Ion pair complexes and anion binding in the solution of a ditopic receptor

Article abstract of DOI:10.1039/c6dt00414h

The synthesis and crystal structures with alkali halides of a ditopic benzo-15-crown-5 bis-urea receptor L have been presented. In addition, the anion binding properties of L and its alkali metal complexes in solution are presented. A comprehensive single-crystal X-ray crystallographic study of L, all together 13 crystal structures, including the ion pair complexes with NaCl, NaBr, NaI, KF, KCl, KBr, KI, RbF, RbCl, and RbI, give a detailed view of how L behaves in the solid-state with different alkali halides depending on the size of the cation and anion. In the solid-state L forms a 1 : 1 complex with a sodium cation and the anion is complexed as a contact (NaCl) or a separate ion pair (NaBr, NaI). With larger potassium and rubidium cations L assembles into a 2 : 1 complex and forms a separated ion pair complex with the anion. Reflecting the crystal structures the L forms a 1 : 1 complex with Na⁺ in solution, and a 2 : 1 complex with K⁺, which were verified by Job's plot analysis in 4 : 1 CDCl₃/dimethyl sulfoxide. The binding strength of the monomeric [L·Na]⁺ and the dimeric [2L·K]⁺ toward chloride, bromide and iodide anions was studied by ¹H NMR titrations in 4 : 1 CDCl₃/DMSO, and a clear turn-on effect of the cation complexation compared to the neutral receptor L alone (K_a with L for Cl^-, Br^- and I^- being 832, 174 and 32 M^-1, respectively) was observed. The monomeric [L·Na]⁺ binds chloride 9, bromide 8, and iodide 12 times stronger than L, while for the dimeric [2L·K]⁺ the corresponding increase in binding is 51 (Cl^-), 84 (Br^-), and 22 (I^-) times with the same stoichiometric ratios as observed for the ion pair complexes in the solid-state.

Full text of DOI:10.1039/c6dt00414h

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Ion pair complexes and anion binding in the
solution of a ditopic receptor†
Cite this: Dalton Trans., 2016, 45,
6481
T. Mäkelä and K. Rissanen*
The synthesis and crystal structures with alkali halides of a ditopic benzo-15-crown-5 bis-urea receptor L
have been presented. In addition, the anion binding properties of L and its alkali metal complexes in solu-
tion are presented. A comprehensive single-crystal X-ray crystallographic study of L, all together 13 crystal
structures, including the ion pair complexes with NaCl, NaBr, NaI, KF, KCl, KBr, KI, RbF, RbCl, and RbI, give
a detailed view of how L behaves in the solid-state with different alkali halides depending on the size of
the cation and anion. In the solid-state L forms a 1 : 1 complex with a sodium cation and the anion is com-
plexed as a contact (NaCl) or a separate ion pair (NaBr, NaI). With larger potassium and rubidium cations L
assembles into a 2 : 1 complex and forms a separated ion pair complex with the anion. Reflecting the
crystal structures the L forms a 1 : 1 complex with Na+ in solution, and a 2 : 1 complex with K+, which were
verified by Job’s plot analysis in 4 : 1 CDCl3/dimethyl sulfoxide. The binding strength of the monomeric
1
[L·Na]+ and the dimeric [2L·K]+ toward chloride, bromide and iodide anions was studied by H NMR titra-
tions in 4 : 1 CDCl3/DMSO, and a clear turn-on effect of the cation complexation compared to the neutral
receptor L alone (Ka with L for Cl−, Br− and I− being 832, 174 and 32 M−1, respectively) was observed. The
monomeric [L·Na]+ binds chloride 9, bromide 8, and iodide 12 times stronger than L, while for the
dimeric [2L·K]+ the corresponding increase in binding is 51 (Cl−), 84 (Br−), and 22 (I−) times with the same
stoichiometric ratios as observed for the ion pair complexes in the solid-state.
Received 29th January 2016,
Accepted 24th February 2016
DOI: 10.1039/c6dt00414h
www.rsc.org/dalton
multiple ion pair complexes in the system.5–7 In addition the
electrostatic and allosteric contributions in the ion pair
Introduction
Cation and anion receptor chemistry has been an integral part binding by synthetic receptors are not thoroughly understood,
of supramolecular chemistry since the establishment of the although detailed studies around this topic have been recently
field. Particularly, the development of cation receptors has presented.8,9 The recent reviews of ion pair receptors reveal
been a heavily investigated area from the beginning.1 A slow innovative molecular designs, and how the aforementioned
rate in the development of anion receptor chemistry originates difficulties have been addressed in specific ion pair
from the more complicated nature of the anions, i.e. larger recognition.10–13 Crown ethers have been studied as building
size, more complicated geometry and higher solvation energy. blocks in ditopic receptors for example by Reinhoudt,14 Beer,15
Nevertheless, the interest toward anion receptor chemistry Smith,16,17 and Sessler,18 and their coworkers. The efficient
within the scientific community has been booming according practical applications in salt extraction19,20 and solubil-
to the large number of developed systems for anion reco- ization,21 membrane transport,22 catalysis,23 and sensors24
gnition within the last decade.2–4 The combination of cation represent the interesting possibilities of utilizing ion pair
and anion recognition sites in a single molecule, i.e. ion pair receptors in various tasks.
receptors, have recently attracted a growing interest due to
We have recently presented structural and alkali metal
their more selective and effective binding toward both the halide recognition studies with ditopic benzo-18-crown-6 bis-
cation and anion. Ion pair recognition has to overcome the urea,25 which manifested efficient binding of halide anions in
difficulty of the ion separation, and also avoid the formation of solution, when complexed with alkali metal cations. We have
also looked into the solid-state ion pair complexation by
benzo-15-crown-5 and benzo-18-crown-6 uranyl salophen
receptors26 with various alkali halides. These receptors are
constructed from simple and well-known structural units, but
University of Jyvaskyla, Department of Chemistry, Nanoscience Center, P.O. Box 35,
FI-40014; University of Jyvaskyla, Finland. E-mail: kari.t.rissanen@jyu.fi
†Electronic supplementary information (ESI) available: Crystal data, crystal
the combination of the binding units into a single molecule
structure figures, Job’s plot figures, titration data and binding isotherms. CCDC
for ion pair recognition is surprisingly less studied. In par-
1425256–1425269. For ESI and crystallographic data in CIF or other electronic
ticular, the solid-state structures of this type of crown ether
format see DOI: 10.1039/c6dt00414h
This journal is © The Royal Society of Chemistry 2016
Dalton Trans., 2016, 45, 6481–6490 | 6481

View Article Online
Paper
Dalton Transactions
Scheme 1 The synthesis of receptor L. (a) HNO3/H2SO4, CHCl3, 77%. (b) NH2NH2·H2O, 10% Pd/C, EtOH, 98%. (c) 4-nitrophenyl isocyanate, CH2Cl2/
DMF, 75%. The protons Ha, Hb, Hc, Hd, and Hf were followed in Job’s plot analysis and 1H NMR titrations and are presented in the structure of L.
