The role of chirality in directing the formation of cup-shaped porphyrins and the coordination characteristics of such hosts

Article abstract of DOI:10.1002/chem.201100755

Cup-shaped porphyrin 1a has four norbornane rings for encircling space and this type of host could be of interest in supramolecular and catalytic chemistry. We used ¹H NMR spectroscopy to investigate the acid-catalyzed (pTsOH in CHCl₃ and TFA in CH₃CN) condensation of racemic, enantioenriched (80-85% enantiomeric excess (ee)), and enantiopure (99% ee) pyrromethanecarbinol 7 into 1a. We found that the oligomerization of racemic 7_rac would give 1a-d in the ratio different from the statistical one, though a minuscule quantity of 1a (<5%) formed. The oligomerization of enantioenriched 7 (80-85% ee), however, led to the formation of greater amounts of 1a (31-47%) along with other stereoisomers 1b-d. Importantly, pTsOH catalyzed the conversion of enantiopure 7 (99% ee) into 1a (>95% diastereomeric excess (de), 25% overall yield) in CHCl₃ although prolonged reaction times or greater concentration of the catalytic acid gave rise to 1b-d at the expense of 1a. The metallation of 1a with Zn(OAc) ₂ led to the formation of Zn^II-1a and we used computational (DFT: RI-BP86/SV(P),TZVP) and experimental (¹H NMR spectroscopy) methods to study the partitioning of smaller N-methylimidazole 13 (94 A³) and bigger 1,5-dicyclohexylimidazole 14 (268 A³) between the inner and outer side of the host. We found that bigger 14 was mostly encapsulated (90%) inside Zn^II-1a at 298.0 K, whereas smaller 13 would equally partition between the two sides of the host. Furthermore, the out/in equilibrium was, in the case of 14-Zn^II-1a, favored by entropy (TΔS°_out/in=3.5±0.1 kcalmol ^-1) indicating that perhaps differential solvation of the coordinated ligand assisted the encapsulation.

Full text of DOI:10.1002/chem.201100755

DOI: 10.1002/chem.201100755
The Role of Chirality in Directing the Formation of Cup-Shaped Porphyrins
and the Coordination Characteristics of such Hosts
Bao-Yu Wang, Daniel A. Turner, Teodora Zujovic´, Christopher M. Hadad, and
Jovica D. Badjic´*[a]
Abstract: Cup-shaped porphyrin 1a
has four norbornane rings for encir-
cling space and this type of host could
be of interest in supramolecular and
catalytic chemistry. We used 1H NMR
spectroscopy to investigate the acid-
catalyzed (pTsOH in CHCl3 and TFA
in CH3CN) condensation of racemic,
enantioenriched (80–85% enantiomer-
ic excess (ee)), and enantiopure
(99% ee) pyrromethanecarbinol 7 into
1a. We found that the oligomerization
of racemic 7rac would give 1a–d in the
ratio different from the statistical one,
though a minuscule quantity of 1a
amounts of 1a (31–47%) along with
other stereoisomers 1b–d. Importantly,
pTsOH catalyzed the conversion of
SV(P),TZVP)
and
experimental
(1H NMR spectroscopy) methods to
study the partitioning of smaller N-
methylimidazole 13 (94 ꢀ3) and bigger
1,5-dicyclohexylimidazole 14 (268 ꢀ3)
between the inner and outer side of the
host. We found that bigger 14 was
mostly encapsulated (90%) inside ZnII-
1a at 298.0 K, whereas smaller 13
would equally partition between the
two sides of the host. Furthermore, the
out/in equilibrium was, in the case of
enantiopure
7
(99% ee) into 1a
(>95% diastereomeric excess (de),
25% overall yield) in CHCl3 although
prolonged reaction times or greater
concentration of the catalytic acid gave
rise to 1b–d at the expense of 1a.
The metallation of 1a with ZnACTHNURTGNENUG(OAc)2
led to the formation of ZnII-1a and we
used computational (DFT: RI-BP86/
14-ZnII-1a,
favored
by
entropy
indi-
(TDS8out/in =3.5Æ0.1 kcalmolÀ1)
Keywords: asymmetric synthesis ·
coordination chemistry · host–guest
systems · molecular recognition ·
porphyrin
cating that perhaps differential solva-
tion of the coordinated ligand assisted
the encapsulation.
ACHTUNGTRENNUNG(<5%) formed. The oligomerization of
enantioenriched 7 (80–85% ee), how-
ever, led to the formation of greater
Introduction
literature, however, it appeared that diastereoselective syn-
thesis of porphyrins from pyrroles is challenging, necessitat-
ing lengthy optimizations.[19] In particular, Lindsey and co-
workers have shown that the position of meso substituents
(A and B in Figure 1B) could, in desired porphyrin, become
easily permuted during its formation through acidolysis of
pyrrole-based precursors (Figure 1B).[30–37] This acid-cata-
lyzed fragmentation of acyclic oligomers is therefore an un-
desired mechanistic feature in any rational synthesis of por-
phyrins with different substituents. For this reason, various
Brønsted (trifluoroacetic acid (TFA) in CH3CN)[33] and
We have designed gated molecular baskets to investigate
quantitative relationships that characterize the kinetics and
thermodynamics of molecular encapsulation.[1–6] An impor-
tant objective of such studies has been centred on under-
standing the role that gating[7,8] could play in controlling the
outcome of chemical reactions. Accordingly, we have sought
to develop a family of gated synthetic hosts with a catalytic
center embedded in the interior for promoting chemical
transformations.[9–17] It occurred to us that a stereoselective
installation of norbornane “walls” around a porphyrin[18–27]
“floor” (Figure 1A) could allow the creation of this type of
supramolecular catalyst. The hypothesis was that head-to-
tail tetramerization of enantiopure, but not racemic, pyrro-
methanecarbinol 7 could predominantly give the cup-shaped
compound 1a having four bicyclic rings[28,29] pointing in the
same direction in space. On the basis of our survey of the
Lewis acids (InCl3, YbCAHTUNGRTNE(NNUG OTf)3, ScAHCTUNRTGE(NGNUN OTf)3, and DyACHTNUGRTEN(NUGN OTf)3 in
CH2Cl2)[35] were optimized to kinetically direct the conden-
sation of dipyrromethanecarbinols toward various meso-sub-
stituted porphyrins. In accord with such reports, we set to in-
vestigate the self-condensation of enantiopure pyrrometha-
necarbinol 7 in the presence of two Brønsted acids (TFA
and toluene sulfonic acid (pTsOH)); note a proposed elec-
tron-pushing mechanism in Scheme 1 describing the process.
Indeed, these Brønsted acids were already used in the syn-
thesis of porphyrins,[29,33,38] but only TFA (in CH3CN)[33] was
demonstrated to inhibit the scrambling of dipyrromethane-
carbinols. Accordingly, the head-to-tail condensation of
enantiopure 7 should give desired 1a (Scheme 1), whereas
the acidolysis of intermediate pyrrole-based oligomers or
the porphyrinogen could lead to the formation of diastereo-
´
[a] B.-Y. Wang, D. A. Turner, T. Zujovic, Prof. C. M. Hadad,
´
Prof. J. D. Badjic
Department of Chemistry, The Ohio State University
100W 18thAvenue, Columbus OH (USA)
Fax : (+1)614-292-1685
E-mail: badjic@chemistry.ohio-state.edu
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201100755.
8870
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8870 – 8881

FULL PAPER
Results and Discussion
Working hypotheses: Tradition-
ally, porphyrins are made by
condensation of pyrroles with
aldehydes in the presence of
acids.[19] The transformation
may necessitate a formal oxida-
tion of the intermediate por-
phyrinogen with 2,3-dichloro-
5,6-dicyanobenzoquinone
(DDQ) to complete the forma-
tion of the porphyrin. Lindsey
and co-workers have investigat-
ed alternative and milder ways
for the preparation of porphyr-
ins demonstrating that acid-cat-
alyzed tetramerization of more
reactive dipyrromethanecarbi-
nols can be precisely controlled
to give targeted porphyrins with
meso substituents in a desired
order.[30–37] Inspired by these
studies, we aimed to examine
the diastereoselectivity of the
macrocyclization of pyrrome-
thanecarbinols (racemic, enan-
tioenriched, and enantiopure)
of type 7 (Scheme 1). The hy-
pothesis was that enantiopure 7
should, in the presence of weak
acids,[35] undergo head-to-tail
tetramerization, so that upon
the oxidation of the intermedi-
ate porphyrinogen it gives 1a as
the sole kinetic product (Fig-
ure 1A). In accord with a plau-
sible electron-pushing mecha-
nism (Scheme 1), our objective
was to find reaction conditions
that would minimize the rever-
sibility of the transformation
and suppress the acidolysis: the
Figure 1. (A) Energy-minimized structures (MMFFs, Spartan) of four diastereomeric porphyrins 1a–d. The
condensation of enantiopure pyrromethanecarbinol 7 (but not racemic 7rac) has been hypothesized to give 1a
as the major product. (B) The acidolysis[33] of pyrrole-based oligomers leads to the scrambling of meso sub-
stituents A and B in the product.
acidolysis
would
allegedly
meric 1b–d. Our study reveals that the acidolysis indeed
occurs but is controllable and is a function of the concentra-
tion of the catalytic acid and the reaction time. The diaste-
reoselectivity of the key tetramerization of enantiopure pyr-
“scramble” the product,[19,32] giving rise to undesired diaste-
reomers 1b–d that are less interesting from the prospect of
host–guest/catalytic chemistry; note that H+ promoted
À
cleavage of C C bonds of oligomers (see dashed gray
romethanecarbinols of type
7
was thus optimized
arrows in Scheme 1) contributes to the scrambling and the
loss of stereochemical information. All things considered,
we decided to vary the reaction time, the nature and the
concentration of the catalytic acid (TFA and pTsOH) as
well as the solvent polarity (CH3CN and CH2Cl2), and
follow the conversion of 7 into 1a–d with quantitative
1H NMR spectroscopy.[39]
ACHTUNGTRENNUNG(>95% de) to even eliminate a need for the chromatograph-
ic separation of the product. We further used computational
and experimental methods to examine at room temperature
the hosting characteristics of zinc(II)-metallated 1a. Inter-
estingly, this cup-shaped host has been shown to accommo-
date bulky 1,5-dicyclohexylimidazole in its cavity, whereas
smaller N-methylimidazole coordinates to both of its faces.
In addition, we postulated that differently sized N-hetero-
cycles would coordinate to the cup-shaped zinc(II)-metallat-
Chem. Eur. J. 2011, 17, 8870 – 8881
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8871

´
J. D. Badjic et al.
Scheme 1. An electron-pushing mechanism for acid-catalyzed oligomerization of enantiopure 7 into porphyrinogen, which upon oxidation gives 1a (see
Figure 1A). The proposed fragmentation–acidolysis (surrounded by gray dashed line) could lead to the loss of stereochemical information and the forma-
tion of 1b–d (see Figure 1).
ed 1a host (see later, Figures 8 and 9).[40–44] The coordination
of guests to ZnII-1a could occur at two of its faces and we
wanted to investigate the selectivity of such axial recogni-
tion.[45–47]
For diastereomeric 1a–d, the meso hydrogens Ha gave
single H NMR resonances (d=10.20–10.26 ppm, Figure 2B
1
and C) corresponding to the symmetric characteristics of
each compound; however, it required additional effort to
complete a full assignment of these signals (Figure 2B and
C). First, we used thin-layer chromatography (SiO2, hexane/
ethyl acetate=15:1) to isolate the two least-polar stereoiso-
mers of 1a–d from the TFA-catalyzed reaction (Rf =0.25
and 0.19, Figure 2A). The 1H NMR spectrum of the less
polar compound (Rf =0.25) was consistent with either 1a or
1d, showing a single resonance for Ha protons (see the Sup-
porting Information, Figure S1).[48] To determine the struc-
ture of this molecule, we first completed its metallation with
The condensation of racemic pyrromethanecarbinol 7rac
:
The synthesis of 7rac (rac=racemic) is shown in Fig-
ure 2A.[48] Benzonorbornadiene 2 was made on multigram
scale in the cycloaddition of cyclopentadiene and 1,2-dibro-
mobenzene. The stereospecific (anti) addition of benzenesul-
fenyl chloride (PhSCl) to 2 gave exo-(phenylthio) compound
[49]
3rac
,
which upon elimination (tBuOK) and then oxidation
of 4rac with MCPBA afforded a,b-unsaturated sulfone 5rac
.
The treatment of 5rac with ethyl isocyanoacetate, in the pres-
ence of tBuOK (the Barton–Zard reaction), led to the for-
mation of pyrrole derivative 6rac. The reduction of 6rac with
LiAlH4 gave racemic pyrromethanecarbinol 7rac that was
subjected to self-condensation with Brønsted acids by the
following conditions: a) TFA (30 mm) in CH3CN and
b) pTsOH (0.39 mm) in CHCl3. The oxidation of the result-
ing porphyrinogen with DDQ gave a mixture of porphyrins
as suggested by TLC chromatography (Figure 2A). In addi-
tion, the 1H NMR spectrum of the product mixture (Fig-
ure 2B and C) also indicated that both acids promoted the
formation of 1a–d.
ZnACTHNGUTERNNUG
(OAc)2.[29] After, we titrated N-methylimidazole to ZnII–
porphyrin (1.0 mm, CDCl3) and followed the process with
1H NMR spectroscopy (Figure 3A). Apparently, the coordi-
nation of the ligand was thermodynamically favored, but
1
fast on the H NMR time scale at 298.0 K. Variable temper-
ature 1H NMR spectroscopic measurements (Figure 3B),
however, revealed a reduction in the symmetry of this com-
1
plex at lower temperatures: each H NMR signal would split
into two having equal intensity. Since a slow exchange of
N-methylimidazole to and from the host would only desym-
metrize ZnII-1d but not ZnII-1a (Figure 3A), we concluded
that the isolated compound must be 1d; note that these ex-
8872
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8870 – 8881

Role of Chirality in Directing the Formation of Cup-Shaped Porphyrins
FULL PAPER
phyrin (Rf =0.19) was consis-
tent with either 1b or 1c (see
the Supporting Information,
Figure S2).[48] Since 1c has an
equal number of norbornane
“side” arms pointing up and
down, it should be less polar
than 1b with a greater retention
time so that the isolated com-
pound (Rf =0.19) was designat-
ed as 1c. Furthermore, the de-
1
convolution of the H NMR sig-
nals (MNova software, Fig-
ure 2B and C) corresponding to
Ha revealed the proportion of
diastereomeric porphyrins 1a–d
in the mixture.
On purely statistical grounds,
one would expect for the self-
condensation of 7rac to give 1a/
1b/1c/1d in the ratio of 1:4:2:1
(%=12.5:50:25:12.5). The ex-
perimental distribution of dia-
stereomeric porphyrins was dif-
ferent (Figure 2B and C),
though only the reaction em-
ploying pTsOH led to the for-
mation of miniscule amounts of
desired 1a (5.5%). The forma-
tion of rather larger proportions
of 1c/d could be explained by
postulating that a smaller acti-
vation energy is associated with
the appearance of these per-
haps more stable diastereomers.
In fact, we note that the yield
of stereoisomeric 1a–d increas-
es as the number of adjacent
norbornanes pointing to the
same
direction
decreases:
1a(4)/1b(3)/1c(2)/1d(0)=
0:15:60:25% (Figure 2B).
The condensation of enantioen-
riched pyrromethanecarbinol 7:
Figure 2. (A) The synthesis of racemic 7rac, and its condensation/oxidation into 1a–d. MCPBA=meta-chloro-
perbenzoic acid, tBuOK=potassium tert-butoxide. TLC retention times of 1b–d (SiO2, hexane/ethyl acetate=
15:1) (B) Partial H NMR spectrum (400 MHz, CDCl3, black) of the product mixture obtained 20 min after in-
To obtain enantioenriched
7
(Figure 4A), we used the meth-
odology for enantiotopic dis-
crimination of sp2-hybridized
carbon atoms in meso-norbor-
nene compounds.[50,51] In partic-
ular, it has been shown that bis-
1
itiation of the self-condensation of 7rac (5.0 mm) with trifluoroacetic acid (TFA, 30.0 mm) in CH3CN and subse-
1
quent oxidation with 2,3-dichloro-5,6-dicyanobenzoquinone (DDQ). (C) Partial H NMR spectrum (400 MHz,
CDCl3, black) of the product mixture obtained 120 min after initiation of the self-condensation of 7rac (5.0 mm)
with para-trifluoroacetic acid (pTsOH, 0.39 mm) in CHCl3 and subsequent oxidation with DDQ; top gray line
(composite) and lower gray line (resolved) 1H NMR signals were obtained from deconvolution of the experi-
mental spectra (black) with Mnova software.
AHCTUNGTRENGN(UN phenyl)sulfonylalkenes of type
9 could be treated with a chiral
diol to give one diastereomeric
periments also allowed us to distinguish the Ha singlet be-
acetal in excess. Accordingly, we attempted the desymmetri-
longing to 1a. The 1H NMR spectrum of the more polar por-
zation of meso-9 with (R)-1,1’-bi-2-naphthol (BINOL).[51]
Chem. Eur. J. 2011, 17, 8870 – 8881
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8873

