Conformational and structural studies of N-methylacetohydroxamic acid and of its mono- and bis-chelated uranium(VI) complexes

Article abstract of DOI:10.1016/j.jinorgbio.2015.06.002

The thermodynamics and kinetics of the cis/trans isomerism of N-methylacetohydroxamic acid (NMAH) and its conjugated base (NMA^-) have been reinvestigated in aqueous media by ¹H NMR spectroscopy. Hindered rotation around the central C

Full text of DOI:10.1016/j.jinorgbio.2015.06.002

JIB-09732; No of Pages 12
Journal of Inorganic Biochemistry xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
Conformational and structural studies of N-methylacetohydroxamic acid and of its
mono- and bis-chelated uranium(VI) complexes
Stéphane Brandès a, Alejandra Sornosa-Ten a, Yoann Rousselin a, Mickael Lagrelette a, Christine Stern a,
Aurélien Moncomble b, Jean-Paul Cornard b, Michel Meyer a,
⁎
a
Institut de Chimie Moléculaire de l'Université de Bourgogne (ICMUB), UMR 6302, CNRS, Université de Bourgogne Franche-Comté, 9 avenue Alain Savary, BP 47870, 21078 Dijon Cedex, France
Laboratoire de Spectrochimie Infrarouge et Raman (LASIR), UMR 8516, CNRS, Université de Lille – Sciences et Technologies, 59655 Villeneuve d'Ascq Cedex, France
b
a r t i c l e i n f o
a b s t r a c t
Article history:
The thermodynamics and kinetics of the cis/trans isomerism of N-methylacetohydroxamic acid (NMAH) and its
conjugated base (NMA−) have been reinvestigated in aqueous media by 1H NMR spectroscopy. Hindered
rotation around the central C–N bond due to electronic delocalization becomes slow enough on the NMR time
scale to observe both rotamers in equilibrium in D2O at room temperature. By properly assigning the methyl
Received 17 February 2015
Received in revised form 29 May 2015
Accepted 1 June 2015
Available online xxxx
group resonances, evidence for the prevalence of the E over the Z form is unambiguously provided [K300
=
[E] / [Z] = 2.86(2) and 9.63(5) for NMAH and NMA−, respectively], closing thereby a long-lasting dispute
about the most stable conformer. To that end, calculations of the chemical shifts by density functional theory
(DFT), which accurately reproduced the experimental data, turned out to be a much more reliable method
than the direct computation of the relative energy for each conformer. The Z ⇌ E interconversion dynamics
was probed at 300 K in D2O by 2D exchange-correlated spectroscopy (EXSY), affording the associated rate
constants [kZE = 9.0(2) s−1 and kEZ = 3.14(5) s−1 for NMAH, kZE = 0.96(3) s−1 and kEZ = 0.10(2) s−1 for
NMA−] and activation barriers at 300 K [ΔG≠ = 68.0 kJ mol−1 and ΔG≠ = 70.6 kJ mol−1 for NMAH,
Keywords:
Hydroxamic acid
Uranium(VI) complexes
EXSY spectroscopy
Raman spectroscopy
X-ray crystal structures
ZE
EZ
ΔG≠ = 73.6 kJ mol−1 and ΔG≠ = 79.2 kJ mol−1 for NMA−]. For the first time, mono- and bis-chelated
ZE
EZ
uranium(VI) complexes of NMA− have been isolated. Crystals of [UO2(NMA)(NO3)(H2O)2] and
[UO2(NMA)2(H2O)] have been characterized by X-ray diffractometry, infrared and Raman spectroscopies.
© 2015 Elsevier Inc. All rights reserved.
1. Introduction
implemented in advanced PUREX flow-sheets, especially to
control the oxidation state of neptunium(IV) and to avoid thereby its
Among the plethora of bidentate O,O′-functional groups, the
naturally-occurring hydroxamic acids of general RCC(_O)N(OH)RN
formula occupy a special place in bioinorganic, medicinal, and coordina-
tion chemistry [1–3]. This bidentate chelating unit is ubiquitous among
siderophores, the so-called biological iron(III) transport agents
produced by bacteria and yeasts in order to acquire this essential nutri-
ent and growth factor [4–7]. Besides, numerous synthetic hydroxamic
acids which display a wide range of biological activities, mostly related
to enzyme inhibition processes, have been described [8–10]. Moreover,
these versatile and outstanding ligands enabled to isolate intricate
supramolecular architectures [11,12], including metallocrowns
behaving as single-molecule magnets [13–16]. Last but not least,
hydroxamic acids have found different technologically-oriented
applications in catalyzed asymmetric synthesis [17], as analytical [3,
18–20], anticorrosion [21], flotation [22], and extracting agents in nucle-
ar fuel reprocessing by liquid–liquid partitioning [23,24]. Formo- (RC =
RN = H) and acetohydroxamic acids (RC = Me, RN = H) have been
co-extraction with plutonium(IV) into the organic phase [25,26].
Acetohydroxamic acid selectively reduces NpO22+ and Pu4+ but has no
redox activity in the presence of UO22+, thus enabling the recovery of
uranium(VI) in the organic phase.
Structurally, these weak acids, characterized by pKa values in the
range of 8–9 in aqueous media, resemble to amides with the N–H
substituted by an N–OH ionizable group and thus can be viewed as
the direct oxidation product of them. Actually, the link between
amides and hydroxamic acids goes beyond this simple analogy:
molybdenum(V)-mediated oxidation of the former has been
proposed as a preparative route of the latter [27], while in turn
palladium-catalyzed hydrogenolysis of benzyl or trityl-protected cyclic
hydroxamic acids was shown to produce in some amounts the
corresponding amide by reduction of the N–OH group [11,28]. Another
important feature shared by both classes of compound is the participa-
tion of several canonical forms in the electron delocalization scheme,
which are responsible for a partial C–N double-bond character. As a
matter of fact, the iminium resonance forms (primed structures in
Scheme 1) further account for the restricted rotation around the central
bond, which can become slow enough on the NMR time scale to allow
⁎
Corresponding author.
E-mail address: michel.meyer@u-bourgogne.fr (M. Meyer).
http://dx.doi.org/10.1016/j.jinorgbio.2015.06.002
0162-0134/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: S. Brandès, et al., Conformational and structural studies of N-methylacetohydroxamic acid and of its mono- and bis-
chelated uranium(VI) complexes, J. Inorg. Biochem. (2015), http://dx.doi.org/10.1016/j.jinorgbio.2015.06.002

2
S. Brandès et al. / Journal of Inorganic Biochemistry xxx (2015) xxx–xxx
(formation of a hydroximate dianion) acids [30,37,39,47–55]. Of course,
for each protonation state or keto/iminol form, the molecule can adopt
either an E or Z conformation.
According to this brief overview, the conformational properties of
hydroxamic acids are still a matter of debates in spite of numerous
efforts undertaken for more than three decades. Probed in silico by
means of theoretical calculations [30,34,37,39–41,43,47–60] and in
solution by NMR spectroscopy [29–32,34–41,43,44,47], conflicting
results appeared over the years in the literature mainly because of
computational uncertainties that often exceed the actual energy differ-
ence between both isomers once solvation is considered and contradic-
tory assignments of NMR resonances, respectively. The present work is
intended to take a new look on this long-lasting debate by focusing our
attention on N-methylacetohydroxamic acid (NMAH, RC = RN = Me)
[32,34–36,43,44,47,51,53,55,56], the closest structural and electronic
model for the binding units incorporated in many siderophores, includ-
ing the widespread desferrioxamine B. Moreover by combining 1D and
2D exchange-correlated spectroscopy (EXSY) 1H NMR measurements,
the scope of the equilibrium and kinetic investigations pertaining to
the cis/trans interconversion process is extended for the first time also
to the conjugated base NMA−. Appearance of a negative charge on
one oxygen atom is shown to strongly affect the E/Z concentration
ratio as well as to significantly slow down the rotation rates around
the C–N bond. As a part of our ongoing actinide chelation studies with
hydroxamic siderophores and abiotic receptors [28] in relation to envi-
ronmental migration issues of radionuclides in contaminated soils, we
also disclose herein the first crystal structures for mono- and bis-
chelated uranium(VI) complexes of NMA−. Their structural features
are discussed in light of those reported in the literature, albeit only
five other uranyl/hydroxamato complexes have been characterized so
far by X-ray diffraction techniques, all of them involving exclusively
aryl-substituted ligands, N-phenylbenzohydroxamate [61–64] or a
more encumbered analog of it [RN = Ph, RC = p-(i-Pr)C6H4] [65].
Scheme 1. Isomerization equilibria and resonance formulae between the E and Z
conformers of NMAH.
the observation of both cis (Z) and trans (E) conformers at room temper-
ature. Actually, in the E form both oxygen atoms are misoriented for
coordinating a metal cation in a bidentate fashion. To achieve the forma-
tion of a chelate cycle around a metal center, one oxygen atom has to
rotate around the C–N bond to bring both of them in close proximity
as found in the Z conformer, which is predisposed for binding. Thus,
the knowledge of the E/Z ratio and of the interconversion kinetics of
hydroxamic acids is an important issue for understanding their metal–
ion binding ability.
The relative stability of both hydroxamic acid rotamers depends on
an intricate manner on the nature of the substituents borne by the
carbonyl carbon (RC) and nitrogen (RN) atoms, the solvent, the concen-
tration, and the temperature [29–44]. As stressed earlier, strong
electron-donating groups at the RN position are intended to stabilize
the partial positive charge located on the nitrogen atom in the iminium
form [35,36,45]. Under such circumstances, the iminium resonance
forms E′ and Z′ gain in weight, enhancing thereby the electron density
on the carbonyl oxygen atom. Its increased basicity translates into
more stable but also more inert metal complexes [2,46]. The E/Z ratio
is also highly solvent dependent. Some authors proposed that the cis
or Z isomer of monoalkylhydroxamic acids (RN = H) is stabilized in
polar solvents (e.g., H2O, DMSO) owing to hydrogen bonding interac-
tions with residual water or the solvent itself [31–33,37,41]. In contrast,
N-substituted hydroxamic acids should favor the intramolecularly
hydrogen-bonded Z form in nonpolar media [30], but it has also
been claimed that this conformer should rather prevail in polar solvents
[38]. Further complications might arise in less-polar solvents owing
to aggregation effects. Intermolecular hydrogen-bonding was suggested
to favor at higher concentrations the E or trans conformation of
various aliphatic hydroxamic acids (RN = Me, RC = alkyl) in CDCl3,
CD2Cl2, C6D6, and toluene-d8 [32]. However, dimerization of Z or cis N-
phenylbenzohydroxamic acid (PBAH, RN = RC = Ph) has been
evidenced at −85 °C in acetone [38]. This hydrogen-bonded dimer
was found to be in equilibrium with the E monomer (~80%), the
proportion of the latter decreasing in more polar solvents (methanol
or water). Further low-temperature NMR studies by the same authors
on benzohydroxamic acid (RN = H, RC = Ph) revealed the predomi-
nance of the E over the Z conformation in a 3:1 ratio in acetone [39,
40], in contrast to other works dealing with N-arylhydroxamic acids
that claimed the prevalence of the cis rotamer in solution [30,33,47].
Moreover, concentration-dependent studies highlighted the formation
of both E–E and Z–Z dimers, but also of 1:1 E–Z aggregates, whose
amounts were influenced by the water content [39,40]. Further compli-
cations arise for primary hydroxamic acids (RN = H) as the latter are in
equilibrium with the tautomeric RCC(OH) = N(OH) iminol form, while
these compounds can undergo two consecutive deprotonation steps,
acting both as N–OH (formation of a hydroxamate anion) and N–H
2. Experimental
2.1. Safety note
Uranium (primary isotope 238U) is a weak α-emitter (4.197 MeV)
with a half-life of 4.47 × 109 years. All sample preparations were carried
out in monitored fume hoods, in a laboratory equipped with α- and β-
counting equipment.
2.2. General
Unless otherwise noted, all solvents and analytical-grade chemicals
were obtained from commercial suppliers and used without further pu-
rification. O-Benzylhydroxylamine [66], N-(tert-butyloxy)-O-benzyl hy-
droxylamine, and N-methyl-O-benzylhydroxylamine were synthesized
following described procedures [67–69]. Aqueous solutions were
prepared with double-deionized high-purity water (18.2 MΩ cm)
obtained from a Maxima (USF Elga) cartridge system designed for
trace analysis. Apparent pH* values in D2O were measured with a
PHM 210 (Radiometer) ionometer fitted with a semi-micro pHc 3006
(Radiometer) Ag/AgCl combined glass electrode filled with a saturated
NaCl solution. The cell was calibrated with commercial aqueous buffers
+
(pH = −log aH = 4, 7, and 10) purchased from Acros. pD values
+
(pD = −log aD ) were estimated according to the formula pD =
pH* + 0.4 [70].
All physical measurements were performed at the “Plateforme
d'Analyses Chimiques et de Synthèse Moléculaire de l'Université de
Bourgogne — Pôle Chimie Moléculaire”, the technological platform for
chemical analysis and molecular synthesis (http://www.wpcm.fr) of
the Institut de Chimie Moléculaire de l'Université de Bourgogne.
1H and 13C{1H} NMR spectra were recorded at 300 K on Bruker
spectro-meters operating either at 300 MHz (Avance III NanoBay) or
Please cite this article as: S. Brandès, et al., Conformational and structural studies of N-methylacetohydroxamic acid and of its mono- and bis-
chelated uranium(VI) complexes, J. Inorg. Biochem. (2015), http://dx.doi.org/10.1016/j.jinorgbio.2015.06.002

S. Brandès et al. / Journal of Inorganic Biochemistry xxx (2015) xxx–xxx
3
500 MHz (Bruker Avance III), using CDCl3 or D2O as solvent. Chemical
2.4. Computational details
shifts on the δ scale (ppm) relative to tetramethylsilane were referenced
internally with respect to either the protio resonance of residual CHCl3
(δH = 7.26 ppm and δC = 77.16 ppm) or to dioxane added as a standard
to all D2O solutions (δH = 3.75 ppm and δC = 67.19 ppm) [71].
Fourier-transform mid-infrared (400–4000 cm−1) spectra (FTMIR)
were recorded at 4 cm−1 resolution on a Bruker VERTEX 70v spectro-
meter fitted with an A225 diamond attenuated total reflection (ATR)
accessory (Bruker) and a DTGS (deuterated triglycine sulfate) detector
(350–4000 cm−1). Solution spectra in CDCl3 were acquired at the
same resolution but in transmission mode from 4000 to 1000 cm−1
on a VECTOR 22 spectrometer (Bruker) using a variable path-length
cell from Perkin-Elmer equipped with IRTRAN-2 (ZnS) windows.
Although the reference spectrum corresponding to pure CDCl3 was
systematically subtracted from the sample spectrum, the contribution
from the solvent could not be fully canceled. Raman spectra were col-
lected with a Renishaw inVia spectrometer equipped with a 632.8 nm
He–Ne laser excitation source, a 1800 grooves/mm grating, and a micro-
scope fitted with a ×20 objective. Wavenumbers were calibrated with
respect to the silicon scattering line at 520(1) cm−1 of an internal
standard, whereas an external Si reference was used periodically to
check for energy drifts over time. Samples of neat NMAH (oil contained
in a glass capillary) or of uranyl compounds (randomly-oriented single
crystals placed on a glass slide) were analyzed in the 100–3200 or
160–1100 cm−1 range by accumulating 2 (extended mode) or 100
scans (static mode), respectively.
All computations have been carried out using the Gaussian 09
software [72]. The PBE0 global hybrid functional was used throughout
[73,74]. All atoms have been represented by the 6-311+G** Pople
basis set [75–77]. The solvent was represented using the polarized
continuum model (PCM) as implemented in Gaussian 09 [78]. Accord-
ingly, water was regarded as a continuous dielectric characterized by a
constant permittivity. Standard algorithms were used for optimizations
without any symmetry constraint and the nature of stationary points
was checked by subsequent vibrational analysis. NMR computations
were carried out in the GIAO formalism [79].
2.5. 1H–1H EXSY spectroscopy
2D EXSY measurements were carried out by applying a standard
pulse sequence for phase-sensitive 1H–1H NOESY (nuclear Overhauser
effect spectroscopy) spectra. A total of 256 complex points were collect-
ed in the indirectly-detected dimension with 4 scans and 1024 points
per increment. Quantitative spectra were acquired with mixing times
(tm) ranging from 3 ms to 1 s. Diagonal and off-diagonal cross peaks
for the methyl resonances were integrated in the 3.2–3.4 ppm range
using the MestReNova 8.1 software, the intensities were normalized,
and the rate constants were calculated using Eq. (1) [80], where k is
the overall exchange rate constant (k = kZE + kEZ) and r is expressed
by Eq. (2).
r þ 1
ln
¼ k tm
ð1Þ
ð2Þ
r−1
2.3. Synthesis of N-methylacetohydroxamic acid (NMAH)
IZZ þ IEE
2
N-Methyl-O-benzylhydroxylamine (7.00 g, 51 mmol) was dissolved
under nitrogen atmosphere in 120 mL of tetrahydrofuran (THF) at 0 °C
to which 10.5 g (75 mmol) of potassium carbonate was added. Then,
acetyl chloride (4.00 g, 51 mmol) was dropped-in and the reaction
mixture was allowed to react for 4 h at room temperature. After filtra-
tion and evaporation of the solvent, the residue was dissolved in
100 mL of dichloromethane and washed successively with a 0.1 M
solution of citric acid and water. The organic phase was dried over
magnesium sulfate and the solvent evaporated. The product was
purified by chromatography on silica gel using dichloromethane as
eluent. After solvent removal, 7.31 g (40.8 mmol, 80%) of a yellowish
oil was obtained. 1H NMR (300 MHz, CDCl3, 300 K): δ = 2.07 (s, 3H,
COCH3), 3.19 (s, 3H, NCH3), 4.82 (s, 2H, OCH2Ph), 7.38 (m, 5H, Ph)
ppm. 13C{1H} NMR (75 MHz, CDCl3, 300 K): δ = 20.3 (COCH3), 33.5
(NCH3), 76.3 (OCH2Ph), 128.8 (Ph), 129.1 (Ph), 129.4 (Ph), 134.6
(Ph), 174.3 (CO) ppm. Elemental analysis calculated for C10H13NO2
(179.22 g/mol): C 67.02, H 7.31, N 7.82; found: C 66.73, H 7.04, N,
7.80.
Successive glassware washings with hydrochloric acid and water
were carried out before performing the debenzylation step. The yellow
oil (7 g, 39.0 mmol) was dissolved in 100 mL of methanol and 200 mg of
Pd/C 10% (activated 3 h under vacuum at 150 °C) was added under
nitrogen atmosphere. The flask was evacuated under vacuum and
then put under 1 bar of dihydrogen. After stirring the reaction mixture
for at least 6 h, the Pd/C catalyst was removed by filtration through a
0.2 μm Whatman™ membrane. Evaporation of the solvent afforded
the desired product (NMAH) as a yellow oil in 95% yield (3.31 g,
37.1 mmol). 1H NMR (500 MHz, D2O, pD = 6.4, 300 K): δ = 2.09 (s,
0.82H, Z conformer, COCH3), 2.12 (s, 2.18H, E conformer, COCH3),
3.22 (s, 2.22H, E conformer, NCH3), 3.36 (s, 0.78H, Z conformer,
NCH3) ppm. 13C{1H} NMR (125 MHz, D2O, pD = 6.4, 300 K): δ =
19.57 (E conformer, COCH3), 20.29 (Z conformer, COCH3), 36.38 (E
conformer, NCH3), 39.65 (Z conformer, NCH3), 170.04 (Z conformer,
CO), 174.34 (E conformer, CO) ppm. Elemental analysis calculated for
C3H7NO2 (89.09 g/mol): C 40.44, H 7.92, N 15.72; found: C 40.11, H
7.92, N 16.00.
r ¼ 4xZxE
−ðxZ−xEÞ
IZE þ IEZ
In Eq. (2), xZ and xE stand for the mole fractions of the Z and E con-
formers, respectively (xZ = 1 − xE), IZZ and IEE for the intensities of the
diagonal peaks assigned to the Z and E conformers, and IZE and IEZ for
the intensities of the off-diagonal cross peaks. Eqs. (1) and (2) are
only valid if the T1 spin–lattice relaxation times for the two exchanging
methyl protons are equal [80]. T1 values were measured using the
standard inversion recovery experiment with a recycle delay (d1) set
to 20 s (Fig. S4). The obtained values for the four methyl groups belong-
ing to both E and Z conformers are roughly identical and range between
4.41 and 4.67 s (Table S1). Hence, both simplified equations can be
safely applied. Moreover, the recycle delay of 20 s was long enough to
allow for proton relaxation (d1 ~ 5 T1). The overall exchange rate
constant k, corresponding to the slope of the linear part of the
ln[(r + 1) / (r − 1)] vs. tm plot, was calculated by linear regression anal-
ysis of the experimental data according to Eq. (1). For large mixing
times, the curve reaches a plateau owing to relaxation effects; hence
the cross peak intensities become insensitive to exchange rates. Individ-
ual first-order rate constants for the cis to trans (kZE) and reverse (kEZ
)
interconversions were derived from Eq. (3), the equilibrium constant
KT being measured for the same solution and at the same temperature
(T) by 1H NMR spectroscopy.
½Eꢀ kZE
kZE
KT
¼
¼
¼
ð3Þ
½Zꢀ kEZ k−kZE
The Gibbs energy of activation for a given process (ΔGi≠) is given by
the Eyring equation [Eq. (4)], where kB, h, and R correspond to the
Boltzmann, the Planck, and the ideal gas constants, respectively, ki is
the first-order rate constant associated with the conformational change,
and T is the absolute temperature.
ꢀ
ꢁ
ki h
ΔGi≠ ¼ −RT ln
ð4Þ
kB
T
Please cite this article as: S. Brandès, et al., Conformational and structural studies of N-methylacetohydroxamic acid and of its mono- and bis-
chelated uranium(VI) complexes, J. Inorg. Biochem. (2015), http://dx.doi.org/10.1016/j.jinorgbio.2015.06.002

4
S. Brandès et al. / Journal of Inorganic Biochemistry xxx (2015) xxx–xxx
2.6. X-ray crystallography
palladium on charcoal afforded the desired product NMAH as a pale-
yellow oil in quantitative yields and pure form, as evidenced by elemen-
tal analysis and NMR spectroscopy.
Crystal data were collected either on a Nonius Kappa ([UO2(NMA)
(NO3)(H2O)2]) or a Bruker D8 VENTURE diffractometer ([UO2(NO3)2
(H2O)2]·4H2O and [UO2(NMA)2(H2O)]) equipped with an Apex II or a
CMOS Photon 100 detector, respectively. A suitable specimen was
selected and mounted on a Mylar loop with oil. The X-ray source was
graphite monochromated Mo-Kα radiation (λ = 0.71073 Å) from a
sealed tube. Data were measured using ϕ and ω scans, under a liquid
nitrogen jet stream (Oxford Cryosystems). Cell parameters were re-
trieved and refined with the SAINT software (release 8.34a) [81]. Data
reduction was performed using the same program which corrects for
Lorentz polarization. Structures were solved by direct methods using
the ShelXT software [82] and then refined by full-matrix least-squares
on F2 using ShelXL (release 2014/7) [83], both routines being imple-
mented in the Olex2 environment [84]. All non-hydrogen atoms were
refined anisotropically. Methylenic hydrogen atoms were placed at
calculated positions and refined using a riding model with C–H =
0.98 Å and Uiso(H) = 1.5Ueq(CH3). Likewise, the protons on water
molecules were allowed to ride and to rotate, with Uiso(H) =
1.5Ueq(OH) according to the procedure implemented in Olex2 [84].
COD-3000043 for [UO2(NO3)2(H2O)2]·4H2O contains the supplementa-
ry crystallographic data for this paper. These data can be obtained free of
charge from The Crystallography Open Database via http://www.
crystallography.net/search.html. CCDC-1042532 for [UO2(NMA)(NO3)
(H2O)2] (1) and 1042608 for [UO2(NMA)2(H2O)] (2) contain the
supplementary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
3.2. Vibrational spectroscopy of NMAH
The ATR-Fourier-transform mid-infrared (ATR-FTMIR) and Raman
scattering frequencies of band maxima of neat NMAH are summarized
in Table 1 (spectra are displayed in Figs. S1 and S2, respectively), togeth-
er with the DFT-computed band assignments and vibrational energies
for both cis and trans conformers. The ATR-FTMIR data of NMAH in di-
lute aqueous solutions (pH = 6 and 12) have already been reported in
the literature and discussed in light of DFT computations [88]. However,
this study was limited to a narrow spectral range (800–1800 cm−1) due
to the presence of water, while subtraction of its contribution to the
measured spectra has led to an information loss characterized by the
disappearance of some bands, such as the intense feature occurring at
1422 cm−1. Moreover, DFT-based band assignment was restricted to
the sole cis form [88]. Owing to these limitations, we recorded the spec-
trum of the neat oil (Fig. S1) and of CDCl3 solutions (c = 0.05 and 0.6 M;
Fig. S3), which allowed us to investigate the typical 400–4000 cm−1
MIR range. In addition, the Raman spectrum was also collected in
order to gain more insights into the vibrational structure of NMAH,
whereas the vibrational data for both cis and trans arrangements were
calculated (Table 1).
For some bands, a careful examination of Table 1 reveals marked dif-
ferences between the calculated vibrational frequencies for both
conformers. This is especially true for the ν(OH) mode, as this stretch
resonates in the cis geometry at lower energies (by ~250 cm−1) in
agreement with the formation of a intramolecular hydrogen bond
with the carbonyl oxygen atom. Likewise, the ν(CO) oscillator strength
is weakened by 52 cm−1 for the cis with respect to the trans arrange-
ment. Table 1 also indicates different assignments for the bands
expected to appear around 1148, 955, and 500 cm−1. According to our
computations, it can be anticipated that a mixture of both isomers
would result in a composite spectrum in which several bands are split.
The differences between the main vibrational frequencies of cis and
trans forms are indeed significantly larger than the spectral resolution
at which the spectrum was recorded (4 cm−1), but it cannot be defini-
tively concluded if both isomers or only one of them is present as
the peak width at half height in the fingerprint area reaches up to
~30 cm−1. Overall, the wavenumbers calculated for both isomers
appear to be in rather good agreement with the experimental data
after applying a very reasonable scaling factor of 0.98 [59]. It should
however be noted that several calculated harmonic frequencies for the
trans form are lower than those measured, while the reverse situation
is expected owing to systematic errors originating especially from
basis set truncation (applying anharmonic corrections leads to comput-
ed frequencies closer but still higher than experimental ones [89]). In
spite of the correction factor, the large discrepancy between computed
and recorded frequencies for the O–H and C_O stretching modes
might be ascribed to a large extend to medium effects (e.g., aggregation
through intermolecular hydrogen bonds) since the experimental spec-
trum was acquired for the neat oil, while the polarized continuum
model (PCM) for water was applied in the DFT calculations.
3. Results and discussion
3.1. Synthesis of NMAH
Several synthetic routes of N-methylacetohydroxamic acid (NMAH)
have been reported in the literature. The classical method proposed by
Blatt rests on the condensation of hydroxylamine with esters of acetic
acid [85]. Direct acylation of unprotected N-methylhydroxylamine has
also been described [86,87], but O-substitution might not be completely
avoided during the reaction. Moreover, the cost of the commercially-
available HNMe(OH) reagent represents a further limitation of this
procedure that prompted us to use an alternative strategy relying on
Boc and benzyl protection. The latter motif has often been chosen as
an advantageous O-protecting group, especially for the total synthesis
of more sophisticated hydroxamic acid based siderophores, such
as aerobactin and schizokinen for example [67], because of its outstand-
ing stability during the acylation step and its ease of removal by
hydrogenolysis. More recently, Sippl and Schenck prepared NMAH
starting from paraformaldehyde and O-benzylhydroxylamine [44] by
adapting a procedure described earlier by Lee and Miller [67]. Our ap-
proach is based on the synthesis of N-methyl-O-benzylhydroxylamine
[67–69] followed by N-functionalization by acylation with acetyl
chloride (Scheme 2). This method is a somewhat more convenient
alternative as it avoids tedious chromatographic purification of the
intermediates and final product, and thus is more amenable to large-
scale preparations. Deprotection by hydrogenolysis over 10%-loaded
Scheme 2. Synthesis of NMAH.
Please cite this article as: S. Brandès, et al., Conformational and structural studies of N-methylacetohydroxamic acid and of its mono- and bis-
chelated uranium(VI) complexes, J. Inorg. Biochem. (2015), http://dx.doi.org/10.1016/j.jinorgbio.2015.06.002

S. Brandès et al. / Journal of Inorganic Biochemistry xxx (2015) xxx–xxx
5
higher energies by 20, 66, and 16 cm−1, respectively. Owing to the
broadness of the carbonyl vibration band and its rather symmetric
shape, FTMIR spectroscopy seems not to be a suitable method for
studying the conformational properties of hydroxamic acids, in contrast
to 1H NMR spectroscopy (vide infra).
Table 1
Most important IR and Raman features along with computed vibrational frequencies
(cm−1) and assignments for the cis and trans isomers of neat NMAH.
Experimental
Calculated
IR
Raman
Cis
Assignmenta
Trans
Assignmenta
3154
1606
1548
1489
1448
1422
1387
1365
1197
1148
1126
1038
1014
955
3549
1661
1603
1492
1471
1464
1424
1377
1266
1175
1145
1045
1044
990
ν(OH)
ν(CO)
ν(N–CO)
δ(CH)
δ(CH)
3795
1713
1525
1469
1464
1459
1403
1372
1234
1182
1142
1045
1026
967
ν(OH)
ν(CO)
ν(N–CO)
δ(CH)
δ(CH)
δ(CH)
ν(CC) + δ(CH)
δ(CH)sym
ν(N–CH3) + ν(NO)
ν(CC) + ν(CH3)
τ(CH)
τ(CH)
τ(CH)
ν(N–CH3) + τ(CH)
Mode a
δ(NCO)
3.3. NMR spectroscopy of NMAH
1604
1490
1444
1428
1390
1361
1200
1146
3.3.1. Equilibrium studies
The 1H NMR spectrum of NMAH recorded at pD = 6.4 in D2O
shows two partially overlapping singlets of unequal intensities attribut-
ed to the C-methyl group at 2.09 ppm (0.82 H) and 2.12 (2.18 H),
respectively. The two other unequal singlets at 3.22 (2.22 H) and
3.36 ppm (0.78 H) are assigned to the N-methyl group of N-
methylacetohydroxamic (NMAH) acid (Fig. 1a). Moreover, relative
line intensities were found to be independent of the total concentration
in the ca. 0.09–0.45 M range. These results are in perfect accordance
with those reported recently by Sippl and Schenck for up to 25-fold
more dilute solutions [44]. Observation of these two pairs of signals is
consistent with the hindered rotation around the C–N bond on the
NMR time scale and with the occurrence of two rotamers of E and Z
conformations in equilibrium at room temperature [31,32,35,36,43].
Moreover, the large splitting observed between both N-methyl reso-
nances (Δδ = 0.14 ppm) can thus be ascribed to the close vicinity of
the corresponding protons with the shielding cone arising from the
carbonyl function for the E conformer, whereas the C-bound methyl
group is not affected by such effects in either conformation (Δδ =
0.03 ppm).
δ(CH)
ν(CC) + δ(CH)
δ(CH)sym
ν(N–CH3) + ν(NO)
ν(CC) + δ(CH)
τ(CH)
τ(CH)
τ(CH)
ν(NO) + τ(CH)
Mode a
δ(NCO)
1014
955
744
599
587
502
736
597
584
498
757
624
609
478
757
595
593
504
τ(skeleton)
δ(OH)out-plane
τ(skeleton)
δ(CCN) + δ(ONC)
a
Vibrational modes are labeled as usual: ν refers to a stretching, δ to a bending, and τ to
a torsion mode. Mode a is described in plain text.
Interestingly, both N–C bonds (N–CH3 and N–CO) resonate at very
different energies, the ν(N–CO) band occurring 350 cm−1 above the
ν(N–CH3) signal, in full accordance with the double-bond character of
the central N–C bond. Calculations also evidenced an original vibrational
mode, labeled “a” in Table 1, which cannot be described by the usual
distortions. It involves a simultaneous symmetrical motion of the
oxygen and both carbon centers linked to the immobile nitrogen
atom. Hence, this vibration looks like a ring respiration that leads to a
marked variation in polarizability and thus affords an intense Raman
line located at 744 cm−1. It can also be noticed that the in-plane
δ(OH) distortion is not listed in Table 1 because it is widely distributed
over a great number of vibrational modes.
Based on what precedes, it seems not possible to conclude on the
actual geometry of neat NMAH simply by comparing the computed
and experimental IR spectra. Therefore, solution spectra at various
total concentrations in CDCl3 were also recorded (Fig. S3). The transmis-
sion FTMIR spectrum of a 0.6 M NMAH solution enabled to distinguish
only the ν(OH), ν(CH3), and ν(CO) bands at 3174, 2935, and
1622 cm−1, the other ones being obscured by those arising from the
solvent in spite of the background correction. In case of the more dilute
solution (c = 0.05 M), only the ν(CO) signal centered at almost the
same wavenumber (1624 cm−1) could be recognized. Compared to
the spectrum of neat NMAH, the three bands are shifted in solution to
At that stage, the major difficulty in data interpretation resides in the
correct identification of the peaks arising from both the major and
minor isomers, as this is the primary source of discrepancies between
previous reports. In the past, authors often based their reasoning on
shielding effects [31,32,36], simple analogies with other hydroxamic
acids or various amides [30,33,40,43], or relied on the prediction of
the most stable structure calculated by quantum chemical methods
[34,35,37,41,43]. However, these approaches have all their own caveats,
related to the high variability of chemical shifts with the nature of
solvent and substituents, the more or less pronounced concentration
dependence of the E/Z ratio [32,38,39], the inherent difficulties of
treating solvation effects by theoretical methods, and the computational
uncertainty that exceeds the actual energy difference of just a few
kJ mol−1 between both forms (vide infra) [33,34,37,39,43,48,51–53,
55,57,59].
Recently, Sippl and Schenck proposed for the first time a direct and
thus reliable assignment of both sets of resonances of NMAH based on
1D NOE experiments performed in D2O, especially NOE build-up curves
[44]. Following these authors, the pair of most intense signals at 2.12
Fig. 1. 1H NMR spectra (500 MHz) of NMAH recorded at 300 K in D2O. (a) pD = 6.4 and (b) pD = 10.6.
Please cite this article as: S. Brandès, et al., Conformational and structural studies of N-methylacetohydroxamic acid and of its mono- and bis-
chelated uranium(VI) complexes, J. Inorg. Biochem. (2015), http://dx.doi.org/10.1016/j.jinorgbio.2015.06.002

6
S. Brandès et al. / Journal of Inorganic Biochemistry xxx (2015) xxx–xxx
Table 2
obviously lower in the trans arrangement than in the cis one. Moreover,
the former layout maximizes also the hydrogen bonding interactions
with the surrounding water molecules in the first solvation shell,
disfavoring intermolecular aggregation. In turn, the Z form is less-well
solvated in water than in chloroform, owing to the expected intramole-
cular hydrogen bond and the location of both methyl groups on the
same side of the molecule, thus creating an unfavorable hydrophobic
surface.
DFT-calculated relative 1H NMR chemical shifts (ppm) for both isomers of N-
methylacetohydroxamic acid.
Conformer
Δδ(C–CH3)
Δδ(N–CH3)
Z (cis)
E (trans)
0.00
1.19 [1.27]a
1.16 [1.13]a
0.07 [0.03]a
a
Experimental value measured in D2O at 300 K.
So far, E/Z equilibrium studies have surprisingly mainly focused on
the acid form of NMAH that prevails in acidic or near neutral media.
To the best of our knowledge, the conformational properties of NMA−
have only been investigated by DFT and ab initio calculations, which
predict the greater stability of the E isomer [53,55]. This prompted us
to analyze also the underlying isomerization process of the conjugated
base under alkaline conditions. According to the pKa value of 8.68
measured in water [I = 0.1 M KNO3, T = 298.2(2) K], the abundance
of the anionic NMA− species exceeds 95% above pH 10.2 in aqueous
solutions [28,90]. The 1H NMR spectrum of the deprotonated form
recorded in D2O at pD 10.6 evidences also two sets of signals (Fig. 1b),
but all of them are shielded by 0.053–0.105 ppm with respect to the
experimental data collected at pD = 6.4 (Fig. 1a). Shifts of similar
magnitude to higher fields with increasing pD values have been found
for the deprotonation of acetohydroxamic acid (RC = CH3, RN = H)
[37]. Two intense resonances occurring at 2.01 (C–CH3) and 3.17 ppm
(N–CH3) are assigned to the E form and one singlet at 3.27 ppm (N–
CH3) is attributed to the Z conformer, the second signal for the Z form
overlapping with the manifold at 2.01 ppm arising from the E conform-
er. Interestingly, the two most intense peaks centered at 2.01 and
3.17 ppm split into quadruplets. Observation of a scalar coupling
between the N–CH3 and C–CH3 methyl protons [5J = 0.74(4) Hz]
supports the occurrence of a rigid structure on the NMR time scale
and further suggests a high weight of the iminium resonance form in
the electron delocalization scheme. Moreover, the close resemblance
of the line-patterns found under acidic and basic conditions provides
additional support to the spectral assignment derived from Sippl and
Schenck's NOE experiments, according to which the N-methyl
group of the Z isomer appears downfield from the more abundant E
conformer.
The E/Z concentration ratio, which equals 9.63(5) at 300 K, is about
three times higher than that found for the neutral NMAH form. Interest-
ingly, the corresponding change in Gibbs free-energy in basic conditions
(ΔG300 = 5.6 kJ mol−1) amounts ~2NkBT. The equilibrium shift towards
the trans isomer upon deprotonation can be easily understood as a
direct consequence of the more pronounced electrostatic repulsions
occurring in the cis rotamer between the carbonyl lone pairs and the
anionic N–O− versus the neutral N–OH oxygen atom. Insertion of a
proton in between the C_O and the N–O− group evidently reduces
the electronic repulsion, thus increasing the proportion of the cis isomer
in the protonated vs. deprotonated state, as evidenced by the signifi-
cantly higher basicity of 1-oxypiperidine-2-one (PIPO−), a cyclic six-
membered analog of NMA− possessing an enforced cis hydroxamate
moiety [28]. Moreover, it can also be argued that the higher stability
of cis-NMAH with respect to cis-NMA− might also be ascribed to some
extent to an intramolecular N–OH⋯O_C hydrogen bond which can
only occur in the former species disfavored in aqueous media.
and 3.22 ppm can be assigned to the E conformer. Confirmation was
provided by our DFT calculations of the 1H NMR spectrum of NMAH
using the GIAO formalism. It should be noted that this approach for
assigning NMR chemical shifts of the hydroxamic acid conformers is
extremely seldom [41]. To avoid the computation of NMR shifts for an
external reference, only relative Δδ values up-field from the most
shielded signal (δ set arbitrarily to 0) are reported in Table 2. Although
only qualitative trends should be considered, as differences in the
measured chemical shifts are usually quite small in comparison with
the typical error of the GIAO formalism, comparison between the
calculated and measured NMR data enables in the present case to
draw two gratifying conclusions. Firstly, the qualitative agreement
between the experimental and the computed chemical shift values is
quite good and allows further investigations. Secondly, the relative
position of both sets of resonances, with the two signals of the E isomer
located in-between those of the Z one, is perfectly reproduced.
According to the peak intensity ratio of both well-resolved N–CH3
resonances, the E isomer of NMAH prevails in an almost 3:1 ratio over
the Z form at 300 K in D2O. The herein derived equilibrium constant
[K300 = [E] / [Z] = 2.86(2)] is in excellent agreement with the values
reported at slightly lower temperatures by Sippl and Schenck (K281
=
3.17, T = 281 K) [44], Birus et al. (K298 = 2.8, T = 298.2 K) [36], and
Caudle and Crumbliss (K298 = 3.5, T = 298.2 K) [35], although the latter
authors erroneously assigned the major species to the Z rotamer.
Noteworthy, these equilibrium constants for isomerization of NMAH
are consistent also with the corresponding values found for
desferrioxamine B [45].
Very recently, Guérard et al. evidenced by 1H NMR spectroscopy the
occurrence of a slight excess of cis- over trans-NMAH in CDCl3 (K =
0.92) at room temperature and corroborated this finding by DFT
calculations [43]. As a matter of fact, their claim that the Z form is the
most stable one in solution suffers from several shortcuts. The herein
derived ΔG300 value (2.6 kJ mol−1) equals the thermal energy
(NkBT = 2.5 kJ mol−1 at 300 K). It turns out that an accurate computa-
tion of ΔG values by quantum chemical methods is quite challenging as
it is beyond DFT accuracy. Moreover, the assertion by Guérard et al. rests
on a single measurement without any indication about the concentra-
tion. According to Brown et al. [32], inversion of the relative E/Z stability
is observed in CDCl3 (K298 = 0.5–5.7) for NMAH concentrations varying
between 0.052 and 0.84 M, while the E conformation remains more
stable over the entire concentration range in C6D6 (K298 = 2.0–6.4)
and toluene-d8 (K298 = 1.9–6.3) [32].
Repeating the measurements in chloroform, we could indeed ascer-
tain a strong concentration-dependence of the E/Z ratio (Fig. S5)
although the derived equilibrium constants (Table S2) deviate slightly
from those reported in the literature [32]. The residual water content
might in part explain the discrepancies. Nevertheless, the main trend
is the same and can be rationalized as follows. In dilute solutions (e.g.,
0.05 M), the Z isomer is stabilized by an intramolecular C_O⋯HON
hydrogen bond, especially in less-dissociating solvents like CDCl3. At
higher concentrations (e.g., 0.6 M), aggregation through two intermo-
lecular hydrogen bonds favors the E conformer [32]. In contrast,
formation of aggregates can be ruled out in D2O at least up to 0.45 M,
as the 1H NMR spectra evidence no concentration dependence. The
prevalence of monomeric NMAH in the E form can be understood in
terms of electrostatics and solvation. Owing to the larger distance, the
electronic repulsion between the oxygen lone pairs of electrons is
3.3.2. Kinetic studies
Although the kinetic parameters for the hindered rotation around
the C–N bond of NMAH have already been estimated by variable-
temperature and pressure 1H NMR spectroscopy in D2O [35,36] and
CDCl3 [34], the interconversion rate between the Z and E conformers
of the conjugated base NMA− has not been reported so far. This incited
us to explore for the first time the conformational rearrangement
processes of hydroxamates by dynamic 2D EXSY as a function of their
protonation state. EXSY has gained popularity since it is a powerful
method for unraveling slow isomerization or exchange processes,
Please cite this article as: S. Brandès, et al., Conformational and structural studies of N-methylacetohydroxamic acid and of its mono- and bis-
chelated uranium(VI) complexes, J. Inorg. Biochem. (2015), http://dx.doi.org/10.1016/j.jinorgbio.2015.06.002

S. Brandès et al. / Journal of Inorganic Biochemistry xxx (2015) xxx–xxx
7
especially for multisite systems for which the more classical line-shape
analysis of variable-temperature spectra or the magnetization transfer
method becomes often quite complex and cumbersome [80,91]. Associ-
ated first-order rate constants in the range from about 0.05 to 20 s−1 are
straightforwardly derived from the respective intensities of the diagonal
and off-diagonal cross correlation peaks, as outlined in the Experimental
section [Eqs. (1)–(3)]. In practice, EXSY and NOESY correlations are de-
termined in the same experiment [80], but they can be easily distin-
guished from each other since cross peaks that correlate the Z and E
conformers can only be related to slow exchange processes between
them and thus belong to the EXSY component. Moreover, the in-phase
EXSY off-diagonal correlations are easily discernable from the anti-
phase NOESY peaks.
2D NOESY/EXSY correlation charts recorded for NMAH using mixing
times higher than 150 ms (Fig. 2) evidence only two sets of off-diagonal
peaks that correlate the signals arising from the E and Z conformers for
each type of inequivalent methyl groups. However, for shorter mixing
times an additional cross peak becomes visible between the N–CH3
and C–CH3 resonances arising from the E form, which results from a
weak scalar long-range coupling (5J COSY artifact) already observed in
the 1D spectrum of NMA− rather than to a through-space NOE effect
[44]. By varying the mixing time from 3 ms to 1 s, the on- and off-
diagonal cross peaks are sufficiently intense and well-resolved to
allow for reliable volume integration and determination of the E ⇌ Z
exchange rate constants according to Eqs. (1) and (2). The slope of the
line obtained by plotting ln[(r + 1) / (r − 1)] vs. the mixing time
(Fig. 3) corresponds to the overall rate constant k, defined as the sum
Fig. 3. EXSY data plotted according to Eq. (1). Slopes at the origin correspond to the sum of
rate constants k = kZE + kEZ, see the Experimental section for details. T = 300 K, pD = 6.4
(●) and pD = 10.6 (▲).
61.0 kJ mol−1 (instead of 68.0 kJ mol−1) for ΔG≠ZE and 71.1 kJ mol−1
(instead of 70.6 kJ mol−1) for ΔG≠EZ. Overall, these findings are also in
excellent agreement with the kinetic parameters derived by Birus
et al. for NMAH [kZE = 8.2(5) s−1, kEZ = 2.8(5) s−1, ΔV≠
=
EZ
12(4) cm3 mol−1 at 298 K] [36] and desferrioxamine B [45] using vari-
able temperature 1H NMR spectroscopy and line-shape analysis. These
authors interpreted the positive activation volume for the trans-to-cis
isomerization process by a sweep out of solvent molecules from the sol-
vation shell towards the bulk when reaching the activated state. In line
with the created disorder and bond dissociations, they also reported
positive activation entropy [ΔS≠ = 31(16) J mol−1 K−1] and enthalpy
[ΔH≠ = 76(5) kJ mol−1] for the overall rotational process, the latter
being also considered as the rate-determining step for the formation
of [Cu(NMA)]+, [La(NMA)]2+, and various desferrioxamine B com-
plexes in acidic media.
of the forward cis-to-trans (kZE) and backward trans-to-cis (kEZ
)
exchange rate constants (k = kZE + kEZ). From the estimated value of
the equilibrium constants KT measured for the same sample at identical
temperature (T = 300 K), the individual rate parameters kZE
=
9.0(2) s−1 and kEZ = 3.14(5) s−1 are obtained. The order of magnitude
found for both rotation rate constants complies with the slow-exchange
condition, k ≪ ΔνZE, where k is the exchange rate constant (3–9 s−1
)
For EXSY experiments carried out at pD = 10.6, the normalized
exchange cross peaks were found to be less intense compared to those
appearing under slightly acidic conditions, suggesting a much slower
rotation around the C–N bond when the anionic hydroxamate form
prevails in solution. Taking into account the higher E/Z concentration
ratio measured for the conjugated base NMA− [K300 = 9.63(5) vs.
2.86(2) for NMAH], the quantitative treatment of the EXSY data array
and ΔνZE is the chemical shift difference expressed in Hz between the
resonances of both rotamers (ΔνZE = 0.14 ppm × 500 MHz = 70 Hz).
Applying the Eyring equation [Eq. (4)], the Gibbs energy of activation
for the cis → trans rotation at 300 K amounts ΔG≠ZE = 68.0 kJ mol−1
,
while the activation barrier for the reverse trans → cis conversion is
ΔG≠EZ = 70.6 kJ mol−1. The DFT computations of ΔG≠ values were satis-
factory as they returned the correct order of magnitude with values of
collected as a function of the mixing time (Fig. 3) gives kZE
=
=
0.96(3) s−1 (ΔG≠ZE = 73.6 kJ mol−1) and kEZ = 0.10(2) s−1 (ΔG≠
EZ
79.2 kJ mol−1) at 300 K. The computations carried out on the anionic
species are less accurate, as values around 88 kJ mol−1 were obtained.
Although the observed trend is respected, the lower accuracy is not un-
expected as computations at the level of theory used herein are less like-
ly to give quantitative results in case of anions. These experimental
exchange rate constants are roughly 10 and
3 times slower,
respectively, than those determined for the protonated species. Such a
decrease of the interconversion rates can be primarily assigned to the
more pronounced electrostatic repulsion between the negatively-
charged N–O− and the neutral carbonyl oxygen atoms, which is also
responsible for the higher proportion of the E conformer, as well as for
the enhancement of the double bond character of the C–N bond in
both iminium resonance forms Z′ and E′ (Scheme 1) [37]. The greater
partial positive charge on the nitrogen atom is stabilized by the methyl
electron-donating substituent at that position, which contributes to
increase the double bond character of the C–N bond and therefore
decreases drastically the rotation rate. The heavy weight of the iminium
form is also fully compliant with the quadruplet splitting pattern
observed for the protons of both methyl groups, which reflects long
range couplings through the C_N bond the in the E′ mesomeric form.
Semi-quantitative DFT computations evidenced a significant decrease
of the Wiberg bond index for the carbonyl group from 1.60 to 1.46
upon deprotonation of NMAH. This confirms the less pronounced
double-bond character of the C_O group in the anionic form and,
Fig. 2. 2D NOESY/EXSY correlation for NMAH recorded in D2O (pD = 6.4) at 300 K with a
mixing time of 500 ms. The diagonal and off-diagonal cross peaks in the 3.2–3.4 ppm range
were integrated to determine the overall interconversion rate constant k between Z and E
conformers.
Please cite this article as: S. Brandès, et al., Conformational and structural studies of N-methylacetohydroxamic acid and of its mono- and bis-
chelated uranium(VI) complexes, J. Inorg. Biochem. (2015), http://dx.doi.org/10.1016/j.jinorgbio.2015.06.002

8
S. Brandès et al. / Journal of Inorganic Biochemistry xxx (2015) xxx–xxx
consequently, the more important double-bond character of the central
C–N bond, which is perfectly in line with all aforementioned results.
158.0(1)° and O1B–U1B–O4B = 157.4(1)°]. The two independent ura-
nium centers U1A and U1B are distant by 6.4240(5) Å and almost
aligned with the O5A and O5B atoms from the coordinated water mol-
3.4. X-ray crystallography of uranyl complexes
ecules [U1A–O5A⋯U1B = 176.7(2)° and O5A⋯U1B–O5B =
174.1(1)°], thus forming a linear chain that parallels the virtual axis
passing through the inversion centers. Both O5 least-squares planes
are rotated by 60.9(1)°. In the crystal lattice, the symmetry-related
mean planes parallel to (O1A, O2A, O3A, O4A, O5A) are much further
apart (7.320 Å) than those which are parallel to (O1B, O2B, O3B, O4B,
O5B) (2.700 Å), avoiding thereby any close contact between the “yl”
oxygen atoms.
The uranyl ions have typical U_O bond lengths, although the slight-
ly shorter distance found for 1 [1.770(1) Å] in comparison to 2
[1.785(3) Å] reflects the weaker equatorial ligand field strength when
a single NMA− anion binds to the metal atom. A similar trend is found
for the crystal structures of uranium(VI) complexes with PBA− reported
in the literature. Indeed, the average U_O distance for the mono-
The uranyl nitrate salt used to prepare the NMA− complexes was
analyzed by X-ray diffractometry in order to assess the molecular
weight and composition. The [UO2(NO3)2(H2O)2]·4H2O formulation
−
with two bidentate nitrate ions [U–ONO = 2.51(1) and 2.52(2) Å]
3
and two water molecules [U–Ow = 2.406(9) Å] coordinated in the
equatorial plane of the linear uranyl cation [U_O = 1.75(2) and
1.76(2) Å, O_U_O = 179.8(8)°] has been established. These structural
characteristics are very similar to the bond metrics derived from neu-
tron diffraction studies [92]. Further details are available in the
Supporting Information.
Single crystals of red-orange color were obtained for both the mono-
chelated [UO2(NMA)(NO3)(H2O)2] (1) and bis-chelated [UO2(NMA)2
(H2O)] (2) species by slow evaporation of aqueous uranyl nitrate
solutions at pH ~ 4–5 containing either one or two equivalents of the
free ligand, respectively. Crystallographic parameters, data collection,
and refinement information are summarized in Table 3. ORTEP plots
of the asymmetric unit found in both crystal structures are displayed
in Fig. 4 together with the atom labeling scheme. Due to the presence
of 0.1 M KNO3 in the mother liquor from which compound 1 crystallized
out, a monodentate nitrate anion binds to the uranyl cation in addition
to a single bidentate NMA− anion and two water molecules. In turn, the
bis-chelated complex 2 was prepared without adding an alkali salt, and
thus only a single water molecule, located in between both chelating
NMA− ligands according to a butterfly-type arrangement, fulfills the
coordination sphere. Interestingly, the asymmetric unit of 2 contains
two structurally distinct [UO2(NMA)2(H2O)] units which are connected
by two pairs of hydrogen bonds (Table S3) involving four protons of the
coordinated water molecules and the hydroxamic N–O oxygen atoms
belonging to the neighboring motif. Each unit cell comprises two such
dimers (Z = 4). It turns out that both ligands adopt the same orientation
around the UO22+ ion, with both pairs of C–O and N–O oxygen
atoms facing each other respectively. The former oxygen donors are
chelated complex [UO2(PBA)(THF)2Cl] is about 0.02
Å shorter
[1.75(1) Å] [61] compared to the mean value of 1.77(2) Å calculated
for the bis-chelated compounds [UO2(PBA)2S] (S = MeOH, DMSO, or
DMF) [62–64]. Likewise, the O_U_O angles (Table 4) are, within the
error limits, identical to those measured for the four PBA− derivatives.
They also fall in the typical range found for the other crystallographical-
ly characterized uranyl complexes, as 80.2% of the O_U_O angles are
in the range 176.63–180° among the 2048 structures for which 3D-
coordinates could be retrieved from the Cambridge Structural Database
(CSD, release 5.35) with R-factors less than 10%.
The nearly linear UO22+ cation which is slightly bent towards a
bound water molecule in 1 and 2, is surrounded by five oxygen donor
atoms forming the equatorial coordination plane, as it is classically
observed for about 60% of uranium(VI) compounds among the 2048 en-
tries found in the CSD (release 5.35). Thus, the coordination polyhedron
around uranium can be best described as a pentagonal bipyramid. Dis-
tortion from the ideal structure is reflected by slight departures from
the ideal 90° of the Oeq–U–Oyl angles between equatorial and apical
“yl” oxygen atoms (Table 4). The near-planar arrangement of the five
equatorial oxygen atoms is also evidenced by deviations from the corre-
sponding least-squares plane which do not exceed 0.144 Å [0.008–0.144
approximately situated trans to each other [O1A–U1A–O4A
=
−
(ONO ) Å for 1, 0.025–0.111 and 0.010–0.069 Å for units A and B in 2, re-
3
spectively], while the U atom lies only by 0.076 Å (1), 0.020 Å (2, unit A),
Table 3
and 0.009 Å (2, unit B) out of the plane. Another informative parameter
that gauges the planarity of the donor atoms, and thus the crowding in
the equatorial mean plane, is the sum (Σeq) of the five Oeq–U–Oeq angles
(Table 4). For a perfectly coplanar arrangement like in a pentagonal
bipyramid, Σeq should equal 360°. Both the mono- and bis-chelated
complexes 1 and 2 evidence no significant deviation from this ideal
value, reflecting the absence of steric congestion imposed by the
hydroxamate backbone, the attached methyl substituents, or the other
ligands present in the coordination sphere. It can be noticed that the
perfectly planar nitrate anion bound to the uranyl cation in 1 forms a
dihedral angle of 67.5(1)° with respect to the equatorial O5 mean
plane. Σeq values reaching almost 360° are also encountered for the
diphenyl-substituted hydroxamato complexes [UO2(PBA)(THF)2Cl]
[61], [UO2(PBA)2S] (S = MeOH or DMF) [62,64], or for [UO2(1,2-
HOPO)2(H2O)] which contains aromatic ligands [93]. According to
Table 4, the average Oeq–U–Oeq bite angle for the bidentate NMA−
ligand in 1 and 2 is 64.6(3)°, which is similar to the value found for
the bis-chelated uranyl complexes of PBA− [65.1(4)°]. However, the
bite angle is larger by about 2° for [UO2(PBA)(THF)2Cl] (67.0°) [61]
and [UO2(1,2-HOPO)2(H2O)] (66.1°) [93].
X-ray crystallographic data for [UO2(NMA)(NO3)(H2O)2] (1) and [UO2(NMA)2(H2O)] (2).
[UO2(NMA)(NO3)(H2O)2] [UO2(NMA)2(H2O)]
Empirical formula
Formula weight
Crystal system
Space group
C3H10N2O9U
456.16
Monoclinic
P 21/c
C6H14N2O7U
464.22
Triclinic
P1
a (Å)
b (Å)
c (Å)
α (°)
β (°)
γ (°)
V (Å3)
Z
6.1772(6)
19.328(2)
9.4005(8)
90
112.643(5)
90
8.5329(6)
10.7559(8)
13.621(1)
106.237(3)
100.528(3)
97.115(3)
1159.6(2)
4
1035.8(2)
4
T (K)
115(2)
2.925
15.706
100(2)
2.659
14.021
ρ
calcd (g cm−3
)
μ (Mo Kα) (mm−1
)
2θ range for data collection (°) 6.31–55.29
5.76–55.19
39,239
Reflections collected
20,929
Independent reflections
Data/restraints/parameters
Goodness-of-fit on F2a
R indices [I N 2σ(I)]b
2385 [Rint = 0.0518]
2385/0/140
1.123
R1 = 0.0216,
wR2 = 0.0518
R1 = 0.0243,
wR2 = 0.0527
1.01/−1.75
5345 [Rint = 0.0481]
5345/0/299
1.179
R1 = 0.0258,
wR2 = 0.0519
R1 = 0.0350,
wR2 = 0.0556
2.90/−1.96
Selected U–O distances are summarized in Table 4. As expected, the
negatively charged hydroxamate oxygen atom in 1 interacts more
strongly with the metal [U–ON = 2.387(3) Å] than the carbonyl oxygen
atom does [U–OC = 2.396(3) Å]. More surprisingly, one water molecule
and the bound nitrate anion give rise to significantly shorter
[2.360(3) Å] and longer [2.474(3) Å] U–O bonds, respectively. Overall,
R indices (all data)b
Largest diff. peak/hole (e Å−3
)
a
GOF = [Σw(|Fo| − |Fc|)2 / (No − Nv)]1/2
.
b
R1 = Σ||Fo| − |Fc|| / Σ|Fo|, wR2 = {Σ[w(F2o − Fc2)2] / Σ[w(Fo2)2]}1/2
.
Please cite this article as: S. Brandès, et al., Conformational and structural studies of N-methylacetohydroxamic acid and of its mono- and bis-
chelated uranium(VI) complexes, J. Inorg. Biochem. (2015), http://dx.doi.org/10.1016/j.jinorgbio.2015.06.002

