Fourier Transform Infrared

Spectroscopy

FTIR DETERMINATION OF MTBE IN GASOLINE

AND ETHANOL IN VODKA AND MOUTHWASH

Last updated: June 17, 2014

FTIR - 1
 Fourier Transform Infrared Spectroscopy
FTIR DETERMINATION OF MTBE IN GASOLINE
AND ETHANOL IN VODKA AND MOUTHWASH

INTRODUCTION
As a part of the 1990 Clean Air Act Amendments, certain urban areas were required to
add oxygenates to gasoline in order to meet attainment levels of carbon monoxide.1 In
California, since June 1996, virtually all gasoline sold has contained MTBE (methyl tert–butyl
ether) as its primary oxygenate. However, there has been controversy over the use of MTBE as
an oxygenate for making cleaner burning gasoline.2-5 The additive has been found to
contaminate ground water supplies by release from leaking gasoline storage tanks. MTBE has
been classified as a possible human carcinogen and drinking water standards for this compound
are being established. As a result, MTBE has been banned from being used in gasoline in
California since 2003,6 and other additives, primarily ethanol, are used as the oxygenate.
However, small quantities of MTBE are typically found in gasoline, even where it is not the
major oxygenate. The amount of MTBE in gasoline samples will be determined in Part I of this
experiment.
Ethanol is a nervous system depressant with a broad variety of physiological effects
based on the blood alcohol level. It is found in various amounts in different alcoholic beverages
and other household items. Ethanol content is most commonly described in terms of proof,
which is just the ethanol volume percentage multiplied by 2. The potency of an alcoholic
beverage used to be tested by putting it on gunpowder and burning it for “proof” it was at least
50% ethanol by volume. The pervasiveness of alcohol consumption in the general populace, and
with high school and college students in particular, is widespread. The effects of alcohol abuse
on death rates, drug abuse, violence, health issues and economic costs are beyond the scope of
this introduction. In Part II of this experiment, the amount of ethanol in vodka and mouthwash
will be measured.

BACKGROUND
I: Qualitative Analysis
The technique of Infrared (IR) Spectroscopy takes advantage of the fact that many
molecules strongly absorb IR radiation and that the degree of absorption is proportional to the
molecular concentration. The wavelength range of the IR region extends from about 780 nm to
1,000 µm, with the relation between energy (E), wavelength (λ) and frequency (ν) shown in
Equations I and II below:

hc
E = hv = Equation I c = λν Equation II
λ

In Equations I and II, h is Planck’s constant (6.626 ×10–34 J s), and c is the speed of light in a
vacuum (taken to be 3.00 ×108 m s–1).
In IR techniques, the absorption or transmission of the IR radiation is commonly
measured as a function of wavenumber. A wavenumber is the reciprocal of the wavelength and is

FTIR - 2
most commonly expressed in units of cm-1. Thus the range of wavenumbers corresponding to
the IR spectrum would be about 12,800 to 10 cm–1. This is broken down into 3 main IR regions:
near-IR (12,800 to 4000 cm–1), mid-IR (4000 to 200 cm–1), and far-IR (200 to 10 cm–1). The
most commonly scanned wavenumbers are from 4000 to 670 cm–1, which encompass
absorptions by the majority of common organic functional groups.
For a molecule to absorb IR radiation, it must change its dipole moment upon vibration,
and the frequency of the radiation must exactly match the natural vibrational frequency of the
molecule, resulting in a change in the amplitude of the vibration. Some simple molecules (O2,
N2, etc.) have no fluctuating dipole moment, and so they do not absorb IR radiation. But many
vibrations of MTBE and ethanol change the dipole moment; such vibrations are said to be IR
active.
The two fundamental types of molecular vibrations are stretching and bending modes.
The stretching mode consists of a change in the distance along the axis of a bond between two
atoms. The bending mode results from a change in the angle between two bonds. There are four
types of bending vibrations: rocking, twisting, wagging and scissoring. Organic functional
groups have particular absorption peaks that can be used in qualitative analysis, varying only by
the molecular environment. For example, the "ether band" of MTBE around 1092 cm-1 is easily
distinguishable from absorptions by other components of gasoline and will be analyzed in Part I
of this lab.
From quantum theory, the vibrational states are quantized and the allowed vibrational
transitions are those in which the vibrational quantum number changes by unity. The more
atoms there are in the molecule, the more complicated the IR spectrum becomes due to increased
vibrational coupling and possible overtone peaks and combination bands. These effects create a
unique IR absorption spectrum for each molecule that can be used as a “fingerprint” in
qualitative experiments.