´
J. D. Badjic et al.
ure S6).[48] Evidently, the stereo-
selectivity of the 9-to-10 step
was not ideal but screening
other chiral alcohols could, per-
haps, help improve it in future.
As expected, the Barton–Zard
conversion of 5 to 6 did not dis-
rupt the actual proportion of
enantiomers (80-85% ee, see
the Supporting Information,
Figure S8).[48]
Enantioenriched 7 contained
at least 90% of one enantio-
mer, and its self-condensation
was examined with Brønsted
acids by using the following
conditions: a) 30 mm TFA in
CH3CN
(Figure 4B)
and
b) 0.39 mm of pTsOH in CHCl3
(Figure 4C). Interestingly, the
1H NMR spectrum of the prod-
uct mixture (after its treatment
with DDQ) would in each case
reveal a greater proportion of
desired
1a
formed
after
120 min (Figure 4B and C).
Concomitantly, we followed the
progress of the TFA reaction
with time (Figure 4C)[48] to find
that the concentration of 1a
was the highest (ꢀ45%) at the
early stages of the reaction, for
example, 3 min. A decrease in
the concentration of 1a and the
accumulation of 1c with time
was also evident (Figure 4C).
Apparently, more dramatic
changes were occurring within
the first three minutes of the re-
action, although these were dif-
ficult to follow, let alone con-
trol. To account for the results,
however, we suggest that the
acidolysis of acyclic intermedi-
Figure 3. (A) Partial 1H NMR spectra (400 MHz) of ZnII-1d (1.0 mm) in CDCl3 showing the perturbation of
Ha resonance with incremental addition of a) 0, b) 0.1, c) 0.2, d) 0.3, e) 0.4, f) 0.6, g) 0.8, h) 1.0, i) 1.2, and j) 1.4
molar equiv of neat N-methylimidazole. (B) Variable temperature 1H NMR spectra (400 MHz) of ZnII-1d
(1.0 mm) in CD2Cl2 showing decoalescence of Hc/d/e signals.
The diastereopure product was indeed isolated, but all at-
tempts for removing (by hydrolyzing) the chiral auxiliary
failed. The reaction of bis(phenylsulfonyl)benzonorborna-
diene 9 with an equimolar amount of sodium salt of (R,R)-
(+)-methoxy-1,2-diphenylethanol, however, gave enol ether
10, which was easily hydrolyzed and reduced into 11.[50]
Upon the mesylation (MsCl) of 11 and then elimination of
the mesyl group from 12, we were able to isolate compound
ates was taking place (Scheme 1) contributing to the forma-
tion of 1b–d.[35] At last, the mixture of diastereomeric por-
phyrins was practically inseparable by column chromatogra-
phy (SiO2) and we decided to prepare enantiopure 7 and ex-
amine its condensation.
The condensation of enantiopure pyrromethanecarbinol 7:
We recrystallized enantioenriched 6 from hexane/ethyl ace-
tate to obtain homochiral crystals of the same compound in
overall 40% yield (>99% ee, see the Supporting Informa-
tion, Figure S9).[48] After the reduction of 6 with LiAlH4
(Figure 4A), we completed the condensation of enantiopure
7 (5.0 mm) in CHCl3 using 0.39 mm pTsOH (Figure 5). Evi-
1
5. Importantly, the H NMR spectrum of this compound was
equivalent to that of previously synthesized 5rac corroborat-
ing its structure.[48] With the assistance of chiral HPLC, we
further found that the enantiomeric excess for 5 was in the
range of 80 to 85% (see the Supporting Information, Fig-
8874
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8870 – 8881

Role of Chirality in Directing the Formation of Cup-Shaped Porphyrins
FULL PAPER
Figure 4. (A) The synthesis of enantioenriched (80–85% ee) and enantiopure 7 (99% ee), and their condensation/oxidation into 1a–d. DMAP=4-dime-
thylaminopyridine, TEA=triethanolamine, MsCl=methanesulfonyl chloride, DBU=1,8-diazabicyclo
[5.4.0]undec-7-ene. (B) Partial 1H NMR spectrum
AHCTUNGTRENNUNG
(400 MHz, CDCl3, black) of the product mixture obtained 120 min after initiation of the self-condensation of enantioenriched 7 (5.0 mm) with TFA
(30.0 mm) in CH3CN and subsequent oxidation with DDQ. (C) Partial 1H NMR spectrum (400 MHz, CDCl3, black) of the product mixture obtained
120 min after initiation of the self-condensation of enantioenriched 7 (5.0 mm) with pTsOH (0.39 mm) in CHCl3 and subsequent oxidation with DDQ
(top gray line (composite) and lower gray line (resolved)). 1H NMR signals were obtained from deconvolution of the experimental spectra (black) with
Mnova software. The change in the concentration of 1a and 1c with time during the self-condensation of enantioenriched 7 (80–85% ee, 5.0 mm) with
TFA (30.0 mm) in CH3CN.
dently, compound 1a formed as a predominant condensation
product within the first 10 min of the reaction (>95% de,
Figure 5A).[48] Diastereomers 1b–d also appeared with time
(Figure 5B and C) although at a sufficiently slow rate (Fig-
ure 6B) that we could “safely” stop the reaction and isolate
the desired 1a. In addition, the overall yield of 1a–d stayed
within 25–30% over 5 h (Figure 6D). The fast, and at the
same time, predominant formation of 1a is in line with the
absence of the acidolysis (Scheme 1) at early stages of the
reaction (Figure 6A and B). The acidolysis, however, oc-
curred at a much greater rate when the concentration of the
acidic catalyst pTsOH was increased to 0.79 mm (Figure 6A
and C).[48] That is, the rapid scrambling led to the formation
of 1b–d (Figure 6C) at the expense of 1a (Figure 6A). Evi-
dently, it is critical to use lower concentrations of pTsOH
and short reaction times for the diastereoselective synthesis
of porphyrins of type 1a (Scheme 1). Given the favorable
outcome of the study, we did not investigate the condensa-
tion of enantiopure 7 with TFA, although this and other
Brønsted acids could as well be effective catalysts.
The dynamic and electronic characteristics of cup-shaped
1
1a: We used H–1H NOESY correlation spectroscopy (see
the Supporting Information, Figure S34)[48] for fully assign-
1
ing the H NMR resonances of 1a (Figure 7A). Interesting-
ly, 1H NMR signals (400 MHz) corresponding to C4v sym-
metric 1a (CDCl3) underwent decoalescence at low temper-
atures, disclosing a change in the symmetry of the com-
pound to C2 (see the Supporting Information, Figure S32).[48]
We reasoned that a double proton transfer between two tau-
tomeric forms of 1a contributed to the process (Fig-
ure 7A).[52,53] It has already been suggested that the mecha-
nism of such proton shuttling would involve a stepwise
motion of the hydrogen atoms with the formation of less
stable “cis” intermediates. The corresponding rate coeffi-
cients (k, sÀ1) for the intramolecular shuttling were mea-
Chem. Eur. J. 2011, 17, 8870 – 8881
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8875

´
J. D. Badjic et al.
between two aromatic carbons
(Figure 7B) and allowing opti-
mization of all other geometric
parameters.[48] The computed
energy profile for such opening/
closing of the cup-shaped por-
phyrin was asymmetric: splay-
ing the norbornane groups
apart is clearly “more expen-
sive” than pushing them close
together (Figure 7B).
This dynamic property of 1a
shall be considered in studies of
molecular recognition when
placing a sizeable guest inside
of 1a.
Host–guest characteristics of
cup-shaped ZnII-1a: Compound
1a comprises an electron-rich
cavity that is open for properly
sized guests to enter/depart
(Figure 8A). The binding affini-
ty (DG8) as well as the constric-
[54,55]
tive binding (DG°
)
in/out
would likely be small for a
range of guests given that weak
van der Waals forces or the sol-
vophobic effect could drive the
association and there is no
“physical barrier” for such a
process. To investigate the host-
ing characteristics of 1a we
metallated this compound with
1
Figure 5. Partial H NMR spectrum (400 MHz, CDCl3, black) of the product mixture obtained (A) 10, (B) 120,
and (C) 2400 min after initiation of the self-condensation of enantiopure 7 (99% ee, 5.0 mm) with pTsOH
(0.39 mm) in CHCl3 and subsequent oxidation with DDQ. For (A) experimental spectra (top black line), gray
line (composite)); for (B) experimental spectra (top black line), top gray line (composite), and lower gray line
(resolved). 1H NMR signals were obtained from deconvolution of the experimental spectra (top black line)
with Mnova software.
ZnACTHNUTRGNEUNG
(OAc)2 to form ZnII-1a (Fig-
ure 8A).[48] Zinc(II)-porphyrin
receptors are known to bind to
N-heterocyclic guests[43] and in
sured, and interestingly the data revealed non-Arrhenius be-
havior (over a greater temperature range) so that tunneling
was suggested to play an important role in the mecha-
nism.[53] Nevertheless, we completed quantitative line-shape
analysis (DNMR5) of our variable temperature 1H NMR
data by fitting the line shapes of He resonance (Fig-
ure 7A).[48] Interestingly, the activation energy (Ea =7.5Æ
0.1 kcalmolÀ1) for the shuttling of NHb protons was, from
the Arrhenius plot (Figure 7A), found to be somewhat
lower than reported for variously substituted porphyrins
(Ea =9.4–10.6 kcalmolÀ1).[52] The activation free energy
(DG° =12.5Æ0.1 kcalmolÀ1 at 298 K) of the proton shut-
tling was, however, within the 12.0–12.5 kcalmolÀ1 range re-
ported for different porphyrins.[52]
this study we examined the affinity of ZnII-1a toward small-
er N-methylimidazole 13 (94 ꢀ3, Figure 8B) and bigger 1,5-
dicyclohexylimidazole 14 (268 ꢀ3, Figure 8B). In particular,
we were curious about the regioselectivity of the axial coor-
dination: are compounds 13 and 14 capable of distinguishing
the inner from the outer side (in/out, Figure 8A) of cup-
shaped ZnII-1a? Furthermore, the results from the binding
studies of simple ligands will be useful for future investiga-
tions of this type of porphyrin hosts.
First, we used density functional theory (DFT: RI-BP86/
SV(P),TZVP) to compute the free energy difference
(DGout/in at 298.0 K) between ligands 13 and 14 bound to the
outside and inside of ZnII-1a (Table 1, Figure 9A and B).[48]
Interestingly, there was a small enthalpic (DHout/in =À0.74 to
À1.02 kcalmolÀ1) preference for placing both ligands inside
the cup-shaped porphyrin; the entropy, however, was some-
what adverse (TDSout/in =À0.01 to À0.42 kcalmolÀ1) for such
out/in equilibration. Nevertheless, the calculations suggested
that both smaller 13 (94 ꢀ3) and larger 14 ligands (268 ꢀ3)
Furthermore, we used density functional theory to com-
pute the energy profile for synchronized inward/outward
motion of the norbornane walls in ZnII-1a (DFT: RI-BP86/
SV(P),TZVP, Figure 7B). That is to say, we completed a
series of DFT calculations by constraining the distance (d)
8876
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8870 – 8881

Role of Chirality in Directing the Formation of Cup-Shaped Porphyrins
FULL PAPER
Figure 6. (A) The formation of 1a (1H NMR spectroscopy, Mnova software) as a function of time during self-condensation of enantiopure 7 (5.0 mm)
1
~
&
with pTsOH (0.39 ( ) and 0.79 mm ( )) in CHCl3. (B) and (C) The formation of 1b–d ( H NMR spectroscopy, Mnova software) as a function of time
during self-condensation of enantiopure 7 (5.0 mm) with 0.39 (B) and 0.79 mm (C) pTsOH in CHCl3. (D) The overall yield of 1a–d (1H NMR spectrosco-
py, C6H6 was used as an internal standard) as a function of time during self-condensation of enantiopure 7 (5.0 mm) with 0.39 mm pTsOH in CHCl3.
Table 1. Computed (DFT, RI-BP86/SV(P),TZVP) and experimental
thermodynamic parameters for the partitioning of N-heterocycles 13 and
14 between the outer (out) and inner (in) side of ZnII-1a (see Figure 9A
and B).
geometries, the ZnII atom was located below the plane of
the porphyrin. This caused the porphyrin ring to bend in
toward the zinc, and thereby bring the sidewalls to a closer
distance (Figure 7B).
[a]
[a]
Ligand
DH8out/in
TDS8out/in
DG8out/in
DEout/in
To experimentally evaluate the overall affinity of 13 and
14 toward ZnII-1a, we titrated a standard solution of each of
these ligands with host ZnII-1a (CD2Cl2, 298.0 K) and fol-
[kcalmolÀ1
]
[kcalmolÀ1
]
[kcalmolÀ1
]
[kcalmolÀ1
]
13out/in (comp) À1.02
14out/in (comp) À0.74
À0.01
À0.42
À1.01
À0.32
À1.00
À0.93
n/a
1
lowed the process with H NMR spectroscopy (Table 2). No-
13out/in (exptl)
14out/in (exptl)
0.72Æ0.01
2.3Æ0.1
0.76Æ0.02 À0.04Æ0.02
3.5Æ0.1
À1.2Æ0.1
n/a
tably, the incremental addition of 13 and 14 caused an up-
field shift in the proton resonances of the host (see the Sup-
porting Information, Figures S35 and S36).[48] A fast equilib-
rium between the free and ligand-bound ZnII-1a was appar-
ent, and the non-linear curve fitting of the binding
isotherms revealed a somewhat greater affinity of bulkier 14
(Ka >105 mÀ1, Table 2) toward ZnII-1a.[48] In fact, smaller
N-methylimidazole 13 had a propensity for coordinating to
ZnII-1a (Ka =1.9Æ0.3ꢂ104 mÀ1, Table 2) comparable to that
measured for 13 or 14 binding to “flat” ZnII-TTP 15 (Fig-
ure 8B, Ka =2.1–2.4Æ0.2ꢂ104 mÀ1). Note that our fitting
model (see the Supporting Information, Figures S39 and
S40) did not take into account the binding of each ligand to
two sites of ZnII-1a with different affinity. We reason that
this particular approximation contributed to rather poor fit-
ting of the measured chemical shifts (see the Supporting In-
formation, Figure S40). Indeed, proper UV/Vis or fluores-
cence titration experiments may give more accurate data,
however, the approximate values for Ka reported in Table 2
shall suffice our discussion. At this point we deduced that
[a] at T=298.0 K.
should have a somewhat greater preference for coordinating
to the inner side of ZnII-1a (DG8out/in =À0.32 to À1.01 kcal
molÀ1). Importantly, the computational studies indicate that
for the examined host–guest systems in the gas phase, there
should not be any considerable adverse interactions perturb-
ing the in/out equilibrium. This further led us to assume that
solvation could, in an experimental setting, play an impor-
tant role for facilitating the equilibrium (see below).
Moreover, the computations revealed that placing 13 and
14 inside the cavity of ZnII-1a (Figure 9A and B) affected
the norbornane side walls forcing them apart (d=7.4 for 13
and 7.6 ꢀ for 14, Figure 7B); note that without any guest,
d=7.2 ꢀ (Figure 7B). The cavity perhaps “stretches” to re-
lieve the steric strain caused by the interaction of the ligand
with the phenyl groups. When the ligands were placed out-
side the cavity of ZnII-1a, the walls moved in the opposite
direction (d=6.8 ꢀ for both 13 and 14). For each of these
Chem. Eur. J. 2011, 17, 8870 – 8881
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8877

