Multistate Switches: Ruthenium Alkynyl-Dihydroazulene/Vinylheptafulvene Conjugates

Article abstract of DOI:10.1002/chem.201600178

Multimode molecular switches incorporating distinct and independently addressable functional components have potential applications as advanced switches and logic gates for molecular electronics and memory storage devices. Herein, we describe the synthesis and characterization of four switches based on the dihydroazulene/vinylheptafulvene (DHA/VHF) photo/thermoswitch pair functionalized with the ruthenium-based Cp?(dppe)Ru ([Ru?]) metal complex (dppe=1,2-bis(diphenylphosphino)ethane; Cp?=pentamethylcyclopentadienyl). The [Ru?]-DHA conjugates can potentially exist in six different states accessible by alternation between DHA/VHF, Ru^II/Ru^III, and alkynyl/vinylidene, which can be individually stimulated by using light/heat, oxidation/reduction, and acid/base. Access to the full range of states was found to be strongly dependent on the electronic communication between the metal center and the organic photoswitch in these [Ru?]-DHA conjugates. Detailed electrochemical, spectroscopic (UV/Vis, IR, NMR), and X-ray crystallographic studies indeed reveal significant electronic interactions between the two moieties. When in direct conjugation, the ruthenium metal center was found to quench the photochemical ring-opening of DHA, which in one case could be restored by protonation or oxidation, allowing conversion to the VHF state.

Full text of DOI:10.1002/chem.201600178

DOI: 10.1002/chem.201600178
Full Paper
&
Molecular Switches
Multistate Switches: Ruthenium Alkynyl–Dihydroazulene/
Vinylheptafulvene Conjugates
Alexandru Vlasceanu, Cecilie L. Andersen, Christian R. Parker, Ole Hammerich,
Thorbjørn J. Morsing, Martyn Jevric, Søren Lindbæk Broman, Anders Kadziola, and
Mogens Brøndsted Nielsen*[a]
Abstract: Multimode molecular switches incorporating dis-
tinct and independently addressable functional components
have potential applications as advanced switches and logic
gates for molecular electronics and memory storage devices.
Herein, we describe the synthesis and characterization of
four switches based on the dihydroazulene/vinylheptaful-
vene (DHA/VHF) photo/thermoswitch pair functionalized
with the ruthenium-based Cp*(dppe)Ru ([Ru*]) metal com-
plex (dppe=1,2-bis(diphenylphosphino)ethane; Cp*=pen-
tamethylcyclopentadienyl). The [Ru*]–DHA conjugates can
potentially exist in six different states accessible by alterna-
tion between DHA/VHF, RuII/RuIII, and alkynyl/vinylidene,
which can be individually stimulated by using light/heat, oxi-
dation/reduction, and acid/base. Access to the full range of
states was found to be strongly dependent on the electronic
communication between the metal center and the organic
photoswitch in these [Ru*]–DHA conjugates. Detailed elec-
trochemical, spectroscopic (UV/Vis, IR, NMR), and X-ray crys-
tallographic studies indeed reveal significant electronic inter-
actions between the two moieties. When in direct conjuga-
tion, the ruthenium metal center was found to quench the
photochemical ring-opening of DHA, which in one case
could be restored by protonation or oxidation, allowing con-
version to the VHF state.
ible oxidation to the RuIII species or protonation to the corre-
sponding vinylidene. This has been used in the formation of
dithienylethene (DTE) photoswitches end-capped with rutheni-
um alkynyl groups such as trans-(dppe)2RuX[7] or Cp*(dppe)Ru[8]
producing multistate systems (dppe=1,2-bis(diphenylphos-
phino)ethane, Cp*=pentamethylcyclopentadienyl anion). The
proximity of the Ru metal center to the thiophene ring was ob-
served to influence the switching behavior and impart proper-
ties unlike those observed for related Fe complexes.[8a] Effective
radical delocalization into the DTE photochromic system could
be suppressed by using a phenylene spacer, thus allowing for
the preparation of a multimode switch exhibiting six states
with distinct non-linear optical (NLO) properties, independently
interconvertible by acid/base (alkynyl versus vinylidene ligand),
electrochemical (RuII versus RuIII), and light (ring-opening/clo-
sure) stimulation.[9] The incorporation of ruthenium alkynyl
moieties was also shown to modulate the coupling between
gold electrodes and a central DTE unit when one such mole-
cule was suspended between two electrodes by anchoring to
sulfur end-groups. In this case, the ruthenium complexes pre-
vented quenching of the DTE photoswitching by the elec-
trodes.[4,10]
Introduction
Molecular switches play a key role in the development of func-
tional molecular electronics devices.[1] In this way, conventional
computing operations could potentially be performed by
single molecules functioning as transistors or logic gates.[2] In
addition, the development of compounds that can exist in
multiple states, exhibiting distinct optical properties, intercon-
vertible by external stimuli has become a research forefront
with considerable prospects in this field.[3] One strategy to-
wards multistate molecules is through the combination of dis-
tinct functional components such as redox-active, photoactive,
or pH-responsive groups.[4] This can be achieved through the
functionalization of organic molecular switches with organo-
metallic moieties, as has been recently reviewed by Akita.[5]
With regard to this, RuII acetylides have been extensively stud-
ied in the literature due to their favorable properties.[6] The
strong electron-donating properties of the RuII metal center in
such complexes can typically be modulated by either revers-
[a] A. Vlasceanu, C. L. Andersen, Dr. C. R. Parker, Prof. Dr. O. Hammerich,
Dr. T. J. Morsing, Dr. M. Jevric, Dr. S. Lindbæk Broman, Prof. Dr. A. Kadziola,
Prof. Dr. M. B. Nielsen
It has recently been shown that suitably functionalized dihy-
droazulene (DHA, 1A)/vinylheptafulvene (VHF, 1D) photo/ther-
moswitches (Scheme 1) can also undergo light- and heat-con-
trolled conductance modulation in molecular junctions.[11] Ad-
ditionally, the incorporation of various functional components
as well as donor and/or acceptor groups through conjugated
or non-conjugated bridges into the DHA/VHF system has pre-
Department of Chemistry and Center for Exploitation of Solar Energy
University of Copenhagen, Universitetsparken 5
2100 Copenhagen (Denmark)
E-mail: mbn@chem.ku.dk
Supporting information and ORCIDs from the authors for this article are
available on the WWW under http://dx.doi.org/10.1002/chem.201600178.