II: Quantitative Analysis

Although infrared spectroscopy is used extensively for qualitative analysis in organic
chemistry, band intensities are related to the concentration and path length of the sample through
the Beer–Lambert Law, shown in Equation III, and so this technique can be used for quantitative
analysis as well.
A = εlC Equation III

Where A is the Absorbance, ε is the molar absorptivity in L/(mol cm), l is the path length in cm
and C is the concentration of analyte solution in moles/L.
If the absorbance of a series of known standard solutions are measured, a plot of
Absorbance as a function of concentration can be made and least-square analyzed. Following
the expected linear dependence format, A = slope × C + offset, the slope of the linear plot would
be equal to εl, allowing determination of the molar absorptivity if the path length of the cell, l is
known (typically 1 cm). Also, an unknown solution’s concentration can be determined after its
Absorbance is measured and applied to the linear least squares fit.

III: The Fourier Transform Technique (for Advanced Readers)

Most IR instruments used today are of the Fourier Transform type. There are three major
advantages of using Fourier Transform techniques in IR spectroscopy.

FTIR - 3
 1) Fourier transform instruments do not need slits to attenuate radiation and have fewer
optical elements. The increased power reaching the detector gives a larger signal to noise ratio.
2) The high resolving power and wavelength reproducibility allow for more accurate
analysis of collected spectra.
3) The multiplex advantage, or faster scanning. In Fourier techniques, all wavelengths
are scanned simultaneously, allowing an entire spectrum to be scanned in 1 second or less. Since
the signal to noise ratio S ( )
N
increases as the number of scans, k, increases (as shown in
Equation IV), then Fourier techniques allow many more scans in less time and much better signal
to noise ratios.

(S N )
k scans
( N)
= S
one scan
× k Equation IV

As you can see, the quality of the spectrum increases in proportion to the square root of the
number of scans.
Fourier techniques basically differ from conventional techniques in that they measure
radiant power as a function of time (time domain) whereas conventional spectroscopy measures
power as a function of frequency (frequency domain). This time domain spectrum is then
mathematically converted into a frequency domain spectrum using a Fourier transform. The
process is so complex that it requires a high speed computer and will not be covered here.
Power variations at the very high frequencies of IR sources (1012 to 1014 Hz) cannot be
measured directly with today's electronics (transducers measure averages instead of variations at
these high frequencies). Therefore the high frequencies must be scaled down to much lower
values in order to measure time domain signals. This is commonly accomplished using a
Michelson Interferometer.
A Michelson Interferometer essentially splits the IR radiation beam from the source (high
frequencies) into two beams using a beam splitter. One beam is directed to a fixed mirror and
the other to a mirror moving at a constant speed, vm. The two beams are then recombined and
directed to the detector. The moveable mirror causes the radiation power at the detector to
fluctuate in a predictable manner based on the constructive and destructive interference patterns
of the recombined beams. These interference patterns are based on the difference in path length
(or retardation, δ) for the two beams. The plot of output power from the detector vs. retardation
is called an interferogram.
The resolution of the spectrometer, which is the difference in wavenumber between
two peaks that can just be separated by the instrument, is equal to the inverse of the retardation.
The relationship between the molecular emitter's frequency, ν, and the interferogram frequency,
f, is based on the moveable mirror speed, vm, according to Equation V below.

vm
f =2 v Equation V
c

Assuming a typical mirror velocity of 1.5 cm/sec and with the speed of light being 3.00 x
8
10 m/sec, then the interferometer reduces the frequency of the source radiation by a factor of
10–10 (i.e.: f = 10–10 ν). This brings the frequency into the audio range and allows transducers
to measure the power variations and thus record a time domain spectrum. The Fourier

FTIR - 4
Transform converts the time domain spectrum back to a frequency domain spectrum rapidly and
with incredible resolution and signal to noise ratio.