based receptors are scarce. The CSD search27 combining ments were performed with Micromass LCT ESI-TOF-MS
benzo-15-crown-5 (B15C5) or benzo-18-crown-6 (B18C6) with instrument by using a lock-mass method.
amide moieties reveals only 21 structures having these
binding sites covalently connected. Out of these structures
only 13 contain ion pairs in the crystal lattice, of which eight
are separate and five are contact ion pairs. In addition, CSD
search28 of urea-functionalized molecules complexed with
alkali halides gives only five solid-state structures, although
the urea-group is heavily incorporated into other anion recep-
tors. From this point of view, there is a true potential to
utilize these simple building blocks to develop new receptor
systems for ion pairs. Due to this, and encouraged by our
recent results,25 our focus has been directed to study a benzo-
15-crown-5 bis-urea receptor L (Scheme 1) to act as a ditopic
receptor in the solid-state and solution. We have conducted
an extensive single-crystal X-ray crystallographic study of the
alkali metal halide complexes of L to gain more insight about
the interactions governing the ion pair binding of L in the
solid-state. As the crystal structures might give a biased view
on the binding stoichiometry in solution, we have also con-
ducted a series of solution studies by determining the stoichi-
ometry and binding affinities of L and its alkali metal
Syntheses (Scheme 1)
Dinitrobenzo-15-crown-5 (2) and diaminobenzo-15-crown-5 (3)
were synthesized according to the literature procedure with
small modifications.25 Details are presented in the ESI.†
Synthesis of L. Diaminobenzo-15-crown-5 (3) (0.163 g,
0.546 mmol) was dissolved in degassed dichloromethane
(20 ml) and DMF (3 ml) under argon. 4-Nitrophenyl isocyanate
(0.128 g, 0.780 mmol) was dissolved in degassed dichloro-
methane (10 ml) and added slowly to 3 through a septum.
After the addition, 20 ml of degassed dichloromethane was
added to the solution. The reaction mixture was stirred for
18 h under argon. The precipitated solid was filtered and
washed with dichloromethane, methanol and diethyl ether.
The second batch of the product was obtained from the filtrate
by concentrating the solution and precipitating the product
with a slow addition of diethyl ether into the solution. The
1
solid was washed as previously; yield 75%. H NMR (500 MHz,
DMSO, 303 K): δ = 3.62 (8H, s, CH2), 3.78 (4H, dd CH2), 4.04
(4H, dd, CH2), 7.23 (2H, s, ArH), 7.70 (4H, d, ArH), 8.12 (2H, s,
NH), 8.18 (4H, d, ArH), 9.74 (2H, s, NH) ppm. 13C NMR
(126 MHz, DMSO, 303 K): δ = 68.79, 68.91, 69.81, 70.33,
110.94, 117.38, 124.31, 125.07, 140.91, 145.57, 146.50,
152.76 ppm. HRMS (+ESI): calcd for C28H30N6O11Na,
[M + Na]+: m/z 649.1865; found: m/z 649.1852 (Δ = 1.3 mDa).
1
complexes toward anions via Job’s plot analysis and H NMR
titrations in moderately competitive 4 : 1 CDCl3/DMSO-d6
solution.
Experimental
Crystallography
Materials and methods
Single-crystal X-ray data were collected with Agilent SuperNova,
All the reagents and solvents were purchased from Aldrich, equipped with a multilayer optics monochromated dual source
Fluka and Altia and used as received. All the yields refer to (Cu and Mo) and a Atlas detector, using Cu-Kα (λ = 1.54184 Å)
spectroscopically homogeneous materials. 1H NMR spectra radiation at 123 K or 134 K. Data acquisitions, reductions, and
were recorded with Bruker Avance 300 and Bruker Avance 500 analytical face-index based absorption corrections were made
instruments at 303 K. 13C spectra were recorded with Bruker using the program CrysAlisPRO.29 Single-crystal X-ray data for
Avance 500 instrument at 303 K. All the spectra were calibrated 2 and L·DMF were collected with a Bruker Kappa Apex II
using the solvent signals of CHCl3 (1H = 7.26 ppm, 13C = diffractometer, using graphite-monochromatized Mo-Kα (λ =
77.16 ppm) or DMSO (1H = 2.50 ppm, 13C = 39.52 ppm) as 0.71073 Å) radiation at 173 K and 120 K, respectively. Collect
internal standards. All high-resolution (HR) mass measure- software30 was used for the data measurement and
6482 | Dalton Trans., 2016, 45, 6481–6490
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
DENZO-SMN31 was used for the data processing. Multi-scan
absorption correction was done with SADABS2008.32 The struc-
tures were solved with ShelXS,33 Superflip34 or Sir200235
Results and discussion
Synthesis and characterization
programs and refined on F2 by full matrix least-squares The receptor L was prepared by a similar three-step synthesis
techniques with ShelXL33 program in Olex236 (v.1.2) program as described previously.25 The starting material was a commer-
package. The non-H atoms were refined anisotropically and all cially available benzo-15-crown-5 compound 1 (Scheme 1). The
hydrogen positions were calculated using a riding atom model starting material was nitrated by a two-phase reaction in
with ShelXL33 default parameters. Solvent masking37 in Olex2 chloroform and a nitric acid/sulfuric acid mixture. The dinitro
program package was used for structures L·NaCl, 2L·KF, 2L·KI, product 2 was obtained in 77% yield. Compound 2 was then
2L·RbF, and 2L·RbCl.
