Effect of a methyl group on the spontaneous resolution of a square-pyramidal coordination compound: Crystal packing and conglomerate formation

Article abstract of DOI:10.1039/c2ce25497b

Reaction of [Cu(pic)₂]·2H₂O (where pic stands for 2-picolinato) with 2-({[2-(dimethylamino)ethyl]amino}methyl)phenol (HL ¹) produces the square-pyramidal complex [CuL¹(pic)] (1), which crystallizes as a conglome

Full text of DOI:10.1039/c2ce25497b

View Online / Journal Homepage
CrystEngComm
Accepted Manuscript
This is an Accepted Manuscript, which has been through the RSC Publishing peer
review process and has been accepted for publication.
CrystEngComm
Accepted Manuscripts are published online shortly after acceptance, which is prior
to technical editing, formatting and proof reading. This free service from RSC
Publishing allows authors to make their results available to the community, in
citable form, before publication of the edited article. This Accepted Manuscript will
be replaced by the edited and formatted Advance Article as soon as this is available.
www.rsc.org/crystengcomm
Volume 12 | Number 5 | May 2010 | Pages 1317–1658
To cite this manuscript please use its permanent Digital Object Identifier (DOI®),
which is identical for all formats of publication.
More information about Accepted Manuscripts can be found in the
Information for Authors.
Please note that technical editing may introduce minor changes to the text and/or
graphics contained in the manuscript submitted by the author(s) which may alter
content, and that the standard Terms & Conditions and the ethical guidelines
that apply to the journal are still applicable. In no event shall the RSC be held
responsible for any errors or omissions in these Accepted Manuscript manuscripts or
any consequences arising from the use of any information contained in them.
HOT ARTICLE
COMMUNICATION
Hall et al.
HIGHLIGHT
Wang, Su et al.
Alginate-mediated routes to the
selective synthesis of complex metal
oxide nanostructures
Long
Ionothermal synthesis of a new
open-framework zinc phosphate
NIS-3 with low framework density
pH effect on the assembly of
metal–organic architectures
www.rsc.org/crystengcomm
Registered Charity Number 207890

CrystEngComm
Page 1 of 8CrystEngComm
Dynamic Article Links►
View Online
Cite this: DOI: 10.1039/c0xx00000x
www.rsc.org/xxxxxx
PAPER
Effect of a methyl group on the spontaneous resolution of a square-
pyramidal coordination compound: crystal packing and conglomerate
formation
Apurba Biswas,aCarolina Estarella,bAntonio Frontera,bPablo Ballester,c,dMichael G. B. Drew,ePatrick
5
Gamez*d,fand Ashutosh Ghosh*a
Received (in XXX, XXX) XthXXXXXXXXX 2012, Accepted Xth XXXXXXXXX 2012
DOI: 10.1039/b000000x
Reaction of [Cu(pic)2]ꢀ2H2O (where pic stands for 2ꢁpicolinato) with 2ꢁ({[2ꢁ
(dimethylamino)ethyl]amino}methyl)phenol (HL1) produces the squareꢁpyramidal complex [CuL1(pic)]
10 (1), which crystallizes as a conglomerate (namely a mixture of optically pure crystals) in the Sohncke
space group P212121. The use of the methylated ligand at the benzylic position, i.e. (±)ꢁ2ꢁ(1ꢁ{[2ꢁ
(dimethylamino)ethyl]amino}ethyl)phenol (HL2), yields the analogous fiveꢁcoordinate complex
[CuL2(pic)] (2) that now crystallizes as a true racemate (namely the crystals contain both enantiomers) in
the centrosymmetric space group P21/c. DFT calculations indicate that the presence of the methyl group
15 indeed leads to a distinct crystallization behaviour, not only for intramolecular steric effects, but also
because its involvement in nonꢁcovalent C−Hꢀꢀꢀπ and hydrophobic intermolecular contacts appears to be
an important factor contributing to the crystalꢁlattice (stabilizing) energy of 2
in a conglomerate form are not well understood, and indeed there
are known examples where comparable complexes have been
reported that generate a true racemate or a conglomerate;23
however, it has been mentioned that the crystal packing of the
50 chiral molecules may be the critical factor,31 associated with
kinetics and thermodynamic issues.32
In the present study, an interesting illustration of the effect of
small variations of the crystal packing on the formation of a
racemic compound or a racemic mixture is described. Actually,
1. Introduction
Optical isomerism is intimately linked to the development of
20 modern coordination chemistry instigated by Alfred Werner at
the end of the nineteenth century.1ꢁ3 In fact, the chirality exhibited
by sixꢁcoordinate complexes represented a proof of the octahedral
configuration proposed by Werner for such coordination
compounds.4ꢁ7
25
The crystallization of chiral molecules may produce a
55 the
reaction
of
2ꢁ({[2ꢁ
crystalline racemate that can be a racemic compound (true
racemate), a racemic solid solution (pseudoracemate) or a
racemic mixture (conglomerate).8 The conglomerate is of
particular interest since it corresponds to a spontaneous resolution
30 of the two enantiomers, which is reflected by a mechanical
mixture of the two pure single stereoisomers. For instance,
Raymond and colleagues have reported an interesting case of
conglomerate for which the two stable enantiomers (even in
solution) could be separated manually.9
(dimethylamino)ethyl]amino}methyl)phenol (HL1; Scheme 1)
with [Cu(pic)2]ꢀ2H2O (where pic stands for 2ꢁpicolinato) and
triethylamine
in
methanol yields
the
conglomerate
[CuL1(pic)](1), while the analogous reaction with (±)ꢁ2ꢁ(1ꢁ{[2ꢁ
60 (dimethylamino)ethyl]amino}ethyl)phenol (HL2; Scheme 1)
produces the true racemate [CuL2(pic)] (2).
35
A remarkable case of a coordination compound giving a
conglomerate is Werner’s cobalt(III) complex cisꢁ
[Co(en)2(NO2)2]Cl (where en stands for 1,2ꢁdiaminoethane),
whose optical isomers have been isolated by Ivan Bernal in
1985.10 Since then, although they are quite rare,11, 12 a number of
Scheme 1 2ꢁ({[2ꢁ(dimethylamino)ethyl]amino}methyl)phenol (HL1)
and(±)ꢁ2ꢁ(1ꢁ{[2ꢁ(dimethylamino)ethyl]amino}ethyl)phenol (HL2)
40 coordination conglomerates have been reported.13 Spontaneous
resolution of sixꢁcoordinate complexes is clearly the most
common,14ꢁ23 but fourꢁcoordinate,24ꢁ26 fiveꢁcoordinate,27 sevenꢁ
coordinate,28eightꢁcoordinate29and nineꢁcoordinate30 systems
have also been described in the literature.
65
The crystal structures of conꢁ1 and racꢁ2 have been analysed
and their distinct crystallization behaviour has been probed by
circular dichroism (CD) and studied theoretically.
45
The reasons why certain (coordination) compounds crystallize
This journal is © The Royal Society of Chemistry [year]
CrystEngComm, 2012, 14, 00–00 |1

CrystEngComm
Page 2 of 8
View Online
concentrated HCl (5 mL) and the solvent was then evaporated
under reduced pressure to dryness. The reduced Schiffꢁbase
60 ligand HL1 was extracted from the solid mass with methanol and
this methanol solution (ca. 20 mL) was used for the preparation
of the complexes. HL2 was synthesized in a similar manner using
2. Experimental section
2.1 Materials and methods
All commercially available reagents were used as received.
Bis(picolinatoꢁN,O)copper(II)dihydrate, i.e.[Cu(pic)2]ꢀ2H2O, was
prepared as described.33 Elemental analyses (C, H, N) were
performed using a PerkinꢁElmer 240C elemental analyzer.
Infrared spectra were recorded in the range 4500−500 cm−1 on a
PerkinꢁElmer RXI FTꢁIR spectrophotometer (in KBr pellets). Xꢁ
ray diffraction data for 1-Λ were collected at 150 K from a single
10 crystal mounted atop a glass fiber (at 50 mm from the CCD) with
an Oxford Diffraction XꢁCalibur CCD diffractometer using
graphite monochromated MoꢁKα (λ= 0.71073 Å) radiation. 321
frames were measured with a counting time of 10 seconds. Data
analysis was carried out with the CrysAlis program.
Data for racꢁ2 were collected at 295 K from a single crystal
mounted on a Bruker SMART diffractometer equipped with a
graphite monochromator and MoꢁKα (λ = 0.71073 Å) radiation.
The structures were solved employing the SHELXSꢁ97 direct
methods program and refined by fullꢁmatrix leastꢁsquares on F2.34
20 The nonꢁhydrogen atoms were refined with anisotropic thermal
parameters. The hydrogen atoms bonded to a carbon were
included in geometric positions and given thermal parameters
equivalent to 1.2 times those of the atom to which they were
attached. Absorption correction for 1-Λ was carried out using the
25 ABSPACK program. The structures were refined on F2 to R1
0.0449, 0.0336; wR2 0.0824, 0.0904 for 3625 and 3635 data for 1
and 2, respectively, with I> 2σ(I). Crystallographic details are
summarized in Table 1.
Circular Dichroism (CD) and UVꢁVis spectroscopic
30 measurements were done with a Chirascan CD Spectrometer
(Applied Photophysics Limited). For the solidꢁstate studies, the
pellets were placed as close as possible to the sensitive surface of
the photomultiplier tube to diminish the scattering effect. Sample
rotation around the incidentꢁlight axis was done to exclude the
35 presence of linear dichroism (LD) and linear birefringence (LB).
All measurements were performed using the following
parameters: single scans, 10 repetitions, continuous scanning
mode, 1 nm SBW, 5 nm data interval, and air baseline had been
done to obtain the UVꢁVis and CD spectra. For the solution
40 studies, the spectra were recorded using methanolic solutions of
the complexes at a concentration of 1 × 10–3 M. For the solidꢁ
state measurements, all the pellets were prepared as follows: in
all the samples, a constant mass of KCl, namely 100 mg, was
used where distinct quantities of the sample to be analysed were
45 added. Thus, to avoid saturation of the detector, the CD spectra
were recorded applying weight% concentrations of 0.3, 0.6 and
1.0.
5
2ꢁhydroxyacetophenone (0.60 mL,
salicylaldehyde.
5
mmol) instead of
65
Synthesis of the complexes [CuL1(pic)] (1) and [CuL2(pic)]
(2): a methanolic (20 mL) suspension of [Cu(pic)2]ꢀ2H2O (1.70 g,
5 mmol) was added to a methanolic solution of HL1 (prepared as
mentioned above). Triethylamine (5 mmol, 0.7 mL) was then
added dropwise to this reaction mixture under constant stirring.
70 Next, the resulting suspension was heated at reflux for 1 h, after
which the solid [Cu(pic)2]ꢀ2H2O complex had almost completely
dissolved. The solution was then cooled to room temperature and
filtered to remove a tiny amount of remaining precursor
[Cu(pic)2]ꢀ2H2O. The darkꢁgreen solution was left unperturbed at
75 room temperature for the slow evaporation of the solvent. After a
few days, green single crystals of conꢁ1 suitable for Xꢁray
diffraction studies were obtained. Deep green single crystals of
racꢁ2 were obtained in the same way using HL2 instead of HL1.
Compound 1: (Yield: 1.285 g, 68 %), Anal. calc. for
80 C17H21CuN3O3: C, 53.89; H, 5.59; N, 11.09 %. found: C, 53.67;
15
H, 5.53; N, 10.89 %. IR (KBr): ν ν(C–N),
(N–H), 3233 cm−1,
1598 cm−1; λmax (nm), [εmax (dm3 mol−1 cm−1)] (methanol), 702
(112), 422 (744).
Compound 2: (Yield: 1.392 g, 71 %), Anal. calc. for
85 C18H23CuN3O3: C, 55.02; H, 5.90; N, 10.69 %. found: C, 54.87;
H, 5.99; N, 10.82 %. IR (KBr): ν
(N–H), 3187 cm−1, ν(C–N),
1597 cm−1; λmax (nm), [εmax (dm3 mol−1 cm−1)] (methanol), 702
(103), 422 (662).
A moderately strong and sharp peak due to N–H stretching
90 vibration at 3233 (1), and 3187 cm−1 (2) indicates that the imine
double bond of the Schiff base has been reduced. The strong
bands at 1650 and 1658 cm−1 are likely to be due to the
antisymmetric stretching mode of the carboxylate group, whereas
the bands at 1358 and 1347 cm−1 can be attributed to the
95 symmetric stretching modes of the carboxylato ligands in 1 and 2,
respectively.
2.3 Theoretical methods
The geometries of all complexes investigated in the present work
were computed at the DFꢁB97ꢁD/SVP level of theory by means
100 of the Gaussianꢁ09 program.35 The optimizations were carried out
using the crystallographic coordinates as the starting model.
Density functional theory (DFT) offers the possibility of tackling
large systems efficiently, and it is less basisꢁset dependent than
37
the more advanced waveꢁfunction methods.36,
However,
105 conventional DFT techniques fail to treat dispersion effects
completely. Recently, the DFT community has developed a
variety of methods for the treatment of van der Waals
(dispersion) interactions, including treatment with specialized
functionals, such as B97ꢁD.38 Hence, we have used the B97ꢁD
110 method together with the SVP basis set39 in order to reach an
acceptable compromise between the size of the system and the
accuracy of the results. In addition, we have used the density
fitting approximation to reduce the time of calculation. DFꢁDFT
method is considerably faster than the DFT and the interaction
115 energies and equilibrium distances are almost identical for both
2.2 Synthesis and characterization
Synthesis of the reduced Schiffꢀbase ligands 2ꢀ[(2ꢀ
50 dimethylaminoꢀethylamino)ꢀmethyl]ꢀphenol (HL1) and 2ꢀ[1ꢀ(2ꢀ
dimethylaminoꢀethylamino)ꢀethyl]ꢀphenol (HL2): HL1 was
prepared by refluxing for one hour a solution of salicylaldehyde
(0.52 mL, 5 mmol) and N,Nꢁdimethylethylenediamine (0.54 mL,
5 mmol) in methanol (30 mL). The reaction mixture was cooled
55 down subsequently to 0 ºC, and solid sodium borohydride (210
mg, 6 mmol) was added slowly under stirring. After completion
of the addition, the resulting solution was acidified with
2
|CrystEngComm, 2012, 14, 00–00
This journal is © The Royal Society of Chemistry [year]

Page 3 of 8
CrystEngComm
View Online
methods.40
3. Results and discussion
3.1 Structure description
Reaction of HL1 or HL2 with [Cu(pic)2]ꢀ2H2O in refluxed
methanol in the presence of a base, namely triethylamine, to
deprotonate the phenolic group provides the corresponding
coordination compounds conꢁ1 and racꢁ2 as crystalline materials.
5
Fig. 1 Representation of the molecular structure of [CuL1(pic)] (1-Λ), as
determined by Xꢁray diffraction. Only the H atom involved in hydrogenꢁ
bonding interactions (amine group N19) is shown for clarity
[CuL1(pic)] (con-1). Compound
conglomerate in the (chiral) Sohncke space groupP212121 (Table
1
crystallizes as
a
40 Table 2 Coordination bond lengths (Å) and angles (º) for 1-Λ and 2
10 1). In our first structure determination, with details reported in
Tables 1 and 2, the refined absolute structure (Flack) parameter
[+0.016(15)] clearly shows that the coordinates correspond to the
absolute structure of the molecule found in the crystal having the
Λ configuration, and the structure will therefore subsequently be
15 called 1-Λ. The coordination bond lengths and angles are listed in
Table 2.
Compound
1-Λ
2
Cu1−O11
Cu1−N19
Cu1−O38
Cu1−N31
Cu1−N22
1.907(2)
1.986(2)
1.993(2)
1.985(2)
2.333(3)
1.914(1)
1.992(2)
2.002(2)
2.013(2)
2.310(2)
Table 1 Summary of the crystallographic data for 1-Λ and 2
O11−Cu1−N19
N19−Cu1−O38
O38−Cu1−N31
N31−Cu1−O11
O11−Cu1−N22
N19−Cu1−N22
O38−Cu1−N22
N31−Cu1−N22
N19−Cu1−N31
O11−Cu1−O38
94.99(9)
90.92(9)
81.99(9)
93.00(10)
107.51(10)
83.30(11)
97.91(10)
93.97(10)
172.01(10)
154.38(11)
93.81(6)
93.25(5)
81.44(6)
90.93(6)
104.24(7)
83.61(7)
95.58(6)
97.72(7)
174.62(6)
159.57(6)
1-Λ
C17H21CuN3O3
378.91
orthorhombic
P212121
7.3729(4)
11.2139(4)
20.1034(8)
90
2
C18H23CuN3O3
392.94
Monoclinic
P21/c
9.126(5)
13.953(5)
14.834(5)
103.643(5)
1836.96(14)
4
Formula
Formula weight
Crystal system
Space group
a/Å
b/Å
c/Å
β
/°
V/Å3
1662.14(13)
4
Z
Dcalc/g cm−3
ꢁ/mm−1
1.514
1.421
1.334
1.21
F(000)
788
820
R(int)
0.0457
0.032
Total Reflections
Flack parameter
Unique reflections
I >2σ(I)
8525
0.016(15)
4439
23296
ꢁ
4633
3635
3625
R1, wR2
Temp (K)
0.0449, 0.0892
150(2)
0.0336, 0.0904
295(2)
Fig. 2 a) Crystal packing of 1-Λ showing the hydrogenꢁbonding network
(orange dotted lines; N19−H19ꢀꢀꢀO39a = 2.853(3) Å; b) C−Hꢀꢀꢀπ
45 interactions (red dotted line; H33ꢀꢀꢀCg(5) = 2.85 Å) observed in the solidꢁ
state structure of 1. C−Hꢀꢀꢀπ contact is also shown in spaceꢁfilling mode.
Only the H atoms involved in hydrogenꢁbonding or C−Hꢀꢀꢀπ interactions
are shown for clarity. Symmetry operation: a = 1−x, −1/2+y,1/2−z
The molecular structure of 1-Λ exhibits a copper(II) centre
20 (Figure 1), which is fiveꢁcoordinate in a squareꢁpyramidal
environment (τ = 0.29),41 whose basal plane is defined by the
nitrogen and oxygen atoms from the picolinato ligand (N31 and
O38), and the nitrogen and oxygen atoms from the deprotonated
ligand L1 (N19 and O11). The axial position is occupied by the
25 nitrogen atom N22, belonging to the N,Nꢁdimethylaminoethyl
arm of tridentate L1, which thus adopts a facial configuration
around the metal ion. The Cu−N and Cu−O bond distances are in
normal ranges for this type of coordination environment.42, 43 The
basal angles of the square pyramid, varying from 81.99(9) to
30 94.99(9)º, reflect a distortion of the coordination geometry that is
most likely due to the small bite angle (O38−Cu1−N31 =
81.99(9)º) of the picolinato ligand. Hence, the basal coordinating
atoms O11, N31, O38, N19 deviate from the leastꢁsquares mean
plane by −0.217(2), 0.240(3),−0.240(2), and 0.217(3) Å,
35 respectively, thus showing a tetrahedral distortion.
Table 3 Supramolecular interactions (H bonds and C−Hꢀꢀꢀπ contacts)
50 present in the solidꢁstate structures of 1-Λ and 2
Hydrogen bonds
Compd.
1-Λ
D–HꢀꢀꢀAa
AꢀꢀꢀH (Å)
2.10
2.22
DꢀꢀꢀA (Å)
2.853(3)
3.078(2)
∠D–H–A (°)
139
N19–H19ꢀꢀꢀO39a
2
N19–H19ꢀꢀꢀO38b
157
C−Hꢂꢂꢂπ contacts
Compd.
1-Λ
X−HꢀꢀꢀCg
C33–H33ꢀꢀꢀCg(5)
CgꢀꢀꢀH (Å) XꢀꢀꢀCg (Å)
∠X–H–Cg (°)
2.85
3.344(4)
114
aSymmetry operations: a = 1−x, −1/2+y,1/2−z; b = 1−x, −y,2−z.
The crystal packing of 1-Λ shows the formation of
supramolecular hydrogenꢁbonded polymeric chains along the
crystallographic b axis, whose building units, namely molecules
55 of 1-Λ, are connected by means of strong hydrogen bonds (N19–
H19ꢀꢀꢀO39a = 2.853(3) Å; Table 3 and Figure 2a). In addition,
This journal is © The Royal Society of Chemistry [year]
CrystEngComm, 2012, 14, 00–00 |3

CrystEngComm
Page 4 of 8
View Online
these supramolecular chains are connected to each other through
C−Hꢀꢀꢀπ bonding contacts between a pyridyl C−H group from a
picolinato ligand and a phenol ring belonging to a neighbouring
L1 ligand (the hydrogenꢀꢀꢀring centroid distance, i.e. H33ꢀꢀꢀCg(5),
is 2.85 Å; Table 3 and Figure 2b), producing a 3ꢁD framework.
As already mentioned above, 1-Λ crystallizes in a chiral space
group, i.e.P212121. Hence, the crystal packing of the coordination
compound obtained by Xꢁray analysis of one single crystal
necessarily reveals the presence of a sole enantiomer, in this first
5
10 case the stereoisomer Λ (Figure 3).
50
Fig. 4 Solidꢁstate UVꢁVis spectra of conꢁ1 recorded at different
concentrations (0.3, 0.6 and 1.0 weight% in KCl)
When a single crystal of conꢁ1, necessarily containing just a
single enantiomer of 1, is dissolved in methanol (at a millimolar
concentration), the corresponding silent CD spectrum of the
55 solution suggests the racemization of the complex (Figure S4),
and therefore confirms its spontaneous resolution (conglomerate
formation) during the course of the crystallization.
Fig. 3 Molecular structure of the Λ enantiomer of compound 1
Since 1 is obtained from achiral ligands (pic and L1), the ꢃ
enantiomer is expected to be formed as well in solution. In fact,
15 several single crystals grown from a solution containing 1 as a
racemic mixture have been examined by Xꢁray diffraction, and
indeed the ꢃ enantiomer of 1 has also been established via a full
structure determination (Figure S1).† Crystallographic and metric
data of this structure 1-ꢀ are identical to those of the Λ
20 stereoisomer (Tables S1 and S2), except for the Flack parameter.
Therefore, racemic compound 1 represents the second reported
example of a fiveꢁcoordinate complex that crystallizes as a
conglomerate.27
To investigate further this unusual case of spontaneous optical
25 resolution, solution and solidꢁstate CD and UVꢁVis spectroscopic
studies have been carried out on 1. The solution studies were
performed using a racemic mixture of 1 while in the solidꢁstate
experiments a conglomerate single crystal was used. Both in the
solidꢁstate (Figure 4) and in solution (Figure S2), the UVꢁVis
30 spectra show an absorption band around 422 nm that can be
attributed to (metalꢁtoꢁligand chargeꢁtransfer) MLCT transitions.
The very weak band(s) observed in the 630−730 nm region of the
solution spectrum (Figure S2) may be ascribed to dꢁd electronicꢁ
Fig. 5 Solidꢁstate spectra of 1 (single crystals) at different concentrations
60 (weight% in KCl), at room temperature. The negative peaks are marked in
red with the positive ones in blue
[CuL2(pic)] (rac-2). The use of ligand HL2 (instead of HL1),
which contains a methyl group at the benzylic carbon atom (red
marked in Scheme 1), gives rise to a drastically distinct
65 crystallization process. Indeed, compound 2 crystallizes in the
achiral centrosymmetric space group P21/c. The molecular
structure of 2 is very similar to that of 1-Λ (see Figures 1 and 6).
transfer absorptions in
a
squareꢁpyramidal CuII (3d9
35 configuration) environment (ground state in C4ν).44 These bands
are not seen in the solidꢁstate spectrum, most likely as a result of
the high dilution of the system. However, the UVꢁVis data
suggest that the coordination geometry observed in the crystal
structure of 1 is maintained in solution.
40
The solidꢁstate CD spectra of compound conꢁ1 have been
recorded at different concentrations (0.3, 0.6 and 1.0 weight% in
KCl) using powdered single crystals in KCl pellets (one single
crystal was used for each measurement) at room temperature
(Figure 5). The (concentrationꢁdependent) CD spectra present
Fig. 6 Representation of the molecular structure of [CuL2(pic)] (2), as
70 determined by Xꢁray diffraction. Only the H atom involved in hydrogenꢁ
bonding interactions (amine group N19) is shown for clarity. The
additional methyl group (C181) and the benzylic carbon atom (C18) are
labelled
45 Cotton effects with negative peaks observed at 338 nm and 682
nm (red marked in Figure 5), and positive ones at 422 nm and
574 nm (blue marked in Figure 5). The CD spectra of both
enantiomers could be obtained (see Figure S3).
However, a closer look at the coordination bond distances and
75 angles reveals some differences (see Table 2), although the two
4
|CrystEngComm, 2012, 14, 00–00
This journal is © The Royal Society of Chemistry [year]

Page 5 of 8
CrystEngComm
View Online
squareꢁpyramidal complexes exhibit almost identical geometries
(τ41 = 0.25 for 2 while it is 0.29 for 1-Λ). For instance, all
coordination bond lengths are slightly larger (by 0.006−0.028 Å),
except the axial Cu1−N22 distance (which is 0.023 Å shorter),
different weight% concentrations (depicted in Figure 8) are
clearly silent (the positive peak observed at the highest
concentration, i.e. 1%, is not due to a Cotton effect but most
likely to a concentration effect45). The CD spectrum of a
5
and the angles N31−Cu1−O11, O11−Cu1−N22, O38−Cu1−N22 55 millimolar methanolic solution of 2 (prepared from one single
and N31−Cu1−N22 vary by about 3º. For 2, the basal
coordinating atoms O11, N31O38, N19 deviate from the leastꢁ
squares mean plane by −0.181(1), 0.123(1),−0.183(1), and
0.100(1) Å, respectively. These smaller structural variations
crystal) is logically CDꢁsilent as well (Figure S5). It should be
noted that the solidꢁstate and solution UVꢁvis spectra of 2 (Figure
S6) are analogous to those of 1. These results are not at all
surprising since both complexes exhibit almost identical
10 compared to those in 1-Λ, are consistent with to steric effects of 60 coordination geometries (see Figures 1 and 6).
the methyl group C181.
It has to be mentioned here that the bulkiness of this methyl
substituent is not the sole distinction between 1-Λ and 2.
Actually, this methyl group converts the benzylic carbon C18 into
15 a chiral centre. Namely, ligand HL2 is asymmetric while ligand
HL1 is not chiral. Hence, the crystal packing of racꢁ2 shows very
interesting features. First, as expected from the fact that racꢁ2
crystallizes in a nonꢁchiral space group, the two enantiomers of
the ligand are found in the solidꢁstate structure obtained from one
20 singleꢁcrystal (R and S; Figure 7). Second, it is noteworthy that
the R ligand exclusively produces the ꢃ enantiomer of the
complex while the S ligand generates the configuration Λ (Figure
7). In addition, the Λ and ꢃ enantiomers form a doubly hydrogenꢁ
Fig. 8 Solidꢁstate CD spectra of racꢁ2 (single crystals) at different
bonded dimer (orange dotted lines in Figure 7; N19–H19ꢀꢀꢀO38b =
25 3.078(2) Å, Table 2).
concentrations (weight% in KCl), at room temperature.
It should be mentioned here that since a racemic mixture of the
ligand has been used and that both the Λ and ꢃ coordination
isomers can be produced, four diastereomeric compounds may be
expected, namely the (R/ꢃ), (S/Λ), (R/Λ) and (S/ꢃ) complexes.
30 Moreover, if these complexes form dimers in the solid state, then
six different combinations could be expected, namely
Thus, in contrast to 1 (conglomerate), complex 2 behaves as a
65 true racemate. Interestingly, two selfꢁrecognition phenomena
(which are certainly linked to each other) occur upon
crystallization. Indeed, each enantiomer of the racemic ligand
mixture generates an optical isomer of the copper complex (ꢃ for
R and Λ for S). Then, the (R/ꢃ) and (S/Λ) compounds selfꢁ
70 assemble by means of hydrogen bonds to yield a racemic dimer,
and therefore a true racemate in which the single crystals contain
both optical isomers. It is interesting to note that the achiral
ligand HL1 leads to a spontaneous resolution of the coordination
stereoisomers while the asymmetric ligand HL2 yields a racemic
75 mixture of the complexes.
(R/ꢃ)ꢀꢀꢀ(S/Λ),
(R/ꢃ)ꢀꢀꢀ(S/ꢃ),
(R/Λ)ꢀꢀꢀ(S/ꢃ),
(R/Λ)ꢀꢀꢀ(S/Λ),
(R/ꢃ)ꢀꢀꢀ(R/Λ) and (S/ꢃ)ꢀꢀꢀ(S/Λ). Therefore, we have collected Xꢁ
ray datasets for six different single crystals from the same
35 reaction batch. Interestingly, all six solidꢁstate structures obtained
belong to the racemic dimer (R/ꢃ)ꢀꢀꢀ(S/Λ). As compound racꢁ2 is
obtained with a yield of 71% (which is well above the maximum
value of 16.7% that would be expected if all six dimers exhibited
equal probability of crystallization), most likely a selfꢁrecognition
40 process takes place upon crystallization that leads to the likely
most stable association, namely the (R/ꢃ)ꢀꢀꢀ(S/Λ) pair (racꢁ2).
Fig. 7 Formation of a racemic dimer of compound 2. The R ligand
generates the ꢃ coordination stereoisomer while the S ligand produces the
45
Λ one
In the solidꢁstate structure, the (R/ꢃ)ꢀꢀꢀ(S/Λ) dimers interact
through weak Csp2−HꢀꢀꢀO (3.184(9) Å) and Csp3−HꢀꢀꢀO (3.286(8)
Å) contacts.
Fig. 9 Left: portion of the crystal structure of compound conꢁ1. Right:
DFꢁB97ꢁD/SVP optimized structure. The green hydrogen atoms in Figure
9 left are those that have been replaced by methyl groups during the
theoretical study to evaluate their influence on the crystallization process
80
The formation of
a racemic compound has also been
50 corroborated by CD spectroscopy. The solidꢁstate spectra of 2 at
This journal is © The Royal Society of Chemistry [year]
CrystEngComm, 2012, 14, 00–00 |5

CrystEngComm
Page 6 of 8
View Online
3.2 Theoretical studies
The present case of conglomerate versus trueꢀracemate
formation for two closely related complexes has been
investigated further using DFT calculations. The theoretical study
has been divided into two central parts. In a first instance, a large
crystallographic fragment of compound 1-Λ has been analysed
(see Figure 9).
5
After optimization at the DFꢁB97ꢁD/SVP level of theory, the
fragment obtained has been compared to that found in the solidꢁ
10 state structure of conꢁ1. The geometry is clearly conserved after
gasꢁphase optimization (Figure 9); for instance, the shortest
CuꢀꢀꢀCu separation distance in the crystal lattice is practically
unchanged (7.37 Å instead of 7.38 Å). Only a negligible
reorientation of the molecules is noticed, which therefore
15 indicates that this arrangement is stable and most likely defines
the crystal packing that leads to spontaneous resolution. Actually,
when one of the hydrogen atoms of the benzylic carbon (depicted
in green in Figure 9) is replaced by a methyl group (that is when
HL1 is converted to the S enantiomer of HL2), a complete
20 reorganization of the molecules occurs, giving rise to a severe
alteration of the initial solidꢁstate pattern (a distinct molecular
assembly is observed). The computational data suggest that the
presence of the methyl groups will generate a distinct packing, as
indeed is noted experimentally.
Fig. 11 Interaction energies of (left) the fragment depicted in Figure 10
and of (right) the racemic, diastereomeric dimer observed in the solidꢁ
45
state structure of racꢁ2.
The interaction energy of the assembly (depicted in Figure 11)
at the B97ꢁD/SVP level of theory amounts to −120.6 kcal/mol. As
shown in Figure 11 (left), a number of weak bonding contacts are
found in the solidꢁstate structure of 2. The racemic dimer
50 exhibits, in addition to the strong N−HꢀꢀꢀO hydrogen bonds, four
C−HꢀꢀꢀO contacts that contribute as well to the large interaction
energy observed, namely −36.7 kcal/mol (Figure 11 (right)).
When the (green) methyl groups are replaced by hydrogen atoms
(i.e. when L2 is changed to L1), the interaction energy for the
55 dimer is −35.1 kcal/mol, therefore suggesting that the formation
of the hydrogenꢁbonded motif is slightly less favourable.
Moreover, the formation energy of the whole fragment weakens
to −111.9 kcal/mol. This notable change of the interaction energy
may be explained by a stabilization effect of the methyl groups in
60 the original structure (depicted in Figure 10). Indeed, the CH3ꢀꢀꢀπ
interactions observed in the solidꢁstate structure of 2 (C−Hꢀꢀꢀπ =
3.13 Å; Figure 11 (left)), may represent an important factor in
stabilizing the packing of the racemic dimers. In addition, these
methyl groups are involved in (intradimer) hydrophobic
65 interactions with the adjacent dimethylamino groups; obviously,
these stabilizing bonding contacts cannot be established when the
methyls are changed to hydrogen atoms, as indicated by the
different calculated interaction energies for the corresponding
dimers (Figure 12).
25
The same calculation pathway has been used with compound
racꢁ2. Thus, a large fragment of the crystal structure of racꢁ2 has
been used in the DFT calculations (Figure 10). This fragment is
characterized by an inversion centre resulting from the hydrogenꢁ
bonded (N−HꢀꢀꢀO =3.078(2) Å, HꢀꢀꢀO = 2.22 Å) formation of a
30 (racemic) dimer between the two enantiomers of a single
diastereoisomer of racꢁ2. The methyl groups connected to the
stereogenic carbon atoms are depicted in green in Figure 10. The
formation energy of this fragment representing the structural
arrangement of the molecules found in the solidꢁstate structure of
35 racꢁ2 has been determined and compared to that of the
hypothetical assembly for which the green methyl groups have
been replaced by hydrogen atoms (Figure 11).
70
Fig. 12 Left: Illustration showing the hydrophobicꢁinteraction region
(lightꢁblue ovals) in the Hꢁbonded dimer of 2 (with its interaction energy).
Right: interaction energy of the Hꢁbonded dimer when the methyl groups
are replaced by hydrogen atoms.
75
Since the presence/absence of the methyl group does not affect
the hydrogenꢁbond strength, each hydrophobic interaction can be
roughly estimated to −0.8 kcal/mol [(−36.7 − (−35.1)) / 2]. Such a
small energy difference (for instance, the hydrophobic
interactions in 2 are characterized by a stabilization energy of
Fig. 10 Portion of the crystal structure of compound racꢁ2 used in the
40 DFT calculations. The methyl groups differentiating HL2 from HL1 are
shown in green
80 −1.6 kcal/mol per dimer) may not be expected to have an
important effect on the crystallization process; however, it has
been shown that the difference in Gibbs free energy between a
6
|CrystEngComm, 2012, 14, 00–00
This journal is © The Royal Society of Chemistry [year]

Page 7 of 8
CrystEngComm
View Online
(2). For ESI and crystallographic data in CIF or other electronic format
see DOI: 10.1039/b000000x/
racemic crystal and a homochiral crystal is in the range 0.1−2.0
kcal/mol.32
60
1. K. H. Ernst, F. Wild, O. Blacque and H. Berke, Angew. Chem.ꢀInt.
Ed., 2011, 50, 10780ꢁ10787.
2. I. Bernal and G. B. Kauffman, J. Chem. Educ., 1987, 64, 604ꢁ610.
3. A. Werner, Z. Anorg. Chem., 1893, 3, 267ꢁ330.
65 4. K. H. Ernst and H. Berke, Chirality, 2011, 23, 187ꢁ189.
5. A. Werner and V. L. King, Ber. Dtsch. Chem. Ges., 1911, 44, 1887ꢁ
1898.
6. I. Bernal, J. Cetrullo and W. G. Jackson, Inorg. Chem., 1993, 32,
4098ꢁ4101.
Conclusions
The clarification of why some molecules resolve spontaneously
while others do not represents an important challenge in
stereochemistry since optical pure compounds are found in many
areas of chemical science, with valuable practical applications for
instance in pharmaceutical and materials chemistry.
5
70 7. G. B. Kauffman, Bull. Hist. Chem., 1997, 20, 50ꢁ59.
8. J. Jacques, A. Collet and S. H. Wilen, Enantiomers, racemates, and
resolutions Krieger Publishing Company, Malabar, FL, 1991.
9. M. Ziegler, A. V. Davis, D. W. Johnson and K. N. Raymond, Angew.
Chem.ꢀInt. Ed., 2003, 42, 665ꢁ668.
It has been proposed that lattice energy together with entropic
10 and kinetic effects have an influence on crystallization processes
and that very small differences may explain the formation of a
true racemate versus a conglomerate. In the present study, an
75 10. I. Bernal, Inorg. Chim. Acta, 1985, 96, 99ꢁ110.
11. M. Avalos, R. Babiano, P. Cintas, J. L. Jimenez and J. C. Palacios,
Orig. Life Evol. Biosph., 2004, 34, 391ꢁ405.
12. H. D. Flack, Helv. Chim. Acta, 2003, 86, 905ꢁ921.
13. L. PérezꢁGarcía and D. B. Amabilino, Chem. Soc. Rev., 2007, 36,
example is reported that shows how
a subtle structural
modification of a ligand coordinated to a copper(II) ion can lead
15 to a distinct crystallization behaviour. Actually, the replacement
of a benzylic hydrogen atom of HL1 by a methyl group,
generating ligand HL2 (Scheme 1), gives rise to a remarkable
situation in which the original Cu/L1 complex that produces
homochiral crystals is converted to the Cu/L2 compound that
20 gives racemic crystals. DFT calculations on compound 1 for
which the key hydrogen atom has been replaced by a methyl
group indicate that the original solidꢁstate structure is not
conserved. Similarly, when computational studies are carried out
on a fragment of compound 2 containing several molecules, its
25 interaction energy is considerably reduced when the methyl group
is replaced by a hydrogen atom, most likely because the CH3ꢀꢀꢀπ
and hydrophobic interactions observed in the experimental
structure cannot be established in the (new) crystal lattice.
80
941ꢁ967 and references therein.
14. J. L. Liu, X. Bao, J. D. Leng, Z. J. Lin and M. L. Tong, Cryst.
Growth Des., 2011, 11, 2398ꢁ2403.
15. A. Lennartson, Inorg. Chim. Acta, 365, 451ꢁ453.
16. W. Huang and T. Ogawa, Polyhedron, 2006, 25, 1379ꢁ1385.
85 17. S. Khatua, H. StoeckliꢁEvans, T. Harada, R. Kuroda and M.
Bhattacharjee, Inorg. Chem., 2006, 45, 9619ꢁ9621.
18. A. Johansson, M. Hakansson and S. Jagner, Chem.ꢀEur. J., 2005, 11,
5311ꢁ5318.
19. X. Ribas, J. C. Dias, J. Morgado, K. Wurst, M. Almeida, T. Parella, J.
90
Veciana and C. Rovira, Angew. Chem.ꢀInt. Ed., 2004, 43, 4049ꢁ4052.
20. W. K. Rybak, A. Skarzynska and T. Glowiak, Angew. Chem.ꢀInt. Ed.,
2003, 42, 1725ꢁ1727.
21. W. K. Rybak and A. Skarzynska, New J. Chem., 2003, 27, 1687ꢁ
1689.
95 22. R. G. Kostyanovsky, V. Y. Torbeev and K. A. Lyssenko,
Tetrahedron: Asymmetry, 2001, 12, 2721ꢁ2726.
23. J. Breu, H. Domel and A. Stoll, Eur. J. Inorg. Chem., 2000, 2401ꢁ
2408.
Acknowledgements
30 We thank CSIR, Government of India for Senior Research
Fellowship to AB, Sanction No. 09/028(0717)/2008ꢁEMRꢁI and
DST FIST for financial support for the Xꢁray diffraction. PG and
PB acknowledge the Institució Catalana de Recerca I Estudis
Avançats (ICREA). We also thank the EPSRC and the University
35 of Reading for funds for the XꢁCalibur diffractometer.
24. W. T. Liu, Y. C. Ou, Z. J. Lin and M. L. Tong, CrystEngComm,
100
2010, 12, 3487ꢁ3489.
25. Q. Z. Sun, Y. Bai, G. J. He, C. Y. Duan, Z. H. Lin and Q. J. Meng,
Chem. Commun., 2006, 2777ꢁ2779.
26. U. Siemeling, I. Scheppelmann, B. Neumann, A. Stammler, H. G.
Stammler and J. Frelek, Chem. Commun., 2003, 2236ꢁ2237.
105 27. A. Lennartson and M. Hakansson, Angew. Chem.ꢀInt. Ed., 2009, 48,
5869ꢁ5871.
28. A. Lennartson, M. Vestergren and M. Hakansson, Chem.ꢀEur. J.,
2005, 11, 1757ꢁ1762.
29. M. Hakansson, M. Vestergren, B. Gustafsson and G. Hilmersson,
Angew. Chem.ꢀInt. Ed., 1999, 38, 2199ꢁ2201.
30. A. Lennartson and M. Hakansson, CrystEngComm, 2009, 11, 1979ꢁ
1986.
31. C. P. Brock, W. B. Schweizer and J. D. Dunitz, J. Am. Chem. Soc.,
1991, 113, 9811ꢁ9820.
Notes and references
aDepartment of Chemistry, University College of Science, University of
Calcutta, 92 A. P. C. Road, Kolkata, 700009, India. Eꢀmail:
ghosh_59@yahoo.com
110
40 bDepartment of Chemistry, Universitat de les Illes Balears, Crta. de
Valldemossa km 7.5, 07122Palma de Mallorca (Baleares), Spain
c Institute of Chemical Research of Catalonia (ICIQ), Avgda. Països
Catalans 16, 43007 Tarragona, Spain
115 32. E. D'Oria, P. G. Karamertzanis and S. L. Price, Cryst. Growth Des.,
2010, 10, 1749ꢁ1756.
dCatalan Institution for Research and Advanced Studies (ICREA), Passeig
45 Lluìs Companys, 23, 08010 Barcelona, Spain
33. J. H. Luo, M. C. Hong, Q. Shi, Y. C. Liang, Y. J. Zhao, R. H. Wang,
R. Cao and J. B. Weng, Transit. Met. Chem., 2002, 27, 311ꢁ315.
34. G. M. Sheldrick, Acta Crystallogr. Sect. A, 2008, 64, 112ꢁ122.
120 35. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A.
Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F.
Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y.
e School of Chemistry, The University of Reading, P. O. BOX 224,
Whiteknights, Reading RG 66AD, UK
fDepartament de Química Inorgànica, Universitat de Barcelona, Martí i
Franquès 1ꢀ11, 08028 Barcelona, Spain. Email: patrick.gamez@qi.ub.es
50
† Electronic Supplementary Information (ESI) available: crystallographic
and refinement data for the the
representation of the molecular structure of the ꢃ enantiomer of 1 (Figure
S1); solution UVꢁvis and CD spectra of (Figures S2 and S4,
55 respectively); solidꢁstate CD spectrum of the ꢃ enantiomer of 1 (Figure
S3); solution CD spectrum of 2 (Figure S5); solution and solidꢁstate UVꢁ
vis spectra of 2 (Figure S6). CCDC 869389 (1-Λ), 869390 (1-ꢀ), 869391
ꢃ enantiomer of compound 1;
125
Honda, O. Kitao, H. Nakai, T. Vreven, J. Montgomery, J. A., J. E.
Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N.
Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K.
Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M.
Cossi, N. Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross, V.
Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O.
1
130
This journal is © The Royal Society of Chemistry [year]
CrystEngComm, 2012, 14, 00–00 |7

CrystEngComm
Page 8 of 8
View Online
Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L.
Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J.
J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B.
Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian 09,
Revision A.1, (2009) Gaussian, Inc., Wallingford CT.
5
36. C. W. Bauschlicher, A. Ricca, H. Partridge and S. R. Langhoff,
Chemistry by density functional theory. In Recent Advances in
Density Functional Methods, Part II, World Scientific, Singapore,
1997.
10 37. J. M. L. Martin, in Density Functional Theory: A Bridge between
Chemistry and Physics, eds. P. Geerlings, F. De Proft and W.
Langenaeker, VUB Press, Brussels, Belgium, 2000.
38. S. Grimme, J. Comput. Chem., 2006, 27, 1787ꢁ1799.
39. F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005, 7,
15
3297ꢁ3305.
40. A. Frontera, D. Quinonero, C. Garau, A. Costa, P. Ballester and P. M.
Deya, J. Phys. Chem. A, 2006, 110, 5144ꢁ5148.
41. A. W. Addison, T. N. Rao, J. Reedijk, J. van Rijn and G. C.
Verschoor, J. Chem. Soc.ꢀDalton Trans., 1984, 1349ꢁ1356.
20 42. A. Mangia, M. Nardelli, C. Pelizzi and G. Pelizzi, J. Chem. Soc.ꢀ
Dalton Trans., 1972, 2483ꢁ2487.
43. M. A. S. Goher and F. A. Mautner, Polyhedron, 1994, 13, 2149ꢁ
2155.
44. W. Han, L. Li, Z. Q. Liu, S. P. Yan, D. Z. Liao, Z. H. Jiang and P. W.
25
Shen, Z. Anorg. Allg. Chem., 2004, 630, 591ꢁ596.
45. L. Ding, L. R. Lin, C. Y. Liu, H. K. Li, A. J. Qin, Y. Liu, L. Song, H.
Zhang, B. Z. Tang and Y. F. Zhao, New J. Chem., 2011, 35, 1781ꢁ
1786.
30
Graphical Abstract
Crystallization of the squareꢁpyramidal copper complex where the R
group is a hydrogen atom yields a mechanical mixture of homochiral
35 crystals while that whose R group is a methyl generates racemic crystals
8
|CrystEngComm, 2012, 14, 00–00
This journal is © The Royal Society of Chemistry [year]