Alkaline-earth metal carboxylates characterized by 43Ca and 87Sr solid-state NMR: Impact of metal-amine bonding

Article abstract of DOI:10.1021/ic402658d

A series of calcium and strontium complexes featuring aryl carboxylate ligands has been prepared and characterized by alkaline-earth (⁴³Ca and ⁸⁷Sr) solid-state NMR experiments in a magnetic field of 21.1 T. In the 11 compounds studied as part of this work, a range of coordination motifs are observed including nitrogen atom binding to Ca²⁺ and Sr ²⁺, a binding mode which has not been investigated previously by ⁴³Ca or ⁸⁷Sr solid-state NMR. ⁴³Ca isotopic enrichment has enabled the full characterization of the ⁴³Ca electric field gradient (EFG) and chemical shift tensors of the two calcium sites in calcium p-aminosalicylate (Ca(pams)), where both NMR interactions are affected by the presence of a nitrogen atom in the first coordination sphere of one of the metal sites. The ⁴³Ca isotropic chemical shift is sensitive to the Ca-N distance as exemplified by the NMR parameters of a second form of Ca(pams) and density functional theory (DFT) calculations. Studies of the strontium analogue, Sr(pams), confirm a similar sensitivity of the ⁸⁷Sr EFG tensor to the presence or absence of nitrogen in the first coordination sphere. To our knowledge, this is the first systematic ⁸⁷Sr NMR study of strontium complexes featuring organic ligands. The |C_Q(⁸⁷Sr)| values are found to be sensitive to the coordination number about Sr²⁺. In general, this work has also established a larger data set of reliable experimental |C_Q( ⁴³Ca)| values which correlate well with those obtained using gauge-including projector-augmented-wave (GIPAW) DFT calculations. It is found that the use of a recently recommended quadrupole moment for ⁴³Ca, -44.4 mbarn, improves the agreement with experimental values. This contribution lays the groundwork for the interpretation of ⁴³Ca and ⁸⁷Sr NMR spectra of more challenging systems, particularly where nitrogen-alkaline earth metal bonding is occurring.

Full text of DOI:10.1021/ic402658d

Article
pubs.acs.org/IC
43
87
Alkaline-Earth Metal Carboxylates Characterized by Ca and Sr
Solid-State NMR: Impact of Metal-Amine Bonding
Kevin M. N. Burgess, Yang Xu, Matthew C. Leclerc, and David L. Bryce*
Department of Chemistry and Centre for Catalysis Research and Innovation, University of Ottawa, 10 Marie Curie Private, Ottawa,
Ontario, Canada K1N 6N5
S
* Supporting Information
ABSTRACT: A series of calcium and strontium complexes featuring aryl
carboxylate ligands has been prepared and characterized by alkaline-earth
(43Ca and 87Sr) solid-state NMR experiments in a magnetic field of 21.1 T.
In the 11 compounds studied as part of this work, a range of coordination
motifs are observed including nitrogen atom binding to Ca2+ and Sr2+, a
binding mode which has not been investigated previously by 43Ca or 87Sr
solid-state NMR. 43Ca isotopic enrichment has enabled the full character-
ization of the 43Ca electric field gradient (EFG) and chemical shift tensors of
the two calcium sites in calcium p-aminosalicylate (Ca(pams)), where both
NMR interactions are affected by the presence of a nitrogen atom in the first
coordination sphere of one of the metal sites. The 43Ca isotropic chemical
shift is sensitive to the Ca−N distance as exemplified by the NMR
parameters of a second form of Ca(pams) and density functional theory
(DFT) calculations. Studies of the strontium analogue, Sr(pams), confirm a similar sensitivity of the 87Sr EFG tensor to the
presence or absence of nitrogen in the first coordination sphere. To our knowledge, this is the first systematic 87Sr NMR study of
strontium complexes featuring organic ligands. The |CQ(87Sr)| values are found to be sensitive to the coordination number about
Sr2+. In general, this work has also established a larger data set of reliable experimental |CQ(43Ca)| values which correlate well with
those obtained using gauge-including projector-augmented-wave (GIPAW) DFT calculations. It is found that the use of a
recently recommended quadrupole moment for 43Ca, −44.4 mbarn, improves the agreement with experimental values. This
contribution lays the groundwork for the interpretation of 43Ca and 87Sr NMR spectra of more challenging systems, particularly
where nitrogen−alkaline earth metal bonding is occurring.
some inorganic materials may hamper the acquisition of NMR
spectra in a reasonable amount of time. Three recent reviews
on alkaline-earth metal SSNMR19−21 have aided in identifying
the major pitfalls encountered when performing calcium and
strontium NMR experiments. As summarized in Table 1,22
both NMR active nuclei (43Ca and 87Sr) have very low Larmor
frequencies (νL(43Ca) = 26.93 MHz and νL(87Sr) = 17.34 MHz
INTRODUCTION
■
The divalent calcium and strontium cations (Ca2+ and Sr2+)
have large abundances in Earth’s crust and play important roles
in many areas of chemistry, biochemistry, and materials science.
For example, both cations are often present in the active sites
associated with many biochemical processes including those in
oxygen-evolving complexes such as photosystem II1−3 and
metalloproteins.4,5 The alkaline-earth metal cations also play
key roles in tunable fluorescent indicators,6 ion sensing,7
macrocyclic ion receptors,8 and metal organic frameworks
(MOFs) for selective CO2 and N2 adsorption.9−11 Various
groups have discussed the economic benefit of Ca2+ and Sr2+
over transition metals as catalysts for C−C or C−N bond
formation,12,13 often with chiral nitrogen-binding ligands.14,15
The potential applications of calcium Grignard reagents16 and
allylcalcium complexes17,18 are also currently under inves-
tigation.
Table 1. Calcium and Strontium Nuclear Magnetic
a
Resonance Properties
property
43Ca
87Sr
spin quantum number, I
7/2
9/2
natural abundance/%
0.14
7.02
17.34
39.01
305
Larmor frequency at B0 = 9.4 T/MHz
Larmor frequency at B0 = 21.1 T/MHz
26.93
60.58
b
c
quadrupole moment, Q/mbarn
−40.8 or −44.4
a
b
Nuclear magnetic resonance properties are from ref 22. Quadrupole
Despite this plethora of applications, alkaline-earth metal
solid-state nuclear magnetic resonance (SSNMR) is still a
rather underdeveloped field specifically when attempting to
relate certain NMR signatures to the structure or activity of a
metal complex. In organic molecular environments, an
increased dilution factor for Ca2+ and Sr2+ ions compared to
c
moments are summarized in ref 23. An updated quadrupole moment
for 43Ca is reported in ref 67.
Received: October 22, 2013
Published: December 20, 2013
© 2013 American Chemical Society
552
dx.doi.org/10.1021/ic402658d | Inorg. Chem. 2014, 53, 552−561

Inorganic Chemistry
Article
when νL(1H) = 400.13 MHz). The main challenge in 43Ca
NMR spectroscopy is its natural abundance of 0.14%
(compared to 7.02% for 87Sr), which is one of the smallest of
all NMR-active nuclides in the periodic table. Powder NMR
line shapes are also broadened in the solid state since both 43Ca
and 87Sr are quadrupolar nuclei with spin quantum numbers, I,
of 7/2 and 9/2, respectively. The nuclear electric quadrupole
moment (Q), an intrinsic property of each nucleus, couples
with the electric field gradient (EFG) resulting in a quadrupolar
interaction (QI) that perturbs the Zeeman splitting and leads to
broadening of the central transition (CT) resonance. The
primary difficulty with 87Sr NMR spectroscopy stems from its
large quadrupole moment of 305 mbarn, whereas the currently
accepted Q value for 43Ca is much smaller (−40.8 mbarn).23
The very low receptivity of each of these nuclides means that it
is often impractical to record NMR spectra in a reasonable
amount of time at standard magnetic field strengths.
