Matrix-isolation and computational study of salicylhydroxamic acid and its photochemical degradation

Article abstract of DOI:10.1039/b418033j

The IR absorption spectrum of salicylhydroxamic acid (sha) isolated in an argon matrix at 12-21 K has been recorded. Calculations of all the possible entgegen and zusammen keto conformers of sha have been performed at the B3LYP/6-311++G(d,p) level of theory. The computed energies, optimized structures, and relative abundances show that sha exists in the matrix as the z,z,z,z-conformer [relative to the C-O(-H), C-C(-N), C-N and N-O bonds, respectively] and is stabilized by two intramolecular H-bonds. Photofragmentation of the compound in matrix and methanol-water solution has been obtained by irradiation with visible and ultraviolet light. o-Hydroxyphenylisocyanate is the main photolysis product in an argon matrix, while salicylamide is the primary photolysis product in methanol-water solution. The Owner Societies 2005.

Full text of DOI:10.1039/b418033j

View Article Online / Journal Homepage / Table of Contents for this issue
R E S E A R C H P A P E R
Matrix-isolation and computational study of salicylhydroxamic
acid and its photochemical degradationw
Agnieszka Kaczor,a Jan Szczepanski,b Martin Valab and Leonard M. Proniewicz*ac
a
´
Faculty of Chemistry, Jagiellonian University, ul. Ingardena 3, 30-060 Krakow, Poland.
E-mail: proniewi@chemia.uj.edu.pl; Fax: þ48 12 634 05 15; Tel: þ48 12 633 63 77 int. 2253
b Department of Chemistry and Center for Chemical Physics, University of Florida, Gainesville,
FL 32611-7200, USA
c Regional Laboratory of Physicochemical Analysis and Structural Research,
´
Jagiellonian University, ul. Ingardena 3, 30-060 Krakow, Poland
Received 1st December 2004, Accepted 15th March 2005
First published as an Advance Article on the web 6th April 2005
The IR absorption spectrum of salicylhydroxamic acid (sha) isolated in an argon matrix at 12–21 K has been
recorded. Calculations of all the possible entgegen and zusammen keto conformers of sha have been performed at
the B3LYP/6-311þþG(d,p) level of theory. The computed energies, optimized structures, and relative abundances
show that sha exists in the matrix as the z,z,z,z-conformer [relative to the C–O(–H), C–C(–N), C–N and N–O
bonds, respectively] and is stabilized by two intramolecular H-bonds. Photofragmentation of the compound in
matrix and methanol–water solution has been obtained by irradiation with visible and ultraviolet light.
o-Hydroxyphenylisocyanate is the main photolysis product in an argon matrix, while salicylamide is the
primary photolysis product in methanol–water solution.
of three isomers.21 In the solid state, sha exists as z-keto
1
Introduction
isomers24 and, in DMSO solutions, as solvent-bound clusters
of the same conformation.25 In this work, the structure of sha
in low-temperature argon matrices has been determined by
IR spectroscopy and supported by density functional theory
calculations.
Compounds containing the hydroxamic group [C(QO)–N
(–H)–O–H)] act as chelating agents for metal ions.1 As compo-
nents of naturally-occurring siderophores, they participate in
microbial iron transport.2,3 As efficient inhibitors of some metal-
loproteinases,4–8 hydroxamic acids may be potential therapeutic
agents in the treatment of tumors and Alzheimer’s disease.
Introduction of a donor substituent (such as amine, oximic
or another hydroxamic) in proximity to a hydroxamic group
facilitates different types of chelation with metal ions.1,9–11
Salicylhydroxamic acid (sha), which contains a phenolic group
ortho to the hydroxamic group, is a derivative of benzhydroxa-
mic acid, a well-known bidentate ligand of metal ions,1,12
studied previously as an inhibitor of peroxidase.13 Solution
studies of simple metallo-complexes and binding of ligands to
myeloperoxidase have shown that sha complexes are quite
stable.5,12 Not only does sha produce tetranuclear metal coro-
nates,14 but it also acts as an efficient ligand in mixed ternary
complexes with ribonucleotides and metal(II) ions.15
Although there is growing interest in therapeutics based on
hydroxamic group-containing species, the structures of the
hydroxamic acids are still the matter of discussion. In the solid
state, the compounds usually form z-keto isomers (with respect
to the C–N bond),16 while in solution substituted hydroxamic
acids usually appear as iminol tautomers.17–20 Interestingly,
one of the simplest hydroxamic acids, acetohydroxamic acid,
has been shown via quantum-chemical calculations21–23 and
low-temperature matrix isolation studies to exist as a mixture
Photolysis of aromatic hydroxamic acid solutions produces
anilides or amides.26 Photodegradation commonly occurs via
the acylaminoxyl radical.26 The photochemical properties of
sha were studied here both in solution and in low-temperature
matrices.
2
Experimental
Salicylhydroxamic acid was synthesized following an estab-
lished literature procedure.27 Elemental analysis (found (calc.))
yielded: C: 54.60 (54.91); H: 4.69 (4.61); N: 8.88 (9.09%).
Fourier transform infrared (FT-IR) absorption spectra were
recorded on a Nicolet Magna 560 spectrometer in the range of
400–4000 cmꢀ1 or 700–4000 cmꢀ1 with a resolution of 1 cmꢀ1
.
Sample matrices were prepared by vaporizing the species from
a quartz oven, followed by trapping on a cryogenically cooled
CsI window with argon isolant gas (Matheson, 99.9995%
purity). Deposition temperatures in the range of 12–21 K were
obtained with a Displex closed cycle cryostat. Deposition times
varied from 70 to 135 min at an oven temperature of B75 1C.
A 100 W medium-pressure mercury lamp (in the presence or
absence of filters (450, 350 or 320 nm cut-off)) was used for
photolysis of matrix-isolated samples.
w Electronic supplementary information (ESI) available: Table S1:
Geometry (bond lengths in A, angles in 1) of the energetically most
stable sha structures. Table S2: Definition of internal coordinates used
in the normal mode analysis of sha (the numbering scheme is presented
in Fig. 1). Table S3: Assignments of experimental spectrum recorded
after photolysis of sha in an Ar matrix at 20 K. Table S4: Assignments
of experimental spectrum recorded after photolysis of sha in an Ar
matrix at 12 K. See http://www.rsc.org/suppdata/cp/b4/b418033j/
Photolysis of sha solutions (1.4 mg mlꢀ1 in 30% methanol–
water) was effected with 300 nm radiation from a Rayonet
reactor (equipped with eight RPR 3000 A lamps, each B21 W)
for 5, 30 and 60 min intervals. The reaction products were
monitored by mass spectrometry using a Finnigan MAT 95S
apparatus equipped with an electrospray (ESI) ionization
source.
1960
P h y s . C h e m . C h e m . P h y s . , 2 0 0 5 , 7 , 1 9 6 0 – 1 9 6 5
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 5

View Article Online
Table 1 The sum of electronic and zero-point energies (DEcorr) and
Gibbs free energies (DG) of salicylhydroxamic acid relative to the most
3
Computational methods
Sixteen sha isomers, with all possible entgegen and zusammen
conformations, were initially optimized using density func-
tional theory with the B3LYP28 functional and 6-31G(d) basis
set (using Gaussian 9829). After optimization, the sixteen
proposed isomers of sha yielded nine independent energy-
minimized geometries. Geometries were subsequently recom-
puted using a triple-zeta basis set with two sets of diffuse
functions and polarization functions [B3LYP/6-311þþG
(d,p)]. These optimized structures were then used to calculate
the normal mode harmonic frequencies [B3LYP/6-311þþG
(d,p)].
All structures of the matrix photoproducts were computed at
the same level of theory. The calculated harmonic frequencies
were scaled by a factor of 0.98 to account for anharmonicity
effects. Potential energy distributions (PEDs) of the normal
modes were computed in terms of natural internal coordi-
nates30 with the Gar2ped program.31
stable monomer and the relative abundance of the given isomer (%IS
)
Structurea
DEcorrb/kJ molꢀ1
DGc/kJ molꢀ1
%
IS
z,z,z,z (1)
e,e,z,z (2)
z,z,e,e/z (3)
z,z,e,e/z (4)
z,e,z,z (5)
e,e,e,e/z (6)
e,z,z,z (7)
z,e,e,e/z (8)
z,e,e,e (9)
a
0.0
16.2
18.8
18.8
21.8
36.8
38.6
39.6
51.9
0.0
15.0
16.4
17.3
22.0
36.9
36.3
39.5
51.3
99.52
0.24
0.13
0.09
0.01
0.00
0.00
0.00
0.00
b
B3LYP/6-311þþG(d,p), all structures in symmetry C1. Ecorr
¼
ꢀ551.375 847 Hartree moleculeꢀ1 for z,z,z,z-keto structure, the con-
version factor 1 Hartree moleculeꢀ1 ¼ 2625.5 kJ molꢀ1
.
G ¼
c
ꢀ551.410 529 Hartree moleculeꢀ1 for z,z,z,z-keto structure, the con-
version factor as for Ecorr
.
4
Results and discussion
Although the agreement between spectra is in general very
good, the agreement in the 3000–3500 cmꢀ1 range is poorer
than in the 700–1700 cmꢀ1 region (the same scaling factor was
used).
4.1 IR spectra of salicylhydroxamic acid in argon matrix
The vibrational spectrum of argon matrix-isolated sha in the
700–3600 cmꢀ1 range is presented in Fig. 1. The theoretical
spectrum of the most stable z,z,z,z isomer (1) is also given in
Fig. 1. Good agreement between calculated and experimental
frequencies can be seen. Table 1 gives the zero point-corrected
energies, Gibbs free energies, and relative abundances calcu-
lated for the nine conformers.
Although the frequency difference between the experimental
band at B1353 cmꢀ1 and its computed value (1377 cmꢀ1
)
seems large, this is in fact only a 2% difference. The band is
connected mainly with the phenolic C–O–H bend (22%). The
inability to reproduce exactly the frequency of the band is
probably a consequence of the formation of the six-membered
H-bond which is a resonance-assisted H-bond (RAHB).35–37
Previously the formation of the RAHB has been proposed for
salicylic acids and its analogues and the inadequacy of the
description of the geometrical parameters of the molecules has
been noticed in the case of HF, DFT and MP2 methods.38
The position of the hydroxyl O–H stretch in sha at 3341
cmꢀ1 provides strong support for H-bonding in this molecule.
Its lower frequency compared to the O–H stretch in mono-
meric (3635 cmꢀ1) and dimeric phenol (3555, 3490 cmꢀ1) in Ar
matrices39 is consistent with this supposition. Correlation of
the hydrogen bond strength in pseudo-ring structures and the
tOH frequency40 suggests that the strength of the sha six-
membered H-bond does not differ significantly from the
H-bond energy in salicylaldehyde, estimated as 43.5 kJ molꢀ1
in the gas phase.41
Optimized monomer geometries are given in Fig. 2, where
the letters e and z denote the conformation with respect to the
C–O(–H), C–C(QO,–N–), and hydroxamic C–N and N–O
bonds. E/z indicates a structure in which the conformation
around the N–O bond lies between entgegen and zusammen.
Corrected energy differences (in kJ molꢀ1) relative to the most
stable z,z,z,z-isomer are given in square brackets. The relative
conformer abundances were computed using DG ¼ RT ln Kc,34
where DG is the difference in Gibbs energies and Kc is the ratio
of the concentrations of two conformers. The relative abun-
dance of conformer 1 is B99.5%, so that the percentages of all
other conformers are negligible. The calculated frequencies
mimic the experimental spectra well. Since no new bands are
observed upon annealing (to 31 ꢁ 2 K) and the ratio of
annealed-to-unannealed bands (excluding monomeric water
bands and some superimposed or low-intensity bands) lie in
the 0.89–1.11 range, the conclusion that only a single confor-
mer is present in Ar matrices is justified. Additionally, no
particular changes were noted for the spectra recorded at
12 and 20 K.
The strong bands at 1650 and 1604 cmꢀ1 (theory: 1643, 1602
cmꢀ1) are assigned to the CQO stretch with contributions
from C–C ring stretching modes. The most intense band, at
1408 cmꢀ1 (theory: 1406 cmꢀ1), involves the in-plane N–O–H
Fig. 1 The experimental (lower) and calculated (upper) spectra of salicylhydroxamic acid in the range of 700–1850 (a) and 3000–3600 cmꢀ1 (b).
Calculations [B3LYP/6-311þþG(d,p)] are based on z,z,z,z-keto sha model. *Denotes water vibrations.32,33
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 5
P h y s . C h e m . C h e m . P h y s . , 2 0 0 5 , 7 , 1 9 6 0 – 1 9 6 5
1961

View Article Online
(O9ꢂ ꢂ ꢂH18 1.969 A, O9ꢂ ꢂ ꢂH18–O16 118.81), the weaker six-
membered bond is present (O10ꢂ ꢂ ꢂH17 1.989 A, O10ꢂ ꢂ ꢂH17–N1
122.91). The strength of a five-membered H-bond may be
estimated as B19 kJ molꢀ1 by comparison of structures 1
and 3.
Other conformers of sha may play important roles when
interacting with enzymes in solutions (or solids). For example,
hydrogen-bonding in the entgegen conformer with respect to
the C–N bond may be responsible for the strong inhibition of
human myeloperoxidase (MPO) activity by sha.43
4.3 Photolysis of salicylhydroxamic acid
Photolysis of sha was carried out in both room-temperature
solutions (30% methanol–water) and low-temperature argon
matrices. Solution photolysis (at 300 nm) was followed by
electrospray mass spectrometry which showed a decrease in the
sha molecular ion signal (m/z ¼ 154.8 ꢁ 1.0) after 30 min and
complete disappearance after 60 min irradiation. Simulta-
neously, a peak at m/z ¼ 138.9 ꢁ 1.0 was found to increase
and, after 30 min, became the predominant peak. Since the
photoreaction of hydroxamic acids in solution is known to
produce amides and since salicylamide has mass of 137.1, we
conclude that irradiation of sha produces salicylamide.