Part I: Determination of MTBE in Gasoline

In the first part of the experiment, you will quantify the amount of MTBE in gasoline
from its absorption of infrared radiation transmitted through the solution. The "ether band" of
MTBE around 1092 cm-1 is easily distinguished from other absorptions due to the hydrocarbon
components of gasoline. A series of MTBE/hexane standards can be used to prepare a linear
calibration plot of absorbance at the ether band vs. concentration of MTBE. From this plot, the
concentration of MTBE in a sample of gasoline can be derived.

Experimental Procedure
Note: Detailed instructions on the start up, use, and shut down of the Jasco FT/IR–615
instrument are provided in the four page handout near the machine in the lab, please read
them carefully before beginning.
1) In the fume hood, prepare a stock solution of 5% (volume/volume) MTBE by carefully
adding hexane to 2500 µL of MTBE until you obtain a total volume of 50.0 mL in a
volumetric flask. There are pipettes available for adding 250 µL of the MTBE. Be sure
to condition the flasks and pipettes first.
2) From this stock solution, make five standard solutions:
Volume of 5% Stock Solution Add Hexane to Total Volume Final Concentration
1.0 mL 10.0 mL 0.5 % (v/v)
3.0 10.0 1.5 %
4.0 10.0 2%
6.0 10.0 3%
8.0 10.0 4%
Be sure to condition each flask and label with tape. Close the flasks after preparation to
avoid evaporation.
3) Dilute the “old” gasoline sample with hexane to make a 25% (volume/volume) solution
by adding hexane to 2500 µL of the gasoline to a total volume of 10.0 mL of solution.
Label the flask and close it. Make a similar solution with “new” gasoline.
4) Set the spectrometer resolution to 1 cm-1 and the number of scans to 16 by clicking on
“Measure” and then “Parameters”.
5) Using a plastic pipette, flush the transmission IR salt crystal cell with the hexane solvent
four times, then fill with hexane. After that, load this cell into the spectrometer
compartment.
6) Take a background spectrum of the hexane solvent and save this spectrum for later use.
To do a single beam spectrum, click on “Measure” then “Parameters + Background” and
under vertical axis choose “single” for the background and “abs” for the sample. Click
“OK” to run the background. (or click on the “B” icon with a box in it as a shortcut).
The spectral analysis software will then automatically ratio the sample spectra with that
of the most recent background spectrum. When the run is complete, the colors of the top
tabs will return.

FTIR - 5
 7) Now take the cell out and flush it four times with the standard solution you are going to
use next. After that fill it with the standard solution and place it back into the
spectrometer. Take a spectrum for your first standard by clicking on the “S” tab with no
box in it (for run Sample) and save it.
8) Scale the spectra to focus on the ether peak around 1092 cm–1. Do this by clicking on
“View” then “Scale” and type in the desired x and y axis ranges (~ 1150 to 1050 cm–1
and 0 to 2, respectively). Use “Peak Find” to locate all peaks (click on “Processing ---->
“Peak Process” ----> “Peak Find” ----> “Execute”) and choose the appropriate peak from
the Table listed.
9) Repeat steps 7-8 for your remaining MTBE standard solutions.
10) Now, record the spectrum of the 25% gasoline samples (old and new). Make sure the
absorbance of the ether band falls on the calibration curve. Determine the absorbance of
the sample from the same ether peak used in the standards as you did in step 8.

Data Analysis
1) Make a Table of MTBE % concentration vs. Peak Absorbance measured at the top of the
ether band around 1092 cm-1 for the standard solutions.

Sample Absorbance

2) Develop a Beer-Lambert Law plot for the MTBE in hexane standards and perform a least
squares analysis of the linear best fit line on a computer or by hand on graph paper if a
computer is not available (Microsoft Excel instructions for graphing are in the Appendix
if needed). This will give you a dependence in the form:

Absorbance = slope × C + offset (y = mx + b)

3) Using the slope and offset parameters determined from your fit, calculate the % MTBE in
the diluted gasoline sample.
4) Calculate the volume percent of MTBE in the original undiluted gasoline samples. Make
sure you take into account the various dilutions.
5) Assign the vibration responsible for the peak at 1092 cm-1 in the MTBE absorbance
spectrum.

