Copper(II) complexes with aromatic o-phosphorylated phenols - Synthesis, crystal structures, and X-ray photoelectron spectroscopy

Article abstract of DOI:10.1002/ejic.201300540

The structural properties of copper complexes with (2-hydroxyphenyl)di-p- tolylphosphane oxide (CuL¹₂) and 3-(2-hydroxyphenyl)-3,4- dihydro-2H-benzo[f][1,5,3]dioxaphosphepine 3-oxide (CuL²₂) are studied in detail. The o-phosphorylated phenols demonstrate both bridging and chelating behaviour, and the main reaction product is a coordination polymer containing dimeric Cu₂O₈ fragments. The structural peculiarities of the o-phosphorylated phenols were revealed by single-crystal X-ray diffraction. The copper complexes are studied by single-crystal and powder X-ray diffraction, and IR, Raman and X-ray photoelectron spectroscopy (XPS). The applicability of XPS spectroscopy to structural studies of coordination compounds with complex organic ligands is demonstrated with CuL¹ ₂ and CuL²₂. The results of ab initio structure solution and Rietveld refinement of the CuL²₂ structure from powder data are augmented by PW-PBE-D calculations. The formation of copper complexes with aromatic o-phosphorylated phenolates with both chelating and bridging configurations was demonstrated by single-crystal and powder X-ray diffraction. Photoelectron spectroscopy was utilized to reveal the coordination polyhedra around the copper ions in an amorphous copper complex. Copyright

Full text of DOI:10.1002/ejic.201300540

Job/Unit: I30540
/KAP1
Date: 09-07-13 16:36:07
Pages: 10
FULL PAPER
Copper Complexes
S. Shuvaev,* I. S. Bushmarinov, I. Sinev,
A. O. Dmitrienko, K. A. Lyssenko,
V. Baulin, W. Grünert, A. Y. Tsivadze,
N. Kuzmina ..................................... 1–10
The formation of copper complexes with
aromatic o-phosphorylated phenolates with
both chelating and bridging configurations
was demonstrated by single-crystal and
powder X-ray diffraction. Photoelectron
spectroscopy was utilized to reveal the co-
ordination polyhedra around the copper
ions in an amorphous copper complex.
Copper(II) Complexes with Aromatic o-
Phosphorylated Phenols – Synthesis, Crys-
tal Structures, and X-ray Photoelectron
Spectroscopy
Keywords: O ligands / Copper / Structure
elucidation / Photoelectron spectroscopy /
Bridging ligands
0000, 0–0
1
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I30540
/KAP1
Date: 09-07-13 16:36:07
Pages: 10
www.eurjic.org
FULL PAPER
DOI:10.1002/ejic.201300540
Copper(II) Complexes with Aromatic o-Phosphorylated
Phenols – Synthesis, Crystal Structures, and X-ray
Photoelectron Spectroscopy
Sergey Shuvaev,*[a] Ivan S. Bushmarinov,[b] Ilya Sinev,[c]
Artem O. Dmitrienko,[b] Konstantin A. Lyssenko,[b]
Vladimir Baulin,[d,e] Wolfgang Grünert,[c] Aslan Yu. Tsivadze,[e] and
Natalia Kuzmina[a]
Keywords: O ligands / Copper / Structure elucidation / Photoelectron spectroscopy / Bridging ligands
The structural properties of copper complexes with (2-hy-
droxyphenyl)di-p-tolylphosphane oxide (CuL12) and 3-(2-hy-
studied by single-crystal and powder X-ray diffraction, and
IR, Raman and X-ray photoelectron spectroscopy (XPS). The
droxyphenyl)-3,4-dihydro-2H-benzo[f][1,5,3]dioxaphosphep- applicability of XPS spectroscopy to structural studies of co-
ine 3-oxide (CuL22) are studied in detail. The o-phosphory- ordination compounds with complex organic ligands is dem-
lated phenols demonstrate both bridging and chelating be-
haviour, and the main reaction product is a coordination
polymer containing dimeric Cu2O8 fragments. The structural
peculiarities of the o-phosphorylated phenols were revealed
by single-crystal X-ray diffraction. The copper complexes are
onstrated with CuL1 and CuL22. The results of ab initio
2
structure solution and Rietveld refinement of the CuL22 struc-
ture from powder data are augmented by PW-PBE-D calcula-
tions.
Introduction
Aromatic ortho-phosphorylated phenols have attracted
considerable attention as potential O,O-donor ligands with
a broad set of possible substituents owing to their wide
range of possible applications. For example, they have
shown encouraging preliminary results as effective antenna
in luminescent lanthanide complexes.[1] From the structural
point of view, they can be considered as a distant relative
of β-diketonates (Figure 1). It stands to reason that any
level of conjugation is not expected for such a system, and
this analogy is valid only as an illustrative example of struc-
tural conformity between both classes.
Figure 1. Structural conformity of the chelating part of enol β-
diketonates (left) and ortho-phosphorylated phenols (right).
The structural mode of ortho-phosphorylated phenols in
coordination compounds is still uncertain, and existing data
is limited only to the synthesis of compounds without any
structural data.[2–4] The main question is whether or not
they can play the role of chelating or bridging ligands (Fig-
ure 2).
The ligand configuration is definitely influenced con-
siderably by the chemical nature of the substituents (R1 and
R2).
For this reason, we chose two ligands, (2-hydroxyphen-
yl)di-p-tolylphosphane oxide (HL1) and 3-(2-hydroxy-
phenyl)-3,4-dihydro-2H-benzo[f][1,5,3]dioxaphosphepine 3-
oxide (HL2), with substituents with substantially different
volumes (Figure 3).
As a starting point, copper(II) complexes were chosen to
reveal the structural potential of the chosen ligands. Cop-
per(II) forms the most stable complexes among 3d metals
[a] Department of Materials Science, Lomonosov Moscow State
University,
119991 Moscow, Leninskie gory 1/3, Russia
Fax: +7-4959390998
E-mail: sergeyshuvaev@gmail.com
Homepage: http://www.inorg.chem.msu.ru/coord/
[b] A.N. Nesmeyanov Institute of Organoelement Compounds,
Russian Academy of Sciences,
(INEOS RAS), GSP-1, 28 Vavilov Street, 119991 Moscow,
Russia
[c] Laboratory of Industrial Chemistry, Ruhr-University Bochum,
Universitätsstraße 150 44801 Bochum, Germany
[d] Institute of Physiologically Active Compounds, Russian
Academy of Sciences,
142432 Chernogolovka, Severny proezd 1, Moscow Region,
Russia
[e] Frumkin Institute of Physical Chemistry and Electrochemistry,
Russian Academy of Sciences,
Leninsky Pr. 31, Moscow 119071, Russia
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201300540.
Eur. J. Inorg. Chem. 0000, 0–0
2
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I30540
/KAP1
Date: 09-07-13 16:36:07
Pages: 10
www.eurjic.org
FULL PAPER
(Figure S4) is disordered and contains orientations of the
2-hydroxyphenyl group with O1–P1–C1–C2 torsion angles
of 4.11(12) and 167.56(11)° in a ratio 0.54:0.46. In the for-
mer orientation, the OH group forms an intramolecular hy-
drogen bond with the P=O group [O···O 2.5407(15) Å, O–
H···O 160.3(2)°], and in the latter it interacts with the P=O
group of a neighbouring molecule [O···O 2.7182(13) Å, O–
H···O 168.35(11)°]. Notably, in most β-diketone crystal
structures, for example, that of dibenzoylmethane, only a
strong intramolecular hydrogen bond is observed in the
enol form,[7,8] which leads to an effective O–C···C–O tor-
sion angle of 0. The ortho-phosphorylated phenols can be
expected to be much more flexible ligands, as the corre-
sponding average O–C···P=O torsion angles are 55.5° in
HL1 and either 2.2(2) or 162.7(2)° in HL2.
Figure 2. Two possible configurations of o-phosphorylated phenol-
ate ligands. Left: chelating conformation. Right: bridging confor-
mation.
Synthesis and Identification of Sodium Salts NaL1 and
NaL2
o-Phosphorylated phenols HL1 and HL2 can be easily
transformed into sodium salts by a simple reaction between
HL and sodium hydroxide in methanol and isolated as solid
products. The formation of NaL1 and NaL2 was confirmed
by elemental analysis and by comparison of their spectral
characteristics, including excitation and luminescence spec-
troscopy in both solution and powder forms, with those of
Figure 3. Aromatic o-phosphorylated phenols HL1 (left) and HL2
(right).
and demonstrates a clear tendency for square-planar coor-
dination, such as in copper(II) β-diketonates.[5] Therefore,
we can expect a similar coordination environment for aro-
matic o-phosphorylated phenols.
1
HL1 and HL2. The H NMR spectra demonstrate the dis-
appearance of the “acidic” hydrogen atom of the hydroxy
group, and the 31P NMR spectra show a tangible upfield
shift of the singlet resonances. At the same time, infrared
and Raman spectroscopy revealed several splittings and
shifts of the bands of the bonds mainly affected during de-
protonation and introduction of sodium ions (P=O and C–
O bonds). Both HL and NaL demonstrate luminescence in
the near-UV region. From HL to NaL, a typical bathoc-
romic shift was revealed (Figure S5), which indicates a de-
crease of the energy gap between the highest occupied mo-
lecular orbital (HOMO) and the lowest unoccupied molec-
ular orbital (LUMO) by formation of sodium salts. Such
a shift often accompanies a charge rearrangement in the
molecular structure and is often observed during a change
of bond type (e.g., if a predominantly covalent bond is sub-
stituted by an ionic bond as in saltlike structures).
Thus, the main goal of present paper is the synthesis of
copper complexes (CuL12 and CuL2 ) and the evaluation of
2
the copper polyhedra by both X-ray diffraction and X-ray
photoelectron spectroscopy. To the best of our knowledge,
the latter approach has been infrequently used for the
analysis of coordination compounds over recent decades.
Results and Discussion
Crystal Structures of o-Phosphorylated Phenols HL1 and
HL2
To date, very little data is available on the crystal struc-
tures of aromatic o-phosphorylated phenols. Previously, the
crystal structure of HL2 was described, but that data needed
additional verification owing to a relatively high R factor.[6]
The mutual orientation of the P=O and C–O bonds (the
O–C/P–O torsion angle) attracts particular interest as the
flexibility of the O=P–C=C torsion angle directly affects
the range of possible structures for coordination com-
pounds containing these ligands.
Synthesis of Copper Complexes CuL1 , CuL1 ·2MeOH and
2
2
CuL2
2
The copper complexes CuL1 and CuL2 were obtained
2
2
by in situ reaction between freshly prepared solutions of
The hydrogen-bond systems in the crystal structures of the sodium salts in methanol and methanol solutions of
HL1 and HL2 are substantially different (Table S1). The Cu(NO3)2·2H2O. After slow evaporation of the methanol,
unit cell of HL1 contains two symmetrically independent greenish powders formed: a dark green amorphous precipi-
molecules (HL1* and HL1Ј, Figure S3), which differ only tate of CuL1 , and a light green polycrystalline precipitate
2
slightly in the mutual orientation of the toluene substituent of CuL2 . To obtain CuL1 in crystalline form, an alterna-
2
2
and the O1–P1–C1–C2 torsion angles [54.0(2) and tive synthetic approach without the intermediate sodium
–59.3(3)°, Figure S1]. Both molecules form intermolecular salt was performed. The reaction between copper(II) acet-
O–H···O=P hydrogen bonds. The crystal structure of HL2 ate and HL1 in m-xylene at reflux followed by the removal
Eur. J. Inorg. Chem. 0000, 0–0
3
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I30540
/KAP1
Date: 09-07-13 16:36:07
Pages: 10
www.eurjic.org
FULL PAPER
of acetic acid by azeotropic distillation resulted in the for- comprise two distorted CuO5 square pyramids joined by a
mation of the same amorphous product CuL1 and several base edge. One ligand with an O–C···P=O torsion angle of
2
single crystals of CuL1 ·2MeOH as a byproduct isolated 0.1(2)° provides the O2 atoms that bridge two CuO5 pyra-
2
from the mother liquid. The structure of CuL1 ·2MeOH mids, and the O1 atoms are the apexes of the pyramids.
2
was determined by single-crystal XRD (see Experimental The other ligand forms a bridge between the dimeric Cu2O8
Section), but the amount of it was too small to allow study fragments with an O–C···P=O torsion angle of –101.1(2)°.
by other methods. The CuL2 specimen contained only one Such coordination is unknown for β-diketonates but is fully
2
phase according to powder XRD, and its structure was de- in line with the flexibility demonstrated by the studied li-
termined from the powder data (see below). The complexes gands in the uncoordinated form.
CuL1 and CuL2 were also identified by IR, Raman, dif-
As the geometry data obtained from the powder diffrac-
fuse-reflectance and X-ray photoelectron spectroscopy tion data often has low precision, we performed a PW-PBE-
2
2
(XPS).
D DFT calculation for comparison. The deviation of the
calculated result from the Rietveld refined structure was
only 0.21 Å. The differences in the Cu–O bond lengths were
more pronounced, as the Cu position was refined without
any restraints, and are summarized in Table 1. Nevertheless,
the geometry remained qualitatively the same, and the Cu–
Cu distance allows spin coupling in line with the experimen-
tal EPR findings.
Structure of CuL1 and CuL2 Complexes
2
2
The structure of CuL1 ·2MeOH contains a Cu atom with
2
square-planar coordination. The Cu atom lies on an inver-
sion centre; thus, the CuO4 polyhedron has zero deviation
from planarity. The Cu1–O1 and Cu2–O2 bond lengths are
1.9420(18) and 1.8951(19) Å, respectively (Figure 4). Such
a coordination was an expected result for a β-diketonate
analogue, but both the O 1s XPS and electron paramagnetic
Table 1. The main bond lengths [Å] in the Rietveld refined and
calculated structures of CuL2 .
2
Rietveld refined
PW-PBE-D
resonance (EPR) data for bulk CuL1 disagree with this
2
Cu1–O2
1.89(3)
1.88(2)
1.96(2)
2.20(3)
1.97(3)
3.05(2)
1.988
1.917
1.973
2.371
2.000
3.172
structure as it should be paramagnetic and contain two
types of oxygen atoms in a 1:2 ratio (see below). The O–
C···P=O torsion angle in this compound is 37.9(2)°, which
shows that the analogy with β-diketonates is only superflu-
ous. The bulk CuL12 was amorphous according to the pow-
der XRD data, and its crystal structure could not be deter-
Cu1–O2A
Cu1–O1A[a]
Cu1–O1[b]
Cu1–O2[b]
Cu1–Cu1[b]
[a] 3 – x, –y, –z. [b] 2 – x, –y, –z.
mined directly. However, CuL2 was crystalline and its
2
structure was determined by ab initio methods from the
powder diffraction data (see below). The structure was
Determination of the Structure of CuL2 from Powder XRD
Data
found to be a coordination polymer formed by CuL2
2
2
chains along the a direction. The Cu atoms form dimeric
Cu2O8 coordination polyhedra (on an inversion centre) that
The powder diffraction pattern (see Experimental Sec-
tion) of CuL22 was indexed by TOPAS 4.2 software[9] in the
¯
triclinic crystal system with the P1 space group, as judged
from the cell volume (expected ZЈ = 1). The lattice param-
eters (after Rietveld refinement) are a = 8.9008(3) Å, b =
10.6655(3) Å, c = 16.4108(8) Å, α = 103.660(4)°, β =
95.975(3)°, γ = 113.527(2)°, V = 1353.17(9) Å3. The struc-
ture solution was performed by using the parallel-tempering
method as implemented in FOX.[10] The model for the solu-
tion and refinement was prepared based on a PBE/L2[11]
calculation of an isolated monomeric CuL22 complex (anal-
ogous to the CuL12 complex from CuL1 ·2MeOH) by using
2
PRIRODA.[12] The search with this model was unsuccess-
ful, so two Cu–O bond constraints corresponding to
Cu···O=P coordination, as well as C–O···Cu angle con-
straints, were removed, and the search was repeated. This
search provided a reasonable position for one of the ligands
and the Cu atom (one run out of 40), so that the Cu–O–
Cu bridge and part of square Cu polyhedron could be iden-
tified. The final solution was obtained by constraining the
Cu atom coordinates and the orientation and position of
the ligand to the found ones, and searching for the position,
conformation and orientation of the other ligand freely. The
Figure 4. Overview of the CuL12 crystal structure. Atoms are repre-
sented by thermal displacement ellipsoids (p = 50%), superscripts
indicate symmetry-generated atoms and correspond to the follow-
ing symmetry transformations. 1: –x, –y, –z. Only one disordered
component of the methanol moiety is labelled.
Eur. J. Inorg. Chem. 0000, 0–0
4
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I30540
/KAP1
Date: 09-07-13 16:36:07
Pages: 10
www.eurjic.org
FULL PAPER
resulting structure contained Cu atoms in 4+1 coordination
polyhedra and was Rietveld refined in TOPAS 4.2 by using
bond lengths and angles from the above model as restraints
for the ligands. The Cu–O bond lengths and O–Cu–O
angles were not restrained. Isotropic thermal parameters
were refined independently for each atom type within each
ligand, and the line-profile anisotropy was refined by using
second order spherical harmonics.[13]
The “Morse” restraint model[14] was applied during the
refinement (see Experimental Section). The analysis of the
deviations of the refined bond lengths from the defined val-
ues within this restraint model has been shown to incor-
rectly identify the Rietveld refined structures. After 150 re-
finements in TOPAS with decreasing penalty function
weight (K1, see Experimental Section), there were no outli-
ers in the bond-length-deviation (Δd) distribution (Fig-
ure 5), which further supports the accuracy of the refined
structure (Figure 6). At K1 = 8, the difference between the
Figure 6. Overview of the CuL22 crystal structure. Atoms are repre-
sented by spheres, superscripts indicate symmetry-generated atoms
and correspond to the following symmetry transformations. 1: 3 –
x, –y, –z; 2: 2 – x, –y, –z; 3: –1 + x, y, z.
Table 2. Refinement indicators for Pawley fit and Rietveld refine-
ment of CuL2 .
2
Pawley fit
Rietveld refinement
K1
–
–
8
RMS Δd [Å]
Rwp[%]
0.029
2.62
13.97
1.91
16.07
0.79
2.17
2.37
11.94
1.66
12.49
0.07
2.28
RwpЈ [%]
Rp [%]
RpЈ [%]
Rbragg [%]
χ2
Figure 5. Multiple boxplot for CuL2 Δd distributions at different
2
values of K1.
Figure 7. Calculated, experimental and difference powder patterns for CuL2 .
2
Eur. J. Inorg. Chem. 0000, 0–0
5
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I30540
/KAP1
Date: 09-07-13 16:36:07
Pages: 10
www.eurjic.org
FULL PAPER
calculated and experimental patterns was featureless (Fig- disappears in the spectrum of CuL22 and provides clear evi-
ure 7). A comparison between the final refinement indi- dence of successful deprotonation.
cators and those for a Pawley fit is summarized in Table 2.
A Raman spectrum was recorded only for CuL2 (Fig-
2
ure 8), as CuL1 melted under laser irradiation. Neverthe-
2
less, several maxima were determined and their values are
listed below in the Experimental Section. The band corre-
sponding to P=O vibration (ca. 1127 cm–1) was substan-
tially broadened and shifted towards higher energies (ca.
1159 cm–1). The same shift is observed for the band as-
cribed to C–O bond vibration (ca. 1322 vs. ca. 1326 cm–1).
In addition, complexation affected the vibrational modes
within the aromatic rings (several bands at ca. 1588 cm–1).
Thermal Stability
To evaluate the thermal behaviour of the synthesized
complexes, thermogravimetric/differential scanning calo-
rimetry (TG-DSC) curves were recorded (Figure S6). Both
complexes are thermally stable to ca. 190 °C, and their
broad exothermic bands corresponding to the decomposi-
tion of complexes on the DSC curve are of similar shape.
Infrared and Raman Spectroscopy
Initial elucidation and assignment of the vibrational
bands of the ligands were performed from theoretically cal-
culated infrared and Raman spectra of HL1 and HL2. A
preliminary comparison of the IR spectra of HL and CuL2
(Figure S7) revealed considerable changes in the position
and shape of the most important bands, ascribed to ν(P=O)
and ν(C–O). For the P=O vibration modes, the most in-
tense band centred at ca. 1120 cm–1 is observed in both
complexes and contains at least two sub-bands (resolved for
CuL2 and unresolved for CuL1 ). In addition to the
Figure 8. Raman spectra of HL2 and CuL2 .
2
2
2
ν(P=O) stretching mode, these bands comprise in-plane
bending β(CCH) and stretching ν(C=C) of the aryl rings,
as well as out-of-plane φ(HCH) twisting of the methylene
bridges for HL2 and CuL2 . In both cases (CuL1 and
Diffuse-Reflectance Spectroscopy of CuL1 and CuL2
2
2
2
2
CuL2 ), the barycentre of these broad bands is slightly
As CuL1 is amorphous, absorption spectroscopy could
2
2
shifted towards lower frequencies, which is typical for coor-
dinated P=O bonds.
be a useful tool in the determination of the copper polyhe-
dra by comparison with the spectrum of CuL2 , for which
2
For the bands corresponding to the C–O vibration mode,
the best solution is to compare the obtained data with pre-
vious IR spectra of ortho-substituted phenols.[15] Here, we
can single out two groups of bands corresponding to the
C–O vibration modes. The first group centred at ca.
1245 cm–1 is unfavourable for analysis because of the over-
lap of the vibrations of several bonds. The second group
centred at ca. 1305 cm–1 is considerably clearer for further
analysis and, therefore, was chosen for comparison. In both
cases, a broad band centred at ca. 1304 cm–1 comprising at
least two sub-bands is observed, in contrast with the nar-
rower bands revealed in the HL spectra. From theoretical
calculations, for HL2 and CuL22 these bands are apparently
the result of an overlap between νsymm(C–C) of the catechol
substituent with the in-plane bending β(OCH) and out-of-
plane φ(HCH) wagging vibrations of the methylene bridges
at lower frequencies, and ν(C–O) at higher frequencies, dis-
played as a shoulder. The lack of methylene bridges in HL1
led to a narrower band comprising only νsymm(C–C) and
ν(C–O) vibrations. The nature of the additional band ob-
served at higher frequencies for the copper complexes can
be attributed to the two types of C–O bonds in the coordi-
nated ligands (1.305 vs. 1.288 Å).
the copper polyhedra are known. However, because of poor
solubility, absorption spectroscopy in solution was substi-
tuted by diffuse-reflectance spectroscopy (Figure 9). The
electronic absorption spectra contain two asymmetrical
bands (one of them is unresolved for CuL2 ) below
2
22000 cm–1 [attributed to d–d transitions of the copper(II)
ions], which indicates the presence of more than one elec-
tronic transition with similar energy values (Figure 9).[16]
The close proximity of the evaluated peak values and their
A relatively narrow band centred at 1420 cm–1 in the
Figure 9. Diffuse-reflectance absorption spectra of CuL1 and
2
spectrum of HL2, ascribed to in-plane β(COH) bending, CuL2 at ambient temperature.
2
Eur. J. Inorg. Chem. 0000, 0–0
6
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I30540
/KAP1
Date: 09-07-13 16:36:07
Pages: 10
www.eurjic.org
FULL PAPER
shapes could indicate similar coordination polyhedra [viz.
pentacoordinate copper(II) ions] and close ligand-field
strength (LFS) values, in line with a minor impact of the
substituents on the charge density localized on the oxygen
atoms participating in complexation.[17] Broad bands at
higher energies are ascribed to charge-transfer states and
intraligand transitions.
X-ray Photoelectron Spectroscopy
To obtain a deeper insight into the coordination polyhe-
dra around the copper(II) ions in amorphous CuL1 , XPS
2
measurements were required. First, the XPS spectra of
CuL22 (with determined crystal structure) were recorded for
use as a starting point for analysis of the spectrum of
Figure 10. (a) Cu 2p photoelectron and (b) Cu LMM spectra.
The O 1s photoelectron spectra of both samples have
major contributions at ca. 531.0 eV binding energy (Fig-
ure 11). These signals originate from the oxygen atoms that
form the polyhedra around the copper ions. The spectrum
of CuL12 also shows a minor contribution at 532.4 eV bind-
ing energy, which most likely belongs to out-of-plane oxy-
gen atoms in polyhedra around the copper ions (with a ra-
tio between both contributions of 75:25). The O 1s photo-
amorphous CuL1 (with unknown structure of the copper
2
polyhedra).
The C 1s photoelectron spectra of both CuL1 and
2
CuL2 are presented on Figure S6. Both spectra demon-
2
strate a major contribution at lower binding energies, which
most likely belong to sp2-hybridized aromatic carbon atoms
and were referenced to 284.5 eV. The high-binding-energy
components can be assigned to the sp3-hybridized carbon
atoms of substituent radicals [two methyl substituents for
(L1)– ligand and two methylene substituents for (L2)–] with
1:11.5 and 1:3 atomic ratios to the main aromatic fragments
in the (L1)– and (L2)– samples, respectively. The latter ratio
considerably contradicts the theoretical value (1:6) and
could be explained by the formation of extra carbon in the
X-ray beam from pump oil and other organic contami-
nation on the surface of the samples.
electron line of CuL2 has a well-defined doublet structure,
2
but it can be described by three contributions with a ratio
of 50:37.5:12.5 positioned at 533.3, 531.0 and 531.8 eV
binding energy, respectively, which is in a good agreement
with the ratio between the in-plane and out-of-plane oxygen
atoms in the polyhedra (1:3) of the obtained crystal struc-
ture. The contribution at 533.3 eV corresponds to the oxy-
gen atoms of the dioxaphosphepine ring, and has a BE
value that corresponds well to the literature data for C–O–
Ph oxygen atoms in polymers (533.0–533.5 eV).[22] The line
at 531.8 eV most likely corresponds to the O1 oxygen atom
at the apex of the CuO5 pyramid. As its bonding with Cu
Figure S9 shows the experimental P 2p photoelectron
lines deconvoluted into a doublet with area ratio con-
strained according to p sublevel multiplicity (1:2). The P
2p 3/2 photoelectron lines are at slightly different binding
energies, depending on the ligand nature (distinctions in
P=O bond lengths). However, their energies (132.0 and
132.3 eV for CuL12 and CuL2 , respectively) are close to the
2
energies reported for phosphorus in different organic ligand
surroundings.[18,19]
The Cu 2p photoelectron and Cu LMM X-ray Auger
electron spectra (Figure 10, a and b, respectively) indicate
that copper is present mostly in the 2+ oxidation state.
There is also asymmetry in the Cu 2p 3/2 line for CuL1
2
and an easily distinguishable shoulder on the high binding
energy (BE) side of the CuL2 , which might indicate the
2
presence of some other copper states. Unfortunately, the
copper XPS signal was not intensive enough even after sig-
nificant signal accumulation, so reliable signal deconvol-
ution cannot be performed. The maxima of the Cu 2p 3/2
envelope position (934.0 and 933.7 eV for CuL1 and
2
CuL2 , respectively) and modified Auger[20] parameters val-
2
ues (1850.4 and 1850.7 eV) comply with those of copper 2+
reference compounds such as CuO.[21]
Figure 11. O 1s photoelectron spectra.
Eur. J. Inorg. Chem. 0000, 0–0
7
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I30540
/KAP1
Date: 09-07-13 16:36:07
Pages: 10
www.eurjic.org
FULL PAPER
is relatively weak because of the significant Cu–O distance compounds. The crystal structures of the aromatic ortho-
of 2.371 Å, its BE should be close to that of the oxygen phosphorylated phenols HL1 and HL2 were determined,
atom in triphenylphosphine oxide (530.9–531.1 eV).[23]
and the influence of substituents on packing motifs has
In the XPS study, the most unexpected result obtained is been analyzed. Main distinctions were found in the mutual
the observation of high BE shoulders for the Cu 2p XPS orientation of the P=O and C–O bonds and, as a result, the
lines. The binding energy for those contributions is rather system of hydrogen bonds. In the crystal structures of the
high, ca. 937.0 eV; such values have only been reported for copper complexes (CuL1 ·2MeOH and CuL2 ), the mutual
2
2
copper fluoride to date. One possible explanation for the orientation of P=O and C–O bonds led to substantial dis-
extra signals would be a differential charging of the speci- tinctions in the type of polyhedra around the copper com-
men surface. This effect is often observed when the sample plex. Such a result clearly demonstrates that this type of
has different parts with the same elements with different prospective ligands can appear in both chelating and bridg-
charges. The corresponding XPS lines have different shifts ing modifications, depending on the chemical nature of sub-
as a result. However, if that were the case, the shake-up stituents. One of the most significant results of this paper
satellites observed next to each of the Cu 2p lines would is the possibility to distinguish between different types of
have the same extra contributions.
coordinated oxygen atoms by X-ray photoelectron spec-
Additionally, it should be noted that the O 1s signal de- troscopy.
convolution performed for CuL2 is not the only one pos-
2
sible, it is even not the simplest. The line envelope can be
Experimental Section
fitted with two Gaussian–Lorentz product components
with the same goodness of fit. The present fit is a result of
constraining the low BE component to 533.3 eV and using
the expected intensity ratios. The same procedure can be
performed by using the 531.0 eV component as the main
one, which results in a shift of the minor component to
higher BEs. A three-component model without constraints
gives atomic ratios significantly different to the predicted
ones, and the results strongly depend on the lineshape used
(Gaussian and Lorentzian functions ratio). Therefore, the
fact that the experimental signal shape can be reproduced
with three components in the expected intensity ratio and
plausible line shape parameters definitely lends experimen-
tal support to this ratio.
Materials: Analytically pure methanol and ethanol (Sigma–Ald-
rich), Cu(NO3)2·2H2O and Cu(CH3COO)2·2H2O (Riedel-de Haën)
were used without further purification.
Measurements: The C, H and N contents were determined by con-
ventional elemental analysis. IR spectra were recorded in the region
4000–400 cm–1 with a Perkin–Elmer Spectrum ONE FTIR spec-
trometer with samples in Nujol. Raman spectra were recorded in
the region 400–2000 cm–1 by using a Renishaw InVia spectrometer.
Diffuse-reflectance spectra were recorded in the region 250–800 nm
with a Perkin–Elmer Lambda 650 spectrometer. Luminescence and
excitation luminescence spectra were recorded in the region 200–
450 nm with a Perkin–Elmer LS-55 spectrometer. 1H and 31P NMR
spectra were obtained with a Bruker Avance-400 {400 MHz, [D6]-
DMSO/tetramethylsilane (TMS)} spectrometer. TG-DSC curves
were obtained with a NETZSCH STA 409 PC/PG analyzer with
argon flow at a heating rate of 10 °C/min in the temperature range
40–1000 °C.
Electron Paramagnetic Resonance
Single-Crystal X-ray Diffraction (SC-XRD): X-ray quality single-
crystals of HL1 and HL2 were grown by recrystallization from
methanol solutions. The single-crystal X-ray diffraction data were
collected with a Bruker SMART APEX II CCD diffractometer
(Mo-Kα, λ = 0.71073 Å, graphite monochromator) at 100 K. The
structures were solved by direct methods and refined with the full-
matrix F2 least-squares technique, as implemented in SHELX.[25]
The hydrogen atoms of OH groups were found in difference Fourier
syntheses. Other H atoms were placed in idealized positions and
refined by using a riding model with fixed isotropic thermal param-
eters. The crystal data for the studied compounds is provided be-
low.
To get additional data concerning the polyhedron around
the copper ion in amorphous CuL1 , a comparison of the
2
EPR spectra of both complexes (CuL12 and CuL2 ) is essen-
2
tial. For CuL2 , the close proximity between the copper ions
2
within the dimeric fragment [3.05(2) Å] can give rise to cou-
pling between the paramagnetic copper ions and overall
diamagnetic behaviour. Actually, in both cases a signal was
not detected (Figure S9), except for a slight contamination
by a paramagnetic impurity in CuL1 , which was ascribed
2
to a square-planar copper complex.[24] Therefore, for
CuL1 , both types of coordination environment (square-
2
HL1: C20H19O2P (M = 322.32): monoclinic, space group P21/c (no.
14), a = 19.090(3) Å, b = 9.6266(14) Å, c = 18.199(3) Å, β =
90.332(3)°, V = 3344.4(8) Å3, Z = 8, T = 100.0(2) K, μ(Mo-Kα) =
0.171 mm–1, Dcalcd. = 1.280 g/mm3, 22663 reflections measured
(2.24 Յ 2θ Յ 56), 7983 unique (Rint = 0.0472) reflections were used
in all calculations. The final R1 was 0.0550 [I Ͼ 2σ(I)] and wR2 was
0.1525 (all data).
planar and pentacoordinate) are present, but the pentaco-
ordinate complex substantially prevails. The obtained re-
sults are entirely in line with our previous assumption about
the proximity between both polyhedra (CuL1 and CuL2 ),
and square-planar coordinated complexes were obtained
only as a minor admixture.
2
2
HL2: C14H13O4P (M = 276.21): monoclinic, space group P21/n (no.
14), a = 9.7074(6) Å, b = 12.3078(7) Å, c = 10.7244(6) Å, β =
105.2600(10)°, V = 1236.14(12) Å3, Z = 4, T = 100(2) K, μ(Mo-
Kα) = 0.229 mm–1, Dcalcd. = 1.484 g/mm3, 12876 reflections mea-
Conclusions
The present paper aims to awake interest in the role of
phosphorylated phenols as anionic ligands in coordination sured (5.04 Յ 2θ Յ 60), 3589 unique (Rint = 0.0512) reflections
Eur. J. Inorg. Chem. 0000, 0–0
8
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I30540
/KAP1
Date: 09-07-13 16:36:07
Pages: 10
www.eurjic.org
FULL PAPER
were used in all calculations. The final R1 was 0.0363 [I Ͼ 2σ(I)]
and wR2 was 0.1021 (all data).
Theoretical Calculation of Infrared and Raman Spectra: The struc-
tures of initial molecules HL1 and HL2 were optimized at the
B3LYP/6-311*G+ level[32,33] by using the Gaussian 09 package. The
infrared and Raman intensities were calculated for optimized geo-
metries in the framework of the same DFT method. All theoretic-
ally calculated frequencies were be real, which indicates the true
minimum of the calculated total energy. To provide an easier com-
parison with experimental spectra, a scaling factor of 0.975 was
introduced. Vibrational assignment of fundamental modes was per-
formed on the ground of the calculated vibrational mode anima-
tion with the ChemCraft program.
CuL1 ·2MeOH: C42H44CuO6P2 (M = 770.25): triclinic, space group
2
¯
P1 (no. 2), a = 9.325(2) Å, b = 10.057(2) Å, c = 10.099(2) Å, α =
99.686(5)°, β = 93.142(5)°, γ = 103.912(5)°, V = 901.7(4) Å3, Z =
1, T = 100(2) K, μ(Mo-Kα) = 0.744 mm–1, Dcalcd. = 1.418 g/mm3,
10799 reflections measured (4.12 Յ 2θ Յ 58), 4780 unique (Rint
=
0.0526) reflections were used in all calculations. The final R1 was
0.0482 [I Ͼ 2σ(I)] and wR2 was 0.1199 (all data).
X-ray Powder Diffraction (XRD): The powder pattern of CuL22 was
measured with a Bruker D8 Advance Vario diffractometer with
a LynxEye detector and Ge (111) monochromator, λ(Cu-Kα1) =
1.54060 Å, θ/2θ scan from 4 to 90°, stepsize 0.0104788°. The mea-
Electron Paramagnetic Resonance (EPR): The measurements were
performed with a Varian E-4 X-band (ν ≈ 9.1 GHz) EPR spectrom-
eter. The temperature measurements were performed with the use
of a platinum resistor transducer (50 Ω at 0 °C) with an accuracy
of Ϯ0.5 K.
surement was performed in transmission mode with CuL2 de-
2
posited on kapton film.
The penalty function in the “Morse” restrained refinement is de-
fined as follows:
Syntheses: Aromatic o-phosphorylated phenols (HL1 and HL2)
were synthesized according to the synthetic route described else-
where.[5] The purity of the synthesized compounds was proved by
1
C, H, N analysis and H and 31P NMR spectroscopy.
in which P is penalty function, K1 is a global penalty function
weighting, κi is the weighting of the individual bond penalty, ai is
a coefficient corresponding to the bond force constant, Di is the
defined length of a given bond and di is its refined length at current
minimization step.
A mixture of HL and NaOH (1:1) was prepared in methanol solu-
tion with stirring and heating at 60 °C for 1 h. Afterwards, a white
precipitate was isolated in good yield (85%) after slow evaporation
of the solvent over several days at room temperature. The isolated
1
product was identified by C,H,N analysis, H and 31P NMR spec-
Periodic Density Functional Calculation of CuL2 : Periodic DFT
2
troscopy and IR spectroscopy.
calculations of crystal structures CuL2 were performed by using
2
NaL1: IR: ν = 1116 (s), 1148 (m, P=O), 1163 (m), 1243 (w), 1267
the VASP 5.212 code.[26–29] The conjugated-gradient technique was
used for optimizations of the atomic positions and minimization of
the total energy. Experimental atomic coordinates (with K1 = 8)
˜
(m, C–O), 1440 (s, P–C) cm–1. 1H NMR: δ = 2.40 (s, 6 H, Me),
6.84 (2ϫd, 2 H, OPh), 7.23 (t, 1 H, OPh), 7.27 (2ϫd, 4 H, Ph),
7.36 (t, 1 H, OPh), 7.50 (q, 4 H, Ph) ppm. 31P NMR: δ = 27.91 (s)
ppm. C20H18NaO2P (344.32): calcd. C 69.77, H 5.23; found C
69.58, H 5.37.
and the cell parameters of CuL2 were taken as the starting point.
2
The projected augmented wave (PAW) method was applied to ac-
count for core electrons, and valence electrons were approximated
by plane-wave expansion with a 400 eV cutoff. Exchange and corre-
lation terms of total energy were described by a PBE exchange-
correlation functional.[30] Kohn–Sham equations were integrated
by using a Γ-point approximation with additional dispersion cor-
rection.[31] At the final step of our calculations, the atomic displace-
ments converged better than 0.03 eVÅ–1, and the energy variations
were less than 10–4 eV. The atomic coordinates for the calculations
can be found in the Supporting Information.
NaL2: IR: ν = 1126 (m), 1168 (m, P=O), 1250 (m, C–O), 1429 (m,
˜
P–C), 1445 (s), 1453 (s), 1489 (s) cm–1. 1H NMR: δ = 4.70–4.75
(2ϫd, JH,H = 14.65, 14.78 Hz, JP,H = 7.07, 7.20 Hz, 2 H, CH2),
4.97–5.01 (d, JH,H = 14.66 Hz, JP,H = 1.14 Hz, 2 H, CH2), 6.84
(2ϫd, 2 H, OPh), 7.23 (t, 1 H, OPh), 7.27 (2ϫd, 4 H, Ph), 7.42–
7.46 (t, JH,H = 7.96, 7.57 Hz, 1 H, OPh), 7.66–7.71 (t, JH,H = 9.17,
9.58 Hz, 1 H, OPh) ppm. 31P NMR: δ = 35.75 (s) ppm.
C14H12NaO4P (298.21): calcd. C 56.39, H 4.06; found C 56.24, H
4.15.
CCDC-935792 (for HL1), -935793 (for HL2), -935794 (for
CuL1 ·2MeOH) and -935795 (for CuL2 ) contain the supplemen-
2
2
Cu(NO3)2·2H2O (0.31 mmol) was added to a mixture of HL and
NaOH in ethanol (molar ratio 1:2:2), and the mixture was stirred
for 1 h at ambient pressure. After evaporation of the solvent in air,
green precipitates of different tints were isolated, rinsed with dis-
tilled water and dried in air over one day, yield ca. 80%. Notably,
the hydrate content of the isolated precipitates (amorphous for
tary crystallographic data for this paper. These data can be ob-
tained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
X-ray Photoelectron Spectroscopy (XPS): Measurements were per-
formed in an ultra-high vacuum (UHV) set-up equipped with a
monochromatic Al-Ka X-ray source (hν = 1486.6 eV), operated at
14.5 kV and 35 mA, and a high-resolution Gammadata-Scienta
SES 2002 analyzer. The base pressure in the measurement chamber
was maintained at ca. 7ϫ 10–10 bar. The measurements were per-
formed in the fixed transmission mode with a pass energy of 200 eV
resulting in an overall energy resolution of 0.25 eV. A flood gun
was applied to compensate the charging effects. High-resolution
spectra for C 1s, O 1s, Cu 2p and P 2p photoelectron lines along
with Cu LMM Auger transitions were recorded. The binding-en-
ergy scales were corrected to the charge shift by referencing the
most intensive sp2-hybridized C 1s contribution to 284.5 eV. The
Casa XPS software with a Gaussian–Lorentzian product function
and Shirley background subtraction was used for peak deconvol-
ution.
CuL12 and polycrystalline for CuL2 ) is not the same for both com-
2
plexes. Both CuL1 and CuL2 were unsolvated without further
2
2
drying.
Cu(CH3COO)2·2H2O (0.31 mmol) was dissolved in methanol and
added to HL1 in m-xylene (molar ration 1:2). The mixture was
heated to reflux for 4 h, and the acetic acid was subsequently re-
moved by azeotropic distillation. After evaporation of the solvent
in air, a dark green amorphous precipitate was isolated, rinsed with
distilled water and dried in air for one day, yield ca. 80%.
CuL1 : C40H38CuO4P2 (708.23): calcd. C 68.03, H 5.14; found C
2
67.87, H 5.19. IR: ν = 1118 (vs, P=O), 1161 (w), 1248 (m), 1266
˜
(m, C–O), 1437 (vs, P–C) cm–1. Raman: ν = 1030, 1128, 1334, 1446,
˜
1461, 1589 cm–1.
Eur. J. Inorg. Chem. 0000, 0–0
9
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I30540
/KAP1
Date: 09-07-13 16:36:07
Pages: 10
www.eurjic.org
FULL PAPER
CuL2 : C28H24CuO8P2 (613.99): calcd. C 54.77, H 3.94; found C
[9] Bruker, TOPAS, v. 4.2, User Manual, Bruker AXS GmbH,
Karlsruhe, Germany, 2009.
2
55.01, H 3.83. IR: ν = 1118 (s, P=O), 1136 (s), 1167 (m), 1238 (m),
˜
ˇ
[10] V. Favre-Nicolin, R. Cerný, J. Appl. Crystallogr. 2002, 35, 734–
1251 (s, C–O), 1435 (s, P–C) cm–1. Raman: ν = 1029, 1311, 1333,
˜
743.
1440, 1472, 1588 cm–1.
[11] D. N. Laikov, Chem. Phys. Lett. 2005, 416, 116–120.
[12] D. N. Laikov, Chem. Phys. Lett. 1997, 281, 151–156.
[13] M. Järvinen, J. Appl. Crystallogr. 1993, 26, 525–531.
[14] I. S. Bushmarinov, A. O. Dmitrienko, A. A. Korlyukov, M. Yu.
Antipin, J. Appl. Crystallogr. 2012, 45, 1187–1197.
[15] Z. Rappoport, in: The chemistry of phenols, Wiley-VCH,
Weinheim, Germany, 2003, p. 333–395.
[16] N. A. Rey, A. Neves, A. J. Bortoluzzi, W. Haase, Z. Tomkow-
icz, Dalton Trans. 2012, 41, 7196–7200.
[17] Z. D. Matovic´, B. Ristic´, M. Joksovic´, S. R. Trifunovic´, G. Pe-
losi, S. Ianelli, G. Ponticelli, Transition Met. Chem. 2000, 25,
720–726.
Supporting Information (see footnote on the first page of this arti-
cle): molecular and crystal structures of HL1 and HL2; hydrogen-
bond parameters of HL1 and HL2; luminescence and luminescence
excitation spectra of HL1, HL2, NaL1 and NaL2; IR spectra of
HL1, HL2, CuL12 and CuL2 ; C 1s and P 2p XPS spectra of CuL1
2
2
and CuL2 ; TG-DSC curves of CuL1 and CuL2 ; EPR spectra of
2
2
2
CuL1 and CuL2 .
2
2
Acknowledgments
[18] W. E. Morgan, W. J. Stec, R. G. Albridge, J. R. van Wazer, In-
org. Chem. 1971, 10, 926–930.
[19] M. Pelavin, D. N. Hendrickson, J. M. Hollander, W. L. Jolly, J.
Phys. Chem. 1970, 74, 1116–1121.
The authors thank Dr. Vladimir Dolzhenko of the Inorganic
Chemistry Department (MSU), Dr. Dmitry Tsymbarenko of the
Material Sciences Department (MSU) and Mr. Dmitry Gil of the
Kurnakov Institute of general and inorganic chemistry of the Rus-
sian Academy of Sciences. This study was partially supported by
Russian Foundation for Basic Research (grant numbers 12-03-
31560, 12-03-33107 and 12-03-00878-a).
[20] C. D. Wagner, L. H. Gale, R. H. Raymond, Anal. Chem. 1979,
51, 466–482.
[21] C. D. Wagner W. M. Riggs L. E. Davis J. F. Moulder, G. E.
Muilenberg, in: Handbook of X-ray Photoelectron Spectroscopy,
Perkin–Elmer Corporation, Physical Electronics Division,
Eden Prairie, 1979.
[22] D. Briggs, G. Beamson, Anal. Chem. 1993, 65, 1517–1523.
[23] W. E. Morgan, W. Stec, R. G. Albridge, J. R. Van Wazer, Inorg.
Chem. 1971, 10, 926–930.
[24] J. R. Pilbrow, Transition Ion Electron Paramagnetic Resonance,
Calendron Press, Oxford, 1990.
[1] S. Shuvaev, O. Kotova, V. Utochnikova, A. Vaschenko, L.
Puntus, V. Baulin, N. Kuzmina, A. Tzivadze, Inorg. Chem.
Commun. 2012, 20, 73–76.
[2] K. A. Petrov, V. M. Chizhov, V. P. Pokatun, S. V. Agafonov, [25] G. M. Sheldrick, SHELXL97, Program package for crystal
Russ. Chem. Rev. 1986, 55, 1042–1053.
structure determination, University of Göttingen, Germany,
1998.
[3] V. I. Evreinov, V. E. Baulin, Z. N. Vostroknutova, Z. V. Saf-
ronova, I. B. Krashakova, Bull. Acad. Sci. USSR Div. Chem.
Sci. (Engl. Transl.) 1992, 40, 1759–1766.
[26] G. Kresse, J. Furthmüller, Comput. Mater. Sci. 1996, 6, 15–50.
[27] G. Kresse, J. Furthmüller, Phys. Rev. B 1996, 54, 11169–11186.
[28] G. Kresse, J. Hafner, Phys. Rev. B 1993, 47, 558–561.
[29] G. Kresse, J. Hafner, Phys. Rev. B 1994, 49, 14251–14269.
[30] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996,
77, 3865–3868.
[31] S. Grimme, J. Comput. Chem. 2006, 27, 1787–1799.
[32] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.
[33] C. Lee, W. Yang, R. G. Parr, Phys. Rev. 1988, 37, 785–789.
Received: April 24, 2013
[4] G. I. Bondarenko, J. Struct. Chem. 1996, 37, 206–209.
[5] R. C. Mehrotra, R. Bohra, D. P. Gaur, in: Metal β-diketonates
and allied derivatives, Academic Press, London, 1978.
[6] E. N. Zvetkov, V. H. Sundukova, V. E. Baulin, Bull. Acad. Sci.
USSR Div. Chem. Sci. (Engl. Transl.) 1989, 38, 135–137.
˙
[7] S. Öztürk, M. Akkurt, S. Ide, Cryst. Mater. 1997, 212, 808–
810.
[8] N. V. Belova, H. Oberhammer, G. V. Girichev, Struct. Chem.
2011, 22, 269–277.
Published Online: ᭿
Eur. J. Inorg. Chem. 0000, 0–0
10
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim