Donor-σ-Acceptor Molecules Incorporating a Nonadecyl-Swallowtailed Perylenediimide Acceptor

Article abstract of DOI:10.1021/jo035409w

Donor-σ-acceptor-lipid molecules were prepared by using perylenetetracarboxylic diimide as the acceptor, starting from perylenetetracarboxylic dianhydride. One imide nitrogen was attached to a "swallowtail" lipid (a long alkyl tail connected at midchain), which imparts enough solubility to make the system tractable and provides a lipophilic region suitable for promoting Langmuir-Blodgett monolayer formation. The other imide link was to a donor group (pyrene, ferrocene, tetramethylphenylenediamine, phenyl) through a short alkyl σ bridge. Features of the ¹H and ¹³C NMR spectra of swallowtailed perylenediimides are interpreted as resulting from restricted rotation about the imide C-N bond; the ¹³C NMR spectra and stereochemistry of these molecules are contrasted with the case of the related bis-(2,5-di-tert-butylphenyl)perylenetetracarboxylic diimide.

Full text of DOI:10.1021/jo035409w

Don or -σ-Accep tor Molecu les In cor p or a tin g a
Non a d ecyl-Sw a llow ta iled P er ylen ed iim id e Accep tor
Lyle D. Wescott and Daniell Lewis Mattern*
Department of Chemistry and Biochemistry, University of Mississippi,
Box 1848, University, Mississippi 38677
mattern@olemiss.edu
Received September 26, 2003
Donor-σ-acceptor-lipid molecules were prepared by using perylenetetracarboxylic diimide as the
acceptor, starting from perylenetetracarboxylic dianhydride. One imide nitrogen was attached to
a “swallowtail” lipid (a long alkyl tail connected at midchain), which imparts enough solubility to
make the system tractable and provides a lipophilic region suitable for promoting Langmuir-
Blodgett monolayer formation. The other imide link was to a donor group (pyrene, ferrocene,
1
tetramethylphenylenediamine, phenyl) through a short alkyl σ bridge. Features of the H and 13C
NMR spectra of swallowtailed perylenediimides are interpreted as resulting from restricted rotation
about the imide C-N bond; the 13C NMR spectra and stereochemistry of these molecules are
contrasted with the case of the related bis-(2,5-di-tert-butylphenyl)perylenetetracarboxylic diimide.
In tr od u ction
Donor-σ-acceptor (D-σ-A) molecules contain an
electron-donor group (D) and an electron-acceptor group
(A), connected through a nonconjugated σ-bond bridge
that prevents the direct overlap of D and A energy
levels.1,2 Aviram and Ratner proposed in 1974 that an
organized monolayer of D-σ-A’s with appropriate orbital
energy levels should exhibit unidirectional electrical
conductivity, that is, rectification of electricity.3 Such
molecular-sized rectifiers hold promise for the miniatur-
ization of electronic circuits as engineering approaches
the scale of nanotechnology.4
For a monolayer to rectify, its molecules must be
aligned in register between two electrodes, MD and MA
(Figure 1a). This enables the molecules to work together
when electrons flow from electrode MA (acting as cathode)
to A, and exit from D to electrode MD (acting as anode);
the extra electron in A’s LUMO then tunnels through the
-σ- bridge to the vacancy in D’s HOMO to complete the
forward-direction flow. Individual molecules would work
at cross purposes if the D and A groups were out of
register (as in Figure 1b), and no rectification would be
expected. One method commonly used to align D-σ-A
molecules is the Langmuir-Blodgett (LB) technique,5
which requires the molecules to have polar and nonpolar
ends. The D-σ-A part constitutes the polar end, while
an alkyl-chain (lipid) tail typically provides the nonpolar
end (Figure 1c). Monolayers of such compounds, formed
F IGURE 1. Alignment of D-σ-A molecules in monolayers.
spontaneously on an aqueous surface, can be lifted onto
one electrode by careful dipping and withdrawal. Re-
peated treatment builds up multilayers. The second
electrode is applied by evaporative metal techniques.3,4
In the past, our group has used pyrene, trialkoxy-
phenyl, and ferrocene derivatives as donors, TCNQ and
dinitrophenyl derivatives as acceptors, and carbamates
as -σ- bridges.6-9 Multilayers of one of our compounds
with a pyrene donor and a dinitrophenyl acceptor gave
hysteresis-free electrical rectification.10
P er ylen ebis(d ica r boxim id e) a s Accep tor . We have
begun the preparation of a series of D-σ-A targets based
(1) Metzger, R. M.; Panetta, C. A. New J . Chem. 1991, 15, 209-
221.
(6) Musa, A.; Sridharan, B.; Lee, H.; Mattern, D. L. J . Org. Chem.
1996, 61, 5481-5484.
(2) Mallouk, T. E.; Lee, H. J . Chem. Educ. 1990, 67, 829-834.
(3) Aviram, A.; Ratner, M. A. Chem. Phys. Lett. 1974, 29, 277-283.
(4) (a) Metzger, R. M. Acc. Chem. Res. 1999, 32, 950-957. (b)
Metzger, R. M. Chem. Rev. 2003, 103, 3803-3834.
(5) Langmuir-Blodgett Films, 1st ed.; Roberts, G., Ed.; Plenum
Press: New York, 1990.
(7) Metzger, R. M.; Wiser, D. C.; Laidlaw, R. K.; Takassi, A.;
Mattern, D. L.; Panetta, C. A. Langmuir 1990, 6, 350-357.
(8) Nadizadeh, H.; Mattern, D. L.; Singleton, J .; Wu, X. L.; Metzger,
R. M. Chem. Mater. 1994, 6, 268-277.
(9) Lee, H. Ph.D. Dissertation, University of Mississippi, 1997.
10.1021/jo035409w CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/03/2003
10058
J . Org. Chem. 2003, 68, 10058-10066

Donor-σ-Acceptor Molecules
on the good11 acceptor perylene-4,5,9,10-bis(dicarboxim-
ide) (3), which can be prepared in principle from com-
mercially available perylenetetracarboxylic dianhydride
(1) by sequential imidizations.
to encourage LB film formation. We report here the
preparation of several of these D-σ-A-lipid targets.18
We also discuss some unusual features of the NMR
spectra that stem from the swallowtails. The electrical
characteristics of these targets will be reported separ-
ately.
Resu lts a n d Discu ssion
Ar yl a n d F er r ocen yl Don or s. According to the
literature, symmetrical 3’s can be obtained by autoclaving
an aqueous solution19 of the dianhydride 1 and a primary
amine, or by heating them in molten imidazole with a
Lewis acid catalyst.13,20 Subsequent semihydrolysis of the
symmetric 3’s can provide 2’s.21,22 (This route is preferable
to obtaining 2’s by monoimidization of 1, which proceeds
poorly if the R groups are large.21) If a single swallowtail
makes 2 adequately soluble, the conversion to unsym-
metric 3 may be feasible by reflux in an aromatic solvent.
In practice, this procedure worked well in the forma-
tion of 4, where R′ contains a pyrene donor group.
Treatment of 1 with the known 10-nonadecanamine
(prepared by RedAl reduction of 10-nonadecanone oxime13)
gave alkyl-swallowtailed 3a , followed by semihydrolysis
to 2a and condensation of 2a with commercial pyrenyl-
methylamine to give the target 4 (Scheme 1).
In addition to being excellent acceptors, the pery-
lenediimides have another striking advantage: the two
synthetic handles of 1 can be exploited to produce the
unsymmetrically disubstituted diimides (3 with R * R′)
needed for constructing D-σ-A-lipid systems. That is,
one anhydride group of 1 can be used to connect the lipid
tail needed for LB film formation (making monoimide 2),
and the other anhydride group can be used to attach the
-σ- bridge and donor.
Sw a llow ta il Solu bilizer s. Along with the advantages
of perylenediimides, however, comes a major disadvan-
tage. Their poor solubility (typically 1-2 mg/L) limits the
amount of useful chemistry that can be accomplished
with them.12 Conventional 3’s are not even capable of
routine crystallizations. Langhals and co-workers have
improved solubility by attaching 2° carbons in the middle
of long chains.13 These create “swallowtail” diimides (like
the 10-nonadecyl-disubstituted 3a ) which are amazingly
soluble, to as much as 350 000 mg/L in heptane.14 Twin
alkyl chains have similarly been used to improve the
solubility of fluorene derivatives.15
We also prepared, in a similar manner, a D-σ-A
target with a four-carbon σ bridge to pyrene (5); this
should allow us to see the effect of donor-acceptor
A wide variety of unsymmetrical 3’s with a single
swallowtail have been reported,16 many coupled to vari-
ous acceptor groups. Some of these groups can act as
donors upon photoexcitation. Only one, however, with a
pyrene donor and a C13 swallowtail,16f is a D-σ-A
molecule of the type discussed above.17 We have chosen
to use C19 swallowtails, as that length gives the strongest
solubilization, and the longer alkyl chains should help
(10) (a) Martin, A. S.; Sambles, J . R. Nanotechnology 1996, 7, 401-
405. (b) Brady, A. C.; Hodder, B.; Martin, A. S.; Sambles, J . R.; Ewels,
C. P.; J ones, R.; Briddon, P. R.; Musa, A. M.; Panetta, C. A.; Mattern,
D. L. J . Mater. Chem. 1999, 9, 2271-2275.
(11) O’Neil, M. P.; Niemczyk, M. P.; Svec, W. A.; Gosztola, D.;
Gaines, G. L., III; Wasielewski, M. R. Science 1992, 257, 63-65.
(12) Review: Langhals, H. Heterocycles 1995, 40, 477-500.
(13) Demmig, S.; Langhals, H. Chem Ber. 1988, 121, 225-230.
(14) Langhals, H.; Demmig, S.; Potrawa, T. J . Prakt. Chem. 1991,
333, 733-748.
(15) Belfield, K. D.; Schafer, K. J .; Alexander, M. D., J r. Chem.
Mater. 2000, 12, 1184-1186.
(16) For example: (a) Langhals, H.; Gold, J . J . Prakt. Chem. 1996,
338, 654-659. (b) Langhals, H.; J ona, W. Eur. J . Org. Chem. 1998,
847-851. (c) Langhals, H.; J ona, W. Angew. Chem., Int. Ed. 1998, 37,
952-955. (d) Langhals, H.; Ismael, R.; Yu¨ru¨k, O. Tetrahedron 2000,
56, 5435-5441. (e) Zang, L.; Liu, R.; Holman, M. W.; Nguyen, K. T.;
Adams, D. M. J . Am. Chem. Soc. 2002, 124, 10640-10641. (f) Langhals,
H.; Saulich, S. Chem. Eur. J . 2002, 8, 5630-5643.
(18) Preliminary report: Wescott, L.; Mattern, D. L. Abstracts of
Papers; 225th National Meeting of the American Chemical Society;
American Chemical Society: Washington, DC, 2003; ORGN 540.
(19) Nagao, Y.; Misono, T. Bull. Chem. Soc. J pn. 1981, 54, 1191-
1194.
(20) Langhals, H. Chem. Ber. 1985, 118, 4641-4645.
(21) Kaiser, H.; Lindner, J .; Langhals, H. Chem. Ber. 1991, 124,
529-535.
(22) Langhals, H.; Sprenger, S.; Brandherm, M.-T. Liebigs Ann.
1995, 481-486.
(17) For an interesting variation where the D-σ-A donor is a
modified perylenediimide, see: Lukas, A. S.; Zhao, Y.; Miller, S. E.;
Wasielewski, M. R. J . Phys. Chem. B 2002, 106, 1299-1306.
J . Org. Chem, Vol. 68, No. 26, 2003 10059

Wescott and Mattern
SCHEME 1
separation on current flow. Such a long bridge could
conceivably allow the donor and acceptor groups to fold
back and interact intramolecularly, but we have success-
fully used 4- and 5-atom bridges previously.7,8,10 Com-
pound 6 contains a ferrocene donor; the required ferro-
cenylethanamine precursor was prepared by LiAlH4
reduction of ferrocenylacetonitrile.23 Condensation of 2a
with phenethylamine gave 7, which can serve as a control
D-σ-A, since the phenyl group should not exhibit donor
properties.
as 9 that has a primary amine with which to attach the
perylenediimide.
An alkyl group substituted on the TMPDA ring would
be expected to diminish its donor properties somewhat
by forcing the adjacent dimethylamino group out of
planarity. For example, the reduction potential of the
dipiperidinyl TMPDA analogue 10 is reported to be 192
Tetr a m eth ylp h en ylen ed ia m in e Don or . We then
turned to the excellent donor tetramethylphenylenedi-
amine (TMPDA, 8), whose nitrogen lone pairs release
mV vs NHE; the methylated derivative 11 has an E1/2 of
351 mV.26 This still represents a good donor, however,
and even tetraethylphenylenediamines with two ring
substituents have similar low reduction potentials.24f
TMPDA has been directly alkylated by using radical
reactions.27 For example, photochemical activation of
alkyl mercury compounds has been reported to give
TMPDA alkylation, as shown in Scheme 2 for the
formation of 12.28 We attempted a more straightforward
generation of radicals,29 by treating the t-BOC-protected
bromoamine 14 (prepared from 3-bromopropylammonium
bromide and tert-butyl dicarbonate30) with tributyl tin
hydride and AIBN. No reaction of 8 occurred with these
electron density into the aromatic π system. Although
TMPDA itself is readily available, it is more difficult to
obtain derivatives. A handful of ring-alkylated TMPDA’s
have been reported,24 but they lack suitable functional
group handles.25 We sought a TMPDA derivative such
(23) Godillot, P.; Korri-Youssoufi, H.; Srivastava, P.; El Kassmi, A.;
Garnier, F. Synth. Met. 1996, 83, 117-123.
(24) For example, see: (a) Srinivas, S.; Taylor, K. G. J . Org. Chem.
1990, 55, 1779-1786 (alkyl). (b) Russell, G. A.; Guo, D.; Khanna, R.
K. J . Org. Chem. 1985, 50, 3423-3425 (benzyl). (c) Barton, D. H. R.;
Finet, J . P.; Gianotti, C.; Halley, F. J . Chem. Soc., Perkin Trans. 1
1987, 241-249 (phenyl). (d) Nakamura, Y.; Iwamura, H. Bull. Chem.
Soc. J pn. 1993, 66, 3724-3728 (aryl). (e) Schwarzenbacher, G.; Evers,
B.; Schneider, I.; de Raadt, A.; Besenhard, J .; Saf, R. J . Mater. Chem.
2002, 12, 534-539 (ethynyl). (f) Zhou, Q.; Swager, T. M. J . Org. Chem.
1995, 60, 7096-7100 (ethynyl and butoxy).
(26) Pearson, A. J .; Gelormini, A. M. Tetrahedron Lett. 1997, 38,
5123-5126.
(27) Srinivas, S.; Taylor, K. G. J . Org. Chem. 1990, 55, 1779-1786.
(28) Russell, G. A.; Guo, D.; Khanna, R. K. J . Org. Chem. 1985, 50,
3423-3425.
(29) Ono, N.; Yoshida, Y.; Tani, K.; Okamoto, S.; Sato, F. Tetrahe-
dron Lett. 1993, 34, 6427-6430.
(25) For D-σ-A compounds formed from a piperidine-pyrazine
analogue of TMPDA, see: Pearson, A. J .; Gelormini, A. M.; Fox, M.
A.; Watkins, D. J . Org. Chem. 1996, 61, 1297-1305.
(30) Zhang, J .; Drugeon, G.; L’hermite, N. Tetrahedron Lett. 2001,
42, 3599-3601.
10060 J . Org. Chem., Vol. 68, No. 26, 2003

Donor-σ-Acceptor Molecules
tion might proceed through such an intermediate,34 we
did not observe successful iodination.
SCHEME 2
We then followed a longer route as shown in Scheme
3, iodinating 4-nitroaniline with iodine and H2O2 as
described by Toth35 and reducing the nitro group of the
resulting 17 with stannous chloride in hot concentrated
HCl36,37 to make 18, followed by Giumanini and co-
workers’ procedure38 for amino group permethylation
with formaldehyde and sodium borohydride in acidic
aqueous THF to give the desired iodo-TMPDA, 16. This
compound was found to be a liquid despite the presence
of the iodine. The iodine is expected to force the adjacent
dimethylamino group out of planarity with the benzene
ring. This is consistent with aromatic NMR chemical
shifts which are about 0.5 ppm downfield compared to
simple predictions for 16; with its resonance capabilities
diminished, the nonplanar amino group apparently ex-
erts a more electron-withdrawing shielding effect than
it would if it were planar.
Heck reaction of 16 with acrylonitrile, palladium(II)
acetate, triethylamine, and triphenylphosphine in a hot
sealed tube gave a mixture of cis and trans alkenes 19.
The alkene mixture was saturated by catalytic hydroge-
nation with Pd/C in a Parr shaker to give the cyanoethyl
compound 20. Attempts to reduce the cyano group
simultaneously with the double bond were not successful,
but it was easily reduced in a separate step with LiAlH4
in ether to give the desired 9, which condensed easily
with imide-anhydride 2a in refluxing toluene to make
the target compound 21.
reagents in refluxing benzene or toluene, so we converted
14 into the more reactive iodide 15 with sodium iodide
in acetone and tried again. However, no reaction was
apparent by TLC analysis.
We therefore turned to an indirect strategy, a pal-
ladium-catalyzed ring alkylation of iodinated TMPDA.
Iodo-TMPDA (16) has been reported in the literature,24d
Di-Sw a llow t a iled P er ylen ed iim id e Con for m a -
tion . The proton NMR spectrum of 3a showed two
signals, well separated at 1.8 and 2.2 ppm, for the eight
methylene hydrogens â to nitrogen in the swallowtail
chains. The literature14 spectrum for 3a reports only one
signal at 2.1 ppm for these methylenes, although the
literature21,22 spectrum for the imide-anhydride 2a shows
two signals for its corresponding four methylene hydro-
gens (in agreement with our findings for 2a ). We thought
that a conformational restriction might be making the
two methylenes of 3a nonequivalent, but a sample
warmed to 105 °C in pyridine-d5 showed no changes in
1
any swallowtail H signals. Upon further consideration,
we realized that the two hydrogens in each â methylene
are in fact diastereotopic, which accounts for their
distinct signals. One can think of these hydrogens as
being in exo and endo positions in the MM2-minimized
extended-chain conformation shown for 3a ′; the extended
swallowtails are based on the crystal structure reported
for 3 with C17 swallowtails.14 The corresponding â carbons
are enantiotopic and give a single NMR signal at 32.0
ppm; this signal shows the expected HMQC correlation
to both methylene H signals.
where it was used in a Pd(0) coupling to an aryl zinc
species, but no information about its preparation or
physical properties was given.
We first attempted to prepare 16 by direct iodination
of TMPDA with I-Cl, following a literature procedure
for 4-nitroaniline.32 In cold acetic acid, at least three
products were apparent in TLC analysis; these were
poorly separated by column chromatography, and no
upfield signal diagnostic for iodinated aromatic carbon
was apparent in the 13C NMR spectrum. With 2 equiv of
H2SO4 added to protonate the amino groups and moder-
ate the process, no reaction was apparent, and in CCl4
there was no reaction at room temperature and tar
formation upon heating. In all cases, a blue-violet color
formed, suggesting some oxidation by I-Cl of TMPDA
to its radical cation, Wurster’s blue.33 Although substitu-
If 3a were completely symmetrical, its four carbonyls
would be identical. Instead, the 13C NMR spectrum of 3a
(33) Michaelis, L.; Schubert, M. P.; Granick, S. J . Am. Chem. Soc.
1939, 61, 1981-1992.
(34) Perrin, C. J . Am. Chem. Soc. 1977, 99, 5516-5518.
(35) Toth, T. Helv. Chim. Acta 1971, 54, 1486-1487.
(36) Nicolet, B. H.; Ray, W. L. J . Am. Chem. Soc. 1927, 49, 1801-
1806.
(37) Lindsay, D. M.; Dohle, W.; J ensen, A. E.; Kopp, F.; Knochel, P.
Organic Lett. 2002, 4, 1819.
(31) Reference deleted in press.
(32) Willgerodt, C.; Arnold, E. Chem. Ber. 1901, 34, 3343-3354.
(38) Giumanini, A. G.; Chiavari, G.; Musiani, M. M.; Rossi, P.
Synthesis 1980, 743-746.
J . Org. Chem, Vol. 68, No. 26, 2003 10061

Wescott and Mattern
SCHEME 3
F IGURE 2. MM2 steric energy of 22 during C-N bond
rotation.
showed two carbonyl signals (164.5 and 163.5 ppm), in
agreement with the literature.14 Further, our 13C NMR
spectrum of 2a showed three carbonyls: one sharp (160.0
ppm) and two broad and weak (163.3 and 164.4 ppm).
Literature reports21,22 for 2a appear to have missed the
weak signals, but this pattern is characteristic of our
mono-swallowtailed products. The doubling of carbonyl
signals from swallowtailed imides suggests that the
swallowtails assume an asymmetrical conformation.
Langhals and co-workers39 have studied the rotational
barrier about the C-N bond of several swallowtailed 3’s
by line shape analysis of the carbonyl 13C NMR signals.
Coalescence temperatures vary depending on the nature
of the swallowtail; rotational barriers are about 15 kcal/
mol. To explore the conformational barrier, we performed
MM2 calculations on 22 as a model for the swallowtail
imide, starting with its swallowtail in the extended, all-
anti conformation, and using a driver to rotate the C-N
bond 360°. Lower energy conformations were found
between 50° and 70° by introducing R,â gauche kinks in
the swallowtail. Figure 2 shows the lowest energy found
at each C-N rotation angle through 90° (further rotation
gives equivalent conformations). The rotation barrier was
11.8 kcal/mol, in reasonable agreement with the experi-
mental barrier. The asymmetry of the low-energy con-
formation (at 0° rotation) accounts for the distinct 13C
signals: one carbonyl is eclipsed with the R hydrogen,
while the other carbonyl is situated between the two
swallowtail alkyls.
The doubling of carbon signals at low temperature and
their exchange at higher temperature call to mind the
(39) Langhals, H.; Demmig, S.; Huber, H. Spectrochim. Acta 1988,
44A, 1189-1193.
10062 J . Org. Chem., Vol. 68, No. 26, 2003

Donor-σ-Acceptor Molecules
SCHEME 4
TABLE 1. Ar om a tic a n d Ca r bon yl 13C NMR Sign a ls a n d Assign m en ts for Sw a llow ta iled P er ylen ed iim id es
av lit.a aromatic
lit.a assignment
of 13C signals
from 3’s
13C signals
for 3a ,
rt (ppm)
13C signals
for 3a ,
60 °C (ppm)
assignment
of 13C signals
from 3a
13C signals
for 3’s (ppm)
164.0
134.4
131.4
129.4
126.3
123.5
122.9
3
3a
4
14b
5a
14c
5
164.5 (bb), 163.5 (b)
134.3
131.8 (b), 131.0 (b)
129.5
126.3
124.0 (b), 123.2 (b)
122.9
164.2 (b)
134.7
131.6 (b)
129.9
126.7
123.9 (b)
123.1
3
5a
4
14b
14c
3a
5
a
b
Reference 39. These signals were broad and weak.
atropisomerism reported by Langhals and co-workers for
bis(di-tert-butylphenyl)-substituted 23.13,20 In that case,
C-N bond rotation would result in interconversion of cis
and trans isomers (Scheme 4). At room temperature,
however, the system is too sterically hindered to rotate,
and standard preparations make a stable mixture of the
cis and trans diastereomers. In principle, the two isomers
could give distinct NMR spectra. However, Langhals
skeleton, the three signals representing the eight interior
carbons were doubled, which seems reasonable since they
are between the interior tert-butyls whose relative posi-
tions change during isomerization. The average differ-
ences in chemical shifts for doubled carbons were 0.05
ppm for perylene skeleton carbons, 0.03 ppm for phenyl,
and 0.01 ppm for tert-butyl. In contrast to 3a , the four
carbonyls in either isomer of 23 are equivalent, because
the preferred conformation about each imide C-N bond
is essentially perpendicular. The carbonyl signals for the
two isomers of 23 were congruent.
1
reported only tiny differences for two of the H signals,20
and reported single values for each of the mixture’s 13C
positions.21 We took 125-MHz spectra of 23 and were able
to distinguish 8 of the 17 carbon signals as distinct in
the two isomers, giving doubled signals.
In principle, an isomerization like 23’s could account
for signal doubling in swallowtailed 3a , since C-N
rotation interconverts diastereomers 3a ′ and 3a ′′ (Scheme
4). However, such isomerism is not possible for single-
swallowtailed imides such as 2a , which also show car-
bonyl doubling. The doubling in swallowtailed imides is
therefore best viewed as a local phenomenon caused by
the asymmetric environment occurring near each swal-
lowtail.
We were able to verify the literature peak assign-
ments21 in the phenyl rings with the help of DEPT,
HMQC, and HMBC spectra, except for two signals:
HMBC coupling from the H6′ singlet at 7.02 ppm to a
CH peak at 126.6 ppm showed that it must be C4′ rather
than C6′, and the return HMBC coupling from H4 to
doubled CH peaks at 127.8 ppm showed that they cannot
be C4′ but must be C6′ instead.40 Assignments in the
perylene ring were made as discussed below for 3a , a
system which enables a final ambiguity to be resolved.
With assignments in hand, the doubled signals could
be mapped onto the structure of 23, as indicated by the
dots in Scheme 4. In the phenyl groups, the more exterior
parts (the C5′ tert-butyl group and C4′) were not affected.
Interestingly, the interior tert-butyl carbon and its ring
attachment C2′ were also unaffected, even though the
carbons surrounding them were doubled. In the perylene
This asymmetry should extend beyond the carbonyl
carbon, and in fact we see doubling of two other carbon
signals in 3a at room temperature. Langhals and co-
workers also saw additional doublings at low tempera-
ture for 3’s with short swallowtails.39 The six aromatic
13C signals of 3 were assigned by the Langhals group39
as shown in Table 1; these are analogous to their
assignments for 23.21 We attempted to confirm those
assignments in 3a , but satisfactory HMBC spectra were
unattainable at room temperature because the expected
correlations from H4 were absent. At 60 °C, however, the
13C spectrum simplified, with the three sets of doubled
peaks becoming broad singlets, and the 3-bond HMBC
(40) The numbering scheme used is the Chemical Abstracts num-
bering for 3, anthra[2,1,9-def:6, 5, 10-d′e′f′]diisoquinoline-1,3,8,10-
(2H,9H)-tetrone.
J . Org. Chem, Vol. 68, No. 26, 2003 10063

Wescott and Mattern
directly purified by column chromatography on silica gel
(CHCl3) to give 930 mg (66% yield) of a deep red solid (mp
correlations from H4 were revealed. HMBC correlation
of the collapsed carbonyl to the downfield (δ 8.5) H
identified it as H4; HMQC correlations to H4 and H5 then
identified C4 and C5. Only C5a was 3-bond-coupled to both
H4 and H5 (in the latter case, it was not the proximal C5a
but its adjacent twin C5b, identical by symmetry, that was
so coupled). C14b was HMBC-coupled to H4 only. The
remaining two carbon signals, C14c and C3a, both showed
HMBC coupling to H5. However, C14c, situated in the
middle of the structure, should experience the same
environment for both rotamers of 3a , and therefore was
distinguished as the signal that was not doubled at room
temperature; this is the distinction that could not be
made by looking only at 23. By this analysis, three of
the perylenediimide aromatic 13C signals have been
misassigned in the literature (Table 1).
With structural assignments in hand, we concluded
that the three doubled signals in 3a represent the
carbons that are closest to the restricted C-N bond; these
are dotted in Scheme 4. Note that C14b, also close to C-N,
should nonetheless be undoubled because it has the same
environment in the two rotamers. We discerned similar
patterns of two doubled carbons in all swallowtailed
derivatives except 4 and 5, where apparent overlaps
obscured expected peaks at 123.2 and 131.0 ppm, respec-
tively. The long swallowtails of 3a appear to significantly
slow C-N rotation, since 3’s with shorter swallowtails
show doubling only at lower temperature.39
Consistent with the problematic HMBC at room tem-
perature was the unusual appearance of the aromatic
protons: H5 was a sharp doublet, but H4 was quite broad.
We attribute the H4 broadening to the influence of the
nearby C-N rotamers, whereas H5, being more distant,
escapes the effect. At 60 °C, H4 becomes a sharp doublet.
Similar spectra are obtained for other swallowtailed
peryleneimides. For example, concentrated samples of 7
at 60 °C show four resolved, sharp doublets for the four
aromatic pairs of perylene protons; however, at room
temperature, the low-field signal broadens into an un-
resolved lump.
1
97-99 °C, lit.14 mp 97 °C). H NMR (CDCl3) δ 0.79 (t, 12H),
1.21 (m, 56 H), 1.84 (m, 4H), 2.23 (m, 4H), 5.16 (m, 2H), 8.51
(d, 4H), 8.59 (br d, 4H). 13C NMR (CDCl3, 77.16) δ 14.2, 22.8,
27.1, 29.4, 29.67, 29.68, 32.0, 32.5, 55.1, 122.9, 123.2, 124.0,
126.3, 129.5, 131.0, 131.8, 134.3, 163.5, 164.5. IR (PFE/PE
cards) 2920 (m), 2852 (m), 1694 (m) cm-1
.
N-(10-Non adecyl)-3,4,9,10-per ylen etetr acar boxylic Acid
3,4-An h yd r id e-9,10-im id e (2a ).21,22 A mixture of 0.24 g (0.26
mmol) of 3a , 60 mg (0.90 mmol) of 85% KOH pellets, and 13
mL of t-BuOH was brought to reflux. After 30 min, an
additional 2 pellets of KOH was added and heating was
continued for an additional hour. The mixture was poured,
with stirring, into a mixture of 13 mL of AcOH and 7 mL of 2
N HCl. The resulting red-brown precipitate was concentrated
by centrifugation, washed twice with water, and dried at 80
°C. Chromatography on silica gel (90% CHCl3/10% AcOH)
yielded 94 mg (53% yield) of 2a . 1H NMR (CDCl3) δ 0.80 (t,
6H), 1.20 (m, 28H), 1.85 (m, 2H), 2.21 (m, 2H), 5.15 (m, 1H),
8.55 (m, 6H), 8.65 (br, 2H). 13C NMR (CDCl3) δ 14.2, 22.8, 27.1,
29.4, 29.7, 32.0, 32.4, 55.0, 118.9, 123.2, 124.0, 124.8, 126.4,
126.7, 129.5, 131.3, 131.8, 132.0, 133.6, 136.3, 160.0, 163.3,
164.4. IR (PE card) 1775 (s), 1732 (m), 1703 (m), 1661 (m) cm-1
.
N-(10-Non a d ecyl)-N′-(1-p yr en ylm eth yl)p er ylen e-3,4,9,-
10-bis(d ica r boxim id e) (4). To 0.10 g (0.37 mmol) of 1-pyrene-
methylamine hydrochloride in 10 mL of MeOH was added,
with stirring, 3 pellets of 85% KOH. An equal volume of water
was added and the cloudy mixture was extracted with meth-
ylene chloride, dried over MgSO4, and concentrated by rotary
evaporation. The resulting free base was dissolved in 10 mL
of toluene containing 85 mg (0.13 mmol) of 2a . The mixture
was brought to reflux, held for 85 min, and then allowed to
stand at room temperature overnight. The resulting precipitate
was filtered and purified by chromatography on silica gel (90%
CHCl3/10% AcOH) to yield 130 mg of crude product. This
material was rechromatographed and then crystallized from
chloroform/hexanes to yield 30 mg (27%) of red solid, mp 275-
290 °C. 1H NMR (CDCl3) δ 0.80 (t, 6H), 1.20 (m, 28H), 1.84
(m, 2H), 2.23 (m, 2H), 5.17 (m, 1H), 6.07 (s, 2H), 7.88 (m, 3H),
8.04 (m, 5H), 8.42 (m, 4H), 8.58 (m, 5H). 13C NMR (CDCl3) δ
14.2, 22.8, 27.2, 29.4, 29.7, 32.0, 32.5, 41.6, 55.0, 123.0, 123.2,
124.1, 124.7, 124.8, 124.9, 125.2, 125.3, 126.0, 126.1, 126.3,
126.4, 127.3, 127.4, 127.9, 128.9, 129.4, 129.5, 130.2, 130.75,
130.82, 131.1, 131.3, 131.8, 131.9, 134.2, 134.8, 163.4 (w),
163.8, 164.7 (w). IR (PE card) 1694 (s), 1658 (s), 1651 (s), 1594
(s) cm-1. Anal. Calcd for C60H58N2 O4‚H2O: C, 81.04; H, 6.80;
N, 3.15. Found: C, 81.22; H, 6.56; N, 3.20.
Con clu sion
Five new D-σ-A targets were prepared with use of
perylenetetracarboxylic diimide as the acceptor. The 10-
nonadecyl swallowtail attached to one of the imide
nitrogens imparted ample solubility to make all of the
products amenable to chromatographic purification, while
providing a lipid tail needed to promote LB film forma-
tion. Donor groups attached to the second imide nitrogen
included a novel TMPDA derivative that was prepared
with a 1° amine synthetic handle. Published NMR data
for 2a , 3a , and 23 were revisited, and the effects of
restricted bond rotation on the spectra were clarified.
Film formation and electrochemical studies of these
products will be reported elsewhere.
Gen er a l P r oced u r e for F or m a tion of Mixed Diim id es.
A mixture of 2a with a 2- to 4-fold molar excess of the
appropriate amine was refluxed in toluene for 1 h. The reaction
mixture was then cooled, washed with 5% HCl followed by 10%
K2CO3, dried over MgSO4, and concentrated by rotary evapo-
ration. Crude products were purified by chromatography on
silica gel with CHCl3 as the eluent to give red solids. Melting
points were difficult to obtain due to the deep color of the
materials and, in some cases, to apparent liquid crystalline
behavior. The values given are where phase changes were first
observed.
N-(10-Non adecyl)-N′-(4-[1-pyr en yl]bu tyl)per ylen e-3,4,9,-
10-bis(d ica r boxim id e) (5). 5 was prepared from pyrenebu-
tylamine (Toronto Research Chemicals) in 52% yield; mp 255-
258 °C. 1H NMR (CDCl3) δ 0.81 (t, 6H), 1.19 (m, 28H), 1.84
(m, 2H), 1.99 (m, 4H), 2.26 (m, 2H), 3.31 (m, 2H), 4.23 (m,
2H), 5.18 (m, 1H), 7.9 (m, 8H), 8.14 (d, 1H), 8.45 (m, 6H), 8.63
(br, 2H). 13C NMR (CDCl3) δ 14.2, 22.8, 27.2, 28.2, 29.2, 29.4,
29.73, 29.76, 32.0, 32.5, 33.5, 40.3, 54.9, 122.2, 122.3, 122.4,
122.8, 123.0, 123.6, 124.36, 124.44, 124.47, 124.50, 124.7,
125.3, 126.1, 126.9, 127.2, 127.3, 127.99, 128.04, 128.4, 129.0,
129.4, 129.9, 130.4, 130.6, 130.9, 131.4, 133.6, 133.7, 136.2,
163.1, 163.6 (w), 164.6 (w). IR (PE card) 1691 (s), 1658 (s),
Exp er im en ta l Section
N,N′-Di(10-n on a d ecyl)p er ylen e-3,4,9,10-bis(d ica r box-
im id e) (3a ).14 A mixture of 606 mg (1.55 mmol) of 3,4,9,10-
perylenetetracarboxylic dianhydride (1), 1.07 g (3.78 mmol) of
10-nonadecanamine, 3.0 g of imidazole, and 214 mg (1.16
mmol) of zinc acetate was heated, with stirring, at 160 °C for
2 h. The mixture was then cooled, dissolved in CHCl3, and
10064 J . Org. Chem., Vol. 68, No. 26, 2003

Donor-σ-Acceptor Molecules
1650 (s), 1596 (m) cm-1. Anal. Calcd for C63H64N2O4: C, 82.86;
H, 7.06; N, 3.07. Found: C, 82.50; H, 7.21; N, 3.02.
CHCl3. The organic layer was dried over MgSO4 and concen-
trated by rotary evaporation. The residue was crystallized from
CHCl3 to give 13.3 g (85%), mp 110-112 °C (lit.35 mp 115 °C,
105 °C). 1H NMR (CDCl3) δ 4.90 (br s, 2 H), 6.67 (d, J ) 15
Hz, 1 H), 8.02 (d × d, J ) 15 Hz, J ) 4 Hz, 1 H), 8.52 (d, J )
4 Hz, 1 H).
2-Iod o-1,4-ben zen ed ia m in e (18). A slurry of 17 (5.27 g,
20.0 mmol) in 27 mL of conc HCl was warmed in a hot water
bath. To this slurry was added, over 9 min and with stirring,
a solution of 15.8 g (83.3 mmol) of SnCl2 in 27 mL of conc HCl.
After the solid dissolved and the yellow color diminished, the
reaction was allowed to cool for 15 min then was placed on
ice. NaOH (50% aq) was added until the mixture was strongly
basic. The resulting thick mixture was cooled to 5 °C and
filtered. The filtrand was crystallized from water to give 3.03
g (65%), mp 115-116 °C (lit.36 mp 110-111 °C). 1H NMR
(CDCl3) δ 3.30 (br s, 2 H), 3.68 (br s, 2 H), 6.57 (m, 2 H), 7.04
(d, J ) 4 Hz, 1 H). 13C NMR (DMSO-d6) δ 85.3, 116.0, 116.5,
123.6, 139.0, 141.1.
N-(10-Non adecyl)-N′-(2-fer r ocen yleth yl)per ylen e-3,4,9,-
10-bis(d ica r boxim id e) (6). The intermediate 2-ferrocenyl-
ethylamine was prepared as reported by Godillot et al.23
A
mixture of 85 mg of LiAlH4 in 5 mL of anhydrous Et2O was
combined with a solution of 296 mg of anhydrous AlCl3 in 7.5
mL of Et2O. To this was added, with stirring, a slurry of 500
mg of ferroceneacetonitrile in 5 mL of Et2O. After 1 h a small
amount of water was carefully added, followed by 7 mL of 6 N
sulfuric acid and 5 mL of water. The ether layer was separated
and the aqueous layer was extracted with Et2O, basified with
KOH pellets, diluted with 30 mL of water, and reextracted.
The resulting ether layer was dried over MgSO4 and concen-
trated by rotary evaporation to give 146 mg (29% yield) of
amine, which was used without further purification. 1H NMR
(CDCl3) δ 1.40 (br s, 2H), 2.44 (t, 2H), 2.76 (t, 2H), 4.04 (m,
9H).
The diimide was prepared from 2-ferrocenylethylamine in
79% yield. The analytical sample was further purified by
recrystallization from toluene and washing with hexanes; mp
214-216 °C. 1H NMR (CDCl3) δ 0.80 (t, 6H), 1.18 (m, 28H),
1.86 (m, 2H), 2.22 (m, 2H), 2.74 (m, 2H), 4.10 (s, 2H), 4.20 (s,
7H), 4.36 (m, 2H), 5.15 (m, 1H), 8.55 (m, 8H). 13C NMR (CDCl3)
δ 14.2, 22.8, 27.2, 27.9, 29.4, 29.7, 32.0, 32.5, 41.3, 55.0, 67.6,
68.3, 68.8, 79.3, 85.2, 123.0, 123.1, 123.5, 124.2, 126.2, 126.4,
126.8, 129.3, 129.5, 131.1, 131.3, 131.9, 134.2, 134.6, 163.2,
163.6 (w), 164.6 (w). IR (PE card) 1695 (s), 1658 (s), 1650 (s),
1596 (m) cm-1. Anal. Calcd for C55H60N2O4Fe: C, 76.02 H, 6.96;
N, 3.22. Found: C, 76.06; H, 7.20; N, 3.05.
2-Iod o-N,N,N′,N′-tetr a m eth yl-1,4-ben zen ed ia m in e (16).
A slurry of 18 (1.00 g, 4.27 mmol) and NaBH4 (2.28 g, 60.2
mmol) in 35 mL of THF was added dropwise to a stirred
solution of 7 mL of 3 M H2SO4 and 5 mL of 37% formalin in
35 mL of THF. The pH rose during the addition, and more
sulfuric acid was added periodically to maintain the pH near
1-2. After 2 h, an additional 5 mL of formalin was added and
the reaction was allowed to continue stirring overnight. Solid
KOH was then added to raise the pH above 10. The organic
layer was separated and the aqueous layer extracted twice
with ether. The combined organic layers were dried over
MgSO4 and concentrated by rotary evaporation to yield an oil,
which was purified by column chromatography (CHCl3) to yield
N-(10-Non a d ecyl)-N′-(2-p h en yleth yl)p er ylen e-3,4,9,10-
b is(d ica r b oxim id e) (7). 7 was prepared from phenethyl-
1
1
amine in 72% yield, mp 225-228 °C. H NMR (CDCl3) δ 0.80
a light yellow oil, 710 mg (57%). H NMR (CDCl3) δ 2.66 (s, 6
(t, 6H), 1.18 (m, 28H), 1.87 (m, 2H), 2.23 (m, 2H), 3.02 (m,
2H), 4.34 (m, 2H), 5.15 (m, 1H), 7.29 (m, 5H), 8.20 (d, 2H),
8.28 (d, 2H), 8.35 (d, 2H), 8.48 (br, 2H). 13C NMR (CDCl3) δ
14.2, 22.8, 27.2, 29.4, 29.7, 32.0, 32.5, 34.3, 42.0, 55.0, 122.96,
123.05, 123.4, 124.1, 125.9, 126.2, 126.3, 126.7, 128.7, 129.2,
129.3, 129.5, 131.0, 131.3, 131.8, 132.3, 134.2, 134.6, 138.8,
163.2, 163.5 (w), 164.6 (w). IR (PE card) 1698 (s), 1651 (s),
1594 (m) cm-1. Anal. Calcd for C51H56N2O4: C, 80.49; H, 7.42;
N, 3.68. Found: C, 80.20; H, 7.12; N, 3.53.
ter t-Bu tyl (3-Br om op r op yl)ca r ba m a te (14). To a sus-
pension of 3-bromopropylammonium bromide (13, 1.00 g, 4.57
mmol) in 20 mL of CHCl3 was added NaHCO3 (390 mg, 4.64
mmol) in 15 mL of water, di-tert-butyl dicarbonate (1.00 g, 4.58
mmol) in CHCl3, and KBr (1.10 g, 9.24 mmol). The mixture
was refluxed for 18 h and then cooled. The layers were
separated and the aqueous layer was extracted twice with
CHCl3. The combined organic layers were dried over MgSO4
and concentrated by rotary evaporation to yield 1.06 g of 14
(96%). 1H NMR (CDCl3) δ 1.41 (s, 9 H), 2.02 (m, 2 H), 3.23 (m,
2 H), 3.41 (t, 2 H), 4.65 (br s, 1 H), in agreement with the
literature41 spectrum.
H), 2.87 (s, 6 H), 6.70 (d × d, J ) 14 Hz, J ) 5 Hz, 1 H), 7.00
(d, J ) 15 Hz, 1 H), 7.20 (d, J ) 5 Hz, 1 H). 13C NMR (CDCl3)
δ 41.0, 45.8, 99.9, 113.7, 120.7, 123.7, 144.8, 148.5. Anal. Calcd
for C10H15N2I: C, 41.40; H, 5.21; N, 9.65. Found: C, 41.10; H,
5.43; N, 9.29.
2-(2-Cyan oeth en yl)-N,N,N′,N′-tetr am eth yl-1,4-ben zen e-
d ia m in e (19). Into a thick-walled Pyrex tube were placed 1.13
g of 16 (3.90 mmol), acrylonitrile (3.28 g, 61.9 mmol), Et3N
(0.59 g, 5.8 mmol), Pd(OAc)2 (40 mg, 0.16 mmol), and PPh3
(0.32 g, 1.3 mmol). The tube was closed with a Teflon plug
and placed in a steam bath overnight. After cooling, water was
added to the mixture, and it was extracted three times with
ether. The organic layers were washed with saturated brine,
dried over MgSO4, and concentrated by rotary evaporation.
The crude product was purified by column chromatography
(CHCl3) to yield a yellow-orange oil, 600 mg (72%), which NMR
analysis indicated to be a 2:1 mixture of E:Z diastereomers.
1H NMR (CDCl3) δ 2.62 (s, 6 H), 2.90 (s, 3 H), 2.93 (s, 3 H),
5.39 (d, J ) 20 Hz, 0.3 H), 5.88 (d, J ) 28 Hz, 0.7 H), 6.71 (d,
J ) 5 Hz, 0.7 H), 6.82 (m, 1 H), 7.03 (m, 1 H), 7.44 (d, J ) 5
Hz, 0.3 H), 7.62 (d, J ) 20 Hz, 0.3 H), 7.85 (d, J ) 28 Hz, 0.7
H). 13C NMR (CDCl3) δ 41.3, 45.8, 95.3, 110.8, 117.1, 119.3,
119.8, 120.5, 128.5, 128.8, 147.5, 149.2. IR (PE card) 2214 (m)
ter t-Bu tyl (3-Iod op r op yl)ca r ba m a te (15). A solution of
560 mg (2.35 mmol) of 14 and 400 mg (2.67 mmol) of NaI in
acetone was stirred for 16 h at room temperature. The mixture
was then filtered and the filtrate was concentrated by rotary
evaporation. The residue was dissolved in CHCl3, washed with
Na2S2O5, dried over MgSO4, and concentrated by rotary
cm-1
.
2-(2-Cya n oeth yl)-N,N,N′,N′-tetr a m eth yl-1,4-ben zen ed i-
a m in e (20). Into a Parr hydrogenation flask were placed 600
mg of 19 (2.79 mmol), 50 mL of EtOH, and a catalytic amount
of 10% Pd/C. After the mixture was shaken overnight under
pressurized H2, the catalyst was filtered off and the EtOH was
concentrated by rotary evaporation to yield 410 mg (67%) of a
colorless oil, which was used without purification in the next
step. 1H NMR (CDCl3) δ 2.57 (s, 6 H), 2.67 (t, 13 Hz, 2 H),
2.89 (s, 6 H), 2.98 (t, J ) 13 Hz, 2 H), 6.56 (d, 5 Hz, 1 H), 6.62
1
evaporation to yield 230 mg of 15 (34%). H NMR (CDCl3) δ
1.41 (s, 9 H), 1.97 (m, 2 H), 3.16 (m, 4 H), 4.62 (br s, 1 H), in
agreement with the literature42 spectrum.
2-Iod o-4-n it r oa n ilin e (17). A mixture of iodine (8.09 g,
31.9 mmol), p-nitroaniline (8.08 g, 58.5 mmol), chlorobenzene
(6 mL), and water (25 mL) was heated to 90 °C. Hydrogen
peroxide (3% aq, 77 g) was added dropwise over a 15-min
period. The mixture was heated to reflux for 25 h, and then
was cooled, basified with NaOH pellets, and extracted with
(d × d, J ) 14 Hz, J ) 5 Hz, 1 H), 7.09 (d, J ) 14 Hz, 1 H). 13
C
NMR (CDCl3) δ 18.5, 28.6, 41.1, 46.2, 112.6, 114.4, 120.2, 121.9,
134.9, 143.0, 148.0. IR (PE card) 2247 (w) cm-1. The analytical
sample was purified by successive column chromatographies,
first with 100:4:1 CHCl3:EtOH:NH3(aq), then with 100:1
CHCl3:NH3(aq) (dried over MgSO4), and finally with CH2Cl2.
(41) Lee, B. H.; Miller, M. J . J . Org. Chem. 1983, 48, 24-31.
(42) Ensch, C.; Hesse, M. Helv. Chim. Acta 2002, 85, 1659-1673.
J . Org. Chem, Vol. 68, No. 26, 2003 10065

Wescott and Mattern
Anal. Calcd for C13H19N3: C, 71.85; H, 8.81; N, 19.34. Found:
C, 71.62; H, 8.88; N, 19.27.
129.5, 129.7, 131.2, 131.5, 132.0, 134.5, 134.7, 137.6, 143.3,
147.6, 163.5, 163.7 (w), 164.8 (w). IR (PE card) 1696 (s), 1659
(s), 1596 (s) cm-1. Anal. Calcd for C56H68N4O4‚2H2O: C, 74.97;
H, 8.09; N, 6.24. Found: C, 75.19; H, 8.18; N, 6.08.
2-(3-Am in op r op yl)-N,N,N′,N′-tetr a m eth yl-1,4-ben zen e-
d ia m in e (9). To a stirred suspension of LiAlH4 (512 mg, 12.8
mmol) in 20 mL of anhydrous ether was added, over a 5-min
period, 410 mg of 20 (1.89 mmol) in 10 mL of anhydrous ether.
After an additional 15 min, 35 mL of 15% NaOH was added,
and the mixture was extracted with ether. The organic layer
was dried over MgSO4 and concentrated by rotary evaporation
to yield 360 mg (86%) of an oil that was used without
N,N′-Bis-(2,5-d i-ter t-bu tylp h en yl)p er ylen e-3,4,9,10-bis-
(d ica r boxim id e) (23).20,21 The commercial material was a
mixture of cis and trans isomers, containing residual quinoline.
1H NMR (CDCl3) δ 1.29 (18H, s, C2′-t-Bu), 1.32 (18H, s, C5′-
t-Bu), 7.02 (2H, s, H6′), 7.46 (2H, d, H4′, J ) 8.4 Hz), 7.59 (2H,
d, H3′, J ) 8.5 Hz), 8.70 (4H, m, H5), 8.76 (4H, m, H4). 13C
NMR (CDCl3) δ 31.37 (C5′-Me); 31.88, 31.89 (C2′-Me); 34.41
(C5′-t-Bu); 35.69 (C2′-t-Bu); 123.45 (CH, C5); 123.83 (C3a);
126.56 (CH, C4′); 126.94, 126.98 (C14c); 127.79, 127.82 (CH, C6′);
128.97, 128.99 (CH, C3′); 130.01, 130.06 (C14b); 132.10 (CH, C4);
132.67, 132.69 (C1′); 135.16, 135.21 (C5a); 143.88 (C2′); 150.30,
150.34 (C5′); 164.58 (CdO).
1
purification in the next step. H NMR (CDCl3) δ 1.52 (br s, 2
H), 1.75 (m, 2 H), 2.59 (s, 6 H), 2.68 (m, 4 H), 2.87 (s, 6 H),
6.57 (s + d, 2 H), 7.04 (d × d, J ) 12 Hz, J ) 3 Hz, 1 H). 13C
NMR (CDCl3) δ 27.9, 35.2, 41.3, 41.7, 46.1, 111.5, 114.6, 120.7,
138.3, 143.4, 147.8.
N-(9-Non a d ecyl)-N′-(3-[2-(N′′,N′′,N′′′,N′′′-t et r a m et h yl-
1,4-b en zen ed ia m in yl)p r op yl])p er ylen e-4,5,9,10-b is(d i-
ca r boxim id e) (21). A solution of 2a (300 mg, 0.457 mmol)
and 9 (300 mg, 1.36 mmol) in 15 mL of toluene was refluxed
for 4 h. The mixture was cooled, concentrated by rotary
evaporation, and purified by column chromatography (CHCl3,
then 9:1 CHCl3:EtOH) to give 320 mg (80%) of a deep red, waxy
MM2 ca lcu la tion s were performed with Chem3D Pro 4.0.
Ack n ow led gm en t. We thank the National Science
Foundation, grant no. DMR-0099674, for financial sup-
port.
1
solid, mp 64-67 °C. H NMR (CDCl3) δ 0.80 (t, J ) 11 Hz, 6
Su p p or tin g In for m a tion Ava ila ble: Procedure for the
preparation of 10-nonadecanamine; IR data for new prepared
compounds; 1H NMR and 13C NMR spectra for 2a ; HMBC
spectra for 3a (aromatic region) and 23. This material is
available free of charge via the Internet at http://pubs.acs.org.
H), 1.17 (m, 28 H), 1.84 (m, 2 H), 2.16 (m, 4 H), 2.58 (s, 6 H),
2.83 (s + m, 8 H), 4.29 (m, 2 H), 5.16 (m, 1 H), 6.52 (d × d, J
) 15 Hz, J ) 5 Hz, 1 H), 6.66 (d, 5 Hz, 1 H), 7.00 (d, 14 Hz, 1
H), 8.61 (m, 8 H). 13C NMR (CDCl3) δ 14.2, 22.8, 27.1, 28.7,
28.8, 29.4, 29.7, 31.1, 32.0, 32.5, 41.0, 41.3, 46.0, 54.9, 111.5,
114.4, 120.8, 123.1, 123.2, 123.3, 123.4, 124.1, 126.5, 126.6,
J O035409W
10066 J . Org. Chem., Vol. 68, No. 26, 2003