Naphthalenediimide dimers and trimers form self-assembling hydrogen-bonded nanotubes of enhanced stability

Article abstract of DOI:10.1080/10610278.2012.731062

Covalent linkage of amino acid-functionalised naphthalenediimides (NDIs) produced a dimer and a trimer that form helical, hydrogen-bonded nanotubes in CHCl₃ solutions and that are more resistant than an analogous monomer to the hydrogen-bond di

Full text of DOI:10.1080/10610278.2012.731062

This article was downloaded by: [University of Arizona]
On: 10 December 2012, At: 02:00
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
Supramolecular Chemistry
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/gsch20
Naphthalenediimide dimers and trimers form self-
assembling hydrogen-bonded nanotubes of enhanced
stability
John-Carl Olsen a , Nora A. Batchelder b , Julia H. Raney c & David E. Hansen c d e
a School of Sciences, Indiana University Kokomo, Kokomo, IN, 46904, USA
b Department of Chemistry, Amherst College, Amherst, MA, 01002, USA
c Keck Science Department, Claremont McKenna College, 925 N. Mills Avenue, Claremont,
CA, 91711, USA
d Pitzer College, Claremont, CA, 91711, USA
e Scripps College, Claremont, CA, 91711, USA
Version of record first published: 16 Oct 2012.
To cite this article: John-Carl Olsen, Nora A. Batchelder, Julia H. Raney & David E. Hansen (2012): Naphthalenediimide
dimers and trimers form self-assembling hydrogen-bonded nanotubes of enhanced stability, Supramolecular Chemistry, 24:12,
841-850
To link to this article: http://dx.doi.org/10.1080/10610278.2012.731062
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to
anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should
be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,
proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in
connection with or arising out of the use of this material.

Supramolecular Chemistry
Vol. 24, No. 12, December 2012, 841–850
Naphthalenediimide dimers and trimers form self-assembling hydrogen-bonded nanotubes
of enhanced stability
John-Carl Olsena, Nora A. Batchelderb, Julia H. Raneyc and David E. Hansenc,d,e
*
b
aSchool of Sciences, Indiana University Kokomo, Kokomo, IN 46904, USA; Department of Chemistry, Amherst College,
Amherst, MA 01002, USA; Keck Science Department, 925 N. Mills Avenue, Claremont McKenna College, Claremont,
c
d
CA 91711, USA; Pitzer College, Claremont, CA 91711, USA; Scripps College, Claremont, CA 91711, USA
e
(Received 16 August 2012; final version received 13 September 2012)
Covalent linkage of amino acid-functionalised naphthalenediimides (NDIs) produced a dimer and a trimer that form helical,
hydrogen-bonded nanotubes in CHCl3 solutions and that are more resistant than an analogous monomer to the hydrogen-
bond disrupting effects of tetrahydrofuran, methanol and heat.
Keywords: organic nanotubes; naphthalenediimide; dimer; trimer; solvophobic
Introduction
bonds between carboxylic acid groups.) The naphthalene
cores of the monomers form the walls of the nanotube,
which are oriented parallel to its long axis. The amino acid
side chains project out from the nanotube walls, and in our
design we focused on the use of these side chains as the
handles for the covalent linkage of NDI subunits.
Given that their ‘inherent geometric features, notably
defined inner and outer surfaces as well as termini, give
rise to a controlled spatial segregation’ (1), organic
nanotubes provide an active area of enquiry in
supramolecular chemistry, and systems based on peptides,
porphyrins, nucleobases and a variety of other small
organic molecules have been described (2–6). Recently,
the Sanders group (7) reported that amino acid-
functionalised naphthalenediimide (NDI) monomers
form helical hydrogen-bonded nanotubes in solvents of
medium polarity and in the solid state. Appealing features
of this remarkable new system include the ease of
monomer synthesis (8), the toleration of different amino
acid side chains (7, 9, 10) and the inclusion of various
guest molecules, such as C60, within the central pore of the
nanotubes (11–13). In addition, the extent of nanotube
formation is readily assayed by circular dichroism (CD)
spectroscopy (14–16).
Results and discussion
To probe further the versatility of the Sanders system, we
have studied the properties of a dimer and a trimer formed
by appropriate covalent linkage of NDI subunits. In
particular, we employed molecular mechanics studies to
explore potential covalent linkers between NDI monomer
side chains separated by one turn of the helix (i and i þ 3),
and as shown in Figure 1(b), a terephthalic diester bridge
through serine side chains appeared to be an effective
modification. Dimer 4 and trimer 6 were thus synthesised
using this linker, and the analogous monomer 2 was
synthesised for use as a positive control (Figure 2 – also
shown are the t-butyl esters 1, 3 and 5, the synthetic
precursors to NDI derivatives 2, 4 and 6, respectively). We
report here that dimer 4 and trimer 6 indeed form
nanotubes in solution and that these nanotubes are
significantly more resistant to THF, methanol and heat
than those formed by monomer 2.
The syntheses of NDI derivatives 2, 4 and 6 relied on
the microwave-heating approach developed by the Sanders
group (8), a method that allows for the efficient conversion
of 1,4,5,8-naphthalenetetracarboxylic dianhydride (NDA)
into symmetric and unsymmetric amino acid-functiona-
lised NDIs. The unsymmetric NDI derivative prepared
from L-Phe-t-butyl ester and L-Ser-t-butyl ester was the
The nanotube architecture formed from NDI mono-
mers derivatised with L-amino acids is depicted in
Figure 1(a) (derived from a crystal structure published
by the Sanders group (7)). Three monomers constitute one
turn of the helix, and adjacent monomers (i and i þ 1) are
connected by hydrogen bonds between carboxylic acid
groups. In addition, CZH· · ·O hydrogen bonds between
monomers four steps apart (i and i þ 3) reinforce the
assembly in the solid state. (These interactions, although
weak, presumably provide the driving force for the
formation of the nanotubes in solution – otherwise
entropically more favourable ‘random coils’ would likely
form, with NDI subunits only linked by the hydrogen
*Corresponding author. Email: dhansen@kecksci.claremont.edu
ISSN 1061-0278 print/ISSN 1029-0478 online
q 2012 Taylor & Francis
http://dx.doi.org/10.1080/10610278.2012.731062
http://www.tandfonline.com

842
J.-C. Olsen et al.
R
(a)
R
R
(b)
i
i + 1
R
R
R
i + 2
i + 3
O
O
O
O
R
R
i + 3
R
R
i
Terephthalic
Ester Linker
R
4, R = Ph
R
Figure 1. NDI nanotube segments. (a) An oblique view (modified from Ponnuswamy et al. (14)) composed of five NDI monomers that
shows hydrogen bonds (black, dashed lines), the sense of the helix (black arrow) and the relative designation and orientation of several
monomers (i, i þ 1, i þ 2 and i þ 3). [Adapted with permission from Ponnuswamy, N.; Pantos¸ G.D.; Smulders, M.M.J.; Sanders, J.K.M.
J. Am. Chem. Soc. 2012, 134, 566–573; Copyright 2012, American Chemical Society.] (b) A molecular model of two adjacent NDI units
(i and i þ 3) of a nanotube, which are connected by a terephthalic ester linker.
key intermediate for the synthesis of the protected
monomer 1 and protected dimer 3. For the synthesis of
protected trimer 5, the symmetric bis(L-Ser-t-butyl ester)
NDI derivative was also employed and served as the
central subunit. Trifluoroacetic acid deprotection of
derivatives 1, 3 and 5 then yielded NDIs 2, 4 and 6,
which were obtained from NDA in overall yields of 44%,
32% and 26%, respectively. (Complete synthetic details
are provided below.)
As shown in Figure 3, the CD spectrum of monomer 2
at room temperature in neat CHCl3 displays the signals
characteristic of nanotube formation, with the most
diagnostic being that at l ¼ 383 nm (the result of an
induced Cotton effect from the NDI naphthalene core
(7, 14)). The spectrum of monomer 2 in the presence of
0.12% methanol lacks these signals, presumably due to
disruption of the hydrogen bonds that hold the nanotube
together. As expected, the CD spectrum for the NDI
O
O O
O
RO
N
O
1 R=t Bu
N
2 R=H
OR
O
O
O O
O
O
O O
N
O
O
N
O O
N
RO
O
OR
3 R=t Bu
4 R=H
N
OR
RO
O
O O
O O
O
O
O
O
O
O O
N
O
N
O
N
O O
N
O O
O
RO
OR
O
HO
N
O
N
OR
RO
O
O
OR
O
O O
O O
O
O O
O
5 R=t Bu
6 R=H
Figure 2. A homologous series of NDI acids – monomer 2, dimer 4 and trimer 6 – and their corresponding t-butyl esters 1, 3 and 5.

Supramolecular Chemistry
843
%MeOH
2
2
1
0.00
0.12
0.00
60,000
40,000
20,000
0
–20,000
–40,000
250
300
350
400
450
l (nm)
Figure 3. CD spectra of the methanol titration at room temperature (228C) of NDI monomer 2 (initial concentration 6.9 £ 1025 M) in
CHCl3. For the CD spectrum of t-butyl-protected NDI monomer 1, the concentration was 7.9 £ 1025 M. Molar ellipticities were
calculated with respect to moles of NDI subunit.
monomer 1, in which the essential hydrogen-bonding
carboxylic acid functionalities are masked as t-butyl
esters, also lacks the signals indicative of nanotube
formation.
dissolution. Though the presence of 0.5% THF completely
disrupts nanotube formation by monomer 2 (data not
shown – see Supplementary Information, available
online), dimer 4 forms nanotubes in this solvent system.
In addition, these nanotubes are far more resistant to
methanol than those formed from monomer 2 (Figure 4).
NDI dimer 4 is not soluble in pure CHCl3, and thus,
0.5% THF in CHCl3 was utilised to effect sample
%MeOH
4
4
4
4
4
4
4
3
0.00
0.12
0.79
2.0
60,000
40,000
20,000
0
3.9
7.4
17
0.00
–20,000
250
300
350
400
450
l (nm)
Figure 4. CD spectra of the methanol titration at room temperature (228C) of NDI dimer 4 (initial concentration 4.6 £ 1025 M) in CHCl3
containing 0.5% THF. The CD spectrum of t-butyl-protected NDI dimer 3 (4.3 £ 1025 M) was recorded in pure CHCl3. Molar ellipticities
were calculated with respect to moles of NDI subunit.

844
J.-C. Olsen et al.
%MeOH
0.00
0.12
0.48
17
6
6
6
6
5
60,000
40,000
20,000
0
0.00
–20,000
–40,000
250
300
350
400
450
λ (nm)
Figure 5. CD spectra of the methanol titration at room temperature (228C) of NDI trimer 6 (initial concentration 1.7 £ 1025 M) in
CHCl3 containing 1.0% THF. The CD spectrum of t-butyl-protected NDI trimer 5 (2.3 £ 1025 M) was recorded in pure CHCl3. Molar
ellipticities were calculated with respect to moles of NDI subunit.
For example, in 0.12% methanol, the intensity of the signal
at l ¼ 383 nm for dimer 4 decreases by only 1%. The
addition of increasing amounts of methanol leads to
further decreases in signal intensity, and in 17% methanol
the nanotube signals completely disappear. As expected,
the CD spectrum of dimer 4 in 17% methanol is similar to
that of the t-butyl ester-protected dimer 3.
titration of the trimer with THF (data not shown – see
Supplementary Information, available online), a result that
suggests that the solvophobic effect seen with methanol
may only occur with protic solvents.
Given the very small decrease noted above in the
intensity of the l ¼ 383 nm signal for dimer 4 in 0.12%
methanol, it is likely that the dimer also experiences a
small stabilising solvophobic effect; however, for the
dimer the disruption of the hydrogen bonding is a more
dominant factor. The larger solvophobic effect for the
trimer 6 may be due to the greater relative percentage of
terephthalic ester linker present (the molar ratio of linker
to NDI subunit is 0.67 for the trimer as compared with 0.50
for the dimer).
To reproducibly obtain homogeneous solutions of
trimer 6, a further increase in THF concentration to 1.0%
was required. Even in the presence of the 1.0% THF,
however, trimer 6 forms nanotubes (Figure 5). Moreover,
the intensity of the CD signal at l ¼ 383 nm increases
upon small additions of methanol, reaching a maximum
intensity at 0.48% methanol with a signal enhancement of
20%. These increases in intensity are indicative of a larger
extent of nanotube formation (14–16). This result might
be explained in terms of a competition between
solvophobic (1, 12) and hydrogen-bonding effects upon
the addition of methanol, with the former favouring and
the latter disfavouring nanotube formation. The nanotube
assembly from trimer 6 may present less hydrophobic
surface to the bulk solvent than the random, hydrogen-
bonded polymer or than the completely dispersed and
solvated trimers. The nanotube structure would thus be
favoured in a solvent of higher polarity, as results from the
addition of methanol. However, the ability of methanol to
compete with the hydrogen bonds between NDI subunits
in trimer 6 prevails at methanol concentrations .0.48%,
leading to the disruption of the nanotube structure.
No increases in signal intensity are observed during the
To probe further the stabilities of NDIs 2, 4 and 6,
variable temperature CD data were collected (Figure 6).
Monomer 2 was dissolved in neat CHCl3, and dimer 4 and
trimer 6 each in 1.0% THF in CHCl3. Molar ellipticity was
determined at l ¼ 383 nm in 18 increments from 5 to
558C. The intensities at this wavelength for a solution of
monomer 2 diminished far more rapidly than for either
dimer 4 or trimer 6 and approached zero by 508C,
indicating essentially complete disruption of the nanotube
structure at this temperature. In contrast, the intensities at
l ¼ 383 nm for the solution of dimer 4 decreased by only
11% from 5 to 558C, whereas for trimer 6 the decrease was
15% over this temperature range. In these experiments, the
concentration of dimer 4 was 1.6 times that of trimer 6, and
thus, the concentrations of NDI subunit were approxi-
mately equal. Given the similar decreases in the l ¼ 383

Supramolecular Chemistry
845
2
4
6
60,000
40,000
20,000
0
5
10
15
20
25
30
35
40
45
50
55
T (oC)
Figure 6. Variable temperature CD measurements of monomer 2 (6.9 £ 1025 M in CHCl3, black trace), dimer 4 (2.2 £ 1025 M, 1.0%
THF in CHCl3, red trace) and trimer 6 (1.4 £ 1025 M, 1.0% THF in CHCl3, blue trace). Signal intensity at l ¼ 383 nm was measured at 18
intervals from 5 to 558C. Molar ellipticity was calculated with respect to moles of NDI subunit (colour online).
signal, the nanotubes formed by dimer 4 and trimer 6
appear to have comparable stabilities under these
conditions. THF titrations of dimer 4 and trimer 6 (1.0–
18% THF) also suggest that nanotubes formed from these
two species – again at approximately equal NDI subunit
concentrations – are comparably stable (data not shown –
Supplementary Information, available online).
modified to prevent further nanotube elongation (i.e. the
nanotubes would be ‘capped’). Our ultimate goal is to
discover structural modifications that allow NDI-based
nanotubes of discrete length to form in polar protic
solvents.
Experimental details
Materials and methods
Conclusion
L-Phenylalanine-t-butyl ester hydrochloride (8) was
purchased from Advanced Chem Tech (product number
IF7427). L-Serine-t-butyl ester hydrochloride (9) was
purchased from Indofine Chemical Company Inc. (product
number 04-2437). Terephthalic acid monoallyl ester (13)
was purchased from Chinglu Pharmaceutical Research
LLC (product number A1160). Nuclear magnetic reson-
ance (NMR) solvents were purchased from Aldrich and
Cambridge Isotope Laboratories. Anhydrous CHCl3
stabilised with amylene was purchased from Fisher
(Acros product number 61028-1000). Flash chromatog-
raphy was performed with silica gel 62 (60–200 mesh).
Anhydrous tetrahydrofuran was obtained via elution
through a solvent column drying system as described by
Pangborn et al. (17). All other compounds were purchased
from Aldrich, VWR or Fisher and used without further
purification.
In conclusion, we have shown that NDI dimer 4 and trimer
6 form hydrogen-bonded nanotubes that, relative to the
corresponding monomer 2, resist the disrupting effects of
THF, methanol and heat. Although additional studies (such
as those recently reported by the Sanders group on their
original system (15)) will be required to determine the
underlying thermodynamic basis of these effects, the
preorganisation of NDI subunits in dimer 4 and trimer 6
clearly leads to the formation of nanotubes of enhanced
stability. Determination of the precise positioning of the
terephthalic acid linker also awaits further work – though
designed to bridge monomers four steps apart (i and
i þ 3), the linker could conceivably instead bridge
adjacent monomers (i and i þ 1), although our molecular
mechanics studies suggest that this arrangement would be
of higher energy; finally, the terephthalic ester might in
fact serve as a cross-linker of arrays of nanotubes.
Microwave reactions were run in a Biotage Initiator
microwave reactor in sealed 20-ml glass tubes capped with
Teflonw septa. Reaction pressure was monitored by a non-
invasive sensor within the lid of the reaction cavity that
In the future, we wish to explore approaches that will
lead to nanotubes of discrete length, in which both the NDI
units are covalently linked and the terminal NDIs are

846
J.-C. Olsen et al.
measured the deformation of the Teflonw seal. Reaction
temperature was monitored by an IR sensor. The
temperatures indicated in the experimentals were the
maximum temperatures reached during the reactions. The
reaction times correspond to the total irradiation times. CD
spectroscopy was performed using a JASCO J-715
spectropolarimeter. Sample solutions were measured in a
quartz cuvette with a 1-cm path length. NMR spectra were
obtained using a Bruker Avance III 500 MHz NMR
spectrometer. Chemical shifts for protons are reported in
parts per million downfield from tetramethylsilane (TMS),
and are referenced to TMS (d ¼ 0 ppm) or, when TMS is
absent, residual protium in the NMR solvent (CDCl3:
d ¼ 7.26 ppm, DMSO-d6: d ¼ 2.50 ppm). Chemical shifts
for carbon are reported in parts per million downfield from
TMS, and are referenced to TMS (d ¼ 0 ppm) or, when
TMS is absent, to the carbon resonances of the solvent
(CDCl3: d ¼ 77.16 ppm, DMSO-d6: d ¼ 39.52 ppm). Data
are represented as follows: chemical shift, integration,
multiplicity (br ¼ broad, s ¼ singlet, d ¼ doublet, t ¼
triplet, q ¼ quartet, dd ¼ doublet of a doublet, m ¼
multiplet). Infrared spectroscopy was performed using a
Thermo-Nicolet Avatar 370 FT-IR spectrometer. Mass
spectrometric data were obtained at the University of
California, Riverside High Resolution Mass Spectrometry
Facility. The acronym LIFDI stands for ‘liquid introduc-
tion, field desorption ionisation’, which is a mass
spectrometric technique described online at http://www.
acif.ucr.edu/mspec/lifdi.html.
1408C. After cooling, the solvent was evaporated. The
residue was dissolved in dichloromethane (50 ml),
transferred to a separatory funnel and rinsed with water
(3 £ 50 ml) and brine (50 ml). The organic layer was
separated and dried over sodium sulphate. The product
was adsorbed onto silica gel by evaporation, dry loaded
onto a silica gel column and separated by flash
chromatography with dichloromethane. Fractions contain-
ing product were combined; the solvent was evaporated
and the product was placed under a high vacuum overnight
to give NDI di-t-butyl ester 10 as a yellow solid (553 mg,
78%); m.p. 115–1208C (dec.); IR (KBr) 3534, 2978, 2937,
1
1739, 1709, 1672, 1344, 1248 and 1158 cm21; H NMR
(500 MHz, CDCl3, 298 K): d ¼ 8.73 (d, J ¼ 7.5 Hz, 2H),
8.67 (d, J ¼ 7.5 Hz, 2H), 7.15–7.04 (m, 5H), 5.93 (dd,
J ¼ 10.5, 5.5 Hz, 1H), 5.72 (t, J ¼ 6.5 Hz, 1H), 4.39 (dd,
J ¼ 11.5, 6.5 Hz, 1 H), 4.02 (dd, J ¼ 6.5, 12.0 Hz, 1 H),
3.67 (dd, J ¼ 14.5, 5.5 Hz, 1 H), 3.49 (dd, J ¼ 14.5,
10.5 Hz, 1 H), 2.647 (bs, 1H), 1.47 (s, 9 H), 1.46 (s, 9H)
ppm; 13C NMR (125 MHz, CDCl3, 298 K): d ¼ 168.0,
167.7, 162.6, 162.2, 137.2, 131.4, 131.2, 129.1, 128.4,
126.8, 126.7, 126.6, 126.4, 126.3, 83.2, 82.5, 60.8, 55.7,
55.6, 34.8, 27.9 ppm; HRMS (ESI þ ): calcd for
C34H34N2O9Na: 637.2157; found: 637.2151 [M þ Na]þ.
NDI monomer 14
NDI di-t-butyl ester 10 (2.48 g, 4.03 mmol), terephthalic
acid monoallyl ester (13) (831 mg, 4.03 mmol), EDC
(850 mg, 4.43 mmol) and DMAP (0.492 mg, 4.03 mmol)
were dissolved in dichloromethane (80 ml) and refluxed
overnight at 408C. The reaction mixture was partitioned
between dichloromethane (300 ml) and water (300 ml).
The organic layer was separated, washed with water
(300 ml) and brine (300 ml) and dried over sodium
sulphate. The solution was filtered through a plug of silica
gel with hexanes:ethyl acetate (5:1). The solvent was
evaporated, and the crude product, which contained the
NDI alkene derivative 15 as an inseparable by-product,
was dried overnight under a high vacuum. This product
mixture (1.95 g), tetrakis(triphenylphosphine)palla-
dium(0) (140 mg, 0.122 mmol) and dimethylbarbituric
acid (398 mg, 2.55 mmol) were dissolved in THF (49 ml).
The solution was stirred at room temperature for 4 h,
adsorbed onto silica gel by evaporation and dry loaded
onto a silica gel column. The product was separated by
flash chromatography with hexanes followed by a mixture
of hexanes, ethyl acetate and acetic acid (200:795:5).
Fractions containing product were combined, the solvent
was evaporated and the sample was placed under a high
vacuum overnight to give terephthalic acid NDI half-ester
14 as a pale, yellow-green powder (1.29 g, 48% for two
steps). m.p. 139–1458C (dec.); IR (KBr) cm21 3374.0,
3083.2, 2978.4, 2933.5, 1732.2, 1710.5, 1673.8 and
Molecular mechanics calculations were made using
HyperChem Release 5.11 Pro for Windows (Hypercube,
Inc.). The MMþ force field with the block-diagonal
Newton–Raphson algorithm was employed, and all
minimisations were terminated at an RMS gradient of
˚
0.1 kcal/(A mol) or less.
Synthetic details
The synthetic routes to monomer 2, dimer 4 and trimer 6
are outlined in Scheme 1.
NDI monomer 10
1,4,5,8-Naphthalenetetracarboxylic dianhydride (300 mg,
1.12 mmol), L-Phe-t-butyl ester hydrochloride (8)
(289 mg, 1.12 mmol), dimethylformamide (15 ml) and
triethylamine (0.312 ml, 2.24 mmol) were added in that
order to a 20-ml pressure tube. The reaction mixture was
sonicated for 30 min and then stirred and heated in a
microwave reactor for 5 min at 1408C. After the reaction
mixture had cooled to room temperature, L-Ser-t-butyl
ester hydrochloride (9) (0.221 g, 1.12 mmol) and triethy-
lamine (0.312 ml, 2.24 mmol) were added. The mixture
was again stirred and heated in a microwave for 5 min at

Supramolecular Chemistry
847
O
O
O
O
N
O
O
N
O
O
N
O
O
N
O
O
HO
O
O
OH
OH
HO
O
O
4
1:1 TFA:CH2Cl2
rt / 4 h
76%
O
O
O
O
O
N
O
O
O
N
O
O
N
O
O
tBuO
O
OtBu
N
O
OtBu
tBuO
O
O
N
O
O
N
O
O
O
O
HO
O
3
OH
O
O
O
0.5 eq
pyr
11
Cl
Cl
2
CH2Cl2 / rt / 15 h
54% 3
O
1:1 TFA:CH2Cl2
rt / 4 h
74%
tBuO
1. a)
O
NH3Cl
1.
12
8
HO
TEA / DMF
µwave
140 oC / 5 min
EDC / DMAP
O
CH2Cl2 / 40 oC / 15 h
O
O
O
O
O
O
O
O
N
O
O
O
O
N
O
O
N
76% 1
tBuO
OH
OtBu
tBuO
O
N
O
OtBu
OH
ClNH3
b)
O
O
O
OtBu
9
O
10
1
7
TEA / DMF
µwave
140 oC / 5 min
78% overall
O
HO
O
13
2. Pd(PPh3)4
DMBA
1.
O
THF / rt / 15 h
48% overall
EDC / DMAP
CH2Cl2 / 40 oC / 15 h
O
O
O
O
N
O
O
O
N
O
O
N
O
O
N
tBuO
OH
tBuO
+
OtBu
OtBu
14
O
O
O
O
O
N
O
O
N
O
O
Elimination Byproduct 15
HO
OtBu
OH
0.5 eq
tBuO
O
16
EDC / DMAP
CH2Cl2 / 40 oC / 15 h
75% 5
O
O
O
N
O
O
O
O
N
O
O
N
O
N
O
O
O
O
N
O
O
tBuO
OtBu
O
tBuO
O
N
OtBu
tBuO
OtBu
O
O
O
O
O
O
O
O
5
1:1 TFA:CH2Cl2
rt / 4 h
91%
O
O
O
N
O
O
O
N
O
O
N
O
O
N
O
O
O
O
N
O
O
N
O
HO
O
OH
O
HO
O
OH
HO
OH
O
O
O
O
O
O
6
Scheme 1. Synthetic routes to NDIs 2, 4 and 6. DMBA, dimethylbarbituric acid; EDC, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide.

848
J.-C. Olsen et al.
1H NMR (500 MHz, CDCl3, 298 K):
1582.4 cm21
;
water (50 ml) and brine (50 ml) and dried over sodium
sulphate. The solution was decanted, and the sample was
adsorbed onto silica gel by evaporation and dry loaded
onto a silica gel column. The product was separated by
flash chromatography with hexanes:ethyl acetate (5:1).
Fractions containing product were combined, the solvent
was evaporated and the sample was placed under a high
vacuum overnight to give NDI di-t-butyl ester 1 as a pale,
blue-green solid (183 mg, 76%). m.p. 105–1098C; IR
(KBr) 3427, 3064, 2978, 2937, 1736, 1711 and
d ¼ 8.73 (d, J ¼ 7.5 Hz, 2H), 8.68 (d, J ¼ 7.5 Hz, 2H),
8.05 (d, J ¼ 8.5 Hz, 2H), 7.97 (d, J ¼ 8.0 Hz, 2H), 7.15–
7.04 (m, 5H), 6.10 (dd, J ¼ 9.3, 4.2 Hz, 1H), 5.94 (dd,
J ¼ 10.0, 4.3 Hz, 1H), 5.21 (dd, J ¼ 11.7, 4.3 Hz, 1H),
4.98 (dd, J ¼ 11.5, 9.5 Hz, 1H), 3.67 (dd, J ¼ 14.5, 5.5 Hz,
1H), 3.48 (dd, J ¼ 14.3, 10.3 Hz, 1H), 1.50 (s, 9H), 1.46 (s,
9H); 13C NMR (125 MHz, CDCl3, 298 K): d ¼ 170.5,
168.0, 165.6, 165.3, 162.5, 162.2, 137.2, 134.1, 133.1,
131.4, 131.2, 130.1, 129.7, 129.1, 128.4, 126.9, 126.7,
126.6, 126.5, 126.2, 83.6, 82.6, 62.8, 55.6, 53.6, 34.8,
27.9 ppm; HRMS (LIFDI): calcd for C33H22N2O10:
762.2419; found: 762.2383 Mþ.
1673 cm21
;
1H NMR (500 MHz, CDCl3, 298 K):
d ¼ 8.73 (d, J ¼ 7.5 Hz, 2H), 8.67 (d, J ¼ 7.5 Hz, 2H),
7.87 (d, J ¼ 7.5 Hz, 2H), 7.49 (t, J ¼ 7.5 Hz, 1H), 7.33 (m,
J ¼ 7.5 Hz, 2H), 7.14–7.04 (m, 5H), 6.11 (dd, J ¼ 9.5,
4.5 Hz, 1H), 5.93 (dd, J ¼ 10.0, 5.5 Hz, 1H), 5.15 (dd,
J ¼ 12.0, 4.0 Hz, 1H), 4.98 (dd, J ¼ 11.5, 9.5 Hz, 1H),
3.67 (dd, J ¼ 14.5, 5.5 Hz, 1H), 3.48 (dd, J ¼ 14.0,
10.5 Hz, 1H), 1.50 (s, 9H), 1.46 (s, 9H); 13C NMR
(125 MHz, CDCl3, 298 K): d ¼ 167.9, 166.2, 165.7, 162.5,
162.3, 137.2, 133.1, 131.4, 131.2, 129.6, 129.1, 128.37,
128.35, 126.9, 126.74, 126.65, 126.4, 126.3, 83.4, 82.5,
62.3, 55.6, 53.7, 34.8, 28.0 ppm; HRMS (LIFDI): calcd for
C41H38N2O10: 718.2521; found: 718.2532 Mþ.
NDI monomer 16
1,4,5,8-Naphthalenetetracarboxylic dianhydride (1.26 g,
4.7 mmol), L-Ser-t-butyl ester hydrochloride (9) (2.05 g,
10.4 mmol), dimethylformamide (15 ml) and triethylamine
(2.62 ml, 18.8 mmol) were added to a 20-ml pressure tube.
The reaction mixture was sonicated for 30 min and then
stirred and heated in a microwave reactor for 8 min at
1408C. The reaction mixture was cooled to room
temperature, and the solvent was evaporated. The residue
was dissolved in dichloromethane (300 ml), transferred to
a separatory funnel and rinsed with water (3 £ 300 ml) and
brine (300 ml). The organic layer was dried over sodium
sulphate. After removal of the drying agent, the solution
was adsorbed onto silica gel by evaporation. The sample
was dry loaded onto a silica gel column, and the product
was separated by flash chromatography with hexanes:ethyl
acetate (2:1 ! 1:1 ! 1:2). Fractions containing product
were combined, the solvent was evaporated and the sample
was placed under a high vacuum overnight to give t-butyl-
protected bis-serine NDI 16 as a yellow solid (2.09 g,
80%). m.p. 160–1658C (dec.); IR (KBr) 3508, 3085, 2978,
NDI monomer 2
NDI di-t-butyl ester 1 (502 mg, 0.699 mmol), trifluoroa-
cetic acid (14 ml) and dichloromethane (14 ml) were
stirred at room temperature for 4 h. The solvent was
evaporated, and the sample was placed under a high
vacuum overnight to give NDI monomer 2 as a yellow
powder (312 mg, 74%). m.p. 159–1638C; IR (KBr) 3491,
1
3202, 3087, 2939, 2589, 1709, 1673 and 1347 cm21; H
NMR (500 MHz, DMSO-d6, 298 K): d ¼ 13.30 (bs, 2H),
8.74 (d, J ¼ 7.5 Hz, 2H), 8.69 (d, J ¼ 7.0 Hz, 2H), 7.81 (d,
J ¼ 6 Hz, 2H), 7.58 (bs, 1H), 7.42 (bs, 2H), 7.19–7.04 (m,
5H), 6.09 (dd, J ¼ 8.0, 4.5 Hz, 1H), 5.92 (dd, J ¼ 9.0,
6.0 Hz, 1H), 5.10 (dd, J ¼ 11.2, 3.8 Hz, 1H), 4.85 (t,
J ¼ 10.0 Hz, 1H), 3.63 (dd, J ¼ 13.7, 5.2 Hz, 1H), 3.37
(dd, J ¼ 13.5, 10.0 Hz, 1H); 13C NMR (125 MHz, CDCl3,
298 K): d ¼ 171.2, 169.3, 166.4, 163.3, 162.9, 138.6,
134.2, 132.2, 130.1, 130.0, 129.9, 129.5, 129.1, 127.3,
127.2, 126.9, 126.6, 63.2, 55.8, 55.5, 53.2, 35.2 ppm;
HRMS (ESI þ ): calcd for C33H22N2O10: 606.1269;
found: 606.1241 Mþ.
1739, 1708, 1670, 1582 and 1452 cm21 1H NMR
;
(500 MHz, CDCl3, 298 K): d ¼ 8.78 (s, 4H), 5.73 (t,
J ¼ 6.5 Hz, 2H), 4.41 (dd, J ¼ 11.5, 6.5 Hz, 2H), 4.04 (dd,
J ¼ 11.5, 6.5 Hz, 2H), 2.76 (bs, 2H), 1.45 (s, 18H); 13C
NMR (125 MHz, CDCl3, 298 K): d ¼ 167.6, 162.6, 131.4,
126.9, 126.5, 83.2, 60.7, 55.7, 27.9 ppm; HRMS (ESI þ ):
calcd for C28H30N2O10Na: 577.1797; found: 577.1799
[M þ Na]þ.
NDI monomer 1
NDI dimer 3
NDI di-t-butyl ester 10 (200 mg, 0.325 mmol), benzoic
acid (43.7 mg, 0.358 mmol), EDC 68.6 mg, 0.358 mmol)
and DMAP (43.7 mg, 0.358 mmol) were dissolved in
dichloromethane (10 ml) and refluxed at 408C overnight.
After cooling, the reaction mixture was partitioned
between dichloromethane (50 ml) and 1 M HCl (aq.)
(50 ml). The organic layer was separated and washed with
NDI di-t-butyl ester 10 (3.23 g, 5.26 mmol), terephthalyl
chloride (0.534 g, 2.63 mmol) and DMAP (0.483 g,
3.95 mmol) were dissolved in dichloromethane (105 ml).
Pyridine (0.466 ml, 5.78 mmol) was added, and the
solution was stirred at room temperature for 48 h. The
reaction mixture was adsorbed onto silica gel by

Supramolecular Chemistry
849
evaporation and dry loaded onto a silica gel column. The
product was separated by flash chromatography with
hexanes:ethyl acetate (5:1 ! 3:1). Fractions containing
product were combined; the solvent was evaporated, and
the sample was placed under a high vacuum overnight to
give NDI tetra-t-butyl ester 3 as a yellow solid (1.95 g,
54%). m.p. 156–1608C; IR (KBr) 3086, 2978, 2936, 1739,
organic layer was separated and washed with water (50 ml)
and brine (50 ml) and dried over sodium sulphate. The
solution was decanted, and the sample was adsorbed onto
silica gel by evaporation and dry loaded onto a silica gel
column. The product was separated by flash chromatog-
raphy with hexanes:ethyl acetate (4:1 ! 2:1). Fractions
containing product were combined, the solvent was
evaporated and the sample was placed under a high
vacuum overnight to give NDI hexa-t-butyl ester 5 as a
pale, blue-green solid (449 mg, 75%). m.p. 182–1858C; IR
1
1711, 1675 and 1247 cm21; H NMR (500 MHz, CDCl3,
298 K): d ¼ 8.72 (d, J ¼ 7.5 Hz, 4H), 8.68 (d, J ¼ 7.5 Hz,
4H), 7.85 (s, 4H), 7.1–7.03 (m, 10H), 6.06 (dd, J ¼ 9.0,
4.0 Hz, 2H), 5.94 (dd, J ¼ 10.4, 5.5 Hz, 2H), 5.16 (dd,
J ¼ 12.0, 4.5 Hz, 2H), 4.95 (dd, J ¼ 11.5, 9.3 Hz, 2H),
3.67 (dd, J ¼ 14.0, 5.5 Hz, 2H), 3.48 (dd, J ¼ 14.5, 10.5,
2H), 1.48 (s, 18H), 1.46 (s, 18H) ppm; 13C NMR
(125 MHz, CDCl3, 298 K): d ¼ 168.2, 165.9, 165.6, 162.8,
162.5, 137.5, 133.8, 131.7, 131.5, 129.9, 129.4, 128.7,
127.2, 127.1, 126.9, 126.8, 126.4, 83.8, 82.8, 63.1, 55.9,
53.9, 35.1, 28.3 ppm; HRMS (ESI þ ): calcd for
C76H70N4O20Na: 1381.4476; found: 1381.4442
[M þ Na]þ.
(KBr) cm21
;
1H NMR (500 MHz, CDCl3, 298 K):
d ¼ 8.71–8.67 (m, 12H), 7.85 (s, 8H), 7.15–7.01 (m,
10H), 6.08–6.04 (m, 4H), 5.96–5.92 (dd, J ¼ 10.5,
5.5 Hz, 2H), 5.19–5.14 (m, 4H), 4.99–4.92 (m, 4H), 3.66
(dd, J ¼ 14.0, 5.5, 2H), 3.48 (dd, J ¼ 14.0, 10.0 Hz, 2H),
1.48 (s, 36H), 1.46 (18H); 13C NMR (125 MHz, DMSO-
d6, 298 K): d ¼ 167.9, 165.55, 165.52, 165.3, 162.5, 162.4,
162.2, 137.3, 133.49, 133.45, 131.5, 131.4, 131.2, 129.6,
129.1, 128.3, 127.0, 126.9, 126.7, 126.6, 126.5, 126.4,
126.1, 83.5, 82.5, 62.85, 62.80, 55.6, 53.65, 53.61, 34.8,
27.9 ppm; HRMS (ESI þ ): calcd for C112H102N6O32Na:
2065.6431; found: 2065.6446 [M þ Na]þ.
NDI dimer 4
NDI trimer 6
NDI tetra-t-butyl ester 3 (1.82 g, 1.34 mmol), triethylsilane
(1.56 ml, 13.4 mmol), trifluoroacetic acid (27 ml) and
dichloromethane (27 ml) were added to a flask and stirred
at room temperature for 2 h. The solvent was evaporated,
and the residue was triturated twice with dichloromethane
and placed under a high vacuum overnight to give NDI
dimer 4 as a yellow-green powder (1.17 g, 76%).
m.p. 220–2288C; IR (KBr) 3522, 3085, 2940, 2602,
NDI hexa-t-butyl ester 5 (193 mg, 0.106 mmol), trifluor-
oacetic acid (5 ml) and dichloromethane (5 ml) were
stirred at room temperature for 4 h. The solvent was
evaporated, and the residue was triturated with dichlor-
omethane and placed under a high vacuum overnight to
give 6 as a pale green powder (164 mg, 91%) m.p. 230–
2348C (dec.); IR (KBr) 3516, 3085, 2940, 2606, 1709,
1673 and 1582 cm21 1H NMR (500 MHz, DMSO-d6,
;
1709, 1670, 1581, 1348 and 1249 cm21 1H NMR
;
298 K): d ¼ 13.43 (bs, 6H), 8.76–8.64 (m, 12H), 7.70–
7.90 (m, 8H), 7.00–7.15 (m, 10H), 6.05–5.98 (m, 4H),
5.90–5.84 (m, 2H), 5.20–5.14 (m, 4H), 4.90–4.77 (m,
4H), 3.52–3.62 (m, 2H), 3.50–3.28 (m, 4H); 13C NMR
(125 MHz, DMSO-d6, 298 K): d ¼ 171.1, 169.2, 165.51,
165.48, 163.29, 163.26, 162.9, 138.5, 134.1, 134.0, 132.2,
130.13, 130.08, 129.8, 129.0, 127.24, 127.19, 126.91,
126.88, 126.5, 63.6, 55.9, 55.4, 53.1, 35.1 ppm; HRMS
(ESI 2 ): calcd for C88H53N6O32: 1705.2710; found:
1705.2703 [M 2 H].
(500 MHz, DMSO-d6, 298 K): d ¼ 13.41 (bs, 2H), 13.15
(bs, 2H), 8.68 (d, J ¼ 7.5 Hz, 4H), 8.64 (d, J ¼ 7.5 Hz,
4H), 7.80 (s, 4H), 7.13–6.96 (m, 10H), 6.01 (dd, J ¼ 8.0,
4.5 Hz, 2H), 5.87 (dd, J ¼ 9.5, 6.0 Hz, 2H), 5.05 (dd,
J ¼ 11.5, 9.0 Hz, 2H), 4.79 (dd, J ¼ 11.5, 9.0 Hz, 2H),
3.57 (dd, J ¼ 14.2, 5.3 Hz, 2H), 3.32 (dd, J ¼ 14.0,
10.0 Hz, 2H); 13C NMR (125 MHz, CDCl3, 298 K):
d ¼ 170.2, 168.3, 164.7, 162.4, 162.0, 137.6, 133.2,
131.3, 129.3, 128.9, 128.1, 126.3, 126.05, 125.99, 125.6,
62.7, 54.9, 54.5, 52.2, 34.2 ppm; HRMS (ESI þ ): calcd
for C60H38N4O20Na: 1157.1972; found: 1157.1960
[M þ Na]þ.
Supplementary material
1H and 13C NMR spectra for all new compounds are
provided, as are CD spectra showing the effect of THF on
nanotube formation by monomer 2, dimer 4 and trimer 6.
The supplementary material is available with the online
version of this paper at http://www.tandfonline.com/gsch.
NDI trimer 5
Terephthalic acid NDI half-ester 14 (500 mg,
0.813 mmol), t-butyl-protected bis-serine NDI 16
(138 mg, 0.312 mmol), EDC (240 mg, 1.25 mmol) and
DMAP (38.1 mg, 13.7 mmol) were dissolved in dichlor-
omethane (13.7 ml) and refluxed at 408C overnight. After
cooling, the reaction mixture was partitioned between
dichloromethane (50 ml) and 1 M HCl (aq.) (50 ml). The
Acknowledgements
The authors are grateful to the National Institutes of Health (R15
GM088748), the Research Corporation for Science Advancement

850
J.-C. Olsen et al.
and the Sydney J. Weinberg, Jr Foundation for financial support.
NMR spectra were recorded on an instrument purchased with
funding from the National Science Foundation (MRI 0922393).
CD spectra were recorded at the UCLA DOE Biochemistry
Instrumentation Facility, and the authors are grateful to Martin
Phillips, the facility’s director, for his assistance. Finally, the
authors thank G. Dan Pantos¸ for his invaluable advice on many
aspects of the work reported here.
(7) Pantos¸, G.D.; Pengo, P.; Sanders, J.K.M. Angew. Chem. Int.
Ed. 2007, 46, 194–197.
(8) Pengo, P.; Pantos¸, G.D.; Otto, S.; Sanders, J.K.M. J. Org.
Chem. 2006, 71, 7063–7066.
(9) Anderson, T.W.; Sanders, J.K.M.; Pantos¸, G.D. Org.
Biomol. Chem. 2010, 8, 4274–4280.
(10) Anderson, T.W.; Pantos¸, G.D.; Sanders, J.K.M. Org.
Biomol. Chem. 2011, 9, 7547–7553.
(11) Pantos¸, G.D.; Wietor, J.-L.; Sanders, J.K.M. Angew. Chem.
Int. Ed. 2007, 46, 2238–2240.
References
(12) Tamanini, E.; Ponnuswamy, N.; Pantos¸, G.D.; Sanders,
J.K.M. Faraday Discuss. 2010, 145, 205–218.
(1) Balbo Block, M.A.; Kaiser, C.; Khan, A.; Hecht, S.
Top. Curr. Chem. 2005, 245, 89–150.
(2) De Greef, T.F.A.; Smulders, M.M.J.; Wolffs, M.; Schen-
ning, A.P.H.J.; Sijbesma, R.P.; Meijer, E.W. Chem. Rev.
2009, 109, 5687–5754.
(3) Brea, R.J.; Reiriz, C.; Granja, J.R. Chem. Soc. Rev. 2010,
39, 1448–1456.
(4) Liu, H.; Xu, J.; Li, Y.; Li, Y. Acc. Chem. Res. 2010, 43,
1496–1508.
(5) Kameta, N.; Minamikawa, H.; Masuda, M. Soft Matter
2011, 7, 4539–4561.
(13) Tamanini, E.; Pantos¸, G.D.; Sanders, J.K.M. Chem. Eur. J.
2010, 16, 81–84.
(14) Bulheller, B.M.; Pantos¸, G.D.; Sanders, J.K.M.; Hirst, J.D.
Phys. Chem. Chem. Phys. 2009, 11, 6060–6065.
(15) Ponnuswamy, N.; Pantos¸, G.D.; Smulders, M.M.J.;
Sanders, J.K.M. J. Am. Chem. Soc. 2012, 134, 566–573.
(16) Ponnuswamy, N.; Stefankiewicz, A.R.; Sanders, J.K.M.;
Pantos¸, G.D. Top. Curr. Chem. 2012, 322, 217–260.
(17) Pangborn, A.B.; Giardello, M.A.; Grubbs, R.H.; Rosen,
R.K.; Timmers, F.J. Organometallics 1996, 15, 1518–1520.
(6) Huang, C.; Wen, L.; Liu, H.; Li, Y.; Liu, X.; Yuan, M.;
Zhai, J.; Jiang, L.; Zhu, D. Adv. Mater. 2009, 21,
1721–1725.