Two new types of π-conjugation between a fullerene sphere and an addend

Article abstract of DOI:10.1039/b920339g

Diazirine addition reactions to C₆₀, followed by an HCl elimination step, yielded [6,6] and [5,6] sp² carbon bridged fullerenes. The [5,6] adducts (alkylidenehomofullerenes) are the first examples of fullerene structures where the ad

Full text of DOI:10.1039/b920339g

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Two new types of p-conjugation between a fullerene sphere
and an addendw
Floris B. Kooistra, Tessa M. Leuning, Enrique Maroto Martinez and Jan C. Hummelen*
Received (in Cambridge, UK) 29th September 2009, Accepted 8th February 2010
First published as an Advance Article on the web 16th February 2010
DOI: 10.1039/b920339g
Diazirine addition reactions to C60, followed by an HCl
elimination step, yielded [6,6] and [5,6] sp2 carbon bridged
fullerenes. The [5,6] adducts (alkylidenehomofullerenes) are
the first examples of fullerene structures where the addend is
attached to the fullerene cage in a true, albeit highly bent,
alternating single–double bond fashion.
push–pull systems is, normally, optimal. With fullerenes, this
has never been achieved for a good reason: it is impossible to
make a tangential type of linearly conjugated addend, unless a
substantial part of the fullerene cage is broken. The closest
possible approximation of an attachment point for a ‘linearly’
conjugated addend to an intact fullerene is an open [5,6] sp2
carbon bridged fullerene, as in alkylidenehomofullerene 7
(Scheme 1). Albeit close to orthogonal because the addend
cannot be tangentially attached, the p-system connecting the
fullerene with the addend in 7 is made up of alternating single
and double C–C bonds. In the isomeric stucture 8, an
alkylidenecyclopropafullerene, one sp3 carbon separates the
two p-systems, that can now interact in a way somewhat
similar to the periconjugation configuration, but with quite
different symmetry. The first method of synthesizing both
types of structures is reported below.
Since Kratschmer et al. first isolated C ,1 fullerenes have
¨
60
found many applications in molecular electronics. Their
unique electron accepting and electron conducting properties
have rendered them very useful for applications in organic
photovoltaic devices.2–5 Over the years many synthetic
methodologies have been developed to functionalize
fullerenes.6–9 The solubility and processability of fullerenes
can be perfectly tuned for the various applications in this way.
Altering the electronic properties of fullerenes by direct inter-
action with addends has been more challenging so far. Yet
such tunability is desired, since the ability to optimize the
electronic properties of molecular components is essential in
the field of molecular electronics. Wudl and co-workers found
a moderate interaction between the addend p-system and the
fullerene moiety in certain spiroannelated methanofullerenes.10
This ‘periconjugation’ results from the overlap between the
addend pz orbitals, which are aligned in a perpendicular way
to the nearby fullerene orbitals, in contrast to spiroconjuga-
tion, where the interacting orbitals are oriented in one plane.11
In the periconjugated construct there is an electronic overlap
between two p-systems, separated by two sp3 carbons. We
have shown that placing electron donating and withdrawing
groups on the phenyl ring of PCBM (phenyl-C61-butyric acid
methyl ester) results in modest variations of the fullerene
LUMO level, through an inductive through-bond effect.12
For all of the above mentioned approaches, in which there is
relatively weak electronic coupling, there is a strict limit for
the electron donating power of the addend, because when
energetically feasible, the compound becomes a donor–acceptor
dyad. In that case, the compound does not function as a
normal electron acceptor anymore, especially not under
illumination because intramolecular photoinduced electron
transfer will become a competing process.
We took diazirines as precursors for the fullerene addition
reaction. Diazirines decay under photolytic or thermolytic
conditions with expulsion of nitrogen. Two interesting inter-
mediate species are formed during this decay: a carbene species
and a diazo species. In order to study this decay pathway,
Nagase et al. used C60 as a mechanistic trap. Carbene species
react with C60 to form [6,6] adducts. Diazo intermediates, on
the other hand, form [5,6] adducts.13 When 3-chloro-3-
alkyldiazirines were used, both the [6,6] and the [5,6] adduct
were obtained, with the chlorine atom at the cyclopropane.14
The substituent on the diazirine ring determined the ratio of
[5,6] and [6,6] addition product.15 We envisioned 3-chloro-3-
phenyldiazirines as candidates for addition reactions to C60.
In order to obtain the strongest possible electronic coupling,
(linear) p-conjugation between two moieties leading to
Molecular Electronics, Zernike Institute for Advanced Materials &
Stratingh Institute for Chemistry, University of Groningen,
Nijenborgh 4, 9747 AG Groningen, The Netherlands.
E-mail: J.C.Hummelen@rug.nl
w Electronic supplementary information (ESI) available: UV/Vis
spectra, NMR data, IR data and mass spectrometry data. See DOI:
10.1039/b920339g
Scheme 1 Synthetic overview ((a) R = H, (b) R = NO2).
Chem. Commun., 2010, 46, 2097–2099 | 2097
ꢀc
This journal is The Royal Society of Chemistry 2010

View Article Online
Table 1 UV-Vis absorption data of the fullerene derivatives
Compound
Absorption/nm
Red shift/nm
11
5a
7a
5b
7b
6a
8a
6b
8b
537, 602 (sh)
548, 603 (sh), 664 (sh)
540, 603 (sh)
547
497
515
490
511
7
18
21
Subsequent HCl elimination would then lead to an exocyclic
CQC bond on the [5,6]-bridge or on the [6,6]-cyclopropane
ring, respectively (Scheme 1).
Fig. 1 Possible p orbital overlap of compounds 7a, 8a and 9, with
different orbital symmetries chosen for maximal overlap.
The diazirine approach allows for the presence of the
leaving group (here ClÀ), necessary for b-elimination.
Alternative precursors for diazo compounds, e.g. hydrazones,9
do not allow this. In the case of the [6,6] adduct, the exocyclic
CQC bond is positioned for a peri-type conjugation with the
fullerene p-system. In the case of the [5,6] adduct, the exocyclic
CQC bond is connected to the fullerene p-system in the
desired alternating single–double bond fashion. Substituting
the phenyl ring with electron donating or withdrawing groups
would further allow the determination of the electronic
coupling in this ‘bent conjugated’ system.
while in the case of 8a both p-systems are separated by only
one sp3 carbon atom. Besides being in closer proximity,
the p orbitals are oriented differently (by 901) in both
compounds.
The overlap of the alkene p orbitals and the fullerene
p orbitals in compounds 8a and 8b is stronger than the overlap
of the phenyl p orbitals and the fullerene p orbitals in
compound 9 (see Fig. 1). Furthermore, the orbital symmetry
of the fullerene part is quite different in 8 and 9. This likely
explains the observed differences between the two types of
periconjugated systems.
[5,6]-Adducts 7a and 7b and [6,6] adducts 8a and 8b were
synthesized via this diazirine route (see Scheme 1). The
diazirines 4a and 4b were prepared according to literature
procedures.16,17 The reaction with C60, performed at 60 1C,
was stopped after 50% conversion (C60 50%, [6,6] 35%, [5,6]
5%). The [5,6] and [6,6] adducts were separated by preparative
HPLC. Subsequently, each of the isomers was subjected to
HCl elimination. Acceptable results were obtained using
KOtBu in ODCB (see Scheme 1). Nevertheless, the yield of
the desired elimination products from [5,6] adducts 5a and 5b
was extremely low. Hence, while 8a, b could be fully
characterized, the [5,6] homofullerenes 7a and 7b could only
be investigated using LC-MS and UV-Vis spectroscopy. In all
cases, after performing the elimination step, MS traces showed
a mass 36 m/z lower than for their precursor fullerenes, clearly
indicating the elimination of HCl. The UV/Vis absorption
spectra are indicative of the extent of electronic coupling in the
elimination products. In the case of the [6,6] isomers, a red
shift of 18 nm (for 8a) and 21 nm (for 8b) was observed in the
500 nm region (see Table 1). The [5,6] isomers showed only a
slight red shift (11 nm and 7 nm for 7a and 7b, respectively) in
the 540 nm region. Furthermore a new absorption at 664 nm,
probably caused by a previously forbidden p–p* transition,
was observed for compound 7a. Interestingly, the optical
bandgap and the first reduction potential of 8a were found
to be very close to those of a normal methano[60]fullerene.18
Hence, the interaction between the fullerene and the addend in
8a appears to happen mainly in orbitals other than the HOMO
and LUMO.
In conclusion, two new types of p-conjugated fullerene–
addend linkages have been realized. The new type of
periconjugation, as present in the [6,6] adducts, is quite
effective. The observed red shift in the [5,6] homofullerenes
is smaller than the ones observed in the corresponding [6,6]
adducts. Hence, the strongly bent ‘linear’ p-conjugated system
in the [5,6] adducts (e.g. 7a in Fig. 1) functions as a slightly
weaker electronic coupler. Because the neighbouring p orbitals
on the bridge and the bridgehead sp2 carbon atoms are almost
orthogonal in the [5,6] adducts, it seems more adequate to
regard this system as periconjugated as well. The fullerene–
addend interaction might be improved using an addend with a
better energy match with respect to the fullerene core. This will
be the subject of future research.
The work of FBK was funded by the Zernike Institute for
Advanced Materials.
Notes and references
1 W. Kratschmer, L. D. Lamp, K. Fostiropoulos and
¨
D. R. Huffman, Nature, 1991, 347, 548.
2 G. Yu, J. Gao, J. C. Hummelen, F. Wudl and A. Heeger, Science,
1995, 270, 1789.
3 N. S. Sariciftci, L. Smilowitz, A. Heeger and F. Wudl, Science,
1992, 258, 1474.
4 C. J. Brabec, N. S. Sariciftci and J. C. Hummelen, Adv. Funct.
Mater., 2001, 11, 15.
5 T. D. Anthopoulos, C. Tanase, S. Setayesh, E. J. Meijer,
J. C. Hummelen, P. W. M. Blom and D. M. de Leeuw, Adv.
Mater., 2004, 16, 2174.
6 M. Maggini, G. Scorrano and M. Prato, J. Am. Chem. Soc., 1993,
115, 9798.
7 R. Gonzalez, J. C. Hummelen and F. Wudl, J. Org. Chem., 1995,
60, 2618.
8 C. Bingel, Chem. Ber., 1993, 126, 1957.
The observed shifts for the [6,6] adducts are attributed to a
rather effective periconjugation between both p-systems. The
effect is most likely due to the close proximity of both
p-systems. Note that in compound 9, synthesized by the Wudl
group, both p-systems are separated by two sp3 carbon atoms,
9 A. Hirsch and M. Brettreich, Fullerenes Chemistry and Reactions,
Wiley-VCH Verlag GmbH & Co. KgaA, Weinheim, 2005.
ꢀc
This journal is The Royal Society of Chemistry 2010
2098 | Chem. Commun., 2010, 46, 2097–2099

