Entropy effects in conformational distribution and conformationally dependent UV-induced photolysis of serine monomer isolated in solid argon

Article abstract of DOI:10.1016/j.molstruc.2005.09.021

Monomeric serine can be trapped in low temperature argon matrices in different conformers, which can be classified in three groups (A, B, C) accordingly to the main intramolecular interaction they exhibit: A (OH_A?N hydrogen bond), B (OH_C?N) and C (OH_A?O{double bond, short}) (subscripts A and C stand for alcohol and carboxylic group, respectively). The OH_C?N intramolecular interaction found in B-type conformers is considerably stronger than both the OH_A?N and OH_A?O{double bond, short} hydrogen bonds, and leads to reduce the abundance of B-type form relatively to A and C forms at high temperatures due to entropy effects. When submitted to UV irradiation (λ>200 nm), the main observed photoprocess is decarboxylation, leading to production of CO₂ and ethanolamine. A less important photochemical process is also observed, where the compound undergoes decarbonylation, with formation of CO, H₂O and acetamide. The two observed photoprocesses were found to be dependent on the conformation assumed by the reactant molecule, with A- and C-type conformers of serine undergoing decarboxylation and B-type conformers decarbonylation.

Full text of DOI:10.1016/j.molstruc.2005.09.021

Journal of Molecular Structure 786 (2006) 175–181
www.elsevier.com/locate/molstruc
Entropy effects in conformational distribution and conformationally
dependent UV-induced photolysis of serine monomer isolated in solid argon
*
S. Jarmelo , R. Fausto
Department of Chemistry, University of Coimbra, 3004-535 Coimbra, Portugal
Received 6 September 2005; received in revised form 27 September 2005; accepted 27 September 2005
Available online 7 November 2005
Abstract
Monomeric serine can be trapped in low temperature argon matrices in different conformers, which can be classified in three groups (A, B, C)
accordingly to the main intramolecular interaction they exhibit: A (OHA/N hydrogen bond), B (OHC/N) and C (OHA/Oa) (subscripts A and
C stand for alcohol and carboxylic group, respectively). The OHC/N intramolecular interaction found in B-type conformers is considerably
stronger than both the OHA/N and OHA/Oa hydrogen bonds, and leads to reduce the abundance of B-type form relatively to A and C forms at
high temperatures due to entropy effects. When submitted to UV irradiation (lO200 nm), the main observed photoprocess is decarboxylation,
leading to production of CO2 and ethanolamine. A less important photochemical process is also observed, where the compound undergoes
decarbonylation, with formation of CO, H2O and acetamide. The two observed photoprocesses were found to be dependent on the conformation
assumed by the reactant molecule, with A- and C-type conformers of serine undergoing decarboxylation and B-type conformers decarbonylation.
q 2005 Elsevier B.V. All rights reserved.
Keywords: Serine; Matrix-isolation infrared spectroscopy; Entropy effect on conformational equilibrium; Conformationally dependent photodecarboxylation
1. Introduction
spectra of monomeric neutral form of serine [11]. In the
gaseous phase, the compound was found to exist in several
different conformers, which can be classified in three groups
(A, B, C) accordingly to the main intramolecular interaction
they exhibit: A (OHA/N hydrogen bond), B (OHC/N) and C
(OHA/Oa) (throughout this article, in references to the OH
groups, the subscripts A and C stand for alcohol and carboxylic
group, respectively). Representatives of each of these groups
could be trapped in low temperature argon matrices and
characterized structurally and spectroscopically [11].
The study of the neutral (non-zwitterionic) form of
aminoacids by matrix isolation infrared spectroscopy has
been receiving considerable interest in recent years [1–9].
Indeed, this experimental approach has been particularly well
succeeded in elucidating many important details of the
molecular physics and chemistry of this type of compounds.
Some notable examples are (i) the first experimental
observation of the third conformer of glycine [1], which until
then had successively escaped to detection by other techniques
[10], (ii) the first direct observation of conformationally
selective aggregation in aminoacids, in the case, N,N-
dimethylglycine [7], and (iii) the identification of the entropy
associated with the intramolecular hydrogen-bonds present in
the possible conformers of aminoacids as an important variable
to determine their relative gas-phase populations [8,9].
In this study, the conformational equilibrium of serine in the
gaseous phase (TZ470 K) was re-investigated by taking into
account entropy effects. In addition, the matrix-isolated serine
was submitted to UV irradiation (lO200 nm) and the influence
of the conformation of the reactant molecule in determining the
preference for two observed reaction channels (decarboxyla-
tion and decarbonylation) was investigated.
Recently, matrix isolation connected with infrared spec-
troscopy was also successfully used to investigate the
molecular structure, preferred conformations and vibrational
2. Experimental and computational details
Both DL-serine and its D,LK3,3-dideuterated (O98% D)
isotopologue used in this study were commercial products,
obtained from Aldrich and Icon Stable Isotopes, respectively.
Matrices were prepared as described elsewhere [11]. The
temperatures of the vapor of the compound prior to deposition
and of the cold tip of the cryostat were 470 and 10 K,
*
Corresponding author. Tel.: C35 1239852080; fax: C35 1239827703.
E-mail address: sjarmelo@qui.uc.pt (S. Jarmelo).
0022-2860/$ - see front matter q 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2005.09.021

176
S. Jarmelo, R. Fausto / Journal of Molecular Structure 786 (2006) 175–181
H
respectively. Decomposition of serine was determined by
differential scanning calorimetry to start only above 490 K and,
accordingly, no evidence of significant decomposition was
found in the registered spectra of the as-deposited matrix. The
matrices were irradiated through the outer quartz window of
the cryostat, using a high-pressure mercury lamp (HBO200)
fitted with a water filter and a cutoff filter transmitting light
with lO200 nm.
All the theoretical calculations were performed using the
Gaussian98 program package [12] at the DFT/B3LYP level of
theory [13,14], with the extended 6-311CCG(d,p) basis set
[15]. The selection of the conformers submitted to investi-
gation took into account the results of our previous extensive
conformational study on the same compound [11]. Thermo-
dynamic quantities were obtained as described in [16].
H
H
H
H
O
N H
H
H
N
N
O
O
H
H
H
H
O
H
O
1
O
H
H
H
H
H
O
O
H
O
H
H
H
O
H
H
O
H
H
H
H
O
H
O
O
H
O
H
H
O
H
2
H
H
N
H
O
N
O
H
H
H
H
N
H
H
H
O
O
H
H
O
O
H
H
O
H
H
H
O
O
H
O
3
H
H
H
H
H
N
H
O
H
N
H
H
N
H
H
H
H
H
H
H
H
N
N
H
O
H
O
N
H
O
H
H
O
O
H
O
3. Results and discussion
H
O
H
4
5
H
O
H
H
H
H
O
H
H
H
A systematic search on the conformational potential energy
surface of serine in the neutral form was presented in our
previous study on this compound [11]. The DFT/B3LYP/6-
311CCG(d,p) theoretical calculations revealed the existence
of 61 distinct minima, with the first nine forms (Fig. 1), which
differ in energy by less than ca. 7 kJ molK1, being predicted to
account to ca. 75% of the total conformational population at
470 K (the temperature of the vapor of the compound
immediately before deposition) [11]. The estimation of the
relative conformational populations made in Ref. [11] assumed
nearly similar entropy contributions to the free energy of the
conformers, and the experimental analysis of the spectra were
shown to be in general good agreement with the theoretical
predictions. However, since entropy effects were more recently
found to be strongly relevant in determining the gas phase
relative conformational populations in phenylalanine [8], a
more detailed thermochemical analysis was undertaken in the
present study. The results are summarized in Table 1.
O
H
H
O
O
O
H
H
H
H
H
H
H
O
O
O
H
O
H
H
N
H
O
N
HN
H
H
H
H
H
H
O
H
O
H
H
H
O
H
H
O
H
H
O
H
O
H
H
6
H
H
O
H
O
H
N
H
H
N
O
N
H
H
H
H
H
O
H
O
H
H
H
H
O
O
H
N
H
O
O
H
H
H
H
7
H
N
N
O
H
H
H
O
H
H
H
O
H
H
H
O
H
H
H
H
O
O
H
O
O
O
H
H
O
O
H
H
H
H
8
9
H
H
N
N
As shown in this table, the most stable conformers of serine
are representatives of bonding schemes A, B and C, whose
dominant intramolecular interactions are the OHA/N,
OHC/N and OHA/Oa hydrogen bonds, respectively.
B-type conformers (e.g. 2, 3, 5, 8 and 9) are stabilized by a
strong OHC/N hydrogen bonding interaction, where the
combination of donor and acceptor groups is the most favorable
among all possible combinations. However, to establish this
H-bond, the conformation adopted by the carboxylic group must
be the less stable trans configuration (dihedral angle OaC–O–H
is ca. 1808) [17–20]. On the other hand, in the A-type conformers
(e.g. 1, 4 and 6), as well as in the C form (7), the intramolecular
hydrogen bonds are weaker than in B forms, but the carboxylic
group adopts the most stable cis conformation (dihedral angle
OaC–O–H ca. 08) [17–20]. The resulting global minimum is a
type A conformer (see Fig. 1).
H
H
O
H
N
H
H
O
H
H
H
H
O
H
H
H
O
H
O
H
H
H
O
O
H
H
H
H
O
H
O
N
NH
H
O
H
N
H
H
H
H
Fig. 1. The nine most stable conformers of serine, drawn in three different
perspectives.
see Table 1). At 470 K, the smaller conformational flexibility
of the B conformers, bearing the strongest intramolecular
hydrogen bond, justifies the smaller entropy contribution (TDS)
to their free energies and, consequently, the increase of the
corresponding free energy differences relatively to the most
stable conformer. This effect is more important for the lowest
energy B-type conformers (2 and 3), which correspond to those
forms where the intramolecular hydrogen bonds are stronger;
in the higher energy B-conformers (5, 8, 9) the entropy
contributions practically do not differ from those found in
A-type conformers, and the free energy differences at 470 K
are only slightly larger than the zero-point corrected relative
As it could be expected, entropy only plays a significant role
at high temperatures, with the relative conformational energies
estimated for low temperatures (either enthalpies, DH, or Gibbs
free energies, DG) being essentially equal to the previously
reported [11] zero-point corrected relative energies (DEZPVE
;

S. Jarmelo, R. Fausto / Journal of Molecular Structure 786 (2006) 175–181
177
energies. Conformer C has also a slightly larger entropy
contribution to the free energy at 470 K, since in this form,
besides the dominant H-bond interaction (OHA/Oa, which
could be expected not to differ significantly in energy from that
found in the A conformers, OHA/N), two additional
interactions take place (NH/OA and NH/OC; see Fig. 1),
contributing to increase the rigidity of the molecule (and, thus,
reducing its entropy).
In spite of the above mentioned entropy effects, in serine,
contrarily to what was observed in phenylalanine [8], the
relative populations of the different conformers are not strongly
affected by this variable, in particular in what concerns the
general characterization of the equilibrium at 470 K. Indeed,
the two most stable conformers are predicted to be the same
(conformers 1 and 2, of A and B type, respectively) with or
without consideration of entropy effects, and the relative
populations of the nine most stable conformers estimated from
the calculated free energies or from the zero-point corrected
energies do not differ significantly, except that the B-type
conformers experiment a general reduction of abundance. This
fact justifies the good agreement observed in our previous
study [11] between the experimental analysis of the vibrational
spectra of the compound and the theoretical predictions based
on the relative populations estimated from zero-point corrected
energies, i.e. without considering entropy effects. Indeed, the
present calculations give further support to the band assign-
ments previously made, in particular with respect to the
identification of the specific conformer that is in the origin of
the different observed spectral features. In turn, this provides a
strong support to the interpretation of the photochemical
experiments reported in the present study.
The matrix isolation infrared spectra of serine and 3,3-
dideutero-serine, in the range 4000–400 cmK1, were discussed
in detail previously [11]. Based on the comparison of the
experimentally observed frequencies with those obtained
theoretically at the DFT/B3LYP/6-311CCG(d,p) level for
the nine most stable conformers of the two isotopologues,
characteristic bands of each family of conformers could then be
identified. In this paper, only the nOH and nCaO stretching
regions will be discussed in detail, which are representative for
the goal of showing the dependence of the observed
photochemical reactions on the conformation assumed by the
reactant molecule.
In the 3700–3100 cmK1 spectral region, besides the nOH
stretching modes, in which our attention is focused, the
theoretical calculations predict that all conformers of practical
relevance should also give rise to bands due to nNH2 stretching
modes. The bands due to nOH stretching vibrations were
predicted to appear at considerably different frequencies in the
three groups of conformers (Fig. 2). In conformers belonging to
group A, the nOHC bands are predicted to occur at slightly
higher frequencies than nOHA, since in these conformers the
alcohol group is H-bonded to the amine group. The predicted
frequency for nOHC is typical of a free cis carboxylic group
and is considerably lower than those of the free alcohol groups
in all conformers of group B but conformer 3 (where OHA is
H-bonded to the carbonyl oxygen; see Fig. 1). In conformer 7

178
S. Jarmelo, R. Fausto / Journal of Molecular Structure 786 (2006) 175–181
Fig. 2. The nO–H and nNH2 stretching region (above 3350 cmK1) of the IR
spectrum of matrix-isolated serine (Ar matrix) and the spectra calculated at the
DFT/B3LYP/6-311CCG(d,p) level for the nine experimentally relevant
conformers. The calculated spectra for conformers belonging to the same group
(A, B or C) were added with individual intensities multiplied by the estimated
relative populations at 470 K. The nO–HC of the H-bonded B-type conformers
are observed as a broad band centered around 3170 cmK1 not shown in this
Figure.
Fig. 3. The nCaO stretching region of the IR spectrum of matrix-isolated serine
(Ar matrix) and the spectra calculated at the DFT/B3LYP/6-311CCG(d,p)
level for the nine experimentally relevant conformers. The calculated spectra
for conformers belonging to the same group (A, B or C) were added with
individual intensities multiplied by the estimated relative populations at 470 K.
(group C), the predicted order of frequencies for the nOHC and
nOHA modes is reversed, since in this case the H-bond
involving the alcohol group is established with the carbonyl
oxygen atom, that is a weaker H-bond acceptor than the amine
nitrogen. In spite of the change in the order of the two nOH
stretching modes in groups A and C, the general pattern of the
spectra for all conformers belonging to these two groups is
predicted to be identical (see Fig. 2). In all conformers
belonging to group B (except conformer 3), where the alcohol
group is free, nOHA is predicted at frequencies above
3600 cmK1. Conformer 3 gives rise to a nOHA vibration with
a frequency similar to those found in groups A and C. In all
B-type conformers the nOHC is predicted at considerably lower
frequencies, due to the involvement of the carboxylic group in
the strongest OHC/N hydrogen bond. In this spectral region,
the experimental spectra nicely fit the theoretical predictions,
as shown in Fig. 2, where the assignments are summarized.
The agreement between the calculated and experimentally
observed spectra is also very good in the nCaO stretching
spectral region (Fig. 3). Five main bands due to monomeric
serine are observed in between 1850 and 1740 cmK1. The
absorptions with maximum at ca. 1720 cmK1, which increased
considerably upon annealing, are originated in aggregates,
appearing at similar frequencies observed for aggregates of
analogous compounds [21]. B-conformers give rise to nCaO
bands at higher frequencies than A-conformers, whereas
conformer 7 (of the C-type) absorbs at a considerably lower
frequency than all the other forms. Conformer 3 (B-type) is
once again an exception to the regularities found for other
B-type conformers, due to the presence in this form of the
relatively stronger OHA/OaH-bond, which reduces the
frequency of the carbonyl stretching mode, making it similar
to those found in the A-group of conformers.
In situ UV irradiation (lO200 nm) of the matrix led to a
general decrease of intensity of the bands ascribed to serine
(see Figs. 2 and 3), while bands due to CO2 (around 2340 and
668 cmK1), CO (at ca. 2138 cmK1) and H2O (both in the 3780–
3600 and 1660–1550 cmK1 regions), appeared in the spectra
(or increased in intensity1), testifying the occurrence of both
1 Indeed, in the spectra of the as-deposited samples, small bands due to traces
of both H2O and CO2, appearing in the matrices as impurities originated mainly
from atmospheric gases adsorbed on the surfaces of the inlet devices and
vacuum line, could already be observed. These were, however, well-isolated
species and do not interfere with the experiment.

S. Jarmelo, R. Fausto / Journal of Molecular Structure 786 (2006) 175–181
179
CH2OH
hν
+
CO2
H2N
C
H
CH2OH
C
cis COOH
H
conformers
H2N
COOH
Ethanolamine
Carbon Dioxide
H
hν
CH3
Serine
trans COOH
CO
Carbon Monoxide
H2O
H2N
C
O
+
+
conformers
Acetamide
Water
Fig. 4. Unimolecular photodecomposition channels observed in the UV (lO200 nm) photochemistry of matrix isolated serine.
photodecarboxylation and photodecarbonylation reactions.
However, the intensities of the bands ascribed to the different
conformers (which are directly related with their relative
populations and, then, with the corresponding rates of
consumption) showed a different pattern of variation with
time of irradiation (see Figs. 2 and 3): conformer 7 (C-type)
was found to react more extensively, followed by A-type
conformers (1, 4 and 6) and then by B-type forms. This is to say
that the experimental observations indicate that the photoreac-
tions are strongly dependent on the conformation of the
reactant molecule, occurring preferentially for those con-
formers where the carboxylic group is in the cis conformation
and, consequently, not engaged in any H-bond as proton donor.
The formation of CO2, as major product, and of CO together
with H2O, as minor products, indicates that the dominant
observed photoreaction is the decarboxylation of serine. This
reaction shall lead to simultaneous production of CO2 and
ethanolamine (Fig. 4), while in the less important reaction
acetamide shall be produced as primary photoproduct, together
with CO.
A new band in the spectra of the irradiated matrices was also
observed at 2276 cmK1 (Fig. 5). This is a characteristic
frequency of the nNaCaO antisymmetric stretching of the
isocyanate moiety [26], and can be due to traces of
methylisocyanate (CH3NCO) present in the matrix as result
of secondary processes. Note that CH3NCO should also absorb
around 776 cmK1, and, indeed, the spectral intensity in the
780–765 cmK1 region was also found to increase after
irradiation. Nevertheless, the assignment of these weak
features to methylisocyanate shall be considered as tentative,
since no previous indication was found of direct production of
Photoproduced CO2
Ar Matrix
UV irradiated
0.15
0.10
0.05
0.00
Photoproduced CO2
Isolated CO2
The observed bands of the photoproduced CO2 and CO
show broad profiles, which strongly differ from those of the
monomeric species isolated in argon (Fig. 5) [22,23], clearly
indicating that these compounds are interacting with other
species produced in the same matrix cage during photolysis of
serine. As mentioned above, these species are, with all
probability, ethanolamine and acetamide, respectively. How-
ever, identification in the spectra of the irradiated matrices of
bands originated in these compounds could not be made
unequivocally because of the low intensity of their infrared
spectra [24,25] and extensive overlap with bands due to the
serine (for example, in the high-frequency nOH stretching
region, the most stable conformer of ethanolamine isolated in
an argon matrix absorbs at 3555 cmK1 [24], which coincides
exactly with the frequency of the most intense group of bands
of serine observed in this spectral region). In addition, for
ethanolamine, there is also the possibility of photoproduction
of different conformers, which leads also to reduce the
intensities of bands due to individual conformers, while the
total amount of photoproduced acetamide is certainly very low,
since this species is produced in the substantially less important
reaction channel.
675
670
665
660
Photoproduced CO
Isolated CO2
Isolated CO
2350
2250
2150
–1
Wavenumber / cm
Fig. 5. CO2 bending (670–660 cmK1), and CO2 asymmetric stretching (2360–
2320 cmK1), CO stretching (ca. 2140 cmK1) spectral regions of the serine
irradiated matrix and typical spectra of free CO2 and CO isolated in argon. The
origin of the small band marked with an asterisk (at 2276 cmK1) is uncertain,
but tentatively assigned to methylisocyanate nNCO antisymmetric stretching
(see text).

180
S. Jarmelo, R. Fausto / Journal of Molecular Structure 786 (2006) 175–181
this species by UV-photolysis of either ethanolamine of
acetamide isolated in matrices [27–30].
solid argon, which depend on the conformation of the reactant
species and develop on the S1 surface. The decarboxylation is
the preferred path starting from a cis carboxylic group
(conformers of A- and C-type), while the decarbonylation
path is preferred when the conformation of the carboxylic
group of the reactant molecule is trans. The first process is also
probably facilitated by the fact that in both A and C-type
conformers the carboxylic moiety is not involved in any
intramolecular hydrogen bond as proton donor, while the
decarbonylation might also be favoured in conformers B by
the participation of the carboxylic group, as proton donor, in
the relatively strong OHC/N intramolecular hydrogen bond,
which weakens the C–O bond. The relative population of the
cis relatively to the trans conformers in matrices also
contributes to the observed prevalence of decarboxylation
process over the decarbonylation process. The causes leading
to the more efficient decarboxylation of the C conformer
relatively to the A forms are not easy to figure out, but they
can be related with the fact that the C conformer has a weaker
C–Ca bond (also consistent with a lower nC–Ca stretching
frequency) than the A-type conformers (C conformer:
The mechanisms of the photolysis processes cannot be
unequivocally established on the basis of the experiments here
described. However the photochemical reactions shall involve
radical intermediates resulting from a-cleavages either of
the C–Ca bond or C–O bond, as found previously for most of
the carboxylic acids that undergo decarboxylation or decarbo-
nylation reactions upon UV irradiation [31–36]. This indicates
that the decomposition reactions shall take place mainly on the
excited state surfaces rather than in the vibrationally excited
ground state. In the case of acetic acid, whose excited states (S1
and T1) surfaces have been extensively studied in relation to
photodecomposition dynamics [37], the transition states for the
C–Ca bond cleavage (associated to the decarboxylation
reaction) in the S1 and T1 surfaces were found to have
geometries that can be correlated with different conformations
of the carboxylic moiety: cis for S1 and trans for T1 [37].
Assuming that in serine an analogous situation occurs, and
taken into consideration that the T1)S0 is a forbidden
transition, the decarboxylation reaction in this compound
shall then be strongly preferred when started from C- and
A-type conformers, as is supported by the observed different
rates of consumption of the three types of conformers of serine.
To test this hypothesis, the amount of CO2 produced was
d(C–Ca)Z1.530 A; n(C–Ca)Z735 cmK1; A conformers:
˚
d(C–Ca)Z1.524 A; n(C–Ca): 852 cmK1, average values [11]).
˚
4. Conclusion
estimated from the intensity of the band at ca. 2340 cmK1
,
Monomeric serine can be trapped in low temperature argon
matrices in different conformers, which can be classified in
three groups (A, B, C) accordingly to the main intramolecular
interaction they exhibit: A (OHA/N hydrogen bond), B
(OHC/N) and C (OHA/Oa). The OHC/N intramolecular
interaction found in B-type conformers is considerably
stronger than both the OHA/N and OHA/Oa hydrogen
bonds and leads to reduce their abundance relatively to the
other forms at high temperatures due to entropy effects.
However, in serine, contrarily to what was recently observed in
phenylalanine [8], the general characterization of the equili-
brium at 470 K is not qualitatively affected by this variable.
When submitted to UV irradiation (lO200 nm), serine
decomposes by two different channels in the S1 excited state
surface: decarboxylation, leading to production of CO2 and
ethanolamine and decarbonylation, with formation of CO, H2O
and acetamide. The two observed photoprocesses were found
to be dependent on the conformation assumed by the reactant
molecule, with A- and C-type conformers undergoing
decarboxylation and B-type conformers decarbonylation, a
result which is consistent with transition states’ geometries of
the carboxylic group for the two processes in the S1 surface
resembling the cis and trans arrangements, respectively.
using the calculated intensity for this mode (712 km molK1) as
normalizing factor, and compared with the amount of serine in
the A and C conformations that was consumed (in this case, the
decrease in the intensity of the 3580–3500 cmK1 spectral
region upon irradiation was considered to make the estimation,
together with the relevant calculated intensities and expected
abundances of the different A and C conformers; this region
contains absorptions due to the nOHA stretching mode of all A
conformers and to the nOHC stretching mode of conformers 1
and 7). The obtained values (in arbitrary units) are equal within
the experimental error (1.4!10K3 and 1.5!10K3, respect-
ively), strongly supporting the idea that the decarboxylation
pathway takes place in the S1 surface and corresponds
essentially to photolysis of serine in the C and A-type initial
conformations.
Very interestingly, analogous calculations made using the
CO band intensity (weighted by the corresponding calculated
intensity: 89 km molK1) and the relevant data accounting for
consumption of B-type conformers (the nOHC band in the
3250–3085 cmK1 region, due to all B-type conformers was
used in this case, together with the corresponding calculated
band intensities and conformer abundances), yield values (3!
10K4 vs. 7!10K4) which indicate that the amount of CO
produced is similar to the amount of serine in the B-type
conformers that were consumed during the photolysis
experiments. This indicates that the transition state for the a
C–O cleavage in serine’s S1 surface shall have a nearly trans
geometry of the carboxylic group, similar to that exhibited by
the B conformers.
Acknowledgements
The authors thank Drs Maciej Nowak and Leszek Lapinski
for their useful suggestions and comments. This work
ˆ
was financially supported by ‘Fundac¸a˜o para a Ciencia e a
Tecnologia’ (FCT-Projects POCTI/QUI/59019/2004 and POC-
TI/QUI/58937/2004), Lisbon, and GRICES (Project 00813).
S.J. also acknowledges the FCT Grant SFRH/BD/6696/2001.
Hence, the results point to the preference by two distinct
photolysis reaction channels of monomeric serine isolated in

S. Jarmelo, R. Fausto / Journal of Molecular Structure 786 (2006) 175–181
[15] C.T. Lee, W.T. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.
181
References
[16] J.W. Ochterski, Thermochemistry in Gaussian, white paper published
electronically: (http://Gaussian.com/g_whitepap/thermo.htm).
[17] N.L. Owen, N. Sheppard, Proc. Chem. Soc. London 1963, p. 264.
[18] J.E. Piercy, S.V. Subrahmanyan, J. Chem. Phys. 42 (1965) 1475.
[19] J. Bayley, A.M. North, Trans. Faraday Soc. 64 (1968) 1499.
[20] R. Fausto, J.J.C. Teixeira-Dias, J. Mol. Struct. (Theochem). 150 (1987)
381.
[1] I.D. Reva, A.M. Plokhotnichenko, S.G. Stepanian, A.Yu. Ivanov,
E.D. Radchenko, G.G. Sheina, Yu.P. Blagoi, Chem. Phys. Lett. 232
(1995) 141.
[2] S.G. Stepanian, I.D. Reva, E.D. Radchenko, M.T.S. Rosado,
M.L.T.S. Duarte, R. Fausto, L. Adamowicz, J. Phys. Chem. A 102
(1998) 1041.
[21] S. Jarmelo, R. Fausto, Phys. Chem. Chem. Phys. 4 (2002) 1555.
[3] D.G. Stepanian, I.D. Reva, E.D. Radchenko, L. Adamowicz, J. Phys.
Chem. A 102 (1998) 4623.
´
[4] A. Gomez-Zavaglia, R. Fausto, Vibrat. Spectrosc. 33 (2003) 105.
¨
[22] T. Svensson, B. Nelander, G. Karlstrom, Chem. Phys. 265 (2001) 323.
[23] H. Abe, H. Takeo, K.M.T. Yamada,, Chem. Phys. Lett. 311 (1999)
153.
[5] R. Ramaekers, J. Pajak, M. Rospenk, G. Maes, Spectrochim. Acta A 61
(2005) 1347.
[24] C. Cacela, M.L. Duarte, R. Fausto, J. Mol. Struct. 482/483 (1999) 591.
[25] Y. Hase, Spectrochim. Acta A 51 (1995) 2561.
[26] R. Fausto, I.D. Reva, L. Lapinski, Proceedings of the XX IUPAC
Symposium on Photochemistry, Granada, Spain, 2004.
[27] F. Duvernay, T. Chiavassa, F. Borget, J.-P. Aycard, J. Phys. Chem. A 109
(2005) 6008.
[6] S.G. Stepanian, I.D. Reva, E.D. Radchenko, L. Adamowicz, J. Phys.
Chem. A 105 (2001) 10664.
´
[7] A. Gomez-Zavaglia, I.D. Reva, R. Fausto, Phys. Chem. Chem. Phys. 5
(2003) 41.
[8] A. Kaczor, I.D. Reva, M. Proniewicz, R. Fausto, J. Phys. Chem. A,
(submitted).
[28] X.-B. Chen, W.-H. Fnag, D.-C. Fang, J. Am. Chem. Soc. 125 (2003)
9689.
[9] B. Lambie, R. Ramaekers, G. Maes, Spectrochim. Acta A 59 (2003) 1387.
[10] K. Iijima, K. Tanaka, S. Onuma, J. Mol. Struct. 246 (1991) 257.
[11] S. Jarmelo, L. Lapinski, M.J. Nowak, P.R. Carey, R. Fausto, J. Phys.
Chem. A 109 (2005) 5689.
¨ ¨
[29] J. Lundell, M. Krajewska, M. Rasanen, J. Mol. Struct. 448 (1998) 221.
¨ ¨
[30] J. Lundell, M. Krajewska, M. Rasanen, J. Phys. Chem. A 102 (1998)
6643.
[12] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb,
J.R. Cheeseman, V.G. Zakrzewski, J.A. Montgomery, Jr., R.E. Stratmann,
J.C. Burant, S. Dapprich, J.M. Millam, A.D. Daniels, K.N. Kudin,
M.C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi,
B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski,
G.A. Petersson, P.Y. Ayala, Q. Cui, K. Morokuma, D.K. Malick, A.D.
Rabuck, K. Raghavachari, J.B. Foresman, J. Cioslowski, J.V. Ortiz, A.
Baboul, B.B. Stefanov, A. Liu, G. Liashenko, P. Piskorz, I. Komaromi, R.
Gomperts, R.M. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng,
A. Nanayakkara, M. Challacombe, P.M.W. Gill, B.J. Johnson, W. Chen,
M.W. Wong, J.L. Andres, C. Gonzalez, M. Head-Gordon, E.S. Replogle,
J.A. Pople, (1998)GAUSSIAN 98, Revision A.9;Gaussian Inc.: Pittsburgh, PA.
[13] A.D. McLean, G.S. Chandler, J. Chem. Phys. 72 (1980) 5639.
[14] A.D. Becke, Phys. Rev. A 38 (1988) 3098.
[31] D. Tevault, M. Lin, M. Umstead, R. Smardzewski, Int. J. Chem. Kinet. 11
(1979) 445.
ˆ
[32] R. Fausto, E.M.S. Mac¸oas, A. Kulbida, J. Mol. Struct. 480/481 (1999)
83.
ˆ
¨ ¨
[33] E.M.S. Mac¸oas, L. Khriachtchev, R. Fausto, M. Rasanen, J. Phys. Chem.
A 108 (2004) 3380.
[34] S.S. Hunnicutt, L.D. Waits, J.A. Guest, J. Phys. Chem. 95 (1991) 562.
[35] D.R. Peterman, R.G. Daniel, R.J. Horwitz, J.A. Guest, Chem. Phys. Lett.
236 (1995) 564.
[36] P.D. Naik, H.P. Hupadhyaya, A. Kumar, A.V. Sapre, J.P. Mittal, Chem.
Phys. Lett. 340 (2001) 116.
[37] W.-H. Fang, R.-Z. Liu, X. Zheng, D. Phillips, J. Org. Chem. 67 (2002)
8407.