Electrochemical syntheses of cyclo[n]pyrrole

Article abstract of DOI:10.1002/chem.201000043

Cyclo[8]pyrrole was obtained efficiently when 3, 3', 4, 4'-tetraethylbipyrrole was subjected to bulk electrolysis, with yields spanning a range from close to 0% with tetra-n-butylammonium fluoride as the electrolyte, to almost 70% when tetra-n-butylammonium hydrogensulfate was used for this purpose. These observations are consistent with the conclusion that the reaction is controlled by anion-related fac-tors such as a specific templating effect. Note that cyclo[8]pyrrole was the only detectable macrocyclic product obtained from 3, 3', 4, 4'-tetraethylbipyrrole under the conditions of the electrochemical oxidation. When similar electrolyses were performed by using 3, 4-diethylpyrrole as the starting material, two products could be isolated, which were identified as being the cyclo[7]pyrrole and cyclo[8]pyrrole, respectively. Detailed analyses of the oxidized forms of cyclo[7]pyrrole and cyclo[8]pyrrole revealed that under the conditions of the electrolysis these latter species are not stable.

Full text of DOI:10.1002/chem.201000043

DOI: 10.1002/chem.201000043
Electrochemical Syntheses of ACTHNUTRGNEUNGCyclo[n]pyrrole
Mihai Buda,[a, c] Adriana Iordache,[a] Christophe Bucher,*[a] Jean-Claude Moutet,[a]
Guy Royal,[a] Eric Saint-Aman,[a] and Jonathan L. Sessler*[b]
Abstract: Cyclo[8]pyrrole was obtained
efficiently when 3,3’,4,4’-tetraethylbi-
pyrrole was subjected to bulk electroly-
sis, with yields spanning a range from
close to 0% with tetra-n-butylammoni-
um fluoride as the electrolyte, to
almost 70% when tetra-n-butylammo-
nium hydrogensulfate was used for this
purpose. These observations are consis-
tent with the conclusion that the reac-
tion is controlled by anion-related fac-
tors such as
a
specific templating
lar electrolyses were performed by
using 3,4-diethylpyrrole as the starting
material, two products could be isolat-
ed, which were identified as being the
cyclo[7]pyrrole and cyclo[8]pyrrole, re-
spectively. Detailed analyses of the oxi-
dized forms of cyclo[7]pyrrole and cy-
clo[8]pyrrole revealed that under the
conditions of the electrolysis these
latter species are not stable.
effect. Note that cyclo[8]pyrrole was
the only detectable macrocyclic prod-
uct obtained from 3,3’,4,4’-tetraethylbi-
pyrrole under the conditions of the
electrochemical oxidation. When simi-
Keywords: anions · electrochemis-
try · electrolysis · oxidative cou-
pling · pyrroles
Introduction
to be amenable to chemical synthesis through the oxidative
coupling of 3,3’,4,4’-alkyl substituted 2,2ꢀ bipyrroles.[6,7] In a
broader sense, chemically driven oxidative coupling reac-
tions between aryl fragments have been extensively em-
ployed in the preparation of numerous compounds including
polycyclic aromatic hydrocarbons,[8] fused porphyrin
arrays,[9] natural products,[10] and porphyrin analogues.[11,12]
However, harsh reaction conditions or expensive organome-
tallic reagents are often required. This has made the use of
electrochemical methods attractive as possible alterna-
tives.[13–21]
Expanded porphyrins have emerged in recent years as an at-
tractive class of porphyrin analogues, with interesting anion-
binding properties,[1,2] unusual metal complexation proper-
ties,[3] and potential utility in the area of molecular electron-
ics device construction.[4,5] Owing to the absence of meso
carbon atoms in their cyclic frameworks, one set of expand-
ed porphyrins, the cyclo[n]pyrroles, represented by struc-
tures [1H6]2+, [2 H7]2+, and [3H8]2+ (Scheme 1), were found
Although widely employed to produce conducting or insu-
lating amorphous polymeric structures[22,23] including poly-
thiophene or polypyrrole, recent findings have provided sup-
port for the highly appealing suggestion that porphyrins and
porphyrin analogues could be prepared directly by using
electrochemical methods. Such an approach was pioneered
by Smith and co-workers who described the first electrosyn-
thesis of porphyrins from a,c-biladienes.[24,25] More recently,
an N-substituted cyclo[6]pyrrole was produced from a steri-
cally constrained hexapyrrolylbenzene by using a covalently
templated electrochemical procedure.[26] In the context of
our own efforts to develop electrochemical oxidations as a
potentially clean and straightforward complement to chemi-
cal-based oxidative procedures, we have reported the elec-
trochemically driven ring-closure of a linear hexapyrrole to
give a [24]hexaphyrin(1.0.1.0.0.0).[27] We have also described
the electrolyte-assisted electrochemical oxidative conversion
[a] Dr. M. Buda, A. Iordache, Dr. C. Bucher, Prof. J.-C. Moutet,
Prof. G. Royal, Prof. E. Saint-Aman
Dꢁpartement de Chimie Molꢁculaire
Laboratoire de Chimie Inorganique Rꢁdox
UMR-5250, Universitꢁ Joseph Fourier, BP 53, 38041
Grenoble Cꢁdex 9 (France)
Fax : (+33)0476-51-4267
E-mail: christophe.bucher@ujf-grenoble.fr
[b] Prof. J. L. Sessler
Department of Chemistry and Biochemistry
University of Texas, 1 University Station-A5300 Austin
Texas 78712-0165 (USA)
Fax : (+1)512-471-5009
E-mail: sessler@mail.utexas.edu
[c] Dr. M. Buda
Permanent address: Dept. of Applied Physical Chemistry
and Electrochemistry, “POLITEHNICA” University of Bucharest
Calea Grivitei 132, 010737, Bucharest (Romania)
6810
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 6810 – 6819

FULL PAPER
effect various anions might
have on the electrochemical
conversion of bipyrrole 4 into
cyclo[n]pyrrole.
Before considering the prob-
lem of synthesis, the effect of
anions on the redox potentials
of
cyclic voltammetry. Initially,
two cyclic voltammograms
4 was assessed by using
Scheme 1.
(CVs), corresponding to the ox-
À
idation of 4 with BrÀ and BF4
(both as the tetrabutylammoni-
um, TBA, salts) as the electrolytes, were recorded. As can
be seen from an inspection of Figure 1, the effect of anions
is significant. Although both CVs are qualitatively alike,
of 3,3’,4,4’-tetraethyl-2,2’-bipyrrole (4, Scheme 1) into the
corresponding cyclo[8]pyrrole (e.g., [3H8]2+).[28]
In our hands the electrochemical-based strategy not only
gave yields that were competitive with those obtained by
using the best chemical oxidation methods,[29,30] it also al-
lowed us to avoid the use of harsh conditions, such as 1 n
acid and ironACHTUNGTRENNUNG(III) or chromium(VI) salts, that are a hall-
mark of these chemical methods. Another appealing aspect
of electrochemical-based methods is that they allow the re-
action conditions to be varied easily. For instance, under
conditions of electrolysis it is possible to adapt the oxidation
potentials to suit specifically the redox properties of a given
pyrrolic precursor and to study the effects of various poten-
tial templating agents. The study of possible templating ef-
fects is of particular interest in the case of cyclo[n]pyrroles.
This is because cyclo[8]pyrrole [3H8]2+ proved to be the ex-
clusive product when the chemical oxidation was carried out
under conditions of chemical oxidation in the presence of
H2SO4; in contrast, a mixture of cyclo [6]-, cyclo[7]-, and cy-
clo[8]pyrrole (i.e., [1H6]2+, [2H7]2+, and [3H8]2+, respective-
ly) was obtained when HCl was used.[6] We thus set out to
explore the electrochemical oxidation of bipyrrole 4 and
pyrrole 5 (Scheme 1) and to investigate the possible tem-
plating effects of various electrolytes. Specific motivation
for this study came from an appreciation that 1) acids other
than H2SO4 and HCl failed to give isolable quantities of any
cyclo[n]pyrrole when bipyrrole 4 was subject to chemical ox-
idation and 2) attempts to effect the direct conversion of
pyrrole 5 to cyclo[n]pyrrole likewise failed completely. As
detailed below, these limitations are relaxed when electro-
chemical methods are employed.
Figure 1. CVs corresponding to the oxidation of 4 (1.5ꢃ10À3 m) in CH2Cl2
in the presence of two different supporting electrolytes tetra-n-butylam-
monium bromide (c=0.15 m TBABr) and tetra-n-propylammonium
tetrafluoroborate (a=0.11 m TPABF4). Pt electrode, Ø=2 mm,
100 mVsÀ1
.
with an irreversible oxidation and a reduction wave of weak
intensity in the reverse scan being observed at all accessible
scan rates (attributed to the formation and reduction of an
electro-generated quaterpyrrole radical cation),[31] an ap-
proximately 100 mV difference in both signals is clearly evi-
dent.[32,33]
The influence of organic and inorganic anions on the
polymerization of pyrrole has been investigated by several
authors and it has been shown, for instance, that the favora-
ble action of tosylate anions in promoting polypyrrole for-
mation can be accounted for in large measure by homogene-
ous processes involving ion pairing and hydrogen bond-
ing.[34] The significant potential shift (ꢀ100 mV; see above)
Results and Discussion
Electrochemical oxidation of 3,3’,4,4’-tetraethyl-2,2’-bipyr-
role (4): In our initial studies, we found that cyclo[8]pyrrole,
specifically [3H8]ACHTUNGTRENNUNG(SO4), was obtained efficiently when bipyr-
role 4 was subjected to bulk electrolysis. However, the effect
of putative anion templates was not investigated, even
though such effects were thought to account for the roughly
3-fold reduction in yield of [3H8]2+ seen when H2SO4 was
replaced by HCl under conditions of chemical oxidation.[6]
A primary goal of the present study was thus to test the
À
observed between the BrÀ and BF4 curves can thus reason-
ably be attributed to strong anion binding effects. Specifical-
ly, a tightly bound (hydrogen-bonded, ion-paired) electron-
rich anionic species would increase the overall electron den-
sity on 4 and 4+· and thus shift the corresponding oxidation
to a less positive value.[35] A similar anion effect has been
Chem. Eur. J. 2010, 16, 6810 – 6819
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6811

C. Bucher, J. L. Sessler et al.
previously noticed by Zotti et al.[34] in the context of study-
ing the anodic coupling of 2,2’-bipyrrole in acetonitrile.
To assess the affect of anions in the context of the electro-
chemical conversions to cyclo[n]pyrroles, several electroly-
ses involving bipyrrole 4 were carried out in dichlorome-
thane in the presence of various supporting electrolytes
(TBA salts containing different anions) by using a platinum
working electrode. The data obtained from these analyses
are summarized in Table 1.
computed from crystal structure data.[36] Although these
values may not represent the actual size of each anion, they
nevertheless reflect accurately the relative size of the anions
involved in our experiments. Therefore, by using these
values, plots of yield versus size were constructed (Figure 2).
Table 1. Yields corresponding to the electrochemical synthesis of [3H8]2+
from 4 in the presence of different supporting electrolytes. The thermo-
chemical radii of each anion are those of Roobottom.[36] Peak potentials
correspond to the pyrrole-centered irreversible oxidation wave.
Supporting electro- Yield
Anion radii
[nm][b]
Oxidation potential
[V][a]
lyte
[%]
TBAF
TBACl
TBABr
TBANO3
TPABF4
TBAHSO4
TEAClO4
ꢀ0
17
33
54
55
68
11
0.126
0.168
0.190
0.200
0.205
0.218
0.225
0.080
–
0.065
0.080
0.165
0.350
–
Figure 2. Plots of the yield of [3H8]2+ obtained through electrochemical
oxidation of 4 against (left) the size of the anion[36] and against (right)
the logarithm of the gas-phase acidity of the conjugate acid of the anion
in question.[37–39]
[a] Irreversible oxidation: the values are the peak potentials measured at
a scan rate of 0.1 VsÀ1 (Pt electrode, E vs. Ag/Ag+). TBA=tetra-n-buty-
lammonium, TPA=tetra-n-propylammonium, TEA=tetraethylammoni-
um. [b] Thermochemical radii were used for complex anions; see text for
details.
In agreement with theoretical calculations predicting that
dicationic pyrrole oligomers tend to wrap around anions,[40]
the resulting graph provides support for the notion that ap-
propriately sized anions, specifically ones that can fit well
into the pocket of [3H8]2+, act as templates and help drive
It is apparent from these results that the choice of anionic
species within the supporting electrolyte has a major effect
on the oxidation potentials and on the efficiency of the cyc-
lization process. Specifically, yields of [3H8]2+ were found to
vary from close to 0% when the electrolysis was carried out
in the presence of TBAF, to almost 70% with TBAHSO4.
Notable is the factor of four reduction in yield when the
electrolysis was effected in the presence of ClÀ instead of
À
the electrochemically driven C C coupling process towards
the cyclo[8]pyrrole product. The trend is quite consistent
À
and shows that ions with very similar sizes such as BF4 and
À
NO3 afford similar yields.
Another important factor that could account for the ob-
served electrolyte versus cyclo[8]pyrrole yield data is the ba-
sicity of the anionic species. It is well known from a wide
range of solid-state structures that cyclo[n]pyrroles, as well
as linear polypyrroles, in their protonated forms can bind
anions.[41,42] As a general rule, more basic anions are expect-
ed to bind more strongly. However, if the anionic species in
question is too basic, deprotonation can occur. During elec-
trolyses, the a,a-coupling process generates free protons
(Scheme 2). However, if the electrolyte is relatively basic, as
is true for FÀ, the anion template effect, relying on hydrogen
bonds and electrostatic interactions, could be inhibited as
the result of deprotonation and formation of well-recog-
nized species such as HF and F2HÀ anions. Depending on
the extent to which this occurs, the a,a coupling process
might not proceed or follow unproductive pathways with
regard to cyclo[8]pyrrole formation.[43]
À
HSO4 . This correlates well with the threefold decrease seen
under conditions of chemical oxidation. Although this is not
proof, such an observation is consistent with the intuitively
appealing conclusion that the reaction in both cases is con-
trolled by anion-related factors such as a specific templating
effect.[6]
Given the above observation, we have attempted to corre-
late the cyclization yield with two key parameters, namely,
the size and basicity of the anions in question. Both of these
factors would be expected to affect the strength of the NH–
anion interaction both in the starting bipyrrole 4 and in the
final cyclo[8]pyrrole [3H8]2+. Although other factors, includ-
ing the strength of the hydrogen bonds and the ability to
form ion pairs in solution, could also play important roles in
regulating the yield, this analysis was expected to provide a
first-order insight into the role anions could be playing in
terms of mediating the conversion of two-pyrrole fragment
The experimental basicities of anions in CH2Cl2 are not
well established. Therefore, for the purposes of comparison,
we used the gas-phase acidities of the corresponding conju-
gate acids, either experimental (HX) or calculated (HNO3,
H2SO4, HClO4, and HBF4).[37–39] By using these values, a
4 into cyclic octapyrrolic product [3H8]2+
Unfortunately, experimental data for the ionic radii of
.
À
À
complex anions such as BF4 and HSO4 are scarce. Thus,
we chose to use the semi-empirical thermochemical radii, as
6812
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 6810 – 6819

Electrochemical Syntheses of Cyclo[n]pyrrole
FULL PAPER
Scheme 2. Possible mechanism leading to the cyclo[8]pyrrole [3H8]2+ from four molar equivalents of bipyrrole 4. The final steps from 8 to 3 have been
detailed from [8H4]2+ but similar coupling processes could potentially be considered from [8H3]+, [8H2], or [8H4]+·
.
plot of yield versus inferred basicity (specifically the log of
the gas-phase acidity) was constructed (Figure 2). In com-
parison with what was seen in the analogous plot relating
the product yield with the size of the anion template (also
given in Figure 2), the correlation is less clear. For instance,
anion geometry can be more safely ignored, then the yield
follows the anion basicity, with FÀ (which is a rather strong
base in CH2Cl2) falling at the extreme limit of the plot. Pre-
sumably, this better correspondence reflects the fact that in
this solvent system there is a good correlation between
halide anionic radius and effective basicity.
À
À
BF4 (which is a weaker base than HSO4 ) gives a lower
À
yield, whereas NO3 and BrÀ (two anions that have almost
Note that cyclo[8]pyrrole [3H8]2+ was the only macrocy-
clic product obtained from 4 under the conditions of the
electrochemical oxidation. Specifically, no smaller analogues
of cyclo[8]pyrrole, namely, [1H6]2+ or [2H7]2+, were ever
identical basicities) produce very different yields. On the
other hand, if we restrict the discussion to the three halide
anions in which complicating factors such as disparities in
Chem. Eur. J. 2010, 16, 6810 – 6819
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6813

C. Bucher, J. L. Sessler et al.
identified, even in trace amounts. This was true for all of the
electrolyses that we performed even when nonideal anions
such as chloride and fluoride were used to make up the elec-
trolyte. This finding stands in marked contrast with what is
seen under conditions of chemical oxidation in the presence
of HCl. For instance, when bipyrrole 4 was subject to oxida-
tion with FeCl3 in HCl (1m), isolable quantities of cyclo[6]-
pyrrole [1H6]2+ and cyclo[7]pyrrole [2H7]2+ were obtained.[6]
The formation of these latter species through chemical syn-
thesis must necessarily reflect the specific reaction condi-
tions, which involve the use of 1) a relatively strong iron-
analogous to those used in the case of the ironACHTUGNTERN(UNG III)-mediat-
ed oxidative coupling.[45] Nevertheless, we feel that one key
to the present electrochemical approach is that it allows
coupling to take place under lower proton concentrations
than those required for chemical coupling.[46]
At present, it is not possible to formulate a detailed mech-
anism for the reactions involving the conversion of bipyrrole
4 to the isolated cyclo[8]pyrrole product [3H8]2+. However,
as detailed below (and summarized in Scheme 2), considera-
ble insights into the details of the process have been ob-
tained. The unambiguous first step of the process is the irre-
versible oxidation of the bipyrrole 4 to form a cation radical
4+·. This radical is then expected to evolve spontaneously
ACHTUNGTRENNUNG(III)-based oxidant, 2) a biphasic reaction set-up, and 3) a
strongly acidic aqueous phase (1m HCl). Presumably, these
conditions influence the anion–pyrrole interactions as well
as, possibly, the bipyrrole–bipyrrole coupling process.[43] In-
terestingly, in the absence of strong acid, isolable quantities
of cyclized product were not obtained when bipyrrole 4 was
treated with FeCl3. Left undetermined by these observations
is whether the high proton concentration used under the
chemical conditions is needed to increase the activity of the
À
towards the quaterpyrrole 8H4 as the result of a C C cou-
pling with the concomitant release of two protons. Although
this latter species will be easier to oxidize than the starting
material 4,[31] the resulting radical [8H4]+· is expected to be
far less reactive than its shorter precursor 4+·,[47] leading us
to propose that the dimerization of 4+· (producing quater-
pyrrole 8H4) proceeds more quickly than various competing
processes.
ironACHTUNGTRENNUNG(III) reagent or to create protonated intermediates that
would be more likely to undergo cyclization. Although it is
difficult to ascertain directly the effective pH under either
the biphasic chemical conditions or the purely organic ones
used for the electrolyses, it is safe to assume that the proton
concentration is lower under the latter conditions. This is
certainly true at the beginning of the electrolyses because
no added acid is present. Although unlikely to rise to the
level provided by contact with a biphasic aqueous HCl (1m)
layer, it is important to appreciate that the proton concen-
tration does increase throughout the electrolysis process.
This is because protons are released as the result of the suc-
Consistent with this hypothesis is our failure to observe
the formation of hexa- or heptapyrrolic products (e.g.,
[1H6]2+ or [2H7]2+) under the conditions of electrochemical
cyclization, even when templating anions of small size such
as chloride are employed. This excludes the involvement of
terpyrrolic or pentapyrrolic intermediates. These same ob-
servations also rule out any pathways that would require the
coupling of quaterpyrrole species such as 8 or its radical
forms with bipyrrole precursors such as 4 or 4+· as being sig-
nificant. On the other hand, the observed anion-mediated
effects are fully consistent with the coupling of bipyrrole 4
to form a tetrapyrrolic intermediate, which then couples
with yet another tetrapyrrole to produce the cyclo[8]pyrrole
product [3H8]2+. What is less clear, however, is the oxidation
state and degree of protonation of the intermediates in
question, 8H4, 8H42+, 8H3, and 8H2.
À
cessive C C couplings. This proton release will displace the
various equilibria towards fully protonated species. This
chemistry is shown in Scheme 2 for the quaterpyrrole inter-
mediates such as [8H4]2+, which are assumed to form at the
working electrode through oxidative coupling of two bipyr-
role precursors (see below). This, in turn, will shift the oxi-
dation potential of the fragments in question towards more
positive values. In a less acidic environment such as that ex-
pected to operate during the early stages of the electrolysis,
the formation of neutral and thus easier to oxidize inter-
mediates such as [8H2] could lead to a different reactivity
profile and product distribution than that operative under
the conditions of chemical oxidation.