S. Brandès et al. / Journal of Inorganic Biochemistry xxx (2015) xxx–xxx
9
Fig. 4. ORTEP views of (a) the mono-chelated [UO2(NMA)(NO3)(H2O)2] (1) and (b) the bis-chelated [UO2(NMA)2(H2O)] (2) complexes found in the asymmetric units, with thermal ellipsoids
drawn at the 50% probability level.
these results contrast with the bond metrics reported for the only other
known monohydroxamato uranyl complex [UO2(PBA)(THF)2Cl] where
U–ON = 2.37(3) Å and U–OC = 2.34(3) Å [61]. In 2, similar discrepan-
cies are noted among the two pairs of chelate cycles, as U–ON and U–
OC distances are in the ranges 2.376–2.401 Å [avg. 2.39(1) Å] and
2.368–2.386 Å [avg. 2.38(1) Å], respectively. These averages compare
well with the corresponding mean values [U–ON = 2.37(3) Å and U–
OC = 2.38(2) Å] calculated for the three bis-chelated compounds
[UO2(PBA)2S] [62–64]. However, owing to the limited set of available
data but probably also to the difficulties of precisely locating light
atoms when solving X-ray structures of coordination compounds that
incorporate elements with high electron densities, no significant trend
seems to emerge once the uncertainties are taken into account.
the short C–N distances [1.289(6) Å for 1 and 1.306(6) Å on average
for 2] clearly reflect the partial double-bond character, in agreement
with the literature data. Hence, an almost flat arrangement of the four
atoms belonging to each binding unit can be anticipated. This is indeed
the case as the O–N–C–O torsion angles (τO–N–C–O) are less than 1° in
both NMA− complexes. The two methyl substituents induce no signifi-
cant steric congestion, considering the almost coplanar arrangement of
both sp3 carbon atoms which deviate from the O–N–C–O least-squares
plane by no more than 0.054 Å for C3 in 1 and 0.141 Å for C3A in 2. In
contrast, the replacement of both methyl groups by bulkier phenyl
substituents results in a backbone distortion of the PBA− moiety, as
τ
O–N–C–O reaches −7.1° in [UO2(PBA)2(MeOH)] [62] and even 11° in
[UO2(PBA)(THF)2Cl] [61], in spite of the tilted orientation adopted by
the aromatic rings in order to minimize the mutual steric interaction.
According to ring-puckering analysis (Table S4) [94], the five-
membered chelate cycle formed by NMA− upon binding to UO22+
adopts in 1 an envelope-type conformation. For the bis-chelated com-
plex 2, each independent uranyl cation is surrounded by one almost
flat chelate cycle, while the second one adopts an envelope conforma-
tion with the metal center siting out of the plane, like in 1. Noteworthy,
the two mostly puckered chelate cycles possess also the mostly
differentiated U–OC and U–ON bond lengths (Δd = 0.03 Å vs. 0.01 and
0.006 Å for the approximately flat U1A, O1A, C2A, N1A, O2A and U1B,
O4B, C5B, N2B, O3B rings, respectively). In comparison, the diphenyl
derivative PBA− forms likewise a bent chelate cycle in the [UO2(PBA)
(THF)2Cl] complex [61], whereas the geometries of those found in
[UO2(PBA)2(MeOH)] [62] are more distorted than those occurring in
2, as they adopt a more twisted conformation due to the higher steric
crowding imposed by the phenyl rings (Table S4). A similar situation
seems to be encountered for [UO2(PBA)2(DMF)] [64], although the dis-
order encountered in one chelate ring does not enable to discuss that
point further.
In spite of some charge delocalization over the O–N–C–O hydro-
xamate group, the average C_O bond distances found in 1 [1.332(5)
Å] and 2 [1.300(8) Å] are significantly shorter than the N–O
ones [1.345(5) and 1.362(8) Å, for 1 and 2, respectively], following the
general trend found among the uranyl PBA− complexes. Moreover,
Table 4
Selected bond distances (Å) and angles (°) for [UO2(NMA)(NO3)(H2O)2] (1) and
[UO2(NMA)2(H2O)] (2).
1
2
Unit A
Unit B
U–O1
U–O2
U–O3
U–O4
U–O5
U_O6
U_O7
2.396(3)
2.386(3)
2.474(3)a
2.420(3)b
2.360(3)b
1.770(3)
1.771(3)
176.7(1)
64.3(1)c
68.8(1)
2.387(4)
2.376(4)
2.397(4)
2.368(4)
2.364(4)b
1.783(4)
1.786(4)
176.9(2)
64.8(1)c
73.0(1)
2.371(4)
2.402(4)
2.377(4)
2.384(4)
2.370(4)b
1.788(4)
1.782(4)
177.7(2)
64.6(1)c
73.4(1)
O6_U_O7
O1–U–O2
O2–U–O3
O3–U–O4
O4–U–O5
O5–U–O1
Sum of O–U–O angles (Σeq
77.9(1)
77.6(1)
71.8(1)
360.4(2)
64.6(1)c
78.4(1)
64.6(1)c
79.0(1)
3.5. Vibrational spectroscopy of uranyl complexes
79.6(1)
360.4(2)
78.6(1)
360.2(2)
The infrared spectrum of NMAH exhibits a shift of the carbonyl
stretching frequency (1606 cm−1 for the neat oil) to higher
(1632 cm−1) and lower (1597 cm−1) wavenumbers upon formation
of the mono- and bis-chelated complexes 1 (Fig. S8) and 2 (Fig. S9),
)
a
Atom belonging to a bound nitrate anion (O3I).
Atom belonging to a bound water molecule.
Bite angle.
b
c
Please cite this article as: S. Brandès, et al., Conformational and structural studies of N-methylacetohydroxamic acid and of its mono- and bis-
chelated uranium(VI) complexes, J. Inorg. Biochem. (2015), http://dx.doi.org/10.1016/j.jinorgbio.2015.06.002

10
S. Brandès et al. / Journal of Inorganic Biochemistry xxx (2015) xxx–xxx
respectively. While a bathochromic shift of ν(CO) is expected with re-
spect to the unbound ligand, as observed for 2 in the crystalline state,
compound 1 gives rise to an apparent blue shift. However, the ν(CO) vi-
brational frequency of NMAH is most likely impacted by hydrogen
bonding interactions involving the carbonyl group. For that reason,
this band seems not to be as informative as expected. The situation is
even less clear for other relevant bands, such as ν(CN) and ν(NO), as
they are much more difficult to identify.
In turn, the symmetric (ν1) and asymmetric (ν3) stretching modes
of the linear uranyl cation are respectively Raman and IR active accord-
ing to point-group theory. Experimentally, the Raman spectrum of the
crystalline uranium nitrate salt [UO2(NO3)2(H2O)2]·4H2O taken as a
reference evidences the ν1 vibrational mode at 869 cm−1 (Fig. S7),
whereas the ATR-FTMIR spectrum (Fig. S6) is dominated by the intense
ν3 band at 941 cm−1 flanked by a very weak peak at 868 cm−1 corres-
ponding to the forbidden ν1 stretch (Table 5), in accordance with
previous findings [95].
Ligand-induced shifts of the symmetric uranyl stretch were investi-
gated by recording Raman scattering spectra for X-ray quality crystals of
1 and 2 under a microscope (Fig. 5 and Table 5). A 633 nm He-Ne laser
was selected to avoid fluorescence effects. The spectrum collected for
[UO2(NMA)(NO3)(H2O)2] (1) evidences two diagnostic bands at 858
and 1042 cm−1. The first one is assigned to the ν1 vibrational mode,
which is also visible in the ATR-FTMIR spectrum (ν1 = 859 cm−1
)
together with the ν3 band at 934 cm−1. Both uranyl stretches are
affected likewise with respect to the reference values measured for
[UO2(NO3)2(H2O)2]·4H2O, as the vibrational energies are lowered by
11 and 7 cm−1, respectively. These data match very well with those
reported for [UO2(PBA)(THF)2Cl] by Smith and Raymond [61]. The
other Raman signal at 1042 cm−1 originates from the coordinated
nitrate anion. The symmetrical vibrational mode of free nitrates, like
in KNO3, is typically observed at 1050 cm−1 [96]. The slight shift by
8 cm−1 to lower energies of ν1(NO−3 ) found for 1 is in line with the
rather weak monodentate binding interaction with the uranium center
−
[U–ONO = 2.474(3) Å] revealed by the crystallographic study. It is also
3
worth to note that the ν1(NO3−) resonance appears at 1042 cm−1 for
Fig. 5. Raman spectra in the uranyl ν1 region of (a) the mono-chelated
[UO2(NMA)(NO3)(H2O)2] (1) and (b) the bis-chelated [UO2(NMA)2(H2O)] (2) complexes
recorded on crystalline material. The bands at 1042 (a) and 1050 cm−1 (b) correspond to
the bound and free (residual impurity) nitrate anions, respectively.
[UO2(NO3)2(H2O)2]·4H2O, although the nitrates are chelating UO22+
in a bidentate fashion but with slightly longer U–O distances [U–
−
ONO = 2.51(1) and 2.52(2) Å]. Replacing one water molecule and the
3
nitrate anion in 1 for a second bidentate NMA− ligand in 2, lowers ν1
further by 30 cm−1, the band occurring at 828 cm−1 in the Raman
spectrum of the bis-chelated complex (830 cm−1 in the FTMIR
spectrum). In turn, the asymmetrical vibrational mode ν3 of UO22+ is
not visible in the Raman spectra recorded for compounds 1 and 2, in
agreement with the selection rules, but gives rise to intense IR bands
at 934 and 896 cm−1, respectively.
Vibrational spectroscopy appears to be a sensitive tool for probing
the equatorial ligand-field strength, as both ν1 and ν3 modes undergo
significant bathochromic shifts of similar magnitude upon complex
formation. The U=O oscillator strength is influenced by several factors
among them, the electron-donation ability of the equatorial ligands,
the coordination number and ligand denticity, as well as the overall
Table 5
Summary of the U_O bond distances, the calculated and measured wavenumbers for the symmetric (ν1) and asymmetric (ν3) stretches of the uranyl cation in nitrato and hydroxamato
complexes characterized by X-ray crystallography.
Complex
U_O (Å)
ν
1 calcd. (cm−1
)
ν
1 exp. (cm−1
)
ν
3 calcd. (cm−1
)
ν
3 exp. (cm−1
)
Ref
[UO2(NO3)2(H2O)2]·4H2O
1.75(2)a
863b
869c
868e
858c
859e
873e
828c
830e
n. r.f
n. r.f
n. r.f
950b
941d
This work
[UO2(NMA)(NO3)(H2O)2]
1.770(3)a
841b
921b
934d
This work
[UO2(PBA)(THF)2Cl]
[UO2(NMA)2(H2O)]
1.75(3)a
863b
826b
950b
899b
934d
896d
[61]
This work
1.785(4)a
[UO2(PBA)2(MeOH)]
[UO2(PBA)2(DMSO)]
[UO2(PBA)2(DMF)]
1.75(1)a
1.777(6)a
1.78(2)a
863b
834b
831b
950b
911b
906b
n. r.f
910d
895d
[62]
[63]
[64]
a
Average crystallographic distance.
Value estimated using the empirical equation of Bartlett and Cooney, see Supporting information.
Value measured by Raman spectroscopy.
Value measured by FTMIR spectroscopy.
Value measured by FTMIR spectroscopy (weak band corresponding to the forbidden transition).
Not reported.
b
c
d
e
f
Please cite this article as: S. Brandès, et al., Conformational and structural studies of N-methylacetohydroxamic acid and of its mono- and bis-
chelated uranium(VI) complexes, J. Inorg. Biochem. (2015), http://dx.doi.org/10.1016/j.jinorgbio.2015.06.002

S. Brandès et al. / Journal of Inorganic Biochemistry xxx (2015) xxx–xxx
11
charge of the complex. Enriching the electron density at the uranium
center by strong electron donating ligands is intended to destabilize
the uranyl HOMO σu orbital and, consequently, to weaken the U_O
bond order as well as its vibrational energy (See Supporting Information
for further details). In turn, out-of-plane distortion and O_U_O bend-
ing seem not to exert a noticeable effect on the electronic proper-
ties [97]. Chelates 1 and 2 are both pentacoordinated neutral species
with a first coordination shell of similar geometry, but differing in the
number of bound NMA− ligands. However, it turns out that the ob-
served ν1 and ν3 values for [UO2(NO3)2(H2O)2]·4H2O, complexes 1
and 2 do not linearly depend upon the number of bound bidentate
hydroxamate moieties. Albeit weakly coordinated, the nitrate anion in
1 is expected to donate some more electron density to the uranium cen-
ter than a neutral water molecule would provide in its place. Based on
that consideration, the individual contribution of each bidentate
NMA− anion to the overall ν1 and ν3 shifts with respect to the free ura-
nyl nitrate salt is expected to be slightly less than ca. 10 cm−1 upon for-
mation of a mono-chelated complex, but close to 20 cm−1 in the case of
the bis-chelated compound 2 if an identical donating power of both li-
gands is assumed.
hydroxamate receptors in aqueous or organic solutions. We are current-
ly probing the practicality of this approach for determining metal-
binding equilibrium constants.
Acknowledgments
This work was financially supported by the Centre National de la
Recherche Scientifique (CNRS), the Conseil Régional de Bourgogne
(programs PARI IME SMT8 and PARI II CDEA), the Groupement
National de Recherche (GNR) PARIS and the program “Défi NEEDS
Environnement” (project ACTISOL). A. S. thanks the Conseil Régional
de Bourgogne for a post-doctoral fellowship (grant n° 2012 —
9201AAO049S02778). This work was granted access to the HPC
resources of CINES and IDRIS under the allocation 2014086933 made
by GENCI (Grand Equipement National de Calcul Intensif). We also
thank the CRI (Centre de Ressources Informatiques) of the Université
de Lille for providing computing time for part of the calculations.
Marie-José Penouilh (ICMUB) is acknowledged for her assistance in
NMR measurements.
As a matter of fact, the Raman and IR resonances for both bidentate
mono- and bis-hydroxamato coordination schemes are easily distin-
guishable as the symmetric ν1 and asymmetric ν3 vibrational modes
of UO22+ are lowered by ca. 10–40 cm−1 with respect to those found
for [UO2(NO3)2(H2O)2]·4H2O, while the signals are well resolved (the
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.jinorgbio.2015.06.002.
ν
1 bandwidths at half-height are less than 10 cm−1). This opens the
References
possibility of identifying both environments in solutions for equilibrium
mixtures of mono- and bis-chelated complexes.
[1] M.J. Miller, Chem. Rev. 89 (1989) 1563–1579.
[2] R. Codd, Coord. Chem. Rev. 252 (2008) 1387–1408.
[3] D.J. Raines, T.J. Sanderson, E.J. Wilde, A.K. Duhme-Klair, Siderophores, Reference
Module in Chemistry, Molecular Sciences and Chemical Engineering, Elsevier, 2015.
[4] A.M. Albrecht-Gary, A.L. Crumbliss, Coordination chemistry of siderophores:
thermodynamics and kinetics of iron chelation and release, in: Metal Ions in Biolog-
ical Systems, A. Sigel, H. Sigel (Eds.), Iron Transport and Storage in Microorganisms,
Plants, and Animals, vol. 35, Marcel Dekker, New York 1998, pp. 239–327.
[5] A. Stintzi, K.N. Raymond, Siderophore chemistry, in: D.E. Templeton (Ed.),
Molecular and Cellular Iron Transport, New York, Marcel Dekker 2001, pp. 273–319.
[6] E.A. Dertz, K.N. Raymond, Biochemical and physical properties of siderophores, in:
J.H. Crosa, A.R. Mey, S. Payne (Eds.), Iron Transport in Bacteria, ASM Press,
Washington, DC 2004, pp. 5–17.
4. Conclusion
A detailed conformational analysis of the NMAH and of its conjugat-
ed base NMA− has been undertaken by combining 1H NMR measure-
ments in D2O solutions with DFT calculations in order to close a
controversy lasting for several decades about the relative stability of
both E and Z rotamers. Unambiguous assignment of the methyl reso-
nances has been substantiated by the calculation of the chemical shifts
using theoretical methods, enabling to ascertain the slightly higher sta-
bility of the E conformer at room temperature in both neutral and an-
ionic states. Owing to the higher electrostatic repulsion energy
between both cis-oriented oxygen atoms of NMA− with respect to
NMAH, deprotonation of the N–OH group results in a shift of the Z ⇌ E
equilibrium towards the E form, with a concentration ratio that reaches
almost 10. Rotation barriers and exchange rate constants were derived
at 300 K by means of 1H EXSY spectroscopy. Results evidence a signifi-
cant slow-down of the hindered rotation around the C–N bond in the
deprotonated state. The complete set of kinetic parameters
determined in the present work for the cis/trans isomerism will be
highly valuable for interpreting forthcoming mechanistic metal-
complex formation or dissociation studies (i.e., assignment of the rate-
determining step). Since NMAH can be considered as the simplest
model for the hydroxamic binding units found in ubiquitous soil
siderophores, such as desferrioxamine B, these data are highly signifi-
cant for a deeper understanding of the role played by these natural
chelators in mobilizing environmentally-relevant metal cations. As
uranium(VI) belongs to our targets, preliminary crystallographic and
vibrational spectroscopic investigations enabled to structurally charac-
terize the mono- [UO2(NMA)(NO3)(H2O)2] and bis-chelated [UO2
(NMA)2(H2O)] complexes. In both structures, the uranyl center is
pentacoordinated in the essentially flat equatorial plane, thus evidenc-
ing no steric congestion between the bidentate NMA− motifs and the
additional monodentate ligands (H2O and/or NO−3 ). Interestingly, the
mono- and bis-chelated species are easily distinguishable by their IR
and Raman signatures. Using the wavenumbers measured in the crystal
state as benchmarks, it can be anticipated that Raman spectroscopy
would be a valuable speciation tool for studying uranyl uptake by
[7] M. Sandy, A. Butler, Chem. Rev. 109 (2009) 4580–4595.
[8] M. Whittaker, C.D. Floyd, P. Brown, A.J.H. Gearing, Chem. Rev. 99 (1999) 2735–2776.
[9] E.M.F. Muri, M.J. Nieto, R.D. Sindelar, J.S. Williamson, Curr. Med. Chem. 9 (2002)
1631–1653.
[10] B. Lou, K. Yang, Mini-Rev. Med. Chem. 3 (2003) 609–620.
[11] P. Jewula, J.-C. Chambron, M.-J. Penouilh, Y. Rousselin, M. Meyer, RSC Adv. 4 (2014)
22743–22754.
[12] A. Pathak, V.L. Blair, R.L. Ferrero, M. Mehring, P.C. Andrews, Chem. Commun. 50
(2014) 15232–15234.
[13] A. Pacco, T.N. Parac-Vogt, E. van Besien, K. Pierloot, C. Görller-Walrand, K.
Binnemans, Eur. J. Inorg. Chem. 2005 (2005) 3303–3310.
[14] G. Mezei, C.M. Zaleski, V.L. Pecoraro, Chem. Rev. 107 (2007) 4933–5003.
[15] J. Jankolovits, C.M. Andolina, J.W. Kampf, K.N. Raymond, V.L. Pecoraro, Angew. Chem.
Int. Ed. 50 (2011) 9660–9664.
[16] M. Tegoni, M. Remelli, Coord. Chem. Rev. 256 (2012) 289–315.
[17] Z. Li, H. Yamamoto, Acc. Chem. Res. 46 (2012) 506–518.
[18] Y.K. Agrawal, S.A. Patel, Rev. Anal. Chem. 4 (1980) 237–278.
[19] J.D. Glennon, M.R. Woulfe, A.T. Senior, N. NiChoileain, Anal. Chem. 61 (1989)
1474–1478.
[20] B. Bodenant, F. Fages, M.H. Delville, J. Am. Chem. Soc. 120 (1998) 7511–7519.
[21] J. Telegdi, T. Szabó, L. Románszki, M. Pávai, The use of nano-/microlayers, self-
healing and slow-release coatings to prevent corrosion and biofouling, in: A.S.H.
Makhlouf (Ed.), Handbook of Smart Coatings for Materials Protection, Woodhead
Publishing, Cambridge 2014, pp. 135–182 (Chap. 7).
[22] T.C. Hughes, US Patent 7007805, 2004.
[23] M.J. Carrott, O.D. Fox, C.J. Maher, C. Mason, R.J. Taylor, S.I. Sinkov, G.R. Choppin,
Solvent Extr. Ion Exch. 25 (2007) 723–745.
[24] P. Tkac, A. Paulenova, G.F. Vandegrift, J.F. Krebs, J. Radioanal. Nucl. Chem. 280 (2009)
339–342.
[25] I. May, R.J. Taylor, I.S. Denniss, G. Brown, A.L. Wallwork, N.J. Hill, J.M. Rawson, R. Less,
J. Alloys Compd. 275–277 (1998) 769–772.
[26] P. Govindan, S. Sukumar, R.V. Subba Rao, Desalination 232 (2008) 166–171.
[27] S.A. Matlin, P.G. Sammes, R.M. Upton, J. Chem. Soc. Perkin Trans. 1 (1979)
2481–2487.
[28] P. Jewula, J.-C. Berthet, J.-C. Chambron, Y. Rousselin, P. Thuéry, M. Meyer, Eur. J.
Inorg. Chem. (2015) 1529–1541.
[29] L. Bauer, O. Exner, Angew. Chem. Int. Ed. Engl. 13 (1974) 376–384.
[30] E. Lipczynska-Kochany, H. Iwamura, J. Org. Chem. 47 (1982) 5277–5282.
Please cite this article as: S. Brandès, et al., Conformational and structural studies of N-methylacetohydroxamic acid and of its mono- and bis-
chelated uranium(VI) complexes, J. Inorg. Biochem. (2015), http://dx.doi.org/10.1016/j.jinorgbio.2015.06.002

12
S. Brandès et al. / Journal of Inorganic Biochemistry xxx (2015) xxx–xxx
[31] D.A. Brown, W.K. Glass, R. Mageswaran, B. Girmay, Magn. Reson. Chem. 26 (1988)
970–973.
[32] D.A. Brown, W.K. Glass, R. Mageswaran, S.A. Mohammed, Magn. Reson. Chem. 29
[64] D.K. Hazra, S. Dinda, M. Helliwell, R. Bhattacharyya, M. Mukherjee, Z. Kristallogr. 224
(2009) 544–550.
[65] M. Hojjatie, S. Muralidharan, P.S. Bag, C.G. Panda, H. Freiser, Iran. J. Chem. Chem. Eng.
(1991) 40–45.
14 (1995) 81–89.
[66] J.T. Welch, K.W. Seper, J. Org. Chem. 53 (1988) 2991–2999.
[67] B.H. Lee, M.J. Miller, J. Org. Chem. 48 (1983) 24–31.
[68] R. Sulsky, J.P. Demers, Tetrahedron Lett. 30 (1989) 31–34.
[69] R. Grigg, Z. Rankovich, M. Thoroughgood, Tetrahedron 56 (2000) 8025–8032.
[70] P.K. Glasoe, F.A. Long, J. Phys. Chem. 64 (1960) 188–190.
[33] D.A. Brown, R.A. Coogan, N.J. Fitzpatrick, W.K. Glass, D.E. Abukshima, L. Shiels, M.
Ahlgren, K. Smolander, T.T. Pakkanen, T.A. Pakkanen, M. Perakyla, J. Chem. Soc.
Perkin Trans. 2 (1996) 2673–2679.
[34] D.A. Brown, L.P. Cuffe, G.M. Fitzpatrick, N.J. Fitzpatrick, W.K. Glass, K.M. Herlihy,
Collect. Czech. Chem. Commun. 66 (2001) 99–108.
[35] M.T. Caudle, A.L. Crumbliss, Inorg. Chem. 33 (1994) 4077–4085.
[36] M. Birus, M. Gabricevic, O. Kronja, B. Klaic, R. van Eldik, A. Zahl, Inorg. Chem. 38
(1999) 4064–4069.
[71] H.E. Gottlieb, V. Kotlyar, A. Nudelman, J. Org. Chem. 62 (1997) 7512–7515.
[72] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G.
Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li,
H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M.
Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J.A. Montgomery, J.E. Peralta, F. Ogliaro, M.J. Bearpark, J.
Heyd, E.N. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K.
Raghavachari, A.P. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, N.J.
Millam, M. Klene, J.E. Knox, 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, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J.
Dannenberg, S. Dapprich, A.D. Daniels, Ö. Farkas, J.B. Foresman, J.V. Ortiz, J.
Cioslowski, D.J. Fox, Gaussian 09, Gaussian, Inc., Wallingford, CT, USA, 2009.
[73] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 3865–3868.
[74] C. Adamo, V. Barone, J. Chem. Phys. 110 (1999) 6158–6170.
[75] W.J. Hehre, R. Ditchfield, J.A. Pople, J. Chem. Phys. 56 (1972) 2257–2261.
[76] P.C. Hariharan, J.A. Pople, Theor. Chim. Acta 28 (1973) 213–222.
[77] T. Clark, J. Chandrasekhar, G.W. Spitznagel, P. von Ragué Schleyer, J. Comput. Chem.
4 (1983) 294–301.
[37] M.L. Senent, A. Niño, C.M. Caro, S. Ibeas, B. García, J.M. Leal, F. Secco, M. Venturini, J.
Org. Chem. 68 (2003) 6535–6542.
[38] B. García, S. Ibeas, A. Muñoz, J.M. Leal, C. Ghinami, F. Secco, M. Venturini, Inorg.
Chem. 42 (2003) 5434–5441.
[39] B. García, S. Ibeas, J.M. Leal, F. Secco, M. Venturini, M.L. Senent, A. Niño, C. Muñoz,
Inorg. Chem. 44 (2005) 2908–2919.
[40] B. García, F. Secco, S. Ibeas, A. Muñoz, F.J. Hoyuelos, J.M. Leal, M.L. Senent, M.
Venturini, J. Org. Chem. 72 (2007) 7832–7840.
[41] A. Kaczor, L.M. Proniewicz, J. Mol. Struct. 704 (2004) 189–196.
[42] A. Tsatsaroni, G. Zoidis, P. Zoumpoulakis, A. Tsotinis, M.C. Taylor, J.M. Kelly, G. Fytas,
Tetrahedron Lett. 54 (2013) 3238–3240.
[43] F. Guérard, Y.-S. Lee, R. Tripier, L.P. Szajek, J.R. Deschamps, M.W. Brechbiel, Chem.
Commun. 49 (2013) 1002–1004.
[44] S.P. Sippl, H.L. Schenck, Magn. Reson. Chem. 51 (2013) 72–75.
[45] M. Birus, M. Gabricevic, O. Kronja, B. Klaic, Inorg. Chem. 34 (1995) 3110–3113.
[46] B. Kurzak, H. Kozłowski, E. Farkas, Coord. Chem. Rev. 114 (1992) 169–200.
[47] A. Bagno, C. Comuzzi, G. Scorrano, J. Am. Chem. Soc. 116 (1994) 916–924.
[48] M. Remko, P. Mach, P. von Ragué Schleyer, O. Exner, J. Mol. Struct. THEOCHEM 279
(1993) 139–150.
[78] J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 105 (2005) 2999–3094.
[79] K. Wolinski, J.F. Hinton, P. Pulay, J. Am. Chem. Soc. 112 (1990) 8251–8260.
[80] C.L. Perrin, T.J. Dwyer, Chem. Rev. 90 (1990) 935–967.
[49] O.N. Ventura, J.B. Rama, L. Turi, J.J. Dannenberg, J. Am. Chem. Soc. 115 (1993)
[81] Bruker, SAINT: area-detector integration sofware, Madison, Wisconsin, USA, 2013.
5754–5761.
[82] G.M. Sheldrick, Acta Crystallogr. A 71 (2015) 3–8.
[83] G.M. Sheldrick, Acta Crystallogr. C 71 (2015) 3–8.
[84] O.V. Dolomanov, L.J. Bourhis, R.J. Gildea, J.A.K. Howard, H. Puschmann, J. Appl.
[50] O.N. Ventura, J.B. Rama, L. Turi, J.J. Dannenberg, J. Phys. Chem. 99 (1995) 131–136.
[51] L.J. Yamin, C.A. Ponce, M.R. Estrada, F.T. Vert, J. Mol. Struct. THEOCHEM 360 (1996)
109–117.
Crystallogr. 42 (2009) 339–341.
[52] D.-H. Wu, J.-J. Ho, J. Phys. Chem. A 102 (1998) 3582–3586.
[53] J. El Yazal, Y.-P. Pang, J. Phys. Chem. A 103 (1999) 8346–8350.
[54] C. Muñoz-Caro, A. Niño, M.L. Senent, J.M. Leal, S. Ibeas, J. Org. Chem. 65 (1999)
405–410.
[55] R. Kakkar, R. Grover, P. Chadha, Org. Biomol. Chem. 1 (2003) 2200–2206.
[56] N.J. Fitzpatrick, R. Mageswaran, Polyhedron 8 (1989) 2255–2263.
[57] L. Turi, J.J. Dannenberg, J. Rama, O.N. Ventura, J. Phys. Chem. 96 (1992) 3709–3712.
[58] M. Remko, J. Phys. Chem. A 106 (2002) 5005–5010.
[59] D.C. Edwards, S.B. Nielsen, A.A. Jarzęcki, T.G. Spiro, S.C.B. Myneni, Geochim.
Cosmochim. Acta 69 (2005) 3237–3248.
[60] D. Kaur, R. Kohli, Int. J. Quantum Chem. 108 (2008) 119–134.
[61] W.L. Smith, K.N. Raymond, J. Inorg. Nucl. Chem. 41 (1979) 1431–1436.
[62] U. Casellato, P.A. Vigato, S. Tamburini, R. Graziani, M. Vidali, Inorg. Chim. Acta 81
(1984) 47–54.
[85] A.H. Blatt, Org. Synth. 2 (1943) 67.
[86] Y. Endo, T. Uchida, S. Hizatate, K. Shudo, Synthesis (1994) 1096–1105.
[87] G. Schäfer, J.W. Bode, Angew. Chem. Int. Ed. 50 (2011) 10913–10916.
[88] J. Yang, P.J. Bremer, I.L. Lamont, A.J. McQuillan, Langmuir 22 (2006) 10109–10117.
[89] P. Sinha, S.E. Boesch, C. Gu, R.A. Wheeler, A.K. Wilson, J. Phys. Chem. A 108 (2004)
9213–9217.
[90] M. Lagrelette, Université de Bourgogne, Dijon, France, 2008.
[91] C.L. Perrin, R.K. Gipe, J. Am. Chem. Soc. 106 (1984) 4036–4038.
[92] J.C. Taylor, M.H. Mueller, Acta Crystallogr. 19 (1965) 536–543.
[93] J. Casellato, P.A. Vigato, S. Tamburini, R. Graziani, M. Vidali, Inorg. Chim. Acta 72
(1983) 141–147.
[94] D. Cremer, J.A. Pople, J. Am. Chem. Soc. 97 (1975) 1354–1358.
[95] C. Caville, H. Poulet, J. Inorg. Nucl. Chem. 36 (1974) 1581–1587.
[96] R. Frech, J.P. Devlin, Chem. Phys. Lett. 38 (1976) 79–82.
[63] S. Chakraborty, S. Dinda, R. Bhattacharyya, A.K. Mukherjee, Z. Kristallogr. 221 (2006)
[97] M.J. Sarsfield, M. Helliwell, J. Raftery, Inorg. Chem. 43 (2004) 3170–3179.
606–611.
Please cite this article as: S. Brandès, et al., Conformational and structural studies of N-methylacetohydroxamic acid and of its mono- and bis-
chelated uranium(VI) complexes, J. Inorg. Biochem. (2015), http://dx.doi.org/10.1016/j.jinorgbio.2015.06.002