Rigidified dendritic structures for imprinting chiral information

Article abstract of DOI:10.1021/om050647y

Chiral, polymerizable P₂Pt(S-BINOL) metallodendrimers constructed with a rigid phenyl benzoate ester linkage were copolymerized into highly porous and cross-linked methacrylate-based polymers. The performance of these polymers with respect to BINOL exchange reaction enantioselectivities was compared to similar metallodendrimers constructed of more flexible benzyl ether repeat units. The increased rigidity of the ester linkage resulted in higher enantioselectivities for a BINOL/Br₂BINOL exchange reaction, though exact structure/ selectivity relationships are difficult to establish. At elevated temperatures, BINOL loss is observed leading to the inactivation of sites, though it is the less selective, and therefore less hindered sites, that are preferentially deactivated. Enantiomeric ratios of up to 84:16 were observed for the substitution of P₂Pt(S-BINOL) with (±)-Br ₂BINOL within the imprinted polymer site, which equals the best yet achieved.

Full text of DOI:10.1021/om050647y

6338
Organometallics 2005, 24, 6338-6350
Rigidified Dendritic Structures for Imprinting Chiral
Information
Sharon M. Voshell and Michel R. Gagne´*
Department of Chemistry, University of North Carolina at Chapel Hill,
Chapel Hill, North Carolina 27599-3290
Received July 28, 2005
Chiral, polymerizable P2Pt(S-BINOL) metallodendrimers constructed with a rigid phenyl
benzoate ester linkage were copolymerized into highly porous and cross-linked methacrylate-
based polymers. The performance of these polymers with respect to BINOL exchange reaction
enantioselectivities was compared to similar metallodendrimers constructed of more flexible
benzyl ether repeat units. The increased rigidity of the ester linkage resulted in higher
enantioselectivities for a BINOL/Br2BINOL exchange reaction, though exact structure/
selectivity relationships are difficult to establish. At elevated temperatures, BINOL loss is
observed leading to the inactivation of sites, though it is the less selective, and therefore
less hindered sites, that are preferentially deactivated. Enantiomeric ratios of up to 84:16
were observed for the substitution of P2Pt(S-BINOL) with (()-Br2BINOL within the
imprinted polymer site, which equals the best yet achieved.
Introduction
the active site is spatially oriented in the protein cavity
and each site is identical. This contrasts the case of
molecularly imprinted active sites since each active site
is constructed of a seemingly random collection of
monomers under conditions that are heavily dependent
on complex phase separations and heterogeneous reac-
tion kinetics.7 The ill-defined nature of matrix formation
necessarily leads to a diversity of sites, each differing
in the degree of encapsulation by the polymer matrix.8
Some of the sites are situated on the surface with no
surrounding polymer matrix, while other sites are
formed with well-defined polymer cavities. Finally, some
sites are completely surrounded by polymer and, thus,
are prevented from interacting with substrate. Although
good activities and selectivities are possible with MIP
catalysts, this problem of site heterogeneity prevents
MIPs from displaying the desirably high structure-
activity correlation that one observes in metalloen-
zymes.9 Another approach to achieving metalloenzyme-
like active sites has been monomolecular imprinting
where the active site is contained within an organic
nanoparticle synthesized by the intramolecular cross-
linking of a dendrimer.10
In recent years, chemists have attempted to synthe-
size functional polymers that rival the sophistication of
the chemical systems found in nature. Molecularly
imprinted polymers (MIPs)1 are particularly suited to
the task of imparting to traditional metal catalysts the
outer-sphere control elements observed in a metalloen-
zyme. A molecularly imprinted polymer is formed when
a template molecule is copolymerized into a highly cross-
linked polymer matrix with a cross-linking monomer in
the presence of a porogen. When the template is
removed, a three-dimensional cavity is formed that is
complementary to the template molecule in size, shape,
and chemical functionality. The guiding notion is thus
that this man-made cavity can be made similar to the
active site in a metalloenzyme. The molecular recogni-
tion properties of these cavities have been utilized most
successfully as chromatographic stationary phases,2 but
applications in sensors,3 synthesis,4 and catalysis5,6 have
also been demonstrated.
Because enzymes are unique molecular structures,
built up of specific interactions between amino acids,
(1) (a) Whitcombe, M. J.; Vulfson, E. N. Adv. Mater. 2001, 13, 467-
478. (b) Wulff, G. Angew. Chem., Int. Ed. Engl. 1995, 34, 1812-1833.
(c) Sellergren, B.; Hall, A. J. In Molecularly Imprinted Polymers;
Sellergren, B., Ed.; Elsevier: Amsterdam, 2001; pp 21-55.
(2) (a) Sellergren, B. J. Chromatogr., A 2001, 906, 227-252. (b)
Andersson, L. I. J. Chromatogr,. B 2000, 745, 3-13. (c) Andersson, L.
I.; Mosbach, K. J. Chromatogr., A 1990, 516, 323-331. (d) Buchmeiser,
M. R. J. Chromatogr., A 2001, 918, 233-266.
(3) Molecularly Imprinted Materials-Sensors and Other Devices;
Shea, K. J., Yan, M., Roberts, M. J., Eds.; Material Research Society
Symposium Proceedings 723; Materials Research Society: Warrendale,
PA, 2002.
(4) (a) Whitcombe, M. J.; Alexander, C.; Vulfson, E. N. Synlett 2000,
6, 911-923. (b) Bru¨ggeman, O.; Haupt, K.; Ye, L.; Yilmaz, E.; Mosbach,
K. J. Chromatogr., A 2000, 889, 15-24.
(5) For reviews see: (a) Wulff, G. Chem. Rev. 2002, 102, 1-27. (b)
Tada, M.; Iwasawa, Y. J. Mol. Catal. 2003, 199, 115-137. (c) Alex-
ander, C.; Davidson, L.; Hayes, W. Tetrahedron 2003, 59, 2025-2057.
(d) Severin, K. Curr. Opin. Chem. Biol. 2000, 4, 710-714.
(6) (a) Liu, J.; Wulff, G. J. Am. Chem. Soc. 2004, 126, 7452-7453.
(b) Svenson, J.; Zheng, N.; Nicholls, I. A. J. Am. Chem. Soc. 2004, 126,
8554-8560. (c) Visnjevski, A.; Yilmaz, E.; Bru¨ggeman, O. Appl, Catal.,
A 2004, 260, 169-174. (d) Wendicke, S. B.; Burri, E.; Scopelliti, R.;
Severin, K. Organometallics 2003, 22, 1894-1897. (e) Nestler, N.;
Severin, K. Org. Lett. 2001, 3, 3907-3909. (f) Polborn, K.; Severin, K.
Chem. Eur. J. 2000, 6, 4604-4611. (g) Cammidge, A. N.; Baines, N.
J.; Bellingham, R. K. Chem. Commun. 1999, 24, 2588-2589. (h) A¨ıt-
Haddou, H.; Leeder, S. M.; Gagne´, M. R. Inorg. Chim. Acta 2004, 357,
3854-3864. (i) Viton. F.; White, P. S.; Gagne´, M. R. Chem. Commun.
2003, 24, 3040-3041. (j) Koh, J. H.; Larsen. A. O.; White, P. S.; Gagne´,
M. R. Organometallics 2002, 21, 7-9.
(7) Sherrington, D. C. Chem. Commun. 1998, 21, 2275-2286.
(8) Brunkan, N. M.; Gagne´, M. R. J. Am. Chem. Soc. 2000, 122,
6217-6225.
(9) Carey, J. R.; Ma, S. K.; Pfister, T. D.; Garner, D. K.; Kim, H. K.;
Abramite, J. A.; Wang, Z.; Guo, Z.; Lu, Y. J. Am. Chem. Soc. 2004,
126, 10812-10813.
10.1021/om050647y CCC: $30.25 © 2005 American Chemical Society
Publication on Web 11/12/2005

Rigidified Dendritic Structures
Organometallics, Vol. 24, No. 26, 2005 6339
Scheme 1
Figure 1. Bond rotation energy profiles for phenyl ben-
zoate (blue), benzyl phenyl ether (green), and diphenyl-
ethane (red) vs φ1,2,3,4, along with the lowest energy
conformation of phenyl benzoate (RHF-6-31G* as imple-
mented in MacSpartan02).
which are consequently capable of adopting a large
number of conformations during the imprinting process.
This was postulated to lead to sites that, although
constitutionally similar, were less selective by virtue of
conformational heterogeneity. We hypothesized that
increasing the rigidity of the dendritic arms might
improve the fidelity of the imprint while retaining the
site homogeneity and, thus, improve the observed drop
in selectivity.
MIPs previously produced in our lab from metal-
lomonomer 1 (Scheme 1) contained a complex distribu-
tion of platinum active sites. The overall enantioselec-
tivity of the polymer for the imprinted enantiomer of
BINOL was 82:18, with some sites achieving selectivi-
ties of up to 97:3.8 To achieve a polymer in which each
site is similarly encapsulated by the polymer matrix,
the environment directly surrounding the template
molecule must be better controlled. To this end, our
laboratory developed polymerizable, dendritic diphos-
phine ligands11 with the expectation that the dendritic
arms of the phosphine ligand would provide an interface
to the bulk polymer that would be similar for each
platinum atom.12
This strategy with metallodendrimers 2, 3, and 4
(Scheme 1) showed less variability in reactivity and
selectivity as a function of temperature (cf. 1), indicative
of a more homogeneous distribution of sites; however,
the overall selectivity of the dendritic MIP systems was
lower than that of the nondendritic MIP. The source of
the lower selectivity was thought to be the flexibility of
the benzyl ether arms of the dendritic phosphine ligand,
Results and Discussion
Conformational Analysis. An approach to achiev-
ing a more rigid template was arrived at by an analysis
of rotational barriers. In alkanes and ethers, the barrier
to rotation around a C-C or C-O bond is less than 5
kcal/mol, and there is only a slight preference for the
anti conformer. In contrast, the C-O bond of an ester
possesses partial double-bond character, which in-
creases the barrier to rotation around the C-O bond
(10-13 kcal/mol).13 Figure 1 shows the rotational pro-
files of 1,2-diphenylethane, benzyl phenyl ether, and
phenyl benzoate as a function of the dihedral angle
defined by atoms 1, 2, 3, and 4 (φ1,2,3,4).14 The barrier to
rotation around the C-O bond in the ester is 11 kcal/
mol, and at room temperature, the thermodynamic
preference for the anti conformer is significant. In
(10) (a) Mertz, E.; Elmer, S. L.; Balija, A. M.; Zimmerman, S. C.
Tetrahedron 2004, 60, 11191-11204. (b) Zimmerman, S. C.; Zharov,
I.; Wendland, M. S.; Rakow, N. A.; Suslick, K. S. J. Am. Chem. Soc.
2004, 125, 13504-13518. (c) Zimmerman, S. C.; Lemcoff, N. G. Chem.
Commun. 2004, 1, 5-14.
(11) Balaji, B. S.; Obora, Y.; Ohara, D.; Koide, S.; Tsuji, Y. Orga-
nometallics 2001, 20, 5342-5350.
(12) Becker, J. J.; Gagne´, M. R. Organometallics 2003, 22, 4984-
4998.
(13) Eliel, E. L.; Wilen, S. H.; Mahler, L. N. Conformation of Acyclic
Molecules. Stereochemistry of Organic Compounds; John Wiley &
Sons: New York, 1994; pp 597-624.
(14) The data were generated by driving the relevant dihedral angle
in 10° increments with full optimization at each step (RHF-6-31G*)
as implemented in Macspartan 2002.

