Spectroscopic and theoretical investigation of the conformational space of a pyrazolo-thiazole precursor of extended dipole diazafulvenium methide intermediates

Article abstract of DOI:10.1016/j.molstruc.2008.12.061

The structure, preferred conformers and vibrational spectra of the pyrazolo-thiazole precursor of extended dipole diazafulvenium methide intermediates, dimethyl 2,2-dioxo-1H,3H-pyrazolo[1,5-c][1,3]thiazole-6,7-dicarboxylate (DPTD) were investigated in low-temperature noble gas matrices (Ar, Xe), low temperature neat amorphous and crystalline phases and in KBr pellet (crystal and melted phases) by infrared spectroscopy, supported by quantum chemical calculations. Two types of conformers were observed spectroscopically in the matrices and in the neat amorphous solid resulting from fast condensation of the vapour of the compound onto the cold (20 K) substrate of the cryostat. These conformers correspond to the two pairs of nearly degenerated structures exhibiting skew/cis (conformers S′C and SC′) and gauche/trans (conformers G′T and GT′) arrangements of the N{double bond, long}C-C{double bond, long}O/C{double bond, long}C-C{double bond, long}O moieties. In the crystalline phase, the vibrational signature of the compound indicates that it exists in a skew/cis arrangement. After melting of the crystal, a conformational mixture is formed, where both skew/cis and gauche/trans forms exist in equilibrium and, with all probability, also conformer G′C′. This latter conformer cannot exist in the low temperature matrices and amorphous state, in view of the very low energy barrier that separates this form from the lower energy SC′ conformer, which can be easily surpassed during deposition of the compound (conformational cooling).

Full text of DOI:10.1016/j.molstruc.2008.12.061

Journal of Molecular Structure 921 (2009) 101–108
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
Spectroscopic and theoretical investigation of the conformational space
of a pyrazolo-thiazole precursor of extended dipole diazafulvenium methide
intermediates
*
Cláudio M. Nunes, Susy Lopes, Teresa M.V.D. Pinho e Melo, Rui Fausto
Department of Chemistry, University of Coimbra, Rua Larga, P-3004-535 Coimbra, Portugal
a r t i c l e i n f o
a b s t r a c t
Article history:
The structure, preferred conformers and vibrational spectra of the pyrazolo-thiazole precursor of
extended dipole diazafulvenium methide intermediates, dimethyl 2,2-dioxo-1H,3H-pyrazolo[1,5-
c][1,3]thiazole-6,7-dicarboxylate (DPTD) were investigated in low-temperature noble gas matrices (Ar,
Xe), low temperature neat amorphous and crystalline phases and in KBr pellet (crystal and melted
phases) by infrared spectroscopy, supported by quantum chemical calculations. Two types of conformers
were observed spectroscopically in the matrices and in the neat amorphous solid resulting from fast con-
densation of the vapour of the compound onto the cold (20 K) substrate of the cryostat. These conformers
correspond to the two pairs of nearly degenerated structures exhibiting skew/cis (conformers S0C and SC0)
and gauche/trans (conformers G0T and GT0) arrangements of the N@C–C@O/C@C–C@O moieties. In the
crystalline phase, the vibrational signature of the compound indicates that it exists in a skew/cis arrange-
ment. After melting of the crystal, a conformational mixture is formed, where both skew/cis and gauche/
trans forms exist in equilibrium and, with all probability, also conformer G0C0. This latter conformer can-
not exist in the low temperature matrices and amorphous state, in view of the very low energy barrier
that separates this form from the lower energy SC0 conformer, which can be easily surpassed during
deposition of the compound (conformational cooling).
Received 4 December 2008
Accepted 19 December 2008
Available online 30 December 2008
Keywords:
Dimethyl 2,2-dioxo-1H,3H-Pyrazolo[1,5-
c][1,3]thiazole-6,7-dicarboxylate
Matrix isolation
Low temperature FTIR spectroscopy
DFT calculations
Conformational analysis
Ó 2008 Elsevier B.V. All rights reserved.
1. Introduction
The pyrazolo[1,5-c][1,3]thiazoles belong to a class of heterocy-
cles that possess two fused rings, a pyrazole and a thiazolidine ring,
Heterocyclic compounds play a central role in chemistry. They
have a wide range of applications and represent about half of the
known organic compounds [1]. The discovery of new heterocyclic
systems and understand their properties and structure are thus ex-
tremely important research topics. We have been studying pericy-
clic reactions of extended dipoles, such as azafulvenium methides
and diazafulvenium methides, in order to evaluate the scope of the
use of these intermediates in the synthesis of heterocyclic com-
pounds [2–7]. In particular, diazafulvenium methides 2a–2d par-
which share a common nitrogen atom. The synthesis of the first
derivatives was reported recently by Storr et al. [8–9]. On the other
hand, pyrrolo[1,2-c][1,3]thiazoles analogues (a pyrrole and a thia-
zolidine fused ring that share the same nitrogen atom) are known
at least since 1980 and some representatives show interesting bio-
logical activities [10–16].
The molecular properties of pyrazolo and pyrrolo-thiazoles are
practically unknown. To the best of our knowledge, only our recent
study
on
dimethyl
2,2-dioxo-5-methyl-1H,3H-pyrrolo[1,2-
ticipate in [8
p
+ 2
p
] cycloadditions giving pyrazolo-annulated
c][1,3]thiazole-6,7-dicarboxylate has been reported hitherto [17].
