Hollow porphyrin prisms: Modular formation of permanent, torsionally rigid nanostructures via templated olefin metathesis

Article abstract of DOI:10.1039/b800063h

Hollow, hexa-porphyrin prisms of two sizes were template-assembled and covalently locked, via cross-olefin metathesis, into permanent, torsionally rigid structures whose active sites (metal sites) can be both accessed and altered in a facile manner. The Royal Society of Chemistry.

Full text of DOI:10.1039/b800063h

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Hollow porphyrin prisms: modular formation of permanent, torsionally
rigid nanostructures via templated olefin metathesisw
Kyoung-Tae Youm, SonBinh T. Nguyen* and Joseph T. Hupp*
Received (in Berkeley, CA, USA) 7th January 2008, Accepted 13th March 2008
First published as an Advance Article on the web 27th June 2008
DOI: 10.1039/b800063h
Hollow, hexa-porphyrin prisms of two sizes were template-
assembled and covalently locked, via cross-olefin metathesis,
into permanent, torsionally rigid structures whose active sites
(metal sites) can be both accessed and altered in a facile manner.
plate to assemble three porphyrins into trigonal shapes and then
permanently ‘‘set’’ the structure using irreversible Glaser coup-
ling.7a Reversible bond formation, referred to as dynamic combi-
natorial chemistry (DCC), has also been used by Sanders et al. to
create more flexible disulfide-linked cyclic porphyrin dimers,
trimers, and tetramers in the presence of the appropriate templa-
tes.7b,c Other notable examples deploy the aforementioned tRCM
idea to form molecular squares.6a,b In each case, however, sig-
nificant torsional freedom remains in the assembled structure after
template removal, potentially reducing its usefulness in applica-
tions where structural rigidity is important.
Discrete multiporphyrin assemblies have received considerable
attention due to their potential in light-harvesting, sensing,
molecular transport, and catalysis.1,2 Ideally, such assemblies
should be: (a) easy to construct, (b) persistent and robust, (c)
topologically well-defined, (d) hollow (to enable molecular siev-
ing, analyte binding, reactant encapsulation, etc.), and (e) highly
functional (luminescent, catalytic, etc.). Interestingly, while a
wide variety of cyclic porphyrin assemblies having one or more
of these characteristics exist, few, if any, possess all of them. Two
examples from our own labs may serve to illustrate this. The
first, a family of M4Por4 ‘‘molecular squares’’ featuring por-
phyrin edges (Por) and coordinated metal corners (M), are
obtainable in high yield by coordinative self-assembly.3 While
suitable for sieving3d,e,h,i and sensing,3c,g they lack the M-por-M
torsional rigidity3b,e,j,4 and resulting well-defined cavity geome-
try (ESIw, Fig. S1) needed for strong binding of guests or for
highly selective catalysis. A second example comprises porphyrin
prisms where replacement of single-porphyrin edges with di-,
tri-, or tetra-porphyrin panels enables redundant coordination
(directly at porphyrin Zn(II) sites), thereby fixing the overall
assemblies and their component panels in well-defined, rigid
geometries.5 Unfortunately, this assembly strategy leaves the
porphyrin metal sites blocked and the cavities occupied. Addi-
tionally, because the Zn-ligand bonds are weak, the assemblies
dissociate in highly polar solvents or at high dilution.
As noted above, we have previously described the reversible
assembly of prisms5 possessing multi-porphyrin panels. Here,
we expand upon this strategy to construct torsionally rigid,
permanent prisms. As proof of principle, we selected panels
composed of two Zn(porphyrin) rings fully conjugated
through a butadiyne linker (Scheme 1). The judicious use of
alkenyloxy and trialkylsilylalkynyl substituents allows these
building blocks and their assembled structures to remain fully
soluble in solvents such as methylenechloride, toluene, and
tetrahydrofuran (THF). As expected, both Zn2–AA and
Zn2–PP dimers readily bind to Lewis bases such as pyridine
and this interaction can be used to pre-organize the panels into
assemblies that are ready for tRCM.
Exposing a 3 : 2 molar ratio mixture of Zn2–AA and Py3T
to Grubbs’ first-generation catalyst8 in CH2Cl2 results in the
formation of the covalently linked, templated trigonal prism
(Zn2–AA)3(Py3T)2 in high yield (77%) after size-exclusion
chromatography.z Similar tRCM of a 3 : 2 molar mixture of
Zn2–PP and tris((4-pyridyl)ethynyl)benzene (TPEB) afforded
76% of the corresponding (Zn2–PP)3(TPEB)2 permanent
prism. The templates are critical to the formation of the
corresponding prisms; only (Zn2–AA)2 or (Zn2–PP)2 (i.e.,
dimers) form in their absence. In addition, the size of the
template must match well that of the corresponding prism; the
yield of the covalently stabilized prism (Zn2–PP)3, for exam-
ple, is very low when assembled via the smaller template, Py3T.
The identities of (Zn2–AA)3(Py3T)2 and (Zn2–PP)3(TPEB)2
were confirmed by NMR spectroscopy, gel-permeation chroma-
tography (GPC), and MALDI-TOF MS. The 1H NMR spectrum
of (Zn2–AA)3(Py3T)2 differs significantly from those for Zn2–AA
and the Py3T template (ESIw, Fig. S2). While Zn2–AA exhibits
distinct olefinic (6.29(m), 5.64(d), and 5.47(d) ppm) and allylic
(4.85(d) ppm) resonances, the olefinic protons of (Zn2–AA)3-
(Py3T)2 appear as broad singlets at 6.47 and 6.34 ppm, indicative
of both E and Z conformations. Accordingly, its allyl protons also
appear as two discrete singlets at 5.37 and 5.22 ppm. Notably, the
chemical shifts of the pyridyl protons of the encapsulated Py3T
template appear at 2.45 and 5.91 ppm, significantly upfield from
Herein, we demonstrate that templated ring-closing metathesis
(tRCM) can be used for the facile preparation of both metallated
and metal-free porphyrin prisms possessing all five of the above-
mentioned properties, in addition to excellent solubility and che-
mical stability in high-polarity solvents. We also show that tRCM-
prepared assemblies can be demetallated and, if desired, remetal-
lated with the same or other ions, without structural degradation.
Cyclic multiporphyrin assemblies have previously been ob-
tained in high yield via a stepwise approach where modular
building blocks are organized into the desired shape by a template
prior to structure-setting.6,7 In an elegant example, Anderson and
coworkers employed 2,4,6-tris(4-pyridyl)triazine (Py3T) as a tem-
Department of Chemistry, Northwestern University, 2145 Sheridan
Road, Evanston, IL 60208-3113, USA. E-mail: stn@northwestern.edu;
j-hupp@northwestern.edu; Fax: +1 847-491-7713;
Tel: +1 847-467-3347
w Electronic supplementary information (ESI) available: Details of
synthesis and spectroscopic data (UV-vis, NMR, GPC, and MALDI-
TOF MS). See DOI: 10.1039/b800063h
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 3375–3377 | 3375

View Article Online
Scheme 1 The tRCM syntheses of porphyrinic trigonal prisms. For clarity, template ligands are omitted from the product assembly.
those observed for free Py3T (8.59 and 8.95 ppm), indicating a
strongly shielded environment due to binding of the template
within the trigonal prism cavity.7a,9,10 Similar to that for
(Zn2–AA)3(Py3T)2, the 1H NMR spectrum of (Zn2–PP)3(TPEB)2
also shows dramatic changes from those for free Zn2–PP and
TPEB (ESIw, Fig. S2). For example, the olefinic protons (6.00(m),
5.21(d), and 5.12(d) ppm) of free Zn2–PP disappear and new
tional dimensions (albeit, much different depth dimensions). It
is also noteworthy that the GPC peak shapes for the prisms
are clearly more symmetrical than those for free panels, with
the largest prism displaying the most symmetrical shape. All
else being equal, a more narrow and symmetrical shape implies
a more rigid structure (i.e., less conformational freedom).
Fig. 1 (left panel) compares the electronic absorption spectra
for the free panel (Zn2–AA), a 3 : 2 molar ratio mixture of
Zn2–AA and Py3T, and the permanent, covalently linked prism
(Zn2–AA)3(Py3T)2. Formation of the labile prism from the free
panel is marked by intensification of the porphyrin By band and
by narrowing and red-shifting of the Qy band. The narrowing is
consistent with dimer panel rigidification and diminution of
rotational isomerization.5,10 The red-shift is consistent with panel
planarization and concomitant enhancement of porphyrin–por-
phyrin electronic coupling, as previously noted by Anderson and
co-workers in their studies of ‘‘ladder’’ assemblies of conjugated-
porphyrin.10 Interestingly, tRCM stabilization of the prism
results in further Q-band red-shifting and sharpening, implying
that the component panels are further planarized and rigidified.
An important consequence of covalently stabilizing cyclic
assemblies should be stronger binding of template ligands or
other guest molecules, relative to binding by non-permanent
structures. While the enthalpy of guest binding should not
change (since Zn–N bond strengths are unaffected), pre-organi-
zation of the host structure via tRCM should considerably lower
olefinic
protons
corresponding
to
olefin-metathesized
(Zn2–PP)3(TPEB)2 appear as broad singlets at 5.80 and
5.70 ppm, indicative of a mixture of E and Z conformers.
The MALDI-TOF mass spectra (dithranol matrix) of both
(Zn2–AA)3(Py3T)2 and (Zn2–PP)3(TPEB)2 exhibited only molecular
ion peaks for the zincated prisms, presumably because of removal
of the templates during ionization (ESIw, Figs. S5 and S8).7b,c
In the gel-permeation chromatograms for purified panels
and prisms (ESIw, Fig. S4)—where retention times scale
inversely with molecule or assembly size—the prisms appear
well before the panels (Zn2–AA and Zn2–PP), with the large-
templated prism (Zn2–PP)3(TPEB)2 eluting first. Comparisons
to a previously reported GPC-based size-calibration curve for
a family of molecular squares11 show that the prisms elute in
essentially the same time as the largest square examined there:
an M4Por4 assembly comprising of Re(CO)3Cl corners and
dipyridyl-porphyrin edges. While direct comparisons are chal-
lenging because of the different assembly shapes, the M4Por4
square and the current prisms all feature similar cross-sec-
Fig. 1 Left: Electronic absorption spectra of Zn2–AA, 3Zn2–AAꢁ2Py3T (3 : 2 mixture) and covalently linked (Zn2–AA)3(Py3T)2. Right:
Electronic absorption spectra of covalently linked (Zn2–AA)3(Py3T)2, free-base (AA)3, and (Zn2–AA)3. All were examined in CH2Cl2 as solvent,
except (AA)3, which was examined in THF.
ꢀc
This journal is The Royal Society of Chemistry 2008
3376 | Chem. Commun., 2008, 3375–3377

View Article Online
the entropy of binding. Enhanced binding could be valuable, for
example, for improving retention of encapsulated catalysts or
for extending host–guest chemistry into higher-polarity solvents.
To test the notion of enhanced binding, we compared the
abilities of labile and permanent prisms to retain Py3T when
challenged with THF, a competitive ligand for Zn(II). Displace-
ment was monitored by taking advantage of the ability of Py3T,
but not THF, to attenuate the fluorescence of Zn2–AA when
excited at 712 nm (labile assembly) or 723 nm (covalent assem-
bly). For the labile prism, 50% recovery of the full fluorescence
emission intensity of Zn2–AA was observed at [THF] ¼ 0.018
M. For the covalently stabilized prism, 50% recovery was
observed at [THF] ¼ 8.3 M, i.e., 460ꢂ higher.
Notes and references
z General tRCM procedure: Under nitrogen, the dimer panel and template
(2 : 3 molar equivalents, B0.3 mM in dimer) were dissolved together in
CH2Cl2 and allowed to stir for 30 min. A solution of Grubbs’ 1st-generation
catalyst (25 mol%) in CH2Cl2 was then added to this reaction mixture. After
stirring at room temperature overnight, the reaction mixture was opened to
air and acetone (10 mL) was added. The volatiles were removed from the
reaction mixture under reduced pressure using a rotary evaporator. The
remaining residue was extracted with CH2Cl2 and subjected to size-exclusion
chromatography (Bio-Rad Bio-Beads S-X1, CH2Cl2) to afford the tem-
plated, covalently linked trigonal prisms as a dark solid (470%).
1. For reviews see: (a) A. K. Burrell, D. L. Officer, P. G. Plieger and D.
C. W. Rei, Chem. Rev., 2001, 101, 2751–2796; (b) M.-S. Choi,
T. Yamazaki, I. Yamazaki and T. Aida, Angew. Chem., Int. Ed.,
2004, 43, 150–158; (c) B. C. Feiters, A. E. Rowan and R. J. Nolte,
Chem. Soc. Rev., 2000, 29, 375–384; (d) M. Nakash and J. K.
M. Sanders, J. Org. Chem., 2000, 65, 7266–7271; (e) J. T. Hupp,
Struct. Bonding, 2006, 121, 145–165; (f) Y. Kobuke, Struct. Bonding,
2006, 121, 49–104; (g) L. Flamigni, V. Heitz and J.-P. Sauvage,
Struct. Bonding, 2006, 121, 217–261; (h) A. Prodi, M. T. Indelli,
C. J. Kleverlaan, E. Alessio and F. Scandola, Coord. Chem. Rev.,
2002, 229, 51–58; (i) J. C. Garno, C. Xu, G. Bazzan, J. D. Batteas
and C. M. Drain, ACS Symp. Ser., 2006, 928, 128–183; (j) L. Badini
and C. M. Hunter, Adv. Inorg. Chem., 2002, 53, 213–259;
(k) S. J. Lee and J. T. Hupp, Coord. Chem. Rev., 2006, 250,
1710–1750; (l) H. L. Anderson, Chem. Commun., 1999, 2323–2330.
2. A related motif, that has proven especially useful for applications in
selective catalysis, comprises reactive sites coordinatively enshrouded with
unlinked monomeric metalloporphyrins. See, for example: (a) V. F Slagt,
J. N. H. Reek, P. C. J. Kamer and P. W. N. M. van Leeuwen, Angew.
Chem., Int. Ed., 2001, 40, 4271–4274; (b) V. F. Slagt, P. C. J. Kamer, P.
W. N. M. van Leeuwen and J. N. H. Reek, J. Am. Chem. Soc., 2004, 126,
1526–1536; (c) M. Kuil, T. Soltner, P. W. N. M. van Leeuwen and J. N.
H. Reek, J. Am. Chem. Soc., 2006, 128, 11344–11345.
3. (a) R. V. Slone and J. T. Hupp, Inorg. Chem., 1997, 36, 5422–5423;
(b) M. L. Merlau, M. d. P. Mejia, S. T. Nguyen and J. T. Hupp, Angew.
Chem., Int. Ed., 2001, 40, 4239–4242; (c) G. A. Mines, B.-C. Tzeng,
K. J. Stevenson, J. I. Li and J. T. Hupp, Angew. Chem., Int. Ed., 2002, 41,
154–157; (d) A. M. Massari, R. W. Gurney, C. P. Schwartz, S. T. Nguyen
and J. T. Hupp, Langmuir, 2004, 20, 4422–4229; (e) M. H. Keefe,
J. L. O’Donnell, R. C. Bailey, S. T. Nguyen and J. T. Hupp, Adv. Mater.,
2003, 15, 1936–1939; (f) K. E. Splan, A. M. Massari and J. T. Hupp,
J. Phys. Chem. B, 2004, 108, 4111–4115; (g) S. H. Chang, K.-B. Chung,
R. V. Slone and J. T. Hupp, Synth. Met., 2001, 117, 215–217;
(h) K. F. Czaplewski, R. Q. Snurr and J. T. Hupp, Adv. Mater., 2001,
13, 1895–1897; (i) M. E. Williams and J. T. Hupp, J. Phys. Chem. B,
2001, 105, 8944–8950; (j) J. Libera, R. Gurney, C. Schwartz, H. Jin,
T.-L. Lee, S. T. Nguyen, J. T. Hupp and M. Bedzyk, J. Phys. Chem. B,
2005, 109, 1441–1450; (k) A. B. F. Martinson, A. M. Massari, S. J. Lee,
R. W. Gurney, K. E. Splan, J. T. Hupp and S. T. Nguyen,
J. Electrochem. Soc., 2006, 153, A527–A532.
4. (a) P. J. Stang, J. Fan and B. Olenyuk, Chem. Commun., 1997,
1453–1454; (b) L. Miljacic, L. Sarkisov, D. E. Ellis and
R. Q. Snurr, J. Chem. Phys., 2004, 121, 7228–7236.
5. S. J. Lee, K. L. Mulfort, J. L. O’Donnell, X. Zuo, A. J. Goshe,
P. J. Wesson, S. T. Nguyen, J. T. Hupp and D. M. Tiede, Chem.
Commun., 2006, 4581–4583.
6. (a) P. C. M. Gerven, J. A. A. W. Elemans, J. W. Gerritsen, S. Speller,
R. J. W. Nolte and A. E. Rowan, Chem. Commun., 2005, 3535–3537;
(b) J. M. Bakker, S. J. Langford, M. J. Latter, K. A. Lee and
C. P. Woodward, Aust. J. Chem., 2005, 58, 757–761; (c) T. Ishida,
Y. Morosaki and Y. Chujo, Tetrahedron Lett., 2006, 47, 5265–5268.
7. (a) J. Webb and J. K. M. Sanders, Inorg. Chem., 2000, 39, 5912–5920
and J. Webb and J. K. M. Sanders, Inorg. Chem., 2000, 39, 5920–5929;
(b) A. L. Kieran, A. D. Bond, A. M. Belenguer and J. K. M. Sanders,
Chem. Commun., 2003, 2674–2675; (c) A. L. Kieran, S. I. Pascu,
T. Jarrosson and J. K. M. Sanders, Chem. Commun., 2005, 1276–1278.
8. R. H. Grubbs, S. J. Miller and G. C. Fu, Acc. Chem. Res., 1995,
28, 446–452.
An additional dividend of covalent stabilization of the prism
assemblies should be the ability to remove and/or replace
otherwise structurally-crucial Zn(II) sites with other metals of
interest. We find that the free-base trigonal prism, (AA)3, can be
readily obtained via TFA-demetallation of (Zn2–AA)3(Py3T)2 in
CH2Cl2 and then purified by size-exclusion chromatography.
Interestingly, the analytical GPC peak of purified (AA)3 appears
at about 6.1 min, 0.2 min later than that for (Zn2–AA)3(Py3T)2
(ESIw, Fig. S4), suggesting that the hollow prism is able to
distort and sample slightly smaller gel pores than is the more
1
rigid templated structure. Demetallation was confirmed by H
NMR spectroscopy (free NH protons at ꢃ2.07 ppm; absence of
pyridyl (template) protons) and by MALDI-TOF MS (absence
of the molecular ion peak for (Zn2–AA)3(Py3T)2 and the pre-
sence of a strong signal for (AA)3 (ESIw, Fig. S5)). Fig. 1 (right
panel) shows that demetallation introduces additional Q bands,
consistent with reductions in porphyrin symmetry upon replace-
ment of Zn(II) with protons. Significantly, there were very little
changes in the location and breadth of the lowest energy Q
band, implying that the porphyrin subunits within each dimeric
panel remain coplanar and rigid following prism demetallation.
Finally, as detailed in the ESIw, the free-base prism (AA)3 can be
easily and completely remetallated with Zn21 or Co21
.
In conclusion, we have demonstrated that highly conjugated
porphyin dimer panels can be template-assembled into trigonal
prisms in solution and permanently ‘‘set’’ into nanoporous
assemblies using olefin metathesis. Depending on the lengths of
the starting vinyl arms, the assembled structures can be template-
adjusted to yield multiporphyrin cavities with various well-
defined sizes. Once the structure is ‘‘set’’ by olefin metathesis,
the template can be readily removed via treatment with Lewis
basic ligands, without detriment to the rigid trigonal prism shape.
The permanent prisms display much more tenacious guest
binding than do their labile counterparts. For the permanent
structures, the porphyrin units can be readily demetallated to
give the corresponding free-base trigonal prisms, which can be
readily remetallated with other functional metals. This strategy
lends itself to the preparation of a variety of porphyrinic trigonal
prisms possessing tunable metal environments that can be used
for molecular recognition, catalysis, or photonic energy transfer.
These studies will be reported in due course.
9. E. B. Fleischer and A. M. Shachter, Inorg. Chem., 1991, 30, 3763–3769.
10. (a) H. L. Anderson, Inorg. Chem., 1994, 33, 972–981; (b) P. N. Taylor,
J. Huuskonen, G. Rumbles, R. T. Aplin, E. Williams and
H. L. Anderson, Chem. Commun., 1998, 909–910; (c) P. N. Taylor
and H. L. Anderson, J. Am. Chem. Soc., 1999, 121, 11538–11545.
11. C. R. Graves, M. L. Merlau, G. A. Morris, S.-S. Sun,
S. T. Nguyen and J. T. Hupp, Inorg. Chem., 2004, 43, 2013–2017.
We thank Rebecca Jensen for carrying out the trigonal ligand
displacement studies. We gratefully acknowledge the U.S. Na-
tional Science Foundation, AFOSR, DTRA/ARO, and the
Korea Research Foundation (award #KRF-214-C00107, post-
doctoral fellowship for K.-T.Y.) for support of our research.
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 3375–3377 | 3377