6340 Organometallics, Vol. 24, No. 26, 2005
Voshell and Gagne´
Figure 2. Histograms for dendritic structures showing the number of local minima (AM1) normalized to the global
minimum (A-G), and the corresponding Boltzmann distribution at 60 °C for each histogram.
contrast, the ether has a lower barrier to rotation (∼7
kcal/mol) and a significantly wider range of accessible
conformers within 2 kcal/mol of the energy minimum
(60-300°). Additionally, the lowest energy conformer of
the ester is further rigidified because it contains a
phenyl ring locked into planarity with the ester group
to preserve conjugation and an -OPh substituent that
has a slight (∼2 kcal/mol) preference to be orthogonal
to the plane of the ester. This analysis therefore
suggested that templates with ester linkages should be
significantly more rigid than templates with ether
linkages. The rigidity of the ester linkage in the den-
dritic structure should serve to limit the conformational
degrees of freedom of the dendritic phosphine ligand and
thus improve the conformational homogeneity of the
imprint.
To experimentally address the effect of increased
rigidity in a metallomonomer imprint, we chose to
replace the benzyl ether linkage in 2, 3, and 4 that was
closest to the metal core. This was for two reasons: (1)
ease of synthesis (vide infra) and (2) previous experi-
ments showing the aryl(4-vinylbenzoate) ester linkage
to be prone to hydrolysis.
informative view. We hypothesized that a dendron with
fewer low-energy conformers would provide a more
effective imprint than a dendron with many such
conformers.
The histograms in Figure 2 show the range of
conformers for several dendritic fragments obtained
from a semiempirical (AM1) Monte Carlo conformer
search.15 As expected, the substitution of an ester
linkage for an ether linkage (A and B) reduced the total
number of conformational minima, while simulta-
neously shifting the distribution of energies closer to the
global minimum. Changing the dendron substitution
pattern from the symmetric 3,5-disubstituted structure
in B to the 2,3-disubstituted in C results in a much
wider distribution of energies, though the total number
of local minima is smaller. This wider distribution in
energy could indicate that this set of structures is more
diverse and thus would provide a less effective imprint.
The 2,4-disubstituted dendron (D) is more similar to B
than C; however, there are a number of higher energy
conformations that become available. In the case of the
trisubstituted dendrons E and F, the increase in steric
(15) The data were generated by first doing a Monte Carlo conformer
distribution analysis with molecular mechanics (MMFF), keeping all
minima within 10 kcal/mol of the global minimum. A semiempirical
(AM1) geometry optimization of each local minimum was then carried
out. The duplicate structures were removed from the data set, and
each was normalized to the global minimum and plotted in histogram
fashion (Macspartan 2002).
To provide a semiquantitative measure of how such
an ester for ether substitution would affect the confor-
mational diversity, we performed a conformational
analysis at the single dendron level, realizing that this
approach would provide a simplistic but, perhaps,

Rigidified Dendritic Structures
Organometallics, Vol. 24, No. 26, 2005 6341
Scheme 2
bulk caused by 2,3,4-trisubstitution results in a very
narrow distribution of conformers, which must signifi-
cantly affect the resulting imprint. The incorporation
of a naphthalene repeat unit in G for the benzene repeat
unit present in B broadens the distribution of conform-
ers.
Although the pictorial representation of the energetic
profiles is simplified by the use of the histograms shown
in Figure 2, it must be stressed that the actual metal-
lomonomers have four dendrons linked to a phosphine
core, which not only multiplies the total number of
conformations, but necessarily affects the relative ener-
gies of the conformers as they are forced to interact with
each other and adopt compact, dense structures. This
is especially true for the ortho-substituted cases (C, D,
F, and G).
A histogram plot of energy and count is also not an
accurate representation of the conformer populations
since lower energy structures will naturally be more
populated than those of higher energy. To continue the
semiquantitative analysis of conformation, we calcu-
lated, based on a Boltzmann distribution at 60 °C, the
polymerization temperature, the contribution of each of
the conformers to the ensemble (Figure 2).16 For the
ether, even though there are many conformers within
5.5 kcal/mol of the global minimum (A), the three
conformers within 0.5 kcal/mol comprise 31% of the
Boltzmann distribution, and the eight conformers within
1.0 kcal/mol comprise 54% (red line, Figure 2). The ester
linkage (B) results in a Boltzmann distribution (green
line, Figure 2) wherein 62% of the conformers (9) lie
within 0.5 kcal/mol of the global minimum and 81% (17)
lie within 1.0 kcal/mol. When the substitution pattern
of the ester dendron is changed from 3,5-disubstituted
to 2,3-disubstituted (blue line, Figure 2), 45% of the
conformers (2) lie within 0.5 kcal/mol and 71% (5) lie
within 1.0 kcal/mol of the global minimum. These data
clearly indicate a more complex structural picture than
the histograms alone; steeper lines mean fewer contrib-
uting conformers.
The contrast between A and B was illuminating. Even
though the ester dendron B contained fewer local
minima than the ether dendron A, they tended to be
closer in energy to the global minimum and, therefore,
contributed more to the Boltzmann distribution (lower
slope, more data points in the 0-0.5 kcal/mol range).
The benzyl ether, by contrast, had many more available
conformations, but they were further from the global
minimum in energy, which minimized their population.
Thus, the benzyl ether dendron may, in fact, be popu-
lated by fewer accessible conformers (higher slope), at
least at this crude level of analysis.
The 2,3-disubstituted case (C), which has the fewest
local minima in the low-energy region, has the highest
slope of the cases shown in Figure 2. Of the three lowest
energy structures (comprising 57% of the conformers),
a superposition of the structures indicates that the
polymerizable benzyl arms project in roughly the same
region of space, a property that, a priori, might be
expected to increase active site homogeneity.
related templates wherein a key ether linkage was
replaced with an ester linkage and the substitution
pattern of the dendron was varied. The necessary
dendritic phosphine ligands were synthesized via a
convergent approach17 in which the phosphine core and
dendritic arms were synthesized separately and then
coupled through the ester linkage. This approach was
particularly convenient in that a large variety of den-
drons could be quickly and easily attached to the
phosphine core. In contrast, the original synthesis of the
benzyl ether metallomonomers (2, 3, and 4) coupled the
dendron to the phosphine ligand though a sensitive P-C
bond formation.12
The synthesis of the hydroxy-functionalized phos-
phine core is shown in Scheme 2. Addition of the
Grignard of 4-bromophenol tert-butyldimethylsilyl (TB-
DMS) ether to 1,2-bis(dichlorophosphino)ethane pro-
vided the desired diphosphine after borane protection
(dppe-OTBDMS). The silyl ether protecting groups
were then cleaved with tetrabutylammonium fluoride
to give hydroxy-functionalized diphosphine ligand dppe-
OH in moderate overall yield. Curiously, the function-
alized phosphine was isolated with 1 equiv of tetrabu-
tylammonium cation, which could not be removed by
repeated precipitation. The absence of a 1H NMR
resonance for the phenol protons suggests that the
product might be monoanionic wherein rapid proton
exchange is occurring; it is depicted as such in Scheme
2.
The carboxylic acid-terminated dendrons were syn-
thesized by reaction of vinyl benzyl chloride with the
appropriate di- and trihydroxy benzoic acids or methyl
benzoates using potassium carbonate, 18-crown-6 ether
as a phase transfer agent, and a catalytic amount of
tetrabutylammonium iodide in refluxing acetone.17 The
resultant dendritic esters were then hydrolyzed with
potassium hydroxide in ethanol18 or 2-propanol to reveal
the desired carboxylic acid terminus after workup.
Scheme 3 shows an example synthesis of dendron G1-
COOH from 3,5-dihydroxy methyl benzoate, as well as
the structures of the other dendrons synthesized in this
manner (G1(2,3)-COOH, G1(2,4)-COOH, G1t-COOH,
G1t(2,3,4)-COOH, and G1n(3,5)-COOH).
The diphosphine dendrons were then accessed by first
utilizing a DCC (1,3-dicyclohexylcarbodiimide)-mediated
Synthesis. The experimental approach to testing the
key hypothesis was to synthesize a variety of closely
(17) (a) Grayson, S. M.; Fre´chet, J. M. J. Chem. Rev. 2001, 101,
3819-3867. (b) Hawker, C. J.; Fre´chet, J. M. J. J. Am. Chem. Soc.
1990, 112, 7638-7647.
(18) van Nunen, J. L. M.; Folmer, B. B. B.; Nolte, R. J. M. J. Am.
Chem. Soc. 1997, 119, 283-291.
(16) Atkins, P. W. Physical Chemistry; W. H. Freeman and Com-
pany: New York, 1994; p 14. See Supporting Information for calcula-
tion details.

6342 Organometallics, Vol. 24, No. 26, 2005
Voshell and Gagne´
Scheme 3
Scheme 5
Scheme 4
was used in previous organometallic MIP experiments
and is known to yield high surface area (460 m2/g)
materials.20 The solutions gelled within 5 min, and after
24 h the polymerizations were complete, yielding yellow
monolithic polymers. After polymerization, the polymers
were washed with dichloromethane in a Soxhlet extrac-
tor for 6 h to remove chlorobenzene and then dried
overnight under vacuum. The theoretical platinum
content of these polymers was calculated to be 60-65
µmol Pt/g dry polymer.
Site Accessibility. The polymerization procedure
caused a small amount of (S)-BINOL to be cleaved from
the metallomonomers. This BINOL was recovered from
the Soxhlet extraction filtrate and quantified by HPLC
analysis. The results are collected in Figure 3 (white
bars) and show that more BINOL loss occurs with the
present metallomonomers than with the nondendritic
coupling of the acid and dppe-OH (the NBu4+ does not
appear to interfere), followed by borane deprotection by
treatment with 1,4-diazabicyclo[2.2.2]octane (DABCO).
Other standard deprotection conditions (e.g., Et2NH,
morpholine, HBF4) were incompatible with the ester
linkage and/or the vinyl functionality. The resultant
diphosphine ligands are not oxidized upon benchtop
exposure to air. The dendrons shown in Scheme 3 as
well as 4-vinylbenzoic acid (G0-COOH) were coupled
to dppe-OH to form diphosphine ligands for use in
MIPs (Scheme 4).
From the deprotected diphosphines, metallomono-
mers 5-11 (Scheme 5) were synthesized in three
efficient steps from (1,4-cyclooctadiene)PtCl2. First, the
diphosphine ligands were coordinated to platinum to
form the corresponding platinum dichloride in excellent
yield. The platinum dichloride was then converted to
the platinum carbonate by silver carbonate in wet
dichloromethane,19 followed by addition of (S)-BINOL
to form the metallomonomer. Excess BINOL was re-
moved by washing with diethyl ether, and the metal-
lomonomers were isolated as yellow and brown powders
in good overall yields.
The desired MIPs were synthesized from metal-
lomonomers 5-11 by heating (60 °C) under nitrogen a
solution of metallomonomer (1.5 mol %) and AIBN (1
mol %) in an equal volume of ethylene glycol dimethacry-
late (97.5 mol %) and chlorobenzene. This composition
Figure 3. BINOL loss during polymerization (white bars)
and site accessibility (gray bars) in MIPs expressed as a
percentage of theoretical platinum sites. Each data point
represents an average of three separate polymer prepara-
tions.
(19) Andrews, M. A.; Gould, G. L.; Klooster, W. T.; Koenig, K. S.;
Voss, E. J. Inorg. Chem. 1996, 35, 5478-5483.

Rigidified Dendritic Structures
Organometallics, Vol. 24, No. 26, 2005 6343
Scheme 6
Figure 4. Percentage of total platinum sites that ex-
changed Br2BINOL for (S)-BINOL at 60 °C (gray) and 120
°C (black). Each data point represents an average of three
separate polymer preparations.
metallomonomer 1. Previous experiments had shown
that the amount of BINOL loss at this stage could be
correlated to the basicity of the diphosphine; more
electron deficient ligands had less loss.12 An ester
moiety, in contrast to the original benzyl ether, pos-
sesses enough electron-withdrawing character to sta-
bilize the metallodendrimer so that BINOL loss is
limited to less than 10% of the platinum sites.
cleaved (S)-BINOL before treatment with hydrochloric
acid to excise all BINOL species for quantification. The
enantiomeric excess of the rebound Br2BINOL proved
to be a key metric for site characterization.
The total number of accessible sites in these polymers
was quantified by adding concentrated hydrochloric acid
to a suspension of the polymer in dichloromethane. This
procedure cleaves the BINOL from the platinum and
thus (after quantification) evaluates the degree of site
availability or, perhaps more precisely, the number of
sites with an available pathway for BINOL egress from
the polymer interior (Figure 3, gray bars). The dendritic
systems 2-11 all have fewer accessible sites than the
nondendritic system presumably due to increased con-
gestion around the metal center. As expected, there is
a generation-dependent decrease in the number of
accessible sites as shown by P-5, P-6, and P-9 similar
to the trend previously shown by P-2, P-3, and P-4.12
Polymers synthesized from the trisubstituted dendrons
(P-4, P-9, and P-10) show the lowest site accessibility,
presumably due to their greater steric protection of the
metal center, while polymers P-2 and P-5 have the
largest percentage of accessible sites.
From the data we conclude that the replacement of
an ether linkage with an ester linkage has little effect
on the overall site accessibility of the MIPs. Substitution
patterns of the dendrons also do not greatly affect the
amount of BINOL cleaved, as evidenced by comparing
P-6, P-7, and P-8 as well as P-9 and P-10.
Site Enantioselectivity. The selectivity of the plati-
num sites in the MIPs was studied by a BINOL/Br2-
BINOL exchange reaction (Scheme 6).8,12 The polymer
was treated with an excess (20 equiv relative to the
theoretical platinum content of the polymer) of racemic
Br2BINOL ((()-6,6′-dibromo-1,1′-bi-2-naphthol) in a 6:1
mixture of chlorobenzene and water at either 60 or 120
°C for 12 h. Those sites that are able to accommodate
the associative exchange mechanism displaced the (S)-
BINOL ligand with either (S)- or (R)-Br2BINOL based
on the selectivity imparted by the polymer surrounding
the platinum site. Some sites were too hindered for
BINOL exchange to occur and thus retained (S)-BINOL.
The polymer was then washed for 16 h in a Soxhlet
extractor to remove unreacted Br2BINOL as well as the
An ideal MIP, one that most closely resembles a
homogeneous enzyme active site, should contain identi-
cal platinum sites that each react to exchange Br2-
BINOL at the same rate and with the same enantiose-
lectivity. Because our MIPs have a heterogeneous
distribution of sites, the reactivity and enantioselectivity
of the sites vary with temperature, with the more open,
less selective, sites reacting at lower temperatures.8,12
We studied the BINOL/Br2BINOL exchange reaction at
two temperatures (60 and 120 °C) and assumed that
the difference in the number of platinum sites to react
and the enantiomeric excess of the rebound Br2BINOL
could be used to estimate the heterogeneity of the
platinum sites. A large difference would indicate a wide
site distribution, while conversely, a small variation
would be most consistent with a narrow site distribu-
tion.
As shown in Figure 4, P-1 synthesized from the
nondendritic metallomonomer showed the expectedly
large increase in the number of sites to react with Br2-
BINOL at 60 and 120 °C.8 In contrast, the polymers
prepared from dendritic metallomonomers tended to
show much smaller reactivity changes with tempera-
ture. As the data show, the ester linkage substitution
has no systematic effect on the differential reactivity
at 60 and 120 °C (compare P-2 with P-5, P-3 with P-6,
and P-4 with P-9). Most notable were the metallomono-
mers with ortho-styryl substituents (7, 8, 10, and 11),
which tended to show diminished site accessibility and
attenuated differences in reactivity between 60 and 120
°C, consistent with more hindered sites.
Unlike accessibility differences, the polymers pre-
pared from ester dendritic metallomonomers 5-11 do,
in general, show higher enantioselectivities for rebind-
ing Br2BINOL (Figure 5). These polymers however also
show a larger increase in enantioselectivity with tem-
perature than their ether counterparts, a phenomenon
that previously we have ascribed to active site hetero-
geneity.
Site Deactivation. The observation that P-10 and
P-11 had fewer sites rebind Br2BINOL at 120 °C than
at 60 °C, and yet had an increase in enantioselectivity,
(20) Santora, B. P.; Gagne´, M. R. Macromolecules 2001, 34, 658-
661.

6344 Organometallics, Vol. 24, No. 26, 2005
Voshell and Gagne´
Table 1. Site Deactivation in MIPs Expressed as a
Percentage of Platinum Sites that Excised BINOL
at 120 °C
polymera
P-5
P-6
accessible Pt Sites (%)b
BINOL lost (%)c
87
14
74
88
81
19
62
80
BINOL cleaved (%)d
total sites (%)e
a The data come from single experiments. b Before thermolysis.
c Quantified by HPLC after thermolysis at 120 °C for 12 h.
e BINOL lost + BINOL cleaved.
was quantitatively analyzed for (S)-BINOL by HPLC.
A typical HCl cleavage experiment was conducted on
the washed polymers, and the excised BINOL was
quantified by HPLC. The results are summarized in
Table 1. It was found that in P-5 14% of the platinum
sites became inactive at 120 °C, while the number of
“accessible” sites decreased from 87% to 74%. P-6
experienced BINOL loss from 19% of platinum sites, and
the site accessibility decreased from 81% to 62%. These
figures are in agreement with the amount of BINOL lost
from P-5 and P-6 under the standard Br2BINOL ex-
change conditions (Figure 6) and are internally consis-
tent with the number of sites that were accessible prior
to thermolysis.
Figure 5. Enantiomeric excess of rebound Br2BINOL at
60 °C (gray) and 120 °C (black). Each data point represents
an average of three separate polymer preparations.
Thus, thermolytic deactivation of the less selective sites
is sufficient to explain the trend of decreasing site
accessibility coupled with increased enantioselectivity
in the more hindered metallomonomers. We have no
evidence that this reactivity pattern holds for the varied
metallomonomers studied to date. Nevertheless, this
possibility adds, unfortunately, contributes a significant
uncertainty to our attempts to establish structure/
activity/selectivity relationships for achieving further
improvements. Equally uncertain is the intriguing
possibility that such an effect could be used as a general
strategy for increasing active site selectivity and homo-
geneity.
To further investigate the mechanism of BINOL loss
at the rebinding conditions, we studied the thermolysis
of 12, a nonpolymerizable analogue of 5. The reaction
of this compound in a 6:1 mixture of chlorobenzene to
water was monitored by phosphorus NMR. After 14 h
at 120 °C, complete decomposition of the P2Pt(BINOL)
was observed. A mixture of unidentified products with
31P chemical shifts ranging from 30 to 50 ppm was
obtained along with platinum black.
Similar thermolysis of (dppe)Pt(S-BINOL), 13, indi-
cated a significantly reduced thermal sensitivity, as only
a trace of 13 had decomposed (apparently to (dppe)-
PtCl2) after 14 h at 120 °C.21 These results indicated
that the compounds are indeed sensitive to solution
thermolysis, but that this is significantly mitigated in
the constrained environment of a rigid cross-linked
polymer. Perhaps the more open sites, those most prone
to deactivation, are those that are also flexible enough
for decomposition (like solution).
Figure 6. Percentage of platinum sites inactivated during
rebinding at 60 °C (gray) and 120 °C (black). Each data
point represents an average of three separate polymer
preparations.
was most unusual and opposite of the normally observed
trend. Equally confusing, but informative, was the fact
that the amount of residual (S)-BINOL after the final
cleavage dropped by 20% on going from 60 to 120 °C,
this despite the lack of compensating bound Br2BINOL
at the higher temperature. Thus, it appears that some
of the active sites became deactivated, i.e., lose (S)-
BINOL but do not rebind Br2BINOL. Since the enan-
tiomeric excess of the rebound Br2BINOL actually
increased, the deactivated sites must therefore be those
that are otherwise less selective for the imprinted
BINOL enantiomer. In other words, the less selective
sites deactivate under the reaction conditions.
The loss of active sites during the rebinding can be
quantified by subtracting from the total number of
accessible sites before thermolysis (Figure 3) the sum
of Br2BINOL and residual (S)-BINOL cleaved after the
thermal exchange. As shown in Figure 6, each of the
polymers had some loss of sites, but it was always larger
at higher temperatures, especially for those metal-
lomonomers with ortho substituents (7, 8, 10, and 11).
In principle, this scenario should be testable by
quantifying the amount of (S)-BINOL that is released
during the Br2BINOL exchange; however the excess Br2-
BINOL overwhelms the residual (S)-BINOL and the
analysis is problematic. Instead, samples of P-5 and P-6
were exposed to the rebinding reaction conditions in the
absence of Br2BINOL (12 h at 120 °C in 6:1 chloroben-
zene/water). The polymers were then washed in a
Soxhlet extractor with dichloromethane, and the extract
(21) The possibility that platinum sites in the polymer were
abstracting chlorine atoms from the chlorinated solvent during the
rebinding process was considered; however, BINOL loss was still
observed when P-5 was heated at 110 °C for 12 h in a 6:1 mixture of
toluene and water.

Rigidified Dendritic Structures
Organometallics, Vol. 24, No. 26, 2005 6345
NMR spectra were recorded on either a Bruker Avance 300
or a Bruker Avance 400 spectrometer. Chemical shifts are
reported in ppm and are referenced to the residual solvent
peaks (1H and 13C NMR) or to an external standard (85%
H3PO4, 31P NMR). The dendritic P2PtCl2 compounds were not
soluble enough in chloroform to allow carbon spectra to be
recorded even overnight. Many of the carbon signals in G1t-
(2,3,4) and the Pt(BINOL)s overlap, and therefore signals were
not detected for all carbons in the compounds. Carbon spectra
for phosphorus-containing compounds were recorded while
decoupling both hydrogen and phosphorus nuclei. Carbon
spectra for dppe-OTBDMS and dppe-OH are shown in the
Supporting Information as evidence of purity because these
compounds did not combust completely for elemental analysis
(δ%C was repeatedly >10%). Microanalysis of the P2Pt-
(BINOL) compounds was difficult because these compounds
tend to retain traces of solvent.
Summary
A number of MIPs were prepared from metallomono-
mers comprised of dendritic phosphine ligands with
comparatively rigid ester linkages containing variable
arrangements of polymerizable arms. The reactivity and
enantioselectivity of these MIPs in a BINOL exchange
reaction were investigated to determine which struc-
tural variables were key to achieving optimum selectiv-
ity and homogeneity. The data clearly showed that the
rigidity of the ester linkage indeed increased the enan-
tioselectivity over more flexible benzyl ether MIPs.
Unfortunately, a mechanism for BINOL cleavage com-
petes with the exchange process and results in inactive
sites, making it impossible to rigorously characterize the
homogeneity of sites contained within these MIPs based
solely on reactivity and selectivity variations with
temperature. Nevertheless, these systems once again
show that it is possible to achieve enantiomer-selective
reactivities in platinum MIPs where none is possible in
solution (the dppe ligand is achiral). Significant chal-
lenges remain to realizing the potential of this approach.
HPLC analysis was performed on a Hewlett-Packard Series
1100 instrument, using a Daicel Chiralcel OD-H column (95%
hexanes/5% ethanol, 0.8 mL/min flow rate). Chromatograms
were recorded at wavelengths of 220, 232, 254, 289, and 333
nm and compared to a calibration curves to determine analyte
concentration.
TBDMS 4-Bromophenol. A solution of 4-bromophenol
(10.0 g, 57.8 mmol) and imidazole (5.9 g, 86.7 mmol) in 200
mL of dimethylformamide was cooled to 0 °C, and tert-
butyldimethylsilyl chloride (TBDMS-Cl, 9.6 g, 63.6 mmol) was
added. The solution was stirred for 12 h and allowed to warm
to room temperature. A saturated aqueous solution of am-
monium chloride (100 mL) was added, and the mixture was
extracted with 200 mL of ethyl acetate. The ethyl acetate was
washed with 100 mL of 0.5 M NaOH and with 100 mL of brine,
dried over magnesium sulfate, and filtered. The solvent was
removed in vacuo to give TBDMS 4-bromophenol (15.5 g, 93%
yield). 1H NMR (400 MHz, CDCl3, δ): 7.34 (d, J ) 8.4 Hz, 2H),
6.74 (d, J ) 8.4 Hz, 2H), 1.02 (s, 9H), 0.22 (s, 6H). 13C NMR
(100.6 MHz, CDCl3, δ): 155.1, 132.6, 122.2, 113.9, 26.0, 18.5,
-4.2. Anal. Calcd for C12H19BrOSi: C, 50.17; H, 6.67. Found:
C, 50.02; H, 6.68.
dppe-OTBDMS. Magnesium turnings (2.0 g, 83.5 mmol)
were flame dried inside a 100 mL Schlenk flask. The flask was
cooled and 2.0 mL of THF was added, followed by 0.25 mL
(4.3 mmol) of methyl iodide. The mixture was stirred for
several minutes until cloudy. The liquid was removed by
cannula, and the magnesium was rinsed three times with 2
mL of THF. A solution of 12.0 g (41.8 mmol) of TBDMS
4-bromophenol in 40 mL of THF was added to the magnesium,
and the mixture was refluxed for 3 h. After cooling, the solution
was filtered via cannula from the magnesium into a flame-
dried 250 mL Schlenk flask. The solution was cooled to -50
°C, and a solution of 1.26 mL (8.35 mmol) of 1,2-bis(dichloro-
phosphino)ethane was added dropwise. The solution was
stirred at -50 °C for 4 h before quenching with 2.0 mL of
deoxygenated methanol. Borane-tetrahydrofuran complex (60
mL of a 1.0 M solution in tetrahydrofuran) was added, and
the solution was allowed to warm to room temperature
overnight. The volume of the solution was reduced in vacuo
until a white solid appeared. Ethanol (25 mL) was added, and
the mixture was stored at -26 °C overnight. The white solid
was collected by vacuum filtration (5.1 g, 64% yield). 1H NMR
(400 MHz, CDCl3, δ): 7.45 (m, 8H), 6.84 (d, J ) 8.4 Hz, 8H),
2.22 (br, 4H), 0.95 (s, 36H), 0.19 (s, 24H). 13C NMR (75.5 MHz,
CDCl3, δ): 158.8, 133.9, 120.8, 120.1, 25.7, 20.3, -4.2. 31P NMR
(161.3 MHz, CDCl3, δ): 15.5.
Experimental Section
General Methods. Racemic Br2BINOL,22 (COD)PtCl2,23
G1-CO2Me, and G1t-CO2Me12 were synthesized according to
literature procedures. The platinum complexes P2PtCl2,
P2PtCO3, and P2Pt(BINOL) were synthesized by procedures
similar to those published for dppe analogues.8,24 All chemicals
were purchased from Aldrich except 1,2-bis(dichlorophosphi-
no)ethane (Strem) and S-BINOL (Kankyo Kagaku Center Co.,
Ltd., Japan) and used as received. Before use, acetone was
distilled from CaSO4. Tetrahydrofuran, dichloromethane, and
toluene were passed through a column of activated alumina.
Chlorobenzene was distilled from P2O5 and freeze-pump-
thaw degassed before use in polymerizations, but was used
as received for rebinding experiments. AIBN (2,2′-azobis-
(isobutyro)nitrile) was recrystallized from methanol, dried in
vacuo, and stored under nitrogen at -35 °C. EDMA (ethylene
glycol dimethacrylate) was washed twice with aqueous 1 M
NaOH and once with brine to remove inhibitor, dried over
MgSO4, filtered, distilled (105 °C/1 mmHg), freeze-pump-
thaw degassed, and stored under nitrogen at -35 °C. Poly-
merizations were performed in an MBraun LabMaster 100
glovebox. Reactions performed under nitrogen were carried out
using standard Schlenk techniques.
dppe-OH. dppe-OTBDMS (3.0 g, 3.17 mmol) was dissolved
in 75 mL of THF and cooled to 0 °C. A solution of tetrabutyl-
ammonium fluoride (1.0 M in THF, 16 mL) was added
dropwise. The solution was allowed to stir 16 h and warm to
room temperature. A white precipitate formed. The volume of
the solution was reduced in vacuo to 10 mL, and 25 mL of
ethanol was added to further precipitate the product. The
(22) Cram, D. J.; Sogah, G. D. Y. J. Am. Chem. Soc. 1979, 101,
3035-3042.
(23) McDermott, J. X.; White, J. F.; Whitesides, G. M. J. Am. Chem.
Soc. 1976, 98, 6521-6527.
(24) (a) Gugger, P.; Limmer, S. O.; Watson, A. A.; Willis, A. C.; Wild,
S. B. Inorg. Chem. 1993, 32, 5692-5696. (b) Brunkan, N. M.; White,
P. S.; Gagne´, M. R. Angew. Chem., Int. Ed. 1998, 37, 1579-1582.

6346 Organometallics, Vol. 24, No. 26, 2005
Voshell and Gagne´
white solid was collected by vacuum filtration and dried in
vacuo. The product contained 1 equiv of tetrabutylammonium
ion that could not be removed by repeated precipitation (1.46
g, 64% yield). 1H NMR (400 MHz, DMSO-d6, δ): 7.25 (m, 8H),
6.70 (d, J ) 8.4 Hz, 8 H), 3.17 (m, 8H), 2.00 (br, 4H), 1.55 (m,
8H), 1.31 (q, J ) 7.2 Hz, 8H), 0.94 (t, J ) 7.2 Hz, 12H). 13C
NMR (75.5 MHz, DMSO-d6, δ): 164.9, 134.2, 117.8, 117.7,
114.5, 58.5, 24.0, 21.3, 20.1, 14.4. 31P NMR (162.1 MHz,
DMSO-d6, δ): 13.1.
(162.1 MHz, CDCl3, δ): 27.1 (JPt-P ) 3621 Hz). Anal. Calcd
for C82H60O10P2Pt: C, 67.35; H, 4.14. Found: C, 66.32; H, 4.43.
G1-COOH. G1-CO2Me12 (3.0 g, 7.49 mmol) was suspended
in 125 mL of absolute ethanol. Crushed potassium hydroxide
pellets (1.1 g, 18.7 mmol) were added to the suspension, and
the mixture was refluxed for 16 h. The ethanol was removed
in vacuo, and the residue was taken up in 100 mL of ethyl
acetate and acidified with an equal volume of 1 M hydrochloric
acid. The aqueous layer was removed, and the ethyl acetate
was washed again with 100 mL of 1 M hydrochloric acid, then
with 100 mL of distilled water, and finally with 100 mL of
brine. After drying over magnesium sulfate, the ethyl acetate
was removed in vacuo. The resulting yellow solid was partially
dissolved in dichloromethane, and hexanes was added to
induce precipitation. Upon collection by filtration, 2.3 g of a
white solid was obtained (75% yield). 1H NMR (400 MHz,
CDCl3, δ): 7.42 (d, J ) 8.0 Hz, 4H), 7.37 (d, J ) 8.4 Hz, 4H),
7.34 (d, J ) 2.4 Hz, 2H), 6.82 (t, J ) 2.4 Hz, 1H), 6.71 (dd, J
) 17.6 Hz, 10.8 Hz, 2H), 5.76 (dd, J ) 17.6 Hz, 0.8 Hz, 2H),
5.25 (dd, J ) 10.8 Hz, 0.8 Hz, 2H), 5.06 (s, 4H). 13C NMR (100.6
MHz, CDCl3, δ): 172.0, 160.1, 137.9, 136.7, 136.2, 131.4, 128.1,
126.8, 114.6, 109.3, 108.6, 70.4. Anal. Calcd for C25H22O4: C,
77.70; H, 5.74. Found: C, 77.73; H, 5.77.
G0 (Method A). dppe-OH (0.25 g, 0.342 mmol), 4-vinyl-
benzoic acid (0.25 g, 1.71 mmol), and 4-(dimethylamino)-
pyridine (0.13 g, 1.03 mmol) were dissolved in 40 mL of
dichloromethane. 1,3-Dicyclohexylcarbodiimide (0.35 g, 1.71
mmol) was added, and the mixture was stirred for 16 h at room
temperature under nitrogen. The reaction was then eluted
through a pad of silica with 75 mL of ethyl acetate. After
removal of the solvent in vacuo, methanol was added to the
resultant oil to precipitate a solid. The solid was collected by
vacuum filtration and rinsed with methanol and hexanes. The
solid was dissolved along with 1,4-diazabicyclo[2.2.2]octane
(0.15 g, 1.37 mmol) in 30 mL of toluene, and the solution was
stirred at 35 °C under nitrogen for 16 h. The solution was
reduced in volume to 5 mL, and 25 mL of methanol was added
to precipitate the product. A white solid was collected by
vacuum filtration, rinsed with methanol, and dried in vacuo
G1 (Method A). G1-COOH (0.26 g, 1.71 mmol). Yield: 0.37
g, 56%. 1H NMR (400 MHz, CDCl3, δ): 7.39 (m, 48H), 7.18 (d,
J ) 8.4 Hz, 8H), 6.82 (t, J ) 2.0 Hz, 4H), 6.70 (dd, J ) 17.6,
10.8 Hz, 8H), 5.74 (d, J ) 17.6 Hz, 8H), 5.24 (d, J ) 10.8 Hz,
8 H), 5.05 (s, 16H), 2.12 (br, 4H). 13C NMR (75.5 MHz, CDCl3,
δ): 164.9, 160.1, 151.9, 137.8, 136.7, 136.1, 135.6, 134.3, 131.5,
128.1, 126.8, 122.4, 122.3, 114.6, 109.2, 70.4, 24.6. 31P NMR
(162.1 MHz, CDCl3, δ): -13.7. Anal. Calcd for C126H104O16P2:
C, 78.16; H, 5.41. Found: C, 77.99; H, 5.29.
1
to give G0 (0.22 g, 65% yield). H NMR (400 MHz, CDCl3, δ):
8.12 (d, J ) 8.4 Hz, 8H), 7.49 (d, J ) 8.0 Hz, 8H), 7.42 (m,
8H), 7.19 (d, J ) 8.4 Hz, 8H), 6.76 (dd, J ) 17.6, 10.8 Hz, 4H),
5.89 (d, J ) 17.6 Hz, 4H), 5.41 (d, J ) 10.8 Hz, 4H), 2.12 (br,
4H). 13C NMR (75.5 MHz, CDCl3, δ): 165.0, 151.9, 143.0, 136.3,
135.5, 134.3, 130.9, 128.8, 126.6, 122.3, 117.3, 24.4. 31P NMR
(162.1 MHz, CDCl3, δ): -12.6. Anal. Calcd for C62H48O8P2: C,
75.76; H, 4.92. Found: C, 75.71; H, 4.95.
(G1)PtCl2 (Method B). G1 (0.39 g, 0.200 mmol). Yield: 0.41
g, 92%. 1H NMR (400 MHz, CDCl3, δ): 7.98 (m, 8H), 7.40 (m,
48H), 6.85 (s, 4H), 6.71 (dd, J ) 17.6, 10.8 Hz, 8H), 5.75 (d, J
) 17.6 Hz, 8H), 5.25 (d, J ) 10.8 Hz, 8H), 5.06 (s, 16H), 2.45
(G0)PtCl2 (Method B). A solution of G0 (0.20 g, 0.200
mmol) in 10 mL of dichloromethane was added to a solution
of (COD)PtCl2 (0.075 g, 0.200 mmol) in 10 mL of dichlo-
romethane. The solution was stirred for 1 h before the solvent
was removed in vacuo. Methanol (20 mL) was added, and the
white solid was collected by vacuum filtration, rinsed with
methanol, and dried in vacuo to give (G0)PtCl2 (0.21 g, 85%
yield). 1H NMR (400 MHz, CDCl3, δ): 8.14 (d, J ) 8.0 Hz, 8H),
7.97 (dd, J ) 11.8, 8.4 Hz, 8H), 7.53 (d, J ) 8.0 Hz, 8H), 7.38
(dd, J ) 8.4, 2.0 Hz, 8H), 6.78 (dd, J ) 17.6, 10.8 Hz, 4 H),
5.91 (d, J ) 17.6 Hz, 4H), 5.43 (d, J ) 10.8 Hz, 4H), 2.42 (m,
4H). 31P NMR (162.1 MHz, CDCl3, δ): 40.7 (JPt-P ) 3592 Hz).
5 (Method C). To a solution of (G0)PtCl2 (0.119 g, 0.095
mmol) in 25 mL of dichloromethane was added silver carbonate
(39.4 mg, 0.143 mmol) and 5 drops of distilled water. The
mixture was stirred at room temperature, protected from light
until all the PtCl2 was converted to PtCO3, as monitored by
31P NMR. The mixture was filtered through a pad of Celite to
remove the silver salts. S-BINOL (30.0 mg, 0.105 mmol) was
added to a solution of the PtCO3 in 25 mL of dichloromethane,
and the solution was stirred at room temperature until 31P
NMR showed complete conversion to the Pt(BINOL). The
dichloromethane was removed in vacuo, and 25 mL of diethyl
ether was added. The yellow solid was collected by vacuum
filtration and rinsed with an additional 25 mL of diethyl ether
to remove excess BINOL. After drying overnight in vacuo 5,
(G0)Pt(S-BINOL) (0.117 g, 84% yield), was obtained. 1H NMR
(400 MHz, CDCl3, δ): 8.19 (d, J ) 8.4 Hz, 4H), 8.08 (m, 8H),
7.81 (m, 4H), 7.68 (d, J ) 8.0 Hz, 2H), 7.54 (m, 6H), 7.47 (d, J
) 8.4 Hz, 4H), 7.35 (d, J ) 7.2 Hz, 4H), 7.26 (d, J ) 7.6 Hz,
4H), 7.06 (m, 2H), 6.96 (m, 4H), 6.79 (dd, J ) 17.6, 10.8 Hz,
2H), 6.74 (dd, J ) 17.6, 10.8 Hz, 2H), 6.54 (d, J ) 8.8 Hz, 2H),
5.93 (d, J ) 17.6 Hz, 2H), 5.87 (d, J ) 17.6 Hz, 2H), 5.45 (d,
(m, 4H). 31P NMR (162.1 MHz, CDCl3, δ): 40.6 ppm (JPt-P
3600 Hz).
)
6 (Method C). (G1)PtCl2 (0.210 g, 0.095 mmol). Yield: 0.186
g, 81%. 1H NMR (400 MHz, CDCl3, δ): 8.11 (m, 2H), 7.85 (m,
2H), 7.70 (d, J ) 8.0 Hz, 2H), 7.55 (d, J ) 8.8 Hz, 2H), 7.38
(m, 48H), 7.08 (m, 2H), 6.88 (m, 2H), 6.83 (m, 2H), 6.72 (m,
8H), 6.56 (d, J ) 8.8 Hz, 2H), 5.77 (d, J ) 17.6 Hz, 4H), 5.74
(d, J ) 17.6 Hz, 4H), 5.26 (d, J ) 10.8 Hz, 4H), 5.24 (d, J )
10.8 Hz, 4H), 4.99 (s, 8H), 4.93 (s, 8H), 2.25 (m, 4H). 13C NMR
(75.5 MHz, CDCl3, δ): 164.7, 164.5, 161.6, 160.3, 160.2, 154.2,
154.1, 137.9, 137.8, 136.7, 136.1, 136.0, 135.5, 135.2, 131.12,
131.08, 129.1, 128.2, 128.1, 127.7, 126.8, 126.3, 125.6, 125.0,
124.6, 123.9, 123.1, 122.9, 121.4, 117.6, 109.4, 109.4, 108.5,
70.5, 70.4, 27.6. 31P NMR (162.1 MHz, CDCl3, δ): 27.0 (JPt-P
) 3580 Hz).
G1(2,3)-COOH. Vinyl benzyl chloride (9.0 mL, 64.9 mmol)
was added to a suspension of 2,3-dihydroxybenzoic acid (1.0
g, 6.49 mmol), potassium carbonate (3.6 g, 26.0 mmol), 18-
crown-6 (0.34 g, 1.30 mmol), and tetrabutylammonium iodide
(0.12 g, 0.324 mmol) in 40 mL of acetone. The mixture was
heated to reflux for 16 h under nitrogen until TLC showed
complete consumption of the benzoic acid. Insoluble salts were
removed by vacuum filtration, and the solvent was removed
from the filtrate in vacuo. The residue was then redissolved
in 50 mL of dichloromethane and washed 3 times with 50 mL
of distilled water, then dried over magnesium sulfate. The
dichloromethane was removed in vacuo to yield a yellow oil.
2-Propanol (50 mL) and potassium hydroxide (1.8 g, 32.4
mmol) were added to the oil, and this mixture was heated to
reflux for 16 h. Then an additional 1.8 g (32.4 mmol) of
potassium hydroxide was added, and reflux was continued for
24 h until TLC showed complete consumption of starting
material. After the 2-propanol was removed in vacuo, 50 mL
of ethyl acetate and 50 mL of 1 M hydrochloric acid were
added, and the mixture was stirred for 30 min. The aqueous
J ) 10.8 Hz, 2H), 5.40 (d, J ) 10.8 Hz, 2H), 2.23 (m, 4H). 13
C
NMR (75.5 MHz, CDCl3, δ): 164.8, 164.6, 161.6, 154.2, 154.1,
143.3, 143.2, 136.2, 136.0, 135.5, 135.1, 134.2, 131.1, 131.0,
129.1, 128.31, 128.27, 127.7, 126.73, 126.68, 126.2, 125.6,
124.9, 124.6, 123.7, 123.1, 122.9, 121.4, 117.6, 27.6. 31P NMR

Rigidified Dendritic Structures
Organometallics, Vol. 24, No. 26, 2005 6347
1
layer was removed and the ethyl acetate was washed again
with 50 mL of 1 M hydrochloric acid, then twice with 50 mL
of distilled water, and finally with 50 mL of brine. The ethyl
acetate was dried over magnesium sulfate and filtered. After
removing the ethyl acetate in vacuo, 10 mL of dichloromethane
was added to dissolve the product followed by 20 mL of
hexanes, and the mixture was stored at -26 °C overnight to
induce precipitation. Upon collection by filtration, 1.5 g of a
yellow solid was obtained (58% yield). 1H NMR (400 MHz,
CDCl3, δ): 7.72 (dd, J ) 8.0, 1.6 Hz, 1H), 7.45 (d, J ) 8.0 Hz,
2H), 7.41 (d, J ) 8.0 Hz, 2H), 7.34 (d, J ) 8.0 Hz, 2H), 7.27 (d,
J ) 8.0 Hz, 2H), 7.22 (d, J ) 1.6 Hz, 1H), 7.17 (t, J ) 8.0 Hz,
1H), 6.74 (dd, J ) 17.6, 10.8 Hz, 1H), 6.67 (dd, J ) 17.6, 10.8
Hz, 1H), 5.78 (d, J ) 17.6 Hz, 1H), 5.74 (d, J ) 17.6 Hz, 1H),
5.29 (d, J ) 10.8 Hz, 1H), 5.27 (d, J ) 10.8 Hz, 1H), 5.23 (s,
2H), 5.15 (s, 2H). 13C NMR (100.6 MHz, CDCl3, δ): 165.5;
151.6; 147.4; 138.9; 138.3; 136.6; 136.5; 135.6; 134.4; 129.9;
128.4; 127.0; 125.4; 124.9; 123.4; 119.4; 115.3; 115.0; 71.7.
Anal. Calcd for C25H22O4: C, 77.70; H, 5.74. Found: C, 77.72;
H, 5.82.
G1(2,3) (Method A). G1(2,3)-COOH (0.26 g, 1.71 mmol).
Yield: 0.32 g, 47%. 1H NMR (400 MHz, CDCl3, δ): 7.52 (dd, J
) 7.8, 1.4 Hz, 4H), 7.37 (m, 24H), 7.28 (m, 16H), 7.13 (m, 16H),
6.72 (dd, J ) 17.6, 10.8 Hz, 4H), 6.63 (dd, J ) 17.6, 10.8 Hz,
4H), 5.76 (d, J ) 17.6 Hz, 4H), 5.66 (d, J ) 17.6 Hz, 4H), 5.26
(d, J ) 10.8 Hz, 4H), 5.17 (d, J ) 10.8 Hz, 4H), 5.11 (s, 16H),
2.12 (br, 4H). 13C NMR (75.5 MHz, CDCl3, δ): 164.5, 153.2,
151.8, 149.3, 137.8, 137.6, 137.1, 136.9, 136.7, 136.3, 135.5,
134.3, 129.2, 128.2, 126.8, 126.5, 126.2, 124.4, 123.7, 122.3,
119.0, 114.6, 114.2, 75.9, 71.5, 24.7. 31P NMR (162.1 MHz,
CDCl3, δ): -13.3. Anal.Calcd for C126H104O16P2: C, 78.16; H,
5.41. Found: C, 77.95; H, 5.36.
dried in vacuo to give 2.0 g (85% yield). H NMR (400 MHz,
CDCl3, δ): 7.87 (d, J ) 8.4 Hz, 1H), 7.38 (m, 8H), 6.71 (dd, J
) 17.6, 10.8 Hz, 2H), 6.56 (m, 2H), 5.76 (d, J ) 17.6 Hz, 1H),
5.75 (d, J ) 17.6 Hz, 1H), 5.26 (d, J ) 10.8 Hz, 1H), 5.24 (d,
J ) 10.8 Hz, 1H), 5.12 (s, 2H), 5.02 (s, 2H), 3.86 (s, 3H). 13C
NMR (100.6 MHz, CDCl3, δ): 166.4, 163.4, 160.4, 137.8, 137.3,
136.7, 136.6, 136.4, 135.9, 134.2, 128.0, 127.2, 126.7, 126.6,
114.6, 113.4, 106.3, 101.7, 70.6, 70.2, 51.9.
G1(2,4)-COOH. G1(2,4)-CO2Me (1.5 g, 3.75 mmol) was
suspended in 75 mL of absolute ethanol. Crushed potassium
hydroxide pellets (0.53 g, 9.36 mmol) were added to the
suspension, and the mixture was refluxed for 16 h. The ethanol
was removed in vacuo, and the residue was taken up in 50
mL of ethyl acetate and acidified with an equal volume of 1 M
hydrochloric acid. The aqueous layer was removed, and the
ethyl acetate was washed again with 50 mL of 1 M hydrochlo-
ric acid, then with 50 mL of distilled water, and finally with
50 mL of brine. After drying over magnesium sulfate, the ethyl
acetate was removed in vacuo. The resulting yellow solid was
partially dissolved in dichloromethane, and hexanes was added
to induce precipitation. Upon collection by filtration, 1.0 g of
1
a white solid was obtained (72% yield). H NMR (400 MHz,
CDCl3, δ): 8.13 (d, J ) 8.8 Hz, 1H), 7.38 (m, 8H), 6.70 (m,
3H), 6.65 (d, J ) 2 Hz, 1H), 5.78 (dd, J ) 17.6, 10.8 Hz, 1H),
5.76 (dd, J ) 17.6, 10.8 Hz, 1H), 5.30 (dd, J ) 10.8 Hz, 0.4 Hz,
1H), 5.27 (dd, J ) 10.8 Hz, 0.4 Hz, 1H), 5.20 (s, 2H), 5.08 (s,
2H). 13C NMR (100.6 MHz, CDCl3, δ): 165.4; 164.3; 159.0;
138.9; 138.2; 136.6; 136.4; 136.0; 135.5; 133.8; 128.5; 128.1;
127.3; 126.9; 115.4; 114.9; 111.4; 108.2; 101.1; 72.3; 70.6. Anal.
Calcd for C25H22O4: C, 77.70; H, 5.74. Found: C, 77.71; H,
5.69.
G1(2,4) (Method A). G1(2,4)-COOH (0.26 g, 1.71 mmol).
Yield: 0.34 g, 49%. H NMR (400 MHz, CDCl3, δ): 8.03 (d, J
1
[G1(2,3)]PtCl2 (Method B). G1(2,3) (0.39 g, 0.200 mmol).
Yield: 0.41 g, 93%. 1H NMR (400 MHz, CDCl3, δ): 7.91 (dd, J
) 11.8, 8,4 Hz, 8H), 7.54 (dd, J ) 8.0, 1.6 Hz, 4H), 7.39 (m,
16H), 7.29 (m, 24H), 7.15 (m, 8H), 6.72 (dd, J ) 17.6, 10.8 Hz,
4H), 6.65 (dd, J ) 17.6, 10.8 Hz, 4H), 5.76 (d, J ) 17.6 Hz,
4H), 5.68 (d, J ) 17.6 Hz, 4H), 5.26 (d, J ) 10.8 Hz, 4H), 5.17
(d, J ) 10.8 Hz, 4H), 5.13 (s, 8H), 5.09 (s, 8H), 2.37 (m, 4H).
31P NMR (162.1 MHz, CDCl3, δ): 40.7 ppm (JPt-P ) 3596 Hz).
7 (Method C). [G1(2,3)]PtCl2 (0.210 g, 0.095 mmol). Yield:
0.184 g, 80%. 1H NMR (400 MHz, CDCl3, δ): 8.07 (m, 4H),
7.81 (m, 4H), 7.66 (d, J ) 8.0 Hz, 2H), 7.61 (dd, J ) 7.6, 1.6
Hz, 2H), 7.52 (m, 4H), 7.35 (m, 34H), 7.10 (m, 20H), 6.74 (dd,
J ) 17.6, 10.8 Hz, 2H), 6.71 (dd, J ) 17.6, 10.8 Hz, 2H), 6.62
(dd, J ) 17.6, 10.8 Hz, 2H), 6.60 (dd, J ) 17.6, 10.8 Hz, 2H),
6.53 (d, J ) 8.8 Hz, 2H), 5.78 (d, J ) 17.6 Hz, 2H), 5.76 (d, J
) 17.6 Hz, 2H), 5.64 (d, J ) 17.6 Hz, 2H), 5.63 (d, J ) 17.6
Hz, 2H), 5.28 (d, J ) 10.8 Hz, 2H), 5.26 (d, J ) 10.8 Hz, 2H),
5.14 (m, 20H), 2.25 (m, 4H). 13C NMR (75.5 MHz, CDCl3, δ):
154.1, 153.3, 153.2, 149.5, 149.4, 137.9, 137.8, 137.7, 137.6,
137.0, 136.7, 136.2, 136.0, 135.4, 135.1, 129.3, 129.2, 129.1,
128.2, 128.1, 127.8, 126.80, 126.77, 126.6, 126.5, 126.3, 126.2,
125.7, 125.6, 125.0, 124.53, 124.46, 123.7, 123.2, 123.0, 121.3,
119.3, 119.2, 114.6, 114.4, 114.3, 75.9, 75.8, 71.5, 27.6. 31P
NMR (162.1 MHz, CDCl3, δ): 27.2 (JPt-P ) 3620 Hz). Anal.
Calcd for C146H116O18P2Pt: C, 72.60; H, 4.84. Found: C, 72.09;
H, 4.43.
G1(2,4)-CO2Me. Methyl 2,4-dihydroxy benzoate (1.0 g, 5.95
mmol), potassium carbonate (2.1 g, 14.9 mmol), 18-crown-6
(0.31 g, 1.19 mmol), and tetrabutylammonium iodide (0.11 g,
0.297 mmol) were suspended in 30 mL of acetone. Vinyl benzyl
chloride (1.8 mL, 12.5 mmol) was then added, and the mixture
was refluxed under nitrogen for 16 h. Insoluble salts were
removed by vacuum filtration, and the solvent was removed
from the filtrate in vacuo. The residue was redissolved in 50
mL of ether and washed three times with 50 mL of distilled
water. The ether layer was dried over magnesium sulfate and
evaporated to dryness in vacuo to give a pale yellow oil.
Hexanes (1 mL) was added, and the mixture was stored at
-26 °C for several days to give a white solid. The solid was
) 8.8 Hz, 4H), 7.38 (m, 42H), 7.15 (d, J ) 8.4 Hz, 8H), 6.67
(m, 16H), 5.76 (d, J ) 17.6 Hz, 4H), 5.68 (d, J ) 17.6 Hz, 4H),
5.26 (d, J ) 10.8 Hz, 4H), 5.18 (d, J ) 10.8 Hz, 4H), 5.07 (s,
8H), 5.05 (s, 8H), 2.10 (br, 4H). 13C NMR (75.5 MHz, CDCl3,
δ): 164.1, 163.7, 161.3, 152.0, 138.0, 137.4, 136.8, 136.6, 136.2,
135.8, 135.1, 134.9, 134.2, 128.1, 127.4, 126.8, 126.7, 122.5,
114.7, 114.3, 112.4, 106.5, 101.6, 70.6, 70.3, 24.7. 31P NMR
(162.1 MHz, CDCl3, δ): -13.6. Anal. Calcd for C126H104O16P2:
C, 78.16; H, 5.41. Found: C, 77.89; H, 5.35.
[G1(2,4)]PtCl2 (Method B). G1(2,4) (0.39 g, 0.200 mmol).
1
Yield: 0.40 g, 90%. H NMR (400 MHz, CDCl3, δ): 8.05 (d, J
) 9.2 Hz, 4H), 7.91 (m, 8H), 7.39 (m, 42H), 6.67 (m, 16H), 5.76
(d, J ) 17.6 Hz, 4H), 5.70 (d, J ) 17.6 Hz, 4H), 5.26 (d, J )
10.8 Hz, 4H), 5.19 (d, J ) 10.8 Hz, 4H), 5.13 (s, 8H), 5.08 (s,
8H). 31P NMR (162.1 MHz, CDCl3, δ): 40.8 ppm (JPt-P ) 3608
Hz).
8 (Method C). [G1(2,4)]PtCl2 (0.210 g, 0.095 mmol). Yield:
1
0.195 g, 85%. H NMR (400 MHz, CDCl3, δ): 8.13 (d, J ) 9.2
Hz, 2H), 8.03 (m, 6H), 7.82 (m, 4H), 7.66 (d, J ) 8.0 Hz, 2H),
7.52 (d, J ) 8.8 Hz, 2H), 7.39 (m, 36H), 7.27 (d, J ) 7.6 Hz,
4H), 7.00 (m, 6H), 6.67 (m, 16H), 6.54 (d, J ) 8.4 Hz, 2H),
5.75 (m, 8H), 5.28 (m, 4H), 5.11 (m, 20H), 2.25 (m, 4H). 13C
NMR (75.5 MHz, CDCl3, δ): 163.3, 163.25, 163.16, 161.8,
161.5, 161.4, 154.4, 154.3, 138.0, 137.9, 137.6, 137.4, 136.6,
136.1, 136.0, 135.7, 135.0, 129.1, 128.1, 128.1, 127.6, 127.4,
126.82, 126.79, 126.7, 126.3, 125.7, 124.9, 124.5, 123.3, 123.1,
121.2, 114.7, 114.4, 114.3, 111.7, 106.6, 101.7, 101.6, 70.6, 70.4,
27.6. 31P NMR (162.1 MHz, CDCl3, δ): 27.3 (JPt-P ) 3621 Hz).
Anal. Calcd for C146H116O18P2Pt: C, 72.60; H, 4.84. Found: C,
71.20; H, 4.19.
G1t-COOH. G1t-CO2Me12 (1.5 g, 2.82 mmol) was suspended
in 50 mL of absolute ethanol. Crushed potassium hydroxide
pellets (0.40 g, 7.04 mmol) were added to the suspension, and
the mixture was refluxed for 16 h. The ethanol was removed
in vacuo, and the residue was taken up in 50 mL of ethyl
acetate and acidified with an equal volume of 1 M hydrochloric
acid. The aqueous layer was removed, and the ethyl acetate
was washed again with 50 mL of 1 M hydrochloric acid, then

6348 Organometallics, Vol. 24, No. 26, 2005
Voshell and Gagne´
with 50 mL of distilled water, and finally with 50 mL of brine.
After drying over magnesium sulfate, the ethyl acetate was
removed in vacuo. The resulting yellow solid was partially
dissolved in dichloromethane, and hexanes was added to
induce precipitation. Upon collection by filtration, 0.91 g of a
white solid was obtained (63% yield). 1H NMR (400 MHz,
CDCl3, δ): 7.38 (m, 10H), 7.32 (d, J ) 8.0 Hz, 2H), 7.27 (d, J
) 8.4 Hz, 2H), 6.72 (dd, J ) 17.6 Hz, 10.8 Hz, 2H), 6.86 (dd,
J ) 16.4 Hz, 10.8 Hz, 1H), 5.76 (d, J ) 17.6 Hz, 2H), 5.72 (d,
J ) 16.4 Hz, 1H), 5.26 (d, J ) 10.8 Hz, 2H), 5.23 (d, J ) 10.8
Hz, 1H), 5.11 (s, 6H). 13C NMR (100.6 MHz, CDCl3, δ): 171.5,
152.9, 143.4, 137.7, 137.6, 137.3, 136.9, 136.8, 136.4, 129.1,
128.1, 126.8, 126.4, 124.4, 114.5, 114.3, 110.0, 75.2, 71.3. Anal.
Calcd for C34H30O5: C, 78.74; H, 5.83. Found: C, 78.60; H,
5.84.
G1t (Method A). G1t-COOH (0.89 g, 1.71 mmol). Yield:
0.41 g, 47%. 1H NMR (400 MHz, CDCl3, δ): 7.47 (s, 8H), 7.37
(m, 40H), (7.28 m, 16H), 7.16 (d, J ) 8.4 Hz, 8 H), 6.69 (m,
12H), 5.74 (d, J ) 17.6 Hz, 8H), 5.72 (d, J ) 17.6 Hz, 4H),
5.24 (d, J ) 10.8 Hz, 8H), 5.23 (d, J ) 10.8 Hz, 4H), 5.10 (s,
24H), 2.12 (br, 4H). 13C NMR (75.5 MHz, CDCl3, δ): 164.8,
152.9, 151.9, 143.3, 137.7, 137.6, 137.2, 136.9, 136.8, 136.3,
135.6, 134.4, 129.1, 128.1, 126.7, 126.4, 124.5, 122.3, 114.5,
114.2, 109.9, 75.2, 71.3, 24.7. 31P NMR (162.1 MHz, CDCl3,
δ): -13.5. Anal. Calcd for C162H136O20P2: C, 78.94; H, 5.56.
Found: C, 78.69; H, 5.43.
(G1t)PtCl2 (Method B). G1t (0.49 g, 0.200 mmol). Yield:
0.51 g, 94%. 1H NMR (400 MHz, CDCl3, δ): 7.95 (m, 8H), 7.48
(s, 8H), 7.39 (m, 56H), 6.70 (m, 12H), 5.75 (d, J ) 17.6 Hz,
8H), 5.73 (d, J ) 17.6 Hz, 4H), 5.25 (d, J ) 10.8 Hz, 8H), 5.24
(d, J ) 10.8 Hz, 4H), 5.14 (s, 24H), 2.42 (m, 4H). 31P NMR
(162.1 MHz, CDCl3, δ): 40.9 (JPt-P ) 3597 Hz).
9 (Method C). (G1t)PtCl2 (0.260 g, 0.095 mmol). Yield:
0.243 g, 87%. 1H NMR (400 MHz, CDCl3, δ): 8.10 (m, 4H),
7.87 (m, 4H), 7.70 (d, J ) 8.4 Hz, 2H), 7.55 (m, 6H), 7.35 (m,
60H), 7.10 (m, 2H), 7.01 (m, 4H), 6.70 (m, 12H), 6.58 (d, J )
8.4 Hz, 2H), 5.74 (m, 12H), 5.25 (m, 12H), 5.18 (s, 8H), 5.17
(s, 4H), 5.12 (s, 4H), 5.11 (s, 8H), 2.27 (m, 4H). 13C NMR (75.5
MHz, CDCl3, δ): 164.6, 164.4, 161.6, 154.2, 154.2, 153.0, 152.9,
143.5, 143.4, 137.8, 137.7, 137.6, 137.6, 137.1, 136.9, 136.7,
136.6, 136.2, 136.2, 136.0, 135.3, 129.11, 129.09, 128.14,
128.09, 126.7, 126.4, 125.0, 124.7, 124.0, 123.1, 122.9, 121.4,
114.5, 114.3, 110.0, 75.24, 75.21, 71.4, 71.3, 27.5. 31P NMR
(162.1 MHz, CDCl3, δ): 27.0 (JPt-P ) 3577 Hz). Anal. Calcd
for C182H148O22P2Pt: C, 74.25; H, 5.07. Found: C, 73.29; H,
4.80.
G1t(2,3,4)-COOH. Vinyl benzyl chloride (8.4 mL, 58.8
mmol) was added to a suspension of 2,3,4-trihydroxybenzoic
acid (1.0 g, 5.88 mmol), potassium carbonate (3.3 g, 23.5 mmol),
18-crown-6 (0.31 g, 1.18 mmol), and tetrabutylammonium
iodide (0.11 g, 0.294 mmol) in 60 mL of acetone. The mixture
was heated to reflux for 16 h under nitrogen until TLC showed
complete consumption of the benzoic acid. Insoluble salts were
removed by vacuum filtration, and the solvent was removed
from the filtrate in vacuo. The residue was then redissolved
in 50 mL of dichloromethane and washed three times with 50
mL of distilled water, then dried over magnesium sulfate. The
dichloromethane was removed in vacuo to yield a yellow oil.
2-Propanol (100 mL) and potassium hydroxide (1.7 g, 29.4
mmol) were added to the oil, and this mixture was heated to
reflux for 16 h. Then an additional 1.7 g (29.4 mmol) of
potassium hydroxide was added, and reflux was continued for
24 h until TLC showed complete consumption of starting
material. After the 2-propanol was removed in vacuo, 100 mL
of ethyl acetate and 100 mL of 1 M hydrochloric acid were
added, and the mixture was stirred for 30 min. The aqueous
layer was removed, and the ethyl acetate was washed again
with 100 mL of 1 M hydrochloric acid, then twice with 100
mL of distilled water, and finally with 100 mL of brine. The
ethyl acetate was dried over magnesium sulfate and filtered.
After removing the ethyl acetate in vacuo, 10 mL of dichlo-
romethane was added to dissolve the product, followed by 20
mL of hexanes, and the mixture was stored at -26 °C
overnight to induce precipitation. Upon collection by filtration,
1.4 g of a yellow solid was obtained (46% yield). 1H NMR (400
MHz, CDCl3, δ): 7.87 (d, J ) 8.8 Hz, 1H); 7.35 (m, 12H); 6.87
(d, J ) 8.8 Hz, 1H); 6.72 (m, 3H); 5.76 (m, 3H); 5.27 (m, 3H);
5.25 (s, 2H); 5.15 (s, 2H); 5.05 (s, 2H). 13C NMR (100.6 MHz,
CDCl3, δ): 165.4; 157.7; 152.2; 141.1; 138.9; 138.14; 138.10;
136.7; 136.6; 136.50; 136.47; 135.5; 134.3; 129.9; 129.3; 128.9;
128.2; 127.0; 126.9; 126.6; 115.7; 115.3; 114.9; 126.6; 115.7;
115.3; 114.9; 114.7; 110.0; 77.8, 76.0, 71.2. Anal. Calcd for
C34H30O5: C, 78.74; H, 5.83. Found: C, 78.94; H, 5.84.
G1t(2,3,4) (Method A). G1t(2,3,4)-COOH (0.89 g, 1.71
mmol). Yield: 0.43 g, 49%. 1H NMR (400 MHz, CDCl3, δ): 7.77
(d, J ) 8.4 Hz, 4H), 7.32 (m, 56H), 7.13 (d, J ) 8.4 Hz, 8H),
6.70 (m, 16H), 5.77 (d, J ) 17.6 Hz, 4H), 5.73 (d, J ) 17.6 Hz,
4H), 5.68 (d, J ) 17.6 Hz, 4H), 5.27 (d, J ) 10.8 Hz, 4H), 5.24
(d, J ) 10.8 Hz, 4H), 5.18 (d, J ) 10.8 Hz, 4H), 5.091 (s, 8H),
5.089 (s, 8H), 5.00 (s, 8H), 2.13 (br, 4H). 13C NMR (75.5 MHz,
CDCl3, δ): 163.7, 157.5, 154.8, 151.9, 143.0, 137.9, 137.69,
137.67, 137.0, 136.9, 136.7, 135.8, 135.3, 134.3, 129.4, 129.3,
128.0, 126.8, 126.5, 126.5, 122.4, 118.0, 114.7, 114.3, 109.0,
76.4, 75.7, 71.0, 24.7. 31P NMR (162.1 MHz, CDCl3, δ): -13.0.
Anal. Calcd for C162H136O20P2: C, 78.94; H, 5.56. Found: C,
78.76; H, 5.49.
[G1t(2,3,4)]PtCl2 (Method B). G1t(2,3,4) (0.49 g, 0.200
mmol). Yield: 0.48 g, 88%. 1H NMR (400 MHz, CDCl3, δ): 7.93
(m, 8H), 7.80 (d, J ) 8.8 Hz, 4H), 7.33 (m, 56H), 6.83 (d, J )
8.8 Hz, 4H), 6.71 (m, 12H), 5.78 (d, J ) 17.6 Hz, 4H), 5.73 (d,
J ) 17.6 Hz, 4H), 5.71 (d, J ) 17.6 Hz, 4H), 5.28 (d, J ) 10.8
Hz, 4H), 5.24 (d, J ) 10.8 Hz, 4H), 5.19 (d, J ) 10.8 Hz, 4H),
5.15 (s, 8H), 5.14 (s, 8H), 5.02 (s, 8H), 2.40 (m, 4H). 31P NMR
(162.1 MHz, CDCl3, δ): 40.9 (JPt-P ) 3596 Hz).
10. To a solution of [G1t(2,3,4)]PtCl2 (0.260 g, 0.095 mmol)
in 25 mL of dichloromethane were added silver carbonate (39.4
mg, 0.143 mmol) and 5 drops of distilled water. The mixture
was stirred at room temperature, protected from light until
all the PtCl2 was converted to PtCO3 as monitored by 31P NMR.
The mixture was filtered through a pad of Celite to remove
the silver salts. S-BINOL (30.0 mg, 0.105 mmol) was added
to a solution of the PtCO3 in 25 mL of dichloromethane, and
the solution was stirred at room temperature until 31P NMR
showed complete conversion to the Pt(BINOL). The dichlo-
romethane was removed in vacuo, and 25 mL of diethyl ether
was added. The ether was decanted from the yellow oil, and
this process was repeated three times. After drying overnight
in vacuo 10 [G1t(2,3,4)]Pt(S-BINOL) (0.233 g, 84% yield) was
obtained as a foamy solid from which complete solvent removal
was impossible. 1H NMR (400 MHz, CDCl3, δ): 8.08 (m, 4H),
7.87 (d, J ) 8.8 Hz, 2H), 7.81 (m, 4H), 7.76 (d, J ) 8.8 Hz,
2H), 7.54 (d, J ) 8.8 Hz, 2H), 7.31 (m, 58H), 7.03 (m, 6H),
6.87 (d, J ) 8.8 Hz, 2H), 6.70 (m, 12H), 6.54 (d, J ) 8.8 Hz,
2H), 5.73 (m, 12H), 5.28 (m, 8H), 5.13 (m, 28H), 2.23 (m, 4H).
13C NMR (75.5 MHz, CDCl3, δ): 155.0, 154.9, 154.3, 154.2,
153.2, 143.1, 143.0, 138.0, 137.9, 137.8, 137.71, 137.66, 137.0,
136.92, 136.85, 136.7, 136.0, 135.8, 135.5, 135.0, 134.0, 129.4,
129.3, 129.1, 128.08, 128.05, 126.8, 126.6, 126.51, 126.47,
125.0, 124.6, 123.2, 123.1, 121.3, 117.5, 117.4, 114.7, 114.3,
109.2, 76.4, 75.71, 75.66, 71.0. 31P NMR (162.1 MHz, CDCl3,
δ): 27.5 (JPt-P ) 3630 Hz).
G1n(3,5)-COOH. Vinyl benzyl chloride (7.0 mL, 49.0 mmol)
was added to a suspension of 3,5-dihydroxy-2-naphthoic acid
(1.0 g, 4.90 mmol), potassium carbonate (2.7 g, 19.6 mmol),
18-crown-6 (0.26 g, 0.979 mmol), and tetrabutylammonium
iodide (91 mg, 0.249 mmol) in 30 mL of acetone. The mixture
was heated to reflux for 16 h under nitrogen until TLC showed
complete consumption of the naphthoic acid. Insoluble salts
were removed by vacuum filtration, and the solvent was
removed from the filtrate in vacuo. The residue was then
redissolved in 50 mL of dichloromethane and washed three
times with 50 mL of distilled water, then dried over magne-

Rigidified Dendritic Structures
Organometallics, Vol. 24, No. 26, 2005 6349
sium sulfate. The dichloromethane was removed in vacuo to
yield a yellow oil. 2-Propanol (50 mL) and potassium hydroxide
(1.4 g, 24.5 mmol) were added to the oil, and this mixture was
heated to reflux for 16 h. Then an additional 1.4 g (24.5 mmol)
of potassium hydroxide was added, and reflux was continued
for 24 h until TLC showed complete consumption of starting
material. After the 2-propanol was removed in vacuo, 50 mL
of ethyl acetate and 50 mL of 1 M hydrochloric acid were
added, and the mixture was stirred for 30 min. The aqueous
layer was removed, and the ethyl acetate was washed again
with 50 mL of 1 M hydrochloric acid, then twice with 50 mL
of distilled water, and finally with 50 mL of brine. The ethyl
acetate was dried over magnesium sulfate and filtered. After
removing the ethyl acetate in vacuo, 10 mL of dichloromethane
was added to dissolve the product, followed by 20 mL of
hexanes, and the mixture was stored at -26 °C overnight to
induce precipitation. Upon collection by filtration, 0.87 g of a
yellow solid was obtained (41% yield). 1H NMR (400 MHz,
CDCl3, δ): 8.74 (s, 1H), 7.81 (s, 1H); 7.45 (m, 9H), 7.32 (t, J )
8.0 Hz, 1H), 6.96 (d, J ) 8.0 Hz, 1H), 6.72 (m, 2H), 5.79 (d, J
) 17.6 Hz, 1H), 5.77 (d, J ) 17.6 Hz, 1H), 5.34 (s, 2H), 5.30
(d, J ) 10.8 Hz, 1H), 5.29 (d, J ) 10.8 Hz, 1H), 5.22 (s, 2H).
13C NMR (100.6 MHz, CDCl3, δ): 165.8; 153.6; 153.5; 138.7;
137.9; 136.6; 136.5; 136.4; 136.1; 134.2; 129.8; 129.1; 128.7;
128.1; 127.2; 126.8; 125.7; 122.1; 118.9; 115.3; 114.8; 108.8;
104.0; 72.4; 70.6. Anal. Calcd for C29H24O4: C, 79.80; H, 5.54.
Found: C, 79.51; H, 5.54.
broken into ∼2 mm pieces with a mortar and pestle. The
polymers contained 60-65 µmol Pt/g polymer.
BINOL Loss Experiment. The solvent from the Soxhlet
extraction filtrate was removed in vacuo, and the residue was
quantitatively transferred into a 5 mL volumetric flask and
diluted to the mark with HPLC grade 2-propanol. The con-
centration of BINOL in the solution was determined by HPLC
analysis.
HCl Cleavage Experiment. Polymer (0.25 g) was weighed
quantitatively and suspended in 20 mL of dichloromethane.
Concentrated hydrochloric acid (10 drops) was added, and the
mixture was stirred for 6 h. The polymer was filtered through
a Soxhlet frit and rinsed with an additional 20 mL of CH2Cl2.
The polymer was Soxhlet extracted for 16 h with the filtrate.
The solvent was removed from the filtrate in vacuo, and the
residue was quantitatively transferred to a 10 mL volumetric
flask and diluted to the mark with HPLC grade 2-propanol.
The concentration of BINOL in the solution was determined
by HPLC analysis.
Rebinding Experiment. Polymer (0.30 g) and rac-Br2-
BINOL (0.15 g) were weighed quantitatively and added to a
Schlenk tube and were dried under vacuum for 1 h. Under
nitrogen, 3.0 mL of chlorobenzene and 0.50 mL of distilled
water were added. The tube was sealed and placed in a 60 or
120 °C oil bath for 12 h. The polymer was filtered through a
Soxhlet frit and rinsed with 40 mL of dichloromethane, then
Soxhlet extracted for 14 h. After drying for 1 h in vacuo, the
polymer was subjected to a standard HCl cleavage experiment.
The residue from the Soxhlet extraction was dissolved in 2.00
mL of 2-propanol and analyzed by HPLC.
dppe-OCO-Ph. To a suspension of dppe-OH (0.25 g, 0.342
mmol) in 20 mL of dichloromethane was added benzoyl
chloride (0.20 mL, 1.71 mmol), followed by triethylamine (0.25
mL, 1.71 mmol). The suspension was stirred for 1 h under
nitrogen until TLC showed the consumption of dppe-OH and
the formation of a product. The mixture was eluted through a
pad of silica with 50 mL of ethyl acetate. The solvent was
removed in vacuo to give a white solid. The solid was collected
by vacuum filtration and rinsed with methanol and hexanes.
The solid was dissolved along with DABCO (0.15 g, 1.37 mmol)
in 30 mL of toluene, and the solution was stirred at 35 °C
under nitrogen for 16 h. The solution was reduced in volume
to 10 mL, and 25 mL of methanol was added to precipitate
the product. A white solid was collected by vacuum filtration,
rinsed with methanol, and dried in vacuo to give dppe-OCO-
Ph (0.18 g, 60% yield) with a small amount of toluene
remaining. 1H NMR (400 MHz, CDCl3, δ): 8.17 (d, J ) 7.2
Hz, 8H), 7.62 (t, J ) 7.6 Hz, 4H), 7.48 (t, J ) 7.6 Hz, 8H), 7.42
(m, 8H), 7.21 (d, J ) 8.4 Hz, 8H), 2.13 (br, 4H). 13C NMR (75.5
MHz, CDCl3, δ):165.3, 151.9, 135.6, 134.4, 130.6, 130.5, 129.7,
128.9, 122.3, 24.7. 31P NMR (162.1 MHz, CDCl3, δ): -13.6.
Anal. Calcd for C54H40O8P2: C, 73.80; H, 4.59. Found: C, 72.90;
H, 4.64.
G1n(3,5) (Method A). G1n(3,5)-COOH (0.75 g, 1.71 mmol).
Yield: 0.41 g, 53%. 1H NMR (400 MHz, CDCl3, δ): 8.46 (s,
4H), 7.72 (s, 4H), 7.32 (m, 56H), 6.91 (d, J ) 7.6 Hz, 4H), 6.75
(dd, J ) 17.6, 10.8 Hz, 4H), 6.66 (dd, J ) 17.6, 10.8 Hz, 4H),
5.79 (d, J ) 17.6 Hz, 4 H), 5.69 (d, J ) 17.6 Hz, 4H), 5.22 (m,
24H), 2.17 (br, 4H). 13C NMR (75.5 MHz, CDCl3, δ): 164.6,
155.0, 153.4, 152.0, 137.7, 137.4, 136.8, 136.7, 136.5, 135.5,
134.3, 133.6, 133.5, 128.9, 127.9, 127.7, 126.8, 126.7, 124.7,
122.4, 121.7, 121.6, 114.6, 114.2, 108.2, 103.8, 103.7, 70.6, 70.4,
25.3. 31P NMR (162.1 MHz, CDCl3, δ): -12.9. Anal. Calcd for
C142H112O16P2: C, 79.68; H, 5.25. Found: C, 79.72; H, 5.26.
[G1n(3,5)]PtCl2 (Method B). G1n(3,5) (0.49 g, 0.200
mmol). Yield: 0.42 g, 88%. 1H NMR (400 MHz, CDCl3, δ): 8.49
(s, 4H), 7.98 (m, 8H), 7.45 (s, 4H), 7.39 (m, 48H), 6.92 (d, J )
7.6 Hz, 4H), 6.72 (m, 8H), 5.79 (d, J ) 17.6 Hz, 4H), 5.71 (d,
J ) 17.6 Hz, 4H), 5.25 (m, 24H), 2.43 (m, 4H). 31P NMR (162.1
MHz, CDCl3, δ): 40.9 (JPt-P ) 3601 Hz).
11 (Method C). [G1n(3,5)]PtCl2 (0.229 g, 0.095 mmol).
1
Yield: 0.171 g, 69%. H NMR (400 MHz, CDCl3, δ): 8.58 (s,
2H), 8.46 (s, 2H), 8.12 (m, 4H), 7.87 (m, 4H), 7.80 (s, 2H), 7.73
(s, 2H), 7.68 (d, J ) 8.0 Hz, 2H), 7.56 (d, J ) 8.8 Hz, 2H), 7.42
(m, 48H), 7.02 (m, 8H), 6.92 (d, J ) 7.6 Hz, 2H), 6.76 (dd, J )
17.6, 10.8 Hz, 2H), 6.74 (dd, J ) 17.6, 10.8 Hz, 2H), 6.64 (dd,
J ) 17.6, 10.8 Hz, 2H), 6.62 (dd, J ) 17.6, 10.8 Hz, 2H), 6.58
(d, J ) 8.4 Hz, 2H), 5.80 (d, J ) 17.6 Hz, 2H), 5.78 (d, J )
17.6 Hz, 2H), 5.66 (d, J ) 17.6 Hz, 2H), 5.65 (d, J ) 17.6 Hz,
2H), 5.25 (m, 24H), 2.29 (m, 4H). 13C NMR (75.5 MHz, CDCl3,
δ): 154.4, 154.3, 153.5, 153.4, 153.3, 137.7, 137.5, 137.4, 136.7,
136.5, 136.0, 135.5, 135.1, 134.0, 133.9, 130.9, 129.4, 129.2,
129.1, 128.93, 128.86, 128.0, 127.8, 126.8, 126.7, 126.3, 125.5,
125.0, 124.9, 124.6, 123.6, 123.3, 123.1, 121.7, 121.4, 121.1,
118.4, 112.6, 114.3, 112.4, 108.4, 104.0, 103.9, 70.7, 70.4, 27.7.
31P NMR (162.1 MHz, CDCl3, δ): 28.2 (JPt-P ) 3645 Hz). Anal.
Calcd for C162H124O18P2Pt: C, 74.39; H, 4.78. Found: C, 71.75;
H, 4.03.
Polymerization. AIBN (6.8 mg, 41 µmol), metallomonomer
(62 µmol), EDMA (0.80 g, 4.0 mmol), and chlorobenzene (0.76
mL) were weighed quantitatively into a scintillation vial. The
vial was sealed, and the mixture was shaken to completely
dissolve the template. The vial was then placed in a 60 °C
heating block for 24 h. The polymers were removed from the
glovebox and manually broken into 3-4 pieces, then Soxhlet
extracted with 40 mL of dichloromethane for 6 h. After
extraction, the polymer was dried in vacuo for 16 h before being
(dppe-OCO-Ph)PtCl2. A solution of dppe-OCO-Ph (0.15 g,
0.17 mmol) in 10 mL of dichloromethane was added to a
solution of (cod)PtCl2 (0.067 g, 0.17 mmol) in 10 mL of
dichloromethane. The solution was stirred for 1 h before the
solvent was removed in vacuo. Methanol (25 mL) was added,
and the white solid was collected by vacuum filtration, rinsed
with methanol, and dried in vacuo to give (dppe-OCO-Ph)PtCl2
1
(0.16 g, 82% yield). H NMR (400 MHz, CDCl3, δ): 8.18 (d, J
) 8.4 Hz, 8H), 7.97 (dd, J ) 11.8, 8.4 Hz, 8H), 7.65 (t, J ) 7.2
Hz, 4H), 7.51 (t, J ) 8.0 Hz, 8H), 7.38 (dd, J ) 8.4, 2.0 Hz,
8H), 2.43 (m, 4H). 31P NMR (162.1 MHz, CDCl3, δ): 40.8 ppm
(JPt-P ) 3598 Hz).
12. To a solution of (dppe-OCO-Ph)PtCl2 (0.100 g, 0.0873
mmol) in 25 mL of dichloromethane was added silver carbonate
(36.1 mg, 0.131 mmol) and 5 drops of distilled water. The
mixture was stirred at room temperature, protected from light
until all the PtCl2 was converted to PtCO3 as monitored by
31P NMR. The mixture was filtered through a pad of Celite to

6350 Organometallics, Vol. 24, No. 26, 2005
Voshell and Gagne´
remove the silver salts. S-BINOL (27.5 mg, 0.0961 mmol) was
added to a solution of the PtCO3 in 25 mL of dichloromethane,
and the solution was stirred until 31P NMR showed complete
conversion to the Pt(BINOL). The dichloromethane was re-
moved in vacuo, and 25 mL of diethyl ether was added. The
yellow solid was collected by vacuum filtration and rinsed with
an additional 25 mL of diethyl ether to remove excess BINOL.
After drying overnight in vacuo (dppe-OCO-Ph)Pt(S-BINOL)
(0.0886 g, 73% yield) was obtained. 1H NMR (400 MHz, CDCl3,
δ): 8.25 (d, J ) 7.2 Hz, 4H), 8.15 (d, J ) 7.2 Hz, 4H), 8.10 (dd,
J ) 11.2, 8.4 Hz, 4H), 7.81 (dd, J ) 11.2, 8.4 Hz, 4H), 7.68 (t,
J ) 7.6 Hz, 4H), 7.61 (t, J ) 7.6 Hz, 2H), 7.55 (m, 10H), 7.48
(t, J ) 7.6 Hz, 4H), 7.34 (d, J ) 7.6 Hz, 4H), 7.29 (d, J ) 7.6
Hz, 4H), 2.26 (m, 4H). 13C NMR (75.5 MHz, CDCl3, δ): 165.0,
164.9, 161.5, 154.3, 154.2, 136.0, 135.6, 135.5, 135.2, 134.3,
130.8, 130.6, 129.32, 129.26, 129.2, 129.0, 127.7, 127.1, 126.3,
125.5, 125.0, 124.6, 123.6, 123.2, 122.9, 121.4, 118.4, 27.749.
31P NMR (162.1 MHz, CDCl3, δ): 27.7 (JPt-P ) 3619 Hz). Anal.
Calcd for C74H52O10P2Pt: C, 65.44; H, 3.86. Found: C, 65.56;
H, 3.84.
Thermolysis of Model Metallomonomer. To a Schlenk
tube was added 47.3 mg of 12 (36.8 mmol), and the compound
was dried under vacuum for 1 h. Under nitrogen, 6.0 mL of
chlorobenzene and 1.0 mL of distilled water were added to form
a yellow solution. The tube was sealed and placed in a 120 °C
oil bath for 14 h. The brown solution was transferred to a flask,
and the solvent was removed in vacuo. Dichloromethane (2
mL) was added, and the thermolysis products were analyzed
by 31P NMR.
Acknowledgment. We wish to thank the National
Science Foundation (CHE-0315203) for support of this
research project.
Supporting Information Available: Sample Boltzmann
distribution analysis for dendron models. 13C NMR spectra for
dppe-OTBDMS and dppe-OH. This material is available free
of charge via the Internet at http://pubs.acs.org.
OM050647Y