FTIR - 6
PART II: Determination of Ethanol in Vodka and Mouthwash
This part of the experiment will show that infrared spectroscopy can be carried out in
water solutions using appropriate infrared-transmitting, but water-insoluble, crystals (of ZnSe in
this case) using the technique of attenuated total reflectance (ATR) FTIR. You will use this
technique to determine the ethanol concentration in vodka and mouthwash.

Figure 1. Schematic diagram of single-bounce ATR accessory.

This technique allows you to quantitatively measure the absorbance of ethanol in water,
even though water is a very strong absorber of infrared radiation itself. As seen in Figure 1, in
the ATR technique the sample is placed on an internal reflection element and the IR beam is
directed into the element. It strikes the internal crystal-air interface at an angle greater than the
critical angle, and as a result undergoes internal reflection inside the crystal. Most radiation is
reflected at the point of internal reflection, but a small fraction is absorbed by molecules present
at the surface of the ATR crystal. This absorption of infrared radiation can then be detected and
measured.
Increased sensitivity can be obtained by using a multipass ATR accessory. Figure 2
shows a schematic of the light path in such a device; the increased number of internal reflections
leads to a proportional increase in the absorbance and hence in the sensitivity.

analyte solution
light in

light out to
detector

Figure 2. Schematic diagram of multiple reflections inside a multipass ATR accessory.

FTIR - 7
Experimental Procedure
1) Prepare a 10 % (by volume) 95% ethanol in water solution by adding nanopure water to 5.0
mL of ethanol to a total volume of 50.0 mL in a volumetric flask.
2) Now place the liquid multi-pass ATR accessory in the sampling compartment. Be sure
not to turn any of the screws on the accessory as they have been tuned to make sure the
infrared beam passes into the crystal and back out to the detector properly. When
properly aligned, you should be able to see red dots where the HeNe laser is reflecting at
the crystal surface along the center of the crystal. Try to count the number of reflections
you can see in the crystal. You may want to do this in the dark, since it makes the spots
easier to see.
3) Carefully fill the top of the crystal with nanopure water using a pipette (DO NOT SPILL
WATER IN THE SAMPLE COMPARTMENT). Take a background absorbance
spectrum using 16 scans and 1 cm-1 resolution (hit the “B” icon with a box in it). Remove
the water carefully with a plastic pipette and then dab the trough with clean, lint-free
tissue. Do not exert any pressure on the glass surface during this procedure.
4) Now condition, then fill the ATR accessory with the 10% ethanol sample. Take a
spectrum of this sample and then take another scan immediately afterwards for
reproducibility. Scale the spectra around the alcohol peak near 1044 cm–1. Find the
absorbance at that peak and record it in your lab book.
5) Prepare a set of standard solutions by diluting the 10% Ethanol solution with Nanopure
H2O.
Volume of 10% Stock Add Nanopure H2O to Final Concentration
Solution Total Volume
1.0 mL 10.0 mL 1.0% (v:v)
3.0 10.0 3.0%
5.0 10.0 5.0%
Label each flask with tape. Close the flasks after preparation to avoid evaporation. Find
the absorbance of each solution as you did for the 10 % sample in step 4. Record the
values in the Table below.

Sample Absorbance

FTIR - 8
 6) Prepare a diluted vodka solution by adding nanopure H2O to 1.0 mL of the original vodka
solution in a 10.0 mL volumetric flask. Record the spectrum of the diluted vodka sample
as before and make any necessary dilutions so the absorbance falls on the calibration
curve. Record the value in the previous Table.
7) Prepare a diluted mouthwash solution by adding nanopure H2O to 2.0 mL of the original
mouthwash solution in a 10.0 mL volumetric flask. Record the spectrum of the resulting
sample as before and make any necessary dilutions so the absorbance falls on the
calibration curve. Record the value in the previous Table.
8) Record the dimensions of the crystal as well as its angle of incidence from the label in the
ATR cabinet.
9) If there is extra time there are 2 possible solutions to test. First, if a non-alcoholic
mouthwash is available, prepare a diluted solution as in Step 7 and scan it. Second, if a
sample of newer (supposedly MTBE free) gasoline is available, make a diluted solution
as in Step 3 in Part I and scan it. See if the MTBE peaks disappear and if the ethanol
peaks appear.

