A self-assembled phthalocyanine dimer

Article abstract of DOI:10.1016/S0040-4039(00)00501-3

The synthesis of phthalocyanines 1a and 1b appended with four glycoluril modules is presented. H NMR spectroscopy showed that 1a and 1b self-assemble by means of hydrogen bonding to form discrete dimers. Capsule formation is solvent dependent where the dimeric assembly is most pronounced in aromatic solvents. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00501-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 3777–3780
A self-assembled phthalocyanine dimer
Arne Lützen, Stephen D. Starnes, Dmitry M. Rudkevich and Julius Rebek Jr. ∗
The Skaggs Institute for Chemical Biology and Department of Chemistry, The Scripps Research Institute, La Jolla,
CA 92037, USA
Received 23 February 2000; revised 17 March 2000; accepted 20 March 2000
Abstract
1
The synthesis of phthalocyanines 1a and 1b appended with four glycoluril modules is presented. H NMR
spectroscopy showed that 1a and 1b self-assemble by means of hydrogen bonding to form discrete dimers. Capsule
formation is solvent dependent where the dimeric assembly is most pronounced in aromatic solvents. © 2000
Elsevier Science Ltd. All rights reserved.
Keywords: phthalocyanine; self-assembly; supramolecular chemistry; hydrogen bonds.
Reversibly formed, self-assembled capsules are recently devised systems that provide new forms of
molecular recognition.1 Increasingly, these systems are incorporating function; one particularly attractive
function is as a reaction chamber.2 It could be advantageous to incorporate metallo catalytic sites into
these chambers, in a manner that has proven successful for other reversibly formed, self-assembled
systems,3 and we describe here our initial steps toward this goal.
Phthalocyanine was selected as framework for receptor scaffolding: the known catalytic properties,4
the C4v symmetry and the extended aromatic π surface that the molecule presents are all attractive featu-
res. The phthalocyanine core, in combination with the workhorse glycoluril and its self-complementary
hydrogen bonding motif, leads to structure 1 (Fig. 1). Inspection of the model suggests that it is pre-
organized for self-assembly and it does indeed give a discrete dimeric capsule in solution.
The synthesis of 1 (Scheme 1) started with the bromination of o-xylene to give 4,5-dimethyl-
1,2-dibromobenzene5 followed by a Rosenmund–von Braun reaction to give 4,5-dimethyl-1,2-
dicyanobenzene.6 A second bromination afforded 4,5-di(bromomethyl)-1,2-dicyanobenzene,7 which
was coupled with the PMB-protected glycoluril 28 to give 3. Deprotection of 3 with ceric ammonium
nitrate (CAN)8 proceeded smoothly to give the phthalocyanine precursor 4, which was cyclized to
the metallated phthalocyanine product 1a and 1b by gentle reflux of 4 in n-pentanol with Ni(OAc)2
or Zn(OAc)2 and a catalytic amount of DBU.9 At this stage, purification of 1a and 1b proved to be
quite tedious and not trivial. This, in part, is due to the necessity to separate the four stereoisomers of
1 produced in the synthesis. Borrowing the terminology popular in porphyrin chemistry, we describe
∗
Corresponding author. E-mail: jrebek@scripps.edu (J. Rebek Jr.)
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00501-3
tetl 16780

3778
Fig. 1. Phthalocyanine 1 and the energy minimized metal free dimeric assembly — side view and top view. CH hydrogens and
C7H15 chains are omitted for viewing clarity
the four stereoisomers as αααα, αααβ, ααββ, and αβαβ with respect to the orientation of the four
glycolurils, with the further understanding that the isomers are not interconvertible. To our delight, it
was found that extraction of the crude reaction mixture with p-xylene enriched the sample in the desired
αααα-isomer, which was then further purified by chromatography. During the course of the synthesis,
it was also found that impure 1 could be derivatized to the less polar Boc-protected derivative which was
purified by chromatography over silica gel; deprotection afforded 1 in return.
Scheme 1. Reagents and conditions: (a) NaH, DMF, rt; (b) 4,5-di(bromomethyl)-1,2-dicyanobenzene; (c) CAN,
CH3CN:THF:H2O 7:3:1, rt; (d) M(OAc)2, cat. DBU, n-pentanol, reflux; (e) Boc-anhydride, THF, DMAP; (f) purify over silica
gel; (g) CH2Cl2, TFA
Molecular modelling of 1 dimer using Macromodel10 and the Amber*11 force field suggested that
the assembly offers a spacious binding cavity. An internal volume of ∼275 Å3 was calculated using
GRASP.12 At least two molecules of benzene can fit comfortably inside. The overall shape can best be
described as a molecular box somewhat similar to the jelly doughnut,13 a structure that was shown to
encapsulate disc-like molecules (benzene and cyclohexane, for example).

3779
1H NMR spectroscopy was used to determine if the molecule self-assembled as predicted, through the
seam of 16 hydrogen bonds. The telltale sign indicating assembly is the chemical shift of the glycoluril-
NH signal — in the assembled state, it is expected downfield of 8 ppm, whereas in the non-assembled
state this signal is expected upfield of 6 ppm.2,8,13,14 From Table 1, which lists the chemical shift of the
glycoluril-NH of 1a and 4 in a variety of solvents, it is apparent that 1 assembles in aromatic solvents.
In p-xylene-d10, for instance, the chemical shift of the glycoluril-NH of 1a and 1b is 9.05 and 8.90 ppm,
respectively, indicative of the self-assembled dimer. In contrast, the glycoluril-NH resonance of 4 falls
at 4.97 ppm in p-xylene-d10. Fig. 2 provides a portion of the 1H NMR spectrum of 1a in p-xylene-d10.
In order to verify the nature of the resonance at 9.05 ppm as belonging to that of the glycoluril-NH,
a drop of D2O was added to the NMR sample, which resulted in the disappearance of the signal at
9.05 ppm (from Fig. 2a). Compounds 1a and 1b have also been shown to self-assemble in toluene-d8,
benzene-d6, and cyclohexane-d12. It is likely that the mere fact that 1 assembles in p-xylene assisted in
the purification of the compound. In the assembled state, 1 exposes 16 solubilizing alkyl groups to the
solvent while shielding two of its four aromatic surfaces. As a result, 1 is more soluble than the other
three stereoisomers in p-xylene and facilitates its own purification.
Table 1
1H NMR chemical shift of the glycoluril-NH of 1a and 4 in different solventsa
Fig. 2. Portion of the 1H NMR spectra of 1a (a) and 4 (b) in p-xylene-d10
Compound 1 does not assemble in CDCl3, CD2Cl2, cis/trans-decaline-d18, or acetone-d6. The lack of
assembly of 1 in these solvents indicates that the dimeric assembly formed in aromatic solvents may enjoy
additional stabilization from interior π-stacking interactions. With two molecules of benzene included
within the cavity, the volume shrinks slightly to 270 Å3 with a packing coefficient of 62% which is
slightly higher than the ideal packing coefficient of ∼55%.15 The two benzene molecules stack with
opposite phthalocyanine units at 3.4–3.5 Å and with each other at 3.8 Å.
The molecular recognition and catalytic properties of this assembly are currently being investigated
and will be reported in due course.

3780
Acknowledgements
We thank the Skaggs Research Foundation and the National Institutes of Health for financial support.
A.L. thanks the Deutsche Akademische Austauschdienst (DAAD) for a postdoctoral fellowship. We also
thank Prof. Dr. R. J. M. Nolte and Dr. A. E. Rowan of the University of Nijmegen for helpful discussions.
References
1. de Mendoza, J. Chem. Eur. J. 1998, 4, 1373–1377. Conn, M. M.; Rebek Jr., J. Chem. Rev. 1997, 97, 1647–1668, and
references cited therein.
2. Kang, J.; Santamaria, J.; Hilmersson, G.; Rebek Jr., J. J. Am. Chem. Soc. 1998, 120, 7389–7390. Kang, J.; Hilmersson, G.;
Santamaria, J.; Rebek Jr., J. J. Am. Chem. Soc. 1998, 120, 3650–3656. Kang, J.; Rebek Jr., J. Nature 1997, 385, 50–52.
3. Kuroda, Y.; Kawashima, A.; Hayashi, Y.; Ogoshi, H. J. Am. Chem. Soc. 1997, 119, 4929–4933. Drain, C. M.; Fischer, R.;
Nolen, E. G.; Lehn, J.-M. J. Chem. Soc., Chem. Commun. 1993, 243–245.
4. Phthalocyanines: Properties and Applications, Vol. 2; Leznoff, C. C.; Lever, A. B. P., Eds.; VCH: New York, 1993.
5. Klingsberg, E. Synthesis 1972, 29–30.
6. Pawlowski, G.; Hanack, M. Synthesis 1980, 287–288.
7. Wang, C.; Bryce, M. R.; Batsanov, A. S.; Stanley, C. F.; Beeby, A.; Howard, J. A. K. J. Chem. Soc., Perkin Trans. 2 1997,
1671–1678.
8. Rudkevich, D. M.; Rebek Jr., J. Angew. Chem., Int. Ed. Engl. 1997, 36, 846–848.
9. Selected MS–MALDI-TOF. Compound 1a: m/e 2628 ([M+H]+, calcd for C160H184N24O8NiH 2628). Compound 1b: m/e
2635 ([M+H]+, calcd for C160H184N24O8ZnH 2635).
10. Mohamadi, F.; Richards, N. G. J.; Guida, W. C.; Liskamp, R.; Lipton, M.; Caufield, C.; Chang, G.; Hendrickson, T.; Still,
W. C. J. Comput. Chem. 1990, 11, 440–467.
11. Allinger, N. L. J. Am. Chem. Soc. 1977, 99, 8127–8134.
12. Nicholls, A.; Sharp, K. A.; Honig, B. Proteins 1991, 11, 281–296.
13. Grotzfeld, R. M.; Branda, N.; Rebek Jr., J. Science 1996, 271, 487–489. O’Leary, B. M.; Grotzfeld, R. M.; Rebek Jr., J. J.
Am. Chem. Soc. 1997, 119, 11 701–11 702.
14. Wyler, R.; de Mendoza, J.; Rebek Jr., J. Angew. Chem., Int. Ed. Engl. 1993, 32, 1699–1701.
15. Mecozzi, S.; Rebek Jr., J. Chem. Eur. J. 1998, 4, 1016–1022.