Irradiation in low-temperature matrices leads to products
different than those in solution. Photolysis of sha in matrices
was probed by IR spectroscopy. The IR absorption spectra of
Fig. 2 The geometries of the optimized conformers of salicylhydroxa-
mic acid. The values in square brackets indicate the difference of the
corrected energy relative to the most stable isomer (in kJ molꢀ1).
bend, phenolic C–O–H bend and C6–C8 stretch. The band at
1018 cmꢀ1 (theory: 1013 cmꢀ1) is primarily due to the N–O
stretch, although it is also coupled to the N–C stretch and
C–C–C ring bends. This band is characteristic of the hydro-
xamic moiety.
Other theoretical models predict neither the frequencies nor
the relative intensities of the experimental spectra correctly. In
particular, differences between calculated and experimental
values are most often found for groups involved in the
H-bonding. Thus, the CQO vibration frequency is strongly
overestimated (1687–1696 cmꢀ1) in structures 5–9. Although
the latter are well reproduced (1642–1647 cmꢀ1) for structures
2–4, significant differences are observed for the phenolic O–H
frequency in 2 (3759 cmꢀ1, exptl.: 3341 cmꢀ1) and the hydro-
xamic O–H frequencies in 3 and 4 (3693 and 3701 cmꢀ1
respectively, exptl.: 3463 cmꢀ1).
,
4.2 Geometry of salicylhydroxamic acid in argon matrix
The most stable conformer (1) adopts a zusammen (z) con-
formation around all bonds. Distorted from planarity, the
H17–N1–O16–H18 torsion angle is equal to 146.01. This confor-
mation enables the formation of two H-bonded pseudo-rings,
one five-membered (O9ꢂ ꢂ ꢂH18 1.955 A, O9ꢂ ꢂ ꢂH18–O16 118.61)
and the other six-membered (O9ꢂ ꢂ ꢂH15 1.740 A, O9ꢂ ꢂ ꢂH15–O10
145.71). These hydrogen bonds involve the carbonyl oxygen with
both the hydroxamic hydroxyl hydrogen and the phenolic
hydrogen. Despite the negative ‘‘cooperativity’’ of these
bonds,42 1 is stabilized with respect to 2 (and 3), in which two
(one) weaker H-bond is formed. This is consistent with the
general principle that H-bonding is stronger in pseudo-six-
membered rings than in pseudo-five-membered ones.40
The most important factor stabilizing the geometry of sha
seems to be the O9ꢂ ꢂ ꢂH15 hydrogen bond. Its absence renders
structures 6–9 less stable relative to 1. Comparison of struc-
tures 1 and 7 allows for a rough estimation of the strength of
the O9ꢂ ꢂ ꢂH15 H-bond (B39 kJ molꢀ1). The two H-bonds
involving the carbonyl oxygen atom stabilize 1 by 16.2 kJ
molꢀ1 relative to 2, where apart from the five-membered bond
Fig. 3 The experimental (lower) and calculated (upper) absorption
spectra in the range of 700–1850 cmꢀ1. Experimental spectrum (1) was
recorded before matrix photolysis, while spectrum (2) was scanned
after 70 min photolysis (with filter transparent for l 4 320 nm
photons) and spectrum (3) after 79 min photolysis (48 and 21 min
with filter transparent for l 4 350 nm and l 4 320 nm photons
without filter), respectively. Spectrum (4) was obtained after 190 min
photolysis without filter. Calculations [B3LYP/6-311þþG(d,p)] are
based on structures A and B. *Denotes water vibrations.32,33
1962
P h y s . C h e m . C h e m . P h y s . , 2 0 0 5 , 7 , 1 9 6 0 – 1 9 6 5
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 5

View Article Online
Fig. 4 The experimental (lower) and calculated (upper) absorption spectra in the range of 2050–2500 (a) and 3350–3800 cmꢀ1 (b). Experimental
spectrum (1) was recorded before matrix photolysis, while spectrum (2) was scanned after 70 min photolysis (with filter transparent for l 4 320 nm
photons) and spectrum (3) after 79 min photolysis (48 and 21 min with filter transparent for l 4 350 nm and l 4 320 nm photons without filter),
respectively. Spectrum (4) was obtained after 190 min photolysis without filter. Calculations [B3LYP/6-311þþG(d,p)] are based on structures A
and B. *Denotes water vibrations.32,33
sha irradiated at wavelengths o320 nm and additionally, at
wavelengths o230 nm are given in Figs. 3 and 4.
Second, Pritchina et al. showed that photolysis of benzoyl
azide (at 254 nm) in low-temperature argon matrices resulted in
a Curtius rearrangement, elimination of the nitrogen molecule,
and formation of phenylisocyanate and its isomer containing a
three-membered CNO ring attached to a phenyl ring.45 The
latter rearranges to phenylisocyanate upon further exposure (at
313 nm).45 The Curtius rearrangement leading from azides to
The most striking feature of the experimental spectrum is the
appearance of very intense bands in the 2240–2280 cmꢀ1 range.
Strong IR peaks in this range are characteristic of species
containing the isocyanate group.44 Bands in the same range,
split due to Fermi resonance, were previously observed for
phenylisocyanate in an Ar matrix.45 In addition, bands asso-
ciated with monomeric water,32,33 carbon monoxide, carbon
dioxide, and the carbon monoxide–water complex46 were ob-
served at longer exposure times.
isocyanates is analogous to the Lossen rearrangement in which
¨
hydroxamic acids are converted to isocyanates. Therefore, a
parallel intermediate is expected here. The differences in zero-
point corrected energies of the products also suggest that B is
an intermediate of the reaction (and is destabilized by 318.4 kJ
molꢀ1 relative to A).
The theoretical frequencies and intensities of B were com-
puted and are shown in Table 2. It is immediately apparent
that the observed band at 1798 cmꢀ1 is in a region where no
bands due to A appear. However, a reasonably strong band is
predicted at 1750 cmꢀ1 (320 km molꢀ1) for B. A number of
other bands may also be assigned to B. We therefore conclude
that photolysis of sha in a cryogenic matrix forms both the
intermediate B and the o-hydroxyphenylisocyanate (A).
In addition to observed bands for structures A and B, two
groups of unassigned bands appeared in the spectra. In the first
group, bands at 1094, 1295, 1512, 2313, 2290 and 3413 cmꢀ1
decreased their intensities when comparing spectrum (3) with
Figs. 3 and 4 also show the calculated spectra for the
proposed species, o-hydroxyphenylisocyanate (A) and struc-
ture B produced after photolysis of sha in an argon matrix. The
correspondence between the observed and calculated frequen-
cies is shown in Table 2.
As Table 2 indicates, a number of medium-to-strong bands
can not be assigned to A. We therefore investigated whether
they might belong to structure B, which is in between a nitrene
and an oxazirene. This suggestion is based on the following
reasoning. First, the photolysis of N-phenylbenzenehydroxa-
mic acid has been shown to yield benzanilide and benzoic
acid.26 The reaction proceeds through an oxaziridine inter-
mediate, which was, however, not isolated in room-tempera-
ture solutions, presumably because of its low stability.26
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 5
P h y s . C h e m . C h e m . P h y s . , 2 0 0 5 , 7 , 1 9 6 0 – 1 9 6 5
1963

View Article Online
Table 2 Assignments of experimental spectrum recorded after photo-
lysis of salicylhydroxamic acid in an Ar matrix at 12 K. Harmonic
frequencies (in cmꢀ1) and integral intensities (in km molꢀ1) are
calculated at B3LYP/6-311þþG(d,p) level. Relative intensities are
given in parenthesis
n
exptl/cmꢀ1
ncalcabc/cmꢀ1
Calc. intensity/
km molꢀ1
Species
723 (0.19)d
754 (0.47)d
749 (0.16)d
836 (0.14)
997 (0.10)
737 (0.19)
743 (0.54)
757 (0.18)
27
77
56
11
17
17
22
17
75
141
149
48
88
47
47
27
30
16
20
40
89
117
54
56
73
320
1707
8
A
A
B
A
B
A
A
A
B
A
B
B
A
A
B
A
B
A
A
B
B
A
B
A
B
B
A
A
A
A
833 (0.08)
1028 (0.05)
1029 (0.12)
1076 (0.16)
1137 (0.12)
1137 (0.23)
1199 (1.00)
1224 (0.46)
1251 (0.15)
1252 (0.63)
1296 (0.33)
1326 (0.05)
1344 (0.20)
1353 (0.09)
1445 (0.12)
1468 (0.14)
1473 (0.12)
1482 (0.28)
1517 (0.83)
1582 (0.17)
1595 (0.40)
1624 (0.23)
1750 (1.00)
2307 (12.10)
3115 (0.05)
3130 (0.07)
3691 (0.69)
1082 (0.26)
1151 (0.15)
1203 (1.00)
1210 (0.18)d
1254 (0.06)d
1260 (0.26)d
1304 (0.55)d
1326 (0.15)
1344 (0.19)
1349 (0.03)d
1428 (0.27)
1456 (0.08)
1477 (0.12)d
1482 (0.29)d
1527 (0.89)d
1589 (0.07)d
1589 (1.30)d
1617 (0.17)d
1789 (0.98)
2258 (17.49)d
3214 (0.01)
3228 (0.11)
3383 (2.61)
a
Fig. 5 Proposed scheme of the photolysis of salicylhydroxamic acid in
the low-temperature matrix and in the solution.
involves the migration of an alkyl group from a carbon atom to
an electron-deficient nitrogen atom.47 Ketooximes (Beckman),
azides (Curtius) and hydroxamic acids (Lossen) are the groups
¨
which undergo rearrangement. An isocyanate (which hydro-
lyses to an amine in aqueous solution) is produced as a result of
a [1,2] shift of a nitrene, which is an intermediate of the
reaction.47 The structure of the intermediate in the benzoyl
azide study (and the intermediate product proposed here) is
analogous to the oxaziridines formed on irradiation of
N-substituted hydroxamic acids (N-phenylbenzenehydroxamic
and N-methyl-2-naphthalenecarbohydroxamic acids).26
9
97
Conclusions
b
The IR spectrum of sha obtained in a low-temperature argon
matrix has been assigned to the z,z,z,z-keto monomer on the
basis of B3LYP/6-311þþG(d,p) normal mode calculations.
The most stable structure, the z,z,z,z-keto monomer, is stabi-
lized by two intramolecular H-bonds which form five- and six-
membered pseudo-rings. The former links the carbonyl oxygen
with the proton of the hydroxamic OH group and is also found
in the most stable structures of oxalodihydroxamic acid and
2-(hydroxy)iminopropanohydroxamic acid.48 The carbonyl
oxygen atom is also involved in a stronger H-bond with the
proton of the phenolic group. This is possible due to the
presence of the phenolic group ortho to the hydroxamic moiety.
The energy of the H-bond in the six-membered pseudo-ring
Frequencies scaled uniformly by factor of 0.98. In A and B struc-
ture the intensities are relative to the 1199 and 1750 cmꢀ1 experimental
c
band integral intensity, respectively. The theoretical bands of relative
d
intensity Z 0.05 were included, Intensity is not reliable due to over-
lapped bands.
(2) in Figs. 3 and 4, while bands at 726, 1135, 1166, 1546, 1758,
1767 and 3636 cmꢀ1 increased in intensity by ca. 90%. More-
over, matrix UV photolysis (without filter) for 190 min gen-
erates more new bands in Figs. 3 and 4, spectrum (4). This
indicates that more photoreaction pathways are triggered by
the 320 nm 4 l 4 230 nm photons.
bond was estimated as 39 kJ molꢀ1
.
The proposed mechanism of photodegradation for the first
stage of photolysis is presented in Fig. 5.
Ultraviolet photolysis of sha in Ar matrices leads to o-hy-
droxyphenylisocyanate as the main reaction product. In addi-
tion, an intermediate compound, whose structure lies between
a nitrene and an oxazirene, has also been identified. Hydro-
In this scheme, hydrogen elimination from the hydroxamic
nitrogen of sha leads to the radical D. This is followed by
detachment of a similarly bonded hydroxyl group. This loss of
water leads to formation of a compound B whose structure is
intermediate between a nitrene and an oxazirene. In low-
temperature matrices, this structure may rearrange to o-hydroxy-
phenylisocyanate (A). In room-temperature solutions, either or
both of two pathways may be followed: (i) A rearranges while
picking up two hydrogens from the solvent, and forms salicyla-
mide (C). (ii) B adds two hydrogens and converts directly to C
(cf. Fig. 5). (A proton is presumed to be added subsequently in
the electrospray step leading to the protonated cation of sali-
cylamide, which is detected via mass spectrometry).
xamic acids are known to undergo the Lossen rearrangement
¨
to yield isocyanates.49 Although isocyanates are unstable in
solution, immediately hydrolyzing to amines,47 they may be
stabilized in low-temperature matrices. The intermediate was
identified on the basis of its IR absorption band pattern. An
analogous intermediate was seen previously upon irradiation
of benzoyl azide in argon matrices which yielded phenyliso-
cyanate via a Curtius rearrangement.45
The solution photolysis of hydroxamic acids generally leads
to the corresponding amides or anilides.26 In a similar fashion,
salicylamide was shown to be the sole photolysis product of sha
in aqueous methanol solutions. Reaction in solutions and
matrices is thought to occur via the same intermediate, identi-
fied in the matrix as the structure in between a nitrene and an
oxazirene.
There is precedent for the proposed mechanism from pre-
vious studies on similar molecules.45
Our proposed mechanism is similar to one proposed pre-
viously by Lossen, and also by Curtius and Beckman, and
¨
1964
P h y s . C h e m . C h e m . P h y s . , 2 0 0 5 , 7 , 1 9 6 0 – 1 9 6 5
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 5

View Article Online
23 J. El Yazal and Y. P. Pang, J. Phys. Chem. A, 1999, 103, 8346.
24 I. K. Larsen, Acta Crystallogr., Sect. B, 1978, 34, 962.
25 A. Kaczor and L. M. Proniewicz, J. Mol. Struct. (THEOCHEM),
2003, 640, 133.
Acknowledgements
The authors thank Mr Z. Urbanek for the synthesis of sha, and
Prof. M. Nowakowska and Dr Sz. Zapotoczny for the solution
UV irradiation of sha. Calculations were done at the Academic
Computer Centre ‘‘Cyfronet’’, Krakow, Poland, and the Inter-
disciplinary Centre for Mathematical and Computational
Modelling, Warsaw University, Warsaw, Poland, which are
acknowledged for computing time. Part of this research was
supported by the grant 4T09A064 22 from the Polish State
Committee for Scientific Research (to L. M. P.).
26 E. Lipczynska-Kochany, Chem. Rev., 1991, 91, 477.
´
27 W. Ruiqin, Y. Xiaolan, F. Zhenye, M. M. I. Haq, M. M. P.
Khurshid, T. J. Rayner, J. A. S. Smith and M. H. Palmer, J. Am.
Chem. Soc., 1989, 111, 114.
28 A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
29 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, N. Rega, P. Salvador, J. J. Dannenberg,
D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J.
Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin,
D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayak-
kara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M.
W. Wong, J. L. Andres, C. Gonzalez, M. Head-Gordon, E. S.
Replogle and J. A. Pople, GAUSSIAN 98 (Revision A. 11.4),
Gaussian Inc., Pittsburgh, PA, 2002.
30 P. Pulay, G. Fogarasi, F. Pang and J. E. Boggs, J. Am. Chem.
Soc., 1979, 101, 2550.
31 J. M. L. Martin and C. Van Alsenoy, Gar2ped, University of
Antwerp, 1995.
32 R. M. Bentwood, A. J. Barnes and W. J. Orville-Thomas, J. Mol.
Spectrosc., 1980, 84, 391.
33 W. B. Person, J. E. Del Bene, W. Szajda, K. Szczepaniak and
M. Szczesniak, J. Phys. Chem., 1991, 95, 2770.
34 H. Meghezzi and A. Boucekkine, J. Mol. Struct. (THEOCHEM),
1992, 257, 175.
35 G. Gilli, F. Bellucci, V. Ferretti and V. Bertolasi, J. Am. Chem.
Soc., 1989, 111, 1023.
36 V. Bertolasi, P. Gilli, V. Ferretti and G. Gilli, J. Am. Chem. Soc.,
1991, 113, 4917.
37 P. Gilli, V. Bertolasi, V. Ferretti and G. Gilli, J. Am. Chem. Soc.,
1994, 116, 909.
38 Y. Kwon, J. Mol. Struct. (THEOCHEM), 2000, 232, 227.
39 A. M. Plokhotnichenko, E. D. Radchenko, Yu. P. Blago and
V. A. Karachevtsev, Low Temp. Phys., 2001, 27, 666.
40 H. Rostkowska, M. J. Nowak, L. Lapinski and L. Adamowicz,
Phys. Chem. Chem. Phys., 2001, 3, 3012.
41 A. Modelli, F. Scagnolari and G. Distefano, Chem. Phys., 1999,
250, 311.
42 S. Scheiner, Hydrogen Bonding, A Theoretical Perspective, Oxford
University Press, New York, 1997.
References
1
B. Kurzak, H. Koz"owski and E. Farkas, Coord. Chem. Rev.,
1992, 114, 169.
M. J. Miller, Chem. Rev., 1989, 89, 1563.
J. B. Neilands, J. Biol. Chem., 1995, 270, 26723.
A. J. Stemmler, J. W. Kampf, M. L. Kirk and V. L. Pecoraro,
J. Am. Chem. Soc., 1995, 117, 6368.
2
3
4
5
6
7
M. Ikeda-Saito, D. A. Shelley, L. Lu, K. S. Booth, W. S. Caughey
and S. Kimura, J. Biol. Chem., 1991, 266, 3611.
S. Parvathy, I. Hussain, E. H. Karran, A. J. Turner and N. M.
Hooper, Biochemistry, 1998, 37, 1680.
R. D. Groneberg, C. J. Burns, M. M. Morrissette, J. W. Ullrich,
R. L. Morris, S. Darnbrough, S. W. Djuric, S. M. Condon, G. M.
McGeehan, R. Labaudiniere, K. Neuenschwander, A. C. Scotese
and J. A. Kline, J. Med. Chem., 1999, 42, 541.
8
9
P. J. Hajduk, G. Sheppard, D. G. Nettesheim, E. T. Olejniczak, S.
B. Shuker, R. P. Meadows, D. H. Steinman, G. M. Carrera, Jr., P.
A. Marcotte, J. Severin, K. Walter, H. Smith, E. Gubbins, R.
Simmer, T. F. Holzman, D. W. Morgan, S. K. Davidsen, J. B.
Summers and S. W. Fesik, J. Am. Chem. Soc., 1997, 119, 5818.
B. Kurzak, L. Nakonieczna, G. Rusek, H. Koz"owski and E.
Farkas, J. Coord. Chem., 1993, 38, 17.
10 D. A. Brown, R. Geraty, J. D. Glennon and N. N. Choileain,
Inorg. Chem., 1986, 25, 3792.
11 A. Dobosz, N. M. Dudarenko, I. O. Fritsky, T. G"owiak, A.
Karaczyn, H. Koz"owski, T. Yu. Sliva and J. Swia˛tek-Koz"owska,
J. Chem. Soc., Dalton Trans., 1999, 743.
12 E. C. O’Brien, E. Farkas, M. J. Gil, D. Fitzgerald, A. Castineras
and K. O. Nolan, J. Inorg. Biochem., 2000, 79, 47.
13 H. Itakura, Y. Oda and K. Fukuyama, FEBS Lett., 1997, 412, 107.
14 R. W. Saalfrank, S. Trummer, U. Reimann, M. M. Chowdhhry, F.
Hampel and O. Waldmann, Angew. Chem., Int. Ed., 2000, 39, 3492.
15 M. M. Khalil, J. Chem. Eng. Data, 2000, 45, 70.
16 I. K. Larsen, Acta Crystallogr., Sect. B, 1988, 44, 527.
17 J. Schraml, M. Kvıcalova, V. Blechta, L. Soukupova, O. Exner,
´ ´ ´
´
43 C. A. Davey and R. E. Fenna, Biochemistry, 1996, 35, 10967.
44 The Aldrich Library of FT-IR Spectra Edition II, Sigma–Aldrich
Co., USA, 1997.
45 E. A. Pritchina, N. P. Gritsan, A. Maltsev, T. Bally, T. Autrey, Y.
Liu, Y. Wang and J. P. Toscano, Phys. Chem. Chem. Phys., 2003,
5, 1010.
H.-M. Boldhaus, F. Erdt and C. Bliefert, Magn. Reson. Chem.,
2000, 38, 795.
18 J. Schraml, M. Kvı
Reson. Chem., 1999, 37, 427.
19 J. Schraml, M. Kvıcalova
O. Exner, J. Organomet. Chem., 2000, 597, 200.
20 J. Schraml, M. Kvıcalova L. Soukupova
O. Exner, J. Phys. Org. Chem., 1999, 12, 668.
calova, L. Soukupova and V. Blechta, Magn.
´ ´ ´
´
´
,
L. Soukupova
´
,
V. Blechta and
46 H. Abe, H. Takeo and K. M. T. Yamada, Chem. Phys. Lett., 1999,
311, 153.
´
´
,
´
, V. Blechta and
47 http://www.ch.ic.ac.uk/widdowson/teach_files/nitrogen/dw1.html.
48 A. Kaczor, J. Szczepanski, M. Vala, H. Koz"owski and L. M.
Proniewicz, Phys. Chem. Chem. Phys., 2003, 5, 2337.
49 L. Bauer and O. Exner, Angew. Chem., Int. Ed. Engl., 1974, 13,
376.
21 A. Kaczor, J. Szczepanski, M. Vala and L. M. Proniewicz, The
2nd International Conference on Advanced Vibrational Spectro-
scopy, Nottingham, UK, 24–29 August 2003, Abstract Book, 227.
22 C. Munoz-Caro, A. Nino, M. L. Senent, J. M. Leal and S. Ibeas,
J. Org. Chem., 2000, 65, 405.
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 5
P h y s . C h e m . C h e m . P h y s . , 2 0 0 5 , 7 , 1 9 6 0 – 1 9 6 5
1965