вход по аккаунту


Picotube Tetraanion A Novel Lithiated Tubular System.

код для вставкиСкачать
Stable Tetraanionic Carbon Tubes
Picotube Tetraanion: A Novel Lithiated Tubular
Noach Treitel, Markus Deichmann, Tamar Sternfeld,
Tuvia Sheradsky, Rainer Herges,* and
Mordecai Rabinovitz*
Carbon nanotubes have attracted considerable attention since
their discovery,[1] and have recently been the focus of
intensive investigations. Studies have revealed that nanotubes
exhibit extraordinary electronic properties.[2, 3] Nanotubes
have been shown to ignite upon exposure to short-range
flashes of light,[2] and recently ultrafast carbon-nanotubebased transistors have been shown to outperform their silicon
The electronic and magnetic properties of such tubes
might be studied with guest atoms present inside the tube,
such as in fullerenes.[4] However, standard nanotubes, as well
as fullerenes, are highly insoluble in most solvents, which
makes such studies rather intricate. Additionally, nanotubes
are end-capped, and the insertion of atoms or cations into the
tube is very difficult. The conjugated tubelike polycyclic
aromatic hydrocarbon (PAH) 5,24:6,11:12,17:18,23-tetra[1,2]benzenotetrabenzo[a,e,i,m]cyclohexadecene (1), dubbed
“picotube”,[5] has successfully been synthesized in gram
amounts through the dimerization metathesis of tetradihydrodianthracene.[6a]
Picotube 1 possesses open ends and is more soluble than
standard nanotubes. In addition, its nanotube-like structure
and fullerene-like ability to include endohedral atoms makes
it an ideal candidate for further study. The picotube could
play a role as a host molecule, by encapsulating one or more
counterions upon charging, while possibly coordinating with
others on the exterior. The specific orientation of such cations
could also reveal information regarding the charge distribution within the molecule.
[*] Prof. Dr. R. Herges, M. Deichmann
Institut fr Organische Chemie
Universitt Kiel
Otto-Hahn Platz 4, 24098 Kiel (Germany)
Fax: (+ 49) 431-8801-558
Prof. M. Rabinovitz, N. Treitel, T. Sternfeld, Prof. T. Sheradsky
Department of Organic Chemistry and
The Lise Meitner Minerva Center for Computational Chemistry
Safra Campus, The Hebrew University of Jerusalem
Givat Ram, Jerusalem 91904 (Israel)
Fax: (+ 972) 2-652-7547
[**] Financial support from the US-Israel Binational Science Foundation
(BSF) and from the Lise Meitner-Minerva Center for Computational
Quantum Chemistry is gratefully acknowledged. We thank Prof.
Silvio Biali for insight and fruitful discussions.
Supporting information for this article is available on the WWW
under or from the author.
2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4210-1172 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2003, 42, No. 10
Picotube 1 has been fully characterized by NMR spectroscopy, mass spectrometry, density functional theory (DFT)
calculations, matrix IR spectroscopy and crystal structure
analysis.[6a–c] The latter reveals that the picotube, which
consists of four anthracene units, has a diameter of 5.4 9
and a length of 8.2 9, and shows D4h symmetry for the neutral
species 1.[6c] However, DFT calculations and low-temperature
IR studies indicate that the D4h symmetry is only a timeaveraged structure of two rapidly interconverting D2d isomers
(Figure 1), obtained by twisting of the quinoid double
Figure 1. Rapid interconversion of neutral 1 (D2d), to yield a D4h timeaveraged structure.
bonds.[7] This lowering of symmetry has been calculated to
provide an energy gain of about 4.5 kcal mol1, relative to D4h
H and 13C NMR spectroscopy[6a] show that picotube 1
indeed possesses D4h symmetry on the NMR timescale, with
only one AA’BB’ pattern in the proton spectrum and four
resonances in the carbon spectrum (Figure 2 a). Previously,[6a]
the freezing of conformational movement could not be
observed down to 203 K, and in the present study, no splitting
of the signals was seen, even at 150 K, which indicates that the
barrier for topomerization of the neutral molecule is lower
than 7.8 kcal mol1,[8] in good agreement with calculated
In view of the interest in introducing alkali metals into
nanotubes, we report that the picotube undergoes slow
chemical reduction by lithium metal to form an unusually
stable charged species.[9] The 1H NMR spectrum of the lightyellow neutral species tends to broaden as the solution
becomes olive green. The spectrum loses its form as the
solution further changes color, first to dark green, and then to
dark blue, most likely as a result of the formation of
paramagnetic species. Finally a dark-brown solution is
obtained which gives a different 1H NMR spectrum (Figure 2 b) to that of the neutral species. Since the olive-green
solution must contain a species that has, at least, been reduced
to a radical anion, and the dark-blue solution contains a
dianionic species, then the diamagnetic dark-brown solution
must contain a species which can be no less reduced than a
tetraanion. Charge-density calculations from 13C NMR spectroscopy show a value for Kc[10] of approximately 131 ppm per
electron, in accordance with the formation of a quadruply
charged species.[11] DFT calculations using gauge-including
atomic orbitals (GIAO) [12] also reinforce the assessment that
the reduced species is a tetraanion (see below).
No diamagnetic species were observed until the formation
of the tetraanion, and further attempts at reduction failed to
yield new products. The dianion is, therefore, likely to be an
easily reduced triplet diradical or unstable singlet. The
tetraanion was fully characterized by NMR spectroscopy,
and charge-density calculations support its nature. According
to the 1H NMR spectrum, quenching of the tetraanion with
oxygen leads to the near quantitative and immediate recovery
Figure 2. 1H and 13C NMR (400 and 100 MHz, respectively, [D8]THF, 298 and 220 K for the neutral and charged species, respectively) of a) the
neutral picotube and b) the charged species.
Angew. Chem. Int. Ed. 2003, 42, No. 10
2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4210-1173 $ 20.00+.50/0
of the neutral compound, which indicates that no bonds are
formed or cleaved during the reduction process.
In contrast to the neutral compound, the tetraanion gives
more complex spectra, with eight proton signals and fourteen
carbon lines observed (Figure 2 b). The picotube is thus
reduced to a species with lower symmetry (C2v, C4, or D2). The
proton signals are found in two groups, and can be shown[13a]
to be representative of one system, the anthracenic unit. Each
hydrogen and carbon atom of any given anthracene unit is
different, and gives a characteristic resonance; bond cleavage
would result in a different number of proton and carbon
signals. The very fast dynamic motion, which is only
detectable by matrix IR spectroscopy in the neutral compound, slows down dramatically in the charged species. The
tetraanion shows no coalescence of signals, even at 325 K.
This result means that the barrier of the “D2dQ[D4h]°QD2d”[14]
dynamic process (Figure 1) increases considerably to give a
lower limit for DG°
for the
325 of approximately 19.3 kcal mol
No decomposition of the tetraanion is seen, even after
standing at room temperature for one week. Furthermore,
after heating the tetraanion to 325 K, the NMR signals only
become slightly broadened, and only disappear when the
temperature is further raised to that of boiling THF; upon
cooling the solution, the signals return to their original
intensity. Only a few PAH tetraanions without stabilizing
substituents that are stable in boiling THF are currently
The tetraanion exhibits a significant charge alternation[16]
encircling the 9,10 positions of the anthracenylidene units.
While the carbon atom in the 9 position shows a 13C NMR
resonance at d = 85.0 ppm, position 10 gives rise to a signal at
d = 143.3 ppm. This difference between the two carbon atoms,
which have identical chemical shifts in the neutral compound,
is significant and indicates a highly irregular charge distribution within the anthracenylidene units (Figure 3 a), which is
contrary to the behavior of anthracene upon reduction[17]
(Figure 3 b). The carbon atom in the 9 position bears a partial
negative charge (ca. 0.4), while that in the 10 position
carries a positive charge of + 0.06. The question as to whether
the anthracene units arrange the charge alternation in a
neutral-neutral-negative-negative or a neutral-negative-neutral-negative fashion was answered by 2D NMR methods;[13b]
the preferred pattern is charge alternation (Figure 4).
Figure 3. Charge distribution within a) the anthracenylidene units of 1
and b) the anthracene dianion[17] , as calculated from 13C NMR chemical shifts. Filled circles represent negative charge; hollow circles represent positive charge.
2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 4. Charge alternation throughout the tetraanion.
In agreement with similar structures investigated by
Schleyer and Pople,[18a] the NMR data strongly suggest that
two of the Li atoms are coordinated to two of the quinoid
“double bonds”[19] and bridge the sp2 carbon atoms. The
remaining two olefin units are uncoordinated according to the
experimental data (charge alternation). Because of the small
size of the cavity, coordination is only possible from outside
the tube. The remaining two Li atoms are located so as to
interact with the aromatic systems, either outside or inside the
tube. A systematic DFT study[12] of numerous conceivable
geometries was performed to find the optimum location of the
other two Li atoms, which led to a global minimum of
C2 symmetry, with two Li atoms positioned inside the tube,
separated by 2.61 9. The endohedral Li atoms coordinate to
the double bonds that are already coordinated by Li+ ions
from outside the tube, to form doubly bridged ethylene units
that are analogous to the 1,2-dilithioethane structure investigated by Schleyer, Pople, et al.[18a] A similar structure has
been observed in a stilbene–dilithium complex.[18c] An even
stronger interaction (CLi 2.14 9) was found with the ipso
carbon atom of a neighboring benzene ring. Our theoretical
finding that two of the Li atoms are bound inside the tube is
further supported by an X-ray structural analysis of the
picotube–Ag+ complex (1:2 stoichiometry), in which two
Ag+ ions occupy similar positions inside the tube[6c] as were
found for Li+ ions in our calculations. Structures investigated
with four Li atoms either all outside or inside the tube were all
found to be considerably less stable than the above C2
structure. According to the calculations, the C2 structure
undergoes a fast racemization to give a time-averaged D2
The proposed structure is corroborated by a comparison
of the theoretically calculated and experimental 13C NMR
spectra of the time-averaged D2 structure (Figure 5). The
calculations[12] confirm the large variance between the carbon
atoms in the 9 and 10 positions, and predict an even larger
difference in Dd than found by experimentation. All other
shifts agree within the error limit of the method (Table 1). The
large deviation between theory and experiment of almost
25 ppm for the anionic centers (position 9) can be accounted
for by solvation of the coordinated Li ions. While the
Li atoms inside the tube are shielded from the solvent, the
1433-7851/03/4210-1174 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2003, 42, No. 10
Table 1: Calculated (B3LYP/6-31G*, GIAO) and experimental 13C NMR shifts of the tetraanion.
Carbon No.
Experimental[a,b] 118.1(11) 123.7(7) 106.6(13) 123.0(8) 128.4(6) 107.7(12) 121.8(9) 121.6(10) 85.0(14) 143.3(1) 139.4(3) 129.4(5) 131.1(4) 141.9(2)
111.3(12) 120.9(9) 110.6(13) 125.8(6) 121.3(7) 112.5(11) 121.1(8) 119.4(10) 60.3(14) 140.2(1) 133.0(2) 126.8(5) 129.5(4) 131.0(3)
Deviation [%][f ]
Rank Deviation[g]
[a] In ppm, relative to TMS. [b] In [D8]THF, at 220 K. [c] Rank order for experimental and calculated values: 1–14 = lowest-to-highest field. [d] B3LYP/6-31G*
optimized GIAO single-point calculation. [e] Averaged values for two independent anthracene units (see Supporting Information). [f] Deviation between
experimental and calculated values. [g] Difference in experimental and calculated ranks.
counterions. Carbanions with Li+ counterions are anything
but free anions.[22] The CLi bond may have a partial covalent
character. 7Li NMR indicates two different lithium cations,
with a sharp signal at d = 3.6 ppm and a broader signal at
d = 0.2 ppm (at 220 K). The latter signal may be attributed to
a lithium cation that is externally coordinated to the picotube,
while the former attests to another type, located within the
tube;[23] support for this type of structure can be gained from
To summarize, a host molecule in a stable tetraanionic
state has been prepared that can host two Li+ cations within
the inner domain of a “tube-like” system resembling the
edges of an open nanotube; this anion exhibits a lower
symmetry than the neutral molecule. The change from a
D2dQ[D4h]°QD2d dynamic process in the neutral system to a
C2Q[D2]°QC2 equilibrium in the tetraanion is mainly caused
by coordination of Li atoms[24] to the inner “walls” of the tube.
Received: October 1, 2002 [Z50277]
Figure 5. Computer-generated optimized structure of the picotube /
4 Li+ complex at the B3LYP/6-31G* level, with black spheres representing the Li+ cations. All LiC distances shorter than 2.4 J are indicated
as dashed lines.
external Li atoms are strongly coordinated to THF. To
estimate the solvent effect on the 13C shifts of these carbon
atoms, we performed calculations on the water-solvated (as a
model for THF) system (fully optimized within C2 symmetry).
With one solvent molecule at each Li atom, the deviation is
reduced to 21 ppm. Further solvation, or even formation of a
solvent-separated ion pair thus accounts for the variation
between theory and experiment.[20]
Consequently, it is most probable that a D2 time-averaged
structure is observed,[21] with a fast C2Q[D2]°QC2 equilibrium
existing in solution. However, this process was not observed
by NMR spectroscopy as the motion was apparently too
rapid; no splitting of the signals was observed, even at 170 K.
The conformational change could arise by small movements
of two opposing anthracene units with respect to each other,
rather than a concerted movement of all four units in the
“D2dQ[D4h]°QD2d” equilibrium.
The reduction in symmetry during the charging process is
most probably a result of charge localization by the Li+
Angew. Chem. Int. Ed. 2003, 42, No. 10
Keywords: carbon · density functional calculations · lithium ·
nanotubes · reduction
[1] S. Ijima, Nature 1991, 354, 56.
[2] P. M. Ajayan, M. Terrones, A. de la Guardia, V. Huc, N. Grobert,
B. Q. Wei, H. Lezec, G. Ramanath, T. W. Ebbesen, Science 2002,
296, 705.
[3] S. J. Wind, J. Appenzeller, R. Martel, V. Derycke, P. Avouris,
Appl. Phys. Lett. 2002, 80, 3817.
[4] a) M. Saunders, H. A. Jimenez-Vazquez, R. J. Cross, S. Mroczkowski, D. I. Freedberg, F. A. L. Anet, Nature 1994, 367, 256;
b) E. Shabtai, A. Weitz, R. C. Haddon, R. E. Hoffman, M.
Rabinovitz, A. Khong, R. J. Cross, M. Saunders, P.-C. Cheng,
L. T. Scott, J. Am. Chem. Soc. 1998, 120, 6389.
[5] In analogy to the larger and fully conjugated nanotubes, and for
[6] a) S. Kammermeier, P. G. Jones, R. Herges, Angew. Chem. 1996,
108, 2834; Angew. Chem. Int. Ed. Engl. 1996, 35, 2669; b) R.
Herges, M. Deichmann, J. Grunenberg, G. Bucher, Chem. Phys.
Lett. 2000, 327, 149; c) S. Kammermeier, P. G. Jones, I. Dix, R.
Herges, Acta Crystallogr. Sect. C 1998, 54, 1078.
[7] Despite torsion of the double bonds in the neutral compound,
the electronic structure is mainly quinoid, rather than benzoid,
and is slightly antiaromatic with 16 p electrons in the periphery.
This is confirmed by the ACID method: R. Herges, A.
Papafilippopoulos, Angew. Chem. 2001, 113, 4809; Angew.
Chem. Int. Ed. 2001, 40, 4671; R. Herges, D. Geuenich, J. Phys.
Chem. A 2001, 105, 3214.
2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4210-1175 $ 20.00+.50/0
[8] Estimated
pffiffiffi with DG = 4.57 T[10.32log(k/T)], where k =
(pDu)/ 2, and using a reasonable minimum proton resonance
difference, Dumin = 5 Hz, in the neutral state.
[9] The anions were prepared in the following manner: Lithium was
directly inserted as a freshly produced wire into the upper part of
an extended NMR tube containing the PAH (2–4 mg). The tube
was first filled with argon, attached to a vacuum line, and flamedried under vacuum. Approximately 0.5 mL of [D8]THF (yielding 6–11 mm solutions), dried over a sodium/potassium alloy
under high vacuum, was vacuum transferred from a reservoir
into the tube. The sample was degassed under vacuum using the
freeze-pump-thaw technique and flame sealed. The solution was
brought into contact with the lithium wire by turning the tube
upside down and the tube was repeatedly inverted at 78 8C.
H NMR spectroscopy detected the formation of the anions.
[10] This parameter has been empirically shown to correlate between
the total change in chemical shift of the 13C NMR spectrum upon
charging, SDd, and the total extra charge added to the p system
of the molecule upon charging, SDqp, as expressed in the
equation: SDd = KcSDqp. In our case, SDd = 524 ppm. For a
review, see: a) S. Braun, H.-O. Kalinowski, S. Berger, 150 and
More Basic NMR Experiments: A Practical Course, VCH-Wiley,
Weinheim, 1998; b) R. Schaefer, W. G. Schneider, Can. J. Chem.
1963, 41, 966; c) H. Spiesecke, W. G. Schneider, Tetrahedron
Lett. 1961, 468; d) P. C. Lauterbur, J. Am. Chem. Soc. 1961, 83,
1838; e) P. C. Lauterbur, Tetrahedron Lett. 1961, 274.
[11] a) A. Ayalon, M. Rabinovitz, P.-C. Cheng, L. T. Scott, Angew.
Chem. 1992, 104, 1691; Angew. Chem. Int. Ed. Engl. 1992, 31,
1636; b) B. Eliasson, U. Edlund, K. MLllen, J. Chem. Soc. Perkin
Trans. 2 1986, 937.
[12] All structures were optimized at the B3LYP/6-31G* level of
DFT theory within the corresponding point group, and confirmed by harmonic frequency analysis to be either minima or
transition states. NMR shifts were calculated using the GIAO
method. An extensive investigation of the energy hypersurface
using the semiempirical PM3 method preceded the DFT
[13] a) Using standard 2D NMR techniques, that is, COSY, NOESY,
HSQCSI, and HMBC. b) Clear nOe signals pertaining to carbon
pairs 4 and 5, as well as 1 and 8, dismissed the possibility that the
alternation is in a neutral-negative-neutral-negative fashion. A
structure of two neutral anthracene units alternating with two
anthracene dianion units is also possible, which gives the same
2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
symmetry. However, DFT calculations confirm the proposed
structure, rather than this possibility.
Quotation marks indicate that the dynamic process still exists in
the tetraanion, although it is too slow to observe.
Deduced from the Eyring equation, DG° = 4.57 T[10.32log(k/
T)]; k = 0.7 s (from T1), from NOESY spectrum at 325 K
(maximum working temperature for THF). See, for example:
J. SandstrOm, Dynamic NMR Spectroscopy, Academic Press,
London, 1982, p. 96.
a) J. Klein, Tetrahedron 1983, 39, 2733; b) Y. Cohen, J. Klein, M.
Rabinovitz, J. Am. Chem. Soc. 1988, 110, 4634.
R. E. Hoffman, N. Treitel, E. Shabtai, R. Benshafrut, M.
Rabinovitz, J. Chem. Soc. Perkin Trans. 2 2000, 1007.
a) A. J. Kos, E. D. Jemmis, P. von R. Schleyer, R. Gleiter, U.
Fischbach, J. A. Pople, J. Am. Chem. Soc. 1981, 103, 4996; For
further olefinic and aromatic lithium complexes, see for example: b) D. Scheschkewitz, M. Menzel, M. Hofmann, P. von R.
Schleyer, G. Geiseler, W. Massa, K. Harms, A. Berndt, Angew.
Chem. 1999, 111, 3116; Angew. Chem. Int. Ed. 1999, 38, 2936;
c) R. Benken, W. Andres, H. GLnther, Angew. Chem. 1988, 100,
1212; Angew. Chem. Int. Ed. Engl. 1988, 27, 1182; d) M.
Walczak, G. Stucky, J. Am. Chem. Soc. 1976, 98, 5531; e) Ab
Initio Molecular Orbital Theory, (Eds.: W. J. Hehre, L. Radom,
P. von R. Schleyer, J. A. Pople), Wiley-Interscience, New York,
1986, p. 450.
In reference to the original double bonds in the neutral
Model calculations of the D2h symmetric dilithio ethylene group
at the same level of theory for both the unsolvated and solvated
forms (with 4 H2O ligands, D2h) revealed an upfield shift of the
sp2 carbon atoms of as much as 45.6 ppm.
C2 symmetry would give rise to 16 proton and 28 carbon
M. Szwarc, Carbanions, living polymers, and electron transfer
processes, Wiley, New York, 1968.
An extreme case in which two types of Li+ cations appear is
reported in: A. Ayalon, A. Sygula, P.-C. Cheng, M. Rabinovitz,
P. W. Rabideau, L. T. Scott, Science 1994, 265, 1065.
The picotube also undergoes reduction with a potassium metal
mirror. However, preliminary results show that potassium
apparently leads to bond cleavage, to yield a species with
different symmetry than that of the [picotube]4/4 Li+ complex.
1433-7851/03/4210-1176 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2003, 42, No. 10
Без категории
Размер файла
143 Кб
picotube, lithiated, tubular, tetraanion, system, novem
Пожаловаться на содержимое документа