Mass spectra of halogenostyrylbenzoxazoles

Article abstract of DOI:10.1016/j.ijms.2012.08.016

Several series of styrylbenzoxazoles of general formula XC ₆H₃(NCO)CHCHC₆H₄Y [X = F, Cl or Br; Y = H, F, Cl, Br, CH₃ or CH₃O] have been investigated by positive ion electrospray and electron ionization mass spectrometry. These compounds, many of which are biologically active or have pharmaceutical potential, show in their electrospray spectra strong peaks for MH⁺ ions, which undergo relatively little fragmentation. The electron ionization spectra are extremely clean, being dominated by the loss of an atom or radical, Y, from the ortho position of the pendant ring, by a rearrangement that may be interpreted as a proximity effect. The resultant [M-Y]⁺ ions are exceptionally stable and rarely undergo further fragmentation. The analytical value of this proximity effect, which is analogous to intramolecular aromatic substitution, in revealing the presence of a substituent in the pendant ring and determining its position, is emphasized. Elimination of a species (including H or F) derived from an ortho substituent in the pendant ring occurs even when apparently more favourable alternative fragmentation is possible by direct cleavage of the CX bond (X = Cl or Br) in the benzoxazole ring.

Full text of DOI:10.1016/j.ijms.2012.08.016

International Journal of Mass Spectrometry 345–347 (2013) 120–131
Contents lists available at SciVerse ScienceDirect
International Journal of Mass Spectrometry
journal homepage: www.elsevier.com/locate/ijms
Mass spectra of halogenostyrylbenzoxazoles
Stephen T. Ayrtona, Jekaterina Panovaa, Adam R. Michalika, William H.C. Martina,
Richard T. Gallagherb, Richard D. Bowena,∗
a Chemical and Forensic Sciences, School of Applied Sciences, University of Bradford, Bradford BD7 1DP, UK
b Oncology IMED, Astrazeneca, Alderley Park, Macclesfield SK10 4TG, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 19 June 2012
Received in revised form 27 July 2012
Accepted 7 August 2012
Available online 20 August 2012
Several series of styrylbenzoxazoles of general formula XC6H3(NCO)CH CHC6H4Y [X = F, Cl or Br; Y = H,
F, Cl, Br, CH3 or CH3O] have been investigated by positive ion electrospray and electron ionization mass
spectrometry. These compounds, many of which are biologically active or have pharmaceutical potential,
show in their electrospray spectra strong peaks for MH+ ions, which undergo relatively little fragmen-
tation. The electron ionization spectra are extremely clean, being dominated by the loss of an atom or
radical, Y•, from the ortho position of the pendant ring, by a rearrangement that may be interpreted as
a proximity effect. The resultant [M−Y]+ ions are exceptionally stable and rarely undergo further frag-
mentation. The analytical value of this proximity effect, which is analogous to intramolecular aromatic
substitution, in revealing the presence of a substituent in the pendant ring and determining its posi-
tion, is emphasized. Elimination of a species (including H• or F•) derived from an ortho substituent in
the pendant ring occurs even when apparently more favourable alternative fragmentation is possible by
direct cleavage of the C X bond (X = Cl or Br) in the benzoxazole ring.
Dedicated to Professors Keith R. Jennings
and Jim Scrivens on the occasion of their
80th and 60th Birthdays, respectively, in
recognition of their contributions to the
development and application of
experimental mass spectrometry in a wide
range of scientific disciplines.
© 2012 Elsevier B.V. All rights reserved.
Keywords:
Benzoxazoles
Styrylbenzoxazoles
Rearrangement
Proximity effect
Halogen atom elimination
1. Introduction
interaction of separate parts of a molecular ion through space can
provide extremely useful structural information by permitting
Mass spectrometry has been applied in structure elucidation for
at least six decades, but it is increasingly regarded in some quarters
merely as a means of obtaining molecular mass or molecular for-
mula information. Thus, even experienced organic chemists often
erroneously believe that mass spectrometry can never reveal the
substitution pattern of aromatic compounds. This misunderstand-
ing is unfortunate, because it prevents maximum advantage being
taken of the analytical value of mass spectrometry.
orthodisubstituted aromatic compounds to be distinguished from
their meta and para isomers.
The importance of fragmentations that involve rearrangement
was recognized early in the development of electron ionization
(historically known as electron impact, EI) mass spectrometry. The
McLafferty rearrangement [1,2] of ionized carbonyl compounds
is a celebrated example (Eq. (1)). Ortho effects, as illustrated by
the loss of H2O from ionized 1,2-disubstituted benzenes such
as anthranilic acid, H2NC6H4CO2H, and related species [3] (Eq.
(2)) are another example of how rearrangements involving the
(1)
(2)
∗
Another interesting rearrangement of this kind can arise when
cyclization of the molecular ion is followed by elimination of
Corresponding author. Tel.: +44 1274 233774; fax: +44 1274 235350.
E-mail address: r.d.bowen@bradford.ac.uk (R.D. Bowen).
1387-3806/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ijms.2012.08.016

S.T. Ayrton et al. / International Journal of Mass Spectrometry 345–347 (2013) 120–131
121
Scheme 1. Loss of H• from PhCH CHCOR•+ species.
a substituent from an existing ring, with the production of an
especially stable product ion in which a new ring has been formed.
The archetypal example was discovered through the observation
[4] of unexpectedly strong [M−H]+ signals in the EI of cinnamic
acid, C6H5CH CHCO2H, and several other compounds containing
the C6H5CH CHC(R)O subunit (R = H, OH, CH3 or C6H5). Deuterium
labelling revealed that the eliminated hydrogen atom originated
from the ortho position of the aromatic ring. The rearrangement
is readily interpreted as shown in Scheme 1. Subsequently, further
support for this mechanism was provided by establishing that
the structure of the fragment ion formed from ionized ben-
zylacetones (R = CH3) was the 2-methylbenzylpyrillium cation
[5].
This class of rearrangement, which can be described as an
intramolecular aromatic substitution [6], is sometimes referred to
as a proximity effect. Its analytical value is by no means restricted
to cinnamyl derivatives. Thus, aurones [7] and other natural prod-
ucts show [M−R]+ ions in their EI mass spectra when the R group
is suitably located in an ortho position. Polycyclic compounds
containing an ArCH CH side chain also tend to show signals in
their mass spectra that may be attributed to a proximity effect. For
example, prominent [M−H]+ and [M−Y]+ signals are seen in the
spectra of a wide range of styrylbenzazoles, 1, of general formula
C6H4(NCZ)CH CHC6H4Y [Z = NH, O or S; Y = H, F, Cl, Br, CH3, CH3O or
NO2] when the substituent, Y, is in the ortho position of the pendant
ring (Scheme 2) [8]. Following isomerization of the trans C C bond
Scheme 2. Loss of Y• from C6H4(NCZ)CH CHC6H4Y+ species grnerated under electron ionization conditions.
Scheme 3. Synthetic routes to 5XYSBOs (with reagents and conditions). (i) HNO3, 16 ◦C, 2 min (X = F); 2 h (X = Br). (ii) H2/Pd (X = F); SnCl2/HCl 4.5 h (X = Br). (iii) Ac2O/ꢀ
(X = F/Cl); CH3C(OCH3)3/reflux, 1.5 h (X = Br). (iv) YC6H4CHO (Y = H, F, Cl, Br, CH3, CH3O), C6H5CH2N(C2H5)3+Cl−/NaOH(aq)/DCM.

