Tải bản đầy đủ - 0trang
3 Polyhydrofullerenes C(60)H(n) and C(70)H(n) (n = 14–60)
pressure chemical ionization (APCI), chemical ionization (CI) or electron impact
(EI) . After some debate , the early assumption, that C60H36 is the major
product of the Birch–Hückel reduction could be confirmed by methods mentioned
above [47, 48]. For the outstanding stability of C60H36 over all the other possible
polyfullerenes two main explanations were given . Firstly, 36 is the number of
hydrogens required to leave a single unconjugated double bond on each pentagon
of C60 and Birch–Hückel reduction should not be able to reduce single double
bonds. Secondly, the increasing number of hydrogens attached to C60 leads to
decreasing bond angle strain, resulting from sp2 to sp3 hybridization, and at the
same time increasing strain due to hydrogen–hydrogen repulsion. Combined strain
reaches a minimum when n = 36. The first explanation proved not to be very
important, since most experimentally and theoretically found stable isomers do
not show isolated double bonds and, for example, the Th isomer with only isolated
double bonds is a high-energy-isomer relative to other possible isomers (Figure 5.10
below, Section 5.3.5). Instead a criterion of stability seems to be the formation of
benzene-like hexagons with conjugated double bonds (Section 5.3.5).
About 6 · 1014 different possible isomers of C60H36 were estimated . This
number, the similarity between this isomers and the instability of hydrofullerenes
make it almost impossible to get the exact structure of the isomers that are included
in the complex mixture of C60H36 isomers obtained by Birch reduction . Based
on 1H and 3He NMR spectroscopy and on calculations (Section 5.3.5) C1-, C3-, D3d-,
S6-isomers and a T-isomer are supposed to be among the preferred isomers [7, 48,
51–53]. Some examples of probable structures are given in Figure 5.10 below
Reduction with Zn/HCl
Contrary to the reduction of C60 with Zn/Cu couple (Section 5.2.2) or with Zn and
half-concentrated HCl  that leads only to oligohydrofullerenes, the reaction with
Zn and concentrated HCl in toluene or benzene solution leads in a smooth and
fast reaction mainly to C60H36 [7, 55, 56]. The reaction takes place within 1 h and at
room temperature and only a small amount of side products were found. EI-mass
spectroscopy, run immediately after the reaction, carried out under a nitrogen
atmosphere gave very good spectra of the product almost without fragmentation
[55, 56]. These spectra revealed that 75% of the product was C60H36. The remaining
25% were assigned to C60H38 and C60H40. The major product C60H36 was first
proposed to be the T-isomer (Figure 5.10, Section 5.3.5) . Later, based on
vibrational spectroscopy, a S6-symmetry was suggested . Proof of the structure
is not yet given but different isomers are formed than with the Birch–Hückel
Under exclusion of light and air C60H36 has good stability towards high temperature . Nevertheless, extended heat treatment of the polyhydrofullerene mixture
leads to an increasing peak of C60H18. Indeed, C60H18 formation can predominate
if the reaction is carried out at high temperature and pressure .
5.3 Polyhydrofullerenes C60Hn and C70Hn (n = 14–60)
Since deuterochloric acid DCl is easily available, the reduction with Zn and conc.
DCl provides an easy access to deuterated fullerenes . Reducing C60 under the
same, already described conditions yields polydeuterofullerenes in good yields and
in short time. Interestingly, the product distribution is different from the HCl
reduction. The C–D bond is more stable than the C–H bond. This is probably why
deuteration yields not only C60D36 but also C60D38 and, in slightly smaller percentage,
also C60D40–44 as major products.
C70 is less reactive in terms of hydrogenation than C60. Zn–HCl reduction was
complete after 1.5 h and gave a pale yellow product, consisting of C70H36 and C70H38
as major products and C70H40–44 as further products . Deuteration of C70 shows
the same shift of the degree of hydrogenation to higher numbers. In this case
C70D42 is the dominant component.
Transfer Hydrogenation of C60 and C70
Hydrogenation to polyhydrofullerenes is also possible by transfer hydrogenation
 using 9,10-dihydroanthracene as a hydrogen source [44, 51, 60, 61]. The
treatment of C60 in a sealed glass tube in a melt of 9,10-dihydroanthracene at 350 °C
under N2 leads to a color change of the dissolved fullerene in 30 min from brown
via rubin-red, orange and yellow to colorless (Scheme 5.3). The resulting product
also shows a base peak in the mass spectra (EI, FAB, FD) centered on 756 u,
corresponding to C60H36 . Extending the reaction time to 24 h recolorizes the
reaction mixture and the isolated reaction product shows a mass distribution,
determined by the same techniques as above, centered at 738 u, corresponding to
C60H18. Based on this observation, conditions were established that allow either
synthesis of nearly pure C60H36 or of nearly pure C60H18 [7, 44]. Impurities are
small amounts of other hydrofullerenes and anthracene. The anthracene impurities
can be removed by sublimation. Polydeuterated fullerenes C60Dn can be obtained
analogously using 9,10-dideuteroanthracene as a deuterium source. The reaction
temperature of the transfer hydrogenation can be lowered to 250 °C by addition of
7H-benzanthrene or 7,7′-dideutero-7H-benzanthrene as a catalyst . In this way,
higher degrees of hydrogenation (n up to 44) are obtained for C60.
The predominance C60H36 and C60H18 formation under the applied conditions
has been conclusively proven via 3He NMR [48, 52], 1H NMR and 13C NMR [51, 61,
Figure 5.6 MALDI-RETOF mass spectrum of C60H36 with matrix
5-methoxysalicylic acid and comatrix NaBF4 (1 : 10 : 10). C60H36
can be detected as the C60H35+ ion at 755 u .
62] spectroscopy and with different mass spectrometric ionization techniques such
as EI , FD [44, 51] and MALDI [47, 63]. Especially with MALDI, by using a
combination of two matrix molecules , it was possible to obtain a mass spectrum
consisting almost entirely of C60H35+, which can be taken as direct evidence for the
exclusive production of C60H36 (Figure 5.6).
The clear solutions of the polyhydrofullerenes in various organic solvents become
inhomogeneous upon the formation of a precipitate. This, together with the broad
peaks in the 1H NMR spectra, shows the instability of these C60Hns. Thermal
treatment of C60Hn in the solid state at 550 °C leads to a complete reversion to C60
(Scheme 5.3). Sublimation of C60H18 and C60H36 at lower temperatures (273–412 °C)
was accompanied by partial loss of hydrogen. Decomposition of C60H36 was
confirmed to be a stepwise process with formation of C60H18 as an intermediate
Compared with the Birch–Hückel reduction of C60, transfer hydrogenation
produces less isomers. As well as mainly C60H36 and C60H18, in good yields, a
smaller number of different structural isomers are also formed. Nevertheless, due
to the huge number of possible isomers, thermal decomposition and, probably,
isomerization and the instability against air and light, identification turned out to
be almost as complicated as with the Birch reduction product mixtures. For C60H36,
structures with the symmetries T [52, 61, 65, 66], D3d [57, 66], S6 [57, 66], C3 [51, 52,
61] and C1 [51, 61] were claimed. Another reason for such various possible structures
for C60H36 may be the dependence of the product distribution on the reaction
conditions . Surprisingly, and almost independent of all these factors, only one
major isomer of C60H18 was found. The structure was elucidated by Taylor and coworkers , who showed with 1H NMR spectroscopy that C60H18 is a crown-shaped
molecule with C3v symmetry that may be considered as a substructure of T-C60H36
(Figure 5.7). All hydrogens are located on one side of the C60-ball, surrounding an
isolated benzene ring and leaving an extended conjugated system on the opposite
In the 3He NMR spectra of a mixture of 3He@C60H36 and 3He@C60H18 and also
of a nearly pure 3He@C60H18 sample prepared via the “Rüchardt”-procedure only
one isomer of 3He@C60H18 with a typical shift of −16.45 ppm was found .
After 10 years of attempts to elucidate the structure of the C60H36 isomers formed
during hydrogenation with dihydroanthracene the work of Billups and co-workers
5.3 Polyhydrofullerenes C60Hn and C70Hn (n = 14–60)
Figure 5.7 (A) Schlegel diagram of C3v-C60H18 (two different views),
which is a substructure of (B) T-C60H36.
lead to a breakthrough . Using HPLC for purification instead of sublimation
and combined 1H, 13C and 3He NMR spectroscopy showed clearly that only two
major isomers are formed and these isomers must have C3 and C1 symmetry
(Figure 5.8). The two major isomers are formed in a ratio of C1 to C3 of about 3 : 1.
This structure elucidation was made on the basis of 80 signals (32 sp2 and 48 sp3
signals) in the 13C NMR spectrum and of only two distinct but very close 3He
signals. The exact structure of the C1- and C3 isomers could not be proven but it
was suggested that they have the structures shown in Figure 5.8. Based on this
work, Gakh and co-workers, using 2D 1H NMR spectroscopy recorded at 800 MHz,
demonstrated that these suggestions were right . Moreover they found a further
minor isomer with T-symmetry (Figure 5.8), which was already calculated to be the
most stable isomer (Section 5.3.5).
Evidently the two major isomers have very similar structures, as shown in the
almost identical 3He NMR shifts of −8.014 and −8.139 ppm. This also shows
impressively how effectively 3He NMR spectroscopy can be utilized to distinguish
different C60 isomers, even if their structures are very similar.
Also, the less reactive C70 can be hydrogenated and deuterated in this way .
Interestingly, the base peak in the EI mass spectra of C70Hn at 876 u shows that, in
this case, a polyhydrofullerene with 36 hydrogens also exhibits an enhanced stability.
Beside the predominant C70H36 substantial amounts of C70H38, C70H40, C70H42,
C70H44 and C70H46 are formed and could be detected by field desorption (FD) mass
Figure 5.8 Structures of the major C60H36 isomers formed during transfer
hydrogenation of C60. Bold lines represent double bonds; bold hexagons
represent a hexagon with three conjugated double bonds.
Reduction with Molecular Hydrogen
A radical-induced hydrogenation of C60 and C70 can be carried out with iodoethane
as the hydrogen radical promoter [67–69]. In this method the fullerenes are placed
in a glass vessel inside an autoclave with an excess of iodoethane and are pressurized
with hydrogen to 6.9 MPa. Hydrogenation is carried out at 400 °C for 1 h. The
polyhydrofullerenes are obtained as a light brown solid. In the absence of iodoethane,
no hydrogenation of the fullerenes takes place. In contrast to the material obtained
by the Birch–Hückel reduction (Section 5.3.1) and transfer hydrogenation (Section
5.3.3) these polyhydrofullerenes are insoluble in many organic solvents and are
only slightly soluble in nitrobenzene. Analysis by fast atom bombardment (FAB)
mass spectrometry revealed a mixture of hydrofullerenes consisting mainly of
C60H36 and C70H36. Reactions at higher temperatures and pressures result in a
lower degree of hydrogenation.
Catalytic hydrogenation of C60 is also possible on activated carbon with Ru as
catalyst in refluxing toluene [70, 71]. Thereby, comparatively high degrees of
hydrogenation (up to C60H50) are obtained. The degree of hydrogenation of C60
increases with increasing hydrogen gas pressure and by elevating the reaction
temperature. A complete reversion to C60 takes place upon the treatment of C60Hn
with DDQ in refluxing toluene. C70 has also been catalytically hydrogenated in this
way, leading to mixtures of C70-polyhydrofullerenes that consist mainly of C70H36.
Dehydrogenation of C70Hn to C70 with DDQ proceeds quantitatively.
Hydrogenation of C60 on alumina-supported nickel leads selectively to C60H36
. Degradation products or other hydrofullerenes such as C60H18 or C60H44 were
not observed. The reduction was carried out in toluene in an autoclave at 50–250 °C
with the hydrogen pressure ranging from 2.5 to 7.5 MPa for 1 to 24 h.
More insight into the activity and selectivity of different catalytically active metals
was gained by a systematic investigation of this reaction . Various metal and
noble metal catalysts are usable for this hydrogenation. All metals were used on
alumina support. Reactions with Ru, Rh and Ir as catalysts give mainly C60H18,
while Pd, Pt, Co and Ni lead predominantly to C60H36, and Au and Fe have very
little activity for C60 hydrogenation. The %-d character of the transitions metals is,
seemingly, responsible for the selectivity, with Ru, Rh and Ir having a larger %-d
character than Pd, Pt, Co and Ni. The best selectivity for C60H36 is with the Ni/Al2O3catalyst .
The advantage of the above-described catalytic procedures is the possibility of
preparing large amounts of hydrofullerenes that can be synthesized selectively
(depending on the metal). Compared with other hydrogenations, the products can
easily be separated by filtration of the catalyst and evaporation of the solvent .
In the absence of a catalyst, hydrogenation can also take place but either very
high pressure or temperature is required to yield any products. High pressure
hydrogenation has been performed at hydrogen or deuterium pressures of 3.0 GPa
and 650–700 K . The major product under these conditions is C60H36. High
temperature but low pressure was used in a novel method, the chemical vapor
5.3 Polyhydrofullerenes C60Hn and C70Hn (n = 14–60)
modification (CVM) technique . C60 is sublimed by exposure to a tungsten
filament, which has a temperature of 1900 K. The substrate temperature at these
conditions is 960 K and a reaction time of 30 min is used. At a hydrogen pressure
of 20 Torr mainly C60H18 is formed.
Several systematic calculations on various structures of the polyhydrofullerenes
C60Hn have been carried out at different theoretical levels [8, 16, 30–32, 35, 38, 51,
53, 76–88]. Since the number of the theoretically possible regioisomers of the several
C60Hn adducts is very high, not every single structure has been calculated. The
situation becomes even more complicated if not only isomers with externally but
also with hydrogens added internally to the C60 cage are considered [77, 79, 80, 89,
90]. However, based on the results available, some trends of the stability of different
polyhydrofullerenes, depending on the addition mode and the degree of hydrogenation of C60 and of C70, can be recognized.
As pointed out in Section 5.2.4, the addition of hydrogen to C60 leading to
hydrofullerenes C60Hn with a low degree of hydrogenation (n ≤ 12) is energetically
favored if it proceeds in a 1,2-mode to the [6,6] bonds of the cyclohexatriene units
of the C60-framework [35, 38]. This mode avoids the energetically unfavorable
introduction of [5,6] double bonds. Conversely, the eclipsing interaction of the
hydrogens resulting from one 1,2-addition costs about 3–5 kcal mol−1. Thus, the
strain energy, due to eclipsing interactions of the hydrogens, is expected to become
ever more important upon increasing the degree of hydrogenation. Therefore, the
polyhydrofullerenes will eventually be unstable. Indeed, calculations of several C60Hn
isomers formed by a 1,2- or a 1,4-addition mode with the semiempirical MNDO
method would predict 1,4-additions to be more favorable than 1,2-additions starting
for n > 12 (Table 5.5) . But, only up to n = 24, exclusive 1,4-additions are possible.
Table 5.5 Calculated MNDO heats of formation of C60Hn formed by 1,2- and 1,4-addition .
ΔHf° 1,2-addition (kcal mol−1)
ΔHf° 1,4-addition (kcal mol−1)
n = 0 corresponds to C60.
Figure 5.9 Schlegel diagram of Th-C60H24 (dots represent C–H units).
Higher degrees of hydrogenation require formal 1,2-additions. The polyhydrofullerene formed by the complete and exclusive 1,4-addition mode is Th-C60H24
To obtain a measure for the dependence of n (number of hydrogens in C60Hn) on
the stability of C60Hn, the normalized  calculated heats of formation ΔHf° were
Δ( ΔHf°)n = [ΔHf°(C60Hn) − ΔHf°(C60H2)]/(n/2)
Whereas, for 1,2-additions, a destabilization begins for n > 12 with Δ( ΔHf°)
becoming endothermic, the opposite is observed for 1,4-additions, where the values
become more exothermic. Even for larger values (n up to 24) the normalized values
are more exothermic than for n = 2 in this addition mode . The large destabilization for C60H60 (n = 60) of Δ( ΔHf°) = 20.8 kcal mol−1 shows, significantly, the
accumulating influence on continuing 1,2-additions to C60.
The polyhydrofullerenes C60H36 and C60H48 are the most stable molecules in the
C60H12n (n = 1–5) series of exo-hydrogenated C60, according to ab initio calculations
at the Hartree–Fock level . For some specific isomers of this series the C–H
bond energies were calculated with an ab initio molecular orbital theory .
Although the employed hydrogenated isomers may not be the most stable isomers,
the averaged C–H bond energy seems to be maximized in C60H36 (Table 5.6).
Table 5.6 C–H bond energies for some isomers of the series C60H12n with n = 1–5 .
C–H bond energy (eV)
5.3 Polyhydrofullerenes C60Hn and C70Hn (n = 14–60)
Extensive AM1 calculations and density functional calculations were performed
for numerous isomers of C60Hn, with n varying from 2 to 60, by Clare and Kepert
[32, 53, 82–85]. They used AM1 for the geometry optimization and either AM1 or
the density functional method B3LYP for single point calculations of the energies.
Among numerous calculated structures, which compared with the existing number
of isomers is still minute, a couple of isomers turn out to have salient stability.
Among other isomers, these are the hydrofullerenes C60H18, C60H36 and C60H48.
Comparing calculations of ΔHf° carried out either with the AM1 method or density
functional methods reveals divergent results in some cases . In AM1 calculations
structures with the highest number of isolated double bonds were calculated as the
most stable structures. In contrast, calculations with density functional methods
result in low energies for structures with a high number of isolated benzenes such
The theoretical investigations support the experimental findings [42, 44, 49, 67,
70, 71] (see previous chapters) that (1) complete hydrogenation of C60 is difficult,
and has not yet been observed, and (2) even polyhydrogenated fullerenes with a
lower degree of hydrogenation, such as C60H36, are not stable and will slowly
decompose. The reasons for the instabilities of polyhydrofullerenes in terms of
simple topological arguments are inter alia (1) enhanced eclipsing interactions of
the H atoms as in the case of cis-1-additions in the 1,2-mode; (2) strain within the
C-network of polyhydrofullerenes, especially due to the deviation from the tetrahedral angle of sp3 C-atoms; and (3) introduction of [5,6] double bonds. Initially
generated reaction mixtures may rearrange to form more stable compounds. This
can cause a predominant occurrence of the species C60H36, C70H36 and C60H18
after thermal treatment of polyhydrofullerenes, which were obtained, for example,
from Birch–Hückel reductions or transfer-hydrogenations. The applied temperature
is very important for this annealing process. During the synthesis of C60H36 via
transferhydrogenation, an equilibrium between different isomers probably exists.
At 340 °C the mixture anneals to a mixture containing only the most stable isomers,
in this case a C1-, C3- and a T-isomer (Section 5.3.3). Higher temperatures (> 340 °C)
lead to the dehydrogenation product C60H18, which appears to be the final
hydrogenation product at elevated temperatures .
To date only one isomer of C60H18 with C3v-symmetry has been experimentally
proven (Section 5.3.3). Also in calculations this isomer appears as a particularly
stable compound [83, 84]. From calculations of the bond length of this crown shaped
isomer the central, isolated benzenoic ring can be considered essentially planar
aromatic . For C60H36, various structures have been suggested and extensively
treated theoretically [42, 51, 83, 85–88, 91, 92]. Some of these structures are shown
in Figure 5.10.
In structures such as 29 most of the double bonds are located in pentagons,
which may be unfavorable, whereas in structures such as 15 only [6,6] double bonds
and pentagons constructed with single bond edges are present. In addition, similar
to Th-C60H12, four benzenoid rings are formed, which may provide further stabilization. Considering all the energy values of the different isomers obtained by different
methods (Table 5.7) reveals the benzenoid ring as a very important stabilizing
Figure 5.10 Schlegel diagram representations of some isomers  of C60H36
(out of 1014 theoretically possible isomers). Bold lines represent double bonds;
bold hexagons represent a hexagon with three conjugated double bonds.
element. The more isolated benzenoid 6-rings the isomer holds the more stable is
this isomer. In most calculations the T-isomer was found to be the most stable
isomer. This derivative has four benzenoid rings, the maximum number. The three
product isomers of the reduction with transfer hydrogenation (Section 5.3.3), whose
structures are proven, have at least three benzenoid rings. Remarkably, instead of
the two major isomers (Figure 5.8) actually formed, the minor isomer with
T-symmetry (Figure 5.8) was predicted to be most stable. The S6 isomer with only
two benzenoid rings, which was often claimed to be among the most stable isomers,
does not follow this trend.
Putting any of the 36 hydrogens inside the fullerene leads to conformers with
higher energies .
5.3 Polyhydrofullerenes C60Hn and C70Hn (n = 14–60)
Table 5.7 Heats of formation, normalized relative to isomer 15 at various levels of theory.
ΔHf° AM1 a)
ΔHf° DF b)
ΔHf° D c)
AM1 with MOPAC 6.0 .
Density functional technique B3LYP/6-31G* with Gaussian 98, (b) , (c) .
Optimized with MNDO, reoptimized at the SCF/3-21G level .
Figure 5.11 All-outside isomer of C60H60 .
The icosahedral C60H60 (Figure 5.11) is neither the only nor the most stable conformer of completely hydrogenated C60. Molecular mechanics calculations (MM3)
on C60H60 show that moving just one hydrogen inside the cage and forming a new
conformer is an exothermic process and decreases the energy by 53 kcal mol−1
A conformer of C60H60 with 10 hydrogens added inside the cage is the lowest
energy isomer (Figure 5.12). This minimum energy isomer with C1-symmetry is
predicted to have a heat of formation 400 kcal mol−1 lower than that for the all-
Figure 5.12 Most stable isomer of C60H60 with ten hydrogens inside .
outside isomer . In another attempt, an isomer with D5-symmetry but also with
10 hydrogens inside the cage was found as the most stable isomer . The alloutside isomer is highly strained due to eclipsing hydrogen–hydrogen interactions
and due to the 120° angles of the sp3 carbons (20 planar cyclohexane rings). Putting
hydrogens inside leads to fewer eclipsing interactions and also to a decrease of
many C–C–C bond angles. In another study, comparison on the basis of ab initio
calculations of the stabilities of totally exo-hydrogenated C60H60 and Th-C60H60,
with 12 hydrogens added inside the cage, predicts the latter conformer to be the
more stable . The calculated energy barrier for the hydrogen penetration from
outside to inside the C60 cage, however, is very high (at least 2.7 eV atom−1). To date
no C60H60 could be isolated, regardless of the hydrogens pointing in or out of the
C. C. Henderson, P. A. Cahill, Science
1993, 259, 1885.
S. Ballenweg, R. Gleiter, W. Krätschmer, Tetrahedron Lett. 1993, 34, 3737.
A. G. Avent, A. D. Darwish, D. K. Heimbach, H. W. Kroto, M. F. Meidine,
J. P. Parsons, C. Remars, R. Roers,
O. Ohashi, R. Taylor, D. R. M. Walton,
J. Chem. Soc., Perkin Trans. 2 1994, 15.
T. F. Guarr, M. S. Meier, V. K. Vance,
M. Clayton, J. Am. Chem. Soc. 1993, 115,
R. G. Bergosh, M. S. Meier,
J. A. L. Cooke, H. P. Spielmann,
B. R. Weedon, J. Org. Chem. 1997, 62,
W. E. Billups, W. Luo, A. Gonzalez,
D. Arguello, L. B. Alemany,
T. Marriott, M. Saunders,
H. A. Jimenez-Vazquez, A. Khong,
Tetrahedron Lett. 1997, 38, 171.
J. Nossal, R. K. Saini, L. B. Alemany,
M. Meier, W. E. Billups, Eur. J. Org.
Chem. 2001, 4167.
A. Hirsch, A. Soi, H. R. Karfunkel,
Angew. Chem. 1992, 104, 808.
A. Hirsch, T. Grösser, A. Skiebe,
A. Soi, Chem. Ber. 1993, 126, 1061.