One-pot reaction as an efficient method for rigid molecular tweezers

Article abstract of DOI:10.1021/ol8018067

(Equation Presented) Rigid molecular tweezers are compounds of increasing scientific interest. As the structural requirements for such compounds are highly specific, few types of these tweezers are thus far known. The preparation of examples of rigid larg

Full text of DOI:10.1021/ol8018067

ORGANIC
LETTERS
2008
Vol. 10, No. 21
4767-4769
One-Pot Reaction as an Efficient Method
for Rigid Molecular Tweezers
Martin Havlík, Vladimír Kra´l, Robert Kapla´nek, and Bohumil Dolensky´*
Department of Analytical Chemistry, Institute of Chemical Technology, Prague,
Technicka´ 5, Praha 166 28, Czech Republic
dolenskb@Vscht.cz
Received August 7, 2008
ABSTRACT
Rigid molecular tweezers are compounds of increasing scientific interest. As the structural requirements for such compounds are highly
specific, few types of these tweezers are thus far known. The preparation of examples of rigid large-pincered molecular tweezers based on
bis Tro¨ger’s bases derived from 1,4-benzenediamine is described. In addition, evidence is presented of the different binding abilities of the
diastereoisomers of such compounds.
Analogous to common tweezers, molecular tweezers are
compounds which consist of two pincers able to bind a guest
compound through the intermolecular interactions occurring
on each side. The pincers are joined together by a tether,
which keeps them more or less preorganized. Most studies
of molecular tweezers are focused on rigid tweezers, i.e.,
molecules consisting of a rigid tether and reasonably large,
electron-rich arenes (π-donor, D, host). Generally, the goal
of such studies is to bind an electron-deficient arene
(π-acceptor, A, guest) via the formation of a D-A-D
π-sandwich.1 To achieve this, the tether should maintain the
pincers in parallel alignment at a distance of about 0.64 to
0.70 nm from each other.1c As many electron-deficient arenes
have been shown to be important for both health and the
environment, molecular tweezers have been studied
intensively, both experimentally and using quantum chemical
calculations.2,3 The first examples of their potential use in
medicine3,4 and analytical applications5 have recently been
published.
As the concept of rigid molecular tweezers is quite young
and the requirement of keeping parallel pincers at a proper
distance from each other is very limiting, there are few
known structural motifs for useful tethers. Until recently,
rigid tweezers were based on derivatives of glycolurils,6
Kagan’s ethers,7 or methanoanthracenes.3,8 However, the
(2) (a) Kla¨rner, F.-G.; Panitzky, J.; Preda, D.; Scott, L. T. J. Mol. Model
2000, 6, 318–327. (b) Chang, C.-E.; Gilson, M. K. J. Am. Chem. Soc. 2004,
126, 13156–13164.
(3) (a) Kla¨rner, F.-G.; Kahlert, B.; Nellesen, A.; Zienau, J.; Ochsenfeld,
C.; Schrader, T. J. Am. Chem. Soc. 2006, 128, 4831–4841. (b) Talbiersky,
P.; Bastkowski, F.; Kla¨rner, F.-G.; Schrader, T. J. Am. Chem. Soc. 2008,
130, 9824–9828.
(1) The molecular tweezers idea: (a) Chen, C.-W.; Whitlock, H. W.
J. Am. Chem. Soc. 1978, 100, 4921–4922. Reviews: (b) Zimmerman, S. C.
Bioorg. Chem. Front. 1991, 2, 33–71. (c) Zimmerman, S. C. Top. Curr.
Chem. 1993, 165, 71–102.
(4) Weiss, S. T.; McIntyre, N. R.; McLaughlin, M. L.; Merkler, D. J.
Drug DiscoVery Today 2006, 11, 819–824.
