Regioselective addition of N-(4-Thiocyanatophenyl)pyrrolidine addends to fullerenes

Article abstract of DOI:10.1002/ejoc.200900228

The regioselectivity of 1,3-dipolar cycloaddition reactions between N-(4-thiocyanatophenyl)glycine and tetrakis[bis(ethoxycarbonyl)methylene]-C ₆₀ (7) was studied. The monoaddition resulted, in two penta-adduct isomers (8 and 9) in a 1:1 ratio.

Full text of DOI:10.1002/ejoc.200900228

FULL PAPER
DOI: 10.1002/ejoc.200900228
Regioselective Addition of N-(4-Thiocyanatophenyl)pyrrolidine Addends to
Fullerenes
Angy L. Ortiz,[a] Danisha M. Rivera,[a] Andreas J. Athans,[a] and Luis Echegoyen*[a]
Keywords: Fullerenes / Nitrogen heterocycles / Cycloaddition / Bingel–Hirsch reaction / Orthogonal transposition /
Regioselectivity / Isomers
The regioselectivity of 1,3-dipolar cycloaddition reactions be-
tween N-(4-thiocyanatophenyl)glycine and tetrakis[bis-
(ethoxycarbonyl)methylene]-C60 (7) was studied. The mono-
addition resulted in two penta-adduct isomers (8 and 9) in a
1:1 ratio. A second addition to each compound led to, in the
case of 8, two hexa-adduct isomer products (10 and 11),
whereas addition to 9 led to 11 and 12. Finally, reaction of 11
and the glycine afforded a third addition, giving rise to
hepta-adduct isomers 13 and 14. Preparative TLC and col-
umn chromatography were used to separate and purify these
compounds. The structures were assigned based on their mo-
lecular symmetries as analyzed by 1H, 13C, and 2D NMR
spectra and MALDI-TOF MS.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
were called C3 and D3.[5] It should be noted that the isola-
tion of these isomers was only achieved after laborious
chromatographic separation and was not a trivial endeavor.
Hirsch also proposed a nomenclature for the positional
relationships of the eight different double bonds in bis-
(ethoxycarbonyl)methylene-C60 [6,6]-bis-adducts with re-
spect to the bond carrying the first addend. The most favor-
able positions for subsequent cyclopropanation of a mono-
adduct were found to be the equatorial and trans-3 posi-
tions (1, Figure 1).[5b]
Introduction
Since Krätschmer, Huffman, and co-workers[1] discov-
ered a bulk fullerene preparation method in 1990, the ad-
ducts of C60 have been studied extensively. One area that
has interested researchers is the use of fullerenes in molecu-
lar electronics applications. There are reports of C60 being
used as a single-molecule transistor by inducing nanomech-
anical oscillations between two gold electrodes,[2a,2b] tunnel-
ing of electrons through a C60 compound between two mag-
netic nickel electrodes[2c] and a field-sensitive double-tunnel
junction with C60 embedded in SiO2 for nonvolatile mem-
ory applications.[2d] Morita and Lindsay[2e] presented the
first example of a C60 derivative as a single-molecule tran-
sistor. Two bis-adducts trans-1 and trans-2 with amino-ter-
minated linkers exhibited enhanced electron-tunneling con-
ductance upon accepting electrons, with the dianion being
the best conductor and the neutral species the worst.[2e]
The presence of 30 equiv. of [6,6] double bonds on the
surface of C60, all of which exhibit identical reactivity, re- Figure 1. Hirsch nomenclature of the eight double bonds in C60
[6,6]-bis-adducts with respect to the bond having the first addend
sults in low regioselectivity under conditions resulting in
R.
multiaddition products.[3] After the formation of the mono-
adduct, successive additions of one, two, and three symmet-
The distribution of the eight possible bis-adducts de-
rical addends yield 8, 46, and 262 possible regioisomers,
pends not only on the steric demands of the addends but
respectively.[4] Hirsch and co-workers published an exten-
also on the type of cycloaddition reaction employed. An-
sive study on the isolation and characterization of the mul-
other cycloaddition[6] that has been extensively studied is
tiple adducts from cyclopropanation of C60 with diethyl
the 1,3-dipolar cycloaddition of azomethine ylides to C60,
bromomalonate. These included the first tris-adducts, which
first reported by Prato and co-workers.[7] This reaction
yields a fullerene product with a pyrrolidine ring attached.
[a] Department of Chemistry, Clemson University,
Wilson and co-workers found that 1,3-dipolar cycload-
ditions are less chemoselective than the corresponding cy-
clopropanations of C60.[6] The reaction resulted in all eight
possible regioisomers if the azomethine ylide was symmet-
219 Hunter Laboratories, Clemson, SC 29634, USA
Fax: +1-864-656-6613
E-mail: luis@clemson.edu
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.200900228.
