Surface-Guided Formation of an Organocobalt Complex

Article abstract of DOI:10.1002/anie.201600567

Organocobalt complexes represent a versatile tool in organic synthesis as they are important intermediates in Pauson-Khand, Friedel-Crafts, and Nicholas reactions. Herein, a single-molecule-level investigation addressing the formation of an organocobalt c

Full text of DOI:10.1002/anie.201600567

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201600567
German Edition: DOI: 10.1002/ange.201600567
Surface Chemistry
Surface-Guided Formation of an Organocobalt Complex
Peter B. Weber, Raphael Hellwig, Tobias Paintner, Marie Lattelais, Mateusz Paszkiewicz,
Pablo Casado Aguilar, Peter S. Deimel, Yuanyuan Guo, Yi-Qi Zhang, Francesco Allegretti,
Anthoula C. Papageorgiou, Joachim Reichert, Svetlana Klyatskaya, Mario Ruben,
Johannes V. Barth, Marie-Laure Bocquet,* and Florian Klappenberger*
[
4]
Abstract: Organocobalt complexes represent a versatile tool in
organic synthesis as they are important intermediates in
Pauson–Khand, Friedel–Crafts, and Nicholas reactions.
Herein, a single-molecule-level investigation addressing the
formation of an organocobalt complex at a solid–vacuum
interface is reported. Deposition of 4,4’-(ethyne-1,2-diyl)di-
benzonitrile and Co atoms on the Ag(111) surface followed by
annealing resulted in genuine complexes in which single
Co atoms laterally coordinated to two carbonitrile groups
undergo organometallic bonding with the internal alkyne
moiety of adjacent molecules. Alternative complexation sce-
narios involving fragmentation of the precursor were ruled out
by complementary X-ray photoelectron spectroscopy. Accord-
ing to density functional theory analysis, the complexation with
the alkyne moiety follows the Dewar–Chatt–Duncanson model
for a two-electron-donor ligand where an alkyne-to-Co
donation occurs together with a strong metal-to-alkyne back-
donation.
CoÀCo bond. This important catalytic intermediate acts as
[5]
a promoter for the Friedel–Crafts acylation and mediates
the [2+2+1] co-cyclization, known as the Pauson–Khand
The closely related cobalt-complexed propargyl
cations are central ingredients in the Nicholas reaction.
The exploration of metal–ligand interactions at well-
defined surfaces presents an important alternative approach
complementing classic solution-based synthetic procedures.
This strategy not only provides deep insight into the under-
lying mechanisms at the single-molecule level by rendering
possible the application of surface-science techniques, but
also triggers the emergence of novel complexation phenom-
ena unique to two-dimensional environments. Notably,
a series of metal–organic coordination networks were cre-
ated, opening up prospects towards new functional surfaces
with technological potential in a variety of fields and
applications, such as host–guest chemistry,
and molecular nanomagnetism.
Herein we employ the so-called STM + XS approach
[
2,6]
reaction.
[3]
[7]
[
8]
[
9]
[10]
[11]
catalysis,
[12]
[8b]
O
rganocobalt complexes are a key reagent in organic
combining scanning tunneling microscopy (STM) with X-ray
spectroscopy (XS) to identify the formation of a novel
organocobalt complex evolving under interfacial confinement
and to elucidate its intricate nature. Our real-space molecular-
level insight reveals that on the Ag(111) surface, a unique
cobalt compound can be synthesized in which a single
Co atom is coordinated to two carbonitrile moieties and an
internal alkyne. By STM we identify a strong distortion of the
alkyne-bonded species, contrasting its originally linear con-
formation, which results from the complexation-induced
rehybridization. Complementary X-ray photoelectron spec-
troscopy (XPS) measurements establish the core-level signa-
ture of the complex phase. A thorough analysis of the bonding
mechanism with density functional theory (DFT) unravels
strong analogies to the Dewar–Chatt–Duncanson ligation
synthesis because they can be used as a versatile tool for
various reactions. In an industrial context these complexes are
central in the multimillion-dollar technology of hydroformy-
[
1]
lation. Furthermore, they are involved in various classic
synthetic concepts, such as the Pauson–Khand or the
Nicholas reactions. Amongst the most successful and
popular compounds are dicobalt hexacarbonyls activating
alkyne triple bonds by formation of a m -alkyne bridging the
[2]
[3]
2
[
*] P. B. Weber, R. Hellwig, T. Paintner, M. Paszkiewicz,
P. Casado Aguilar, P. S. Deimel, Dr. Y. Guo, Dr. Y.-Q. Zhang,
Dr. F. Allegretti, Dr. A. C. Papageorgiou, Dr. J. Reichert,
Prof. Dr. J. V. Barth, Dr. F. Klappenberger
Physik Department E20, Technische Universität München
James-Franck-Str., 85748 Garching (Germany)
E-mail: florian.klappenberger@tum.de
[13]
model. However, contrary to the known isocyano organo-
[14]
cobalt complexes, herein we observe a nitrogen-lone-pair-
mediated coordination of the carbonitriles. Furthermore, we
demonstrate the self-assembly of chains of complexes origi-
nating from the ditopic nature of the employed ligand
molecules.
Dr. M. Lattelais, Dr. M.-L. Bocquet
Ecole Normale SupØrieure, PSL Research University
DØpartement de Chimie, CNRS UMR
8
640 PASTEUR, 75005 Paris (France)
E-mail: marie-laure.bocquet@ens.fr
For our study we employed the de novo synthesized
Dr. S. Klyatskaya, Prof. Dr. M. Ruben
Karlsruher Institut für Technologie
4
,4’-(ethyne-1,2-diyl)dibenzonitrile (DBNE), depicted in
Scheme 1, consisting of two benzonitrile moieties connected
through a central alkyne unit (for the synthetic details see the
Synthesis Section in the Supporting Information). The length
of the molecule, defined as the distance between the nitrogen
atoms, measures 14.9 Š according to DFT calculations.
In a first step, a sub-monolayer coverage of DBNE was
Kaiserstraße 12, 76131 Karlsruhe (Germany)
Prof. Dr. M. Ruben
IPCMS-CNRS, UniversitØ de Strasbourg
2
3 rue de Loess, 67034 Strasbourg (France)
Supporting information for this article can be found under:
http://dx.doi.org/10.1002/anie.201600567.
5
754
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 5754 –5759

Angewandte
Communications
Chemie
explained by three-fold Co coordination, in analogy to earlier
work with related, carbonitrile-terminated linker spe-
[9b,15c,17]
cies.
In contrast to the previous studies in which
Scheme 1. Chemical structure of 4,4’-(ethyne-1,2-diyl)dibenzonitrile
DBNE).
highly crystalline arrangements with translational symmetry
were observed, here the honeycomb network exhibits varying
pore shapes. A simple analysis of large domains employing
ball-and-stick models (Figure 1d) and assuming a strictly
hexagonal geometry yields an unreasonably small CoÀN
(
evaporated onto the Ag(111) surface at 270 K followed by
cooling and transfer to the microscope, where investigation
took place at cryogenic temperature. We observed a regular
molecular network (Figure 1a) dominating on the sample,
whereby domains typically exhibit lateral dimensions as large
as 500 nm. A superposition of scaled ball-and-stick models of
DBNE onto a high-resolution image (Figure 1b) demon-
strates a supramolecular packing with interdigitating, anti-
parallel aligned CN groups and a distance of about 2 Š from
terminal N atoms to the nearest hydrogen neighbor under the
assumption of a completely planar conformation for the
adsorbates. Similar self-assembly has been reported and
thoroughly analyzed in previous work on carbonitrile-termi-
binding distance of 1.0 Š. We conclude that the introduction
of the central alkyne unit results in unfavorable epitaxy with
respect to the underlying silver surface. Accordingly, the
network adapts its geometry by deviations from the perfect
six-fold symmetry, apparently facilitated by the increased
flexibility provided by the carbon–carbon triple bond as
compared to a purely phenylene backbone. Recent reports
showed similar adaptive coordination networks constructed
[
18]
from linkers featuring a central butadiyne moiety.
Following mild annealing to 250 K, the network trans-
formed into a different phase (Figure 2a), where the pores
appear elongated and line up in rows that are separated by
zigzag-shaped double lines. This change results from the
formation of a novel scissor-like structural motif (dashed
lines) beside the three-fold coordination nodes (green with
black lines). Further annealing to 350 K drives the formation
of large domains entirely consisting of zigzag-shaped lines
(Figure 2b) and areas of bare metal surface (not shown)
where no residual porous structures persist. Figure 2c, show-
ing an expanded portion of Figure 2b, demonstrates that the
regular zigzag lines incorporate exclusively the scissor-like
[9f, 15]
nated species on the same substrate.
Accordingly, we
associate the observed pattern with dipole–dipole interac-
[16]
tions between CN groups and proton-acceptor ring inter-
[
15e]
actions.
For exploring the metal-directed assembly, DBNE depo-
sition followed by evaporation of Co atoms was carried out at
a sample temperature of 135 K (see also the Supporting
Information). In this case, an open-porous honeycomb-like
network (Figure 1c) prevails on the surface. The network is
Figure 1. STM topographs of organic and metal–organic phases before
complex formation. a) Dense packed network of DBNE molecules prior
to Co deposition (bias voltage=0.75 V, tunneling current=65 pA).
b) Expanded portion of (a) with overlaid molecular models. The
distance between terminal N atoms and adjacent H atoms is approx-
imately 2 Š. c) Metal-directed assembly of an open network after
Co deposition at 135 K (À0.03 V, 30 pA). d) Expanded view of a few
pores in (c) with superimposed models of Co (green) and DBNE
Figure 2. a) STM topograph of mixed-phase network (À0.3 V, 30 pA)
obtained after mild annealing to 250 K. b,c) STM topographs of the
complex network (0.32 V, 47 pA) obtained after further annealing to
350 K. In (c) a distorted DBNE is marked with a solid black line
whereas the new scissor-shaped complex is outlined by the dashed
black line. d) Schematic models of complexes superimposed on an
STM topograph. The unit cell is indicated with a dashed outline.
(À1.1 V, 0.305 nA).
Angew. Chem. Int. Ed. 2016, 55, 5754 –5759
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5755

Angewandte
Communications
Chemie
motif (dashed line). This high-resolution version of the STM
image allows us to clearly identify that this structure contains
two bright and straight protrusions joined through a darker
bridge to a curved feature (indicated with a black solid line).
The curved protrusions suggest that a significant chemical
change occurred for a portion of the molecules, which now
adopt a modified, significantly bent conformation. The
absence of Co clusters on the entire surface implies that the
transition-metal atoms are involved in the scissor-shaped
motifs. Accordingly, a qualitative model for the motif can be
put forward suggesting a Co center surrounded by two
straight and one strongly distorted unit of DBNE connected
through the carbonitrile and alkyne groups, respectively
(
Figure 2d).
To better assess the chemical nature of the scissor motif
we carried out XPS measurements (see the Supporting
Information) in which we compared the three phases,
namely the purely organic phase, the metal–organic phase,
and the novel phase tentatively assigned to the organometal-
lic complex motif. From the detected N 1s core-level signa-
tures (see Figure S1 in the Supporting Information) we
conclude that the metal–organic structure is indeed stabilized
by CNÀCo coordination and that the scissor motif is chemi-
cally distinct from the metal–organic phase. Furthermore, the
N 1s signature of the organometallic complex phase contains
two chemically inequivalent nitrogen species consistent with
the tentative assignment in Figure 2d. The well-defined XPS
signature also rules out more dramatic reaction scenarios
based on complete disruption of the CꢀC triple bond
Figure 3. DFT-optimized models of complexes M1, M2, and M3 on the
Ag(111) surface, showing the top (left panels) and side (right panels)
views. Atom colors: Ag=gray; Co=green; C=black; N=blue;
H=white. The side views are drawn assuming a direction-of-view as
indicated by the arrows in the left panels and a glancing angle of 258.
(
Figure S2). Finally, based on the Co 2p data (Figure S1h),
the Co atom in the scissor motif adopts a quasi-neutral state
0
close to Co within the complex.
bottom), is inspired by the classical co-cyclization of triple
bonds catalyzed by cobalt complexes: the two CN and alkyne
units are cyclized through a [2+2+2] cycloaddition forming
a pyrazine cycle. Here, the Co atom is situated below the
center of the pyrazine between the substrate and the plane of
the organic ring (see the side view in Figure 3, bottom) at
a similar adsorption height as in the M2 case. In contrast, in
the first model M1, the atom is pulled up from the surface and
resides almost at the same height as the organic ligands. The
Co heights above the surface (Ag atom centers) are 2.4 Š in
M1, 1.7 Š for M2 and M3, and 1.6 Š for the isolated
Co atoms.
Next, an investigation of the thermal stability of the
scissor motifs by a series of STM experiments carried out at
room temperature (Figure S3) demonstrated the structural
integrity at 300 K. Neither three-fold coordination nodes nor
hexagonal pores were detected, corroborating that the bond-
ing of the scissor motif is of a different nature and is stronger
than the three-fold coordination. Annealing temperatures
above 373 K restore the pristine metal surface, signaling that
dissolution of the complexes allowing desorption of the
individual parts is possible and suggesting an, at least partially,
reversible character of the scissor motif.
To determine the chemical nature of the complexes
Figures 2b–d), we carried out extensive DFT calculations,
The formation energies, that is, the energy of the
combined system minus the energy of the individual constit-
uents, of the different adsorbed complexes M1, M2, and M3
amount to À3.8, À2.6, and À3.6 eV, respectively. Thus, of the
three possibilities, M1 seems energetically favorable. Inter-
estingly, the energy balance of complex M3 comes close to
that of M1, which results from the aromaticity of the formed
six-membered ring. Note that these energies do not reflect the
barriers of the transition states to be overcome during
coupling and therefore neglect reaction kinetics.
(
whereby we optimized different potential coordination/cova-
lent bonding schemes. Notably, the three configurations
displayed in Figure 3 were considered. The DFT model of
the first complex, M1 (Figure 3, top), assumes that two
pristine DBNE molecules coordinate to the Co center
through the N lone pair of the CN group and the third
ligand is distorted and attached through the central alkyne
unit. The second possibility M2 (Figure 3, middle) is inspired
by the 5-membered ring intermediate of a Co-catalyzed
To distinguish between the three models, the XPS
signatures do not provide conclusive information. This results
[19]
[
2+2+2] cycloaddition reaction and represents a variation
of M1 where two DBNE ligands are covalently linked by a CÀ from the lack of reports on related complexes and the
C coupling reaction of the a-carbon atoms of the CN groups,
in addition to concomitant N coordination leading to a 1,2-
diimine-like ligand. Finally, the third complex, M3 (Figure 3,
pyrazine N 1s binding energy being similar to that of the CNÀ
[20]
Co type. We focused instead on geometric considerations
and carefully analyzed the structural characteristics of specific
5
756
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 5754 –5759

Angewandte
Communications
Chemie
configurations in nonperiodic structures (Figures S4, S5). The
observed variety of the scissor motif opening angles is
unlikely to appear for models M2 and M3 which feature
covalently connected “blades”. Furthermore, the CoÀCo
distances in coordinated and complexed structures (see
Figure S4) are equal within the experimental resolution and
agree well with the length expected for intact DBNE
coordinatively connecting Co atoms (see the Supporting
Information), whereas significant structural disagreement
well beyond the experimental error bar is detected for M2
and M3 (Figure S5).
Additional support for the reversible bonding nature
behind the scissor motif is provided by manipulation experi-
ments. We were able to induce a transformation from the
asymmetric organocobalt complex (Figure S6a) to the three-
fold symmetric coordination node (Figure S6b) by the
combination of current pulses and movement of the STM
tip. The reversible character is more likely to be compatible
with the M1 configuration than with the covalent linking
involved in M2 and M3.
To further substantiate our interpretation of the observed
complexes according to model M1, simulated STM images
were calculated of the three models and compared with the
experimental topography data (Figure S7). The best agree-
ment is obtained for model M1 (Figure 4).
Figure 5. a) Chemical structure of complex M1. b) Charge-transfer map
calculated from the difference between the charge density in the
structure and the charge density of the isolated Co atom (the DBNE
molecules and the Ag(111) surface are in the same position as in the
composite structure). Red and yellow isosurfaces represent electron
À3
loss and gain, respectively, with absolute contour values of 0.04 e Š
.
0
c) Schematic representation of the established Dewar–Chatt–Duncan-
son model for the Co–alkyne bonding: the metallacyclopropene (left)
and presumed interactions characteristic of a 2p-electron donor with
p-donation (middle) and p-back-donation (right).
[1,23]
pentadienylcobalt complexes and related compounds.
Cyano cobalt catalysts, in which the CN moiety is connected
to the cobalt center through the carbon atom, are wide-
[14,24]
spread.
However, catalytic cobalt complexes with nitriles
coordinated through the N lone pair are much more scarce.
Complexes of this type have been reported in single-nitrile-
[25]
coordinated compounds. Because a twofold carbonitrile-
coordinated organocobalt complex, as identified here, has not
been previously reported, it is suggested that the underlying
surface plays a crucial role in its formation.
The detailed DFT analysis of the electronic interaction
provides further insights. Figure 5b shows the charge-transfer
map calculated from the difference between the charge
density of M1 and the charge densities of the isolated
Co atom. The map displays spatial zones of electron gain
(yellow) and electron loss (red) upon Co bonding to the
molecular ligands on the Ag substrate. On the alkyne, there is
clearly an electron depletion zone resembling one of the out-
of-plane states of the alkynes along the z direction (see also
the side view shown in Figure S8). The electron accumulation
zone nearby mimics an in-plane state along the y direction.
Furthermore, the electron loss of the in-plane d state (dxy)
indicates electron back-donation. For interpreting the charge
transfer, we combine molecular orbitals (MOs) of the
Co center (d orbital manifold) with the alkyne MOs of
matching symmetry (Figure 5c). These redistribution maps
are in agreement with the MO interaction schemes given in
Figure 5c. Classically, the metal–alkyne bonding in organo-
cobalt complexes can be described by one resonance struc-
ture, specifically the metallacyclopropene shown in Figure 5c.
The alkyne ligand has four p-electrons with two sets of
occupied p-states and unoccupied p*-states. The interaction
with the Co atom follows the Dewar–Chatt–Duncanson
model for a two-electron-donor ligand: an alkyne-to-Co
electron donation with a strong metal-to-alkyne back-dona-
tion. The strong back-donation is characteristic of electron-
Figure 4. a) Simulated STM image of the complex at a bias voltage of
À1 V. b) Actual STM image of complex (À1 V, 143 pA).
Considering these findings, we conclude that complex M1
is fully consistent with the combined experimental observa-
tions, including spectroscopic, structural, and electronic
aspects.
Complex M1 can be interpreted in terms of the chemical
structure shown in Figure 5a, in which formal conversion of
the carbon–carbon triple bond into a double bond connected
with the establishment of two carbon–cobalt bonds is
assumed. Indeed, the DFT-calculated CÀC bond lengths of
the internal alkynes of the two intact DBNE (1.23 Š) and the
distorted species (1.33 Š) assume values near those of
prototypical triple and double bonds (1.20 Š for ethyne,
1
.34 Š for ethene) and thus corroborate the suggested
rehybridization. Indeed, similar organometallic bonding
accompanied by carbon rehybridization in the vicinity of
CO-coordinated Co centers is known for organocobalt
[21]
complexes from solution chemistry.
Various types of organocobalt-based catalysts exist involv-
ing combinations of oxygen- and nitrogen-coordinated
[
1,22]
Co centers.
Furthermore, cycloaddition reactions involv-
ing nitriles are known to proceed in the presence of cyclo-
Angew. Chem. Int. Ed. 2016, 55, 5754 –5759
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5757

Angewandte
Communications
Chemie
[
26]
rich transition metals such as Co
and explains the
Gauss Centre for Supercomputing/Leibniz Supercomputing
Centre (Grant ID: pr85la) and the Pole Scientifique de
ModØlisation Numerique (PSMN) in ENS de Lyon.
occurrence of the metallocyclopropene form of the complex.
In this form the alkyne is mostly hybridized to an alkene
sp trigonal geometry) as a result of the weakening of one p-
2
(
bond. This weakening originates from the back-donation
process with electrons occupying one antibonding p*-orbital
of the alkyne.
Keywords: alkynes · density functional calculations ·
organocobalt complexes · scanning tunneling microscopy ·
surface chemistry
In principle, even though not detected on the surface, an
alternative complex structure where one Co center binds
simultaneously to two ethynylene bridges can be hypothe-
sized. To elucidate why the mixed complex M1 is preferred,
we evaluated the binding energetics of two purely organo-
metallic complexes in the gas phase, C1 and C2, and
compared them with C3, the gas-phase variant of M1
How to cite: Angew. Chem. Int. Ed. 2016, 55, 5754–5759
Angew. Chem. 2016, 128, 5848–5853
[1] I. Omae, Appl. Organomet. Chem. 2007, 21, 318 – 344.
[2] P. L. Pauson, Tetrahedron 1985, 41, 5855 – 5860.
[
[
[
3] K. M. Nicholas, Acc. Chem. Res. 1987, 20, 207 – 214.
4] W. G. Sly, J. Am. Chem. Soc. 1959, 81, 18 – 20.
5] D. Seyferth, A. T. Wehman, J. Am. Chem. Soc. 1970, 92, 5520 –
(
Figure S9). All three complexes were geometrically opti-
mized without substrate. For C1, the alkyne ligands lie in the
same plane and for C2 the alkynes are oriented perpendicular
to each other. We find that the perpendicular structure is
5
522.
6] a) S. Laschat, A. Becheanu, T. Bell, A. Baro, Synlett 2005, 2547 –
570; b) K. M. Brummond, J. L. Kent, Tetrahedron 2000, 56,
3263 – 3283.
[
2
1
.2 eV more stable than the planar structure (total binding
energy of À3.4 eV for C1 versus À4.6 eV for C2) in which the
Co atom is displaced from the alkyne center, suggesting
interfering steric repulsion between the phenyl groups. The
stabilization of the perpendicular structure could be due to
the favorable interactions between one triple CꢀC bond and
[7] a) J. V. Barth, Surf. Sci. 2009, 603, 1533 – 1541; b) J. A. A. W.
Elemans, S. B. Lei, S. De Feyter, Angew. Chem. Int. Ed. 2009, 48,
7
298 – 7332; Angew. Chem. 2009, 121, 7434 – 7469; c) N. Lin, S.
Stepanow, M. Ruben, J. V. Barth, Top. Curr. Chem. 2009, 287, 1 –
4.
4
[
8] a) S. M. Barlow, R. Raval, Surf. Sci. Rep. 2003, 50, 201 – 341; b) F.
Klappenberger, Prog. Surf. Sci. 2014, 89, 1 – 55.
the d orbital of Co and between the other triple CꢀC bond
xz
and the dyz orbital of Co. In comparison, the energy gained by
the formation of C3 (5.6 eV) exceeds that of both purely
organometallic complexes, which is consistent with the
absence of complexes of C1 and C2 on the metallic surface.
In conclusion, our investigation demonstrates intricate
organometallic complexation at well-defined surfaces under
ultra-high-vacuum conditions, whereby novel complexes can
be stabilized. Inspections at the single-molecule level com-
bined with state-of-the-art theoretical modeling provide
detailed insight into the pertaining structural and electronic
properties. The reversible character of the bond formation, an
essential property for its success as a catalytic intermediate in
various synthetic pathways, is demonstrated by the ability to
dissolve the metal–organic complexes with rather mild
temperatures and molecular manipulation experiments. Fur-
thermore, our results are equally important for the rapidly
evolving field of on-surface covalent synthesis. This work will
facilitate the expansion of the previously employed direct-
[9] a) S. Stepanow, N. Lin, D. Payer, U. Schlickum, F. Klappen-
berger, G. Zoppellaro, M. Ruben, H. Brune, J. V. Barth, K. Kern,
Angew. Chem. Int. Ed. 2007, 46, 710 – 713; Angew. Chem. 2007,
1
19, 724 – 727; b) U. Schlickum, R. Decker, F. Klappenberger, G.
Zoppellaro, S. Klyatskaya, M. Ruben, I. Silanes, A. Arnau, K.
Kern, H. Brune, J. V. Barth, Nano Lett. 2007, 7, 3813 – 3817; c) F.
Cicoira, C. Santato, F. Rosei, Top. Curr. Chem. 2008, 285, 203 –
267; d) G. Pawin, K. L. Wong, D. Kim, D. Z. Sun, L. Bartels, S.
Hong, T. S. Rahman, R. Carp, M. Marsella, Angew. Chem. Int.
Ed. 2008, 47, 8442 – 8445; Angew. Chem. 2008, 120, 8570 – 8573;
e) M. Marschall, J. Reichert, A. Weber-Bargioni, K. Seufert, W.
Auwärter, S. Klyatskaya, G. Zoppellaro, M. Ruben, J. V. Barth,
Nat. Chem. 2010, 2, 131 – 137; f) S. Klyatskaya, F. Klappenberger,
U. Schlickum, D. Kühne, M. Marschall, J. Reichert, R. Decker,
W. Krenner, G. Zoppellaro, H. Brune, J. V. Barth, M. Ruben,
Adv. Funct. Mater. 2011, 21, 1230 – 1240; g) D. Écija, J. I. Urgel,
A. C. Papageorgiou, S. Joshi, W. Auwarter, A. P. Seitsonen, S.
Klyatskaya, M. Ruben, S. Fischer, S. Vijayaraghavan, J. Reichert,
J. V. Barth, Proc. Natl. Acad. Sci. USA 2013, 110, 6678 – 6681;
h) A. Shchyrba, C. Wäckerlin, J. Nowakowski, S. Nowakowska,
J. Bjçrk, S. Fatayer, J. Girovsky, T. Nijs, S. C. Martens, A.
Kleibert, M. Stçhr, N. Ballav, T. A. Jung, L. H. Gade, J. Am.
Chem. Soc. 2014, 136, 9355 – 9363.
[27]
coupling mechanisms using alkynes to more sophisticated
pathways via controlled catalytic reaction intermediates
and transition-metal centers incorporated in specific
environments.
[10] S. Stepanow, M. Lingenfelder, A. Dmitriev, H. Spillmann, E.
Delvigne, N. Lin, X. B. Deng, C. Z. Cai, J. V. Barth, K. Kern, Nat.
Mater. 2004, 3, 229 – 233.
[
11] S. Fabris, S. Stepanow, N. Lin, P. Gambardella, A. Dmitriev, J.
Honolka, S. Baroni, K. Kern, Nano Lett. 2011, 11, 5414 – 5420.
12] P. Gambardella, S. Stepanow, A. Dmitriev, J. Honolka, F. M. F.
de Groot, M. Lingenfelder, S. Sen Gupta, D. D. Sarma, P.
Bencok, S. Stanescu, S. Clair, S. Pons, N. Lin, A. P. Seitsonen,
H. Brune, J. V. Barth, K. Kern, Nat. Mater. 2009, 8, 189 – 193.
13] a) J. S. Dewar, Bull. Soc. Chim. Fr. 1951, 18, C71 – C79; b) J.
Chatt, L. A. Duncanson, J. Chem. Soc. 1953, 2939 – 2947.
Acknowledgements
[
Funding was provided by the EU through the ERC Advanced
Grant MolArt (Grant 247299), the Munich Center for
Advanced Photonics, the TUM Institute of Advanced Study
[
(
IAS) funded by the German Excellence Initiative, and the
[
[
14] W. P. Fehlhammer, M. Fritz, Chem. Rev. 1993, 93, 1243 – 1280.
15] a) U. Schlickum, R. Decker, F. Klappenberger, G. Zoppellaro, S.
Klyatskaya, W. Auwärter, S. Neppl, K. Kern, H. Brune, M.
Ruben, J. V. Barth, J. Am. Chem. Soc. 2008, 130, 11778 – 11782;
b) F. Klappenberger, D. Kühne, W. Krenner, I. Silanes, A.
German Research Foundation (DFG) through KL 2294/3-1.
M.L., M.-L.B. and J.V.B. thank the collaborative transnational
grant in the ANR-DFG 2011 call (ANR-11-INTB-1014-01)
for support. Computer resources have been provided by the
5
758
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 5754 –5759

Angewandte
Communications
Chemie
Arnau, F. J. Garcia de Abajo, S. Klyatskaya, M. Ruben, J. V.
[22] a) K. Gao, P. S. Lee, T. Fujita, N. Yoshikai, J. Am. Chem. Soc.
Barth, Nano Lett. 2009, 9, 3509 – 3514; c) D. Kühne, F. Klappen-
berger, R. Decker, U. Schlickum, H. Brune, S. Klyatskaya, M.
Ruben, J. V. Barth, J. Phys. Chem. C 2009, 113, 17851 – 17859;
d) F. Klappenberger, D. Kühne, M. Marschall, S. Neppl, W.
Krenner, A. Nefedov, T. Strunskus, K. Fink, C. Wçll, S.
Klyatskaya, O. Fuhr, M. Ruben, J. V. Barth, Adv. Funct. Mater.
2010, 132, 12249 – 12251; b) P. S. Lee, T. Fujita, N. Yoshikai, J.
Am. Chem. Soc. 2011, 133, 17283 – 17295.
[23] N. E. Schore, Chem. Rev. 1988, 88, 1081 – 1119.
[24] T. Kauffmann, Angew. Chem. Int. Ed. Engl. 1996, 35, 386 – 403;
Angew. Chem. 1996, 108, 401 – 418.
[25] J. H. Hall, R. L. Delavega, W. L. Purcell, Inorg. Chim. Acta 1985,
102, 157 – 162.
2
011, 21, 1631 – 1642; e) E. Arras, A. P. Seitsonen, F. Klappen-
berger, J. V. Barth, Phys. Chem. Chem. Phys. 2012, 14, 15995 –
6001.
[26] B. Fang, W. S. Ren, G. H. Hou, G. F. Zi, D. C. Fang, L. Maron,
M. D. Walter, J. Am. Chem. Soc. 2014, 136, 17249 – 17261.
[27] a) Y.-Q. Zhang, N. Kep cˇ ija, M. Kleinschrodt, K. Diller, S.
Fischer, A. C. Papageorgiou, F. Allegretti, J. Bjçrk, S. Klyat-
skaya, F. Klappenberger, M. Ruben, J. V. Barth, Nat. Commun.
2012, 3, 1286; b) H. Y. Gao, H. Wagner, D. Y. Zhong, J. H.
Franke, A. Studer, H. Fuchs, Angew. Chem. Int. Ed. 2013, 52,
4024 – 4028; Angew. Chem. 2013, 125, 4116 – 4120; c) F. Beben-
see, C. Bombis, S. R. Vadapoo, J. R. Cramer, F. Besenbacher,
K. V. Gothelf, T. R. Linderoth, J. Am. Chem. Soc. 2013, 135,
2136 – 2139; d) O. Díaz Arado, H. Mçnig, H. Wagner, J.-H.
Franke, G. Langewisch, P. A. Held, A. Studer, H. Fuchs, ACS
Nano 2013, 7, 8509 – 8515; e) J. Eichhorn, W. M. Heckl, M.
Lackinger, Chem. Commun. 2013, 49, 2900 – 2902; f) J. Liu, P.
Ruffieux, X. L. Feng, K. Müllen, R. Fasel, Chem. Commun.
2014, 50, 11200 – 11203; g) B. Cirera, Y.-Q. Zhang, J. Bjçrk, S.
Klyatskaya, Z. Chen, M. Ruben, J. V. Barth, F. Klappenberger,
Nano Lett. 2014, 14, 1891 – 1897; h) F. Klappenberger, Y. Zhang,
J. Bjçrk, S. Klyatskaya, M. Ruben, J. V. Barth, Acc. Chem. Res.
2015, 48, 2140 – 2150.
1
[
16] a) T. Yokoyama, S. Yokoyama, T. Kamikado, Y. Okuno, S.
Mashiko, Nature 2001, 413, 619 – 621; b) Y. Okuno, T.
Yokoyama, S. Yokoyama, T. Kamikado, S. Mashiko, J. Am.
Chem. Soc. 2002, 124, 7218 – 7225; c) E. Y. Vedmedenko, N.
Mikuszeit, ChemPhysChem 2008, 9, 1222 – 1240.
[
[
17] N. Henningsen, R. Rurali, C. Limbach, R. Drost, J. I. Pascual,
K. J. Franke, J. Phys. Chem. Lett. 2011, 2, 55 – 61.
18] a) C. S. Kley, J. Cechal, T. Kumagai, F. Schramm, M. Ruben, S.
Stepanow, K. Kern, J. Am. Chem. Soc. 2012, 134, 6072 – 6075;
b) J. Cechal, C. S. Kley, T. Kumagai, F. Schramm, M. Ruben, S.
Stepanow, K. Kern, Chem. Commun. 2014, 50, 9973 – 9976.
19] a) K. P. C. Vollhardt, Angew. Chem. Int. Ed. Engl. 1984, 23, 539 –
[
5
56; Angew. Chem. 1984, 96, 525 – 541; b) J. H. Hardesty, J. B.
Koerner, T. A. Albright, G. Y. Lee, J. Am. Chem. Soc. 1999, 121,
055 – 6067.
6
[
[
20] U. W. Hamm, V. Lazarescu, D. M. Kolb, J. Chem. Soc. Faraday
Trans. 1996, 92, 3785 – 3790.
21] a) P. Magnus, L. M. Principe, Tetrahedron Lett. 1985, 26, 4851 –
4
854; b) P. J. Low, R. Rousseau, P. Lam, K. A. Udachin, G. D.
Enright, J. S. Tse, D. D. M. Wayner, A. J. Carty, Organometallics
999, 18, 3885 – 3897; c) P. Magnus, C. Exon, P. Albaughrobert-
son, Tetrahedron 1985, 41, 5861 – 5869.
1
Received: January 19, 2016
Published online: April 6, 2016
Angew. Chem. Int. Ed. 2016, 55, 5754 –5759
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5759