(5) Kamieth, M.; Burkert, U.; Corbin, P. S.; Dell, S. J.; Zimmerman,
S. C.; Kla¨rner, F.-G. Eur. J. Org. Chem. 1999, 2741–2749.
10.1021/ol8018067 CCC: $40.75
Published on Web 10/01/2008
2008 American Chemical Society

beginning of the 21st century saw the introduction of new
molecular tweezers based on Tro¨ger’s base derivatives,9
bisTB.10,11 The unique property of bisTB, its ability to
isomerize between tweezer diastereoisomers (syn-bisTB) and
nontweezer diastereoisomers (anti-bisTB) in acid media,12
has potential for future applications. The requirements for
rigid molecular tweezers (vide supra) are best met by bisTB
regioisomer 1 derived from 1,4-benzenediamine. However,
until now, the preparation of bisTBs 1 with reasonably large
pincers has not been successfully achieved.13
Scheme 1
To date, bisTBs 1 have been prepared entirely via the
troegeration14 of corresponding amine intermediates prepared
exclusively by the reduction of equivalent amides. The TB
units can be constructed simultaneously in one reaction step11
or in separate steps.10 The one-step simultaneous formation
of both TB units from the corresponding tetraamine is
obviously the shortest way of preparing symmetrical bisTBs.
However, we recently discovered that the preparation of the
corresponding amides is problematic. Furthermore, due to
their instability in relation to reduction agents, we found it
impossible to reduce the matching amides derived from large-
arene-amines.15 Another approach involves the mixed tro-
egeration of 1,4-benzenediamine and an arylamine. However,
the scope of this reaction is unknown (having only been
described with 4-methoxyaniline12), and the isolation of the
bisTB from the complex reaction mixture is highly prob-
lematic. In this article, we present synthetic pathways for
the preparation of the first examples of the rigid large-
pincered molecular tweezers, syn-1b and syn-1c, as well as
the first evidence of the different binding abilities of syn-1b
and anti-1b.
3 being followed by monobromide 4 (6:5). (The use of longer
reaction times and/or higher amounts of bromination agent
led to even more complex mixtures.) The treatment of the
mixture with an excess of methyl 6-amino-2-naphthoate
(brom to amine 1:4.7, temp 75 °C, time 5 h) converted
monobromide 4 into monoarylamine 5 as expected,16 but
dibromide 3 produced no desired disubstitution product 6
but cyclic amine 7 only. In contrast, the treatment of
hexakis(bromomethyl)benzene in neat aniline has been
shown to give a hexasubstitution product in a preparative
yield of 47% (bromine to amine 1:18, temp 150 °C, time
27 h), wherein no cyclic products were reported.17
The aim of our second approach was to bypass the
intramolecular attack (leading to cyclic amine 7) via the
treatment of dibromide 3 with the sodium salt of amine.
However, we were aware that such an approach would
inevitably lead to acid-base equilibrium with protons of the
BocNH groups and, thereby, complicate the reaction. There-
fore, we protected the BocNH groups by additional Boc
groups (Scheme 2).
First, we tried to prepare bisTB 1 using a method similar
to the successful preparation of bisTBs derived from 1,3-
benzenediamine.15 Correspondingly (Scheme 1), the starting
bis(BocNH)xylene 2 (Boc ) t-butoxycarbonyl) was prepared
from commercially available 2,3-dimethylaniline in five steps
(overall yield 29%). However, the subsequent bromination
resulted in only partial conversion, the expected dibromide
In accordance with the known methods,18 the treatment
of bis(BocNH)xylene 2 with Boc2O (di-t-butyl dicarbonate)
gave bis(Boc2N)xylene 8. Contrary to the bromination of
bis(BocNH)xylene 2 (vide supra), the bromination of
bis(Boc2N)xylene 8 gave the expected pure dibromide 9. The
obtained dibromide was subsequently treated with sodium
salt of the corresponding arylamine, resulting in a mixture
of products. NMR analysis showed that some protection
groups were lost18,19 but that the expected substitution took
place, meaning that partially deprotected tetraamines 10 were
formed. As the protection of amino groups is not required
for the subsequent troegeration steps, we used the obtained
mixture without purification. The obtained mixture contained
both diastereoisomers of the expected bisTB 1. The new
compounds, syn and anti diastereoisomers of the naphthalene
bisTB derivatives 1b,c, were isolated in good yields by
(6) (a) Smeets, J. W. H.; Sijbesma, R. P.; Niele, F. G. M.; Spek, A. L.;
Smeets, W. J. J.; Nolte, R. J. M. J. Am. Chem. Soc. 1987, 109, 928–929.
(b) Rowan, A. E.; Elemans, J. A. A. W.; Nolte, R. J. M. Acc. Chem. Res.
1999, 32, 995–1006.
(7) Kla¨rner, F.-G.; Kahlert, B. Acc. Chem. Res. 2003, 36, 919–932.
(8) Harmata, M. Acc. Chem. Res. 2004, 37, 862–873.
(9) The most recent review: Dolensky´, B.; Elguero, J.; Kra´l, V.; Pardo,
C.; Val´ık, M. AdV. Heterocycl. Chem. 2007, 93, 1–56.
(10) Pardo, C.; Sesmilo, E.; Gutierrez-Puebla, E.; Monge, A.; Elguero,
J.; Fruchier, A. J. Org. Chem. 2001, 66, 1607–1611
(11) Val´ık, M.; Dolensky´, B.; Petr˘´ıc˘kova´, H.; Kra´l, V. Collect. Czech.
Chem. Commun. 2002, 67, 609–621
.
.
(12) (a) Dolensky´, B.; Val´ık, M.; Mateˇjka, P.; Herdtweck, E.; Kra´l, V.
Collect. Czech. Chem. Commun. 2006, 71, 1278–1302. Preparations of
different oligoTB derivatives by mixed troegeration were published recently.
(b) Dolensky´, B.; Val´ık, M.; Sy´kora, D.; Kra´l, V. Org. Lett. 2005, 7, 67–
ˇ
70. (c) Val´ık, M.; Cejka, J.; Havl´ık, M.; Kra´l, V.; Dolensky´, B. Chem.
(16) Note that troegeration of monoarylamine 5 gave an unidentified
compound consisting of one TB unit.
Commun. 2007, 37, 3835–3837.
(13) Havl´ık, M.; Kra´l, V.; Dolensky´, B. Org. Lett. 2006, 8, 4867–4870.
(14) Troegeration is defined as any treatment of an arylamine derivative
with a source of formaldehyde (e.g., aqueous solution, paraformaldehyde,
hexamethylenetetraamine, dimethoxymethane, trioxane) in acidic condition
(e.g., HCl, TFA) that leads to a Tro¨ger’s base derivative, at least in the
case of troegerable arylamine derivatives (i.e., arylamine derivative giving
a Tro¨ger’s base derivative under troegeration).
(17) Hardy, A. D. U.; MacNicol, D. D.; Wilson, D. R. J. Chem. Soc.,
Perkin Trans. 2 1979, 1011–1019.
(18) (a) Boger, D. L.; McKie, J. A.; Nishi, T.; Ogiku, T. J. Am. Chem.
Soc. 1997, 119, 311–325. (b) Dambrough, S.; Mervic, M.; Condon, S. M.;
Burns, C. J. Synth. Commun. 2001, 31, 3273–3280.
(19) Boc2N groups are unstable in basic condition, e.g.: (a) Herzig, S.;
Kritter, S.; Lu¨bbers, T.; Marquardt, N.; Peters, J.-U.; Weber, S. Synlett 2005,
3107–3108. (b) Grehn, L.; Gunnarsson, K.; Ragnarsson, U. Acta Chem.
(15) Havl´ık, M.; Kra´l, V.; Dolensky´, B. Collect. Czech. Chem. Commun.
2007, 72, 392–402.
Scand., Ser. B 1987, 41, 18–23
.
4768
Org. Lett., Vol. 10, No. 21, 2008

able 2,3-dimethylaniline (6% 1b, 4% 1c), from the interme-
diate dibromide 9 (26% 1b, 18% 1c), and from commercially
available 1,4-benzenediamine (one-pot preparation: 8% 1a,
13% 1b, 6% 1c) shows that, in spite of the poor yields, one-
pot preparation is efficient and suitable for upscaling.
The prepared compounds syn-1a-c represent rigid mo-
lecular tweezers; e.g., syn-1b has large pincers close to
parallel (32°) and at an appropriate distance (0.67-0.80 nm)
from each other.20 In addition, a tweezer effect was observed
when colorless chloroform solutions (4.3 × 10-3 M) of 1b
diastereoisomers were doped with equimolar amounts of
tetracyanoethylene, the syn-1b solution turning darker purple
than anti-1b. The formations of charge-transfere complexes21
were followed by UV-vis absorption spectra (Figure 1).
Scheme 2
column chromatography (16% anti-1b, 10% syn-1b, 13%
anti-1c, 5% syn-1c).
For our third approach, we attempted one-pot mixed
troegeration. 1,4-Benzenediamine, hexamethylenetetraamine,
and arylamine a-c (in the molar ratio 10:19:28) were treated
in TFA at room temperature for five days. With bisTB 1
standards in hand, repeated column chromatography was used
to isolate the expected bisTB compounds from the corre-
sponding complex reaction mixtures. Thus, 4-methoxyaniline
yielded 5% of anti-1a and 3% of syn-1a; naphthyl-2-amine
gave 9% of anti-1b and 4% of syn-1b; and methyl 6-amino-
2-naphthoate produced 4% of anti-1c and 2% of syn-1c.
In all cases, diastereoisomers (anti-1 and syn-1) were
formed in a ratio of approximately 2:1. The difference in
the chemical shifts of N-CH2-N protons (measured in
CDCl3) was twice as great (∆δ ) 0.09 - 0.13 ppm) for the
isomers with higher Rf values (silica, CH2Cl2/CH3OH 95:5)
than for the isomers with lower Rf values (∆δ ) 0.04 -
0.05 ppm). Since the stereostructures of both anti-1a and
syn-1a diastereoisomers are known from X-ray diffraction,12
we assume that the isomers with higher Rf values are the
anti diastereoisomers, and those with lower Rf values are
the syn diastereoisomers.
Figure 1. Charge-transfer bands of tetracyanoethylene complexes
with syn-1b (upper spectrum, left cuvette) and anti-1b (lower
spectrum, right cuvette).
Further detailed studies of the binding abilities of bisTBs
are already underway.
Acknowledgment. This work was supported by the
Ministry of Education, Youth and Sports of the Czech
Republic (LC06077, LC512, and MSM 6046137307) and
by the Grant Agency of the Czech Republic (203/08/1445).
Supporting Information Available: Experimental pro-
cedures and spectroscopic data for all presented compounds
and complexes. This material is available free of charge via
the Internet at http://pubs.acs.org.
In conclusion, we have described two synthetic pathways
for the preparation of new bisTB 1 derivatives. Although
both step-by-step and one-pot preparations should be further
optimized, a preliminary comparison can be made. Step-by-
step preparation is time-consuming (involving nine reaction
steps) and expensive compared with one-pot preparation. The
comparison of preparative yields from commercially avail-
OL8018067
(20) Measurements were made on the molecular model of syn-1b,
wherein the geometry was optimized by MM2. The angle is formed by
planes of naphthalene moieties, and the distances were measured between
the centroids of the naphthalene rings (see the abstract picture).
(21) Masnovi, J. M.; Kochi, J. K.; Hilinski, E. F.; Rentzepis, P. M. J.
Phys. Chem. 1985, 89, 5387–5395.
Org. Lett., Vol. 10, No. 21, 2008
4769