Chem. Eur. J. 2016, 22, 7514 – 7523
7514
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 1. DHA/VHF interconversion and DHA atom numbering.
sented a way to tune both the light-induced ring-opening of
DHA and the thermal ring-closure of VHF.[12]
To take advantage of multiple functional groups and expand
the scope of the DHA/VHF system, functionalization with RuII
alkynyl groups, based on Cp*(dppe)Ru ([Ru*]) complexes, was
undertaken. Herein, we describe the synthesis, characterization,
and multimode switching behavior of four bridged [Ru*]–DHA
conjugates (2A–5A) shown in Figure 1. The position of the
Scheme 2. Conversion between the six states of a [Ru*]–DHA complex.
functionalized DHA proceeding via a vinylidene intermediate,
which is subsequently deprotonated, and ii) by Sonogashira
cross-coupling[13b,14] between Cp*(dppe)RuC2H and a halide-
functionalized DHA (Scheme 3). The two protocols encompass
Scheme 3. Metalation and Sonogashira cross-coupling protocols for assem-
bling the [Ru*]–DHA conjugates. X=halogen.
both possibilities for forming either the [Ru*]–alkynyl or alkyn-
yl–DHA bond in the final step. The two ruthenium precursors 6
and 7 were readily prepared according to known procedures[15]
as were the two iodo-functionalized DHAs 8 and 9 (Fig-
ure 2),[12d,16] which are precursors for 2A and 3A. Ethynyl-func-
tionalized DHAs 10 and 11 were prepared by following a strat-
egy starting from alkyne-functionalized acetophenone deriva-
tives described in the Supporting Information (Section 2) and
similar to previous reports.[12j,16] DHAs 12 and 13 (precursors
for 4A and 5A) were prepared from 1A (readily obtained on
a large scale[17]) by employing a bromination/elimination/cross-
coupling strategy.[12f,g,h,18] An isomerization reaction in the syn-
thesis of 4A and 5A occurred under the metalation conditions
(see below).
Figure 1. [Ru*]–DHA conjugates 2A–5A.
[Ru*] acetylide functionality, either meta or para on the phenyl-
ene ring to the DHA core (2A and 3A), at the DHA 6-position
with direct attachment (4A), or with a p-phenylene spacer
(5A) was found to strongly influence the properties of the
system. The proposed switching behavior of a [Ru*]–DHA com-
plex is shown in Scheme 2, which shows that the system can
reach six different states including two neutral (A/D), two vinyl-
idene (B/C), and two dimeric (F/E) forms, interchangeable by
a combination of light/heat, oxidation/reduction, and acid/
base stimulation. The latter, dimeric states (F/E), are tentatively
assigned as the result of redox stimulation, which is discussed
in more detail below.
Results and Discussion
Synthesis
The coupling between the ruthenium and DHA moieties of
conjugates 2A–5A was undertaken via two methods: i) by
a metalation approach,[13] using Cp*(dppe)RuCl and an ethynyl-
Figure 2. DHA building blocks.
Chem. Eur. J. 2016, 22, 7514 – 7523
www.chemeurj.org
7515
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
The successful preparation of 2A and 3A by the two differ-
ent routes is shown in Scheme 4. It was found that the metala-
tion of ethynyl DHAs required elevated temperatures overnight
to reach completion, whereas simpler literature-reported sub-
strates, for example, phenylacetylene, could be metalated
a result of a light–heat cycle (ring-opening/ring-closure), yield-
ing mixtures of the two isomers.[12g,h] In the present case, how-
ever, the isomerization does not involve irradiation. Such a pro-
cess has been previously noted in a sterically strained DHA
whereby a substituent shift was observed to occur from the 8-
to the 5-position in the absence of photochemical stimu-
lation.[12i] To test the possibility of isomerization in the absence
of Ru, DHA 12 was submitted to overnight reflux in NH4PF6/
MeOH but no signs of any reaction or isomerization were seen.
To further investigate the nature of this phenomenon, the met-
alation reaction between 6 and 12 was interrupted before de-
protonation, allowing for the isolation of the 6-substituted vi-
nylidene intermediate (14; Scheme 6). No evidence of any 7-
substituted derivative could be found. Treatment of 14 with
NEt3 gave product 4A, which could be protonated by trifluoro-
acetic acid (TFA) to generate the trifluoroacetate salt of 14
1
(4B), confirmed by H-1H COSY NMR spectroscopy.
Scheme 4. Synthesis of [Ru*]–DHA conjugates 2A and 3A.
within 1 h.[13a] This may occur as a result of competitive, yet re-
versible, coordination of the nitrile groups of DHA to the
[Cp*(dppe)Ru]+ metal complex. The displacement of chloride
from Cp*(dppe)RuCl in favor of MeCN coordination, promoted
by NH4PF6 in MeOH, has been reported to yield [Cp*(dppe)Ru-
(MeCN)]PF6, which exhibits a characteristic singlet in the
31P NMR (CD2Cl2) spectrum at d=75.52 ppm.[19] The combina-
tion of Cp*(dppe)RuCl (6) with 1A under analogous conditions
leads to a reaction mixture from which a yellow/green solid
can be isolated. The 31P NMR (CD2Cl2) spectrum of this product
includes an AB system, d=75.37 (d, J=20.1 Hz), 76.02 ppm (d,
J=20.1 Hz) as well as a singlet (d=75.97 ppm), which likely
correspond to a diastereoisomeric mixture of [Cp*(dppe)Ru-
(DHA)]PF6 salts.
Scheme 6. Trapping the 6-substituted vinylidene intermediate (14).
From the reaction mixture obtained through the coupling of
6 and 12, several interesting [Ru*]–DHA and [Ru*]–azulene by-
products were isolated during column chromatographic purifi-
cation (Figure 3). Azulene 15 and furoazulene 16 were both
The syntheses of 4A and 5A were accomplished via the
metalation strategy by combining 6 with 12 and 13, respec-
tively (Scheme 5). Interestingly, both reactions proceeded
through an isomerization by which the ruthenium–alkynyl sub-
stituent shifted from position 7 to position 6 on the DHA core.
Partial 7/6-isomerization has previously been observed as
Figure 3. Byproducts from the metalation reaction between 6 and 12.
characterized, the latter also by X-ray crystallography (see
below). The formation of 16 might be a result of nucleophilic
attack by residual water at the susceptible a-carbon of vinyli-
dene 14, which subsequently undergoes cyclization with the
cycloheptatriene ring. Such nucleophilic attack has more often
been observed for less bulky Ru complexes.[20] The presence of
the O atom at the 7-position of compound 16 suggests that
the vinylidene had already isomerized to the 6-position before
the nucleophilic attack by water took place. The azulene moi-
eties of 15 and 16 are most likely formed as a result of HCN
elimination, a phenomenon commonly observed in DHA syn-
thesis.[16]
Scheme 5. Synthesis of [Ru*]–DHA conjugates 4A and 5A.
Chem. Eur. J. 2016, 22, 7514 – 7523
www.chemeurj.org
7516
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 5. Reference complexes Cp*(dppe)RuC2PhX (19–21)[13b] and 22–23.[21]
Scheme 7. Sonogashira reaction between 7 and 17, which gave, among
other unidentifiable products, minute quantities of azulene 18.
Table 1. Selected bond lengths (crystallographic data).[a]
[b]
[Ru*]ÀC [Š]
CꢀC [Š]
CalkyneÀCX [Š] C1ÀC8a [Š]
A Sonogashira reaction between 7 and 7-bromo-substituted
DHA 17 was also attempted (Scheme 7). Although no evidence
of the desired product was found, column chromatographic
purification of the crude mixture revealed a variety of prod-
ucts. Among these, X-ray crystallography (see below) revealed
one complex that could be assigned to the azulene structure
18 in which the alkyne substituent has moved to position 4
and a cyano group has been incorporated at position 7, allud-
ing to more complex reactions.
1A[c]
10
–
–
–
–
1.575(1)
1.584(2)
1.572(7)
1.580(3)
1.577(3)
1.569(2)
1.595(5)
1.582(7)
1.420(3)
1.416(5)
–
1.201(3)
1.191(3)
1.179(4)
1.182(3)
1.180(3)
1.218(4)
1.220(5)
1.393(4)[d]
1.212(6)
1.221(8)
1.225(8)
1.215(5)
1.202(4)
1.432(3)
1.448(3)
1.440(3)
1.441(3)
1.441(3)
1.433(4)
1.421(5)
1.414(3)
1.407(6)
1.437(8)
1.413(8)
1.431(5)
1.444(4)
11
13
–
–
2A 2.005(2)
4A 1.979(3)
16
18
2.031(2)
1.960(4)
19[e] 1.997(6)
2.004(6)
20[e] 2.011(4)
21[e] 2.026(3)
–
–
Characterization
[a] When more than one molecule is present in the asymmetric unit, the
length of the indicated bond is reported for each molecule. [b] X=DHA,
Ar, or azulene. [c] Ref. [17]. [d] Bond length corresponding to [Ru*]-C=C
of the furan ring. [e] Ref. [13b].
Key to the potential of multimode switches in molecular elec-
tronics devices is the electronic communication between the
different functional components of the system. Here, we par-
tially describe the geometric (X-ray crystallographic), spectro-
scopic (NMR, IR, UV/Vis), and electrochemical characteristics of
[Ru*]–DHA dyads 2A–5A in their neutral (A/D), vinylidene (B/
C), and oxidized (F/E) forms. Vinylidine analogs 2B–5B could
all be quantitatively formed in situ by treatment of 2A–5A
with TFA (see the Supporting Information, Section 2). Charac-
terization data for literature-reported complexes 19–23[13b,21]
(Figure 5) are also reported here as references for comparison.
Single-crystal X-ray crystallography was used to reveal the
structures of [Ru*]–DHA conjugates 2A and 4A, which are
shown in Figure 4. Table 1 lists selected bond lengths in com-
parison to parent DHA 1A[17] and ruthenium complexes 19–
21.[13b] Longer CꢀC triple bonds are observed in the [Ru*]–
DHA conjugates relative to the uncomplexed DHA–alkynes (10,
11, and 13), which is indicative of partial electronic communi-
cation with the metal center. The molecular structures of azu-
lene byproducts 16 and 18 are shown in Figure 6.
Electrochemistry was used to determine the standard redox
potentials of all [Ru*]–DHA conjugates both in the acetylide
(2A–5A) and vinylidene (2B–5B) forms. Cyclic voltammograms
(CVs) were acquired between approximately À2 and +1 V and,
for 2A–5A, were found to exhibit electrochemically reversible
oxidation waves at approximately À0.2 V versus Fc/Fc+, corre-
sponding to the RuII/RuIII redox event (Table 2). As expected,
these potentials are comparable to those for 19–23. The re-
Figure 4. Molecular structures of 2A and 4A with displacement ellipsoids
drawn at the 50% probability level for non-hydrogen atoms.
Figure 6. Molecular structures of 16 (left) and 18 (right).
Chem. Eur. J. 2016, 22, 7514 – 7523
www.chemeurj.org
7517
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
It is hypothesized that the initial oxidations of RuII–DHA
complexes 2A–5A lead to significantly delocalized RuIII radical
cations, which undergo further chemical reactions and are thus
not observed on the timescale of the spectroscopic experi-
ments conducted here. The fleeting stability of such RuIII com-
plexes is well-established in the literature. For analogous metal
complexes, experimental observations, supported by calcula-
tions, have identified Cb of the alkynyl moiety as a site very
prone to further reactivity.[21–24] Thus, one possible explanation
for the chemical reaction observed here, resulting from the ox-
idation of 2A, could be that of a Cb–Cb radical dimerization be-
tween two oxidized [Ru*]–DHA species. A more detailed inves-
tigation into the oxidative dimerization products of
Cp(PPh3)2RuC2Ph with [FeCp2]PF6 was recently undertaken by
Bruce et al.[22g] Solvent-dependent redox chemistry in an analo-
gous oxidation using AgI salts on Cp(PPh3)2RuC2Ph was also ex-
plored, in one case leading to the isolation of divinylidene
[{Cp(PPh3)2Ru}2(m-C4Ph2)][PF6]2.[23] In other trials, oxidations of
Cp*(dppe)RuC2PhX using AgOTf also resulted in unstable radi-
cal cations leading to more complex product mixtures.[13b]
Such behavior, however, was not observed for the analogous
Fe derivatives, the oxidized states of which show greater stabil-
ity.[24] This is thought to be the result of a more localized and
shielded, metal-centered FeIII radical.
Table 2. Cyclic voltammetry data.[a]
[c]
[d]
Complex
Eo(1)[b] [V]
À0.17
DEp [V]
ipc/ipa
Ep(2)[e] [V]
Ep(1)[f] [V]
2A
2B
0.06
>0.90
>0.95
>0.90
>0.93
0.68
0.76
0.86
0.52
0.62
3A
3B
4A
4B
5A
5B
À0.18
À0.22
À0.21
0.06
0.07
0.08
0.72
0.55
0.65
19[g]
20[g]
21[g]
22[h]
23[h]
À0.06
À0.23
À0.41
À0.15
À0.10
0.09
0.09
0.09
0.09
0.11
>0.95
>0.90
>0.85
1.0
0.13
0.71
0.82
1.0
[a] Obtained at a glassy carbon working electrode, at room temperature,
using 0.5 mm [Ru*]–DHA (see the Supporting Information, Section 4.2)
and 0.1m Bu4NPF6 supporting electrolyte in CH2Cl2, at a sweep rate of
0.1 VsÀ1. Potentials are given versus the ferrocene/ferrocenium redox
couple (Fc/Fc+). [b] Formal oxidation potential for the first reversible one-
electron oxidation taken as (Epa +Epc)/2. [c] Peak separation of Eo(1) taken
as jEpaÀEpc j. [d] Peak current ratio of Eo(1). [e] Peak potential for the
second, irreversible oxidation event. [f] Peak potential for the first, irre-
versible oxidation of complexes 2B–5B. [g] From Ref. [13b]; recorded at
208C at a Pt working electrode. The formal oxidation potentials have
been converted by the present authors to the Fc/Fc+ scale by subtraction
of 0.460 V. [h] From Ref. [21]; recorded at 208C at a Pt working electrode.
The formal oxidation potentials have been converted by the present au-
thors to the Fc/Fc+ scale by subtraction of 0.460 V.
The chemical process following the oxidation of 2A was fur-
ther investigated by conducting bulk electrolysis of this com-
plex. The electrochemical experiment was monitored by cyclic
voltammetry and found to result in complete consumption of
2A, consistently leading to the quantitative formation of
a single, bright-yellow product (2F). Subjecting 2A to chemical
oxidation by using AgOTf or [FeCp2]PF6, however, resulted in
a more complex mixture of 2F and other products, not investi-
gated further. Thus, the clean and quantitative formation of
2F, from 2A, could only be achieved by electrochemical
means and, although 2F was observed to be particularly
stable in the electrolytic medium (0.1m Bu4NPF6 in CH2Cl2 or
CD2Cl2), standard purification techniques were unsuccessful in
isolating the pure complex. Bulk electrolysis of 2A in CD2Cl2
(see the Supporting Information, Section 5) allowed for de-
tailed spectroscopic (1H, 13C, 31P NMR, IR, and UV/Vis) character-
ization in situ. Furthermore, the oxidative dimerization of 2A
to 2F was found to be readily reversible by reduction, almost
quantitatively regenerating starting material 2A. This intercon-
version could reliably be followed by UV/Vis spectroelectro-
chemical experiments (see below) as well as by cyclic voltam-
metry over the course of bulk electrolytic experiments. The
sults indicate that 4A and 5A are marginally easier to oxidize
than 2A and 3A, which is not surprising as the [Ru*]–acetylide
moiety is attached to the partially electron-donating seven-
membered ring. A second, irreversible oxidation can be ob-
served at approximately +0.6 V for 2A–5A as was also ob-
served for reference complexes 22–23.
The reversibility of the 2A oxidation was found to be depen-
dent on the concentration of the sample, providing strong evi-
dence for a second-order intermolecular dimerization (Table 3).
The reaction was observed to occur on the second/minute
timescale for 2A, which displays scan-rate-dependent ipc/ipa re-
versibilities. Electrolysis of all complexes 2A–5A resulted in
a chemical event (but not corresponding to dimerization for
3A–5A, see below) that quantitatively consumed the transient
RuIII species, giving rise to indicative color changes.
1
latter also provided H and 31P NMR spectroscopic evidence of
Table 3. Concentration (c) and sweep rate (n) dependence of the CV
peak current ratio, ipc/ipa, for the RuII/RuIII redox couple of complex 2A.[a]
a clean reductive fragmentation from 2F back to 2A.
Analogous electrochemical oxidations of 3A–5A, however,
were not found to furnish similar oxidation products (3F–5F),
as observed for 2A. These were instead, quite surprisingly,
identified by NMR spectroscopy as the vinylidene species 3B–
5B (see the Supporting Information, Section 5). Although this
redox process was also found to be reversible, regenerating
the A species (see the Supporting Information, Section 4), the
mechanistic details of the electrochemical interconversion be-
tween 3A–5A and 3B–5B could not be elucidated further.
c
ipc/ipa
ipc/ipa
ipc/ipa
ipc/ipa
[mM]
(n=0.05 VsÀ1
)
(n=0.1 VsÀ1
)
(n=0.2 VsÀ1
)
(n=0.5 VsÀ1
)
4.6
2.3
1.15
0.58
0.29
0.84
0.92
0.95
0.98
1.00
0.90
0.95
0.98
0.99
1.00
0.95
0.98
0.99
1.00
1.00
0.99
0.99
1.00
1.00
0.99
[a] Recorded at a glassy carbon working electrode (d=3 mm) in CH2Cl2
with 0.1m Bu4NPF6 as supporting electrolyte.
Chem. Eur. J. 2016, 22, 7514 – 7523
www.chemeurj.org
7518
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
IR spectroscopy also provides strong evidence for electronic
communication between the DHA and the Ru metal center,
which acts as a strong electron donor. As expected, lower
energy CꢀC stretches in the IR region are observed for conju-
gates 2A–5A at approximately 2050 cmÀ1, similar to those of
19–23 (2041–2073 cmÀ1),[13b,21] in comparison to the trialkylsil-
yl-terminated derivatives of 10, 11, and 13 at approximately
2150 cmÀ1 (Table 5). The CꢀN stretching vibrations were
found at higher energies. The CꢀC stretching vibrations of the
free alkynes (10–13) could not be observed. Vinylidene deriva-
tives 2B–5B exhibited characteristic IR absorptions in the
region 1700–1800 cmÀ1. Notably, literature investigations
reveal that the RuIII radical cations of reference complexes 20
and 22–23 all show CꢀC stretching vibrations at approximate-
ly 1920 cmÀ1. The absence of such a signal in the spectrum of
2F indicates the absence of a RuIII acetylide motif. The pres-
ence of a stretch at 1708 cmÀ1, however, supports a vinyli-
dene-like structure for this complex.
Table 4. Selected 13C and 31P NMR spectroscopic data.[a]
Solvent
[Ru*]–C–C dC [ppm]
dppe dP [ppm]
2
2A
2B
C6D6
CD2Cl2
C6D6
140.22 (t, JCP =24.5 Hz)
n.o.
81.36 (s)
81.14 (s)
72.37 (s)
2
353.14 (t, JCP =15.9 Hz)
CD2Cl2
CD2Cl2
C6D6
n.o.
n.o.
72.59 (s)
72.26 (s)
81.69 (d, 2JPP =13.9 Hz)
81.56 (d, 2JPP =13.9 Hz)
72.90 (s)
81.52 (d, 2JPP =14.7 Hz)
81.28 (d, 2JPP =14.7 Hz)
72.67 (d, 2JPP =19.6 Hz)
72.05 (d, 2JPP =19.6 Hz)
2F[b]
3A
2
133.23 (t, JCP =24.6 Hz)
2
3B
4A
C6D6
C6D6
349.93 (t, JCP =16.1 Hz)
142.66 (t, JCP =24.7 Hz)
2
2
4B
5A
C6D6
C6D6
355.14 (t, JCP =16.2 Hz)
2
n.o.
81.71 (d, JPP =13.9 Hz)
81.51 (d, 2JPP =13.9 Hz)
72.87 (s)
2
5B
15
C6D6
C6D6
354.07 (t, JCP =16.1 Hz)
136.40 (t, JCP =25.0 Hz)
164.25 (t, JCP =23.9 Hz)
2
81.03 (s)
2
[a] n.o.=not observed. [b] Complex 2F was characterized as the crude
product obtained from the electrolysis experiment in 0.1m Bu4NPF6/
CD2Cl2 (see the Supporting Information, p. S7). [c] Ref. [13b]. [d] Ref. [21].
UV/Vis absorption spectroscopy of the [Ru*]–DHA conju-
gates reveals characteristic DHA absorptions at approximately
350 nm (Table 6). Higher energy shoulders near 250 nm are
found for all compounds including 2A–5A (Figure 7), parent
DHA 1A, and reference complexes 19–23.[13b,21] Similar to
these references, [Ru*]–DHA conjugates 2A–5A also exhibit
lower energy absorptions in the region between 300 and
500 nm. For 19–21, Paul et al.[13b] have attributed these transi-
tions to metal-to-ligand charge-transfer (MLCT) bands based
on analogy with the Fe analogs.[24] Spectroelectrochemical UV/
Vis/NIR studies of 20 and 22–23 by Fox et al.[21] have revealed
that the absorption bands in this region are increasingly red-
shifted with the increasing size of the aromatic ligand. Time-
dependent density functional theory (TD-DFT) calculations on
20, 22, and 23 have likened the origin of these transitions to
(d/p)–phenyl p* charge-transfer bands (HOMO![LUMO+3])
rather than purely MLCT bands. The absorptions exhibited by
2A–5A in this region are thus hypothesized to arise from
NMR spectroscopy revealed characteristic signals of which
selected data are provided in Table 4. Conjugates 2A–5A all
present expected 31P NMR signals near 81 ppm, corresponding
to the two phosphorous atoms of the dppe ligand, analogous
to those observed for reference complexes 19–23[13b,21] (CDCl3,
d=81.1–82.5). The two P nuclei of the [Ru*]–DHA conjugates
are inherently diastereotopic because of the stereocenter at
C8a of the DHA unit, which thus explains the AB doublets ob-
served for complexes 3A, 4A, and 5A in C6D6. The 31P NMR
spectrum of complex 2A, however, exhibits only a singlet, sug-
gesting that the chiral environment is not as strongly felt by
the two phosphorus nuclei. The 13C NMR spectra of 2A–5A ex-
hibit the expected triplet resonances, characteristic of the a-
acetylide carbon atoms, near 140 ppm, with coupling con-
2
stants of approximately JCP =25 Hz. In comparison, vinylidene
analogs 2B–5B all present strongly downfield-shifted 13C trip-
lets near d=350 ppm for the a-carbon of vinylidene, upfield-
shifted 31P NMR resonances around d=72 ppm, as well as a b-
1H singlets near d=4.5 ppm. Notably, the 31P NMR resonances
are observed as singlets for complexes 3B and 5B in contrast
to the pair of AB doublets observed for acetylides 3A and 5A
in C6D6.
Table 5. Selected IR absorptions (ATR, neat).[a]
n(CꢀC) or n(C=C) [cmÀ1
]
n(CꢀCN) [cmÀ1
]
1A
10-SiMe3
11-Si(iPr)3
13-Si(iPr)3
2A
–
2244 (vw)
2249 (w)
2254 (w)
2246 (w)
n.o.
[b]
2155 (s)
2156 (w)
2154 (m)
2053 (s)
[b]
[b]
Complex 2F, the oxidation product of 2A, was obtained and
spectroscopically characterized as the crude product of elec-
trolysis experiments in 0.1m Bu4NPF6/CD2Cl2. The 31P NMR
spectrum of 2F presents a sharp singlet at d=72.23 ppm, simi-
lar to vinylidene 2B at d=72.59 ppm. Complex 2A was found
to be notably sensitive to its surroundings. Although relatively
stable in C6D6 and neutralized CD2Cl2, any residual acid in a so-
lution of the latter resulted in a delicate equilibrium arising be-
tween the neutral (2A), vinylidene (2B), and oxidized (2F)
states, which varied according to the specific conditions. Figur-
es S66 and S67 (in the Supporting Information, p. S100) show
1H and 31P NMR spectra of a mixture of these three species,
thereby supporting their individual identities.
2B
1780 (m), 1717 (m)
1708 (m)
2057 (s)
1781 (m), 1717 (m)
2035 (s)
1782 (m), 1717 (m)
2064 (s)
n.o.
n.o.
n.o.
n.o.
2279 (w)
n.o.
2279 (w)
n.o.
2F[c]
3A
3B
4A
4B
5A
5B
15
1781 (m), 1716 (m)
2008 (s)
2194 (w)
[a] Relative signal intensity: vs=very strong, s=strong, m=medium, w=
weak, vw=very weak. n.o.=not observed. [b] Trialkylsilyl-terminated
alkyne. [c] The complex was deposited as a thin film by evaporation from
a 0.1m solution of Bu4NPF6 in CD2Cl2 obtained from electrolysis experi-
ments.
Chem. Eur. J. 2016, 22, 7514 – 7523
www.chemeurj.org
7519
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
is reasonable as we would expect dimer 2F, for steric reasons,
to exhibit characteristics of two non-interacting (out of plane)
vinylidene units.
Table 6. Selected UV/Vis absorption maxima (lmax [nm]) associated with
1A/D, the different states of [Ru*]–DHA conjugates 2A–5A, and refer-
ence complexes 19–23. Where available, values are reported in MeCN
and/or approximately 0.1m Bu4NPF6/CH2Cl2.[a]
The NMR (upfield 31P-shifts), IR (Ru=C=C stretches), electro-
chemical (redox behavior), and UV/Vis (blueshifted absorption
maxima) data strongly suggest that 2F exhibits a vinylidene-
like structure and properties. These observations are in good
accord with the hypothesized dimerization mechanism, and
with the proposed divinylidene [Ru*]–DHA dimer 2F (Figure 8).
Similar dimerization phenomena have previously been ob-
served for analogous Ru and other metal complexes.[22] In par-
ticular, the aforementioned divinylidene [{Cp(PPh3)2Ru}2-
(m-C4Ph2)][PF6]2, isolated by Bruce et al. by treatment of
Cp(PPh3)2RuC2Ph with AgC2Ph,[23] presents an analogous struc-
ture to that hypothesized for 2F.
A[b]
B
C
D
E
F[b]
1
2
(353) 357
–
(348) 402
(472) 480
–
(356) 358
(328) 334
(522) 528
(360) 364
(464) 480
322
–
–
–
–
(470) 475
(345) 328
(474) 474
(600) 600
(332) 336
(474) 474
–
–
–
–
–
–
–
–
–
470
–
–
–
–
–
–
–
–
–
–
–
–
396
–
–
–
–
–
–
–
–
–
–
474
893
–
336
495
909
(398)
–
–
(356) 358
(430) 428
–
(465)
–
–
3
4
5
(478) 480
–
–
–
–
–
–
–
(400) 398
–
–
–
–
19[c]
20[c]
20[d]
500
318
339
21[c]
22[d]
304
299
–
–
–
–
–
–
–
–
23[d]
382
–
–
–
–
[a] Values measured in MeCN are reported in brackets whereas those
measured in 0.1m Bu4NPF6/CH2Cl2 are without brackets. [b] Assignments
A–F do not apply to complexes 19–23; for these, the most intense neu-
tral RuII (A) and radical cation RuIII (F) absorptions are reported.
[c] Ref. [13b]. [d] Selected values are adapted to nm from cmÀ1 and were
measured in 0.1m Bu4NPF6/CH2Cl2; Ref. [21].
Figure 8. Hypothesized structure of [Ru*]–DHA dimer 2F.
Multimode switching studies
Each of the [Ru*]–DHA conjugates were subjected to detailed
photoswitching experiments by using UV/Vis absorption spec-
troscopy to follow the conversions. Full, six-state switching
was successfully achieved for [Ru*]–DHA complex 2A by using
redox, acid/base, light, and thermal stimulation (Scheme 8).
However, although each individual state could be achieved in
one or more steps, the outcome of the light stimulus depend-
ed on the state of the Ru moiety.
Figure 7. UV/Vis spectra of [Ru*]–DHA dyads 2A–5A in MeCN.
a combination of such processes. Notably, the meta-coupling
of complex 3A seems to greatly weaken the interaction of the
Ru metal center with the DHA fragment. The most redshifted
absorption is exhibited by complex 4A, where the [Ru*]–alkyn-
yl group is directly attached to the DHA core.
It is also notable that upon oxidation from RuII to RuIII, com-
plexes 20 and 22–23 systematically present redshifted UV/Vis
absorption maxima corresponding to a RuIII alkynyl system. In
contrast to this, oxidation product 2F presents a blueshifted
absorption maximum analogous to vinylidene 2B and similar
to 3B–5B as well. This indicates strong structural similarities
between the chromophoric systems of these complexes, which
Scheme 8. Conversion between the six states of [Ru*]–DHA complex 2A The
numbers inside the dotted boxes indicate the percentage regeneration of
the initial state (A) after a cycle of the type A!B!C!D!A or A!F!E!
D!A.
Chem. Eur. J. 2016, 22, 7514 – 7523
www.chemeurj.org
7520
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
The studies were conducted according to two methods.
Redox chemistry was performed in an optically transparent
thin layer electrochemical (OTTLE) cell having an approximately
0.2 mm path length in CH2Cl2 by using Bu4NPF6 as the support-
ing electrolyte. On the other hand, acid/base chemistry was
performed in a Schlenk-adapted UV cuvette with a path length
of 1 cm by using MeCN as the solvent. The spectroscopic re-
sults of the multimode switching experiments on 2A are sum-
marized in Figure 9.
tion, p. S65) compared with reference DHA 1A.[25] Both the
electrochemical oxidation of RuII to RuIII (2A to 2F) and proto-
nation of the acetylide to the vinylidine (2A to 2B) were suc-
cessful. In each case, the electronic communication across the
acetylide bridge has been altered and the strong electron-do-
nating power of the Ru metal center into the DHA core is ef-
fectively turned off. As the resulting chromophores are pre-
sumed to be structurally similar, it is not surprising to observe
strong similarities in the UV/Vis absorption spectra of 2B and
2F (Figure 9).
Thus, protonation or oxidation of 2A was found to re-estab-
lish the photoactivity of the DHA system. Photochemical iso-
merization of both 2B to 2C and 2F to 2E, could be accom-
plished by irradiation at their respective absorption maxima. It
was also observed that 2C and 2E present analogous features
in their UV/Vis absorption spectra, again suggesting structural-
ly analogous chromophoric features. Although the initial elec-
trochemical oxidation product of 2A can be assigned to the
hypothesized dimeric structure 2F, the subsequent photo-
switching process and structural assignment of state 2E is only
based on UV/Vis characterization.
A direct photoinduced ring-opening of 2A to 2D did not
proceed and may be rationalized by quenching of the photo-
excited state by interaction with the metal center.[12b] This has
previously been observed in DHA systems incorporating func-
tional groups such as anthraquinone,[12c] ferrocene,[12b] tetra-
[12j]
thiafulvalene,[12f] amines,[12d,g,h] and C60
and is presumably
either a result of light-induced electron transfer involving the
redox-active unit or electronic perturbation of the DHA chro-
mophore. Alternatively, the cross-conjugated system of 3A did
in fact allow photochemical isomerization to 3D; however,
with a much lower photoactivity (see the Supporting Informa-
Finally, stimulation of either 2F by reduction or of 2B by de-
protonation leads to the formation of 2D, which in both cases
exhibits the same two absorption maxima as well as a charac-
teristically broad shoulder (lmax =600 nm) in the UV/Vis absorp-
tion spectrum (Figure 9). In both cases, 2D thermally reverts to
2A, completing the cycle. The regeneration of 2A proceeds in
78% and 59% yield for the acid/base and electrochemical
cycles, respectively (illustrated in Scheme 8).
Although the acid/base switching experiments were con-
ducted in an excess of these reactants, neither the photochem-
ical ring-opening nor the thermal ring-closure is expected to
be affected by these conditions. Previous studies have indeed
demonstrated this for DHA 1A and others.[12g,26]
Overall, six states of switching have been achieved for [Ru*]–
DHA conjugate 2A. Not surprisingly, several of these resemble
each other in their UV/Vis absorption spectra even though
they were reached by fundamentally different pathways.
[Ru*]–DHA conjugates 3A, 4A, and 5A did not exhibit the
same degree of switchability as complex 2A. It was found that
electrochemical oxidation of these complexes in fact resulted
in the quantitative formation of vinylidene species 3B–5B (see
the Supporting Information, Section 5). Complex 3B could
thus be generated from 3A by protonation (see the Support-
ing Information, p. S66), or under electrochemically oxidizing
conditions. Furthermore, 3A was found to be photoactive in
both neutral and protonated (3B) forms, thus convertible to
3D and 3C, respectively. States 3C and 3D could subsequently
be interconverted by either acid/base or electrochemical stim-
ulation. A maximum of four states could thus be achieved for
complex 3A (Scheme 9).
Complex 4A could only be stimulated to quantitatively
achieve a maximum of two states (Scheme 9). Conversion of
4A to vinylidene 4B could be achieved by chemical (see the
Supporting Information, p. S71) and electrochemical methods
(see the Supporting Information, Section 5). In the case of 4A,
electrochemical oxidation also resulted in features in the UV/
Figure 9. Normalized UV/Vis spectra (A226nm =1) illustrating the multiple
states of [Ru*]–DHA dyads 2A, 4A, and 5A. Spectra of states in MeCN were
generated by chemical stimulation (acid/base) and acquired in a UV cuvette,
whereas spectra of states in CH2Cl2 were generated by electrochemical stim-
ulation in a 0.1m Bu4NPF6/CH2Cl2 medium and acquired in an OTTLE cell
(see the Supporting Information, Section 4).
Chem. Eur. J. 2016, 22, 7514 – 7523
www.chemeurj.org
7521
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
by single-crystal X-ray diffraction crystal structure determina-
tion.
Complexes 2A–5A were studied to determine the effect
and functionality of the ruthenium metal center on the proper-
ties of the DHA system and vice versa. Results from spectro-
scopic interpretations indicate significant metal–ligand interac-
tions between these two moieties compared with reference
and literature values. This could be seen not only from their
UV/Vis absorption spectra, but also by crystallographic (length-
ening of CꢀC bonds), IR (lowering of n(CꢀC) energy), and
13C NMR (downfield shift of the [Ru*]–C triplet) spectroscopic
studies. As the magnitude of these interactions increases, the
complexes exhibit intense UV/Vis absorptions with increasingly
redshifted absorption maxima.
Scheme 9. Multimode switching behavior of [Ru*]–DHA conjugates 3A, 4A,
and 5A.
Vis absorption spectra that might be assigned to a RuIII radical
cation, but this was not investigated in more detail (see the
Supporting Information, p. S88). Photochemical isomerization
of 4A to 4D was unfortunately unsuccessful, as was that of 4B
to 4C. Complex 5A behaved analogously to 4A (Figure 9). The
protonation of [Ru*]–DHA conjugates 2A–5A was, in all cases,
accompanied by clear color changes as shown in Figure 10.
Table 6 summarizes the UV/Vis absorption data for the achiev-
able states of each system (2A–5A) as well as reference com-
pounds 1A and 19–21.
Incorporation of the ruthenium acetylide moiety at varying
positions on the DHA core was found to have a great effect on
the light-induced photoswitching of the DHA to VHF, effective-
ly quenching the ring-opening isomerization for complexes
2A, 4A, and 5A in the neutral state. Interestingly, when at-
tenuating the electronic interaction through a meta-phenylene
bridge, the photoactivity was maintained (complex 3A). In
regard to the influence of cross-conjugation, it has previously
been observed that the kinetics of the VHF ring-closure reac-
tion are significantly less influenced by an electron-accepting
group when placed in cross-conjugation to the VHF compared
with direct linear conjugation.[27]
Conclusions
The synthesis and characterization of four organometallic
[Ru*]–DHA photoswitches (2A–5A) was accomplished. The
coupling of the ruthenium and DHA motifs in the last synthetic
step was found to be more reliably conducted through a metal-
ation reaction rather than a Sonogashira cross-coupling reac-
tion. For the two complexes where the intended 7-substituted
isomers were the desired outcome (4A, 5A), the metalation
route instead led to the 6-substituted products as well as
a few unexpected complexes, some of which were identified
Electrochemical and spectroelectrochemical studies employ-
ing cyclic voltammetry and UV/Vis absorption spectroscopy
support the full six-state switching potential of complex 2A,
whereas only four distinct and interconvertible states could be
obtained for 3A, and just two could be assigned for 4A and
5A. Oxidation of 2A is thought to transiently form the RuIII
radical cation, which subsequently dimerizes to species 2F.
Experimental Section
Unless otherwise stated, all reactions and manipulations of air- and
moisture-sensitive materials were performed under an argon or ni-
trogen atmosphere, with all glassware having been pre-dried in an
oven at 2008C overnight or flame-dried prior to use. No special
precautions to exclude air or water were taken during the subse-
quent workup procedures, which were conducted under atmos-
pheric conditions. All light-sensitive compounds, reactions, and
manipulations were excluded from light by either conducting the
procedures in a very dimly lit room, at night, or by masking the
glassware and equipment with aluminum foil. Common solvents
were dried and degassed prior to use (see the Supporting Informa-
tion, p. S2).
All commercially available reagents were used as received without
further purification. Nuclear magnetic resonance (NMR) spectra
were acquired with either a Bruker Ultrashield Plus 500 (1H at
500 MHz, 13C at 126 MHz) or a 500 MHz Varian spectrometer (1H at
500 MHz, 31P at 202 MHz), the former was equipped with a non-in-
verse cryoprobe. Chemical shift values are provided relative to in-
1
ternal solvent references for H and 13C NMR spectra and an inter-
nal reference tube containing H3PO4 (d=0.00 ppm) for 31P NMR
spectra. Infrared spectroscopy (IR) data was acquired with a Bruker
Alpha FTIR spectrometer, equipped with ALPHA’s Platinum ATR
single-reflection diamond ATR module instrument. Samples were
Figure 10. From left to right: Solutions of 2A, 3A, 4A, and 5A (top) and
protonated complexes 2B, 3B, 4B, and 5B (bottom) (ca. 6”10À5 m in
MeCN).
Chem. Eur. J. 2016, 22, 7514 – 7523
www.chemeurj.org
7522
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Hu, Y. Liu, G. C. Solomon, H. G. Kjaergaard, T. Bjørnholm, M. B. Nielsen,
M. W. Laursen, K. Nørgaard, Adv. Mater. 2013, 25, 4164–4170.
[12] a) J. Daub, J. Salbeck, T. Knçchel, C. Fischer, H. Kunkely, K. M. Rapp,
Angew. Chem. Int. Ed. Engl. 1989, 28, 1494–1496; Angew. Chem. 1989,
101, 1541–1542; b) J. Daub, S. Gierisch, J. Salbeck, Tetrahedron Lett.
1990, 31, 3113–3116; c) J. Achatz, C. Fischer, J. Salbeck, J. Daub, J.
Chem. Soc. Chem. Commun. 1991, 504–507; d) L. Gobbi, P. Seiler, F. Die-
derich, Helv. Chim. Acta 2001, 84, 743–777; e) M. Š. Petersen, L. Zhu,
S. H. Jensen, A. S. Andersson, A. Kadziola, K. Kilså, M. B. Nielsen, Adv.
Funct. Mater. 2007, 17, 797–804; f) M. Š. Petersen, A. S. Andersson, K.
Kilså, M. B. Nielsen, Eur. J. Org. Chem. 2009, 1855–1858; g) S. L. Broman,
M. Š. Petersen, C. G. Tortzen, A. Kadziola, K. Kilså, M. B. Nielsen, J. Am.
Chem. Soc. 2010, 132, 9165–9174; h) M. Š. Petersen, S. L. Broman, K.
Kilså, A. Kadziola, M. B. Nielsen, Eur. J. Org. Chem. 2011, 1033–1039; i) L.
Skov, M. Š. Petersen, S. L. Broman, A. D. Bond, M. B. Nielsen, Org.
Biomol. Chem. 2011, 9, 6498–6501; j) M. Santella, V. Mazzanti, M. Jevric,
C. R. Parker, S. L. Broman, A. D. Bond, M. B. Nielsen, J. Org. Chem. 2012,
77, 8922–8932; k) S. L. Broman, M. Jevric, M. B. Nielsen, Chem. Eur. J.
2013, 19, 9542–9548.
loaded directly or by evaporation from a suitable solvent. IR ab-
sorptions are reported in units of wavenumbers (cmÀ1). UV/Vis ab-
sorption methods are detailed in the Supporting Information (Sec-
tion 4). Thermal ellipsoid plots of individual molecules were gener-
ated by using Mercury 1.4.2 for Windows. Thin layer chromatogra-
phy (TLC) was conducted by using commercially available, precoat-
ed plates (silica 60) with a fluorescence indicator. With respect to
the DHA compounds, TLC was carried out in the absence of light,
and where a color change from yellow to red upon exposure to
UV light indicates a conversion to VHF. All melting points are un-
corrected. CCDC 1434224 (2A), 1434949 (4A), 1434948 (16), and
1434954 (18) contain the supplementary crystallographic data for
this paper. These data are provided free of charge by The Cam-
bridge Crystallographic Data Centre.
Acknowledgments
[13] a) C. Bitcon, M. W. Whiteley, J. Organomet. Chem. 1987, 336, 385–392;
b) F. Paul, B. G. Ellis, M. I. Bruce, L. Toupet, T. Roisnel, K. Costuas, J. F.
Halet, C. Lapinte, Organometallics 2006, 25, 649–665.
[14] K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Lett. 1975, 16, 4467–
4470.
The University of Copenhagen is acknowledged for financial
support.
[15] M. I. Bruce, B. G. Ellis, P. J. Low, B. W. Skelton, A. H. White, Organometal-
lics 2003, 22, 3184–3198; b) M. I. Bruce, M. A. Fox, P. J. Low, B. K. Nichol-
son, C. R. Parker, W. C. Patalinghug, B. W. Skelton, A. H. White, Organo-
metallics 2012, 31, 2639–2657.
[16] S. L. Broman, M. Jevric, A. D. Bond, M. B. Nielsen, J. Org. Chem. 2014, 79,
41–64.
Keywords: acetylides · azulenes · fulvenes · organometallic
compounds · photochromism
[1] a) A. R. Pease, J. O. Jeppesen, J. F. Stoddart, Y. Luo, C. P. Collier, J. R.
Heath, Acc. Chem. Res. 2001, 34, 433–444; b) A. Troisi, M. A. Ratner,
Small 2006, 2, 172–181; c) N. Weibel, S. Grunder, M. Mayor, Org. Biomol.
Chem. 2007, 5, 2343–2353; d) R. Klajn, J. F. Stoddart, B. A. Grzybowski,
Chem. Soc. Rev. 2010, 39, 2203–2237; e) S. Jan van der Molen, P. Liljer-
oth, J. Phys. Condens. Matter 2010, 22, 133001; f) S. J. van der Molen, P.
Liljeroth, Conductance Properties of Switchable Molecules, in Molecular
Switches, Vol. 2 (Eds.: B. L. Feringa, W. R. Browne), Wiley-VCH, Weinheim,
2011, pp. 719–777; g) S. L. Broman, M. B. Nielsen, Phys. Chem. Chem.
Phys. 2014, 16, 21172–21182; h) D. Kim, H. Jeong, H. Lee, W. T. Hwang,
J. Wolf, E. Scheer, T. Huhn, H. Jeong, T. Lee, Adv. Mater. 2014, 26, 3968–
3973.
[17] S. L. Broman, S. L. Brand, C. R. Parker, M. Š. Petersen, C. G. Tortzen, A.
Kadziola, K. Kilså, M. B. Nielsen, Arkivoc 2011, 4, 51–67.
[18] For a recent account of the synthesis and functionalization of DHA, see:
M. Cacciarini, S. L. Broman, M. B. Nielsen, Arkivoc 2014, 1, 249–263.
[19] a) P. M. Treichela, D. A. Komara, P. J. Vincentia, Synth. React. Inorg. Met.–
Org. Chem. 1984, 14, 383–400; b) F. Morandini, A. Dondana, I. Munari,
G. Pilloni, G. Consiglio, A. Sironi, M. Moret, Inorg. Chim. Acta 1998, 282,
163–172.
[20] D. Pilette, K. Ouzzine, H. Le Bozec, P. H. Dixneuf, C. E. F. Rickard, W. R.
Roper, Organometallics 1992, 11, 809–817.
[2] a) J. AndrØasson, U. Pischel, Chem. Soc. Rev. 2010, 39, 174–188; b) A. P.
de Silva, T. P. Vance, B. Wannalerse, M. E. S. West, Molecular Logic Sys-
tems, in Molecular Switches, Vol. 2 (Eds.: B. L. Feringa, W. R. Browne),
Wiley-VCH, Weinheim, 2011, pp. 669–696.
[21] M. A. Fox, R. L. Roberts, W. M. Khairul, F. Hartl, P. J. Low, J. Organomet.
Chem. 2007, 692, 3277–3290.
[22] a) R. S. Iyer, J. P. Selegue, J. Am. Chem. Soc. 1987, 109, 910–911; b) S.
Rigaut, F. Monnier, F. Mousset, D. Touchard, P. H. Dixneuf, Organometal-
lics 2002, 21, 2654–2661; c) D. A. Valyaev, O. V. Semeikin, N. A. Ustynyuk,
Coord. Chem. Rev. 2004, 248, 1679–1692; d) M. I. Bruce, B. G. Ellis, B. W.
Skelton, A. H. White, J. Organomet. Chem. 2005, 690, 792–801; e) M. I.
Bruce, B. G. Ellis, B. W. Skelton, A. H. White, J. Organomet. Chem. 2005,
690, 1772–1783; f) N. A. Ustynyuk, O. V. Gusev, L. N. Novikova, M. G. Pe-
terleitner, L. I. Denisovich, T. A. Peganova, O. V. Semeikin, D. A. Valyaev, J.
Solid State Electrochem. 2007, 11, 1621–1634; g) M. A. Fox, J. D. Farmer,
R. L. Roberts, M. G. Humphrey, P. J. Low, Organometallics 2009, 28,
5266–5269; h) M. I. Bruce, A. Burgun, F. Gendron, G. Grelaud, J.-F. Halet,
B. W. Skelton, Organometallics 2011, 30, 2861–2868; i) P. A. Schauer, P. J.
Low, Eur. J. Inorg. Chem. 2012, 390–411; j) P. J. Low, S. Bock, Electrochim.
Acta 2013, 110, 681–692.
[3] B. J. Coe, Chem. Eur. J. 1999, 5, 2464–2471.
[4] F. Pina, A. J. Parola, R. Gomes, M. Maestri, V. Balzani, Multistate/Multifunc-
tional Molecular-Level Systems: Photochromic Flavylium Compounds, in
Molecular Switches, Vol. 1 (Eds.: B. L. Feringa, W. R. Browne), Wiley-VCH,
Weinheim, 2011, pp. 181–226.
[5] M. Akita, Organometallics 2011, 30, 43–51.
[6] J. Manna, K. D. John, M. D. Hopkins, Adv. Organomet. Chem. 1995, 38,
79–154.
[7] a) B. Li, J.-Y. Wang, H.-M. Wen, L.-X. Shi, Z.-N. Chen, J. Am. Chem. Soc.
2012, 134, 16059–16067; b) H.-M. Wen, B. Li, J.-Y. Wang, L.-X. Shi, C.-N.
Chen, Z.-N. Chen, Organometallics 2013, 32, 1759–1765; c) Y. Liu, C. M.
Ndiaye, C. Lagrost, K. Costuas, S. Choua, P. Turek, L. Norel, S. Rigaut,
Inorg. Chem. 2014, 53, 8172–8188.
[8] a) Y. Tanaka, T. Ishisaka, A. Inagaki, T. Koike, C. Lapinte, M. Akita, Chem.
Eur. J. 2010, 16, 4762–4776; b) Y. Lin, H. Cui, P. Wan, J. Yin, S. Liu, Mol.
Cryst. Liq. Cryst. 2015, 608, 55–61.
[23] M. I. Bruce, G. A. Koutsantonis, M. J. Liddell, E. R. T. Tiekink, J. Organomet.
Chem. 1991, 420, 253–269.
[24] a) K. Costuas, F. Paul, L. Toupet, J.-F. Halet, C. Lapinte, Organometallics
2004, 23, 2053–2068; b) R. Denis, L. Toupet, F. Paul, C. Lapinte, Organo-
metallics 2000, 19, 4240–4251.
[9] K. A. Green, M. P. Cifuentes, T. C. Corkery, M. Samoc, M. G. Humphrey,
Angew. Chem. Int. Ed. 2009, 48, 7867–7870; Angew. Chem. 2009, 121,
8007–8010.
[25] H. Goerner, C. Fischer, S. Gierisch, J. Daub, J. Phys. Chem. 1993, 97,
4110–4117.
[10] F. Meng, Y.-M. Hervault, L. Norel, K. Costuas, C. Van Dyck, V. Geskin, J.
Cornil, H. H. Hng, S. Rigaut, X. Chen, Chem. Sci. 2012, 3, 3113–3118.
[11] a) S. Lara-Avila, A. V. Danilov, S. E. Kubatkin, S. L. Broman, C. R. Parker,
M. B. Nielsen, J. Phys. Chem. C 2011, 115, 18372–18377; b) S. L. Broman,
S. Lara-Avila, C. L. Thisted, A. D. Bond, S. Kubatkin, A. Danilov, M. B. Niel-
sen, Adv. Funct. Mater. 2012, 22, 4249–4258; c) T. Li, M. Jevric, J. R.
Hauptmann, R. Hviid, Z. Wei, R. Wang, N. E. A. Reeler, E. Thyrhaug, S. V.
Petersen, J. A. S. Meyer, N. E. Bovet, T. A. J. Vosch, J. Nygård, X. Qiu, W.
[26] C. R. Parker, C. G. Tortzen, S. L. Broman, M. Schau-Magnussen, K. Kilså,
M. B. Nielsen, Chem. Commun. 2011, 47, 6102–6104.
[27] K. Fjelbye, T. N. Christensen, M. Jevric, S. L. Broman, A. U. Petersen, A.
Kadziola, M. B. Nielsen, Eur. J. Org. Chem. 2014, 7859–7864.
Received: January 14, 2016
Published online on April 26, 2016
Chem. Eur. J. 2016, 22, 7514 – 7523
www.chemeurj.org
7523
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim