Synthesis and characterisation of macrobicyclic tetrathiafulvalene-bridged cage molecules

Article abstract of DOI:10.1002/(SICI)1099-0690(199809)1998:9<1743::AID-EJOC1743>3.0.CO;2-J

A series of new macrobicyclic tetrathiafulvalenophanes of type 1 and 2 with three tetrathiafulvalene bridges has been prepared under high-dilution conditions using a stepwise selective protection-deprotection of tetrathiafulvalenethiolates. All the macrobicyclic tetrathiafulvalenes, along with the intermediate compounds 5 and 6 and the unexpected tetrathiafulvalene pentamers 17, were studied by cyclic voltammetry. An electrochemical investigation using the Bard-Anson equation and thin-layer cyclic voltammetry has been carried out, allowing an estimate of the number of electrons involved in each redox process of these multi-redox compounds. The X-ray crystal structure showing the unusual crystal packing of 2a is also presented.

Full text of DOI:10.1002/(SICI)1099-0690(199809)1998:9<1743::AID-EJOC1743>3.0.CO;2-J

FULL PAPER
Synthesis and Characterisation of Macrobicyclic Tetrathiafulvalene-Bridged
Cage Molecules
Philippe Blancharda,b, Niels Svenstrupa, Joe¨lle Rault-Berthelotb, Ame´de´e Riouc, and Jan Becher*a
Department of Chemistry, Odense Universitya,
Campusvej 55, DK-5230 Odense M, Denmark
Fax: (internat.) ϩ 45-66/158780
E-mail: JBE@chem.ou.dk
IMMO, UMR CNRS 6501, Faculte´ des Sciences of Angersb,
2 Bd Lavoisier, F-49045 Angers, France
Laboratoire d’Electrochimie Organique, UMR CNRS 6510, Universite´ Rennes Ic,
Campus de Beaulieu, Av. Ge´ne´ral Leclerc, F-35042 Rennes, France
Received April 17, 1998
Keywords: Cyclophanes / Electrochemistry / Tetrathiafulvalenes / Macrocycles / Crystal structure
A series of new macrobicyclic tetrathiafulvalenophanes of
type 1 and 2 with three tetrathiafulvalene bridges has been
by cyclic voltammetry. An electrochemical investigation
using the Bard-Anson equation and thin-layer cyclic
prepared under high-dilution conditions using a stepwise voltammetry has been carried out, allowing an estimate of
selective protection-deprotection of tetrathiafulvalen-
ethiolates. All the macrobicyclic tetrathiafulvalenes, along
the number of electrons involved in each redox process of
these multi-redox compounds. The X-ray crystal structure
showing the unusual crystal packing of 2a is also presented.
with the intermediate compounds
5 and 6 and the
unexpected tetrathiafulvalene pentamers 17, were studied
Introduction
interesting to study such molecular objects, and in addition
to their potential use in conducting organic materials or in
host-guest chemistry, these molecules can be seen as models
for the study of electronic interactions between several re-
dox-active centers.
The molecular design of three-dimensionally bridged
macropolycyclic compounds is currently a challenge in syn-
thetic chemistry[1] and incorporation of redox-active groups
into such molecules is of interest for the preparation of
macropolycyclic receptor molecules and “smart” catalytic
systems.[2] With the proper design, systems of this type can
in principle be designed to signal electrochemically the
selective binding of any charged or neutral guest. Tetrathia-
fulvalene (TTF) is a well-known redox-active compound
whose electrochemical properties have been extensively and
principally used for the generation of conducting or even
superconducting organic materials.[3] However, since the
first tetrathiafulvalenocyclophane prepared by Staab et al.
in 1980,[4] the focus of attention has gradually shifted
towards the development of more complex tetrathiafulva-
lene systems. For example, various tetrathiafulvalene-like
cyclophanes[5][6] such as double-bridged or even quadruple-
bridged tetrathiafulvalenophanes (“belts”)[7] with two tetra-
thiafulvalene units have been described recently. The recent
combination of tetrathiafulvalene chemistry and supramo-
Most of the tetrathiafulvalenes incorporated into macro-
cyclic compounds mentioned above have been prepared by
a coupling reaction between two 1,3-dithiole moieties, gen-
erating the central fulvalene double bond of the tetrathia-
fulvalene in the final synthetic step, but a number of com-
plications arise from the use of this strategy, most notably
the lack of control over the macrocyclisation, the harsh re-
agents employed in the coupling reaction and the long syn-
thetic routes involved in making analogous members of a
series.[3a] Only few compounds of this type have previously
been prepared using preformed mono- or bifunctionalised
tetrathiafulvalenes. Thus, although a lot of complex mol-
ecules containing tetrathiafulvalene building blocks have
been successfully prepared, further improvements in this re-
search area requires new synthetic tools.
We have recently developed a strategy for the incorpor-
lecular chemistry has led to the construction of new elabo- ation of preformed tetrathiafulvalenes into macrocyclic
rated systems[8] such as crown ethers,[9] aza-[5c][9d] and thia- compounds.[15] This methodology is based on the facile pro-
crowns,[10] a copper(I) [2]-catenate incorporating one tetra- tection-deprotection of tetrathiafulvalenethiolates and their
thiafulvalene unit,[11] a tetrathiafulvalene cryptand,[12]
a
subsequent in situ alkylation. We have earlier successfully
tetrathiafulvalene rotaxane,[13] and cyclodextrin-TTF as- applied this strategy for the synthesis of larger assemblies
semblies.[14] But until now, the preparation of highly sym- such as macrocyclic tetrathiafulvalenes,[15b][16] oligomeric
metrical macrobicyclic molecules containing three or more tetrathiafulvalenes,[17] tetrathiafulvalene-based catenanes
tetrathiafulvalene units has not been reported. It would be and rotaxanes,[18] and stable macrocyclic donor-acceptor
Eur. J. Org. Chem. 1998, 1743Ϫ1757
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1434Ϫ193X/98/1743Ϫ1757 $ 17.50ϩ.50/0
1743
EurJOC 172/98

P. Blanchard, N. Svenstrup, J. Rault-Berthelot, A. Riou, J. Becher
FULL PAPER
systems.[19] As an extension of this strategy, we have pre- needed to prevent the formation of linear oligomeric by-
viously reported our preliminary findings on the synthesis products, and consequently a stepwise strategy appeared to
of the first macrobicyclic tetrathiafulvalenophanes with be necessary.
three tetrathiafulvalene bridges grafted on aromatic spa-
The synthesis of the macrobicyclic tetrathiafulvaleno-
cers.[20] In this paper, we report the full experimental details phanes of type 1 and 2 is outlined in Scheme 2. The first
on the synthesis of molecules 1 and 2, including several new step of our reaction scheme involves the selective mono-
examples which have increased the range of molecules in deprotection of one of two different bis-protected tetrathi-
this series, and report for the first time the crystal structure afulvalenethiolates recently described,[15a][b] namely 2,6(7)-
of 2a, as well as cyclic voltammetry studies of the macro- bis(2Ј-cyanoethylthio)-3,7(6)-bis(methylthio)tetrathia-
cycles.
fulvalene 3 (mixture of cis and trans isomers) and 2,3-bis(2Ј-
cyanoethylthio)-6,7-bis(methylthio)tetrathiafulvalene 4.
One thiolate is generated and S-alkylated with the desired
tris(bromomethyl) compound in the appropriate stoichi-
ometry affording tris(tetrathiafulvalenes) of type 5 and 6.
Finally, the three latent thiolates of compounds 5 and 6
were generated using caesium hydroxide monohydrate, and
the macrobicyclic triply bridged cyclophanes were syn-
thesised by a tripod-tripod coupling with the corresponding
tris(bromomethyl) compound under high-dilution con-
ditions.
Scheme 1
Scheme 2
Results and Discussion
Synthesis
As mentioned in the previous section, our strategy is
based on the facile deprotection of tetrathiafulvalenethio-
lates protected by cyanoethyl groups. The cyanoethyl
groups are easily eliminated as acrylonitrile on treatment
with caesium hydroxide monohydrate generating the thiol-
ates which can subsequently be functionalised by reaction
with a wide range of electrophiles. In a first attempt, the
“simpler” and faster strategy of construction of the macro-
bicyclic compounds,[1] a synthetic strategy involving a six-
fold bond formation under high-dilution conditions was
tested, more specifically the reaction of three identical tetra-
i) CsOH·H2O (1.05 equiv.), MeOH, DMF, room temp., N2; ii) tri-
bromide (0.33 equiv.), DMF, room temp., N2; iii) CsOH·H2O
(3.15Ϫ3.6 equiv.), MeOH, DMF, room temp., N2; iv) tribromide (1
equiv.), DMF, high dilution, room temp., N2.
The selectivity of the mono-deprotection of tetrathiaful-
thiafulvalenedithiolates with two identical tris-electrophiles. valenes of type 3 and 4 is different, however (Scheme 3),
The actual synthesis performed was the reaction between with the mono-deprotection of 4 being the cleanest by far.
three tetrathiafulvalenebis(thiolates) and two aromatic spa- A simple experiment shows that slow addition at room tem-
cers bearing three bromomethyl groups, in this case tris- perature of CsOH·H2O (1.05 eq.), dissolved in methanol,
(bromomethyl)benzene, in an attempt to prepare 1a. Unfor- to an N,N-dimethylformamide solution of 3 (1 eq.), fol-
tunately, it results in the precipitation of an insoluble yellow lowed by methylation using an excess of iodomethane, gives
powdery oligomeric or polymeric product which proved dif- a mixture of three products which can be separated by chro-
ficult to characterise, and the desired product 1a was ob- matography on silica gel. The desired 2-(2Ј-cyanoethylthio)-
served only in trace amounts along with significant 3,6,7-tris(methylthio)tetrathiafulvalene 7 is obtained in 67%
amounts of other by-products in the mother liquor. Evi- yield along with tetrakis(methylthio)tetrathiafulvalene 8 re-
dently, more control in the macrocyclisation step was sulting from a bis-deprotection, and unconverted starting
1744
Eur. J. Org. Chem. 1998, 1743Ϫ1757
EurJOC 172/98

Macrobicyclic Tetrathiafulvalene-Bridged Cage Molecules
FULL PAPER
material 3. On the contrary, by subjecting 4 to the same Scheme 4
conditions, 7 is formed as essentially the only product in
94% isolated yield as shown by us earlier.[17] Thus, tris(te-
trathiafulvalene) 5 will inherently be produced in lower
yield and will involve a more elaborate purification step
than does tris(tetrathiafulvalene) 6.
Scheme 3
i) CsOH·H2O (1.1 equiv.), MeOH, DMF, room temp., N2; ii) MeI
excess.
The preparation of tris(bromomethyl)aryl compounds
has been described in the literature (Scheme 4). 1,3,5-Tris-
(bromomethyl)benzene (10) was readily obtained starting
from trimethyl 1,3,5-benzenetricarboxylate (9) using the
convenient one-pot synthesis of aralkyl bromides by re-
ductive halogenation of aromatic carbonyl compounds de-
i) LiAlH4, Et2O, reflux 1 h, N2; ii) HBr(g), Ϫ30°C to room temp.;
iii) NBS, AIBN, CCl4, reflux 3 h; iv) NBS, AIBN, hν, CCl4, reflux
1.5 h.
scribed by Demerseman et al.[21] 1,3,5-Tris[4-(bromometh-
yl)phenyl]benzene[22] (13), 1,3,5-tris[3-(bromomethyl)phen-
yl]benzene[22] (14), and 1,1,1-tris[4-(bromomethyl)phenyl]-
ethane[23] (16) were prepared by radical bromination of
their corresponding methyl derivatives 11, 12, and 15 using
N-bromosuccinimide (3 equiv.) following the procedure re-
ported by Vögtle et al. The NBS bromination route is much
less convenient than the Demerseman bromination proto-
col, as the tris(bromomethyl) compounds prepared by NBS
bromination are always contaminated by products of asym-
Table 1). Under the same conditions, but starting from
tetrathiafulvalene 4, tris(tetrathiafulvalenes) 6a؊d were pre-
pared in significantly higher yields than the tris(tetrathiaful-
valenes) 5, for the reasons mentioned above. Note that com-
pounds 5 were obtained as a mixture of four cis/trans iso-
mers which could not be separated by column chromatogra-
phy
on
silica
gel.
Whereas
purification
of
1
metric bromination and overbromination (detected by H-
tris(tetrathiafulvalene) 6 was easily achieved (essentially one
spot in TLC), several by-products were observed during
chromatographic separation of tris(tetrathiafulvalene) 5.
The most prominent by-product resulting from alkylation
of a combination of one bis-deprotected and four mono-
deprotected tetrathiafulvalenes by the tribromide com-
pound afforded pentakis(tetrathiafulvalene) 17 as a mixture
of isomers (Scheme 5). Although this reaction seems to be
general during the preparation of compounds of type 5 as
evidenced by TLC, we only isolated and characterised com-
pounds 17a and 17b in 18% and 11Ϫ16% yield, respec-
tively. These oligomeric tetrathiafulvalenes are fully charac-
NMR and electron-impact mass spectrometry) due to the
unselective nature of the reagent. Purification of the reac-
tion mixture employing the experimental procedure already
reported, afforded compounds 13, 14, and 16 in 68%, 60%,
72% of purity respectively [HPLC analysis, reverse phase
Nucleosil 5 C18, CH3CN/H2O (9:1), 274 nm]. Further puri-
fication by column chromatography and recrystallisations
(see Experimental Section) allowed us to isolate the desired
compounds 13 and 14 in 93% and 69% purity respectively.
Cage Syntheses
For the synthesis of the tetrathiafulvalene cages we em- terised and stable under atmospheric conditions. Recently,
ployed a strategy developed in our group, based on thiolate Bryce et al. prepared and characterised a new air-stable
alkylation of an in situ generated thiolate function. Thiolate dendritic molecule incorporating thirteen tetrathiafulvalene
alkylations are highly efficient reactions, which have proven units[25a] while in a previous paper, a dodecakis(tetrathiaful-
to be generally applicable for the synthesis of macropoly- valene) derivative was unstable.[25b] The pentakis(tetrathia-
cyclic compounds.[24] Thus, upon treatment of a DMF solu- fulvalenes) 17, having four protected thiolates, may be used
tion of a mixture of cis/trans isomers of the tetrathiafulva- as precursors to dendritic molecules containing multiple re-
lene 3 (1 equiv.) with caesium hydroxide monohydrate (1.05 dox units.[26][27][28]
equiv.) in methanol and subsequent alkylation with 10, 13,
Construction of the macrobicyclic compounds 1 and 2
14, and 16 (0.32 equiv.) in DMF, tris(tetrathiafulvalenes) was achieved by the simultaneous addition of a DMF/
5a؊d were obtained in the range of 27Ϫ37% yields after MeOH solution of the fully deprotected tetrathiafulvalene-
purification by column chromatography on silica gel (see thiolate 5 or 6 (1 equiv.) and a DMF solution of the appro-
Eur. J. Org. Chem. 1998, 1743Ϫ1757
1745
EurJOC 172/98

P. Blanchard, N. Svenstrup, J. Rault-Berthelot, A. Riou, J. Becher
Scheme 6
FULL PAPER
Table 1. Isolated yields after purification by chromatography for 1,
2, 5, and 6
Compound
Yield (%)
b
a
c
d
5
6
1
2
27Ϫ30
74Ϫ77
43
34Ϫ36
75
37
73
37
22
36
84
40
30
39
68
32
i) CsOH·H2O (1.05 equiv.), MeOH, DMF, room temp., N2; ii) tri-
bromide 10 (0.33 equiv.), DMF, room temp., N2; iii) CsOH·H2O
(1.05 equiv.), MeOH, DMF, room temp., N2; iv) tribromide 10
(0.33 equiv.), DMF, room temp., N2.
Scheme 5
valene) 6a was immediately treated with caesium hydroxide
to cleave off the remaining protecting groups. The resulting
trithiolate was realkylated by the dropwise addition 0.33
equiv. of 1,3,5-tris(bromomethyl)benzene (10) to give 2a as
the only product in 68% yield after the usual purification
by column chromatography. This result demonstrates the
efficiency of each individual step since a one-pot reaction
of this type would be impossible if an incomplete synthetic
step was involved. The synthesis of 2a is an example of an
assisted self-assembly reaction, since all the individual com-
ponents of the macrocycle appear to be pre-programmed
for cyclisation due to their complementary geometry. It is
rare that a reaction, in which 6 new covalent bonds and 2
macrocyclic rings are formed, proceeds in such a high yield.
priate tribromide 10, 13, 14, or 16 (1 equiv.) to neat DMF
under high dilution conditions using a perfusor pump. The
new macrobicyclic tetrathiafulvalenes were isolated as yel-
Characterisation by Spectral Analyses
1H-NMR spectra of compounds 5 (Table 2) show the
low or orange amorphous powders after purification by col- presence of cis/trans isomers, demonstrated by two different
umn chromatography on silica gel using CH2Cl2/light signals for each methylene group of the cyanoethylthio
petroleum ether (1:1) as eluent. The synthetic yields ob- group and two signals also for the methylthio groups in the
tained for 1a؊d and 2aϪd are collected in Table 1. Except β-position of the cyanoethylthio substituents. Compounds
1
for 2a, which was obtained in exceptionally high yield, the 6 displays much simpler H- and 13C-NMR spectra. One
yields of 1 were only slightly better than those obtained may note that the proton signals of the cyanoethylthio
for 2.
groups of 6 are slightly shifted upfield compared to those
Considering the high yield in the individual steps of the of 5 (on average 0.27 ppm for SCH2 and 0.14 ppm for
synthesis of the macrocycle 2a, a one-pot reaction was en- CH2CN) indicating a stronger influence of the shielding ef-
visaged (Scheme 6). Thus, using the same reaction con- fect by aromatic spacers in 6. That is maybe why the triplets
ditions as before, the clean reaction mixture (as evidenced of 6b and 6c assigned to the SCH2 moiety of the cyano-
by TLC) containing the freshly generated tris(tetrathiaful- ethylthio group are observed at lower chemical shifts than
1
Table 2. H-NMR chemical shifts of tris(tetrathiafulvalenes) 5 and 6
Com-
pound
Chemical shifts (δ values relative to TMS)
CH3 (s)
SCH3 (s)
SCH3 (s)[b]
SCH2 (t)
CH2CN (t)
ArCH2S (s)
Aromatic H
5a
5b
5c
Ϫ
Ϫ
Ϫ
2.32
2.47 and 2.48
2.44 and 2.45
2.42 and 2.43
2.71 and 2.72
2.67 and 2.68
2.65 and 2.66
3.03 and 3.04
2.99 and 3.00
2.97 and 2.99
3.97
4.07
4.12
7.18 (s)
2.27 and 2.28
2.27
7.43 (d), 7.64 (d), 7.74 (s)
7.35 (d), 7.45 (dd), 7.62
(d), 7.67 (s), 7.79 (s)
7.02 (d), 7.21 (d)
7.22 (s)
5d
6a
6b
2.11
Ϫ
Ϫ
2.26 and 2.27
2.44
2.42 (18 H)
2.45 and 2.46
2.45
2.34
2.68 and 2.69
2.48
2.84
3.03 and 3.15
2.90
4.09
3.98
4.02
7.42 (d),
7.67 (d), 7.77 (s)
4.14
6c
6d
[a]
Ϫ
2.38
2.41
[b]
2.41
2.43
2.36
2.48
2.82
2.89
7.34 (d), 7.45 (dd), 7.66
(d), 7.69 (s), 7.82 (s)
7.04 (d), 7.23 (d)
2.14
4.03
250 MHz, CDCl3 at 25°C. Ϫ In β position of the cyanoethylthio group.
1746
Eur. J. Org. Chem. 1998, 1743Ϫ1757
EurJOC 172/98

Macrobicyclic Tetrathiafulvalene-Bridged Cage Molecules
FULL PAPER
1
Table 3. H-NMR chemical shifts of macrobicycles 1 and 2
Compound
Chemical shifts (δ values relative to TMS)
CH3
SCH3
ArCH2S
Aromatic H
1a
1b
1c
Ϫ
Ϫ
Ϫ
2.38Ϫ2.49 (m)
2.01Ϫ2.26 (m)
2.14Ϫ2.25 (m)
3.83-4.09 (m)
3.94-4.08 (m)
3.95-4.00 (m)
6.99Ϫ7.23 (m, 6 H)
7.06Ϫ7.63 (m, 30 H)
7.23Ϫ7.42 (m, 12 H), 7.53Ϫ7.58
(m, 12 H), 7.70Ϫ7.75 (m, 6 H)
1d
2a
2b
1.25Ϫ2.46 (m, 24 H)
Ϫ
Ϫ
3.75-3.96 (m)
2.44 (s)
2.42 (s)
6.61Ϫ7.23 (m, 24 H)
3.84 (s)
3.99 (s)
7.06 (s, 6 H)
7.16 (d, 12 H), 7.28 (d, 12 H), 7.33 (s, 6
H)
2c
Ϫ
2.24 (s)
3.70 (s)
7.15Ϫ7.43 (m, 30 H)
6.91 (d), 7.06 (d)
2d (“out-out”)
2d (“in-out”)
2.06 (br. s, 6 H)
1.17 (br. s, 3 H “in”) (m)
2.10 (br. s, 3 H “out”)
2.42 (s)
3.85 (s)
2.40 and 2.42 (2s)
3.94 and 4.07 (s)
6.96Ϫ7.02 (m), 7.15Ϫ7.22 (m)
[a]
250 MHz, CDCl3 at 25°C.
the primary methyl signal of the methylthio groups. Due to in guest complexation selectivity between two isomeric “in-
1
the cis/trans isomers, it is difficult to assign the H-NMR out” and “out-out” host molecules.[30a] In our case, sepa-
spectra of the macrobicyclic systems 1 (Table 3). However, 1 ration by column chromatography of the two isomers was
displays three different main multiplets assigned to ϪSCH3, not possible.
ϪSCH2Ϫ, and aromatic protons (Figure 1). CPK models
Figure 1. 1H-NMR (250 MHz) spectra in CDCl3 at 25°C of 1a (top)
show that protons of different cis/trans isomers can be sub-
and 2a (bottom)
jected to a shielding or a deshielding effect due to the dia-
magnetic anisotropy of the aromatic spacers. Consequently,
each of the three types of signal (ϪSCH3, ϪSCH2Ϫ, and
aromatic H) give broad multiplets. In the case of the macro-
bicyclic compounds 2, the aromatic protons are shielded
compared to 6 and have relatively low chemical shifts com-
pared to those of 1 (except for 2d) indicating relatively
stronger interactions between the aromatic spacers in 2 due
to the small size of the cavity of 2. The high symmetry of
1
2a is clearly evident in its extremely simple H-NMR spec-
trum, which displays three sharp singlets at δ ϭ 7.06 (aro-
matic H), δ ϭ 3.84 (ϪSCH2Ϫ) and δ ϭ 2.44 (ϪSCH3) in
1
the ratio of 1:2:3. This H-NMR spectrum is only consist-
ent with the macrobicyclic compound 2a having an overall
D3h symmetry in CDCl3 solution on the NMR time scale at
room temperature. The 13C-NMR spectrum of 2a likewise
reflects the high symmetry of the molecule: 8 lines in total
(2 in the aliphatic and 6 in the aromatic region) is evident
of the high symmetry of 2a. On the contrary, compound 2d
exhibits a more complicated spectrum, as 2d is an approxi-
mately 2:1 mixture of two different cage compounds
(Scheme 7), namely one symmetrical cage (D3h symmetry)
with two methyl groups pointing out of the cavity (“out-
out” isomer) and one unsymmetrical cage with one methyl
Mass-spectrometry studies confirmed the structure of
group pointing into the cavity and one pointing out of the
our compounds.[*] Figure 2 shows the PDMS spectrum of
1d.
cavity (“in-out” isomer). The “in” methyl group is signifi-
cantly shielded compared to the “out” methyl group as evi-
denced by its signalЈs upfield shift (δin ϭ 1.17 and δout
ϭ
X-ray Structure
2.10). CPK models suggest that the construction of such
an “in-out” isomer is sterically possible whereas the “in-in”
isomer cannot be prepared, except in 1d. Although not
often observed, the in/out isomerism is a well-known fea-
ture of bridged bicyclic structures containing small or me-
dium-sized rings[29] as reported by Vögtle et al,[30] who,
Attempts to grow single crystals of the compounds 1 and
2 were unsuccessful except in the case of 2a, for which thin
[*] As commonly reported in the case of tetrathiafulvalene deriva-
tives, small amounts of tetrathiafulvalene S-oxides (Mϩ• ϩ 16)
were detected in the case of macrobicycles 1 and 2 using Plasma
using the same aromatic spacers, demonstrated differences Desorption Mass Spectrometry (PDMS).
Eur. J. Org. Chem. 1998, 1743Ϫ1757
1747
EurJOC 172/98

P. Blanchard, N. Svenstrup, J. Rault-Berthelot, A. Riou, J. Becher
FULL PAPER
Scheme 7
twisted, while the other half seemed to keep their appear-
ance, suggesting the presence of two different crystalline
phases. X-ray structure determination performed on the
stable needles revealed the structure of 2a (Figure 3). The
structure belongs to the P21/n molecular point group (see
Experimental Section). Several unusual aspects of this
1
structure are noteworthy. Contrary to the H-NMR study,
where a D3h symmetry was observed in CDCl3 solution, it
appears that the molecule 2a adopts a different confor-
mation in the solid state. Indeed, two tetrathiafulvalene
units (A and B) are approximately parallel whereas the third
one (C) is orientated almost perpendicularly to the parallel
tetrathiafulvalene pair. Each tetrathiafulvalene framework
adopts a different conformation. Tetrathiafulvalene unit A
is the most planar; its external dithiole ring folds only by
ca. 13° along the S···S vector. In the case of tetrathiafulva-
lene units B and C, the dithiole moieties close to the aro-
matic spacer exhibit an envelope conformation giving rise
to an angle of ca. 25° for both of them between the central
tetrathioethylene mean plane of the tetrathiafulvalene core
and the dithioethylene mean plane of the dithiole moiety.
Furthermore the benzene rings do not overlap face to face
and hence, they tilt 33° away from each other. The shortest
distance between two carbon atoms of two different ben-
long orange needles were finally obtained after slow dif-
fusion of ethanol into a CS2 solution of 2a. Nevertheless,
after a few hours under N2 or atmospheric conditions, ap-
proximately half of the needles became amorphous and
˚
zene rings is 4.21 A which is much more than in molecules
with stacked benzene rings.[31] There is no interaction be-
tween the two tetrathiafulvalenes lying parallel since the
shortest sulfurϪsulfur distances are comprised between
Figure 2. Plasma desorption mass spectra of 1d
1748
Eur. J. Org. Chem. 1998, 1743Ϫ1757
EurJOC 172/98

Macrobicyclic Tetrathiafulvalene-Bridged Cage Molecules
FULL PAPER
Figure 3. X-ray structure of macrobicycle 2a
Figure 4a. Formation of tetrathiafulvalene columnar arrays along
the a axis via bimolecular assemblies; dotted lines represent S···S
˚
distances shorter than the sum of the van der Waals radii (3.70 A)
molecules fit together giving rise to a bimolecular assembly
showing an inversion centre where four tetrathiafulvalene
units lie roughly parallel. Inside the molecular unit, tetrathi-
afulvalenes A and BЈ as well as tetrathiafulvalenes B and
AЈ show a ring-over-double bond overlap. This molecular
arrangement is stabilised by interactions between parallel
tetrathiafulvalenes as evidenced by the presence of numer-
ous short S···S distances (Figure 4a). Figure 4b shows a
view of the unit cell. The repetition of the bimolecular unit
approximately along the a axis and the short S···S distances
observed between tetrathiafulvalenes AЈ and AЈЈ, lead to
the existence of columnar arrays of tetrathiafulvalenes.
Moreover, short S···S distances are observed between tetra-
thiafulvalenes of type C along the c axis [d(S22ϪSЈ19) ϭ
˚
6.59 and 7.11 A. On the other hand, there is one S···S intra-
˚
˚
molecular contact shorter than the sum of the van der
Waals radii (3.70 A) between two perpendicular tetrathiaf-
ulvalenes [d(S15ϪS22) ϭ 3.54 A]. This short distance cannot
explain the loss of the C3 symmetrical axis. A better expla-
3.68 A, d(S19ϪSЈ20) ϭ 3.61 A] giving rise to the formation
of tetrathiafulvalene dimers.
˚
˚
Electrochemistry
nation can be found in the peculiar mode of packing of
The redox behaviours of 1, 2, 5, 6, and 17 were investi-
compound 2a in the crystal. As illustrated in Figure 4a, two gated by cyclic voltammetry (Table 4) using a mixture of
Figure 4b. Packing of the molecules in the crystal
Eur. J. Org. Chem. 1998, 1743Ϫ1757
1749
EurJOC 172/98

P. Blanchard, N. Svenstrup, J. Rault-Berthelot, A. Riou, J. Becher
FULL PAPER
solvents CH2Cl2/CH3CN (3:1).[*] The new macrobicycles Figure 5. Cyclic voltammograms of macrobicycles 1a (ϪϪ) and 2a
(···) (CH2Cl2/CH3CN, 3:1; nBu4NPF6, 0.1 mol·lϪ1; Ag/AgCl; sweep
1b؊d, 2b؊d, and the new tris(tetrathiafulvalenes) 5b؊d
rate 100 mV sϪ1
)
and 6b؊d exhibit two reversible redox processes. As re-
ported for the analogues of 5 and 6[6a][32][33][34] and as-
suming the same peak area for each redox process, we can
expect that the first and the second waves result from simul-
taneous formation of three radical cations followed by three
dications at higher potentials. The fact that the two redox
waves are well resolved, shows that the large and rigid aro-
matic spacers (b, c, d) prevent Coulombic through-space in-
teraction between the three redox groups.
Table 4. Cyclic voltammetry data
Compound
Eox [V]/Ered [V]
TTF[b]
TTF
1a
1b
1c
0.37/0.28
0.41/0.32
0.775/0.69
0.88/0.79
0.89/0.85
0.86/0.80
0.87/0.79
0.88/0.80
0.86/0.78
0.86/0.79
0.91/0.80
0.89/0.82
0.91/0.84
0.91/0.83
0.875/0.815
0.87/0.81
0.89/0.82
0.89/0.82
0.88/0.82
0. 88/0.81
0.89/0.81
0.91/0.82
0.87/0.795
0.88/0.82
0.88/0.80
0.43/0.38, 0.53/0.47, 0.62/0.56
0.53/0.49
0.56/0.47
0.57/0.48
the cyclic voltammograms of 1a and 2a, approximately
three different redox waves are observed for the successive
formation of a mono-, a bis-, and a tris(radical cation) spe-
cies (E11/2 ϭ 0.405 V, E21/2 ϭ 0.50 V, E31/2 ϭ 0.59 V vs. Ag/
AgCl) in the case of 1a. Due to the architecture of macrocy-
cle 1a (CPK models), the three redox units are in close
proximity and can interact easier and stronger than in the
more strained macrobicycles 2a where the electroactive
groups are pointing out of the cavity. Consequently, the ex-
pected “first” redox waves of 1a are broader than for 2a,
giving rise to almost three redox waves in the case of 1a
and only two for 2a. The low values of the oxidation poten-
tials of the first two waves of 1a are consistent with a stabi-
lisation by folding of one radical cation tetrathiafulvalene
1d
2a
2b
2c
0.48/0.43, and 0.57/0.49
0.54/0.47
0.58/0.47
2d
3
0.58/0.48
0.59/0.52
4
0.595/0.525
5a
5b
5c
0.51 (sh) and 0.60/0.47 (br.)[c]
0.56/0.495
0.58/0.50
0.575/0.495
5d
6a
6b
6c
0.52 (sh) and 0.61/0.49 (br.)
0.58/0.51
0.59/0.50
6d
7
17a
17b
0.59/0.51
0.555/0.49
0.585 (br.)/0.42 (sh) and 0.51
0.57/0.50
moiety and a close neutral tetrathiafulvalene to form a di-
[a]
Reference electrode: Ag/AgCl; working and counter electrodes:
•ϩ [33]
meric radical species (TTF)2
.
Whereas a folding of the
platinum; sweep rate: 100 mV sϪ1; solvent: CH2Cl2/CH3CN (3:1);
supporting electrolyte: nBu4NPF6 (0.1 mol·lϪ1). Ϫ In CH3CN. three tetrathiafulvalene groups is possible in 1a, this is not
[b]
[c]
Ϫ
sh: shoulder, br.: broad.
the case for 2a where only two tetrathiafulvalene groups can
come close (CPK models). In the last case, the broader wave
observed results from the successive removal of electrons
leading to the 2a3•ϩ species probably in two steps. The
single well-defined wave obtained for the generation of the
six-fold charged macrobicycles 1a6ϩ supports the fact that
tris(radical cation) species generated from 1a, as is also the
case for 1b؊d, 2a؊d, 5a؊d, and 6a؊d, all have perfect C3
symmetry in solution.
On the contrary, in the case of 1a, 2a, 5a, and 6a, where
the tetrathiafulvalenes are linked by the small 1,3,5-tri-
methylenebenzene spacer(s), cyclic voltammograms show
broader waves, shoulders or additional waves for the gener-
ation of the radical cation state below 0.7 V (Table 4), simi-
lar to related oligomeric tetrathiafulvalenes.[6a][33] This elec-
trochemical behaviour is a result of the flexibility in the
molecule and a close proximity of the three redox moieties
which allow Coulombic through-space interactions. How-
ever at higher potentials (Eox > 0.7 V) a single sharp wave
is observed with the same peak area than the sum of the
previous ones. Assuming that the first processes involved
three electrons, a simultaneous loss of three electrons occurs
for the formation of a six-fold charged tris(dicationic) state;
at this stage, the absence of interactions between electroac-
tive centres is commonly observed.[33] In Figure 5, showing
The pentakis(tetrathiafulvalene) 17b contains five inde-
pendent electroactive tetrathiafulvalene units as evidenced
by its cyclic voltammogram, which exhibits two well-re-
solved reversible redox processes which can be assigned to
the successive formation of a five-fold charged and a ten-
fold charged species.[**]
TLCV[25a][35] studies in 1,1,2-trichloroethane solution in
the presence of chloranil as an internal reference have been
performed on compounds 2a and 17b. Since chloranil gives
a one-electron reduction peak, the number of electrons ex-
[*] When CH2Cl2 only was used, in some cases the second oxidation
waves (Eox > 0.7 V) were distorted due to adsorption processes at
the electrode, as commonly observed with large donor molecules.[32] [**] Using Thin Layer Cyclic Voltammetry (TLCV), Bryce et al. have
The sufficient addition of CH3CN allows a complete solubility of recently shown that an analogous tetrathiafulvalene pentamer gave
[25a]
the adsorbed species giving a well-defined redox process.
rise to two five-electron single waves upon electrooxidation.
1750
Eur. J. Org. Chem. 1998, 1743Ϫ1757
EurJOC 172/98

Macrobicyclic Tetrathiafulvalene-Bridged Cage Molecules
FULL PAPER
changed per oxidation wave can be calculated by integrating
the voltammetric waves if the concentrations of the sample
and chloranil are known. The calculated value (4.8 ± 0.2
from different experiments) at the first oxidation potential
of 17b is close to the expected value. In the case of 2a,
experiments were complicated by interactions with the in-
ternal reference (chloranil, dichloronaphthoquinone). How-
ever, when studied alone, 2a oxidation presents two waves,
the first one exhibiting a shoulder. The area of each wave
is identical. Furthermore, the use of ferrocene as internal
reference solves the interaction problems observed before.
The oxidation potential of ferrocene lies in the potential
range of 2a first oxidation wave. Calculations taking into
account the integration of the second wave (belonging to
2a) and the first wave (belonging both to ferrocene and 2a)
show that each wave is a 0.75 electron process. Conse-
quently, three electrons are exchanged for two molecules of
2a. This results suggest the occurrence of strong intermolec-
ular interactions between tetrathiafulvalene units in the thin
layer, which is not surprising when looking at the X-ray
structure of 2a. Figure 6 shows the TLCV of compound
17b.
and molar mass respectively. This equation was recently
used for the determination of the number of electrons in-
volved in the first reduction step of a multi-C60 deriva-
tive.[37]
The subscript p refers here to our compounds 1, 2, 5,
6, and 17, and the subscript m corresponds to a reference
compound chosen for similarity reasons to be either tetra-
kis(methylthio)tetrathiafulvalene (8) (in the case of 1 and 2)
or 2-(2Ј-cyanoethylthio)-3,6,7-tris(methylthio)tetrathiaful-
valene (7) (in the case of 5, 6 and 17). For one redox pro-
cess, i is defined by Equation (2) where ipa is the peak an-
odic current at Epa and ipc the peak cathodic current at Epc,
and where ia and ic correspond to the anodic and cathodic
currents of the couple solvent/supporting electrolyte deter-
mined with a solution free of electroactive compounds at
Epa and Epc, respectively . When interactions between tetra-
thiafulvalenes occur at the first redox process, np has been
calculated using the second redox process. In order to ob-
tain more accurate concentrations, these experiments were
achieved in a mixture of solvents CH2Cl2/CH3CN (9:1),
thus allowing a better solubilisation of our compounds. The
experimental values i, c, and np are collected in Table 5.
In the case of 5 and 6 these results confirm our previous
interpretation, i.e. a three-electron transfer for the forma-
tion of a tris(radical cation) and then, another three-elec-
tron transfer for the formation of a six-fold charged species.
Hence, this study shows that each of the two redox waves
of 17 involves a five-electron transfer leading to a ten-fold
charged compound. In the case of 17a, the second redox
wave showed the presence of adsorption at the electrode
during the reduction process, giving rise to an overestimated
np value (6.9). Thus, despite its slight broadening, the first
redox wave of 17a is more appropriate for the determination
of np which is found to be close to five electrons. Conse-
quently, these data reveal that, within an experimental error
of ca. 10%, all the tetrathiafulvalene units present in each
molecule can be oxidised.[38] On the other hand, while np is
close to three exchanged electrons for 1b, 2b, and 2d, the
cases of 1a and 2a, where tetrathiafulvalenes can interact,
remain ambiguous.
Figure 6. Thin layer cyclic voltammograms of compound 17b
(7.5·10Ϫ5 mol·lϪ1) in the presence of chloranil (6.90·10Ϫ4 mol·lϪ1
)
;
as internal reference (1,1,2-trichloroethane; nBu4NPF6, 1 mol·lϪ1
reference vs. SCE; sweep rate 5 mV sϪ1
)
Charge-Transfer Complexation
Charge-transfer complexation experiments were investi-
gated with the macrobicycles 1a, 1b, 2a, 2b, and also with
the tris(tetrathiafulvalene) 6b by UV/Vis spectroscopy.
Thus, CH2Cl2 solutions of each compound rapidly became
In order to gain further insights into the determination dark in the presence of iodine. The formation of charge-
of the number of electrons np exchanged in the redox pro- transfer complexes was confirmed by the emergence of a
cesses of our new compounds, cyclic voltammetric measure- broad band in the UV/Vis spectra: λmax ϭ 826, 839, 862,
ments were made using Equation (1) of Bard and Anson 869, 827 nm, for 1a, 1b, 2a, 2b, and 6a, respectively . Ad-
which was originally applied for polymeric molecules con- dition of equimolar amounts of either AgClO4 or CuCl2 ·2
[36]
taining identical non-interacting electroactive centers,
H2O, dissolved in CH3CN, to a CH2Cl2 solution of 2a led
and where i, c, and M represent the intensity, concentration, to the formation of black powders corresponding to charge-
Eur. J. Org. Chem. 1998, 1743Ϫ1757
1751
EurJOC 172/98

P. Blanchard, N. Svenstrup, J. Rault-Berthelot, A. Riou, J. Becher
FULL PAPER
Table 5. Determination of the number of electrons np involved in lene) oligomers for the determination of the number of elec-
each redox process using the Bard-Anson equation
trons exchanged in a non-interacting electrochemical pro-
cess. Preliminary host-guest chemistry with our macro-
bicyclic compounds gave poor results suggesting the syn-
thesis of new molecules using alternative spacers and new
functionalised tetrathiafulvalene derivatives.
Com- Amount Mol. c [mol·lϪ1] i [µA] E1/2 [V] np ±
[b]
pound taken [mg] mass
vs. SCE ∆np
8
0.82
Љ
388
Љ
5.28·10Ϫ4 2.36 0.50
1.00
Љ
2.32 0.82
1.00
1a
1b
2a
2b
2d
7
0.50
0.67
0.70
1.00
0.89
1.65
Љ
0.80
0.73
0.72
0.91
1.45
0.87
Љ
1310
1767
1310
1767
1671
427
Љ
4.24·10Ϫ5 0.31 0.88[c]
4.74·10Ϫ5 0.36 0.52
5.34·10Ϫ5 0.34 0.86[c]
5.96·10Ϫ5 0.46 0.52
6.66·10Ϫ5 0.49 0.52
4.83·10Ϫ4 1.73 0.535
2.3 ± 0.2
2.6 ± 0.3
2 ± 0.2
2.6 ± 0.3
2.5 ± 0.3
1.00
Experimental Section
All solvents were purified/dried by standard methods prior to
use. The petroleum ether used had a boiling range of 60Ϫ80°C. All
reagents used were obtained from Aldrich or Fluka and used with-
out purification. Microanalyses were performed at the Microana-
lytical Laboratory, University of Copenhagen. The silica gel used
was Merck 60, 230Ϫ400 mesh. Ϫ NMR spectra were recorded with
a Bruker AC 250; chemical shifts are given in ppm relative to TMS.
Ϫ EI mass spectra were recorded with a Varian MAT 311A spec-
trometer. PDMS and FAB spectra were recorded with a BioIon
10K and a Kratos MS50TC, respectively. Ϫ IR spectra were re-
corded with a Perkin-Elmer 580 spectrophotometer and UV/Vis
spectra with a Shimadzu 160A spectrophotometer. Ϫ Melting
points were determined with a Büchi apparatus and are uncor-
rected. Ϫ CV measurements were carried out with nBu4NPF6 as
supporting electrolyte, with a sweep rate of 100 mV/s. Counter and
working electrodes were made of Pt, and the reference electrode
was Ag/AgCl, unless otherwise stated. Ϫ The thin layer cyclic vol-
tammetry cell used in this work was constructed as described pre-
viously.[35b]
Љ
1.76 0.865
1.00
5a
5b
5d
6a
6b
17a
1355
1.102·10Ϫ4 0.98 0.87[c]
3.3 ± 0.3
3.3 ± 0.3
3.1 ± 0.3
3 ± 0.3
3 ± 0.3
1583.5 7.68·10Ϫ5 0.63 0.54
1535
1355
8.75·10Ϫ5 0.69 0.55
1.36·10Ϫ4 1.07 0.86[c]
1583.5 1.22·10Ϫ4 0.91 0.55
2244
Љ
2700
7.25·10Ϫ5 0.82 0.55[d] 5 ± 0.5
Љ
1.15 0.87[c][e] 6.9 ± 0.7
5.23·10Ϫ5 0.60 0.54
5.3 ± 0.5
17b
1.13
[a]
Reference electrode: SCE; working and counter electrodes: plati-
num; sweep rate: 100 mV sϪ1; solvent: CH2Cl2/CH3CN (9:1); sup-
porting electrolyte: nBu4NPF6 (0.1 mol·lϪ1). Ϫ
∆np takes into
[b]
account all experimental errors and was estimated to be ∆np ϭ 10%
[c] n
[d]
np. Ϫ
was calculated using the second redox wave. Ϫ Broad
p
[e]
wave. Ϫ Slight adsorption at the electrode during the reduction
process.
transfer complexes. Electrocrystallisations of macrobicycles
Ϫ
X-ray Structural Analysis: Crystal data for 2a: C42H36S24, Mw
ϭ
2a (9 mg) at 0°C under 1 µA using PF6 as anion and a
1310.19 g·molϪ1, monoclinic, P21/n, Z ϭ 4, a ϭ 14.226(7), b ϭ
CH2Cl2/EtOH (9:1) mixture as solvent, gave black plates in
the bottom of the cell after nine days, however unsuitable
for X-ray diffraction.
8.964(2), c ϭ 42.297(9) A, β ϭ 92.02(9)°, V ϭ 5398(1) A3, ρcalcd. ϭ
˚
˚
1.612 g·cmϪ3, λ(Mo-Kα) ϭ 0.71073 A. Ϫ Data collection: Data
˚
collection was carried out by zig-zag ω scan technique (2° < θ <
25°) with an Enraf-Nonius Mach III diffractometer. Conditions of
measurements were tmax ϭ 40 s, range h, k, l (h 0 to10; k 0 to 16;
l Ϫ50 to 50). Intensity-control reflections were measured every 2 h
without appreciable decay (0.15%). 10597 independent reflections
were collected from which 2701 correspond to I > 3σ(I). Ϫ Struc-
ture refinement: After Lorentz and polarisation corrections, the
structure was solved by direct methods (SIR) which reveal all the
non-hydrogen atoms. After anisotropic refinement of all the C and
S atoms, the coordinates of H atoms were determined from the
HYDRO program. The whole structure was refined by full-matrix
least-squares techniques {refinement on F, x, y, z, Uij for S atoms,
x, y, z and U for C atoms; 390 variables and 2701 observations,
weighting scheme: non-Poisson contribution with w ϭ 1/σ(Fo)2 ϭ
Conclusion
In this work, we have established an efficient method-
ology for the assembly of the first macrobicyclic cyclotetra-
thiafulvalenophanes 1a؊d and 2a؊d. They have been pre-
pared using a two-step strategy based on the deprotection
of protected thiolates and subsequent S-alkylation. It is
worth noting that a relatively complicated molecule, such
as 2a, can be obtained in good yield in a one-pot reaction
starting from the easily available tetrathiafulvalene deriva-
tive 4. Isolation of the by-products, tetrathiafulvalene pen-
tamers 17, allowed a better understanding of the reaction
leading to tris(tetrathiafulvalene) 5. 1H-NMR and cyclic
voltammetry studies of these new compounds gave precise
information on their conformational structures in solution.
In the case of 2a, where the crystal structure was estab-
lished, a completely different conformation was observed in
4Fo2/[σ(I)2 ϩ (0.04 Fo2)2] with the resulting R ϭ 0.099, Rw
ϭ
0.119}. All calculations were performed using the MolEN package.
2-(2Ј-Cyanoethylthio)-3,6,7-tris(methylthio)tetrathiafulvalene (7)
by Monodeprotection of the cis/trans Mixture of 2,6(7)-Bis(2Ј-
cyanoethylthio)-3,7(6)-bis(methylthio)tetrathiafulvalene (3): Com-
the solid state as a result of S···S intermolecular interac- pound 3 (0.466 g, 1.0 mmol) was dissolved in anhydrous degassed
DMF (50 ml) under N2 at room temp., and a solution of
CsOH·H2O (0.179 g, 1.05 mmol) in anhydrous degassed MeOH
(10 ml) was added dropwise over a period of 1 h. After stirring for
another 1 h, MeI (1.0 ml, excess) was added in one portion and
the solution was stirred overnight. After concentration in vacuo,
the residue was redissolved in CH2Cl2 (150 ml), the solution was
washed with water (3 ϫ 50 ml), dried (MgSO4), and concentrated.
Compound 7 was isolated after column chromatography on silica
gel using CH2Cl2 as eluent (0.287 g, 67%); Rf ϭ 0.55 (CH2Cl2);
m.p. 103Ϫ104°C (ref.[17] 102Ϫ104°C from toluene/cyclohexane). Ϫ
tions. The reaction protocol described in this paper can be
applied to the synthesis of a number of new macrobicycles
either by introduction of new substituents (lipophilic or
hydrophilic) instead of the methylthio group, or by incor-
poration of new spacers. Furthermore intermediate tris(te-
trathiafulvalenes) 5a؊d and 6a؊d or pentakis(tetrathiaful-
valenes) 17a, b with potential thiolates can be seen as new
building blocks for the preparation of new electron rese-
voirs.[26][27] In addition, we have shown that the equation of
Bard and Anson could be applied to multi(tetrathiafulva- 1H NMR (CDCl3): δ ϭ 2.43 (s, 6 H, SCH3), 2.47 (s, 3 H, SCH3),
1752
Eur. J. Org. Chem. 1998, 1743Ϫ1757
EurJOC 172/98

Macrobicyclic Tetrathiafulvalene-Bridged Cage Molecules
FULL PAPER
2.70 (t, 2 H, SCH2, J ϭ 7.2 Hz), 3.02 (t, 2 H, CH2CN, J ϭ 7.2
leum ether, 1:2); 3.79 g (45%), m.p. 185Ϫ187°C (ref.[22] m.p.
Hz). Ϫ 13C NMR (CDCl3): δ ϭ 18.74, 19.13, 19.21, 31.23, 109.45, 186Ϫ187°C). However, HPLC analysis showed a 60% purity of the
112.57, 117.54, 120.13, 127.41, 135.13. Ϫ Two by-products were desired product and the presence of two other products of muti-
1
identified by H-NMR spectroscopy and TLC as 2,3,6,7-tetrakis(-
bromination in 35% and 5%. A sample of higher purity was ob-
methylthio)tetrathiafulvalene (8) (0.073 g, 19%, Rf ϭ 0.82 in tained by column chromatography on silica gel using a mixture
CH2Cl2) and the starting material 3 (0.053 g, 11% recovered, Rf ϭ CH2Cl2/petroleum ether (1:2) as eluent, followed by two successive
0.26 in CH2Cl2).
recrystallisations from toluene/acetic acid (4:3) and one recrystalli-
sation from CH2Cl2/petroleum ether. The final sample showed only
1,3,5-Tris(bromomethyl)benzene (10):[21] Starting material 9 was
prepared by reaction of the corresponding acid chloride with meth-
anol in the presence of trimethylamine in refluxing chloroform in
93% yield. White crystals. M.p. 144Ϫ144.5°C (Aldrich
145Ϫ147°C). Compound 10 was prepared according to a pro-
cedure described in the literature to give white crystals in 50%
yield; m.p. 97Ϫ97.5°C (CH2Cl2/petroleum ether, 1:1; ref.[21] 96°C
from petroleum ether). Ϫ 1H NMR (CDCl3): δ ϭ 4.45 (s, 6 H,
CH2Br), 7.35 (s, 3 H, arom. H). Ϫ 13C NMR (CDCl3): δ ϭ 32.27,
129.68, 139.20. Ϫ MS (70 eV, EI); m/z (%): 360 (2), 358 (8), 356 (8),
354 (2), 279 (54), 277 (100), 275 (48), 198 (34), 196 (33), 115 (37).
a
69% purity in HPLC and a higher melting point; m.p.
1
191Ϫ194°C. Ϫ H NMR (CDCl3): δ ϭ 4.58 (s, CH2), 7.40Ϫ7.83
(m, arom. H).
1,1,1-Tris[4-(bromomethyl)phenyl]ethane (16):[23] A mixture of
15[23] (10 g, 33.28 mmol), NBS (19.5 g, 109.56 mmol), and AIBN
(0.016 g) in dry CCl4 (125 ml) was allowed to reflux under UV
light (200 W). When the reaction mixture began to boil strongly,
refluxing and light were maintained for 90 min. The hot solution
was filtered, and the isolated white precipitate was washed with hot
CCl4 (50 ml). The filtrate was concentrated in vacuo to give a white
residue which was dissolved in CHCl3 (200 ml). This solution was
washed with a saturated aqueous solution of NaHCO3 (3 ϫ 50
ml), dried (MgSO4), and concentrated. The resulting powder was
purified by fractional recrystallisations from EtOAc to give 5.37 g
(30%) of white crystalline powder; m.p. 192Ϫ195°C (ref.[23]
38Ϫ42% yield with m.p. 185Ϫ191°C; for an analytical sample: m.p.
199Ϫ200°C). HPLC analysis showed a 72% purity. Ϫ 1H NMR
(CDCl3): δ ϭ 2.13 (s, 3 H, CH3), 4.48 (s, 6 H, CH2Br), 7.05 (d, 6
H, arom. H, J ϭ 8.4 Hz), 7.29 (d, 6 H, arom. H, J ϭ 8.4 Hz). Ϫ
13C NMR (CDCl3): δ ϭ 30.43, 33.24, 52.31, 128.80, 129.13, 135.77,
148.83. Ϫ MS (70 eV, EI); m/z (%): 537 [Mϩ•] (20), 523 (49), 521
(64), 459 (40), 457 (100), 455 (48), 442 (11), 361 (32), 282 (68), 191
(47), 141 (67)
1,3,5-Tris[4-(bromomethyl)phenyl]benzene (13):[22] A mixture of
11[22] (5 g, 14.35 mmol), NBS (8.05 g, 45.23 mmol), and AIBN (ca.
0.010 g) in dry CCl4 (57 ml) was refluxed for 3 h. The hot solution
was filtered, and the filter was washed with hot CCl4 (50 ml). The
filtrate was concentrated in vacuo to give a white residue which
was dissolved in CHCl3 (120 ml). This solution was washed with a
saturated aqueous solution of NaHCO3 (3 ϫ 50 ml), dried
(MgSO4), and concentrated. Recrystallisation from toluene/acetic
acid (20 ml/15 ml) afforded, after 2 h at room temp., 4.87 g of white
needles. HPLC analysis (reverse-phase Nucleosil 5 C18; CH3CN/
H2O, 9:1; 274 nm) exhibited a 68% purity. A sample of 1 g was
purified by column chromatography on silica gel (d ϭ 5 cm, h ϭ
44 cm) using CH2Cl2/petroleum ether (1:2). As the crude product
was not soluble in this solvent system, it was first adsorbed on
silica gel and then deposited on the chromatography column: 1 g
of product and 7 g of silica gel were mixed in 50 ml of CH2Cl2.
This purification resulted in 0.61 g of a white powder which was
recrystallised from CH2Cl2/petroleum ether to afford 0.56 g of
white needles. Repeating these purification steps with the remaining
3.87 g of crude product, we finally obtained 2.73 g (32%) of white
needles sufficienly pure for further reactions; m.p. 193Ϫ194°C.
HPLC analysis showed a 90% purity of the final product. It was
possible to increase the purity to 93% after another column chro-
matography on silica gel as described above; m.p. 200Ϫ200.5°C
(ref.[22] 192Ϫ194°C). Ϫ 1H NMR (CDCl3): δ ϭ 4.56 (s, 6 H,
CH2Br), 7.50 (d, 6 H, arom. H, J ϭ 8.2 Hz), 7.65 (d, 6 H, arom.
H, J ϭ 8.2 Hz), 7.74 (s, 3 H, arom. H). Ϫ 13C NMR (CDCl3): δ ϭ
33.17, 125.30, 127.74, 129.63, 137.29, 141.05, 141.79. Ϫ MS (70 eV,
EI); m/z (%): 588 (4), 587 (4), 586 (11), 585 (7), 584 (12), 583 (5),
582 (4), 507 (44), 505 (89), 503 (54), 426 (19), 424 (20), 345 (67),
213 (37), 212 (41), 173 (100).
General Procedure for the Preparation of Tris(tetrathiafulvalene)
5: To a solution of a mixture cis/trans isomers of 3 (0.44 g, 0.94
mmol) in anhydrous degassed DMF (40 ml) under N2 was added
dropwise a solution of CsOH·H2O (0.168 g, 1 mmol) in dry and
degassed MeOH (10 ml) over a period of 2 h at room temp. Then
a solution of the tris(bromomethyl) compound (0.30 mmol) in an-
hydrous degassed DMF (30 ml) was added dropwise during 5 h at
room temp. using a perfusor pump. After an additional 9 h of stir-
ring at room temp., the reaction mixture was concentrated in vacuo.
The resulting residue was dissolved in CH2Cl2 (150 ml) and the
solution was washed with water (3 ϫ 50 ml), dried (MgSO4), and
concentrated to give a brown residue which was subjected to col-
umn chromatography on silica gel using CH2Cl2 as eluent. Several
by-products were isolated and identified. The desired products were
obtained as oils or as foamy solids.
1,3,5-Tris({[3Ј(4Ј)-cyanoethylthio-4,4Ј(3Ј)-bis(methylthio)-
tetrathiafulvalen-3-yl]thio}methyl)benzene (5a): Starting from 10
(0.107 g, 0.30 mmol), 0.11Ϫ0.12 g (27Ϫ30% yield) of an orange
foam was obtained. M.p. 53Ϫ55°C (CH2Cl2). Rf ϭ 0.35 (CH2Cl2).
1,3,5-Tris[3-(bromomethyl)phenyl]benzene (14):[22] A mixture of
12[22] (5 g, 14.35 mmol), NBS (7.67 g, 43.09 mmol), and AIBN (ca.
0.010 g) in dry CCl4 (250 ml) was refluxed for 3 h. The hot solution
was filtered, and the filter was washed with hot CCl4 (50 ml). The
filtrate was left overnight at room temp. during which time a white
product precipitated. After filtration with suction, the product was
dissolved in CH2Cl2 (150 ml) and the solution was washed with a
saturated aqueous solution of NaHCO3 (2 ϫ 50 ml), dried
(MgSO4), and concentrated in vacuo. The resulting residue was
recrystallised from CCl4 (180 ml) to give 3.43 g of white crystals
after crystallisation overnight. The mother liquor was concentrated
and the residue recrystallised twice affording an additional crop of
0.36 g of white crystals with the same purity (TLC; CH2Cl2/petro-
1
Ϫ H NMR (CDCl3): δ ϭ 2.32 (s, 9 H, SCH3), 2.47 and 2.48 (2 s,
9 H, SCH3), 2.71 and 2.72 (2 t, 6 H, SCH2, J ϭ 7.2 Hz), 3.03 and
3.04 (2 t, 6 H, CH2CN, J ϭ 7.2 Hz), 3.97 (s, 6 H, SCH2Ar), 7.18
(s, 3 H, arom. H). Ϫ 13C NMR (CDCl3): δ ϭ 18.81, 19.11, 19.19,
19.29, 31.27, 40.21, 109.84, 111.99, 117.60, 120.12, 123.48, 123.75,
129.04, 132.92, 133.18, 135.05, 135.11, 137.42. Ϫ PDMS; m/z:
1355.4 [Mϩ•]; calcd. for C42H39N3S24 1355.23. Ϫ IR (KBr): ν ϭ
˜
2918, 2250 (CN), 1632, 1421, 889, 772 cmϪ1. Ϫ C42H39N3S24: calcd
C 37.22, H 2.90, N 3.10, S 56.78; found C 37.45, H 3.04, N 3.20,
S 56.45.
Tetrathiafulvalene Pentamer 17a: 0.06 g (18% yield) of an orange
1
foam. M.p. 71Ϫ73°C (CH2Cl2). Rf ϭ 0.21 (CH2Cl2). Ϫ H NMR
Eur. J. Org. Chem. 1998, 1743Ϫ1757
1753
EurJOC 172/98

P. Blanchard, N. Svenstrup, J. Rault-Berthelot, A. Riou, J. Becher
FULL PAPER
(CDCl3): δ ϭ 2.32 (br. s, 18 H, SCH3), 2.47 and 2.48 (2 s, 12 H, isotope pattern calcd. for C56H51N3S24 1535. Ϫ IR (KBr): ν ϭ
˜
SCH3), 2.71 and 2.72 (2 t, 8 H, SCH2, J ϭ 7.1 Hz), 3.02 and 3.03 2919, 2250 (CN), 1631, 1508, 1418, 1018, 890, 773 cmϪ1
( 2t, 8 H, CH2CN, J ϭ 7.1 Hz), 3.97 (br. s, 12 H, SCH2Ar), 7.18
.
General Procedure for the Preparation of Tetrathiafulvalene Mac-
robicycles 1: Under N2, a solution of CsOH·H2O (3.6 equiv.) in
anhydrous degassed MeOH (5 ml) was added to a solution of a
cis/trans mixture of tris(tetrathiafulvalene) (1 equiv.) in anhydrous
degassed DMF (25 ml) over a period of 10 min at room temp. This
solution and a solution of the tribromide (1 equiv.) in anhydrous
degassed DMF (30 ml) were added simultaneously, during 10 h at
room temp. with stirring, to 50 ml of anhydrous degassed DMF
under high-dilution conditions using a perfusor pump. Stirring was
continued for additional 5 h and the reaction mixture was concen-
trated in vacuo. The residue was then dissolved in CH2Cl2 (200 ml).
The solution was washed with water (3 ϫ 50 ml), dried (MgSO4),
and concentrated in vacuo. The resulting residue was firstly puri-
fied by filtration through a short column of silica gel using CH2Cl2
as eluent and then by column chromatography on silica gel using
CH2Cl2/petroleum ether (1:1) as eluent to afford the tetrathiafulva-
lene macrobicycles as a mixture of four cis/trans isomers.
(br. s, 6 H, arom. H). Ϫ MS (FAB); m/z: 2243.3 [Mϩ•]; isotope
pattern calcd. for C70H64N4S40 2243. Ϫ IR (KBr): ν ϭ 2918, 2250
˜
(CN), 1631, 1421, 888, 772 cmϪ1. Ϫ C70H64N4S40: calcd. C 37.47,
H 2.88, N 2.50, S 57.16; found C 37.07, H 2.71, N 2.50, S 54.86.
1,3,5-Tris[4-({[3Ј(4Ј)-cyanoethylthio-4,4Ј(3Ј)-bis(methylthio)-
tetrathiafulvalen-3-yl]thio}methyl)phenyl]benzene (5b): Starting
from 13 (92.5% pure; 0.190 g, 0.30 mmol), 0.16Ϫ0.17 g (34Ϫ36%
yield) of yellow foam was obtained. M.p. 81Ϫ83°C (CH2Cl2). Rf ϭ
1
0.47 (CH2Cl2). Ϫ H NMR (CDCl3): δ ϭ 2.27 and 2.28 (2 s, 9 H,
SCH3), 2.44 and 2.45 (2 s, 9 H, SCH3), 2.67 and 2.68 (2 t, 6 H,
SCH2, J ϭ 7.2 Hz), 2.99 and 3.00 (2 t, 6 H, CH2CN, J ϭ 7.2 Hz),
4.07 (s, 6 H, SCH2Ar), 7.43 (d, 6 H, arom. H, J ϭ 8.1 Hz), 7.64
(d, 6 H, arom. H, J ϭ 8.1 Hz), 7.74 (s, 3 H, arom. H). Ϫ 13C NMR
(CDCl3): δ ϭ 18.75, 18.77, 19.07, 19.17, 29.69, 31.21, 40.41, 109.16,
112.30, 117.53, 120.02, 120.20, 124.23, 124.55, 124.97, 127.51,
129.67, 132.68, 132.95, 135.03, 135.18, 136.27, 140.24, 141.93. Ϫ
MS (FAB); m/z: 1583 [Mϩ•]; isotope pattern calcd. for C60H51N3S24
Macrobicycle 1a: Starting from 5a (0.31 g, 0.23 mmol),
CsOH·H2O (0.139 g, 0.83 mmol), and 10 (0.082 g, 0.23 mmol),
0.13 g (43% yield) of 1a was obtained as yellow powder which
slowly decomposed between 155Ϫ160°C without melting. Ϫ 1H
NMR (CDCl3): δ ϭ 2.38Ϫ2.49 (m, 18 H, SCH3), 3.83Ϫ4.09 (m,
12 H, SCH2Ar), 6.99-7.23 (m, 6 H, arom. H). Ϫ 13C NMR
(CDCl3): δ ϭ 19.08, 19.15, 19.26, 19.33, 19.38, 39.74, 40.02, 40.08,
40.16, 40.30, 112.28, 112.39, 123.35, 129.22, 129.27, 129.48, 129.57,
136.79, 136.91, 137.39, 137.46, 138.04. Ϫ PDMS; m/z: 1309.1;
calcd. for C42H36S24 1310.19. Ϫ IR (KBr): ν˜ ϭ 2917, 1603, 1455,
1425, 1227, 887, 772, 708 cmϪ1. Ϫ C42H36S24: calcd. C 38.54, H
2.77, S 58.68; found C 38.26, H 2.91, S 58.19.
1582.7. Ϫ IR (KBr): ν ϭ 2917, 2249 (CN), 1596, 1510, 1418, 1394,
˜
888, 830, 773 cmϪ1
.
Tetrathiafulvalene Pentamer 17b: 0.045Ϫ0.065 g (11Ϫ16% yield)
of orange foam. M.p. 104Ϫ107°C (CH2Cl2). Rf ϭ 0.33 (CH2Cl2).
1
Ϫ H NMR (CDCl3): δ ϭ 2.23 (s, 6 H, SCH3), 2.26Ϫ2.28 (m, 12
H, SCH3), 2.43 and 2.45 (2 s, 12 H, SCH3), 2.665 and 2.670 (2 t,
8 H, SCH2, J ϭ 7.1 Hz), 2.99 and 3.00 (2 t, 8 H, CH2CN, J ϭ 7.1
Hz), 4.06 (s, 12 H, SCH2Ar), 7.40 (d, 12 H, arom. H, J ϭ 8 Hz),
7.58Ϫ7.65 (m, 12 H, arom. H), 7.69 and 7.72 (2 s, 6 H, arom.
H). Ϫ MS (FAB); m/z: 2700.4 [Mϩ•]; isotope pattern calcd. for
C106H88N4S40 2700. Ϫ IR (KBr): ν ϭ 2920, 2853, 2249 (CN), 1598,
˜
1510, 1419, 889, 828, 773 cmϪ1. Ϫ C106H88N4S40: calcd. C 47.15,
Macrobicycle 1b: Starting from 5b (0.17 g, 0.11 mmol),
CsOH·H2O (0.065 g, 0.39 mmol) and 13 (90% pure; 0.070 g, 0.11
mmol), 0.074 g (39% yield) of 1b was obtained as yellow powder
which slowly decomposed above 185°C without melting. Ϫ 1H
NMR (CDCl3): δ ϭ 2.01Ϫ2.26 (m, 18 H, SCH3), 3.94Ϫ4.08 (m,
12 H, SCH2Ar), 7.06Ϫ7.63 (m, 30 H, arom. H). Ϫ PDMS; m/z:
1767.1 [Mϩ•]; calcd. for C78H60S24 1766.77. Ϫ IR (KBr): ν˜ ϭ 2920,
H 3.28, N 2.07, S 47.49; found C 47.51, H 3.14, N 2.16, S 46.05.
1,3,5-Tris[3-({[3Ј(4Ј)-cyanoethylthio-4,4Ј(3Ј)-bis(methylthio)-
tetrathiafulvalen-3-yl]thio}methyl)phenyl]benzene (5c): Starting
from 14 (69% pure; 0.255 g, 0.30 mmol), 0.175 g (37% yield) of
yellow foam was obtained. M.p. 75Ϫ79°C (CH2Cl2). Rf ϭ 0.41
(CH2Cl2). Ϫ 1H NMR (CDCl3): δ ϭ 2.27 (s, 9 H, SCH3), 2.42 and
2.43 (2 s, 9 H, SCH3), 2.65 and 2.66 (2 t, 6 H, SCH2, J ϭ 7.1 and
7.2 Hz), 2.97 and 2.99 (2 t, 6 H, CH2CN, J ϭ 7.1 and 7.2 Hz), 4.12
(s, 6 H, SCH2Ar), 7.35 (d, 3 H, arom. H, J ϭ 7.6 Hz), 7.45 (dd, 3
H, arom. H, J ϭ 7.6 and 7.6 Hz), 7.62 (d, 3 H, arom. H, J ϭ 7.6
Hz), 7.67 (s, 3 H, arom. H), 7.79 (s, 3 H, arom. H). Ϫ 13C NMR
(CDCl3): δ ϭ 18.73, 18.75, 19.09, 19.21, 31.22, 40.68, 109.36,
112.02, 117.55, 123.99, 124.30, 125.24, 126.69, 128.13, 128.37,
129.23, 132.79, 133.08, 135.02, 135.16, 137.39, 141.34, 142.02. Ϫ
MS (FAB); m/z: 1582; isotope pattern calcd. for C60H51N3S24
1630, 1511, 1421, 831, 775 cmϪ1
.
Macrobicycle 1c: Starting from 5c (0.14 g, 0.09 mmol),
CsOH·H2O (0.053 g, 0.32 mmol) and 14 (69% pure; 0.075 g, 0.09
mmol), 0.058 g (37% yield) of 1c was obtained as yellow powder;
1
m.p. 175°C (dec.). Ϫ H NMR (CDCl3): δ ϭ 2.14Ϫ2.25 (m, 18 H,
SCH3), 3.95Ϫ4.00 (m, 12 H, SCH2Ar), 7.23Ϫ7.42 (m, 12 H, arom.
H), 7.53Ϫ7.58 (m, 12 H, arom. H), 7.70Ϫ7.75 (m, 6 H, arom. H).
Ϫ PDMS; m/z: 1766.0 [Mϩ•]; calcd. for C78H60S24 1766.77. Ϫ IR
˜
(KBr): ν ϭ 2918, 1595, 1582, 1489, 1421, 889, 876, 793, 772, 705
cmϪ1. Ϫ C78H60S24: calcd. C 53.07, H 3.43, S 43.50; found C 52.68,
1582.7. Ϫ IR (KBr): ν ϭ 2918, 2250 (CN), 1630, 1596, 1420, 889,
˜
772, 706, 622 cmϪ1
.
H 3.48, S 42.84.
1,1,1-Tris[4-({[3Ј(4Ј)-cyanoethylthio-4,4Ј(3Ј)-bis(methylthio)-
tetrathiafulvalen-3-yl]thio}methyl)phenyl]ethane (5d): Starting
Macrobicycle 1d: Starting from 5d (0.19 g, 0.12 mmol),
CsOH·H2O (0.075 g, 0.45 mmol) 16 (72% pure; 0.093 g, 0.12
from 16 (72% pure; 0.225 g, 0.30 mmol), 0.167 g (36% yield) of mmol), 0.08 g (40% yield) of 1d was obtained as yellow powder;
orange foam was obtained. M.p. 79Ϫ82°C (CH2Cl2). Rf ϭ 0.38 m.p. 200Ϫ205°C (dec.). Statistically, this is a mixture of twelve iso-
1
(CH2Cl2). Ϫ H NMR (CDCl3): δ ϭ 2.11 (s, 3 H, CH3), 2.26 and mers considering the cis/trans isomers and the out-out, in-out, and
2.27 (2 s, 9 H, SCH3), 2.45 and 2.46 (2 s, 9 H, SCH3), 2.68 and in-in isomers where the methyl groups of the benzylic positions can
1
2.69 (2 t, 6 H, SCH2, J ϭ 7.1 and 7.2 Hz), 3.005 and 3.015 (2 t, 6
be outside or inside the cavity of the cage. Ϫ H NMR (CDCl3):
H, CH2CN, J ϭ 7.2 and 7.1 Hz), 3.98 (s, 6 H, SCH2Ar), 7.02 (d, δ ϭ 1.25Ϫ2.46 (m, 24 H, CH3 and SCH3), 3.75Ϫ3.96 (m, 12 H,
6 H, arom. H, J ϭ 8.3 Hz), 7.21 (d, 6 H, arom. H, J ϭ 8.3 Hz). Ϫ
SCH2Ar), 6.61Ϫ7.23 (m, 24 H, arom. H). Ϫ PDMS; m/z: 1670.3
13C NMR (CDCl3): δ ϭ 18.77, 19.13, 19.29, 29.69, 31.26, 40.20, [Mϩ•]; calcd. for C70H60S24 1670.69. Ϫ IR (KBr): ν ϭ 2918, 1630,
˜
52.06, 109.00, 117.54, 120.16, 124.35, 128.73, 128.92, 132.72, 1508, 1419, 1018, 879, 773 cmϪ1. Ϫ C70H60S24: calcd. C 50.37, H
132.94, 134.52, 135.13, 148.19, 124.63. Ϫ MS (FAB); m/z: 1536;
3.63, S 46.01; found C 50.59, H 3.52, S 45.17.
Eur. J. Org. Chem. 1998, 1743Ϫ1757
1754
EurJOC 172/98

Macrobicyclic Tetrathiafulvalene-Bridged Cage Molecules
FULL PAPER
General Procedure for the Preparation of Tris(tetrathiafulvalene) H, SCH2Ar), 7.04 (d, 6 H, arom. H, J ϭ 8.3 Hz), 7.23 (d, 6 H,
6: To a solution of 4 (0.42 g, 0.90 mmol) in anhydrous degassed
DMF (35 ml) under N2, was added dropwise a solution of
CsOH·H2O (0.159 g, 0.95 mmol) in anhydrous degassed MeOH
arom. H, J ϭ 8.3 Hz). Ϫ 13C NMR (CDCl3): δ ϭ 18.54, 19.20,
19.22, 31.31, 40.32, 52.08, 109.18, 112.22, 117.57, 125.92, 127.37,
127.65, 128.79, 128.97, 131.98, 134.25, 148.34. Ϫ MS (FAB); m/z:
(10 ml) over a period of 45 min at room temp. After an additional 1534.5 [Mϩ•]; isotope pattern calcd. for C56H51N3S24 1535. Ϫ IR
15 min of stirring, a solution of the tribromide (0.30 mmol) in (KBr): ν ϭ 2917, 2249 (CN), 1630, 1507, 1416, 888, 772, 730 cmϪ1
.
˜
anhydrous degassed DMF (25 ml) was added dropwise during 1 h Ϫ C56H51N3S24: calcd. C 43.80, H 3.35, N 2.74, S 50.11; found C
at room temp. The reaction mixture was stirred overnight at room 43.42, H 2.57, N 2.83, S 49.64.
temp. and then concentrated in vacuo to afford a residue which
was dissolved in CH2Cl2 (200 ml). This solution was washed with
General Procedure for the Preparation of Tetrathiafulvalene Mac-
robicycles 2: Under N2, a solution of CsOH·H2O (3.15 equiv.) in
water (3 ϫ 50 ml), dried (MgSO4), and concentrated in vacuo. The
anhydrous degassed MeOH (10 ml) was added to a solution of 6
resulting oil was subjected to column chromatography on silica gel
(1 equiv.) in anhydrous degassed DMF (40 ml) over a period of 10
using CH2Cl2 as eluent to give the desired tris(tetrathiafulvalene)
min at room temp. This solution and a solution of the tribromide
as an orange oil or foam.
(1 equiv.) in anhydrous degassed DMF (50 ml) were added simul-
taneously, during 17 h at room temp. with stirring, to 50 ml of
anhydrous degassed DMF under high-dilution conditions using a
perfusor pump. Stirring was continued for an additional 1 h, and
the reaction mixture was concentrated in vacuo. The residue was
dissolved in CHCl3 (200 ml). The solution was washed with water
(3 ϫ 50 ml), dried (MgSO4), and concentrated in vacuo affording
a residue which was first purified by filtration through a short col-
umn of silica gel using CH2Cl2 as eluent and then by column chro-
matography on silica gel using a mixture of CH2Cl2/petroleum
ether (1:1) as eluent.
1,3,5-Tris({[4-cyanoethylthio-3Ј,4Ј-bis(methylthio)tetrathia-
fulvalen-3-yl]thio}methyl)benzene (6a): Using 10 (0.107 g, 0.3
mmol), we obtained 0.300Ϫ0.312 g (74Ϫ77% yield) of orange oil
which becomes solid upon standing. M.p. 58Ϫ62°C (CH2Cl2). Ϫ
1H NMR (CDCl3): δ ϭ 2.44 (s, 9 H, SCH3), 2.45 (s, 9 H, SCH3),
2.48 (t, 6 H, SCH2, J ϭ 7.2 Hz), 2.90 (t, 6 H, CH2CN, J ϭ 7.2
Hz), 4.02 (s, 6 H, SCH2Ar), 7.22 (s, 3 H, arom. H). Ϫ 13C NMR
(CDCl3): δ ϭ 18.54, 19.19, 19.26, 31.37, 40.25, 108.70, 113.51,
117.69, 126.56, 127.30, 127.53, 129.18, 131.25, 137.61. Ϫ MS
(FAB); m/z: 1354.6 [Mϩ•]; isotope pattern calcd. for C42H39N3S24
1355. Ϫ IR (KBr): ν ϭ 2916, 2249 (CN), 1631, 1601, 1417, 887,
˜
Tetrathiafulvalene Macrobicycle 2a: Starting from 6a (0.29 g, 0.21
mmol), CsOH·H2O (0.114 g, 0.68 mmol) and 10 (0.077 g, 0.21
mmol), 0.19 g (68% yield) of 2a was obtained as orange powder;
772, 728, 709 cmϪ1. Ϫ C42H39N3S24: calcd. C 37.22, H 2.90, N
3.10, S 56.78; found C 36.84, H 2.69, N 2.97, S 55.92.
1
m.p. 240Ϫ243°C (dec.). Ϫ H NMR (CDCl3): δ ϭ 2.44 (s, 18 H,
1,3,5-Tris[4-({[4-cyanoethylthio-3Ј,4Ј-bis(methylthio)tetrathia-
fulvalen-3-yl]thio}methyl)phenyl]benzene (6b): Using 13 (93%
pure; 0.190 g, 0.30 mmol), we obtained 0.355 g (75% yield) of or-
ange foam. M.p. 87Ϫ89°C (CH2Cl2). Ϫ 1H NMR (CDCl3): δ ϭ
2.34 (t, 6 H, SCH2, J ϭ 7.3 Hz), 2.42 (s, 18 H, SCH3), 2.84 (t, 6
H, CH2CN, J ϭ 7.3 Hz), 4.09 (s, 6 H, SCH2Ar), 7.42 (d, 6 H,
arom. H, J ϭ 8.2 Hz), 7.67 (d, 6 H, arom. H, J ϭ 8.2 Hz), 7.77 (s,
3 H, arom. H). Ϫ 13C NMR (CDCl3): δ ϭ 18.21, 19.14, 31.29,
40.41, 108.86, 112.58, 117.58, 125.07, 126.05, 127.29, 127.62,
129.59, 131.88, 136.13, 140.38, 141.64. Ϫ MS (FAB); m/z: 1582.7
[Mϩ•]; isotope pattern calcd. for C60H51N3S24 1582.7. Ϫ IR (KBr):
SCH3), 3.84 (s, 12 H, SCH2Ar), 7.06 (s, 6 H, arom. H). Ϫ 13C
NMR 125.7 MHz (CDCl3): δ ϭ 19.24, 40.31, 110.05, 111.88,
127.64, 128.17, 128.95, 138.29. Ϫ MS (FAB); m/z: 1309; isotope
pattern calcd. for C42H36S24 1309.6. Ϫ IR (KBr): ν˜ ϭ 2916, 1603,
1419, 969, 886, 772, 706 cmϪ1. Ϫ C42H36S24: calcd. C 38.54, H
2.77, S 58.68; found C 38.56, H 2.59, S 56.76.
Tetrathiafulvalene Macrobicycle 2b: Starting from 6b (0.34 g, 0.21
mmol), CsOH·H2O (0.114 g, 0.68 mmol) and 13 (90% pure; 0.14
g, 0.21 mmol), 0.12 g (32% yield) of 2b was obtained as yellow-
orange powder; m.p. 202Ϫ207°C (dec.). Ϫ 1H NMR (CDCl3): δ ϭ
2.42 (s, 18 H, SCH3), 3.99 (s, 12 H, SCH2Ar), 7.16 (d, 12 H, arom.
H, J ϭ 8.1 Hz), 7.28 (d, 12 H, arom. H, J ϭ 8.1 Hz), 7.33 (s, 6 H,
arom. H). Ϫ 13C NMR (CDCl3): δ ϭ 19.14, 39.78, 110.17, 110.73,
124.27, 127.18, 127.24, 127.52, 128.92, 136.77, 139.72, 141.12. Ϫ
PDMS; m/z: 1765.8; calcd. for C78H60S24 1766.77. Ϫ IR (KBr): ν˜ ϭ
2918, 1631, 1599, 1511, 1419, 888, 833, 775 cmϪ1. Ϫ C78H60S24:
calcd. C 53.07, H 3.43, S 43.50; found C 53.38, H 3.51, S 41.97.
ν ϭ 2917, 2250 (CN), 1631, 1597, 1510, 1418, 888, 831, 773 cmϪ1
.
˜
Ϫ C60H51N3S24: calcd. C 45.51, H 3.25, N 2.65, S 48.59; found C
45.73, H 3.35, N 3.31, S 48.67.
1,3,5-Tris[3-({[4-cyanoethylthio-3Ј,4Ј-bis(methylthio)tetrathia-
fulvalen-3-yl]thio}methyl)phenyl]benzene (6c): Using 14 (69% pure;
0.255 g, 0.30 mmol), we obtained 0.347 g (73% yield) of orange
foam. M.p. 77Ϫ80°C (CH2Cl2). Ϫ 1H NMR (CDCl3): δ ϭ 2.36 (t,
6 H, SCH2, J ϭ 7.4 Hz), 2.38 (s, 9 H, SCH3), 2.41 (s, 9 H, SCH3),
2.82 (t, 6 H, CH2CN, J ϭ 7.4 Hz), 4.14 (s, 6 H, SCH2Ar), 7.34 (d,
3 H, arom. H, J ϭ 7.6 Hz), 7.45 (dd, 3 H, arom. H, J ϭ 7.6 and
7.6 Hz), 7.66 (d, 3 H, arom. H, J ϭ 7.6 Hz), 7.69 (s, 3 H, arom.
H), 7.82 (s, 3 H, arom. H). Ϫ 13C NMR (CDCl3): δ ϭ 18.24, 19.16,
31.26, 40.71, 108.76, 112.54, 117.60, 125.22, 126.28, 126.88, 127.28,
127.63, 128.08, 128.28, 129.33, 131.87, 137.38, 141.35, 141.86. Ϫ
MS (FAB); m/z: 1582.6 [Mϩ•]; isotope pattern calcd. for
Tetrathiafulvalene Macrobicycle 2c: Starting from 6c (0.30 g, 0.19
mmol), CsOH·H2O (0.100 g, 0.60 mmol) and 14 (69% pure; 0.161
g, 0.19 mmol), 0.074 g (22% yield) of 2c was obtained as yellow-
1
orange powder; m.p. 174Ϫ178°C (dec.). Ϫ H NMR (CDCl3): δ ϭ
2.24 (s, 18 H, SCH3), 3.70 (s, 12 H, SCH2Ar), 7.15Ϫ7.43 (m, 30 H,
arom. H). Ϫ 13C NMR (CDCl3): δ ϭ 19.08, 41.42, 109.31, 110.37,
124.61, 126.49, 127.48, 127.91, 128.04, 129.18, 131.34, 137.87,
140.74, 141.26. Ϫ MS (FAB); m/z: 1765.6 [Mϩ•]; isotope pattern
C60H51N3S24 1582.7. Ϫ IR (KBr): ν ϭ 2919, 2250 (CN), 1595,
˜
˜
calcd. for C78H60S24 1766. Ϫ IR (KBr): ν ϭ 2918, 1596, 1582, 1420,
1582, 1419, 890, 772, 707 cmϪ1. Ϫ C60H51N3S24: calcd. C 45.51, H
889, 874, 794, 772, 705 cmϪ1. Ϫ C78H60S24: calcd. C 53.07, H 3.43,
3.25, N 2.65, S 48.59; found C 45.56, H 3.09, N 2.56, S 45.61.
S 43.50; found C 52.43, H 3.23, S 43.39.
1,1,1-Tris[4-({[4-cyanoethylthio-3Ј,4Ј-bis(methylthio)tetrathia-
fulvalen-3-yl]thio}methyl)phenyl]ethane (6d): Using 16 (72% pure;
Tetrathiafulvalene Macrobicycle 2d: Starting from 6d (0.29 g, 0.19
mmol), CsOH·H2O (0.100 g, 0.60 mmol) and 16 (72% pure; 0.14
0.225 g, 0.30 mmol), we obtained 0.385 g (84% yield) of orange g, 0.19 mmol), 0.094 g (30% yield) of 2d was obtained as yellow
foam. M.p. 74Ϫ77°C (CH2Cl2). Ϫ 1H NMR (CDCl3): δ ϭ 2.14 (s, powder; m.p. 190Ϫ193°C (dec.). The 1H-NMR spectra showed a
3 H, CH3), 2.41 (s, 9 H, SCH3), 2.43 (s, 9 H, SCH3), 2.48 (t, 6 H, ca. 2:1 mixture of two different macrobicycles, one symmetrical
SCH2, J ϭ 7.2 Hz), 2.89 (t, 6 H, CH2CN, J ϭ 7.2 Hz), 4.03 (s, 6 cage with two methyl groups outside the cavity (“out-out” or “o-
Eur. J. Org. Chem. 1998, 1743Ϫ1757
1755
EurJOC 172/98

P. Blanchard, N. Svenstrup, J. Rault-Berthelot, A. Riou, J. Becher
FULL PAPER
[7] [7a]
K. Matsuo, K. Takimiya, Y. Aso, T. Otsubo, F. Ogura,
o”) and one unsymmetrical cage with one methyl group inside and
one outside the cavity (“in-out” or “i-o”). Ϫ 1H NMR (CDCl3):
δ ϭ 1.17 (br. s, in CH3 of “i-o”), 2.06 (br. s, out-CH3 of “o-o”),
2.10 (br. s, out-CH3 of “i-o”), 2.40 (s, SCH3 of “i-o”), 2.42 (s, SCH3
of “o-o” and “i-o”), 3.85 (s, SCH2Ar of “o-o”), 3.94 (s, SCH2Ar
of “i-o”), 4.07 (s, SCH2Ar of “i-o”), 6.91 (d, arom. H of “o-o”,
J ϭ 8.3 Hz), 6.96Ϫ7.02 (m, arom. H of “i-o”), 7.06 (d, arom. H of
“o-o”, J ϭ 8.3 Hz), 7.15Ϫ7.22 (m, arom. H of “i-o”). Ϫ 13C NMR
(CDCl3): δ ϭ 19.18, 30.35, 30.52, 39.62, 40.50, 52.14, 52.39, 109.81,
110.23, 110.54, 110.61, 121.13, 126.94, 127.49, 127.90, 128.13,
128.54, 128.59, 128.94, 129.04, 120.20, 129.65, 133.10, 134.63,
134.90, 135.02, 147.76, 147.88, 148.08. Ϫ MS (FAB); m/z: 1669.5
[7b]
Chem. Lett. 1995, 523. Ϫ
K. Takimiya, Y. Shibata, K. Im-
amura, A. Kashibara, Y. Aso, T. Otsubo, Tetrahedron Lett.
[7c]
1995, 36, 5045. Ϫ
J. Tanabe, T. Kudo, M. Okamoto, Y. Ka-
wada, G. Ono, A. Izuoka, T. Sugawara, Chem. Lett. 1995, 579.
[7d]
Ϫ
K. Takimiya, K. Imamura, Y. Shibata, Y. Aso, F. Ogura,
T. Otsubo, J. Org. Chem. 1997, 62, 5567.
T. Jørgensen, T. K. Hansen, J. Becher, Chem. Soc. Rev. 1994,
23, 41.
[8]
[9] [9a]
T. Otsubo, F. Ogura, Bull. Chem. Soc. Jpn. 1985, 58, 1343.
[9b]
Ϫ
T. K. Hansen, T. Jørgensen, J. Becher, J. Chem. Soc.,
[9c]
Chem. Commun. 1992, 1550. Ϫ
T. K. Hansen, T. Jørgensen,
P. C. Stein, J. Becher, J. Org. Chem. 1992, 57, 6403. Ϫ [9d] T. K.
Hansen, T. Jørgensen, F. Jensen, P. H. Thygesen, K. Christi-
ansen, M. B. Hursthouse, M. E. Harman, M. A. Malik, B. Gir-
[Mϩ•]; isotope pattern calcd. for C70H60S24 1670. Ϫ IR (KBr): ν ϭ
may, A. E. Underhill, M. Begtrup, J. D. Kilburn, K. Belmore,
˜
[9e]
P. Roepstorff, J. Becher, J. Org. Chem. 1993, 58, 1359. Ϫ
P.
2919, 1631, 1508, 1418, 1018, 888, 773 cmϪ1. Ϫ C70H60S24: calcd.
D. Beer, J. P. Danks, D. Hesek, Polyhedron 1995, 14, 1327. Ϫ
[9f] R. Dieing, V. Morisson, A. J. Moore, L. M. Goldenberg, M.
R. Bryce, J-M. Raoul, M. C. Petty, J. Garin, M. Saviron, I. K.
Lednev, R. E. Hester, J. N. Moore, J. Chem. Soc., Perkin Trans.
2 1996, 1587.
C 50.37, H 3.63, S 46.01; found C 50.32, H 3.40, S 44.94.
One-Pot Procedure for the Preparation of Tetrathiafulvalene Mac-
robicycle 2a: To a solution of 4 (0.42 g, 0.90 mmol) in anhydrous
degassed DMF (35 ml) under N2, was added dropwise a solution
of CsOH·H2O (0.159 g, 0.95 mmol) in anhydrous degassed MeOH
(10 ml) over a period of 45 min at room temp. After an additional
15 min of stirring, a solution of 10 (0.107 g, 0.30 mmol) in anhy-
drous degassed DMF (25 ml) was added dropwise during 1 h at
room temp. The reaction mixture was stirred at room temp. for 2
h before addition of a solution of CsOH·H2O (0.159 g, 0.95 mmol)
in anhydrous degassed MeOH (10 ml). After 0.5 h of stirring, a
solution of 10 (0.107 g, 0.30 mmol) in anhydrous degassed DMF
(25 ml) was added over 30 min at room temp. After the work-up
procedure described above, 0.267 g (68% yield) of macrocycle 2a
was obtained.
[10] [10a]
T. Jørgensen, B. Girmay, T. K. Hansen, J. Becher, A. E.
Underhill, M. B. Hursthouse, M. E. Harman, J. D. Kilburn, J.
Chem. Soc., Perkin Trans. 1 1992, 2907. Ϫ
[10b]
M. Wagner, D.
Madsen, J. Markussen, S. Larsen, K. Schaumburg, K-H. Lu-
bert, J. Becher, R-M. Olk, J. Chem. Soc., Perkin Trans. 1
1996, 1995.
[11]
[12]
[13]
[14]
T. Jørgensen, J. Becher, J-C. Chambron, J-P. Sauvage, Tetra-
hedron Lett. 1994, 35, 4339.
R. Gasiorowsky, T. Jørgensen, J. Møller, T. K. Hansen, M. Pie-
traszkiewicz, J. Becher, Adv. Mater. 1992, 4, 568.
D. Philp, A. M. Z. Slawin, N. Spencer, J. F. Stoddart, D. J.
Williams, J. Chem. Soc., Chem. Commun. 1991, 1584.
´
Y. Le Bras, M. Salle, P. Leriche, C. Mingotaud, P. Richomme,
J. Møller, J. Mater. Chem. 1997, 12, 2393.
[15] [15a]
N. Svenstrup, K. M. Rasmussen, T. K. Hansen, J. Becher,
[15b]
Synthesis 1994, 809. Ϫ
J. Becher, J. Lau, P. Leriche, P.
Mørk, N. Svenstrup, J. Chem. Soc., Chem. Commun. 1994,
[15c]
[1]
2715. Ϫ
monsen, P. Mørk, G. J. Kristensen, J. Becher, Synthesis 1996,
K. B. Simonsen, N. Svenstrup, J. Lau, O. Si-
C. Seel, F. Vögtle, Angew. Chem. Int. Ed. Engl. 1992, 31, 528;
Angew. Chem. 1992, 104, 542.
[15d]
[2] [2a]
P. D. Beer, Adv. Inorg. Chem. 1992, 39, 79. Ϫ [2b] P. D. Beer,
407. Ϫ
J. Becher, Z-T. Li, P. Blanchard, N. Svenstrup, J.
Lau, M. Brøndsted Nielsen, P. Leriche, Pure Appl. Chem. 1997,
O. Kocian, R. J. Mortimer, P. Spencer, J. Chem. Soc., Chem.
Commun. 1992, 602. Ϫ [2c] P. D. Beer, Adv. Mater. 1994, 6, 607.
69, 465.
[16]
[17]
[2d]
J. Lau, P. Blanchard, A. Riou, M. Jubault, M. P. Cava, J. Be-
cher, J. Org. Chem. 1997, 62, 4936.
Ϫ
D. J. Cram, J. M. Cram, Container Molecules and Their
Guests, Monographs in Supramolecular Chemistry (Ed.: J. F.
Stoddart), 1994.
J. Lau, O. Simonsen J. Becher, Synthesis 1995, 521.
[3] [3a]
[3b]
[18] [18a]
G. Schukat, E. Fanghänel, Sulfur Rep. 1993, 14, 245. Ϫ
Z.-T. Li, P. C. Stein, N. Svenstrup, K. H. Lund, J. Becher,
[3c]
M. R. Bryce, Chem. Soc. Rev. 1991, 20, 355. Ϫ
J. Gar´ın,
J. M. Williams, J.
Angew. Chem. Int. Ed. Engl. 1995, 34, 2524; Angew. Chem. 1995,
[3d]
[18b]
Adv. Heterocycl. Chem. 1995, 62, 249. Ϫ
107, 2719. Ϫ
Z.-T. Li, P. C. Stein, J. Becher, D. Jensen, P.
[18c]
R. Ferraro, R. J. Thorn, K. D. Carlson, U. Geiser, H. H. Wang,
A. M. Kini, M.-H. Whangbo, Organic Superconductors, Pren-
tice Hall, New Jersey, 1992.
Mørk, N. Svenstrup, Chem. Eur. J. 1996, 2, 624. Ϫ
Z.-T.
[18d]
Li, J. Becher, J. Chem. Soc., Chem. Commun. 1996, 639. Ϫ
M.B. Nielsen, Z.-T. Li, J. Becher, J. Mater. Chem. 1997, 7, 1175.
[4]
[5]
[18e]
H. A. Staab, J. Ippen, C. Tao-pen, C. Krieger, B. Starker, An-
gew. Chem. Int. Ed. Engl. 1980, 19, 66; Angew. Chem. 1980,
92, 49.
Ϫ
Z.-T. Li, J. Becher, Synlett 1997, 557.
[19]
[20]
K. B. Simonsen, K. Zong, R. D. Rogers, M. P. Cava, J. Becher,
J. Org. Chem. 1997, 62, 679.
[5a] J. Ippen, C. Tao-pen, B. Starker, D. Schweitzer, H. A. Staab,
P. Blanchard, N. Svenstrup, J. Becher, J. Chem. Soc., Chem.
Commun. 1996, 615.
Angew. Chem. Int. Ed. Engl. 1980, 19, 67; Angew. Chem. 1980,
[5b]
[21]
[22]
92, 51. Ϫ
J. Röhrich, P. Wolf, V. Enkelmann, K. Müllen,
C. Bilger, R. Royer, P. Demerseman, Synthesis 1988, 902.
N. Sendhoff, W. Kiβener, F. Vögtle, S. Franken, H. Puff, Chem.
Ber. 1988, 121, 2179.
G. Hohner, F. Vögtle, Chem. Ber. 1977, 110, 3052.
L. Rossa, F. Vögtle, Top. Curr. Chem. 1983, 113, 1.
Angew. Chem. Int. Ed. Engl. 1988, 27, 1377; Angew. Chem. 1988,
[5c]
100, 1429. Ϫ
B. Girmay, J. D. Kilburn, A. E. Underhill, K.
S. Varma, M. B. Hursthouse, M. E. Harman, J. Becher, G. Bo-
[23]
[24]
[5d]
jesen, J. Chem. Soc., Chem. Commun. 1989, 1406. Ϫ
F. Ber-
tho-Thoraval, A. Robert, A. Souizi, K. Boubekeur, P. Batail, J.
[25] [25a]
C. Wang, M. R. Bryce, A. S. Batsanov, L. M. Goldenberg,
[5e]
Chem. Soc., Chem. Commun. 1991, 843. Ϫ
F. Vögtle, F.
[25b]
J. A. K. Howard, J. Mater. Chem. 1997, 7, 1189. Ϫ
M. R.
Alfter, M. Nieger, E. Steckhan, S. Mavili, Chem. Ber. 1991, 124,
Bryce, W. Devonport, A. J. Moore, Angew. Chem. Int. Ed. Engl.
1994, 33, 1761; Angew. Chem. 1994, 106, 1862.
A.J. Bard, Nature 1995, 374, 13.
897. Ϫ [5f] J. Röhrich, K. Müllen, J. Org. Chem. 1992, 57, 2374.
[5g]
Ϫ
M. Adam, V. Enkelmann, H.-J. Räder, J.Röhrich, K.
[26]
[27]
Müllen, Angew. Chem. Int. Ed. Engl. 1992, 31, 309; Angew.
Chem. 1992, 104, 331. Ϫ [5h] K. Boubekeur, C. Lenoir, P. Batail,
R. Carlier, A. Tallec, M. P. Le Paillard, D. Lorcy, A. Robert,
Angew. Chem. Int. Ed. Engl. 1994, 33, 1379; Angew. Chem. 1994,
104, 1448. Ϫ [5i] P. Hascoat, D. Lorcy, A. Robert, K. Boubekeur,
M. R. Bryce, W. Devonport, Adv. Dendritic Macromol. 1996,
3, 115.
[28]
´
C. Valerio, J-L. Fillault, J. Ruiz, J. Guittard, J-C. Blais, D. As-
truc, J. Am.Chem. Soc. 1997, 119, 2588.
[29]
R. W. Alder, S. P. East, Chem. Rev. 1996, 96, 2097.
P. Batail, R. Carlier, A. Tallec, J. Chem. Soc., Chem. Commun.
[30] [30a]
[5j]
J. Franke, F. Vögtle, Angew. Chem. Int. Ed. Engl. 1985, 24,
1995, 1229. Ϫ
P. Hascoat, D. Lorcy, A. Robert, R. Carlier,
[30b]
219; Angew. Chem. 1985, 96, 224. Ϫ
A. Wallon, J. Peter-
A. Tallec, K. Boubekeur, P. Batail, J. Org. Chem. 1997, 62,6086.
[6]
[6a]
Katalinic, U. Werner, W. M. Müller, F. Vögtle, Chem. Ber. 1990,
For articles see:
M. Adam, K. Müllen, Adv. Mater. 1994, 6,
[6b]
123, 375.
439. Ϫ
8, 203.
T. Otsubo, Y. Aso, K. Takimiya, Adv. Mater. 1996,
[31]
W. Grimme, H. T. Kämmerling, J. Lex, R. Gleiter, J. Heinze,
1756
Eur. J. Org. Chem. 1998, 1743Ϫ1757
EurJOC 172/98

Macrobicyclic Tetrathiafulvalene-Bridged Cage Molecules
FULL PAPER
R. Carlier, J. Simonet, Bull. Soc. Chim. Fr. 1988,
[35b]
M. Dietrich, Angew. Chem. Int. Ed. Engl. 1991, 30, 205; Angew.
Chem. 1991, 103, 215.
129. Ϫ
831.
[32]
[36]
[37]
[38]
M. Fourmigue´, I. Johannsen, K. Boubekeur, C. Nelson P. Ba-
J. B. Flanagan, S. Margel, A. J. Bard, F. C. Anson, J. Am. Chem.
Soc. 1978, 100, 4248.
tail, J. Am.Chem. Soc. 1993, 115, 3752.
[33]
M. Jørgensen, K. A. Lerstrup, K. Bechgaard, J. Org. Chem.
E. Cloutet, Y. Gnanou, J-L. Fillault, D. Astruc, J. Chem. Soc.,
Chem. Commun. 1996, 1565.
1991, 56, 5684.
[34] [34a] M. R. Bryce, G. J. Marshallsay, A. J. Moore, J. Org. Chem.
F. B. Kaufman, A. H. Schroeder, E. M. Engler, S. R. Kramer,
J. Q. Chambers, J. Am. Chem. Soc. 1980, 102, 483.
[98172]
[34b]
1992, 57, 4859. Ϫ
M. Iyoda, M. Fukuda, M. Yoshida, S.
Sasaki, Chem. Lett. 1994, 2369.
[35] [35a]
A. T. Hubbard, F. C. Anson, Elctroanal. Chem. 1970, 4,
Eur. J. Org. Chem. 1998, 1743Ϫ1757
1757
EurJOC 172/98