Influence of a methyl substituent on the Raman spectrum of but-3-enyl methyl ether

Article abstract of DOI:10.1016/j.saa.2012.02.059

The Raman spectrum of but-3-enyl methyl ether, CH₂=CHCH ₂CH₂OCH₃ is reported and compared with those of its homologues in which a methyl group is substituted for a hydrogen atom on one of the carbon atoms of the alkenyl chain. Attention is focused on the influence of this methyl group on the bands in the spectrum associated with specific C-H, skeletal stretching and bending vibrations. The use of ab initio DFT quantum mechanical calculations to assist in making these assignments reveals a high degree of mode-mixing in the skeletal vibrations. The value of model studies of this kind in refining the correlations between the presence and absence of specific bands in a Raman spectrum with molecular structure is emphasised.

Full text of DOI:10.1016/j.saa.2012.02.059

Spectrochimica Acta Part A 93 (2012) 26–32
Contents lists available at SciVerse ScienceDirect
Spectrochimica Acta Part A: Molecular and
Biomolecular Spectroscopy
journal homepage: www.elsevier.com/locate/saa
Influence of a methyl substituent on the Raman spectrum of
but-3-enyl methyl ether
Richard D. Bowen , Howell G.M. Edwards , Tereza Varnalib
a,∗
a
a
Chemical and Forensic Sciences, School of Life Sciences, University of Bradford, Bradford, West Yorkshire BD7 1DP, United Kingdom
Department of Chemistry, Bogazici University, 34342 Bebek, Istanbul, Turkey
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 February 2012
Accepted 10 February 2012
The Raman spectrum of but-3-enyl methyl ether, CH2 CHCH CH OCH is reported and compared with
2
2
3
those of its homologues in which a methyl group is substituted for a hydrogen atom on one of the carbon
atoms of the alkenyl chain. Attention is focused on the influence of this methyl group on the bands
in the spectrum associated with specific C H, skeletal stretching and bending vibrations. The use of ab
initio DFT quantum mechanical calculations to assist in making these assignments reveals a high degree of
mode-mixing in the skeletal vibrations. The value of model studies of this kind in refining the correlations
between the presence and absence of specific bands in a Raman spectrum with molecular structure is
emphasised.
Keywords:
Raman spectroscopy
Alkenyl
Ethers
Molecular assignments
©
2012 Elsevier B.V. All rights reserved.
1
. Introduction
bands may not be complete unless systematic work on suitable
model compounds, including their isotopically labelled analogues,
where appropriate, has been undertaken. Secondly, the vibrations
of functional groups of major interest in organic chemistry may
only be infrared-active or, at best, show only a weak Raman effect.
A few functional groups, particularly the C C entity, are more
amenable to analysis by Raman spectroscopy, especially for sym-
metrically substituted alkenes where the stretching vibration of
this bond is theoretically inactive in the infrared but is strongly
Raman-active. Much supporting structural information may then
be provided by the appropriate analysis of the CH stretching and
bending modes of these species. Attention has been given in a sys-
tematic study of 29 butenyl, pentenyl and hexenyl methyl ethers
[7] to the effect of the substitution pattern, position, and stereo-
chemistry of this double bond on the wavenumber of the ꢀ(C C)
band. In addition, analysis of the number of bands associated with
The routine application of Raman spectroscopy [1–3] to the
determination of the molecular structure of complex molecules is
dependent upon the correct attribution of molecular vibrations to
key features in the spectrum; whereas in many cases this approach
relies upon the assignment of functional groups and molecular fin-
gerprinting for the characterisation of unknown organic materials,
the CH modes are often primarily ignored. In some instances, how-
ever, detailed structural information is thereby lost, especially for
molecules that possess little functionality and for which compar-
atively minor variations in the nature of the CH , CH2 and CH
3
entities may be of critical importance. Raman spectroscopy has
many applications in the realms of forensic science, art and archae-
ology, as illustrated by its potential in distinguishing authentic
ancient artefacts from modern imitations [4]. Further applications
in the biomedical and pharmaceutical arenas are exemplified by
in vivo studies of skin, nail and other tissues [5,6]. These and related
investigations show how fuller and unambiguous assignments of
the CH modes in the Raman spectra can facilitate the identifica-
tion of impurities arising from degradation and adulteration of
the pure material, or the presence of structural polymorphs of
drugs.
