Oxidative electrochemical switching in dithienylcyclopentenes, part 2: Effect of substitution and asymmetry on the efficiency and direction of molecular switching and redox stability

Article abstract of DOI:10.1002/chem.200500163

The electrochemical and spectroelectrochemical properties of a series of C5-substituted dithienylhexahydro- and dithienylhexafluorocyclopentenes are reported. The effect of substitution at C5 of the thienyl moiety on the redox properties is quite dramatic

Full text of DOI:10.1002/chem.200500163

DOI: 10.1002/chem.200500163
Oxidative Electrochemical Switching in Dithienylcyclopentenes, Part 2:
Effect of Substitution and Asymmetry on the Efficiency and Direction of
Molecular Switching and Redox Stability**
Wesley R. Browne, JaapJ. D. de Jong, Tibor Kudernac, Martin Walko, Linda N. Lucas,
Kingo Uchida, Jan H. van Esch, and Ben L. Feringa*[a]
Abstract: The electrochemical and
spectroelectrochemical properties of a
series of C5-substituted dithienylhexa-
hydro- and dithienylhexafluorocyclo-
pentenes are reported. The effect of
substitution at C5 of the thienyl moiety
on the redox properties is quite dra-
matic, in contrast to the effect on their
photochemical properties. The efficien-
cy of electrochemical switching is de-
pendent both on the central cyclopen-
tene unit and on the nature of the sub-
stituents, whereby electron-donating
moieties favour oxidative electrochemi-
cal ring-closure and vice versa. Asym-
metrically substituted dithienylcyclo-
pentenes were investigated to explore
the ring-closure process in more detail.
The results indicate that electrochemi-
cally induced ring-closure occurs via
the monocation of the open form. In
the presence of electroactive groups at
C5 of the thienyl ring (e.g., methoxy-
phenyl) initial oxidation of these
groups is followed by intermolecular
electron transfer, which drives ring-clo-
sure of the open forms.
Keywords: cyclization
·
electro-
chemistry · photochromism · redox
chemistry · UV/Vis spectroscopy
Introduction
tems. These studies have focussed on understanding the fac-
tors which control their photophysical and electrochemical
properties. Although the photochromic behaviour of materi-
als based on diarylethenes, fulgides[7] and spiropyrans has
been investigated extensively in recent years and has been
In recent years, photo- and electrochromic materials have
received considerable attention for potential application in
visual-display and molecular information-storage technolo-
gies.[1] Systems based on transition metals such as rutheniu-
m(ii)[2,3] and organic systems based on spiropyrans[4] and di-
arylethenes[5] have proven to be especially suitable for such
applications. The effectiveness of molecular-based materials
requires that their properties can be tuned readily to suit
particular functions, a requirement which has prompted
many studies on transition metal- and organic-based[2,6] sys-
adapted to drive changes in bulk material properties,[8–13]
a
detailed understanding of the electrochemical properties of
these systems is not yet available, and relatively few studies
have been reported in the literature.[14–16]
In Part 1, we examined the electrochemical and spectro-
electrochemical properties of two representative sets of di-
thienylethene switches (1Ho/1Fo and 2Ho/2Fo, Scheme 1)
and investigated the mechanism and solvent dependence of
the electrochemically driven ring-opening/closing process-
es.[17] It is clear from our earlier work that electrochemical
switching processes are complex and involve interconversion
between several cationic species,[16] so assignment of the
mechanism involved (e.g., ring-closure via mono- or dicat-
ionic species) requires the investigation of asymmetrically
substituted systems. Furthermore, although the influence of
peripheral substituents (i.e., at C5 of the thienyl ring) on
photochemical ring opening/closing has been examined in
detail, the influence of these substituents on the electro-
chemical properties of these compounds has not been estab-
lished to date. Here we extend our investigations to examine
[a] Dr. W. R. Browne, J. J. D. de Jong, T. Kudernac, M. Walko,
Dr. L. N. Lucas, Prof. K. Uchida, Dr. J. H. van Esch,
Prof. Dr. B. L. Feringa
Organic and Molecular Inorganic Chemistry
Stratingh Institute, University of Groningen
Nijenborgh 4, 9747 AG, Groningen (The Netherlands)
Fax : (+31)50-363-4296
E-mail: b.l.feringa@rug.nl
[**] For Part 1 see: W. R. Browne, J. J. D. de Jong, T. Kudernac, M.
Walko, L. N. Lucas, K. Uchida, J. H. van Esch, B. L. Feringa, Chem.
Eur. J. 2005, 11, 6414–6429; DOI: 10.1002/chem.200500162.
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
6430
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6430 – 6441

FULL PAPER
procedures reported previous-
ly.[18] Compounds 4H, 5H, 7H
and 11H–13H were prepared
from 8H (see Scheme 2), and
7F, 9F and 13F were prepared
from 8F. Compound 11H was
prepared by Suzuki coupling of
the bis(boronic ester) A, ob-
tained from 8H, with 2-bromo-
thiophene (Scheme 2). Similar-
ly, 5H was prepared, via A,
from 8H and 1-bromo-4-meth-
ylsulfanylbenzene, but the
same reaction failed with
either 4-bromobenzenethiol or
4-bromobenzenethioacetate,
Scheme 1. Dithienylcyclopentene switches described in the text: suffix o denotes open form, c denotes closed
form; H and F denote hexahydrocyclopentene- and hexafluorocyclopentene-based compounds, respectively.
possibly due to deactivation of
palladium catalyst by the free
thiol groups. To circumvent
the effect of substitution of the dithienylethene switches at
C5 of the thienyl rings (Scheme 1).[18] Of particular interest
is the ability to tune the redox properties of the dithienyl-
ethenes independently or at least semi-independently from
their electronic and photochemical properties. In addition,
asymmetrically substituted dithienylcyclopentenes (i.e., 7H/
F and 13H/F) are investigated to explore the effect of redox
asymmetry on the electrochemical properties of these com-
pounds and to establish the role of peripheral redox groups
in the electrochemical processes observed.
this, 4H was prepared[19] by Suzuki coupling of A with 1-
bromo-4-tert-butylthiobenzene (B) to yield C (45%). Subse-
quent substitution of the tBu groups by acetyl groups pro-
vided 4H.[20] The dialdehyde 9F was obtained by lithia-
tion[18] of 8F, followed by quenching with DMF. Compound
12H was obtained by esterification of the corresponding
diacid in ethanol with a catalytic amount of 30% aqueous
HCl. The asymmetric derivatives 7H and 13H were pre-
pared in two steps starting from 8H. The first Suzuki cou-
pling with bromobenzene was performed under five times
more dilute conditions than employed for the symmetric de-
rivatives (to minimise disubstitution) to yield 13H. After pu-
rification of the monosubstituted product 13H, the second
Suzuki coupling was performed under standard conditions
to yield 7H (63%).[21] Similarly, 7F was prepared from 8F
by an analogous route, but lithiation was performed in di-
Results and Discussion
Synthesis: The symmetric derivatives 1H–3H, 6H, 8H–
10H, 1F–3F and 8F (Scheme 1) were prepared according to
Scheme 2. Synthesis of asymmetric 5-aryl-hexahydrothienylcyclopentenes. a) nBuLi, THF (or Et2O for hexafluorocyclopentene-based compounds),
298 K; b) B(OBu)3, THF, 298 K; c) ArBr, [Pd(PPh3)4], 2m aqueous Na2CO3, ethylene glycol, THF, reflux. The hexafluorocyclopentene analogues were
prepared by similar routes.
Chem. Eur. J. 2005, 11, 6430 – 6441
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6431

B. L. Feringa et al.
ꢀ
ethyl ether to minimise side reactions involving C F bonds
of the hexafluorocyclopentene ring. All compounds were
prepared in up to 500 mg scale and were characterised by
1H and 13C NMR spectroscopy and HRMS (see Experimen-
tal Section for details).
Electronic properties: In an earlier contribution the elec-
tronic properties (in toluene) of several of the compounds
were reported.[18] Here a more detailed examination of the
electronic spectra in acetonitrile solution is presented. The
effect of incorporating a hexahydrocyclopentene group in-
stead of a hexafluorocyclopentene group on the absorption
spectra of the dithienylethenes (e.g., 1Ho: lmax =278 nm,
1Fo: lmax =285) is quite modest, especially considering the
very large difference in the redox properties (vide infra) of
the hexahydrocyclopentene- (1Ho–13Ho) and hexafluoro-
cyclopentene-based (1Fo–13Fo) compounds. In contrast, the
effect of substitution at C5 of the thiophene ring is much
more pronounced: lmax changes by up to 50 nm (ca.
5000 cmꢀ1). The UV absorption spectra of 1Ho–6Ho are
shown in Figure 1a. The data suggest that in the open form
the major influence on the HOMO–LUMO energy gap is
provided by substitution at C5.
In the closed form (i.e., 1Hc–6Hc, Figure 1b) the effect
of substitution at C5 of the thienyl ring is similar to that ob-
served in the open state. The molar absorptivity shows a
marked increase accompanied by a bathochromic shift with
increasing electron-donating character of the substituent in
the para position of the phenyl ring (Table 1).[22] The
changes observed in the near-UV region of the spectrum
(250–400 nm) on substitution are considerable, and the
changes in the band at about 360 nm are similar to those of
the band at about 550 nm. The bands at higher energy show
Figure 1. UV/Vis absorption spectra of a) 1Ho–6Ho and b) 1Hc–6Hc.
a different dependence on substituents than the visible ab-
sorption bands. For the non-aryl-substituted compounds
(e.g., 8H–10H, 8F, 9F) only the chloro substituted com-
Table 1. Electronic and photochemical properties of open and closed forms in CH3CN.
Substituents
PSS[a]
Open form lmax(abs) [nm]
(e [103 cmꢀ1 mꢀ1])[b]
Closed form lmax(abs) [nm]
(e [103 cmꢀ1 mꢀ1])[b]
1H
1F
2H
2F
3H
3F
4H
5H
6H
7H
7F
8H
8F
9H
9F
R1 =R2 =Ph
0.99
0.92
0.99
0.99
0.99
0.99
0.96
278 (18), 303 (S)
285 (33)
284 (28), 308 (S)
296 (38)
267 (15), 287 (S), 331 (I, 13), 349 (5.2), 360 (6.4)
307 (I, 22), 308 (22), 366 (8.8), 380 (9.1)
237 (14), 273 (S), 303 (24), 329 (12), 345 (9.5), 348 (S)
321 (I, 21), 344 (25), 376 (S)
287 (36), 366 (I, 11), 387 (14)
341 (I, 9.4), 374 (5.6), 395 (S)
344(I, 19), 364 (S)
345 (I), 364 (S)
340 (I, 17), 355 (18), 366 (18)
217 (23), 299 (25), 329 (I, 13), 346 (13), 363 (S)
273 (15), 313 (21), 317 (I, 20), 338 (20), 362 (S), 379 (S)
527 (8.8)
588 (12)
519 (13)
593 (18)
575 (22)
586 (5.9)
540 (18)
541
532 (24)
524 (18)
591 (18)
444
R1 =R2 =Ph
R1 =R2 =PhOMe
R1 =R2 =PhOMe
R1 =R2 =PhCN
R1 =R2 =PhCN
R1 =R2 =PhSMe
R1 =R2 =PhSAc
R1 =R2 =PhBr
R1 =Ph, R2 =PhOMe
R1 =Ph, R2 =PhOMe
R1 =R2 =Cl
229 (31), 300 (35)
232 (17), 315 (9.0)
295 (S), 320 (35)
295 (49), 322 (49)
282 (47), 313 (S)
280 (33), 312 (S)
256 (S), 292 (36)
236 (19)
242 (22), 305 (4.1)
269 (25), 315 (8.2)
263 (29), 294 (18)
234 (22), 273 (S)
281 (14), 295 (18)
312
[c]
–
0.99
0.99
0.94
[d]
–
R1 =R2 =Cl
0.45
0.95
0.95
264 (I, 3.9), 287 (I, 2.0), 305 (I, 2.0), 326 (2.7), 338 (S)
218 (I, 11), 236 (I, 12), 336 (I, 5.7), 372 (9.3)
239 (I, 11), 320 (I, 4.5), 344 (5.1), 395 (6.0)
504 (2.1)
577 (8.2)
614 (6.4)
R1 =R2 =CHO
R1 =R2 =CHO
R1 =R2 =Me
[d]
[e]
10H
11H
11F[f]
12H
13H
13F
–
–
R1 =R2 =thienyl
R1 =R2 =thienyl
R1 =R2 =CO2Et
R1 =Cl R2 =Ph
R1 =Cl R2 =Ph
0.98
229 (10), 311 (18), 347 (8.5), 356 (9.5), 369 (S)
519 (13)
605
540 (21)
485 (9.8)
548 (9.6)
0.78
0.75
0.99
258 (27), 297 (S)
256 (18), 273 (18), 310 (S)
255 (22), 285 (19)
316 (I, 21), 354 (21)
246 (19), 280 (21), 317 (I, 12), 317 (S)
240 (15), 298 (I, 17), 298 (17), 348 (7.8), 360 (S)
[a] PSS=maximum percentage of closed state formed by photolysis at l=313 nm in CH3CN. [b] I=isosbestic point, S=shoulder. [c] Very low solubility
in CH3CN precluded accurate measurement by 1H NMR spectroscopy. [d] PSS not determined due to overlap of the signals for the open and closed
forms in 1H NMR spectrum. [e] Not determined; no clear signal changes occur during irradiation at 313 nm. [f] From reference [16a].
6432
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6430 – 6441

Oxidative Electrochemical Switching in Dithienylcyclopentenes
FULL PAPER
pounds (8H/F) display markedly different (hypsochromi-
cally shifted) spectroscopic properties compared to the aryl-
substituted compounds.
states achieved indicate that, as in toluene,[18] in acetonitrile
the quantum yields for photochemical ring opening are con-
siderably lower than those of ring closure. For 8F, the very
low PSS is most likely due to the weak absorption of 8Fo
and the relatively strong absorption of 8Fc at the wave-
length employed for ring closing (l=313 nm).
For all C5-substituted derivatives a bathochromic (red)
shift of 2200ꢁ250 cmꢀ1 is observed on substitution of the
hexahydrocyclopentene unit by the hexafluorocyclopentene
unit (i.e., 1Hc!1Fc, 3Hc!3Fc, etc.). For the majority of
C5-substituted compounds the effect of substitution (i.e.,
the shift in the lowest absorption band compared with 1Hc
or 1Fc) is similar and indicates a reduction in the HOMO–
LUMO gap. The very modest bathochromic shift of cyano-
phenyl-substituted hexafluoro compound 3F compared with
3Hc suggests that the cyanophenyl group contributes to
either the HOMO or LUMO orbital (Table 1).
Redox properties: In our previous contribution,[17] the elec-
trochemical and spectroelectrochemical properties of 1H/F
and 2H/F were explored. In particular, the dependence of
the direction of oxidative switching was examined and a de-
tailed picture of the various electrochemical processes was
developed (see Scheme 3in reference [17]). Here these
studies are extended to examine the effect of substituents
on the electrochemical properties of the dithienylethene
core and to establish the general applicability of the model
developed in our earlier contribution.[17] In addition, asym-
metrically substituted dithienylethene switches were exam-
ined to answer two key questions: first, does ring closing/
opening occur via the mono (e.g., 1Ho+, 1Fc+) or dication-
ic species (e.g., 1Ho2+, 1Fc2+) and secondly, does oxidation
of peripheral groups (e.g., the methoxyphenyl groups of
2Fo) lead to immediate ring closure, or is a model involving
intramolecular electron transfer more appropriate to ration-
alise the observations.
Electrochemical data for the hexahydro and hexafluoro
compounds in both open and closed forms are presented in
Table 2.[23] All redox processes undergo an anodic shift
(210–790 mV) on substitution of the hexahydro- by the
hexafluorocyclopentene unit.[17] The anodic shift observed
for the open forms is similar to that observed for the first
oxidation process in the closed forms, with the exception of
2Ho/2Fo, 3Ho/3Fo and 7Ho/7Fo. The smaller anodic shift
Photochemistry: Ring-opening (e.g., 1Hc!1Ho) and ring-
closing (e.g., 1Ho!1Hc) of all compounds in acetonitrile
(by irradiation at l>460 nm and at l=313 nm, respectively)
was monitored by UV/Vis spectroscopy. Photostationary
1
states (PSS; Table 1) were characterised by H NMR spec-
troscopy. On irradiation at l=313 nm, absorption bands in
the visible region appeared (300–700 nm), and on subse-
quent irradiation of the closed form (at l>460 nm) com-
plete recovery of the open form was observed. This process
could be repeated at least three times with minimal degra-
dation (<5%). However, dependence of the extent of deg-
radation on solvent was observed, whereby solvent purity
was an important factor. This suggests that the degradation
observed may not, for the most part, be inherent to the di-
thienylethene systems. With the exception of 5H, 8H/F,
12H and 13H, all compounds show PSS between 90 and
99% on 313 nm irradiation, despite the nonzero overlap in
the absorptions of the open and closed forms. The high PSS
Table 2. Redox properties of dithienylcyclopentenes (0.3–1.1 mm in 0.1m TBAP/CH3CN).
Open form [V vs SCE]
Closed form E1/2 [V vs SCE]
(Ep,a where irr)[a] (Ep,c where irr)[a]
c/c+/c2+
Ep,a
Ep,c
DE [mV][b]
1H
1F
2H
2F
3H
3F
4H
5H
6H
7H
7F
8H
8F
9H
9F
R1 =R2 =Ph
1.16 (irr)
1.59 (irr)
0.99 (irr)
1.20 (irr)
1.29 (irr)
1.46 (irr)
0.89 (irr)
0.97 (irr)
ꢀ2.53(irr)
ꢀ1.75 (irr)
0.67, 0.43
0.85 (qr)
0.45, 0.32
0.67
0.78, 0.53
1.05 (irr)
0.44, 0.30
0.50, 0.35
0.69, 0.47
0.56, 0.37
0.80 (qr)
0.815 (irr), 0.63
1.15 (irr)
1.13(irr), 0.86 (qr)
1.35 (irr)
0.47, 0.37
0.53, 0.40
0.82
ꢀ1.74, ꢀ2.03240
ꢀ1.13(qr)
R1 =R2 =Ph
R1 =R2 =PhOMe
R1 =R2 =PhOMe
R1 =R2 =PhCN
R1 =R2 =PhCN
R1 =R2 =PhSMe
R1 =R2 =PhSAc
R1 =R2 =PhBr
R1 =Ph R2 =PhOMe
R1 =Ph R2 =PhOMe
R1 =R2 =Cl
ꢀ1.84 (irr), ꢀ2.16 (irr)
ꢀ1.16 (irr), ꢀ1.46 (irr)
ꢀ1.4 (irr)
130
250
ꢀ1.70 (irr)
ꢀ2.06 (irr)
ꢀ0.83(irr)
140
150
220
190
ꢀ1.70 (irr)
1.15 (irr)
1.68 (irr), 1.08 (irr)
1.54 (irr), 1.34 (E1/2
1.37 (irr)
2.03(irr)
1.46 (irr)
2.25 (irr)
1.105 (irr)
1.06 (irr)
1.41 (irr)
ꢀ1.81 (irr), ꢀ2.11 (irr)
)
ꢀ1.73(irr)
ꢀ1.70 (irr)
ꢀ1.50 (irr)
125
240
R1 =R2 =Cl
ꢀ1.10 (irr)
ꢀ0.42 (irr)
ꢀ2.28 (qr)
R1 =R2 =CHO
R1 =R2 =CHO
R1 =R2 =Me
10H
11H
11F[c]
12H
13H
13F
100
130
R1 =R2 =thienyl
R1 =R2 =thienyl
R1 =R2 =CO2Et
R1 =Cl, R2 =Ph
R1 =Cl, R2 =Ph
1.53(irr), 1.43(irr)
1.57 (irr), 1.26 (irr)
1.87 (irr), 1.67 (irr)
1.12 (irr), 0.81
0.67 (irr), 0.52 (qr)
1.03(irr), 0.92 (irr)
ꢀ1.22, ꢀ1.36
280
ꢀ1.67 (irr), ꢀ1.89 (irr)
ꢀ1.04 (irr), ꢀ1.45 (irr)
-1.62 (irr)
[a] irr=irreversible, qr=quasireversible.[23] [b] DE=separation between the first and second oxidation processes of the closed form. Where not otherwise
stated, DE<50 mV. [c] Values for 11F are from reference [16a].
Chem. Eur. J. 2005, 11, 6430 – 6441
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6433