122
S.T. Ayrton et al. / International Journal of Mass Spectrometry 345–347 (2013) 120–131
in 1•+, cyclization of 1cis•+ (shown as 1c•+ in Scheme 2) occurs to
form 2•+, which may then lose Y•, the radical corresponding to the
substituent that was initially present in the ortho position of the
pendant ring, to give the exceptionally stable tetracyclic polyaro-
matic ion, 3. The localized lone pair in the plane of the ring system
on the nitrogen atom, which possesses nucleophilic (and basic)
properties, undoubtedly plays a vital role in the key cyclization
step.
Recent work illustrates the importance of styrylbenzoxazoles
[SBOs, 1, Z = O], styrylbenzimidazoles [SBIs, 1, Z = NH] and styryl-
benzothiazoles [SBTs, 1, Z = S] in a variety of industrial and
medicinal contexts [9–12]. Therefore, a means of determining the
presence and position of bromo and other halogeno substituents in
the dihalogenoSBOs, ideally at high sensitivity, is of considerable
contemporary interest. In order to establish the general value of
the proximity effect in these systems, several series of disubstituted
SBOs with at least one halogen in the fused ring have been prepared
and investigated by both electrospray (ESI) and EI mass spectrom-
etry.
2. Experimental
2.1. Synthesis
The SBOs were prepared by the base-catalysed condensation of
the appropriate 5-halogeno-2-methylbenzoxazole with the requi-
site aromatic aldehyde under phase transfer conditions. In a typical
experiment, equimolar quantities (5 mmol) of the starting materi-
als were dissolved in dichloromethane (20–50 ml) in the presence
of benzyltriethylammonium chloride (3 mmol) and stirred mag-
netically under a nitrogen atmosphere as an aqueous solution of
sodium hydroxide (50%, w/v, 5 ml) was added dropwise over a
period of 10 min. After being stirred for 2–36 h until analytical
thin layer chromatography indicated that the reaction was com-
plete, the mixture was diluted with water (50 ml) and the SBO
was extracted with dichloromethane (3× 20 ml), dried (MgSO4), fil-
tered, evaporated under reduced pressure and recrystallized from
aqueous methanol or ethanol. This method is a modification of
that reported previously [8,13]. All the products had satisfactory
1H NMR and IR spectra.
Fig. 1. CID spectra of MH+ (C16H1079BrFNO+) obtained under positive ion electro-
spray conditions. Mass spectra (top to bottom) 5F2ꢀBrSBO, 5F3ꢀBrSBO and 5F4ꢀBrSBO.
auto-sampler (AcquityTM Sample Manager, Waters Ltd, Elstree, UK).
The LC and MS systems were both controlled by Xcalibur software,
version 2.1.0 (Thermo Scientific Ltd, Hemel Hempstead, UK).
The Liquid Chromatography column was a Waters AcquityTM
UPLC BEH C18, 1.7 mm, 2.1 mm × 100 mm. The sample was
dissolved in 50:50 (v/v) water/methanol at ∼0.01 mg/ml
The 5-halogeno-2-methylbenzoxazoles were prepared by the
treatment of the corresponding 2-amino-4-halogenophenol with
either acetic anhydride (X = F or Cl) [14] or trimethylorthoac-
etate (X = Br) [15]. 2-Amino-4-chlorophenol was commercially
available; 2-amino-4-fluorophenol and 2-amino-4-bromophenol
were obtained by reduction of the corresponding 2-nitro-4-
halogenophenol with H2/Pd [16] and SnCl2·2H2O [17], respec-
tively. Attempts to reduce 2-nitro-4-bromophenol with H2/Pd
resulted in debromination of the aromatic ring. The 2-nitro-
4-halogenophenols were prepared in batches from the 4-
halogenophenols by careful nitration in acetic acid at 15–20 ◦C [18]
on a small scale (2–5 g of halogenophenol). Nitration of fluorophe-
nol required only 2 min, but nitration of bromophenol required
2–16 h (typically around 3 h); attempts to scale up the reaction
resulted in reduced yields. Scheme 3 summarizes these synthetic
routes.
2.2. Mass spectrometry
Liquid Chromatography–Mass Spectrometry analysis under
ESI conditions [19] was carried out on a hybrid Linear Ion
Trap-OrbiTrap mass spectrometer system (Thermo Scientific
LTQ OrbiTrap XL) fitted with an Ultra High Performance Liquid
Chromatography (UHPLC) system. The UHPLC system consisted
of a binary pump (AcquityTM Binary Solvent Manager) and an
Fig. 2. CID spectra of MH+ (C16H1079BrFNO+) obtained under positive ion electro-
spray conditions. Mass spectra (top to bottom) 5Br2ꢀFSBO, 5Br3ꢀFSBO and 5Br4ꢀFSBO.

S.T. Ayrton et al. / International Journal of Mass Spectrometry 345–347 (2013) 120–131
123
concentration; 10 ␮l of the resultant solution was injected
into the system. The mobile phase was set at a flow rate of
0.45 ml/min. Mobile phase A was 0.05% aq. formic acid and mobile
phase B was 0.05% formic acid in methanol. The following gradient
was used: T = 0.0, A = 90%, B = 10%, T = 8.0, A = 20%, B = 80%, T = 8.01,
A = 5%, B = 95%, T = 11.0, A = 5%, B = 95%. At T = 11.01 min the system
reverted to the starting conditions and was held for 4 min to
recondition the column.
The ESI source conditions used were: positive ion mode, with
capillary voltage at +4 kV, the sheath gas flow rate was set to 40 and
the auxiliary flow rate to 20 (arbitrary units). The source capillary
temperature was set to 250 ◦C.
Direct probe EI mass spectrometry analysis was carried out on
a Shimadzu QP-2010 quadrupole MS system fitted with a heated
solids probe and controlled by ‘GCMS solutions’ software, version
2.0 (Shimadzu UK Ltd., Milton Keynes, UK). Some solid samples
were placed directly in a disposable glass vial. Alternatively, the
sample was dissolved in methanol at ∼1 mg/ml concentration; 1 ␮l
of this solution was then placed into the glass vial. The glass vial was
inserted into the end of the solids probe and then placed directly
into the mass spectrometer ion source via a vacuum lock. The probe
was then heated from ambient to 320 ◦C over 10 min. The mass
spectrometer used 70 eV electrons to ionize the thermally desorbed
sample; data were acquired over the m/z 50–600 at a scan speed of
1250 m/z units/s at unit mass resolution.
The OrbiTrap mass spectrometer collected data every ∼0.25 s,
recording a mass spectrum over the m/z range 100–900 at
a resolution of 7500 (FWHM, Full width at half maximum)
3. Results and discussion
followed by an MS2 product ion scan at
a resolution of
7500 (FWHM) from the most intense ion detected in the MS
scan.
Three main series of SBOs have been investigated, with a fluoro,
chloro or bromo substituent in the 5 position of the benzoxazole
Fig. 3. Electron ionization mass spectra of (top to bottom) 5Cl2ꢀBrSBO, 5Cl3ꢀBrSBO and 5Cl4ꢀBrSBO.

124
S.T. Ayrton et al. / International Journal of Mass Spectrometry 345–347 (2013) 120–131
ring (Scheme 2). This substituent is identified by the numeric pre-
fix “5”, followed by the symbol for the halogen in question. Thus,
“5FSBOs” are 5-fluorostyrylbenzoxazoles with a fluorine atom in
the 5 position of the fused ring. The position of any substituent Y
in the pendant ring is indicated by a second prefix that is distin-
guished from the first by being given a prime. Thus, “5F3ꢀBrSBO”
is 5-fluoro-2-(3-bromophenylethenyl)benzoxazole. The series of
5FSBOs, 5ClSBOs and 5BrSBOs were prepared in order to study the
influence of the nature and position of X and Y on the ESI and EI
mass spectra of the disubstituted SBOs. A specific objective of this
study was to investigate the competition between loss of Y• from
the pendant ring by the proximity effect and elimination of X• by
direct cleavage of the C X bond in the benzoxazole ring. In particu-
lar, does loss of F• by the proximity effect dominate loss of Cl• or Br•
from the benzoxazole ring? Loss of F• is a comparatively uncommon
fragmentation of an ionized monofluorocompound (elimination
of HF or C2H2 from aliphatic and aromatic species, respectively,
usually occurs far more readily), but elimination of Cl• and, espe-
cially, Br•, are frequently observed. This contrast reflects, at least
in part, the great strength of the C F bond (mean thermochemi-
cal bond energy = 485 kJ mol−1), whereas the C Cl and C Br bonds
are (much) weaker (bond energy = 327 and 285 kJ mol−1, respec-
tively) [20]. Parallel remarks apply to the competition between loss
of Cl• from the pendant ring and elimination of Br• from the ben-
zoxazole ring. Other representative series, with a fluoro or chloro
substituent in the fused ring and either a methyl or methoxy group
in the pendant ring, were also prepared, as were a limited num-
ber of analogues with two different halogen atoms in the pendant
ring.
3.1. ESI spectra
As expected given the structure of the benzoxazole entity,
in which the nitrogen atom possesses a localized lone pair of
electrons orthogonal to the aromatic ␲-system, protonation of the
SBOs occurs readily under positive ESI conditions to give spectra
that are dominated by intense MH+ signals. These MH+ signals
permit the relative molecular mass and molecular formula to be
Fig. 4. Electron ionization mass spectra of (top to bottom) 5F2ꢀClSBO, 5F3ꢀClSBO and 5F4ꢀClSBO.

S.T. Ayrton et al. / International Journal of Mass Spectrometry 345–347 (2013) 120–131
125
easily determined with very high sensitivity. In each case inves-
tigated in this work, the measured m/z value was within 1 ppm of
the theoretical value; the deviation from the theoretical value was
often less than 0.1 ppm. Bearing in mind the comparatively low
relative molecular mass of the SBOs, this mass accuracy is amply
sufficient to determine the molecular formula, particularly in
cases when the presence of an atom of chlorine and/or bromine is
disclosed by the appearance of the 37Cl or 81Br isotope satellite(s)
in the low resolution spectrum.
occurs less readily than elimination of HBr, as would be expected
intuitively given the strength of the C F bond.
Although these trends indicate that protonation on nitrogen
under ESI conditions effectively suppresses the strong proximity
effect that occurs for ionized SBOs, it remains significant that elim-
ination of a halogen atom from the benzoxazole ring of the MH+
ion generally is less important under CID conditions than loss of
hydrogen halide from the pendant ring. This tendency persists even
when a bromo substituent is present in the benzoxazole ring and
a fluoro substituent is in the ortho position of the pendant ring.
In addition, loss of hydrogen halide from MH+ occurs more readily
when the halogeno substituent is in the ortho position. However,
these trends are not sufficiently strong to be used as a general
means of obtaining reliable structural information on isomeric
SBOs.
These ESI [19] spectra are exceptionally clean, containing rel-
atively few product ion signals. Even under collision induced
dissociation (CID) conditions [21], the only product ion with
a
relative abundance (RA) greater than about 10% of MH+
is [MH−HHal]+, where Hal is the halogen in the pendant
ring. Representative CID spectra are shown in Figs. 1 and 2
for the three isomeric 5FBrSBOs and the complementary set of
5BrFSBOs. There is a slightly greater preference for loss of HBr
from MH+ formed from 5F2ꢀBrSBO, for which the ratio of RA
of [MH−HBr]+:MH+ is ∼2.2, whereas the corresponding ratio for
5F3ꢀBrSBO and 5F4ꢀBrSBO is ∼1.0 and 0.9, respectively. Similar
behaviour is found for the complementary 5BrFSBOs, which show
a much lower [MH−HF]+:MH+ ratio (∼0.3, 0.1 and 0.1, respectively,
for 5Br2ꢀFSBO, 5Br3ꢀFSBO and 5Br4ꢀFSBO). In the CID spectra of
5Br3ꢀFSBO and 5Br4ꢀFSBO, the [MH−HBr]•+ and [MH−HF]+ signals
have quite similar RIs. It seems clear, therefore, that loss of HF
3.2. EI spectra
In contrast to the ESI spectra, which are dominated by MH+
signals, the EI spectra contain a limited number of strong prod-
uct ion peaks, which are of obvious analytical value. In common
with the ESI spectra, these EI spectra are very clean, typically
comprising only two or three peaks of high relative intensity
(RI). Illustrative spectra of sets of isomeric species are shown in
Figs. 3–8.
Fig. 5. Electron ionization mass spectra of (top to bottom) 5Br2ꢀFSBO, 5Br3ꢀFSBO and 5Br4ꢀFSBO.

126
S.T. Ayrton et al. / International Journal of Mass Spectrometry 345–347 (2013) 120–131
3.2.1. Comparison with earlier work
is typical of SBOs with Y in either a meta or para position. To some
extent, this systematic variation in the M•+:[M−H]+ ratio reflects
the fact that only one ortho hydrogen atom is present in the SBOs
with Y in the other ortho position, thus reducing the probability of
cyclization and eventual loss of a hydrogen atom. However, other
factors also must operate, because the M•+:[M−H]+ ratio is greatest
when Y = Br and least when Y = F.
The spectra reported in this work show similar general patterns
to those discussed previously [8]. The principal signals correspond
to M•+, [M−H]+ and [M−Y]+ ions, with very few significant peaks
for other singly charged ions.
3.2.2. Molecular ion signals
Significant M•+ signals are always present, but more intense
peaks are found for either [M−H]+ or [M−Y]+ (Y =/ H). Several infor-
mative trends are found in the ratios of the RI of these signals
(Tables 1 and 2).
Similar trends are evident in the FCH3SBO, FCH3OSBO, ClCH3SBO
and ClCH3OSBO series, in which the M•+:[M−H]+ratio is invariably
higher when a methoxy group is present in the pendant ring. The
ratio lies in the range 0.44–0.53 when the Y = CH3O in the meta
or para position, but it is 2.2–2.7 when this substituent is in the
ortho position. The analogous value when Y = CH3 lies in the range
0.34–0.37 and 1.1–1.3, respectively. The relative insensitivity of
these ratios to the nature of the halogeno substituent in the benzox-
azole ring indicates that local influences in and around the pendant
ring are dominant in determining the rate of hydrogen atom loss.
More spectacular variations in the opposite sense are evident
in the M•+:[M−Y]+ratio, which is always considerably lower, by
First, the M•+:[M−H]+ratio is always appreciably greater, usu-
ally by a factor of at least 2, in the spectra of SBOs with an ortho
substituent [Y = F, Cl, Br, CH3 or CH3O], than in the spectra of their
meta and para isomers, regardless of the nature of Y. A ratio close
to or above unity (ranging in the dihalogeno series from 0.82 for
5Cl2ꢀFSBO to 2.0 for 5F2ꢀBrSBO) is indicative of a substituent in the
ortho position. In contrast, a ratio close to or below 0.5 (ranging in
the dihalogeno series from 0.30 for 5F4ꢀClSBO to 0.48 for 5F3ꢀBrSBO)
Fig. 6. Electron ionization mass spectra of (top to bottom) 5F2ꢀCH3SBO, 5F3ꢀCH3SBO and 5F4ꢀCH3SBO.

S.T. Ayrton et al. / International Journal of Mass Spectrometry 345–347 (2013) 120–131
127
at least a factor of 10 and sometimes by a factor of over 100, for
SBOs with an ortho substituent [Y = F, Cl, Br, CH3 or CH3O] irrespec-
tive of the nature of Y. In the dihalogeno series, a ratio below 0.5
(ranging from 0.11 for 5Cl2ꢀBrSBO to 0.45 for 5Cl2ꢀFSBO) is indica-
tive of a halogeno substituent in the ortho position, whereas a
value above unity (ranging from 1.7 for 5F3ꢀBrSBO to at least 75 for
5Br2ꢀFSBO) is typical of pendant rings with no halogeno substituent
in an ortho position. Similar trends are observed in the M•+:[M−Y]+
ratio in the spectra of the sets of isomeric FCH3SBOs, FCH3OSBOs,
ClCH3SBOs and ClCH3OSBOs. This ratio lies in the range 0.08–0.26
for 5-HalSBOs with either a methoxy or methyl group in the ortho
position, but it is never less than 8 and sometimes 25 or more for
isomeric species with the substituent in either the meta or para
position. The main reason for the more pronounced variations in
the M•+:[M−Y]+ ratio than in the M•+:[M−H]+ ratio is that elim-
ination of Y• usually is preferable to loss of H• by the proximity
effect.
this ratio does show a systematic dependence on the nature of
the halogeno substituent in the pendant ring: this ratio increases
from 0.11 or 0.12 (Y = Br) to 0.22 or 0.26 (Y = Cl) to 0.38 or 0.45
(Y = F) as the electronegativity of the halogen in the ortho position
increases (and its size decreases). Both steric and electronic fac-
tors must be at work here because an even lower ratio of 0.08 or
0.09 is found when Y = OCH3, whereas the higher ratio of 0.25 or
0.26 found when Y = CH3 is roughly the same as that found when
Y = Cl.
Apart from its obvious value in identifying SBOs with a sub-
stituent in an ortho position, the M•+:[M−Y]+ offers a more subtle
means of distinguishing between isomeric SBOs with a substituent
in either a meta or para position. The ratio is always greater for para
than for meta isomers (usually by a factor of about 2). However,
deductions based on this ratio must be treated with caution: when
a methyl group is present in the pendant ring, the difference in the
ratio is barely greater than the experimental error. Nevertheless,
if both isomeric species were available for analysis under identical
conditions, a larger M•+:[M−Y]+ ratio would be characteristic of the
para isomer.
As with the M•+:[M−H]+ ratio, the M•+:[M−Y]+ ratio does not
show a marked dependence on the nature of the halogeno sub-
stituent (F or Cl) in the benzoxazole ring. On the other hand,
Fig. 7. Electron ionization mass spectra of (top to bottom) 5Cl2ꢀCH3OSBO, 5Cl3ꢀCH3OSBO and 5Cl4ꢀCH3OSBO.

128
S.T. Ayrton et al. / International Journal of Mass Spectrometry 345–347 (2013) 120–131
Fig. 8. Electron ionization mass spectra of (top to bottom) 5F2ꢀCl4ꢀBrSBO and 5F2ꢀBr4ꢀClSBO.
3.2.3. [M−H]+ signals
that of the M•+ peak when there is no ortho substituent (other
than hydrogen) in the pendant ring, are in themselves analyti-
cally useful. In many such cases, [M−H]+ is the base peak in the
spectrum.
Loss of a hydrogen atom from ionized molecules of reasonable
size is a comparatively rare fragmentation. Therefore, the strong
[M−H]+ signals, which typically have an RI approximately twice
Table 1
Relative intensities of M•+, [M−H]+, [M−X]+ and [M−Y]+ signals in the electron ionization mass spectra of mono- and dihalogenostyrylbenzoxazoles.
Compound
Relative intensity (RI) (%)a,b
[M]•+
[M−H]+
[M−X]+
[M−Y]+
[M−Y]+/[M−X]+
[M−Y]+/[M−H]+
[M]•+/[M−H]+
[M]•+/[M−Y]+
5FSBO
33
12
45
47
100
6
97
98
5
–
100
26
–
>200
>52
–
0.33
2.0
0.46
0.48
–
5F2ꢀBrSBO
5F3ꢀBrSBO
5F4ꢀBrSBO
<0.5
<0.5
<0.5
16.6
0.27
0.16
0.12
1.73
2.94
16
>32
5F2ꢀClSBO
5F3ꢀClSBO
5F4ꢀClSBO
26
43
30
22
100
100
<0.5
<0.5
<0.5
100
16
5
>200
>32
>10
4.5
0.16
0.05
1.2
0.43
0.30
0.26
2.69
6
5ClSBO
37
11
34
29
100
6
73
74
<0.5
<0.5
<0.5
<0.5
–
100
19
–
>200
>38
–
0.37
1.8
0.43
0.35
–
5Cl2ꢀBrSBO
5Cl3ꢀBrSBO
5Cl4ꢀBrSBO
4.5
0.26
0.14
0.11
1.79
2.9
10
>20
5Cl2ꢀFSBO
5Cl3ꢀFSBO
5Cl4ꢀFSBO
45
43
35
55
100
100
2
0.5
0.5
100
1
<0.5
50
2
<1
1.8
0.01
<0.005
0.82
0.43
0.35
0.45
43
>70
5BrSBO
38
38
45
38
100
36
97
18
15
20
20
–
100
3
–
6.6
0.15
<0.025
–
2.7
0.03
<0.005
0.38
1.1
0.46
0.39
–
0.38
15
5Br2ꢀFSBO
5Br3ꢀFSBO
5Br4ꢀFSBO
98
<0.5
>76
5Br2ꢀClSBO
5Br3ꢀClSBO
5Br4ꢀClSBO
22
29
28
13
74
73
1
12
13
100
7
4
100
0.5
0.3
7.96
0.09
0.05
1.7
0.39
0.38
0.22
4.14
7
a
RI = relative intensity, measured by peak height and normalized to a value of 100 units for the base peak. The M•+ signal is assigned as corresponding to the combination
of the most abundant isotopes of each element (e.g., 79Br35Cl). This peak is not always the strongest for a particular type of ion because the “M+2” peak in the BrClSBOs is a
combination of the 81Br35Cl and 79Br37Cl isotope satellites of M•+ Parallel remarks apply to the [M−H]+ and some other signals.
b
No allowance is made for the contribution to the M•+ signal from the 13C isotope satellite of [M−H]+ or the [M−H]+ signal from the 81Br or 37Cl isotope satellites of M•+
;
either of these contributions may be significant, but do not appreciably alter discussion of the underlying trends.

S.T. Ayrton et al. / International Journal of Mass Spectrometry 345–347 (2013) 120–131
129
Table 2
Relative intensities of M•+, [M−H]+, [M−X]+ and [M−Y]+ signals in the electron ionization mass spectra of 5-halogenostyrylbenzoxazoles with a methyl or methoxy group in
the pendant ring.
Compound
Relative intensity (RI) (%)a,b
[M]•+
[M−H]+
[M−X]+
[M−Y]+
[M−Y]+/[M−X]+
[M−Y]+/[M−H]+
[M]•+/[M−H]+
[M]•+/[M−Y]+
5Cl2ꢀOMeSBO
5Cl3ꢀOMeSBO
5Cl4ꢀOMeSBO
8
44
50
3
100
100
<0.5
<0.5
<0.5
100
5
2
>200
>10
>4
33.3
0.05
0.02
2.67
0.44
0.50
0.08
8.8
25
5Cl2ꢀMeSBO
5Cl3ꢀMeSBO
5Cl4ꢀMeSBO
26
37
34
20
100
100
2
2
2
100
4
3
50
2
1.5
5
0.04
0.03
1.3
0.37
0.34
0.26
9.25
11.33
5F2ꢀOMeSBO
5F3ꢀOMeSBO
5F4ꢀOMeSBO
9
44
53
4
100
100
<0.5
<0.5
<0.5
100
4
2
>200
>2
>1
20
0.04
0.02
2.25
0.44
0.53
0.09
11
26.50
5F2ꢀMeSBO
5F3ꢀMeSBO
5F4ꢀMeSBO
25
36
34
22
100
100
<0.5
<0.5
<0.5
100
4
2.5
>200
2
>5
4.55
0.04
0.025
1.14
0.36
0.34
0.25
9
13.60
a
RI = relative intensity, measured by peak height and normalized to a value of 100 units for the base peak. The M•+ signal is assigned as corresponding to the combination
of the most abundant isotopes of each element (e.g., 79Br35Cl). This peak is not always the strongest for a particular type of ion because the “M+2” peak in the BrClSBOs is a
combination of the 81Br35Cl and 79Br37Cl isotope satellites of M•+ Parallel remarks apply to the [M−H]+ and some other signals.
b
No allowance is made for the contribution to the M•+ signal from the 13C isotope satellite of [M−H]+ or the [M−H]+ signal from the 81Br or 37Cl isotope satellites of M•+
;
either of these contributions may be significant, but do not appreciably alter discussion of the underlying trends.
5.6, respectively. Even in the most favourable case, where Br• may
be lost by simple cleavage, elimination of H• via fairly extensive
rearrangement is so energetically advantageous that it dominates
the fragmentation of the ionized SBOs with no ortho substituent.
Table 3
Relative intensities of M•+, [M−H]+, [M−X]+ and [M−Y]+ signals in the electron ion-
ization mass spectra of styrylbenzoxazoles with two halogeno substituents in the
pendant ring.
Compound
Relative intensity (RI) (%)a,b
ꢀ
ꢀ
ꢀ
ꢀ
+
+
+
[M]•+
[M−Y2
]
[M−Y4
]
[M−Y2 ]+/[M−Y4
]
3.2.4. [M−X]+ and [M−Y]+ signals
2ꢀCl4ꢀBrSBO
2ꢀBr4ꢀClSBO
12
20
100
100
6
5
16.6
20
A major objective of this study was to ascertain whether or not
the formation of [M−Y]+ ions by the proximity effect remains of
diagnostic value in determining the position of a substituent in the
pendant ring even when an apparently more labile halogeno sub-
stituent is present in the benzoxazole ring. The [M−Y]+:[M−X]+
ratio allows the competition between the proximity effect and
direct cleavage of the C X bond in the benzoxazole entity to be
probed, both when X and Y are different halogens and when X is a
halogen and Y is a methyl or methoxy group.
2ꢀCl4ꢀFSBO
2ꢀF4ꢀClSBO
19
45
100
100
<0.5
67
>200
1.49
a
RI = relative intensity, measured by peak height and normalized to a value of
100 units for the base peak. The M•+ signal is assigned as corresponding to the com-
bination of the most abundant isotopes of each element (e.g., 79Br35Cl). This peak
is not always the strongest for a particular type of ion because the “M+2” peak in
the BrClSBOs is a combination of the 81Br35Cl and 79Br37Cl isotope satellites of M•+
Parallel remarks apply to the [M−H]+ and some other signals.
b
No allowance is made for the contribution to the M•+ signal from the 13C isotope
satellite of [M−H]+ or the [M−H]+ signal from the 81Br or 37Cl isotope satellites of
M•+; either of these contributions may be significant, but do not appreciably alter
discussion of the underlying trends.
It is immediately clear from these data that elimination of an
atom or group from the ortho position of the pendant ring invari-
ably dominates the loss of a halogen atom from the benzoxazole
ring. The [M−Y]+ signal is always the base peak in the spectra
of SBOs with a Y substituent in the ortho position. This trend is
evident when the pendant ring is unsubstituted, in which case
[M−H]+normally is the base peak of the spectrum, even when a
bromine atom is present in the benzoxazole ring. An apparent
exception to this rule occurs for BrClSBOs, but only because the
strongest signal in the spectra of these compounds corresponds
to a combination of isotope satellites of [M−H]+ containing either
The contrast in the [M−H]+:M•+ ratio (the inverse of the
M
•+:[M−H]+ quoted in Tables 1 and 2) when Y is in an ortho position
(typically 0.3–0.8, though values up to about 1.2 can occur when
Y = F) and when it is in a meta or para position (usually at least
2.0, though the value for 5F4ꢀOCH3SBO is only 1.9) is quite striking.
However, an even stronger contrast is found in the [M−H]+:[M−Y]+
ratio, which is much greater when no ortho substituent is present
in the pendant ring. The smallest ratio measured in such a case is 3.7
(for 5F3ꢀBrSBO), but values in the range 8–20 are more typical and
the ratio often exceeds 100 (for 5Cl3ꢀFSBO and 5Cl4ꢀSBO, for exam-
ple). Ratios well below unity are found when there is a substituent
in the ortho position. The highest ratio in the disubstituted series
is 0.82 (for 5Cl2ꢀFSBO), but a much smaller value (for instance, 0.08
for 5F2ꢀBrSBO and 0.03 for 5Cl2ꢀOCH3SBO) is frequently found.
The systematic increase in the [M−H]+:[M−Y]+ ratio for SBOs
with no ortho substituent, even when Y = Cl or Br, reveals that the
loss of H• by rearrangement via the proximity effect is so energeti-
cally favourable that it dominates the apparently more favourable
simple cleavage that would lead to elimination of Cl• or Br• from the
pendant ring. Similar remarks apply to the competition between
loss of a halogen atom from the benzoxazole ring by simple cleavage
and elimination of H• via the proximity effect. The [M−H]+:[M−Y]+
ratio in the spectra of 5FSBO, 5ClSBO and 5BrSBO is >200, ∼25 and
79Br and 37Cl or 81Br and 35Cl, whereas Table 1 shows the RIs of M•+
,
[M−H]+, [M−X]+ and [M−Y]+ signals containing the combination of
most abundant isotopes of each element (that is, 79Br and 35Cl).
Other atoms (F• or Cl•) and radicals (CH3 or CH3O•) are also
•
lost from the ortho position in preference to cleavage of the
C
Cl or C Br bond in the fused ring. The spectrum of 5Br2ꢀFSBO
(Fig. 5, top) is particularly striking. Loss of F• is by no means a
favourable process in ionized aromatic compounds. However, the
[M−Br]+:[M−F]+ratio is 0.15. This strong preference for expelling
a fluorine atom is even more striking when the characteristic frag-
mentations of C6H5Hal•+ species are considered. Whereas C6H5Br•+
and C6H5Cl•+ preferentially lose Br• and Cl•, respectively, with
appearance energies of 12.02 and 13.20 eV, the lowest energy frag-
mentation of C6H5F•+ is not elimination of F• but loss of C2H2, with
an appearance energy of 14.73 eV [22]. Despite these facts, ionized
5Br2ꢀFSBO eliminates F• rather than Br•, thus emphasizing the ease

130
S.T. Ayrton et al. / International Journal of Mass Spectrometry 345–347 (2013) 120–131
Table 4
Relative intensities of M•+, [M−H]+, [M−X]+ and [M−Y]+ signals in the electron ionization mass spectra of trihalogenostyrylbenzoxazoles.
Compound
Relative intensity (RI) (%)a,b
ꢀ
ꢀ
ꢀ
ꢀ
+
ꢀ
+
+
[M]•+
[M−F]+
[M−Hal2
]
[M−Hal4
]
[M−Hal2 ]+/[M−Hal4
]
[M−Hal2 ]+/[M−F]+
5F2ꢀCl4ꢀBrSBO
5F2ꢀBr4ꢀClSBO
23
10
<0.5
<0.5
100
100
8
<0.5
12.5
>200
>200
>200
a
RI = relative intensity, measured by peak height and normalized to a value of 100 units for the base peak. The M•+ signal is assigned as corresponding to the combination
of the most abundant isotopes of each element (e.g., 79Br35Cl). This peak is not always the strongest for a particular type of ion because the “M+2” peak in the BrClSBOs is a
combination of the 81Br35Cl and 79Br37Cl isotope satellites of M•+ Parallel remarks apply to the [M−H]+ and some other signals.
b
No allowance is made for the contribution to the M•+ signal from the 13C isotope satellite of [M−H]+ or the [M−H]+ signal from the 81Br or 37Cl isotope satellites of M•+
;
either of these contributions may be significant, but do not appreciably alter discussion of the underlying trends.
ꢀ
may be distinguished by focusing attention on the [M−Hal2
]
and
+
of fragmentation by the proximity effect and the analytical value of
this process.
ꢀ
ꢀ
[M−Hal4
]
signals, especially when Hal2 and Hal4ꢀ = Br or Cl.
+
Although loss of a species derived from the ortho substituent
by the proximity effect generally occurs so readily that it pre-
empts most other fragmentations, it is clear that some Y• atoms
or radicals are lost more readily than others. The expected order
of halogen atom loss (Br• > Cl• > F•) is evident in the ease of
elimination of a halogen atom from the ortho position. Thus,
the [M−Br]+:[M−F]+ ratio in the spectrum of 5F2ꢀBrSBO (Fig. 5,
top) is greater than 200. Parallel trends are found when Y = CH3
or CH3O: the [M−CH3]+:[M−Cl]+ ratio is 50 for 5Cl2ꢀCH3SBO;
the corresponding [M−CH3O]+:[M−F]+ ratio exceeds 200 for
5F2ꢀCH3OSBO.
It would seem at first sight that there is little possibility of dis-
tinguishing between isomeric SBOs with a substituent in the meta
and para position. However, as noted previously in the discussion
of [M−H]+ signals, closer inspection of the data in Tables 1 and 2
reveals some secondary trends that may be applied to achieve this
objective, provided that the spectra of both compounds recorded
under identical conditions are available. The [M−Y]+:[M−X]+ratio
is always lower when the substituent Y is in the para position of
the pendant ring. Thus, the ratio is 32 for 5F3ꢀClSBO, but only 10
for 5F4ꢀClSBO. This trend is not pronounced, but it may be applied
with caution in certain circumstances.
4. Conclusion
The 39 substituted styrylbenzoxazoles studied in this work are
readily analysed by mass spectrometry. Molecular mass and molec-
ular formula information is most easily acquired by electrospray
ionization under positive ion conditions, but may also be obtained
by electron ionization, provided that care is taken to recognize
the strong [M−H]+ and [M−Y]+ signals that often appear in the
spectra of styrylbenzoxazoles with a substituent in the pendant
ring. Complementary structural information may be acquired from
the fragmentation of the M•+ ions, which generally occurs with
high selectivity by the elimination of a radical or atom from the
ortho position of the pendant ring, even when apparently more
favourable simple bond cleavages could occur in the benzoxazole
ring. Both the presence and the position of the substituent, Y, in the
pendant ring can be determined by mass spectrometry by careful
consideration of the M•+, [M−H]+ and [M−Y]+ signals.
Acknowledgement
The contribution of Miss Sanam Habib, who did exploratory
work on the preparation of 2-amino-4-fluorophenol by the nitra-
tion of 4-fluorophenol and hydrogenation of the intermediate
4-fluoro-2-nitrophenol, is gratefully acknowledged.
3.2.5. Substituted styrylbenzoxazoles with two halogeno
substituents in the pendant ring
In order to investigate the competition between loss of a halogen
atom from the ortho position of the pendant ring and elimination of
a halogen atom from the meta or para position, two complementary
pairs of SBOs with different halogenosubstituents were prepared. In
addition, one such isomeric pair with a third halogeno substituent
in the benzoxazole ring was synthesized and investigated. Illustra-
tive spectra and data on the ratio of the RI of [M−Y]+:[M−X]+ are
shown in Fig. 8 and Tables 3 and 4, respectively.
In these more complex systems, elimination of a halogen
atom from the ortho position of the pendant ring remains the
dominant process, as illustrated by the [M−Cl]+:[M−Br]+ ratio of
17 for 2ꢀCl4ꢀBrSBO; the corresponding [M−Br]+:[M−Cl]+ ratio of
20 for 2Br4ꢀClSBO is slightly greater, as might have been expected.
Similar effects are seen in the spectra of the analogous trihalogeno
compounds, 5F2ꢀCl4ꢀBrSBO and 5F2ꢀBr4ꢀClSBO (Fig. 8). However,
loss of a fluorine atom competes less effectively with expulsion of a
more massive halogen atom, as illustrated by the [M−F]+:[M−Cl]+
ratio of only 1.5 for 2ꢀF4ꢀClSBO; the corresponding [M−Cl]+:[M−F]+
ratio for the isomeric compound, 2ꢀCl4ꢀFSBO, is far larger (67),
thus emphasizing the enhanced tendency of Cl• to be lost from the
ortho position of the pendant ring. Earlier work has already noted
similar trends in the spectra of other dihalogenobenzazoles [8] in
which [M−F]+ signals were sometimes significantly weaker than
might have been expected. Nevertheless, it is clear that isomeric
SBOs with two different halogeno substituents in the pendant
ring and a third halogeno substituent in the benzoxazole ring
References
[1] F.W. McLafferty, Mass spectrometric analysis broad applicabil-
ity to chemical research, Analytical Chemistry 28 (1956) 306–316,
http://dx.doi.org/10.1021/ac60111a005.
[2] F.W. McLafferty, F. Turecek, Interpretation of Mass Spectra, 4th ed., University
Science Books, Mill Valley, CA, 1993, ISBN 0-935702-25-3, pp. 72–73.
[3] F.W. McLafferty, F. Turecek, Interpretation of Mass Spectra, 4th ed., University
Science Books, Mill Valley, CA, 1993, ISBN 0-935702-25-3, pp. 77–78.
[4] J. Royayne, J.H. Bowie, D.H. Williams, Studies in mass spectrometry. XIX.1
Evidence for the occurrence of aromatic substitution reactions upon elec-
tron impact, Journal of the American Chemical Society 88 (1966) 4980–4984,
http://dx.doi.org/10.1021/ja00973a037.
[5] B. Schaldach, H.-F. Grutzmacher, Formation of 2-methylbenzopyrylium
ions from the molecular ions of benzalacetones, International Jour-
nal of Mass Spectrometry and Ion Processes 31 (1979) 257–270,
http://dx.doi.org/10.1016/0020-7381(79)83027-2.
[6] H.-F. Grutzmacher, Intra- and intermolecular reactions of aromatic
radical cations: an account of mechanistic concepts and methods in
mass spectrometry, Organic Mass Spectrometry 28 (1993) 1375–1387,
http://dx.doi.org/10.1002/oms.1210281205.
[7] R.J. Goldsack, J.S. Shannon, Proximity effects in the electron impact mass spec-
tra of aurones and related compounds, Organic Mass Spectrometry 15 (1980)
545–555, http://dx.doi.org/10.1002/oms.1210151102.
[8] T. Chantler, V.L. Perrin, R.E. Donkor, R.S. Cawthorne, R.D. Bowen, Proximity
effects in the electron impact mass spectra of 2-substituted benza-
zoles, International Journal of Mass Spectrometry 236 (2004) 65–80,
http://dx.doi.org/10.1016/ijms.2004.05.012.
[9] S.-H. Kim, J.-Z. Chu, J.-Y. Park, E.-M. Han, S.M. Park, Dyes and Pigments 59 (2003)
245–250.
[10] K. Ishikawa, Y. Kudo, N. Nishida, T. Suemoto, T. Sawada, T. Iwaki, K. Doh-
ura, Styrylbenzoazole derivatives for imaging of prion plaques and treatment

S.T. Ayrton et al. / International Journal of Mass Spectrometry 345–347 (2013) 120–131
131
of transmissible spongiform encephalopathies, Journal of Neurochemistry 99
(2006) 198–205, http://dx.doi.org/10.1111/j.1471-4159.2006.04035.x.
[17] M. von Wantoch Rekowskia, A. Pyriochoub, N. Papapetropoulosb, A. Stößela, A.
Papapetropoulosb, A. Giannis, Synthesis and biological evaluation of oxadiazole
derivatives as inhibitors of soluble guanylyl cyclise, Bioorganic and Medicinal
Chemistry 18 (2010) 1288–1296, http://dx.doi.org/10.1016/j.bmc.2009.12.027.
[18] C. Boomfield, A.K. Manglik, R.B. Moodie, K. Schofield, G.D. Tobin, Electrophilic
aromatic substitution. Part 28. The mechanism of nitration of some 4-
substituted anisoles and phenols, and of rearrangement of the intermediate 4-
nitro-4-substituted-cyclohexa-2,5-dienones, Journal of the Chemical Society,
Perkin Transactions 2 (1983) 75–82, http://dx.doi.org/10.1039/P29830000075.
[19] N.B. Cech, C.G. Enke, Practical implications of some recent studies in elec-
trospray ionization fundamentals, Mass Spectrometry Reviews 20 (2001)
362–387, http://dx.doi.org/10.1002/mas.10008.
[20] F.A. Cotton, G. Wilkinson, C.A. Murillo, M. Bochmann, Advanced Inorganic
Chemistry, 6th ed., J Wiley and Sons, New York, 1999, ISBN 0-471-19957-5,
p. 223.
[21] K.R. Jennings, The changing impact of the collision induced decomposition of
ions on mass spectrometry, International Journal of Mass Spectrometry 200
(2000) 479–493, http://dx.doi.org/10.1016/S1387-3806(00)00325-0.
[22] I. Howe, D.H. Williams, Calculation and qualitative prediction of mass spec-
tra. Mono- and para-disubstituted benzenes, Journal of the American Chemical
Society 91 (1969) 7137, http://dx.doi.org/10.1021/ja00411053.
[11] I. Caleta, M. Grdisa, D. Mrvos-Sermek, M. Cetina, V. Tralic-Kulenovic, K. Pavelic,
G. Karminski-Zamola, Synthesis, crystal structure and antiproliferative eval-
uation of some new substituted benzothiazoles and styrylbenzothiazoles,
Farmaco 59 (2004) 297–305, http://dx.doi.org/10.1016/j.farmac.2004.01.008.
[12] G. Vitale, P. Corona, M. Lorgia, A. Carta, G. Paglietti, G. Gilberti,
R. Loddo, E. Masrongiu, P. La Colla, Styrylbenzimidazoles. Synthesis
and biological activity
– Part 3, Medicinal Chemistry 6 (2010) 70–78,
http://dx.doi.org/10.2174/157340610791321514.
[13] D.V. Ramana, P. Mahalakshmi, Proximity interactions in 2,2ꢀ-disubstituted
diphenyl ethers under electron impact conditions, Indian Journal of Chemistry.
Section B (1993) 651–657.
[14] M.A. Philips, The preparation of 1-methylbenzoazole, Journal of the Society of
Chemical Industry 56 (1937) 302T, http://dx.doi.org/10.1002/jctb.5000563215.
[15] J.H. Musser, H. Jones, S. Sciortino, K. Bailey, S.M. Coutts, A. Khandwala, P.
Sonnino-Goldman, M. Leibowitz, P. Wolf, E.S. Neiss, Synthesis and antialler-
gic activities of 1,3-oxazolo[4,5-h]quinolines, Journal of Medicinal Chemistry
28 (1985) 1255–1259, http://dx.doi.org/10.1021/jm00147a023.
[16] S. Yoshida, T. Watanabe, Y. Sato, Regulatory molecules for the 5-HT3 recep-
tor ion channel gating system, Bioorganic and Medicinal Chemistry 15 (2007)
3515–3523, http://dx.doi.org/10.1016/j.bmc.2007.02.054.