Title Page
 Table of Contents
 List of Tables
 List of Figures
 Apparatus and experimental...
 Discussion of results
 Biographical note

Title: study of the hydrogen stretching region in the infrared.
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00091624/00001
 Material Information
Title: study of the hydrogen stretching region in the infrared.
Series Title: study of the hydrogen stretching region in the infrared.
Physical Description: Book
Creator: Knapp, Kenneth T.
 Record Information
Bibliographic ID: UF00091624
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000424012
oclc - 11069216


This item has the following downloads:

Binder1 ( PDF )

Table of Contents
    Title Page
        Page i
        Page ii
    Table of Contents
        Page iii
    List of Tables
        Page iv
    List of Figures
        Page v
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
    Apparatus and experimental technique
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
    Discussion of results
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
    Biographical note
        Page 95
        Page 96
Full Text
a dissertation presented to the graduate council of
the university of florida in partial fulfillment of the requirements for the degree of doctor of philosophy

The author wishes to express sincere thanks to his research director, Dr. Armin H. Gropp, for his advice and encouragement in this research. He is also indebted to the other members of his committee.
The helpful suggestions of some of his colleagues were most appreciated.
The compounds contributed by Dr. Paul Tarrant, Dr. H. C. Brown, Dr. J. A. Young, and their graduate students were invaluable.
The author also wishes to acknowledge the financial assi ance of the General Motors Corporation which made this research possible.

Preparation and Purification of Compounds

Table Page
I Boiling and Melting Points of the Fluorocarbon
Hydrides 32
II O-H Stretching Bands of Unsubstituted Alcohols
from Carbon Tetrachloride Solutions 36
III O-H Stretching Bands of Unsubstituted Alcohols
from the Pure Compounds 39
IV O-H Stretching Bands of Fluorinated Alcohols
from Carbon Tetrachloride Solutions 41
V O-H Stretching Bands of Fluorinated Alcohols
from the Pure Compounds 42
VI O-H Stretching Bands of Carboxylic Acids from
Carbon Tetrachloride Solutions 53
VII N-H Stretching Bands from the Pure Compound 61
VIII N-H Stretching Bands from Carbon Tetrachloride
Solutions 63
IX C-H Stretching Bands of Hydrocarbons from the
Pure Compounds 72
X C-H Stretching Bands of Alcohols from both the
Pure Compounds and Carbon Tetrachloride Solutions 75
XI C-H Stretching Bands of Cyclobutanes from the
Pure Compounds 78
XII C-H Stretching Bands of Aromatic s from the
Pure Compounds 80

Figure Page
1. Comparison of Resolving Power of Calcium Fluoride and Sodium Chloride Prisms Using CF2=CFCH2CH2Br 24
2. Comparison of the Spectra of CF^CF CH^OH with the Spectra of the Carbon Tetrachloride Solution 43
3. Comparison of the Spectra of CF CF CH OH
and CH^CH^CH^OH 44
3 2 2
4. Comparison of the Spectra of CF COOH and

Molecules are no longer considered to be rigid. They are groups of atoms held together by certain forces, called bonds. These bonds are not in themselves rigid but are flexible enough to go through certain changes without destroying themselves. One of these changes is that the bond may stretch like a spring, and energy is need for this transformation. However, only certain discrete energies can be used for this change. The energy necessary for the stretching vibration of a bond is quantized, and the frequency of the vibration is discrete or single valued.
The appearance of these vibrations as bands, rather than discrete
lines, is briefly discussed by McKisson* and in more detail by Z 3, 4
Heraberg and others. The frequency of the vibration depends on
the forces of the bond and the kinds of atoms. If there are more than
two atoms in the molecule, the frequency also depends on which atom,
or group of atoms, is attached to the atoms making up the bond. It
has been known from the early work in infrared that certain functional
groups have characteristic frequencies. These frequencies vary
with the influences on the functional groups, but the variations are
confined to a rather narrow range.

The present study is concerned with the influence of fluorination on the hydrogen stretching vibrations of the oxygen-hydrogen, nitrogen-hydrogen, and carbon-hydrogen bonds. The general range of thj frequency of these vibrations is between 2 and 4 microns, or in the infrared portion of the electromagnetic spectrum.
As early as 1395 publications appeared on the oxygen-hydrogen
stretching vibration, followed in 1905 with publications on both the
carbon-hydrogen and nitrogen-hydrogen stretches. In this early
work, however, the frequencies reported were only approximate and
their limits were larger than those of later works.
Intense studies of the vibration of the O-II bond began with the
work of Kiibert, Wulf, Hendricks, and Liddel and today this is one of the most studied bonds. The earlier work was mainly concerned with the first overtone region rather than the fundamental frequencies. The overtones are integral multiples of the fundamental frequencies. The O-H stretching vibrations can be divided into two classes. One is the free vibration of a hydrogen atom which has no bonds through associations with neighboring molecules, and the other is the associated vibration of a hydrogen atom which has hydrogen bonding with one or more neighboring molecules. Most of the early work was concerned with hydrogen bonding, and the overtone region was used mainly because of its easier detection.

The present study is concerned only with alcohol and carbox-ylic acid O-H bonds, and the following discussion on the O-H bond will be limited to these types of compounds.
The pure alcohols in the liquid or solid state have enough molecules so close together that there is adequate attraction available for hydrogen bonding. In fact, these attractions are so great that in most cases only the associated band appears. Therefore, in order* to obtain the free band the molecules must be far apart. This can be realised either in the gaseous state or in dilute solution where the
solvent has negligible attraction for the hydrogen of the O-H and
should be a non-polar solvent. Bellamy states that the free band is obtained only under these conditions. The fluorinated alcohols of this study, however, display this band without these conditions.
The overall range of the free O-H stretching vibration ie
1 1
visually quoted as being from 3800 cm. to 3500 cm." This
range, however, includes the band given by water, which occurs -1
at 3760 cm. Leaving out water, the range would be considerably
narrower, and tor aicohoU it i. listed by B.W ae being froxn
-1 X l
3650 cm. to 3590 cm. An even narrower region of 3636 cm.
-1 7 8
to 3618 cm. was given the free O-H band by Fox and Martin, '
With the exception of methanol, Kuhn set the limits of the free
-1 -l
O-H band as 3644 cm. to 3605 cm. His value for the free

U-H band of methanol in carbon tetrachloride was given as 3462 cm.
Using the free O-H band of phenols and car boxy lie acid, Goulden
showed a relationship between the frequencies of these bands and the
pKa of the compounds. Haszeldine has also used this band to determine the acid strength of the halogenated alcohols and to show that they are more acidic than the non-halogenated alcohols.
Other workers have shown that it is possible to differentiate between primary, secondary and tertiary alcohols using the frequency
of the free O-M band. Thia, however, only holds for the simple
12 -1
alcohols. The values given by Cole and Jefferie b are 3642 cm,
-1 _i '
for primary, 3629 cm, for secondary and 3618 cm. for tertiary,
The frequency of the free O-H band is shifted under various
conditions. The example given above shows that the position of the
O-H group varies or shifts the frequency of the free O-H band. The
group or type of the bond next to the O-H may also cause a shift.
Such a shift is produced when the O-H is attached to a carbonyl 5
group (forming a carboxylic acid) and the band appears near -1
3520 cm, A double bond causes a slight shift, and steric effects sometimes produce a shift. The temperature, as well as solvent
*The author of the present work feels this is probably a printing error and should be 3662 cm, *K The value 3462 cm, ~l for the free O-H band is less than that given for the single-bridge O-H band (3525 cm.*"*)> also the present research value for this band is 3658 cm, -1,

effects, may also cause a slight shift. No appreciable solvent
effect was noted in the present study, and the temperature was approxi-
mately constant. Haszeidine has reported a slight shift due to
o^-fluorine substitution, and this is discussed in more detail in a later section. In the case of fluorinated alcohols c< substitution means substitution on the carbon next to the -CH^OH group since the -CH^OH group is taken to be the functional group by fluorine chemists rather than the -OH group.
As mentioned above the pure compounds in the solid or liquid states are primarily associated through hydrogen bonding and exist mainly as weak polymers with a few dimer and still fewer free molecules. It is believe than an equilibrium exists between the polymeric, the dimer, and the free molecular forms. This equilibrium shifts toward the dimer and free molecular forms as the concentration of
the molecules of the alcohols becomes less in a given volume. Klitz 13
and Price showed that the shift from the associated band to the
free band can be demonstrated by using carbon tetrachloride solutions.
As in the case of other equilibria, Fox and Martin suggested there should be a temperature effect, and they postulated two rules: that the shift of the jcionomer-dimer equilibrium with temperature changes the intensity of the O-H bands, and that no chango occurs in the
frequency with temperature. In some later work by Finch and 14
lippincott, using the higher resolving L1F prism, they were able

to show a slight shift due to temperature; The maximum shift of -i
5. 5 cm, per 10"C was obtained with butanol.
In general, hydrogen bonding can be represented as X-H* Y. This leads to two possible types of hydrogen bonding, the intermole-cular and the intramolecular, depending on the source of Y. In the intramolecular case, there is generally the possibility of only a single bridge. In the intermoiecular case, however, there can be both single bridges and multi bridges or polymeric association. The intramolecular and intermoiecular bands may generally be distinguished from each other in two ways. The best way to distinguish them is
that the
intramolecular band shows little or no change on dilution,
while the intermoiecular bridges are broken on dilution. The second
way is the interpretation of the appearance of the band with regard
to size, shape, and frequency, Bellamy states that the Intermoiecular hydrogen bands are sharp bands with a range of 3550 cm. 1 to
3450 cm. for a single-bridge compound. The polymeric associated
-1 -1
compounds produce broad bands between 3400 cm. and 3200 cm.
The intramolecular hydrogen bands are sharp and range from 3570 cm. -1
to 3450 cm. A third type is also given, the chelate compounds with
-1 -1
a range from 3200 cm. to 2500 cm. ; these bands are weak and very broad,
Xt is noted above that the range and shape given for the single-bridge intermoiecular hydrogen band and the intramolecular hydrogen

band are approximately the same. Inmost cases the single-bridge band for intermoiecular hydrogen bonds will not appear when the pure compound is run in the liquid or solid state. There is, however, room for doubt. This leaves the first method of distinguishing the intermoiecular and the intramolecular hydrogen bonding the more reliable.
Due to the general broad appearance of the associated O-H band, it is difficult to determine when the band has shifted unless this change is very large. Although the present study is interested in any shift due to fluorination, it places more emphasis on the free O-H, which has a very sharp appearance, than on the broad associated O-H band.
The association in carboxylic acids is indeed great, and the O-H
bands, due to this association, are very broad in the pure compounds.
In fact, they are so broad that very little information, other than the
identification of the compound as an acid, can be obtained. Much
work, however, has been done on carboxylic acids as pure liquids or
solids. Usually the interest is in a wider range, generally from -1-1
5000 cm. to 660 cm. rather than only the O-H stretching region.
Corish and Chapman ran the infrared spectra of the first 10 members of the monocarboxylic acid series in both the liquid and solid states over such a range. The spectrum of the given acid was different in the two states especially in the O-H stretching region, Not only was there a change in the intensity of the band in the O-H stretching region,

but there was also a slight shift in the frequencies of the bands. As the chain length increased tins difference in intensity became less. This, however, did not hold for the frequency shift. A few additional bands appeared in the O-H region when the acid was run in the solid state that did not appear when the acid was run as a pure liquid. The authors of the above mentioned work felt that this slight shift and the additional bands resulted from crystallization*
These shifts can arise either from perturbation effects of the crystalline field or from a small change in hydrogen bond distances. The additional bands may arise from combinations of the O-H
stretching frequency with lattice vibrations or with low frequencies
associated with the dimeric ring form of the acids. These extra
peaks may be due to summation bands of the combination of the free
O-H stretching band and the single-bridged hydrogen bond band.
The O-H stretching region of carboxylic acids is more complex than
this region is in alcohols. Not only do the acids have the free and
associated O-H stretching bands, but they also have from one to
three additional bands, referred to as submaxima or satellite bands.
These bands appear at a lower frequency than the associated O-H
stretching bands. The appearance of these bands according to the 17
Flett, is the surest way to ascertain that the compound is a carboxylic acid. One treatment of the submaxima bands was given by Davies and Evans*** and carried further by Fuson and Josien.***

This explanation states that these submaxima peaks appear at regular frequencies and are referred to as summation bands. These frequencies are the differences between the frequency of the fundamental free O-H stretch and the single-bridge* or dimer O-H hydrogen bond bands. The relationship aSSOCi a free -n A^ where n is an
integer greater than one and A^ is the average spacing between submaxima bands, gives the frequencies of these submaxima bands, The important question that immediately arises is why there is no band with n equal to one. in most cases of frequency combinations that band with n equal to one of the strongest and most important band.
An interesting phenomenon observed by Fuson and Josien was that if an acid having an -OD group, rather than an -OH group, was used, the value of /\^was cut approximately in half* They suggested a possible explanation based on the multiple quantized ground-state energy level hypothesis since the atomic mass of
deuterium is about tv/ice that of ordinary hydrogen.
Bratoz, Hadiz and Sheppard have also done work comparing the light and heavy {-OH group and -OD group) carboxylic acids. They found that the width of the associated band, measured half way up the band, was twice as wide for the OH group as for the -OD group. Three explanations of the origin of the submaxima or satellite bands, were listed. The first was that the bands could be caused by the second O-H fundamental frequency of the dimeric unit. The second

was that the band could correspond to a different frequency of the typo u where -jj' is a fundamental of an approxi-
mate frequency of 300 cm. \ This is generally referred to as a combination band. The third explanation offered was that these bands could be due to the summation frequencies of the lower fundamentals of the COOH groups. This, too, would be a combination band explanation. One additional point offered was that the greater intensity observed over that which should be expected could be due to Fermi resonance with the O-H vibration frequency.
The first and second explanations were ruled out. This left the third explanation which the authors felt was in general agreement with the experimental results. The original paper should be consulteo for a detailed explanation of why the first and second explanations were discarded.
A study of the monomer-dimer equilibrium for acetic, the three
chloroacetic and benzoic acids was carried out by Harris and Hobbs. They were able to calculate the equilibrium constants for these acids from their data, A combination of the Bouger-Beer law and the expression for the monome r dime r equilibrium constant was used,
Davies and Sutherland have reported a study of acetic, benzoic, and lauric acids in which an attempt was made to determine both the association equilibrium constant and the heats of association. A corrected integrated area of the O-H monomer absorption regions was

used. There was excellent agreement between these two studies for the value of the association equilibrium constant for benzoic acid. Other than this value for the benzoic acid constant, Davieg and Sutherland gave little explicit data for comparison.
Hydrogen bonding also occurs with, the -NH and -NH^ groups. As in the case of the -OH group, this leads to two types of bands, the free and the associated. The -NH group adds another complication since it has both symmetric and asymmetric vibrations.
Fluorinated amines are very unstable; in fact, the primary amines are so unstable that not a single one has every been isolated. A few fluorinated secondary amines have been isolated. However, they decompose on standing by splitting out HF. Because of their unstable nature, the fluorinated amines were not studied. Therefore, the discussion of the N-H stretching bands will be limited to amides and similar compounds.
While the -NH groups have hydrogen bonding associated with ,
them, this hydrogen bonding is somewhat weaker than that of the -OH
groups. The -NH hydrogen bond differs from the -OH hydrogen bond
in other ways as well, Oshida, Ooshika, and Miyasaka mentioned
three types of hydrogen bonding and discussed the possible resulting
resonance in amides. A more extensive study of the hydrogen bonds
in amides was carried out by Cannon. Using various polypeptide alternations, Cannon found a large shift in the frequency of the carbonyl

band but little or no shift in the frequency of the N-H stretching band.
This implied that there was a dipole-dipole interaction of the -OCN
dipole which contributed more toward the interaction energy than the
potentially weaker hydrogen bond. Carrying his studies further 24
Gannon used carbon tetrachloride solutions to show that while -COH forms hydrogen bonds with molecules donating protons, -NH is comparatively inert, not only with respect to accepting protons but also donating them. He concluded that the -CONH group was unlike -COOH and forms a peptide group association by dipole interaction rather than by hydrogen bonding. He extended his reasoning to proteins and gave a quantum mechanical consideration to hydrogen bonding.
The space configuration ot amides has also aroused much interest. 25
Senti and Harker gave an elaborate treatment for the structure of acetamide. They determined that the molecule was planar and that the -NH bond lies in the plane of the molecule. They then calculated the dipole moment using vector addition of component bond moments. The intramolecular bond distances along with the N-H* *Q bridge distance were also given. From their investigation they also felt that the molecules exist in the keto form.
Damon and Rudall were also concerned with the structure and orientation of amides and similar compounds. Their method of investigation involved using polarised radiation in which they rotated a film of the substance under study. The change of intensity of a band

due to the rotation of the film gave evidence of the orientation of that group. From this information, and with the aid of X-ray and other means, the amide molecules were found to be planar.
It is known that there are two possible forms in which a ketone or aldehyde can exist, these being the keto and enol forms. This possibility also exists with amides and has been of great concern to many investigators.
In 1931 Hantzsch, using the ultra-violet spectra of amides in
various solvents, concluded that the enolic form predominates. This*
however, did not satisfy other workers and Buswell, Rodebush and 28
Roy took up the study of this problem. They concluded that if the
enolic form exists in amides the OH group must "be present, and it
should give the typical O-H stretching bands. They found no such
bands for the monosubstituted amides and suggested that the enolic
form of these amides did not exist. However, a large peak at -1
3200 cm. for primary amides was found, and they concluded that this was an O-H stretching band. This implied that an enolic form does exist in primary amides. This 3200 cm. band rapidly disappeared as solutions of the primary amides were diluted and indicated that the enolic form may exist only at high concentrations
when the primary amides exist as polymers through hydrogen bonding.
Richards and Thompson felt that if the enolic OH group exists it should give the sharp free O-H stretching band. This band can be

obtained when OH containing compounds are run in dilute solution
with a non-polar solvent. In dilute solutions of various amides two
-1 -1
sharp bands were found at about 3400 cm. and 3520 cm, The
-1 -1
free O-H stretching band appears between 3650 cm. and 3590 cm,
for alcohol and phenoU and about 3550 em.for carbolic acids.
Only the halogenated carboxylic acids give the free O-H stretching
band as low as 3520 cm. Using this evidence Richards and Thompson
decided that the amides do not exist in the enolic form, Letaw and
Gropp resolved the controversy in their work. Using both mono-substituted and disubstituted amides, they deduced that the band at about 1650 cm. which exists in the spectra of all amides, was the carbonyl band, This band only exists with the ketonic form. Because of the great intensity of these bands, it was concluded that the ketonic form predominates.
Buswell and co-workers also stated that two bands should
exist for the -NH., group as in the case of H20. There should be
both symmetric and an asymmetric stretching band. They reported
two such bands for propanamide, the symmetric at 3531 cm. and the asymmetric at 3424 cm. "V They concluded that the primary amide existed as a polymer in the pure form, since the bands due to hydrogen bonding changed at different rates upon dilution of the carbon tetrachloride solution. The change of the bands at different rates upon dilution implies there is more than one type of hydrogen

bonding, probably both polymeric and dimeric. Evidence is given
that the mono substituted amides exist in the dimeric form, This is
30 31
also Supported by Letaw and Gropp and Klemperer and co-workers.
Another problem which has created much interest is whether
the mono substituted amides exist with the proton of the -NH group
cis or trans, or both, to the oxygen of the carbonyl group. With
higher resolving instruments the peak in the range of 3420 cm. ~* to -1
3460 cm, of mono substituted amides was found to be split into two peaks for some of the amides.
The first explanation offered was that these peaks were perhaps overtones. The following arguments were used to rule this out. First,
the only intense band that could be considered was the carbonyl band '
-1 -1
at about 1660 cm. In order for the 3420 cm. band to be due to the
first overtone of this band, there would have to be a large negative anharmonicity which is not the normal case. Furthermore, if Fermi resonance between this Overtone and the N-H stretching band compensated for the intensity, it would make the anharmonicity worse. The second, and more conclusive, objection was that when sulfur
replaced the oxygen of the carbonyl group, and shifted the carbonyl
-1 32
frequency, the 3420 cm, band remained.
32 -1
Russell and Thompson assigned the band between 3420 cm.
and 3440 cm. as the band due to the cis form and the band between 3440 cm. and 3460 cm. as the band due to the trans form. They

investigated 33 compounds and, using the intrinsic intensities of these bands, calculated the per cent of cis and trans forms. Most of the compounds showed both forms; however, the imides gave only the cis form, and the lactam and aromatic amides with restricted rotation gave only the trans form.
Determining the dipole moment, using vector addition as Senti
25 33
and Marker did for acetamide, Worsham and Hobbs compared the
calculated moments with observed values obtained from infrared
spectra. Good agreement was obtained in all but one case. Using
this comparison, they were able to predict which band was due to the
cis form and v/hich band was due to the trans form. Darmon and 34
Sutherland studied the infrared spectra of polyamides such as nylon and proteins. They gave both a chain and a cyclic cis form. They also found two types of hydrogen bonding in some of these compounds.
Through the vigorous effort of many investigators using the 3000 cm. range of the infrared spectra, some very important problems have been solved. Because of this work, one can now deduce many facts about amides from this region of their infrared spectra.
The 5000 cm. to 2500 cm. _1 region (2 to 4 micron) also contains the C-H stretching frequencies. Since there is generally no hydrogen bonding associated with the -CH group, a less complicated system than that of either the -NH or -OH groups exists. There is,

however, the complication arising from the possibility of symmetric and asymmetric vibrations,
In organic compounds the carbon atom forms four bonds, In some cases there may be two or even three bonds to the same aw<^.. With the exception of methane this leads to three types of hydrogen* carbon groups depending on the number of hydrogens attached to a given carbon. In general, these groups are; three hydrogens on the carbon, called the methyl group; two hydrogens on the carbon, called the methylene group; and a single hydrogen on the carbon. The number of bands, the frequencies, and the intensities of the C-H stretching vibration differs among these groups, The properties of these vibrations also depend on the kinds of atoms which are attached to the carbon atom and what kinds of bonds hold these atoms to the carbon atom.
Many intensive studies of the C-H stretching vibrations have
7 8 35
been carried out. Fox and Martin, in several publication, have treated all the carbon-hydrogen groups. Not only did they assign frequencies to the various bands, but they also gave a theoretical treatment with illustrations to the different types of vibrations. A large number of hydrocarbons containing methyl groups
were examined by them, and two strong bands at 2962 cm, and
' -1
2872 cm, were found in all cases. They concluded that these two bands should be due to the symmetric and the asymmetric vibrations

of the methyl group. In saturated hydrocarbons the position 01 these bands and their intensities per ^CH^ group were fairly constant. This, however,, .was not the case with unsaturated compounds with the.double bond adjacent to the methyl group. In these cases both the position and intensities varied from one compound to another. These compounds
the two bands 2872 cffl,""l
The methylene group can either have two single bonds to two
7 8
atoms or one double bond to one atom. Fox and Martin considered
both types. The vibrations of the sCH2 group were assigned the values
*1 -l
3079 cm. and 2978 cm, for the asymmetric and the symmetric
vibrations, respectively. They assigned the frequencies of the vibrations of the grouP a* slightly lower values. The values
-1 -I
given were 2926 cm. for the asymmetric vibration and 2853 cm.
for the symmetric vibration.
In the case of the single hydrogen on the carbon, there are four possible types of bonding. There are, three single bonds, one single bond.,' and one-double bond as in: olefins, the aromatic bond, and'one triple bond. Fox and Martin investigated the first two types intensely, but only briefly mentioned the last two* The CU group was assigned the value 2890 cm." for its stretching vibration, and the =CH group in olefins was given the value 3019 cm,for its stretching vibration,

Since Fox and Martin's work, many investigations on special groups of compounds have been carried out. In most cases where the molecules were not under great strain, the values reported for the C-H stretching vibrations agreed fairly well with Fox and Martin's values. There is, however, some disagreement on the assignment of the
various bands, Another extensive coverage of the C-H stretching
frequencies was given by Colthup.
Using sulfuriaed and oxygenated compounds Poaefsky and Cogge-37
shall studied the C-H stretching frequencies. ; On the average, they found the frequencies highest for the oxygenated series* intermediate in the sulfuriaed series aiii lowest in the hydrocarbons. The variations of the assignment of the frequencies from those of Fox and Martin were, in general* not very great. No band was observed that these authors felt could be ascribed to the -^GH group.
Careful studies of the intensity of the C-H stretching band have
been conducted. Fox; and Martin* as has already been noted, found that the intensities of the and -CHg bands were directly pro-
portional to the number of these groups present in the molecule of the long-chain hydrocarbons. They also noted that the intensities of the
^CI-I^ bands changed by a. steady increment as the number of this group- increased.:... Pozef sky and Coggeshall found that, this. intensity relation did not hold for oxygen and sulfur containing cpm*
pounds, This was later confirmed by Francis. Mi rone and

Fabbri, however, found that primary alcohols did follow the in* tensity additive law. In the case of secondary and tertiary alcohols the intensities were lower than that which the additive law had predicted.
One of the first questions that arose out of the assignment of a
band as to the vibration it represented came from the spectra of
acetone* Acetone gives three peaks of comparable intensities.
The explanation offered by Francis was that this extra band might be due to the degenerate vibration of the free "CH^ group perhaps split from the interaction of the two methyl groups. A more elab-orate study was carried out by Nolin and Jones on ketones, in which the hydrogen of the methyl groups, the methylene groups, or both, were replaced by deuterium. The deuterium substitution was conducted in such a way as to eliminate either the methyl or methylene C-H stretching vibration or both. When the methylene hydrogens were replaced, three peaks were observed, and when the methyl hydrogens were replaced, only one band was observed. This led Nolin and Jones to assign three bands to the methyl C-H stretching vibration and only one to the methylene C-H stretching vibration. It was found that the greater the strain of a ring hydrocarbon,
the greater were the changes in the intensity and frequencies of its
C-H stretching vibrations. Roberts and Chambers found, starting with a six membered ring, that as the number of carbon atoms in a

ring decreased, the intensity decreased. They showed this by giving
the extinction coefficients for the various ring compounds.
An extensive work on cyclobutane was conducted by Rathgens 42
and co-workers. They determined the symmetry and the entropy
value using infrared and Raman spectra and electron diffraction.
The aromatic C-H stretching vibrations were also determined by
Fox and Martin,8 They found bands near 3038 cm, which they
assigned to these C-H stretching vibrations. In general, three bands
are found unless the compound is heavily substituted. In nearly all
cases one of the bands in this region is considerably more intense 5
than the others. The C-H stretching frequencies for multiple-ring
systems has also been studied, Fuson and Josien, who studied substituted benzanthracenes, discussed the possible correlations between the C-H frequencies and the carcinogenic activities of these compounds,
Halogenation of hydrocarbons also produces shifts in the C-H
stretching vibrations. Theimer and Nielsen have investigated the
spectra of fluorinated ethanes and gave the value 3017 cm. to the C-H stretching vibration for pentahalo-ethane as compared to Fox
and Martin's value of 2890 cm, for stretching vibration of the
\' 45
pVL group. Weissman, Meister, and Cleveland found similar
shifts using halogenated methanes. Their average value was given -1
as 3026 cm. Even higher C-H stretching frequencies were reported

reported by Mann, Acquista, and Plyler using halogenated ethylene.
Their highest value reported was 3130 cm. Fluorinated aromatics
4.7, 4g
were also studied by Nielsen and co-workers, "* and they assigned
the band at 3095 cm. as due to the C-H stretching vibration.
Many other informative studies on the 2 to 4 micron region have also been conducted. The above discussion was limited to those works which were of interest in the present study. For a more extensive bibliography on infrared studies a text on that subject should
be consulted, Bellamy lists many references at the end of each chapter in his text. Most of the articles consulted in this study gave further references.
Although much work has been done on the 2 to 4 micron region of the infrared spectra, it still holds many unanswered questions. The present study was carried out in the hope that some of these problems might be elucidated.

A Perkin-Elmer Model 21 double-beam infrared recording spectrophotometer was used in this research* This spectrophotometer is normally equipped with a sodium chloride (NaCl) prism, but is designed in such a way that the prism can be changed. It was decided to use the calcium fluoride (CaF^) prism, which has better resolution than the NaCl prism in the range of this investigation. Figure 1 illustrates the difference in resolving power of the NaCl and CaF, prisms.
The range of the spectrophotometer with the CaF_, prism is from 5000 cm. to 1050 cm.* ;. This study, however, was concerned only with the hydrogen stretching region which lies between 5000 and 2500 cm,"1.
For a more detailed description of the spectrophotometer and the ranges and resolving power of the various prisms, the manuals prepared by the Perkin-Elmer Corporation should be consulted.
Whenever the prism of the spectrophotometer is changed, several calibrations must be made- These are also described in

< Frequency in cm.
5000 3333 2500 5000 3333
Sodium Chloride Calcium Fluorid
Figure 1. Comparison of Resolving Power of Calcium Fluoride and Sodium Chloride Prisms Using CF =CFCH CH_Br

the manuals which are supplied with the prism and its interchange unit.
Because the research was mainly interested in the position of the bands, rather than their relative intensities, the most important of these calibrations was that of the wavelength. Two changes
were made in the procedure given in the manual for this calibration.
Since the research was concerned only with the range of 5000 cm. -1
to 2500 cm. the first of these changes was that the long wavelength band calibration was omitted. The band usually used for this
calibration is the 1350 cm. -1 H.,0 band. The other change was,
-1 -1
instead of using the 3337 cm. NH^ band, the 3759 cm. atmospheric
HLjO band and the 3062 cm, ~* C-H stretching band of a 0, 07 mm. polystyrene film supplied with the instrument were used.
A frequent check was made of the wavelength using the 3759 cm. atmospheric Hj,0 band. The wavelength stayed constant, and no adjustment was needed until the prism was changed.
All spectra were determined with the gain set as illustrated in the instruction manuals. The scanning time was generally about 11/2 minutes per micron. However, when a major band appearea, the speed was reduced and the instrument stopped at the peak of the band, and the wavelength was then recorded from the counter on the prism drive. The region of these peaks was predetermined by a rapid preliminary scan of the spectrum. This preliminary scan

also served to determine if the concentrations were large enough to give an absorption between 50% and 85% for the major peaks. In some of the solution work the solubility and available cell size prevented obtaining such an absorption. This will be discussed in more detail in the sections dealing with the various types of pounds.

Because the compounds under investigation were in the gaseous, liquid, and solid states, a variety of techniques was needed. The gaseous samples were run in two gas cells, a 5 cm. path length cell and a 10 cm. path length cell. The cell was first evacuated, and then the gaseous sample was allowed to fill the cell to the desired pressure, measured with a manometer attached to the gas cell filling apparatus* The pressure at which the samples were run depended on that required to give the major bands an absorption be* tween 50% and 85%* This pressure varied among the samples depending on the infrared activity of the compound being studied* The extremes were 5 cm* Hg pressure in the 5 cm* cell to 60 cm* Hg pressure in the 10 cm. cell*
The choice of cell also depended on keeping the absorption of the major bands between 50% and 85%* The 10 cm* ceil was used when the 5 cm* cell required a pressure greater than about 60 cm. of Hg pressure.
The majority of the compounds investigated were liquid. This included compounds with varying degrees of absorption, and therefore various sample sizes were needed. The compounds with high absorption, such as the fluorinated alcohols, were run in demountable cells. The rapid evaporation of some of the lower boiling compounds made it difficult to keep a sample in the beam. To overcome

this, a 0. 015 mm. spacer was placed between the salt plates of the demountable cell, and the spectra were scanned faster. A preliminary run was made to determine the approximate strong bands, and then the final run was made ranidly up to these regions. Since the absorption of the fluorinated hydrocarbons was weaker than that of the alcohols, fixed cells were used. The cell size was determined by obtaining the desired amount of absorption* Three different cell sizes were used to run the pure compounds. These were the 0* 027 mm., 0. 05 mm. and 0.1 mm. fixed cells. As the number of halogen atoms per molecule increased, a larger ceil was needed to obtain the desired amount of absorption.
The compounds containing OH and NH were run in carbon tetrachloride solutions. These solutions were treated in the same manner as the pure liquid, except that the size of the cell used was larger. Due to the limited solubility of some of these compounds, the 3.0 mm cell, which is the largest cell available at this laboratory, had to be used.
Generally when compounds are run in solution, two cells of approximately the same size are used. One contains the solution and is placed in the sample beam, while the other contains only the solvent and is placed in the reference beam. This allows the instrument to compensate for any band due to the solvent. Carbon
tetrachloride, however, has no major band between 5000 cm, and

2500 cm. In the largest cell, the 3.0 mm. cell, only one very weak peak at about 3850 cm,"1 was observed. This band has a lower wavelength than any other band observed in the compounds of this investigation and no compensation was needed. This band did not appear when any cell smaller than the 3. 0 mm. was used.
The solid compounds required considerably more time for investigation than either the liquids or gases. There are several
methods for obtaining the spectra of solids. A list of most of these
methods is given by Silas, In this investigation only three methods were used. The solids were run as melts, mulls, and in solution. The compounds that had melting points below 2Q0C were run as melts, In this method, some of the solid is placed between the salt plates of a demountable cell and is heated until it melts, The cell is then tightened to give the desired thickness while the substance is still in the liquid state. The method works well as long a compounds do not decompose, change color and char, or form an opaque film when the cell cools.
The mull method suspends a finely ground sample in some mulling oil or grease. In this investigation, fluorinated hydrocarbons with all of the hydrogen replaced by fluorine or chlorine were used. These are normally called Kel-F oils. Several of the
?Trade mark of Minnesota Mining & Manufacturing Company. However, this name is generally used by spectroscopists.

Kel-F oils used varied in viscosity. Although one oil would have been satisfactory, several were tried in order to determine if one produced better mulls than the others. Some compounds made mulls easier with a more viscous oil, while others did better in a lighter oil. The mull technique was used when a compound was hard enough to give a fine powder upon grinding and when the melting point of the compound was too high to run as a melt.
Carbon tetrachloride solutions of the solids containing OH and NH were run as described above. One series of NH compounds, the perfluorinated amides of the dibasic acids, was insoluble in carbon tetrachloride, and no appropriate solvent could be found.

Preparation and Purification of Compounds
Most of the compounds used in this investigation were obtained either from laboratories of this University or commercial companies. One series, however, had to be prepared by the investigator. These compounds were the hydrocarbons with all but one hydrogen remacea by fluorine. These are sometimes referred to as the fluorocarbon hydrides. The general method is to prepare the sodium salt of the perfluorinated acid, then decarboxvlate it in the presence of ethylene glycol between 170*G. and 190*C., and thus obtain the corre-50
sponding hydride. The decarboxylation of sodium perfluoracetaio may be represented by the following equation:
2CF,COONa + HOCH,CH0OH 2CF.H + 2CO, + NaOCH,CH,ONa 3 2 2 3 2 2 2
Table I lists the boiling and melting points of the hydrides prepared. It can be noted that the boiling point of CFH is below that of "Dry Ice"-acetone mixture, therefore, liquid air had to be used to collect tne product. In the other two cases, the "Dry Ice" -acetone mixture was used to obtain the product.
H. C. Brown and J, A. Young of Reed Laboratory of the University of Florida supplied the nitrogen-containing fluorocarbons. P. Tarrant and his group in the Chemistry Department supplied most of the fluorinated hydrocarbons and some of the fluorinated alcohols and acids. The remaining compounds were either obtained as samples from various companies or purchased from them.


Compound Boiling Point (*C.) Melting Point CO
CF3H- -84.4 -160
C2HF5 -48 -103
C HF -18.5
3 7
Adapted Irom "Aliphatic Fluorine Compounds"
A few of the compounds had to be repurified. Distillation at reduced pressure was used since allof them were liquids.

In order to facilitate a better comparison of the spectra of the fluorinated compounds* the spectra of the unsubstituted compounds were run whenever they were available. The frequencies obtained for the various vibrations from the spectra of these unsubstituted compounds are used as the standard for comparison. In most the values obtained in this study were in good agreement with those reported in the literature. The values of the various frequencies are listed in tables in this chapter, and the following notations are used to indicate the relative band intensities: vs = very strong, _s strong, m = medium, and w = weak. The above notations are used when the bands are well defined sharp peaks with an accuracy of , 4 cm. \ Two other notations are used when the bands are not so well defined and the accuracy may be no better thani. 15 cm. These notations are: shfor a band that appears as a shoulder on a stronger band; br for very broad medium to strong bands.
This discussion is divided into three section: the O-H stretching region, the N-H stretching region, and the C-H stretching region.

Some compounds are found in two sections, but only the particular bands of these compounds belonging to any one section are discussed in that section.
In all regions fluorination caused a shift in the frequencies. In some cases, the shift was to a lower frequency, while in others it was to a higher frequency. Another observation was that additional bands appeared in the fluorinated compounds that did not appear in the unsubstituted compounds. Most of these bands were unexpected and are believed to be overtones or combination bands of the carbon to halogen, mainly fluorine, bands. When these bands are more than just very weak bands, they are discussed under the various sections.

The Oxygen-Hydrogen Stretching Region
Two classes of compounds were investigated which gave bands associated with the oxygen-hydrogen stretching vibrations. They were alcohols and carboxylic acids, and they will be discussed in that order.
As previously mentioned in Chapter I, the -OH group affords two types of bands, the free O-H stretching vibration and the associated O-H stretching vibration. Carbon tetrachloride solutions were used in order that the free O-H stretching band could be clearly resolved. Table II lists the frequencies of bands obtained from the spectra of the carbon tetrachloride solution of the unsubstituted alcohols. With the exception of methanol, the average frequency of the free O-H stretching band for the unsubstituted primary alcohols was 3649 cm. The maximum deviation for all these alcohols was dL 2 cm. This value is only slightly higher than the value Hated
12 i'
by Cole and Jefferies, who listed it as 3642 cm. for primary alcohols. Three unsubstituted secondary alcohols were run with two of them giving a. vaiue lor the free O-H stretching vibrations
of 3636 cm. and the other giving a value for this vibration of
3635 cm, Only one tertiary alcohol was studied, and its value
for the free O-H stretching vibration was 3629 cm, Two other unsubstituted alcohols were also investigated. One of these alcohols was unsaturated and the other was a diol. The values obtained

Compound Frequency in -1 cm.
CHOH 3 3658s 3355s
CH.CH OH 3 2 3647 s 3503w 3353s
CH CH CH OH o c* c* 3647s 3508w 3339s
CH{CH_),CH,OH 3648 s 3508w 3344s
CH3(CH2)3CH2OH 3650s 3508w 3349s
CHJCHJ CR,OH 3 2 4 2 3648 s 3508w 3353s
CHJCHJcCH,OH 3 2 5 2 3650s 3508v/ 3357s
CHJCH ),CH,OH 5 o 2 3648s 3508w 3344s
CH (CHJ _CH_OH 3 2 7 2 3651s 3508w 3368s
CH3(CH2)8CH2OH 3651s 3508w 3356s
CH3(CH2)9CH2OH 3651s 3508w 3357s
CH CH(OH)CH 3 3 3636s 3496w 3350s
CH CH,CH(OH)CH 5 c 5 3635s 3496w
CH3(CH2)2CH(OH)CH3 3636s 3502w
(CH rCHCH_,OH 5 c. c 3654s 3508w
(CH_).COH 3 5 3629s 3504w
CH =CHC(CH ) OH 2 3'2 3673sh 3623s 3492s 3404sh
HOCHJCH,) CH^OPI 2 2 3 2 3650s 3364br

from their spectra for the free O-H band are listed in Table II.
The spectra of the above alcohols were determined at various concentrations in carbon tetrachloride and, in most cases, the various runs for a given alcoftol were consistent. In any case where there was a difference of more wan 2 cm, for the free O-H peak of different concentrations of a given alcohol, new solutions of that alcohol were made, and the spectra of these new solutions were determined
A weak peak, not nearly as well defined as the free O-H peak, was noted at 3508 cm. 1 for all the unsubstituted primary alcohols. This band did not give a sharp peak for the absorption in the solution
where it appeared as a separate band. It was in some cases as broad
-1 -1
at the top as 10 cm. The value of 3508 cm. was the mid-point
in these cases. In the more concentrated solutions, this band appeared
as a shoulder on the strong, broad polymeric association band and
could not be determined with any accuracy from the spectra of these
' 9
solutions. Kuhn, as mentioned in Chapter I, gives the range of
3525 3472 cm."1 for dimeric or single-bridge association. Since
this observed value of 3508 cm. falls in the middle of this range, the peak is accredited to the dimeric or single-bridge form of the associated alcohols. Both the secondary and tertiary alcohols also gave peaks in this range. These values can also be found in Table II.

In the more concentrated solutions, the polymeric associated form of the alcohols still appeared as strong broad bands, but these peaks were weaker in the less concentrated solutions and disappeared completely in the most dilute solution. Because of the broad nature of these peaks, it is difficult to determine their frequencies with a high degree of accuracy* These peaks also appear to shift slightly to a higher frequency as the solutions are diluted. This could be an actual shift due to the formation of different polymeric forms of fewer molecules, or it may be just a change in the appearance of the band because of the formation of the dimeric shoulder, The shift is probably due to a combination of both of the above reasons. The values listed in the last column of Table II are assigned to the polymeric association and are taken from the spectra of the solutions with a concentration sufficient to produce a well defined peak*
The frequencies listed in Table III are those obtained from the spectra of the same unsubstituted alcohols which were run as pure liquids. Except for methanol and ethanol, only one frequency is listed for each alcohol. This frequency is assigned to the polymeric associated O-H stretching vibration which lies in the interval 3340 7 cm, The secondary and tertiary alcohols gave an average of 3357 cm. with a maximum deviation of J: 1 cm. These values, however, are the mid-points of broad bands with the top of the peak as wide as 10 cm. "1 in some cases. The two additional peaks listed for

Compound Frequency in -1 cm.
CH3OH 3691vv 3352s
CH,CH0OH 5 c 3654sh 3340 s
CH,CH,CH.OH 3 2 2 3343s
CH (CH ) CH OH 3341s
CH(CH )CH.,OH 3 2'3 2 3342s
CH3(CH2)4CH2OH 3346s
CH (CH ) CH OH 5 co c 3336s
CH (CH ) .CH OH 3 co c 3333s
CH3(CH2)7CH2OH 3333s
CH3(CH2)8CH2OH 3336s
CH,(CKJ CH OH 3 2 7 2 3346s
CH,CH(OH)CH 3 3 3356s
CH3CH2CH(OH)CH3 3358s
CH3(CH2)2CH(OH)CH3 3358s
(CH 1 COH 3357s
(CH,),CHCHnOH 5 c c 3342s
CH =CHC(CH ) OH 6* Jj 3404s

methanol and ethanoi are probably due to traces of water absorbed from the atmosphere, since the humidity in these laboratories is a problem. In the case of the perfluorinated acids, which are known to be very hygroscopic, a check on such a band was conducted, and this band was found to be due to water. The method of checking for water is explained below with the discussion of these perfluorinated acids.
The frequencies of the O-H vibrations for the fluorinated alcohols are listed in Table IV and Table V, Table IV lists the values for the carbon tetrachloride solutions, while Table V lists the values for the pure compounds. As in the case of the unsubstituted alcohols, the spectra of the carbon tetrachloride solutions of the fluorinated alcohols were determined at various concentrations. The values listed are from the dilution v/hich gave the best defined peaks.
A comparison of the spectra of a pure compound, and its carbon tetrachloride solution for a fluorinated alcohol, is given in Figure 2. Figure 3 compares the spectra of a fluorinated alcohol and an unsubstituted alcohol.
A problem existed in the fluorinated alcohols which did not occur in the unsubstituted alcohols. This was the limited solubility of these fluorinated alcohols in carbon tetrachloride. Because of this, the saturated solutions of these alcohols in carbon tetrachloride were the first concentrations used. After obtaining the spectra of these

Compound Frequency in cm.
CF CH OH c* 3634s 3489w 3372br
CFCF_CH,OH 3 2 2 3634s 3378s
CF {CF ) CH OH 3635s 3413br
CF3(CF2)6CH2OH 3632s 3498w
CHF CF CH OH & C* Ct 3634s 3435m
CHF2(CF2)3CH2OH 3632s 3500w
CHF (CF ) CH OH 3631s 349 7w
CHF (CF ) CH OH C* Ct ( Ct 3632s 3505w
CHF(CF.)nCH_OH 3631s
CF,CH(OH)CH, 3 3 3623s 3416br
CF CF,CH(OH)CH 3 2 3 3623s 3405br
CF3(CF2)2CH(OH)CH3 3623 s 3420s
(CH3)2C(OH)CF3 3622s 3419br
CH =CHC(CH )(OH)CF * 3 J 3613s 3484m 3309w
HGCH (CF0) CH,OH 2 2 3 2 3628s 3346s

Compound Frequency in cm.
CF3CH2OH 3643sh 3397s
CF0CF_CH,OH J 2 2 3631sh 3359s
CFJCFJ.CH.OH 3 c c c 3631ah 3360s
CF3(CH2)6CH2OH 3634sh 3360s
CHF2CF2CH2OH 3631w 3372s
CHFJCFJCH^OH 2 2 3 2 3630w 3385s
CHF,(CFJ_CH_OH C CO c 3639w 3385s
CHF2(CF2)?CH2OH 3636w 3403br
CHF_ CF,CH(OH)CH, 3 3 3623sh 3389s
CF,CF,CH(OH)CH, 3 2 3 3639w 3386s
CF3(CF2)2CH(OH)CH3 361Ish 3385s
(CH3)2C(OH)CF3 3609w 3396s
CH =CHC(CHJ(OH)CF, 2 3 3 36l2w 3458s
HOCH,(CF_),CH_OH ii CO c 3636w 3413sh 3367s

Figure 2. Comparison of the Spectra of CF^F^H^OH with the Spectra of the Carbon Tetrachloride Solution
Frequency in cm.

5000 3333 2500 5000 3333 2500
r i-, i-1-1
Figure 3. Comparison of the Spectra of CF CF CH OH and CH CH CH OH 3 22
Frequency in cm.

saturated solutions, various dilutions were made and their spectra determined. The agreement among the various solutions was very-good and, in most cases, did not vary more than one or two cm. \ The values for the free O-H stretching vibration were quite consistent within the various groups. There was, however, a disagreement with the values listed by Haszeldine* for some of these fluorinated compounds. Wherever a disagreement occurred, compounds were re-run after the instrument had been recalibrated. The values obtained from these re-runs were within+, 2 cm. 1 of the previously determined values for the particular alcohol.
The primary fluorinated alcohols gave an average value for the free O-H stretching vibration of 3633 cm. and all the alcohols were within a range of +, 2 cm. about this average value. This average value for the primary fluorinated alcohols, when compared to that given by the unsubstituted alcohols for the free O-H band, shows a shift of 16 cm. 1 to a lower frequency.
A plausible explanation for the cause of this frequency shift is that the highly electronegative fluorine, replacing the less electronegative hydrogen of the alcohol, changes the distribution of the electrons between the oxygen and hydrogen. The fluorine atoms, with their great electron-attracting power, pull at the electrons of the carbon, to which they are attached, causing this carbon atom to acquire a slight positive charge. In order to overcome the electron

deficiency, this carbon atom attracts the electrons of the next carbon atom which, in turn* attracts the electrons between itself and the oxygen atom. Since the oxygen has only a slightly lower electronegativity than fluorine, it will not release its electrons without compensation for the loss, Thus, the oxygen atom pulls the electrons closer to itself from the bond between it and the hydrogen of the -OH group. This displacement of electrons changes the forces acting on the bond, and it also changes the strength of the bond. The displacement of these bonding electrons allows the hydrogen to vibrate easier, thus, requiring less energy and giving a lower vibrational frequency.
This weaker oxygen-hydrogen bond implies that the hydrogen can be removed easier and these fluorinated alcohols should be stronger acids than the similar unsubstituted alcohols. This v/as found to be the case by Haszeidine, 11
The single fluorine atom is more electronegative than the -CF^ group. This would imply that, if the fluorine were replaced by the -CF^ group, some of the electron-attracting power, causing the displacement of the electrons of the -OH group, would be lost and the shift in frequency would be less. This was, however, not observed, probably because the replacement of one fluorine causes a shift too small to bs detected by presently available instruments.
The secondary fluorinated alcohols did not shift the frequency of the free O-H vibration as much as the primary fluorinated alcohols.

The shift of the frequency for the free O-H vibration was even less for the tertiary fluorinated alcohol. These shifts were 13 cm. for the secondary fluorinated alcohols, and 7 cm. for the tertiary fluorinated alcohol respectively. An extension of the reasoning used above can explain these smaller shifts. The -CH^ group is a better electron-donating group than the hydrogen atom. All the fluorinated secondary alcohols had fluorine atoms on only one side chain with the -CH^ group as the other side chain. The tertiary alcohol also had fluorine on only one chain with two -CH^ groups making up the other side chains. This, then, leaves this -CK^ group of the secondary alcohols free to absorb some of the electron-attracting forces of the fluorine, and thereby reduces the effect on the oxygen-hydrogen bond, thus causing a smaller frequency shift. Furthermore, in the case of the tertiary alcohol, the effect on the oxygen-hydrogen bond was even less because of the addition of the second -CH, group.
For some of the fluorinated alcohols, a weak band was found near 3500 cm. With other fluorinated alcohols, the concentration needed to resolve clearly a peak near this value was not obtained. Some of these compounds gave a shoulder near this frequency. However, the frequency of this band could not be determined with any accuracy and no values are listed in Table IV. The average value obtained from five of the primary fluorinated alcohols, which gave fairly well defined peaks for this band, was 3498 cm. Three of these compounds

gave values within 2 cm. 1 while one was 9 cm. low and the other 7 cm. high from the average value, This band had the same general appearance of the band assigned to the dimeric form for the unsubstituted alcohols, and it is, therefore, assumed to be due to the dimeric form. U five compounds gave a lower frequency than the average value of the primary unsubstituted alcohols, The average value for these primary fluorinated alcohols was 10 cm. lower than the average value for the primary unsubstituted alcohols.
The introduction of fluorine atoms increases the acidity of the alcohol.This implies that there is an increase in the effective positive charge on the hydrogen atom which will increase the strength of hydrogen bonding. This change in the hydrogen bonding might account for the frequency shift.
One other fluorinated alcohol gave a distinct peak which can be accredited to the dimeric form. This alcohol was an unsaturated secondary alcohol and gave a value of 3484 cm. for the frequency of the band. This value, compared to that given by the equivalent unsubstituted alcohol, shows only a slight shift of 8 cm. 1 to a lower frequency,
The remaining frequencies listed in Table IV are attributed to various polymeric associated forms. These values are not used for comparison since they were too varied among the different dilutions which were used,

The values listed in Table V are for these same fluorinated alcohols run as the pure compounds. At least tv/o values are given for each alcohol. The first value listed is that for a weak peak or a sharp shoulder on a stronger band. The average value found for these peaks from the primary fluorinated alcohols was 3635 cm. This frequency has already been attributed to the free O-H stretching vibration. The average value for this band obtained from the spectra of the pure compounds differs from the average value given by the carbon tetrachloride solution of these compounds by only 2 cm. This implies that there is little or no solvent effect. Similar results were obtained for the secondary and tertiary fluorinated alcohols.
Only one other frequency is listed in Table V fdr all but two of the fluorinated alcohols. In these two cases, the additional bands were distinct shoulders on the stronger peaks. Both of these alcohols were solids, and this could account for the additional peak. The major peak listed in Table V is accredited to the polymeric associated O-H stretching vibration. The agreement was only fair if the two longest chain alcohols, which are solids at ordinary temperatures, and trifluoroethanoi are included. Excluding these
three alcohols, an average value for fluorinated primary alcohol
-1 -1
was found to be 3371 cm, with a maximum deviation of . 14 cm.
The values for the three alcohols not included can be found in Table V.
All had frequencies at least 25 era, higher.

The higher frequencies of the longer chain fluorinated alcohols can be attributed to the fact that their spectra were determined as melts, while the spectra of the other fluorinated alcohols were run as liquid films. The apparant deviation of trifluoroethanol may be attributed to that compound having an additional fluorine atom close to the -OH group.
A shift of 31 cm. was found upon comparing the average values for the fluorinated and unsubstituted primary alcohols. Trifluoro-ethano! gave a shift of 57 cm. '\ The shifts produced by fluori-nation were to higher frequencies.
This shift to a higher frequency would not be expected from the explanation offered for the shift of the frequency of the dimeric form. This implies that there must be additional effects.
One additional effect, due to fiuorination, is that the basicity of the oxygen atom is reduced. This loss of basicity causes a reduction in the strength of the hydrogen bonding. Apparently when there is a strong polymeric association, the reduction of the basicity of the oxygen atom produces a greater change in the hydrogen bonding of the compound than the change caused by the increase in the effective positive charge on the hydrogen atom. Steric effects are also possible. However, they are not likely to occur in these cases since the fluorine atom is only slightly larger than the hydrogen atom which it has replaced.

Again it was noted that fluorination caused a smaller shift in
the secondary and tertiary alcohols than in the primary alcohols.
The buffering of the effects of the fluorine by the -CH groups, as
discussed in the explanation of the shift in the free O-H stretching vibration, is again thought to be the case.
Hydrogen bonding through the fluorine atoms is also a possibility for theae fluorinated alcohols. This should lead to another absorption peak. The hydrogen bonding arising from the association of the fluorine and hydrogen atom is somewhat less than that association attributed to the oxygen and hydrogen atom. This weaker hydrogen bond would have less effect on the O-H stretching vibration, .-.'.id would lead to a smaller shift in the frequency at which the vibration would absorb. The absorption of the O-H vibration due to this fluorine hydrogen bond should appear somewhere between the absorption of the free O-H stretching vibration and the absorption due to the polymeric association through hydrogen bonding involving the oxygen atom.
Many spectra of these fluorinated alcohols were examined, including both the spectra of the pure compounds and various dilutions of their carbon tetrachloride solutions. No band was found that could be assigned to hydrogen bonding involving the fluorine atoms. If this band does exist, it probably occurs at a frequency so close to that of the strong broad band assigned to the polymeric association,

that it is completely incorporated in this broad band. This band was probably not resolved in solution because the fluorine-hydrogen bond has a weaker effect than the polymeric association of the oxygen attracted hydrogen bond, and it is more readily broken than the polymeric associated bond.
The O-H stretching frequencies of carboxylic acids were also studied. This study not only included the fluorinated and non-halogenated acids, but it also included several chlorinated acids and a chlorinated and fluorinated acid* The spectra of many of these acids were determined for the pure substances. These spectra, however, contained very little useful information for this study. This discussion is limited to the carbon tetrachloride solutions of various concentrations of these acids. The values obtained from these solutions are listed in Table VI.
A weak, but very sharp, peak was obtained from the spectra of the dilute carbon tetrachloride solutions of the non-halogenated aliphatic acids. The average value for this peak was 3542 cm. 1 with a maximum deviation of 5 cm. The peak has the general appearance of the free O-H stretching vibration band, and the peak is considered to be produced by this vibration,
In the halogenated acids two peaks close to this value were observed. The peak occurring at the highest frequency was believed to have been caused by water. In order to elucidate this, water was

Compound Frequency in cm. "
CH COOH 3 3537w t 3093sh 3024sh 2999br 2735sb. 2681s 2627s 2552m
CH CH COOH i 2 3547w 2755sh 2653br 2551m
CH (CH^) COOH 3541w 3028sh 2755sh 2646m 2604sh
CH {CH ) COOH 3 2 4 3544vv 3025sh 2641m
CH{CHJ,CGOH 3 2 5 354lw 3028sh 2755sh 2657m
CH CH=CHCOOH 3 3541 w 3040sh 2963s 2900s 2699br 2591m 2540m
C.HCOOH 6 5 3544w 3049s 3020br 29.67oh 2873s 2670s 2589s! 2540m
p-CH C.KXOOH 1 3 6 4 3552w 3076s 3038br 2998sh 2877br 2674s 2597m 2545s
CI-I CiCOOH 2 3529w 3106sh 3011s 2926br 2690m 2579m
CHC1 COOH 3520m 3106sh 3012s 2907br 2635m 2565m
CClCOOH 3 3511s 3094br 3032sh 2890sh 2618s 2432m
CC1F COOH 3506s 3076a 3048sh 2933sh 2679s 2534s

Frequency in cm.
CF3COOH 3505a 3098br 3075sh 2909sh 2732br 2577m
CF CF COOH 5 C* 3503s 3075br 3040sh 2923sh 2686m 2532m
CF3(CF2)2COOH 3500s 3075br 2960sh 266 7w.
CFJCFICOQH 5 2 o 3503s 3076br 2967sh 2680w 2528w
C1(CF -CFCi) CF COOH Ct Ct c* 3500s 3070br 2921sh 2681w 253 lw
C1(CF -CFCl) CF COOH 2 3 2 3504s 3070br 2945sh 2680m 2529m
CH3CH=CFCOOH 3540m 3049br 2941sh 2895br 2690m 2646m 2586m 2534m
(CH ) C=CFCOOH J Ct 3546m 3033br 2904br 2886sh 2708m 2660m 2565m 2522m
C,H_CH=CFCOOH 6 5 3540w 3058sh 3028s 2881sh 2673m 2611m 2531m
p-FC.HXOOH 3546w 3074sh 2998br 2865br 2667s 2597sh 2535s
p-CBrF CFCIC,H,,COOH c Z 6 4 3532w 3090sh 2990br 2882br 2667s 2540 s

added to the carbon tetrachloride solutions of these acids to determine whether there was any change in this band. The method used was to prepare the carbon tetrachloride solutions of these acids and then determine the spectra of the solutions. Using the same solutions, small amounts of water were added and the spectra of the new solutions were determined. In all cases the addition of water increased the intensity of the band,
A second unexpected effect was observed. This was that the intensity of the other peaks in the region under study (5000 cm, 1 to 2500 cm. *) became less. At first it was thought that this addition of water had affected the vibrations in this region, and perhaps an explanation of the submaxima or satellite peaks would evolve. Since the intensities of both the submaxima and the O-H stretching bands had decreased, this might imply that these submaxima peaks were related to the O-H stretching peaks through some type of combination band. In an effort to resolve this point, a check was made on other bands by determining the spectrum at longer wavelengths. It was felt that if the carbonyl band remained constant the water must have had a definite effect on the O-H stretching vibration. The results obtained indicated that the intensity of the carbonyl band decreased at approximately the same rate as the O-H stretching band. This did not definitely prove that the O-H stretching vibrations were not affected. It did indicate that the major effect was the replacement of

some of the acid molecules in the radiation beam by the water molecules. There was an indication that more work is needed on some of the questions arising from the addition of water to these acids. However, the point that the questionable bands were due to water was resolved, The values obtained for this band are not included in Table VI.
The other band appearing near 3500 cm, was a stronger band and is assigned to the free O-H stretching vibration. The three chlorinated acids, the mono, di, and tri substituted, showed an increase in intensity and a decrease in frequency for this vibration with a given concentration, as the number of chlorine atoms on the molecules increased. The fluorinated acids, except for the unsaturated and aromatic fluorinated acids, gave an average value of 3503 cm. 1 with a maximum deviation of 3 cm. A comparison of this average value with that of the unsubstituted acids indicates a shift of 39 cm. < to a lower frequency, The same explanation offered for the shift of the free O-H stretching vibration of the fluorinated alcohols should apply here.
The loosening of the oxygen-hydrogen bond, resulting from the changing of the atom or atoms on or near the carbon on which the -OH group is attached, is well known through the study of the ionization constant. The following series illustrates this relationship. The ionization constant increases from left to right in the series:

and CF^COOH. The ease of removing the hydrogen from the -OH group depends on the ionization constant* This implies that the greater the ionization constant the looser the oxygen-hydrogen bond* The frequencies of the free O-H stretching vibration also decrease from left to right in this series. The lower the frequency the easier the bond vibrates, and this indicates a looser bond.
The unsaturated acids investigated had only one fluorine atom per molecule, which was located on one of the carbon atoms of the double bond. There was no noticeable shift in the frequency of the free O-H stretching band in these compounds. This may be due to the absorption of the effects of the fluorine atoms by the double bond. Furthermore, the fluorinated aromatic acids showed little or no shift.
In most of the non-halogenated acids, the C-H stretching vi-
brations, which have frequencies near 2900 cm. interfere with
the O-H associated vibrations when the spectra are determined in
carbon tetrachloride solutions. Almost all of the frequencies given
in Table VI with values near 2900 cm. for these non-halogenated acids are values obtained from shoulder bands, therefore, no good comparison in this region can be made. There are no values listed for CHCH COOH in this region because the C-H stretching bands
-15 Ct
were so strong in the concentrations used that they incorporated all the O-H stretching bands.