View Article Online
10 M. Eiermann, R. C. Haddon, B. Knight, Q. Chan Li, M. Maggini,
N. Martin, T. Ohno, M. Prato, T. Suzuki and F. Wudl, Angew.
Chem., Int. Ed. Engl., 1995, 34, 1591.
14 T. Wakahara, Y. Niino, T. Kato, Y. Maeda, T. Akasaka,
M. T. H. Liu, K. Kobayashi and S. J. Nagase, J. Am. Chem.
Soc., 2002, 124, 9465.
11 H. Durr and R. Gleiter, Angew. Chem., Int. Ed. Engl., 1978, 17,
559.
12 F. B. Kooistra, J. Knol, F. Kastenberg, L. M. Popescu,
W. J. H. Verhees, J. M. Kroon and J. C. Hummelen, Org. Lett.,
2007, 9, 551.
15 M. T. H. Liu, Y. H. Choe, M. Kimura, K. Kobayashi,
S. J. Nagase, T. Wakahara, Y. Niino, M. O. Ishitsuka,
Y. Maeda and T. Akasa, J. Org. Chem., 2003, 68, 7471.
16 M. T. H. Liu, N. H. Chishti, M. Tencer, H. Tomoka and Y. Izawa,
Tetrahedron, 1984, 40, 887.
13 T. Akasaka, M. T. H. Liu, Y. Niino, Y. Maeda, T. Wakahara,
M. Okamura, K. Kobayashi and S. J. Nagase, J. Am. Chem. Soc.,
2000, 122, 7134.
17 W. H. Graham, J. Am. Chem. Soc., 1965, 87, 4396.
18 PCBM: Eg = 1.73 eV; E1/21 = À1.078 V. 6a: E1/21 = À1.047 V. 8a:
Eg = 1.74 eV; E1/2 = À1.033 V (vs. Fe/Fe+).
1
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 2097–2099 | 2099