reduced with hydrazine monohydrate and 10% palladium on
activated charcoal under anaerobic conditions under an argon
atmosphere to give diamine product 3. The yield was nearly
Job’s plot analysis
Job’s plot measurements were performed in a 4 : 1 CDCl3/ quantitative, and the diamine was used without further purifi-
DMSO solvent mixture. 2.5 mM stock solutions of L and the cation due to the instability of the product. Condensation reac-
guest were prepared, and for ion pair investigations 2.5 mM tion between 3 and 4-nitrophenyl isocyanate under argon
stock solution of L with 1 equivalent of MBPh4 (M = Na or K) atmosphere afforded the desired product L, which precipitated
was prepared. Samples with mole fractions X = 0, 0.1, 0.2, 0.3, out from the reaction mixture as a yellow solid with total yield
1
0.4, 0.5, 0.6, 0.7, 0.8, 0.9 according to the guest were prepared of 75%. Products 2 and L were analyzed by H and 13C NMR,
in NMR tubes with a total volume of 500 µl. The 1H NMR HR-ESI-MS, and single-crystal X-ray analysis (see ESI† and
spectra were recorded with a Bruker Avance 500 spectrometer Fig. 1). The diamine 3 was characterized by 1H NMR.
at 303 K. The spectra were calibrated using the CHCl3 signal
Crystal structures of L and its alkali halide complexes
(δ = 7.26 ppm) as an internal standard. The Job’s plots were
obtained by plotting (Δδ*[H])/([H] + [G]) against the mole In the course of this work, three solvate structures of L were
fraction of the guest (H = L, G = guest). The chemical shifts obtained. Slow evaporation of acetone solution of L with NaF
of the receptor’s urea protons Ha, Hb, aromatic proton Hc and added in aqueous solution gave crystals with a dimeric assem-
aliphatic protons Hd and Hf were followed (see Scheme 1 and bly of L (Fig. 1a and S7;† structure is discussed later in the
Fig. S20† for proton assignments).
text). Crystallization of L from dimethylformamide (DMF) or
dimethyl sulfoxide (DMSO) solutions by evaporation afforded
the respective solvate structures, which both show hydrogen
1H NMR titrations
All titrations were performed in a 4 : 1 CDCl3/DMSO solvent bonding between the solvate molecule and one urea group in
mixture. For anion titrations 0.05 M stock solution of L was L (Fig. 1b and S8,† Fig. 1c and S9†).
prepared in DMSO, and the sample was diluted with DMSO
The solid-state ion pair recognition of L was comprehen-
and CDCl3 to give a final sample concentration of 2.5 mM in sively studied with single-crystal X-ray crystallography. The ten
500 µl. For anion titrations 0.125 M stock solution of corres- obtained crystal structures of L complexed with NaCl, NaBr,
ponding TBA-salt in 4 : 1 CDCl3/DMSO was prepared. For ion NaI, KF, KCl, KBr, KI, RbF, RbCl, and RbI revealed interesting
pair titrations with the [L·Na]+ complex, 0.05 M stock solutions solid-state behavior of the receptor. Each obtained ion pair
were prepared for L and NaBPh4. Equimolar amounts of the complex with a sodium cation shows a 1 : 1 complex formation
stock solutions were measured in a NMR tube, and the sample between L and Na+, with either contact or separate ion pair for-
was diluted to give 2.5 mM 1 : 1 [L·Na]+ complex concentration mation with the anion depending on the size (and polarizabil-
in 500 µl. For ion pair titrations with the [2L·K]+ complex, ity) of the anion. When L is complexed with larger potassium
0.05 M stock solutions of L and KBPh4 were prepared. Suitable or rubidium cations, it forms a dimeric assembly,39,40 binding
amounts of L and KBPh4 were measured in a 2 : 1 ratio to give the anion separately from the cation, regardless of the nature
5 mM L concentration resulting in approximately 2.5 mM of the anion. The formation of dimeric assembly is not sur-
[2L·K]+ complex concentration in 500 µl. For ion pair titrations prising when considering the previously published solid-state
0.125 M stock solution of the corresponding TBA-salt in 4 : 1 structures with benzo-15-crown-5 (B15C5) based molecules,
CDCl3/DMSO was prepared. 1H NMR spectra of the titrations which all show the dimer formation with potassium (39 struc-
were measured with a Bruker Avance 500 spectrometer at tures) or rubidium (3 structures).41 Also sodium is capable of
303 K and all titrations were performed with 18 measurements forming a dimeric assembly with B15C5, with 11 structures
with the following amounts of the guest added: 0, 0.1, 0.2, 0.3, (out of 34, 32%) showing this kind of behavior. However,
0.4, 0.5, 0.6, 0.8, 1.0, 1.2, 1.4, 1.6, 2.0, 2.5, 3.0, 4.0, 6.0, 10.0 cation induced dimerization was not observed in any of the
equiv. The spectra were calibrated using the CHCl3 signal (δ = Na+ complexes of L.
7.26 ppm) as an internal standard. Titration data was fitted
Single-crystals of L·NaCl were obtained by slow diffusion of
into a desired binding model with HypNMR2008.38 The chemi- diethyl ether into solution of L in acetone/DMSO with excess
cal shifts of urea protons Ha and Hb, and aromatic proton Hc of NaCl added in water. The complex crystallized in the mono-
were followed during the titration experiments (see Scheme 1 clinic space group P21/n. The dimeric structure of L with com-
and Fig. S20† for proton assignments).
plexed NaCl is presented in Fig. 2. The dimerization results
This journal is © The Royal Society of Chemistry 2016
Dalton Trans., 2016, 45, 6481–6490 | 6483

View Article Online
Paper
Dalton Transactions
Fig. 1 Single-crystal X-ray structures of (a) L [unit cell parameters: monoclinic, a = 15.41305(15) Å, b = 23.7285(2) Å, c = 18.50903(16) Å, β = 95.2966(9)°],
(b) L·DMF [unit cell parameters: triclinic, a = 10.71590(10) Å, b = 11.3107(2) Å, c = 15.4992(3) Å, α = 101.4580(10)°, β = 95.3000(10)°, γ =
100.4910(10)°], and (c) L·DMSO [unit cell parameters: triclinic, a = 8.8481(3) Å, b = 10.1334(4) Å, c = 18.5466(6) Å, α = 86.562(3)°, β = 89.252(3)°, γ =
80.827(3)°]. Water molecules hydrogen bonded to the crown ether ring are omitted from the picture of L. The nonbonding hydrogen atoms from
structures L·DMF and L·DMSO, and other solvate molecules are omitted from the picture.
from hydrogen bonding between the urea groups in the adja- the crown ether, and carbonyl oxygen O2′ of the adjacent mole-
cent receptor molecules (details are presented in the ESI, cule and acetonitrile solvate molecules fill the coordination
Fig. S10†). The chloride anion forms a contact ion pair with sphere of the cation (Fig. 3b).42,43 Positioning of the receptors
the crown ether-complexed sodium cation. The chloride is forms a urea proton-coated binding pocket hosting two
further hydrogen bonded by the urea group of the adjacent bromide anions, capped by the acetonitrile molecule co-
dimer (N3A⋯Cl1 distance 3.178 Å, N4A⋯Cl1 distance 3.130 Å, ordinated to the sodium cation in the adjacent complex (not
N1B⋯Cl2 distance 3.258 Å, and N2B⋯Cl2 distance 3.079 Å, shown). Bromide is interacting with the two receptor mole-
Fig. 2).
cules through four hydrogen bonds, three of them being
Interestingly, this structure is nearly identical to the formed to urea groups in L [N1⋯Br1 distance 3.594 Å,
dimeric L structure (Fig. 1a and S7†), in which water molecule N2⋯Br1 distance 3.350 Å, and N4⋯Br1 distance 3.499 Å], and
occupies the same position as the chloride anion in L·NaCl one to the urea group in the adjacent L′ molecule (N3′⋯Br1
complex (Fig. S7b†). The water molecule forms hydrogen distance 3.397 Å).
bonds with the crown ether oxygens and urea nitrogens, result-
The complex L·NaI was crystallized by slow diffusion of
ing in the same kind of positioning of the receptor molecules diethyl ether into an acetone solution of L with NaI added in
(Fig. S10c†), and nearly identical packing between the two excess in water. The complex crystallized in the monoclinic
structures. The dimerization of the two L molecules is reminis- space group I2/a, consisting of L and a separated ion pair
cent of the general solid-state binding motif with benzo-18- complex with NaI. L forms a 1 : 1 complex with the sodium
crown-6 bis-urea receptor presented previously.25 In that (Fig. 4), and the iodide is hydrogen bonded with three urea
binding motif the dimerization resulted from hydrogen protons [N1⋯I1 distance 3.934(3) Å, N3⋯I1 distance 3.515(3) Å,
bonding between the receptor and the anion, and ion–dipole and N4⋯I1 distance 3.646(3) Å]. Interestingly, one urea
interactions between the complexed cation and the nitro group nitrogen is hydrogen bonded to the nitro group of an adjacent
oxygen. In the L·NaCl complex the dimerization results solely receptor molecule [N2⋯O5′ distance 3.088(3) Å, N2⋯O6′ dis-
from the hydrogen bonding between the urea groups, and tance 3.240(3) Å] leading to a different packing of the L mole-
there is no coordinative bond between the nitro group oxygen cules in the crystal lattice compared to the L·NaBr structure.
and the sodium cation. The strong point charges of the small There is a coordinative bond between sodium and the carbonyl
and weakly polarized ions result in a strong ion pair, and the oxygen (Fig. S12b†), resulting in the formation of a 1D coordi-
weak interactions present in the solid-state might not be nation polymer along the crystallographic b-axis. The coordi-
powerful enough to separate the ions, thus resulting in a nation sphere of sodium is filled by the seventh coordinative
contact ion pair complex. L·NaCl was the only solid-state bond to the nitro group oxygen in the adjacent receptor
complex of L where this type of contact ion pair was observed.
The L·NaBr complex was crystallized with slow diffusion of
(not shown).
All the crystal structures of L obtained with potassium and
diethyl ether into acetonitrile solution of L with NaBr added in rubidium consist of a dimeric assembly of the receptors.
excess in water. The complex crystallized in the triclinic space Dimerization leads automatically to a separated ion pair
ˉ
group P1. In this structure L forms a separated ion pair complex, and the anion is located in a urea-hydrogen functio-
complex with NaBr (Fig. 3 and S11†). Sodium is coordinated to nalized binding pocket. Fig. 5a shows the crystal structure
6484 | Dalton Trans., 2016, 45, 6481–6490
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
Fig. 3 Crystal structure of L·NaBr. (a) L forms a separated ion pair
complex with NaBr in the solid-state. Bromide anions are located in a
urea proton-coated binding pocket. (b) The coordination sphere of
sodium is filled by a coordinative bond to carbonyl oxygen O2’ of the
adjacent receptor and an acetonitrile solvate molecule [unit cell para-
meters: triclinic, a = 9.4682(3) Å, b = 12.1247(3) Å, c = 16.1742(5) Å, α =
72.838(3)°, β = 87.307(2)°, γ = 87.437(2)°].
Fig. 2 Crystal structure L·NaCl. (a) Hydrogen bonding between the
urea functionalities results in dimerization of two L molecules. Chloride
forms a contact ion pair with the sodium complexed in the crown ether
ring. (b) Chloride anion is further hydrogen bonded with the adjacent
dimer. Only one molecule of the dimer is shown [unit cell parameters:
monoclinic, a = 15.83850(18) Å, b = 23.7838(3) Å, c = 18.5462(2) Å, β =
90.8013(11)°].
tance 2.891 Å, N4B⋯O1B′ distance 2.750 Å]. Details of the
packing are presented in the ESI (Fig. S14b†).
2L·KCl, which was obtained by the slow evaporation of the
In the other fluoride and chloride complexes of L with pot-
acetone solution of L with excess KCl added in water. The assium and rubidium, there are solvent molecules also inter-
ˉ
complex crystallized in the triclinic space group P1. The dimer acting with the anion. In 2L·KF (Fig. 6 and S13†), 2L·RbF
is formed through cation coordination by ten crown ether (Fig. S17†), and 2L·RbCl (Fig. S18†) structures, there are either
oxygens, and intermolecular hydrogen bonds between O2B water or methanol molecules hydrogen bonded with the
and urea nitrogens N3A and N4A [N3A⋯O2B distance 2.892 Å, anion, resulting in complex solvent-mediated hydrogen
N4A⋯O2B distance 2.942 Å]. The dimerization leads to a favor- bonding networks. Interestingly, this is not observed in any of
able orientation of the urea nitrogens N1A, N2A, N1B, and N2B the structures obtained with bromide or iodide. The dimeric
toward the center of the cavity formed by the receptor arms. assemblies of L with KBr (Fig. S15†), KI (Fig. S16†), and RbI
The chloride anion is hydrogen bonded with these urea nitro- (Fig. S19†) are structurally very similar (Fig. 5b), differing only
gens [N1A⋯Cl1 distance 3.389 Å, N2A⋯Cl1 distance 3.171 Å, slightly in the receptor arm orientations and unit cell dimen-
N1B⋯Cl1 distance 3.264 Å, and N2B⋯Cl1 distance 3.194 Å], sions. Surprisingly, crystals of L with RbBr were not obtained
the distances being close to the average N⋯Cl distance in com- in spite of numerous attempts. Taking into consideration the
plexes with chloride hydrogen bonded to the urea group similarities in the structures with bromide and iodide salts of
(3.203 Å).44 The dimers are further packed by hydrogen bonds potassium and rubidium, probably this complex would not
between the urea groups in adjacent dimers [N3B⋯O1B′ dis- differ much from the already obtained structures.
This journal is © The Royal Society of Chemistry 2016
Dalton Trans., 2016, 45, 6481–6490 | 6485

View Article Online
Paper
Dalton Transactions
Fig. 4 Crystal structure of L·NaI. Structure consists of a separated ion
pair complex, with sodium complexed in the crown ether and iodide
anion hydrogen bonded with three urea hydrogen bonds. Fourth urea
proton forms hydrogen bonds to the adjacent receptor’s nitro group
[unit cell parameters: monoclinic, a = 22.1422(5) Å, b = 7.22931(19) Å,
c = 42.6101(13) Å, β = 100.291(3)°].
Although all the solid-state structures of L with potassium
and rubidium have basic similarities in the dimer formation
and ion pair complexation, all the structures are not identical
to one another. The most strikingly different structure was
obtained with L and KF. The X-ray quality crystals of 2L·KF
were grown by slow diffusion of diethyl ether into methanol/
DMF solution of L with excess KF added in water. The complex
crystallized in the orthorhombic space group Fdd2, having an
extremely large unit cell (63 500 Å3) and large voids (8.6% of
the unit cell)45 in the crystal lattice (Fig. S13b†). The structure
can be formulated as a tetramer, viz. dimer of two 2L·KF com-
plexes (Fig. 6). The fluoride anion is bound inside the binding
pocket by
a water-mediated hydrogen bonding network
(Fig. 6c). The hydrogen bond lengths of the urea-groups vary
between 2.633 Å–2.957(6) Å for F1, and 2.707 Å–2.871 Å for F2.
Both fluoride anions are further hydrogen bonded with water
molecules [O12⋯F1 distance 2.557(8) Å, O16⋯F2 distance
2.658(6) Å]. The dimers are hydrogen bonded via urea-groups
resulting in dimer of dimers [N1D⋯O2B distance 3.017(6) Å,
N2D⋯O2B distance 2.866(6) Å, N1B⋯O2D distance 3.039(6) Å,
and N2B⋯O2D distance 2.798(6) Å]. These interactions are
reminiscent of the urea group interactions responsible for
dimer formation in structures L and L·NaCl. The structure con-
sists of a large number of solvent molecules, but because of
the large unit cell and porous structure, all the solvent mole-
cules within the structure could not be successfully modelled.
Fig. 5 (a) Crystal structure of 2L·KCl. Complexation of K+ in
15C5 moiety leads to dimerization of two receptor molecules. Dimeriza-
tion is further enhanced by intermolecular hydrogen bonds between
carbonyl oxygen O2B and urea nitrogens N3A and N4A [unit cell para-
meters: triclinic, a = 10.5214(2) Å, b = 15.4120(4) Å, c = 21.1832(5) Å, α =
105.004(2)°, β = 97.408(2)°, γ = 109.292(2)°]. (b) Overlay of structures
2L·KBr (yellow) and 2L·KI (purple) showing very similar conformation of
the receptor molecules in the dimeric assemblies.
M = Na and K) as the cation source and the tetrabutyl-
ammonium halides (TBAX, X = Cl, Br, and I) as the anion
−
source. The BPh4 and TBA+ ions do not interact with the very
Anion binding studies in solution
similar benzo-18-crown-6 bis-urea receptor and are thus not
affecting the binding event.25 In addition to solubilizing L in
weakly polar CDCl3, DMSO was used to create more competi-
tive media for the binding, which is known to be strong in less
competitive solvent systems.25 The host and guest concen-
trations and the binding induced chemical shifts were used in
The stoichiometry of the complexation in solution with L and
a set of alkali metal cations, anions and ion pairs were studied
with Job’s plot analysis. In addition, the behavior of L and its
alkali metal complexes as anion receptors was studied via H
NMR titrations. These studies were performed in 4 : 1 CDCl3/
DMSO-d6, using alkali metal tetraphenylborates (MBPh4,
1
6486 | Dalton Trans., 2016, 45, 6481–6490
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
Fig. 6 Crystal structure of 2L·KF. (a) Side view of the structure. It consists of dimer of dimers formed through hydrogen bonding between urea
groups in the neighboring dimers. (b) Top view of the structure. The interactions of “inner” molecules resemble to those seen in structures L and
L·NaCl. (c) The fluoride anion is located inside the binding pocket hydrogen bonded with the urea protons and water molecules [unit cell para-
meters: orthorhombic, a = 52.080(2) Å, b = 43.9225(8) Å, c = 27.7395(6) Å].
the HypNMR2008 program,38 utilizing global analysis46 with receptor, and a reference bis-urea receptor without the crown
the help of Ha, Hb, and Hc proton (see Scheme 1) signals to ether unit.25 Details of the binding constant calculations and
calculate the binding constants (Ka).
the binding isotherms are given in the ESI (Fig. S24–S26†).