2
vibrations of the ꢀ(sp C H) bonds and the wavenumber of these
bands led to an important correction in the previously accepted
assignments of these vibrations. The revised assignments were con-
firmed by examining the spectra of CD2 CHCH CH CH OCH and
2
2
2
3
CH2 CDCH CH CH OCH . This work established general correla-
2
2
2
3
tions that were analytically valuable in determining the nature of
2
the C C bond and the associated ꢀ(C C) and ꢀ(sp C H) bands in
The scope of Raman spectroscopy for the detailed analysis of
organic compounds remains limited by two factors. Firstly, the cor-
relation between the molecular structure and the observed Raman
the Raman spectrum of compounds which contain a simple alkenyl
moiety. Careful analysis of the bands associated with the deforma-
tion modes later led to further correlations with the geometry of
the hexenyl chain [8].
It is clearly of interest to extend this investigation to include
the skeletal modes of other alkenyl systems since many com-
pounds of biological significance contain one or more C C groups,
∗ Corresponding author.
E-mail address: r.d.bowen@bradford.ac.uk (R.D. Bowen).
1
386-1425/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.saa.2012.02.059

R.D. Bowen et al. / Spectrochimica Acta Part A 93 (2012) 26–32
27
H2
OCH3
OCH3
OCH3
H4
CH3
H3
H1
CH3
1-1CH3
1-4CH3
1
CH3
OCH3
OCH3
CH3
1-3CH3
1
-2CH3
Scheme 1. Structures of but-4-enyl methyl ether and four homologues in which a hydrogen atom in the butenyl chain has been replaced with a methyl group.
often with at least one methyl group attached to a nearby car-
bon atom. The analytical utility of Raman spectroscopy would
thus be further enhanced if the presence and position of such a
methyl group could be inferred reliably by considering the skele-
tal modes or related features of the spectrum. The influence of a
methyl group at each of the positions in the alkenyl skeleton may
be probed by considering the Raman spectra of an appropriate
set of ethers, each of which is derived from a common parent. In
this paper a series of substrates have been studied beginning with
the archetypal but-3-enyl methyl ether, CH2 CHCH CH OCH , 1,
2.3. Raman spectroscopy
The Raman spectra were recorded on a Bruker IFS66 infrared
instrument with an FRA 106 Raman module attachment, with an
3+
Nd /YAG laser operating at 1064 nm as the excitation source. A
sample of each ether contained in a sealed glass capillary tube
was mounted in a specially constructed holder. The laser beam
was focussed to a spot of 100 ␮m diameter at the sample; laser
powers of up to 500 mW were used. A liquid nitrogen-cooled ger-
manium detector with extended bandwidth was used over the scan
2
2
3
−
1
which may be considered to be the parent species from which the
four pentenyl methyl ethers, CH2 CHCH CH(CH )OCH , 1-1CH ,
range of 50–3500 cm . 2000 consecutive scans were accumulated
−
1
at 4 cm resolution in a typical run to enhance the signal-to-noise
ratio. The observed band wavenumbers are considered correct to
2
3
3
3
CH2 CHCH(CH )CH OCH , 1-2CH , CH C(CH )CH CH OCH , 1-
3
2
3
3
2
3
2
2
3
−
1
3
CH3, and CH CH CHCH CH OCH , 1-4CH3, are derived by
better than ± 1 cm
.
3
2
2
3
1
2
3
substitution of a methyl group for a hydrogen atom (H , H , H
and H , respectively) at each of the four distinct sites in the carbon
skeleton, Scheme 1.
4
2.4. Vibrational spectroscopy
The highest molecular symmetry displayed by the C H12O iso-
6
mers with a terminal C C group shown in Fig. 1 is Cs. The 51
vibrational modes will be distributed as follows:
2
. Experimental
2.1. Specimens
ꢀ
ꢀꢀ
ꢁtotal = 35A + 16A
The suite of specimens selected for this spectroscopic analysis
and
comprised the parent but-3-enyl methyl ether (C H7OCH ) and
4
3
four homologues in which a methyl group is attached to one of
the carbon atoms in the butenyl chain. This system is the simplest
in which replacement of a hydrogen atom with a methyl group is
possible in four distinct sites (each end of the C C bond, the adja-
cent allylic position and the more distant homoallylic position),
thus permitting the influence of methyl substitution on and in the
vicinity of the double bond to be investigated. In addition, sev-
eral isomers with a different structural relationship between the
alkene and the methyl group may be considered, thus shedding
further light on the correlation between substitution pattern and
the associated Raman bands.
ꢀ
ꢀꢀ
ꢁskeletal = 11A + 4A
ꢀ
ꢀꢀ
ꢁC H = 24A + 12A
All of these C H modes are non-degenerate; therefore, all will
be active in both infrared absorption and Raman scattering spectra.
The skeletal vibrational modes in the A symmetry class
comprise C C stretching, 2 C O stretching (symmetric and asym-
metric), 3 C C stretching and 5 in-plane bending modes of type
ꢀ
C
A
C C, C C C, C C O and C O C. The skeletal modes of the
ꢀ
ꢀ
symmetry class will consist of 4 out-of-plane bending modes
2
.2. Synthesis
i, NaH in
triglyme
1
2
n-C
n
H
2n-1OH
n-C
n
H
2n-1OCH
3
[n = 4 or 5]
The alkenyl methyl ethers were synthesised by the base-
ii, CH 3I
catalysed condensation of the alkoxide anion derived from the
corresponding alkenol with a slight deficiency of iodomethane.
Most of the alkenols were commercial samples. Pent-3-en-1-ol,
which was not commercially available, was prepared from pent-3-
enoic acid by acid-catalysed esterification with methanol, followed
CH OH/HCl
3
CH CH=CHCH CO H
CH CH=CHCH CO CH
3 2 2 3
3
2
2
Reflux
i, LiAlH 4
in ether
◦
by reduction at low temperature (−78 C) of the resultant methyl
pent-3-enoate with lithium aluminium hydride in diethyl ether.
Representative examples of these synthetic procedures, which are
summarised in Scheme 2, have been reported in more detail else-
where [9,10].
CH CH=CHCH CH OH
3
2
2
ii, NaOH
in water
Scheme 2.

2
8
R.D. Bowen et al. / Spectrochimica Acta Part A 93 (2012) 26–32
Fig. 1. FT-Raman spectrum of parent but-3-enyl methyl ether (3-BME), labelled 1
in Scheme 1; wavenumber range 100–3400 cm , spectral resolution 4 cm , exci-
tation wavelength 1064 nm, 1000 spectral scans accumulated.
Fig. 2. FT-Raman spectrum of 1-methyl substituted 3-BME, labelled 1-1CH3 in
Scheme 1; conditions as for Fig. 1.
−1
−1
associated with C
C C, C C C, C C O and C O C groups. These
skeletal modes will undoubtedly be mixed with each other.
The parts of the spectra in which the bands that may be ascribed
to C H stretching, bending and rocking modes are expected to be
ꢀ
ꢀꢀ
rather complex since the 36 vibrational modes (24A + 12A ) will
be distributed as follows:
ꢀ
ꢀꢀ
ꢀꢀ
CH2 C: 4A + A ; 2 C H stretches, 1 deformation, 2 rocking
ꢀ
CH: 2A + A ; 1 C H stretch, 2 deformation
ꢀ
ꢀꢀ
CH : 12A + 6A ; 6 C H stretches, 6 deformations, 6 rocking
CH : 3A + 6A ; 3 C H stretches, 2 deformations, 6 rocking, 1 tor-
2
ꢀ
ꢀꢀ
3
sion
ꢀ
ꢀꢀ
CH/CH synchronised: A + 2A ; 1 C H stretch, 2 CH deformations.
2
2
2.5. Ab initio DFT calculations
Ab initio DFT quantum mechanical calculations were carried out
on the molecular samples set for the skeletal modes only using the
Gaussian package program [11] with a 6-31G** basis set in order to
explore the derived molecular ground states for comparison with
the experimental spectroscopic data. The lowest energy optimised
structures were taken for derivation of the electronic energy lev-
els, orbital sets and predicted Raman spectral wavenumbers and
relative intensities.
Fig. 3. FT-Raman spectrum of 2-methyl substituted 3-BME, labelled 1-2CH3 in
Scheme 1; conditions as for Fig. 1.
ways. This comparative approach permits differences in the spec-
tra of two (or more) ethers with slightly different alkenyl moieties
to be ascribed to specific features that are absent (or present) in
one ether, but present (or absent) in the other(s). Secondly, the
3
. Results and discussion
The Raman spectra of 1, 1-1CH , 1-2CH , 1-3CH , and 1-4CH
3
3
3
3
are shown in Figs. 1–5, respectively. Assignments of the bands in
the spectra are summarised in Table 1.