Data Analysis
1) Calculate the theoretical number of reflections, N, along the crystal given the formula

N = l cot θ
2t

where l is the crystal length, θ is the angle of incidence (determined by the optical
configuration and provided by the manufacturer of the ATR accessory) and t is the
thickness of the crystal.
2) Develop a Beer-Lambert plot for the ethanol in water standards. Use the least squares fit
to determine the % ethanol in the diluted vodka sample, as well as for the Listerine.
Follow the same instructions as the plot made in Part I.
3) Calculate the volume percent of ethanol in the original undiluted vodka and mouthwash
samples. Make sure you take into the account the various dilutions. You can show your
work calculations below.
4) Determine the proof of the original vodka sample.

IF THERE IS STILL TIME LEFT, DO THE FOLLOWING EXTRA SECTION:

PART III: Instrumental Noise

A standard method for improving signal-to-noise (S/N) is to increase the number of
scans. The S/N should improve by the square root of the number of scans as described in the
background section.

FTIR - 9
Experimental Procedure
1. Set the spectrometer to capture an interferogram. Set the number of scans to be 1 and the
resolution at 1 cm-1. Make sure there is no accessory inside the sample compartment of
the spectrometer.
2. Take a background spectrum with no cell in the sample compartment. Convert the
interferogram into a single beam spectrum.
3. Now take another spectrum with no cell in the sample compartment. Convert the
interferogram into a single beam spectrum and ratio this spectrum to your background
spectrum. The resulting spectrum will be a transmittance spectrum of the noise of the
instrument. Convert this transmittance spectrum to absorbance.
4. View the spectrum of noise from 2100 to 2000 cm-1 to find the lowest valley and the
highest peak in this region. The difference gives you the peak-to-peak noise. Mark the
valley and peak on the spectrum and print out.
5. Now set the number of scans to be 4 (leave the resolution at 1 cm-1). Repeat step one
through four to find the peak-to-peak noise with 4 scans.
6. Now set the number of scans to be 64 (leave the resolution at 1 cm-1). Repeat step one
through four to find the peak-to-peak noise with 64 scans.

Data Analysis
1) Make a table showing the number of scans and the peak-to-peak noise for each.
2) Quantitatively compare the change in the noise with the number of scans and compare to
theoretical expectations.

REFERENCES
1. J. G. Calvert, J. B. Heywood, R. F. Sawyer and J. H. Seinfeld, “Achieving Acceptable Air
Quality: Some Reflections on Controlling Vehicle Emissions”, Science, 1993, 261, 37.
2. "Air Toxics Program Summary: Adding Oxygenates to Fuel", Health Effects Institute,
1997.
3. http://tsrtp.ucdavis.edu/mtberpt/homepage.html
4. Reuter, J. E.; Allen, B. C.; Richards, R. C.; Pankow, J.; Goldman, C. R.; Scholl, R. L.;
Seyfried, J. S. Environ. Sci. Technol. 1998, 32, 3666.
5. Johnson, R.; Pankow, J.; Bender, D.; Price, C.; Zogorski, J. Environ. Sci. Technol. 2000,
34, 210A.
6. http://www.calepa.ca.gov/programs/mtbe/eotasks.htm
7. Griffiths, P. R.; Fuller, M. P. In Advances in Infrared and Raman Spectroscopy; Clark, R.
J. H., Hester, R. E., Eds.; Heydon and Sons: London, 1982; Vol. 9, Ch. 2, pp. 63-129.

Additional References:
1. D. A. Skoog, F. J. Holler and T. A. Nieman, Principles of Instrumental Analysis, 5th Ed.,
Harcourt Brace, Philadelphia, 1998, pp. 380-401, 404-421.
2. P. Griffiths and J. DeHaseth, Fourier Transform Infrared Spectrometry, Wiley, 1986.

FTIR - 10