In that study, the conformational space of the compound was
investigated using low temperature infrared spectroscopy (for
the matrix-isolated monomers, as well as for the compound in
the neat amorphous solid and crystalline state) and computational
methods [17]. In this study, the conformational space of dimethyl
2,2-dioxo-5-methyl-1H,3H-pyrazolo[1,5-c][1,3]thiazole-6,7-dicar-
boxylate 1a (DPTD) was investigated using a similar approach in
order to set further insight on the structural and spectroscopic
properties of this type of extended dipole diazafulvenium methide
intermediates precursor.
heterocycles 3 and undergo intramolecular sigmatropic [1,8]H
shifts affording vinyl-1H-pyrazoles 4–5, some of which with poten-
tial biological activity (Scheme 1) [6,7]. The diazafulvenium meth-
ides 2a–2d can be generated from the thermal extrusion of sulfur
dioxide from 2,2-dioxo-1H,3H-pyrazolo[1,5-c][1,3]thiazoles 1a–
1d, making relevant the gathering of information on the chemistry
and structure of 2,2-dioxo-1H,3H-pyrazolo[1,5-c][1,3]thiazoles.
* Corresponding author. Tel.: +351 239 852080; fax: +351 239 827703.
E-mail address: rfausto@ci.uc.pt (R. Fausto).
0022-2860/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2008.12.061

102
C.M. Nunes et al. / Journal of Molecular Structure 921 (2009) 101–108
CO2Me
2 R1
R
CO2Me
O2S
N
N
1a R1 = R2 = R3 = H
R3
1b R1 = R2 = H; R3 = Me
1c R1 = R2 = Me; R3 = H
1d R1 = R2 = R3 = Me
MeO2C
Me
CO2Me
-SO2
2b R1 = R2 = H; R3 = Me
N
N
4
CO2Me
CO2Me
MeO2C
CO2Me
R1
R2
Dipolarophile
Sigmatropic [1,8]H Shift
N
[8p+2p] Cycloaddition
N
3
N N
R3
2a R1 = R2 = R3 = H
R3
MeO2C
Me
CO2Me
N
2b R1 = R2 = H; R3 = Me
N
5
2c R1 = R2 = Me; R3 = H
2d R1 = R2 = R3 = Me
R3
Scheme 1. Reactivity of diazafulvenium methides (2a–2d) generated from thermal extrusion of sulfur dioxide from 2,2-dioxo-1H,3H-pyrazolo[1,5-c][1,3]thiazoles (1a–1d).
or xenon N48; both obtained from Air Liquid) onto the CsI optical
substrate of the cryostat cooled to 10 K (for argon matrices) or 20 K
(for xenon matrices). The sublimation temperatures used were
ꢁ383 K, in xenon experiments and ꢁ403 K, in argon and neat DPTD
amorphous/crystalline state experiments. After depositing of the
compound, annealing experiments were performed by increasing
the temperature up to 35 K for, Ar matrices, and 75 K, for Xe matri-
ces. The spectra of the neat amorphous solid of DPTD was obtained
using a similar procedure, but with the flux of the cryogenic gas cut
off. The solid layer was then allowed to anneal at a slowly increas-
ing temperature up to 298 K. Infrared spectra were collected dur-
ing this process every 10–20 K of temperature change. After
temperature reached 298 K, the substrate was cooled back to
20 K, and spectra were again collected each 10–20 K.
In all experiments performed, an APD Cryogenics closed-cycle
helium refrigeration system with a DE-202A expander was used
and the temperature was measured directly at the sample holder
by a silicon diode temperature sensor, connected to a digital con-
troller (Scientific Instruments, model 9650-1), with an accuracy
of 0.1 K.
2. Experimental
2.1. Synthesis
Dimethyl
dicarboxylate (DPTD) 1a was prepared using a known synthetic
procedure (Scheme 2) [9]. Thiazolidine-4-carboxylic acid 6, ob-
tained for condensation of L-cysteine and formaldehyde in aqueous
solution, was nitrosated under standard conditions to give N-nitr-
osothiazolidine-4-carboxylic acid 7. Treatment of 7 with TFAA (tri-
fluoroacetic anhydride) in anhydrous ether furnished the bicyclic
mesoionic ring sydnone 8 in good yield. This later underwent a
2,2-dioxo-1H,3H-pyrazolo[1,5-c][1,3]thiazole-6,7-
4p + 2p cycloaddition with dimethyl acetylenedicarboxylate
(DMAD) to give the fused pyrazolo[1,5-c][1,3]thiazole 9 in high
yield. Finally, oxidation with MCPBA (m-cloroperoxybenzoic acid)
gives the sulfone 1a in almost quantitative yield.
2.2. Infrared spectroscopy
For matrix isolation and low temperature studies on neat DPDT,
the IR spectra were recorded using a Mattson Infinity (60AR series)
or Nicolet 6700 Fourier transform infrared spectrometer, equipped
with a deuterated triglycine sulphate (DTGS) detector and a Ge/KBr
beam splitter, with 0.5 cmꢀ1 spectra resolution. Infrared studies at
high temperature (298–418 K) were undertaken for the compound
as a potassium bromide pellet using a Specac temperature varia-
tion cell in a BOMEM (MB40) spectrometer, which has a Zn/Se
beam splitter and a DTGS detector, with 4 cmꢀ1 spectral resolution.
To avoid interference from atmospheric H2O and CO2, a stream of
dry nitrogen continuously purged the optical path of the
spectrometers.
2.3. Computational methodology
The quantum chemical calculations were performed with
Gaussian 03 [18] at the DFT level of theory, using the standard 6-
31+G(d) and 6-311++G(d,p) basis sets [19], and the three-parame-
ter density functional B3LYP, which includes Becke’s gradient ex-
change correction [20] and the Lee, Yang and Parr correlation
functional [21]. The conformational space of the DPTD molecule
were examined through a systematic independent variation of
the two conformationally relevant dihedrals (C2C1C14@O16 and
N28C4C15@O18; see Fig. 1) from ꢀ180° to 180°, with increments
of 15°. Optimization of the input structures was initially done at
the B3LYP/6-31+G(d) level of theory, the minimum energy struc-
To obtain the matrix spectra, DPTD was sublimated using a spe-
cially designed mini-furnace, which was placed inside the cryostat,
and co-deposited with a large excess of the matrix gas (argon N60
CO2Me
CO2Me
O
CO2Me
CO2Me
CO2H
CO2H
NH
MCPBA
96%
NaNO2
HCl
63%
O
TFAA
Et2O
76%
DMAD
xylene, Δ
90%
S
N
S
O2S
N N
S
N N
S
N N
NO
1a
6
7
9
8
Scheme 2. Synthesis of DPTD (1a).

C.M. Nunes et al. / Journal of Molecular Structure 921 (2009) 101–108
103
sent a general rule for molecules bearing the carboxylic acid or car-
boxylic ester groups [30–34]. The DPTD molecule can then be
conformationally characterized by two internal rotation axes (rota-
tions around the C4–C15 and C1–C14 bonds), which can be expected
to give rise to different low energy conformers. Internal rotations
around the O19–C20 and O17–C21 bonds would lead to the methyl
ester groups adopting high energy trans conformations (over
20 kJ molꢀ1 above the most stable cis forms), which can be safely
assumed to be irrelevant in practical terms. The identification of
the low energy conformers of DPTD can then be made by system-
atic investigation of the DFT(B3LYP)/6-31+G(d) potential energy
surface (PES) of the molecule, through incremental variation of
the two conformationally relevant dihedral angles (C2C1C14@O16
and N28C4C15@O18) in steps of 15°, with all remaining geometric
parameters being optimized at each point. Fig. 2 presents a contour
map of the PES calculated as described above. Six energy minima
(S0C, SC0, G0C0, GC, G0T and GT0) were found on this PES, all belonging
to the C1 symmetry point group. The geometries of these conform-
ers reoptimized at the DFT(B3LYP)/6-311++G(d,p) level of theory
are shown in Fig. 3 (the optimized geometrical parameters for all
structures are provided as Supporting Information; Table S1). The
calculated relative energies of the six minima as well as the rela-
tive energies of these structures with zero-point vibrational energy
corrections are given in Table 1.
All studied minimum energy conformations have a planar pyr-
azole ring and an envelope-like geometry for the thiazolidine ring
(with the C5, C2, N3, and C7 atoms nearly in the same plane and the
sulphur atom in the apex position, similarly to what was previ-
ously found for dimethyl 2,2-dioxo-5-methyl-1H,3H-pyrrolo[1,2-
c][1,3]thiazole-6,7-dicarboxylate[17]).
The differences in the geometrical parameters calculated for the
various minima do not in general exceed 0.6 pm for the bond
lengths (except for carbon oxygen bonds, in which the difference
can be of 1.9 pm) or 6–7° for the bond angles (again with the
exception of bond angles of the ester groups). In these forms, the
two conformationally relevant dihedral angles [N28C4C15@O18
(Phi) and C2C1C14@O16 (Psi)] adopt only two different orientations.
Phi assumes absolute values of ꢁ60° and ꢁ133°, here designated as
gauche and skew, and Psi absolute values of ꢁ10° and ꢁ170°, des-
ignated as cis and trans. According to the orientation of these two
H23
20
H22
H24
O19
15
H11
7
H10
N28
4
N3
O18
O9
1
2
S6
O8
H12
O17
5
H26
H25
14
21
H27
H13
O16
Fig. 1. Adopted numbering scheme for the DPTD molecule. Carbon atoms are
represented by their numbers.
tures being subsequently re-optimized and the vibrational fre-
quencies calculated at the B3LYP level with the /6-311++G(d,p)
split-valence triple-f basis set.
The underestimation of the theoretically predicted frequencies
for S@O stretching vibrations has been reported for different types
of compounds when calculated at the DFT (B3LYP) level with the 6-
311++G(d,p) basis set [22–25]. On the other hand, a good agree-
ment between experimental and predicted frequencies was ob-
served for both dimethyl sulphite [23] and sulphate [24] when
the B3LYP method was used with the aug-cc-pVQZ augmented
split valence quadruple-f basis set, whereas for 3-chloro-1,2-benz-
isothiazole-1,1-dioxide an accurate reproduction of the frequen-
cies associated with the S@O bond stretching modes could also
be achieved at the B3LYP/6-311++G(3df,3pd) level [25]. Indeed it
has been noticed that an extensive set of polarization functions is
necessary to correctly reproduce frequencies of hypervalent S@O
bonds, at least when the B3LYP functional is used [25]. Since the
size of the DPTD molecule makes calculations with an extensive
set of polarization functions unpractical the following strategy
was used in the present study to optimize the calculated values
for the S@O stretching modes in the DPTD molecule: firstly, the
vibrational frequencies of these modes calculated at the B3LYP/6-
311++G(3df,3pd) level for 3-chloro-1,2-benziothiazole-1,1-dioxide
[25] were compared to those obtained using the smallest B3LYP/6-
311++G(d,p) basis set; then, a proper scaling factor was obtained
taking into account the frequency ratios [B3LYP/6-311++G(d,p) to
B3LYP/6-311++G(3df,3pd)] for the antisymmetric and symmetric
S@O stretching vibrations (1.046 and 1.051, respectively); the
mean value (1.048) was used as scaling factor for the B3LYP/6-
311++G(d,p) S@O stretching frequencies calculated for DPTD. For
all other frequencies, a scaling factor of 0.990 was used.
Transition state structures and energy barriers for conforma-
tional interconversion were obtained using the synchronous tran-
sit-guided quasi-Newton (STQN) method [26]. Normal coordinate
analysis was undertaken in the internal coordinates space, as de-
scribed by Schachtschneider [27], using the program BALGA [28]
and the optimized geometries and harmonic force constants
resulting from the DFT(B3LYP)/6-311++G(d,p) calculations. The
internal coordinates used in this analysis were defined as recom-
mended by Pulay et al. [29].
3. Results and discussion
3.1. Geometries and energies
The DPTD molecule consists of two adjacent ester groups bound
to positions
6 and 7 of the 2,2-dioxo-1H,3H-pyrazolo[1,5-
Fig. 2. DFT(B3LYP)/6-31+G(d) potential energy (kJ molꢀ1) contour map for DPTD as
c][1,3]thiazole nucleus (C1 and C4 in the adopted numbering
scheme; see Fig. 1). These esters groups shall adopt a cis arrange-
ment (CH3–O–C@O approximately 0°) since the preference for
the cis orientation around the C–O bond has been found to repre-
a function of internal rotation about
N28C4C15@O18 and C2C1C14@O16 dihedral
angles. The two dihedral angles were changed incremently in steps of 15°, and all
the remaining coordinates were fully optimized at each point. Minima are indicated
by an asterisc and the first-order saddle points by a white square.

104
C.M. Nunes et al. / Journal of Molecular Structure 921 (2009) 101–108
Fig. 3. DFT(B3LYP)/6-311++G(d,p) calculated minima for conformers of DPTD. The values of zero-point corrected energy (in kJ molꢀ1) relative to conformer S0C are given in
parentheses. Dihedral angles N28C4C15@O18 (Phi) and C2C1C14@O16 (Psi) are also provided.
Table 1
Calculated relative energies (
PES minima of DPTD.
to the relative stability of the DPTD minima is the repulsive inter-
D
E in kJ molꢀ1) and dipole moments (
l in Debyes) of the
action between O18 (one of the carbonyl oxygen atoms; see Fig. 1)
and the nitrogen atom of the pyrazole ring (N28). Indeed, in S0C/SC0,
the N28ꢂꢂꢂO18 (ꢁ344 pm) repulsive interaction is weaker than those
existing in the other pairs of structures: GC/G0C0, N28ꢂꢂꢂO18 (ꢁ303
pm), and G0T/GT0, N28ꢂꢂꢂO18 (ꢁ303 pm). However, this repulsive
interaction is partially compensated in energetic grounds by a
more important repulsive interaction between N28 and O19 in the
S0C/SC0 forms (ꢁ280 pm) than in the remaining forms (GC/G0C0:
ꢁ332 pm; G0T/GT0: ꢁ333 pm).
a
Minimum
D
E
D
EZPVE
l (D)
S0C
SC0
0.00
0.78
2.24
2.24
3.84
4.04
0.00
0.63
2.08
2.11
3.40
3.57
3.51
4.53
4.03
4.85
2.88
4.27
G0C0
GC
G0T
GT0
a
Relative energies including zero point energy corrections.
3.2. Vibrational spectra of matrix-isolated DPTD
dihedrals, three types of minima can be defined: skew/cis (S0C and
SC0), gauche/cis (GC and G0C0) and gauche/trans (G0T and GT0). It
should be emphazised that the pair of minima forming a given type
of family according to the above classification is not mirror-image
because the non-planarity of the thiazolidine ring. Nevertheless,
these minima were found to have very similar energies (0.0/0.6,
2.1/2.1 and 3.4/3.6 kJ molꢀ1, for S0C/SC0, GC/G0C0 and G0T/GT0,
respectively), reflecting their close structural similarity.
The most significant factor that determines the relative stability
of the DPTD minima is the steric repulsion between the ester
groups, in particular those between the lone electron pairs of the
oxygen atoms. For the lowest energy S0C/SC0 (skew/cis) structures,
only one lone-electron pairs’ repulsive interaction exists [O17ꢂꢂꢂO18
(ꢁ300 pm)]. For the higher energy forms, two oxygen–oxygen
lone-electron pairs’ repulsions are significant: O17ꢂꢂꢂO19 (ꢁ305
pm) and O17ꢂꢂꢂO18 (ꢁ375 pm), for GC/G0C0 (gauche/cis), and O16ꢂꢂꢂO19
(ꢁ309 pm) and O16ꢂꢂꢂO18 (ꢁ385 pm), for G0T/GT0 (gauche/trans). The
slightly larger higher energies predicted for G0T and GT0 can be ac-
counted by the fact that in these forms it is the more negative car-
bonyl oxygen atom (O16) that is closest to the second carboxylic
group, whereas in the GC and G0C0 forms it is the ester O17 atom,
which bears a smaller negative charge. Another factor contributing
In order facilitate the interpretation of the infrared spectra of
matrix-isolated DPTD, it is very important to know the landscape
of the PES of the molecule in the relevant regions, in particular
the energy barriers for conformational interconversion between
the low energy conformers. Indeed, the calculated energy barriers
for conformational isomerization can be expected to be a good
approximation to the real isomerization barriers of the DPTD mol-
ecule in cryogenic matrices. Though very frequently these energy
barriers are large enough to allow the gas phase equilibrium con-
formational population to be efficiently trapped in the cryogenic
matrices, when they are very low (lower than a few kJ molꢀ1) the
situation is considerably more complex. In this case, the energy
barriers can easily be surpassed during deposition of the matrix
and higher energy conformers may relax into lower energy con-
formers. This phenomenon known as conformational cooling is
quite frequent and has been discussed in detail in some of our pre-
vious studies [35–38]. Its main consequence is to make the abun-
dances of conformers in a matrix different from those existing in
equilibrium in the gas phase prior to matrix deposition.
The B3LYP/6-311++G(d,p) calculated conformational isomeriza-
tion energy barriers are given in Table 2. The locations of the differ-
ent transition states (Figure S1; Supporting Information) on the

C.M. Nunes et al. / Journal of Molecular Structure 921 (2009) 101–108
105
Table 2
Calculated energies (kJ molꢀ1; relatively to the reactant species, presented in the left column) of the transition state structures for the interconversion between the energy minima
on the DFT(B3LYP)/6-311++G(d,p) potential energy surface of DPTDa.
S0C
SC0
GC
2.37 (1.88)
G0C0
G0T
GT0
S0C
SC0
GC
G0C0
G0T
GT0
–
7.03 (6.62)
–
14.95 (14.65)
–
6.25 (5.99)
0.13 (ꢀ0.21)
–
–
–
–
1.90 (1.65)
10.76 (10.80)
–
14.85 (14.72)
–
–
–
13.76 (13.70)
16.95 (16.68)
–
–
0.44 (0.17)
10.77 (10.77)
–
15.16 (15.20)
16.45 (16.00)
–
10.91 (10.77) 10.55 (10.58)
11.11 (11.25)
–
–
11.11 (10.94) 10.74 (10.75
10.50 (10.77)
a
Values in parentheses correspond to the relative energies of the transition state structures to the reactant minimum energy conformation taking into account zero point
energy corrections.
PES are shown in Fig. 2, which also shows the positions of the 6 low
energy minima. The transition states TS_G and TS_H are associated
with the interconversion of conformers gauche/cis into conformers
skew/cis. They define the energy barriers for the GC ? S0C and
G0C0 ? SC0 conversions, which were found to be extremely low:
0.13 and 0.44 kJ molꢀ1, respectively (2.37 and 1.90 kJ mol in the re-
verse direction). In fact, when zero point energy corrections were
considered, the G0C0 ? SC0 energy barrier reduces to only
0.17 kJ molꢀ1 and the GC ? S0C process becomes barrierless, with
the energy of TS_G being 0.21 kJ molꢀ1 below that of the GC form.
In practical terms, this means that GC shall not be considered a real
conformer, but a vibrationally excited form of conformer S0C. On
the other hand, it can be also anticipated that during matrix depo-
sition the G0C0 conformer should relax promptly to conformer SC0.
Since as detailed below all the other relevant energy barriers for
conformational isomerization are large enough to prevent any con-
version to take place during matrix deposition, it can be concluded
that two types of conformers, skew/cis (S0C/SC0) and gauche/trans
(G0T/GT0), are then expected to be present in the matrices. Con-
former SC0 should have a population given by the sum of its gas
phase population at the temperature of deposition with that of
the G0C0 form. The estimated populations for the four experimen-
tally relevant conformers (in matrices: S0C: SC0: G0T: GT0), at the
temperatures used in the experiments undertaken in this study,
are 33%: 44%: 11%: 11% (within an unit of percentage, the popula-
tions are identical at 383 and 403 K, the two experimentally rele-
vant temperatures; see Section 2).
formers into the lowest energy one, could also in principle take
place upon annealing of the matrices to higher temperatures. We
shall turn to this point below.
The spectra of DPTD isolated in argon (T = 10 K) and xenon
(T = 20 K) matrices are presented in Fig. 4. All four relevant con-
formers of DPTD belong to the C1 symmetry point group, with 78
fundamental vibrations, all of them expected to be active in the
infrared. The calculated spectra for these conformers and the cor-
responding results of normal coordinates analysis are provided as
Supporting Information (Tables S2–S6). Since the two fused rings
in the molecule are not significantly affected by the changes in ori-
entation of the ester groups, their associated vibrations are similar
and give rise to bands practically coincident in all conformers. In
addition, within each family of conformers (S0C/SC0 and G0T/GT0)
the ester fragments were also found to give rise to practically
coincident spectra. On the other hand, some of these vibrations
0.6
xenon matrix
(T= 20K)
0.4
0.2
0.0
1800 1650 1500 1350 1200 1050 900
750
600
450
As mentioned above, all other relevant barriers to conforma-
tional isomerization in DPTD are considerably larger than those be-
tween the gauche/cis and skew/cis structures. The SC0 ? S0C
isomerization (between the two skew/cis conformers; associated
0.4
0.2
argon matrix
(T= 10K)
with transition state TS_F) has
a
barrier of 5.99 kJ molꢀ1
(6.62 kJ molꢀ1 in the reverse direction), whereas that interconvert-
ing GC and G0C0 (the two gauche/cis structures; TS_E) amounts to ca.
10.8 kJ molꢀ1 in both directions. In turn, the predicted energy bar-
riers for conversion between gauche/trans conformers (GT0 ? G0T)
are 10.58 and 10.77 kJ molꢀ1 (10.75 and 10.94 kJ molꢀ1 in the re-
verse direction), corresponding to two different reaction pathways
(associated with transition states TS_C and TS_D, respectively; see
Fig. 2). The pathway for conversion of conformers gauche/trans into
skew/cis (i.e., G0T ? S0C and GT0 ? SC0) have energy barriers dic-
tated by the transitions states TS_I and TS_J, with values of 11.25
and 10.77 kJ molꢀ1, respectively. TS_A and TS_B could also in prin-
ciple allow the interconversion of forms gauche/trans into less
energetic conformers. However, the corresponding barriers are
considerably larger (ꢁ15 kJ molꢀ1) than those corresponding to
alternative pathways, which make these processes of no interest
in practical terms.
0.0
80
1800 1650 1500 1350 1200 1050 900
750
600
450
Simulated
60
40
20
0
1800 1650 1500 1350 1200 1050 900
Wavenumbers/cm-1
750
600
450
Fig. 4. Infrared spectra of DPTD isolated in solid argon and xenon (as-deposited
matrices; substrate temperatures: xenon, 20 K; argon, 10 K) and simulated spec-
trum built by adding the DFT(B3LYP)/6-311++G(d,p) calculated spectra of con-
formers S0C, SC0, G0T and GT0 scaled by their relative populations at the sublimation
temperatures (estimated from calculated relative energies and assuming the
Boltzmann distribution; see text). The theoretical spectra were simulated using
Lorenzian functions centered at the calculated frequencies and with a bandwidth-
It shall be noticed that all the energy barriers now discussed are
only moderately high (less than ca. 10 kJ molꢀ1) and, then, conver-
sions of the gauche/trans conformers into the skew/cis forms, as
well as between the higher energy member of each family of con-
at-half-height equal to 4 cmꢀ1
.

106
C.M. Nunes et al. / Journal of Molecular Structure 921 (2009) 101–108
are sufficiently different in the two families and could be used suc-
cessfully for their spectroscopic identification.
The theoretically simulated spectrum, obtained by taking into
account the calculated spectra for conformers S0C, SC0, G0T and
GT0 weighted by their expected populations in the as-deposited
matrices, is also shown in Fig. 4, for comparison. As it can be seen,
the experimental spectra of DPTD in both argon and xenon matri-
ces are well reproduced by the simulated spectrum.
Xe
Identification of bands due to each family of conformers (skew/
cis and gauche/trans) would be relatively easy if we could change
their population in the matrix. This could in principle be done in
two ways: varying the temperature of the gas phase prior to
deposition, or taking advantage of conformational cooling during
deposition of the matrix or upon annealing of the deposited ma-
trix to higher temperatures. Very unfortunately, none of these
strategies were found to be successfully in this case. Changing
the temperature of the gas prior to deposition was not efficient,
since within the usable temperature range (from the temperature
at which the compound starts to sublimate to that at which the
compound starts to decompose, i.e., from ca. 360 K to ca. 450 K)
the change in populations are too small (less than ca. 3%) to be
noticeable. On the other hand annealing of the argon matrices
up to the maximum possible temperature (ca. 35 K; from that
on, the argon matrix starts to evaporate considerably and be-
comes unusable) was found not to be enough to provide the re-
quired energy for the isomerization reactions to occur and the
spectrum did not change. In turn, the annealing experiments per-
formed in the xenon matrices (up to 75 K) lead to spectral
changes for the high temperatures investigated, but these are
mainly ascribable to aggregation. Even if it took place simulta-
neously with aggregation, conformational isomerization could
not be spectroscopically proved without any doubt. Furthermore,
the experiments in which the temperature of the substrate was
increased during deposition of the xenon matrices led essentially
to the same results as the annealing experiments in the same
type of matrix. Hence, the identification of the two types of con-
formers of DPTD in the matrices had to be done by simple direct
comparison between the experimental spectra and their calcu-
lated spectra. Fortunately, there are a few spectral regions where
the bands due to the two types of conformers are predicted to ap-
pear enough well separated. Figs. 5 and 6 show the 750–850 cmꢀ1
and 1050–1275 cmꢀ1 regions, where bands ascribable to both
types of experimentally relevant conformers can be observed. In
these figures, the simulated spectra correspond to the sum of
the calculated spectra of the two conformers belonging to the
same family (dashed lines: S0C and SC0; solid lines: G0T and GT0)
with the intensities weighted by their expected populations in
the matrices.
Ar
Calc.
825
800
775
750
Wavenumbers/cm-1
Fig. 5. Experimental spectra (750–850 cmꢀ1 range) of DPTD isolated in xenon
(20 K) and argon (10 K) matrices, and simulated spectra constructed from the
DFT(B3LYP)/6-311++G(d,p) calculated spectra of the two pairs of conformers
(dashed lines: S0C and SC0; solid lines: G0T and GT0). In the simulated spectra, the
intensities of the calculated spectra of individual conformers were weighted by
their expected populations in the matrices (S0C: SC0: G0T: GT0 = 33%: 44%: 11%: 11%;
see text).
The full list of observed bands both in argon and xenon matrices
and the proposed assignments are summarized in Table S7.
3.3. Vibrational spectra of neat condensed phases of DPTD and of DPTD
in KBr pellet
Figure S2 (Supporting Information) displays the infrared spectra
of DPTD in the neat condensed phases and in KBr pellet: polycrys-
talline DPTD in a KBr pellet, at room temperature (298 K); melted
DPTD in KBr pellet (418 K); low temperature (20 K) amorphous
phase resulting from fast deposition of the vapour (at 403 K) of
the compound onto the cold substrate of the cryostat and crystal-
line state resulting from warming the previous sample at 298 K
and subsequently cooling it to 20 K. As seen in this Figure, the
room temperature spectrum of the polycrystalline DPTD in a KBr
pellet is identical to the spectrum of the crystalline form obtained
by heating the amorphous film resulting from deposition of the va-
pour of the compound at low temperature, despite this latter
exhibits the expected band narrow effect due to the low tempera-
ture. Also the spectrum of the neat low temperature amorphous
state and melted phase in KBr pellet are similar. Though a defini-
tive answer to the conformation assumed by DPTD in the crystal
should rely on X-ray data, the comparison of spectrum of the crys-
tal with those calculated for the different conformers of DPTD
strongly suggests that in the crystalline state the DPTD molecules
exist in a skew/cis conformation. In Fig. 7, the 750–850 cmꢀ1 spec-
In the low frequency range (Fig. 5), the intense bands around
820 cmꢀ1 and the two main bands in the 785–770 cmꢀ1 range
are essentially due to the S0C and SC0 forms, the first group of bands
being mainly due to the dOCO and d(T-ring 2) modes and the sec-
ond to dOCO0 and the two
cC@O rocking modes (Tables S3 and S4).
On the other hand, the lower intensity bands in the 805–790 cmꢀ1
region and at ca. 766 cmꢀ1, as well as the shoulder ca. 775 cmꢀ1 in
the spectra of the xenon matrix are mainly due to the G0T and GT0
conformers and ascribable to vibrations involving predominantly
the dOCO and cC@O coordinates (Tables S5 and S6).
In the higher frequency region (Fig. 6) there is a more pro-
nounced overlap between the bands originated in the two pairs
of conformers. However, the bands at ca. 1215 and 1125 cmꢀ1 (in
particular the one at 1215 cmꢀ1) can be safely taken as being orig-
inated mostly in the less populated G0T and GT0 conformers, the
coupled cCH3(2)/m m(N–N P-ring)/twCH2 vibrations being
C–O0 and
the dominating contributors to these bands, respectively (see Ta-
bles S5 and S6).

C.M. Nunes et al. / Journal of Molecular Structure 921 (2009) 101–108
107
Xe
Exp.
Ar
Calc.
Calc.
1275 1250 1225 1200 1175 1150 1125 1100 1075 1050
Wavenumbers/cm-1
Fig. 6. Experimental spectra (1275–1050 cmꢀ1 range) of DPTD isolated in xenon
(20 K) and argon (10 K) matrices, and simulated spectra constructed from the
DFT(B3LYP)/6-311++G(d,p) calculated spectra of the two pairs of conformers
(dashed lines: S0C and SC0; solid lines: G0T and GT0). In the simulated spectra, the
intensities of the calculated spectra of individual conformers were weighted by
their expected populations in the matrices (S0C: SC0: G0T: GT0 = 33%: 44%: 11%: 11%;
see text).
825
800
775
750
Wavenumbers/ cm-1
tral range of the spectrum of the crystal in KBr pellet is shown in
detail (top panel, dashed line). In this spectrum, the characteristic
bands of the gauche/trans forms in the 805–790 cmꢀ1 region and at
ca. 766 cmꢀ1 and 775 cmꢀ1 are absent. Upon increasing the tem-
perature of the KBr pellet to 145 °C (418 K), the melt is obtained,
the phase transition being easily observed spectroscopically. Par-
ticularly interesting is the appearance of new bands in the spec-
trum, in particular those that are characteristic of the gauche/
trans conformers, which doubtlessly indicate that in this phase
DPTD exists as a mixture of different conformers. The 750–
850 cmꢀ1 spectral range, shown in Fig. 7, clearly reveals the pres-
ence of both skew/cis and gauche/trans conformers in the melted
state, through simple comparison with the calculated spectra for
these types of conformers that are also presented in the figure,
for comparison. The calculated spectra were simulated summing
the theoretically obtained spectra for the conformers of each fam-
ily (dashed line: S0C and SC0; thick solid line: G0T and GT0) with the
intensities weighted by their expected populations at 418 K. Note
that in the melted state, we could also expect a contribution from
conformer G0C0, since no conformational cooling can take place in
this case. However, considering the similarity of the spectrum of
this conformer and those of the gauche/trans conformers (see
Fig. 7) it is not possible to confirm this in a definitive way. It shall
be noticed that in the spectrum of the low temperature amorphous
neat compound, conformational cooling is expected to take place
and in this case only the skew/cis and gauche/trans conformers shall
contribute to the spectrum. Taking this into account, comparison of
the spectra of the neat amorphous state and melted phase could, in
principle, give us some indication of the features in this latter spec-
trum that are given rise by the G0C0 form. However, the bands in
both spectra are considerably broad and since, as already men-
tioned, the spectrum of the G0C0 conformer is very similar to those
of the gauche/trans conformers, it was not possible in practice to
identify with certainty any band in the spectrum of the melt that
can be ascribed to the G0C0 form.
Fig. 7. Top: experimental infrared spectra (750–850 cmꢀ1 range) of DPTD in KBr
pellet: polycrystalline, at room temperature (298 K) – dashed line; melted phase
(418 K)
– solid line; Bottom: simulated spectra obtained by summing the
theoretically obtained spectra for the conformers of each family (dashed line: S0C
and SC0; thick solid line: G0T and GT0; thin solid line: G0C0) with the intensities
weighted by their expected populations at 418 K according to the calculations
(S0C:SC0:G0T:GT0:G0C0 = 32:27:18:12:11%).
Table S8 (Supporting Information) presents the proposed
assignments for the spectra obtained in KBr pellet (for both crystal
a melted states) and for the low temperature neat amorphous and
crystalline phases.
4. Conclusions
The pyrazolo-thiazole precursor of extended dipole diazafulve-
nium methide intermediates, dimethyl 2,2-dioxo-1H,3H-pyrazol-
o[1,5-c][1,3]thiazole-6,7-dicarboxylate, DPTD, was studied by
infrared spectroscopy under different experimental conditions (in
both argon and xenon matrices, as a polycrystalline sample and
melted phase in KBr pellet, and in the neat amorphous and crystal-
line solid states). The experimental data was interpreted taking
into account results obtained by quantum chemical calculations
performed at the DFT/B3LYP level of theory using both the 6-
31+G(d) and 6-311++G(d,p) basis sets. Six low energy minima were
identified in the resulting potential energy surfaces (S0C, SC0, GC,
G0C0, G0T and GT0) with relative energies within ca. 4.5 kJ molꢀ1
.
Consideration of zero-point energies and barriers to conforma-
tional isomerization; however, led to establish the practical rele-
vance of only two pairs of conformers for the compound under
the low temperature experimental conditions considered in this
investigation: skew/cis (conformers S0C and SC0) and gauche/trans
(conformers G0T and GT0).

108
C.M. Nunes et al. / Journal of Molecular Structure 921 (2009) 101–108
[13] D. Lavé, C. James, H. Rajoharison, P.E. Bost, I. Cavero, Drugs Future 14 (1989)
In consonance with the theoretical analysis of the potential en-
891.
ergy surface of DPTD, the detailed analysis of the matrix-isolated
DPTD infrared spectra reveal the presence in the matrices of both
skew/cis and gauche/trans conformers. The same conformers of
DPTD were also found in both the amorphous phase resulting from
fast deposition of the sole vapour of the compound onto the cold
(20 K) substrate of the cryostat and in the melted compound at
high temperature. In this latter phase, the presence of the G0C0 con-
former could not be neither proved nor excluded, though in view of
the theoretical results its presence in this phase appears to be quite
probable. On the other hand, in the neat crystalline state the vibra-
tional signature of the compound indicates that it exists only in a
skew/cis arrangement.
[14] J. Fabre, D. Farge, C. James, D. Lavé, US Patent 4529728, 1985.
[15] Fabre, J.; James, C.; Lavé, D. US Patent 4786645, 1988.
[16] S.K. Davidsen, J.B. Summers, D.H. Albert, J.H. Holms, H.R. Heyman, T.J. Magoc,
R.G. Conway, D.A. Rhein, G.W. Carter, J. Med. Chem. 37 (1994) 4423.
[17] A. Kaczor, T.M.V.D. Pinho e Melo, Maria I.L. Soares, R. Fausto, J. Phys. Chem. A
110 (2006) 6531–6539.
[18] Frisch, M.J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J.
C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi,
M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara,
M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.;
Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.;
Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev,
O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.;
Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.;
Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck,
A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.;
Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.;
Wong, M. W.; González, C.; Pople, J. A. Gaussian 03, revision B.01; Gaussian,
Inc.: Wallingford, CT, 2004..
Acknowledgements
This work has been funded by the Portuguese Science Founda-
tion (FCT, Lisbon), under research project PTDC/QUI/71203/2006.
S.L. and C.M.N. also thank FCT for the Ph.D. Grants #SFRH/BD/
29698/2006 and #SFRH/BD/28844/2006.
[19] M. Frisch, M. Head-Gordon, Pople, J. Chem. Phys. Lett. 166 (1990) 281.
[20] A. Becke, Phys. Rev. A 38 (1988) 3098.
[21] C. Lee, W. Yang, R. Parr, Phys. Rev. B 37 (1988) 785.
[22] A.H. Zeng, L. Yu, Y. Wang, Q.Y. Kong, Q. Xu, M.F. Zhou, J. Phys. Chem. A 108
(2004) 6656.
Appendix A. Supplementary data
[23] A. Borba, P. Simões, R. Fausto, J. Phys. Chem. A 109 (2005) 3578.
[24] A. Borba, A. Gómez-Zavaglia, P. Simoes, R. Fausto, Spectrochim. Acta A – Mol.
Biomol. Spectrosc. 61 (2005) 1461.
[25] A. Kaczor, R. Almeida, A. Gómez-Zavaglia, M.L.S. Cristiano, R. Fausto, J. Mol.
Struct. 876 (2008) 77.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.molstruc.2008.12.061.
[26] C. Peng, H.B. Schlegel, Isr. J. Chem. 33 (1994) 449.
[27] J. Schachtschneider, Technical Report; Shell Development Co.: Emeryville, CA,
1969.
[28] M. Nowak, L. Lapinski, BALGA computer program for PED calculations.
[29] P. Pulay, G. Fogarasi, F. Pang, J.E. Boggs, J. Am. Chem. Soc. 110 (1979)
2550.
[30] A. Gómez-Zavaglia, A. Kaczor, A.L. Cardoso, T.M.V.D. Pinho e Melo, R. Fausto, J.
Phys. Chem. A 110 (2006) 10742.
[31] A. Gómez-Zavaglia, A. Kaczor, A.L. Cardoso, T.M.V.D. Pinho e Melo, R. Fausto, J.
Phys. Chem. A 110 (2006) 8081.
[32] E.M.S. Maçôas, L. Khriachtchev, M. Pettersson, R. Fausto, M. Räsänen, J. Phys.
Chem. A 109 (2005) 3617.
References
[1] T.R. Gilchrist, Heterocyclic Chemistry, third ed., Longman, Essex, 1997.
[2] T.M.V.D. Pinho e Melo, M.I.L. Soares, A. M. d’A. Rocha Gonsalves, H. McNab,
Tetrahedron Lett. 45 (2004) 3889.
[3] T.M.V.D. Pinho e Melo, Maria I.L. Soares, A.M.d’A. Rocha Gonsalves, J.A. Paixão,
A. Matos Beja, M. Ramos Silva, J. Org. Chem. 70 (2005) 6629.
[4] T.M.V.D. Pinho e Melo, M.I.L. Soares, A.M.d’A. Rocha Gonsalves, Tetrahedron
Lett. 47 (2006) 791.
[5] T.M.V.D. Pinho e Melo, M.I.L. Soares, C.M. Nunes, Tetrahedron 63 (2007) 1833.
[6] T.M.V.D. Pinho e Melo, C.M. Nunes, M.I.L. Soares, J.A. Paixão, A. Matos Beja, M.
Ramos Silva, J. Org. Chem. 72 (2007) 4406.
[7] M.I.L. Soares, T.M.V.D. Pinho e Melo, Tetrahedron Lett. 49 (2008) 4889.
[8] O.B. Sutcliffe, R.C. Storr, T.L. Gilchrist, P. Rafferty, A.P.A. Crew, Chem. Comm.
(2000) 675.
[33] J.-H. Lii, J. Phys. Chem. A 106 (2002) 8667.
[34] K.B. Wiberg, K.E. Laidig, J. Am. Chem. Soc. 109 (1987) 5935.
[35] I.D. Reva, S.G. Stepanian, L. Adamowicz, R. Fausto, Chem. Phys. Lett. 374 (2003)
631.
[36] A. Gómez-Zavaglia, R. Fausto, Phys. Chem. Chem. Phys. 5 (2003) 52.
[37] I.D. Reva, A.J. Lopes Jesus, M.T. Rosado, R. Fausto, M.E. Eusébio, J.S. Redinha,
Phys. Chem. Chem. Phys. 8 (2006) 5339.
[9] O.B. Sutcliffe, R.C. Storr, T.L. Gilchrist, P. Rafferty, Tetrahedron 56 (2000) 10011.
[10] J.M. Kane, J. Org. Chem. 45 (1980) 5396.
[11] W.K. Anderson, R.H. Mach, J. Med. Chem. 30 (1987) 2109.
[12] D. Ladureé, J.-C. Lancelot, M. Robba, E. Chenu, G. Mathé, J. Med. Chem. 32
(1989) 456.
[38] A. Borba, A. Gómez-Zavaglia, R. Fausto, J. Mol. Struct. 794 (2006) 196.