B. L. Feringa et al.
of these compounds is due to the non-innocent nature of the
substituted aryl groups of the hexafluoro compounds in the
open form (vide infra).[17] For the aryl-substituted com-
pounds (2H–6H) a correlation between the inductive effect
of the substituent (sI),[24] the redox properties of the open
and closed forms (see Supporting Information, Figure S1
and Table 2) and the lack of any correlation with resonance
parameters suggest that delocalisation of the HOMO over
the aryl ring is not an important feature of these systems. In-
terestingly, the trend holds both for the neutral compounds
and for the monocations, and this suggests that even in the
oxidised state delocalisation of the SOMO over the phenyl
rings is not significant.[25]
Overall the anodic shift observed on substitution of the
hexahydro unit by the hexafluoro unit for the first oxidation
process is less than that observed for the first reduction pro-
cess (vide infra). This implies that a reduction in the
HOMO–LUMO gap occurs and is in agreement with the
bathochromic shift observed by UV/Vis spectroscopy. For
3Hc/3Fc the anodic shift is comparable for both the first re-
duction (530 mV) and first oxidation processes (520 mV),
which implies a modest (400 cmꢀ1) red shift between 3Hc
and 3Fc. This is in agreement with the shift observed
(330 cmꢀ1) in the UV/Vis spectra (vide supra).
band represents the absorption to the Franck–Condon state
from the lowest vibronic state. The relationship holds re-
gardless of the nature of the cyclopentene group (i.e., hexa-
hydro vs hexafluoro).
lmaxðabsÞ=cmꢀ1 ¼ Ec=cþ =cmꢀ1ꢀEc=cꢀ cmꢀ1 þ 2300 cmꢀ1
ð1Þ
Substituent dependence of reversibility of redox processes,
solvent effects and the comproportionation constant Kc: The
effect of substituent on the reversibility of the redox proc-
esses is quite striking. In general the reversibility of the first
and second oxidation processes improves with increasing
electron-donor strength of the substituent. In agreement
with the solvent dependence of the redox properties ob-
served for 1Hc/1Ho,[17] the strong dependence of the stabili-
ty of the various oxidation states and of electrochemical ring
closure/opening on the thienyl C5 substituent is related to
the electronic stabilisation of the cationic species formed
(e.g., Hc+, Hc2+ etc). Irreversibility was found to be due to
chemical instability (thermally related) rather than to intrin-
sic properties of the electrochemical process itself. The use
of strongly electron withdrawing groups resulted in almost
complete irreversibility in the oxidation of the closed form.
For the hexafluorocyclopentene-based compounds the irre-
versibility was more pronounced, and only the methoxy-
phenyl- (2F) and phenyl-substituted (1F) compounds show
significant reversibility.[17] The solvent dependence of the
redox properties of 1Ho and 1Hc were described earlier.[17]
The effect of donor substituents on the solvent depend-
ence of redox and electronic properties was explored for
5Ho/5Hc. Compound 5H was chosen due to its redox sta-
bility in the closed form and the relatively large effect of the
acetylthiophenyl substituent on the oxidation potentials of
both the open and closed forms. As for 1Hc, a correlation
was observed between DE of 5Hc and Guttmann solvent
donor numbers[28] (see Supporting Information, Figure S2
and Table S2); DE decreases with increasing solvent donor
strength.
Spectroelectrochemical correlation: Within the present data-
set, the potentials of the first oxidation (Ec/c+), the first re-
duction (Ec/cꢀ) process and the lowest energy visible absorp-
tion band lmax(abs) for a sizeable group of ring-closed sub-
stituted dithienylhexahydro- and dithienylhexafluorocyclo-
pentenes are available (Tables 1 and 2). As has been ob-
served for a wide range of redox-active systems,[26] lmax(abs)
varies systematically (ꢁ1000 cmꢀ1) with the HOMO–
LUMO energy gap determined electrochemically (Figure 2).
In addition, the empirical constant of 2300 cmꢀ1 [Eq. (1)] is
close to that reported previously for inorganic systems.[27]
The difference between the HOMO–LUMO energy gap
measured electrochemically and that determined from elec-
tronic data is due to the fact that lmax of the absorption
Importantly, however, the magnitude of DE for 5Hc is
consistently lower than for 1Hc and the difference increases
with decreasing solvent donor strength, that is, 5Hc is less
sensitive to solvent than 1Hc. The reduction in DE with in-
creasing electron-donating power of the substituent can be
rationalised on the basis that the electron-donating proper-
ties of the thioacetate group serve to stabilise the positive
charge of the monocation and thereby reduce the potential
influence of solvent and electrolyte in stabilising the mono-
cation.
For 1Ho and 1Hc, temperature, solvent and electrolyte
dependence of the formation of 1Hx (see Scheme 3in refer-
ence [17] for details) was apparent. The formation of 1Hx
from 1Hc2+ was found to be a thermally activated process
and was inhibited by increasing solvent donor strength. For
5Ho/5Hc, the formation of an equivalent species (5Hx) was
not as apparent as for 1Ho/1Hc. In acetonitrile, the forma-
tion of 5Hx was not observed at 298 K and, although the
Figure 2. Correlation of the electrochemical and spectroscopic properties
of (1–3, 5, 7, 11–13)Hc and (1–3, 8, 9, 11, 13)Fc (R2 =0.85, intercept
2300 cmꢀ1).
6434
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6430 – 6441

Oxidative Electrochemical Switching in Dithienylcyclopentenes
FULL PAPER
formation of 5Hx was observed in solvents such as THF and
dichloromethane, the rate of formation was at least one
order of magnitude lower than for 1Hx.
The formation of species such as 1Hx and 5Hx appears to
be a common feature of dithienylethenes. Its formation
competes with that of the closed form (e.g., 1Hc2+ and
5Hc2+), and by choice of solvent, temperature and the
nature of the substituents, the formation of this species can
be controlled. The very low oxidation potential (e.g., 1Hx
ca. ꢀ0.1 V and 1Fx ca. 0.3V) indicates that in this state the
HOMO of this species is destabilised compared with both
the closed and open forms. Hence, whilst in the dicationic
form, the high HOMO level leads to stability, it is clear that
on reduction it becomes very unstable and undergoes iso-
merisation to 1Hc.
The effect of substituents on the stabilisation of the first
oxidation process with respect to disproportionation [1/Kc,
Eqs. (2)–(4), Swc =dithienylcyclopentene] can be determined
by examining the separation DE (Table 2) between the first
and second oxidation processes of the closed form of the di-
thienylcyclopentenes. Overall, electron-donating groups
(2Hc, 4Hc, 5Hc and 11Hc) reduce the magnitude of DE in
comparison to that of 1Hc, while electron-withdrawing
groups increase the magnitude of DE (12Hc). For 3Hc and
5Hc only a modest effect was observed, presumably due to
the balance of electron-donating (resonance) and -withdraw-
ing (inductive) effects of the substituents. For 8Hc (Cl) and
10Hc (Me), the similarly low values of DE (ca. 100 mV) are
surprising considering the opposite electronic character of
these substituents. This can be rationalised, however, by
considering the origin of DE (and hence the comproportio-
nation constant Kc). There are two contributing factors to
the magnitude of DE: the extent to which the initial oxida-
tion step involves the entire dithienylethene system and
electrostatic interaction between the oxidised and unoxi-
dised trans-butadiene systems (see Scheme 3). From an elec-
trostatic viewpoint, increasing the electron-donating proper-
ty of the substituents will stabilise the positive charge
formed during the first oxidation process and thereby
reduce the barrier to the second oxidation process and
hence decrease DE. Similarly, electron-withdrawing substitu-
ents in the C5 position would stabilise the HOMO of the
thienyl groups and increase the HOMO-mediated interac-
tion between the two thienyl moieties.
Scheme 3. Delocalisation/localisation of the SOMO in the closed state.
Depending on the nature of the substituent, the system will favour either
of the mesomeric states.
(Scheme 1). For the asymmetric compounds two limiting sit-
uations can be envisaged: 1) if the oxidation steps involve
independent oxidation of each half of the bis-substituted di-
thienylcyclopentene and no communication between the
two halves occurs, then the properties of the compound will
be a superimposition of the properties of the parent mole-
cules (localised model); 2) if the redox processes involve the
entire dithienylethene system, the properties of the asym-
metric compounds will be an average of those of the two
symmetric parent compounds (delocalised model). It would
be expected that neither of these limiting cases would be ob-
served, but rather an intermediate situation is present. A
further complication are the ring opening/closing processes,
for which it is unclear whether the switching processes occur
via the mono- or dicationic species. This aspect is potentially
addressable by investigating the asymmetrically substituted
compounds. Overall, the central question is to what extent
delocalisation of charge occurs in the closed and open
forms.
Electrochemical properties of 7H/7F: For 7Ho the first oxi-
dation step at 1.08 V is a one-electron process (Figure 3),
while a second oxidation occurs at 1.68 V. The potential of
the first process is intermediate between those of the parent
symmetric compounds 1Ho and 2Ho (1.16 and 0.99 V). For
7Hc two reversible oxidation process are observed at 0.37
and 0.56 V and, as for 7Ho, the potential of each process
and the separation between the first and second oxidation
processes (DE=190) are intermediate between those of the
parent compounds 1Hc (DE=240) and 2Hc (DE=130).
Similarly, the reduction potentials for 7Hc are intermediate
between those of 1Hc and 2Hc. The intermediacy of the
electrochemical properties of 7H between those of 1H and
2H (ꢀ1.81 versus ꢀ1.74 and ꢀ1.84 V) suggests that the oxi-
dation process is delocalised (at least partially) over the
Swc þ Sw2cþ Ð 2 Swþc
Kc ¼ ½Swþc ꢂ2=½Swcꢂ½Swc2þꢂ
Kc ¼ expðDE=25:69Þ
ð2Þ
ð3Þ
ð4Þ
Asymmetrically substituted dithienylethenes: To probe fur-
ther the effect of substituents on the electrochemistry of the
dithienylcyclopentenes, the asymmetrically substituted com-
pounds 7H/7F (R1 =Ph, R2 =PhOMe), 13H/13F (R1 =Cl,
R2 =Ph) and 12H (R1 =Cl, R2 =Ph) were investigated
Chem. Eur. J. 2005, 11, 6430 – 6441
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6435

B. L. Feringa et al.
(2Fo!2Fc). The difference in the chemical reactivity of the
oxidation products of 1F and 2F can be ascribed to the in-
volvement of the methoxyphenyl unit in 2F. For 7F a more
complex electrochemical behaviour is observed than for 7H.
For 7F two closely separated oxidation processes are ob-
served at 1.34 and 1.54 V (Figure 4). The first process is
electrochemically reversible (at high scan rates, >5 Vsꢀ1,
the subsequent chemical reaction is avoided, Figure 5)
whilst the second process at 1.54 V is fully irreversible.
Figure 3. a) Cyclic voltammograms of 1 equiv ferrocene and 1 equiv 7Ho
with a 10 mm Pt microelectrode in 0.1m TBAP/CH3CN, 0.01 Vsꢀ1 show-
ing Fc/Fc+ redox couple at 0.39 V and first oxidation process of 7Ho at
1.08 V (a), and at a 2 mm glassy carbon electrode at 0.1 Vsꢀ1 in the pres-
ence (b: bottom) and absence (b: top) of 1 equivalent of ferrocene.
entire molecule and not on individual thienyl units. If the
oxidation processes were localised then the first redox pro-
cess for both 7Ho and 7Hc would be expected to be close
to those observed for 2Ho and 2Hc, respectively.
Figure 4. Cyclic voltammogram of 7Fo at 0.1 Vsꢀ1 (a) and 10 Vsꢀ1 (b),
ꢀ0.2 V to 1.5 V (solid) and ꢀ0.2 V to 1.8 V (dashed) in 0.1m TBAP/
CH3CN.
As with 1Ho and 2Ho, electrochemically induced ring
closure is observed for 7H, which indicates that the proper-
ties of the asymmetric dithienylhexahydrocyclopentene 7H
are largely those observed for the parent compounds. The
one-electron nature of the first redox process observed for
7Ho suggests that after single-electron oxidation of the
open form ring closure occurs, that is, from 7Ho+ to 7Hc+.
This is in agreement with the observation of significant
amounts of 1Hc+ and 2Hc+ in spectroelectrochemical in-
vestigations.[17] As for 2Ho, the first oxidation process of
7Ho is assigned to the dithienylcyclopentene unit on the
basis of comparison with 1Ho. The second oxidation process
at 1.68 V is assigned tentatively to oxidation of the methoxy-
phenyl unit.
For 7Fo at low scan rates (ca. 0.1 Vsꢀ1), on the return
scan after the second oxidation process (1.54 V), a new elec-
trochemically reversible (Ep,aꢀEp,c =60 mV) reduction is ob-
served at 0.31 V, and a very minor subsequent oxidation at
about 0.80 V, coincident with 7Fc (Figures 4 and 5). At
higher scan rates (ca. 10 Vsꢀ1) formation of the closed form
7Fc becomes the dominant electrochemical process follow-
ing the second oxidation step of 7Fo.
The redox and photochemical processes for 7F are sum-
marised in Scheme 4. It is clear from the reversibility at
higher scan rates that the first oxidation process (7Fo!
7Fo+) is reversible. Based on the potential of the process
(compared with 2Fo), it is assigned to one-electron oxida-
tion of the methoxyphenyl group of 7Fo (7Fo-methoxy-
phenyl+, Scheme 4). The second irreversible oxidation pro-
In contrast to the hexahydro compounds 1H, 2H and 7H,
which undergo electrochemical ring closure, compounds 1F
and 2F exhibit very different electrochemical properties.[17]
Specifically, 1F undergoes electrochemical ring opening
(1Fc!1Fo), while 2F exhibits electrochemical ring closure
6436
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6430 – 6441

Oxidative Electrochemical Switching in Dithienylcyclopentenes
FULL PAPER
cess at 1.54 V cannot be assigned to a second oxidation of
the phenylmethoxy group and is more likely to be an oxida-
tion process localised on the phenyl-substituted thienyl ring
(7Fo+-methoxyphenyl+, Scheme 4). The oxidation of the
thienyl ring is then followed by formation of the closed
form, which, due to its lower redox potential (0.8 V), under-
goes intramolecular electron transfer from the methoxy-
phenyl group to form 7Fc2+. At higher scan rates (>5 Vsꢀ1)
7Fc2+ is reduced to 7Fc (via 7Fc+), but a competitive pro-
cess (i.e., formation of 7Fx) occurs at slower scan rates
(<5 Vsꢀ1).
At cathodic potentials 7Fo undergoes two-electron fully
irreversible reduction at ꢀ1.73V, again at a potential inter-
mediate beteween those of 1Fo and 2Fo. This suggests that
the reduction process involves the entire ring system. For
7Fc a single electrochemically reversible redox process is
observed at ꢀ0.80 V, which is very similar to that observed
for 1Fo but with improved chemical stability. As for the
hexahydrocylcopentene-based compound the improved sta-
bility of this process is attributed to the electron-donating
effect of the methoxyphenyl group, which stabilises the
mono- and dications (7Fc+ and 7Fc2+). The stabilisation is
less than that observed for 2Fc, but this is expected on the
basis that 2Fc has twice the electron donor stabilisation
compared with 7Fc.
For both 7Ho and 7Fo, the first oxidation processes can
be assigned to one-electron oxidation of the dithienylethene
group and methoxyphenyl group, respectively. The differ-
ence in the location of the first oxidation process is in agree-
Figure 5. Scan-rate dependence of the oxidation processes of 7Fo
(0.1 Vsꢀ1 to 10 Vsꢀ1) in 0.1m TBAP/CH3CN. a) Spectra offset for clarity,
I=0 is indicated for each trace by an asterisk. b) Overlay of 0.1, 05 and
5.0 Vsꢀ1 traces.
Scheme 4. Redox and photochemical processes observed for 7F.
Chem. Eur. J. 2005, 11, 6430 – 6441
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6437

B. L. Feringa et al.
ment with the assignments made for 2Ho and 2Fo.[17] In
contrast to 2Fo, in which ring closure occurred after the first
oxidation process, for 7Fo, the first oxidation (at 1.34 V)
was found to be reversible and localised on the methoxy-
phenyl moiety. The second oxidation process (1.54 V) then
takes place at the dithienylethene moiety and is responsible
for ring-closure. It is clear that for 7Fo the second oxidation
process is localised at the dithienylethene core; hence, in
agreement with 7Ho, one-electron oxidation of the dithienyl-
ethene appears to be sufficient to allow ring-closure to take
place. This has implications for 2Fo, in that it supports the
mechanism whereby double intramolecular electron transfer
to the dithienylethene core occurs, together with stabilisa-
tion of the dication (2Fc2+) by the electron-donating prop-
erties of the two methoxyphenyl groups.
effect of perturbing the electron density on the thienyl rings,
and hence affects the stabilisation of the cationic species,
and of introducing separate redox processes, which can
bypass the oxidative isomerisation processes of the thienyl
core. The introduction of asymmetry though non-redox-
active peripheral groups (i.e., the chloro group in 13H/F) re-
sults in only minor perturbation of the redox properties of
each thienyl unit. In terms of electrochemical switching the
direction of ring switching is very dependent on the substitu-
ents chosen. The first oxidation process of 13Ho is assigned
to the phenyl-substituted thienyl ring, but ring closure is not
observed (even if it occurs at all) due to the effect of the
chloro group, which renders the dication of the closed form
(13Hc2+) unstable.
Electrochemical properties of 13H/13F: For 13H and 13F
(R1 =Cl, R2 =Ph) the first oxidation processes are expected
to be located on the phenyl-substituted thienyl ring on the
basis of comparison of 1H/F and 8H/F (Table 2). For 13Ho
two oxidation processes are observed at 1.26 and 1.57 V.
The first oxidation process (13Ho!13Ho+) is close to that
of 1Ho, and the second process (13Ho+!13Ho2+) is con-
siderably more anodic (by 200 mV) than that of 8Ho, that
is, significant communication exists between the two thienyl
groups. This implies that the second oxidation step is made
more difficult by oxidation of the first thienyl ring. No ring
closure is observed even after the second oxidation step of
13Ho (i.e., above 1.6 V). For 13Hc, two oxidation processes
occur, the second of which is irreversible, at potentials inter-
mediate between those of the parent compounds.
For 13Fo two oxidation processes are observed at poten-
tials intermediate between those of the parent compounds
1Fo and 8Fo. The first process is assigned to oxidation of
the phenyl-substituted thienyl group, whilst the second pro-
cess is assigned to the chloro-substituted thienyl group. The
intermediacy of the potential of the oxidation processes of
13Fo between those of 1Fo and 8Fo indicates that some in-
teraction between the thienyl rings occurs. For 13Fc, two ox-
idation processes are observed (both fully irreversible), and
as for 13Fo the potentials of the first and second processes
lie between those of the parent compounds 1Fc and 8Fc.
For 13Fc, two reduction processes are observed at cathodic
potentials at ꢀ1.04 and ꢀ1.45 V. The observation of two
processes is unexpected considering that for 1Fc and 8Fc a
single reduction process at about ꢀ1.1 V is observed. This
suggests that for the reduced species significant interaction
between the thienyl groups is present.
Conclusion
Overall, the results obtained demonstrate that the nature of
the redox properties of the dithienylethenes can be tuned
towards electrochemical ring-closing or ring-opening by sta-
bilisation and destabilisation of the mono- and dication of
the closed state, respectively. Alternatively, redox-active
groups may be employed to change the direction of electro-
chemical switching. Nevertheless, to design molecules which
exhibit the desired direction of switching it is important to
recognise the mechanism by which these processes occur.
Although ring closure occurs immediately after oxidation of
the open form, the ring-closed product must subsequently
be reduced. The formation of species such as 1Hx and 5Hx
from 1Hc2+ and 5Hc2+, respectively, by a thermally activat-
ed rearrangement must be inhibited either by environmental
control (e.g., solvent and temperature) or by the introduc-
tion of electron-donating groups. The latter approach must
be taken with caution given that electron-donating groups
frequently exhibit redox chemistry themselves (e.g., in 2F).
The driving force for ring opening and closing appears to
lie in the ability of the bridging cyclopentene moiety to
allow for stabilisation of the various cationic species. The
observation of a reversible methoxyphenyl oxidation process
for 7Fo and the characterisation of the first oxidation pro-
cess of 7Ho as being one-electron, taken together with the
observation of the monocation in significant quantities
during spectroelectrochemistry,[17] provide compelling evi-
dence that ring-closure occurs through the monocation (i.e.,
7Ho+, in which the positive charge is localised on the thien-
yl ring system). The driving force for ring-closure is stabili-
sation of the monocation though (partial) delocalisation of
the charge on the second ring. Where the communication
between the rings is poor, as is the case in the hexafluoro
compounds, the stabilisation achieved does not compensate
for the loss of ring stabilisation (aromaticity), and hence
ring opening of the monocation/dication of the closed form
is favoured.
Overall for 7H/F and 13H/F it is clear that in both open
and closed states the thienyl moieties are not independent
of each other and that some interaction (either inductive or
by delocalisation) is observed. The effect of asymmetry
itself is very much dependent on the substituents employed,
and although the redox properties of 7H and 7F are broadly
intermediate between those of the parent compounds 1H/
2H and 1F/2F, the introduction of peripheral redox-active
groups (e.g., the methoxyphenyl group) has the additional
A key feature of both the dithienylhexafluoro- and dithie-
nylhexahydrocyclopentenes is their propensity to undergo
ring closing and opening photochemically and, more recent-
6438
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6430 – 6441

Oxidative Electrochemical Switching in Dithienylcyclopentenes
FULL PAPER
1
86%). H NMR (300 MHz, CDCl3): d=1.25–1.28 (m, 9H), 7.34–7.47 ppm
(m, 4H); 13C NMR (75.4 MHz, CDCl3): d=30.9 (q), 46.1 (s), 123.4 (s),
131.6 (d), 131.8 (s), 138.9 ppm (d).
ly, the possibility of using chemical and electrochemical oxi-
dation to achieve similar processes has been highlighted.[16]
However, the mechanism and factors which influence the
oxidative ring-opening/closing of the switches were unclear
and, in particular, an understanding of how substituents
drive the direction of switching was lacking, both in terms of
the effect of variation in the cyclopentene bridging unit and
in terms of the substituent at the C5 position of the thienyl
rings. In addition, the mechanism by which redox-active
groups attached to the thienyl rings influence the ring-
switching process directly had not been investigated. In our
previous[17] and present studies these issues were addressed
systematically by variation of the bridging cyclopentene
group, by variation of substituents at C5 of the thienyl rings
and by the introduction of asymmetry in the substitution
pattern (e.g., 7H). The data set now available provides a
considerable basis for the further development of this dis-
tinct class of photo/electrochromic compounds in functional
devices and systems.
1,2-Bis[5’-(4’’-tert-butylsulfanylphenyl)-2’-methylthien-3’-yl]cyclopentene
(C): nBuLi (6.0 mL, 1.6m in hexane, 7.0 mmol) was added to a solution
of 8H (1.00 g, 3.05 mmol) in THF (40 mL) under an inert atmosphere.
After 1 h, B(OBu)3 (2.50 mL, 6.5 mmol) was added to produce bis(boron-
ic ester) intermediate A. A separate flask was charged with 4-bromo-tert-
butylsulfanylbenzene (1.00 mg, 6.2 mmol), [Pd(PPh3)4] (400 mg,
0.083mmol), THF (25 mL), aqueous Na 2CO3 (4 mL, 2m) and ethylene
glycol (5 drops). The mixture was heated to 808C and the preformed bor-
onic ester was added slowly by syringe. The reaction mixture was heated
at reflux for 3h, after which it was diluted with diethyl ether (50 mL)
and washed with brine (50 mL). The brine solution was washed with ad-
ditional diethyl ether (50 mL), and the combined organic phases were
dried over Na2SO4 and concentrated. Chromatography over silica
(hexane/diethyl ether 100/1) produced C (800 mg, 1.35 mmol, 45%, 95%
pure). 1H NMR (300 MHz, CDCl3): d=1.26–1.29 (m, 18H), 2.00 (s, 6H),
2.06 (m, 2H), 2.82 (t, J=7.8 Hz, 4H), 7.03(s, 2H), 7.24–7.49 ppm (m,
8H).
1,2-Bis[5’-(4’’acetylthio)-2’-methylthien-3’-yl]cyclopentene (4H): BBr3
(0.08 mL, 1.6 mmol) was added to a solution of C (500 mg, 0.80 mmol)
and AcCl (0.40 mL) in CH2Cl2 (10 mL) under an inert atmosphere.[20]
The reaction mixture was stirred overnight and then diluted with diethyl
ether (10 mL) and poured over ice (5 g). The phases were separated, the
aqueous layer was extracted with additional diethyl ether (20 mL) and
the combined organic phases were dried with Na2SO4 and concentrated.
Chromatography over silica (pentane/diethyl ether 500/1) produced 4H
(130 mg, 0.23 mmol, 29%). M.p. 61–648C; 1H NMR (300 MHz, CDCl3):
d=1.99 (s, 6H), 2.08 (m, 2H), 2.41 (s, 6H), 2.83(t, J=7.8 Hz, 4H), 7.06
(s, 2H), 7.35 (d, J=8.1 Hz, 4H), 7.51 ppm (d, J=8.1 Hz, 4H); 13C NMR
(75.4 MHz, CDCl3): d=14.4 (q), 23.0 (t), 30.2 (q), 38.4 (t), 124.8 (d),
125.8 (d), 126.0 (s), 134.7 (s), 134.8 (d), 135.5 (s), 135.6 (s), 136.8 (s),
138.6 (s), 194.2 ppm (s); MS (EI): 560 [M+]; HRMS calcd for C31H28O2S4
560.097, found 560.097.
Experimental Section
For all spectroscopic measurements Uvasol-grade solvents (Merck) were
employed. All reagents employed in synthetic procedures were of re-
agent grade or better, and used as received unless stated otherwise. The
symmetric compounds 1H–3H, 6H, 8H–10H, 1F–3F and 8F were pre-
pared by previously reported procedures.[18] The intermediate structures
in the syntheses are numbered A, B, C and D, with short characterisation
for clarity. 1H NMR spectra were recorded at 200, 300, 400 or 500 MHz;
13C NMR spectra at 50.3, 75.4 or 125.7 MHz; and 19F NMR spectra at
188.2 or 470.3MHz. All spectra were recorded at ambient temperature,
with the residual proton signals of the solvent as an internal reference.
Chemical shifts are reported relative to TMS.
1,2-Bis[5’-(4’’-methylsulfanylphenyl)-2’-methylthien-3’-yl]cyclopentene
(5H): nBuLi (1.4 mL, 1.6m in hexane, 2.3mmol) was added to a solution
of 8H (250 mg, 0.76 mmol) in THF (10 mL) under an inert atmosphere.
After 1 h, B(OBu)3 (0.60 mL, 2.3mmol) was added to produce the bis-
(boronic ester) intermediate. A separate flask was charged with 2-bromo-
thioanisole (310 mg, 1.5 mmol), [Pd(PPh3)4] (96 mg, 0.083mmol), THF
(5 mL), aqueous Na2CO3 (4 mL, 2m) and ethylene glycol (5 drops). The
mixture was heated to 808C and the preformed boronic ester was added
slowly by syringe. The reaction mixture was heated at reflux for 3h and
then diluted with diethyl ether (50 mL) and washed with brine (50 mL).
The brine solution was washed with additional diethyl ether (50 mL) and
the combined organic phases were dried over Na2SO4 and concentrated.
Chromatography over silica (hexane/CH2Cl2 10/1) and stirring in hexane
(excess)/CH2Cl2 produced 5H (120 mg, 0.26 mmol, 35%). M.p. 162–
1648C; 1H NMR (300 MHz, CDCl3): d=1.97 (s, 6H), 2.06 (m, 2H), 2.47
(s, 6H) 2.82 (t, J=7.2 Hz, 4H), 6.97 (s, 2H), 7.20 (d, J=8.4 Hz, 4H),
7.38 ppm (d, J=8.4 Hz, 4H); 13C NMR (75.4 MHz, CDCl3): d=14.4 (q),
16.0 (q), 23.0 (t), 38.4 (t), 123.7 (d), 125.6 (d), 127.0 (d), 131.5 (s), 134.2
(s), 134.6 (s), 136.6 (s), 137.0 (s), 139.1 ppm (s); MS (EI): 504 [M+];
HRMS calcd. for C29H28S4: 504.108, found: 504.108.
Mass spectrometry was performed with CI, DEI or EI ionisation proce-
dures. For several of the dithienylethenes, sublimation was difficult and
only DEI (desorption electron ionisation) provided mass spectra. Mass
spectra were recorded on a Jeol JMS-600 mass spectrometer in the scan
range of m/z 50–1000 with an acquisition time between 300 and 900 ms
and a potential between 30 and 70 V. Derivatives synthesized starting
from compound 8H are light-sensitive and were therefore handled exclu-
sively in the dark by using brown glassware. Column chromatography
(Aldrich silica gel grade 9385, 230–400 mesh) was performed under
yellow light. For UV/Vis analyses, each spectrum was recorded by sum-
mation of 20 scans. UV/Vis absorption spectra (accuracy ꢁ2 nm) were
recorded on a Hewlett-Packard UV/Vis 8453spectrometer. Electrochem-
ical measurements were carried out as described previously.[17]
1,2-Bis[5’-di-n-butoxyboryl-2’-methylthien-3’-yl]cyclopentene (A): Com-
pound 8H (70 mg, 0.2 mmol) was dissolved in anhydrous THF (8 mL)
under a nitrogen atmosphere, and nBuLi (0.31 mL, 1.6m in hexane,
0.5 mmol) was added by syringe. This solution was stirred for 30 min at
room temperature, and B(nOBu)3 (0.18 mL, 0.6 mmol) was added. The
resulting solution was stirred for 1 h at room temperature and used di-
rectly in the Suzuki cross coupling-reactions without workup due to hy-
drolysis of the boronic ester A during isolation. Starting from 8F, the lith-
1-[5-(4-Methoxyphenyl)-2-methylthien-3-yl]-2-(2-methyl-5-phenylthien-3-
yl)cyclopentane (7H): tBuLi (0.40 mL, 1.5m in pentane, 0.593mmol) was
added to a solution of 13H (200 mg, 0.539 mmol) in THF (7 mL) under
an inert atmosphere. After 1 h, B(OBu)3 (0.22 mL, 0.81 mmol) was
added and the mixture was stirred for 1 h to produce the bis(boronic
ester) intermediate. A separate flask was charged with 4-bromoanisole
(0.135 mL, 1.08 mmol), [Pd(PPh3)4] (19 mg, 0.016 mmol), THF (3mL),
aqueous Na2CO3 (2m, 3mL) and ethylene glycol (3drops). The mixture
was heated to 808C and the performed boronic ester was added slowly.
The reaction mixture heated at reflux overnight, diluted with diethyl
ether (40 mL) and washed with water (40 mL). The aqueous layer was
washed with additional diethyl ether (40 mL) and the combined organic
phases were dried over Na2SO4 and concentrated. Chromatography on
ꢀ
iation was carried out in diethyl ether, due to less substitution of the C F
bonds.
1-Bromo-4-tert-butylsulfanylbenzene (B): A catalytic amount of AlCl3
was added to a solution of bromothiophenol (2.50 g, 13.3 mmol) and tert-
butyl chloride (3g, 2 equiv) in acetonitrile (25 mL). [19] The solution was
heated at reflux for 72 h, cooled to room temperature, extracted with
water, dried over Na2SO4 and concentrated. Chromatography over silica
(hexane) produced the desired compound as an oil (2.80 g, 11.5 mmol,
Chem. Eur. J. 2005, 11, 6430 – 6441
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6439

B. L. Feringa et al.
silica gel (hexane) afforded 7H as a sticky oil (150 mg, 63%). 1H NMR
(300 MHz, CDCl3): d=2.05 (s, 3H), 2.07 (s, 3H), 2.10–2.16 (m, 2H),
ture was cooled to room temperature, H2O (5 mL) was added, the organ-
ic layer separated and the water layer extracted with ethyl acetate (2”
20 mL). The combined organic layers were dried over Na2SO4, and the
solvent evaporated. The product was purified by column chromatography
(SiO2, hexane/toluene 9/1) to give 7F as a white solid (123mg, 45%).
M.p. 127–1288C; 1H NMR (400 MHz, CDCl3): d=1.94 (s, 3H), 1.97 (s,
3H), 3.84 (s, 3H), 6.91 (d, J=8.8 Hz, 2H), 7.16 (s, 1H), 7.27–7.33 (m,
2H), 7.39 (t, J=7.3Hz, 2H), 7.47 (d, J=8.8 Hz, 2H), 7.54 ppm (d, J=
7.3Hz, 2H); 13C NMR (100.6 MHz, CDCl3): d=14.6 (q), 14.7 (q), 55.5
(q), 116.4 (t), 114.5 (d), 118–119 (m), 121.3(d), 122.5 (d) 125,7 (d), 125.8
(s), 126.0 (s), 126.3 (s), 127.0 (d), 128.0 (d), 129.1 (d), 133.5 (s), 140.4 (s),
141.4 (s), 142.3(s), 159.6 ppm (s); 19F NMR (188.2 MHz, CDCl3): d=
ꢀ111.09 (t, J=5.6 Hz, 4F), ꢀ132.88 ppm (m, 2F); MS (EI): 550 [M+];
HRMS calcd for C28H20F6OS2: 550.086, found 550.085.
2.88–2.93(m, 4H), 3.85 (s, 3H), 6.93(d,
J=8.8 Hz, 2H), 6.99 (s, 1H),
7.12 (s, 1H), 7.20–7.30 (m, 1H), 7.36–7.41 (m, 2H), 7.49 (d, J=8.4 Hz,
2H), 7.57 ppm (d, J=7.3Hz, 2H); 13C NMR (75.4 MHz, CDCl3): d=11.9
(q), 12.0 (q), 20.6 (t), 36.0 (t), 52.9 (q), 111.8 (d), 120.5 (d), 121.6 (d),
122.9 (d), 124.1 (d), 124.5 (d), 125.0 (s), 126.3 (d), 132.0 (s), 132.1 (s),
132.1 (s), 132.3 (s), 134.1 (s), 134.3 (s), 137.1 (s), 156.4 ppm (s); MS (EI):
442 [M+]; HRMS calcd for C28H26OS2: 442.143, found: 442.141.
1,2-Bis[5’-(thien-2-yl)-2’-methylthien-3’-yl]cyclopentene (11H): 2-Bromo-
thiophene (0.04 mL, 0.4 mmol) was dissolved in THF (5 mL), and after
addition of [Pd(PPh3)4] (15 mg, 0.012 mmol), the solution was stirred for
15 min at room temperature Aqueous Na2CO3 (1 mL, 2m) and 6 drops of
ethylene glycol were added, and the resulting biphasic system was heated
in an oil bath at reflux. The solution of A was added dropwise by syringe,
after which the reaction mixture was heated at reflux for 2 h, and then al-
lowed to cool to room temperature Diethyl ether (50 mL) and H2O
(50 mL) were added, and the organic layer was collected and dried
(Na2SO4). After evaporation of the solvent the product was purified by
column chromatography (SiO2, hexane) to gave a purple solid (32 mg,
38%). M.p. 1478C (decomp); 1H NMR (300 MHz, CDCl3): d=1.94 (s,
6H), 2.04 (m, 2H), 2.79 (t, J=12.5 Hz, 4H), 6.87 (s, 2H), 6.95 (t, J=
6.5 Hz, J=3.3 Hz, 2H), 7.03 (d, J=4.5 Hz, 2H), 7.13ppm (d, J=8.5 Hz,
2H); 13C NMR (75.4 MHz, CDCl3): d=14.3(q), 22.9 (t), 38.5 (t), 122.9
(d), 123.7 (d), 124.5 (d), 127.6 (d), 133.0 (s), 134.0 (s), 134.5 (s), 136.3 (s),
137.7 ppm (s); MS (EI): 424 [M+]; HRMS calcd for C23H20S4: 424.045,
found: 424.042.
1,2-Bis(5’-formyl-2’-methylthien-3’-yl)hexafluorocyclopentene
Under the same conditions as described for 11H, nBuLi (1.6m in hexane,
0.13mL, 1.8 mmol) was added to stirred solution of 8F (30 mg,
(9F):
a
0.06 mmol) in anhydrous diethyl ether (5 mL) under nitrogen at room
temperature over 30 min, after which the mixture was quenched with an-
hydrous dimethylformamide (0.05 mL, 0.6 mmol). Trituration from
hexane/CH2Cl2 afforded the compound as a brown-orange solid (20 mg,
1
66%). M.p. 1828C; H NMR (200 MHz, CDCl3): d=2.02 (s, 6H), 7.73(s,
2H), 9.85 ppm (s, 2H); 13C NMR (50.3MHz, CDCl 3): d15.2 (q), 110.1 (t),
115.7 (t), 125.8 (d), 135.3 (s), 136.5 (t) 142.2 (s), 151.3 (s), 181.5 ppm (d);
19F NMR (188.2 MHz, CDCl3): d=ꢀ111.97 (t, J=4.9 Hz, 4F),
ꢀ133.55 ppm (quintet, J=4.9 Hz, 2F); MS (EI): 424 [M+]; HRMS calcd
for C17H10F6O2S2 424.003, found 424.004.
1-(5-Chloro-2-methylthien-3-yl)-2-(2-methyl-5-phenylthien-3-yl)hexa-
fluorocyclopentene (13F): Compound 8F (2.19 g, 5 mmol) was dissolved
in anhydrous diethyl ether (50 mL) under nitrogen and nBuLi (3.2 mL,
1.6m in hexane, 5 mmol) was added slowly by syringe. This solution was
stirred at room temperature for 1 h and B(nOBu)3 (1.63mL, 6 mmol)
was added in one portion to yield A. After the mixture had been stirred
for 1 h at room temperature, THF (50 mL), aqueous Na2CO3 (10 mL,
2m), iodobenzene (0.78 mL, 7 mmol) and [Pd(PPh3)4] (58 mg, 0.05 mmol)
were added and the mixture was heated at reflux for 16 h. The reaction
mixture was cooled to room temperature, H2O (20 mL) added, the organ-
ic layer separated and the aqueous layer extracted with ethyl acetate (2”
50 mL). The organic layers were combined and dried over Na2SO4,
evaporated and the crude product purified by column chromatography
(SiO2, hexane) to give a yellowish oil which solidified on standing (1.32 g,
71%). M.p. 80–818C; 1H NMR (400 MHz, CDCl3): d=1.86 (s, 3H), 1.98
(s, 3H), 6.94 (s, 1H), 7.26 (s, 1H), 7.31 (t, J=7.3Hz, 1H), 7.39 (t, J=
7.3Hz, 2H), 7.55 ppm (d, J=7.3Hz, 2H); 13C NMR (100.6 MHz,
CDCl3): d=14.5 (q), 14.7 (q), 111.1 (t), 113.8 (t), 116.1 (t), 122.3 (d),
124.5 (s), 125.7 (d), 125.8 (d), 127.9 (s), 128.1 (d), 129.2 (d), 133.3 (s),
140.6 (s), 141.4 (s), 142.6 ppm (s); 19F NMR (188.2 MHz, CDCl3): d=
ꢀ111.4 (t, J=5.1, 4F), ꢀ133.0 ppm (m, 2F); MS (EI): 478 [M+]; HRMS
calcd for C21H13F6S2Cl: 478.005, found: 478.001.
1,2-Bis[5’-ethoxycarbonyl-2’-methylthien-3’-yl]cyclopentene (12H): Com-
pound D (75 mg) was dissolved in EtOH (50 mL) and a catalytic amount
of 30% aqueous HCl was added. After the mixture was stirred overnight,
the solvent was removed to yield 12H as a light brown solid in quantita-
tive yield. M.p. 121–1228C; 1H NMR (300 MHz, CDCl3): d=1.32 (t, J=
6.9 Hz, 6H), 1.87 (s, 6H), 2.00–2.07 (m, 2H), 2.76 (t, J=7.5 Hz, 4H), 4.28
(q, J=6.9 Hz, 4H), 7.49 ppm (s, 2H); 13C NMR (75.4 MHz, CDCl3): d=
14.3 (q), 14.8 (q), 22.8 (t), 38.6 (t), 61.0 (t), 129.6 (s), 134.2 (s), 134.7 (d),
136.5 (s), 142.5 (s), 162.1 ppm (s); MS (EI): 404 [M+]; HRMS calcd for
C21H24O4S2: 404.112, found: 404.112.
1-(5-Chloro-2-methylthien-3-yl)-2-(2-methyl-5-phenylthien-3-yl)cyclopen-
tene (13H): nBuLi (4.70 mL, 1.6m in hexane, 7.51 mmol) was added to a
solution of 8H (2.25 g, 6.83mmol) in THF (100 mL) under an inert at-
mosphere. After 1 h, B(OBu)3 (2.77 mL, 10.3mmol) was added and the
mixture was stirred for a further hour to produce bis(boronic ester) inter-
mediate A. A separate flask was charged with bromobenzene (2.86 mL,
27.3mmol), [Pd(PPh 3)4] (0.237 g, 0.21 mmol), THF (23 mL), aqueous
Na2CO3 (2m, 18 mL) and ethylene glycol (20 drops). The mixture was
heated to 808C and the preformed boronic ester was added slowly. The
reaction mixture was heated at reflux overnight, cooled to room tempera-
ture, diluted with diethyl ether (200 mL) and washed with water
(200 mL). The aqueous layer was extracted with additional diethyl ether
(200 mL) and the combined organic phases were dried over Na2SO4 and
concentrated. Subsequent chromatography on silica gel (hexane) afford-
ed 13H as a sticky oil (1.95 g, 77%). 1H NMR (300 MHz, CDCl3): d=
1.94 (s, 3H), 2.05 (s, 3H), 2.08–2.16 (m, 2H), 2.78–2.89 (m, 4H), 6.68 (s,
1H), 7.05 (s, 1H), 7.30 (t, J=7.0 Hz, 1H), 7.37–7.42 (m, 2H), 7.55 ppm
(d, J=7.3Hz, 2H); 13C NMR (75.4 MHz, CDCl3): d=14.1 (q), 14.3(q),
22.9 (t), 38.4 (t), 38.5 (t), 123.8 (d), 125.0 (s), 125.3 (d), 126.8 (d), 127.0
(d), 128.8 (d), 133.2 (s), 133.7 (s), 134.4 (s), 135.1 (s), 135.3 (s), 136.3 (s),
139.8 ppm (s); MS (EI): 370 [M+]; HRMS calcd for C21H19S2Cl: 370.062,
found: 370.063.
Acknowledgements
The authors thank Dr. Johan Hjelm (Dublin City University, Ireland) for
his invaluable advice and comments during the preparation of this manu-
script. The authors acknowledge the Materials Science Centre (MSC+)
(J.J.D.J., M.W.), Ryukoku University (K.U.), FOM (T.K.) and the Dutch
Economy, Ecology, Technology (EET) program (W.R.B.) for financial
support. Zeon Corporation is acknowledged for their generous gift of oc-
tafluorocyclopentene. Finally, we thank Dr. L. P. Candeias (Delft Techni-
cal University) for useful discussions.
1-[5-(4-Methoxyphenyl)-2-methylthien-3-yl]-2-(2-methyl-5-phenylthien-3-
yl)hexafluorocyclopentene (7F): Compound 13F (0.24 g, 0.5 mmol) was
dissolved in anhydrous diethyl ether (5 mL) under nitrogen and nBuLi
(0.32 mL, 1.6m in hexane, 0.5 mmol) was added slowly by syringe. This
solution was stirred at room temperature for 1 h and B(nOBu)3 (0.16 mL,
0.6 mmol) was added in one portion. After the mixture had been stirred
for 1 h at room temperature, THF (5 mL), aqueous Na2CO3 (1 mL, 2m),
4-bromoanisole (0.09 mL, 7 mmol) and [Pd(PPh3)4] (6 mg, 0.005 mmol)
were added, and the mixture heated at reflux for 16 h. The reaction mix-
[1] a) H. Durr, H. Bouas-Laurent, Photochromism: Molecules and Sys-
tems, Elsevier, Amsterdam, 1990; b) M. Blank, L. M. Soo, N. H.
Wasserman, B. F. Erlanger, Science 1981, 214, 70–72; c) S. Shinkai,
Pure Appl. Chem. 1987, 59, 425–430; d) M. Irie, Adv. Polym. Sci.
6440
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6430 – 6441

Oxidative Electrochemical Switching in Dithienylcyclopentenes
FULL PAPER
1990, 94, 27–67; e) M. Irie, Applied Photochromic Polymer Sys-
tems, (Ed.: C. B. McArdle), Blackie, Glasgow, 1991, 174–206; f) T.
Ikeda, O. Tsutsumi, Science 1995, 268, 1873–1875; g) Y. Yu, M.
Nakano, T. Ikeda, Nature 2003, 425, 145–145; h) G. H. Brown,
Photochromism, Wiley-Interscience, New York, 1971.
[2] V. Balzani, F. Scandola, Supramolecular Photochemistry, Horwood,
Chichester, 1991.
[3] S. Welter, K. Brunner, J. W. Hofstraat, L. De Cola, Nature 2003, 421,
54–57.
[4] G. Berkovic, V. Krongauz, V. Weiss, Chem. Rev. 2000, 100, 1741–
1754.
[5] a) O. A. Fedorova, Y. V. Fedorov, E. N. Andryukhina, S. P. Gromov,
M.V. Alfimov, R. Lapouyade, Org. Lett. 2003, 5, 4533–4535; b) B.
Qin, R. Yao, H. Tian, Org. Biomol. Chem. 2003, 1, 2187–2191;
c) S. H. Kawai, Tetrahedron Lett. 1998, 39, 4445–4448; d) M. Take-
shita, M. Irie, J. Org. Chem. 1998, 63, 6643–6649; e) M. Irie, O.
Miyatake, K. Uchida, T. Eriguchi, J. Am. Chem. Soc. 1994, 116,
9894; f) H. Tian, B. Qin, R. X. Yao, X. L. Zhao, S. J. Yang, Adv.
Mater. 2003, 15, 2104–2107.
[6] a) S. H. Kawai, S. L. Gilat, R. Ponsinet, J.-M. Lehn, Chem. Eur. J.
1995, 1, 275–284; b) S. H. Kawai, S. L. Gilat, R. Ponsinet, J.-M.
Lehn, Chem. Eur. J. 1995, 1, 285–293; c) V. I. Minkin, Chem. Rev.
2003, 103, 2751–2776; d) J. Bonvoisin, J.-P. Launay, M. van der Au-
weraer, F. C. de Schryver, J. Phys. Chem. 1994, 98, 5052.
[7] Y. Yokoyama, Chem. Rev. 2000, 100, 1717–1740.
[8] J. J. D. de Jong, L. N. Lucas, R. M. Kellogg, J. H. van Esch, B. L. Fer-
inga, Science 2004, 304, 278–281.
[9] a) Special issue: Photochroism: Memories and Switches: Chem. Rev.
2000, 100, 1683–1890; b) Molecular Switches (Ed.: B. L. Feringa),
Wiley-VCH, Weinheim, 2001; c) H. Tian, S. Yang, J. Chem. Soc.
Rev. 2004, 33, 85–97.
Hartschuh, H. Port, J. Am. Chem. Soc. 2000, 122, 3037–3046; c) c.-
W. Chiu, T. J. Chow, C.-H. Cheun, H.-M. Lin, T. J. Chow, Chem.
Mater. 2003, 15, 4527–4532.
[16] a) A. Peters, N. R. Branda, J. Am. Chem. Soc. 2003, 125, 12, 3404–
3405; b) A. Peters, N. R. Branda, Chem. Commun. 2003, 4, 954–
955; c) B. Gorodetsky, H. D. Samachetty, R. L. Donkers, M. S.
Workentin, N. R. Branda, Angew. Chem. 2004, 116, 2872–2875;
Angew. Chem. Int. Ed. 2004, 43, 2812–2815; d) X.-H Zhou, F.-S.
Peng Yuan, S.-Z. Pu, F.-Q. Zhao, C.-H. Tung, Chem. Lett. 2004, 33,
1006–1007.
[17] For Part 1, see:W. R. Browne, J. J. D. de Jong, T. Kudernac, M.
Walko, L. N. Lucas, K. Uchida, J. H. van Esch, B. L. Feringa, Chem.
Eur. J. 2005, 11, in press; DOI: 10.1002/chem.200500162.
[18] a) J. J. D. de Jong, L. N. Lucas, R. M. Kellogg, B. L. Feringa, J. H.
van Esch, Eur. J. Org. Chem. 2003, 1887–1893; b) L. N. Lucas,
J. J. D. de Jong, R. M. Kellogg, J. H. van Esch, B. L. Feringa, Eur. J.
Org. Chem. 2003, 155–166.
[19] R. Adams, A. Ferretti, J. Am. Chem. Soc. 1959, 81, 4927–4931.
[20] N. Stuhr-Hansen, J. B. Christensen, N. Harrit, T. Bjornholm, J. Org.
Chem. 2003, 68, 1275–1282.
[21] N. Miyaura, A. Suzuki, Chem. Rev. 1995, 95, 2457–2483.
[22] M. Irie, K. Sakemura, M. Okinaka, K. Uchida, J. Org. Chem. 1995,
60, 8305–8309.
[23] In the present report, reversibility, quasireversibility and irreversibil-
ity refer to the chemical stability of the oxidised or reduced species.
In all cases electrochemical reversibility (Ep,aꢀEp,c <80 mV) is ob-
served between 0.01 and 2.0 Vsꢀ1, where the chemical stability of
the oxidised/reduced species is sufficient to observe the return wave.
Where the Ip,a peak is not significantly different to the Ip,c peak, the
term reversible is applied; similarly, quasireversibility refers to situa-
tions where chemical reversibility is dependent on scan rate (i.e., ob-
served only at higher scan rates).
[24] C. Hansch, A. Leo, R. W. Taft, Chem. Rev. 1991, 91, 165–195.
[25] Owing to the limited data set available and the involvement of
redox-active substituents, a similar correlation for the hexfluorocy-
clopentene-based switches was not attempted.
[10] A. Doron, E. Katz, G. Tao, I. Willner, Langmuir 1997, 13, 1783–
1790.
[11] D. Duliæ, S. J. van der Molen, T. Kudernac, H. T. Jonkman, J. J. D.
de Jong, T. N. Bowden, J. van Esch, B. L. Feringa, B. J. van Wees,
Phys. Rev. Lett. 2003, 91, Art. No. 207402.
[12] a) O. A. Fedorova, Y. V. Fedorov, E. N. Andryukhina, S. P. Gromov,
M.V. Alfimov, R. Lapouyade, Org. Lett. 2003, 5, 4533–4535; b) B.
Qin, R. Yao, H. Tian, Org. Biomol. Chem. 2003, 1, 2187–2191;
c) S. H. Kawai, Tetrahedron Lett. 1998, 39, 4445–4448; d) M. Take-
shita, M. Irie, J. Org. Chem. 1998, 63, 6643–6649; e) H. Tian, B.
Qin, R. X. Yao, X. L. Zhao, S. J. Yang, Adv. Mater. 2003, 15, 2104–
2107.
[13] a) N. P. M. Huck, W. F. Jager, B. de Lange, B. L. Feringa, Science
1996, 273, 1686–1688; b) C. Denekamp, B. L. Feringa, Adv. Mater.
1998, 10, 1080–1082; c) see reference [6] for further examples.
[14] a) R. H. Mitchell, Z. Brkic, V. A. Sauro, D. J. Berg, J. Am. Chem.
Soc. 2003, 125, 7581–7585; b) S. Fraysse, C. Coudret, J.-P. Launey,
Eur. J. Inorg. Chem. 2000, 1581–1590.
[26] a) M. J. Cook, A. P. Lewis, G. S. G. McAuliffe, V. Skarda, A. J.
Thomson, J. L. Glasper, D. J. Robbins, J. Chem. Soc. Perkin Trans. 2
1984, 1293–1303; b) M. J. Cook, A. P. Lewis, G. S. G. McAuliffe, V.
Skarda, A. J. Thomson, J. L. Glasper, D. J. Robbins, J. Chem. Soc.
Perkin Trans. 2 1984, 1303–1309; c) G. Roelfes, V. Vrajmasu, K.
Chen, R. Y. N. Ho, J.-U. Rohde, C. Zondervan, R. M. La Crois, E. P.
Schudde, M. Lutz, A. L. Spek, R. Hage, B. L. Feringa, E. Munck, L.
Que, Jr., Inorg. Chem. 2003, 42, 2639–2653.
[27] S. Goswami, R. Mukherjee, A. Chakravorty, Inorg. Chem. 1983, 22,
2825–2832.
[28] V. Guttmann, G. Resch, W. Linert, Coord. Chem. Rev. 1982, 43,
133–164.
[15] a) R. T. F. Jukes, V. Adamo, F. Hartl, P. Belser, L. de Cola, Inorg.
Chem. 2004, 43, 2779–2792; b) J. M. Endtner, F. Effenberger, A.
Received: February 15, 2005
Published online: August 11, 2005
Chem. Eur. J. 2005, 11, 6430 – 6441
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6441