Accurate molecular structure of nickel phthalocyanine (NiN 8C32H16): Gas-phase electron diffraction and quantum-chemical calculations

Article abstract of DOI:10.1016/j.molstruc.2012.05.066

The molecular structure of the nickel phthalocyanine (NiPc) was determined using the combination of gas-phase electron diffraction (GED), mass spectrometry and quantum chemical calculations. The DFT calculations with employing different functionals and ba

Full text of DOI:10.1016/j.molstruc.2012.05.066

Journal of Molecular Structure 1023 (2012) 227–233
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
Accurate molecular structure of nickel phthalocyanine (NiN8C32H16): Gas-phase
electron diffraction and quantum-chemical calculations
a
b
Natalya V. Tverdova a, Oleg A. Pimenov a, Georgiy V. Girichev a, , Sergey A. Shlykov , Nina I. Giricheva ,
⇑
Vladimir E. Mayzlish a, Oscar I. Koifman a
a Ivanovo State University of Chemistry and Technology, Department of Physics, Engels Av. 7, 153000 Ivanovo, Russia
b Ivanovo State University, Department of Physical Chemistry, Ermak St. 39, 153025 Ivanovo, Russia
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 6 June 2012
The molecular structure of the nickel phthalocyanine (NiPc) was determined using the combination of
gas-phase electron diffraction (GED), mass spectrometry and quantum chemical calculations. The DFT
calculations with employing different functionals and basis sets found the molecular structure with
D4h symmetry that agrees satisfactorily to the one found in experiment at 776 5 K. The important bond
lengths and bond angles according to GED (total errors in parentheses) are: rh1(Ni–N) = 1.913(5) Å, rh1(N–
Keywords:
Nickel phthalocyanine
Electron diffraction
Mass spectrum
DFT calculations
Molecular structure
Cp) = 1.385(4) Å,
rh1(Cp–Nm) = 1.327(5) Å,
r
h1(Cp–Cp) = 1.462(6) Å,
rh1(Cp–C) = 1.399(3) Å,
rh1(C–
C) = 1.395(3) Å, \NiNCp = 126.6(2)°, \NCpNm = 128.4(4)°, \CpNmCp = 120.1(5)°, \NCpCp = 110.0(4)°,
\CpCpC = 121.2(1)°. The nickel–nitrogen bond distance obtained is by 0.04 Å longer than the one found
for this compound in the literature. The structures of the crystalline and gaseous NiPc complex are com-
pared. The effect of the metal (Ni, Cu, Zn) on structural features of the MPc complexes is discussed. Using
NBO-analysis, the correlation between the metal–ligand bond distance and the total energy of donor–
acceptor orbital interactions per one M–N bond was found in the MPc complexes.
Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction
complexes (ZnPc, SnPc, CuPc) have been re-investigated by GED
[9,12,13]. It should be noted that the re-investigation of the ZnPc,
The increased interest in the investigation of metal phthalocya-
nines (MPcs) is based on their industrial and biological applica-
tions. At present these compounds are broadly used as dyes [1,2]
in paint and varnish industry or color printing. The electrical and
optical properties of metal phthalocyanines are actively used in
photovoltaic solar cells [3], field-effect transistors and light-emit-
ting diodes [4], liquid crystals [5], etc.
The high chemical and thermal stability of these complexes
makes it possible to form the thin phthalocyanine films by OMBD
(Organic Molecular Beam Deposition) method, having prospects in
production of electronic devices [6].
Despite a great interest in the MPc complexes, they have not
been extensively studied by the GED method because of low vola-
tility, which leads to certain difficulties for the experiment.
The gas-phase structures of high-temperature MPc complexes
(where M = Sn, Mg, Zn, Cu and Ni) were first reported by Fink
and co-workers more than 10 years ago [7,8]. Since that time the
molecular structure of only four Pc macrocycles (F-PcZn, H2Pc,
and Cl-MPc, where M = Al, Ga) [9–11] have been studied, and three
SnPc, CuPc complexes resulted in a considerably higher accuracy
molecular structures, as compared with the previous investiga-
tions demonstrating the reliability of the GED method when being
associated with high-level quantum chemical calculations for the
structure determination of the MPc complexes.
According to the study [8] the Ni–N bond distance was found to
be 1.871(13) Å and is well consistent with values 1.883 Å and
1.887(6) Å obtained by DFT calculations [8] and X-ray study [14],
respectively. Here it is necessary to note that in the work [14]
the results of X-ray investigation of nickel–iodine complex, NiPcIꢀ,
are presented. However, the Ni–N bond distance calculated in later
works [15–17] is about 0.04 Å longer than the experimental value
reported in the study [8]. Moreover, the disagreement factor be-
tween the experimental and theoretical sM(s) functions for the
NiPc was found unusually high, Rf = 11.7%, approximately twice
as large as that for the four other MPc complexes, M = Sn, Mg, Zn,
Cu [7,8].
In this connection we re-investigated the NiPc complex in order
to determine the accurate structure. Here we report the results of
the structure investigation of NiPc by synchronous GED and mass
spectrometry (GED/MS) along with high level quantum chemical
calculations.
⇑
Corresponding author. Tel.: +7 4932 359874; fax: +7 4932 417995.
E-mail addresses: girichev@isuct.ru, g.v.girichev@mail.ru (G.V. Girichev).
0022-2860/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molstruc.2012.05.066

228
N.V. Tverdova et al. / Journal of Molecular Structure 1023 (2012) 227–233
Table 1
The conditions of the synchronous gas-phase electron diffraction and mass spectro-
metric experiment.
Nozzle-to-plate distance, mm
338
598
Electron beam current,
Ionizing voltage, V
lA
1.4
50
0.9
50
Accelerating voltage, kV
72
71
Temperature of effusion cell, °C
Exposure time, s
505(5)
99
502(5)
57
Residual gas pressure diffraction chamber/
MS-block, torr
2 ꢁ 10–6
/
3 ꢁ 10–6
/
6 ꢁ 10–7
7 ꢁ 10–7
smin ꢀ Smax
;
D
s, Åꢀ1
2.4–26.8; 0.1
5
1.1–15.0; 0.1
6
Number of films
2. Experimental
The nickel phthalocyanine was prepared by heating nickel ace-
tate tetrahydrate (0.68 g, 0.028 mol) with phthalonitrile (0.01 mol)
at 453–458 K. The resulting mixture was re-precipitated from con-
centrated sulfuric acid upon dilution. The colored precipitate was
collected by filtration, washed with water and dried at 373–
383 K. The yield of product, NiPc, 4.28 g (75%). Anal. calc. for
C32H16N8Ni,%: C, 67.28; H, 2.82; N, 10.28. Found, %: C, 67.15; H,
3.01; N, 10.20. UV–vis (ClN): kmax, nm (log ): 671 (5.09).
e
Fig. 1. Molecular structure and atom numbering of NiPc (D4h – symmetry).
The electron diffraction patterns and the mass spectrum were
recorded simultaneously using a combined GED/MS unit based
on EMR-100 and APDM-1 devices [18–20]. The sample NiPc was
evaporated at 776 5 K, as measured W/Re-5/20 thermocouple,
from a molybdenum effusion cell. The electron wavelength was
determined from the diffraction patterns of ZnO recorded before
and after the GED experiments on NiPc. The conditions of the
GED/MS experiment are shown in the Table 1. The electron diffrac-
tion patterns were scanned by a modified [21] MD-100 microden-
sitometer. For each film, a rectangular area of 10 ꢁ 130 mm2 was
scanned at a diagonal direction (33 equidistant lines with a step
of 0.1 mm along each line). The total intensity curves were ob-
tained from two nozzle-to-plate distances L1 = 598 mm and
L2 = 338 mm.
Simultaneously with recording the electron diffraction patterns,
the mass spectra of the vapor were recorded. The relative abun-
dances of ion currents are listed in the Table 2. These data indicate
that only the monomeric molecules of NiPc are presented in the
vapor. No volatile admixtures were detected at a level of 1%, at
least. It is worthwhile to note that the mass spectrum of NiPc is
similar to those recorded for other transition metal phthalocya-
nines [13,22] (see Table 2).
one-parameter hybrid functional of Perdew, Burke and Ernzerhof
PBE0 [26], the gradient-corrected exchange-correlation GGA func-
tional of Becke and Perdew BP86 [24,27], which has been com-
bined with different basis sets: 6-31Gꢂ (N, C, H) [28,29], 6-311Gꢂ
(Ni) [30,31], cc-pVTZ (N, C, H) [32], TZV (N, C, H) [33] with the addi-
tion of d-polarization functions for C, N, H [34,35] and, in case of
nitrogen atoms, with the addition of p-diffuse functions [36]. The
core electron shells (1s2s2p) of the Ni atom were described by
the relativistic effective core potential ECP [37]. For the description
of the nickel valence shell the basis set (8s7p6d2f1g/6s5p3d2f1g)
[38] has been used. All calculations were performed with the
Gaussian 03 program [39].
The molecular structure of NiPc (1A1g low spin electronic state)
was optimized under D4h symmetry. The model of the NiPc mole-
cule is shown in the Fig. 1. No imaginary frequencies were found
assuring that the structure of D4h symmetry corresponds to the
minimum of potential energy surface. Structural parameters of
the NiPc molecule obtained by DFT calculations are summarized
in the Table 3.
The force field from B3LYP/(ECP(Ni), cc-pVTZ (N, C, H)) calcula-
tion was used to calculate the set of the vibrational corrections
D
r = rh1 ꢀ ra and the root-mean-square amplitudes for the temper-
3. Computational methods
ature of the GED experiment by the program Shrink [40–42] using
harmonic approximation and taking into account the non-linear
interrelation between internal and Cartesian vibrational
DFT calculations on the NiPc complex were carried out using
the Becke’s three-parameter hybrid functional B3LYP [23–25], the
Table 2
Mass spectra of the MPc complexes.
Abundance
Ion
FePc
CoPc
NiPc
CuPc
ZnPc
Ref.
[22] (70)a
[22] (70)
[22] (70)
[This work]b (50)
[22] (70)
[13]b (50)
[22] (70)
[MPc]+
100
30.8
1.2
7.6
19.7
0.7
100
23.6
0.8
9.0
10.5
1.9
100
40.3
1.4
5.6
1.0
1.6
7.6
100
27.8
–
10.2
<1
100
30.0
–
11.5
<1
100
31.5
100
22.9
100
32.9
[MPc]2+
[MPc]3+
[M+1/4Pc]+
[M+1/2Pc]2+
[1/4Pc]+
[M]+
8.9
0.7
3.6
16.1
17.9
5.8
11.4
4.4
<1
10.6
<1
11.6
3.7
14.4
10.3
10.1
19.3
a
Energy of ionizing electrons is given in parentheses in eV.
Mass spectra recorded during the electron diffraction experiment: for long (left) and short (right) camera distances.
b

N.V. Tverdova et al. / Journal of Molecular Structure 1023 (2012) 227–233
229
Table 3
Structural parameters of NiPc as obtained by DFT calculations.
Parametera
B3LYP
PBE0
BP86
6-31Gꢂ
6-311Gꢂ (Ni), 6-31Gꢂ (N, C, H)
ECP (Ni), TZV (N, C, H)b
ECP(Ni), cc-pVTZ (N, C, H)
Ni–N1
N1–C1
C1–N2
C1–C2
C2–C7
C2–C3
C3–C4
1.905
1.381
1.318
1.452
1.400
1.396
1.393
1.410
1.086
126.7
127.7
121.2
106.5
110.5
106.3
121.4
117.4
121.2
121.8
132.4
120.8
1.911
1.379
1.319
1.454
1.401
1.396
1.394
1.410
1.086
126.6
127.7
121.3
106.8
110.4
106.3
121.4
117.4
121.2
121.9
132.4
120.8
1.921
1.376
1.316
1.453
1.399
1.393
1.390
1.406
1.082
126.3
127.7
122.0
107.3
110.0
106.4
121.3
117.5
121.2
122.4
132.4
120.8
1.916
1.374
1.314
1.451
1.396
1.391
1.387
1.404
1.081
126.4
127.7
122.0
107.3
110.0
106.3
121.3
117.5
121.2
122.3
132.4
120.9
1.902
1.367
1.310
1.446
1.391
1.388
1.384
1.401
1.083
126.5
127.8
121.5
107.0
110.3
106.2
121.4
117.3
121.3
121.9
132.4
120.9
1.912
1.387
1.321
1.454
1.404
1.398
1.396
1.410
1.090
126.6
127.8
121.1
106.7
110.2
106.4
121.3
117.4
121.2
121.9
132.3
120.8
C4–C5
C3–Hc
\NiN1C1
\N1C1N2
\C8N3C9
\C1N1C8
\N1C1C2
\C1C2C7
\C2C7C6
\C2C3C4
\C3C4C5
\N2C1C2
\C1C2C3
\C2C3H
a
b
c
Bond distances in Å, angles in degrees. See Fig. 1 for atom numbering.
B3LYP/Stuttgart ECP(Ni), TZV (N, C, H) with the addition of d-polarization functions, in case of N with the addition of p-diffuse functions.
Average value.
Table 4
Structural parameters of NiPc according to the gas electron diffraction and DFT calculations.
Internuclear distancesa
DFTb
GED
rf
c
f
re
rh1
rg
D
lexp
lcalc.
Ni–N1g
N1–C1g
C1–N2g
NiꢃꢃꢃN2
C1–C2a
C2–C7
C2–C3g
C3–C4
C4–C5
C3–Hg
1.916
1.374
1.314
3.350
1.451
1.396
1.391
1.387
1.404
1.081
1.913(5)
1.385(4)
1.327(5)
3.385(8)
1.462(6)
1.399(3)
1.399(3)
1.395(3)
1.399(3)
1.103(10)
1.905(5)
1.382(4)
1.332(5)
3.363(8)
1.463(6)
1.403(3)
1.398(3)
1.396(3)
1.400(3)
1.102(10)
0.01018
0.00484
–0.00298
0.02497
0.00075
–0.00184
0.00223
0.00129
0.00094
0.00529
0.069(4)p2
0.055(2)p1
0.049(2)p1
0.094(4)p4
0.057(2)p1
0.052(2)p1
0.052(2)p1
0.052(2)p1
0.053(2)p1
0.078(2)p1
0.072
0.053
0.047
0.088
0.054
0.049
0.050
0.049
0.051
0.075
a
Angles
\N1NiN4
90.0
\NiN1C1g
126.4
126.6(2)
\C2C7C6g
121.3
\N1C1N2
127.7
128.4(4)
\C2C3C4
117.5
\C8N3C9
122.0
120.1(5)
\C3C4C5
121.2
\C1N1C8
107.3
106.8(4)
\N2C1C2
122.3
\N1C1C2g
110.0
110.0(4)
\C2C3Hg
120.9
DFTb
GED
90.0d
a
Angles
\C1C2C7
106.3
106.0(5)
DFTb
GED
121.2(1)
117.4(1)
121.4(4)
121.2(4)
119.6(34)
Rfe = 3.81%
a
Internuclear distances in Å, angles in degrees. See Fig. 1 for atom numbering. Error limits given in parentheses were estimated as
and = 3rLS for vibrational amplitudes and angles.
B3LYP/ECP(Ni), cc-pVTZ (N, C, H).
r
= ((0.002r)2 + (2.5
r
LS)2)1/2 for distances
r
b
c
pi – is a group of vibrational amplitudes refined in the LS analysis, where i is the group number.
d
Fixed value.
P
nxi ðs Þ½s Mexp ðs ÞꢀkM s Mtheor ðsi Þꢄ
2
i
i
i
i
e
i
P
Rf
¼
.
n xi ðs Þ½s Mexp ðs Þꢄ
2
i
i
i
Vibrational ciorrections
D
r = rh1 ꢀ ra and root mean square amplitudes l were estimated using the program SHRINK [40–42].
f
g
Independent parameters.
coordinates. The results are collected in Table 4. The calculated
vibrational amplitudes are compared with the experimental values.
were calculated on the base of the force field adopted from the
quantum chemical calculations using SHRINK program [40–42]
with non-linear correlation between Cartesian and internal
coordinates taken into account. The molecular model included 11
independent parameters: (1) six bond distances: Ni–N1, N1–C1,
C1–N2, C1–C2, C2–C3, C3–H; (2) five valence angles: \N1NiX
(the angle was included in the set of independent parameters to
describe the displacement of the Ni atom out of plane of the mol-
ecule, X is a dummy atom positioned at the center of the nitrogen
atoms plane; for the D4h model this parameter was fixed at 90°),
\NiN1C1, \N1C1C2, \C2C7C6, \C2C3H. The differences between
4. Structural analysis
The analysis of the electron diffraction data was performed with
the assumption that the vapor contained the molecules NiPc with
C4 symmetry axis giving the possibility to test D4h and C4v
structures. The set of independent parameters was used for
geometric constraining rh1 structure of molecule. The needed
vibration corrections
D
r = rh1 ꢀ ra to the internuclear distances

230
N.V. Tverdova et al. / Journal of Molecular Structure 1023 (2012) 227–233
the C2–C3 and C3–C4, C2–C7, C4–C5 bond lengths and also be-
tween \C2C7C6 and \C7C6C5, \C2C3H and \C3C4H valence an-
gles were taken from the results of the quantum chemical
calculations, B3LYP/(ECP(Ni), cc-pVTZ (N, C, H)), and fixed in the
LS-analysis. Starting values of independent parameters and of the
root-mean-square vibrational amplitudes were taken from the
above mentioned DFT computation. The amplitudes were refined
in groups corresponding to the different peaks on the radial distri-
bution curve. The weight function WE = exp[ꢀ(si ꢀ sm)2] was ap-
plied in the region s = 25.0–26.8 Å.
and long camera) showing a stable behavior of NiPc at evaporation.
The detected set of ions is typical for metal phthalocyanines and
corresponds to monomeric vapor composition (Table 2).
The model of D4h symmetry of NiPc is in good agreement with
the experimental intensities. The disagreement factor Rf obtained
between the experimental and theoretical sM(s) functions is 3.81%.
The authors [7,8] of the GED study of the NiPc, CuPc, ZnPc mol-
ecules in a gas phase have tested the nonplanar model in which the
metal atom is shifted out of plane of the molecule. Following [7,8]
we included the angle \N1NiX, describing the displacement of the
Ni atom out of plane of the molecule, in the set of independent
parameters. Structure refinement involving the \N1NiX angle at
variation yielded the planar structure in which the metal atom
lay above the molecule plane by h = 0.00(17) Å. Additionally, in or-
der to estimate the height of the pyramid (the h parameter), the
Hamilton’s criterion [43] was applied. Results of the structure
refinements at fixed h-values are plotted in Fig. 4. It may be con-
cluded that the most probable structure of the molecule is planar
with uncertainty of the displacement of the metal atom out of
plane of the macrocyclic ligand less than 0.17 Å.
The parameters of the NiPc complex obtained in two indepen-
dent GED studies, [8] and ours, have noticeable disagreements,
particularly for the Ni–N, C1–N2 and C2–C7 bond lengths. From
the Table 5, it can be seen that the Ni–N bond distance in our study
is significantly longer, by about 0.04 Å, in comparison with the va-
lue determined in the study [8]. The Fig. 3 shows that there is a dis-
tinct peak on the radial distribution curve at r about 1.91 Å. It
should be noted, that according to the GED and DFT results the
peak at r ꢅ 1.91 Å corresponds to the only distance, r(Ni–N), and
the reliability of the determination of the r(Ni–N) and the l(Ni–N)
parameters should therefore be high. The Fig. 3 illustrates also
the sensitivity of GED data to the value of the Ni–N bond distance:
the theoretical radial distribution function f(r) with Ni–
N = 1.913(5) Å fits with the experimental f(r) curve noticeably bet-
ter than the theoretical f(r) corresponding to the fixed Ni–N dis-
tance at 1.871 Å, recommended by authors [8]. It is worth to
note that there is a good agreement between the experimentally
determined value of the Ni–N in this work and those predicted
by DFT calculations (Table 3 and 4).
The Fig. 2 and 3 display the plots of sM(s) and f(r) experimental
functions along with their theoretical analogs. Refined geometric
and vibrational parameters are listed in Table 4.
5. Discussion
5.1. The molecular structure determined by GED (gas-phase structure)
The Table 2 demonstrates a satisfactory similarity of ion sets
and ion current relative abundances in the mass spectra which
had been recorded in two independent experiments (for short
sM(s)
ΔsM(s)
(a)
(b)
0
5
10
15
s, Å-1
20
25
Unlike the results [8], the parameters of the macrocyclic ligand
found in this work are well consistent with those obtained from
the GED studies of the similar molecules CuPc [13] and ZnPc [9].
This observation supports the conclusion that the structural
parameters of the Pc ligand are weakly affected by the metal atoms
Zn, Cu or Ni (Table 5).
Fig. 2. Experimental (dots) and theoretical (line) molecular scattering intensities
sM(s) and their difference sM(s) for the models: with the Ni–N bond distance
D
refined (a) and fixed at 1.871 Å (b). The theoretical molecular intensity is shown for
the model (a).
1.06
1.04
1.02
f(r)
f(r)
(a)
0.05 significance level
(b)
1.00
0
2
4
6
8
10
12
14
16
r, Å
0.00
0.05
0.10
0.15
0.20
0.25
0.30 h, Å
Fig. 3. Experimental (dots) and theoretical (line) radial distribution curves f(r) and
their difference f(r) for the models: with the Ni–N bond distance refined (a) and
fixed at 1.871 Å (b). The theoretical f(r) is shown for the model (a).
Fig. 4. The disagreement factors ratio Rh/Rmin against the fixed h value (perpen-
dicular displacement of the nickel atom out of the N8C8 plane). The dashed
horizontal line corresponds to the Hamilton’s criterion [43].
D

N.V. Tverdova et al. / Journal of Molecular Structure 1023 (2012) 227–233
231
Table 5
Ni–N distance in a free NiPc molecule [8] by ꢅ0.02 Å. According
to the results of our study this distance is longer, by ꢅ0.02 Å, than
that in crystalline NiPc [14,45–47].
The observed «crystal-gas» difference for the Ni–N distance can
be caused by effect of the packing of the molecules in crystal, and
by different physical meaning of parameters which evaluated from
GED and X-ray experiments.
In the X-ray structural analysis, a structure is determined from
the positions of maxima on the electron density map of the com-
pound. If an atom in a molecule possesses a lone electron pair,
the gravity center of electron density is shifted from the atomic nu-
cleus towards this pair [49]. In the NiPc molecule the nitrogen
atoms have the lone electron pairs directed to the Ni atom. More-
over, even at room temperature X-ray data are characterized by
noticeable effective shrinkage (ꢅ0.03 Å) of internuclear distances
as a result of lattice vibrations [49]. The experiments reported in
[14,46,47] were carried out at 283–303 K. That is, according to
the X-ray data, the Ni–N bond may seem to be shorter than the dis-
tance between these atomic nuclei.
Comparison of the structural parameters of the MPc complexes obtained by GED.
a
b
Parameters
NiPcb [this work] NiPcc [8]
rh1
1.913(5)
CuPc [13] ZnPcd [9]
rh1 re
1.949(5)
r
M–N1
1.871(13)
3.742(26)
1.391(16)
1.296
1.982(6)
3.964(12)f
1.380(5)
1.336(5)
1.455(6)
1.407(5)
1.390(5)
1.389(5)
1.402(5)
1.069(12)
N4ꢃ ꢃ ꢃN5
3.826(10)
1.385(4)
1.327(5)
1.462(6)
1.399(3)
1.399(3)
1.395(3)
1.399(3)
1.103(10)
3.898(10)
1.381(5)
1.325(5)
1.459(5)
1.417(11)
1.399(4)
1.397(4)
1.429(18)
1.091(9)
N1–C1
C1–N2
C1–C2
C2–C7
C2–C3
C3–C4
C4–C5
C3–H
\N1C1N2
\C8N3C9
\C1N1C8
\N1C1C2
\C1C2C7
\C2C7C6
\C2C3C4
\C3C4C5
\C2C3H
1.460(18)
1.470(31)
1.374(8)
1.375
1.379
1.083(37)
128.4(4)
127.9(15)
120.5
106.3(13)
111.8
105.0
120.2(12)
117.4
128.2(5)
127.4
123.9
108.6(5)
109.3
106.6
121.1(3)
117.7(8)
121.2(4)
120.7
120.1(5)
106.8(4)
110.0(4)
106.0(5)
121.2(1)
117.4(1)
121.4(4)
119.6(34)
121.9(7)
108.3(5)
109.5(5)
106.4(4)
121.3(2)
117.7(2)
121.1(4)
123.2(44)
This conclusion is in agreement with our results on the struc-
ture of free NiPc molecule but is not with the results of [8] accord-
ing to which the Ni–N bond is shorter than this one in the crystal
structures [14,45–47].
a
Internuclear distances in Å, angles in degrees. See Fig. 1 for atom numbering.
b
Error limits given in parentheses were estimated as
for distances and = 3rLS for angles.
Uncertainties were estimated as 2rLS
Parenthesized values are estimated error limits defined as 2.5(r2LS + (0.001r)2)1/2
r = ((0.002r)2 + (2.5r
LS)2)1/2
r
c
.
d
,
5.3. The effect of the metal on the structural features of the MPc
complexes
for distances and as 2.5
Estimated in this work as 2r(M–N1).
r
LS for angles.
f
The results of the gas-phase electron diffraction studies of MPc,
where M = Ni, Cu, Zn [9,13], this work show that all complexes are
characterized by a planar structure and by significant elongation of
the M–N bond in the series from Ni to Zn. Such behavior of the me-
tal–ligand bond length was found in the bis-complexes of metal
beta-diketonates [50]. As in the bis-complexes of metal beta-diket-
onates, the behavior of the metal–ligand bond in the MPc com-
plexes can be explained in terms of the crystal field theory.
Comparison of the structures of the MPc complexes (M = Ni, Cu,
Zn) demonstrates the macrocycle core expansion in the series
Ni ? Zn. It has to be noted, that the Pc ligand is adjusted under
the metal size at the cost of changes of valence angles rather than
bond distances. Indeed, as the metal size increase from Ni to Zn,
the bond distances (N2–C1), (C2–C7), angles \C1N1C8, \C8N3C9
increase and the angle \N1C1C2 decreases. However, these
changes are small: about 0.005 Å for the bond distances and about
1.5° for the bond angles (Table 5) and according to GED results
these bond distances remain unchanged within the error limits.
Similar situation observed for the DFT calculations: the differences
5.2. Solid state structure
The crystal structure of the NiPc was reported by Robertson and
Woodward in 1937 [44]. This is the only study of crystalline NiPc
by X-ray diffraction. Authors [44] noted that the NiPc molecule is
planar; and the structure of macrocyclic ligand is closely similar
to that of the metal-free phthalocyanine, but some structural
changes in NiPc were, nevertheless, observed: each of the four iso-
indole nitrogen atoms are shifted towards the nickel atom by about
0.09 Å; the C–N bond distances are slightly longer than the values
found for the metal-free phthalocyanine.
Several works on the crystal structures of NiPc(Brꢀ), NiPcðI3ꢀÞ,
NiPcðAsFꢀ6 Þ and NiPcðSbFꢀ6 Þ [14,45–47] whose composition in-
cludes the NiPc complex are available from the Cambridge Crystal-
lographic Database [48]. Selected structural parameters of these
complexes are presented in Table 6. The average Ni–N distance
for the above mentioned complexes in cꢀrystals varies from
1.887 Å in NiPcðIꢀ3 Þ to 1.899 Å in NiPcðAsF6 Þ and exceeds the
Table 6
Comparison of selected structural parameters of the NiPc complex obtained by the X-ray and GED.
Parametera
X-rayb
GED
NiPcðAsFꢀ6 ) [46]
NiPcðSbFꢀ6 Þ [47]
NiPc(Brꢀ)[45]
167
NiPcc [this work] rh1
776
NiPcd [8] r
748
NiPcðIꢀ3 ) [14]
Temperature, K
283–303
283–303
283–303
Ni–N1
N1–C1
C1–N2
C1–C2
C2–C7
1.89(1)
1.35(2)
1.33(2)
1.45(2)
1.39(2)
1.887(6)
1.379(12)
1.320(9)
1.456(7)
1.392(10)
1.899
1.378
1.325
1.462
1.377
1.895
1.381
1.321
1.458
1.380
1.913(5)
1.385(4)
1.327(5)
1.462(6)
1.399(3)
1.871(13)
1.391(16)
1.296
1.460(18)
1.470(31)
\NiN1C1
\N1C1N2
\C8N3C9
\C1N1C8
\N1C1C2
126.5
126.8(5)
128.5(2)
119.2(6)
106.3(6)
110.6(4)
126.8
127.8
120.0
106.3
110.3
126.9
127.8
120.1
106.2
110.3
126.6(2)
128.4(4)
120.1(5)
106.8(4)
110.0(4)
130.0
127.9(15)
120.5
106.3(13)
111.8
129(1)
118(1)
107(1)
110(1)
a
b
c
Distances in Å, angles in degrees. See Fig. 1 for atom numbering in the complexes MPc.
Average parameters. For averaging, the parameters were taken from Cambridge Crystallographic Database [48].
Uncertainties given in parentheses were estimated as ((0.002r2 + (2.5 LS)2)1/2 for bond distances and 3rLS for angles.
r
d
Uncertainties were estimated as 2rLS
.

232
N.V. Tverdova et al. / Journal of Molecular Structure 1023 (2012) 227–233
Table 7
Structural parameters of the MPc complexes, where M = Ni, Cu, Zn as obtained by the DFT–B3LYP calculations.
a
Parameter
NiPc
(I)b
CuPc
(I)b
ZnPc
(I)b
(II)c
1.917
(II)c
1.964
(II)c
1.997
[9]d
2.000
M–N1
1.920
1.967
1.998
N4ꢃꢃꢃꢃN5
3.841
1.376
1.316
1.453
1.399
1.393
1.390
1.406
3.833
1.374
1.314
1.451
1.396
1.391
1.387
1.404
3.934
1.371
1.322
1.457
1.404
1.393
1.391
1.405
3.928
1.369
1.320
1.455
1.402
1.390
1.388
1.403
3.996
1.370
1.327
1.460
1.408
1.392
1.391
1.404
3.994
1.367
1.325
1.458
1.406
1.390
1.388
1.402
4.000
1.367
1.326
1.459
1.406
1.390
1.389
1.402
N1–C1
N2–C1
C1–C2
C2–C7
C2–C3
C3–C4
C4–C5
\MN1C1
\N1C1N2
\C8N3C9
\C1N1C8
\N1C1C2
\C1C2C7
\C7C2C3
\C2C3C4
\C3C4C5
\N2C1C2
\C1C2C3
126.3
126.4
125.6
125.6
125.1
125.1
125.1
127.7
122.0
107.3
110.0
106.4
121.3
117.5
121.2
122.4
132.4
127.7
122.0
107.3
110.0
106.3
121.3
117.5
121.2
122.3
132.4
127.6
123.7
108.9
109.1
106.5
121.2
117.7
121.2
123.4
132.4
127.6
123.7
108.9
109.1
106.5
121.1
117.7
121.2
123.3
132.4
127.4
125.0
109.9
108.5
106.6
121.0
117.8
121.2
124.1
132.4
127.5
125.0
109.9
108.4
106.6
121.0
117.8
121.2
124.1
132.4
127.4
125.1
110.0
108.4
106.6
121.0
117.9
121.1
124.2
132.4
a
b
c
Internuclear distances in Å, angles in degrees. Atom numbering is given in Fig. 1.
(I) – B3LYP/Stuttgart ECP(M), TZV (N, C, H) with the addition of d-polarization functions, in case of N with the addition of p-diffuse functions.
(II) – B3LYP/Stuttgart ECP(M), cc-pVTZ (N, C, H).
d
B3LYP/cc-pVTZ-NR(Zn), cc-pVTZ (N, C, H).
do not exceed 0.01 Å for the C–C and the C–N distances and ꢅ2° for
the valence angles (Table 7).
125
124
123
122
121
120
119
118
ZnPc [9]
Using results of X-ray studies and DFT calculations the authors
[8] determined the linear relationship between some structural
parameters in the MPc complexes. Reliability of determination of
the Ni–N bond in the NiPc molecule can be estimated from the cor-
relation between (M–N) and \C8N3C9 in the series Ni ? Zn. The
plot in the Fig. 5 clearly indicates that the Ni–N distance recom-
mended in work [8] is underestimated.
It is generally known [9,12,51,52] that introduction of small or
large central atoms into the coordination cavity causes a conforma-
tion stress of a relatively rigid macrocycle. This strain can be re-
duced by moving the isoindole nitrogen atoms towards or away
from the cavity center, by the displacement of the central metal
atom out of the N4C8 plane or by the ‘‘ruffling’’ of the macrocycle
core.
CuPc [13]
NiPc [8]
NiPc [this work]
1.86
1.88
1.90
1.92
1.94
1.96
1.98
2.00
From the analysis of the structural data for 18 metal porphy-
rines obtained from X-ray determination, Hoard [51] evaluated
the smallest size of the cavity the complex can keep its planarity
to be 2.01 Å (the M–N distance). At the same time, to keep up
the planar macrocycle core for the metal phthalocyanines the
shortest M–N bond distance has to be 1.90 Å [53]. In the NiPc com-
plex the bond distance rh1(Ni–N) = 1.913(5) Å as obtained from
GED intensities in this work insignificantly exceeds the border va-
lue of 1.90 Å and the planar structure of the macrocycle core for
this molecule seems, thus, preferable.
r (M-N), Å
Fig. 5. The correlation between the M–N bond distance and \C8N3C9 valence angle
for nickel, copper and zinc phthalocyanines as provided by GED. The correlation
coefficient is 0.998.
Table 8
Atomic charges (q) of the MPc, M = Ni, Cu, Zn complexes yielded by NBO analysis.
NiPc
CuPc
ZnPc
q(M)
+1.027
ꢀ0.556
ꢀ0.437
+0.373
ꢀ0.077
ꢀ0.175
ꢀ0.210
+1.32
+1.624
ꢀ0.703
ꢀ0.445
+0.381
ꢀ0.080
ꢀ0.176
ꢀ0.211
q(N1)
q(N2)
q(C1)
q(C2)
q(C3)
q(C4)
ꢀ0.630
ꢀ0.442
+0.378
ꢀ0.078
ꢀ0.175
ꢀ0.210
5.4. NBO analysis
NBO analysis for the MPc complexes, where M = Zn, Cu, Ni was
carried out at the level B3LYP/Stuttgart ECP(Ni), TZV (N, C, H) with
the addition of d-polarization functions, in case of N with the addi-
tion of p-diffuse functions level. The calculated NBO atomic
charges on the metal and donor atoms indicate that the net charges
on the metal in CuPc and NiPc are considerably smaller than that at
the Zn atom. The net charge of +1.62 at the Zn atom is reasonably
close to the ionic limit of +2 (Table 8). These results are consistent
with the calculated energies of donor–acceptor interactions E(2) be-
tween the donor ‘‘hsp2’’ hybrid orbital of the nitrogen atom and the
acceptor ‘‘3d’’ and ‘‘4s’’ orbitals of the metal (Table 9). The lower
value of the donor (N) ? acceptor (M) interaction energy
(E(2)N?M) was found in the ZnPc and the higher in the NiPc. The
experimental (M–N) bond distances correlate clearly with the total
donor–acceptor interaction energy per one M–N bond and show
that the (Ni–N) bond is stronger and shorter (Fig. 6).

N.V. Tverdova et al. / Journal of Molecular Structure 1023 (2012) 227–233
233
Table 9
[9] T. Strenalyuk, S. Samdal, H.V. Volden, J. Phys. Chem. A 111 (2007) 12011.
[10] T. Strenalyuk, S. Samdal, H.V. Volden, J. Phys. Chem. A 112 (2008) 4853.
[11] T. Strenalyuk, S. Samdal, H.V. Volden, J. Phys. Chem. A 112 (38) (2008) 9075.
[12] T. Strenalyuk, S. Samdal, H.V. Volden, J. Phys. Chem. A 112 (40) (2008) 10046.
[13] N.V. Tverdova, G.V. Girichev, N.I. Giricheva, O.A. Pimenov, Struct. Chem. 22
(2011) 441.
[14] C.J. Schramm, R.P. Scaringe, D.R. Stojakovic, B.M. Hoffman, J.A. Ibers, T.J. Marks,
J. Am. Chem. Soc. 102 (1980) 6702.
[15] M.-S. Liao, S. Scheiner, J. Comput. Chem. 23 (2002) 1391.
[16] A.V. Soldatova, J. Kim, X. Peng, A. Rosa, G. Ricciardi, M.E. Kenney, M.A.J.
Rodgers, Inorg. Chem. 46 (2007) 2080.
[17] Z. Liu, X. Zhang, Y. Zhang, J. Jiang, Spectrochem. Acta A 67 (2007) 1232.
[18] G.V. Girichev, A.N. Utkin, Yu.F. Revichev, Prib. Tekh. Eksp. 2 (1984) 187.
[19] G.V. Girichev, S.A. Shlykov, Yu.F. Revichev, Prib. Tekh. Eksp. 4 (1986) 167.
[20] G.V. Girichev, S.A. Shlykov, V.N. Petrova, N.Yu. Subbotina, S.B. Lapshina, T.G.
Danilova, Izv. Vyssh. Uchebn. Zaved., Khim. Khim. Tekhnol. 31 (8) (1988) 46.
[21] E.G. Girichev, A.V. Zakharov, G.V. Girichev, M.I. Bazanov, Izv. Vyssh. Uchebn.
Zaved., Tekstiln. Prom. 2 (2000) 142.
Second order perturbation energies E(2) (kcal/mol) of orbital interactions in the metal
phthalocyanines (see section ‘NBO analysis’ for details).
P
M N direct donation
E(2)
E
(2)(M N)
NiPc
CuPc
hsp1.90(N) ? 4s(Ni)
49.34
39.24
hsp1.90(N) ? 3d(x2–y2)(Ni)
88.58
a
spin
hsp2.02(N) ? 4s(Cu)
23.17
28.46
16.13
b spin
hsp1.97(N) ? 4s(Cu)
hsp1.97(N) ? 3d(x2–y2)(Cu)
67.76
53.42
ZnPc
hsp2.12(N) ? 4s(Zn)
53.42
95
90
85
80
75
70
65
60
55
50
45
[22] D.D. Eley, D.J. Hazeldine, T.F. Palmer, Chem. Phys. 69 (1973) 1808.
[23] A.D. Becke, J. Chem. Phys. 98 (1993) 5648.
[24] A.D. Becke, Phys. Rev. A 38 (1988) 3098.
[25] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.
[26] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 78 (1997) 1396.
[27] J.P. Perdew, Phys. Rev. B 33 (1986) 8822.
[28] G.A. Petersson, A. Bennett, T.G. Tensfeldt, M.A. Al-Laham, W.A. Shirley, J.
Mantzaris, J. Chem. Phys. 89 (1988) 2193.
[29] G.A. Petersson, M.A. Al-Lah, J. Chem. Phys. 94 (1991) 6081.
[30] R. Krishnan, J.S. Binkley, R. Seeger, J.A. Pople, J. Chem. Phys. 72 (1980) 650.
[31] M.J. Frish, J.A. Pople, J.S. Binkley, J. Chem. Phys. 80 (1984) 3265.
[32] T.H. Dunning, J. Chem. Phys. 90 (1989) 1007.
NiPc
CuPc
[33] T.H. Dunning, J. Chem. Phys. 55 (1971) 716.
[34] M. Urban, V. Kello, P. Carsky, Theor. Chim. Acta (Berl.) 45 (1977) 205.
[35] P.C. Hariharan, J.A. Pople, Theor. Chim. Acta (Berl.) 28 (1973) 213.
[36] T.H. Dunning, P.J. Hay, in: H.F. Schaefer III (Ed.), vol. 2, Plenum Press. New York,
1977, p. 1.
ZnPc
[37] M. Dolg, U. Wedig, H. Stoll, H. Preuss, J. Chem. Phys. 86 (1987) 866.
[38] J.M.L. Martin, A. Sundermann, J. Chem. Phys. 114 (2001) 3408.
[39] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
V.G. Zakrzewski, J.A. Montgomery, Jr., R.E. Stratmann, J.C. Burant, S. Dapprich,
J.M. Millam, A.D. Daniels, K.N. Kudin, M.C. Strain, O. Farkas, J. Tomasi, V.
Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J.
Ochterski, G.A. Petersson, P.Y. Ayala, Q. Cui, K. Morokuma, D.K. Malick, A.D.
Rabuck, K. Raghavachari, J.B. Foresman, J. Cioslowski, J.V. Ortiz, B.B. Stefanov,
G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R.L. Martin, D.J. Fox,
T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, C. Gonzalez, M.
Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, J.L. Andres, C.
Gonzalez, M. Head-Gordon, E.S. Replogle, J.A. Pople, Gaussian 03, Revision B04,
Gaussian Inc., Pittsburgh, PA, 2003.
1.91 1.92 1.93 1.94 1.95 1.96 1.97 1.98 1.99 2.00
r (M–N), Å
Fig. 6. Experimental (GED) M–N bond distance and theoretical (DFT) total energy of
donor–acceptor orbital interactions per one M–N bond for nickel, copper and zinc
phthalocyanines. The correlation coefficient is 0.997. See NBO analysis section for
details.
Acknowledgement
[40] V.A. Sipachev, J. Mol. Struct. 567–568 (2001) 67.
[41] V.A. Sipachev, J. Mol. Struct. 121 (1985) 143.
[42] V.A. Sipachev, in: I. Hargittai, M. Hargittai (Eds.), Advances in Molecular
Structure Research, JAI Press, New York, 1999.
This work was supported by the Russian Foundation for Basic
Research (Grant No. 10-03-00884a).
[43] W.C. Hamilton, Acta Cryst. 18 (1965) 502.
[44] J.M. Robertson, I. Woodward, J. Chem. Soc. (1937) 219.
[45] Sharon M. Palmer, Judith M. Stanton, Brian M. Hoffman, James A. Ibers, Inorg.
Chem. 25 (14) (1986) 2296.
References
[46] K. Yakushi, H. Yamakado, M. Yoshitake, N. Kosugi, H. Kuroda, T. Sugano, M.
Kinoshita, A. Kawamoto, J. Tanaka, Bull. Chem. Soc. Jpn. 62 (1989) 687.
[47] K. Yakushi, H. Yamakado, M. Yoshitake, N. Kosugi, H. Kuroda, A. Kawamoto, J.
Tanaka, T. Sugano, M. Kinoshita, S. Hino, Synth. Met. 29 (1989) F95.
[48] Cambridge Crystallographic Date Centre (CCDC), <http://www.ccdc.cam.
ac.uk>.
[1] B.D. Berezin, Coordination Compounds of Porphyrins and Phthalocyanines,
Moscow, 1978.
[2] B.I. Stepanov, Introduction in the Chemistry and Technology of Organic Dyes,
third ed., Moscow, 1984.
[3] T.A. Yourre, L.I. Rudaya, N.V. Klimova, V.V. Shamanin, Semiconductors 37 (7)
(2003) 835.
[49] A. Domenicano, I. Hargittai (Eds.), Molecular Structures [Russian translation],
Mir, Moscow, 1997. p. 672.
[50] G.V. Girichev, N.I. Giricheva, N.V. Belova, N.P. Kuzmina, Coord. Chem. (Russian)
25 (12) (1999) 892.
[4] R.F. Service, Science 310 (2005) 1762.
[5] J. Simon, C. Sirlin, Pure Appl. Chem. 61 (1989) 1625.
[6] S. Heutz, C. Mitra, W. Wu, A.J. Fisher, A. Kerridge, A.M. Stoneham, A.H. Harker, J.
Gardener, H.-H. Tseng, T.S. Jones, Ch. Renner, G. Aeppli, Adv. Mater. 19 (2007)
3618.
[51] J.L. Hoard, Science 174 (1971) 1295.
[52] R. Mason, G.A. Williams, J.Chem, Soc. Dalton (1979) 676.
[53] J.L. Hoard, Ann. N. Y. Acad. Sci. 206 (1973) 18.
[7] V. Chong-Yu Ruan, M. Mastryukov, J. Fink, Chem. Phys. 111 (1999) 3035.
[8] V. Mastryukov, J. Mol. Struct. 556 (2000) 225.