The values listed for the halogenated acids have good agreement. All these acids, except the unsaturated and aromatic fluorinated acids, gave bands between 3100 cm. and 3070 cm. Almost all of these bands were rather broad in appearance, and their mid-points are the values listed.
The region betv/een 2800 cm. and 2500 cm. where there is no interference from the C-H stretching vibrations, contains additional peaks, generally referred to as submaxima or satellite peaks. Not only did the frequencies and the intensities of these bands vary among the compounds, but the number of bands appearing also varied. Figure 4 illustrates these vibrations, since it shows an acid with only two submaxima peaks and an acid with four submaxima peaks,
As previously mentioned in the discussion in Chapter I, these submaxima have caused great concern and many investigators have tried to explain them. A shift in the frequency of the submaxima peaks near 2650 cm. resulting from fluorination, is indicated. The shift is to a higher frequency similar to that given by the associated O-H stretching vibration. This might imply that there is a relation between the submaxima bands and the associated O-H stretching band. However, there is no definite proof of such a relation. No further explanations are offered in this discussion for these additional bands.

Figure 4. Comparison of the Spectra of CF COOH and (CH ) C=CFCOOH

The Nitrogen-Hydrogen Stretching Region
The present study was somewhat limited in the investigation of nitrogen-containing compounds, since only one type of compound was investigated. This was the type which has the nitrogen attached to at least one carbonyl group, A complete list of the nitrogen-containing compounds investiaged is given in Table VII, This table also lists the frequencies obtained from the spectra of the pure compounds. Table VIII lists those nitrogen-containing compounds which were soluble enough in carbon tetrachloride to produce a spectra capable of being interpreted. The frequencies of the band3 obtained from these spectra are also listed in Table VIII.
Almost all of these nitrogen-containing compounds were amides. Several problems were encountered in attempting to obtain good spectra from the amides. The limited solubility of these amides was the major problem. Some of the amides were so insoluble in carbon tetrachloride that no spectra could be obtained. Other solvents were tried, but no suitable solvent could be found. In order to use a solvent, it had to have several specific properties, The three main properties were: the solvent had to dissolve enough of the amide to give a well defined spectrum; the solvent could not dissolve the NaCl cell used to hold the solution while the spectrum was being determined; and the solvent should not give any major absorption band in the range to be studied. Several freons, as well as tetrafluorodichloroacetone, were tried without success.

Frequency in era.
CH3CONH2 3351s 3205sh 3194s
CH,CH0CONH0 3 2 2 3366s 3313sh 3207s
CH3(CH2)2CONH2 3360s 3311sh 3205sh 3194s
H NOCCH CONH C* c C 3341s 3160s
H2NOC(CH2)2CONH2 3340s 3164s
H2NOC(CH2)4CONH2 3368s 3173s
CF CONH 3508s 336lsh 3312s 3205s
CF CF CONH J Ct Ct 3506m 3380br 3279br 32Q5sh
CF (CF_),CONH_ J Ct Ct Ct 3511s 33663 3311sh 3200s 3042m
H2NOCCF2CONH2 3412 3240sh 3192s
H2NOC(CF2)2CONH2 3350s 3194s
H2NOC{CF2)3CONH2 3381s 3240sh 3195s
2811m 2824m

Frequency in cm.
H NOC(CF ) CONH 3405s
Ct Ct Q Ct
OC(CF0)0CONH 3521m
I 3 3 |
CFJCF_),C(:NH)NH_ 3383s
3 c. o c.
(CFJXONH 3336sh
CF CONH(CH ) 3472m 3323s
j Ct
3248sh 3201s
3268s 3163s 2845m
3200s 3014br 2876sh 2825sh
3282s 3044s
3120m 2952m 2810m
3083s 2094s 2769m
3098s 2884s 2723br

Compound Frequency in cm.
CH CONH 3551s 3513s 3425s 3172v/
CH_CH_CONH, 3551s 3512s 3424s 3186w
CH3(CH2)2CONH2 3550s 3512s 3424s 3l6lw
CF.CONH, 3 2 3545s 3502s 3425s 3353w 3238m 3179m
CF3CF2CONH2 3546s 3505s 3426s 3342w 3238m 3175m
CF3(CF2)2CONH2 3546s 3501s 3425s 3333w
CO(CFJCONH i 3 d i 3554m 3350s 3248m 3155m
(CF ) CONH ,24 i 3534w 3488m 3416m 3303w
CF,(CF_),C(:NH)NH, 5 C o Cm 3530w 3409w

The other major problem that these amides presented was that most of them were hard solids with high melting points. This prevented the determination of the spectra of these compounds as melts, In order to overcome this, the spectra of the high melting amides were determined as mulls, using Kel-F oils (see footnote, page 29). The commonly used mulling oils could not be used in making these mulls since these mulling oils absorb in the 5000 cm, to 2500 cm, region. The Kel-F oils used have no hydrogen and, therefore, have no major bands in this region.
As in the case of the -OH group, the -NH^ group gives both free and associated hydrogen stretching vibrations, when the spectra of the pure compounds are determined, only the associated band usually appears. In most cases, the free N-H band is only realized in solutions where the molecules are far apart. Like the fluorinated alcohols, which showed the free O-H stretching vibration in the spectra of the pure compounds, some of the fluorinated amides showed the free N-H stretching vibration in the spectra of the pure compounds. v/ The first three members of both the fluorinated and non-fluori-nated mono-basic aliphatic amides were soluble enough in carbon tetrachloride to obtain suitable peaks from their spectra. The first three frequencies listed in Table VIII for these compounds are from very sharp peaks with approximately equal intensities. The remaining values are from bands due to various polymeric or dimeric

associations. These bands were not nearly as well defined as the first three bands, and will not be discussed further, The other compounds listed are fluorinated compounds with no hydrogen analogue studied. The spectra from these compounds showed mostly medium to weak bands. It appeared that the solubility of these compounds was less than that of the mono-amides.
The first frequencies listed in Table VIII give an average value of 3551 cm. for the non-fluorinated amides and an average value of 3546 cm. for the fluorinated amides. Only one compound in each series gave frequencies that varied from the average values, and in both cases the deviation was only 1 cm, low. Because of the excellent agreement and the sharpness of the bands, the difference of 5 cm. 1 between the two series is thought to be a true shift to a lower frequency, instead of just an experimental deviation.
This first frequency is close to the value assigned by Busweil and 28
co-v/orkers to the symmetric free N-H stretching vibration.
The agreement among the second listed frequencies is not as good as that of the first values. However, there appears to be a definite shift of 10 cm. 1 to a lower frequency when the compounds containing fluorine were run. No corresponding band was listed by Busweil. The last of the three major peaks all have frequency within 2 cm. 1 of each other, thus, there appears to be no shift in this band. The frequency of this band agrees with the asymmetric N-H stretching band listed by Busweil.

Bellamy mentions that additional bands have been observed in solutions of moderate concentration, and he suggests that different types of association are occurring simultaneously.
In order to check the frequencies of the band and the appearance of these additional bands, new carbon tetrachloride solutions of these simple amides were prepared, and the spectra of these new solutions were determined. There was excellent agreement both with respect to the frequencies and the appearance of the extra peaks. This check was necessary since there was only the spectra of one solution, the saturated solution, for each compound determined.
The shifts in the first and the second bands and not in the third band might suggest that there is some relationship between the first two bands. The shift to a lower frequency, as in the case of the free O-H stretching vibration, implies that the electron-attracting power of the fluorine has caused a change in the nitrogen to hydrogen bond. Apparently different amounts of energy are required, since separate bands appear for the free N-H stretching vibration.
Fluoridation of the compounds produces different effects on these two vibrations. A possible explanation of these effects can be found on examining the directions in which the atoms move in the vibrations. The following diagrams illustrate the symmetric and asymmetric vibrations that are infrared active for these atoms:

symmetric vibration
asymmetric vibration
The pull on the electrons that the fluorine atoms exert is directed away from the nitrogen atom similarly to the vibrational notation in the symmetric vibration in the diagram above. The additional pull in that direction may aid the vibration, and cause the vibration to occur at a lower frequency. In the case of the asymmetric vibration, the pull due to the fluorine would be at right angles to the movement of the nitrogen atom, thus producing only a small resulting effect in the same direction as the nitrogen atom moves. There is, possibly, a small effect on the asymmetric vibration, too small to be detected by present means. If this explanation is valid it supports Buswell's assignments of symmetric and asymmetric vibrations.
No definite explanation is offered for the additional bands between the symmetric and asymmetric vibration. However, since the band indicated a shift when the molecules were fluorinated, it is felt that the additional band might be some type of combination band involving the symmetric free N-H stretching vibration.
Some doubt that the combination band idea is the answer to these additional peaks arises from the results of the spectra obtained from the pure compounds. The simple fluorinated amides gave peaks with

values very close to the frequency of the additional bands. The frequency of the band obtained from the spectra of the pure compound was only about 5'cm. from the values given by the middle band in the carbon tetrachloride solutions. It was approximately 40 cm. ^ less than the symmetric free N-H stretching vibrations and about 80 cm. greater than the asymmetric free N-H stretching vibrations. Even though this band was almost 40 cm. from the position of the symmetric free N-H stretching vibration, it could still be this band. This shift in frequency can be attributed to the difference in the states of the compounds determined. The pure amides were all solids, and their spectra were determined as either melts or mulls. A change in the state of a compound has been known to cause such a shift.
A comparison of these fluorinated amides to the fluorinated alcohols, which gave the free O-H stretching vibrations in the spectra of the pure compound, indicates that perhaps a shift has occurred and that this is one of the free N-H stretching vibrations. The other frequencies listed for both the fluorinated and non-fluorinated simple amides, determined from the pure compounds, had very similar values, and they indicated no shift had occurred in the associated N-H stretching vibrations of these compounds.
This was not the case, however, for the diamides. Although the agreement among the fluorinated diamides was not too good for

the highest frequencies listed, there is an indication that a shift to a higher frequency for this associated vibration occurred. This apparent shift to a higher frequency is probably due to the loss of strength in the hydrogen bonding, and the explanation offered for the similar shift in the polymeric associated O-H stretching vibration should also apply here.
No band corresponding to the second values listed for the fluorinated diamides appeared in the spectra of the non-fluorinated diamides. One possible explanation is that these peaks, which only occur in the fluorinated diamides, might be due to hydrogen bonding involving the fluorine atoms.
The third major peak listed for the fluorinated diamides showed an increase in the frequency as the chain length of the amide increased. A similar trend was observed for the non-fiuorinated diamides. A definite shift in this frequency was indicated by comparing the fluorinated and the non-fluorinated compounds. The same reasoning which was applied to the shift of the first associated N-H stretching vibration is offered for this shift.
Only the non-fluorinated simple aliphatic amides and diamides were obtained, therefore, none of the other fluorinated nitrogen-containing compounds could be compared, and there could be no assurance that the frequency shift was due to the fluorination of these molecules. However, interesting trends can be found by comparing

the various types of fluorinated nitrogen-containing compounds. Two
secondary amides containing phosphorous and chlorine, when compared
to a methyl substituted secondary amide, indicated a tremendous shift
of the N-H stretching vibrations. The major band with the highest
frequency for these phosphorous compounds appeared at a frequency
of only about 3090 cm, while the highest frequency for a major
band in the spectrum of the methyl secondary amide appeared at
3323 cm. The frequency for the band of the methyl secondary
amide is well within the range listed by Bellamy for the associated N-H stretching vibration. If the band obtained from the spectra of the phosphorous compounds is the N-H stretching vibration, the addition of this phosphorous group has had a very great effect on the nitrogen-hydrogen bond, The bands from the spectra of the phosphorous compounds are very similar in appearance to the band obtained from the methyl secondary amide.
The cyclic nitrogen-containing compounds also produced interesting spectra. The possibility of frequency shift3 can be seen by examining the values listed for the frequencies of these compounds in Table VII.

The Carbon-Hydrogen Stretching Region
A large number of compounds containing various types of carbon-hydrogen groups were studied. The fluorinated alcohols, discussed in the O-H stretching section, also contain carbon-hydrogen groups. All of these alcohols had at least one methylene group since they contained the -CH^OH group, Both aromatic and aliphatic fluorinated hydrocarbons were studied. Some of the compounds contained other halogens. The aliphatic series included a group of cyclo-butanes, some unsaturated hydrocarbons, and both branched and straight-chained saturated hydrocarbons. All types of tin carbon-hydrogen groups except the =CH were represented.
In order that a better comparison could be made with the spectra of the fluorocarbon hydrides, the spectra of some of the hydrocarbons containing the other halogens were determined. The spectra of the first four members of the alhane series and ethylene were also run.
Table IX lists the frequencies of the C-H stretching vibrations for the saturated and unsaturated hydrocarbons. The frequencies of the C-H stretching vibrations for the fluorinated and non-fluorinated alcohols are given in Table X. Table XI and Table XII list the values for the C-H stretching vibrations for the cyclobutanes and the aromatic compounds respectively. A few of the compounds did not give suitable spectra. The values from these spectra are not included in

Compound Frequency in cm. ~
CHF3 3037s
CHCl3 3026s
CHBr, 3026s 3034a
CHI, 3 3018s
CHF CF Ct 3001s
CHCl.CCl, 2 5 2993s
CHF2CF2CF3 3001s
CH/ 4 3016s
CH_CH_ 3 3 3008s 2969s 2930sh 2893m
CH CH_CH 5 2966s 2900m 2873sh
CH (CH2) CH 2966s 2892sh 2880m
CF2BrCFClCHFCH2Br 3068s 3005m 2969m

Compound Frequency in cm.
CF.CICHCICH, 2 3 3005s 2984ah 2952s 2878m 2827w
CHCHFCH CI 3 2 2939s 2964s 2943s 2878w
CF2CICH2CHC12 3008s 2959m
CF_ClCH0CHBrCH 2994s 2940m 2894sh 2835vv
CFCl2CH2CHBrCH2Cl* 2968s 2356w
CF2BrCH2CH2Br* 3026m 2977s 2933s 2862w
CF ClCH CFBrCH 3042sh 3001s 2965sh 2895w
CF _CICH(CH,)CH,CH a 3 2 3 2969vs 2945s 2884m
CF2BrCH CF Br* 3017s 2979m
CF2BrCH2CF CH CF Br* 3020s 2973m
CCl,CHFCF,Br* 3 2 3069m 3012m 2966s
CCl,CH,CHFBr 3 2 3061m 2992s 2954s
CH2=CH2 3130s 3086s 3012s 2964m

Frequency in cm.
CH = CF_
CF =CFCH_CH_Br 2 2 2
3160ah 3123s 3086s
3l60w 3090sh 3070 s
3166s 3125s
3117sh 3094s 3056s 3002vs 2940m
3089s 3018s 2989s 2963s 2936s
These compounds may contain more than one isomer. They are addition products from reactions that have two possible ways of addition. However, the formulae listed above are for the more favorable isomer.

Compound Frequency in cm. ~
Pure Solution
CH OH 2941vs 2360sh 2825s 2946s 2920sh 2832sh
CH CH OH 2976a 2919sh 2882s 2977vs 2926sh 2880s
CtLCILCH OH 3 2 2 2965s 2935s 2376s 2967s 2935s 2379s
CH3(CH2)2CH2OH 2962s 2933s 2370s 2966s 2933s 2873s
CHJCH ) CH OH 3 c% j Ct 2958s 2933s 2866m 2963vs 2933v3 2372s
CH_(CH_) .CH.OH 3 2 4 2 2956sh 2929s 2859s 2959s 2932vs 2860s
CH.(CH,)_CH_OH J CO c 2950sh 2911vs 2856s 2958sh 2933vs 2859s
CH3(CH2)6CH2OH 2952sh 2924vs 2856s 2958sh 2933vs 2359s
CH (CH ) CH OH 3 etc 2947vs 2915s 2868s 296lvs 2927s 2370s
CH3(CH2)8CH2OH 2953vs 2924vs 2869s 2963vs 2927s 2873s
CH3(CH2)9CH2OH 2959s 2927s 2859m 2964s 2929va 2865s
CH,CH(OH)CH, 3 3 2976s 2926m 2883m 2971s 2930m 2879m

Compound Frequency in cm.
Pure Solution
CH,CH_CH(OH)CH, 3 2 3 2969s 2928m 2879m 2971vs 2927s 2877m
CH (CH ) CH(OH)CH 3 2 2 3 2966vs 2929s 2874s 2964vs 2931s 2865m
(CH,LCOH 3 3 2966s 2874sh 2976s 2933sh 2875sh
(CH3)2CHCH2OH 2953s 2929sh 2874s 2962vs 29333h 2865s
CH =CHC(CH ) OH 3091w 2976vs 2931m 2871sh 3091v/ 2976vs 2931m 2876sh
HOCH,(CHJ0CH.,OH 2X 2'3 2 2938m 2364m 2938s 2869s
CF CH OH 3 2 2962s 2389m 2976sh 2949s 2884w
CF CF,CH OH .5 i-# & 2981o 2945sh 2887sh 2978s 2935m 2867s
CF3(CF2)2CH2OH 2936s 2950s 2884s 2984s 2937m 2879sh
CF fCFJ,CH,OH 3 2 o 2 2958m 2837br 2973sh 2943m 2885w
CHF CF CH OH u O 2974sh 2948m 2882sh 2983sh "2943m 2888sh
CHF2(CF2)3CH2OH 3003sh 2950m 2890w 2999sh 2940m 2882w
CHF_(CF,)_CHo0H 2 2 5 2 2999sh 2956m 2889w 2994sh 2941m 2884w
CHF (CF ) CH OH 2 CiC 2939sh 2952m 2889w 2994m 2942m 2884w

Compound Frequency m cm.
Pure Solution
CHF2(CF2)9CH2OH. 2992ah 2952m 2884\v 2987sh 2935m
CF CH(OH)CH 3 3 2903s 2940sh 2895sh 2975s 2921sh
CF0CF.,CH(OH)CH 3 2 3 2992s 2950m 2908sh 2977s 2940sh 2898sh
CF3(CF2)2CH(OH)CH3 2995m 2952sh 2903sh 2990m 2941sh 2886oh
(CH3)2C(OH)CF3 2998sh 2946s 2887m 2982sh 2943s 2870m
CH =CHC(CH J(OH)CI<' 2 5 5 3104m 3002m 2949w 3103m 2999vs 2945s
HOCH (CF ) CH OH C* Ct .5 C* 2955m 2885m 2962m

Frequency in cm.
CHC,Hr I 6 5
3095sh 3071s 3037s
3022s 2993s 2968m
3028s 3001s 2981sh 2952m
CH -CHC,H i 2 i 6 5
3098sh 3069s 3039vs
3027s 2997s 2072t
.CHC,Hr I 6 5
3096s 3068s 3039s
% Ct \ C* Ct
CH-CHBrCH. Br | 2 2

TABLE XI (continued)
Compound Frequency in cm.
CH -C(CHJ, 2986vs 2944s 2880m
j *L | $ Ct

Frequency in cm,
C.H_F 6 5
C,HCC1 6 5
C6H5Br C6H5I
p-C,H F
" 6 4 2
m-C6H4F2 p-C6H4FI
C,H CF 6 5 3
m-CF C.H Br
3 Q 4
^ 3 6 4
3094s 3073s
3071s 3070s 3062s
3093s 3067s
3083sh 3060sh 3081sh
2992w 2957m 2924s 2854m
3187sh 3157sh 3103:
3076s 3050s
2920w 2855w 3028s 2950sh 2924s 2868m
3035s 2979s 2946sh 2927vs 2369m
2927w 2921w 2931m

Compound Frequency in cm. ~1
C.ILNH, 3075m 3039a
6 5 2
m-CF C,H.NH 3063s 3038s 2927w
3 6 4 2
o-C6H4(CH3)2 3107sh 3055sh 3018s 2970s 2937s 2873sh
o-C.H/CFJ, 3196s 3143sh 3079s 3023m 2918m
6 4 3 2
C.H (CH ), 3130v/ 3041 oh 3006s 2970s 2922va 2865sh
6 3 3 3
G,H_(CH_), 3003m 2962sh 2923s 2866m
6 2 3 4
CH,C,H,(CF 1 3079sh 3053m 2964a 2934s 2874sh
3 o 3 3d.
(CH3)2C6H2(CF3)2 3047w 2976s 2950s 2928s 2379m

the tables. All the frequencies listed for the C-H stretching vibrations were obtained from well-defined sharp peaks unless indicated by sh, in which case, the frequencies were taken from shoulder bands. In most cases, these shoulder bands were well-defined, and they are taken to be accurate within 3 cm, ,
A comparison of the C-H stretching frequencies listed for the alcohols in Table X was made. The fluorinated compounds gave higher frequencies than the non-fluorinated for all vibrations obtained from the pure compounds. They also gave higher frequencies for all but three vibrations from the comparison of the carbon tetrachloride solutions. The fluorinated alcohols studied contained five types of carbon-hydrogen bonds. These types were; -CH^, ^CH -^CH, =CH, and =CH
The compounds of the type CHX9, where X is a halogen, showed a decrease in frequency of the very sharp single band, observed in their spectra, with the changing of X from fluorine through the family to iodine. Similarly, CHF CF, gave a higher frequency than CHCl CCl However, both gave lower frequencies than any of the CHX compounds. CHF CF CF gave a value identical to that of
J C* La $
Some very interesting results were obtained from the series of fluorinated ethylenes. The compound CH_,=CHF gave three bands with frequencies of 3160 cm. \ 3123 cm. and 3086 cm. Replacing

the single hydrogen on the carbon with the fluorine atom with another fluorine atom, the band with the frequency of 3123 cm. did not appear in the spectrum. When a third fluorine was added to yield the compound CHF=CF2, this band was again present but the band at 3086 cm. was not present. In all three compounds, the band at 3160 cm. appeared. A similar band also appeared in the spectrum of ethylene but with a lower frequency of 3130 cm, *. There was no band at or near 3123 cm, in the spectrum of ethyl: >.-, This leads to the conclMlon that the band at 3K3 i. due to the >CH
group's hydrogen stretching vibratica, and the band at 3086 cm. is the result of the =CH group's hydrogen stretching vibration.
A similar situation appeared in the substituted ethylenes that appeared in the free N-H stretching vibration. This was that one vibration revealed a shift while another vibration showed no such shift. The same explanation may hold here that was offered for the NH case since the space configuration for the hydrogens is similar.
The spectra of most of the additional halogenated compounds containing the ^CH group gave bands with frequencies very close
to those obtained from the fluorocarbon hydrides, The range for
-1 -1 5
this band is from 3000 cm. to 3040 cm. Bellamy lists the
value for this band as 2390 10 cm. All of the fluorocarbon
hydrides and many of the other -^CH containing compounds gave
no bands near this value. Some of the investigators of the C-H

stretching vibrations have questioned the value of 2890 cm. for the CH stretching band,
All except one of the mixed haiogenated hydrocarbons gave an absorption peak near 2965 cm, and many of these peaks were strong, sharp bands. The shape and intensities of these peaks implies that they are major absorption bands. No assignment for the peak is offered, since it appeared in compounds containing only the
^CH type of hydrogen and in compounds containing only the methylene type of hydrogen. Various other bands appeared in the spectra of these compounds but no definite trends developed.
Several trends were observed in the cyclobutanes studied. There was, however, no comparison possible since no non-halogenated cyclobutanes were run. The method of preparation for the cyclobutanes and the evidences for the structural for aulae of these com-
pounds are given by Johnson who prepared these compounds.
Two possible types of carbon-hydrogen groups existed in the cyclobutanes. They were a single hydrogen attached to one of the carbon atoms making up the ring, and two hydrogens on the ring carbon. All of the eye lie compounds investigated contained the two hydrogens on the ring. About one-half of the eye lie compounds contained the single hydrogen on the cyclic carbon, while the remaining half did not have this group. A band with a frequency of 3097 A 2 cm. was observed in the spectra of all but one of the

compounds with the single hydrogen to carbon type, The compound that was the exception had a -CN group attached to the same carbon which had the single hydrogen. This might account for the discrepancy. The compounds that did not have the single hydrogen group gave no such band at 3097 cm. This frequency is assigned to the vibration from this single hydrogen on a cyclic carbon. The frequency for this band is slightly higher than those given for the
-^CH group of the other halogenated compounds. The shift in the frequency of the C-H stretching vibration of the -^CH group is probably caused by the strain of the four membered ring.
Two frequencies of the approximate values of 3030 cm. and 2960 cm. 1 appeared in all but one cyclic compound. These bands
are thought to be from the symmetric and asymmetric C-H stretching
vibrations of the CH^ group. It has been stated that even though the ring strain causes a shift in the C-H stretching frequencies, the frequencies for the C-H vibrations of cyclobutane are still below 3000 cm. This implies that the fluorination of the cyclic compounds has caused a frequency shift.
Some of the other groups attached to the ring of these cyclobutane s contained hydrogen. Ail of the compounds containing a phenyl group had a band near 3070 cm. in their spectra. This peak is probably from the aromatic C-H stretching vibration. Several other frequencies are listed, all arising from spectra of compounds con-

taining additional carbon-hydrogen groups. No correlation was found among these remaining frequencies,
The frequency given by Fox and Martin for the aromatic C-K stretching vibration is 3038 cm. The major peak from the spectra of benzene in this study also gave this value. The mono-halogenated aromatic compounds all gave a higher frequency. In this case, fluorine gave the lowest frequency. This may be attributed to a size factor. Fluorine is more nearly the same size as the carbon atoms. Therefore, the p_-orbitals of fluorine and carbon are approximately the same. Because of these similar p_-orbitals, there is a better chance for resonance involving fluorine and carbon than the other halogens and carbon. The difference in resonance could explain the frequency variation.
The difluorobenzenes gave even higher frequencies than the mono-halobenzenes. This would bs expected since the number of halogens is doubled. The meta and para difluorobenzenes both produced bands with nearly the same frequencies.
Other increases in frequencies can be noted in Table XII.
Interesting shifts are produced when the -CH group of toluene is
replaced by the -CF^ group. Similar results were obtained when other -CH^ groups were replaced by -CF^ groups. Not only do the aromatic C-H stretching frequencies shift, but so do those attributed to the -CH groups.

McKisson* has shown that the addition of fluorine to a molecule causes a tightening of the bonds near the point of substitution. Other investigators have found similar results with the other halogens. In the case of the N-H and O-H bonds, a different situation exists as was explained in the sections dealing with these bonds. The shift to a higher frequency of the C-H stretching vibration indicates that the carbon-hydrogen bond has been tightened as a result of fluorination.
The shift as the halogens change in the CHX^ compounds may be attributed to the decrease in electronegativity as the higher members of the halogen family are used. The shifts due to the replacement of a fluorine atom in CHF by a -CF group may also be attributed to a change in electronegativity. A possible explanation for the lack of frequency shift, when one of the fluorines of the -CF group is replaced by another -CF^ grouP may be that the -CF^ group has approximately the same electronegativity as the -CF CF group.

In this study, the infrared spectra of fluorinated compounds containing the oxygen-hydrogen, the nitrogen-hydrogen, and the carbon-hydrogen bond were compared to similar non-fluorinated compounds. The region of interest was limited to the 5000 cm. to 2500 cm. range which includes the hydrogen stretching vibrations of these bonds.
Both fluorinated alcohols and carboxylic acids were examined to determine the effect of fluorination on the O-H stretching vibrations. The spectra of the alcohols were determined from both the pure compounds and the carbon tetrachloride solutions, while only the carbon tetrachloride solutions of the carboxylic acids were examined. Interesting shifts in both the free and the associated O-H stretching vibrations were noted. The free O-H and the dimer associated O-H stretching vibrations were shifted to a lower frequency v/hen the compound contained fluorine. In contrast to this, the polymeric associated O-H stretching vibration was shifted to a higher frequency upon fluorination. The strong inductive effect of the highly electronegative fluorine atom was suggested as the possible

cause for these shifts. The shifts to a lower frequency was attributed to the loosening of the oxygen-hydro gen bond of the -OH group, while the shift to the higher frequency was presumed to be caused by the loss of strength of the hydrogen bonding.
The nitrogen-containing fluorinated compounds also produced frequency shifts. An unexpected difference occurred with the symmetric and asymmetric free N-H stretching vibrations obtained from the spectra of the carbon tetrachloride solutions. The frequency of the symmetric vibration shifted while the frequency of the asymmetric vibration did not shift. The difference was attributed to the manner in which the atoms move in the two vibrations.
An additional peak located between the symmetric and the asymmetric free N-H stretching vibrations was observed in the spectra of the carbon tetrachloride solutions. No definite reason was offered for the additional peak. However, it was thought that the peak might be due to some type of combination band involving this peak and the symmetric free N-H stretching vibration, since both bands showed a similar frequency shift.
As in the case of the associated O-H stretching vibrations, the associated N-H stretching vibrations showed a shift to a higher frequency upon fluorination. The same explanation offered for the O-H cases was given for the N-H cases.

All the frequency shifts due to fluorination in the C-H stretching vibrations were to a higher frequency. This was attributed to the tightening of the carbon-hydrogen bond caused by fluorination.
A comparison among the fluorinated compounds was made. Several interesting trends were found, and some frequencies were assigned.

1. McKisson, E. W,, Thesis, University of Florida (1954),
2. Herzberg, G., "Infrared and Raman Spectra of Polyatomic Molecules, D. Van Nostrand Company, Inc., New York (1945).
3. Wilson, E. B., Jr., Decius, J. C, and Cross, P. C., "Molecular Vibrations, McGraw-Hill Book Company, Inc., New York (1955).
4. Moelwyn-Hughes, E. A., "Physical Chemistry, Pergamon Press, New York (1957).
5. Bellamy, L. J., "The Infrared Spectra of Complex Molecules, Second Edition, John Wiley and Sons, Inc., New York (1958).
6. Hilbert, G, E., Wulf, O. R., Hendricks, S. B, and Liddel, U,, Nature, 135, 147 (1935).
7. Fox, J. J. and Martin, A. E., Proc. Roy. Sox., A162, 419 (1937).
8. Fox, J. J. and Martin, A. E., Froc. Roy. Soc., Al 75, 208, (1941).
9. Kuhn, L. P., J. Am. Chem. Soc, T4t 2492 (1952).
10. Goulden, J. D. S., Spectrochim. Acta, jb, 129 (1954).
11. Haszeldine, R. N., J. Chem. Soc, 1953, 1757.
12. Cole, A. R. H. and Jefferies, P. R., J. Chem. Soc., 19 j6, 4391.
13. Kletz, T. A. and Price, W. C., J. Chem. Soc, 1947, 644.
14. Finch, J. N. and Lippincott, E. R., J. Chem. Phys., 24, 90 8, (1956),

15. Corish, P. J., and Chapman, D., J. Chem. Soc., 1957, 1746.
16. Davies, M. M., and Evans, J. C., J. Chem. Phys., 20, 342, (1957).
17. Flett, M. St, C, J. Chem. Soc, 1951, 962.
18. Fuson, N, and Josien, M., J. Opt. Soc. Am,, 43, 1102 (1953),
19. Bratoz, S,, Hadai, D., and Sheppard, N., Spectrochim. Acta, 8, 249 (1956),
20. Harris, J. T., Jr., and Hobbs, M. E., J. Am. Chem. Soc,, 76, 1419 (1954),
21. Davies, M. M., and Sutherland, G. B. B. M., J Chem, Phys,, 6, 755, 767 (1938).
22. Oshida, I., Ooshika, Y., and Miyasaka, R., J. Phys. Soc, Japan, 1_0, 849 (1955).
23. Cannon, C. G., Mikrochem. Acta, 1955, 555.
24. Cannon, C. G., J. Chem. Phys., 24, 491 (1956).
25. Senti, F. and Harker, D., J. Am, Chem, Soc, 6j2, 2008 (1940).
26. Darmon, S. E., and Rudall, K. M., Discucc. Faraday Soc,, A206, 192 (1950).
27. Hantzsch, A., Eer., _64, 661 (1931),
28. Buswell, A. M., Rodebush, V/. H., and Roy, M. F., J. Am. Chem. Soc, j6_0, 2444 (1938).
29. Richards, R. F. and Thompson, H. V/., J. Chem. Soc,, 1947, 1248.
30. Letaw, H. and Gropp, A. H., J. Chem. Phys. 21, 1621 (1953),
31. Klemperer, W., Cronyl, M. W., Maki, A. H., and Pementel, G. C, J. Am. Chem. Soc, Jh, 5846 (1954).
32. Russell, R. A. and Thompson, H. W., Spectrochim. Acta, 8_, 138 (1956).

33. Wo r sham, J. E., Jr. and Hobbs, M. E., J, Am. Chem. Soc, 73, 2151 (1951).
34. Darmon, S, E. and Sutherland, G, B, B. M,, Nature 164, 440-1 (1949).
35. Fox, J, J. and Martin,A, E., Proc. Roy. Soc., A167, 257, (1938).
36. Colthup, N. B., J. Opt. Soc. Am,, 40, 397 (1950),
37. Poaefsky, A, and Coggeshall, N. D,, Analyt. Chem., 20, 812 (1948).
38. Francis, S. A., J. Chem. Phys,, _19, 942 (1951).
39. Mirone, P. and Fabbri, G. F., Gaza. Chim., 84, 187 (1954),
40. Nolin, B. and Jones, R. N., J. Am, Chem. Soc, 75, 5626 (1953).
41. Roberts, J. D. and Chambers, V. C,, J, Am, Chem. Soc, 73, 5030 (1951).
42. Rathgens, G, W., Jr., Freeman, N. K., Gwinn, V/. D., and Pitaer, K. S., Jr., J. Am. Chem. Soc, 75, 5634 (1951).
43. Fuson, N. and Josien, M., J, Am. Chem. Soc, 78, 3049 (1956),
44. Theimer, R. and Nielsen, J. R., J. Chem. Phys., 27, 887 (1957).
45. VVeissman, H. B., Meister, A. G., and Cleveland, F F., J. Chem. Phys., 29, 72 (1958).
46. Mann, D. E., Acquista, N., and Plyler, E. K., J. Chem. Phys., 23, 2122 (1955),
47. Harris, F. W., Narasimham, N. A., and Nielsen, J. R., J. Chem. Phys., 24, 1232 (1956).
48. Narasimham, N. A., Nielsen, J. R., and Theimer, R., J. Chem. Phys., 27, 740 (1957).

49. Silas, R. S., Thesis, University of Florida (1955).
50. Lovelace, A, M., Rausch, D. A., and Postelnek, W., "Aliphatic Fluorine Compounds, Reinhold Publishing Corp., New York (1958).
51. Johnson, R. V.'., Jr., Thesis, University of Florida (1959).

Kenneth T. Knapp was born June 9, 1930, at Jacksonville, Florida, and received his early education in the public schools o that city. Upon graduation from the Andrew Jackson High School hi February, 1949, he entered Jacksonville Junior College. After his graduation from Jacksonville Junior College, he continued his education at the University of Florida where he received the degree of Bachelor of Science in Chemistry in February, 1954.
From 1954 until 1956, he served in the Signal Corps of the United States Army and was stationed in Germany. After his release from the Army, he returned to the University of Florida where he entered Graduate School, While a graduate student, he was employed as Graduate Assistant and, later, as a Teaching Assistant in the Department of Chemistry. In September, 1959, he was awarded a General Motors Fellowship, and in June I960, he received a College of Arts and Sciences Fellowship.
While attending college he was a member of Alpha Chi Sigma, He is a member of the American Chemical Society,

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs