Fragmentation and skeletal rearrangements of 2-arylylamino-5-aryl-1,3,4- oxadiazoles and their noncovalent complexes with cobalt cation and cyclodextrin studied by mass spectrometry

Article abstract of DOI:10.1002/jms.990

Mass spectrometric fragmentation pathways of title compounds were studied by electron ionization (EI) and electrospray ionization (ESI) as methods of ion generation. To explain the observed complex skeletal rearrangements, tandem mass spectrometry, accurate mass measurement and isotope labeling (compounds containing one ¹³C atom in oxadiazole ring) were used. Loss of CO, N₂ and H atoms under EI conditions led to the formation of 9,10-dihydroacridine-type ions, loss of NH₃ under ESI conditions yielded the 4-phenylphthalazinone-type ions and the loss of HNCO under ESI conditions produced N-arylamino-benzonitrilium ions; however, this process can be affected by the electron-donor/electron-withdrawing properties of groups substituted at the phenyl rings. The ESI was used to study the complexes of the compounds with cobalt as well as with cyclodextrin. It was found that the compounds studied tend to form inclusion complexes with cyclodextrin of stoichiometry 1:1 and complexes of different stoichiometries with cobalt, although those of stoichiometry 6:1 and 4:1 are favored and the attachment of counter ion may stabilize the complexes 3:1 and 2:1. Copyright

Full text of DOI:10.1002/jms.990

JOURNAL OF MASS SPECTROMETRY
J. Mass Spectrom. 2006; 41: 312–322
Published online 18 January 2006 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jms.990
Fragmentation and skeletal rearrangements of
2
-arylylamino-5-aryl-1,3,4-oxadiazoles and their
noncovalent complexes with cobalt cation and
cyclodextrin studied by mass spectrometry
∗
Rafał Fra n´ ski, BłazPej Gierczyk and Grzegorz Schroeder
Adam Mickiewicz University, Faculty of Chemistry, Grunwaldzka 6, 60-780 Pozna n´ , Poland
Received 22 November 2005; Accepted 26 November 2005
Mass spectrometric fragmentation pathways of title compounds were studied by electron ionization (EI)
and electrospray ionization (ESI) as methods of ion generation. To explain the observed complex skeletal
rearrangements, tandem mass spectrometry, accurate mass measurement and isotope labeling (compounds
1
3
containing one C atom in oxadiazole ring) were used. Loss of CO, N
led to the formation of 9,10-dihydroacridine-type ions, loss of NH
-phenylphthalazinone-type ions and the loss of HNCO under ESI conditions produced N-arylamino-
2
and H atoms under EI conditions
under ESI conditions yielded the
3
4
benzonitrilium ions; however, this process can be affected by the electron-donor/electron-withdrawing
properties of groups substituted at the phenyl rings. The ESI was used to study the complexes of the
compounds with cobalt as well as with cyclodextrin. It was found that the compounds studied tend to form
inclusion complexes with cyclodextrin of stoichiometry 1 : 1 and complexes of different stoichiometries
with cobalt, although those of stoichiometry 6 : 1 and 4 : 1 are favored and the attachment of counter ion
may stabilize the complexes 3 : 1 and 2 : 1. Copyright  2006 John Wiley & Sons, Ltd.
KEYWORDS: 1,3,4-oxadiazoles; mass spectrometry; fragmentation pathways; skeletal rearrangements; noncovalent
complexes
INTRODUCTION
Taking into account the biological activities of 2-amino-
-aryl-1,3,4-oxadiazole derivatives, their complexes with
5
Five-member heterocyclics containing oxygen and nitrogen
atoms reveal interesting mass spectrometric fragmentation
biologically active metals like copper and cobalt as well
as with cyclodextrin have been studied. It is worth adding
that 1,3,4-oxadiazoles have been used to obtain interesting
1
pathways. Among this class of compounds, 2-amino-
5
-aryl-1,3,4-oxadiazole derivatives are of special interest
15
bimetallic complexes with cobalt. The complexes of the
2
–10
because they show different biological activities.
Their
compounds studied with cobalt as well as with cyclodextrin
have also been studied in this work by ESI, since it is well
electron ionization- (EI-)induced decomposition has been
1
1–13
discussed.
The EI mass spectrum of 2-arylamino-5-
16
known that this MS technique is the ‘softest’ and the most
1
4
aryl-1,3,4-oxadiazole has already been published. Here,
we report the fragmentation pathways of the title com-
pounds (1–9, Scheme 1) generated by EI and electrospray
ionization (ESI). As stated further in the text, a number of
unusual skeletal rearrangements, important from the point
of view of gas-phase ion chemistry, have been observed.
To analyze the above processes, the isotope labeled com-
pound 1a, containing C at the 5-position of oxadiazole
ring, has been synthesized. Compounds 2–9, the four pairs
of isomers, have been synthesized to study the effect of
the electron-donor/electron-withdrawing properties of the
group substituted at the phenyl rings on the mass spectro-
metric decompositions.
1
7,18
suitable to study the supramolecular systems.
EXPERIMENTAL
The compounds studied were synthesized as follows:
To a solution of corresponding hydrazide (10 mmol) in
anhydrous methanol (25 ml), aryl isothiocyanate (10 mmol)
was added and the mixture was refluxed for 2 h. On cooling,
the solid obtained was filtered and recrystalized from
methanol to obtain pure thiosemicarbazide. The solution of
thiosemicarbazide (5 mmol) in methanol (10 ml) was basified
with 2 M sodium hydroxide to a pH of 9.0, and a solution
of iodine in potassium iodide was added by stirring and
cooling until the color of iodine persisted. The precipitated
product was filtered, dried and recrystalized from ethanol.
The general scheme of their synthesis is shown in Scheme 2.
1
3
Ł
Correspondence to: Rafał Fra n´ ski, Adam Mickiewicz University,
Faculty of Chemistry, Grunwaldzka 6, 60-780 Pozna n´ , Poland.
E-mail: franski@main.amu.edu.pl
Copyright  2006 John Wiley & Sons, Ltd.

Mass spectrometric study of 1,3,4-oxadiazoles
313
1
O
O
C
5
C
2
C
R1
NH
NH2
+
S
S
R2
R1
NH
R2
C
N
4
"
1"
1'
4'
N
3
"
2"
N
2'
3'
4
3
Comp.
R1
R2
O
C
1
H
H
H
R1
NH NH C NH
R2
2
3
OCH3
H
I2/NaOH
H3CO
H S
2
4
5
H
N(CH3)2
H
O
R1
NH
R2
(H3C)2N
N
N
6
7
H
Cl
H
Scheme 2. General scheme of the synthesis of compounds
Cl
studied.
8
9
H
NO2
H
O2N
6 5
H
13COOH (Aldrich), as
by using labeled benzoic acid C
shown in Scheme 3.
The NMR spectrum of the isotope labeled 1a showed the
1
2
following signals: H NMR ([ H]
6
0
-DMSO, υ): 10.68 ppm (s,
O
C
13C
0
NH
1H, NH); 7.91 ppm (m, 2H, H-2 ); 7.63 ppm (d, 2H, H-2 );
.58 ppm (m, 3H, H-3 , 4 ); 7.37 ppm (t, 2H, H-3 ); 7.03 ppm
t, 1H, H-4 ) and C NMR ([ H]
C-2); 157.7 ppm (C-5); 138.7 ppm (C-1 ); 130.6 ppm (C-4 );
129.1 ppm (C-3 ); 128.8 ppm (C-3 ); 125.2 ppm (d, C-2 );
23.6 ppm (d, C-1 ); 121.6 ppm (C-4 ); 116.8 ppm (C-2 ).
9,10-Dihydroacridine was obtained by the reduction of
0
0
00
0
7
(
N
N
0
13
2
6
-DMSO, υ): 160.2 ppm
1a
0
00
00
(
0
0
0
Scheme 1. Compounds studied.
0
0
0
0
1
A detailed description of their synthesis will be published
later. Their structures were confirmed by NMR spectroscopy.
The isotope labeled 1a has been prepared in a different way,
acridine (Aldrich) according to the procedure described
1
9
elsewhere.
F
F
O
13
+
C
OH
HO
F
H N NH2
+
S
C
N
2
F
F
F
F
F
O
S
13
C
O
F
+
H2N NH
C
HN
F
HOC6F5
O
S
C
1
3
C HN NH
HN
I2/NaOH
H S
2
O
C
13
C
NH
N
N
Scheme 3. Scheme of the synthesis of isotope labeled compound 1a.
Copyright  2006 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2006; 41: 312–322

314
R. Fra n´ ski, B. Gierczyk and G. Schroeder
The EI mass spectra were recorded on an AMD-402
two-sector mass spectrometer (AMD Intectra, Germany) of
Table 1. Summary of the MS/MS results obtained for 1-cobalt
complexes
B/E geometry with the acceleration voltage of 8 kV, electron
energy 70 eV and an ion source temperature of 200 °C. The
Selected parent
ion m/z
Detected product
ions m/z
(% relative abundance)
Collision
energy
compounds studied were introduced by a direct insertion
probe. The B/E-linked scan mass spectra of metastable ions
and exact mass measurements (high-resolution data) for the
ions of interest were recorded on the same instrument. High-
resolution data were obtained by using V/E high-resolution
scan in relation to perfluorokerosene, with an error less than
(
% relative abundance)
61 C Co]2C 740.5 (25%) [41 C Co–H] 1006 (45%)
C
[
10
2C
[
[
[
[
51 C Co] 622 (45%)
2C
41 C Co] 503.5 (100%)
C
21 C H] 475 (15%)
1
0 ppm for all ions discussed.
C
1 C H] 238 (35%)
The MS/MS spectra of [M C H]C and [M C H–HNCO]
C
2
C
C
[
51 C Co] 622 (25%)
[41 C Co–H] 1006 (5%)
9
2
C
ions were carried out on a Qstar mass spectrometer
[41 C Co] 503.5 (100%)
C
(Applied Biosystems, Darmstadt, Germany, hybrid QqTOF
[1 C H] 238 (7%)
2
C
C
C
instrument). In order to generate the abundant [M C
[41 C Co] 503.5 (15%) [31 C Co–H] 769 (30%)
15
C
H–HNCO] ions, the declustering potential in the ESI
[21 C Co–H] 532 (20%)
2
C
source was increased to 100 V (its standard value is 50 V).
The compounds studied were dissolved in methanol at a
[31 C Co] 385 (5%)
C
[1 C H] 238 (100%)
ꢀ
5
[31 C Co]2C 385 (40%)
C
concentration of about 10 M and the solutions were directly
[21 C Co–H] 532 (75%)
13
ꢀ
1
C
infused into the ESI source at a flow rate of 5 µl min . The
MS/MS experiments were performed with collision energies
of 25 V. This value allowed the achievement of reasonable
abundances of both parent and daughter ions for each of
the compounds studied. The collision gas used was nitrogen
with a purity of >99.5%. The pressure of the collision gas was
[1 C H] 238 (100%)
ꢀ
5
set to 4, which resulted in a pressure of about 3.3 ð 10 torr
within the collision cell, and this value was maintained in all
fragmentation experiments.
C
In order to determine the structure of [M C H–NH
3
]
n
ions, the MS experiments were carried out on a Fourier
transform ion cyclotron resonance (FT-ICR) mass spectrom-
eter (Finnigan LTQ FTMS, Thermo Electron Co.) At first the
spectra were taken in the ion-trap mode (low-resolution data)
and then, in order to determine the elemental composition of
n
the product ions formed, the MS experiments were repeated
with the use of FT-ICR. The resolution was 100 000. The colli-
C
C
sion energy was 45 V for [1 C H] ion, 35 V for [1 C H–NH
3
]
C
ion and 30 V for [1 C H–NH
3
–CO] ion. The noncovalent
complexes with the cobalt cation and cyclodextrin were also
studied by using the above instrument. At first, the full-scan
mass spectra were recorded in the ion trap mode, and then
the MS/MS spectra were taken in both the ion trap mode
and with the use of FT-ICR. The collision energies used for
2
C
[
n1 C Co] ions are shown in Table 1 and the collision ener-
C
ꢀ
ꢀ
gies for [1 C ˇ-CD C Na] , [1 C ˇ-CD ꢀ H] and [1 ꢀ H]
ions were 17, 15 and 25 V, respectively. The compounds
studied were dissolved in methanol at a concentration of
ꢀ
5
about 10 M and the solutions were directly infused into the
ESI source at a flow rate of 2 µl min
ꢀ
1
.
Figure 1. EI mass spectra of 1 and 1a.
of C H N composition was 180.08133, thus the error was
RESULTS AND DISCUSSION
EI-induced fragmentation pathways
1
3
10
The EI mass spectra of all compounds studied showed
the peak corresponding to the loss of mass 57 from
molecular ions. The accurate mass measurement revealed
that this fragmentation pathway involved the loss of CO,
only ꢀ1.7 ppm. A similar process has been already observed
for 2,5-diphenyl-1,3,4-oxadiazoles leading to the formation
2
0
of fluorine-type ions. The fact that no intermediates were
C
observed (e.g. [1 ꢀ 28] ion) may be due to a slow first step,
N
2
and H atoms. For example, the obtained mass of
which is followed by a sequence of faster consecutive steps.
[
1 ꢀ 57]C ion was 180.08163, whereas the theoretical mass
A comparison of the EI mass spectra of 1 and 1a (Fig. 1)
Copyright  2006 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2006; 41: 312–322

Mass spectrometric study of 1,3,4-oxadiazoles
315
clearly indicates that the eliminated CO contains the C-2
carbon atom.
ion of 9,10-dihydroacridine (Fig. 2). The B/E mass spectrum
C
C
of [1 ꢀ 57] ion is not identical to that of [M ꢀ 1] ion
originating 9,10-dihydroacridine (Fig. 2). Because of the
different internal energies of parent ions, there are minor
differences in the relative ion abundances.
Theoretically, a number of structures can be proposed for
C
C
[
M ꢀ 57] ion, e.g. [M ꢀ H] ion derived from benzophenone
imine. However, the B/E mass spectra of metastable ions
C
recorded for [M ꢀ 57] ion have shown the abundant
Although it is difficult to propose a detailed mechanism
for the loss of mass 57 from the compounds studied, it seems
product ion formed by the loss of mass 28 (e.g. at m/z
C
1
52 for 1, Fig. 2). An analogous fragmentation pathway has
to be reasonable that the dominant structure of [M ꢀ 57] ion
C
21
C
been observed for [M ꢀ H] ion of 9,10-dihydroacridine.
is that of [M ꢀ 1] ion originating from 9,10-dihydroacridine
C
Therefore, the B/E mass spectrum was taken for [M ꢀ H]
as shown in Scheme 4.
Acheson et al. have found that 9,10-dihydroacridine
derivatives show preferences for expulsion of the sub-
a
21
stituents from position 9. Therefore, the structure of
C
[
1a ꢀ H] ion shown in Scheme 4 can be regarded as 9,10-
dihydroacridine deprived of one hydrogen atom from posi-
tion 9. A number of resonance structures can be proposed
for this ion, two of which are shown in Scheme 4.
The molecular ions of the compounds studied lose
ž
the NH–C
6
H
4
–R
2
radical. This is a rather trivial process
occurring without further skeletal rearrangement. As shown
by the B/E-linked scan mass spectra, the decomposition of
the formed fragment ion (e.g. at m/z 145 for 1) involves the
ž
loss of mass 28 (CO molecule), mass 42 ( NCO radical), mass
C
5
6 (CO and N
product ion (e.g. at m/z 77 for 1).
The loss of OCN–C –R molecule is more interesting.
2 1 6 4
molecules) and formation of [R –C H ]
6
H
4
1
As shown by the B/E linked scan mass spectra, the
b
c
Figure 2. EI mass spectrum of 9,10-dihydroacridine (a), B/E
mass spectrum of [M ꢀ H]C ion of 9,10-dihydroacridine (b), B/E
mass spectrum of [M ꢀ 57]C ion derived from 1 (c).
Figure 3. EI mass spectra of 2 and 3.
J. Mass Spectrom. 2006; 41: 312–322
Copyright  2006 John Wiley & Sons, Ltd.

316
R. Fra n´ ski, B. Gierczyk and G. Schroeder
NH
1
3C
+
+ H
O
1
3C
C
N
NH
-
CO -N2 -H
N
1
NH
0
4
5
8
+
3
2
6
7
1
3C
9
1
H
Scheme 4. Loss of CO, N2 and H atom (loss of mass 57) from 1a molecular ion.
decomposition of the fragment ion formed (e.g. at m/z
18 for 1) involves consecutive losses of HCN molecules
as well as the loss of two HCN molecules and H atom
product ions at m/z 91, 64 and 63 for 1). These fragmentation
pathways resemble those described for benzimidazole and
The phenyl ring substitution strongly affects the EI-
1
induced fragmentation pathways. For example, the molec-
C
ular ion of 2 yields abundant [HN–C
6
H
4
–OCH
3
]
product
(
ion at m/z 122 and the dominant fragmentation pathway
žC
of 3 is the formation of [H
3
CO–C
6
H
4
–CN] product ion at
2
2,23
C
indazole very well.
However, the decompositions of these
m/z 133 (Fig. 3). For both 2 and 3, the [M ꢀ 57] ion at m/z
compounds are very similar; therefore, it is not possible
to determine the structure (benzimidazole or indazole) of
210 is followed by the product ion at m/z 180. The latter is
2
formed by elimination of the H CO molecule from the for-
the product ions formed by the loss of the OCN–C
molecule.
6
H
4
–R
1
mer and/or by elimination of CO, N
from molecular ions.
2
and methoxy radical
Figure 4. MS/MS spectra of [1 C H] , [1a C H] , [6 C H] and [7 C H]C ions.
C
C
C
Copyright  2006 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2006; 41: 312–322

Mass spectrometric study of 1,3,4-oxadiazoles
317
Figure 4. (Continued).
+
Fragmentation pathways of [M + H] ions
been observed (e.g. at m/z 111 for 7). On the grounds of the
C
C
C
C
The MS/MS spectra of [1 C H] , [1a C H] , [6 C H] and
above data, a plausible mechanism of the [1 C H–NH
3
] ion
C
[
7 C H] ions are shown in Fig. 4 as representative examples.
formation and its further fragmentation is proposed, and is
shown in Scheme 5.
There are two peaks of product ions formed as a result
of the complex skeletal rearrangements. They correspond
Protonation of amine nitrogen atom can induce loss of
the aniline molecule, and hence the formation of m/z 145
fragment ion (Fig. 3). On the other hand, formation of an
ion–molecule complex, accompanied by the breaking of the
C–O bond of oxadiazole ring, seems to be an alternative,
reasonable pathway (Scheme 5). It is not possible to explain
to the loss of mass 17 (NH
4
3
loss) and to the loss of mass
3 (HNCO loss). The MS experiments (performed on FT-
n
C
ICR mass spectrometer) for [1 C H] ion show that the
C
main decomposition pathway of [1 C H–NH
3
]
involves
2
the loss of mass 28 (N or CO elimination) leading to
the fragment ion at m/z 193. The obtained exact mass
of this fragment ion was 193.0757, indicating the CO
loss (the theoretical mass is 193.0766, error 4.5 ppm) and
in detail the process of ammonia loss (aryne mechanism or
C
electrophile attack); however, the structure of [1 C H–NH
3
]
C
ion can be regarded as [M ꢀ H] ion derived from 4-
C
decomposition of [1 C H–NH
3
–CO] results in the HCN
phenylphthalazinone. In spite of the fact that the loss of
2
4
C
molecule loss yielding the product ion at m/z 166. The
data obtained for the isotope-labeled compound 1a show
that the eliminated CO molecule contains the C-2 carbon
atom and the consecutively eliminated HCN contains the
C-5 carbon atom. The next efficient fragmentation pathway
involves the formation of [C
at m/z 77. This ion has also been observed for 2, 4, 6 and
, but for 3, 5, 7 and 9, the respective [R
hydrogen is not a favored process, the [M ꢀ H] ion is very
abundant in the EI mass spectrum of 4-phenylphthalazinone,
indicating its high stability. The product ions at m/z 193, 166
2
5
C
C
C
and 77 ([M ꢀ H–CO] , [M ꢀ H–CO–HCN] and [C
6
H ]
5
ion, respectively) are also formed with high efficiency from
C
C
25
of ion [1 C H–NH
3
]
6
H
5
]
ion
4-phenylphthalazinone. Therefore, the proposed structure
C
of [1 C H–NH
3
]
ion is reasonable. It seems to be very
C
8
1
–C
6
H
4
]
ion has
likely that after CO loss the product ion formed (m/z
Copyright  2006 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2006; 41: 312–322

318
R. Fra n´ ski, B. Gierczyk and G. Schroeder
O
+
+
NH
+
H N
2
N
N
N
N
H
-
NH3
m/z 238
C
N
N
C
O
-
+
O
m/z 221
N
ion-molecule complex
CO
+
HCN
N
N
+
[
C H ]
+
6 5
m/z 77
N
m/z 193
m/z 166
Scheme 5. Plausible mechanism of [1 C H–NH3]C ion formation and its further decomposition.
1
93) isomerizes to produce the structure corresponding to
of structure a, the loss of the benzonitrile molecule would
C
C
the [M ꢀ H] ion derived from 1-phenyl-benzimidazole (as
proceed, yielding the respective [HN–C
6
H
4
–R
2
]
product
shown in Scheme 5), because in the EI mass spectrum of
ions (e.g. at m/z 92 for 1, m/z 122 for 2), by analogy
to the fragmentation pathway of N-arylbenzonitrilium
C
this compound the product ion at m/z 166 ([M ꢀ H–HCN]
2
6
32,33
C
ion) is the most abundant. The product ion at m/z 166 is
ions.
The abundant [HN–C
6
H
4
–R
2
C
]
product ion has
not formed from 2-phenylbenzimidazole or from 3-phenyl-
indazole.
been observed for all [M ꢀ HNCO] ions except those
2
7,28
originating from 5 and 8. Figure 5 presents the MS/MS
C
C
C
The next abundant product ion derived from [M C H]
spectra of [5 ꢀ HNCO] and [8 ꢀ HNCO] ions as well
C
ions is that formed by the loss of mass 43. This loss involves
a complex skeletal rearrangement related to the loss of the
HNCO molecular loss. A similar process has already been
discussed in detail for 2,5-diaryl-1,3,4-oxadiazoles.29–32 As
seen from Fig. 3, the eliminated HNCO molecule contains the
as [M ꢀ HNCO] ions derived from respective isomers
C
C
([4 ꢀ HNCO] and [9 ꢀ HNCO] ions).
C
The lack of [HN–C
6
H
4
C
–R
2
]
product ions for [5 ꢀ
C
HNCO] and [8 ꢀ HNCO] ions can be explained on
the grounds of the electron-donor/electron-withdrawing
C
C-2 carbon atom. For [1a C H] , loss of mass 43 is observed
properties of R
properties of ꢀCH
1
/R
2
substituents. The strong electron-donor
N– group ensure that protonation of
C
(
analogously as for [1 C H] ), thus the eliminated HNCO
3
ꢁ
2
molecule does not contain C-5 atom. The presence of the
NH- group at the C-2 carbon atom favors the protonation
N-4 nitrogen atom is favored over that of N-3 atom for 5
since the stable quinoid resonance structure can be formed,
as shown in Scheme 7.
—
of the N-3 nitrogen atom because of a possible resonance
C
structure, and the [1a C H–HNCO] product ion should
As a consequence, the dominant structure of the
C
rather have the structure of a (N-arylamino-benzonitrilium
ions) than of b (C-arylamino-phenylisonitrilium ions), as
shown in Scheme 6.
[5 ꢀ HNCO] product ion formed is b (Scheme 6). By
analogy with the mass spectrometric decomposition of N-
arylbenzonitrilium ions, the loss of phenylisonitrile leading
C
In order to confirm that the structure a is the dominant
form of [M ꢀ HNCO]C ions, the MS/MS experiments were
performed for them. It was strongly expected that for ions
to the formation of [HN–C
6
H
5
]
ion at m/z 92 cannot
32,33
be expected.
The dominant fragmentation pathway of
[5 ꢀ HNCO] product ion involves the loss of CH
C
4
molecule.
O
+
13
C
13
C
NH
C
N
NH
N
N+
H
structure a - favored
HNCO
O
+
13
1
3
+
N
C
NH
C
C
N
H
NH
N
structure b
Scheme 6. Loss of HNCO molecule from [1 C H]C ion.
Copyright  2006 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2006; 41: 312–322

Mass spectrometric study of 1,3,4-oxadiazoles
319
Figure 5. MS/MS spectra of [4 ꢀ HNCO] , [5 ꢀ HNCO] , [8 ꢀ HNCO] and [9 ꢀ HNCO]C ion.
C
C
C
probably, other decomposition directions are much more
favored.
O
(
CH ) N
C
+N
H
C
N
NH
3
2
Noncovalent complexes
The complexes of 2,5-diaryl- and 2,5-dipyridyl-1,3,4-oxadia-
zoles with copper were discussed in previous papers,
reporting the formation of complexes of the stoichiometry
O
+
(
CH ) N
C
N
H
C
N
NH
3
2
3
4–36
1
: 1, 2 : 1 and 2 : 2.
Similar results were obtained for
2
C
other metal cations, e.g. Co . For 2-arylamino-5-aryl-1,3,4-
oxadiazoles, the formation of copper cation complexes of
the stoichiometry 2 : 1 (and 3 : 1 of lower abundances) was
observed under ESI conditions. On the other hand, the results
obtained for cobalt cation complexes with the compounds
studied are much more interesting as shown in Fig. 6.
The results presented in Fig. 6 clearly indicate that
the complexes of the stoichiometry 6 : 1 and 4 : 1, between
Co and neutral molecule of 1, are favored over those of
stoichiometry 5 : 1 and 3 : 1, while those of the stoichiometry
2 : 1 are not favored. This conclusion can also be drawn
on the grounds of the MS/MS experiments summarized in
C
Scheme 7. Stabilization of [5 C H] ions, formed after protona-
tion of N-4, by quinoid structure.
Analogously, it can be easily deduced that because of
the strong electron-withdrawing properties of the nitro
group, for 8 the protonation of N-4 nitrogen atom is also
2
C
favored over that of N-3 atom. As a consequence, the
C
[
8 ꢀ HNCO] product ion formed has the structure b rather
C
than a, thus the [HN ꢀ C
6
H
4
–NO
2
]
ion at m/z 137 has
2
C
not been formed. The dominant fragmentation pathway
Table 1. The loss of two ligand molecules from [61 C Co]
C
of [8 ꢀ HNCO] ion involves the loss of NO
2
molecule.
ion proceeds with higher efficiency than the loss of one
ligand molecule. We also deal with the so-called charge
C
The peak of [C
6
H
5
–NCH] ion (protonated phenylisoni-
C
trile) at m/z 104 is clearly seen (Fig. 4). Protonation of
benzonitrile is not favored, since phenylisonitrile is a
stronger Lewis base than benzonitrile. Thus the forma-
–CNH] ion cannot be expected. It is difficult
to rationalize why the respective [ꢀH
ion at m/z 147 was not formed from [5 ꢀ HNCO] ion; most
separation reaction, namely the formation of [41 C Co–H]
3
0
C
C
and [21 C H] product ions. Because the [51 C Co–H] has
3
3
C
not been detected, the [1 C H] product ion is formed by
C
C
tion of [C
6
H
5
subsequent decomposition of [21 C H] ions. The dominant
C
2C
3
Cꢁ
2
N–C
6
H
4
–NCH]
process observed for [51 C Co] ion is the loss of one
C
2C
ligand molecule. On the other hand, for [41 C Co] ion,
Copyright  2006 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2006; 41: 312–322

320
R. Fra n´ ski, B. Gierczyk and G. Schroeder
Figure 6. ESI mass spectrum obtained for methanol solution containing 1 and CoCl2.
the loss of one ligand molecule is a minor process. Its
main fragmentation pathway involves a charge separation
often observed in the ESI mass spectra), they were subjected
to MS/MS experiments. If they were inclusion complexes, the
C
38
reaction, consisting of the formation of [31 C Co–H]
decomposition of the guest molecule should be observed.
C
C
and [1 C H] product ions. Because [21 C H] has not
As shown in Fig. 7(a,b), besides the loss of intact 1 molecule,
C
C
been detected, the [21 C Co–H] product ion is formed
the loss of mass 119 is observed for both [1 C ˇ-CD C Na]
C
by subsequent decomposition of [31 C Co–H] ions. For
ꢀ
and [1 C ˇ-CD ꢀ H] ions. The loss of mass 119 is the
2
C
[
31 C Co] ion, the charge separation reaction is the only
ꢀ
main fragmentation pathway that occurred for [1 ꢀ H]
observed process indicating that the formation of the
: 1 complexes between the cobalt cation and the neutral
molecule of 1 is completely unlikely. It is also worth adding
ions, producing fragment ion at m/z 117, as shown in
2
Fig. 7(c).
ꢀ
This decomposition for [1 C ˇ-CD ꢀ H] ions is expected,
C
C
that the presence of [21 C CoCl] ion and [31 C CoCl] ion
indicating that the deprotonated molecule of 1 is inside the
cavity of intact ˇ-CD molecule. On the other hand, the
2
C
(
the latter having twice the abundance of [31 C Co] ion,
Fig. 6) indicates that the attachment of the counter ion,
C
loss of mass 119 from [1 C ˇ-CD C Na] ion is unexpected
ꢀ
namely Cl , stabilizes the complexes of stoichiometry 3 : 1
and indicates that the deprotonated molecule of 1 is inside
the cavity of protonated and sodiated ˇ-CD molecule.
and 2 : 1.
The molecules containing aliphatic chain fit well into
the ˛-cyclodextrin (˛-CD) cavity, while the molecules
containing phenyl groups fit better into the larger cavity
C
The elemental compositions of [1 C ˇ-CD C Na] and [1 C
ꢀ
ˇ-CD ꢀ H] ions and the product ions derived from them
3
7
have been confirmed by their exact masses obtained in the
FT-ICR mode. For example, the obtained mass of the product
of ˇ-cyclodextrin (ˇ-CD). The noncovalent complexes
between ˇ-CD and aromatic molecules present in the
C
ion formed from [1 C ˇ-CD C Na] is 1275.4162, while its
solution can be successfully detected by ESI, especially if
the guest molecule has at least one polar group.37 Thus if the
theoretical mass is 1275.4126, so the error is ꢀ2.8 ppm. It also
has to be stressed that the ESI mass spectrum recorded
for a solution containing 1 and sucrose does not show
compounds studied in this work are prone to form inclusion
complexes with ˇ-CD, they will be detected under the ESI
conditions. The ESI mass spectra recorded for a solution
containing ˇ-CD and 1 show intense peaks corresponding to
ꢀ
the [1 C M ꢀ H] ion (M stands for a neutral molecule of
C
sucrose). Only in the positive ion mode the [1 C M C Na]
C
ꢀ
the [1 C ˇ-CD C Na] (m/z 1394) and [1 C ˇ-CD ꢀ H] ion
ion was observed; however, the loss of mass 119 was not
detected in its MS/MS mass spectrum. The only processes
(
m/z 1370), in positive and negative ion mode respectively.
C
In order to verify whether the observed ions were really the
inclusion complexes (not nonspecific electrostatic adducts
observed were its dissociation that yielded [M C Na] ion
and sucrose rings decomposition.
Copyright  2006 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2006; 41: 312–322

Mass spectrometric study of 1,3,4-oxadiazoles
321
a
b
c
Figure 7. MS/MS spectra of (a) [1 C ˇ-CD C Na] , (b) [1 C ˇ-CD ꢀ H] and (c) [1 ꢀ H]ꢀ ion obtained in ion trap mode.
C
ꢀ
CONCLUSIONS
REFERENCES
2
-Arylamino-5-aryl-1,3,4-oxadiazoles under mass spectro-
1. Fra n´ ski R, Eitner K, Schroeder G, Szwajka OP. The abundances
of fragment ions formed via skeletal rearrangements from 2,5-
disubstituted-1,3,4-oxadiazoles and their theoretical calculated
metric conditions show a number of complex skeletal
rearrangements. Under the EI conditions, they lose CO,
N
eliminated CO molecule contains the C-2 atom of the oxa-
diazole ring. The protonated molecule of the compounds
3
studied (ESI conditions) lose NH , leading to the formation
stabilities. Eur. J. Mass Spectrom. 2004; 10: 495.
2
and H atoms giving 9,10-dihydroacridine-type ions. The
˜
2
. Ates O, Kocabalkanli A, Cesur N, Ot u¨ k G. Synthesis and
antimicrobial activity of some 5-aryl-2-[(N,N-disubstituted
thiocarbamoylthio)acylamino]-1,3,4-oxadiazoles. Il Farmaco
1
998; 53: 541.
˜
3
. Kocabalkanli A, Ates O, Ot u¨ k G. Synthesis and evaluation of
of phenylphthalazinone-type ions. These product ions lose
the CO molecule containing the C-2 atom as well. The sec-
ond complex rearrangement of [M C H]C ions involves the
loss of the HNCO molecule. The eliminated HNCO molecule
also contains the C-2 atom and the product ions formed
can be regarded as N-arylylamino-benzonitrilium ions. It
was found that the structure of the product ions formed
can be affected by the electron-donor/electron-withdrawing
properties of the groups substituted at the phenyl rings. The
compounds studied are very prone to form complexes with
cyclodextrin (inclusion complexes) and with cobalt. In the
latter case, the complexes of stoichiometry 6 : 1 and 4 : 1 are
favored, and the attachment of the counter ion may stabilize
the complexes 3 : 1 and 2 : 1.
antibacterial activity of some 2-[[˛-(4-substituted benzoyloxy)-
˛
-phenylacetyl or methylacetyl]amino]-5-(4-methoxyphenyl)-
1,3,4-oxadiazoles. Il Farmaco 2001; 56: 975.
4. Zou X-J, Lai L-H, Jin G-Y, Zhang Z-X. Synthesis, fungicidal
activity, and 3D-QSAR of pyridazinone-substituted 1,3,4-
oxadiazoles and 1,3,4-thiadiazoles. J. Agric. Food Chem. 2002;
50: 3757.
5
. Palaska E, Sahin G, Kelicen P, Turlu NT, Altinok G. Synthesis
and anti-inflammatory activity of 1-acylthiosemicarbazides,
1,3,4-oxadiazoles, 1,3,4-thiadiazoles and 1,2,4-triazole-3-thiones.
Il Farmaco 2002; 57: 101.
. Sahin G, Palaska E, Kelicen P, Demirdamar R, Altinok G.
Synthesis of some new 1-acylthiosemicarbazides, 1,3,4-
oxadiazoles, 1,3,4-thiadiazoles and 1,2,4-triazole-3-thiones and
their anti-inflammatory activities. Arzneim.-Forsch./Drug Res.
2001; 51: 478.
6
Copyright  2006 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2006; 41: 312–322

322
R. Fra n´ ski, B. Gierczyk and G. Schroeder
7
8
9
. Lokanatha Rai KM, Linganna N. Synthesis and evaluation
of antimitotic activity of alkylated 2-amino-1,3,4-oxadiazole
derivatives. Il Farmaco 2000; 55: 389.
. Omar FA, Mahfouz NM, Rahman MA. Design, synthesis and
antiinflammatory activity of some 1,3,4-oxadiazole derivatives.
Eur. J. Med. Chem. 1996; 31: 819.
. Sahin G, Palaska E, Ekizoglu M, Ozalp M. Synthesis and
antimicrobial activity of some 1,3,4-oxadiazole derivatives. Il
Farmaco 2002; 57: 539.
ˇ
2
4. McLafferty FW, TureSek F. Interpretation of Mass Spectra, 4th ed.
University Science Books: Mill Valley, CA, 1993.
2
5. Bowie JH, Cooks RG, Donaghue PF, Halleday JA, Rodda HJ.
Electron impact studies. XVIII. Mass spectra of pyridazines,
phthalazines and related compounds. Aust. J. Chem. 1967; 20:
2677.
2
6. Efremov YuA, Fedyainov NV, Filatov VP, Azarova EG. Study of
substituted benzimidazoles during electron impact. Khim. Get.
Soedin. 1984; 6: 819.
7. Ramana DV, Kantharaj E. Mass spectrometer as a probe in the
synthesis of 2-substituted benzimidazoles. Tetrahedron 1994; 50:
1
0. Fra n´ ski R. Biological activities of the compounds bearing 1,3,4-
oxa(thia)diazole ring. A. J. Chem. 2005; 17: 2063.
1. Alekseeva VYa, Boikov YuA, Viktorovskii IV, V’yuonov KA.
Conversion of 2-amino-5-aryl-1,3,4-oxadiazoles into 5-alkoxy-
2
1
2485.
2
8. Taylor EC, Sobieray DM. Bicyclodiazepinones from 3-oxo-1,2-
diazetidinium hydroxide, inner salts. Tetrahedron 1991; 47: 9599.
9. Fra n´ ski R, Schroeder G, Rybachenko V, Szwajka OP. Loss of
isocyanic acid from internal oxadiazole ring of protonated
molecules of some 2,5-diaryl-1,3,4-oxadiazoles. Rapid Commun.
Mass Spectrom. 2002; 16: 390.
3-aryl-1,2,4-triazoles. Khim. Get. Soedin. 1986; 11: 1553.
1
1
1
1
2. Alekseeva VYa, Vergizov SN, Dymshits VA, Viktorovskii IV,
V’yuonov KA. Dual reactivity of 2-amino-5-aryl-1,3,4-
oxadiazoles. Zh. Org. Khim. 1989; 25: 2216.
3. Kraft R, Henklein P, Etzold G, Z o¨ pfl HJ, Westphal G, Schmidt R.
Application of mass spectrometry to the identification of
aminooxadiazole isomers. Org. Mass Spectrom. 1976; 11: 304.
2
30. Fra n´ ski R,
Eitner K,
Schroeder G,
Szwajka OP.
Mass
spectrometric study of some protonated and lithiated 2,5-
disubstituted-1,3,4-oxadiazoles. J. Am. Soc. Mass Spectrom. 2003;
4. Froeyen P.
A synthesis of 1,3,4-oxadiazoles via the aza-
Wittig reaction of N-acylamino iminotriphenylphosphoranes.
Phosphorus, Sulfur Silicon Relat. Elem. 1991; 57: 11.
14: 289.
3
1. Fra n´ ski R, Gierczyk B, Schroeder G. Mass spectrometric
fragmentation pathways of isotope labeled 2,5-disubstituted-
5. Incarvito C, Rheingold AL, Gavrilova AL, Qin JC, Bosnich B.
Bimetallic reactivity. One-site addition two-metal oxidation
reactions using a di-Co(II) complexes of a binucleating ligand
with 5- and 6-coordinate sites. Inorg. Chem. 2001; 40: 4101.
6. Cole RB. Some tenetspertaining top electrospray ionizationmass
spectrometry. J. Mass Spectrom. 2000; 35: 763.
7. Schalley CA. Supramolecular chemistry goes gas phase: the mass
spectrometric examination of noncovalent interactions in host-
guest chemistry and molecular recognition. Int. J. Mass Spectrom.
1
,3,4-oxadiazoles and thiadiazoles. Int. J. Mass Spectrom. 2004;
231: 47.
3
3
3
3
3
2. Fra n´ ski R, Zalas M, Beck S, Springer A, Linscheid M. Mass
spectrometric decomposition of N-arylbenzonitrilium ions. Int.
J. Mass Spectrom. 2005; 242: 1.
3. Riter LS, Fraley DF, Cooks RG. Denitration of nitroaromatic
compounds by arylnitrile radical cations. J. Am. Soc. Mass
Spectrom. 2000; 11: 33.
1
1
2000; 194: 11.
4. Fra n´ ski R. Electrospray ionization mass spectrometric study of
1
8. Schalley CA. Molecular recognition and supramolecular
chemistry in the gas phase. Mass Spectrom. Rev. 2001; 20: 253.
9. Yudin LG, Kost AN, Berlin YuA, Shipov AE. Reduction with
formic acid and its derivatives. VII. Reduction of compounds
containing the pyridine nucleus. Zh. Obshch. Khim. 1957; 27: 3021.
0. Balen’kii LI, Luiksaar SI, Poddubnyi IS, Krayushkin MM. Novel
syntheses of symmetrical 2,5-diaryl-1,3,4-oxadiazoles and 1,4-
phenylenebis-1,3,4-oxadiazoles. Izv. Akad. Nauk SSSR, Ser. Khim.
1
,3,4-oxadiazole-copper complexes. J. Mass Spectrom. 2004; 39:
272.
1
5. Fra n´ ski R. Electrospray ionization mass spectrometric study
of 1,3,4-thiadiazole-copper complexes – comparison with 1,3,4-
oxadiazole derivatives. J. Mass Spectrom. 2004; 39: 705.
6. Fra n´ ski R, Gierczyk B. Self-assembly process of copper cation
and 2,5-bis(3-pyridyl)-1,3,4-oxadiazole under electrospray
ionization conditions. Rapid Commun. Mass Spectrom. 2005; 19:
2
1998; 11: 2309.
585.
2
2
2
1. Acheson RM, Aplin RT, Bolton RG. Electron impact induced
alkyl-group fragmentation on the acridine nucleus. Org. Mass
Spectrom. 1977; 12: 518.
2. Lawesson SO, Schroll G, Bowie JH, Cooks RG. Electron impact
studies – XXII. Mass spectra of substituted benzimidazoles.
Tetrahedron 1968; 24: 1875.
3. Ruchardt C, Hassmann V. Aromatic diazonium salts, X. A one-
pot procedure for the Jacobson synthesis of indazole. Liebigs
Ann. Chem. 1980; 6: 908.
3
7. Gabelica V, Galic N, De Pauw E. On the specificity of
cyclodextrin complexes detected by electrospray mass
spectrometry. J. Am. Soc. Mass Spectrom. 2002; 13: 946.
3
8. Guo M, Song F, Liu Z, Liu S. Characterization of non-covalent
complexes of rutin with cyclodextrins by electrospray ionization
tandem mass spectrometry. J. Mass Spectrom. 2004; 39: 594.
Copyright  2006 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2006; 41: 312–322