph (Figure 11) proves the beers law which states that concentration is directly proportional to absorbance. As seen in Figure 10, the 0.150Wt% butanone solution with its highest concentration has the highest absorbance. The high R2 values found in Lineast function of Figure 11 illustrates the accuracy and linearity of the measurements. Through the equation obtained from the calibration curve (Figure 11), y=3.255x+0.1534. The concentration of the unknown 2-butanone solution was found to be 0.137±0.0068 wt %( see calculations). This value seems reasonable because according to Figure 10, the concentration of the unknown should be between 0.125 wt% and 0.150wt%. The unknown concentration of 0.137wt% falls between this range and proves that this experiment was successful.
Theoretically, the y-intercept; 0.1534 of the equation; y=3.255x+0.1534 should be equal to zero. This is because when there is no concentration, there should be no absorbance. However, as seen in the equation; the y intercept obtained from the experiment is not zero. The R2 value obtained from the lineast function is 0.981828. Theoretically, this value represents linearity and accuracy of the measurements and should be equal to one.
The discrepancies mentioned above might have resulted due to a number of reasons. One reason might be the salt plates. The salt plates lose their transparency and appear to be fogged as they get exposed to moisture. The values obtained for transmittance in the IR spectrum rely highly on the condition of the salt plates. As the windows get exceedingly fogged, the signal strength decreases. Another reason that might have caused this error is erroneous concentrations of the 2-butanone solutions. The solutions tend to evaporate quicker results in disparity between the actual and recorded concentrations.

Works Cited
1. http://gmi-inc.com/AnyLab/nicoletimpact400ir.htm

2. http://books.google.com/books?hl=en&lr=&id=0oWMbfciZzYC&oi=fnd&pg=PR13&dq=Fourier+transform+spectrometer&ots=wnsroEOOwS&sig=TT9h_e1-stvSz27LDm6xoxK0PTE#v=onepage&q=Fourier%20transform%20spectrometer&f=false

3. http://www.google.com/patents?hl=en&lr=&vid=USPAT7733493&id=X3HRAAAAEBAJ&oi=fnd&dq=Fourier+transform+spectrometer&printsec=abstract#v=onepage&q&f=false

4. http://www2.chemistry.msu.edu/faculty/reusch/VirtTxtJml/Spectrpy/InfraRed/infrared.htm

5. http://biosyntrx.com/dynimages/friday_pearl/spectrum.jpg

6. http://www.wcaslab.com/tech/tbftir.htm

7. http://orgchem.colorado.edu/procedures/IR/IRprepother.html

8. http://gmi-inc.com/AnyLab/nicoletimpact400ir.htm

9. http://www.chem.ucla.edu/~webspectra/irintro.html

10. http://mmrc.caltech.edu/FTIR/FTIRintro.pdf

11. http://www.ccrs.nrcan.gc.ca/resource/tutor/fundam/chapter2/04_e.php
12. http://www.irgas.com/app_ftir_res.html