hydroxy-5-nitrobenzoic acid (15), which could be converted
to azide 17 by reduction using ammonium ferrous sulfate in
concentrated ammonia and subsequently diazotized in the
presence of NaN3. Protecting the acid group in 17 to give
ester 18, which was alkylated to yield 19. Thus, through the
“click” coupling[20] of the two essential precursors men-
tioned above, monomer bearing 1,2,3-triazole group in the
middle of two phenyl groups and two function groups of
BocNH and ester at both ends can be obtained. A segment
doubling strategy involving selective deprotections and cou-
plings via acid chlorides was then adopted to generate oligo-
mers 1, 2 and 3. Note that to protect the Boc group, the acid
was activated to the corresponding acid chloride under mild
conditions using the Ghosez reagent (1-chloro-N,N,2-trime-
thylpropenylamine).[21] Compound 4 was obtained after cou-
pling the monomer acid chloride 20c with 9.
Scheme 1. Oligomers studied herein.
amide N-H group is widely used as hydrogen-bond donor to
construct anion receptors.[4f–i] These facts inspired us to
design oligo(phenyl-amide-triazole)s 1–4, which possess two
phenyl rings, a triazole and an amide group in the repeat
unit that is supposed to provide moderate affinities for
halide ions. According to molecular modelling (Figure S1 in
the Supporting Information), oligomer 1 exists as a helical
conformation in one turn in the presence of halide ions. The
longer oligomers such as 2 and 3, in principle, were designed
to yield two and three helical turns, respectively. We hy-
pothesized that longer oligomers could bind more than one
anion like chloride ions. Short oligomer 4 was designed to
assess the affinity between phenyl-amide-triazole motif and
halide ions and also to study the conformation of their com-
plex in the solid state.
Binding properties: The folding behaviors of oligomers 1–4
1
1
were investigated by H NMR spectroscopy. H NMR spec-
tra of 1 and 4 initially recorded in CDCl3 show one set of
sharp signals indicative of nonaggregation taking place. Un-
expectedly, the longer oligomers 2 and 3 were found to ag-
gregate under the identical conditions. This aggregation of 2
can clearly be detected by concentration-dependent
1H NMR spectra and 2D DOSY experiments (Figures S2, S3
and S4). To avoid this aggregation, several other commer-
cially available deuterated solvents were tested, and
[D5]Pyridine ([D5]Pyr) and the mixture of [D6]DMSO/
[D5]Pyr were chosen for investigations as indicated (see
below and Figures S5 and S6).
The 1H NMR spectrum of 4 in [D5]Pyr shows considerable
changes in the chemical shift upon addition of halide ions,
indicative of strong interactions between the phenyl-amide-
triazole motif and halide ions. In details, upon addition of
Bu4N+ClÀ, the chemical shifts of the triazole and amide pro-
tons shifted downfield up to Dd=1.8 and 1.1 ppm, respec-
tively, as a result of the formation of hydrogen bonding (Fig-
ure S7). The titrations of bromide also produced a large
change in the chemical shift of the protons at triazole and
amide moieties but less than that of chloride ions, and the ti-
trations of iodide ions gave the smallest change in chemical
shift (Figures S10 and S13). The fitting analyses of binding
data[22] yield the association constants (K) of 540, 83 and
11mÀ1 in a 1:1 binding model for chloride, bromide and
iodide ions, respectively (Table 1), suggesting the binding is
highly dependent on the inducing capability of halide ions.
Although an attempt to co-crystallize 4 and chloride ion
failed, the results mentioned above still unambiguously sug-
gest the interaction between the phenyl-amide-triazole motif
and halide ions is directional and sufficiently strong so as to
induce the longer oligomer to fold into a helical conforma-
tion.
Results and Discussion
Synthesis: The preparations of the proposed oligomers 1–4
consisting of two functional group in the repeated units, that
is, amide and 1,2,3-triazole, are depicted in Scheme 2. The
key issue in this approach is to obtain two essential precur-
sors, that is, N-Boc-3-ethynyl-5-isobutoxyaniline and methyl
3-azido-5-isobutoxybenzoate. Initially, 3,5-dibromophenol
(6), which was prepared from the comercially available pen-
tabromophenol (5) according to the method reported by
Tour,[18] was alkylated under an alkaline condition to pro-
duce 3,5-dibromo isobutoxybenzene (7). By treating with
nBuLi and TsN3, one of the bromo groups in 7 was convert-
ed to azide to yield 8 at À788C in an acetone bath. Com-
pound 8 was reduced by NaBH4 to generate amine com-
pound 9, which was protected by tert-butoxycarbonyl (Boc)
group via (Boc)2O to provide 10. Then, Sonogashira reac-
tion[19] of 10 with excess 2-methylbut-3-yn-2-ol at 788C yield-
ed 11. In this reaction, we selected 2-methylbut-3-yn-2-ol in-
stead of trimethylsilylacetylene to avoid the tube-sealing op-
eration due to the fact that the boiling point of trimethyl-
silylacetylene (538C) is lower than the reaction temperature.
Treatment of 11 with NaH in anhydrous toluene produced
12. On the other hand, 14 can be prepared from 3,5-dinitro-
benzoic acid as previously reported.[15] The amine group in
14 went through diazotation and then hydrolyzed to yield 3-
The folding of oligomer 1 in the presence of chloride ions
1
were first demonstrated by H NMR titration experiments.
The marked signals of 1 were assigned on the basis of
COSY, HSQC, TOCSY and 2D NOESY experiments (see
1
Supporting Information). It can be found that the H NMR
614
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 613 – 619