The Raman spectroscopic data may be conveniently subdi-
−
1
vided into three regions. First, the 2600–3200 cm region, where
the C H stretching bands of the methine, methylene and methyl
−1
groups occur. Secondly, the 1600–1800 cm region, which con-
tains the C C stretching band, which is especially important in the
−
1
spectra of alkenes and related species. Thirdly, the 700–1500 cm
region, which contains the bands associated with C H bending
vibrations, together with the bands arising from the stretching and
deformation modes of the skeletal bonds (including not only sim-
ple C C and C O modes, but also more complex vibrations of larger
structural subunits such as CCC, CCO and COC entities).
Spectral assignments were made initially by three general meth-
ods. First, by comparing the spectra of 1, 1-1CH , 1-2CH , 1-3CH ,
3
3
3
and 1-4CH3 with those of butenyl and pentenyl methyl ethers
Fig. 4. FT-Raman spectrum of 3-methyl substituted 3-BME, labelled 1-3CH3 in
with structures that are similar, but which differ in well-defined
Scheme 1; conditions as for Fig. 1.

R.D. Bowen et al. / Spectrochimica Acta Part A 93 (2012) 26–32
29
Table 1
Vibrational wavenumbers (cm 1) and assignments of bands in the Raman spectra of but-3-enyl methyl ether and selected methyl substituted homologues.a,b
−
Compound
Probable assignment
1
1-1CH3
1-2CH3
1-3CH3
1-4CH3
3
082 ms
3080 ms
3082 ms
3077 m
ꢀ(CH2) asym, CH2
C
3004 ms
ꢀ(CH), trans CH CH
3
2
002 s
982 s
2999 s
2979 s
2996 s
2983 s
2985 s
2965 s
ꢀ(CH2) sym, CH2
ꢀ(CH), CH CH
ꢀ(CH2/CH3)
C
2982 s
2965 sh
2965 s
962 s
2
ꢀ(CH2/CH3)
2931 sh
2909 s
2872 s
2853 s
2825 s
2807 sh
2933 vs
2915 sh
2870 s
2860 s
2821 s
2930 s
2880 s
2825 s
2935 sh
2919 vs
ꢀ(CH2/CH3)
ꢀ(CH2/CH3)
ꢀ(CH2/CH3)
ꢀ(CH2/CH3)
2912 vs br
2856 s
2826 s
2810 sh
2857 s
2826 s
2808 sh
ꢀ(CH3 O) sym
2758 m
2731 m
2734 m
2
731 mw
2734 w, br
2730 sh
2665 w, br
1
1
673 ms
655 m
ꢀ(C C) trans 1,2-disub
ꢀ(C C) cis 1,2-disub
ꢀ(C C) 1,1-disub
1650 s
1
1
1
1
642 s
1642 s
1641 s
ꢀ(C C) monosub
482 mw
447 m, br
416 m, br
1481 mw
1443 m, br
1416 m, br
1477 m, br
1455 ms, br
1417 m, br
1451 mw
1453 m, br
1416 m, br
ı(CH of C CH2 O)
1396 m, h, l
1
382 w
1378 mw
1385 w
ı(CH of C C CH2 C)
1342 w
1336 w
1349 w
1
301 mw
1307 m
ı(CH of CH3 O)
ı(CH of CH3 O)
1
293 m, h
184 w
1293 m, h
1293 m, h
1
282 mw
1
248 w
202 w
1230 w
1
1
1180 w
154 vw
1
1143 mw, br
1
120 mw, br
055 w
1120 w
1119 mw
1074 mw
1113 mw, br
1123 w
1074 w
1040 w
996 w
1
1040 sh
1018 w
9
77 mw
2
9
9
64 mw
20 w
963 mw
931 w
955 m
967 mw
ꢂ(sp CH2)
927 mw
898 w
890 w
8
8
7
77 w
879 w
879 w
837 mw, l
793 mw, d
8
54 mw
37 br
27 w
833 m
772 m
829 mw, d
ı(CH3 C)
799 mw
63 w, h, l
707 w
6
630 mw, br
546 w
553 w, br
531 w
503 w
484 mw
478 mw
481 w
445 w
424 mw
4
06 mw, br
63 mw, br
394 vw
3
72 br
3
349 m
343 mw, br
96 mw, br
330 mw
2
332 m, br
283 w, br
a
See text for explanation of abbreviation system for butenyl methyl ether and homologues with a methyl group attached to the butenyl chain.
b
“vs”, “s”, “ms”, “m”, “mw”, “w” and “vw” denote “very strong”, “strong”, “medium strong”, “medium”, “medium weak”, “weak” and “very weak”, respectively; “br” and
“sh” denote “broad” and “shoulder”, respectively; “d” denotes “doublet”; “h” and “l” indicate that the peak has a shoulder at high and low wavenumber, respectively.

3
0
R.D. Bowen et al. / Spectrochimica Acta Part A 93 (2012) 26–32
significantly lower wavenumber is discernible. This peak appears in
−
1
the 2995–3000 cm range in the spectra of 1, 1-1CH and 1-2CH ,
3
3
−
1
but it is found at significantly lower wavenumber (2985 cm
)
in the spectrum of 1-3CH . Both the lower wavenumber and the
3
greater intensity of the band indicate that it is associated with
the symmetrical C H stretching vibration of the H C C subunit.
2
This interpretation is consistent with the presence of a band at
−
215 cm in the spectrum of CD2 CH(CH ) OCH [7,12]; the ratio
2 3 3
1
2
of the wavenumber of the bands that may be assigned to the
H and C D vibrations in the spectra of CH2 CH(CH ) OCH and
C
2
3
3
CD2 CH(CH ) OCH is 1.35, which is typical of the shift induced
2
3
3
by deuteriation.
The individual components of the envelope of bands in the
−
1
3
2
800–2980 cm region are attributable to various sp (C H) vibra-
tions. It is not always easy to define these assignments because
of the overlapping bands and the similarities in the numerous
3
sp C H bonds, but it is possible to associate the strong band
−
1
at 2820–2826 cm
probably the symmetrical C H vibration on account of its strong
intensity).
with the vibrations of the methoxy group
Fig. 5. FT-Raman spectrum of 4-methyl substituted 3-BME, labelled 1-4CH3 in
Scheme 1; conditions as for Fig. 1.
(
features identified in the spectra of the ethers were correlated with
those reported in the literature for related compounds. This lit-
erature approach was employed with caution, not least because
application of the comparative method when analysing the spectra
−1
3
.2. The 1600–1800 cm region
This part of the spectrum contains the analytically useful band
associated with the ꢀ(C C) vibration. Variations in the wavenum-
ber of this band reflect changes in the level of substitution and/or
the stereochemistry of the double bond.
The moderately strong signal at 1641 cm in the spectrum of
is typical of that found for a terminal (monosubstituted) alkene
7]. The corresponding band in the spectra of 1-1CH3 and 1-2CH
of CD2 CH(CH ) OCH and CH2 CD(CH ) OCH , and numerous
2
3
3
2
3
3
unlabelled alkenyl methyl ethers [7], had already shown that
some previously accepted assignments were not correct. Thirdly,
reinforcement or modification of the proposed assignments was
achieved by comparison of the experimental data and the informa-
tion acquired from the ab initio calculations.
−1
1
[
3
,
−1
appears at closely similar wavenumber (at 1642 and 1641 cm
respectively), as would be expected because the addition of a
methyl group to position 1 or 2 of the butenyl chain of 1 does not
affect the substitution pattern of the double bond. In contrast, the
−
3
.1. The 2600–3200 cm 1 region
−
1
It is immediately apparent that a band of medium to strong
intensity is present at 3077–3082 cm in the spectrum of 1, 1-
CH , 1-2CH3 and 1-3CH ; this band is not, however, seen in the
analogous band in the spectrum of 1-3CH3 appears at 1650 cm
;
−1
this significant increase in wavenumber reflects the change from
a monosubstituted to a 1,2-disubstituted double bond when 1 is
homologated by addition of a methyl group in position 3. Similarly,
when a methyl group is added to the 4-position of 1, to produce 1-
1
3
3
spectrum of 1-4CH . This clear correlation indicates that this dis-
3
tinctive band at highest wavenumber must be associated with a
vibration of the terminal sp2 methylene group, which is found in
4CH , the main band that may be attributed to the ꢀ(C C) vibration
3
−
1
−1
each of the ethers except 1-4CH . This band, which is logically
attributed to the asymmetric stretching vibration of the H C
entity, is of obvious analytical value.
This interpretation is consistent with related work on
CH2 CH(CH ) OCH and selected partially deuteriated analogues
is seen at 1673 cm , over 30 cm above the analogous band in the
spectrum of 1. This shift is typical of the effect of converting a mono-
substituted double bond to a trans-1,2-disubstituted entity [7].
3
C
2
−
1
The presence of a subsidiary band at 1655 cm
in the spec-
trum of 1-4CH3 is, at first sight, an unfortunate complication.
2
3
3
[
12]. Thus, whereas the spectra of CH2 CH(CH ) OCH and numer-
This signal arises because this pentenyl methyl ether was pre-
2
3
3
3
ous labelled analogues with deuterium atoms attached to sp
carbon atoms contain a band at approximately 3080 cm , the
spectrum of CD CH(CH ) OCH does not; instead a signal at
pared from 3-pentenoic acid (CH CH CHCH CO H) in which the
3 2 2
−
1
stereochemistry of the double bond was predominantly, but not
exclusively, trans. The clear presence of the less intense band at
2
2
3
3
−1
−1
2
308 cm
is present. This pronounced shift to (corresponding
1655 cm
reveals that a minor proportion of 1-4CH3 has a cis-
to a ratio of 1.34:1 for the wavenumber of the C H and C
D
1,2-disubstituted alkenyl moiety. The position of this band is again
typical of that found in other systems with similar stereochemistry
[7].
The systematic dependence of the wavenumber of the band
associated with the ꢀ(C C) vibration serves to illustrate the value
of Raman spectroscopy in probing the structure of species contain-
ing a C C bond. Such analytical data are not so readily accessible
from infrared spectroscopy because the ꢀ(C C) vibration often is
less active than it is in the Raman.
vibrations) is consistent with the influence of deuteriation on
the asymmetrical stretching vibration of the terminal methylene
group.
Further evidence for the presence of a trans-disubstituted C
C
double bond only in 1-4CH is found in the unique signal at
3
−
1
3
004 cm in the spectrum of this ether. This feature is attributable
to the C H stretching vibration of the trans-CH CH entity. A sim-
ilar, though somewhat less conspicuous, feature at approximately
−
000 cm was detected in the spectrum of CH2 CH(CH ) OCH ;
2 3 3
1
3
−
1
the corresponding band in the spectrum of CH CD(CH ) OCH
3.3. The 700–1500 cm region
2
2
3
3
−1
appeared at 2250 cm , so confirming that this signal is associ-
ated with the C H(D) stretching vibration of the penterminal CH
subunit.
The major features in this wavenumber range are seen
at 1482–1477, 1455–1443, 1417–1416, 1301–1282, 1123–1113,
−
1
Although the spectra of 1, 1-1CH , 1-2CH and 1-3CH3 do not
967–955 and 837–827 cm . Raman bands arising from C H defor-
mation, CH2 rocking and C C stretching vibrations are expected to
3
3
−
1
contain a clearly resolved signal at 3004 cm , a stronger band at

R.D. Bowen et al. / Spectrochimica Acta Part A 93 (2012) 26–32
31
appear in this region. The DFT calculations indicate strong mix-
ing between skeletal modes and up to five different modes, thus
complicating or preventing the assignment of individual modes to
specific bands in the spectra. However, some patterns and trends
may be identified.
appears in the spectrum of 1. The ab initio DFT calculations show
strongly mixed deformation modes involving CCC, CCO and COC
vibrations. Their absence from the spectrum of 1 suggests that
they are associated with CH3
C C, CH3 C C and CH3 C O mixed
mode deformations. Moreover, as the methyl group is progressively
situated further from the oxygen atom, the wavenumber of these
bands decreases quite substantially.
A band of medium to weak intensity that is sometimes broad in
−
1
appearance is found between 1482–1477 cm in the spectra of 1,
-2CH3 and 1-3CH , but not 1-1CH , or 1-4CH . Consequently, it is
1
3
3
3
logical to attribute this band to a C-H deformation mode associated
with substrates in which the butenyl group is either unsubstitued or
else has a methyl group in the 2 or 3-position. This interpretation
provides some analytical information because the band does not
appear if the ether has an unbranched alkenyl carbon chain. On
the other hand, all the compounds studied show a band in their
4
. Conclusions
The series of isomeric pentenyl methyl ethers studied in this
work corresponds to a spectroscopically complex system in which
assignment of bands, particularly at low wavenumber, to specific
vibrational modes can present serious difficulties. Nevertheless,
careful comparison of the Raman spectra of compounds containing
closely related alkenyl groups, together with the insight furnished
by ab initio calculations, permits significant conclusions to be
drawn concerning several important C H and skeletal modes. Thus,
unequivocal analytical information may be obtained from the rel-
atively small but highly significant variations in the wavenumber
−1
spectra in the 1455–1443 cm range; this signal may, therefore,
be attributed to a CH deformation mode because such a structural
2
subunit is present in each of the ethers studied in this work.
The presence of a broad band of moderate intensity at 1416 or
−
417 cm in the spectra of 1, 1-2CH , 1-3CH and 1-4CH , but not
3 3 3
1
1
1
-1CH , establishes that this signal is associated with deformation
3
of the CH2 group of the alkenyl chain that is directly attached to
oxygen in the homoallylic position because 1-1CH3 is unique in
possessing a CHCH3 entity, rather than a methylene group, at this
site. A parallel deduction may be made, though with somewhat less
−1
of the comparatively strong band in the 1640–1670 cm region
that is associated with the ꢀ(C C) mode, as has been empha-
sised previously [7]. The presence of a methyl substituent in
the allylic or homoallylic position does not alter the wavenum-
−
1
confidence, about the weaker band in the 1385–1378 cm range
in the spectra of 1, 1-1CH and 1-3CH but not 1-2CH ; this feature
−1
ber of this band by more than ± 1 cm , thus indicating that the
3
3
3
localised vibration approximation holds good in what must be a
relatively “pure” mode. The attachment of a methyl group directly
to the C C group raises the wavenumber of the band by approxi-
may be attributable to deformation of the methylene group in the
allylic position, separated from the methoxy group by an interven-
ing carbon atom. However, in this case, the band is absent not only
−1
mately 10 cm if the double bond has a 1,1-substitution pattern,
but by appreciably more (15 and 30 cm , respectively, for a cis
and trans 1,2-substitution pattern). Moreover, these characteristic
wavenumbers may have at least semi-quantitative value, as shown
by the presence of bands at both 1673 and 1655 cm in the spec-
trum of 1-4CH , which was prepared from pent-3-enol that was
known to be predominantly trans but to contain a minor amount of
the cis geometric isomer. The relative intensities of the associated
Raman bands suggest that these isomers were present in the ratio
of ∼3:1. Although less obvious trends are found in the presence or
absence of the weaker bands at lower wavenumber, chiefly because
of extensive mode mixing, some information may be deduced, par-
ticularly from the bands near 1120, 960 and 830 cm , which may
be attributed to vibration of a C C and adjacent C C subunit, the
ꢂ(CH ) mode of the H C C entity, and a ı(CH ) mode of a carbon
bound methyl group, respectively.
from the spectrum of 1,2CH , where a CHCH subunit has replaced
−1
3
3
the CH2 group, but also from the spectrum of 1-4CH , which might
3
be expected to show this band.
The complex band or bands of medium to weak intensity at var-
−1
−1
ious positions in the 1301–1282 cm range probably is(are) best
associated with CH3 deformation mode(s) of the methoxy group
3
−
1
that is present in each ether. The band at 1293 cm with a shoulder
at higher wavenumber the spectra of 1-1CH , 1-3CH and 1-4CH3,
3
3
but not 1 or 1-2CH , may indicate that no methyl group is present at
3
the allylic position or at the more distant end of the butenyl chain.
However, this interpretation must be treated with caution because
of the weakness of the band.
−1
The (moderately) weak band that is present in the
−1
1
123–1113 cm
be associated with a vibration of the C
present in the spectrum of 1-3CH , in which this feature is not
range in the spectrum of each ether cannot
2
2
3
C
H entity because it is
3
present. Nor can it be assigned to a specific vibration of the H C
C
2
Acknowledgements
subunit because it is present in the spectrum of 1-4CH . It may
3
be associated with a ꢀ(C C) mode near to a C C bond and/or a
methyl substituent.
