Gas-phase structure and conformations of copper(II) 2,9,16,23-tetra-tert-butyl phthalocyanine

Article abstract of DOI:10.1007/s11224-015-0619-3

The molecular structure of copper(II) 2,9,16,23-tetra-tert-butyl phthalocyanine (Cu(pc ^t )) was investigated by gas-phase electron diffraction with mass spectrometric control of vapor composition. Two conformers of C _4h symmetry and three conformers of C _s symmetry are predicted by quantum chemical calculations using the hybrid DFT method UB3LYP with 6-31G and cc-pVTZ basis sets. According to the relative energies at the temperature of GED experiments (436°C), the two C _4h symmetric conformers occur in 1.1 and 2.3 % abundance in the gas phase. The highest mole fraction (66.6 %) and lowest R-factor (4.25 %) correspond to one of the C _s symmetric conformers; however, the two C _4h symmetric models cannot be disproved by the Hamilton R-factor ratio test. The GED structural data of the three models mentioned reliably confirm approximate local D _4h symmetry of the phthalocyanine ligand core.

Full text of DOI:10.1007/s11224-015-0619-3

Struct Chem
DOI 10.1007/s11224-015-0619-3
ORIGINAL RESEARCH
Gas-phase structure and conformations of copper(II) 2,9,16,23-
tetra-tert-butyl phthalocyanine
1
2
3
4
•
Oleg A. Pimenov
Norbert W. Mitzel
•
Nina I. Giricheva
•
•
Sebastian Blomeyer
1
•
Vladimir E. Mayzlish
3
Georgiy V. Girichev
Received: 16 June 2015 / Accepted: 19 June 2015
Ó Springer Science+Business Media New York 2015
Abstract The molecular structure of copper(II) 2,9,16,-
2
R-factor (4.25 %) correspond to one of the C symmetric
s
t
3-tetra-tert-butyl phthalocyanine (Cu(pc )) was investi-
conformers; however, the two C4h symmetric models
cannot be disproved by the Hamilton R-factor ratio test.
The GED structural data of the three models mentioned
reliably confirm approximate local D4h symmetry of the
phthalocyanine ligand core.
gated by gas-phase electron diffraction with mass spec-
trometric control of vapor composition. Two conformers of
C4h symmetry and three conformers of C symmetry are
s
predicted by quantum chemical calculations using the
hybrid DFT method UB3LYP with 6-31G* and cc-pVTZ
basis sets. According to the relative energies at the tem-
perature of GED experiments (436 °C), the two C4h sym-
metric conformers occur in 1.1 and 2.3 % abundance in the
gas phase. The highest mole fraction (66.6 %) and lowest
Keywords Copper(II) Á 2,9,16,23-tetra-tert-butyl
phthalocyanine Á Molecular structure Á Gas-phase electron
diffraction Á Mass spectrum
Introduction
Dedicated to Prof. Magdolna Hargittai on the Occasion of her 70th
birthday.
The metal phthalocyanines (M(pc)) are unique coordina-
tion compounds. Numerous publications are dedicated to
the understanding of the important properties of M(pc)
such as high thermal stability [1, 2], nonlinear optical
activity [3, 4], catalytic [5] and semiconductor capability
Electronic supplementary material The online version of this
article (doi:10.1007/s11224-015-0619-3) contains supplementary
material, which is available to authorized users.
&
Oleg A. Pimenov
oleg.pimenov1988@mail.ru
[
6]. A particular focus has been placed on water-soluble
phthalocyanine derivatives. They are known to be capable
of accumulating selectively in tumors and are therefore
used in medicine for photo-diagnostics and phototherapy of
cancerous diseases [7].
Nina I. Giricheva
g.v.girichev@mail.ru
Sebastian Blomeyer
sebastian.blomeyer@uni-bielefeld.de
Insertion of peripheral and non-peripheral substituent
groups (e.g., methyl, ethyl and tert-butyl) at the 16 reactive
sites in the benzene rings of the pc macrocycle [8, 9] can
have a pronounced influence on the properties, e.g.,
resulting in an increase in solubility and volatility by using
substituents that constrict intermolecular attraction. This
conclusion has been clearly supported from work [10], that
contains results of mass spectrometric (MS) investigations
of the vapor composition of tetra-tert-butyl substituted
1
Department of Physics, Ivanovo State University of
Chemistry and Technology, Engels Av. 7, 15300 Ivanovo,
Russia
2
3
Department of Organic and Physical Chemistry, Ivanovo
State University, Ermak St. 39, 153025 Ivanovo, Russia
Faculty of Chemistry and Center for Molecular Materials
CM2, Bielefeld University, Universit a¨ tsstr. 25,
3
3615 Bielefeld, Germany
4
Research Institute of Macroheterocycles, Ivanovo State
University of Chemistry and Technology, Engels Av. 7,
t
phthalocyanines M(pc ) with M = Cu(II), VO are pre-
1
5300 Ivanovo, Russia
sented. Investigations of the dependence of vapor pressure
123

Struct Chem
on temperature have revealed that tetra-tert-butyl substi-
tuted phthalocyanines have higher volatilities than the
unsubstituted ones. The authors have explained this fact
with the presence of bulky tert-butyl groups preventing
coplanar p–p interaction between phthalocyanine mole-
cules in the solid state and thus leading to weaker inter-
molecular interactions.
Taking into account the circumstances described above,
we attempted to carry out experimental (GED/MS) and
theoretical (DFT) structural investigation of copper(II)
t
2,9,16,23-tetra-tert-butyl-29H,31H-phthalocyanine (Cu(pc ))
in order to compare the results with those for the unsubsti-
tuted copper phthalocyanine (Cu(pc)).
The influence of peripheral substitutions on the electronic
structure of Cu(II) phthalocyanine (Cu(pc)), copper(II)
t
,9,16,23-tetra-tert-butyl-29H,31H-phthalocyanine (Cu(pc ))
Experimental part
Synthesis
2
and copper(II) 1,2,3,4,8,9,10,11,15,16,17,18,22,23,24,25-
F
hexadecafluoro-29H,31H-phthalocyanine (Cu(pc) ) in vari-
ousmatriceshasbeeninvestigatedusingcombinedcontinuous
wave and pulse electron paramagnetic resonance (EPR) of
solutions of these complexes in toluene and sulfuric acid as
well as density functional theory (DFT) [9]. According to the
EPRdata,thein-planemetal–ligandr-bondingandtheout-of-
Thoroughly triturated mix 0.920 g (5.0 mmol) of 4-tert-
butyl phthalonitrile (Aldrich 42322-1G) and 0.300 g
(1.5 mmol) of copper acetate monohydrate is kept for 4 h
at 175–180 °C. Resulting copper(II) 2,9,16,23-tetra-tert-
butyl phthalocyanine was extracted with chloroform and
twice chromatographed on aluminum oxide. The solvent
was removed under vacuum, and the resulting compound
was dried under vacuum at 80° C. Yield: 0.480 g (48 %);
Found, %: C 71.85; N 14.04; H 6.33. C H CuN . Cal-
t
plane p-bonding are more covalent for Cu(pc ) in toluene than
in sulfuric acid solution. Note that the covalent nature of this
bonding is increased upon substitution with tert-butyl groups.
This means that the introduction of peripheral groups can lead
to changes in the covalent/ionic ratio of the metal–ligand bond
in phthalocyanines [9]. The change in electron density distri-
bution in chemical bond involves variations in bond lengths,
and consequently the introduction of peripheral groups has an
influence on the geometry of the coordination cavity. This
conclusion seems reasonable for solutions, but to the best of
our knowledge there is so far no experimental evidence of
structural changes in phthalocyanine coordination cavities
affected by peripheral substitutions for the gaseous phase.
On the other hand, the effects of alkyl substitutions
4
8
48
8
culated, %: C 72.02; N 14.00; H 6.04. ES: kmax (lg e), (nm)
in heptane: 671 (4.44); 640 (4.88); 606 (4.83).
GED/MS experiment
t
In work [10], it was shown that the sample Cu(pc ) sublimes
in the form of monomers and without thermal decomposi-
tion up to 600 °C. The mass spectrum consists of the
molecular ion Cu(pc ) ([M] : m/z = 800) and doubly
t 2?
t ?
?
charged Cu(pc )
.
(
methyl) on the structure of metal porphyrins Mp (M = Cu,
However, a carried out pilot mass spectrometry experi-
ment indicates slow change in the mass spectrum during a
sample heating. The time dependence of the mass spectra
monitoring the vapor composition over the heated sample
is presented in Fig. 1. The most significant change with
proceeding time observes for the ratio between the two
peaks at 799 and 742 Da, corresponding to the molecular
Sn; p = C H N ) based on DFT investigations are pre-
20 12 4
sented in the works [11, 12]. The authors have performed
calculations of unsubstituted (Cup and Snp) and methyl-
substituted porphyrin molecules (Cu(omp) and Sn(omp);
omp = C H N ) for comparison of structural parameters.
2
8 28 4
The substitution of the pyrrole hydrogen atoms by donor-
type methyl groups has a weak influence on the coordination
cavity geometry of the porphyrins. The most significant
changes are taking place in the carbon–carbon bonds of the
pyrrole ligand fragment. Furthermore, a comparison of the
experimental porphyrin structures Cu(omp) and Sn(omp)
?
?
ion [M] and the one with stoichiometry [M–C H ] .
4
9
?
Because for the first 15 min the molecular ion [M] is
dominant in the vapor over solid sample heated to 442 °C
(see Fig. 1), and it is enough to perform GED experiments.
The electron diffraction patterns and the mass spectra
were recorded simultaneously using a combined GED/MS
unit based on EMR-100 and APDM-1 devices [13–15]. A
sample was evaporated from a molybdenum effusion cell at
436 ± 8 °C, as measured with a W/Re-5/20 thermocouple.
The electron wavelength was determined from the diffrac-
tion patterns of ZnO recorded before and after the GED/MS
experiments. The conditions of the GED/MS experiment are
given in Table 1. The diffraction patterns were recorded
(
gas-phase electron diffraction data; GED) with molecular
structures of octaethylporphyrins Cu(oep) and Sn(oep) in
their crystals has been reported [11, 12]. It should be noted
that despite systematic differences between X-ray and
electron diffraction geometries caused by different princi-
ples of scattering and the presence of eight ethyl substituents
in crystalline Sn(oep) and Cu(oep), the geometry of the
coordination cavity in Sn(oep) and Cu(oep) remains practi-
cally the same as in the free Cu(omp) and Sn(omp)
molecules.
from two nozzle-to-plate distances (L = 598 mm and
1
L2 = 338 mm) on Kodak Electron Image films SO-163. The
1
23

Struct Chem
Fig. 1 Time dependence of the
mass spectrum monitoring the
vapor composition over the
heated copper tetra-tert-butyl
Tave =442 °C
+
[
M-C H ]
4
9
t
phthalocyanine (Cu(pc ))
+
[
M]
8
0
2
+
time, min
5
7
[M]
4
6
2
1
6
0
m/z 400
600
800
1000
Table 1 Conditions of the synchronous gas-phase electron diffrac-
tion experiment and mass spectrometric experiment
Table 2 Relative abundances of ions in mass spectrum of vapor over
solid sample
Camera distance (L) (mm)
338
598
Ion
m/z
I
rel. (%)
Uioniz = 50 eV) (Uioniz = 50 eV)
(
Electron beam current (lA)
Ionization voltage (V)
1.74
1.69
T = 444 °C
L = 338 mm
T = 428 °C
L = 598 mm
50(1)
50(1)
Temperature of effusion cell (°C)
˚
Wavelength of electrons (A)
444(5)
0.04381(4)
100
428(5)
0.04443(2)
40
?
4 9
[C H ]
57
143
384
399
649
664
679
694
709
724
742
754
769
784
14
7
15
3
?
[
3 2
1/4TTBPc–(CH ) C]
Exposition time (s)
2
?
-6
-6
[
M–CH
3
]
13
13
11
15
22
26
30
38
54
45
42
60
12
12
4
Residual gas pressure (mmHg)
1
5.4 9 10
6.8 9 10
2?
[M]
-
˚
s
min–smax/Ds (A
)
3.2–27.0/0.1
3
1.2–15.2/0.1
5
?
[
[
[
[
[
[
[
[
[
[
[
3
M–10CH ]
Number of plates
?
?
?
?
?
M–9CH
M–8CH
M–7CH
M–6CH
M–5CH
3
]
]
]
]
]
5
3
3
3
3
8
rectangular area of 10 mm 9 130 mm was scanned at a
diagonal direction (1299 point in line, 33 equidistant lines
with a step of 0.1 mm along each line) by a modified [16]
Carl Zeiss Jena MD-100 microdensitometer.
12
16
21
31
30
31
44
100.0
?
4 9
M–C H ]
Simultaneously with performance of the electron
diffraction experiments, the mass spectra of the vapor over
solid sample were recorded. The relative intensities of ion
currents are listed in Table 2, and visualization of these
data is presented in Fig. 2.
?
M–3CH
3
]
?
M–2CH
3
]
?
3
M–CH ]
?
M]
799 100.0
There are processes of dissociative ionization with
progressive segregation of CH3 groups under electron
impact presented in the vapor. Additionally, the presence
t
the presence of only Cu(pc ) in the vapor during the GED/
MS experiment.
2
?
of the doubly charged molecular ion [M] and the ion
corresponding to the quarter of macrocyclic ligand
Quantum chemical calculations
Computational methods

TTBPc (without tert-butyl fragment) in the vapor can be
observed, the latter being typical for metal phthalocyanines
1, 17–19]. There is no significant change in the ratio
[
between the peaks in the mass spectra during the GED/MS
experiment, as long as the general time of exposure does
not exceed 11 min. These circumstances allow assuming
t
DFT geometry optimizations of Cu(pc ) were carried out
using the Becke’s three-parameter hybrid functional
123

Struct Chem
Irel,
%
+
M]
[
100
9
8
7
6
5
4
3
2
1
0
+
[
M-CH ]
3
0
0
0
0
0
0
0
0
0
+
[
M-2CH ]
3
+
[
M-3CH ]
3
+
[
M-C H ]
4 9
+
[
M-5CH ]
3
+
[M-6CH ]
3
+
[
M-7CH ]
3
+
[
M-8CH ]
3
+
+
[
1/4TTBPc-(CH ) C]
3
2
[M-9CH3]
+
2+
2+
[
C H ]
[M-CH3]
[M]
4
9
+
[
M-10CH3]
1
00
200
300
400
500 600
700
800
m/z
Fig. 2 Average relative abundances of the vapor over solid sample
GED/MS experiment)
(
B3LYP [20–22] and the one-parameter hybrid functional of
Perdew, Burke and Ernzerhof PBE [23], using both unre-
stricted (UB3LYP) and restricted open-shell (ROB3LYP)
treatment of odd number of electrons. Both DFT func-
tionals were combined with 6-31G* [24–26] and cc-pVTZ
(
N,C,H) [27] basis sets. For the latter, the core electrons of
2
2
6
the copper atom (1s 2s 2p ) were described by a rela-
tivistic effective core potential (ECP) of the Stuttgart group
[
28]. For the description of the copper valence shells
8s7p6d2f1g/6s5p3d2f1g), the basis sets described in the
(
literature [28, 29] have been used. We will refer to this
combination of ECP and basis set as ECP10MWB. All
calculations were performed using the Gaussian 03 W
program package [30]. Analysis of the natural bonding
t
orbitals (NBO) for Cu(pc ) and Cu(pc) was done on
B3LYP/ECP10MWB (unrestricted and restricted open-
shell approaches) level of theory using the NBO 3.1 pro-
gram [31] implemented in the GAUSSIAN 03 W program
package. The visualization of molecular models and elec-
tronic orbitals is performed by CHEMCRAFT [32] program.
Electronic structure analysis in terms of Bader’s quantum
theory atoms in molecules (QTAIM) [33] was performed
using the AIMALL program [34].
Fig. 3 Models of copper tetra-tert-butyl phthalocyanine: conformer 1
and 2 (C4h symmetry)
According to frequency analysis, both conformers corre-
spond to minimum on potential energy surface (PES). The
theoretical structural parameters of the two conformers of
t
The theoretical models for the Cu(pc ) molecule
t
According to above-mentioned work [9], the EPR results
t
for Cu(pc) and Cu(pc ) indicate D core symmetry in the
Cu(pc ) (see Fig. 3) are presented in Table S1. According
4
h
to DFT calculations, the energy gap between both con-
-1
formers is more than 1 kcal mol , and conformer 2 is
energetically preferred.
2
molecules in their doublet B electronic ground states.
1
g
t
The molecular structure of Cu(pc ) was optimized under
2
C4h symmetry for the B electronic state. According to
g
For investigating the conformational properties of
t
Cu(pc ), the energy barrier to internal rotation of one tert-
preliminary UB3LYP/6-31G* calculations, there are two
t
Cu(pc ) conformers of C symmetry shown in Fig. 3. The
4
h
butyl group around the bond C4–C10 is determined by
performing a relaxed energy surface scan with a step size of
numbering of atoms is presented in Fig. 5 as well.
1
23

Struct Chem
ΔE, kcal/mole
calculated. It should be noted that the transition from
conformer 2 to conformer 2–1 increases the total electronic
energy DE in contrast to decreasing DG.
This fact can be explained by the entropy contribution
being a lot higher for the asymmetrical conformers. Con-
former 2–1 was predicted to occur in the highest mole
fraction (v = 66.6 %) in the vapor composition, and
0
0
0
0
0
0
0
0
0
.8
.7
.6
.5
.4
.3
.2
.1
.0
1
2
therefore, this model of C symmetry should be taken into
s
account in the experimental data refinement procedure. The
conformer 2–1 of C symmetry and numbering of atoms is
s
presented in Fig. 5.
The quadratic force fields of conformers 1 and 2 of
t
Cu(pc ) (UB3LYP/ECP10MWB) were used to calculate the
-0.1
-20
0
20
40
60
80
100
120
τ (C12-C10-C4-C5), deg.
set of vibrational corrections Dr = rh1 - r and the root
a
mean square amplitudes at the temperature of the electron
Fig. 4 Potential function (UB3LYP/6-31G*) of internal rotation of
tert-butyl group around bond C4–C10 with a step along the torsion
angle 10°. Conformer 1 corresponds to a torsion angle of 60°, whereas
conformer 2 exhibits a torsion angle of 0°
diffraction experiment in second approximation using the
program SHRINK [35–37]. For conformer 2–1,
a
UB3LYP/6-31G* force field was used, but calculation of
this was based on initial structural parameters from
UB3LYP/ECP10MWB calculation of conformer 2.
1
0° along the torsion angle s(C12–C10–C4–C5) (UB3LYP/
-31G*, Fig. 4). The barrier of internal rotation for one tert-
6
butyl group could be determined as V = 0.79 kcal/mol.
Since the thermal energy for the temperature of the exper-
Structural analysis
-1
iment (T = 709 K) is kT = 1.48 kcal mol , the internal
t
rotation of tert-butyl groups is not negligible in Cu(pc ).
Analysis of the electron diffraction data was performed,
assuming that the vapor consists only of molecules having
C4h symmetry (1 and 2) or Cs symmetry (2–1). The
molecular models included 15 independent parameters: (1)
seven bond lengths, namely r(Cu–N1), r(N1–C1), r(C1–N2),
r(C7–C8), r(C2–C7), r(C4–C10) and r(C3–H); (2) seven
valence angles, videlicet \(N1–Cu–X), \(Cu–N1–C1),
\(N1–C8–C7), \(C2–C7–C6), \(C12–C10–C4), \(C2–
C3–H) and \(C10–C12–H); and (3) single torsion angle
Based on the obtained potential function DE(s) and
taking into account the independent rotation of the tert-
t
butyl substituent molecule Cu(pc ), we found five different
stable conformers given in Table 3. The denomination
‘‘conformer 2–1’’ means that one tert-butyl group is rotated
by 180° starting from conformer 2 and so on. The con-
formers are arranged in ascending order of relative elec-
tronic energy DE. For each conformer, frequencies and
amplitudes of vibration at T = 709 K, as well as the Gibbs
energy DG, mole fractions v (according to Boltzmann’s
s (C12–C10–C4–C5) that determines the rotation of the tert-
1
butyl group relative to the plane of the phthalocyanine
distribution), entropy S and molar heat capacity C , were
V
macrocycle; during the LS analysis, this parameter was fixed
t
Table 3 Relative full energies DE, relative Gibbs energies DG and mole fractions v of Cu(pc ) conformers obtained by DFT calculations
(
UB3LYP/6-31G*) according to GED experiment temperature T = 709 K
Conformer 2
Conformer 2–1
Conformer 2–2
Conformer 2–3
Conformer 1
-
1)
DE (kcal mol
0.0
2.8
2.3
0.29
0.0
0.51
1.6
0.78
2.9
1.1
3.8
1.1
-
1
DG
T
(kcal mol
)
v (%)
66.6
21.7
8.3
123

Struct Chem
L = 598 mm
sM(s)
L = 338 mm
Δ
sM(s)
(
a)
b)
(
(c)
0
2
4
6
8
10 12 14 16 18 20 22 24 26 28
-1
s, Å
t
Fig. 6 Molecular intensity functions sM(s) of conformer 2–1 Cu(pc )
dots experimental; solid line model) and difference functions
DsM(s) for molecular models: (a) conformer 2–1; (b) conformer 2;
c) conformer 1
(
(
Fig. 5 Models of copper tetra-tert-butyl phthalocyanine (C
try) and atomic numbering conformer 2–1
s
symme-
at 0°. The \(N1–Cu–X) angle was included in the set of
independent parameters to describe the displacement of the
Cu atom out-of-plane of the molecule (X is a dummy atom
positioned at the center of the coordination cavity plane that
was fixed at 90°).
f(r)
Δ
f(r)
The differences in similar C–C bond lengths and also in
(
a)
(b)
c)
\
(C–C–C), \(C–C–H) angles were fixed to those obtained
(
from the quantum chemical calculations UB3LYP/
ECP10MWB. Starting values of independent parameters
and also of the root mean square vibrational amplitudes
were taken from the above-mentioned DFT calculations.
The amplitudes were refined in groups corresponding to the
peaks on the radial distribution curve. All amplitudes for
0
2
4
6
8
10
12
14
16
18
20
22
r, Å
t
Fig. 7 Radial distribution function f(r) of conformer 2–1 Cu(pc )
(dots experimental; solid line model) and difference functions
Df(r) for molecular models: (a) conformer 2–1; (b) conformer 2;
(
c) conformer 1
˚
distances greater than 15 A were kept fixed to initial
values.
Figures 6 and 7 display the plots of molecular intensity
sM(s) and radial distribution f(r) experimental functions
along with their model analogues and difference curves
DsM(s) and Df(r) for all three conformers. The independent
geometric parameters for all of the three above-mentioned
molecular models are listed in Table 5. The independent
geometric parameters, vibrational amplitudes and vibra-
tional corrections Dr for the all three molecular models are
presented in Tables S2, S3 and S4. Tables S5, S6 and S7
give the matrixes of parameter correlation. The Cartesian
Discussion
Molecular structure by DFT calculations
The structural parameters of conformers 1 and 2 yielded by
three theoretical approximations UB3LYP/6-31G*,
UB3LYP/ECP10MWB and UPBE/ECP10MWB differ
only for the benzene ring C2–C3, C3–C4 and C4–C5 dis-
tances (see Table S1). The largest differences in these
˚
distances amount to 0.005 A, regardless of the theory level
coordinates of the atoms for the least squares refinement r
h1
t
structures of three conformers Cu(pc ) are also available in
of calculation. The same is observed to the angle \(C5–
C4–C10) corresponding to the position of the tert-butyl
group in the plane of the macrocycle. By rotation of the
Supplementary Material.
1
23

Struct Chem
tert-butyl groups from conformer 1 to 2, this angle
increases by 3°, regardless of the theoretical approach. All
other structural parameters, in particular the Cu–N1 and
CuÁÁÁN2 distances, remain almost unchanged for both
conformers. Thus, having analyzed the structures of con-
formers 1 and 2 by DFT calculations, we conclude that
internal rotation of the tert-butyl groups has no appreciable
effect on the structure of the coordination complex cavity,
but a small one on the geometry of the benzene rings of the
phthalocyanine ligand.
Table 4 Independent structure parameters (r /\ ) of conformers 1,
t
h1 h1
2
and 2–1 of Cu(pc ) obtained by gas electron diffraction
a,b
Parameters
1
2
2–1
r
h1
r(Cu–N1)
r(N1–C1)
r(C1–N2)
r(C7–C8)
r(C2–C7)
r(C4–C10)
r(C3–H)
1.966(5)
1.371(3)
1.325(4)
1.454(5)
1.408(3)
1.542(4)
1.086(5)
1.964(5)
1.376(3)
1.317(4)
1.452(4)
1.405(3)
1.540(4)
1.084(4)
1.964(5)
1.378(3)
1.318(4)
1.448(5)
1.405(3)
1.542(4)
1.084(4)
c
c
c
c
\
(Cu–N1–C1)
\(N1–C8–C7)
(C2–C7–C6)
\(C2–C3–H)
(C10–C12–H)
125.5
125.6
125.6
The molecular structure determined by GED
109.9(8)
120.8(2)
123.1(62)
112.8(79)
114.4(9)
0.0
109.7(11)
120.7(2)
124.3(63)
111.1(100)
112.6(16)
180.0
109.9(5)
\
120.6(2)
During the LS optimization, the models of conformers 1, 2
and 2–1 resulted in almost the same agreement of the
experimental and theoretical functions sM(s): R = 4.46 %
122.9(60)
112.2(73)
112.0(13)
(1)0.0/(3)180.0
4.25
\
f
\(C12–C10–C4)
s1(C12–C10–C4–C5)
c
for 1, R = 4.34 % for 2 and R = 4.25 % for 2–1. As
f
f
shown in Table 5, the independent parameters for both C4h
d
Rf (%)
4.46
4.34
and C symmetry models are consistent with each other
s
a
˚
The values of bond lengths (A) and angles (°)
In parentheses, the values of total errors: for bond lengths
within the error limits. Moreover, according to the statis-
tical Hamilton criterion [38] at a significance level of 0.05,
b
2
2 1/2
r = (rscal. ? (2.5rLS) ) , where rscal. = 0.002r, and for the angles
r = 3rLS; the correlation coefficients higher than 0.6 are: conformer
1: r(C2–C7)/r(C7–C8) = –0.64; conformer 2: \(C2C7C6)/
the ratio of R-factors of compared models, e.g., R (1)/
f
Rf(2–1) = 4.46/4.25 = 1.049, is less than the critical value
of 1.0718. Taking into account R (2)/R (2–1) = 4.34/
\
(N1C8C7) = 0.62, \(C12C10C4)/\(C10C12H) = 0.62; conformer
f
f
2
–1: r(C2–C7)/r(C7–C8) = –0.63
4
.25 = 1.021, we come to the conclusion that the all
c
Fixed value at LS analysis
models of conformers 1, 2 and 2–1 can be presented as
possible gas-phase structures (Table 4).
P
n
2
xiðsiÞ½siMexpðsiÞÀkM siMtheorðsiފ
d
i
Rf ¼
n
P
A comparison of structural parameters of the complexes
t
Cu(pc ) (conformer 2–1) and Cu(pc), obtained from DFT
2
xiðsiÞ½siMexpðsiފ
i
calculations and LS processing of GED data, is presented
in Table 5. According to DFT calculations, the introduc-
tion of tert-butyl groups has a weak influence on the
geometry of the coordination cavity. The maximal changes
occur in the C3–C4 and C4–C5 distances of the benzene
NBO analysis results
t
An NBO analysis of Cu(pc ), at the UB3LYP/ECP10MWB
level of theory, provided bond orders and net atomic
charges for conformers 1 and 2 as well as for the unsub-
stituted Cu(pc). Additionally, for Cu(pc) NBO analyses
were performed in unrestricted and restricted open-shell
approaches for comparison purposes. The above-mentioned
NBO analyses show that the bond orders for all C–C and
C–N bonds in the substituted and unsubstituted phthalo-
cyanines are close to 1.5 (except for C1–C2 and C7–C8 in
the isoindole fragments), which is consistent with the
presence of p-conjugation in those fragments. The C–C
bond orders in tert-butyl substituents are, as expected, very
close to one. Atomic charge analyses for both conformers
˚
rings and amount to 0.01 and 0.005 A, respectively.
Concerning the bond angles, the largest difference of
2
.9° is observed for the \(C3–C4–C5) which is larger in
Cu(pc) than in Cu(pc ) angle. In contrast to the calcula-
t
t
tions, the experiments for Cu(pc) and Cu(pc ) are in mutual
agreement within experimental error for all bond lengths
except for Cu–N1 and C1–C2. The Cu–N1 distance has a
multiplicity of four, but does not appear as a distinct peak
or shoulder in the radial distribution curve f(r). Therefore,
its value cannot be determined as reliably as the corre-
sponding one in Cu(pc) [17].
t
of Cu(pc ) provide evidence for a significant role of
The difference between error limits of bond lengths C2–
t
C7 as well as C4–C5 in Cu(pc ) and Cu(pc) is the result of
covalent interaction in Cu–N bonding, and the same is true
for the unsubstituted complex Cu(pc). Thus, the charge of
the central copper ion for the studied complexes is sub-
stantially less than its formal magnitude of ?2 (see
Table 6).
different set of structural parameters used for geometrical
description of independent parameters in molecular model.
The same relates to the bond angles \(C1–C2–C7) and
\
(C3–C4–C5).
123

Struct Chem
Table 5 Comparison of the structural parameters of the molecules
symmetry) and Cu(pc) obtained by
GED and DFT
Table 6 Natural atomic charges (q) and Wiberg bond indexes (Q) in
t
Cu(pc ) (conformer 2–1 of C
t
Cu(pc ) of C4h symmetry conformers and Cu(pc) as obtained by NBO
analysis
s
a,b
t
Cu(pc )
t
Cu(pc) [17] Cu(pc )
t
Cu(pc )
Parameters
Cu(pc) [17]
Atom/bond
Cu(pc)
c
DFT , r
GED, rh1
e
Conformer 1
UB3LYP
Conformer 2
ROB3LYP
r(Cu–N1)
r(N1–C1)
r(C1–N2)
r(CuÁÁÁN2)
r(C1–C2)
r(C2–C7)
r(C2–C3)
r(C3–C4)
r(C4–C5)
1.964(5)
1.378(3)
1.318(4)
3.369(7)
1.448(5)
1.405(3)
1.399(3)
1.401(3)
1.421(3)
1.543(3)
1.550(3)
1.084(4)
90.0
1.949(5)
1.381(5)
1.325(5)
3.385(9)
1.459(5)
1.417(11)
1.399(4)
1.397(4)
1.429(18)
–
1.965 1.964
1.369 1.369
1.320 1.320
3.363 3.362
1.457 1.455
1.400 1.402
1.386 1.390
1.398 1.388
1.408 1.403
a
q(Cu)
?1.32
–0.66
–0.47
?0.41
–0.06
–0.18
?0.004
–0.21
–0.16
–0.08
–0.003
–0.58
–0.59
0.20
?1.32
–0.66
–0.47
?0.41
–0.06
–0.19
?0.004
–0.20
–0.156
–0.08
–0.004
–0.58
–0.59
0.20
?1.32
–0.66
–0.47
?0.41
–0.07
–0.17
–0.19
–0.19
–0.17
–0.07
–
?1.32
–0.66
–0.47
?0.40
–0.07
–0.17
–0.19
–0.19
–0.17
–0.07
–
q(N1)
q(N2)
q(C1)
q(C2)
q(C3)
q(C4)
q(C5)
r(C10–C12)
r(C10–C11)
r(C3–H)
1.536
1.543
–
–
q(C6)
–
q(C7)
1.091(9)
90.0
1.081 1.081
q(C10)
d
\
\
\
\
\
\
\
\
\
\
\
a
(N1–Cu–N4)
(Cu–N1–C1)
(N1–C1–N2)
(C8–N3–C9)
(C1–N1–C8)
(N1–C1–C2)
(C1–C2–C7)
(C2–C7–C6)
(C2–C3–C4)
(C3–C4–C5)
(C2–C3–H)
90.0
90.0
q(C11)
–
–
d
125.6
125.9(2)
128.2(5)
121.9(7)
125.5
127.7
123.7
108.9
109.0
106.4
120.2
119.4
118.3
119.2
125.6
127.6
123.7
108.9
109.1
106.5
121.1
117.7
121.2
120.8
q(C12)
–
–
127.5(4)
123.8(8)
Q(Cu–N1)
Q(N1–C1)
Q(C1–N2)
Q(C1–C2)
Q(C2–C7)
Q(C2–C3)
Q(C7–C6)
Q(C3–C4)
Q(C6–C5)
Q(C4–C5)
Q(C4–C10)
Q(C10–C12)
Q(C10–C11)
a
0.20
1.23
1.38
1.09
1.30
1.36
1.36
1.47
1.47
1.40
–
0.20
1.23
1.38
1.09
1.30
1.36
1.36
1.47
1.47
1.40
–
1.23
1.23
108.8(10) 108.3(5)
1.38
1.38
107.6(5)
108.6(8)
120.6(2)
117.5(7)
109.5(5)
106.4(4)
121.3(2)
117.7(2)
1.09
1.09
1.29
1.30
1.38
1.35
1.37
1.35
123.0(10) 121.1(4)
122.9(60) 123.2(44)
1.42
1.45
1.46
1.49
˚
The values of bond lengths (A) and angles (°)
In parentheses, the values of total errors for bond lengths are given
1.37
1.35
b
0.98
0.98
2
2 1/2
as r = (rscal. ? (2.5rLS) ) , where rscal. = 0.002r, and for the
angles r = 3rLS
1.00
1.00
–
–
0.99
0.99
–
–
c
UB3LYP/ECP10MWB
d
The atomic charges are presented in the portions of elementary
Fixed value at LS analysis
charge e
The supposed charge transfer interaction responsible for
this effect could be proved by NBO analysis, which shows
a donor–acceptor interaction between the donor ‘‘hsp2’’
hybrid orbitals of the nitrogen atom and the acceptor
to an increase in the lengths of these bonds, which leads a
decrease in repulsion between electron densities on orbitals
r(C3–C4) and r(C5–C4) and finally to decrease in \(C3–
t
C4–C5) angle in Cu(pc ) as compared with those in Cu(pc)
‘‘3dx2–y2’’ and ‘‘4s’’ orbitals of the copper atom. Interaction
t
energies E(2) are listed in Table 7 for both Cu(pc ) and
(see Table 5).
Cu(pc).
It should be noted that the introduction of a tert-butyl
substituent has no remarkable effect on either interaction
energy or the distribution of electron density in the coor-
dination cavity, but evokes a perturbation in the charges on
atoms C3, C4 and C5. According to Table 6, the changes in
the atomic charges on C3, C4 and C5 in the transition
AIM analysis results
Another set of atomic charges and bond ellipticities as a
measure for bond orders could be obtained from QTAIM
analysis of the molecular wave functions for both con-
t
formers of Cu(pc ) and the unsubstituted Cu(pc) at
t
Cu(pc) ? Cu(pc ) (which is accompanied by a decrease in
Wiberg bond indexes from 1.47 to 1.42 for C3–C4 bond
UB3LYP/ECP10MWB approach (see Table 8). The ten-
dencies for most of the atomic charges calculated in this
approach are as expected. Though the central copper atom
and from 1.40 to 1.37 for C4–C5 bond) are corresponding
1
23