The Job’s plot analysis confirms the expected 1 : 1 complex
In order to estimate the halide binding affinity of the cat-
ionic complexes of L, some simplifications were made in the of L and Na+ (Fig. S21a†) and [L·Na]+ + Cl− (Fig. 7a) in 4 : 1
binding constant calculations. The guests are used as their CDCl3/DMSO. The halide binding affinity of the cationic
BPh4− and TBA+ salts which are considered as “innocent” complex [L·Na]+ was measured via 1H NMR titrations in the
counter ions. The effect of these counter-ions is considered to same 4 : 1 CDCl3/DMSO solvent system. The chemical shift
be so small that it can be safely neglected from the binding differences observed upon stepwise addition of chloride in
constant calculation. The above is supported by the fact that titrations of L + Cl− and [L·Na]+ + Cl− are presented in Fig. 7b.
the ion pairing of NaBPh4 and TBABPh4 has been quantified
The turn-on effect of the sodium cation on the chloride
in a 4 : 1 CDCl3/CD3CN solvent system, where the ion pair for- binding is clearly seen from the chemical shift differences
mation between Na+ and BPh4 was found to be extremely between L and [L·Na]+ (Fig. 7b). The more drastic chemical
−
weak,8 and weak for TBABPh4.8,47 Considering the similarity of shift changes of [L·Na]+ with small chloride concentrations
the solvent system in this study (4 : 1 CDCl3/DMSO), Na+ can indicate stronger chloride binding. The binding constant for
be considered to be a non-contact ion pair and free from the the [L·Na]+ + Cl− complexation calculated with the 1 : 1
−
BPh4 counter-anion, and thus its interaction with the binding stoichiometry was Ka = 7609 M−1, being ca. nine times
B15C5 moiety is not perturbed by ion pairing. Complexation of larger than with L alone (Fig. S27†). Titration of [L·Na]+ with
the cation into the crown ether moiety is known to be strong,39 bromide and iodide gave binding constants of 1411 and 374
and can be considered to induce a turn-on effect, resulting M−1, respectively (Fig. S28 and S29†), being eight and 12 times
in stronger anion binding compared to the neutral larger than with L alone. These eight to twelve times larger
receptor.21,25,48,49
The anion binding stoichiometry of the neutral receptor L by the sodium complexation in the crown ether part.
alone was studied with Job’s plot analysis (Fig. S22†), and then Based on the X-ray structures discussed above, the larger
binding constants clearly manifest the turn-on effect induced
the anion binding affinity was measured via 1H NMR titrations potassium cation is expected to form a dimer and thus show
in 4 : 1 CDCl3/DMSO. The binding constant calculation using different binding properties in solution. To verify the stoichio-
1 : 1 binding stoichiometry (verified by the Job’s plot analysis) metry of the L + K+ in solution, a Job’s plot analysis with L
resulted in Ka = 828 for chloride, 175 for bromide, and 32 M−1 and KBPh4 was done (Fig. S21b†) showing clearly a 2 : 1
for iodide. These values are in very close agreement with those complex, viz. [2L·K]+. The stoichiometry of the anion complexa-
obtained previously for a benzo-18-crown-6 based bis-urea tion of [2L·K]+ was then done via Job’s plot analysis. As is
This journal is © The Royal Society of Chemistry 2016
Dalton Trans., 2016, 45, 6481–6490 | 6487

View Article Online
Paper
Dalton Transactions
it is only 96 M−1 for the 1 : 2 complex (Fig. S30†). The large
binding constant for chloride is logical considering the pre-
organization of the dimeric assembly to host a suitably sized
anion inside the binding pocket, and larger number of inter-
acting groups in the dimer for anion binding. The second
binding event with excess chloride is explainable by hydrogen
bonding interactions of the chloride with the urea-groups
outside of the binding pocket.
The corresponding titration of [2L·K]+ with bromide
resulted in similar chemical shift changes as with chloride.
The binding behavior was modeled in similar manner giving
binding constants Ka1 = 14 713 M−1 for the 1 : 1 complex and
51 M−1 for the 1 : 2 complex (Fig. S31†). The smaller binding
constants obtained for bromide are in agreement with weaker
hydrogen bond acceptor nature of the bromide. The chemical
shift changes obtained from titrations between iodide and
[2L·K]+ were clearly smaller and due to the weakest hydrogen
bond acceptor character of iodide the 1 : 2 complex was absent.
The simple 1 : 1 binding model produced binding constant
Ka = 717 M−1 (Fig. S32†).
In summary, the results obtained from 1H NMR titrations
do clearly show the turn-on effect of the cation complexation
in the anion recognition behavior of L. Receptor L forms a 1 : 1
complex with sodium in solution, and this positively charged
complex has larger binding affinity toward halide anions com-
pared to L alone. When complexed with K+ in solution, dimeri-
zation of L receptors via crown ether complexation is observed.
For binding constant calculations, the active receptor is con-
sidered as the [2L·K]+ complex, which shows very interesting
solution-state behavior toward chloride (and bromide) with a
Fig. 7 Job’s plot analysis of (a) [L·Na]+ + Cl− and [2L·K]+ + Cl− com-
plexations. (b) The chemical shift differences of urea-protons Ha (solid
line) and Hb (dashed line) in L (red), [L·Na]+ (blue), and [2L·K]+ (green) very strong 1 : 1 complex formation with the anion. With these
upon stepwise addition of Cl−.
anions also a secondary complex formation (1 : 2) has to be
taken into account for successful modelling of the observed
chemical shifts. However, binding constants for 1 : 2 complexa-
clearly evident from the Job’s plot presented in Fig. 7a, the
same 2 : 1 stoichiometry of L persists also in the anion com-
plexation when L is complexed with potassium.
tions are negligible when compared with the 1 : 1 binding, and
thus the 1 : 1 complex between [2L·K]+ and the anion is the
main component (>99%) with chloride concentrations up to
one equivalent. The obtained anion binding constants with
the fitting errors are presented in Table 1.
The halide binding behavior of the dimeric complex [2L·K]+
1
in solution was done via H NMR titrations in the same 4 : 1
CDCl3/DMSO solvent system. As is evident from the chemical
shift differences presented in Fig. 7b, the dimerization via the
potassium cation complexation has clearly a different effect on
the binding mode of L toward chloride. This is seen from the
different effect on chemical shifts of Ha and Hb in [2L·K]+
complex compared to [L·Na]+ upon chloride binding.
Table 1 Binding constants (Ka) obtained from for L, [L·Na]+, and [2L·K]+
1H NMR titrations in 4 : 1 CDCl3/DMSO-d6 at 303 K
The crystal structure indicates that the stoichiometry for
[2L·K]+ + Cl− complexation is 1 : 1 (Fig. 5a), but the chemical
shift changes observed in the titration between [2L·K]+ and
chloride are best modeled with a mixed 1 : 1 + 1 : 2 binding
stoichiometry (where 1 : 1 is considered as [2L·K]+ + Cl−, and
1 : 2 as [2L·K]+ + 2Cl− with both species in equilibrium).
However, the 1 : 1 complex has >99% abundancy with the sec-
ondary 1 : 2 complex being present <1% with up to ca. 1 equiv.
of chloride (Fig. S30d†). The binding constant calculation
using the above model for [2L·K]+ + Cl– complexation gives
binding constant Ka1 = 42 247 M−1 for the 1 : 1 complex while
L
[L·Na]+
[2L·K]+
Cl−
Br−
I−
Ka
828 22
—
—
7609 576
—
—
—
a
b
b
Ka1
Ka2
42 247 20 877
96 29
a
Ka
175
—
—
3
1411 69
—
—
—
a
b
Ka1
Ka2
14 713 2657
51 21
a
Ka
32
3
374 16
717 58
a 1 : 1 (Ka) binding model used. b 1 : 1 (Ka1) + 1 : 2 (Ka2) binding model
used (see the text and ESI for details).
6488 | Dalton Trans., 2016, 45, 6481–6490
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
Conclusions
References
In this work we have comprehensively studied the behavior of
a ditopic benzo-15-crown-5 bis-urea receptor L as an ion pair
receptor in the solid-state and the anion binding properties of
the receptor’s sodium and potassium complexes in solution.
The synthesized benzo-15-crown-5 receptor with bis-urea func-
tionality (L) has binding sites for both the cation and anion,
respectively. A total of 13 crystal structures, from which 10 are
complexes with alkali halides, show how L acts as a ditopic ion
pair receptor in the solid-state. With smaller (and with better
size-fit) sodium cation, L forms a 1 : 1 complex, and binds the
anion as a contact (Cl−) or separate (Br− and I−) ion pair,
depending on the size and the polarizability of the anion.
When complexed with larger potassium or rubidium cations, L
forms a dimeric structure with intermolecular hydrogen
bonding enhancing the dimerization. This dimeric complex
binds all halide anions (F−, Cl−, Br−, and I−) as a separated ion
pair in the solid-state, and the anion is always bound via
hydrogen bonds formed to the urea groups. Although
similar in the dimerization behavior, the structures obtained
with potassium and rubidium halides also show differences
to one another. For example, complex 2L·KF is actually a
tetramer, formed by dimer of dimers with the fluoride anion
being bound through a complex solvent-mediated hydrogen
1 J. W. Steed and J. L. Atwood, Supramolecular Chemistry,
Wiley, Chichester, UK, 2nd edn, 2009.
2 N. Busschaert, C. Caltagirone, W. van Rossom and
P. A. Gale, Chem. Rev., 2015, 115, 8038, DOI: 10.1021/acs.
chemrev.5b00099.
3 P. A. Gale, N. C. Busschaert, J. E. Haynes,
L. E. Karagiannidis and I. L. Kirby, Chem. Soc. Rev., 2014,
43, 205, DOI: 10.1039/c3cs60316d.
4 N. H. Evans and P. D. Beer, Angew. Chem., Int. Ed., 2014, 53,
11716, DOI: 10.1002/anie.201309937.
5 J. W. Jones and H. W. Gibson, J. Am. Chem. Soc., 2003, 125,
7001, DOI: 10.1021/ja034442x.
6 F. Huang, J. W. Jones, C. Slebodnick and H. W. Gibson,
J. Am. Chem. Soc., 2003, 125, 14458, DOI: 10.1021/
ja036606f.
7 F. Huang, J. W. Jones and H. W. Gibson, J. Org. Chem.,
2007, 72, 6573, DOI: 10.1021/jo070792g.
8 B. Qiao, A. Sengupta, Y. Liu, K. P. McDonald, M. Pink,
J. R. Anderson, K. Raghavachari and A. H. Flood,
J. Am. Chem. Soc., 2015, 137, 9746, DOI: 10.1021/
jacs.5b05839.
9 S. Roelens, A. Vacca, O. Francesconi and C. Venturi, Chem.
– Eur. J., 2009, 15, 8296, DOI: 10.1002/chem.200900342.
bonding network, found also in a few other obtained struc- 10 S. K. Kim and J. L. Sessler, Acc. Chem. Res., 2014, 47, 2525,
tures. In solution the Job’s plot analyses of L with Na+ and K+
DOI: 10.1021/ar500157a.
confirmed the formation of 1 : 1 and 2 : 1 complexes, 11 A. J. McConnell and P. D. Beer, Angew. Chem., Int. Ed.,
respectively. The 1H NMR titrations performed with mono-
2012, 51, 5052, DOI: 10.1002/anie.201107244.
meric [L·Na]+ and the dimeric [2L·K]+ with chloride, bromide, 12 S. K. Kim and J. L. Sessler, Chem. Soc. Rev., 2010, 39, 3784,
and iodide show that the anion binding affinity of [L·Na]+ and
DOI: 10.1039/c002694h.
[2L·K]+ is clearly higher than with receptor
alone 13 G. J. Kirkovits, J. A. Shriver, P. A. Gale and J. L. Sessler,
L
showing clear positive cooperativity and turn-on effect in
J. Inclusion Phenom. Macrocyclic Chem., 2001, 41, 69, DOI:
anion binding upon cation complexation. Stronger anion
10.1023/A:1014468126351.
binding of [L·Na]+ compared to L alone is explainable by 14 D. M. Rudkevich, J. D. Mercer-Chalmers, W. Verboom,
electrostatic interactions between the cationic receptor
complex and the anion. The even stronger binding of all the
R. Ungaro, F. de Jong and D. N. Reinhoudt, J. Am. Chem.
Soc., 1995, 117, 6124, DOI: 10.1021/ja00127a027.
studied halides with the [2L·K]+ complex is explainable by the 15 P. D. Beer and S. W. Dent, Chem. Commun., 1998, 825, DOI:
formation of an excellent anion binding site upon dimeriza-
tion, having a larger number of urea groups to form more 16 M. J. Deetz, M. Shang and B. D. Smith, J. Am. Chem. Soc.,
hydrogen bonding interactions between the receptors and the 2000, 122, 6201, DOI: 10.1021/ja994487r.
anion, as seen in the solid-state complexes of L with potassium 17 R. Shukla, T. Kida and B. D. Smith, Org. Lett., 2000, 2,
and rubidium halides. Thus, the differences in solution behav- 3099, DOI: 10.1021/ol0063104.
ior between [L·Na]+ and [2L·K]+ is explainable by the structures 18 J. L. Sessler, S. K. Kim, D. E. Gross, C.-H. Lee, J. S. Kim and
10.1039/A800356D.
formed upon cation complexation as also seen in the solid-
state.
V. M. Lynch, J. Am. Chem. Soc., 2008, 130, 13162, DOI:
10.1021/ja804976f.
19 B. Akhuli and P. Ghosh, Chem. Commun., 2015, 51, 16514,
DOI: 10.1039/c5cc07291c.
20 M. P. Wintergerst, T. G. Levitskaia, B. A. Moyer, J. L. Sessler
and L. Delmau, J. Am. Chem. Soc., 2008, 130, 4129, DOI:
10.1021/ja7102179.
Acknowledgements
We thank the Academy of Finland (K. R.: Project no. 263256 21 Z. Sun, F. Pan, Triyanti, M. Albrecht and G. Raabe,
and 265328) and Graduate School of Univ. of Jyväskylä (T.M.)
for financial support. Johanna Lind, Dr Elina Kalenius, and
Eur. J. Org. Chem., 2013, 7922, DOI: 10.1002/
ejoc.201301032.
Esa Haapaniemi, M.Sc. are gratefully acknowledged for their 22 C. J. E. Haynes and P. A. Gale, Chem. Commun., 2011, 47,
help with the MS and NMR experiments.
8203, DOI: 10.1039/c1cc12061a.
This journal is © The Royal Society of Chemistry 2016
Dalton Trans., 2016, 45, 6481–6490 | 6489

View Article Online
Paper
Dalton Transactions
23 K. Brak and E. N. Jacobsen, Angew. Chem., Int. Ed., 2013, 38 C. Frassineti, S. Ghelli, P. Gans, A. Sabatini, M. S. Moruzzi
52, 534, DOI: 10.1002/anie.201205449.
24 P. Molina, A. Tarraga and M. Alfonso, Dalton Trans., 2014,
43, 18, DOI: 10.1039/c3dt52459k.
25 T. Mäkelä, E. Kalenius and K. Rissanen, Inorg. Chem., 2015,
54, 9154, DOI: 10.1021/acs.inorgchem.5b01577.
and A. Vacca, Anal. Biochem., 1995, 231, 374, DOI: 10.1006/
abio.1995.9984.
39 G. J. Kirkovits, R. S. Zimmerman, M. T. Huggins,
V. M. Lynch and J. L. Sessler, Eur. J. Org. Chem., 2002,
3768,
DOI:
10.1002/1099-0690(200211)2002:22<3768:
26 T. Mäkelä, M.-E. Minkkinen and K. Rissanen, Inorg. Chem.,
2016, 55, 1339, DOI: 10.1021/acs.inorgchem.5b02780.
27 CSD version 5.36 (November 2014); 1.9.2015. CSD search
AID-EJOC3768>3.0.CO;2-R.
40 M. Barboiu, D. Dumitrescu and A. van der Lee, Cryst.
Growth Des., 2014, 14, 3062, DOI: 10.1021/cg500323r.
defined as benzo-15-crown-5 or benzo-18-crown-6 (B18C6) 41 CSD version 5.36 (November 2014); 1.9.2015. CSD search
containing Z–(CO)–NH–Z (Z
fragment.
28 CSD version 5.36 (November 2014); 1.9.2015. CSD search
=
any group except H)
defined as benzo-15-crown-5 fragment with Na+, K+, or Rb+.
42 D. Song, W. You, Y. Zhao, T. Hu, Q. Du and X. Zheng,
J. Mol. Struct., 2008, 888, 337, DOI: 10.1016/
j.molstruc.2008.01.003.
defined as Z–NH–(CO)–NH–Z (Z = any group except H) frag-
ment with M+ and X− (M+ = alkali cation, X− = halide anion). 43 V. Amendola, D. Esteban-Gómez, L. Fabbrizzi, M. Licchelli,
29 CrysAlisPro, version 171.37.31, Agilent Technologies, Ltd,
E. Monzani and F. Sancenón, Inorg. Chem., 2005, 44, 8690,
Yarnton, England, 2014.
DOI: 10.1021/ic050871f.
30 R. W. W. Hooft, COLLECT Data Collection Software, Bruker 44 CSD version 5.36 (November 2014); 1.9.2015. CSD search
AXS, Delft, Netherlands, 2008.
31 Z. Otwinowski and W. Minor, Methods Enzymol. Macromol.
Crystallogr., Part A, 1997, 276, 307.
32 G. M. Sheldrick, SADABS Version 2008/2, University of
Göttingen, Göttingen, Germany, 2008.
33 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystal-
logr., 2008, 64, 112, DOI: 10.1107/S0108767307043930.
34 L. Palatinus and G. J. Chapuis, Appl. Crystallogr., 2007, 40,
786, DOI: 10.1107/S0021889807029238.
defined as distances between Cl− and hydrogen in Z-NH-
(CO)-NH-Z (Z = any group except H) fragment.
45 Mercury 3.5. Calculated using Void-module, Contact
Surface with 1.2 Å probe radius and 0.7 approximated grid
spacing; C. F. Macrae, I. J. Bruno, J. A. Chisholm,
P. R. Edgington, P. McCabe, E. Pidcock, L. Rodriguez-
Monge, R. Taylor, J. van de Streek and P. A. Wood,
J. Appl. Crystallogr., 2008, 41, 466, DOI: 10.1107/
S0021889807067908.
35 M. C. Burla, M. Camalli, B. Carrozzini, G. L. Cascarano, 46 P. Thordarson, Chem. Soc. Rev., 2011, 40, 1305, DOI:
C. Giacovazzo, G. Polidori and R. J. Spagna, Appl. Crystal-
logr., 2003, 36, 1103, DOI: 10.1107/S0021889803012585.
36 O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard
10.1039/c0cs00062k.
47 A. J. Fry, J. Org. Chem., 2015, 80, 3758, DOI: 10.1021/
jo502575s.
and H. J. Puschmann, Appl. Crystallogr., 2009, 42, 339, DOI: 48 J. V. Gavette, J. Lara, L. L. Reling, M. M. Haley and
10.1107/S0021889808042726.
D. W. Johnson, Chem. Sci., 2013, 4, 585, DOI: 10.1039/
37 P. van der Sluis and A. L. Spek, Acta Crystallogr., Sect. A:
c2sc21501b.
Fundam. Crystallogr., 1990, 46, 194, DOI: 10.1107/ 49 J. Romański and P. J. Piątek, Org. Chem., 2013, 78, 4341,
S0108767389011189.
DOI: 10.1021/jo4003322.
6490 | Dalton Trans., 2016, 45, 6481–6490
This journal is © The Royal Society of Chemistry 2016