C,C- and N,C-coupled dimers of 2-aminotetraphenylporphyrins: Regiocontrolled synthesis, spectroscopic properties, and quantum-chemical calculations

Article abstract of DOI:10.1002/chem.201304169

β,β-Bisporphyrins are intrinsically chiral porphyrin dimers with fascinating properties. The configurational stability at their axes can be directed by variation of the central metal atoms. Herein, we present a regioselective functionalization of the monomeric 2-amino-tetraphenyl-porphyrin as a versatile substrate for dimerization by oxidative coupling. By simple variation of the reaction conditions (solvent and oxidant), the oxidation selectively gave either the axially chiral C,C-coupled diaminobisporphyrin in high yields or, under Ullmann conditions, the twofold N,C-linked achiral dimer, also in good yields. A generalized mechanism for the coupling reaction is proposed based on DFT calculations. The axially chiral β,β-coupled porphyrin dimers were isolated as racemic mixtures, but can be resolved by HPLC on a chiral phase. TDDFT and coupled-cluster calculations were used to explain the spectroscopic properties of the aminoporphyrins and their dimers and to elucidate the absolute configurations of the C,C-coupled bisporphyrins. All in the spin: Depending on the oxidative coupling conditions (Ag or Fe vs. Cu), aminoporphyrin monomers can be selectively dimerized to give axially chiral C,C-coupled β,β-diaminobisporphyrins or an N,C-coupled pyrazin-fused dimer (see figure). DFT calculations suggest that the oxidant in combination with the solvent has a strong impact on the spin density of the intermediate radical cation and, consequently, on the regioselectivity of the reaction.

Full text of DOI:10.1002/chem.201304169

DOI: 10.1002/chem.201304169
Full Paper
&
Oxidative Coupling
C,C- and N,C-Coupled Dimers of 2-Aminotetraphenylporphyrins:
Regiocontrolled Synthesis, Spectroscopic Properties, and
Quantum-Chemical Calculations
Torsten Bruhn,[a] Franziska Witterauf,[a] Daniel C. G. Gçtz,[a, c] Carina T. Grimmer,[a]
Max Wꢀrtemberger,[b] Udo Radius,[b] and Gerhard Bringmann*[a]
Abstract: b,b’-Bisporphyrins are intrinsically chiral porphyrin
dimers with fascinating properties. The configurational sta-
bility at their axes can be directed by variation of the central
metal atoms. Herein, we present a regioselective functionali-
zation of the monomeric 2-amino-tetraphenyl-porphyrin as
a versatile substrate for dimerization by oxidative coupling.
By simple variation of the reaction conditions (solvent and
oxidant), the oxidation selectively gave either the axially
chiral C,C-coupled diaminobisporphyrin in high yields or,
under Ullmann conditions, the twofold N,C-linked achiral
dimer, also in good yields. A generalized mechanism for the
coupling reaction is proposed based on DFT calculations.
The axially chiral b,b’-coupled porphyrin dimers were isolat-
ed as racemic mixtures, but can be resolved by HPLC on
a chiral phase. TDDFT and coupled-cluster calculations were
used to explain the spectroscopic properties of the amino-
porphyrins and their dimers and to elucidate the absolute
configurations of the C,C-coupled bisporphyrins.
Introduction
More detailed kinetic investigation on the atropisomerization
process showed that the choice of the central metal has
a high impact on the stereochemical stability of the axis. Addi-
tional amino substituents in the proximity of the porphyrin–
porphyrin axis should not only increase the rotational barrier
of such b,b’-dimers, but would also provide possibilities for fur-
ther functionalization of the system. Thus, exocyclic amino
functions might offer the possibility to construct multimodal-
coordinated metal centers on the periphery of the porphy-
rin,[11,12] in analogy to the related, but smaller, molecule 2,2’-di-
amino-1,1’-binaphthalene (BINAM), which is used as a ligand in
palladium-catalyzed coupling reactions.[13] In this paper, we
report the highly regioselective synthesis of C,C- or N,C-linked
porphyrin dimers, controlled by simple variation of the reac-
tion conditions, and on the spectroscopic and chiroptical prop-
erties of the dimers thus obtained.
Chiral porphyrins are increasingly important as chelating li-
gands, as metal catalysts in asymmetric synthesis,[1,2] and as
substrates for chiral recognition.[3,4] The origin of chirality in
such systems can be manifold. In addition to the introduction
of stereogenic centers or axes, or planar chirality in the periph-
ery of the tetrapyrrole macrocycles,[5] there is also the possibili-
ty to link porphyrins through a chiral bridge.[6] Documented ex-
amples of chiral axes caused by a direct porphyrin–porphyrin
linkage are quite rare, but are gaining attention.[7,8] We de-
scribed the synthesis of the first b,b’-bisporphyrins with intrin-
sic axial chirality, by using a Suzuki coupling with moderate to
good yields, providing the dimers with a large variety of cen-
tral metals.[9,10] The determination of their absolute configura-
tions was accomplished on the basis of quantum-chemical CD
calculations in combination with HPLC-CD measurements.[10]
Results and Discussion
[a] Dr. T. Bruhn, Dipl.-Chem. F. Witterauf, Dr. D. C. G. Gçtz, C. T. Grimmer,
Prof. Dr. G. Bringmann
In a first approach we wanted to synthesize such b,b’-diamino-
bisporphyrins by homocoupling of the respective monomers
under Ullmann conditions (Scheme 1). This approach has been
achieved for the synthesis of biaryls and heterobiaryls previ-
ously,[14] however, only for aromatic substrates with amides or
with secondary or tertiary amino functions, not with primary
amino groups. Moreover, no direct Ullmann coupling of por-
phyrins had been described. It was therefore important to pre-
pare the previously unknown 2-bromo-3-aminotetraphenylpor-
phyrin 2 as a promising precursor for coupling reactions.
Institute of Organic Chemistry, University of Wꢀrzburg
Am Hubland, 97074 Wꢀrzburg (Germany)
Fax: (+49)931-31-84755
E-mail: bringman@chemie.uni-wuerzburg.de
[b] Dipl.-Chem. M. Wꢀrtemberger, Prof. Dr. U. Radius
Institute of Inorganic Chemistry, University of Wꢀrzburg
Am Hubland, 97074 Wꢀrzburg (Germany)
[c] Dr. D. C. G. Gçtz
Present address: Bayer Pharma AG
Process Research & Chemical Development
Friedrich-Ebert-Str. 217-333, 42117 Wuppertal (Germany)
The b-positions in monosubstituted tetraphenylporphyrins
(TPPs) are electronically (and sterically) difficult to differentiate,
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201304169.
Chem. Eur. J. 2014, 20, 3998 – 4006
3998
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 2. Attempted coupling of the brominated aminoporphyrin 2a
under Ullmann type coupling conditions leading to the pyrazine-bridged
dimer 4.
boxylate (CuTC) as the coupling agent. However, instead of the
desired C,C-coupling to give 3a, the formation of a twofold
N,C-coupled dimer was observed (Scheme 2). The main prod-
uct was characterized as the pyrazine-fused porphyrin dimer
4a, which was formed in 58% yield. Under the more common
conditions for an Ullmann coupling,[16] viz. treatment of the
nickelated 2-bromo-3-amino-TPP (2a) with Cu powder (freshly
activated) in N,N-dimethylformamide (DMF) at room tempera-
ture, the pyrazine-fused porphyrin dimer 4 was obtained in as
much as 78% yield. The synthesis of 4 by an oxidative cou-
pling using 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ)
had previously been described by Shinokubo et al., but was
formed in low yields, with low selectivity, and was accompa-
nied by formation of the C,C-coupled b,b’-diaminobisporphyrin
3a.[17] The latter coupling protocol had been adapted from
Osuka et al., who had built up meso,meso-coupled porphyrin
dimers under these conditions.[18] More recently, Mandoj and
co-workers synthesized the biszincated derivative of 4 in mod-
erate yields by condensation of a ZnII diaminoporphyrin with
diethyl oxalate or with a ZnII a-dioneporphyrin derivative; how-
ever, in this case two different buildings blocks were re-
quired.[19]
Scheme 1. Planned synthesis of b,b’-diaminobisporphyrins 3 under Ullmann
conditions.
Table 1. Regioselective synthesis of the halogenated b-aminoporphyrin
2.
Entry
Educt
M
NXS
X
Yield [%]
1
2
3
4
5
1a
1a
1a
1b
1b
Ni
Ni
Ni
Cu
Cu
NBS
NCS
NIS
NBS
NCS
Br
Cl
I
Br
Cl
97 (2a)
76 (2b)
decomp.
94 (2c)
92 (2d)
making regioselective b-functionalization of unsymmetric TPPs
very challenging.[15] Thus, the first synthetic goal to establish-
ing a route towards the 2-bromo-3-amino-TPPs 2 from 2-ami-
noporphyrins such as 1 (Table 1), did not initially seem very
promising. Nevertheless, bromination of the common nickelat-
ed 2-amino-TPP (1a) with N-bromosuccinimide (NBS) quantita-
tively gave a single monobrominated compound (entry 1), and
even chlorination of 1a with NCS succeeded in yields of 76%
(entry 2). The structure of 2a, which confirmed the regioselec-
tive b-functionalization of 1a adjacent to the amino group,
was established on the basis of COSY-NMR spectroscopic anal-
ysis (see the Supporting Information). The same halogenation
reactions with 2-amino CuTPP (1b) by using NBS or NCS gave
the respective halides 2c and 2d in yields of more than 90%
(entries 4 and 5), with the same remarkable regioselectivity. All
attempts to isolate the analogous 2-iodo-3-amino-TPP failed
(entry 3), probably due to the sensitivity of the resulting iodi-
nated porphyrins to light and oxidants. The regioselective
high-yield synthesis of the halogenated b-aminoporphyrins 2
was an important prerequisite for the planned bisporphyrin
syntheses, but these novel intermediates are clearly of value
for other functionalization reactions on porphyrin monomers
in general.
As an alternative approach, the synthesis of 3a was attempt-
ed by directed C,C-coupling of the respective aminoporphyrins
2; to this end, a Suzuki protocol was chosen (not shown), as
described by us for the synthesis of b,b’-dimers.[9,10] However,
the additional amino group present in 2a increased the steric
bulk in the b-position, so that instead of borylation of the bro-
minated aminoporphyrin 2a, only hydrodebromination was
observed.
With the results described above in hand, an oxidative ap-
proach was again considered for the synthesis of the desired
b,b’-diaminobisporphyrin 3a. Surprisingly, treatment of the hal-
ogen-free aminoporphyrin 1a with AgPF6, as previously used
for the meso,meso-coupling of ZnDPP[18] (and likewise for C,C-
linkage), exclusively provided the desired b,b’-diaminobispor-
phyrin 3a in high yields. For the success of the reaction the
use of anhydrous solvent and an inert gas atmosphere was
critical, otherwise decomposition of the starting material
occurred.
Attempts to couple the now easily available key intermedi-
ate 2a were performed under mild Ullmann conditions[14] at
room temperature, using N-methyl-2-pyrrolidinone (NMP) as
the solvent and 2.5 equivalents of copper(I) thiophene-2-car-
To further enhance the yields and to gain more insight into
the reasons for the regioselectivity of this reaction, several con-
Chem. Eur. J. 2014, 20, 3998 – 4006
www.chemeurj.org
3999
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 2. C,C- versus N,C-coupling of aminoporphyrin 1a (50 mg,
72.8 mmol; at RT) to give bisporphyrin 3a and/or pyrazine-bridged dimer
4.
Table 3. Directed synthesis of differently metalated b,b’-diaminobispor-
phyrins 3b–d by oxidative C,C-coupling of 1b–d.
Entry Oxidant
[Equiv] Solvent Volume Time Yield of Yield of
[mL]
[h]
3a [%] 4 [%]
1
2
3
4[a]
5
6
7
8
9
10
11
12
13
AgPF6
AgPF6
AgPF6
AgPF6
AgPF6
AgPF6
AgCO2CF3 1.2
AgCO2CF3 1.2
AgCO2CF3 1.2
AgCO2CF3 5.0
FeCl3
FeCl3
FeCl3
FeCl3
AgPF6
AgPF6
AgPF6
1.2
1.2
0.6
0.6
1.2
1.2
CHCl3
CHCl3
CHCl3
THF
THF
THF
CHCl3
CHCl3
THF
15
80
80
80
15
80
15
80
80
80
15
80
80
80
80
80
80
4
54
93
94
20
50
32
34
14
0
0
0
0
6
0
10
16
16
16
16
0
0
0
8
24
16
14
3
36
36
4
68
20
42
200
5
Entry Substrate (M) Oxidant [Equiv] Time [h] Product Yield [%]
1
2
3
4
5
6
7
8
9
1b (Cu)
1b (Cu)
1b (Cu)
1b (Cu)
1c (Pd)
1c (Pd)
1c (Pd)
1d (Zn)
1d (Zn)
AgPF6
AgPF6
FeCl3
FeCl3
AgPF6
FeCl3
FeCl3
AgPF6
FeCl3
0.6
1.2
0.6
1.2
0.6
0.6
1.2
0.6
0.6
16
16
20
4
14
18
4
3b
3b
3b
3b
3c
3c
3c
3d
3d
94
98
91
78
84
87
94
10
30
THF
0
76
1.0
1.0
0.6
1.2
0.6
5.0
1.0
CHCl3
CHCl3
CHCl3
THF
CHCl3
CHCl3
CHCl3
5
2.5 85
3.5 95
160
20
36
30
14
22
0
0
15[b]
16[b]
17[c]
48
48
44
[a] 25 mg (50%) starting material recovered. [b] 200 mL of NEt3 were
added. [c] 500 mL of DIPEA were added.
synthesis of 3b gave comparable conversion rates using AgPF6
or FeCl3 (entries 1–4), the coupling of 1c and 1d was better ac-
complished with FeCl3 (entries 5–9). The ZnII dimer 3d was,
however, only accessible in 30% yield (entry 9); in this case,
a smaller amount (up to 0.6 equiv) of the oxidizing agent was
used because the ZnII derivative 1d was highly sensitive to oxi-
dation and/or light and easily decomposed during the
reaction.
ditions were tested (Table 2). All the parameters were found to
have significant influence on the outcome of the reaction. In
general, higher amounts (one equivalent or more) of the oxi-
dizing agents clearly decreased the reaction times but also
gave slightly lower yields, probably due to decomposition by
over-oxidation. The best yields (up to 95%) were achieved
when 0.6 equivalents of AgPF6 or FeCl3 (entries 3 and 13) were
used in a diluted chloroform solution. With tetrahydrofuran
(THF) as the solvent the N,C-coupled annulated system 4 was
formed as a by-product (entries 4 and 6). Compound 4 was
also found under all tested reaction conditions when
AgCO2CF3 was used (entries 7–10). With THF as the solvent it
was even formed as the main product, albeit with low yield
(16%; entries 9 and 10).
Crystals of 3a and 3c that were suitable for X-ray structure
analysis were obtained when the respective solution in chloro-
form was layered with n-hexane (Figure 1). Both axially chiral
Clearly, the solvent had a large impact on the outcome of
the reaction. In THF, generally, both coupling types were ob-
served. In chloroform, however, C,C-coupling was found almost
exclusively, whereas addition of NEt3 or Hꢀnig’s base (DIPEA)
to the chloroform again led either to N,C-coupling (Table 2, en-
tries 15 and 16) or to both coupling types (entry 17), when
using AgPF6 as the oxidant. Interestingly, just half the stoichio-
metric amount of oxidant (e.g., 0.6 equiv) was sufficient to
obtain the high coupling yields. It is possible that, under these
conditions, only the radical cation of the dimer was formed
during the synthesis and the fully oxidized dimer was formed
during workup. Investigations on this topic are ongoing.
Having optimized the reaction conditions for the dimeriza-
tion of metalated aminoporphyrin 1a, several derivatives of 3
differing in their central metals were synthesized (Table 3) start-
ing from the respective metalated aminoporphyrins 1b–d,
which were prepared according to published procedures.[20,21]
For the CuII dimer 3b and the PdII dimer 3c the reactions suc-
ceeded with high yields above 90% (entries 1–7). Whereas the
Figure 1. ORTEP plot (thermal ellipsoids at 50% probability) of dimers 3a
(top) and 3c (below), arbitrarily only the M-atropo-enantiomers are shown.
The outer phenyl substituents and all hydrogen atoms have been omitted
for clarity.
Chem. Eur. J. 2014, 20, 3998 – 4006
www.chemeurj.org
4000
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
compounds crystallized as a racemic mixture in the space
group P1. In contrast to a previously published crystal struc-
ture of 3a,[17] the crystals obtained in this case contained a mol-
ecule of water instead of chlorobenzene in the asymmetric
unit of the crystal. A comparison of our crystal structure with
the published structure revealed some significant differences,
which proved to stem from an imprecise description in that ar-
ticle.[17] A direct comparison with the crystal structure data
from the CCDC showed fewer differences. Therefore, prior to
comparing the structure of 3a with that of 3c, we here first
describe the most important (revised) structural features of 3a.
Remarkably, the geometries of the two subunits of the nicke-
lated porphyrin 3a were significantly different, and thus the
molecule clearly deviated from C2 symmetry in the crystal
structure. One porphyrin moiety showed a ruffled conforma-
tion, whereas the second had a distorted but clearly saddle-
shaped ring structure. The phenyl substituents directly adja-
cent to the axis displayed clear p-stackings with the neighbor-
ing macrocycles, with average distances between the respec-
tive pyrroles and phenyls of 3.486 and 3.265 ꢂ. This is con-
Scheme 3. AgI-mediated directed cross-coupling to the b,meso-porphyrin
dimer 6.
1
firmed by the clearly shifted H NMR peaks of the correspond-
ing phenyl substituents. The angle at the central axis, when
defined as that between the porphyrin subunits, was found to
be about 548, but, when defined as the dihedral angle be-
tween the directly connected pyrroles, was 768.
different, otherwise meso,meso-coupling would also be expect-
ed. Such directed oxidative cross-couplings of aromatic com-
pounds had been reported before.[22] The mixed coupling of
1a and 5a was clearly influenced by the metal atom in the
meso-building block. Whereas the zincated triarylporphyrin 5a
reacted very well, no reaction to a b,meso-dimer was observed
in the case of the nickelated derivative 5b, for which only the
b,b’-diaminobisporphyrin 3a was formed. Similar observations
had previously been made for meso,meso-couplings of diphe-
nylporphyrins (DPP) with NiII as the central atom.[18] A simple
explanation was found by analyzing the orbital coefficients of
NiDPP and ZnDPP and their averaged localized ionization
energy (ALIE) surfaces[23] (see the Supporting Information);
thus, only in the zincated educt were the HOMO and an ioniza-
tion energy minimum clearly located at the meso-positions, so
that nucleophilic attack from the meso-position could only
occur for this DPP.
In the bispalladated dimer 3c, the porphyrin subunits were
nearly planar and only slightly saddled, and the phenyl sub-
stituents at the b,b’-axis were no longer centered above the
macrocycles as in 3a, but rather peripheral. Nevertheless, the
average distances of these phenyl groups to the macrocycle of
3.428 and 3.388 ꢂ evidenced a clear p-stacking (also visible in
1
the H NMR spectra). The angle between the porphyrin moiet-
ies of 3c was 608 and the dihedral angle between the pyrrole
units at the axis was 848, thus both angles were significantly
larger than in 3a. The sizes of the angles at the central axis are
reflected in the UV spectra of the b,b’-diaminobisporphyrins 3,
which we discuss below.
Synthesis of b,meso-dimers and b,meso,b-trimers
The successfully established b,meso-linkage encouraged us
to use this reaction principle to elaborate an easy access even
to b,meso,b-trimers. Such achiral and chiral trimers, as previ-
ously prepared by a twofold Suzuki coupling,[24] are of interest
because, even when achiral, they can occur as stable cis/trans-
atropo-diastereoisomers. For the synthesis of the envisaged
b,meso,b-coupled porphyrin trimer 9 (Scheme 4), aminopor-
phyrin 1a was used as the b-building block and ZnDPP (7) as
the meso-unsubstituted precursor. As expected from the previ-
ous results, again no meso,meso-homocoupling was observed
in this reaction. In addition to the symmetric b,b’-diaminobis-
porphyrin 3a (12%) and the cross-coupled intermediate
b,meso-dimer 8 (29%), the desired b,meso,b-trimer 9 (bTAP-
5,15-A2-bTAP type[24]) was found in a yield of 10%. The latter
was isolated as a single diastereomer, most probably the steri-
cally less demanding trans-isomer, however, this could not be
proven by NMR spectroscopic analysis due to the symmetry of
the molecule.[24]
Because the conditions used for oxidative b,b’-couplings were
similar to those used for the previously described[18] meso,me-
so-couplings, the construction of a directed b,meso-linkage
seemed to be a rewarding task. Indeed, treatment of an equi-
molar mixture of the nickelated aminoporphyrin 1a and a tri-
arylporphyrin (viz. the zincated tris(3,5-di-tert-butylphenyl)por-
phyrin 5a) with AgPF6 (Scheme 3) led to the formation of the
symmetrical b,b’-diaminobisporphyrin 3a in 36% yield and the
b,meso-coupled dimer 6 (A3-bTAP type[24]) in 15% yield. Inter-
estingly, no meso,meso-dimer was found. The use of an excess
(3 equiv) of the meso-precursor 5a increased the yield of the
unsymmetrical b,meso-coupled compound 6 to 45%, whereas
the b,b’-diaminobisporphyrin 3a was isolated in only trace
amounts; the remaining starting material 5a (2 equiv) was re-
isolated nearly completely.
It seemed that 1a was not only oxidized more easily than
5a, but also that the nucleophilicities of the two educts were
Chem. Eur. J. 2014, 20, 3998 – 4006
www.chemeurj.org
4001
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 5. Possible mechanism for the formation of 3 and 4.
Scheme 4. Synthesis of the b,meso,b-coupled trimer 9 and the intermediate
dimer 8.
Mechanism
The results described above permit several conclusions to be
drawn regarding the mechanism of the oxidative coupling. In
the initial halogenation step from 1 to 2, the amino group al-
ready has a strong directing influence, with the electrophile ex-
clusively attacking the carbon atom next to the carbon bearing
the amino function (i.e., C-3), showing that this enamine
carbon had by far the highest nucleophilicity in the 2-amino-
porphyrins 1. It has been reported that oxidative coupling in
the meso-position of DPP takes place via a DPP radical cation,
with subsequent C,C-bond formation by its electrophilic attack
to an unoxidized porphyrin, and final oxidation of the primary
coupling product to give the meso,meso-dimer.[18,25] In the case
of 2-aminoporphyrin 1, the reaction of the corresponding radi-
cal cation should occur exclusively at C-3, that is, the b-carbon
atom next to the carbon carrying the amino function. Howev-
er, as shown above, not only C,C-, but also N,C-coupling can
occur upon oxidation, that is, there are two positions from
which the electrophilic attack of the 2-aminoporphyrin radical
cation can take place. These positions can be rationalized by
the mesomeric structures of the intermediate radical cation I
(Scheme 5), which show that spin density can be expected at
the amino nitrogen (Scheme 5, Ib) and at the enamine carbon,
that is, in the 3-position (Scheme 5, Ic, Id).
Figure 2. Calculated spin densities of the intermediate radical cation I a) in
the absence or b) in the presence of a chloride anion (green sphere), and
c) average local ionization energy (ALIE) surface of the nucleophilic educt
1a.
wB97X-D/6-31G* (6-311G* for Ni) method. Surprisingly, the spin
density was mainly present at the amino group, and none was
found at C-3 (Figure 2a). Because the experimental results had
shown an influence of the solvent, the anion of the oxidizing
agent, and of the additives to the oxidative coupling, their
roles were investigated by further calculations. Including the
To validate these conclusions and to further clarify the
mechanism, the spin density in the radical cation I of 1a was
determined by quantum-chemical calculations by using the
Chem. Eur. J. 2014, 20, 3998 – 4006
www.chemeurj.org
4002
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
solvents THF and chloroform in the calculations, as expected,
showed no influence on the spin densities, whereas computa-
tions in the presence of an additional chloride anion in the
proximity of the radical revealed spin density at the b-carbon
C-3 (Figure 2b).
Soret band, as clearly observed in most b,b’-bisporphyrins
without amino functions, did not occur in the UV spectra of
the nickelated or palladated dimers 3a and 3c. The absorp-
tions of the metalated porphyrin dimers 3b–d were quite simi-
lar to that of 3a, although the PdII dimer 3c had a slightly less
broadened Soret band and the CuII dimer 3b showed a small
shoulder at 420 nm and a lower intensity of the Soret band.
The shoulder in the Soret band was also observed for 3d and
may hint at a Davydov splitting in these 2-aminoporphyrin
dimers. As in previous cases,[28,29] the angle between the por-
phyrin subunits had a strong impact on the UV spectrum of
b,b’-bisporphyrins: the smaller this angle, the more broadened
the expected Soret band. The crystal structures of 3a and 3c
explained the observed differences in the Soret band: 3c had
a larger angle than 3a and thus it displayed a narrower ab-
sorption band.
To assess the nucleophilicity of the second reaction partner,
the nonoxidized aminoporphyrin 1a, the average local ioniza-
tion energy (ALIE)[23] surface was calculated from the wB97X-D/
6-31G*-optimized structures. These results were in agreement
with the experimental findings, confirming clearly that the
only position expected to undergo electrophilic attack was the
carbon atom next to the carbon having the amino function,
that is, C-3 (Figure 2c).
For the outcome of the reaction merely the distribution of
the spin density was decisive: When the spin was exclusively
localized on the nitrogen atom (Scheme 5, Ib and Figure 2a),
only the N,C-coupling was possible, whereas in the other cases
(Scheme 5, Ic, Id and Figure 2b) the thermodynamically pre-
ferred C,C-coupling occurred. The ALIE surfaces showed no
clear dependence of the nucleophilicities on the different type
of central metal atoms in the aminoporphyrins 1, in contrast to
the cases of NiDPP and ZnDPP (see the Supporting Informa-
tion).
As described above, the additional amino group should in-
crease the rotational barrier at the central axis of the b,b’-cou-
pled dimers 3, making it possible to resolve the presumed rac-
emic mixtures of 3 by HPLC on a chiral phase. After intense
optimization of the chromatographic system, the resolution of
all dimers 3a–d succeeded with a Chiralpak IA phase (Daicel
Chemical Industries, Ltd.; 3a–c: CH2Cl2/n-hexane 20:80; flow:
1.0 mLminÀ1; 3d: CH2Cl2/n-hexane 40:60; flow: 1.2 mLminÀ1),
permitting measurement of full CD spectra of the enantiomers
by HPLC-CD in the stopped-flow mode. This made it rewarding
to likewise calculate the CD curves for the enantiomers of 3a
quantum-chemically and to compare with these experimental
CD spectra, to assign their absolute configurations.
The fact that treatment of aminoporphyrin 1a with AgPF6 in
THF also gave rise to a small amount of N,C-coupling, might
be explained by the known[26] complex formation of the silver
salt with THF. This complexation has been reported to lower
the oxidation potential of the silver salt.[26] The absence of the
anion as a consequence of the complexation significantly
changes the position of the spin density in the radical cation I
(as shown in Figure 2a and 2b for I with and without a chloride
anion). Addition of NEt3 to AgPF6 in chloroform had a similar
effect and, thus, again N,C-coupling occurred. It is well-known
that oxidative aromatic coupling is very sensitive to the elec-
tron density of the substrate[27] and in the case of the amino-
porphyrins it is clearly possible to easily change the spin densi-
ty by varying the solvent and/or the oxidizing agent.
To find a suitable computational method, we first calculated
the UV spectra with different density functionals. When com-
puting the UV spectra of the monomer 1a and of the dimer
3a using TDCAM-B3LYP in combination with the 6-31G* (for Ni
6-311G*) basis set, severe problems were found in reproducing
the experimental UV spectrum for the monomer, although this
method had been used successfully for the computation of ex-
cited states of porphyrins before.[28,30] The calculated spectrum
had a general blueshift of 47 nm and two excitations (calculat-
ed at 397 and 385 nm) were observed that had an even higher
intensity than the Soret band. Electron-density difference plots
showed that these excitations had a charge-transfer (CT) char-
acter, with transitions from the amino group and the corre-
sponding pyrrole to the porphyrin macrocycle, which might
explain the problems of the TDDFT method.
Spectroscopic properties
The UV spectrum of diaminobisporphyrin 3a in CH2Cl2
(Figure 3) had a broadened Soret band at 421 nm, which was
red-shifted by 10 nm compared with that of the aminopor-
phyrin monomer 1a (411 nm). The Davydov splitting of the
To gain a deeper insight into whether these CT excitations
really exist and if at least their calculated energies were cor-
rect, RI-SCS-CC2 calculations were performed. Although this
coupled-cluster method showed a high hypsochromic shift of
50 nm, the overall agreement with the experiment was much
better. The additional transitions found in the TDDFT computa-
tions also occurred when applying the RI-SCS-CC2 method at
comparable energies but with a much lower intensity in the
absorption spectrum (see the Supporting Information). This
high-level method showed that the CT character of the excita-
tions was slightly overestimated by TDCAM-B3LYP but the gen-
eral level of agreement of the excitation energy with the RICC2
result was acceptable. However, the TDDFT oscillator strengths
Figure 3. UV spectra of the dimers 3a–d in CH2Cl2.
Chem. Eur. J. 2014, 20, 3998 – 4006
www.chemeurj.org
4003
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
were highly overestimated and
this problem was also observed
in the dimer calculations, for
which RICC2 computations were
not feasible. Thus, the determi-
nation of a reliable UV shift,
simply by comparison of the ex-
perimental UV spectra with the
calculated spectra, was difficult.
This problem was solved by
identifying the calculated Soret
bands by analyzing the main
configurations of the excitations.
Transitions from HOMO and
HOMO-1 to LUMO and LUMO+
1 contribute to a high extent
to the Soret band, as is known
for nonaminated porphyrins,
making it easy to find the cor-
rect excitations and to deter-
mine a reliable UV shift.
Figure 4. Determination of the absolute configuration of the two enantiomers of dimer 3a by comparison of their
CD spectra measured online with those calculated quantum-chemically.
The
TDCAM-B3LYP/6-31G*//
wB97X-D/6-31G* (6-311G* for Ni) calculations of the M-config-
ured diaminobisporphyrin 3a yielded a CD spectrum that gave
a very good match with the experimental spectrum of the
faster eluting enantiomer, whereas the curve of the more
slowly eluting compound fit very well with the spectrum calcu-
lated for the P-enantiomer (Figure 4). Clearly, the calculated ro-
tational-strength values did not suffer from the problem identi-
fied for the oscillator-strength values.
In a similar way, quantum-chemical calculations were per-
formed for the dimers with PdII, CuII, or ZnII as the central
metal atoms. All CD curves of the faster eluting peaks showed
nice agreement with the CD spectra calculated for the respec-
tive M-configured enantiomers, whereas the CD curves of the
more slowly eluting peaks matched nearly perfectly with those
calculated for the P-enantiomers (see Figure 5 and Figure S1–
S3 in the Supporting Information). The enantiomeric similarity
indices[31] (DESI, calculated in the range from 350 to 500 nm)
were above 90% for 3a–c and above 60% for 3d, showing
the high reliability of these configurational assignments by
using CD calculations. Interestingly, the metal centers had
a substantial impact on the intensity of the CD curves. The pal-
ladated dimer 3c showed the strongest signal and the nicke-
lated bisporphyrin 3a the weakest, whereas the CD intensities
of 3b (M=Cu) and 3d (M=Zn) were in between.
Figure 5. Experimental CD spectra (offline) of the faster eluting M-configured
enantiomers of 3a–d.
the TDB3LYP or TDBHLYP calculated spectra was impossible
(see the Supporting Information).
All attempts to conduct TDB3LYP and TDBHLYP calculations
did not yield a satisfactory agreement between the calculated
and the experimental CD curves. At first glance the computed
UV curves showed a better match because the ‘CT’ excitations
mentioned above had a significantly lower absorption, but
their excitation energies were much lower (closer to the Q
bands) in the case of the TDB3LYP calculation and much
higher in the case of TDBHLYP (even higher than the Soret
band). This eventually corrupted the computed CD spectra to
such an extent that a reliable determination of the absolute
configuration by comparison of the experimental spectra with
The CD spectra of the M-configured b,b’-diaminobisporphyr-
ins 3 were nearly mirror images of the curves of the M-config-
ured and rotationally stable b,b’-bisporphyrins without amino
substituents[10] (see the Supporting Information). Thus, the
common method used to elucidate the absolute configuration
by comparison of the CD spectra of new compounds with
those of structurally closely related compounds would have
led to wrong results in this case, highlighting the importance
of quantum-chemical CD calculations.
To gain more detailed insight into the stereochemical stabili-
ty of the axis, the resolved pure enantiomers of 3a were inves-
Chem. Eur. J. 2014, 20, 3998 – 4006
www.chemeurj.org
4004
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tropolarimeter for the online-CD investigations (scanning rate:
200 nmminÀ1, bandwidth: 1.0 nm, response time: 0.25 s, number
of accumulations: 3). All enantiomeric resolutions were carried out
at RT with a constant flow rate using an isocratic solvent system
consisting of dichloromethane and n-hexane. Semipreparative res-
olutions of the racematic mixtures were performed with a Jasco
HPLC system (pump PU1580, gradient unit LG-2080–04, degasser
DG-2080–54, UV detector MD-1510) equipped with a Chiralpak IA
column (Chiral Technologies Europe, 10ꢃ250 mm, 5 mm) as the
chiral phase.
tigated by HPLC at different temperatures after different inter-
vals, but even at 1008C no rotation was observed and at
higher temperatures the dimers decomposed. This already
showed that, in comparison to the (stereochemically unstable)
b-unsubstituted nickelated porphyrin dimers,[10] the rotational
barrier had been significantly increased by the additional b-
amino groups.
Conclusion
Unless otherwise noted, materials obtained from commercial sup-
pliers were used without further purification except for N-bromo-
succinimide, which was recrystallized from H2O. Anhydrous CHCl3
was obtained by distillation from CaH2. Column chromatography
was performed on silica gel (0.063–0.2 mm). Thin-layer chromatog-
raphy was carried out on aluminum sheets coated with fluores-
cence-active silica gel 60 F254 (Merck).
A series of novel metalated b,b’-coupled diaminobisporphyrins
3 were synthesized in excellent yields and high regioselectivi-
ties by the simple methodology of an oxidative coupling of 2-
aminoporphyrins 1. These are most valuable substrates be-
cause they can be halogenated in a highly regioselective
manner. In a similar way, the amino group has a strong direct-
ing influence in oxidative coupling reactions, which can be se-
lectively directed towards N,C-coupling, leading to pyrazine-
fused dimers 4, or to C,C-coupling to give axially chiral bispor-
phyrins 3. This regioselectivity can be convincingly rationalized
by ALIE-surface and spin-density calculations.
Computational details
All optimizations were performed by using the wB97X-D[32] func-
tional in combination with the 6–31G*[33] basis set for C, H, and O
atoms, 6–311G*[34] for Ni, Cu, Zn, and Cl, and SDD[35] (and pseudo
potentials) for Pd. The conformers arising from different dihedral
angles at the phenyl axes were ignored as they had no impact on
the CD behavior. Thus, for most of the dimers only one significant
conformation was found. Dimer 3a was the only exception, with
two possible conformers, for which the spectra were added up
after a Boltzmann-statistical weighting.[36] For the optimizations of
the radical cations, solvent effects were considered (CPCM,[37] sol-
vent=chloroform). Subsequently, TDCAM-B3LYP[38] calculations
were done using the same combinations of basis sets and CPCM
(solvent=dichloromethane) for solvent effects. In case of the RI-
SCS-CC2[39] calculations, def2-SV(P)[40] (def2-TZVP[40] for Ni) was uti-
lized. All computations were done with Gaussian09[41] except for
the coupled-cluster approach, which was performed with Turbo-
mole.[42] Visualization of the spin densities was carried out with
Avogadro[43] in combination with PovRay. ALIE surfaces were calcu-
lated using Multiwfn 3.1,[44] and visualized with VMD[45] and PovRay.
Processing of the TD calculations and comparison with the experi-
mental spectra was done with SpecDis.[31,46] The following UV
shifts[36] and s values were used: 32 nm and 0.14 eV for 3a, 45 nm
and 0.1 eV for 3b, 40 nm and 0.11 eV for 3c, 47 nm and 0.16 eV for
3d.
The b,b’-coupled, intrinsically chiral aminoporphyrin dimers
3 were found to be rotationally stable at the central axis. Reso-
lution of their enantiomers was achieved by HPLC with a chiral
phase. By HPLC-CD coupling, in combination with quantum-
chemical CD calculations, the absolute configurations of the
enantiomers were determined. Investigations on the rotational
barrier showed the high impact of the additional amino
groups. Whereas the nickelated b,b’-bisporphyrins without
amino groups have been reported to racemize at room tem-
perature,[10] the pure enantiomers of the novel diamino-substi-
tuted analogue 3a were found to be stereochemically stable.
b,b’-Diaminobisporphyrins might find use as ligands for
metal coordinations as described for N-inverted porphyrin
dimers[11] or even as catalysts in metal-assisted coupling reac-
tions with the porphyrin dimer as a kind of “super BINAM”.[13]
Experimental Section
General experimental procedures
UV/Vis spectra were obtained with a Cary 50 spectrophotometer
(Varian). IR spectra were recorded with a JASCO FTIR-410 spectrom-
eter. H and 13C NMR spectra were taken with Bruker Avance 400
1
Acknowledgements
or DMX 600 (400 and 600 MHz) spectrometers using CDCl3 as inter-
nal standard (CDCl3; d=7.26 and 77.00 ppm, respectively). HRMS
(ESI) spectra were obtained with a microTOF-focus mass spectrom-
eter (Bruker Daltonik GmbH) equipped with an APCI ion source
(Agilent G1947–60101). Due to the isotopic distribution over
a broad m/z region caused by zinc and palladium, the signal of
monoisotopic signals was too small in intensity for some com-
pounds for an accurate mass measurement. In these cases, typical-
ly the most intense signal (X+n) of this isotopic distribution was
taken as described and compared to the respective calculated
value. Enantiomeric resolutions on an analytical scale were per-
formed with a Jasco HPLC system (pump PU1580, gradient unit
LG-980–02S, degasser DG-2080–53, UV detector MD-2010Plus)
equipped with a Chiralpak IA column (Chiral Technologies Europe,
4.6ꢃ250 mm) as the chiral phase and coupled to a Jasco 715 spec-
The authors gratefully acknowledge Johannes Ahrens and Prof.
Dr. Martin Brçring (Institute of Inorganic and Analytical Chemis-
try, Technical University of Braunschweig, Germany) for the
measurement and the interpretation of the CV spectra.
This work was supported by the Degussa-Stiftung and the
Studienstiftung des deutschen Volkes e.V. (fellowships to
Dr. Daniel C. G. Gçtz).
Keywords: atropisomerism · configuration determination ·
oxidative coupling
calculations
·
porphyrins
·
quantum-chemical
Chem. Eur. J. 2014, 20, 3998 – 4006
www.chemeurj.org
4005
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[1] E. Rose, B. Andrioletti, S. Zrig, M. Quelquejeu-Ethꢄve, Chem. Soc. Rev.
2005, 34, 573–583.
[2] J.-C. Marchon, R. Ramasseul, in The Porphyrin Handbook, Vol. 11: Bioinor-
ganic and Bioorganic Chemistry (Eds.: K. M. Kadish, K. M. Smith, R. Gui-
lard), Academic Press, San Diego, 2003, pp. 75–132.
[29] M. Kullmann, A. Hipke, P. Nuernberger, T. Bruhn, D. C. G. Gçtz, M. Sekita,
D. M. Guldi, G. Bringmann, T. Brixner, Phys. Chem. Chem. Phys. 2012, 14,
8038–8050.
[30] Z.-L. Cai, M. J. Crossley, J. R. Reimers, R. Kobayashi, R. D. Amos, J. Phys.
Chem. B 2006, 110, 15624–15632.
[3] H. Ogashi, T. Mizutani, T. Hayashi, Y. Kuroda, in The Porphyrin Handbook,
Vol. 6: Applications: Past, Present, and Future (Eds.: K. M. Kadish, K. M.
Smith, R. Guilard), Academic Press, San Diego, 2000, pp. 279–340.
[4] H. Ogoshi, T. Mizutani, Acc. Chem. Res. 1998, 31, 81–89.
[5] A. C. Tomꢅ, A. M. S. Silva, I. Alkorta, J. Elguero, J. Porphyrins Phthalocya-
nines 2011, 15, 1–28.
[31] T. Bruhn, A. Schaumlçffel, Y. Hemberger, G. Bringmann, Chirality 2013,
25, 243–249.
[32] J.-D. Chai, M. Head-Gordon, Phys. Chem. Chem. Phys. 2008, 10, 6615–
6620.
[33] a) P. C. Hariharan, J. A. Pople, Theor. Chim. Acta 1973, 28, 213–222;
b) W. J. Hehre, R. Ditchfield, J. A. Pople, J. Chem. Phys. 1972, 56, 2257–
2261.
[34] a) A. J. H. Wachters, J. Chem. Phys. 1970, 52, 1033–1036; b) A. D.
McLean, G. S. Chandler, J. Chem. Phys. 1980, 72, 5639–5648.
[35] a) T. H. Dunning, Jr., P. J. Hay, in Modern Theoretical Chemistry, Vol. 3:
Methods of Electronic Structure Theory (Ed.: I. H. F. Schaefer), Plenum,
New York, 1977, pp. 1–27; b) D. Andrae, U. Hꢈußermann, M. Dolg, H.
Stoll, H. Preuß, Theor. Chim. Acta 1990, 77, 123–141.
[6] Q. Ouyang, Y.-Z. Zhu, Y.-C. Li, H.-B. Wie, J.-Y. Zheng, J. Org. Chem. 2009,
74, 3164–3167.
[7] N. Yoshida, A. Osuka, Tetrahedron Lett. 2000, 41, 9287–9291.
[8] P. J. Chmielewski, B. Durlej, M. Siczek, L. Szterenberg, Angew. Chem.
2009, 121, 8892–8895; Angew. Chem. Int. Ed. 2009, 48, 8736–8739.
[9] G. Bringmann, S. Ruedenauer, D. C. G. Goetz, T. A. M. Gulder, M. Reichert,
Org. Lett. 2006, 8, 4743–4746.
[10] G. Bringmann, D. C. G. Gçtz, T. A. M. Gulder, T. H. Gehrke, T. Bruhn, T.
Kupfer, K. Radacki, H. Braunschweig, A. Heckmann, C. Lambert, J. Am.
Chem. Soc. 2008, 130, 17812–17825.
[11] a) P. J. Chmielewski, M. Siczek, L. Szterenberg, Inorg. Chem. 2011, 50,
6719–6736; b) P. J. Chmielewski, L. Szterenberg, M. Siczek, Chem. Eur. J.
2011, 17, 1009–1020.
[36] G. Bringmann, T. Bruhn, K. Maksimenka, Y. Hemberger, Eur. J. Org. Chem.
2009, 2717–2727.
[37] M. Cossi, N. Rega, G. Scalmani, V. Barone, J. Comput. Chem. 2003, 24,
669–681.
[38] T. Yanai, D. P. Tew, N. C. Handy, Chem. Phys. Lett. 2004, 393, 51–57.
[39] a) C. Hꢈttig, F. Weigend, J. Chem. Phys. 2000, 113, 5154–5161; b) O.
Christiansen, H. Koch, P. Jørgensen, Chem. Phys. Lett. 1995, 243, 409–
418.
[12] S. Richeter, C. Jeandon, J.-P. Gisselbrecht, R. Graff, R. Ruppert, H. J.
Callot, Inorg. Chem. 2004, 43, 251–263.
[13] S. G. Telfer, R. Kuroda, Coord. Chem. Rev. 2003, 242, 33–46.
[14] S. Zhang, D. Zhang, L. S. Liebeskind, J. Org. Chem. 1997, 62, 2312–2313.
[15] Some of the rare known examples of a regioselective b-functionaliza-
tion (e.g., by using the directing effect of a nitro group) can be found
in: a) M. J. Crossley, L. G. King, I. A. Newsom, C. S. Sheehan, J. Chem. Soc.
Perkin Trans. 1 1996, 2675–2684; b) K. M. Shea, L. Jaquinod, R. G.
Khoury, K. M. Smith, Chem. Commun. 1998, 759–760; c) L. Jaquinod,
R. G. Khoury, K. M. Shea, K. M. Smith, Tetrahedron 1999, 55, 13151–
13158.
[16] F. Monnier, M. Taillefer, Angew. Chem. 2009, 121, 7088–7105; Angew.
Chem. Int. Ed. 2009, 48, 6954–6971.
[17] M. Akita, S. Hiroto, H. Shinokubo, Angew. Chem. Int. Ed. 2012, 51, 2894–
2897.
[18] A. Osuka, H. Shimidzu, Angew. Chem. 1997, 109, 93–95; Angew. Chem.
Int. Ed. Engl. 1997, 36, 135–137.
[19] F. Mandoj, S. Nardis, R. Pudi, L. Lvova, F. R. Fronczek, K. M. Smith, L.
Prodi, D. Genovese, R. Paolesse, Dyes Pigm. 2013, 99, 136–143.
[20] J. E. Baldwin, M. J. Crossley, J. DeBernardis, Tetrahedron 1982, 38, 685–
692.
[21] H. K. Hombrecher, V. M. Gherdan, S. Ohm, J. A. S. Cavaleiro, M. da GraÅa,
P. M. S. Neves, M. de Fꢆtima Condesso, Tetrahedron 1993, 49, 8569–
8578.
[22] a) T. Dohi, M. Ito, I. Itani, N. Yamaoka, K. Morimoto, H. Fujioka, Y. Kita,
Org. Lett. 2011, 13, 6208–6211; b) E. Faggi, R. M. Sebastiꢆn, R. Pleixats,
A. Vallribera, A. Shafir, A. Rodrꢇguez-Gimeno, C. Ramꢇrez de Arellano, J.
Am. Chem. Soc. 2010, 132, 17980–17982.
[40] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297–3305.
[41] Gaussian 09, Revision B.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Men-
nucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. J. A. Montgomery, J. E. Peralta, F. Ogliaro, M.
Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T. Keith, R.
Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S.
Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Strat-
mann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L.
Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dan-
nenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz,
J. Cioslowski, D. J. Fox, Gaussian, Inc., Wallingford CT, 2010.
[42] TURBOMOLE Version 6.2, R. Ahlrichs, M. K. Armbruster, R. A. Bachorz, M.
Bꢈr, H.-P. Baron, R. Bauernschmitt, F. A. Bischoff, S. Bçcker, N. Crawford,
P. Deglmann, F. Della Sala, M. Diedenhofen, M. Ehrig, K. Eichkorn, S. El-
liott, F. Furche, A. Glçß, F. Haase, M. Hꢈser, C. Hꢈttig, A. Hellweg, S. Hçf-
ener, H. Horn, C. Huber, U. Huniar, M. Kattanek, W. Klopper, A. Kçhn, C.
Kçlmel, M. Kollwitz, K. May, P. Nava, C. Ochsenfeld, H. ꢉhm, M. Pabst, H.
Patzelt, D. Rappoport, O. Rubner, A. Schꢈfer, U. Schneider, M. Sierka,
D. P. Tew, O. Treutler, B. Unterreiner, M. von Arnim, F. Weigend, P. Weis,
H. Weiss, N. Winter, TURBOMOLE GmbH, Karlsruhe, 2010.
[43] M. Hanwell, D. Curtis, D. Lonie, T. Vandermeersch, E. Zurek, G. Hutchi-
son, J. Cheminf. 2012, 4, 17.
[23] J. J. Brown, S. L. Cockroft, Chem. Sci. 2013, 4, 1772–1780.
[24] D. C. G. Gçtz, T. Bruhn, M. O. Senge, G. Bringmann, J. Org. Chem. 2009,
74, 8005–8020.
[25] K. M. Smith, G. H. Barnett, B. Evans, Z. Martynenko, J. Am. Chem. Soc.
1979, 101, 5953–5961.
[44] T. Lu, F. Chen, J. Comput. Chem. 2012, 33, 580–592.
[45] W. Humphrey, A. Dalke, K. Schulten, J. Mol. Graph. 1996, 14, 33–38.
[46] SpecDis, Version 1.61, T. Bruhn, A. Schaumlçffel, Y. Hemberger, G. Bring-
mann, University of Wꢀrzburg, Wꢀrzburg, 2013.
[26] N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 877–910.
[27] M. Grzybowski, K. Skonieczny, H. Butenschçn, D. T. Gryko, Angew. Chem.
2013, 125, 10084–10115.
Received: October 25, 2013
[28] S. Cho, M.-C. Yoon, J. M. Lim, P. Kim, N. Aratani, Y. Nakamura, T. Ikeda, A.
Osuka, D. Kim, J. Phys. Chem. B 2009, 113, 10619–10627.
Published online on February 26, 2014
Chem. Eur. J. 2014, 20, 3998 – 4006
www.chemeurj.org
4006
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim