In situ direct photoproduction of ketenes from substituted coumarins isolated in solid argon: The case of N-(2-oxo-2H-chromen-3-yl)acetamide

Article abstract of DOI:10.1016/j.molstruc.2008.10.013

In this study, the infrared spectrum of N-(2-oxo-2H-chromen-3-yl)acetamide (3-acetamidocoumarin; 3AC) isolated in solid argon, at 10 K, was obtained and assigned. In consonance with the relative energies of the three conformers predicted theoretically, on

Full text of DOI:10.1016/j.molstruc.2008.10.013

Journal of Molecular Structure 924–926 (2009) 81–88
0
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
In situ direct photoproduction of ketenes from substituted coumarins isolated
in solid argon: The case of N-(2-oxo-2H-chromen-3-yl)acetamide
a
a
N. Kußs a,b, , S. Breda , R. Fausto
*
a Department of Chemistry, University of Coimbra, P-3004-535 Coimbra, Portugal
b Department of Physics, Anadolu University, 26440 Eskißsehir, Turkey
a r t i c l e i n f o
a b s t r a c t
Article history:
In this study, the infrared spectrum of N-(2-oxo-2H-chromen-3-yl)acetamide (3-acetamidocoumarin;
3AC) isolated in solid argon, at 10 K, was obtained and assigned. In consonance with the relative energies
of the three conformers predicted theoretically, only the most stable form was observed experimentally.
This conformer is stabilized by two intramolecular hydrogen bonds and is similar to the structural unit of
3AC found in crystalline phase. Upon in situ UV (k > 215 nm) irradiation of the matrix-isolated compound,
the characteristic IR intense band due to the antisymmetric stretching vibration of the ketene (AC@C@O)
group was observed, indicating occurrence of the ring-opening isomerization reaction to the open-ring
ketene isomeric of 3AC. In consonance with the theoretical structural predictions for the most stable iso-
mers of this photoproduct, the experimental data indicates that it is produced in the E arrangement of the
(O@)CAC@CAC(@C@O) fragment. There were also experimental indications pointing to occurrence of a
second photoreaction channel, corresponding to decarbonylation. On the other hand, contrarily to what
Received 25 July 2008
Accepted 6 October 2008
Available online 21 October 2008
Keywords:
Matrix isolation
Infrared spectroscopy
DFT calculations
UV-induced photochemistry
3-Acetamidocoumarin
is generally observed for
production of Dewar isomer of 3AC was observed. This last result, follows the trend observed for 2-pyr-
one-3-carboxylate, and seems to be a quite general rule for matrix-isolated -pyrones bearing relatively
a-pyrones derivatives, including unsubstituted coumarin, no photochemical
a
volumous substituents at the position 3, as a consequence of the unfavorable relaxation of the matrix
around the guest molecule that would be required to accommodate the Dewar isomers of these com-
pounds, whose structure deviates strongly from planarity, thus mismatching the primarily occupied
matrix sites.
Ó 2008 Elsevier B.V. All rights reserved.
1. Introduction
the Dewar isomer formation is much more effective for most of the
substituted derivatives, while the ring-opening photoreaction, lead-
Coumarins are important natural occurring and synthetic com-
pounds which show several relevant applications. In particular,
they exhibit different biological activities, including anticoagu-
lant, spasmolytic, diuretic, anthelmintic and hypoglucemic actions
[1–8].
ing to the isomeric ketene, proceeds easier for unsubstituted a-pyr-
one. The ring-opening reaction leading to the ketene species is
*
p character,
believed [15,16] to originate from excited states with n
whereas the process leading to formation of the Dewar isomers
*
should start from pp states. In
a
-pyrones, the lowest excited sin-
*
*
The properties of coumarins are in large extent determined by
glet state has n
p
character [17]. Hence, the n
p
-type photochemis-
their constituting
a very rich photochemistry [9–14]. In general,
a
-pyrone unit, which has been shown to present
-pyrones undergo
try is favored for these compounds when they are free of any
substituents. On the other hand, conjugative type substitution with
a
*
two main competitive photochemical reactions: ring-opening, lead-
ing to formation of the isomeric open-ring ketenes, and ring-con-
traction to the corresponding Dewar isomers [9–14]. However, the
precise photochemistry and relative importance of these two
photochannels is strongly dependent on the substituents present
in the heterocyclic ring. For example, it has been demonstrated that
groups such as AOH, ACH3 or AOCH3, blue-shifts the S1 S0 (n
p
)
*
transition and red-shifts the S2 S0 (pp ) transition [17,18], reduc-
ing the gap between the S1 and S2 states and favoring the process
leading to the Dewar formation. Note that the ring-opening reaction
in S1 is not a barrierless process [19,20] and some excess of excita-
tion energy is needed to promote the a-bond cleavage. Hence, this
process does not necessarily need to dominate, even though the
*
lowest of the excited states is of n
tween the n
p character. If the energy gap be-
* Corresponding author. Address: Department of Chemistry, University of Coim-
bra, P-3004-535 Coimbra, Portugal.
*
*
p and pp states is not too large, the processes typical
E-mail address: nkus@anadolu.edu.tr (N. Kußs).
0022-2860/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2008.10.013

82
N. Kußs et al. / Journal of Molecular Structure 924–926 (2009) 81–88
*
for pp photochemistry can compete with the
even dominate. Isomerizations to the Dewar forms observed as
dominating processes for 4,6-dimethyl- -pyrone, 4-hydroxy- and
-pyrones are good examples of such behav-
a
-bond cleavage or
(T = 10 K) mounted on the tip of the cryostat. All experiments were
done on the basis of an APD Cryogenics close-cycle helium refrig-
eration system with a DE-202A expander. The matrices were irra-
diated through the outer quartz window of the cryostat, using a
200 W output power of a 500 W Hg(Xe) lamp (Oriel, Newport). Dif-
ferent cut-off filters were used: k > 337, 315, 285, 235 nm and, fi-
nally, unfiltered radiation (k > 215 nm).
a
4-methoxy-6-methyl-
ior [9,10].
Very interestingly, for the matrix-isolated compounds, when
the substituent is a volumous group at the position 3 of the -pyr-
a
a
one ring, the ring-contraction reaction becomes less easy and, at
least in some cases, it does not occur at all. This has been recently
observed for matrix-isolated 3-methyl coumalate, where only the
ring-opening reaction was observed [13,14]. On the other hand,
the 5-substituted analogue (5-methyl coumalate) shows the ex-
pected dominance of the Dewar isomerization process over the
2.2. Computational methodology
The equilibrium geometries for 3AC and its photoproducts were
fully optimized at the DFT level of theory with the 6-311++G(d,p)
split valence triple-f basis set. The DFT calculations were carried
out with the three-parameter B3LYP density functional, which in-
cludes Becke’s gradient exchange correction [23] and the Lee, Yang,
Parr correlation functional [24]. All calculations were carried out
using the Gaussian 03 program package [25].
Vibrational spectra were calculated at the same level of theory.
Transformations of the Cartesian force constants to molecule-fixed
symmetry adjusted internal coordinates allowed the ordinary nor-
mal coordinate analysis to be performed as described by Schach-
tschneider [26]. The symmetry internal coordinates used in this
analysis were defined as recommended by Pulay et al. [27] and
are provided as Supplementary data (Table S1). Potential energy
distribution (PED) matrices [28] have been calculated and the
elements of these matrices greater than 10% were used to charac-
terize the different vibrations in terms of the chosen symmetry
coordinates.
ring-opening reaction, as found for the other substituted
already studied.
a-pyrones
Recently, we have described the photochemistry of matrix-iso-
lated coumarin [21]. This compound was shown to react according
to three main reaction channels, two of them corresponding to
those previously observed for matrix-isolated
a-pyrones: (a)
decarboxylation to form bicyclo[4.2.0]octa-1,3,5,7-tetraene and
CO2, with the Dewar form of coumarin as intermediate; (b) isom-
erization leading to the conjugated ketene isomer. The third ob-
served reaction pathway, decarbonylation, with production of a
CO/benzofuran complex, follows the dominant UV-induced photo-
chemistry of coumarin in gas phase [22].
In the present study, we focused our attention on N-(2-oxo-2H-
chromen-3-yl)acetamide (3-acetamidocoumarin; 3AC. Scheme 1),
which is a substituted coumarin bearing a relatively volumous sub-
stituent at the position 3 of the a-pyrone ring. As it will be shown, for
this compound isolated in a cryogenic inert matrix the photochem-
3. Results and discussion
ical channel leading to formation of the Dewar isomer is also closed.
3.1. Geometries and energies
2. Materials and methods
Isolated molecule of 3AC has never been studied previously
either experimentally or theoretically. To the best of our knowl-
edge, only two reports appeared previously on this compound.
The first one is a 1H and 13C NMR study, which provides chemical
shifts for 3AC in DMSO-d6 solution at room temperature, and sug-
gests that the compound should exist under those experimental
conditions predominantly in a conformation where the two car-
bonyl groups are anti to each other [29]. The second study [30]
considered the room temperature crystalline phase of the com-
pound. The crystals were found to be monoclinic (a = 10.0907,
b = 4.8785, c = 19.096 Å), space group P2(1)/c, with 4 molecules
per unit cell. In the crystal, the molecules exhibit a conformation
where the two carbonyl groups are anti to each other and are inter-
acting by strong intermolecular NAH. . .O@C(ring) hydrogen bonds
forming anti-parallel dimers. The dimers are aligned forming twin-
2.1. Experimental details
N-(2-oxo-2H-chromen-3-yl)acetamide (3-acetamidocoumarin;
3AC) was purchased from Sigma (purity 99%) and used without
any further purification. The infrared spectra were obtained using
a Mattson (Infinity 60AR Series) Fourier transform infrared spec-
trometer, equipped with a deuterated triglycine sulphate (DTGS)
detector and a KBr beamsplitter, with 0.5 cmꢀ1 spectral resolution.
Necessary modifications of the sample compartment of the spec-
trometer were done in order to accommodate the cryostat head
and allow purging of the instrument by a stream of dry nitrogen
to remove water vapors and CO2. A solid sample of 3AC was placed
in a specially designed mini-oven assembled within the cryostat
and sublimated by heating the compound to 125 °C. The vapor of
3AC was then codeposited together with large excess of the host
matrix gas (argon N60, from Air Liquide) onto a cold CsI window
columns through weak
p-stacking interactions and weaker
CAH. . .O(ring) hydrogen bonds [30]. The common element to these
previous studies is the found preference for the molecule of 3AC to
assume a conformation where the two carbonyl groups are anti to
each other.
H 22
H 23
H 24
H 18
In the present study, we performed a detailed conformational
search on the B3LYP/6-311++G(d,p) potential energy surface of
the 3AC molecule. From this search, three different conformers
could be identified. These conformers are depicted in Fig. 1. Se-
lected geometrical parameters for these forms are provided as Sup-
plementary data (Table S2) where they are compared with those
obtained by X-ray [29] and also with those calculated at the same
level of theory for the parent compound, coumarin [21].
According to the calculations, the most stable conformer of 3AC
corresponds to the structure observed in the crystalline state [29].
For the isolated molecule, two intramolecular hydrogen bonds char-
acterize this conformer, the first established between the NH group
O 20
21
19
H 17
H
13
10
4
16H
N
12
9
8
5
6
3
2
1
O
7
O
15H
11
H
14
Scheme 1. 3AC with atom numbering scheme adopted in this study.
and the a-pyrone ring carbonyl oxygen atom (with the NH. . .O dis-

N. Kußs et al. / Journal of Molecular Structure 924–926 (2009) 81–88
83
respectively. In conformer II, the deviation of the two fragments
from being in the same plane is essentially determined by the
repulsion between the methyl group and the hydrogen atom
bound to C4, while in form III it results mainly from the sterically
more relevant interaction between the methyl group and the
pyrone ring carbonyl group (see Fig. 1).
a-
3.2. Infrared spectrum of the as-deposited matrix
Taking into consideration the predicted relative energies of the
3 conformers of 3AC, it can be easily estimated that conformer I ac-
counts for more than 99% of the total population in gas phase at
125 °C, the temperature of sublimation of the compound used to
prepare the matrices. Hence, this form can be expected to be the
sole form present in the as-deposited matrix. Fig. 2 shows the
infrared spectrum of the as-deposited 3AC in argon matrix, at
10 K, and the B3LYP/6-311++G(d,p) predicted spectrum for con-
former I. The full set of calculated frequencies and infrared intensi-
ties and normal coordinate analysis results for the 3 conformers of
3AC are provided in the Supplementary data (Tables S3–S5). As it
can be seen in Fig. 2, the calculated spectrum of conformer I agrees
very well with the experimentally observed spectrum. The assign-
ment of the observed spectrum is given in Table 1. The following
observations deserve further discussion:
Fig. 1. Conformers of 3AC and their relative energies (with zero-point vibrational
corrections), as calculated at the DFT(B3LYP)/6-311++G(d,p) level of theory.
(a) The experimental spectrum reveals existence of multiple
trapping sites of the 3AC molecule in the matrix. This can be ob-
served in all spectral regions, but it is very clearly evidenced, for in-
stance, in the NAH and C@O stretching regions, around 3400 cmꢀ1
and in the 1750–1700 cmꢀ1 range (Fig. 2 A and B). Multiple matrix
trapping sites are commonly observed for heterocyclic compounds
[32,33] and they were found to be particularly numerous for mol-
tance and \NAH. . .O angle equal to 2.186 Å and 107.8°, respec-
tively), and the second between the C4AH and the amide oxygen
(C4H. . .O, 2.242 Å; \C4AH. . .O, 116.8°). These two intramolecular
hydrogen bonds strongly stabilize this conformer, which is pre-
dicted to be more stable than the conformers II and III by 27.4 and
37.9 kJ molꢀ1, respectively (zero-point energy corrected energy dif-
ferences). In conformer II, the NH. . .O hydrogen bond is kept
(NH. . .O, 2.167 Å; \NAH. . .O, 108.1°), but the second H-bond inter-
action is replaced by a repulsive interaction between the methyl
group and the hydrogen atom bound to C4 (see Fig. 1), whereas in
conformer III no intramolecular hydrogen bonds are present.
The presence of the two H-bonds in conformer I is clearly re-
vealed by the relative values of the C@O bond lengths calculated
for the different conformers (C2@O: 1.210 Å in I and 1.207 Å in II,
where the hydrogen bond are present, and 1.201 Å in form III,
where it does not exist; C19@O: 1.219 Å in I, vs. 1.216 Å in both
II and III). It is also interesting to note that the presence of the
NH. . .O bond in conformers I and II of 3AC makes the O1AC2 bond
ecules containing the
a-pyrone moiety, including coumarin itself
[21]. Explanation for this fact is certainly out of the scope of this
study, but appears as an interesting question that might eventually
be successfully addressed by using appropriate theoretical ap-
proaches (e.g., molecular dynamics).
(b) According to the theoretical data, one can expected to ob-
serve two intense bands in the 1375–1355 cmꢀ1 spectral range.
The highest frequency band should correspond to the symmetric
methyl bending mode and have an intensity less than half of that
of the lowest frequency band, which should correspond to a vibra-
tion with the largest contribution located in the d (C4AH) oscillator
(see Table 1). On the other hand, the relative intensities of the ob-
served bands in the same spectral range (ca. 1372 and 1358 cmꢀ1
)
(a to the carbonyl) becoming much shorter than in unsubstituted
coumarin (1.365 Å vs. 1.396 Å; in conformer III, this bond is
1.386 Å, much closer to the value found in coumarin; Table S2),
since it leads to a more electronegative carbonyl carbon atom,
which then favors the electron delocalization from the O1 lone
electron pairs to the O1AC2 bond, increasing its double bond char-
acter. On the other hand, as it could be expected, the O1AC6 bond
length varies in the opposite direction, attaining its maximum va-
lue in 3AC form I (1.374 Å) and minimum value in coumarin
(1.365 Å) (see Table S2).
The calculations predicted all the conformers to belong to the C1
point group, though the deviation of form I from Cs symmetry is
very small, corresponding only to a slight skewing (ca. 4°) of the
methyl group from the alignment with the molecular plane (Table
S2). A similar small geometric distortion in the CH3AC(@O) moiety
from the symmetric Cs geometry was recently reported for hydrox-
yacetone [31]. As found for hydroxyacetone [31], the energy differ-
ence between the Cs structure and the two symmetrically
equivalent minima it separates stays below the zero-point vibra-
tional level of these later, which means that in practical terms
the relevant (most probable) structure is the symmetric one. In
conformers II and III, the planes of the coumarin and acetamide
fragments make angles (C4@C3AN12AC19) of 14.5° and 128.2°,
appear with the opposite relative intensities (Fig. 2). Similar situa-
tions involving vibrations mainly localized in the C4AH bond can
also be noticed in the 1205–1160 cmꢀ1 and 925–910 cmꢀ1 spectral
ranges (see Table 1). As mentioned before, in the experimentally
relevant conformer of 3AC (conformer I) the C4AH bond is in-
volved in an intramolecular H-bond. The observed inconsistencies
can then be correlated with the expected greater sensitivity of the
vibrations associated with the H-bond interacting groups to media
effects. The assignments for these bands proposed in Table 1 were
based on the relative intensities of the bands.
(c) The predicted position for the
mNAH stretching band is ca.
3500 cmꢀ1 (scaled value). However, the corresponding experimen-
tal band is observed at
a considerably lower frequency
(ꢁ3400 cmꢀ1). This is an indication that the calculations underesti-
mate in some extent the strength of the NH. . .O intramolecular
hydrogen bond. Taking this into consideration, one can expect that
the frequency of the cNAH rocking mode is underestimated by the
calculations, since this mode is also well known to be very sensitive
to the H-bond strength, increasing its frequency with the strength of
the hydrogen bond [34–36]. Indeed, the calculations predict the
c
NAH rocking mode at 670 cmꢀ1, while the band in the experimen-
tal spectrum ascribable to this vibration is observed at 778 cmꢀ1
.

84
N. Kußs et al. / Journal of Molecular Structure 924–926 (2009) 81–88
A
B
40
30
20
10
0
300
250
200
150
100
50
0
1800
1700
1600
1500
1400
1300
3600
3400
3200
3000
0.4
0.3
0.2
0.1
0.0
1.0
0.8
0.6
0.4
0.2
0.0
3600
3400
3200
3000
1800
1700
1600
1500
1400
1300
Wavenumber / cm-1
Wavenumber / cm-1
C
100
50
25
50
0
0
1300
1200
1100 1000
800
600
400
0.6
1.0
0.5
0.0
0.4
0.2
0.0
1300
1200
1100
1000
Wavenumber / cm-1
800
600
400
Fig. 2. Experimental (as-deposited argon matrix; 10 K) and calculated spectrum of 3AC in different spectral regions: (A) 3600–2900 cmꢀ1; (B) 1800–1300 cmꢀ1; (C) 1300–
1100 and 1050–400 cmꢀ1. The calculated frequencies were scaled by 0.987; the spectrum was then simulated by Gaussian functions (with 4 cmꢀ1 halfband width) centred at
the calculated scaled frequencies.
Note that the assignment of the 778 cmꢀ1 experimental band to the
NAH rocking mode was also facilitated by the fact that no other
band is predicted by the calculations to occur at this frequency
the ring-opening (to isomeric ketene) and ring-contracting (to De-
war isomer) reactions being generally observed and their relative
importance being influenced by the substituents present in the
heterocyclic ring. An interesting point was the observation that
c
(see Fig. 2).
the presence of a volumous group at the position 3 of the a-pyrone
3.3. Photochemical experiments
ring strongly reduces the importance of the ring-contraction reac-
tion for the matrix-isolated compounds, in view of the energeti-
cally demanding reorganization of the matrix host atoms that
would be required to accommodate the newly formed strongly
As mentioned in Section 1, the
shown to exhibit an interesting photochemistry [9–14], with both
a-pyrone moiety has been

N. Kußs et al. / Journal of Molecular Structure 924–926 (2009) 81–88
85
Table 1
Observed infrared frequencies and intensities of 3AC in argon matrix and B3LYP/6-311++G(d,p) calculated spectrum and potential energy distributions (PED) for conformer I.a
a Frequencies in cmꢀ1, calculated intensities in km molꢀ1
S1 for definition of symmetry coordinates.
. m, bond stretching; d, bending; c, rocking; s torsion; s, symmetric; as, anti-symmetric; n.o., not observed. See Table
b Experimental intensities given were obtained by normalizing the observed relative absorbances in such a way that the total intensity of the observed bands is equal to the
total intensity of the calculated bands having an experimental counterpart.
c
Theoretical positions of absorption bands were uniformely scaled by 0.978. The full calculated spectrum and results of normal coordinate analysis are presented in the
Supplementary data Table S3.
d
PED’s lower than 10% are not included.

86
N. Kußs et al. / Journal of Molecular Structure 924–926 (2009) 81–88
non-planar Dewar species [13,14]. Since 3AC is a molecule bearing
a relatively volumous substituent at the position 3 of the -pyrone
a
ring, the photochemical ring-contraction reaction could be ex-
pected not to take place easily for the matrix-isolated compound.
Another interesting a priori question related with the possibility
or not of 3AC also undergo decarbonylation reaction in a similar
way to what was observed for coumarin [21]. Furthermore, there
was also the question of the photochemical stability of the amide
moiety under the present experimental conditions. Lundell et al.
[37] reported the first photolysis experiments of formamide in
both argon and xenon matrices through irradiation with an exci-
mer laser operating at 193 nm (ArF), and observed two major
channels, leading to the formation of CO and HNCO, complexed
with NH3 and H2, respectively. The vacuum UV photolysis
(k > 160 nm) of acetamide was also studied at low temperature
(in argon matrix) by Duvernay et al. [38], which found HNCO:CH4
and CO:CH3NH2 molecular complexes as main products, together
with acetimidic acid.
C
K(Z)
K(E)
CO
2100
2000
1900
1800
1700
The irradiation experiments performed in the present study
used a similar approach to that used in our previous study of cou-
marin [21]. In particular, different cut-off filters were used: k > 337,
315, 285, 235 nm and, finally, unfiltered radiation (k > 215 nm). No
changes in the spectra were observed upon irradiation with the fil-
tered light. On the other hand, unfiltered irradiation led to appear-
ance of new bands growing in the 2150–2125 cmꢀ1 region. Fig. 3
shows the 2200–1650 cmꢀ1 range of the spectra of the as-depos-
ited and irradiated (k > 215 nm; 215 min.) matrix of 3AC, showing
the appearance of the bands due to photoproducts and consump-
tion of the bands of the reactant. This figure also presents data
we previously obtained for unsubstituted coumarin subjected to
identical irradiation experiments, for comparison.
3AC
K(E)
*
CO
The first conclusion that can be extracted from this figure is
that, as found for coumarin, upon photolysis with k > 215 nm ma-
trix-isolated 3AC undergoes a ring-opening Norrish type I a-cleav-
2100
2000
1900
1800
1700
Wavenumber/ cm-1
Fig. 3. 2200–1650 cmꢀ1 spectral region of the spectra of 3AC (bottom) and
coumarin (C; top) before (thin lines) and after (bold lines) UV irradiation
(k > 215 nm). For 3AC total time of irradiation was 215 min.; for coumarin,
90 min. The band marked with asterisk in the spectra of 3AC is due to an unknown
trace impurity. K(Z) and K(E) indicate bands due to ketene forms with the
(O@)CAC@CAC(@C@O) fragment in the Z and E conformation, respectively.
age to its isomeric ketene. Indeed, the band observed in the spectra
of the irradiated matrix at 2138 cmꢀ1 can be assigned to the
mC@C@O anti-symmetric stretching mode of the ketene in a geom-
etry with the (O@)CAC@CAC(@C@O) fragment in the E conforma-
tion. This is exactly the frequency for the same vibration observed
in the spectra of the irradiated matrix of coumarin, whose assign-
ment has been discussed in detail in Ref. [21]. Very interestingly,
the band at 2120 cmꢀ1 observed in the spectra of the irradiated
matrix of coumarin (see Fig. 3) and assigned to the Z isomers of
the ketene [21], has no equivalent in the case of 3AC, as it could
be expected taking into account that all the low energy
(D ) conformers of the 3AC ketene have an E
E < 49 kJ molꢀ1
arrangement of the (O@)CAC@CAC(@C@O) fragment, because in
the Z forms a strong steric repulsion between the acetamido-ke-
tene substituent and the carbonyl group of the ketene ring exists
(see Fig. 4, where the geometries and relative energies of the most
stable forms of the 3AC ketene isomer are shown).
The second observation is that CO is also produced in the
photolysis of 3AC. In fact, like for coumarin, we can expect decarb-
onylation reaction to take place in 3AC, to give rise to the corre-
sponding substituted benzofuran. In addition, once produced the
benzofuran shall form a complex with the extruded CO in the
matrix cage initially occupied by the reactant molecule. In this
complex, the relative orientation of the CO molecule with respect
to benzofuran may be different and the shoulders observed at
2146 and 2142 cmꢀ1 (Fig. 3) are with all probability due to these
species. Indeed, these shoulders appear at the same positions of
the small bands previously [21] assigned to the benzofuran/CO
complexes obtained from photolysis of matrix-isolated coumarin
(see Fig. 3). Very unfortunately, the benzofuran moiety does not
give rise to any strong or medium intensity characteristic infrared
band that can be clearly visible in the spectra of the irradiated
Fig. 4. Lowest energy conformers of the isomeric ketene of 3AC, calculated at the
B3LYP/6-311++G(d,p) level of theory. Forms
I
to IV have their
(O@)CAC@CAC(@C@O) fragment in the conformation, while conformer
corresponds to the lowest energy conformer of the ketene possessing the
E
V
Z
arrangement of the (O@)CAC@CAC(@C@O) fragment.

N. Kußs et al. / Journal of Molecular Structure 924–926 (2009) 81–88
87
matrix at the level of concentration resulting from the photoin-
duced reaction. Such observation would certainly strenghtened
the conclusion regarding occurrence of the decarbonylation photo-
process. Note that, on the other hand, the band centered at
2138 cmꢀ1 cannot be attributed to CO, even if a minor contribution
of this species to the total intensity of the band cannot be excluded,
since the carbon monoxide absorption is weak in infrared and, if all
area below the absorption band at 2138 cmꢀ1 would originate in
CO, this would require the amount of photoproduced CO molecules
to be more than three times the amount of consumed 3AC. On the
other hand, the ketene moiety gives rise to an extremely strong
infrared absorption (above 1000 km molꢀ1) in this region.
The third conclusion that can be extracted from Fig. 3 is related
with the relative facility of the two compounds (3AC and couma-
rin) to react under the experimental conditions used. For coumarin,
it was observed a decrease in the band intensities indicating that
38% of the initially present compound has already been consumed,
while only 21% of 3AC was transformed after 215 min. irradiation.
There are two main justifications for the smaller ability of 3AC to
react: (i) Firstly, the number of opened reaction photochannels is
different in 3AC and coumarin. In coumarin, besides the ring-open-
ing reaction to the ketene form and the decarbonylation reaction,
isomerization to the Dewar form was also observed, which after
experimental indications pointing to occurrence of photochemical
decarbonylation of 3AC, leading to formation of a benzofuran/CO
complex. On the other hand, no photochemical production of De-
war isomer of 3AC was observed, contrarily to what is generally
observed for
a-pyrones derivatives, including unsubstituted cou-
marin. This later result, came in the line of the results obtained
for the methyl coumalates [13,14], and reinforced the idea that
the isomerization of matrix-isolated
a-pyrones bearing relatively
volumous substituents at the position 3 to the corresponding De-
war form is a disfavored process. This can be interpreted as being
a consequence of the unfavorable relaxation of the matrix around
the guest molecule that in these cases isomerization to the Dewar
form implies, in view of the strong deviation from the planarity of
these later species that then strongly mismatch the primarily occu-
pied matrix sites. The relative ability of 3AC and coumarin were
also compared and an explanation proposed for the smaller reac-
tivity found for 3AC under the used experimental conditions.
Acknowledgement
S. Breda thanks the Portuguese Science Foundation (Fundação
para a Ciência e a Tecnologia (FCT, Portugal) for the Grant
#SFRH/BD/16119/2004. Preliminary studies on 3AC were done un-
der the projects POCI/QUI/58937/2004 (FCT, Portugal) and
#200519010 (Eskißsehir Osmangazi University, Turkey) and the
European Union Erasmus Programme.
decarboxylation led to the final observed product,
a bicy-
clo[4.2.0]octa-1,3,5,7-tetraene/CO2 complex [21]. On the other
hand, there are no indications in the spectra of the irradiated ma-
trix of 3AC of production of these species, a result that is also in
agreement with the fact that when the a-pyrone moiety has a vol-
Appendix A. Supplementary data
umous substituent at the position 3, the ring-contraction photore-
action leading to the Dewar form is strongly unfavoured in low
temperature matrix [13,14]. (ii) Secondly, the ring-opening reac-
tion process can also be expected to be intrinsically less favored
in 3AC than in coumarin, due to the presence in the first molecule
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.molstruc.2008.10.013.
References
of the amido substituent. As stated in Section 1 conjugative type
*
[1] A. Burger, Burger‘s Medicinal Chemistry, in: M.E. Wolf (Ed.), Wiley-
Interscience, New York, 1995.
[2] I.E.M. Bultink, W.F. Lems, P.J. Kostense, B.A.C. Dijkmans, A.E. Voskuyl, Arthritis
Rheum. 52 (2005) 2044.
substitution in the
transition and red-shifts the S2 S0
a
-pyrone ring, blue-shifts the S1 S0 (n
p
)
*
(
pp ) transition [17,18],
reducing the gap between the S1 and S2 states and disfavoring
the process leading to the ketene formation relatively to that lead-
ing to isomerization to the Dewar form. So, in 3AC the ketene for-
mation channel can be expected to be less efficient, whereas by the
reasons already pointed out the channel leading to the Dewar form
is not active for the matrix-isolated reactant. In addition, there has
been noticed that the transition states for decarbonylation and ke-
tene formation for coumarin [21] are located very closely on the
potential energy surface which may imply the existence of an effi-
cient ‘‘cross-road” between the isomerization and decarbonylation
photochannels. It can be expected that a similar situation occurs
for 3AC, which could then also be a factor contributing to the smal-
ler reactivity found for 3AC when compared to coumarin.
[3] R.A. O’Reilly, J. Clinic. Investig. 46 (1967) 829.
[4] H. Zhao, N. Neamati, H. Hong, A. Mazumder, S. Wang, S. Sunder, G.W.A. Milne,
Y. Pommier, T.R. Burke, J. Med. Chem. 40 (1997) 242.
[5] J. Hirsh, J.E. Dalen, D.R. Anderson, L. Poller, H. Bussey, J. Ansell, D. Deykin, Chest
119 (2001) 8S.
[6] Y. Kashman, K.R. Gustafson, R.W. Fuller, J.H. Cardellina, J.B. McMahon, M.J.
Currens, R.W. Buckheit, S.H. Hughes, G.M. Cragg, M.R. Boyd, J. Med. Chem. 35
(1992) 2735.
[7] K.C. Fylaktakidou, D.J. Hadjipavlou-Litina, K.E. Litinas, D.N. Nicolaides, Curr.
Pharm. Design 10 (2004) 3813.
[8] R.N. Gacche, D.S. Gond, N.A. Dhole, B.S. Dawane, J. Enz. Inhib. Med. Chem. 21
(2006) 157.
[9] S. Breda, L. Lapinski, R. Fausto, M.J. Nowak, Phys. Chem. Chem. Phys. 5 (2003)
4527.
[10] S. Breda, L. Lapinski, I. Reva, R. Fausto, J. Photochem. Photobiol. A 162 (2004)
139.
[11] S. Breda, I. Reva, L. Lapinski, R. Fausto, Phys. Chem. Chem. Phys. 6 (2004) 929.
[12] S. Breda, I.D. Reva, L. Lapinski, L. Frija, M.L. Cristiano, R. Fausto, J. Phys. Chem. A
110 (2006) 6415.
4. Conclusions
[13] I.D. Reva, M.J. Nowak, L. Lapinski, R. Fausto, Chem. Phys. Lett. 429 (2006)
382.
[14] I.D. Reva, M.J. Nowak, L. Lapinski, R. Fausto, Chem. Phys. Lett. 452 (2008) 20.
[15] N.J. Turro, Modern Molecular Photochemistry, University Science Books,
Sausalito, California, USA, 1991.
[16] M. Klessinger, J. Michl, Excited States and Photochemistry of Organic
Molecules, VCH, New York, USA, 1995.
[17] J.S. Seixas de Melo, G. Quinteiro, J. Pina, S. Breda, R. Fausto, J. Mol. Struct. 565/
566 (2001) 59.
In this study, the infrared spectrum of 3AC isolated in solid ar-
gon was recorded and assigned. Theoretical calculations predicted
the existence of three different conformers of 3AC. However, in
consonance with the theoretically estimated relative energies of
the conformers, only the most stable form was observed experi-
mentally. This conformer is stabilized by two intramolecular H-
bonds and similar to the structural unit found in crystalline 3AC
[30]. UV (k > 215 nm) irradiation of the matrix-isolated compound
was found to lead to isomerization of the compound to its open-
ring isomeric ketene. In consonance with the theoretical structural
predictions for the most stable isomers of this photoproduct, the
experimental data indicated that it is produced in the E arrange-
ment of the (O@)CAC@CAC(@C@O) fragment. There were also
[18] R.S. Becker, Theory and Interpretation of Fluorescence and Phosphorescence,
Wiley/Interscience, New York, USA, 1969.
[19] E.W.G. Diau, C. Kotting, A.H. Zewail, ChemPhysChem 2 (2001) 273.
[20] E.W.G. Diau, C. Kotting, A.H. Zewail, ChemPhysChem 2 (2001) 294.
[21] N. Kusß, S. Breda, I. Reva, E. Tasal, C. Ogretir, R. Fausto, Photochem. Photobiol. 83
(2007) 1237. Erratum: Photochem. Photobiol. 83 (2007) 1541.
[22] T. Yatsuhashi, N. Nakashima, J. Phys. Chem. A 104 (2000) 1095.
[23] A.D. Becke, Phys. Rev. A 38 (1988) 3098.
[24] C.T. Lee, W.T. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.

88
N. Kußs et al. / Journal of Molecular Structure 924–926 (2009) 81–88
[25] M.J. Frisch, G.W. Trucks, H. B. Schlegel, G.E. Scuseria, M.A. Robb, J.R.
Cheeseman, J.A. Montgomery, Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M.
Millam, S.S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N.
Rega, G.A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J.
Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.
E. Knox, H.P. Hratchian, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R.
Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W.
Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G.
Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D.
Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S.
Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I.
Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A.
Nanayakkara, M. Challacombe, P.M. W. Gill, B. Johnson, W. Chen, M.W.
Wong, C. Gonzalez, J.A. Pople, Gaussian 03, Revision C.02, Gaussian, Inc.,
Wallingford CT, 2004.
[29] F.J. Martínez-Martínez, I.I. Padilla-Martínez, J. Trujillo-Ferrara, Magn. Reson.
Chem. 39 (2001) 765.
[30] E.V. García-Baez, F.J. Mariınez-Martínez, H. Hopfl, I.I. Padilla-Martínez, Crystal
Growth & Design 3 (2003) 35.
[31] A. Sharma, I.D. Reva, R. Fausto, J. Phys. Chem. A 112 (2008) 5935.
[32] S. Breda, I.D. Reva, L. Lapinski, M.J. Nowak, R. Fausto, J. Mol. Struct. 786 (2006)
193.
[33] R. Fausto, I. D. Reva, A. Gómez-Zavaglia, in: T.M.V.D. Pinho e Melo, A.M.d.A.
´
Rocha Gonsalves (Eds.), Recent Research Developments in Heterocyclic
Chemistry , 37/661 (2), Fort P.O., Trivandrum-695 023, Kerala, India, 2007.
[34] M. Rozenberg, G. Shoham, I.D. Reva, R. Fausto, Spectrochim. Acta A 62 (2005)
233.
[35] M. Rozenberg, G. Shoham, I.D. Reva, R. Fausto, Phys. Chem. Chem. Phys. 7
(2005) 2376.
[36] M. Rozenberg, G. Shoham, I.D. Reva, R. Fausto, Spectrochim. Acta A 60 (2004)
463.
[37] J. Lundell, M. Krajewska, M. Räsänen, J. Phys. Chem. A 102 (1998) 6643.
[38] F. Duvernay, P. Chatron-Michaud, F. Borget, D.M. Birney, T. Chiavassa, Phys.
Chem. Chem. Phys. 9 (2007) 1099.
[26] J.H. Schachtschneider, Technical Report, Shell Development Co., Emeryville,
CA, 1969.
[27] P. Pulay, G. Fogarasi, F. Pang, J.E. Boggs, J. Am. Chem. Soc. 101 (1979)
2550.
[28] G. Keresztury, G. Jalsovszky, J. Mol. Struct. 10 (1971) 304.