´
J. D. Badjic et al.
Figure 7. (A) 1H NMR (400 MHz, 298.0 K) spectrum of 1a in CDCl3 and the Arrhenius plot corresponding to double-proton shuttling between two C2
symmetric forms of 1a. (B) Energy profile for the “folding” of ZnII-1a was computed using density functional theory (DFT: RI-BP86/SV(P),TZVP).[48]
Table 2. Affinity constants Ka (298.0 K, CD2Cl2) for the coordination of
N-heterocycles 13 and 14 to ZnII-1a and ZnII-TTP 15 were obtained from
nonlinear curve-fitting of 1H NMR binding isotherms (See the Supporting
Information, Figures S37–S40).[48]
greater perturbation of the juxtaposed proton nuclei
(1H NMR spectroscopy) in the host, thereby revealing its lo-
cation. The binding of 13 to ZnII-1a perturbed the Hf pro-
tons (Dd=0.056 ppm, Figure 8A and C) at the outer side to
a greater degree than the inner Hc/d protons (Dd=0.01–
0.02 ppm, Figure 8A and C). On the contrary, the binding of
14 to ZnII-1a caused a smaller change in the chemical shift
of the outer Hf protons (Dd=0.032 ppm, Figure 8A and C),
but a greater shift of the inner Hc/d protons (Dd=0.033–
0.039 ppm, Figure 8A and C). Importantly, proton nuclei Ha
(Figure 8A), in between the two faces of ZnII-1a, experi-
enced almost identical perturbation upon the coordination
of either 13 or 14 (Dd=0.017–0.019 ppm, see the Supporting
Information, Figures S35 and S36).[48] Once again, the exper-
imental results suggested that 1,5-dicyclohexylimidazole 14
prefers coordinating to ZnII-1a from its inner face.
[a]
Host
Guest
Ka [MÀ1
]
ZnII-1a
13
14
13
14
1.9Æ0.2ꢂ104
ZnII-1a
>ꢂ105
ZnII-TTP 15
ZnII-TTP 15
2.1Æ0.2ꢂ104
2.4Æ0.2ꢂ104
[a] at T=298.0 K.
the observed high affinity of 1,5-dicyclohexylimidazole 14
toward ZnII-1a could perhaps be related to a greater degree
of entrapment.
We further reasoned that in the case of “in” versus “out”
coordination (Figure 9A and B), a guest should cause a
8878
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8870 – 8881

Role of Chirality in Directing the Formation of Cup-Shaped Porphyrins
FULL PAPER
Figure 8. (A) Electrostatics potential surface of 1a (EPS, AM1 Spartan) and energy-minimized structure (DFT: RI-BP86/SV(P),TZVP) of ZnII-1a.
1
(B) Chemical structures of 13, 14, and ZnII-TTP 15. (C) Relative H NMR chemical shifts (Dd, ppm) of proton nuclei Hc/d/f in ZnII-1a that occurred upon
incremental addition of 13 (left) and 14 (right) to its CD2Cl2 solution at 298.0 K.
1
Variable temperature H NMR measurements (400 MHz,
CD2Cl2) of pseudo C4 symmetric 13-ZnII-1a and 14-ZnII-1a
(Figure 9A and B) showed decoalescence of proton reso-
nances at lower temperatures (see Figure 9C and the Supp-
porting Information, Figures S41 and S42).[48] That is to say,
titration calorimetry (ITC) measurements are typically rec-
ommended for determining thermodynamic parameters of
binding, these would be very challenging to examine here
for clarifying the out/in equilibrium. The entrapment of 14
by ZnII-1a was indeed favorable at 298.0 K (DG8out/in
=
1
each H NMR signal would split into two signals with the in-
À1.2Æ0.1 kcalmolÀ1, Table 1), although the out/in equilibri-
um was favored by entropy (TDS8out/in =3.5Æ0.1 kcalmolÀ1).
We postulate that such favorable entropy must be related to
the desolvation of 1,5-dicyclohexylimidazole as it moves
from the outer to the inner side of ZnII-1a whereby it be-
comes partly surrounded by the host (Figure 9B). The out/
in conversion of N-methylimidazole 13 (Figure 9A), having
a smaller contacting surface, was accordingly accompanied
with less favorable entropic change (TDS8out/in =0.76Æ
0.02 kcalmolÀ1, Table 1). The observed adverse enthalpy
(DH8out/in =0.7–2.3 kcalmolÀ1) was, in both cases, likely a
result of differential solvation as our computational studies
indicate that placing 13 and 14 in the interior of ZnII-1a
should be enthalpically inexpensive (DH8out/in =À0.74 to
À1.02 kcalmolÀ1, Table 1).
tensity ratio being a function of the external temperature
(Figure 9C). Evidently, the chemical exchange of ligands 13
and 14 to and from ZnII-1a was, at lower temperatures,
slowed down to display the out/in stereoisomerism. On the
basis of our prior measurements, we assumed that 1,5-dicy-
clohexylimidazole 14 would predominantly populate the in-
terior of ZnII-1a at room temperature so the Hd resonance
(Figure 9C) that increased in intensity at higher tempera-
tures was assigned to Hd(in); note that proton resonances cor-
responding to bound 13 and 14 were broadened and close to
the base line, which prevented their full assignment and
study. Furthermore, the integration of Hd(in/out) signals at var-
ious temperatures, for both 13-ZnII-1a and 14-ZnII-1a, al-
lowed us to evaluate the ratio of out/in stereoisomers as a
function of temperature and create the corresponding van ꢃt
Hoff plots (Figure 9A and B, Table 1). Whereas isothermal
Chem. Eur. J. 2011, 17, 8870 – 8881
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8879

´
J. D. Badjic et al.
toward developing gated 1a
and understanding the host as
well as the catalytic characteris-
tics.
Acknowledgements
This work was financially supported
with funds obtained from the National
Science Foundation under CHE-
1012146. We thank the Ohio Super-
computer Center for generous compu-
tational resources.
[1] X. Bao, S. Rieth, S. Stojanovic,
C. M. Hadad, J. D. Badjic, Angew.
Chem. 2010, 122, 4926; Angew.
Chem. Int. Ed. 2010, 49, 4816.
[2] S. Rieth, X. Bao, B.-Y. Wang,
C. M. Hadad, J. D. Badjic, J. Am.
Chem. Soc. 2010, 132, 773.
[3] S. Rieth, B.-Y. Wang, X. Bao,
J. D. Badjic, Org. Lett. 2009, 11,
2495.
[4] B.-Y. Wang, X. Bao, S. Stojanovic,
C. M. Hadad, J. D. Badjic, Org.
Lett. 2008, 10, 5361.
[5] B.-Y. Wang, X. Bao, Z. Yan, V.
Maslak, C. M. Hadad, J. D.
Badjic, J. Am. Chem. Soc. 2008,
130, 15127.
[6] B.-Y. Wang, S. Rieth, J. D. Badjic,
J. Am. Chem. Soc. 2009, 131,
7250.
[7] S. Rieth, K. Hermann, B.-Y.
Wang, J. D. Badjic, Chem. Soc.
Rev. 2011, 40, 1609.
[8] S. Rieth, J. D. Badjic, Chem. Eur.
J. 2011, 17, 2562.
[9] R. J. Hooley, J. Rebek, Chem.
Biol. 2009, 16, 255.
[10] R. Cacciapaglia, S. Di Stefano, L.
Mandolini, Acc. Chem. Res. 2004,
37, 113.
[11] S. Litvinchuk, G. Bollot, J.
Mareda, A. Som, D. Ronan,
M. R. Shah, P. Perrottet, N. Sakai,
S. Matile, J. Am. Chem. Soc.
2004, 126, 10067.
Figure 9. Energy-minimized structures (DFT: RI-BP86/SV(P),TZVP) of (A) N-methylimidazole 13 and
(B) 1,5-dicyclohexylimidazole 14 coordinated to the inner/outer side of ZnII-1a. The van ꢃt Hoff plots for each
out/in equilibrium (far right) were obtained from variable temperature 1H NMR data. (C) The decoalescence
of the Hd signal in 14-ZnII-1a occurred at lower temperatures.
Conclusion
[12] M. S. Masar III, N. C. Gianneschi,
C. G. Oliveri, C. L. Stern, S. T. Nguyen, C. A. Mirkin, J. Am. Chem.
Soc. 2007, 129, 10149.
The diastereoselective installation of bicyclic rings around a
porphyrin macrocycle can be precisely controlled by using
the here developed protocols to form cup-shaped hosts,[56–58]
which are of interest in supramolecular and catalytic chemis-
try. The coordination of ligands to ZnII-metallated 1a oc-
curred at both of its faces. At room temperature, however,
the bulkier 1,5-dicyclohexylimidazole 14 became fully en-
capsulated, whereas smaller N-methylimidazole 13 was
equally partitioned in between the inner and outer side of
ZnII-1a. On the basis of the binding thermodynamics, we de-
duced that differential solvation of ligands contributed to
their distinct partitioning. Our interest is now directed
[13] J. Meeuwissen, J. N. H. Reek, Nat. Chem. 2010, 2, 615.
[14] J. Meeuwissen, N. H. Reek Joost, Nat. Chem. 2010, 2, 615.
[15] A. Nakamura, Y. Inoue, J. Am. Chem. Soc. 2003, 125, 966.
[16] M. D. Pluth, R. G. Bergman, K. N. Raymond, J. Am. Chem. Soc.
2008, 130, 11423.
[17] J. K. M. Sanders, Chem. Eur. J. 1998, 4, 1378.
[18] J. T. Groves, K. Shalyaev, J. Lee, Porphyrin Handb. 2000, 4, 17.
[19] J. S. Lindsey, Acc. Chem. Res. 2010, 43, 300.
[20] S. Shanmugathasan, C. Edwards, R. W. Boyle, Tetrahedron 2000, 56,
1025.
[21] P. Thordarson, E. J. A. Bijsterveld, A. E. Rowan, R. J. M. Nolte,
Nature 2003, 424, 915.
8880
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8870 – 8881

Role of Chirality in Directing the Formation of Cup-Shaped Porphyrins
FULL PAPER
[22] R. van Hameren, P. Schoen, A. M. van Buul, J. Hoogboom, S. V.
Lazarenko, J. W. Gerritsen, H. Engelkamp, P. C. M. Christianen,
H. A. Heus, J. C. Maan, T. Rasing, S. Speller, A. E. Rowan,
J. A. A. W. Elemans, R. J. M. Nolte, Science 2006, 314, 1433.
[23] J. P. Collman, X. Zhang, V. J. Lee, E. S. Uffelman, J. I. Brauman, Sci-
ence 1993, 261, 1404.
[42] T. Mizutani, T. Kurahashi, T. Murakami, N. Matsumi, H. Ogoshi, J.
Am. Chem. Soc. 1997, 119, 8991.
[43] H. Ogoshi, T. Mizutani, T. Hayashi, Y. Kuroda, Porphyrin Handb.
2000, 6, 279.
[44] T. Hayashi, T. Aya, M. Nonoguchi, T. Mizutani, Y. Hisaeda, S. Kita-
gawa, H. Ogoshi, Tetrahedron 2002, 58, 2803.
[24] Z. Fang, R. Breslow, Org. Lett. 2006, 8, 251.
[25] S. J. Lee, S.-H. Cho, K. L. Mulfort, D. M. Tiede, J. T. Hupp, S. T.
Nguyen, J. Am. Chem. Soc. 2008, 130, 16828.
[45] H. Imai, S. Nakagawa, E. Kyuno, J. Am. Chem. Soc. 1992, 114, 6719.
[46] Y. Uemori, A. Nakatsubo, H. Imai, S. Nakagawa, E. Kyuno, Inorg.
Chem. 1992, 31, 5164.
[26] W. Liu, J. T. Groves, J. Am. Chem. Soc. 2010, 132, 12847.
[27] P. Bhyrappa, J. K. Young, J. S. Moore, K. S. Suslick, J. Mol. Catal. A
1996, 113, 109.
[28] H. Uno, H. Watanabe, Y. Yamashita, N. Ono, Org. Biomol. Chem.
2005, 3, 448.
[47] Y. Uemori, S. Takinami, A. Takahashi, H. Munakata, H. Imai, S. Na-
kagawa, E. Kyuno, Inorg. Chim. Acta 1994, 224, 157.
[48] For additional details, see the Supporting Information.
[49] P. Peluso, C. Greco, O. De Lucchi, S. Cossu, Eur. J. Org. Chem.
2002, 4024.
[29] T. Okujima, Y. Hashimoto, G. Jin, H. Yamada, H. Uno, N. Ono, Tet-
rahedron 2008, 64, 2405.
[50] S. Cossu, O. De Lucchi, P. Peluso, R. Volpicelli, Tetrahedron Lett.
1999, 40, 8705.
[30] B. J. Littler, Y. Ciringh, J. S. Lindsey, J. Org. Chem. 1999, 64, 2864.
[31] G. R. Geier, 3rd, Y. Ciringh, F. Li, D. M. Haynes, J. S. Lindsey, Org
Lett. 2000, 2, 1745.
[32] J. S. Lindsey, Porphyrin Handb. 2000, 1, 45.
[33] P. D. Rao, S. Dhanalekshmi, B. J. Littler, J. S. Lindsey, J. Org. Chem.
2000, 65, 7323.
[51] S. Cossu, O. De Lucchi, P. Pasetto, Angew. Chem. 1997, 109, 1598;
Angew. Chem. Int. Ed. Engl. 1997, 36, 1504.
[52] S. S. Eaton, G. R. Eaton, J. Am. Chem. Soc. 1977, 99, 1601.
[53] J. Braun, M. Koecher, M. Schlabach, B. Wehrle, H.-H. Limbach, E.
Vogel, J. Am. Chem. Soc. 1994, 116, 6593.
[54] Y. Liu, R. Warmuth, Org. Lett. 2007, 9, 2883.
[55] M. C. T. Fyfe, F. M. Raymo, J. F. Stoddart, Stimul. Concepts Chem.
2000, 211.
[34] P. D. Rao, B. J. Littler, G. R. Geier III, J. S. Lindsey, J. Org. Chem.
2000, 65, 1084.
[35] G. R. Geier, III, J. B. Callinan, P. D. Rao, J. S. Lindsey, J. Porphyr-
ins Phthalocyanines 2001, 5, 810.
[56] C.-H. Lee, H. Miyaji, D.-W. Yoon, J. L. Sessler, Chem. Commun.
2008, 24.
[36] G. R. Geier III, B. J. Littler, J. S. Lindsey, J. Chem. Soc. Perkin
Trans. 2 2001, 712.
[57] G. Gil-Ramꢄrez, J. Bene tBuchholz, E. C. Escudero-Adan, P. Balles-
ter, J. Am. Chem. Soc. 2007, 129, 3820.
[37] G. R. Geier III, B. J. Littler, J. S. Lindsey, J. Chem. Soc. Perkin
Trans. 2 2001, 701.
[38] M. J. Crossley, P. Thordarson, J. P. Bannerman, P. J. Maynard, J. Por-
phyrins Phthalocyanines 1998, 2, 511.
[39] C. B. Nielsen, F. C. Krebs, Tetrahedron Lett. 2005, 46, 5935.
[40] D. M. Rudkevich, W. Verboom, D. N. Reinhoudt, J. Org. Chem.
1995, 60, 6585.
[58] For hosts resembling our cup-shaped porphyrins, see: a) Y. Ramo-
denc, R. Schwenninger, T. Phan, B. Krꢅutler, K. Gruber, C. Kratky,
Angew. Chem. 1994, 106, 939; Angew. Chem. Int. Ed. Engl. 1994, 33,
889; b) J. Schlogl, B. Krꢅutler, Synlett 1999, S1, 969; c) R. Schwen-
ninger, Y. Ramondenc, K. Wurst, J. Schlogl, B. Krꢅutler, Chem. Eur.
J. 2000, 6, 1214.
[41] O. Middel, W. Verboom, D. N. Reinhoudt, J Org Chem. 2001, 66,
3998.
Received: March 11, 2011
Published online: June 15, 2011
Chem. Eur. J. 2011, 17, 8870 – 8881
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8881