Struct Chem
(
Table 7 Second-order perturbation energies E (kcal mol ) of
2)
-1
t
Table 8 Atomic charges and bond ellipticities of Cu(pc ) C4h sym-
metry conformers and Cu(pc) as obtained by QTAIM analysis
t
orbital interactions in the metal phthalocyanines: Cu(pc) and Cu(pc )
(
conformer 1)
t
Cu(pc )
Atom/bond
Cu(pc)
P
(
2)
(2)
E (M / N)
M / N direct donation
E
Conformer 1
Conformer 2
UB3LYP
t
Cu(pc )
Atomic charges
Cu
?1.14
-1.17
?0.99
?0.99
-1.21
-0.02
-0.02
?0.01
?0.00
-0.02
-0.01
?0.09
?0.05
?0.05
?0.05
?1.14
-1.17
?0.99
?0.99
-1.21
-0.02
-0.02
?0.01
?0.00
-0.02
-0.01
?0.09
?0.05
?0.05
?0.05
?1.15
-1.17
?0.99
?0.99
-1.21
-0.02
-0.02
?0.01
?0.01
-0.01
-0.01
–
2
.08
a Spin
b Spin
hsp (N) ? 4s(Cu)
23.7
28.5
16.7
N1
2.03
hsp (N) ? 4s(Cu)
C1
2.03
hsp (N) ? 3d(x2–y2)(Cu)
68.9
67.9
C8
Cu(pc)
a Spin
b Spin
N2
2.05
hsp (N) ? 4s(Cu)
23.4
28.1
16.4
C7
1.99
hsp (N) ? 4s(Cu)
C2
1.99
hsp (N) ? 3d(x2–y2)(Cu)
C6
C3
C5
exhibits a positive charge, its magnitude at ?1.14e is less
than the formal charge of ?2e due to N ? Cu electron
donation. Indeed, the positive charge hereby induced in
atom N1 is, according to QTAIM analysis, further delo-
calized over the adjacent carbon atoms C1 and C8 and their
neighboring N2 atom. This results in atomic charges of
C4
C10
C11
–
C12
–
C13
–
Bond ellipticities
Cu–N1
C5–C4
C3–C4
C6–C5
C2–C3
C7–C6
C7–C2
a
-
1.17e for N1, ?0.99e for C1 and C8 and -1.21e for N2.
0.03
0.18
0.20
0.20
0.18
0.18
0.16
0.03
0.17
0.21
0.21
0.17
0.17
0.17
0.03
0.18
0.20
0.20
0.18
0.18
0.16
These results are in very good agreement with those for
Cu(pc) with ?1.15e (N1), ?0.99e (C1/C8) and
-
1.21e (N2), respectively. The bond ellipticities for the
t
Cu–N1 bonds in both Cu(pc ) and Cu(pc) are at 0.03, very
close to that of a single bond (0.00 [39]). These bonds
exhibit a strongly ionic character, since the electron density
at the bond critical point (BCP) is calculated to be
-
3
-3
The atomic charges are presented in the portions of elementary
˚ ˚
.61 e A , whereas it amounts to 2.0-2.7 e A for all the
0
charge e
C–C and C–N bonds. This observation applies to both
t
conformers of Cu(pc ) as well as Cu(pc). Concerning the
electron density distribution in the benzene moieties of the
phthalocyanine ligand in both conformers, there is no
significant difference observed for the carbon atoms sepa-
rated from the substitution site by minimum one atom,
namely C2, C6 and C7. For the transition of
tendencies of charge perturbation resulting from the tert-
butyl substitution. Both NBO and QTAIM analyses show
almost no remarkable difference in the bonding situation of
t
the coordination cavity in Cu(pc) and the Cu(pc ) con-
t
Cu(pc) ? Cu(pc ), the atomic charges of the ipso and
formers. According to QTAIM analysis, the benzene
moiety or rather the atoms C3, C4 and C5 just become
more electron rich while undergoing tert-butyl substitution,
whereas NBO for those atoms reliably reproduce the first-
semester organic chemistry rule of a positively inductive
substituent increasing the electron density in ortho-position
and decreasing it in ipso- and meta-position. The same
conclusions can be drawn from the results of the above-
mentioned calculations in ROB3LYP, UHF and ROHF
approaches as well as from the calculations of the free
ortho carbon atoms C3, C4 and C5 undergo changes, as
they are all smaller in magnitude by approximately 0.01.
Altogether it should be noted that with QTAIM analyses no
differences in the electron density distribution between the
t
Cu(pc ) conformers 1 and 2 can be observed.
Bader atomic charges are known to overestimate the
charge separation between atoms of different effective
nuclear charges or electronegativities, frequently summa-
rized under the term Perrin effect [40]. Therefore, we
refrain from comparing the atomic charges obtained from
NBO and AIM analyses, respectively, but instead want to
focus on the similarities which can mostly be seen in the
2
-
t 2-
ligand double anions pc and (pc ) in B3LYP approach.
The results of these calculations are summarized in
Tables S8 and S9.
123

Struct Chem
1
1. Girichev GV, Giricheva NI, Golubchikov OA, Mimenkov YV,
Semeikin AS, Shlykov SA (2010) Octamethylporphyrin copper,
Conclusion
C
porphyrins in gas phase. J Mol Struct 978:163–169
12. Girichev GV, Giricheva NI, Koifman OI, Mimenkov YV,
Pogonin AE, Semeikin AS, Shlykov SA (2012) Molecular
28
H
28
N
4
Cu—a first experimental structure determination of
By combination of experimental and theoretical results, the
coordination center in copper(II) 2,9,16,23-tetra-tert-butyl
phthalocyanine, CuC N H , was determined to adopt D
4h
4
8
8
48
structure and bonding in octamethylporphyrin tin(II), SnN
28. Dalton Trans 41:7550–7558
4 28-
C
symmetry in the gas phase. According to GED and DFT
results, the internal rotation of tert-butyl groups has no
significant influence on the phthalocyanine ligand struc-
ture. Comparison of the DFT results for the structures of
H
1
1
1
3. Girichev GV, Utkin AN, Revichev YF (1984) Prib Tekh Eksp
(Russian) 2:187–190
4. Girichev GV, Shlykov SA, Revichev YF (1986) Prib Tekh Eksp
(Russian) 4
5. Shlykov SA, Girichev GV (1988) Prib Tekh Eksp (Russian)
t
Cu(pc) and Cu(pc ) shows that the tert-butyl substituents in
the ring have a minor effect on the geometry of the coor-
dination center. According to NBO and AIM analyses of
2
:141–142
16. Girichev EG, Zakharov AV, Girichev GV, Bazanov MI (2000)
Izv Vysh Uchebn Zaved Technol Text Prom (Russian) 2
7. Tverdova NV, Girichev GV, Giricheva NI, Pimenov OA (2011)
Accurate molecular structure of copper phthalocyanine (CuN8-
t
the electron density distribution in Cu(pc) and Cu(pc ),
1
both molecules are close to ideal ionic compound where
2
the central copper cation Cu surrounded by the anionic
?
32
C H16) determined by gas-phase electron diffraction and quan-
2
-
t 2-
ligand pc and (pc ) . The deviation of the calculated
atomic charges from this ideal model is evidence for the
partly covalent nature of the Cu–N bonds.
tum-chemical calculations. Struct Chem 22(2):319–325
8. Tverdova NV, Pimenov OA, Girichev GV, Shlykov SA, Gir-
icheva NI, Mayzlish VE, Koifman OI (2012) Accurate molecular
1
8 32
structure of nickel phthalocyanine (NiN C H16): gas-phase
electron diffraction and quantum-chemical calculations. J Mol
Struct 1023(12):227–233
9. Tverdova NV, Girichev GV, Krasnov AV, Pimenov OA, Koif-
man OI (2013) The molecular structure, bonding, and energetics
of oxovanadium phthalocyanine: an experimental and computa-
tional study. Struct Chem 24(3):883–890
0. Becke AD (1988) Density-functional exchange-energy approxi-
mation with correct asymptotic behavior. Phys Rev A 38:3098
1. Becke AD (1993) Density-functional thermochemistry. III. The
role of exact exchange. J Chem Phys 98:5648–5652
2. Lee C, Yang W, Parr RG (1988) Development of the Colle-
Salvetti correlation-energy formula into a functional of the
electron density. Phys Rev B Condens Matter 37:785–789
Acknowledgments This work was supported by the Russian
Foundation for Basic Research (Grant No. 13-03-00975a), the Rus-
sian Scientific Foundation (Agreement 14-23-00204), the Deutsche
Forschungsgemeinschaft (core facility GED@BI, Grant Mi477/21-1),
the DAAD ‘‘Mikhail Lomonosov’’ program 2014–2015. Computa-
tional resources were provided by the ‘‘Regionales Rechenzentrum
der Universit a¨ t zu K o¨ ln (RRZK)’’.
1
2
2
2
References
1
. Eley DD, Hazeldine DJ, Palmer TF (1973) Mass spectra, ion-
ization potentials and related properties of metal-free and tran-
sition metal phthalocyanines. J Chem Soc Faraday Trans 2 Mol
Chem Phys 69:1808
. Basova T, Semyannikov P, Plyashkevich V, Hassan A, Igumenov
I (2009) Volatile phthalocyanines: vapor pressure and thermo-
dynamics. Crit Rev Solid State Mater Sci 34:180–189
. Gasstevens MK, Samoc M, Pfleger J, Prasad PN (1990) J Chem
Phys 92:2019
23. Perdew JP, Burke K, Ernzerhof M (1996) Generalized gradient
approximation made simple. Phys Rev Lett 77:3865
24. Hehre WJ, Ditchfield R, Pople JA (1972) Self—consistent
molecular orbital methods. XII. Further extensions of Gaussian—
type basis sets for use in molecular orbital studies of organic
molecules. J Chem Phys 56:2257–2261
25. Dill JD, Pople JA (1975) Self-consistent molecular orbital
methods. XV. Extended Gaussian-type basis sets for lithium,
beryllium, and boron. J Chem Phys 62:2921–2923
26. Hariharan PC, Pople JA (1973) The influence of polarization
functions on molecular orbital hydrogenation energies. Theor
Chim Acta (Berl) 28(3):213–222
2
3
4
. Guo L, Ma G, Liu Y, Mi J, Qian S, Qiu L (2002) Optical and non-
linear optical properties of vanadium oxide phthalocyanine films.
Appl Phys B 74:253–257
5
6
. Baumann F, Bienert B, Rosch G, Vollmann H, Wolf W (1956)
Angew Chem Int Ed Engl 68:133–150
27. Dunning THJ (1989) Gaussian basis sets for use in correlated
molecular calculations. I. The atoms boron through neon and
hydrogen. J Chem Phys 90(2):1007–1023
28. Dolg M, Wedig U, Stoll H, Preuss H (1987) Energy-adjusted
ab initio pseudopotentials for the first row transition elements.
J Chem Phys 86(2):866–872
29. Martin JML, Sundermann A (2001) Correlation consistent
valence basis sets for use with the Stuttgart–Dresden–Bonn rel-
ativistic effective core potentials: the atoms Ga–Kr and In–Xe.
J Chem Phys 114(8):3408–3421
30. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA,
Cheeseman JR, Montgomery JA, Vreven JT, Kudin KN, Burant
JC, Millam JM, Iyengar SS, Tomasi J, Barone V, Mennucci B,
Cossi M, Scalmani G, Rega N, Petersson GA, Nakatsuji H, Hada
M, Ehara M, Toyota K, Fukuda R, Hasegawa J, Ishida M,
Nakajima T, Honda Y, Kitao O, Nakai H, Klene M, Li X, Knox
JE, Hratchian HP, Cross JB, Adamo C, Jaramillo J, Gomperts R,
Stratmann RE, Yazyev O, Austin AJ, Cammi R, Pomelli C,
. Kurmaev EZ, Shamin SN, Galakhov VR, Moewes A, Otsuka T,
Koizume S, Endo K (2001) Electronic structure of thiophenes and
phtalocyanines. Phys Rev B 64(4):452111–452117
. Berezin BD (1978) Coordination compounds of porphyrins and
phthalocyanines. Nauka, Moscow
. Mastryukov V, Ruan Chong-yu, Fink M (2000) The molecular
structure of copper- and nickel-phthalocyanine as determined by
gas-phase electron diffraction and ab initio/DFT computations.
J Mol Struct 556:225–237
. Finazzo C, Calle C, Stoll S, Van Doorslaer S, Schweigery A
(2006) Matrix effects on copper(II)phthalocyanine complexes. A
combined continuous wave and pulse EPR and DFT study. Phys
Chem Chem Phys 8:1942–1953
7
8
9
1
0. Plyashkevich V, Basova T, Semyannikov P, Hassan A (2010)
Vapour pressure of tetra-tert-butyl substituted phthalocyanines.
Thermochim Acta 501:108–111
1
23

Struct Chem
Ochterski JW, Ayala PY, Morokuma K, Voth GA, Salvador P,
34. Keith TA (2014) AIMAll (Version 13.01.27)
35. Sipachev VA (1985) Calculation of shrinkage corrections in
harmonic approximation. J Mol Struct (Theochem) 121(1–2):
143–151
36. Sipachev VA (1999) Vibrational effects in diffraction and
microwave experiments: a start on the problem. Advanced in
Molecular Structure Research/I Hargittai, M Hargittai- New
York: JAI Press 5:263–311
37. Sipachev VA (2001) Local centrifugal distortions caused by
internal motions of molecules. J Mol Struct 567–568:67–72
38. Hamilton WC (1965) Significance tests on the crystallographic R
factor. Acta Crystallogr 18:502–510
Dannenberg JJ, Zakrzewski VG, Dapprich S, Daniels AD, Strain
MC, Farkas O, Malick DK, Rabuck AD, Raghavachari K,
Foresman JB, Ortiz JV, Cui Q, Baboul AG, Clifford S, Cio-
slowski J, Stefanov BB, Liu G, Liashenko A, Piskorz P,
Komaromi I, Martin RL, Fox DJ, Keith T, Al-Laham MA, Peng
CY, Nanayakkara A, Challacombe M, Gill PMW, Johnson B,
Chen W, Wong MW, Gonzalez C, Pople JA (2003) Gaussian 03,
Revision B.03. Gaussian Inc, Pittsburgh
3
1. Carpenter JE, Weinhold F (1988) Analysis of the geometry of the
hydroxymethyl radical by the ‘‘different hybrids for different
spins’’ natural bond orbital procedure. J Mol Struct (Theochem)
1
69:41–62
39. Bader RFW, Slee TS, Cremer D, Kraka E (1983) J Am Chem Soc
105:5061–5068
40. Perrin CL (1991) J Am Chem Soc 113:2865–2868
3
3
2. Zhurko GA, Zhurko DA ChemCraft version 1.6 (build 312);
version 1.6 (build 312) ed. http://www.chemcraftprogcom/index.
html
3. Bader RFW (1990) Atoms in molecules—a quantum theory.
Oxford University Press, Oxford
123