chemical shift of Hc moves downfield from d 11.8 to d 12.4. Hb, on
the other hand, shifts even more substantially from d 8.1 to d 10.2
(yddimer for Hb calculated from dilution studies), in accordance
with its intermolecular hydrogen bonding. Also significantly, the
amino protons N(Ha/a9) decoalesce at y5 uC. The chemical shift of
proton Ha9 appears at d y5.9 and remains there as the
temperature is lowered,19 while Ha, participatory in dimer
formation, moves downfield from d 8.4 (5 uC) to d 9.3 (255 uC).
Given that the UDAP derivatives dimerize somewhat more
weakly (by 10-fold) than the most comparable DADA (self-
complementary) quadruple hydrogen bonding system (the alkyl
ureidotriazines of Meijer and co-workers9,20), there is additional
optimization to do and subtleties to be understood. We would
expect the Kdim of 1 to be affected by (a) the interaction of the urea
and N(9) substituents and (b) the use of aryl rather than the more
commonly employed alkyl ureas.9,21 We initiated explorations of
‘‘(a)’’ by preparing the N(7)-alkylated regioisomer of 1a (ESI{).
We were surprised to find that this regioisomer does not dimerize
by quadruple hydrogen bonding in CDCl3;22 hence, interaction
between the N(9) and urea substituents appears to be an important
(and modifiable) parameter in these systems. Along the same lines,
we studied the intrinsic 1N1 vs. 1N3 conformational preference by
computation. When a substituent in the N(9) position is too small
to interact appreciably with the phenylurea group, such as methyl
(1, where R 5 H), computation (MP2/6–31G*//HF/6–31G*)
shows that the 1N1 and 1N3 conformers are essentially isoenergetic,
y0.55 kcal mol21 in favour of the desired 1N3 conformer in the
gas phase. This leaves the monomer conformational equilibrium,
and also likely Kdim, easily perturbed. We are currently developing
the synthetic chemistry to test ‘‘(b)’’.
1 R. P. Sijbesma, F. H. Beijer, L. Brunsveld, B. J. B. Folmer,
J. Hirschberg, R. F. M. Lange, J. K. L. Lowe and E. W. Meijer,
Science, 1997, 278, 1601–1604; F. H. Beijer, R. P. Sijbesma,
H. Kooijman, A. L. Spek and E. W. Meijer, J. Am. Chem. Soc.,
1998, 120, 6761–6769; F. H. Beijer, H. Kooijman, A. L. Spek,
R. P. Sijbesma and E. W. Meijer, Angew. Chem., Int. Ed., 1998, 37,
75–78; S. H. M. So¨ntjens, R. P. Sijbesma, M. H. P. van Genderen and
E. W. Meijer, J. Am. Chem. Soc., 2000, 122, 7487–7493.
2 V. G. H. Lafitte, A. E. Aliev, P. N. Horton, M. B. Hursthouse, K. Bala,
P. Golding and H. C. Hailes, J. Am. Chem. Soc., 2006, 128, 6544–6545.
3 This nomenclature is used throughout the paper: D 5 hydrogen bond
donor; A 5 hydrogen bond acceptor.
4 L. Brunsveld, B. J. B. Folmer, E. W. Meijer and R. P. Sijbesma, Chem.
Rev., 2001, 101, 4071–4097; A. W. Bosman, L. Brunsveld, B. J. B.
Folmer, R. P. Sijbesma and E. W. Meijer, Macromol. Symp., 2003, 201,
143–154; R. P. Sijbesma and E. W. Meijer, Chem. Commun., 2003, 5–16
and references therein.
5 D. W. Guo, R. P. Sijbesma and H. Zuilhof, Org. Lett., 2004, 6,
3667–3670.
6 T. Park, S. C. Zimmerman and S. Nakashima, J. Am. Chem. Soc., 2005,
127, 6520–6521; T. Park, E. M. Todd, S. Nakashima and S. C.
Zimmerman, J. Am. Chem. Soc., 2005, 127, 18133–18142; H. C. Ong
and S. C. Zimmerman, Org. Lett., 2006, 8, 1589–1592; T. Park and
S. C. Zimmerman, J. Am. Chem. Soc., 2006, 128, 11582–11590.
7 For multimeric hydrogen-bonded assemblies of guanines (e.g.
G-quartets) see: J. T. Davis, Angew. Chem., Int. Ed., 2004, 43,
668–698 and references therein.
8 Not shown are the two lowest energy syn-conformers (non-hydrogen-
bonded), calculated at the molecular mechanics level (Amber* and
MM3* in CHCl3) to be y5 kcal mol21 higher in energy (for R 5 H in
Fig. 1).
9 These features have been discussed for various urea-functionalized
heterocycles. See ref. 1 and F. H. Beijer, R. P. Sijbesma, J. Vekemans,
E. W. Meijer, H. Kooijman and A. L. Spek, J. Org. Chem., 1996, 61,
6371–6380.
10 Eur. Pat., EP 543095, 1993.
11 E. Camaioni, S. Costanzi, S. Vittori, R. Volpini, K. N. Klotz and
G. Cristalli, Bioorg. Med. Chem., 1998, 6, 523–533.
12 The regioisomeric mixture depends to some extent on the size of the
electrophile, as the equivalent alkylation with 3,5-diheptyloxybenzyl
bromide yields the desired N(9) isomer in over 80% yield (ESI{).
13 Poor reactivity has also been observed with guanine derivatives (see
ref. 6). In our experience, the nucleophilicity of the amino group on C(2)
is very sensitive to the C(6) substituent. With a deactivating –Cl group in
this position (i.e. 3a), even reactivity with aryl isocyanates is poor.
Strongly activating groups, such as N(CH3)2 (preparation of 5), see
conversion to the urea in minutes at room temperature.
14 S. Porcher and S. Pitsch, Helv. Chim. Acta, 2005, 88, 2683–2704.
15 Ribose groups may be accommodated but have yet to be completely
explored due to their chemical lability during our initial studies (for
similar observations, see ref. 6): A. M. Martin and R. E. Giessert,
unpublished work.
16 E. A. Meyer, R. K. Castellano and F. Diederich, Angew. Chem., Int.
Ed., 2003, 42, 1210–1250.
17 The upfield shift of Hc with dilution demonstrates weakening, through
induction, of the intramolecular hydrogen bond upon dimer decom-
plexation (see ref. 6 for related observations).
18 B. R. Peterson, Associate 1.6, PhD thesis, University of California, Los
Angeles, 1994.
19 This demonstrates that the Hoogsteen edge of the molecule is not
engaged in measurable hydrogen bonding at these concentrations and
temperatures (see ref. 6).
To conclude, the first self-complementary, quadruply hydrogen
bonding purines have been prepared from the readily-available 2,6-
diaminopurine scaffold. The monomers are routinely synthesized
and should offer unique handles (and bioinspired strategies) for the
control of association strength through their multiple substitution
sites, some remote from the hydrogen bonding interface. Their
expanded p-surfaces will facilitate their association into stacked
assemblies and more complex architectures. Explorations in these
directions are currently under way.
We are grateful to the University of Florida and the Research
Corporation (Research Innovation Award (RI-1198) to R. K. C.)
for financial support. R. S. B. and R. E. G. were funded through
University of Florida Alumni Fellowships. K. A. A. wishes to
acknowledge the National Science Foundation and the University
of Florida for funding the X-ray equipment. We are grateful to
Prof. Blake R. Peterson (Penn State) for providing a copy of
Associate 1.6 for the calculation of dimerization constants and
Prof. Adrian Roitberg (UF) for computational time.
20 Comparisons between multiply (.3) hydrogen-bonded systems are not
straightforward. For a discussion, see: O. Lukin and J. Leszczynski,
J. Phys. Chem. A, 2002, 106, 6775–6782.
21 Although the arylurea Hc proton is more acidic (the pKa (DMSO) for
urea is 26.7 and for diphenylurea is 19.6; see: F. G. Bordwell, Acc.
Chem. Res., 1988, 21, 456–463), its carbonyl is a correspondingly weaker
base.
22 While Hc is intramolecularly hydrogen-bonded (d 11.5), Hb is only
modestly deshielded (d 7.8). Weaker DA-type hydrogen bonding may be
operative due to a preferred 1N1 conformation and/or distortion at the
C(6) amino group.
Notes and references
{ Crystal data for 5: C21H21N7O (M 5 387.45), monoclinic, space group
P21/n, radiation type 5 Mo-Ka, l 5 0.71073 s, a 5 9.0629(6),
b 5 19.8629(13), c 5 11.3182(7) s, a 5 c 5 90, b 5 106.862 (1)u,
V 5 1949.9(2) s3, Z 5 4, m 5 0.087 mm21, Dc 5 1.320 g cm23
,
F(000) 5 816, T 5 173(2) K, 4406 independent reflections (Rint 5 0.0337),
final R indices (272 parameters) [I . 2s(I)] were R1 5 0.0374, wR2 5 0.0966
(using 3194 reflections), GOF 5 1.063. Refinement was done using F2.
CCDC 615504. For crystallographic data in CIF or other electronic format
see DOI: 10.1039/b610239e
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 4413–4415 | 4415