Advancements in NMR technologies have resulted in the
emergence of ultrahigh-field magnets (B0 > 18.8 T) and thus an
increase in the amenability of exotic low-γ nuclei such as 43Ca
and 87Sr to NMR experiments. An array of studies are available
in which Ca2+ ions present in biological materials have been
probed by 43Ca magic-angle spinning (MAS) NMR. These
include materials containing alkyl/aryl boronate24,25 and
phosphonate ligands,26 kidney stones27 and calcium oxalate
salts,28 and other various organic ligands.29 One impressive
study from Laurencin et al. on calcium benzoate trihydrate
(referred to in this work as Ca(ben)) employed 43Ca isotopic
labeling and experiments probing heteronuclear correlations
such as transfer of population double resonance (TRAPDOR)
between 43Ca and 13C nuclei.30 Successful studies on other
biologically relevant systems such as phosphates31 and
hydroxyapatite32,33 have paved the way for the study of more
challenging samples like bones, teeth,34 and cement.35 The 87Sr
nucleus, however, has not been studied as extensively because
of its larger quadrupole moment. The most recent study by
Bonhomme et al.36 employs wideband, uniform rate, and
smooth truncation (WURST) pulses coupled with quadrupolar
Carr-Purcell-Meiboom-Gill (QCPMG) acquisition and 87Sr
isotopic labeling at high magnetic field strengths to characterize
Sr2+ environments in Sr-doped Ca-silicate based bioglasses. In
addition, the authors examined a number of organic and
inorganic crystalline strontium phases.
Scheme 1. Ligands Coordinated to Alkaline-Earth Metals in
This Study
a
a
The abbreviations are as follows: benzoate (ben), p-fluorobenzoate
(pF), p-chlorobenzoate (pCl), salicylate (sal), p-aminosalicylate
(pams), p-nitrobenzoate (pNO2), and ethylenediaminetetraacetate
(EDTA).
examine Mg2+ centers featuring nitrogen coordination, for
example, in porphyrins42 and Mg3N2,43 although these studies
did not specifically investigate the role of nitrogen on the
alkaline earth metal NMR parameters.
We also corroborate our experimental findings with gauge-
including projector-augmented-wave (GIPAW) density func-
tional theory (DFT) calculations of the NMR parameters. This
combined experimental-theoretical approach in NMR has
found widespread usage especially in the field of alkaline-
earth metal SSNMR because of the difficulties outlined above
in interpreting NMR spectra of low-γ nuclei.20 Because of the
large data set of calcium compounds characterized by 43Ca
SSNMR in this study, in addition to currently available 43Ca
quadrupolar coupling constants (CQ), we are able to establish a
meaningful correlation between the calculated and experimental
CQ(43Ca) values.
EXPERIMENTAL DETAILS
■
Sample Preparation. Commercially available starting materials
were used for the syntheses presented in this work. p-Fluorobenzoic
acid (pF, ≥ 98.0%), p-chlorobenzoic acid (pCl, ≥ 99.0%), p-
nitrobenzoic acid (pNO2, ≥ 98.0%), sodium benzoate (Na(ben), ≥
99.5%), sodium salicylate (Na(sal), ≥ 99.5%), sodium p-amino-
salicylate (Na(pams), ≥ 97.0%), ethylenediaminetetraacetic acid
(H4EDTA, ≥ 98.5%), calcium carbonate (CaCO3, ≥ 99.0%),
strontium carbonate (SrCO3, ≥ 98.0%), calcium chloride (CaCl2),
strontium chloride (SrCl2, ≥ 96.0%), and calcium acetate (Ca(OAc)2·
H2O, ≥ 99.0%) were all purchased from Sigma-Aldrich and used
without further purification.
Studies to date involving 43Ca and 87Sr SSNMR have focused
mainly on oxygen coordination environments.20 For example,
researchers have correlated the average Ca−O distance of the
first coordination sphere to the 43Ca isotropic chemical shift,
δiso.37,38 There is, however, a relative dearth of studies involving
coordination of Ca2+ and Sr2+ to other atoms, namely, nitrogen.
Amino acids have been shown previously to bind to Ca2+ via
their nitrogen atoms,39−41 and so it is important to study such
environments if we are to understand how 43Ca and 87Sr NMR
spectroscopy can be useful in characterizing metalloproteins,
which are constituted mainly of nitrogen-containing amino
acids. In this work, we have prepared a series of Ca2+ and Sr2+
complexes containing various aryl and carboxylate ligands,
illustrated in Scheme 1. The p-aminosalicylate (pams) and
ethylenediaminetetraacetate (EDTA) ligands bind to the metal
centers via their nitrogen atoms. 43Ca and 87Sr SSNMR
experiments under stationary and/or MAS conditions are used
in an attempt to correlate the presence of coordinated nitrogen
atoms to changes in the alkaline-earth metal NMR parameters.
We note that 25Mg SSNMR has been used previously to
The abbreviations AE(ligand), where AE = Ca or Sr and the ligands
are presented in Scheme 1, are used for conciseness. Several of the
samples are hydrates and the actual molecular formulas for each
alkaline-earth metal complex are provided in the Supporting
Information, Table S1. Ca(pF), Ca(pCl), Sr(pCl), and Sr(pNO2)
were synthesized using a previously reported method by Arlin et al.44
wherein a 1:1 stoichiometric amount of substituted benzoic acid (1 g)
and metal carbonate were dissolved in 20 mL of distilled water. The
solution was stirred overnight, and the residual solids were filtered over
a pad of Celite. The collected mother liquor was slowly evaporated at
room temperature over several days or weeks to yield the appropriate
crystalline salt.
Natural isotopic abundance samples of Ca(sal), Ca(pams), Sr(ben),
Sr(sal), and Sr(pams) were prepared44 by dissolving the appropriate
amount of sodium salt (1 g) and alkaline-earth metal chloride in a 1:1
stoichiometric ratio in a minimum amount of distilled water. The
mixture was stirred for several minutes and filtered over cotton. The
resulting solution was slowly evaporated at room temperature until the
appearance of transparent crystals. Isotopically enriched 43Ca(pams)
was synthesized in a similar way as described above using 43CaCl2 (50
553
dx.doi.org/10.1021/ic402658d | Inorg. Chem. 2014, 53, 552−561

Inorganic Chemistry
Article
a
Table 2. Experimental and Calculated 43Ca NMR Parameters for Organic Calcium Complexes
c
compound
site
|CQ(43Ca)| /MHz
ηQ
δiso/ppm
Ω/ppm
κ
b
Ca(ben)
1.10(0.05)
1.12
0.7(0.1)
0.68
−2.5(0.2)
−7
15(1)
0.75(0.10)
38
0.59
Ca(pF)
1.2(0.1)
0.94
0.6(0.2)
0.75
−2.3(0.2)
−4.9
37
39
0.53
0.61
Ca(pCl)
Ca(sal)
1.3(0.1)
0.95
0.75(0.10)
0.85
−3.5(0.5)
−6.3
−4.0(0.5)
−5.9
1.55(0.10)
1.58
0.50(0.10)
0.13
42
−0.21
Ca(pams)
1
1.63(0.03)
1.61
0.35(0.05)
0.48
−3.2(0.5)
−6.0
−5.5(0.5)
−6.8
30(5)
37
−1.0(0.1)
−0.65
0.6(0.1)
1
d
2
2.60(0.05)
2.37
0.60(0.03)
0.64
65(5)
70
2
0.52
Ca(pams)(OAc)
1.15(0.05)
1.24
0.7(0.1)
0.39
5.4(0.2)
5.8
27
34
18
0.36
e
Ca2(EDTA)·7H2O
1
PQ = 0.8(0.2)
0.97
6.1(0.8)
3.4
1
0.31
0.75
−0.24
−0.20
e
2
PQ = 0.7(0.3)
0.99
15.1(0.4)
15.7
2
a
Errors are given in parentheses. Experimental 43Ca NMR parameters are in boldface and GIPAW DFT calculated results are italicized.
b
c
Experimental 43Ca NMR parameters are from ref 30. Calculations were performed as part of this work. The calculated σiso values were converted to
d
chemical shifts using the relationship found in ref 63 (i.e., δiso = −0.8032σiso + 901.6 ppm). For this site, it was possible to determine the Euler
angles which dictate the relative orientation of the EFG and CS tensor principal axis systems (α, β, γ = 210(20), 20(10), 110(10)°). For site 1 and
other compounds, these angles were set to (α, β, γ = 0, 0, 0°) as the simulated line shape was not sensitive the values used. Because of the absence
e
of discernible line shapes, PQ and δiso parameters were calculated using the method described in ref 20. An error of 1 ppm was used for both sites at
21.1 T and errors of 2 and 3 ppm (widths at approximately 90% peak height) were used for sites 1 and 2 respectively at 9.4 T. Standard deviations as
a result of these errors are in parentheses.
20
mg) and 2.05 equiv of Na(pams), which were dissolved in a minimum
amount of H2O. To synthesize 43CaCl2, a 43Ca-labeled polymer chelate
(donated by an industrial collaborator) was stirred in a 1 M HCl
solution. The nondissolved polymer was filtered away and, upon
solvent evaporation, a yellow oil was obtained. Twenty-five mg of this
oil along with 25 mg of natural abundance CaCl2 was used in the
synthesis. We estimate the enrichment level of 43Ca to be 4% based on
comparison with NMR experiments performed at natural abundance
for Ca(pams).
The Ca(pams)(OAc) analogue was synthesized in a similar way as
Ca(pams), with Ca(OAc)2·H2O as a source of calcium and 2 equiv of
Na(pams) (1 g).45 Ca2(EDTA)·7H2O was synthesized according to a
literature procedure.46 The phase purity of all synthesized alkaline-
earth metal carboxylates was verified by comparing experimental
powder X-ray diffraction patterns to those simulated with the available
crystal structures (see Supporting Information, Figures S1 and S2 for
Ca and Sr compounds, respectively).
M aqueous solution of CaCl2 or SrCl2. The CT-selective π/2 pulse
lengths were those of the standard scaled by a factor of 1/(I + 1/2) =
1/4 for 43Ca and 1/5 for 87Sr. Selective π/4 pulses were used in certain
cases for natural abundance 43Ca NMR experiments because of the
uncertainty in the longitudinal relaxation time constant, T1. In the case
of Ca2(EDTA)·7H2O, the pulse length was calibrated directly on the
sample.
Additional experiments were performed either at 9.4 T (Bruker
AVANCE III, νL(1H) = 400.13 MHz, νL(43Ca) = 26.9 MHz) with a
Bruker 7 mm MAS low-frequency double-resonance (HX) probe or at
11.75 T (Bruker AVANCE, νL(1H) = 500.13 MHz, νL(43Ca) = 33.7
MHz) with a Bruker 4 mm MAS low-frequency double-resonance
(HX) probe. In all cases, single-pulse experiments were used. The
chemical shift referencing and pulse calibrations were carried out
analogously to the procedure described above. Specific experimental
details pertaining to each compound in this work may be found in the
Supporting Information, Tables S2 and S3.
Spectral Processing and Simulations. All spectra were recorded
and processed using the Bruker TopSpin software (ver. 3.0). Most
43Ca NMR spectra were left-shifted a small number of data points to
remove signal due to probe ringing. The first echoes in all 87Sr
WURST−QCPMG experiments were removed also due to probe
ringing, and spectra were analyzed in magnitude mode in the
frequency domain. All individual pieces in the Solomon echo spectra
of Sr(pams) are presented in magnitude mode and coadded in the
frequency domain to yield the full spectrum. 43Ca and 87Sr spectra
were simulated using the WSolids1 program (ver. 1.20.15).50 Graphics
were prepared with the aid of DMFit (ver. 2011).51
Computational Details. GIPAW DFT calculations were
performed using ver. 4.1 of the CASTEP-NMR code (Accelrys Inc.,
San Diego, CA)52,53 as implemented in the Materials Studio (Accelrys
Inc.) software package (ver. 3.2). The available single crystal X-ray
structure crystallographic information files (.cif) for each compound
were used as input structures. The structures of Ca(pCl), Ca(pams),
Sr(ben), Sr(pCl), Sr(pNO2), and Sr(pams) are all from ref 44; those
of Ca(sal) and Sr(sal) were reported by Debuyst et al.;54 the structures
of Ca(pF)55 and Ca(ben)56 were previously reported. The structure of
Solid-State NMR Experiments. Solid-state NMR experiments
were performed on all calcium and strontium samples at the National
Ultrahigh-Field NMR Facility for Solids in Ottawa using a 21.1 T
magnet (Bruker AVANCE II, νL(1H) = 900.13 MHz, νL(43Ca) = 60.6
MHz, νL(87Sr) = 39.0 MHz). All 43Ca NMR experiments at natural
abundance were performed with a Bruker 7 mm MAS low-frequency
double-resonance (HX) probe using a single pulse MAS experiment.
In the case of 43Ca(pams), a Bruker 4 mm MAS low-frequency double-
resonance (HX) probe was used for one-pulse experiments under both
1
MAS and stationary conditions. Continuous wave H decoupling was
used in certain cases. Experiments for strontium samples were
performed using a home-built 7 mm low-frequency double-resonance
(HX) probe and the WURST−QCPMG47 pulse sequence with H
1
decoupling. For Sr(pams), it was necessary to perform a Solomon
echo experiment48 to observe a broader Sr2+ resonance because of its
smaller spin−spin relaxation time constant, T2. In this case, multiple
pieces were acquired with varying transmitter offset frequencies using
the variable offset cumulative spectra (VOCS) method.49 These
experiments were acquired using a modified Bruker broadband single
channel probe. Chemical shifts were referenced to 0.00 ppm with a 1.0
554
dx.doi.org/10.1021/ic402658d | Inorg. Chem. 2014, 53, 552−561

Inorganic Chemistry
Article
shows that both the |CQ(43Ca)| values (ranging from ∼0.7 MHz
in site 2 of Ca2(EDTA)·7H2O to 2.60 MHz in site 2 of
Ca(pams)) and δiso values (a total range of 22 ppm) change
considerably depending on the nature of the carboxylate ligand
bound to Ca2+. This observation is supported by GIPAW DFT
calculations (vide infra). Some specific insights obtained when
relating the 43Ca NMR parameters of the calcium carboxylates
to their respective crystal structures are outlined below.
Ca2(EDTA)·7H2O can be found in ref 46; some water molecules
present in the lattice (i.e., not bound to Ca2+) were assumed to have
full occupancy for the calculations as these are not expected to affect
greatly the calculated 43Ca NMR parameters. Our structure for
Ca(pams)(OAc), which has similar unit cell parameters to a previously
reported one,45 was used for the calculations. The Pnma space group
for Ca(pams)(OAc) has a mirror plane bisecting the aryl ring, which
creates disorder in the positioning of the hydroxyl group on the pams
ligand. The disorder was removed manually and a modified .cif file was
used for the calculations. The structures for Ca(D-gluconate)·5H2O
and Ca(C2O4)·3H2O are taken from references 57 and 58,
respectively. The generalized gradient approximation (GGA)59
exchange-correlation functional of Perdew, Burke, and Ernzerhof
(PBE) was used in all cases. On-the-fly generated pseudopotentials
were employed for all atoms. All hydrogen atom positions were
optimized prior to performing the NMR calculations. Many structures
required a “coarse” setting for kinetic energy cutoff values and k-point
grid because of the large unit cell volumes. Specific considerations for
each calculation may be found in the Supporting Information, Table
S4.
In Figure 2b is shown the natural abundance 43Ca MAS
NMR spectrum for Ca(sal) at 21.1 T. Also depicted in this
For nonperiodic DFT calculations, an isolated molecular model for
Ca(pams) was built where both Ca2+ sites are present and benzoate
ligands were converted to formate ligands. The nitrogen atom bound
to calcium was hydrogen capped to make an ammonia ligand. A
depiction of this model is found in Figure 5. The Gaussian ‘09 software
package60 was used for the DFT computations on the Ca(pams)
model using the B3LYP hybrid functional and the 6-31+G* basis set
on all atoms. Hydrogen positions were optimized prior to NMR
calculations. Multiple NMR calculations were performed where the
Ca−N distance was varied between 2.50 Å and 2.90 Å in 0.05 Å
increments. The EFGShield program61 was used to extract NMR
parameters from the Gaussian ‘09 and CASTEP-NMR output files.
Figure 2. Natural abundance 43Ca solid-state NMR spectra (in black)
for Ca(sal) (b) and Ca(pams) (a) recorded at B0 = 21.1 T under MAS
conditions (νrot = 5 kHz). A dashed blue rectangle highlights a second
broader 43Ca resonance. In red is the analytical simulation for Ca(sal)
in (b). The immediate coordination environment surrounding the
Ca2+ sites in both compounds is also depicted. Calcium atoms are
pink, oxygen atoms are red, and nitrogen atoms are blue.
RESULTS AND DISCUSSION
■
43Ca SSNMR Spectroscopy of Calcium Aryl Carbox-
ylates. In Table 2 are presented the experimental and GIPAW
DFT calculated 43Ca EFG tensor and isotropic chemical shift
parameters for all of the calcium carboxylate compounds
considered in this work. Because of the absence of a discernible
line shape at the higher magnetic field, some spectra were
acquired at two magnetic field strengths. As for Ca(pCl) and
Ca(pF), both structures pack similarly to Ca(ben), which has
already recently been extensively studied by 43Ca SSNMR.30 It
is therefore not surprising that these three compounds exhibit
similar experimental and calculated EFG tensor parameters (see
Figure 1 for the NMR spectra). Inspection of all data in Table 2
figure is the immediate coordination environment of each Ca2+
ion from the reported single crystal X-ray structure.54 The 43Ca
NMR spectrum for Ca(sal) is characterized by a single
resonance, consistent with one crystallographic Ca site, with a
|CQ(43Ca)| value of 1.55 MHz. This resonance is very similar in
shape to the more intense signal in the natural abundance 43Ca
NMR spectrum of Ca(pams) shown in Figure 2a. A qualitative
comparison between the first coordination spheres about both
Ca sites in Ca(pams) to that of Ca(sal) can be made: site 1 in
Ca(pams) is very similar to Ca(sal) (i.e., both are 8-coordinate
with oxygen atoms which form a distorted square face bicapped
trigonal prism geometry about Ca2+). It is also possible to
observe a second broader resonance at natural abundance for
Ca(pams) after quite a long experiment time of nearly 42 h
(indicated by a dashed rectangle in Figure 2a). At this point, it
was possible to assign this broader resonance to Ca site 2 in
Ca(pams), which contains a nitrogen atom in its first
coordination sphere. Since 43Ca SSNMR has not been used
previously to characterize Ca environments where nitrogen
binding is present, we sought to isotopically label this sample
(see Experimental Details) to accurately and precisely measure
the EFG and CS tensor parameters for both Ca sites in
Ca(pams) and relate these values to the calcium environments.
Both MAS and stationary NMR spectra acquired at 21.1 T,
along with a MAS spectrum at 11.75 T, are presented in Figure
3. From the MAS spectra, it is clear that two crystallographically
unique sites are present in the structure in a 1:1 ratio with
differing |CQ(43Ca)| values of 1.63 and 2.60 MHz for sites 1 and
2, respectively. Both sites, however, have very similar
experimental and calculated δiso values. It was also possible in
Figure 1. Natural abundance 43Ca solid-state NMR spectra (in black)
for Ca(pCl) and Ca(pF) recorded at B0 = 21.1 T under MAS
conditions (νrot = 5 kHz). Analytical simulations (see Table 1) are in
red. 9.4 T data are shown in the Supporting Information.
555
dx.doi.org/10.1021/ic402658d | Inorg. Chem. 2014, 53, 552−561

Inorganic Chemistry
Article
Figure 3. Experimental 43Ca solid-state NMR experiments on
isotopically enriched (∼4%) 43Ca(pams) acquired at B0 = 21.1 T
(b) and 11.75 T (a) under MAS conditions (νrot = 10 kHz) using
single pulse experiments. Their respective analytical simulations are in
red. In (e) is the 43Ca NMR spectrum of the same sample acquired
Figure 4. Natural abundance 43Ca solid-state NMR spectra recorded at
B0 = 9.4 (b and e) and 21.1 T (c and f) under MAS conditions (νrot
=
5 kHz) for Ca(pams)(OAc) [left] and Ca2(EDTA)·7H2O [right]
respectively. Analytical simulations are in red. The structure of
Ca(pams)(OAc) is depicted in (a), where a p-aminosalicylate ligand is
bound to Ca via its nitrogen atom. The immediate coordination
environment surrounding both Ca2+ sites in Ca2(EDTA) is in (d). The
two distinct Ca sites are clearly resolved for Ca2(EDTA). Calcium
atoms are pink, oxygen atoms are red, and nitrogen atoms are blue.
1
under stationary conditions with H decoupling at B0 = 21.1 T. The
analytical simulation for the stationary NMR experiment is in (d), and
a deconvolution is shown in blue and green for sites 1 and 2,
respectively. The number of parameters required for the simulations in
(d) was reduced as the quadrupolar parameters and δiso were
constrained to values obtained from MAS experiments. In (c) is the
line shape predicted from GIPAW DFT calculations.
light of the GIPAW DFT calculations to characterize the CS
tensor span, Ω, for both Ca sites in 43Ca(pams). Very few 43Ca
SSNMR studies have reported chemical shift tensor parameters
to date.19−21 Widdifield et al.62 have reported a Ω value of 31
ppm for anhydrous CaCl2 whereas very characteristic chemical
shift tensor parameters have been found for the three different
anhydrous polymorphs of CaCO3.63 Site 2 in Ca(pams) (with
N-coordination) exhibits one of the largest known Ω values of
65 ppm, whereas site 1 has a value of 30 ppm. Qualitatively, it is
therefore clear that N-binding ligands can significantly alter
both the 43Ca EFG and the CS tensor parameters relative to the
case where only oxygen atoms interact with the Ca2+ cation.
To further investigate how N-coordination may affect the
43Ca NMR parameters more generally, we have synthesized a
different Ca complex with the pams ligand and Ca(OAc)2·H2O
as a source of Ca2+ ions. Ca(pams)(OAc) has one crystallo-
graphically unique Ca site which is 8-coordinate, including one
N atom (see Figure 4a), like site 2 in Ca(pams). The OAc−
counteranions are also present in the lattice. Natural abundance
43Ca MAS spectra acquired at 9.4 and 21.1 T confirm the
presence of one Ca site (see Figures 4b and 4c respectively).
Interestingly, a large decrease in the |CQ(43Ca)| value (to 1.15
MHz) and an increase by approximately 10 ppm in the δiso
value (5.4 ppm) are observed compared to site 2 in Ca(pams).
Again, GIPAW DFT calculations (see Table 1) reproduce these
experimental changes.
featureless peaks that are clearly separated by approximately
10 ppm (see Figure 4f). Only the quadrupolar product,
|PQ(43Ca)| (i.e., PQ = CQ(1 + (ηQ)2/3)1/2), and the δiso values
could be quantified even after acquiring a MAS spectrum at 9.4
T (see Figure 4e). The NMR parameters are reported in Table
2, and a significant difference in δiso values between both Ca
sites (6.1 and 15.1 ppm for sites 1 and 2 respectively) is
observed and reproduced with GIPAW DFT calculations. It
was also possible to assign both resonances to the crystal
structure on the basis of chemical shifts; site 2, which contains
Ca−N bonds, corresponds to the more deshielded resonance.