3396
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 3396–3403

Addition of N-(4-Thiocyanatophenyl)pyrrolidine Addends to Fullerenes
ric. The yield of trans-1 bis-adducts is only 1–6% of the moved thermally.[13b] In closely related works Kräutler and
total yield and is the isomer formed in the lowest amount.[8] co-workers reported the first example of an approach they
Additionally, the isolation of isomerically pure bis-adducts named “orthogonal transposition” in which C60 was pro-
is challenging because bis- and tris-adducts co-elute during tected with two thermally labile anthracene groups in the
chromatography due to their similar polarities.[9] One way trans-1 positions followed by symmetric addition of four
in which the regiochemistry of multiple additions to fuller- malonate groups to the equatorial belt of C60. The anthra-
enes has been controlled is the tether-directed remote multi- cene moieties were then removed by a thermal retro-Diels–
functionalization approach.[11] Specifically, a covalent tem- Alder reaction.[14] Using Kräutler’s approach,[14] in 2006 we
plate is used to functionalize the fullerene with reversibly reported the synthesis of trans-1 bis(terpyridylyl)pyrrolidine
removable addends. Diels–Alder reaction and cyclopropan- C60-tetramalonate 2 and trans-1 pyridylpyrrolidine/terpyr-
ation are the two most common routes that have been used idylylpyrrolidine C60-tetramalonate 3.[15] The malonates
in this approach.[11] One of the advantages is that the regi- were subsequently removed by controlled potential electrol-
ochemistry of bis-adducts can be controlled by the length, ysis (CPE) to give the desired trans-1 bis(terpyridylyl) com-
geometry and rigidity of the tether. In some examples equa- pound 4 (Figure 2).[15] An interesting result was that the
torial,[10] cis-2,[11] trans-2[12a] or trans-1,[12b,12c] structures electrochemistry became reversible upon removal of the
could be obtained almost exclusively.
malonates. In the present work, we present a study of mul-
In 1999, Th- and D3-hexakis-adducts of fulleropyrrolid- tiple additions of the thiocyanatophenyl group (C6H4SCN)
ine hexanitroxide were reported by Rubin and co- to tetrakis[bis(ethoxycarbonyl)methano]-C60, (7). We report
workers.[13a] In 2006, Rubin and co-workers also reported a a family of compounds that can potentially complex dif-
cis-1 bis-adduct temporarily blocked by a tethered 1,3- ferent metal ions such as CuII,[16a] Ag, Pt, and Au upon
diene, which activated the LUMO orbital coefficients at the cleavage of the S–CN bond, giving rise to species identical
[trans-4,trans-4,trans-4] positions. Later, the tether was re- to those seen in free thiol assemblies.[16b,16c]
Figure 2. Structures of compounds 2, 3, and 4.
Scheme 1. Synthesis of C60-tetramalonate 7.
Eur. J. Org. Chem. 2009, 3396–3403
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3397

A. L. Ortiz, D. M. Rivera, A. J. Athans, L. Echegoyen
FULL PAPER
The crux of the orthogonal transposition approach in- at two different positions. It can either add to the unique
volves a protection/deprotection scheme in order to prepare double bond on a hemisphere (called symmetric in Fig-
an equatorially protected C60 compound containing four ure 3) or to one of the other four equivalent double bonds
addends, which leave the antipodal hemispheres of the ful- located on the same hemisphere (called unsymmetric in Fig-
lerene available for further reaction.[14] This is done by first ure 3). If the addition reaction occurred statistically the
preparing trans-1 bis(anthracene) adduct 5 of C60 in a solid- product ratio should be 1:4 in favor of the unsymmetrical
state reaction,[17] followed by multiple Bingel reactions in product 9 vs. 8. Since the observed ratio was 1:1, the reac-
order to prepare hexakis-adduct 6.[14] Finally, thermally in- tion at the unique double bond is favored fourfold, probably
duced retro-Diels–Alder reaction to remove the anthracene due to steric hindrance of the malonates around the unsym-
moieties gives compound 7 (Scheme 1).[14] The protection/ metric bonds. Thanks to electronic effects as well,[3,18] the
deprotecion scheme was further extended by treating com- symmetric double bond (Figure 3) will be favored because
pound 7 with terpyridylylglycine and paraformaldehyde, it is an equatorial position with respect to each cyclopro-
which resulted in mono- and bis-addition to yield 2.[15]
On the basis of the results obtained in preparing 2–4,
further studies to develop this strategy for the regioselective
synthesis of pentakis-, hexakis-, and heptakis-adducts of
C60-fullerene were designed, and the results are presented
in the present work.
pane ring.
Results and Discussion
Synthesis of 8, 9 and 10
Compound 7, N-[(4-thiocyanato)phenyl]glycine and pa-
raformaldehyde were allowed to react in a 1,3-dipolar cy-
cloaddition reaction to give hexakis-adducts 10 and 11 in a
single step. This is different from the protocol used pre-
viously[15] where the pentakis-adduct products were reco-
Figure 3. Two different sites of addition to 7 leading to symmetric
(8) or unsymmetric (9) pentakis-adducts.
The (C2v and C1) symmetries of 8 and 9 were determined
1
vered, purified and allowed to react again under the same by NMR spectroscopy. The H NMR spectrum of 8 (Fig-
conditions to obtain the hexakis-adducts (Scheme 2).
ure 4) showed three sets of triplets for CH3 protons, with
After 30 min, three molecular ion peaks corresponding integration ratios of 6:6:12, between δ = 1.40 and 1.30 ppm.
to pentakis-, hexakis- and heptakis-adducts of C60-fullerene These were assigned to the malonate groups for compound
were observed by MALDI-TOF MS. The crude reaction 8. The signals of the methylene protons from the malonate
mixture was subjected to column chromatography with groups overlapped with those of the protons from the pyr-
CH2Cl2 as eluant, and six fractions were isolated rolidine ring. Two doublets integrating for two protons each
(Scheme 2). The first two fractions had the same molecular were observed for the four aromatic protons. The C2v sym-
mass (S25, Supporting Information) corresponding to the metry of compound 8 was also corroborated by the 13C
pentakis-adducts 8 and 9, and the subsequent three frac- NMR spectrum (S14, Supporting Information). Three sig-
tions had the same molecular mass corresponding to hexa- nals for carbonyl groups and three for methyl groups indi-
kis-adducts 10, 11 and 12. The last fraction was a mixture cated that two groups were in different chemical environ-
of heptakis-adducts. The first two pentakis fractions were ments and two were equivalent. Twelve sp2-C signals of the
1
characterized by means of H and 13C NMR spectroscopy fullerene cage were observed between δ = 154 and 117 ppm.
and were determined to be the isomeric compounds 8 and Seven sp3-C signals were also observed, three between δ =
9, respectively. The addition of the C6H4SCN group occurs 69.85 and 61.40 ppm, assigned to the sp3-fullerene carbon
Scheme 2. Synthesis of fullerene derivatives with 4-thiocyanatophenyl groups.
3398
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 3396–3403

Addition of N-(4-Thiocyanatophenyl)pyrrolidine Addends to Fullerenes
atoms. One of these signals was from the two carbon atoms
of the pyrrolidine ring, and the other two were from the
two different carbon atoms of the four cyclopropane rings.
The remaining four sp3-C signals left in this region were
assigned to methylene groups, where one signal was for the
pyrrolidine ring and three were for the four malonate
groups.
1
Figure 5. (a) H NMR spectrum of asymmetric pentakis-adduct 9
(500 MHz, CDCl3) and expanded parts from (b) δ = 4.8 to 4.2 ppm
and (c) δ = 1.6 to 1.2 ppm.
Figure 4. (a) 1H NMR spectrum of symmetric pentakis-adduct 8
(500 MHz, CDCl3) and expanded parts from (b) δ = 4.5 to 4.2 ppm
and (c) δ = 1.5 to 1.2 ppm.
1
The analysis of 9 by H NMR spectroscopy (Figure 5)
showed the expected lower symmetry (Figure 5). The aro-
matic proton signals were shifted downfield when compared
to those of 8. Seven triplets were observed between δ = 1.52
and 1.31 ppm due to eight different methyl groups (one
triplet had twice the intensity of the others) and multiplets
for the methylene protons indicated NMR non-equivalency.
The 13C NMR spectrum (S16, Supporting Information) of
9 indicated C1 symmetry with 49 sp2-C signals from the
fullerene cage appearing between δ = 151 and 111 ppm, 7
signals for carbonyl groups and seven for methyl groups.
Nine signals between δ = 72.22 and 62.00 ppm were as-
signed to the sp3-C atoms of the fullerene cage, whereas
seven signals were assigned to the cyclopropane rings, and
two were assigned to the pyrrolidine ring. Finally, nine sig-
nals were assigned to the methylene moieties of the malon-
ate groups.
1
Figure 6. H NMR spectrum of (a) compound 10, (b) compound
11 (CDCl3).
1
Similar H NMR results were previously obtained for 2,
which possesses the same D2h symmetry as 10.[15] The 13C
NMR spectrum of 10 showed two peaks for carbonyl
groups, two peaks for methyl groups, and six sp2-C peaks
for the fullerene cage. Finally, four signals between δ =
66.87 and 61.00 ppm for the sp3-C atoms of the fullerene
cage and methylene groups were observed (see Supporting
Information). The NMR spectrum of 10 indicated D2h sym-
metry because all of the addends are located at the pseudo-
octahedral positions on the fullerene. The structures of the
other two hexakis-adducts from fractions b and c were diffi-
cult to assign by 1H and 13C NMR spectroscopy and COSY
because of their structural complexity and lack of sym-
metry. Hence, their structures were elucidated by using an
alternate synthetic route, discussed below.
The next three fractions (a, b, and c), with a mass ratio
of 1:2.2:1.3, were purified by preparative TLC on silica gel
using dichloromethane as eluant. The plate was dried mul-
tiple times and new eluant used until an optimal separation
was obtained. Fraction a was identified as hexakis-adduct
1
10 after characterization by H NMR spectroscopy (Fig-
ure 6, a), 13C NMR spectroscopy, COSY, and MALDI-
1
TOF. The H NMR spectrum (Figure 6, a) exhibited two
doublets integrating for four protons each. A sharp peak at
δ = 4.59 ppm was assigned to the eight equivalent pyrroli-
dine protons. Two quartets at δ = 4.35–4.30 and 4.30–
4.26 ppm integrating for eight protons each were assigned
Synthesis of 10, 11, and 12
After allowing 8, N-(4-thiocyanatophenyl)glycine, and
to the two sets of methylene moieties in the malonate paraformaldehyde to react for 15 min (Scheme 3), two hexa-
groups. Two of the triplets at δ = 1.35–1.32 and 1.28– kis-adducts and one heptakis-adduct were detected by TLC.
1.25 ppm corresponded to twelve protons each and were Three fractions were isolated by preparative TLC using
assigned to the two different methyl groups from the malon- dichloromethane as eluant. The first two fractions were
ate groups.
characterized as hexakis-adducts 10 and 11, respectively.
Eur. J. Org. Chem. 2009, 3396–3403
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3399

A. L. Ortiz, D. M. Rivera, A. J. Athans, L. Echegoyen
FULL PAPER
Scheme 3. Synthesis of [N-(4-thiocyanatophenyl)pyrrolidino]fullerene derivatives 10 and 11.
Not surprisingly, 10 and 11 were obtained in a ratio of 1:2.4
instead of the expected 1:4 ratio (see Figure 3).
Two sets of doublets for each ring were observed in the
1H NMR spectrum of 11 (Figure 6, b), two of them over-
lapped and integrated to four protons. The other doublets
correspond to two protons each, for a total of four different
aromatic protons. The protons of the pyrrolidine ring and
the malonate groups overlapped and gave rise to a mul-
tiplet.
Figure 7 shows the COSY spectrum for 11. Three corre-
lations were observed between (1) CH2 and CH3 of the ma-
lonate groups, (2) the protons of the pyrrolidine ring, and
(3) the protons of the C6H4SCN ring. The lack of symmetry
results from addition at one of the four unsymmetric bonds
on the unoccupied hemisphere of 8. The controlled synthe-
sis of the hexakis-adducts by using 8 as the starting material
confirmed the hypothesis that an addition to the same side
of the first symmetric addend was not feasible and does not
occur. Thus, two possible isomers can be obtained when the
synthesis starts with this pentakis-adduct (Figure 8, a). The
NMR spectra and the Rf value of fraction b are consistent
with those expected for 11. This confirmed that fraction b
corresponds to compound 11. The reaction between 9, N-
Figure 7. COSY spectrum of 11 (500 MHz, CDCl3).
(4-thocyanatophenyl)glycine,
and
paraformaldehyde
(Scheme 4) was stopped after 10 min when heptakis-ad-
ducts were observed by TLC. After purification, two hexa-
kis-adducts were isolated (11 and 12) in a mass ratio of
1:1.4. The NMR spectra and the Rf value of fraction c are
consistent with those of 12. When compound 12 was sub-
jected to HPLC by using toluene as the mobile phase in a
Buckyprep column, seven peaks were observed, the reten-
tion time of one of them matching that of 11. From the
seven isomers, two of them had an area three times larger
than those of the other five products (S22, Supporting In-
formation). The next addition of the C6H4SCN group to 9
during the synthesis of 12 resulted in up to seven isomers
(Figure 8, b).
Figure 8. (a) Schematic representation of two possible isomers for
a second addition to compound 8 (mono-symmetric adduct). (b)
Schematic representation of seven possible isomers for a second
addition to compound 9 (mono-unsymmetric adduct). Dotted
circles represent possible addition sites, and bold-faced bonds rep-
resent adduct positions.
From these reactions (Schemes 3 and 4) it is possible to
conclude that 9 is more reactive than 8. After 10 min of occurs on an unsymmetric double bond, the resulting com-
reaction, 9 gave rise to more products than 8. Also, the pound is more reactive, which decreases the yield. Thus, the
amount of 9 recovered after the reaction was much less than reactivity order is 11 Ͼ 10 Ͼ 9 Ͼ 8, with 11 being the most
in the case of 8. This means that every time an addition reactive.
3400
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 3396–3403

Addition of N-(4-Thiocyanatophenyl)pyrrolidine Addends to Fullerenes
Scheme 4. Synthesis of [N-(4-thiocyanatophenyl)pyrrolidino]fullerene derivatives 11 and 12.
Synthesis of 13 and 14
After purification and characterization of 11, the synthe-
sis of heptakis-adducts was pursued by using 11 as the
starting material. According to the previous results, the ad-
dition of a new group should not occur on the same six-
membered ring that already has a group attached. This
leads to the possibility of just two regioisomers (Figure 9).
Figure 10. 1H NMR spectrum of (a) compound 13 and (b) com-
pound 14 (500 MHz, CDCl3).
In an attempt to determine the position of the third ad-
dition of the C6H4SCN group, HMBC (Heteronuclear Mul-
tiple Bond Correlation) experiments were performed (Fig-
ure 11), to determine whether any correlations between the
pyrrolidine protons and the sp2-hybridized carbon atoms
were present. Figure 11 shows that in the HMBC one set of
the pyrrolidine protons correlated with one sp2 carbon
Figure 9. Two different bonds available for the heptakis-adducts of
compound 11.
The reaction of 11, N-(4-thiocyanatophenyl)glycine, and
paraformaldehyde in o-dichlorobenzene resulted in a mix-
ture of heptakis-adducts (Figure 9). After 30 min of reac-
tion, the crude material was subjected to preparative TLC
with CH2Cl2 as eluant, giving two fractions. The first two
fractions had the molecular ion peak corresponding to hep-
takis-adducts, defined as heptakis-1 and heptakis-2, respec-
1
tively. The H NMR spectrum of heptakis-1 (Figure 10, a)
showed two doublets in the aromatic region which inte-
grated for two protons each, and two “triplets” correspond-
ing to four protons each (which were two overlapped dou-
blets) for a total of six different aromatic protons.
Figure 11. HMBC NMR spectrum of 13 (500 MHz, CDCl3).
Eur. J. Org. Chem. 2009, 3396–3403
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3401

A. L. Ortiz, D. M. Rivera, A. J. Athans, L. Echegoyen
FULL PAPER
atom as well as with an sp3 carbon atom of the fullerene.
However, the two isomers did not show significant differ-
ences in their correlations.
Although the identity of the two regioisomers cannot be
established at this point, if the polarity of the molecules is
considered, 13 (side addition 1) should be less polar than
14 (side addition 2), since the third addend is opposite the
trans-2 position (Figure 9). Therefore, heptakis-1 can be
tentatively assigned as 13 and heptakis-2 as 14.
dichlorobenzene (300 mL) was heated to 175 °C under Ar for
30 min. The resulting mixture was cooled and concentrated with
an N2 stream. Separation of the 7 (200 mg, material recovered), 8
(71 mg, 6.3%), 9 (70 mg, 6.2%), 10 (24.4 mg, 2.2%), and 11
(51.9 mg, 4.6%) was achieved by column chromatography on silica
gel using dichloromethane as eluant. Better purification of 10 and
11 was achieved by preparative TLC on silica gel using dichloro-
methane as eluant. 8: 1H NMR (500 MHz, CDCl3, 25 °C): δ = 7.50
(d, 3JH,H = 8.9 Hz, 2 H, ArH), 6.93 (d, 3JH,H = 8.9 Hz, 2 H, ArH),
4.45–4.41 (q, 4 H, -CH2-), 4.36–4.31 (m, 16 H, -CH2-), 1.40–1.30
(m, 24 H, CH3) ppm. 13C NMR (500 MHz, CDCl3, 25 °C): δ =
163.83, 163.73, 163.64, 153.64, 152.36, 149.41, 147.17, 146.14,
145.98, 145.47, 145.30, 144.26, 143.41, 143.15, 142.29, 138.58,
138.20, 133.83, 117.01, 111.80, 69.85, 69.11, 68.22, 67.75, 62.98,
Conclusions
Herein we described the regioselective synthesis and
characterization of mono-, bis-, and tris(4-thiocyanato-
phenyl)pyrrolidino-C60-tetramalonate adducts, that are po-
tentially useful compounds for molecular electronic applica-
tions. The syntheses were achieved by using a selective pro-
tection/deprotection strategy. The low yield (mass ratio 1:1)
and reaction time for unsymmetric pentakis-adduct 9 indi-
cate that it is more reactive than symmetric heptakis-adduct
8. The hexakis-adducts isomers trans-1 10 and trans-2 11
were obtained in a ratio of 1:2.4, which indicated the prefer-
ence of addition at the unsymmetric positions. Formation
of two heptakis-adducts, 13 and 14, from hexakis-adduct
trans-2 11 was detected in a ratio of 1:1.3, which indicated
that competition was no longer present. Therefore, the ad-
dition of a new group seems to be disfavored on the same
six-membered ring that already has a group attached.
62.94, 61.40, 45.39, 44.88, 14.17, 14.09, 14.06 ppm. UV/Vis: λmax
=
291, 309, 339, 476, 545 nm. MS (MALDI): m/z = 1527 [M+ – 1],
1369, 1352. 9: 1H NMR (500 MHz, CDCl3, 25 °C): δ = 7.53 (d,
3
3JH,H = 8.9 Hz, 2 H, ArH), 7.02 (d, JH,H = 8.9 Hz, 2 H, ArH),
4.71–4.27 (m, 20 H, -CH2-), 1.52–1.31 (8 t, 24 H, CH3) ppm. 13C
NMR (500 MHz, CDCl3, 25 °C): δ = 164.21, 164.11, 163.86,
163.70, 163.65, 163.15, 150.88, 149.45, 149.28, 149.00, 148.66,
148.34, 147.74, 147.44, 147.07, 147.02, 146.95, 146.08, 145.78,
145.47, 145.13, 144.42, 144.30, 144.23, 144.15, 143.78, 143.47,
143.32, 143.27, 142.99, 142.82, 142.74, 142.52, 142.34, 142.11,
141.95, 141.90, 141.74, 141.61, 141.30, 141.11, 140.99, 140.87,
140.79, 140.45, 140.33, 140.25, 140.07, 138.69, 138.64, 138.34,
138.19, 135.18, 133.89, 131.81, 130.60, 125.81, 116.98, 111.86,
111.51, 72.22, 70.77, 70.50, 69.17, 69.05, 68.71, 66.34, 66.07, 65.08,
63.27, 63.13, 63.05, 63.03, 62.98, 62.78, 62.65, 62.39, 62.00, 46.33,
43.94, 43.31, 41.37, 14.29, 14.25, 14.20, 14.18, 14.11, 14.09,
14.08 ppm. UV/Vis: λmax = 271, 330, 548 nm. MS (MALDI): m/z
= 1528 [M+], 1369, 1352.
Hexakis-Adducts 10 and 11: A solution of 8 (20 mg, 0.013 mmol),
N-(4-thiocyanatophenyl)glycine (14 mg, 0.065 mmol), and para-
formaldehyde (4 mg, 0.130 mmol) in 1,2-dichlorobenzene (2 mL)
was heated at 175 °C under Ar for 15 min. The mixture was cooled,
and the solvents were evaporated. Separation of 8 (6 mg, material
recovered), 10 (2 mg, 9%) and 11 (5 mg, 22%) was achieved by
preparative TLC on silica gel using dichloromethane as eluant. 10:
Experimental Section
Synthesis of N-(4-Thiocyanatophenyl)glycine: Ethyl bromoacetate
(1.10 mL, 9.99 mmol) was added to a solution of 4-aminophenyl
thiocyanate[19] (3.0 g, 19.97 mmol) in anhydrous 1,4-dioxane
(6 mL). The mixture was heated to reflux for 1 h. After the solution
had cooled, water (10 mL) was added, and the mixture was ex-
tracted with chloroform. The organic layer was removed and
washed with saturated NaHCO3 twice, brine, dried with MgSO4,
then concentrated. The resulting product was purified by column
chromatography on silica gel (chloroform/acetone, 9:1 as eluant) to
give ethyl N-(4-thiocyanatophenyl)glycinate (4.0 g, 60%). 1H NMR
(500 MHz, CDCl3, 25 °C): δ = 7.41 (m, 2 H, ArH), 6.60 (m, 2 H,
ArH), 4.64 (br., 1 H, NH), 4.26 [q, 3J(H,H) = 7.1 Hz, 2 H,
3
1H NMR (500 MHz, CDCl3, 25 °C): δ = 7.54 (d, JH,H = 9.0 Hz,
3
4 H, ArH), 7.04 (d, JH,H = 9.0 Hz, 4 H, ArH), 4.59 (s, 8 H,
3
NCH2-), 4.35–4.30 (q, JH,H = 7.1 Hz, 8 H, -CH2-), 4.30–4.26 (q,
3
3JH,H = 7.1 Hz, 8 H, -CH2-), 1.35–1.32 (t, JH,H = 7.1 Hz, 12 H,
3
CH3), 1.28–1.25 (t, JH,H = 7.1 Hz, 12 H, CH3) ppm. 13C NMR
(500 MHz, CDCl3, 25 °C): δ = 163.87, 163.83, 152.34, 149.45,
146.11, 145.79, 145.53, 145.40, 143.88, 143.66, 143.47, 141.62,
141.13, 140.99, 140.33, 139.72, 133.89, 117.19, 111.82, 70.83, 69.38,
69.08, 66.87, 66.11, 62.85, 61.99, 61.12, 46.11, 45.44, 45.38, 44.71,
14.27, 14.06 ppm. UV/Vis: λmax = 275, 332, 546 nm. MS (MALDI):
m/z = 1704 [M+], 1669, 1596, 1511, 1352. 11: 1H NMR (500 MHz,
3
-COCH2-], 3.90 (d, 2 H, -CH2CO-), 1.32 (t, JH,H = 7.1 Hz, 3 H,
CH3) ppm. 13C NMR (500 MHz, CDCl3, 25 °C): δ = 170.30
(C=O), 148.81, 134.58, 114.05, 112.28, 109.21, 61.69, 45.19,
14.19 ppm. MALDI-MS: m/z = 237 [MH+]. Ethyl N-(4-thiocyana-
tophenyl)glycinate (2.5 g, 10.6 mmol) was dissolved in 5  HCl/di-
oxane[20] (11 mL) and heated to reflux under Ar for 15 min. After
the solution had cooled, the N-(4-thiocyanotophenyl)glycine pre-
cipitated, was filtered and washed with dioxane to give a white
solid (2.8 g, 76%). 1H NMR (500 MHz, [D6]DMSO, 25 °C): δ =
12.65 (br., 1 H, CO2H), 7.39 (m, 2 H, ArH), 6.66 (m, 2 H, ArH),
3.86 (s, 2 H, -CH2-), 3.50 (br., 1 H, NH) ppm. 13C NMR ([D6]-
DMSO, 500 MHz): δ = 172.5 (C=O), 151.0, 135.1, 114.0, 113.0,
106.3, 44.55 ppm. MALDI-MS: m/z = 209 [MH+].
3
CDCl3, 25 °C): δ = 7.55 (d, JH,H = 9.0 Hz, 2 H, ArH), 7.52 (d,
3
3JH,H = 9.0 Hz, 2 H, ArH), 7.06 (d, JH,H = 8.8 Hz, 2 H, ArH),
3
6.98 (d, JH,H = 8.9 Hz, 2 H, ArH), 4.64–4.51 (m, 4 H, -CH2-),
4.47–4.21 (m, 20 H, -CH2-), 1.34–1.26 (m, 24 H, CH3) ppm. 13C
NMR (500 MHz, CDCl3, 25 °C): δ = 164.00, 163.95, 163.86,
163.83, 163.50, 163.20, 163.03, 153.84, 153.17, 152.06, 151.60,
151.13, 149.45, 149.32, 148.66, 147.80, 146.92, 146.66, 146.61,
146.51, 146.42, 146.29, 146.21, 145.99, 145.90, 145.79, 145.71,
145.51, 145.33, 145.25, 144.37, 144.12, 143.50, 143.18, 143.00,
142.48, 142.39, 142.34, 142.25, 142.04, 141.85, 141.43, 141.32,
140.94, 140.68, 139.71, 138.82, 138.53, 136.75, 136.68, 136.59,
136.18, 136.12, 133.93, 133.88, 130.27, 125.23, 117.09, 117.02,
111.89, 111.75, 111.52, 70.55, 70.37, 70.20, 69.15, 68.89, 67.86,
Pentakis-Adducts 8, 9 and Hexakis-Adducts 10 and 11: A solution
of 7 (1 g, 0.74 mmol), N-(4-thiocyanatophenyl)glycine (385 mg,
1.85 mmol), and paraformaldehyde (111 mg, 3.7 mmol) in 1,2-
3402
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 3396–3403

Addition of N-(4-Thiocyanatophenyl)pyrrolidine Addends to Fullerenes
E. C. Kan, Appl. Phys. Lett. 2008, 92, 153109/1–153109/3; e) T.
Morita, S. Lindsay, J. Phys. Chem. B 2008, 112, 10563–10572.
[3] L. Pasimeni, A. Hirsch, I. Lamparth, A. Herzog, M. Maggini,
M. Prato, C. Corvaja, G. Scorrano, J. Am. Chem. Soc. 1997,
119, 12896–12901.
67.74, 67.56, 67.37, 66.99, 64.84, 63.76, 62.98, 62.92, 62.89, 62.84,
62.80, 62.75, 62.56, 62.52, 61.85, 61.57, 61.50, 61.34, 45.55, 43.77,
43.46, 42.03, 14.14, 14.09, 14.06, 14.04 ppm. UV/Vis: λmax = 286,
482, 525 nm. MS (MALDI): m/z = 1706 [M+ + 2], 1671, 1553
[M+ – C6H4SCN], 1354 [M+ – 2 C6H4SCN].
[4] P. W. Fowler, J. Chem. Soc. Faraday Trans. 1995, 91, 2241–
2247.
Heptakis-Adducts 13 and 14: A solution of 11 (11 mg, 0.064 mmol),
N-(4-thiocyanatophenyl)glycine (19 mg, 0.093 mmol), and para-
formaldehyde (6 mg, 0.190 mmol) in 1,2-dichlorobenzene (1 mL)
was heated at 175 °C under Ar for 30 min. The mixture was cooled,
and the solvents were evaporated. Separation of 11 (4.7 mg, mate-
rial recovered), 13 (2.2 mg, 18%) and 14 (1.7 mg, 14%) was
achieved by preparative TLC on silica gel using dichloromethane
[5] a) I. Lamparth, A. Hirsch, J. Chem. Soc., Chem. Commun.
1994, 116, 1727–1728; b) A. Hirsch, I. Lamparth, H. R. Kar-
funkel, Angew. Chem. Int. Ed. Engl. 1994, 33, 437–438.
[6] Q. Lu, D. I. Schuster, S. R. Wilson, J. Org. Chem. 1996, 61,
4764–4768.
[7] M. Maggini, G. Scorrano, M. Prato, J. Am. Chem. Soc. 1993,
115, 9798–9799.
1
[8] K. Kordatos, S. Bosi, T. Da Ros, A. Zambon, V. Lucchini, M.
Prato, J. Org. Chem. 2001, 66, 2802–2808.
[9] S. Marchesan, T. Da Ros, M. Prato, J. Org. Chem. 2005, 70,
4706–4713.
[10] L. Isaacs, R. F. Haldimann, F. Diederich, Angew. Chem. Int.
Ed. Engl. 1994, 33, 2339–2342.
as eluant. 13: H NMR (500 MHz, CDCl3, 25 °C): δ = 7.55 (m, 4
3
H, ArH), 7.48 (d, JH,H = 8.1 Hz, 2 H, ArH), 7.05 (m, 4 H, ArH),
3
6.93 (d, JH,H = 8.1 Hz, 2 H, ArH), 4.65–4.10 (m, 28 H, -CH2-),
1.42–1.05 (m, 24 H, CH3) ppm. UV/Vis: λmax = 274, 520 nm. MS
(MALDI): m/z = 1881 [M+ + 1], 1723, 1705 [M+ – C6H4SCN],
1547, 1529 [M+ – 2 C6H4SCN], 1371. 14: 1H NMR (500 MHz,
[11] F. Diederich, R. Kessinger, Acc. Chem. Res. 1999, 32, 537–545.
[12] a) E. Dietel, A. Hirsch, E. Eichhorn, A. Rieker, S. Hackbarth,
B. Röder, Chem. Commun. 1998, 1981–1982; b) J.-P. Bourgeois,
L. Echegoyen, M. Fibbioli, E. Pretsch, F. Diederich, Angew.
Chem. Int. Ed. 1998, 37, 2118–2121; c) W. Qian, Y. Rubin, An-
gew. Chem. Int. Ed. 1999, 38, 2356–2360; d) Z. Zhou, D. I.
Schuster, S. R. Wilson, J. Org. Chem. 2006, 71, 1545–1551.
[13] a) G. Schick, M. Levitus, L. Kvetko, B. A. Johnson, I. Lam-
parth, R. Lunkwitz, B. Ma, S. I. Khan, M. A. Garcia-Garibay,
Y. Rubin, J. Am. Chem. Soc. 1999, 121, 3246–3247; b) S.-C.
Chuang, S. I. Khan, Y. Rubin, Org. Lett. 2006, 8, 6075–6078.
[14] R. Schwenninger, T. Müller, B. Kräutler, J. Am. Chem. Soc.
1997, 119, 9317–9318.
3
CDCl3, 25 °C): δ = 7.54 (m, 4 H, ArH), 7.48 (d, JH,H = 8.8 Hz, 2
3
H, ArH), 7.03 (m, 4 H, ArH), 6.93 (d, JH,H = 8.8 Hz, 2 H, ArH),
4.64–4.10 (m, 28 H, -CH2-), 1.37–1.15 (m, 24 H, CH3) ppm. UV/
Vis: λmax = 287, 497, 529 nm. MS (MALDI): m/z = 1880 [M+],
1722, 1704 [M+ – C6H4SCN], 1546, 1529 [M+ – 2 C6H4SCN], 1367.
Supporting Information (see footnote on the first page of this arti-
cle): MALDI mass spectra, UV/Vis spectra and NMR spectra of
new compounds.
Acknowledgments
[15] S. Zhang, O. Lukoyanova, L. Echegoyen, Chem. Eur. J. 2006,
12, 2846–2853.
Financial support from the National Science Foundation (NSF)
(Grant number DMR-0809129) to A. J. A. and L. E. is greatly ap-
preciated. This material is based on work supported by the
National Science Foundation while L. E. was working there. Any
opinion, findings and conclusions or recommendations expressed
in this material are those of the authors and do not necessarily
reflect the views of the NSF.
[16] a) M. Kabesˇová, B. Bocˇa, M. Melník, D. Valigura, M. Dunaj-
Jurcˇo, Coord. Chem. Rev. 1995, 140, 115–135; b) J. W. Ciszek,
J. M. Tour, Chem. Mater. 2005, 17, 5684–5690; c) J. W. Ciszek,
Z. K. Keane, L. Cheng, M. P. Stewart, L. H. Yu, D. Natelson,
J. M. Tour, J. Am. Chem. Soc. 2006, 128, 3179–3189.
[17] a) B. Kräutler, T. Müller, J. Maynollo, K. Gruber, C. Kratky,
P. Ochsenbein, D. Schwarzenbach, H.-B. Bürgi, Angew. Chem.
Int. Ed. Engl. 1996, 35, 1204–1206; b) A. Duarte-Ruiz, K.
Wurst, B. Kräutler, Helv. Chim. Acta 2001, 84, 2167–2177.
[18] A. Hirsch, I. Lamparth, G. Schick, Liebigs Ann. 1996, 11,
1725–1734.
[1] W. Krätschmer, L. D. Lamb, K. Fostiropoulos, D. R. Huffman,
Nature 1990, 347, 354–358.
[2] a) C. Joachim, J. K. Gimzewski, H. Tang, Phys. Rev. B 1998,
58, 16407–16417; b) H. Park, J. Park, A. K. L. Lim, E. H. An-
derson, A. P. Alivisatos, P. L. McEuen, Nature 2000, 407, 57–
60; c) A. N. Pasupathy, R. C. Bialczak, J. Martinek, J. E. Grose,
L. A. K. Donev, P. L. McEuen, D. C. Ralph, Science 2004, 306,
86–89; d) T.-H. Hou, H. Raza, K. Afshari, D. J. Ruebusch,
[19] J. S. Yadav, B. V. S. Reddy, S. Shubashree, K. Sadashiv, Tetrahe-
dron Lett. 2004, 45, 2951–2954.
[20] R. M. Valerio, P. F. Alewood, R. Johns, Synthesis 1988, 786–
790.
Received: March 3, 2009
Published Online: May 27, 2009
Eur. J. Org. Chem. 2009, 3396–3403
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3403