One possibility is that a p dimer[48] is formed between two
tetrapyrrolic radicals such as [8H4]+·, and that such a species
would be the intermediate giving rise to the final cyclic
product. To the extent that this occurs, it would lead to the
direct (and exclusive) formation of the final cyclo[8]pyrrole
product [3H8]2+. It is known that oligopyrrole radical cations
tend to form p dimers,[33,49] and the production of such spe-
cies would certainly be consistent with the exclusive forma-
tion of cyclo[8]pyrrole under conditions of electrolysis.
However, such a p dimer is likely to have its geometry set
by orbital interactions; therefore, its transformation into
[3H8]2+ is unlikely to be affected to a significant degree by
the nature of the anion present in the electrolyte. However,
the anions do play a critical role (see above), and we believe
that at least one nonradical tetrapyrrolic species is involved
in the final coupling step. In fact, based on spectroelectro-
chemical analyses (see below), and on existing data that
show that the (unsubstituted) tetrapyrrole radical cation has
Strong evidence of the effect of acidity on pyrrole–oligo-
mer electrochemistry has been put forward recently by
Kadish and co-workers.[44] Specifically, these researchers
showed that an a-protected tetrapyrrole derivative analo-
gous to 8 could be protonated in the presence of a large
excess of TFA in CH2Cl2 and that this protonation, in turn,
affects quite dramatically the electrochemical activity of this
particular open-chain tetrapyrrolic species. Unfortunately,
pyrrole and oligopyrroles such as 8, which are not a-protect-
ed can polymerize spontaneously in the presence of Brçnst-
ed or Lewis acids. Thus, we were unable to carry out electro-
chemical oxidations of 4 under acidic conditions directly
a
rather low p-dimerization equilibrium constant
(ꢀ70 LmolÀ1),[33,49] we conclude that such dimeric species
6814
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 6810 – 6819

Electrochemical Syntheses of Cyclo[n]pyrrole
FULL PAPER
are not important intermediates in the conversion of 4 into
425 nm. Although the absence of unambiguous evidence
leaves the issue of mechanism unresolved, our failure to ob-
serve higher-order intermediates during the course of ob-
taining cyclo[8]pyrrole [3H8]2+ leads us to suggest that this
product is formed through an anion-assisted cyclization pro-
cess. Specifically, we propose that 3H8 is formed from first
bipyrrolic and then tetrapyrrolic species, without the in-
volvement of higher-order oligopyrrolic species such as 7.
[3H8]2+
.
To rule out the significant participation of intermediates
other than those related to 4 and 8, a bulk electrolysis ex-
periment involving bipyrrole 4 and tetra-n-butylammonium
hydrogensulfate as the electrolyte was carried out in di-
chloromethane. The results of this electrolysis reaction was
followed in situ by using a 1 mm UV/Vis immersion probe
(Figure 3). The signature spectral feature growing in the
Electrochemical oxidation of 3,4-diethylpyrrole: To the
extent that the above mechanism is correct, any process that
produces bipyrrolic species such as 4 would be expected to
generate cyclo[8]pyrrole, as long as side reactions are kept
to a minimum. Unfortunately, all efforts to produce cyclo[8]-
pyrrole [3H8]2+ directly from 3,4-diethylpyrrole (5) through
chemical oxidation have so far failed. Presumably, this re-
flects the fact that highly oxidizing conditions are needed to
oxidize the monopyrrole to the corresponding bipyrrole,
which allows for side reactions to occur that can compete
with the subsequent coupling to produce the requisite tetra-
pyrrole.
In contrast, the relatively mild nature of the electrolysis
conditions used to produce [3H8]2+ led us to consider that
such an approach could be employed to convert 3,4-diethyl-
pyrrole 5 directly to cyclo[8]pyrrole. The advantage of this
putative electrochemical process is that it would prevent the
need to synthesize a b-substituted bipyrrole such as 4. Bipyr-
roles of this type, although convenient building blocks for
expanded porphyrin construction, are typically obtained in
four steps from 3,4-diethylpyrrole ethyl ester through a
copper-mediated Ulmann coupling and are not routinely
available on a large scale.[50–52] On the other hand, b-alkylat-
ed pyrroles such as 5 are easier to prepare and, not surpris-
ingly, they have already been studied as precursors of poly-
meric materials. However, to the best of our knowledge,
none of the investigations involving polymerization led to
the identification and/or isolation of well-defined cyclic spe-
cies.[53–57] In support of our assumption that 4 and subse-
quently cyclo[n]pyrrole might be produced electrochemical-
ly from 3,4-diethyl pyrrole 5, we note a recent contribution
from Kita and co-workers,[58] wherein 4 was produced direct-
ly from 5 through the oxidative coupling of the correspond-
ing cation radicals formed in situ by treatment with phenyl-
iodine bis(trifluoroacetate) (PIFA) and bromotrimethylsi-
lane.
As above, prior to carrying out the electrolysis reactions,
the electrochemical properties of 3,4-diethylpyrrole were an-
alyzed. In agreement with previous studies,[53,55–57] the CV of
5 recorded in CH2Cl2 containing TBAHSO4 (0.1m; Figure 4)
revealed the characteristic irreversible wave of pyrrole oxi-
dation at approximately 0.95 V. Owing to its limited conju-
gation pathway and lower number of electron-donating b-
alkyl groups, the pyrrole-centered oxidation process is ob-
served at a much more positive potential in 5 than in 4. This
is very much as expected and is fully consistent with the fail-
ure to obtain cyclo[8]pyrrole from 5 through chemical oxida-
tion; the required conditions simply appear too harsh.
Figure 3. Left: Electrolysis of a solution of 4 (5.10À4 m) followed by UV/
Vis spectroscopy (l=1 mm; CH2Cl2 +TBAHSO4 (0.25m); Eapp =0.3 V vs.
Ag/Ag+; Pt plate working electrode; the initial spectrum is virtually in-
distinguishable from the x-axis). Right: Electrolysis of
a solution of
[3H8]SO4 (1.25 10À4 m) followed by UV/Vis spectroscopy (l=1 mm;
CH2Cl2 +TBAHSO4 (0.25m); Eapp =0.2 V vs. Ag/Ag+; Pt plate working
electrode; c=initial and c=final).
first stage of the experiment, namely, a broad band above
900 nm and a narrower one at 425 nm, was attributed to the
targeted cyclo[8]pyrrole [3H8]2+, which evolves at the end of
the electrolysis towards the mono oxidized species [3H8]3+
.
This assignment is based on the finding that a species with
an identical spectral signature is observed when isolated cy-
clo[8]pyrrole is subject to oxidation under spectroelectro-
chemical conditions (Figure 3). The spectroelectrochemistry
experiments shown in Figure 3 are thus fully in accord with
the efficient formation of [3H8]SO4 without the formation of
significant quantities of any open-chain oligopyrrolic inter-
mediates containing more than four pyrrolic subunits.
A similar conclusion was reached from analogous experi-
ments carried out by using a thin-layer spectroelectrochemi-
cal cell (an electrochemical cell with a 1 mm optical path),
which also failed to reveal spectral features ascribable to
nonbipyrrolic or nontetrapyrrolic intermediates. In fact, se-
quential UV/Vis spectra recorded during the course of
single CVs carried at sweep rates of either 50 mVsÀ1 or
500 mVsÀ1 revealed features that matched what was ob-
served during the first stages of the bulk electrolysis experi-
ments described above. Specifically, a broad band above
900 nm was seen, as well as a narrower feature at lmax
=
Chem. Eur. J. 2010, 16, 6810 – 6819
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6815

C. Bucher, J. L. Sessler et al.
material. Two products could be isolated after workup,
which were identified as being the cyclo[7] and cyclo[8]pyr-
role products [2H7]2+ and [3H8]2+, respectively. Unfortu-
nately, the yields for these two species were found to range
from 1 to 3%, with the best reproducible yield of [3H8]2+
(2.5% isolated yield) being obtained for a charge corre-
sponding to 1.5 electrons per pyrrole equivalents. In the
case of [2H7]2+, the best yield (3% isolated) was obtained
when the electrolysis was stopped after passing a charge cor-
responding to 0.5 electrons per pyrrole units. Unfortunately,
all efforts to increase the yields (such as changes in electro-
lyte, solvent, and temperature) met with failure.[60] Never-
theless, it is important to appreciate that, to the best of our
knowledge, the present study defines the first time wherein
stable and clearly defined cyclic expanded porphyrin species
have been isolated as products from the electrolysis of a
simple pyrrole derivative.
Figure 4. CV of 5 in CH2Cl2 +TBAHSO4 (0.2m; ꢀ4.5ꢃ10À3 m; c) and
CH2Cl2 +TBAClO4 (0.1m; ꢀ2.5ꢃ10À3 m; a). Pt electrode, Ø=2 mm,
100 mVsÀ1
.
Spectroelectrochemical investigations of [2H7]2+ and
[3H8]2+: Because the yields in [2H7]2+ and [3H8]2+ turned
out to be low, we questioned the stability of their oxidized
forms generated in situ as the result of the electrosyntheses.
Importantly, under the conditions used for our CV analy-
sis (CH2Cl2; TBAHSO4 (0.1m)), no evidence of polymer
film being deposited onto the electrode was observed; this
was true even after repetitive cycling. Although no polymer-
ic materials were evidently formed under the conditions of
cyclic voltammetry, the irreversibility of the oxidation wave
is consistent with the species produced on oxidation, 5+·,
evolving quickly towards soluble oligomers by successive
coupling reactions. The rather large difference between the
peak potentials measured on both irreversible curves shown
in Figure 4 (ꢀ0.95 V for the TBAHSO4 electrolyte com-
pared with 0.72 V for the TBAClO4 electrolyte)[59] also
serves to highlight a significant anion effect, similar to what
was observed when bipyrrole 4 was subject to electrooxida-
tion. This led us to consider that appropriate anion-templat-
ing conditions could be found, which would allow for the
direct formation of 3 from 5.
A preparative-scale electrooxidation of 5 was conducted
in CH2Cl2 by using a large platinum electrode (see the Ex-
perimental Section for details). In this study, the current was
maintained at <0.15–0.20 mAcmÀ2 and a relatively low con-
centration of 5 (<10 mm; typically 4–4.5 mm) was employed.
Under these conditions the electrolysis usually yielded
almost exclusively soluble products with very little polymer
being deposited onto the electrode. At higher concentra-
tions of 5 (>10 mm) the formation of an insoluble polymer
film on the platinum electrode became clearly visible.
Because it is known that poly(5) usually exhibits low con-
ductivity[59] and that, therefore, even thin films would serve
to passivate the working electrode, special care was taken to
minimize polymerization. It was found that the easiest way
of avoiding polymer formation was to oxidize 5 at an initial
concentration of 4–5 mm by potential cycling starting at low
potentials (at the foot of the oxidation wave) and gradually
increasing the lower and upper limits as 5 was consumed
(see the Experimental Section for details).
Table 2 summarizes the reversible oxidation and reduction
[6]
potentials recorded for [2H7]2+ and [3H8]2+
.
Table 2. Summary of the electrochemical data for [2H7]2+ and [3H8]2+ in
CH2Cl2 +TBAP (0.1m). Irreversible oxidation waves are also observed
with both species at approximately 0.7 V.
A
[3H8]2+
ACHTUNGTRENNUNG
E0’
E0’
E0’
0.08 (1 eÀ)
0.35 (1 eÀ)
0.37 (1 eÀ)
–
ox1
ox2
red1
À0.53 (2 eÀ)
À0.41 (2 eÀ)
Species [3H8]2+ shows two reversible one-electron oxida-
tion waves. Moreover, bulk oxidation of [3H8]2+ carried out
at 0.200 V induced drastic changes in the UV/Vis signature,
with intense absorption bands growing in between 700 and
900 nm and a clear splitting of the Soret-like band at ap-
proximately 400 nm likewise occurring (Figure 5). These
latter spectral features were attributed to the oxidized form
[3H8]3+, a species that proved to be stable on the time scale
of the electrolysis as evidenced by the fact that the spectral
signature of the starting material [3H8]2+ could be quantita-
tively recovered after a back-reduction was performed at
0 V.
The evolution of the UV/Vis spectrum of [3H8]2+ ob-
served as oxidation was performed at 0.525 V, conditions in
which the doubly oxidized species [3H8]4+ should be
formed, is shown in Figure 5. The spectrum recorded after
removing 2 eÀ per molecule is quite similar to that of
[28]
[3H8]3+
,
with intense bands being observed at 743 and
831 nm. It is thus concluded that [3H8]3+ is formed as an in-
termediate. However, three new bands can also be seen at
297, 360, and 600 nm, the intensities of which increase as the
columbic charge is raised. After passing 6 eÀ per molecule, a
point at which the signature features of [3H8]3+ were still
Following the protocol detailed in the Experimental Sec-
tion, electrolyses were performed by using 5 as the starting
6816
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 6810 – 6819

Electrochemical Syntheses of Cyclo[n]pyrrole
FULL PAPER
Figure 7 shows the evolution of the UV/Vis spectrum of
[2H7]2+ recorded during a bulk oxidation conducted at
0.400 V. The trend is similar to what was observed during
Figure 5. Left: UV/Vis absorbance spectra recorded for a solution of
[3H8]2+ in CH2Cl2 (1.25ꢃ10À4 m)+TBAHSO4 (0.25m) before (a) and
after (c) bulk oxidation conducted at 0.200 V. The arrow in the inset
CV points to the 0.200 V value. Right: Electrooxidation of [3H8]2+ in
CH2Cl2 (4ꢃ10À5 m)+TBAHSO4 (0.15m) at 0.525 V under nitrogen (in a
dry-box). The arrow in the inset CV points to the 0.525 V value; c=
initial, d=1 eÀ per molecule, a=2 eÀ per molecule, and c=6 eÀ
per molecule.
Figure 7. Left: Electrooxidation of [2H7]2+ in CH2Cl2 (5ꢃ10À5 m)+
TBAHSO4 (0.15m) at 0.400 V under nitrogen (in a dry-box). Right: Com-
+
parison between the initial absorbance (g) of [2H7]2 and the spec-
trum of the solution after its oxidation at 0.400 V and re-reduction at
À0.1 V (c=0.6 eÀ per molecule and c=1.0 eÀ per molecule). The
arrows in the inset CV point to the 0.400 V (solid arrow) and À0.100 V
(dotted arrow) values.
clearly visible, the working electrode potential was set to
À0.10 V to effect reduction.
The spectrum was then recorded after full reduction and
was compared to the initial spectrum of [3H8]2+ (Figure 6).
The bands at 297 and 360 nm are still clearly visible, with
the band at 600 nm converted into a shoulder. A new band
is also observed at approximately 570 nm. On this basis it is
concluded that roughly half of the initial starting material
[3H8]2+ is lost over the course of this bulk oxidation/reduc-
tion process. It is thus inferred that [3H8]4+, although stable
enough to permit the recording of CV scans, is not stable on
the time scale of a normal electrolysis reaction.
the oxidation of [3H8]2+ to [3H8]3+ with the initial band at
425 nm splitting into two less-intense signals centered at 425
and 459 nm.[28] As the electrolysis proceeds, the shoulder at
845 nm and the intense band at approximately 950 nm
almost disappear, and two new bands (attributed to [2H7]3+
by analogy to [3H8]3+) appear at 673 and 725 nm. However,
these changes occur monotonically only up to about 0.6 eÀ
per molecule; afterwards, the newly formed bands start to
decrease and a new band at 370 nm starts to grow in.
After passing 1 eÀ per molecule, the solution was subject
to reduction by setting the electrode potential at À0.100 V
(Figure 7, right). After reduction, analysis of the UV/Vis
1
spectrum revealed that about = of the initial [2H7]2+ was
3
lost (as inferred from changes in the spectral intensity). On
the other hand, the band at 370 nm is still visible as a
shoulder. We thus conclude that [2H7]3+ is not stable on the
time scale of the electrolysis, at least under our experimen-
tal conditions. The fact that neither this species nor [3H8]2+
is stable at the high oxidation potentials used to effect mac-
rocycle synthesis provides a reasonable explanation for why
the yields of the cyclic product obtained from 5 are low.
Conclusion
In summary, we have established that appropriately sized
anions act as templates that favor the electrochemical oxida-
tive conversion of 3,3’,4,4’-tetraethyl-2,2’-bipyrrole into the
corresponding cyclo[8]pyrrole. For the first time we have
also isolated cyclic expanded porphyrins, namely, cyclo[8]
Figure 6. Comparison between the initial absorbance (a) of [3H8]2+
and the spectrum of the solution after its oxidation at 0.525 V and re-re-
duction at À0.10 V (c). The arrows in the inset CV point to the
0.525 V (solid arrow) and À0.10 V (dotted arrow) values.
Chem. Eur. J. 2010, 16, 6810 – 6819
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6817

C. Bucher, J. L. Sessler et al.
sulting solid was washed with 2ꢃ60 mL aqueous solution of H2SO4
(0.1m) to remove the supporting electrolyte, filtered, and finally rinsed
with approximately 25 mL of distilled water. The dark insoluble solid was
recovered and dissolved in CH2Cl2. The solution was then dried over an-
hydrous Na2SO4, filtered, and the solvent was evaporated off to afford a
dark solid. Chromatography over silica gel by using CH2Cl2 +CH3OH
(1–5%) as the eluent allowed for the isolation of crude fractions contain-
ing [3H8]2+ and [2H7]2+, respectively. The crude fractions were further
purified by using silica-gel column chromatography that used CHCl3 +
CH3OH (1%) as the eluent for [5H8]2+ and CHCl3 +CH3OH (4%) as
and cyclo[7]pyrrole, as products from the electrolysis of a
simple pyrrole derivative. This approach, relying on the use
of a monomeric heterocyclic building block 3,4-diethylpyr-
role, still needs to be developed further if preparatively
useful quantities of the cyclic products are to be obtained.
Our current efforts are aimed at effecting this optimization
and to generalizing the present electrochemical-based strat-
egy such that viable quantities of new and extant expanded
porphyrin-type macrocycles can be obtained from cheap and
readily available starting materials.
the eluent for [2H7]2+
.
Electrochemical oxidation of 3,3’,4,4’-tetraethyl-2,2’-bipyrrole: A similar
approach was used for experiments involving the oxidation of bipyrrole,
except that the potential was not scanned. Rather, the potential was
maintained initially at À0.1 V and then slowly increased up to 0.025–
0.05 V.[35] A typical experiment involved approximately 20–25 mg of 2 in
approximately 60 mL of the solvent mixture (CH2Cl2 +supporting elec-
trolyte (0.15m, or 0.11m in the case of TPABF4)). The concentration of 2
was thus approximately 1.5 mm. After passing the required charge (2.4–
2.5 eÀ per molecule), the reaction was worked up by using a procedure
similar to the one employed in the case of the electrooxidation reactions
involving 5 (see above). The supporting electrolyte was removed by
washing with a dilute (ꢀ0.1m) aqueous solution of the corresponding
acid, except for the TBABr and TBAF, in which an aqueous solution of
H2SO4 (ꢀ0.05m) was used. In all other cases, care was taken to avoid
any contact with sulfate anions (e.g., drying was achieved by adding large
amounts of pentane followed by azeotropic distillation of water).
The yields of [2H7]2+ and [3H8]2+, especially in the case of electrooxida-
tions involving 5 in which small amounts of product are recovered
(ꢀ1 mg, but in some cases <0.3 mg), were calculated by using data from
UV/Vis spectroscopy (i.e., extinction coefficients) rather than weighing.
Thus, after isolation, samples of [2H7]2+ or [3H8]2+ were dissolved in
CH2Cl2 to produce solutions that were roughly 10À4 m in expanded por-
phyrin (as calculated from weighing when the quantity of material per-
mitted this) and the UV/Vis spectrum was recorded. The absorbance
values at 425 ([2H7]2+) and 428 nm ([3H8]2+) were measured and the
amount of substance in the solution was calculated by using the reported
data for the molar absorptivity[6,7] of [2H7]2+ and [3H8]2+. For the elec-
trolyses involving TBAHSO4, the yield was found to be reproducible
within 3% (average of three electrolyses) and only one electrolysis was
performed by using the other anions.
Experimental Section
General methods and materials: Pyrroles 4[50] and 5[61] were synthesized
according to procedures reported in the literature. The solvents used for
chromatography (chloroform (HPLC grade, Carlo Erba) and methanol
(SDS, anhydrous)) were used as received. The solvents used for electro-
chemical and spectroelectrochemical analyses (dichloromethane (SDS,
anhydrous), dichloroethane (98%), acetonitrile (Rathburn, HPLC
grade), and nitromethane (Acros Organics, 99+ %)) were used as re-
ceived. Tetrabutylammonium fluoride trihydrate (TBAF, Fluka, 98%),
tetrabutylammonium chloride (TBACl, Fluka, 99%), tetrabutylammoni-
um bromide (TBABr, Fluka, >98%), tetrabutylammonium nitrate
(TBANO3, Fluka, 97%), tetrabutylammonium hydrogensulfate
(TBAHSO4, Fluka, 99%), and tetraethylammonium perchlorate
(TEAClO4, Fluka, 97%) were used as received.
Tetrapropylammonium tetrafluoroborate (TPABF4) was prepared by
neutralizing tetrapropylammonium hydroxide (TPAOH) with an aqueous
solution of HBF4 (49.5–50.5%, Aldrich). The white precipitate obtained
in this way was collected by filtration, washed three times with small
amounts of water, and dried. The white solid that remained was dissolved
in a minimum amount of chloroform and the resulting solution was then
passed through a filter and evaporated to dryness. The resulting white
powder was dried for two days by using a vacuum pump at room temper-
ature and it was used without further purification.
All electrochemical data discussed in this report were obtained by using
platinum working electrodes. These had either a large area (55 cm2) in
the case of the electrolyses, or consisted of a small disk (CH-instrument,
2 mm diameter) for the collection of cyclic voltammetry data. All elec-
trolyses were carried out on a PAR 273 potentiostat, whereas cyclic vol-
tammetry data were obtained by using a CHI model 620 electrochemical
Spectroscopic data for [2H7]2+
:
1H NMR (CDCl3, 400 MHz,): d=À2.18
(brs; NH), 1.80 (t, J=7.28 Hz, 42H; CH2CH3), 4.25 ppm (brs, 28H;
CH2CH3); MS (ESI): m/z: 882.5 [MH2Cl]+; UV/Vis (CH2Cl2): lmax
(loge)=426 nm (4.63 molÀ1 dm3 cmÀ1). These data are in accord with the
published values.[7]
workstation. Spectroelectrochemical data were recorded by using
a
Spectroscopic data for [3H8]2+
:
1H NMR (CDCl3, 250 MHz, 293 K): d=
PAR 173 potentiostat-galvanostat and a Zeiss MCS 500 UV/NIR spec-
trometer.
3
À
À
1.59 (t, J=7.3 Hz, 48H; CH2CH3, ), 4.10 ppm (brs, 32H; CH2CH3);
MS (DCl/NH3): 969 [M+1]+; MS (ESI): m/z: 1065.7 [M+H2SO4]+; UV/
Vis (CH2Cl2): lmax (loge)=428 nm (4.76 molÀ1 dm3 cmÀ1). These data are
in accord with the published values.[6]
Typical experimental details for the electrochemical oxidation of 3,4-di-
ethylpyrrole: In a magnetically stirred three-compartment electrolysis
cell, 30–35 mL (ꢀ27–32 mg) of 5 were dissolved in CH2Cl2 (ꢀ60 mL)+
TBAHSO4 (0.2m; ꢀ4.5 mm of 5). The solution was purged with argon
for approximately 20 min prior to pyrrole addition (argon was bubbled
continuously until the end of the electrolysis). A large (ꢀ55 cm2) plati-
num mesh was used as the working electrode in the anodic compartment
with a carbon foam as counterelectrode in the cathodic compartment. A
double-junction Ag/AgNO3 (10À2 m) in CH3CN+tetrabutylammonium
perchlorate (0.1m) reference electrode was used in all experiments (all
potentials are referred to it). This reference electrode was placed in the
anodic compartment as close as possible to the working electrode to min-
imize the ohmic drop. After the addition of 5, the potential of the work-
ing electrode was scanned (50 mVsÀ) from 0.4 to 0.75 V, with a waiting
time of 3–12 s at the upper potential limit. The potential limits and wait-
ing times were gradually increased during the electrolysis (the upper
value was limited to 0.95 V) as 5 is consumed and the current drops.
After passing the required charge, the contents of both the anodic and
middle compartments (some mixing between these compartments occurs
during electrolysis) were collected and the solvent evaporated. The re-
Acknowledgements
The authors thank the Agence National de la Recherche for financial
support (ANR-09-JCJC-0083-01). We also thank the CNRS for providing
a “Poste Rouge” for M.B. The work in Austin was supported by the NSF
(grant CHE-074971 to J.L.S.) and the Robert A. Welch Foundation (F-
1018).
´
[1] L. R. Eller, M. Ste˛pien, C. J. Fowler, J. T. Lee, J. L. Sessler, B. A.
Moyer, J. Am. Chem. Soc. 2007, 129, 11020–11021.
[2] J. L. Sessler, S. Camiolo, Coord. Chem. Rev. 2003, 240, 17–55.
6818
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 6810 – 6819

Electrochemical Syntheses of Cyclo[n]pyrrole
FULL PAPER
[3] P. J. Melfi, S. K. Kim, J. T. Lee, F. Bolze, D. Seidel, V. M. Lynch,
J. M. Veauthier, A. J. Gaunt, Neu, M. P. Z. Ou, K. M. Kadish, S. Fu-
kuzumi, K. Ohkubo, Inorg. Chem. 2007, 46, 5143–5145.
[4] H. Xu, Y. Wang, G. Yu, W. Xu, Y. Song, D. Zhang, Y. Liu, Chem.
Phys. Lett. 2005, 414, 369–373.
up to 0.25 V. It is likely that in this case a larger junction-potential
value shifts the potential scale to more positive values compared
with the TBA-containing solutions.
[36] H. K. Roobottom, H. D. B. Jenkins, J. Passmore, L. Glasser, J.
Chem. Educ. 1999, 76, 1570–1573.
[5] H. Xu, G. Yu, W. Xu, Y. Xu, G. Cui, D. Zhang, Y. Liu, D. Zhu,
Langmuir 2005, 21, 5391–5395.
[37] NIST, Chemistry WebBook, http://webbook.nist.gov/chemistry/.
[38] I. A. Koppel, P. Burk, I. Koppel, I. Leito, T. Sonoda, M. Mishima, J.
Am. Chem. Soc. 2000, 122, 5114–5124.
[39] J. E. Bartmess, J. A. Scott, R. T. McIver, Jr., J. Am. Chem. Soc. 1979,
101, 6046–6056.
[40] T. Tamm, J. Tamm, M. Karelson, Int. J. Quantum Chem. 2002, 88,
296–301.
[41] P. J. Anzenbacher, K. Jursꢅkovꢆ, J. L. Sessler, J. Am. Chem. Soc.
2000, 122, 9350–9351.
[6] T. Kçhler, D. Seidel, V. Lynch, F. O. Arp, Z. Ou, K. M. Kadish, J. L.
Sessler, J. Am. Chem. Soc. 2003, 125, 6872–6873.
[7] D. Seidel, V. Lynch, J. L. Sessler, Angew. Chem. 2002, 114, 1480–
1483; Angew. Chem. Int. Ed. 2002, 41, 1422–1425.
[8] M. D. Watson, A. Fechtenkotter, K. Mullen, Chem. Rev. 2001, 101,
1267–1300.
[9] A. Tsuda, A. Osuka, Science 2001, 293, 79–82.
[10] R. B. Herbert, A. E. Kattah, A. J. Murtagh, P. W. Sheldrake, Tetrahe-
dron Lett. 1995, 36, 5649–5650.
[42] J. L. Sessler, A. Aguilar, D. Sanchez-Garcia, D. Seidel, T. Kçhler, F.
Arp, V. M. Lynch, Org. Lett. 2005, 7, 1887–1890.
[11] T. K. Chandrashekar, S. Venkatraman, Acc. Chem. Res. 2003, 36,
676–691.
[43] S. Sadki, P. Schottland, N. Brodie, G. Sabourad, Chem. Soc. Rev.
2000, 29, 283–293.
[12] J. L. Sessler, A. Gebauer, S. J. Weghorn in The Porphyrin Hand-
book, Vol. 2 (Eds.: K. S. K. M. Kadish, R. Guilard), Academic Press,
San-Diego, 2000, pp. 55–124.
[44] M. Zhang, W. E. K. Ohkubo, D. Sanchez-Garcia, D.-W. Yoon, J. L.
Sessler, S. Fukuzumi, K. M. Kadish, J. Phys. Chem. A 2008, 112,
1633–1642.
[13] G. Tourillon, F. Garnier, J. Electroanal. Chem. 1982, 135, 173–178.
[14] R. J. Waltman, A. F. Diaz, J. Bargon, J. Phys. Chem. 1984, 88, 4343–
4346.
[15] R. J. Waltman, J. Bargon, Can. J. Chem. 1986, 64, 76–95.
[16] O. Inganꢄs, B. Liedberg, W. Chang-Ru, Synth. Met. 1985, 11, 239–
249.
[45] M. Can, H. ꢇzaslan, ꢇ. Is¸ıldak, N. ꢇ. Pekmez, A. Yıldız, Polymer
2004, 45, 7011–7016.
[46] P. Audebert, J.-M. Catelt, G. Coustumer, V. Duchenet, P. Hapiot, J.
Phys. Chem. 1995, 99, 11923–11929.
[47] G. Zotti, S. Martina, G. Wegner, A.-D. Schlꢈter, Adv. Mater. 1992, 4,
798–801.
[17] S. Naitoh, Synth. Met. 1987, 18, 237–240.
[48] T. Nishinaga, K. Komatsu, Org. Biomol. Chem. 2005, 3, 561–569.
[49] J. J. Apperloo, R. A. J. Janssen, Synth. Met. 1999, 101, 373–374.
[50] J. L. Sessler, M. C. Hoehner, Synlett 1994, 211–213.
[51] R. Guilard, M. A. Aukauloo, C. Tardieux, E. Vogel, Synthesis 1995,
1480–1482.
[18] J. R. Reynolds, A. R. Katritzkyf, J. Soloducho, S. Belyakov, G. A.
Sotzing, M. Pyot, Macromolecules 1994, 27, 1225.
[19] J. P. Ferraris, R. G. Andrus, D. C. Hrncir, J. Chem. Soc. Chem.
Commun. 1989, 1318–1320.
[20] Z.-H. Zhou, T. Maruyama, T. Kanbara, T. Ikeda, K. Ichimura, T. Ya-
mamoto, K. Tokuda, J. Chem. Soc. Chem. Commun. 1991, 1210–
1212.
[21] M. V. Joshi, C. Hemler, M. P. Cava, J. L. Cain, M. G. Bakker, A. J.
McKinley, R. M. Metzger, J. Chem. Soc. Perkin Trans. 2 1993, 1081–
1086.
[22] A. Deronzier, J.-C. Moutet, Acc. Chem. Res. 1989, 22, 249–255.
[23] G. Sabouraud, S. Sadki, N. Brodie, Chem. Soc. Rev. 2000, 29, 283–
293.
[52] M. R. Johnson, C. Slebodnick, J. A. Ibers, J. Porphyrins Phthalocya-
nines 1997, 1, 87–92.
[53] H. D. Tabba, K. M. Smith, J. Org. Chem. 1984, 49, 1870–1875.
[54] G. H. Hansen, R. M. Henriksen, F. S. Kamounaha, T. Lunda, O.
Hammerich, Electrochim. Acta 2005, 50, 4936–4955.
[55] J. Rꢈhe, C. Krohnke, T. A. Ezquerra, F. Kremer, C. Wegner, Ber.
Bunsen-Ges. 1987, 91, 885–888.
[56] C. P. Andrieux, P. Audebert, P. Hapiot, J.-M. Savꢁant, J. Am. Chem.
Soc. 1990, 112, 2439–2440.
[24] D. Jeyakumar, K. M. Snow, K. M. Smith, J. Am. Chem. Soc. 1988,
110, 8562–8564.
[25] T. P. Wijesekera, D. Dolphin, Chemtracts: Org. Chem. 1990, 3, 375–
377.
[57] R. Casas, A. Dicko, J. M. Ribo, M. A. Valles, N. Ferrer-Anglada, R.
Bonnett, N. Hanley, D. Bloor, Synth. Met. 1990, 39, 275–280.
[58] T. Dohi, K. Morimoto, A. Maruyama, Y. Kita, Org. Lett. 2006, 8,
2007–2010.
[26] M. Lazerges, M. Jouini, P. Hapiot, P. Guiriec, P.-C. Lacaze, J. Phys.
Chem. A 2003, 107, 5042–5048.
[27] A. Iordache, P. Melfi, C. Bucher, M. Buda, J.-C. Moutet, J. L. Sess-
ler, Org. Lett. 2008, 10, 425–428.
[28] C. Bucher, C. H. Devillers, J.-C. Moutet, J. Pꢁcaut, J. L. Sessler,
Chem. Commun. 2006, 3891–3893.
[29] J. L. Sessler, D. Seidel, V. Lynch, J. Am. Chem. Soc. 1999, 121,
11257–11258.
[30] J. L. Sessler, D. Seidel, A. E. Vivian, V. Lynch, B. L. Scott, D. W.
Keogh, Angew. Chem. 2001, 113, 611–614; Angew. Chem. Int. Ed.
Engl. 2001, 40, 591–594.
[31] C. P. Andrieux, P. Hapiot, P. Audebert, L. Guyard, M. Nguyen Din-
h An, L. Groenendaal, E. W. Meijer, Chem. Mater. 1997, 9, 723–729.
[32] Different liquid-junction potentials also contribute to this potential
shift, but its magnitude does suggest an important anion effect.
[33] L. Guyard, P. Hapiot, P. Neta, J. Phys. Chem. B 1997, 101, 5698–
5706.
[59] S. Ito, H. Watanabe, H. Uno, T. Murashima, N. Ono, C. T. Yu, C.
Yu, R. G. Compton, Tetrahedron Lett. 2001, 42, 707–710.
[60] In one experiment 5 was oxidized at 3 mA (ꢀ55 mAcm2) constant
current, with no yield improvement. An electrolysis at 0.46 mA
(ꢀ8.5 mAcm2) in 1,2-dichloroethane+0.15m TBAHSO4 was carried
out over the course of 29 h and virtually only poly(5) was formed
onto the electrode. Low-temperature electrolyses were also attempt-
ed, suggested by the successful synthesis of 2,2’-bipyrrole from pyr-
role at temperatures below À408C (Ref. [57]). No yield improve-
ments were obtained when electrolyses were conducted between À5
and 08C in CH2Cl2 (TBAHSO4 (0.3m)). An electrolysis at À408C
was also performed in a 1:1 CH2Cl2:CH3CN mixture (TBAHSO4
(0.3m)) yielding only poly(5) with virtually no soluble products. Ni-
tromethane was substituted for acetonitrile, because solvents with
low donor number reportedly inhibit the polymerization of pyrrole,
but poly(5) was formed almost exclusively in this case also.
[61] D. H. R. Barton, S. Z. J. Zard, J. Chem. Soc. Chem. Commun. 1985,
1098–1100.
[34] G. Zotti, G. Schiavon, S. Zecchin, Chem. Mater. 1995, 7, 1464–1468.
[35] With tetrapropylammonium tetrafluoroborate as the supporting
electrolyte, the initial potential was set to 0 V and slowly increased
Received: January 8, 2010
Published online: May 3, 2010
Chem. Eur. J. 2010, 16, 6810 – 6819
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6819