It is not immediately clear how the 43Ca δiso values are
systematically affected by nitrogen coordination. In Ca(pams),
the Ca2+ site with a nitrogen ligand is shielded by 2.3 ppm
relative to the site which is fully coordinated by oxygen atoms,
while in Ca2(EDTA)·7H2O, a deshielding of 9.0 ppm is
observed for the site with two nitrogen ligands compared to the
site which is fully coordinated by oxygen atoms. We were
curious to see how changes in the Ca−N distance may affect
the NMR parameters, as Ca(pams) (dCa−N = 2.7085 Å) and
Ca(pams)(OAc) (dCa−N = 2.5813 Å) have δiso values of −5.5
and +5.4 ppm, respectively. Such a large difference in chemical
shifts cannot be entirely accounted for by the change in the
average Ca−O bond lengths for the two compounds (∼0.03 Å)
and the slope of −154 ppm/Å from model calculations on
calcium organic compounds.38 An isolated molecular model of
Ca(pams) was built (see Figure 5a) and, after geometry
optimization of the hydrogen atom positions, the Ca−N
distance, dCa−N, was varied systematically from 2.50 to 2.90 Å
(this is a representative range for Ca−N bond lengths in other
similar complexes39,64,65) and the changes in the calculated
NMR parameters were monitored. In Figure 5b is shown how
the individual principal components of the 43Ca CS tensor are
affected by changing dCa−N. First, it is evident that a large
The EDTA ligand is known for its binding of various cations
including alkaline-earth metals. In the case of Ca2+ chelation,
the crystal structure of the Ca2(EDTA)·7H2O complex has two
distinct Ca sites, where site 1 is fully coordinated by oxygen
atoms and site 2 contains two nitrogen atoms in its first
coordination sphere (see Figure 4d). The natural abundance
43Ca MAS NMR spectrum was acquired at 21.1 T in a relatively
short amount of experimental time (4 h) and has two
556
dx.doi.org/10.1021/ic402658d | Inorg. Chem. 2014, 53, 552−561

Inorganic Chemistry
Article
decrease in δ33 is observed compared to the changes in δ11 and
δ22. This in turn decreases the value of δiso as the Ca−N bond
length is increased. This demonstrates the sensitivity of 43Ca
NMR toward small structural differences within the first
coordination sphere of Ca2+ specifically when nitrogen atoms
are involved in metal binding. While the 43Ca chemical shift
cannot be easily used to determine the number of nitrogen
atoms in the first coordination sphere, it does appear to be
diagnostic of the calcium−nitrogen distance. To compare the
effect on the computed 43Ca chemical shift of changing a single
Ca−O distance vs a single Ca−N distance, the ammonia ligand
in the model shown in Figure 5 was replaced by a water
Figure 6. GIPAW DFT computed |CQ(43Ca)| values versus the
experimental values from this work and previously characterized Ca
environments for which reliable 43Ca SSNMR data are available (see
main text for relevant references). Two plots were generated wherein
the 43Ca quadrupole moment is −40.8 mbarn (|CQ(43Ca, calc.)| =
0.861|CQ(43Ca, exp.)|, R2 = 0.910) or −44.4 mbarn (|CQ(43Ca, calc.)| =
0.937|CQ(43Ca, exp.)|, R2 = 0.910). Only the latter of the linear
regressions is shown above for clarity. In both cases, the (0,0) data
point was included in the linear regression.
One plot was generated using each Q value (only the graph
where Q = −44.4 mbarn is shown in Figure 6). The slopes of
the linear regressions are 0.861 and 0.937 for Q = −40.8 and
−44.4 mbarn, respectively (recall, CQ = eV33Q/h). It is
therefore recommended to use the updated quadrupole
moment of −44.4 mbarn when performing calculations to
improve the accuracy of the predicted 43Ca EFG tensor
parameters. This analysis will serve as a useful tool in future
studies including those involving crystal structure refinements.
87Sr WURST−QCPMG NMR of Strontium Aryl Carbox-
ylates. Given the structural similarities between some calcium
and strontium complexes, as well as the dearth of 87Sr NMR
information for strontium complexes bearing organic ligands,
we next extended our work to the study of analogous strontium
aryl carboxylate complexes. In total, 87Sr SSNMR spectra were
recorded for five strontium compounds. Spectra for Sr(ben),
Sr(sal), Sr(pCl), and Sr(pNO2), along with their respective
analytical simulations, are shown in Figure 7. The experimental
87Sr EFG tensor parameters and chemical shifts are reported in
Table 3. We note that in all analytical simulations of 87Sr NMR
spectra, chemical shift anisotropy was not considered because
of the dominance of the quadrupolar interaction. All spectra
illustrated in Figure 7 were acquired using the WURST−
QCPMG method at a magnetic field strength of 21.1 T.
WURST pulses provided uniform and broadband excitation of
the entire CT line shape in one transmitter offset setting (in
our spectra, some line shapes spanned over 150 kHz at 21.1
T).47 The very low receptivity of the 87Sr nucleus prompted us
to use the QCPMG method, which leads to the acquisition of
multiple echoes in every scan and concentrates the signal
intensity into spikelets to provide a spectrum with an increased
signal-to-noise ratio when compared to the acquisition of a
single echo. Because of increased quadrupolar broadening of
the CT line shape and the low resonance frequency,
experiments at lower magnetic field strengths (such as 9.4 T)
were not feasible. The |CQ(87Sr)| values, which range from 11.7
MHz for Sr(pNO2) to 33.8 MHz for site 2 in Sr(pams), were
found to be particularly sensitive to the local Sr2+ environment.
Figure 5. Calculated principal components of the 43Ca chemical shift
tensor as a function of the Ca−N distance (B3LYP/6-31+G*) (part
(b)). A model system (part (a)) was built from the available crystal
structure of Ca(pams) wherein both Ca sites are present and
coordinated by water, formate, and ammonia ligands. All chemical
shifts are relative to a calculation on the original structure of
Ca(pams). An approximate range of 10 ppm for δiso is observed for a
range of Ca−N distances between 2.50 and 2.90 Å. δiso changes by
about −36 ppm/Å over this range. When visualizing the principal
tensor components with respect to the molecular frame, δ33 is
observed to be nearly perpendicular to the Ca−N bond vector, which
explains the larger change in the value of δ33.
molecule and the Ca−O distance was systematically varied over
a range of typical values (2.30 to 2.70 Å). The 43Ca chemical
shift was found to be approximately equally sensitive to the
varied distance for N-binding (−36 ppm/Å) and O-binding
(−29 ppm/Å).
As summarized in a recent review by Moudrakovski,21
reliable |CQ(43Ca)| values are scarce, and many studies report
only PQ rather than CQ values. Given the accurate character-
ization of the 43Ca EFG tensor parameters in a larger data set of
compounds as part of this work, a plot of the GIPAW DFT
calculated |CQ(43Ca)| values versus the experimentally
determined values yields a useful correlation (Figure 6). This
plot includes only reliable experimental |CQ(43Ca)| values
reported to date from the current and previous studies, which
are summarized in the Supporting Information, Table
S5.28,29,3133,35,62,63,66 Furthermore, only data for compounds
with high-quality structures from single-crystal X-ray or neutron
diffraction are included. A rather good correlation is observed
(R2 = 0.910), but GIPAW DFT calculations seem to
consistently underestimate the experimental values. Recently,
a new value of Q for 43Ca of −44.4 mbarn67 has been proposed
by combining the measured electric quadrupole hyperfine-
structure constant and relativistic coupled-cluster calculations.
557
dx.doi.org/10.1021/ic402658d | Inorg. Chem. 2014, 53, 552−561

Inorganic Chemistry
Article
Figure 7. Experimental 87Sr NMR spectra acquired at 21.1 T under stationary conditions using the WURST−QCPMG pulse sequence. Spikelet
separations are 2 kHz for all samples. Partially excited satellite transitions are visible beyond the high-frequency discontinuity in the spectrum of
Sr(ben). Red traces represent analytical simulations and NMR parameters are summarized in Table 3. Chemical shift anisotropy was not included in
the simulations.
Figure S3). The 87Sr NMR spectrum for Sr(ben) demonstrates
one of the limitations of the WURST−QCPMG method in
resolving sites which have very similar quadrupolar and
chemical shift parameters. Sr(sal) (P21 space group) is
characterized by a larger |CQ(87Sr)| value of 26.1 MHz
compared to Sr(pCl) (P1 space group) and Sr(ben) (C2/c
space group) with |CQ(87Sr)| values of 22.2 and 22.5 MHz,
respectively. Qualitative inspection of the coordination
geometry about Sr2+ in each of the three compounds does
not reveal any obvious reason for the moderate variability in the
87Sr quadrupolar coupling constants; indeed it has been shown
recently that geometrical considerations about the immediate
coordination sphere of Sr2+ alone do not fully account for
changes in the value of CQ(87Sr).36 However, GIPAW DFT
calculations prove to predict quite accurately these changes in
the EFG tensor parameters (see Table 3).
The spectrum of Sr(pNO2) is shown in Figure 7 and has the
narrowest 87Sr line shape considered in this work. It is
characterized by a much smaller |CQ(87Sr)| value of 11.7 MHz
and hence fewer spikelets define the CT line shape. GIPAW
DFT calculations also predict a lowering of the |CQ(87Sr)| value
for Sr(pNO2) relative to the compounds discussed above, but
the calculations overestimate the experimental datum by several
MHz (Table 3). This overestimation might be due to the
presence of a larger number of coordinated water molecules
(i.e., 7 in this case) about the Sr2+ cation compared to the other
compounds considered in this work. Dynamics of coordinated
water molecules are not accounted for in our calculations
(performed at 0 K) and have been shown to affect quadrupolar
parameters in previous SSNMR studies involving Mg2+
coordinated to organic-functionalized carboxylate ligands.68
More interestingly, Sr(pNO2) is 9-fold coordinated with oxygen
atoms whereas Sr(ben), Sr(sal), Sr(pCl), and site 1 of Sr(pams)
are all coordinated to 8 oxygen atoms. A lower value of
|CQ(87Sr)| for Sr(pNO2) is consistent with a few previously
characterized 9-oxygen coordinated strontium sites such as
those in SrCO3 (|CQ(87Sr)| = 8.91 MHz)69 and a micaceous
mineral (|CQ(87Sr)| = 9.02 MHz).70 However, the previously
studied strontium malonate36 is 9-coordinate71 and has a larger
|CQ(87Sr)| value of 31.5 MHz. Both oxygen coordination
polyhedra surrounding Sr(pNO2) and strontium malonate are
monocapped square antiprisms. The latter complex has,
Table 3. Experimental and Calculated 87Sr NMR Parameters
for Strontium Aryl Carboxylates
a
space
group
|CQ(87Sr)|
b
compound
Sr(ben)
site
/MHz
ηQ
δiso/ppm
C2/c
22.5(0.5)
20.7
0.87(0.02)
0.76
0(50)
−52
−58
1
2
20.8
0.84
Sr(pCl)
Sr(sal)
P1
22.2(0.2)
22.3
0.71(0.02)
0.69
0(50)
̅
−65
30(20)
−25
P21
26.1(0.3)
32.2
0.18(0.03)
0.45
Sr(pNO2)
Sr(pams)
P21/c
P21/n
11.70(0.3)
15.3
0.25(0.05)
0.31
10(20)
−30
−50(30)
−5
1
1
2
2
22.0(0.5)
22.0
0.92(0.02)
0.86
33.8(0.3)
0.85(0.03)
150(50)
35.1
0.65
−30
a
Errors are given in parentheses. Experimental 87Sr NMR parameters
are in boldface, and GIPAW DFT calculated results are italicized. The
calculated σiso values were converted to chemical shifts using the
relationship found in ref 36 (i.e., δiso = −0.8750σiso + 2574 ppm).
b
Isotropic chemical shifts could not be measured as precisely as
in our 43Ca experiments given the acquisition method and the
87Sr spectral breadth. The experimental errors in the analytical
simulations for δiso range from 20 to 50 ppm, which represents
10 to 25% of the total range (i.e., 200 ppm, see Table 3) of
chemical shifts characterized as part of this work. Therefore, the
establishment of meaningful correlations between δiso and the
Sr2+ coordination environment is not feasible.
The WURST−QCPMG 87Sr NMR spectrum of Sr(ben)
shows one discernible Sr site characterized by a |CQ(87Sr)| value
of 22.5(0.5) MHz and an asymmetry parameter, ηQ, of 0.87.
However, according to the X-ray crystal structure reported in
the literature,44 two distinct strontium sites are expected.
GIPAW DFT calculations on this structure reveal very similar
EFG tensor parameters for both sites, with |CQ(87Sr)| values of
20.7 and 20.8 MHz and ηQ values of 0.76 and 0.84, for
strontium sites 1 and 2, respectively. Structurally, both sites
have nearly identical oxygen coordination environments but are
mirror images of one another (see the Supporting Information,
558
dx.doi.org/10.1021/ic402658d | Inorg. Chem. 2014, 53, 552−561

Inorganic Chemistry
Article
qualitatively, a much larger degree of distortion, which explains
the large difference between the |CQ(87Sr)| values.
QCPMG experiment. The reason for this is likely a smaller T2
value for site 2. Instead, a solid echo experiment was
successfully used to acquire the spectrum of both strontium
sites (see Figure 8c). For this spectrum, multiple transmitter
offset settings were required to obtain the full CT line shape,
with a spectral breadth of nearly 400 kHz. The singularities in
this spectrum confirm the presence of a second Sr site with a
larger |CQ(87Sr)| value of 33.8 MHz. The observation of this
broader site is in agreement with the GIPAW DFT calculations
and thus the larger CQ value can be attributed to the presence
of a nitrogen atom coordinated to Sr2+ as was shown for the
Ca(pams) analogue. As shown with this example, care must be
taken when performing WURST−QCPMG experiments on Sr
samples as T2 values may affect our ability to observe Sr2+ in
certain chemical environments.
The final Sr compound in our series, Sr(pams), has a
molecular and crystal structure that resembles closely its
calcium analogue, Ca(pams) (see Supporting Information,
Figure S4 for a comparison). Both structures crystallize in the
P21/n space group and have two crystallographically unique
metal cation sites. As is the case for Ca(pams), site 1 in
Sr(pams) is fully coordinated by oxygen atoms, and site 2 has
seven oxygen atoms and 1 nitrogen atom in the first
coordination sphere. Our first attempt at acquiring this 87Sr
spectrum was in the same manner as for the other Sr
compounds in this study. The 87Sr WURST−QCPMG
spectrum is shown in Figure 8b and resembles quite closely
CONCLUSIONS
■
In summary, we have presented a 43Ca and 87Sr solid-state
NMR study on the Ca2+ and Sr2+ cations present in various
organic molecular environments which may serve as models for
metal organic frameworks, metal ion receptors, and biological
systems. The benzoate, salicylate, and ethylenediaminetetraa-
cetate (EDTA) ligands employed in this study allow for a
variety of coordination motifs to be probed by alkaline-earth
metal solid-state NMR spectroscopy including nitrogen binding
to the metal centers. It was shown with the examples of
Ca(pams), its derivative, Ca(pams)(OAc), and Sr(pams) that
both the electric field gradient and chemical shift tensors are
sensitive to the presence of alkaline earth metal−nitrogen
interactions. For example, DFT computations on a model for
Ca(pams) demonstrated an inverse correlation between the
calcium chemical shift and the Ca−N bond length which is
consistent with the trend observed experimentally for Ca-
(pams) and Ca(pams)(OAc). It was also possible to fully
characterize the 43Ca chemical shift tensors of both crystallo-
graphically unique sites in Ca(pams) with the aid of 43Ca
isotopic labeling. A structural assignment could be made with
the aid of GIPAW DFT computations that predicted larger
chemical shift anisotropy and |CQ(43Ca)| values for the Ca2+
interacting with a nitrogen atom in its first coordination sphere.
Many previous studies have been unsuccessful in fully
characterizing accurate 43Ca EFG tensor parameter (i.e., by
only reporting PQ values for example), thus prohibiting until
now the establishment of a meaningful correlation between
DFT calculated and experimental |CQ(43Ca)| values for future
crystal structure refinement purposes. Our study has also
allowed for the accurate characterization of both the 43Ca
quadrupolar coupling constants and asymmetry parameters.
The best agreement was observed between the GIPAW DFT
calculated |CQ(43Ca)| values and experiment when using the
newly proposed quadrupole moment for 43Ca, Q = −44.4
mbarn, suggesting that this value be used in such studies going
forward.
Figure 8. Experimental 87Sr WURST−QCPMG NMR spectrum of
Sr(pams) acquired at 21.1 T (b) and simulation of the GIPAW DFT
calculated 87Sr NMR spectrum under stationary conditions (a). Only
the NMR parameters for the narrower Sr line shape can be estimated
using this method. A stepped solid echo experiment was performed at
21.1 T (c, black trace) and a simulation with two strontium sites in a
1:1 ratio is shown in red. The deconvolution of the simulation from
(c) is depicted in (d) where site 1 is in blue and site 2 is in green.
Dashed lines indicate transmitter offset frequency settings for VOCS
data acquisition.
the spectrum observed for Sr(ben) with an asymmetry
parameter near unity and a |CQ(87Sr)| value of ∼22 MHz.
The observation of a single Sr site is in disagreement with the
reported crystal structure, which has two crystallographically
unique Sr2+ cations. Furthermore, unlike for Sr(ben), the
GIPAW DFT calculations predict substantially different
|CQ(87Sr)| values for Sr(pams); these are 22.0 and 35.1 MHz
for sites 1 and 2 respectively (see Figure 8a for a simulation
based on the calculated NMR parameters). The larger
|CQ(87Sr)| value is in agreement with what is observed for
Ca(pams) where site 2, which contains the alkaline-earth metal-
bound nitrogen atom, is characterized by a CQ value about 1.5
times larger than that for site 1. We were unable to observe the
second broader site in the 87Sr NMR spectrum even when
stepping the transmitter offset frequency in the WURST−
Finally, 87Sr WURST−QCPMG NMR was used to character-
ize a series of analogous organic strontium complexes. It was
found that, in general, for 87Sr NMR, the EFG tensor is the
most sensitive parameter to minute structural changes as was
evidenced by both Sr(pCl) and Sr(sal). This is consistent with
the recent findings of Bonhomme et al.36 In the case of
Sr(pNO2), the smaller |CQ(87Sr)| value was attributed to a more
symmetric 9-oxygen coordinate monocapped square antiprism
geometry. As for Sr(pams), changes in the 87Sr NMR
acquisition method allowed us to observe the second broader
559
dx.doi.org/10.1021/ic402658d | Inorg. Chem. 2014, 53, 552−561

Inorganic Chemistry
Article
(9) Banerjee, D.; Zhang, Z.; Plonka, A. M.; Li, J.; Parise, J. B. Cryst.
Growth Des. 2012, 12, 2162−2165.
NMR line shape associated with the nitrogen-binding Sr site.
Good agreement between experiment and GIPAW DFT
calculations was also noted for 87Sr NMR parameters. This
contribution has allowed for a deeper understanding of how
subtle structural changes about alkaline-earth metal cations are
manifested in their NMR parameters, and may find use in
future studies involving these essential elements which play key
roles in metalloproteins and more affordable alkaline-earth
metal catalysts, to name but a few applications.
(10) Zhang, X.; Huang, Y.-Y.; Zhang, M.-J.; Zhang, J.; Yao, Y.-G.
Cryst. Growth Des. 2012, 12, 3231−3238.
(11) Yeh, C.-T.; Lin, W.-C.; Lo, S.-H.; Kao, C.-C.; Lin, C.-H.; Yang,
C.-C. CrystEngComm 2012, 14, 1219−1222.
(12) Begouin, J.-M.; Niggemann, M. Chem.Eur. J. 2013, 19, 8030−
8041.
(13) Brinkmann, C.; Barrett, A. G. M.; Hill, M. S.; Procopiou, P. A. J.
Am. Chem. Soc. 2012, 134, 2193−2207.
(14) Tsubogo, T.; Yamashita, Y.; Kobayashi, S. Chem.Eur. J. 2012,
ASSOCIATED CONTENT
* Supporting Information
18, 13624−13628.
■
(15) Kobayashi, S.; Yamashita, Y. Acc. Chem. Res. 2011, 44, 58−71.
S
(16) Koler, M.; Langer, J.; Fischer, R.; Gorls, H.; Westerhausen, M.
̈
̈
Powder X-ray diffractograms for all Ca and Sr compounds;
immediate coordination polyhedra for Sr(ben), Ca(pams), and
Sr(pams); compound abbreviations and full formulas; exper-
imental 43Ca and 87Sr SSNMR conditions; spectra obtained at
9.4 T; computational details for GIPAW DFT calculations;
summary of data for Figure 6. This material is available free of
charge via the Internet at http://pubs.acs.org.
Chem.Eur. J. 2013, 19, 10497−10500.
(17) Jochmann, P.; Dols, T. S.; Spaniol, T. P.; Perrin, L.; Maron, L.;
Okuda, J. Angew. Chem., Int. Ed. 2009, 48, 5715−5719.
(18) Jochmann, P.; Maslek, S.; Spaniol, T. P.; Okuda, J. Organo-
metallics 2011, 30, 1991−1997.
(19) Bryce, D. L. Dalton Trans. 2010, 39, 8593−8602.
(20) Laurencin, D.; Smith, M. E. Prog. Nucl. Magn. Reson. Spectrosc.
2013, 68, 1−40.
(21) Moudrakovski, I. L. Annu. Rep. NMR Spectrosc. 2013, 79, 129−
240.
AUTHOR INFORMATION
Corresponding Author
*E-mail: dbryce@uottawa.ca. Phone: +1 613 562 5800, ext.
2018. Fax: +1 613 562 5170
■
(22) Harris, R. K.; Becker, E. D.; Cabral de Menezes, S. M.;
Goodfellow, R.; Granger, P. Pure Appl. Chem. 2001, 73, 1795−1818.
(23) Pyykko, P. Mol. Phys. 2008, 106, 1965−1974.
̈
Notes
(24) Reinholdt, M.; Croissant, J.; Di Carlo, L.; Granier, D.; Gaveau,
́
P.; Begu, S.; Devoisselle, J.-M.; Mutin, P. H.; Smith, M. E.;
The authors declare no competing financial interest.
Bonhomme, C.; Gervais, C.; van der Lee, A.; Laurencin, D. Inorg.
Chem. 2011, 50, 7802−7810.
ACKNOWLEDGMENTS
■
(25) Sene, S.; Reinholdt, M.; Renaudin, G.; Berthomieu, D.;
Zicovich-Wilson, C. M.; Gervais, C.; Gaveau, P.; Bonhomme, C.;
K.M.N.B. and Y.X. thank the Natural Sciences and Engineering
Research Council (NSERC) of Canada for graduate and
undergraduate scholarships, respectively. D.L.B. thanks
NSERC, the Canada Foundation for Innovation, and the
Ontario Ministry of Research and Innovation for funding. We
thank Dr. Glenn A. Facey, Dr. Ilia Korobkov, Dr. Tara Kell, and
Dr. Victor Terskikh for technical support, and Dr. Igor
Moudrakovski for helpful comments. Access to the 900 MHz
NMR spectrometer was provided by the National Ultrahigh-
Field NMR Facility for Solids (Ottawa, Canada), a national
research facility funded by the Canada Foundation for
Filinchuk, Y.; Smith, M. E.; Nedelec, J.-M.; Beg
Laurencin, D. Chem.Eur. J. 2013, 19, 880−891.
(26) Sene, S.; Bouchevreau, B.; Martineau, C.; Gervais, C.;
Bonhomme, C.; Gaveau, P.; Mauri, F.; Begu, S.; Mutin, P. H.;
́
u, S.; Mutin, P. H.;
́
Smith, M. E.; Laurencin, D. CrystEngComm 2013, 15, 8763−8775.
(27) Bowers, G. M.; Kirkpatrick, R. J. Cryst. Growth Des. 2011, 11,
5188−5191.
(28) Colas, H.; Bonhomme-Coury, L.; Diogo, C. C.; Tielens, F.;
Babonneau, F.; Gervais, C.; Bazin, D.; Laurencin, D.; Smith, M. E.;
Hanna, J. V.; Daudon, M.; Bonhomme, C. CrystEngComm 2013, 15,
8840−8847.
́
Innovation, the Ontario Innovation Trust, Recherche Quebec,
(29) Wong, A.; Howes, A. P.; Dupree, R.; Smith, M. E. Chem. Phys.
Lett. 2006, 427, 201−205.
(30) Laurencin, D.; Gervais, C.; Wong, A.; Coelho, C.; Mauri, F.;
Massiot, D.; Smith, M. E.; Bonhomme, C. J. Am. Chem. Soc. 2009, 131,
13430−13440.
(31) Gervais, C.; Laurencin, D.; Wong, A.; Pourpoint, F.; Labram, J.;
Woodward, B.; Howes, A. P.; Pike, K. J.; Dupree, R.; Mauri, F.;
Bonhomme, C.; Smith, M. E. Chem. Phys. Lett. 2008, 464, 42−48.
(32) Laurencin, D.; Wong, A.; Dupree, R.; Smith, M. E. Magn. Reson.
Chem. 2008, 46, 347−350.
(33) Wong, A.; Laurencin, D.; Dupree, R.; Smith, M. E. Solid State
Nucl. Magn. Reson. 2009, 35, 32−36.
(34) Laurencin, D.; Wong, A.; Chrzanowski, W.; Knowles, J. C.; Qiu,
D.; Pickup, D. M.; Newport, R. J.; Gan, Z.; Duer, M. J.; Smith, M. E.
Phys. Chem. Chem. Phys. 2010, 12, 1081−1091.
(35) Moudrakovski, I. L.; Alizadeh, R.; Beaudoin, J. J. Phys. Chem.
Chem. Phys. 2010, 12, 6961−6969.
(36) Bonhomme, C.; Gervais, C.; Folliet, N.; Pourpoint, F.; Diogo,
C. C.; Lao, J.; Jallot, E.; Lacroix, J.; Nedelec, J.-M.; Iuga, D.; Hanna, J.
V.; Smith, M. E.; Xiang, Y.; Du, J.; Laurencin, D. J. Am. Chem. Soc.
2012, 134, 12611−12628.
(37) Dupree, R.; Howes, A. P.; Kohn, S. C. Chem. Phys. Lett. 1997,
276, 399−404.
(38) Wong, A.; Laurencin, D.; Wu, G.; Dupree, R.; Smith, M. E. J.
Phys. Chem. A 2008, 112, 9807−9813.
the National Research Council Canada, and Bruker BioSpin
and managed by the University of Ottawa (www.nmr900.ca).
NSERC is acknowledged for a Major Resources Support grant.
REFERENCES
■
(1) Gatt, P.; Petrie, S.; Stranger, R.; Pace, R. J. Angew. Chem., Int. Ed.
2012, 51, 12025−12028.
(2) Tsui, E. Y.; Tran, R.; Yano, J.; Agapie, T. Nat. Chem. 2013, 5,
293−299.
(3) Park, Y. J.; Cook, S. A.; Sickerman, N. S.; Sano, Y.; Ziller, J. W.;
Borovik, A. S. Chem. Sci. 2013, 4, 717−726.
(4) Babu, C. S.; Dudev, T.; Lim, C. J. Am. Chem. Soc. 2013, 135,
6541−6548.
(5) Hoi, H.; Matsuda, T.; Nagai, T.; Campbell, R. E. J. Am. Chem. Soc.
2013, 135, 46−49.
(6) Collot, M.; Loukou, C.; Yakovlev, A. V.; Wilms, C. D.; Li, D.;
́
Evrard, A.; Zamaleeva, A.; Bourdieu, L.; Leger, J.-F.; Ropert, N.; Eilers,
J.; Oheim, M.; Feltz, A.; Mallet, J.-M. J. Am. Chem. Soc. 2012, 134,
14923−14931.
(7) Bar-Shir, A.; Gilad, A. A.; Chan, K. W. Y.; Liu, G.; van Zijl, P. C.
M.; Bulte, J. W. M.; McMahon, M. T. J. Am. Chem. Soc. 2013, 135,
12164−12167.
(8) Sleem, H. F.; Dawe, L. N.; Georghiou, P. E. New J. Chem. 2012,
36, 2451−2455.
560
dx.doi.org/10.1021/ic402658d | Inorg. Chem. 2014, 53, 552−561

Inorganic Chemistry
Article
(39) Fox, S.; Busching, I.; Barklage, W.; Strasdeit, H. Inorg. Chem.
(66) Huang, Y.-C.; Mou, Y.; Tsai, T. W.-T.; Wu, Y.-J.; Lee, H.-K.;
Huang, S.-J.; Chan, J. C. C. J. Phys. Chem. B 2012, 116, 14295−14301.
(67) Sahoo, B. K. Phys. Rev. A 2009, 80, 012515.
(68) Burgess, K. M. N.; Xu, Y.; Leclerc, M. C.; Bryce, D. L. J. Phys.
Chem. A 2013, 117, 6561−6570.
̈
2007, 46, 818−824.
(40) Kim, E. E.; Sicignano, A.; Eriks, K. J. Am. Chem. Soc. 1985, 107,
6042−6046.
(41) Schmidbaur, H.; Bach, I.; Wilkinson, D. L.; Muller, G. Chem.
̈
(69) Bowers, G. M.; Lipton, A. S.; Mueller, K. T. Solid State Nucl.
Magn. Reson. 2006, 29, 95−103.
Ber. 1989, 122, 1439−1444.
(42) Wu, G.; Wong, A.; Wang, S. Can. J. Chem. 2003, 81, 275−283.
(43) Pallister, P. J.; Moudrakovski, I. L.; Ripmeester, J. A. Phys. Chem.
Chem. Phys. 2009, 11, 11487−11500.
(70) Bowers, G. M.; Ravella, R.; Komarneni, S.; Mueller, K. T. J. Phys.
Chem. B 2006, 110, 7159−7164.
(71) Briggman, B.; Oskarson, Å. Acta Crystallogr., Sect. B: Struct. Sci.
1977, 33, 1900−1906.
(44) Arlin, J.-B.; Florence, A. J.; Johnston, A.; Kennedy, A. R.; Miller,
G. J.; Patterson, K. Cryst. Growth Des. 2011, 11, 1318−1327.
(45) Cole, L. B.; Holt, E. M. Inorg. Chim. Acta 1989, 160, 195−203.
(46) Barnett, B. L.; Uchtman, V. A. Inorg. Chem. 1979, 18, 2674−
2678.
(47) (a) O’Dell, L. A.; Schurko, R. W. Chem. Phys. Lett. 2008, 464,
97−102. (b) O’Dell, L. A.; Rossini, A. J.; Schurko, R. W. Chem. Phys.
Lett. 2009, 468, 330−335.
(48) Solomon, I. Phys. Rev. 1958, 110, 61−65.
(49) Massiot, D.; Farnan, I.; Gautier, N.; Trumeau, D.; Trokiner, A.;
Coutures, J. P. Solid State Nucl. Magn. Reson. 1995, 4, 241−248.
(50) Eichele, K.; Wasylishen, R. E. WSOLIDS NMR Simulation
Software, Version 1.17.30; Universitat Tubingen: Tubingen, Germany,
̈
̈
̈
2001.
(51) Massiot, D.; Fayon, F.; Capron, M.; King, I.; Le Calve,
́
S.;
Alonso, B.; Durand, J.-O.; Bujoli, B.; Gan, Z.; Hoatson, G. Magn.
Reson. Chem. 2002, 40, 70−76.
(52) Segall, M. D.; Lindan, P. J. D.; Probert, M. J.; Pickard, C. J.;
Hasnip, P. J.; Clark, S. J.; Payne, M. C. J. Phys.: Condens. Matter 2002,
14, 2717−2744.
(53) Pickard, C. J.; Mauri, F. Phys. Rev. B 2001, 63, 245101.
(54) Debuyst, R.; Dejehet, F.; Dekandelaer, M.-C.; Declercq, J.-P.;
Germain, G.; van Meerssche, M. J. Chim. Phys. Phys.-Chim. Biol. 1979,
76, 1117−1124.
(55) Karipides, A.; McKinney, C.; Peiffer, K. Acta Crystallogr., Sect. C:
Cryst. Struct. Commun. 1988, 44, 46−48.
(56) Senkovska, I.; Thewalt, U. Acta Crystallogr., Sect. C: Cryst. Struct.
Commun. 2005, 61, m448−m449.
(57) Wieczorek, M. W.; Błaszczyk, J.; Krol
Sect. C 1996, 52, 1193−1198.
́
, B. W. Acta Crystallogr.,
(58) Deganello, S.; Kampf, A. R.; Moore, P. B. Am. Mineral. 1981, 66,
859−865.
(59) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77,
3865−3868.
(60) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
̈
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
Revision A.02; Gaussian, Inc.: Wallingford, CT, 2009.
(61) Adiga, S.; Aebi, D.; Bryce, D. L. Can. J. Chem. 2007, 85, 496−
505.
(62) Widdifield, C. M.; Bryce, D. L. Can. J. Chem. 2011, 89, 754−
763.
(63) Bryce, D. L.; Bultz, E. B.; Aebi, D. J. Am. Chem. Soc. 2008, 130,
9282−9292.
(64) Deng, Z.-P.; Kang, W.; Zhu, Z.-B.; Huo, L.-H.; Zhao, H.; Gao, S.
Dalton Trans. 2012, 41, 8354−8360.
(65) Plonka, A. M.; Banerjee, D.; Parise, J. B. Cryst. Growth Des.
2012, 12, 2460−2467.
561
dx.doi.org/10.1021/ic402658d | Inorg. Chem. 2014, 53, 552−561