Financial support from the British Mass Spectrometry Society (a
grant from the General Fund) to permit the purchase of alkenols
from which the alkenyl methyl ethers were prepared is gratefully
acknowledged.
In contrast, the band of moderate intensity between 967 and
−1
9
55 cm in the spectra of all the ethers apart from 1-4CH3 can be
assigned with greater certainty. Its absence from the spectrum of
-3CH3 leads to a clear association with a vibration of the terminal
1
References
H C C entity, presumably the ꢂ(CH ) mode.
2
2
−
1
The band of variable intensity and appearance near 830 cm
[
1] P.J. Hendra, C. Jones, G. Warnes, Fourier Transform Raman Spectroscopy, Ellis
Harwood, Chichester, UK, 1991.
that is absent from the spectrum of 1 may logically be ascribed to
a deformation mode of the carbon bound methyl group. Since it
[2] D.B. Chase, J.F. Rabolt, Fourier Transform Raman Spectroscopy: From Concept
to Experiment, Academic Press, San Diego and New York, USA, 1994.
is weaker in the spectrum of 1-1CH , which has the methyl sub-
3
[
3] D.A. Long, The Raman Effect, J. Wiley and Sons, Chichester, UK, 2002.
stituent in the homoallyl position, it may well be preferentially
associated with a methyl group that is either directly attached to a
double bond or is adjacent to it in the allylic position.
[4] T. Karapanayiotis, S.E. Jorge Villar, R.D. Bowen, H.G.M. Edwards, Analyst 129
(2004) 613–618.
[
[
5] A.C. Williams, H.G.M. Edwards, B.W. Barry, J. Raman Spectrosc. 95 (1994) 25.
6] H.G.M. Edwards, M. Gniadecka, S. Peterson, J.P. Hart Hansen, O.F. Nielsen, D.H.
Christensen, H.C. Wulf, Vib. Spectrosc. 28 (2002) 3–15.
−
1
The part of the spectra between about 700 and 300 cm con-
tains two bands that vary in wavenumber with a discernable
pattern. The wavenumber of these bands decreases from 484 to
[7] R.D. Bowen, H.G.M. Edwards, D.W. Farwell, I. Rusike, D.M. Saunders, J. Chem.
Res. (S) (1998) 426, J. Chem. Res. (M) (1998) 1901.
[
[
8] R.D. Bowen, H.G.M. Edwards, D.W. Farwell, J. Mol. Struct. 976 (2010) 36–41.
9] R.D. Bowen, A.D. Wright, S.R. Thomas, K.R. Jennings, A. Maccoll, Org. Mass
Spectrom. 26 (1991) 421.
−
1
4
06 and 349 to 283 cm , respectively, in increments between 13
−1
and 39 cm on progressing from 1-1CH3 to 1-4CH . Neither band
3

3
2
R.D. Bowen et al. / Spectrochimica Acta Part A 93 (2012) 26–32
[
[
10] R.D. Bowen, P. Clifford, R.T. Gallagher, Eur. Mass Spectrom. 2 (1996) 233.
11] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
V.G. Zakrewski, J.A. Montgomery, R.E. Stratmann, J.C. Burant, S. Dappriach, J.M.
Millam, A.D. Daniels, K.N. Kudin, M.C. Strain, O. Farkas, J. Tomasi, V. Barone,
M. Cossi, R. Cammi, B. Mennuci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski,
G.A. Petersson, P.Y. Ayala, Q. Cui, K. Morokuma, D.K. Malick, A.D. Rabuck, K.
Raghavachari, J.B. Foresman, J. Cioslowski, J.V. Ortiz, B.B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, A. Komaromi, R. Gompertz, R.L. Martin, D.J. Fox, T. Keith,
M.A. Al-laham, J.Y. Peng, A. Nanayakkara, C. Gonzales, M. Challacombe, P.M.W.
Gill, B.G. Johnson, W. Cheng, M.W. Wong, J.L. Andres, M. Head-Gordon, E.S.
Replogle, J.W. Pople, Gaussian 98 Revision A3, Gaussian Int., Pittsburgh, PA,
1998.
[12] R.D. Bowen, H.G.M. Edwards, K.A. Hope, in preparation.